help button home button Endocrine Society Endocrinology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Nakamura, I.
Right arrow Articles by Duong, L. T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Nakamura, I.
Right arrow Articles by Duong, L. T.
Endocrinology Vol. 139, No. 12 5182-5193
Copyright © 1998 by The Endocrine Society

ARTICLES

Echistatin Inhibits the Migration of Murine Prefusion Osteoclasts and the Formation of Multinucleated Osteoclast-Like Cells

Ichiro Nakamura, Hirofumi Tanaka, Gideon A. Rodan and Le T. Duong

Department of Bone Biology and Osteoporosis Research, Merck Research Laboratories, West Point, Pennsylvania 19486

Address all correspondence and requests for reprints to: Le T. Duong, Ph.D., Department of Bone Biology and Osteoporosis Research, Merck Research Laboratories, West Point, Pennsylvania 19486.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
The vitronectin receptor {alpha}vß3 is highly expressed in osteoclasts and was shown to play a critical role in osteoclast function in vivo. The objective of this study was to examine the role of {alpha}vß3 integrin in osteoclast formation in vitro using the inhibitory disintegrin echistatin, an RGD-containing snake venom. We documented by immunocytochemistry and Northern blot analysis that during murine osteoclast-like cell (OCL) formation in a coculture of mouse osteoblastic MB1.8 cells and bone marrow cells there is increased expression of the {alpha}v and ß3 integrin subunits. Echistatin binds preferentially to the membrane fraction of isolated enriched OCLs (IC50 = 0.6 nM), and this binding is inhibited by vitronectin receptor-blocking polyclonal antibodies. Additionally, cross-linking of radiolabeled echistatin to OCLs, followed by immunoprecipitation with antibodies to vitronectin or fibronectin receptors, shows that {alpha}vß3 integrin is the predominant receptor for echistatin in this system. In this coculture, echistatin completely inhibits the formation of multinucleated OCLs, but not that of mononuclear prefusion OCLs (pOCs). This inhibition is RGD and dose dependent (IC50 = 0.7 nM). We tested the hypothesis that inhibition of OCL formation may be due to interference with pOC migration and found that echistatin inhibited macrophage colony-stimulating factor-induced migration and fusion of pOCs (IC50 = 1 and 0.6 nM, respectively). Echistatin inhibition of pOCs migration and fusion is also RGD dependent. These results suggest that the integrin {alpha}vß3 plays a role in pOC migration, which can explain the inhibitory effect of echistatin on multinucleated osteoclast formation in vitro.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
OSTEOCLASTS are the primary cells responsible for bone resorption. These multinucleated cells of hemopoietic origin are formed by fusion of mononuclear precursors (1, 2, 3). Osteoclastic bone resorption involves a number of sequential differentiation and activation steps regulated by several cell types in bone. These include 1) proliferation and homing of hemopoietic precursors to bone, 2) their differentiation into mononuclear osteoclasts, and 3) fusion of these cells to form multinucleated osteoclasts (2, 3, 4). Osteoclast activation involves migration to sites of subsequent resorption, attachment to the bone surface, formation of a tight seal between the osteoclast membrane and the bone matrix, known as the clear zone, and polarized secretion of acid and proteases into the resorption lacuna (5, 6, 7).

Integrins are transmembrane heterodimeric glycoproteins that mediate cell-cell and cell-matrix interactions (8). Several studies have shown that both mature osteoclasts and osteoclast-like cells (OCLs) generated in tissue culture highly express the vitronectin receptor, {alpha}vß3 integrin, and the collagen/laminin receptor, {alpha}2ß1 integrin (9, 10, 11). There are differing reports on the localization of {alpha}vß3 integrin in osteoclasts; one study localized the receptor to the basolateral and ruffled border membranes (12, 13), whereas other reports also described its presence in the sealing zone of resorbing osteoclasts (10, 14). Its role in osteoclast function is thought to relate to osteoclast attachment to bone via recognition of RGD-containing bone matrix proteins (15, 16). Monoclonal antibodies against the {alpha}vß3 integrin inhibit osteoclast attachment to bone and bone resorption in vitro (17, 18). Furthermore, osteoclasts adhere to plastic dishes coated with ligands for the {alpha}vß3 integrin (vitronectin, fibrinogen, von Willebrand factor, osteopontin, and bone sialoprotein) in an RGD-dependent manner via the ß3 integrin (16, 19, 20). Moreover, the RGD-containing disintegrin echistatin, a snake venom peptide of 49 amino acids isolated from the saw-scaled viper Echis carinatus, was previously reported to inhibit integrin {alpha}IIbß3-mediated platelet aggregation (21) and bone resorption in vitro (15) and in vivo (22); moreover, echistatin has been recently reported to inhibit bone resorption in vivo in estrogen-deficient mice and rats (23) and in mice with secondary hyperparathyroidism (24).

In this study, echistatin was used to examine the role of {alpha}vß3 integrin in osteoclast differentiation in vitro. {alpha}vß3 integrin was suggested to be the primary receptor for echistatin in mature osteoclasts (15, 25). We used the murine coculture system of mouse osteoblastic cells and bone marrow cells, which was developed to study osteoclast development in vitro (26). In this system, osteoclast-like multinucleated cells are formed in the presence of osteotropic hormones, such as 1{alpha},25-dihydroxyvitamin D3 [1{alpha},25-(OH)2D3] or PTH. We found herein that echistatin binds to the mononuclear osteoclasts and blocks the migration of these cells, leading to the inhibition of osteoclast multinucleation. Our data support the hypothesis that {alpha}vß3 integrin is the predominant receptor for echistatin, and this integrin plays an important role in the formation of multinucleated osteoclasts, probably through its function in cell migration.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Chemicals and animals
Echistatin and its analogs were provided by Dr. V. Garsky (Merck Research Laboratories, West Point, PA). [{alpha}-32P]Deoxy (d)-CTP, [{alpha}-32P]dATP, Bolton and Hunter reagent, and radiolabeled protein standards were obtained from Amersham (Arlington Heights, IL). Tissue culture media were purchased from Life Technologies (Grand Island, NY), FBS was obtained from JRH Bioscience (Lenexa, KS), collagenase was obtained from Wako Chemicals (Richmond, VA), and dispase was obtained from Boehringer Mannheim (Mannheim, Germany). Antisera to human vitronectin receptor and human fibronectin receptor were purchased from Life Technologies. Antibodies to the cytoplasmic domains of {alpha}v, {alpha}5, and ß5 integrin subunits were purchased from Chemicon (Temecula, CA). Rabbit antiserum to the ß3 cytoplasmic domain was a gift from Dr. M. Ginsburg (Scripps Research Institute, La Jolla, CA). 1{alpha},25-(OH)2D3 was a gift from Dr. Milan R. Uskokovic (Hoffman-La Roche, Inc., Nutley, NJ). BALB/c mice were obtained from Taconic Farms, Inc. (Germantown, NY). All animals were cared for and housed under conditions stated in the Institutional Animal Care and Use Committee (IACUC) Guide for the Care and Use of Laboratory Animals, and the studies were reviewed and approved by the Merck Research Laboratories Institutional Animal Care and Use Committee.

Coculture system and enrichment of OCLs
Murine osteoclast-like multinucleated and mononuclear cells (OCLs) were prepared from BALB/c mice in a coculture system, as reported previously with slight modifications (26). In short, a mouse osteoblastic MB1.8 cell line was used instead of primary osteoblastic cells (27). MB1.8 cells were plated at 1 x 104 cells/cm2. After 24 h, bone marrow cells isolated from tibiae of 6- to 8-week-old male mice were added (2.5 x 104 cells/cm2) to the monolayer of MB1.8 cells. The coculture was maintained in {alpha}MEM containing 10% FBS and 10 nM 1{alpha},25-(OH)2D3. OCLs were formed within 7 days and identified by the staining for tartrate-resistant acid phosphatase (TRAP), which is a marker enzyme for osteoclasts (26). TRAP-positive cells with three or more nuclei were counted as multinucleated cells. These cells were shown to possess calcitonin receptors and form resorption pits on bovine cortical bone slices (28). The formation of OCLs in this coculture system was quantitated as the number of TRAP-positve multinucleated cells per well of a 24-well tissue culture plate.

The cocultures were maintained in 150-mm tissue culture dishes for 7 days, then washed twice with PBS and treated with 0.1% collagenase and 0.1% dispase in PBS for 20 min at 37 C. The dishes were vigorously washed with PBS (three times). The nonadhering cells were collected by this treatment and washed with PBS (three times) before membrane preparation. The adhering cells were immediately stained for TRAP activity, and the purity of the enriched OCLs was determined by the number of all TRAP-positive cells over total cell number. Membranes were prepared by scraping the cells into 10 mM HEPES, pH 6.5, and 1 mM EDTA containing leupeptin (20 µg/ml), soybean trypsin inhibitor (20 µg/ml), and phenylmethylsulfonylfluoride (1 mM). The crude membrane fraction was then collected by two-step centrifugation as described previously (29). As a control, the calvarial osteoblastic cells were also cultured without bone marrow cells and with 1{alpha},25-(OH)2D3 treatment for 7 days, from which membranes were prepared as described above.

The mononuclear prefusion osteoclasts (pOCs) were prepared as described previously (27) with the following modifications. The murine coculture of bone marrow cells and MB1.8 cells was maintained in 150-mm tissue culture dishes for 6 days. pOCs were then detached using 10 mM EDTA solution after removing MB1.8 cells with collagenase-dispase, followed by washing with 0.1% BSA in {alpha}MEM (three times). Preparations of pOCs (~95% pure as determined by TRAP staining) were used for cell fusion assays and cell migration assays.

RNA isolation and Northern blot analysis
Total cellular RNA was isolated by guanidine isothiocyanate and phenol extraction (30). Twenty micrograms of total RNA were separated using formaldehyde-agarose gel electrophoresis, followed by transfer onto nylon filters (Hybond-N, Amersham, Arlington Heights, IL). Human {alpha}v and ß3 complementary DNA (cDNA) probes were cloned from a human umbilical vein endothelial cell cDNA library by PCR using the published human sequences (31, 32). Mouse {alpha}5 and human ß1 and ß5 cDNA probes were also cloned by PCR from a mouse OCL cDNA library and a human placenta cDNA library, respectively (33, 34, 35). Human glyceraldehyde 3-phosphate dehydrogenase cDNA was a gift from Dr. Robert W. Allen (American Red Cross, St. Louis, MO). cDNAs were labeled with [{alpha}-32P]dCTP using a random primer a DNA labeling kit (Pharmacia Biotech, Piscataway, NJ). A 40-mer oligo DNA (AGCCGTTGTCGACGAC CAGCGCAGCGATATCGTCATCAT) for mouse ß-actin was synthesized and labeled with [{alpha}-32P]dATP using a DNA tailing kit (Boehringer Mannheim). Hybridizations were performed in 40% formamide, 5 x SSC (standard saline citrate), 0.1% SDS, 0.1% Ficoll, 0.1% polyvinylpyrolidone, 0.1% BSA, and 200 µg/ml sonicated salmon sperm DNA at 42 C overnight and washed twice (30 min each time) at 55 C in 0.1 x SSC and 0.1% SDS. The filters were dried and exposed to XAR-2 films (Eastman Kodak Co., Rochester, NY).

Immunocytochemistry
To examine localization of the {alpha}vß3 integrin by immunofluorescence, cells were fixed in 4% paraformaldehyde at room temperature for 15 min. Cells were blocked in 1% BSA in PBS and stained with a hamster antimouse ß3 integrin monoclonal antibody (PharMingen, San Diego, CA), at a 1:200 dilution for 1 h at room temperature, followed by the addition of fluorescein-conjugated goat antihamster IgG (Organon Teknika, Durham, NC). All slides were mounted with Fluorotec mounting medium (Accurate Chemical Co., Westbury, NY). Fluorescent samples were viewed under a Microphot-FX microscope (Nikon, Garden City, NY).

Binding and localization of Echistatin
Iodination of echistatin. Peptide (10 µg) dissolved in 0.1 M borate buffer (pH 8.5) was iodinated using Bolton and Hunter reagent according to the method described by the manufacturer. The reaction mixture was purified on a Sep-Pak C18 cartridge (Millipore Corp., Milford, MA). The cartridge was washed with H2O and then eluted with 60% acetonitrile-0.1% trifluoroacetic acid. Labeled echistatin was immediately neutralized and stored in 0.1 mM HEPES (pH 7.4) containing 1% BSA. The specific activity was 2200 µCi/mmol.

Autoradiography of [125I]echistatin binding. Osteoblastic cells and bone marrow cells were cocultured on glass coverslips (12 mm in diameter) in the presence of 10 nM 1{alpha},25-(OH)2D3 as described above. The cells were washed twice with prewarmed {alpha}MEM containing 0.5% BSA and incubated with [125I]echistatin (20 nCi) in {alpha}MEM containing 0.5% BSA in the presence or absence of 1 µM cold echistatin. After incubation for 10 min at room temperature, cells were washed with cold {alpha}MEM (four times) containing 0.5% BSA and fixed in 0.1% glutaraldehyde and 4% paraformaldehyde, followed by TRAP activity staining as described above. The coverslips were mounted on a glass slide, coated with NTB2 emulsion (Eastman Kodak Co.), incubated in the dark at 4 C, and developed in Kodak D-19 developer. The pictures were taken in the same field under either phase contrast or episcopic polarization illumination, using IGS filter block (Nikon, Inc., Garden City, NY) with GIF prefilter (Nikon, Inc.) on a Microphot-FX microscope.

Cross-linking of [125I]echistatin. Enriched preparations of TRAP-positive OCLs were washed twice with PBS and once with HEPES buffer (50 mM HEPES, 140 mM NaCl, 1 mM MgCl2, and 1 mM CaCl2, pH 7.4) and harvested in HEPES buffer containing 1 mM phenylmethylsulfonylfluoride and leupeptin (20 µg/ml). The cells were incubated with [125I]echistatin (20 nCi) in the presence or absence of 1 µM echistatin for 30 min at room temperature. Bis(sulfosuccinimidyl) suberate (Pierce, Rockford, IL) was added at final concentration of 0.2 mM, and cells were incubated for 30 min at room temperature to cross-link [125I]echistatin. After the addition of Tris (pH 8.0) at a final concentration of 50 mM to stop the reaction, cells were washed three times with 10 mM Tris and 140 mM NaCl, pH 7.4, and solubilized in 10 mM Tris and 140 mM NaCl, pH 7.4, containing 2% Nonidet P-40 for 30 min before centrifugation. The supernatant was incubated with 0.5% BSA and 0.05% Tween-20 in the presence or absence of antihuman vitronectin receptor antiserum for 2 h at 4 C, followed by protein A-agarose (Pharmacia Biotech) for 1 h at room temperature. The polyclonal antivitronectin receptor ({alpha}vß3 integrin) antiserum was preabsorbed overnight (4 C) over a confluent monolayer of a human embryonic kidney 293 cell line that was stably transfected with human {alpha}vß5 (36). As 293 cells were previously shown to lack {alpha}vß3 expression (37), we found that this preabsorption step removed all cross-reactivity of this antiserum to other {alpha}v-associated integrins, in particular to {alpha}vß5. The preabsorbed antiserum was then used in immunoprecipitation and blocking studies. The immunoprecipitated products were then separated on a 6% SDS-PAGE (38). The gels were fixed, dried, and exposed to XAR-2 films with intensifying filters at -70 C.

Echistatin membrane binding. Membrane fractions prepared from OCLs were resuspended (40 µg/ml) in 10 mM HEPES, pH 7.5, and 150 mM NaCl containing 0.5% BSA. [125I]Echistatin (10 nCi/tube) was added to membrane suspension and incubated for 1 h at room temperature. Specific binding was determined by adding 1 µM unlabeled echistatin. Inhibition of echistatin binding by antibodies was performed by preincubating membranes with preimmune serum, antihuman vitronectin receptor antibodies, or antihuman fibronectin antibodies for 30 min (25 C) before the addition of radiolabeled echistatin. In this experiment, IgG fractions were purified from all antisera using IgPure isolation kit (UNISYN, Milford, MA).

Cell fusion assay
pOCs (10,000 cells/well) were plated on 96-well culture plates with 5 nM macrophage colony-stimulating factor (M-CSF) in the presence or absence of graded concentrations of echistatin (10-8–10-11 M) for 15 h, then fixed and stained for TRAP activity. In another set of experiments, pOCs were plated at increasing cell densities (3.0 x 103–3.0 x 105 cells/well) for 1 h in the absence of M-CSF or for 15 h in the presence of M-CSF and 10 nM echistatin. The number of multinucleated TRAP-positive cells with more than 10 nuclei was quantified using an image-analyzing system (Empire Imaging Systems, Milford, NJ). Results are expressed as a mean value of triplicate samples.

Cell migration assay
Osteoclast migration was assayed using a Boyden chamber-type apparatus (Neuro Probe, Inc., Cabin John, MD) using the pOC preparation, as reported previously with slight modification (36). Before assay, pOCs were preincubated with the graded concentrations of echistatin (10-7–10-11 M) in 0.1% BSA in {alpha}MEM for 20 min. M-CSF (5 nM), a chemotactic inducer of osteoclast migration (39), and 10% FBS in {alpha}MEM were placed in the bottom chamber. pOCs were then added to the upper chamber at a density of 10,000 cells/well and allowed to migrate through a polycarbonate filter (Neuro Probe) for 15 h in a humidified incubator at 37 C. After cells on the top of the filter were wiped away, cells migrating to the bottom of the filter were stained for TRAP activity. The number of TRAP-positive cells was counted, using an image-analyzing system (Empire Imaging Systems). Results are expressed as a mean value of triplicate or quadruplicate samples.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
The time course of OCL formation induced by 1{alpha},25-(OH)2D3 in the coculture of mouse osteoblasts and bone marrow cells was previously reported (3, 40). Multinucleated TRAP-positive cells appear on day 4, and their number peaks around day 7. In this case, the presence and regulation of vitronectin receptors {alpha}vß3 and {alpha}vß5 were initially examined using Northern analysis of total RNAs from the coculture. Two transcripts of the {alpha}v subunit, a major form at 9 kb and a minor form at 4 kb, were detected on day 2. Their levels gradually increased until days 5–6, when they peaked (3 times greater for the 9-kb transcript; Fig. 1Go). The ß3 transcript (6.9 kb) was barely detected on days 2 and 3, increased markedly from days 4–6, then slightly decreased on day 7 (Fig. 1Go). In contrast, there was no significant change in the expression of ß5 messenger RNA (mRNA; 3.5 kb) during the entire coculture period (Fig. 1Go). As a control, we also examined the expression of the fibronectin receptor in this coculture system. The mRNA levels of both subunits {alpha}5 (4.4 kb) and ß1 (3.2 kb) did not change during the coculture period (Fig. 1Go).



View larger version (55K):
[in this window]
[in a new window]
  		Figure 1. Time course of integrin subunit {alpha}v, ß3, ß5, {alpha}5, and ß1 mRNA expression in a coculture system. The murine osteoblastic cell line MB1.8 and bone marrow cells were cocultured, and at the indicated times, RNA was isolated as described in Materials and Methods. Total RNA (20 µg/lane) was subjected to Northern blot analysis using corresponding cDNA probes. The same filters were rehybridized with ß-actin probes. The sizes of the mRNA species were determined relative to 28S and 18S ribosomal RNA, as indicated on the right.

 
The presence of the integrin {alpha}vß3 in OCLs was also examined using enriched OCLs. OCLs were prepared by treating the coculture with collagenase and dispase to remove the majority of the osteoblastic MB1.8 cells. TRAP-positive cells that appear as reddish stained cells are shown in the coculture on day 6, before and after enzyme digestion (Fig. 2Go, A, C, and E). Both TRAP-positive mononuclear and multinucleated cells attach tightly to the plates and survive this treatment. The purity of the enriched culture, quantitated by the number of TRAP-positive cells over the total cell number, was approximately 71.2 ± 5.4%. As this figure refers to multinucleated cells, it is an underestimate of the purity of the enriched culture. When the area of the cells is measured, the fraction of TRAP-positive cell surface is greater than 90% (data not shown). The staining for TRAP activity results in severe quenching of immunofluorescence; we therefore used parallel cultures of OCLs for immunolocalization of the integrin {alpha}vß3 with an anti-ß3 integrin monoclonal antibody. As shown in Fig. 2BGo, in the coculture on day 6, we observed many mononuclear and large multinucleated cells expressing {alpha}vß3 integrins. After collagenase and dispase treatment, the {alpha}vß3-positive cells remained tightly attached to the culture dishes (Fig. 2Go, D and F). These cells included both mononuclear and multinucleated cells. The mononuclear cells appeared to express {alpha}vß3 at a higher level than the multinucleated cells. In multinucleated cells, the highest levels of integrin {alpha}vß3 were found in the microlamelopodia structure. A number of OCLs underwent fusion (Fig. 2Go, D and F), but {alpha}vß3 integrin was not concentrated at the site of fusion. These observations confirmed the presence of the {alpha}vß3 integrin in OCLs in the coculture system.



View larger version (124K):
[in this window]
[in a new window]
  		Figure 2. Coenrichment of TRAP-positive OCLs and {alpha}vß3 integrin-expressing cells from the coculture of murine osteoblastic cell line MB1.8 and bone marrow cells. The cocultures on day 7 were treated with collagenase-dispase as described in Materials and Methods. Parallel cultures were fixed and subjected to either TRAP staining or immunostaining using an monoclonal anti-ß3 integrin antibody as described in Materials and Methods. TRAP-positive mononuclear and multinucleated cells were present before (A) and after (C and E) the enzyme digestion treatment. Similarly, cells expressing ß3 integrin were present in the cocultures before (B) and after (D and F) treatment. TRAP-positive cells in four enriched cultures were quantitated, and their purity was approximately 71%. Bar = 50 µm.

 
The selective expression of {alpha}vß3 integrins in mononuclear and multinucleated OCLs was further supported by the comparison of {alpha}vß3 integrin expression in the enriched OCLs vs. all of the cocultured cells by Northern analysis. Enrichment of OCLs increased mRNA levels of both {alpha}v and ß3 integrins by about 5-fold and 10 fold, respectively, whereas ß5 mRNA levels were reduced 7-fold (Fig. 3Go). The level of ß-actin mRNA was not changed, which was used as a reference for quantitation in this experiment.



View larger version (35K):
[in this window]
[in a new window]
  		Figure 3. Enrichment of OCLs increases the expression level of {alpha}vß3 integrin and reduces that of {alpha}vß5 integrin. The coculture of the murine osteoblastic cell line MB1.8 and bone marrow cells on day 7 was treated with collagenase-dispase digestion as described in Materials and Methods. Expression levels of integrin subunits {alpha}v, ß3 and ß5 were analyzed using total RNA (20 µg/lane) prepared from the cells before (lanes 1, 3, and 5) or after (lanes 2, 4, and 6) enzyme treatment.

 
Next, to directly identify the target cells of echistatin, the binding of [125I]echistatin to cells at different stages of the coculture was examined. Figure 4Go shows reflection light microscopy images, in which yellow dots represent silver grains in the autoradiography of [125I]echistatin binding, side by side with phase contrast images of the same fields. TRAP-positive mononuclear cells appeared as early as day 3 in the coculture, and some of them showed weak binding of [125I]echistatin (data not shown). On day 4, significant echistatin binding colocalized with many TRAP-positive mononuclear cells (Fig. 4Go, A and B). In the early stage of the coculture, a small percentage (<10%) of TRAP-positive mononuclear cells did not bind echistatin. On day 5, [125I]echistatin bound to both mononuclear and some small multinucleated cells (Fig. 4Go, C and D). On day 7, the majority of echistatin binding colocalized with large multinucleated cells (Fig. 4Go, E and F). The presence of excess unlabeled echistatin completely competed radiolabeled echistatin binding to these cells (Fig. 4Go, G and H), suggesting that the target cells of echistatin are mononuclear and multinucleated OCLs.



View larger version (126K):
[in this window]
[in a new window]
  		Figure 4. Evidence for TRAP-positive mononuclear and multinucleated cells as [125I]echistatin-binding cells in the coculture system. Colocalization of TRAP-positive cells and echistatin binding cells in the coculture of the murine osteoblastic cell line MB1.8 and bone marrow cells was described in Materials and Methods. On day 4 (A and B), day 5 (C and D), and day 7 (E and F), cells were incubated with [125I]echistatin, followed by TRAP staining. Radioiodinated echistatin binding was competed with 1 µM unlabeled echistatin (G and H) in a coculture on day 7. TRAP-positive cells are shown as phase contrast images (A, C, E, and G), and autoradiographic images of the same optical field are shown (B, D, F, and H). A few TRAP-positive mononuclear cells do not bind echistatin (white arrow). Retraction of the osteoblastic monolayer was observed over several large multinucleated TRAP-positive cells (E), which caused uneven echistatin binding to these cells. Bar = 100 µm.

 
As echistatin binding correlated with the cell population that expressed {alpha}vß3 integrin, this integrin was a candidate for the echistatin receptor. This was further tested by measuring echistatin binding to membrane fractions prepared from the cocultured cells. As shown in Fig. 5AGo, echistatin bound specifically to the membrane fraction prepared from enriched OCLs. In contrast, equal amounts of membrane prepared from the nonadhering cells removed by enzyme digestion did not bind echistatin significantly. Most importantly, when MB1.8 cells were cultured in the presence of 1{alpha},25-(OH)2D3 alone, their membranes exhibited little echistatin binding. The specificity of echistatin binding to {alpha}vß3 in OCLs was demonstrated when membrane fractions of enriched OCLs were preincubated with either polyclonal antibodies raised against {alpha}vß3 or fibronectin receptor {alpha}5ß1 before [125I]echistatin binding. As shown in Fig. 5BGo, anti-{alpha}vß3 antibodies specifically blocked echistatin binding to osteoclast membranes, whereas anti-{alpha}5ß1 antibodies had no effect.



View larger version (12K):
[in this window]
[in a new window]
  		Figure 5. Evidence for {alpha}vß3 integrin-mediated echistatin binding in OCLs. A, Specific echistatin binding to membranes of OCLs. From left to right, binding of [125I]echistatin to membrane fractions (5 µg) prepared from MB1.8 cells (OB) treated with 1{alpha},25-(OH)2D3 for 7 days, from cells lifted (LC) from the cocultures (on day 7) by collagenase-dispase treatment, and from the enriched preparation of TRAP-positive cells (OC) that remained attached to the culture dishes. Specific echistatin binding was performed in the absence or presence of 1 µM unlabeled echistatin. B, [125I]Echistatin binding to OCL membranes in the absence (control) or presence of preimmune rabbit IgG (300 µg; +Preimm), with anti-{alpha}vß3 integrin polyclonal antibodies (300 µg; +VNRAb), or with anti-{alpha}vß5 integrin polyclonal antibodies (300 µg; +FNRAb). The results are expressed as the mean ± SD of triplicate samples.

 
Moreover, when radiolabeled echistatin was chemically cross-linked to the OCLs, two major bands on SDS-PAGE were identified (Fig. 6Go). Under nonreducing conditions, one was approximately 135 kDa and the other was 89 kDa (Fig. 6AGo, lane 2). Under reducing conditions, these proteins shifted to 123 and 104 kDa, respectively (Fig. 6BGo, lane 2). Several faint bands around 75 kDa and above 200 kDa were also detected. All cross-linked products were completely competed by excess unlabeled echistatin (Fig. 6Go, A and B, lane 3), suggesting specific binding of echistatin to these proteins. When the cross-linking products were immunoprecipitated with anti-{alpha}vß3 antibodies, the major immunoprecipitated proteins had the same apparent molecular masses (Fig. 6Go, A and B, lane 4). These results strongly suggest that echistatin binds to the {alpha}vß3 integrins in OCLs in the coculture system. We, therefore, examined the effect of echistatin on the coculture system to investigate the involvement of {alpha}vß3 integrins in osteoclast formation in vitro.



View larger version (42K):
[in this window]
[in a new window]
  		Figure 6. Immunoprecipitation of the cross-linking products of OCLs with [125I]echistatin by anti-{alpha}vß3 integrin polyclonal antibodies. An enriched preparation of TRAP-positive cells was incubated with [125I]echistatin in the presence (lanes 1 and 3) or absence (lanes 2, 4, and 5) of 1 µM cold echistatin for 30 min before cross-linking as described in Materials and Methods. The cell extract was immunoprecipitated with (lanes 3 and 4) or without (lanes 5) anti-{alpha}vß3 integrin polyclonal antibodies. The whole cell extract (lanes 1 and 2) or the immunoprecipitates (lanes 3, 4, and 5) were separated on 6% SDS-PAGE under nonreducing (A) or reducing (B) conditions. Molecular mass marker proteins as shown (myosin, 200 kDa; phosphorylase b, 92.5 kDa; BSA, 69 kDa).

 
When echistatin (100 nM) was present in the coculture for the 7-day culture period, formation of TRAP-positive multinucleated cells was completely inhibited, whereas formation of the TRAP-positive mononuclear cells was not impeded (Fig. 7AGo). As the formation of multinucleated cells is not observed until day 4 (Fig. 4Go), echistatin was added during days 0–4 and days 4–7. Addition of echistatin for the first 4 days had no effect on the subsequent formation of multinucleated cells; however, the presence of echistatin during the last 4 days completely inhibited the formation of multinucleated OCLs (Fig. 7BGo). These results suggest that echistatin may block the fusion of mononuclear osteoclasts into multinucleated osteoclasts.



View larger version (71K):
[in this window]
[in a new window]
  		Figure 7. Echistatin inhibits the late phase of OCL formation. The murine osteoblastic cell line MB1.8 and bone marrow cells were cocultured with or without echistatin treatment (100 nM). A, TRAP-positive cells are shown in parallel cultures in the absence (a) and presence (b) of echistatin for 7 days. Bar = 260 µm. B, Number of TRAP-positive multinucleated cells were quantitated in cultures on day 7, either without treatment (Control) or after echistatin was added for 7 days (0–7d), for the first 4 days (0–4d), or for the last 3 days (4 5 6 7 ). The results are expressed as the mean ± SD of four cultures.

 
We used peptide analogs of echistatin to correlate the specificity of echistatin inhibition of OCL formation and inhibition of {alpha}vß3-mediated osteoclast membrane binding. Echistatin inhibited the formation of multinucleated OCLs with an IC50 of 0.7 nM, whereas A24-echistatin, in which the arginine residue in RGD is substituted with an alanine, had no effect at 1 mM (Table 1Go). Further structure-activity relationship studies showed that the oxidation of methionine at position 28 [M(O)28-echistatin] or substitution with phenylalanine for aspartic acid at position 27 or of leucine for methionine at position 28 also reduced the affinity of echistatin analogs to its receptor 400- and 20-fold, respectively. Surprisingly, deletion of only six residues from the carboxyl-terminus of echistatin [echi(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43)] resulted in a 10-fold decrease in echistatin affinity to the receptor, indicating that changes in the sequence proximal to and distal to the RGD moiety result in a significant loss in receptor affinity. This suggests that the RGD moiety alone was necessary, but not sufficient, to maintain the high affinity and selectivity of echistatin binding to {alpha}vß3 integrins. Furthermore, a direct role of {alpha}vß3 integrins in the formation of OCLs was suggested by the similar affinities (IC50 values) for the binding of echistatin analogs to OCL membranes and their inhibition of TRAP-positive multinucleated cell formation (Table 1Go).


View this table:
[in this window]
[in a new window]
  		Table 1. Inhibition of TRAP(+) multinucleated cell formation and membrane binding by echistatin

 
We confirmed the inhibitory effect of echistatin on the fusion of mononuclear osteoclasts using pOCs, which are about 95% pure mononuclear osteoclasts (27). As it has been recently reported that M-CSF can induce osteoclast fusion (41), we used M-CSF (5 nM) to induce pOC fusion. As shown in Fig. 8Go, echistatin inhibited M-CSF-induced pOC fusion dose dependently (IC50 = 0.6 nM), whereas A24-echistatin did not. This IC50 is in agreement with the inhibitory effect of echistatin on the formation of multinucleated OCLs in the coculture system. In the absence of M-CSF, pOCs alone could not survive for more than 4 h (data not shown); we therefore could not determine the effect of echistatin on pOC fusion in the absence of this growth factor.



View larger version (57K):
[in this window]
[in a new window]
  		Figure 8. Echistatin inhibits the fusion of mononuclear pOCs induced by M-CSF. pOCs were prepared as described in Materials and Methods. pOCs were added to 96-well plates in the presence of increasing concentrations of echistatin or A24-echistatin. Cell fusion was induced by M-CSF (5 nM). After culture for 15 h, cells were stained for TRAP activity. The number of multinucleated TRAP-positive cells with more than 10 nuclei was counted. Results are expressed as the mean of triplicate samples ± SD.

 
As cell fusion involves migration, cell-cell recognition, and membrane fusion, we examined the effect of echistatin on the migration of pOCs. As shown in Fig. 9Go, echistatin inhibited pOC migration induced by M-CSF in a dose-dependent manner with an IC50 of 1.0 nM. Similar to the inhibitory effect of echistatin on the formation of multinucleated OCLs, inhibition of pOC migration was RGD dependent, as demonstrated by the A24-echistatin analog. Taken together, the above observations suggested that the blocking effect of echistatin on OCL formation may be due to the inhibition of pOC migration.



View larger version (61K):
[in this window]
[in a new window]
  		Figure 9. Echistatin inhibits the migration of mononuclear pOCs induced by M-CSF. pOCs were prepared as described in Materials and Methods. pOCs were added to the upper well of the migration chamber in the presence of increasing concentrations of echistatin or A24-echistatin and assayed for migration toward the lower wells containing 5 nM M-CSF. After culture for 15 h, migrating pOCs were stained for TRAP activity. The number of cells was counted. Results are expressed as the mean of triplicate samples ± SD (A). Migrating pOCs are stained for TRAP in the presence of 10 nM A24-echistatin (B), whereas few pOCs with 10 nM echistatin cannot migrate (C). Bar = 20 µm.

 
To further test this hypothesis, we examined the relationship between pOC plating density and fusion, reasoning that migration distance decreases at higher density. We found that pOCs plated for 1 h in the absence of M-CSF at a plating density exceeding 3.0 x 104 cells/well spontaneously fused, to form OCLs with more than 10 nuclei (Fig. 10AGo). Furthermore, echistatin (10 nM) had no effect on pOC fusion at the high cell density (Fig. 10BGo) compared with the complete inhibition of osteoclast multinucleation at the lower cell density (Fig. 8Go). This result suggested that {alpha}vß3 integrins play a major role in the migration of osteoclast precursors that precedes the fusion of pOCs in vitro.



View larger version (70K):
[in this window]
[in a new window]
  		Figure 10. A, Increase in the cell density of mononuclear pOCs leads to induced their spontaneous fusion. pOCs were plated in 96-well plates at the indicated densities for 1 h, followed by staining for TRAP activity. The number of multinucleated TRAP-positive cells with more than 10 nuclei was counted. Results are expressed as the mean of triplicate samples ± SD. Note that the majority of pOCs plated at high density (3 x 104 cells/well) fused (b, inset), whereas pOCs at a low density (3 x 103 cells/well) remained as mononuclear TRAP-positive cells (a, inset). Bars = 10 µm. B, pOCs fusion at high cell density is not blocked by echistatin. pOCs were plated into 96-well plates at the indicated densities in the presence of 10 nM echistatin and 5 nM M-CSF. After 15 h in culture, the number of multinucleated TRAP-positive cells with more than 10 nuclei was counted. Results are expressed as the mean of triplicate samples ± SD.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Two integrins are highly expressed in osteoclasts, the collagen/laminin receptor, {alpha}2ß1 integrin (very late antigen-2), and the vitronectin receptor, {alpha}vß3 integrin (9). The {alpha}vß3 integrin was proposed to play a role in osteoclast migration, attachment, and, therefore, bone resorption (16, 17, 18, 19). Mononuclear osteoclast precursors also express {alpha}vß3 integrin, but its function in these cells is not clear (42, 43). In this report, we demonstrate that {alpha}vß3 integrin is the major vitronectin receptor in mouse multinucleated and mononuclear OCLs. We also present data that strongly suggest the involvement of this receptor in one of the later phases of osteoclast maturation in vitro, the formation of multinucleated osteoclasts from mononuclear osteoclasts.

First, we confirmed that the expression of {alpha}vß3 integrin is up-regulated during osteoclastogenesis in a coculture system and is highly increased in the enriched preparation of OCLs, examined by mRNA levels and immunolocalization, which suggests that this receptor is expressed in OCLs. The expression of the ß5 integrin subunit mRNA was detected in cocultured cells (OCLs and MB1.8 cells), but its level was not changed during the coculture period and was drastically reduced in the enriched preparation of OCLs, indicating that {alpha}vß5 is not a major integrin in murine osteoclasts and appears to be the vitronectin receptor in MB1.8 cells. These findings support previous observations showing that {alpha}vß5 is expressed mainly in osteoblastic cells (10). Another RGD- dependent integrin, the fibronectin receptor {alpha}5ß1, was also expressed in the cocultured cells, and its level was not changed in the enriched OCL preparation (data not shown), suggesting its expression in both cell types.

The possible role of {alpha}vß3 integrin in osteoclast formation in this system was studied by using the snake venom echistatin after demonstration of its specific and selective binding to the vitronectin receptor in OCLs. Echistatin was previously colocalized with {alpha}vß3 integrin in both chicken and rat osteoclasts (25) and was shown to inhibit osteoclast attachment to bone (44) and bone resorption in vitro (15) and in vivo (22, 23, 24). In this study, we demonstrate that both multinucleated and mononuclear OCLs are major echistatin-binding cells, and no binding to osteoblasts or other bone marrow-derived cells was significantly detected. This finding was further supported by the specific binding of echistatin to membranes prepared from enriched OCLs. Furthermore, the echistatin receptor in OCLs was shown to be the {alpha}vß3 integrin, as echistatin binding to osteoclast membranes was blocked by anti-{alpha}vß3 integrin polyclonal antibodies and was not affected by antifibronectin receptor antibodies. To further substantiate that {alpha}vß3 integrin is the major echistatin receptor in this coculture system, we show that virtually all the [125I]echistatin cross-linked proteins in OCLs were immunoprecipitated by anti-{alpha}vß3 integrin antibodies. We also ruled out directly the possible involvement of two other RGD-dependent integrins, {alpha}vß5 and {alpha}5ß1, in echistatin binding to OCLs. Northern analysis reveals that osteoblasts express the majority of {alpha}vß5 and {alpha}5ß1 mRNA in the coculture, and the membrane fraction prepared from osteoblasts did not bind echistatin. Furthermore, separate observations in our laboratory show that echistatin does not inhibit cell attachment to vitronectin via {alpha}vß5 integrin or to fibronectin via {alpha}5ß1 integrin (unpublished observations). Taken together, these findings strongly suggest that in this coculture system, echistatin binds specifically to {alpha}vß3 integrin, which is highly expressed in TRAP-positive mononuclear and multinucleated cells. We concluded therefore that its effects in this coculture are related to the function of {alpha}vß3 integrin.

Echistatin inhibited the formation of TRAP-positive multinucleated cells (IC50 = 0.7 nM), but did not inhibit the appearance of TRAP-positive mononuclear cells on days 4–5 of the coculture, suggesting that echistatin inhibits osteoclast multinucleation. This effect was reversible, as multinucleation proceeded again after echistatin washout (data not shown). At the concentrations that inhibit osteoclast formation, echistatin did not cause detachment of either the osteoblastic MB1.8 cells or pOCs from the culture dishes, even when echistatin was added to the cocultures from the time of initial plating. This finding indicates that the effects of echistatin are not due to cell removal or cellular toxicity of osteoclast progenitors or MB1.8 cells. In fact, expression of ß3 integrin was not detected in the coculture until day 4, coincident with the appearance of prefusion osteoclasts. Furthermore, the addition of echistatin to the coculture from the time of plating did not decrease the appearance of mononuclear osteoclast precursors. By morphological analysis, we observed that pOCs attached tightly to the surrounding MB1.8 cells. Interestingly, the multinucleated TRAP-positive cells formed underneath the monolayer of MB1.8 cells, as shown in Fig. 2Go. This suggests that in an intact coculture system, adhesion of pOCs and OCLs might be mediated by other adhesion receptors, including the collagen receptor {alpha}2ß1 and the fibronectin receptors (11).

We also confirmed the inhibitory effect of echistatin on osteoclast multinucleation, using pOCs that are virtually pure mononuclear osteoclasts. Echistatin inhibited M-CSF-induced fusion of pOCs in a dose-dependent manner (IC50 = 0.6 nM). The effects of echistatin were RGD dependent, as documented by substituting alanine for arginine. It has been previously reported that RGD-containing peptides inhibit bone resorption in rat bone organ culture (45) and block the appearance of TRAP-positive multinucleated osteoclasts in calcified matrix, whereas the TRAP-positive mononuclear cells were unaffected in the periosteum. The researchers suggested that integrins, possibly {alpha}vß3, play an important role in osteoclast multinucleation preceding the actual resorption activity. This observation is consistent with our findings and provides ex vivo evidence that supports the hypothesis that {alpha}vß3 integrin may play a role in the later phase of osteoclast formation. The IC50 of the inhibitory effect of echistatin on multinucleation of osteoclasts is similar to that for echistatin inhibition of bone resorption by rat osteoclasts (0.1 nM) (15) and for the shape changes (retraction) produced in rat or chick osteoclasts on serum-coated glass (0.78 nM) (44). This is also consistent with echistatin binding to the membranes of OCLs (Kd = 0.6 nM) and further supports the hypothesis that echistatin interferes with {alpha}vß3 integrin function in this system. By comparison, the IC50 for echistatin inhibition of platelet aggregation is 30 nM (21).

The formation of multinucleated osteoclasts is proposed to involve at least three basic events: migration of precursor cells, homotypic cell-cell recognition, and membrane fusion in conjunction with the structural reorganization of the multinucleated cell. The role of {alpha}vß3 integrin in cell migration has been extensively characterized in endothelial cells, smooth muscle cells, and neutrophils (46, 47, 48, 49). By analogy, a likely target of echistatin action is the migration of mononuclear osteoclasts, although effects on others events cannot be excluded. As several lines of evidence, including this study, have shown that mononuclear osteoclasts in culture express {alpha}vß3 integrin (27, 41, 42), we examined the effect of echistatin on the migration of mononuclear osteoclasts stimulated by M-CSF, which was reported to induce osteoclast migration (39). Indeed, echistatin inhibited pOC migration in a RGD- and dose-dependent manner, similar to that found to inhibit the formation of multinucleated OCLs in the coculture system. Simpson and Horton (50) reported that vitronectin receptors are expressed by periosteal mononuclear cells and suggested that these cells home to bone and fuse to form osteoclasts. Based on the above observations, we propose that one of the major roles of the integrin {alpha}vß3 is mediation of osteoclast migration.

The following observations favor a role for {alpha}vß3 integrins in the migration, rather than the fusion, of osteoclast precursors. First, in migration studies, echistatin treatment decreases the number of mononuclear TRAP-positive cells on the membrane surface facing the lower Boyden chamber; second, increased pOC cell density induces their spontaneous fusion, suggesting that the proximity of these cells can result in fusion. The fusion of osteoclast precursors at high cell density appears to be insensitive to echistatin treatment. Therefore, we suggest that in the coculture system, the primary effect of {alpha}vß3 integrin on osteoclast formation is based on its role in the migration of osteoclast precursors, rather than direct involvement in the fusion process.

This inhibitory effect of echistatin on osteoclastogenesis is different from that of osteoprotegrin, which was identified as a soluble receptor of TRANCE/RANK ligand/osteoclast differentiation factor (51, 52, 53, 54). Osteoprotegrin inhibits the appearance of TRAP-positive cells in the coculture system, whereas echistatin does not inhibit the differentiation process from hemopoietic progenitors to TRAP-positive cells, but blocks the later step, that is their fusion into multinucleated TRAP-positive cells.

Interestingly, we recently reported that echistatin inhibits bone resorption in vivo in estrogen-deficient mice and rats (23) and in mice with secondary hyperparathyroidism (24) without decreasing osteoclast number on the bone surface. This observation indicates that in vivo echistatin does not inhibit the multinucleation of osteoclasts. In addition, no gross change was detected in osteoclastic ultrastructure, including the organization of the ruffled border and sealing zone, in echistatin-treated mice (24). These observations indicated that in vivo {alpha}vß3 integrin plays a rate-limiting role in osteoclast functions other than in cell differentiation and cell adhesion. The present study leads by analogy to the supposition that {alpha}vß3 integrins may mediate the migration not only of osteoclast precursors, but mature osteoclasts as well. Inhibition of mature osteoclast migration during the resorption cycle would result in less efficient bone resorption. The echistatin effect on cell migration could thus explain both the in vitro and in vivo effects on osteoclasts. However, it should be borne in mind that in vivo there is no reduction in osteoclast number, suggesting that at the given concentrations, echistatin does not inhibit the in vivo preosteoclast recruitment.

The ligand of {alpha}vß3 integrin with which echistatin competes remains unknown. Echistatin inhibits the binding of {alpha}vß3 integrin to all of its natural ligands, including osteopontin and bone sialoprotein, two RGD-containing extracellular matrix proteins abundant in bone (55). Osteopontin is produced by both osteoblasts (56) and osteoclasts (57, 58). However, in preliminary experiments, soluble osteopontin or antiosteopontin antibodies did not inhibit the formation of multinucleated OCLs (data not shown).

The intracellular signals that mediate the effects of echistatin remain to be determined. Integrins activate multiple signaling pathways, including the elevation of intracellular Ca2+, lipid turnover, and tyrosine phosphorylation, leading to cytoskeletal rearrangement and de novo gene expression (59). Adhesion of neutrophils via ß2 integrins and of endothelial cells via {alpha}vß3 integrins induced a transient increase in the intracellular Ca2+ concentration (47, 60, 61). In isolated rat osteoclasts and OCLs generated in this coculture, soluble ligands of {alpha}vß3 integrin, including echistatin, also increased the intracellular Ca2+ concentration (62, 63). In human OCLs, calcium-mediated intracellular signaling was reported to be involved in osteoclast chemotaxis (64, 65). Further studies are required to clarify the role of changes in intracellular Ca2+ and the involvement of other signal transduction pathways in the effects of echistatin.

In conclusion, we have demonstrated that the target of echistatin in osteoclasts is {alpha}vß3 integrin and that this integrin plays a crucial role in the multinucleation of osteoclasts in vitro, possibly via cell migration. Many key questions remain to be answered regarding the role of {alpha}vß3 integrin in osteoclast function in vivo, because, as mentioned above, infusion of inhibitors of {alpha}vß3 to hypercalcemic or ovariectomized rodents blocked bone resorption, but had little effect on the number of osteoclasts on the bone surface (23, 24). These observations suggest that inhibition via this integrin did not affect osteoclast differentiation, at least within the short duration of treatment (up to 4 weeks). As efficient bone resorption requires osteoclast migration from site to site, both in vitro and in vivo echistatin effects can be explained by a direct role of the {alpha}vß3 integrin in osteoclast migration.


   Acknowledgments
 
We are grateful to Dr. V. M. Garsky for echistatin and its analogs, to Dr. Keiko O. Simon for advice on the migration assay, and to Gregory Wesolowski and Rose M. Nagy for providing technical support during this study.

Received April 28, 1998.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Athanasou NA 1996 Cellular biology of bone-resorbing cells. J Bone Joint Surg 78A:1096–1112
  2. Roodman GD 1996 Advances in bone biology: the osteoclast. Endocr Rev 17:308–332[Abstract]
  3. Suda T, Udagawa N, Takahashi N 1996 Cells of bone: osteoclast generation. In: Bilezikian JP, Raisz LG, Rodan, GA (eds) Principles of Bone Biology. Academic Press, San Diego, pp 87–102
  4. Nijweide PJ, Grooth R 1992 Ontogeny of the osteoclast. In: Rifkin BR, Gay CV (eds) Biology and Physiology of the Osteoclast. CRC Press, Boca Raton, pp 81–104
  5. Baron R, Neff L, Louvard D, Courtoy JP 1985 Cell-mediated extracellular acidification and bone resorption: evidence for a low pH in resorbing lacunae and localization of a 100-kD lysozomal membrane protein at the osteoclast ruffled border. J Cell Biol 101:2210–2222[Abstract/Free Full Text]
  6. Blair HC, Kahn AJ, Crouch EC, Jeffrey JJ, Teitelbaum SL 1986 Isolated osteoclasts resorb the organic and inorganic components of bone. J Cell Biol 102:1164–1172[Abstract/Free Full Text]
  7. Väänänen HK, Karhukorpi EK, Sundquist K, Wallmark B, Roininen I, Hentunen T, Tuukkanen J, Lakkakorpi PT 1990 Evidence for the presence of a proton pump of the vacuolar H+-ATPase type in the ruffled borders of osteoclasts. J Cell Biol 111:1305–1311[Abstract/Free Full Text]
  8. Hynes RO 1992 Integrins: versatility, modulation, and signaling in cell adhesion. Cell 69:11–25[CrossRef][Medline]
  9. Horton MA, Davies J 1989 Perspectives: adhesion receptors in bone. J Bone Miner Res 4:803–807[Medline]
  10. Hultenby K, Reinholt FP, Heinegård D 1993 Distribution of integrin subunits on rat metaphyseal osteoclasts and osteoblasts. Eur J Cell Biol 62:86–93[Medline]
  11. Nesbitt S, Nesbit A, Helfrich M, Horton M 1993 Biochemical characterization of human osteoclast integrins: osteoclasts express {alpha}vß3, {alpha}2ß1, and {alpha}vß1 integrins. J Biol Chem 268:16737–16745[Abstract/Free Full Text]
  12. Lakkakorpi PT, Horton MA, Helfrich MH, Väänänen HK 1991 Vitronectin receptor has a role in bone resorption but does not mediate tight sealing zone attachment of osteoclasts to the bone surface. J Cell Biol 115:1179–1186[Abstract/Free Full Text]
  13. Lakkakorpi PT, Helfrich MH, Horton MA, Väänänen HK 1993 Spatial organization of microfilaments and vitronectin receptor, {alpha}vß3, in osteoclasts. A study using confocal laser scanning microscopy. J Cell Sci 104:663–670[Abstract]
  14. Nakamura I, Gailit J, Sasaki T 1996 Osteoclast integrin {alpha}vß3 is present in the clear zone and contributes to cellular polarization. Cell Tissue Res 286:507–515[CrossRef][Medline]
  15. Sato M, Sardana MK, Grasser WA, Garsky VM, Murray JM, Gould RJ 1990 Echistatin is a potent inhibitor of bone resorption in culture. J Cell Biol 111:1713–1723[Abstract/Free Full Text]
  16. Helfrich MH, Nesbitt SA, Dorey EL, Horton MA 1992 Rat osteoclasts adhere to a wide range of RGD (Arg-Gly-Asp) peptide-containing proteins, including the bone sialoproteins and fibronectin, via a ß3 integrin. J Bone Miner Res 7:335–343[Medline]
  17. Chambers TJ, Fuller K, Darby JA, Pringle JAS, Horton MA 1986 Monoclonal antibodies against osteoclasts inhibit bone resorption in vitro. Bone Miner 1:127–135[Medline]
  18. Horton MA, Taylor ML, Arnett TR, Helfrich MH 1991 Arg-Gly-Asp (RGD) peptides and the anti-vitronectin receptor antibody 23C6 inhibit dentine resorption and cell spreading by osteoclasts. Exp Cell Res 195:368–375[CrossRef][Medline]
  19. Flores ME, Norgård M, Heinegård D, Reinholt FP, Andersson G 1992 RGD-directed attachment of isolated osteoclasts to osteopontin, bone sialoprotein, and fibronectin. Exp Cell Res 201:526–530[CrossRef][Medline]
  20. Nakamura I, Takahashi N, Sasaki T, Jimi E, Kurokawa T, Suda T 1996 Chemical and physical properties of the extracellular matrix are required for the actin ring formation in osteoclasts. J Bone Miner Res 11:1873–1879[Medline]
  21. Gan Z, Gould RJ, Jacobs JW, Friedman PA, Polokoff MA 1988 Echistatin: a potent platelet aggregation inhibitor from the venom of the viper, Echis carinatus. J Biol Chem 263:19827–19832[Abstract/Free Full Text]
  22. Fisher JE, Caulfield MP, Sato M, Quartuccio HA, Gould RJ, Garsky VM, Rodan GA, Rosenblatt M 1993 Inhibition of osteoclastic bone resorption in vivo by echistatin, an "arginyl-glycyl-aspartyl" (RGD)-containing protein. Endocrinology 132:1411–1413[Abstract]
  23. Yamamoto M, Fisher JE, Gentile M, Seedor JG, Leu C-T, Rodan SB, Rodan GA 1998 The integrin ligand echistatin prevents bone loss in overiectomized mice and rats. Endocrinology 139:1411–1419[Abstract/Free Full Text]
  24. Masarachia P, Yamamoto M, Leu C-T, Rodan G, Duong LT 1998 Histomorphometric evidence for echistatin inhibition of bone resorption in mice with secondary hyperparathyroidism. Endocrinology 139:1401–1410[Abstract/Free Full Text]
  25. Sato M, Garsky VM, Majeska RJ, Einhorn TA, Murray J, Tashjian Jr AH, Gould RJ 1994 Structure-activity studies of the s-echistatin inhibition of bone resorption. J Bone Miner Res 9:1441–1449[Medline]
  26. Takahashi N, Akatsu T, Udagawa N, Sasaki T, Yamaguchi A, Moseley JM, Martin TJ, Suda T 1988 Osteoblastic cells are involved in osteoclast formation. Endocrinology 123:2600–2602[Abstract]
  27. Wesolowski G, Duong LT, Lakkakorpi PT, Nagy RM, Tezuka K, Tanaka H, Rodan GA, Rodan SB 1995 Isolation and characterization of highly enriched, refusion mouse osteoclastic cells. Exp Cell Res 219:679–686[CrossRef][Medline]
  28. Akatsu T, Tamura T, Takahashi N, Udagawa N, Tanaka S, Sasaki S, Yamaguchi A, Nagata N, Suda T 1992 Preparation and characterization of a mouse osteoclast-like multinucleated cell population. J Bone Miner Res 7:1297–1306[Medline]
  29. Duong LT, Hadac EM, Miller LJ, Vlasuk GP 1989 Purification and characterization of the rat pancreatic cholecystokinin receptor. J Biol Chem 264:17990–17996[Abstract/Free Full Text]
  30. Chomczynski P, Sacchi N 1987 Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 162:156–159[Medline]
  31. Suzuki S, Argraves WS, Arai H, Languino LR, Pierschbacher MD, Ruoslahti E 1987 Amino acid sequence of the vitronectin receptor alpha subunit and comparative expression of adhesion receptor mRNAs. J Biol Chem 262:14080–14085[Abstract/Free Full Text]
  32. Frachet P, Uzan G, Thevenon D, Denarier E, Prandini MH, Marguerie G 1990 GPIIb and GPIIIa amino acid sequences deduced from human megakaryocyte cDNAs. Mol Biol Rep 14:27–33[CrossRef][Medline]
  33. Holers VM, Ruff TG, Parks DL, McDonald JA, Ballard LL, Brown EJ 1989 Molecular cloning of a murine fibronectin receptor and its expression during inflammation. J Exp Med 169:1589–1605[Abstract/Free Full Text]
  34. Argraves WS, Suzuki S, Arai H, Thompson K, Pierschbacher MD, Ruoslahti E 1987 Amino acid sequence of the human fibronectin receptor. J Cell Biol 105:1183–1190[Abstract/Free Full Text]
  35. Suzuki S, Huang ZS, Tanihara H 1990 Cloning of an integrin ß subunit exhibiting high homology with integrin ß-3 subunit. Proc Natl Acad Sci USA 87:5354–5358[Abstract/Free Full Text]
  36. Simon KO, Nutt EM, Abraham DG, Rodan GA, Duong LT 1997 The {alpha}vß3 integrin regulates {alpha}5ß1-mediated cell migration toward fibronectin. J Biol Chem 272:29380–29389[Abstract/Free Full Text]
  37. Bodary SC, McLean JW 1990 The integrin ß1 subunit associates with the vitronectin receptor {alpha}v subunit to form a novel vitronectin receptor in a human embryonic kidney cell line. J Biol Chem 265:5938–5941[Abstract/Free Full Text]
  38. Laemmli UK 1970 Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680–685[CrossRef][Medline]
  39. Fuller K, Owens JM, Jagger CJ, Wilson A, Moss R, Chambers TJ 1993 Macrophage colony-stimulating factor stimulates survival and chemotactic behavior in isolated osteoclasts. J Exp Med 178:1733–1744[Abstract/Free Full Text]
  40. Tanaka S, Takahashi N, Udagawa N, Tamura T, Akatsu T, Stanley ER, Kurokawa T, Suda T 1993 Macrophage colony-stimulating factor is indispensable for both proliferation and differentiation of osteoclast progenitors. J Clin Invest 91:257–263
  41. Amano H, Yamada S., Felix R 1998 Colony-stimulating factor-1 (CSF-1) stimulates the fusion process in osteoclasts. J Bone Miner Res 13:846–853[CrossRef][Medline]
  42. Kurihara N, Gluck S, Roodman GD 1990 Sequential expression of phenotype markers for osteoclasts during differentiation of precursors for multinucleated cells formed in long term human marrow culture. Endocrinology 127:3215–3221[Abstract]
  43. Prallet B, Male P, Neff L, Baron R 1992 Identification of a functional mononuclear precursor of the osteoclast in chicken medullary bone marrow cultures. J Bone Miner Res 7:405–414[Medline]
  44. Horton MA, Doray EL, Nesbitt SA, Samanen J, Ali FE, Stadel JM, Nichols A, Greig R, Helfrich MH 1993 Modulation of vitronectin receptor-mediated osteoclast adhesion by Arg-Gly-Asp peptide analogs: a structure-functional analysis. J Bone Miner Res 8:527–533P[Medline]
  45. Van der Pluijm G, Mouthaan H, Baas C, De Gro