Beta1 integrin/focal adhesion kinase-mediated signaling induces intercellular adhesion molecule 1 and receptor activator of nuclear factor kappaB ligand on osteoblasts and osteoclast maturation.

We have assessed characteristics of primary human osteoblasts, shedding light on signaling mediated by beta1 integrin. beta1 integrins are major receptors for these matrix glycoproteins. 1) Integrins beta1, alpha2, alpha3, alpha4, alpha5, alpha6, and alphav were highly expressed on primary osteoblasts. 2) Engagement of beta1 integrins on osteoblasts by cross-linking with specific antibody or ligand matrices, such as fibronectin or collagen, augmented expression of intercellular adhesion molecule 1 (ICAM-1) and receptor activator of nuclear factor kappaB ligand (RANKL) on the surface. 3) Up-regulation of ICAM-1 and RANKL on osteoblasts by beta1 stimulation was completely abrogated by pretreatment with herbimycin A and genistein, tyrosine kinase inhibitors, or transfection of dominant negative truncations of focal adhesion kinase (FAK). 4) Engagement of beta1 integrins on osteoblasts induced tartrate-resistant acid phosphatase-positive multinuclear cell formation in the coculture system of osteoblasts and peripheral monocytes. 5) Up-regulation of tartrate-resistant acid phosphatase-positive multinuclear cell formation by beta1 stimulation was completely abrogated by transfection of dominant negative truncations of FAK. Our results indicate that beta1 integrin-dependent adhesion of osteoblasts to bone matrices induces ICAM-1 and RANKL expression and osteoclast formation via tyrosine kinase, especially FAK. We here propose that beta1 integrin/FAK-mediated signaling on osteoblasts could be involved in ICAM-1- and RANKL-dependent osteoclast maturation.

Bone metabolism in health and disease is based on a selfregulating cellular event. The two major processes of bone remodeling, bone formation and resorption, are closely regulated by intercellular signaling involving soluble factors, systemic hormones, and cellular adhesion (1)(2)(3)(4)(5). Osteoblasts play a central role in bone formation by synthesizing multiple bone matrix proteins and by differentiation into osteocytes. However, osteoblasts also regulate osteoclast maturation by producing bone-resorbing cytokines and by direct cell attachment, resulting in bone resorption (6 -8). Cell adhesion of osteoblasts and osteoclastic precursors of hematopoietic origin is a prerequisite for osteoclast maturation. Several studies have demonstrated that interaction of receptor activator of nuclear factor B ligand (RANKL) 1 on osteoblasts and RANK on osteoclast precursors provides an essential signal to osteoclast precursors for their maturation into resorbing cells (9 -11). We have previously reported that human osteoblasts express intercellular adhesion molecule (ICAM)-1 and that interaction between ICAM-1, expressed on osteoblasts, and leukocyte function-associated antigen (LFA)-1, expressed on monocytes, is required for osteoclast maturation by RANKL on osteoblasts (12).
Adhesion molecules play a fundamental role in cell-to-cell and cell-to-extracellular matrix (ECM) interactions. However, recent findings have indicated that certain adhesion molecules not only function as glue but also regulate several cellular functions by transducing signaling. We have reported that ICAM-1 on rheumatoid synovial cells induced transcription of interleukin-1␤ by activation of a nuclear factor, AP-1, and that stimulation of ␤ 1 integrin up-regulated ICAM-1 and Fas, and Fas mediated apoptosis of rheumatoid synovial cells through focal adhesion kinase (FAK) (13,14). These results prompted us to investigate the adhesion molecules involved in regulating the expression of other adhesion molecules, such as ICAM-1 and RANKL, on human osteoblasts. Cell adhesion to matrices is primarily mediated by integrins, cell surface receptors that comprise an expanding family of transmembrane heterodimers of ␣ and ␤ subunits (15)(16)(17)(18). Interaction of integrins with their protein ligands increases tyrosine phosphorylation and triggers the assembly of cytoskeletal proteins, signaling complexes including FAKs, and their substrates into membrane-substratum junctions referred to as focal adhesions (19 -23).
Although osteoblasts are always surrounded by and encounter ECMs including type I collagen and fibronectin mainly through ␤ 1 integrin, the relevance of ␤ 1 integrin to the intracellular signaling and functions in osteoblasts remains unclear. It is well established that osteoblast differentiation and maturation are regulated by their interaction with ECMs such as type I collagen (24,25). However, such an adhesive interaction may act on osteoblasts to modulate bone metabolism, not only bone formation by activating osteoblasts to proliferate and * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
¶ To whom correspondence should be addressed. synthesize bone matrix protein but also bone resorption by indirectly activating osteoclast function and differentiation mediated through osteoblasts. The aim of the present study was to determine the role of ␤ 1 integrin-mediated signaling in the regulation of cell surface adhesion molecules on osteoblasts. Our results demonstrate that engagement of ␤ 1 integrin by a specific antibody or ligand matrices up-regulated ICAM-1 and RANKL expression on osteoblasts and induced osteoclast formation via tyrosine kinase, especially FAK.

EXPERIMENTAL PROCEDURES
The study protocol was approved by the Human Ethics Review Committee of the University of Occupational and Environmental Health, Japan, and a signed consent form was obtained from each subject prior to taking tissue samples used in the present study.
Purification of Human Osteoblastic Cells-Osteoblast-like cells were purified from metaphyseal trabecular bone in the proximal femur of four osteoarthritis patients during total hip arthroplasty by the established procedures of Russell and colleagues (26,27). After removing pieces of cortical bone, articular cartilage, and soft connective tissue, the fragments were cut into small pieces and washed extensively. The bone explants were cultured in Dulbecco's modified Eagle's medium (DMEM, Invitrogen) containing 10% fetal calf serum (FCS, Invitrogen) in 25-cm 2 culture flasks (Falcon, Lincoln Park, NJ) in a humidified 5% CO 2 atmosphere. When cell monolayers were confluent after the 6 -8week culture, the explants were removed, and the cells were replated and incubated, resulting in new cellular outgrowth and eventually a confluent monolayer of cells. At confluence, the cells were trypsinized, passaged at a 1:3 split ratio, and recultured. The medium was changed twice each week, and the cells were used after three to seven passages. The obtained cells showed a flattened polygonal shape with multiple spindle legs and possessed characteristics of osteoblast-like phenotype including osteocalcin, bone sialoprotein, type I collagen, and bone alkaline phosphatase as described previously (27).
Stimulation of Osteoblasts by ␤ 1 Integrin Using mAbs and Substrates-After cells were cultured to subconfluence, the medium was changed to DMEM containing 1% FCS on the day before assay. The obtained cells were then incubated with anti-CD29 (␤ 1 integrin) mAbs such as MAB13 and control mAbs (10 g/ml) in DMEM without FCS for 30 min at 37°C. Signal inhibitors were added at the indicated concentration for a 30-min incubation prior to cell stimulation. After washing the cells three times, 1 g/ml goat anti-rat IgG-Fc was added as the second Ab for ␤ 1 cross-linking. The cells were then incubated in DMEM without FCS at 37°C for the indicated duration. Purified fibronectin (10 g/ml), collagen type I (10 g/ml), and control bovine serum albumin (10 g/ml, Wako) were applied to 6-well plastic plates in Ca 2ϩ /Mg 2ϩfree PBS at 4°C overnight. Binding sites on plastic were subsequently blocked with Ca 2ϩ /Mg 2ϩ -free PBS, 2.5% bovine serum albumin for 2-3 h at 37°C to reduce nonspecific attachment of the cells. Subsequently plates were washed three times with PBS, and the cells were added to each well as described above and were incubated in DMEM without FCS at 37°C for the indicated duration. After the incubation, the plates were washed twice with PBS and treated with trypsin for 1 min at 37°C. DMEM containing 10% FCS was added to stop trypsinization. After harvesting from the wells, the obtained cells were washed with PBS and were settled in medium suitable for the following experiments.
Transfection of Plasmids-A human wild-type FAK expression plasmid (VSV-FAK), a FAT expression plasmid (VSV-FAT), and a FRNK expression plasmid (VSV-FRNK) were introduced into osteoblasts using a cationic liposome-mediated transfection method. Plasmids (2.5 g) dissolved in 100 l of serum-free medium (OPTI-MEM, Invitrogen) were mixed with 5 l of Lipofectin reagent (LipofectAMINE 2000 TM , Invitrogen) in the same volume of OPTI-MEM and incubated for 15 min at room temperature. The plasmids and liposome complex were added to osteoblasts plated in a 6-well culture dish, incubated for 3 h in OPTI-MEM, and then replaced with DMEM containing 10% FCS for 24 h. The expression of VSV-FAK, VSV-FAT, and VSV-FRNK in osteoblasts was confirmed by flow cytometric analysis using anti-VSV Ab after their transfection. High expression of the VSV was observed, and 50 -80% of the cells were transfected by intracellular flow cytometric analysis. A marked difference in the transfection efficiency among all three vectors was not observed in osteoblasts, and none of these vectors produced cytotoxic effects on osteoblasts as confirmed by trypan blue staining (data not shown).
Flow Microfluorometry-Staining and flow cytometric analysis of osteoblasts were conducted by standard procedures as described previously (18) using a FACScan (BD Biosciences). In brief, 2 ϫ 10 5 cells were incubated with negative control mAb thy1.2, integrin ␤ 1 mAb, ␤ 2 mAb, ␤ 3 mAb, ␣ 1 mAb, ␣ 2 mAb, ␣ 3 mAb, ␣ 4 mAb, ␣ 5 mAb, ␣ 6 mAb, ␣ v mAb, CD54 (ICAM-1) mAb, RANKL mAb, or CD106 (VCAM-1) mAb in FACS medium consisting of Hanks' balanced salt solution (Nissui, Tokyo, Japan), 0.5% human serum albumin (Yoshitomi, Osaka, Japan), and 0.2% NaN 3 (Sigma) for 30 min at 4°C. After washing the cells three times with FACS medium, they were further incubated with fluorescein isothiocyanate-conjugated goat anti-mouse IgG Ab, goat anti-rabbit IgG Ab, or rabbit anti-goat IgG Ab for 30 min at 4°C. The staining of cells with mAbs was detected using FACScan. Quantification of cell surface antigens on single cells was calculated using standard beads (QIFKIT, Dako Japan, Kyoto, Japan) as already described (30,31). The data were used for the construction of the calibration curve (mean fluorescence intensity (MFI)) against antibody binding capacity (ABC). The cell specimen was analyzed on the FACScan, and ABC was calculated by interpolation on the calibration curve. When the green fluorescence laser detector was set at the 450 level in the FACScan, ABC ϭ 414.45 ϫ exp(0.0092 ϫ MFI) (R 2 ϭ 0.9999). Subsequently specific ABC was obtained after corrections for background, apparent ABC of the negative control mAb thy1.2. Specific ABC corresponds to the mean number of accessible antigenic sites per cell, referred to as antigen density and expressed in sites/cell.

Coculture of Osteoblasts and Monocytes and Subsequent
Staining of Tartrate-resistant Acid Phosphatase-Osteoblasts were seeded onto 24well multiwell dishes (1 ϫ 10 5 cells/well) and cultured to subconfluence. Then osteoblasts, with or without transfection of several plasmids as mentioned above, were stimulated by cross-linking with specific mAb for 6 h. Purified peripheral blood CD14 ϩ monocytes from healthy donors (1 ϫ 10 6 cells/well) were added to osteoblasts, and they were cocultured for 9 days in DMEM containing 10% heat-inactivated FCS in the presence of 10 Ϫ7 M 1␣,25(OH) 2 D 3 . Osteoclasts were cytochemically stained for tartrate-resistant acid phosphatase (TRAP, Sigma) as described previously (32). The number of TRAP-positive multinuclear cells (MNCs) that contained more than three nuclei was counted by light microscopy.
Statistical Analysis-Data were expressed as mean Ϯ S.D. of the number of indicated patients. Differences from the control were examined for statistical significance by analysis of variance followed by posthoc Scheffe's F-test. A p value less than 0.05 denoted the presence of a statistically significant difference.

␤ 1 Integrin Was Highly Expressed on Primary Human
Osteoblastic Cells-Initially we assessed the expression of various cell surface functional molecules on primary human osteoblastic cells using FACScan. Fig. 1 shows the number of 12 representative molecules, including integrins, ICAM-1, and RANKL, on osteoblasts. Among the screened molecules, ␤ 1 integrin was highly expressed on osteoblasts. Among ␣ subunits of integrins, ␣ 2 , ␣ 3 , ␣ 4 , ␣ 5 , ␣ 6 , and ␣ v were expressed on osteoblasts. A histogram implied that the vast majority of osteoblasts expressed ␤ 1 . We therefore assumed that ␤ 1 integrin, which is consistently highly expressed on osteoblasts, might play a functional role in primary osteoblasts. ICAM-1 and RANKL were moderately expressed on osteoblasts. Estimation from histograms of multiple donors showed that one-third to twothirds of osteoblasts expressed ICAM-1 and RANKL without stimulation.
Cross-linking of ␤ 1 Integrin Up-regulated ICAM-1 and RANKL Expression on Osteoblasts-To characterize the function of ␤ 1 integrin on osteoblasts, we assayed the cell surface molecule expression by ␤ 1 cross-linking using a specific mAb and a second cross-linker Ab. Flow cytometry showed that expression of ICAM-1 and RANKL on the surface was markedly augmented by the ␤ 1 cross-linking on osteoblasts. As shown in Fig. 2, ICAM-1 and RANKL were moderately expressed on unstimulated osteoblasts. However, ␤ 1 cross-linking significantly up-regulated ICAM-1 and RANKL expression, whereas cross-linking of major histocompatibility complex class I using their specific mAbs and second cross-linker Ab had no effect. In contrast, cross-linking of ␤ 1 integrin on osteoblasts did not induce VCAM-1 expression (Fig. 2). The results were consistent in osteoblasts derived from four donors. Time course experiments showed that ICAM-1 and RANKL expression reached maximum levels within 3 h after ␤ 1 cross-linking (Fig.   2). These results suggest that ␤ 1 integrin appears to play a pivotal role in ICAM-1 and RANKL up-regulation on osteoblasts.

Engagement of ␤ 1 Integrin by Ligand Matrix Glycoproteins Augmented ICAM-1 and RANKL Expression on Osteoblasts-
Fibronectin and type I collagen are major ligands for cell surface ␤ 1 integrin. We next assessed the biological activities of fibronectin and type I collagen on osteoblasts. Expression of both ICAM-1 and RANKL were markedly induced by incubation of these cells on fibronectin-or collagen-coated plastic plates (Fig. 3). In contrast, no change was noted when these cells were incubated on bovine serum albumin-coated plates (Fig. 3). Furthermore ICAM-1 and RANKL expression induced by fibronectin and type I collagen was completely inhibited by pretreatment of cells with anti-␤ 1 mAb (Fig. 3). These data suggest that fibronectin and type I collagen are possible ligands involved in ␤ 1 integrin-induced ICAM-1 and RANKL expression on osteoblasts.
Involvement of Tyrosine Kinases in ␤ 1 Integrin-mediated Upregulation of ICAM-1 and RANKL on Osteoblasts-Next, to determine signaling pathways involved in ␤ 1 integrin-induced induction of ICAM-1 and RANKL expression on osteoblasts, we pretreated the cells with or without different concentrations of various inhibitors of intracytoplasmic signaling, and then ICAM-1 and RANKL expression by ␤ 1 cross-linking was determined by FACScan (Fig. 4). Pretreatment of cells with the tyrosine kinase inhibitors herbimycin A or genistein completely inhibited ␤ 1 -induced up-regulation of ICAM-1 and RANKL expression on osteoblasts. In contrast, H7 or staurosporine (protein kinase C inhibitors) and wortmannin (a PI 3-K inhibitor) did not influence ␤ 1 integrin-induced ICAM-1 and RANKL expression. These results suggest that ␤ 1 integrin-induced ICAM-1 and RANKL expression on osteoblasts is mediated mainly through tyrosine kinases, consistent with previous reports from our laboratories and those of others (14,33,34).
Involvement of FAK in ␤ 1 -mediated Signaling Inducing ICAM-1 and RANKL on Osteoblasts-FAK is known to be an important mediator of the integrin-mediated signaling by adhering to ECM proteins (28). ICAM-1 and RANKL expression was up-regulated by cross-linking of ␤ 1 integrins with a specific mAb on osteoblasts that did or did not express VSV-FAK, a wild-type FAK. However, on the cells expressing VSV-FAT or VSV-FRNK (a dominant negative truncation of FAK), such induction of ICAM-1 and RANKL by stimulation of ␤ 1 was completely reduced to the expression seen on the unstimulated cells (Fig. 5). These results indicate that FAK is involved in signaling via ␤ 1 stimulation, which leads to up-regulation of ICAM-1 and RANKL expression on osteoblasts.

␤ 1 -mediated Signaling Increased TRAP ϩ MNC Formation in
Osteoblasts from Peripheral Monocytes-In a coculture system using osteoblasts with peripheral monocytes, ␤ 1 -stimulated osteoblasts that did or did not express VSV-FAK significantly induced the formation of TRAP ϩ MNCs from the monocytes compared with untreated osteoblasts. However, on the cells expressing VSV-FAT or VSV-FRNK, such induction of TRAP ϩ MNCs by stimulation of ␤ 1 was markedly inhibited (Fig. 6). These results also indicate that FAK is involved in ␤ 1 stimulation, and this helps commit hematopoietic precursors toward osteoclast development. DISCUSSION Integrins are a superfamily of cell surface receptors involved in cell-cell and cell-matrix adhesion. Signals from matrices transduced by integrins play critical roles in regulating gene expression, tissue-specific differentiation, and survival of primary osteoblasts and fibroblasts (6,35,36). It has been reported that human osteoblasts express a diverse range of integrins, particularly of the ␤ 1 integrin, including ␣ 1 through ␣ 5 subunits. While our and others' studies show different patterns of integrin expression, this may reflect the heterogeneity of osteoblast-like populations and the different stages of osteoblasts from fetal or adult bone (37)(38)(39)(40). Although osteoblasts express ␣ v integrins, they differ from osteoclasts in that ␤ 1 integrins appear to have the major functional role, which has been underscored by in vivo data (41). ␤ 1 integrins are major adhesion receptors mediating interactions between osteoblasts and ECMs found in bone, such as collagen, fibronectin, osteopontin, thrombospondin, and vitronectin. The ␣ 2 ␤ 1 integrins bind to type I collagen, which is the dominant bone matrix protein, and several studies have reported that ␣ 2 ␤ 1 integrinmediated binding to ECM regulates osteoblastic differentiation (24,25). The ␣ 5 ␤ 1 integrin, the selective fibronectin receptor, is important in both the development and maintenance of bone. Globus et al. (42) have demonstrated that ␣ 5 ␤ 1 ligand binding is necessary for cell survival and that receptor blockade leads to cell apoptosis in mature osteoblasts. Thus, the accumulating evidence indicates that ␤ 1 integrin-mediated adhesion to bone matrix induces proliferation, differentiation, and bone matrix synthesis of osteoblasts and that ␤ 1 integrin-mediated signaling on osteoblasts could be involved in bone formation. However, based on the results of the present study, we propose a new concept that stimulation of the adhesion molecule ␤ 1 integrin plays a pivotal role in the regulation of ICAM-1 and RANKL expression and osteoclastogenesis. We deduced this from the following novel findings. 1) Engagement of ␤ 1 integrins on purified human osteoblasts by cross-linking using a specific antibody as well as ligand matrices markedly augmented surface ICAM-1 and RANKL expression. 2) Up-regulation of both ICAM-1 and RANKL by ␤ 1 integrin stimulation was mediated by the tyrosine kinase signaling pathway, especially involving FAK. 3) Engagement of ␤ 1 integrins on osteoblasts induced TRAP-positive MNC formation in the coculture system of osteoblasts and peripheral monocytes. 4) Up-regulation of TRAP ϩ MNC formation by ␤ 1 stimulation was completely abrogated by transfection of dominant negative truncations of FAK. Therefore, we further propose that ␤ 1 integrin-dependent adhesion to bone matrix proteins in osteoblasts can transduce signaling to induce ICAM-1 and RANKL expression through tyrosine kinase involving FAK-and RANKL-induced osteoclast formation.
As RANKL, a member of the tumor necrosis factor family expressed on the cell surface membrane of COS cells and osteoblasts/stromal cells, induces osteoclast formation from its precursor through cognate interaction between osteoblasts and osteoclast precursors, it is thereby required for RANKL-induced osteoclastogenesis (9 -11). Thus, higher affinity adhesion between osteoblasts and osteoclast precursors is emerging as a prerequisite for interaction of membrane-bound RANKL to be efficiently presented to its receptor, RANK. However, the binding affinity between tumor necrosis factor family molecules and tumor necrosis factor receptor family molecules, including CD40/CD40L, CD30/CD30L, Fas/FasL, and RANKL/RANK binding, is not sufficient to support static or firm cell adhesion (43). Furthermore we reported that anti-RANKL antibody did not inhibit the adhesion of osteoblasts to osteoclast precursors, whereas anti-LFA-1 antibody completely blocked the adhesion in a human cell culture system (12). Thus, based on our in vitro study, it can be assumed that up-regulation of ICAM-1 and RANKL expression on osteoblasts by ␤ 1 integrin-mediated signaling could affect cellular adhesion between osteoblasts and osteoclast precursors through the ICAM-1/LFA-1 and RANKL/ RANK pathways and lead to differentiation of osteoclast progenitors to osteoclasts in vivo.
During bone remodeling processes, adhesion-dependent interaction among osteoblasts and osteoclasts causes an imbalance in bone metabolism by favoring bone resorption through the expression of RANKL, ICAM-1, and other factors involved in cellular interaction. Although several studies have reported that ␤ 1 integrin-mediated adhesion to bone matrix induces proliferation, differentiation, and bone matrix synthesis of osteoblasts, our novel findings suggest that ␤ 1 integrin/FAKmediated signaling on osteoblasts could be involved in ICAM-1and RANKL-dependent osteoclast maturation. Thus, it can be assumed that ␤ 1 integrin-mediated signaling on osteoblasts could be involved in high turnover on bone metabolism through two paradoxical features of bone formation and bone resorption. However, it is as yet unclear how the same signaling pathway controls such diverse cellular events. After ligation of ␤ 1 integrins with surrounding ECMs, the integrins are found in focal adhesion plaques where various cytoskeletal proteins accumulate. Engagement of ␤ 1 integrins leads to initiation of intracellular signal transduction through accumulated cytoskeletal signaling kinases, resulting in the activation, differentiation, development, and mobility of various cell types (15,16). Several studies have established that among various cytoskeletal proteins, FAK, a cytoplasmic protein-tyrosine kinase that localizes focal adhesions, is an important mediator of integrin-mediated signaling and that the initial events triggered by the stimulation of ␤ 1 integrin are tyrosine phosphorylation and activation of FAK. In the present study, we observed that ␤ 1 integrin-mediated induction of ICAM-1 and RANKL expression was completely inhibited when the cells were pretreated with tyrosine kinase inhibitors. Furthermore ␤ 1 integrin-induced up-regulation of ICAM-1 and RANKL on osteoblasts expressing FRNK or FAT (dominant negative truncations of FAK) was completely inhibited. These findings suggest that ␤ 1 integrin-induced up-regulation of both ICAM-1 and RANKL expression was brought about by the signaling pathway of tyrosine kinases, specifically involving FAK. Phosphorylated FAK activates several transduction molecules including Src and Grb2, which may cause the activation of mitogen-activated protein kinase or PI 3-K via Ras (22). Recent evidence indicates that small guanine nucleotide-binding regulatory proteins (Gproteins) control signaling pathways critical for diverse cellular functions. Among several small G-proteins, Ras proteins are molecular switches that act as a "hub," which radiates multiple signaling pathways critical for diverse cellular functions, including Raf-1/mitogen-activated protein kinase and PI 3-K (30,44,45). We have reported the relevance of H-Ras and its downstream effectors to functions of osteoblasts and proposed that H-Ras signals, especially those followed by the Raf-1/mitogenactivated protein kinase pathway, but not PI 3-K, induce cell cycle arrest and subsequent apoptosis via Fas up-regulation and Bcl-2 down-regulation (46). Although further evidence is required, there is a possibility that such diverse cellular regulation by the ␤ 1 integrin-bone matrix interaction may be mediated by these G-protein signaling cascades and that FAK is FIG. 6. ␤ 1 -mediated signaling increased TRAP ؉ MNC formation in osteoblasts from peripheral monocytes. In a coculture system using osteoblasts with peripheral monocytes, ␤ 1 -stimulated osteoblasts transfected with or without control vectors encoding VSV-FAK, VSV-FAT, or VSV-FRNK were evaluated for the formation of TRAP ϩ MNCs from the monocytes. The data are representative of four similar experiments. Numbers Ϯ S.D. represent the number of TRAP ϩ MNCs in one well. *, p Ͻ 0.05 compared with controls. the immediate transducer of ␤ 1 integrin-mediated signaling. In conclusion, our results suggest a novel mechanism of ␤ 1 integrin-bone matrix cross-talk and a pivotal role in osteoclastogenesis. Further studies will be required to understand ␤ 1 integrin function in bone metabolism.