Bone Morphogenetic Protein 2 Stimulates Osteoclast ... - CiteSeerX

0013-7227/01/$03.00/0 Printed in U.S.A.

Endocrinology 142(8):3656 –3662 Copyright © 2001 by The Endocrine Society

Bone Morphogenetic Protein 2 Stimulates Osteoclast Differentiation and Survival Supported by Receptor Activator of Nuclear Factor-␬B Ligand KANAMI ITOH, NOBUYUKI UDAGAWA, TAKENOBU KATAGIRI, SHUNICHIRO IEMURA, NAOTO UENO, HISATAKA YASUDA, KANJI HIGASHIO, JULIAN M. W. QUINN, MATTHEW T. GILLESPIE, T. JOHN MARTIN, TATSUO SUDA, AND NAOYUKI TAKAHASHI Department of Biochemistry, Showa University School of Dentistry (K.I., N.U., T.K., T.S., N.T.), Tokyo 142-8555; Division of Morphogenesis, Department of Developmental Biology, National Institute for Basic Biology (S.I., N.U.), Okazaki 444-8585; Center for Experimental Medicine, Institute of Medical Science, University of Tokyo (H.Y.), Tokyo 108-8639; Snow Brand Milk Products Co., Ltd. (K.H.), Tochigi 329-0512, Japan; and St. Vincent’s Institute of Medical Research (J.M.W.Q., M.T.G., T.J.M.), Fitzroy, Victoria 3065, Australia Bone is a major storage site for TGF␤ superfamily members, including TGF␤ and bone morphogenetic proteins. It is believed that these cytokines are released from bone during bone resorption. Recent studies have shown that both RANKL and macrophage colony-stimulating factor are two essential factors produced by osteoblasts for inducing osteoclast differentiation. In the present study we examined the effects of bone morphogenetic protein-2 on osteoclast differentiation and survival supported by RANKL and/or macrophage colonystimulating factor. Mouse bone marrow-derived macrophages differentiated into osteoclasts in the presence of RANKL and macrophage colony-stimulating factor. TGF␤ superfamily members such as bone morphogenetic protein-2, TGF␤, and activin A markedly enhanced osteoclast differentiation induced by RANKL and macrophage colony-stimulating factor, although each cytokine alone failed to induce osteoclast differentiation in the absence of RANKL. Addition of a soluble form of bone morphogenetic protein receptor type IA to the culture markedly inhibited not only osteoclast formation in-

I

T IS BELIEVED that bone-resorbing osteoclasts are derived from hemopoietic progenitors of the monocytemacrophage lineage. We have developed a mouse coculture system of hemopoietic cells and primary osteoblasts to investigate the regulatory mechanism of osteoclast formation (1, 2). A series of experiments has established the concept that osteoblasts/stromal cells are essentially involved in the differentiation of hemopoietic progenitors into osteoclasts (1– 4). Macrophage colony-stimulating factor (M-CSF; also called CSF-1) was first shown to be an essential factor produced by osteoblasts/stromal cells for osteoclast development (5). It was also hypothesized that osteoblasts/stromal cells express another essential factor, called osteoclast dif-

Abbreviations: BMP, Bone morphogenetic protein; BMPR, BMP receptor; 1␣,25-(OH)2D3, 1␣,25-dihydroxyvitamin D3; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; GM-CSF, granulocyte-macrophage colony-stimulating factor; M-CSF, macrophage colony-stimulating factor; MNC, multinucleated cell; NF-␬B, nuclear factor-␬B; ODF, osteoclast differentiation factor; sBMPR, soluble BMPR; TRAP, tartrateresistant acid phosphatase.

duced by RANKL and bone morphogenetic protein-2, but also the basal osteoclast formation supported by RANKL alone. Either RANKL or macrophage colony-stimulating factor stimulated the survival of purified osteoclasts. Bone morphogenetic protein-2 enhanced the survival of purified osteoclasts supported by RANKL, but not by macrophage colony-stimulating factor. Both bone marrow macrophages and mature osteoclasts expressed bone morphogenetic protein-2 and bone morphogenetic protein receptor type IA mRNAs. An EMSA revealed that RANKL activated nuclear factor-␬B in purified osteoclasts. Bone morphogenetic protein-2 alone did not activate nuclear factor-␬B, but rather inhibited the activation of nuclear factor-␬B induced by RANKL in purified osteoclasts. These findings suggest that bone morphogenetic proteinmediated signals cross-communicate with RANKL-mediated ones in inducing osteoclast differentiation and survival. The enhancement of RANKL-induced survival of osteoclasts by bone morphogenetic protein-2 appears unrelated to nuclear factor-␬B activation. (Endocrinology 142: 3656 –3662, 2001)

ferentiation factor (ODF), as a membrane-bound cytokine in response to bone-resorbing factors such as 1␣,25-dihydroxyvitamin D3 [1␣,25-(OH)2D3] and PTH. We recently succeeded in molecular cloning of ODF from a cDNA library of ST2 cells, which supports osteoclast formation in the cocultures (6). The deduced amino acid sequence of ODF showed that this factor is a member of the TNF ligand family and is identical to RANKL, TNF-related activation-induced cytokine, and OPG ligand, which were independently identified by other studies (7–9). A soluble form of RANKL together with M-CSF induced osteoclast differentiation from mouse hemopoietic cells and human peripheral blood mononuclear cells even in the absence of osteoblasts/stromal cells (6, 9 –11). Thus, RANKL is a cytokine essential for inducing osteoclast differentiation. This idea was further supported by the findings that targeted disruption of the gene encoding either RANKL or RANK similarly leads to severe osteopetrosis with a complete absence of osteoclasts in the deficient mice (12–14). OPG is a soluble decoy receptor for RANKL that inhibits osteoclast

3656

Itoh et al. • Role of BMP in Osteoclast Formation and Function

differentiation and function induced by RANKL (6, 9 –11, 15–18). OPG knockout (⫺/⫺) mice exhibited severe osteoporosis (19, 20). Administration of OPG to mice or rats strongly inhibited osteoclastic bone resorption and increased bone mineral density (16, 18). These findings indicate that OPG functions as a negative regulator of osteoclast differentiation and function. Interestingly, the expression of RANKL, M-CSF, and OPG is recognized in many types of tissues, indicating that the expression of RANKL is not specific to bone tissues (15–18). This suggests that a factor(s) other than RANKL could determine the precise appearance of osteoclasts in bone. Bone morphogenetic proteins (BMPs) were first identified as cytokines that induce ectopic bone formation in vivo when implanted into muscular tissues (21). The deduced amino acid sequence of BMPs has indicated that they are members of the TGF␤ superfamily. Calcified tissues such as bone and dentine contain a large amount of TGF␤ and BMPs. It was shown that TGF␤ and BMPs are released from bone during osteoclastic bone resorption (22). The receptors for TGF␤ and BMPs are members of a family of transmembrane serine/ threonine kinases (23). The intracellular signals of the TGF␤ superfamily are transduced via specific sets of type I and type II. Two type I receptors (BMPR-IA and BMPR-IB) and one type II receptor (BMPR-II) have been identified for BMP-2 and BMP-4 (24 –26). The extracellular domain of the type I receptor is sufficient for stable binding to BMPs and subsequent formation of a heteromeric complex with the intact type II receptors (24, 27, 28). We have established methods for obtaining a large amount of a soluble form of the extracellular domain of BMPR-IA (sBMPR-IA) using silkworm expression system (29) and the Novegen expression system (Novegen Inc., Madison, WI). This sBMPR-IA was in a monomer form in solution and bound to BMP-4, but not to activin or TGF␤1 (29). Alkaline phosphatase activity induced by BMP-2 in the mouse osteoblastic cell line MC3T3-E1 and in the bone marrow stromal cell line ST2 was markedly inhibited by sBMPR-IA added simultaneously (29). Although the role of BMPs in osteoblast differentiation has been extensively investigated, their action on osteoclast differentiation and function has not been elucidated. In the present study we explored the role of BMPs in osteoclast formation and function. BMP-2 dramatically increased osteoclast formation in bone marrow-derived macrophage cultures treated with RANKL and M-CSF. Like OPG, sBMPR-IA strikingly inhibited osteoclast formation supported by RANKL with and without BMP-2. BMP-2 also enhanced the survival of purified osteoclasts supported by RANKL. Both BMP-2 and BMPR-IA mRNAs were expressed not only in osteoclast progenitors, but also in mature osteoclasts. The present findings suggest that the potentiating effects of BMP-2 on osteoclast formation and activation may be important for the precise appearance of osteoclasts in bone tissues. Materials and Methods Drugs Recombinant human M-CSF (Leukoprol) was obtained from Yoshitomi Pharmaceutical Co. (Osaka, Japan). Recombinant human BMP-2 was provided by Yamanouchi Pharmaceutical Co., Ltd. (Tokyo, Japan).

Endocrinology, August 2001, 142(8):3656 –3662

3657

Recombinant sBMPR-IA was prepared by fusing the extracellular domain of BMPR-IA (amino acids 20 –152) to the Flag-tag sequence using the Novagen expression system (29a). Recombinant human TGF␤1 and activin A were purchased from Genzyme/Techne (Cambridge, MA). Recombinant human OPG and a soluble form of mouse RANKL were prepared as described previously (6, 18). Type I collagen gel solution (cell matrix type IA) was obtained from Nitta Gelatin, Inc. (Osaka, Japan). Bacterial collagenase and 1␣,25-(OH)2D3 were obtained from Wako Pure Chemical Industries, Ltd. (Osaka, Japan). Other chemicals and reagents used were of analytical grade.

Mouse bone marrow macrophage cultures Five- to 8-wk-old male ddY mice were obtained from Sankyo Co., Ltd. (Tokyo, Japan). Bone marrow cells prepared from the tibia of ddY mice were suspended in ␣MEM containing 10% FBS (JRH Biosciences, Lenexa, KS) and cultured in 48-well plates (1.5 ⫻ 105 cells/0.3 ml䡠well) in the presence of M-CSF (100 ng/ml). After culturing for 4 d, nonadherent cells were completely removed from the culture by pipetting (30). Almost all of the adherent cells expressed macrophage-specific antigens such as Mac-1, Moma-2, and F4/80 (30). These macrophages were further cultured for 3 d with vehicle (control), BMP-2 (300 ng/ml), TGF␤ (10 ng/ml), or activin A (10 ng/ml) in the presence or absence of RANKL (100 ng/ml) and M-CSF (100 ng/ml). Some cultures were simultaneously treated with OPG (10 ng/ml) or sBMPR-IA (1000 ng/ml). Cells were then fixed and stained for tartrate-resistant acid phosphatase (TRAP; a marker enzyme of osteoclasts) as described previously (31). TRAP-positive multinucleated cells (MNCs) containing more than three nuclei were counted as osteoclasts under microscopic examination. Some cultures were stained for both TRAP and alkaline phosphatase (a marker enzyme of osteoblasts) as previously described (32).

Survival assay of mature osteoclasts Primary osteoblasts were prepared from the calvaria of newborn ddY mice (31). Osteoblasts and freshly prepared bone marrow cells were cocultured in ␣MEM containing 10% FBS and 1␣,25-(OH)2D3 (10-8 M) in 100-mm-diameter dishes precoated with collagen gels (31). Osteoclasts were formed within 6 d in the coculture, and all cells were removed from the dishes by treatment with 0.2% collagenase. The purity of osteoclasts in this preparation was about 5%. To purify osteoclasts, the crude osteoclast preparation was replated in culture dishes (24-well dishes). After culture for 8 h, osteoblasts were removed with PBS containing 0.001% pronase E (Calbiochem, La Jolla, CA) and 0.02% EDTA as described previously (31, 33). The purity of osteoclasts in this preparation was about 95%. Some cultures were then stained for TRAP. The other cultures were further incubated for the indicated periods in the presence or absence of RANKL (100 ng/ml) and/or BMP-2 (300 ng/ml), and stained for TRAP. TRAP-positive MNCs were counted as living osteoclasts.

PCR amplification of reverse transcribed mRNA For semiquantitative RT-PCR analysis, total cellular RNA was extracted from bone marrow-derived macrophages, osteoblasts, and purified osteoclasts. Purified osteoclasts were prepared from cocultures of osteoblasts and bone marrow cells. Macrophages were treated with RANKL (100 ng/ml) and/or BMP-2 (300 ng/ml) for 24 h. Total cellular RNA was extracted using TRIzol solution (Life Technologies, Inc., Grand Island, NY). First strand cDNA was synthesized from the total RNA with random primers and subjected to PCR amplification with EX Taq polymerase (Takara Biochemicals, Shiga, Japan) using specific PCR primers: mouse RANK, 5⬘-GCAAACCTTGGACCAACTGCAC-3⬘ (forward, nucleotides 533–554) and 5⬘-AATCCACCGTGCTTTCAGTCCC-3⬘ (reverse, nucleotides 1186 –1207); mouse BMP-2, 5⬘-GATTGACTCCATTGGCCCTA-3⬘ (forward, nucleotides 4202– 4221) and 5⬘-GGCTAGTTTCTGGGCAGTTG-3⬘ (reverse, nucleotides 4401– 4420); mouse BMPRIA, 5⬘-GGGTCGTTACAACCGTGATT-3⬘ (forward, nucleotides 699 –718) and 5⬘-CGCCATTTACCCATCCATAC-3 (reverse, nucleotides 902–921); mouse granulocyte-macrophage colony-stimulating factor (GM-CSF), 5⬘-CTTTGTGCCTGCGTAATGA-3⬘ (forward, nucleotides 500 –519) and 5⬘-GAGTCAGCGTTTTCAGAGGG-3 (reverse, nucleotides 592– 611); mouse calcitonin receptor, 5⬘-TTTCAAGAACCT-

3658



TAGCTGCCAGAG-3⬘ (forward, nucleotides 1023–1046) and 5⬘-CAAGGCACGGACAATGTTGAGAAG-3 (reverse, nucleotides 1564 –1586); mouse c-Fms, 5⬘-AACAAGTTCTACAAACTGGTGAAGG-3⬘ (forward, nucleotides 2653–2677) and 5⬘-GAAGCCTGTAGTCTAAGCATCTGTC-3⬘ (reverse, nucleotides 3381–3405); mouse glyceraldehyde-3-phosphate dehydrogenase (GAPDH), 5⬘-ACCACAGTCCATGCCATCAC-3⬘ (forward, nucleotides 566 –585) and 5⬘-TCCACCACC-CTGTTGCTGTA-3⬘ (reverse, nucleotides 998-1017). Preliminary experiments were performed to ensure that the number of PCR cycles was within the exponential phase of the amplification curve. PCR products were separated by electrophoresis on a 2% agarose gel.

EMSA Purified mouse osteoclasts were treated with RANKL (100 ng/ml) and/or BMP-2 (300 ng/ml) for 30 min. Nuclear extracts were then prepared from osteoclasts as previously described (33). A nuclear factor-␬B (NF-␬B) binding oligonucleotide sequence (5⬘-AGCTTGGGGACTTTCCGAG-3⬘) was used as a radioactive DNA probe. The DNA binding reaction was performed at room temperature in a volume of 30 ␮l, which contained the binding buffer [10 mm Tris-HCl (pH 7.5), 1 mm EDTA, 4% glycerol, 100 mm NaCl, 5 mm dithiothreitol, and 100 mg/ml BSA], 3 ␮g poly(dI-dC), 1 ⫻ 105 cpm of a 32P-labeled probe, and 8 ␮g nuclear proteins. After incubation for 15 min, the samples were electrophoresed on native 5% acrylamide/0.25 ⫻ TBE gels. The gels were dried and exposed to x-ray film.

Statistical analysis The data were analyzed by one-factor ANOVA and t test (StatView, Abacus Concepts, Inc., Berkeley, CA). The t test was performed when the ANOVA indicated significance at the P ⬍ 0.0001 level. All values are the mean ⫾ sem of quadruplicate cultures.

Results

We modified a mouse bone marrow culture system to obtain highly purified osteoclast progenitors (30). Mouse bone marrow cells were cultured for 3 d with M-CSF, then nonadherent cells were completely removed from the cultures. The remaining adherent cells expressed macrophagespecific antigens such as Moma-2, Mac 1, and F-4/80, and there were very few alkaline phosphatase-positive stromal cells in this preparation (30). The bone marrow-derived macrophages differentiated into TRAP-positive MNCs when they were cultured with RANKL (100 ng/ml) and M-CSF (100 ng/ml) for 3 d (Fig. 1). BMP-2 (300 ng/ml) markedly stimulated the formation of TRAP-positive MNCs supported by RANKL and M-CSF (Fig. 1, A and C). Bone marrowderived stromal cells were rarely observed in the macrophage preparation even after culture for 3 d. Most TRAP-positive cells were formed apart from alkaline phosphatase-positive stromal cells in the presence and absence of BMP-2 (Fig. 1B). The TRAP-positive MNCs induced by RANKL together with BMP-2 were also positive for calcitonin receptors (data not shown). TGF␤1 (10 ng/ml) or activin A (10 ng/ml) enhanced TRAP-positive MNC formation induced by RANKL and M-CSF (Fig. 1C). However, none of the TGF␤ superfamily cytokines induced osteoclast differentiation in the absence of RANKL. The addition of OPG completely inhibited osteoclast formation induced by RANKL and BMP-2 (Fig. 1C). A soluble form of the extracellular domain of BMPR-IA, sBMPR-IA, has been shown to antagonize BMP activity in the target cells. Therefore, we examined the effects of sBMPR-IA on osteoclast formation in bone marrow macrophage cul-

FIG. 1. Potentiating effects of BMP-2 on osteoclast formation in bone marrow-derived macrophage cultures induced by RANKL and MCSF. Mouse bone marrow cells were cultured in the presence of M-CSF (100 ng/ml) for 4 d to prepare adherent macrophages. Macrophages were further cultured for 3 d with M-CSF (100 ng/ml) in the presence or absence of BMP-2 (300 ng/ml) and/or RANKL (100 ng/ml). A, Cells were fixed and stained for TRAP. Bar, 100 ␮M. B, Cells were fixed and stained for both TRAP and alkaline phosphatase. The inset shows alkaline phosphatase staining of primary osteoblasts cultured for 3 d as the positive control. Bar, 100 ␮M. C, Adherent macrophages were treated with vehicle (control), BMP-2 (300 ng/ml), TGF␤ (10 ng/ml), or activin A (10 ng/ml) in the presence of RANKL (100 ng/ml) and M-CSF (100 ng/ml). Some cultures were simultaneously treated with OPG (10 ng/ml). Cells were then fixed and stained for TRAP, and the number of TRAP-positive MNCs was scored. Values are expressed as the mean ⫾ SEM of quadruplicate cultures. Similar findings were obtained in four independent sets of experiments. *, Significantly different from the control cultures (P ⬍ 0.01).

tures. sBMPR-IA (1000 ng/ml) strongly suppressed osteoclast formation induced not only by the combination of RANKL and BMP-2, but also by RANKL alone (Fig. 2). We next examined the expression of mRNAs encoding BMP-2 and BMPR- IA using RT-PCR. Both bone marrow macrophages and purified mature osteoclasts expressed mRNAs of RANK, BMP-2, and BMPR-IA at levels comparative to those in osteoblasts (Fig. 3A). Treatment of bone


FIG. 2. Effects of sBMPR-IA on osteoclast formation induced by RANKL in bone marrow-derived macrophage cultures. Mouse bone marrow cells were cultured in the presence of M-CSF (100 ng/ml) for 4 d to prepare adherent macrophages. Macrophages were further cultured for 3 d with M-CSF (100 ng/ml) and RANKL (100 ng/ml) in the presence or absence of BMP-2 (300 ng/ml). Some cultures were simultaneously treated with sBMPR-IA (1000 ng/ml). Cells were then fixed and stained for TRAP, and the number of TRAP-positive MNCs was scored. Values are expressed as the mean ⫾ SEM of quadruplicate cultures. Similar findings were obtained in four independent sets of experiments. *, Significantly different from the cultures with control (P ⬍ 0.01).

marrow macrophages with RANKL and/or BMP-2 for 24 h showed no significant changes in the expression level of RANK, c-Fms (M-CSF-receptor), and GM-CSF mRNAs (Fig. 3B). Expression of those mRNAs remained unchanged in the macrophage cultures treated with RANKL and/or BMP-2 for 3 d, although osteoclasts were formed in some cultures (data not shown). These findings indicate that stimulation of osteoclast formation by BMP-2 was not due to changes in RANK, c-Fms, and GM-CSF expression in bone marrow macrophages. Purified osteoclasts spontaneously died in a time-dependent manner in the absence of osteoblasts (Fig. 4, A and B). We have shown that three cytokines, RANKL, IL-1, and M-CSF, directly supported the survival of purified osteoclasts (33, 34). BMP-2 (300 ng/ml) also enhanced the survival of purified osteoclasts in the presence of RANKL (Fig. 4). BMP-2 alone showed no stimulatory effect on the survival of osteoclasts (Fig. 4A). BMP-2 did not enhance the survival of osteoclasts supported by M-CSF (Fig. 4C). sBMPR-IA (1000 ng/ml) strongly inhibited RANKL-induced survival of osteoclasts (Fig. 4C). These findings suggest that the BMP-meditated signaling pathway plays an important role in the RANK-RANKL interaction not only in osteoclastogenesis but also in the survival of osteoclasts. An EMSA revealed that RANKL activated NF-␬B in purified osteoclasts (Fig. 5). BMP-2 itself failed to activate NF␬B, but, rather, inhibited NF-␬B activation induced by RANKL in purified osteoclasts (Fig. 5). This indicated that enhancement of RANKL-induced survival of osteoclasts by BMP-2 was not due to the activation of NF-␬B. Discussion

We previously reported that recombinant BMP-2 stimulated differentiation of pluripotent immature stromal cells (C3H10T1/2 cells) into osteoblasts (35, 36). Concomitantly,


3659

BMP-2 inhibited the differentiation of C2C12 myoblasts into multinucleated myotubes and induced the expression of typical phenotypes of osteoblasts (37). These findings indicated that the BMP-mediated signaling pathway induced osteoblastic differentiation not only of immature mesenchymal progenitor cells, but also of myogenic cells. However, the effects of BMPs on the differentiation of osteoclasts, which is controlled by osteoblasts in normal bone via membranebound expression of both M-CSF and RANKL (38), have not been extensively investigated. Abe et al. (39) reported that BMP-2 increased osteoclast formation by 4- to 6-fold in PTHtreated bone marrow cultures that contained both hemopoietic and osteoblastic stromal cells. Koide et al. (40) showed that the combination of BMP-2 and IL-1␣ elevated the expression of both cyclooxygenase-2 and RANKL mRNAs in osteoblasts, resulting in enhancement of osteoclast differentiation. However, we report here that BMP-2 directly enhanced osteoclastic differentiation of the progenitor cells in bone marrow macrophage cultures treated with RANKL and M-CSF. Osteoclast formation induced by RANKL and BMP-2 was suppressed by the addition of sBMPR-IA. sBMPR-IA also inhibited RANKL-induced osteoclast formation even in the absence of exogenous BMP-2. These findings suggested that BMP-mediated signals were involved not only in osteoblastic bone formation, but also in osteoclastic bone resorption. Sells Galvin et al. (41) first reported that TGF␤ enhanced osteoclast differentiation in hemopoietic cells in the presence of RANKL and M-CSF. Neutralizing anti-TGF␤ antibody abrogated osteoclast formation from macrophages induced by RANKL, TGF␤, and M-CSF (42). Fuller et al. (43) also reported that activin A synergistically stimulated RANKLinduced osteoclast differentiation from the hemopoietic progenitors. Moreover, osteoclast formation induced by RANKL was completely abolished by soluble activin receptor type IIA or soluble TGF␤ receptor II (43, 44). These findings clearly explained the observations that TGF␤2 transgenic mice developed osteoporosis due to enhanced osteoclast formation (45), and that transgenic mice expressing a cytoplasmically truncated TGF␤ receptor type II showed marked reduction of osteoclast formation (46). Consistent with those findings, sBMPR-IA inhibited RANKL-induced osteoclast formation even in the absence of exogenous BMP-2. We also determined that both osteoclast progenitors and purified osteoclasts expressed BMP-2 mRNA. This suggests that endogenous production of BMP-2 by osteoclast progenitors is involved in their differentiation into osteoclasts induced by RANKL in the present cultures. It has also been demonstrated that TGF␤1 mRNA was constitutively expressed by mouse bone marrow macrophages and by osteoclasts in human giant cell tumor of bone (47). Although the mechanism by which these soluble forms of TGF␤ receptor superfamily members similarly inhibit RANKL-induced osteoclast differentiation is not known, TGF␤ superfamily cytokines appear to be important for recruiting osteoclasts as autocrine or paracrine regulators. Kanatani et al. (48) first reported that BMPR-IA mRNA was expressed in hemopoietic blast cells supported by GM-CSF. It was found in the present study that both bone marrow macrophages and purified mature osteoclasts expressed BMPR-IA mRNA. When osteoblasts were removed from the

3660



FIG. 3. Expression of mRNAs of RANK, c-Fms, calcitonin receptors, BMP-2, and BMPR-IA in bone marrow macrophages, primary osteoblasts, and purified osteoclasts (A) and effects of RANKL and BMP-2 on the expression of RANK, c-Fms, and GM-CSF mRNAs in bone marrow macrophages (B). Mouse bone marrow cells were cultured in the presence of M-CSF (100 ng/ml) for 4 d to prepare adherent macrophages. Primary calvarial osteoblasts and purified osteoclasts were prepared as described in Materials and Methods. A, Total RNA was extracted from bone marrow macrophages, purified osteoclasts, and osteoblasts and amplified by PCR for mouse RANK (30 cycles), c-Fms (30 cycles), calcitonin receptors (30 cycles), BMP-2 (32 cycles), BMPR-IA (32 cycles), and GAPDH (25 cycles) using the respective specific primer pairs. PCR was performed under conditions determined to be in the linear range of product formation. B, Bone marrow macrophages were cultured for 24 h with or without RANKL (100 ng/ml) and/or BMP-2 (300 ng/ml) in the presence of M-CSF (100 ng/ml). Total RNA was then extracted from macrophages and subjected to semiquantitative PCR analysis for RANK, c-Fms, GM-CSF, and GAPDH using respective specific primer pairs. Similar findings were obtained in four independent sets of experiments.

cocultures, osteoclasts rapidly died within 48 h (49, 50). RANKL and M-CSF potentiated the survival of osteoclasts though their respective receptors. Interestingly, BMP-2 enhanced the RANKL-induced survival of purified osteoclasts, but it never stimulated the survival of osteoclasts in the absence of RANKL. In addition, BMP-2 failed to stimulate the M-CSF-supported survival of osteoclasts. These findings suggested that BMP receptor-induced signals cross-communicated with RANK-mediated signals, but not with c-Fms (M-CSF receptor)-mediated signals, in inducing the survival of mature osteoclasts. It is therefore suggested that BMP-2induced enhancement of osteoclast differentiation is caused by cross-communication between BMP receptor-mediated signals and RANK-mediated signals. BMP-2 increased the number of resorption pits formed on dentine slices in mouse bone marrow cultures treated with M-CSF and RANKL (data not shown). The effects of BMP-2 on both the differentiation and survival of osteoclasts may result in an increase in the number of resorption pits formed on the slices. Recently, Kaneko et al. (51) reported that mature rabbit osteoclasts expressed BMP receptors, and BMP-2 directly stimulated their pit-forming activity even in the absence of exogenous RANKL. The difference between their findings and those of the present study may be due to the different species-derived osteoclasts used. Despite this difference, these findings suggest that BMP-mediated signals are involved not only in osteoclast differentiation, but also in osteoclast function. GM-CSF is a potent inhibitor of osteoclast differentiation

from their progenitors in the mouse culture system (52, 53). Wani et al. (54) reported that PGE2 enhanced osteoclast formation induced by RANKL in M-CSF-dependent bone marrow macrophage cultures. Subsequently, using mice lacking PG G/H synthase-2, Okada et al. (55) showed that enhancement of RANKL-induced osteoclast formation by PGE2 in mouse bone marrow cultures was caused by inhibition of GM-CSF expression. BMP-2 did not reduce GM-CSF mRNA expression in osteoclast progenitors in the present study. Furthermore, no significant change in mRNA expression of RANK and c-Fms was observed in osteoclast progenitors treated with BMP-2. Therefore, stimulation of osteoclast differentiation by BMP-2 seems to be independent from the changes in RANK, c-Fms, and GM-CSF expression in osteoclast progenitors. We have previously shown that activation of NF-␬B was involved in the survival of osteoclasts supported by IL-1 or RANKL (33, 34). BMP-2, however, did not stimulate, but, rather, inhibited, RANKL-induced activation of NF-␬B in purified osteoclasts. This suggests that signals other than NF-␬B are involved in the survival of osteoclasts induced by RANKL and BMP-2. Both Smad1 and Smad5 are involved in the BMP signals, whereas Smad2 and Smad3 are involved in TGF␤ signals in the target cells (56). However, BMP and TGF␤ showed similar effects on osteoclast progenitors in the present study. This suggests that signaling pathways other than Smad-mediated pathways are involved in the enhancement of RANKL-induced osteoclast differentiation by TGF␤ superfamily members. The differentiation and survival of



3661

FIG. 5. Effects of RANKL and/or BMP-2 on the activation of NF-␬B in purified osteoclasts. Purified osteoclasts were prepared in culture dishes as described in Materials and Methods. Purified osteoclasts were treated for 1 h with RANKL (100 ng/ml), BMP-2 (300 ng/ml), or RANKL (100 ng/ml) plus BMP-2 (300 ng/ml). Nuclear extracts were prepared and analyzed using an EMSA with an NF-␬B-binding oligonucleotide probe as described in Materials and Methods. Similar findings were obtained in four independent sets of experiments.

osteoclasts are tightly regulated by MAPK-mediated signals (33, 57). Therefore, MAPK pathways rather than Smadmediated ones may be involved in the osteoclastic bone resorption enhanced by TGF␤ superfamily members. How TGF␤ superfamily members potentiate RANKL-induced osteoclastogenesis is unknown, but the members appear to be important regulators of bone resorption as well as bone formation. These data provide evidence for a direct role of TGF␤ superfamily members on osteoclast progenitors and mature osteoclasts. As bone has an abundant store of latent TGF␤ that is released and activated as a consequence of bone resorption, regulation of TGF␤ activity is paramount to maintenance of osteoclast formation and function. Notwithstanding the effects of TGF␤ on the osteoblast, the abundance of local activated TGF␤ coincident with the pathologies observed in cancer-induced metastasis and rheumatoid arthritis would favor the recruitment and survival of osteoclasts. These activities of TGF␤ and BMPs would contribute to the enhanced bone loss observed in these conditions. Acknowledgments

FIG. 4. Time course of change in the survival of osteoclasts induced by BMP-2 (A) and effects of M-CSF, RANKL, BMP-2, and sBMPR-IA on the survival of osteoclasts (B and C). Purified osteoclasts were prepared on culture dishes as described in Materials and Methods. A, Purified osteoclasts were incubated with BMP-2 (300 ng/ml) and/or RANKL (100 ng/ml) for the indicated periods. Cells were then fixed and stained for TRAP, and the number of TRAP-positive osteoclasts was scored. Values are expressed as the mean ⫾ SEM of quadruplicate cultures. B, Purified osteoclasts were cultured for 36 h with vehicle (control), RANKL (100 ng/ml), and RANKL (100 ng/ml) plus BMP-2 (300 ng/ml). Cells were then stained for TRAP. Bar, 100 ␮M. C, Purified osteoclasts were treated with M-CSF (100 ng/ml), RANKL (100 ng/ml), BMP-2 (300 ng/ml), or sBMPR-IA (1000 ng/ml). After culture for 48 h, cells were fixed and stained for TRAP, and the number of TRAP-positive osteoclasts was scored. Values are expressed as the mean ⫾ SEM of quadruplicate cultures. Similar findings were obtained in four independent sets of experiments.

Received December 29, 2000. Accepted April 5, 2001. Address all correspondence and requests for reprints to: Dr. Naoyuki Takahashi, Department of Biochemistry, Showa University School of Dentistry, 1-5-8 Hatanodai, Shinagawa-ku, Tokyo 142-8555, Japan. Email: [email protected]. This work was supported in part by grants-in-aid (11470393, 12470393, 12137209, and 12877299), the High-Technology Research Center Project from the Ministry of Education, Science, Sports, and Culture of Japan, and grants-in-aid from the Hamaguchi Foundation for the Advancement of Biochemistry.

References 1. Takahashi N, Akatsu T, Udagawa N, et al. 1988 Osteoblastic cells are involved in osteoclast formation. Endocrinology 123:2600 –2602 2. Udagawa N, Takahashi N, Akatsu T, et al. 1990 Origin of osteoclasts: mature monocytes and macrophages are capable of differentiating into osteoclasts under a suitable microenvironment prepared by bone marrow-derived stromal cells. Proc Natl Acad Sci USA 87:7260 –7264 3. Suda T, Takahashi N, Martin TJ 1992 Modulation of osteoclast differentiation. Endocr Rev 13:66 – 80

3662


4. Udagawa N, Takahashi N, Katagiri T, et al. 1995 Interleukin (IL)-6 induction of osteoclast differentiation depends on IL-6 receptors expressed on osteoblastic cells but not on osteoclast progenitors. J Exp Med 182:1461–1468 5. Yoshida H, Hayashi S, Kunisada T, et al. 1990 The murine mutation osteopetrosis is in the coding region of the macrophage colony stimulating factor gene. Nature 345:442– 444 6. Yasuda H, Shima N, Nakagawa N, et al. 1998 Osteoclast differentiation factor is a ligand for osteoprotegerin/osteoclastogenesis-inhibitory factor and is identical to TRANCE/RANKL. Proc Natl Acad Sci USA 95:3597–3602 7. Anderson DM, Maraskovsky E, Billingsley WL, et al. 1997 A homologue of the TNF receptor and its ligand enhance T-cell growth and dendritic-cell function. Nature 390:175–179 8. Wong BR, Rho J, Arron J, et al. 1997 TRANCE is a novel ligand of the tumor necrosis factor receptor family that activates c-Jun N-terminal kinase in T cells. J Biol Chem 272:25190 –25194 9. Lacey DL, Timms E, Tan HL, et al. 1998 Osteoprotegerin ligand is a cytokine that regulates osteoclast differentiation and activation. Cell 93:165–176 10. Matsuzaki K, Udagawa N, Takahashi N, et al. 1998 Osteoclast differentiation factor (ODF) induces osteoclast-like cell formation in human peripheral blood mononuclear cell cultures. Biochem Biophys Res Commun 246:199 –204 11. Quinn JM, Elliott J, Gillespie MT, Martin TJ 1998 A combination of osteoclast differentiation factor and macrophage-colony stimulating factor is sufficient for both human and mouse osteoclast formation in vitro. Endocrinology 139:4424 – 4427 12. Kong YY, Yoshida H, Sarosi I, et al. 1999 OPGL is a key regulator of osteoclastogenesis, lymphocyte development and lymph-node organogenesis. Nature 397:315–323 13. Dougall WC, Glaccum M, Charrier K, et al. 1999 RANK is essential for osteoclast and lymph node development. Genes Dev 13:2412–2424 14. Hsu H, Lacey DL, Dunstan CR, et al. 1999 Tumor necrosis factor receptor family member RANK mediates osteoclast differentiation and activation induced by osteoprotegerin ligand. Proc Natl Acad Sci USA 96:3540 –3545 15. Tsuda E, Goto M, Mochizuki S, et al. 1997 Isolation of a novel cytokine from human fibroblasts that specifically inhibits osteoclastogenesis. Biochem Biophys Res Commun 234:137–142 16. Simonet WS, Lacey DL, Dunstan CR, et al. 1997 Osteoprotegerin: a novel secreted protein involved in the regulation of bone density. Cell 89:309 –319 17. Kwon BS, Wang S, Udagawa N, et al. 1998 TR1, a new member of the tumor necrosis factor receptor superfamily, induces fibroblast proliferation and inhibits osteoclastogenesis and bone resorption. FASEB J 12:845– 854 18. Yasuda H, Shima N, Nakagawa N, et al. 1998 Identity of osteoclastogenesis inhibitory factor (OCIF) and osteoprotegerin (OPG): a mechanism by which OPG/OCIF inhibits osteoclastogenesis in vitro. Endocrinology 139:1329 –1337 19. Bucay N, Sarosi I, Dunstan CR, et al. 1998 Osteoprotegerin-deficient mice develop early onset osteoporosis and arterial calcification. Genes Dev 12:1260 –1268 20. Mizuno A, Amizuka N, Irie K, et al. 1998 Severe osteoporosis in mice lacking osteoclastogenesis inhibitory factor/osteoprotegerin. Biochem Biophys Res Commun 247:610 – 615 21. Wozney JM, Rosen V, Celeste AJ, et al. 1988 Novel regulators of bone formation: molecular clones and activities. Science 242:1528 –1534 22. Oreffo RO, Mundy GR, Seyedin SM, Bonewald LF 1989 Activation of the bone-derived latent TGF beta complex by isolated osteoclasts. Biochem Biophys Res Commun 158:817– 823 23. Miyazono K, ten Dijke P, Yamashita H, Heldin CH 1994 Signal transduction via serine/threonine kinase receptors. Semin Cell Biol 5:389 –398 24. Suzuki A, Thies RS, Yamaji N, et al. 1994 A truncated bone morphogenetic protein receptor affects dorsal-ventral patterning in the early Xenopus embryo. Proc Natl Acad Sci USA 91:10255–10259 25. Yamaji N, Celeste AJ, Thies RS, et al. 1994 A mammalian serine/threonine kinase receptor specifically binds BMP-2 and BMP-4. Biochem Biophys Res Commun 205:1944 –1951 26. Liu F, Ventura F, Doody J, Massague J 1995 Human type II receptor for bone morphogenic proteins (BMPs): extension of the two-kinase receptor model to the BMPs. Mol Cell Biol 15:3479 –3486 27. Graff JM, Thies RS, Song JJ, Celeste AJ, Melton DA 1994 Studies with a Xenopus BMP receptor suggest that ventral mesoderm-inducing signals override dorsal signals in vivo. Cell 79:169 –179 28. ten Dijke P, Yamashita H, Sampath TK, et al. 1994 Identification of type I receptors for osteogenic protein-1 and bone morphogenetic protein-4. J Biol Chem 269:16985–16988 29. Natsume T, Tomita S, Iemura S, Kinto N, Yamaguchi A, Ueno N 1997 Interaction between soluble type I receptor for bone morphogenetic protein and bone morphogenetic protein-4. J Biol Chem 272:11535–11540 29a.Hatta T, Konishi H, Katoh E, Natsume T, Ueno N, Kobayashi Y, Yamazaki T 2000 Identification of the ligand-binding site of the BMP type IA receptor for BMP-4. Biopolymers 55:399 – 406 30. Kobayashi K, Takahashi N, Jimi E, et al. 2000 Tumor necrosis factor ␣ stimulates osteoclast differentiation by a mechanism independent of the ODF/ RANKL-RANK interaction. J Exp Med 275–286 31. Suda T, Jimi E, Nakamura I, Takahashi N 1997 Role of 1␣,25-dihydroxyvitamin D3 in osteoclast differentiation and function. Methods Enzymol 282:223–235


32. Takahashi N, Yamana H, Yoshiki S, et al. 1988 Osteoclast-like cell formation and its regulation by osteotropic hormones in mouse bone marrow cultures. Endocrinology 122:1373–1382 33. Jimi E, Nakamura I, Ikebe T, Akiyama S, Takahashi N, Suda T 1998 Activation of NF-␬B is involved in the survival of osteoclasts promoted by interleukin-1. J Biol Chem 273:8799 – 8805 34. Jimi E, Akiyama S, Tsurukai T, et al. 1999 Osteoclast differentiation factor acts as a multifunctional regulator in murine osteoclast differentiation and function. J Immunol 163:434 – 442 35. Katagiri T, Yamaguchi A, Ikeda T, et al. 1990 The non-osteogenic mouse pluripotent cell line, C3H10T1/2, is induced to differentiate into osteoblastic cells by recombinant human bone morphogenetic protein-2. Biochem Biophys Res Commun 172:295–299 36. Yamaguchi A, Katagiri T, Ikeda T, et al. 1991 Recombinant human bone morphogenetic protein-2 stimulates osteoblastic maturation and inhibits myogenic differentiation in vitro. J Cell Biol 113:681– 687 37. Katagiri T, Yamaguchi A, Komaki M, et al. 1994 Bone morphogenetic protein-2 converts the differentiation pathway of C2C12 myoblasts into the osteoblast lineage. J Cell Biol 127:1755–1766 38. Itoh K, Udagawa N, Matsuzaki K, et al. 2000 Importance of membrane- or matrix-associated forms of M-CSF and RANKL/ODF in osteoclastogenesis supported by SaOS-4/3 cells expressing recombinant PTH/PTHrP receptors. J Bone Miner Res 15:1755–1766 39. Abe E, Yamamoto M, Taguchi Y, et al. 2000 Essential requirement of BMPs-2/4 for both osteoblast and osteoclast formation in murine bone marrow cultures from adult mice: antagonism by noggin. J Bone Miner Res 15:663– 673 40. Koide M, Murase Y, Yamato K, Noguchi T, Okahashi N, Nishihara T 1999 Bone morphogenetic protein-2 enhances osteoclast formation mediated by interleukin-1␣ through upregulation of osteoclast differentiation factor and cyclooxygenase-2. Biochem Biophys Res Commun 259:97–102 41. Sells Galvin RJ, Gatlin CL, Horn JW, Fuson TR 1999 TGF-␤ enhances osteoclast differentiation in hematopoietic cell cultures stimulated with RANKL and M-CSF. Biochem Biophys Res Commun 265:233–239 42. Kaneda T, Nojima T, Nakagawa M, et al. 2000 Endogenous production of TGF-␤ is essential for osteoclastogenesis induced by a combination of receptor activator of NF-␬B ligand and macrophage-colony-stimulating factor. J Immunol 165:4254 – 4263 43. Fuller K, Bayley KE, Chambers TJ 2000 Activin A is an essential cofactor for osteoclast induction. Biochem Biophys Res Commun 268:2–7 44. Fuller K, Lean JM, Bayley KE, Wani MR, Chambers TJ 2000 A role for TGF␤(1) in osteoclast differentiation and survival. J Cell Sci 113:2445–2453 45. Erlebacher A, Derynck R 1996 Increased expression of TGF-␤2 in osteoblasts results in an osteoporosis-like phenotype. J Cell Biol 132:195–210 46. Filvaroff E, Erlebacher A, Ye J, et al. 1999 Inhibition of TGF-␤ receptor signaling in osteoblasts leads to decreased bone remodeling and increased trabecular bone mass. Development 126:4267– 4279 47. Zheng MH, Fan Y, Wysocki SJ, et al. 1994 Gene expression of transforming growth factor-␤ 1 and its type II receptor in giant cell tumors of bone. Possible involvement in osteoclast-like cell migration. Am J Pathol 145:1095–1104 48. Kanatani M, Sugimoto T, Kaji H, et al. 1995 Stimulatory effect of bone morphogenetic protein-2 on osteoclast-like cell formation and bone-resorbing activity. J Bone Miner Res 10:1681–1690 49. Jimi E, Shuto T, Koga T 1995 Macrophage colony-stimulating factor and interleukin-1␣ maintain the survival of osteoclast-like cells. Endocrinology 136:808 – 811 50. Udagawa N, Takahashi N, Jimi E, et al. 1999 Osteoblasts/stromal cells stimulate osteoclast activation through expression of osteoclast differentiation factor/RANKL but not macrophage colony-stimulating factor: receptor activator of NF-␬B ligand. Bone 25:517–523 51. Kaneko H, Arakawa T, Mano H, et al. 2000 Direct stimulation of osteoclastic bone resorption by bone morphogenetic protein (BMP)-2 and expression of BMP receptors in mature osteoclasts. Bone 27:479 – 486 52. Shuto T, Jimi E, Kukita T, Hirata M, Koga T 1994 Granulocyte-macrophage colony stimulating factor suppresses lipopolysaccharide-induced osteoclast-like cell formation in mouse bone marrow cultures. Endocrinology 134:831– 837 53. Udagawa N, Horwood NJ, Elliott J, et al. 1997 Interleukin-18 (interferon-␥inducing factor) is produced by osteoblasts and acts via granulocyte/macrophage colony-stimulating factor and not via interferon-␥ to inhibit osteoclast formation. J Exp Med 185:1005–1012 54. Wani MR, Fuller K, Kim NS, Choi Y, Chambers T 1999 Prostaglandin E2 cooperates with TRANCE in osteoclast induction from hemopoietic precursors: synergistic activation of differentiation, cell spreading, and fusion. Endocrinology 140:1927–1935 55. Okada Y, Lorenzo JA, Freeman AM, Tomita M, Morham SG, Raisz LG, Pilbeam CC 2000 Prostaglandin G/H synthase-2 is required for maximal formation of osteoclast-like cells in culture. J Clin Invest 105:823– 832 56. Massague J, Wotton D 2000 Transcriptional control by the TGF-␤/Smad signaling system. EMBO J 19:1745–1754 57. Miyazaki T, Katagiri H, Kanegae Y, et al. 2000 Reciprocal role of ERK and NF-␬B pathways in survival and activation of osteoclasts. J Cell Biol 148:333–342