- Email: [email protected]
Calcitonin induces dephosphorylation of Pyk2 and phosphorylation of focal adhesion kinase in osteoclasts
Bone Vol. 31, No. 3 September 2002:359 –365
Calcitonin Induces Dephosphorylation of Pyk2 and Phosphorylation of Focal Adhesion Kinase in Osteoclasts Z. ZHANG,1,2 L. NEFF,1 A. L. M. BOTHWELL,2 R. BARON,1,3,4 and W. C. HORNE1,4 1
Departments of Cell Biology and Orthopaedics, 2Section of Immunobiology, 3Yale Cancer Center, and 4Yale Core Center for Musculoskeletal Disorders, Yale University School of Medicine, New Haven, CT, USA
Introduction Calcitonin induces the association and tyrosine phosphorylation of focal adhesion kinase (FAK), paxillin, and HEF1 in HEK-293 cells that overexpress the calcitonin receptor (C1a-HEK), but the hormone’s effect on these adhesion-related proteins in osteoclasts is not known. We therefore studied the effect of calcitonin on the tyrosine phosphorylation and subcellular distribution of paxillin, HEF1, FAK, and Pyk2, a FAK-related tyrosine kinase, in osteoclasts. Osteoclasts expressed both Pyk2 and FAK, with Pyk2 much more highly expressed. The two tyrosine kinases and paxillin were prominently associated with small punctate structures that were most densely clustered in the region of the peripheral F-actin-rich ring. Some of the punctate structures stained either for Pyk2 alone or FAK alone. Treatment with calcitonin disrupted the actin ring and induced the loss of the peripheral staining of paxillin, Pyk2, and FAK. In calcitonin-treated osteoclast-like cells, the tyrosine phosphorylation of paxillin and FAK increased, whereas the tyrosine phosphorylation of Pyk2 decreased. Calcitonin also induced increased phosphorylation of Erk1 and Erk2 in osteoclasts, as it did in the C1a-HEK cells. The unexpected dephosphorylation of Pyk2 correlated with decreased phosphorylation of Tyr402, the autophosphorylation site of Pyk2. The calcitonin-induced dephosphorylation of Pyk2 was not observed in C1a-HEK cells transfected with Pyk2, suggesting that the reduced phosphorylation seen in osteoclasts may be specific to these cells. Treatment of osteoclast-like cells with 12-phorbol 13-myristate acetate increased the tyrosine phosphorylation of both Pyk2 and FAK, and calphostin C, an inhibitor of protein kinase C, blocked calcitonin-stimulated FAK phosphorylation. Increasing intracellular calcium with ionomycin caused a decrease in the tyrosine phosphorylation of Pyk2 and the loss of the actin ring in a manner similar to the effect of calcitonin. Ionomycin had no effect on FAK tyrosine phosphorylation. Calcitonin (CT)-induced changes in Pyk2, FAK, and Erk1/2 phosphorylation were independent of c-Src. (Bone 31:359 –365; 2002) © 2002 by Elsevier Science Inc. All rights reserved.
Osteoclasts (OCs) attach to extracellular matrix or the bone surface via specialized attachment structures called podosomes, which form a prominent F-actin-rich ring that is thought to correspond to the sealing zone of resorbing OCs.3 The dynamic assembly and disassembly of podosomes are important for OC function, as seen in Src⫺/⫺ OCs, which exhibit decreased cell motility and adhesion structures that are abnormal in terms of both intracellular location and turnover.25 These defects may contribute to the reduced resorbing activity of Src⫺/⫺ OCs. We and others have suggested that Pyk2, a member of the focal adhesion kinase (FAK) family that is highly expressed in OCs, plays a key role in the Src-dependent regulation of the adhesion and motility of OCs. Pyk2 localizes in the podosomes of OCs and binds to c-Src upon activation of ␣v3 integrin, the main integrin expressed in OCs.8,25 Pyk2-deficient OCs and OCs expressing antisense Pyk2 exhibit abnormal adhesion structures and decreased bone resorption in vitro and in vivo, whereas Pyk2/Src double knockout mice develop more severe osteopetrosis and deficient bone remodeling than mice with only the Pyk2 or Src gene deletion.9,31 Pyk2 and FAK have similar domain structures, with a central kinase (catalytic) domain, multiple phosphorylated tyrosines, two proline-rich regions in the C-terminus, and a focal adhesion targeting domain. For full kinase activity, both proteins require the autophosphorylation of homologous sites (Pyk2 Tyr402, FAK Tyr397) that serve as binding sites for Src family kinases. Although the two proteins are 61% identical overall, the N- and C-terminal regions outside of the kinase domain are only 42% and 36% identical, respectively,18 suggesting that there are likely to be both similarities and differences in the signaling functions of the two proteins. Consistent with this, Pyk2 compensates for FAK’s role in fibronectin-induced cell spreading, Grb2-Shc association, and Src-dependent Erk2 activation in FAK⫺/⫺ cells,30 and both proteins can be activated by integrin ligation and activation of G-protein-coupled receptors in a variety of cells.1,12,14,24 However, Pyk2 and FAK activation have been reported to differ in activated platelets in terms of time course and dependence on integrin engagement,22 and the two proteins respond differently to the treatment of rat aortic smooth muscle cells, transfected Chinese hamster ovary cells, and chicken embryo cells with various soluble factors.26,40 Calcitonin (CT) inhibits OC motility, induces OC retraction, and disrupts the actin-ring structure of OCs,17,33,34,38 suggesting that it might regulate the activity of proteins such as FAK and Pyk2 that are involved in the regulation of osteoclast adhesion.
Key Words: calcitonin, osteoclast, Pyk2, focal adhesion kinase, paxillin.
Dr. Zhang is the recipient of a fellowship from the Arthritis Foundation. Address for correspondence and reprints: Dr. William C. Horne, Department of Orthopaedics, Yale University School of Medicine, P.O. Box 208044, New Haven CT 06520-8044. E-mail: [email protected] © 2002 by Elsevier Science Inc. All rights reserved.
359
8756-3282/02/$22.00 PII S8756-3282(02)00834-7
360
Z. Zhang et al. Calcitonin regulates Pyk2 and FAK phosphorylation in osteoclasts
Our laboratory previously reported that CT stimulates tyrosine phosphorylation and association of FAK, paxillin, and HEF1 in HEK-293 cells that overexpress the C1a isoform of the CT receptor (C1a-HEK).39 Moreover, induction of these responses involves activation of protein kinase C (PKC) and increased cytosolic calcium, two signaling effectors known to activate Pyk2 and increase Pyk2 phosphorylation.4,18,21,25 We therefore examined the effect of CT on Pyk2, FAK, paxillin, and HEF1 in OCs. Paxillin, Pyk2, and FAK were all relatively enriched in the region of the F-actin-rich podosome ring in authentic OCs and OC-like cells (OCLs) derived from rabbit bone marrow cultures, and treatment with CT induced their dispersion. The tyrosine phosphorylation of paxillin and FAK increased transiently following CT treatment, whereas Pyk2 phosphorylation decreased transiently, suggesting that Pyk2 and FAK, the two related kinases involved in cellular adhesion structures, play different roles downstream of CT receptor activation on OCs. Materials and Methods Reagents and Antibodies Salmon calcitonin was purchased from Peninsula Laboratories, Inc. (Belmont, CA). Calphostin C and ionomycin were from Calbiochem-Novabiochem International (San Diego, CA). Phorbol 12-myristate 13-acetate (PMA), prostaglandin E2, and fetal bovine serum (FBS) were from Sigma. The anti-FAK polyclonal antibody (Cat. No. 06-543) used for immunoprecipitation and immunostaining and the anti-phosphotyrosine antibody (Clone 4G10) were from Upstate Biotechnology, Inc. (Lake Placid, NY). The anti-FAK antibody used for western blotting, the monoclonal anti-paxillin antibody and the anti-Pyk2 antibody were from Transduction Laboratories. The antibody against phospho-Pyk2 was from Biosource International (Camarillo, CA). Anti-HEF1 antibody was from Santa Cruz Biotechnology (Santa Cruz, CA). Antibodies against phosphorylated Erk1/2 and total Erk1/2 were purchased from New England Biolabs (Beverly, MA). Stock solutions of calphostin C, ionomycin, and PMA were prepared in dimethylsulfoxide (DMSO). Isolation of Osteoclasts and Osteoclast-like Cells Rabbit authentic osteoclasts and osteoclast-like cells. Bone cells were isolated and cultured as described elsewhere,7 with modifications. Briefly, the long bones and scapulae of neonatal New Zealand white rabbits (60 –90 g) were dissected free of soft tissue and minced in ␣-modified minimal essential medium with 0.55 g/L sodium bicarbonate and 10 mmol/L HEPES (pH 7.10). The bone fragments were then allowed to settle, the cell suspension was collected, and the cells were pelleted by centrifugation at 1500g for 5 min at 4°C. To isolate mature OCs, the cell pellets were resuspended in medium supplemented with 10% FBS, aliquoted (100 L) onto coverslips in six well dishes, and cultured for 1 h at 37°C in humidified air with 5% CO2. Then, 2 mL of medium supplemented with 5% FBS was added to each well. After an additional 3 h incubation, the cultures were washed with fresh medium to remove nonadherent cells, and the remaining cells were cultured for another 2–3 h. The cells were then changed to fresh medium with 1% FBS and cultured for another 0.5 h before stimulation. To produce rabbit OCLs, the resuspended cells were cultured in medium supplemented with 10% FBS and 10⫺8 mol/L 1,25dihydroxyvitamin D3 (Roussel-UCLAF, Romainville, France) for 9 or 10 days. The medium was changed every other day. Cells were cultured in medium with 1% FBS for 18 h before stimulation.
Bone Vol. 31, No. 3 September 2002:359 –365
Murine osteoclast-like cells. Murine OCLs were produced by coculturing calvarial osteoblasts with spleen and bone marrow cells, as described elsewhere.25 Transient Transfection C1a-HEK cells were cultured and transiently transfected as described elsewhere.39 Immunoprecipitation and Immunoblotting Prior to treatment, rabbit and murine cocultures were incubated for 18 h in medium containing 1% FBS. The serum-starved OCLs were stimulated with CT or other agents in the presence of the cocultured mesenchymal cells at 37°C. The dishes were then transferred onto ice and the OCLs purified at 0°C by peeling away the layer of less adherent, mainly stromal cells. The purified OCLs were processed for immunoprecipitation and western blotting, as previously described,39 with slight modifications. The purified OCLs and C1a-HEK cells were lysed in 1% IGEPAL CA-630, 150 mmol/L NaCl, 50 mmol/L Tris-HCl (pH 8.0), 5 mmol/L ethylene-diamine tetraacetic acid (EDTA), 1 mmol/L phenylmethylsulfonylfluoride, 5 mmol/L iodoacetamide, 10 mmol/L NaF, 0.4 mmol/L Na2VO4, and 10 g/mL aprotinin. The lysates were incubated with specific antibodies and protein G-Sepharose, and the immune complexes were washed with PBS containing 1 mmol/L phenylmethylsulfonylfluoride, 5 mmol/L iodoacetamide, 10 mmol/L NaF, 0.4 mmol/L Na2VO4, and 10 g/mL aprotinin. The samples were then processed for immunoblotting as described elsewhere.39 Immunostaining and Confocal Microscopy Authentic OCs on coverslips were fixed in PBS containing 3.7% formaldehyde for 10 min at room temperature, and washed in PBS. Coverslips for actin labeling were extracted in ice-cold acetone for 3–5 min, and returned to PBS. All other coverslips were permeabilized in 0.05% saponin for 30 min. The coverslips were blocked in 5% normal goat serum for 30 min, then incubated in the appropriate primary antibody for 2 h, washed in PBS, incubated for 1 h in the appropriate fluorescein- or rhodamine-conjugated secondary antibody, and washed. Those coverslips used for actin labeling were incubated in a 1:40 dilution (in PBS) of rhodamine phalloidin (Molecular Probes) for 20 min and washed. All coverslips were mounted in FluorSave. Cells were examined using a scanning laser confocal imaging system (MRC-600; Bio-Rad Laboratories). Images were recorded, composite images were compiled, and image enhancements performed using Adobe Photoshop 4.0. Results Calcitonin Modulates Tyrosine Phosphorylation of Podosome-associated Proteins in Osteoclasts We initially sought to determine if CT induced changes in the tyrosine phosphorylation of FAK family members, paxillin and HEF1, in OCLs as it does in the HEK-293 cells stably transfected with the CT receptor (CTR). Unlike the HEK cells, the predominant FAK family member in OCs is Pyk2,8,25 whereas the expression of FAK in OCs is more controversial, with some investigators reporting its presence and others failing to detect it.8,35 We therefore first examined the expression of Pyk2 and FAK in OCs by western blot (Figure 1A) and confocal immunofluorescence microscopy (Figure 2A,B). Both FAK and Pyk2 were detected in lysates of rabbit (Figure 1 and Figure 3) and
Bone Vol. 31, No. 3 September 2002:359 –365
Z. Zhang et al. Calcitonin regulates Pyk2 and FAK phosphorylation in osteoclasts
361
Double staining of FAK and Pyk2 (Figure 2B) showed that, although the locations of the two proteins were similar in the untreated OCs, at least as many of the individual punctate structures stained only for Pyk2 or FAK as stained for both. Treatment with 1 nmol/L CT induced the retraction of OCs, which could be detected by 3 min, and was maintained until at least 30 min (Figure 2C). CT also disrupted the actin ring, as previously reported,34 and dispersed the peripherally enriched Pyk2, FAK, and paxillin (Figure 2A, lower images), which became diffusely distributed throughout the OCs. To examine the effect of CT treatment on the tyrosine phosphorylation of Pyk2, FAK, paxillin, and HEF1, OCLs were treated with 1 nmol/L CT for time durations as long as 30 min, then washed and lysed. The individual proteins were immunoprecipitated and analyzed for phosphotyrosine content by western blot. In contrast to what we previously found in the C1a-HEK cells, OCLs expressed relatively low amounts of HEF1 and there was little change in the tyrosine phosphorylation after CT treatment (Figure 1B). The tyrosine phosphorylation of both paxillin (Figure 1C) and FAK (Figure 1D) increased transiently following CT addition, with the phosphotyrosine content of both proteins peaking at 1–3 min and returning to near-basal levels by about 10 min. Unexpectedly, CT induced a decrease in the tyrosine phosphorylation of Pyk2, which reached a minimum by 3–10 min and returned to the basal level by 30 min (Figure 1E). This decrease in Pyk2 tyrosine phosphorylation was correlated with decreased phosphorylation of Tyr402 (Figure 1F), the autophosphorylation site of Pyk2.
Calcium, But Not PKC, Induces Decreased Tyrosine Phosphorylation of Pyk2 in Osteoclasts Figure 1. Calcitonin regulates the tyrosine phosphorylation of paxillin, FAK, and Pyk2 in osteoclasts. (A) To examine the expression of FAK in OCs, rabbit OCLs were generated and lysed, and the total cell lysates processed for western blotting with anti-Pyk2 (top left) and anti-FAK (top right) antibodies, as described in Materials and Methods. The membranes were then stripped and reblotted with anti-actin antibody to demonstrate equal loading (bottom). (B)–(F) Rabbit OCLs were cultured overnight in low serum, then treated with 1 nmol/L CT for the indicated times and purified and lysed, as described in Materials and Methods. HEF1 (B), paxillin (C), FAK (D), and Pyk2 (E) were immunoprecipitated from the lysates and the immune complexes were western blotted with anti-phosphotyrosine antibody [(B–E) top]. The membranes were then stripped and reblotted with antibody to the appropriate antigen (bottom). The amounts of phosphotyrosine and protein in the blots were quantified by densitometry, and the phosphotyrosine was corrected for the sample load based on the amount of protein and expressed as fold increase or decrease relative to the control lane (the numbers below each pair of blots). (F) Total cell lysates were immunoblotted with anti-Pyk2 phospho-Tyr402 antibody (top). The membranes were then stripped and reblotted with anti-Pyk2 (bottom). In (B) and (F), the OCLs were treated with CT for 3 min. Each blot is representative of three or four replicate experiments.
murine (Figure 4) OCLs by western blot, although the level of Pyk2 was much higher, consistent with other reports. In untreated authentic OCs, Pyk2-positive punctate structures were densely present in the peripheral F-actin-rich ring and similar structures were also more sparsely distributed throughout the rest of the cell (Figure 2A, upper right image). The pattern of the FAK staining (Figure 2A, upper center image) was similar to that of Pyk2 (and to that of paxillin, Figure 2A, upper left image), but the intensity of FAK staining was much weaker than that of Pyk2, consistent with the relative levels seen in the western blots.
The CT-induced decrease in Pyk2 tyrosine phosphorylation in OCLs was unexpected, because we and others have shown that signaling from the CTR activates PKC and increases cytosolic calcium,6,19,29,33,36,38 and it has been generally observed that stimuli that activate PKC or increase cytosolic calcium, including the activation of some G-protein-coupled receptors, stimulate Pyk2 phosphorylation.12,18 We therefore examined the roles of PKC and calcium in the regulation of Pyk2 phosphorylation in OCLs and C1a-HEK cells. In the OCLs (Figure 3A), PMA caused the tyrosine phosphorylation of both Pyk2 (left) and FAK (middle) to increase moderately, in contrast to the effect of CT on Pyk2 phosphorylation. Ionomycin, on the other hand, induced a decrease in Pyk2 tyrosine phosphorylation (left), similar to the effect of CT and in contrast to what has been reported in the literature for other cell types. Ionomycin did not change FAK phosphorylation (right). Thus, the CT-induced decrease in Pyk2 phosphorylation in OCLs appears to be mediated by increased cytosolic calcium, suggesting that calcium activates different signaling effectors in OCs than in many other cells. C1a-HEK cells were transiently transfected with Pyk2 cDNA, because Pyk2 was not detectable by western blot (data not shown). As predicted from the published reports and our earlier findings, the tyrosine phosphorylation of the transiently expressed Pyk2 was strongly induced in the C1a-HEK cells by CT, PMA, and ionomycin (Figure 3B). We also examined the effect of ionomycin on the actin ring. Treatment of authentic mature osteoclasts with 1 mol/L ionomycin for 5 min caused a loss of the actin ring, similar to the effect of CT (Figure 3C).
362
Z. Zhang et al. Calcitonin regulates Pyk2 and FAK phosphorylation in osteoclasts
Bone Vol. 31, No. 3 September 2002:359 –365
Figure 2. Effect of calcitonin on osteoclast morphology and the distribution of paxillin, Pyk2, and FAK in osteoclasts. Authentic rabbit OCs were isolated and plated on coverslips as described in Materials and Methods. (A) Cells treated with 1 nmol/L CT for 3 min (lower images), or not (upper images), were fixed and stained for F-actin (not shown) and either paxillin, FAK, or Pyk2, then examined by confocal immunofluorescence microscopy. The optical slice in these and the other images was taken at the level of cell attachment to the substrate, in order to visualize attachment structures. In untreated OCs, all three proteins were enriched in the peripheral actin-rich ring. Treatment with 1 nmol/L CT for 3 min disrupted the peripheral actin ring (not shown) and induced the redistribution of all three proteins to a more diffuse pattern throughout the cell interior. (B) Double labeling of untreated OCs with monoclonal anti-FAK (red) and polyclonal anti-Pyk2 (green). In an enlarged portion of the merged image (lower panel) with the red channel enhanced to increase FAK intensity, numerous punctate structures that stain only for FAK can be seen (arrowheads), in addition to similar structures that stain only for Pyk2 (green) or both proteins (yellow). (C) The CT-induced cell retraction and loss of the actin-rich peripheral podosome ring in authentic rabbit OCs. Cells were treated with 1 nmol/L CT for the indicated times, fixed, and labeled with rhodamine-conjugated phalloidin to visualize F-actin distribution.
Calcitonin-induced Changes in Pyk2, FAK, and Erk1/2 Are Independent of c-Src c-Src is required for normal OC function,32 in part because of the role that it plays in modulating OC attachment and motility.25 c-Src also plays a role in modulating FAK and Pyk2 activity.14 We therefore examined the effect of deleting c-Src on the CT-induced changes in the tyrosine phosphorylation of FAK and Pyk2 in OCs (Figure 4). For these experiments, OCLs were prepared by coculturing osteoblastic cells derived from wild-type mouse calvariae with bone marrow and spleen cells isolated from Src⫺/⫺ mice or their normal littermates. OCLs were treated with CT, lysed, and the phosphotyrosine content of FAK and Pyk2 analyzed. As in the rabbit OCLs, treatment of wild-type murine OCLs with CT decreased Pyk2 phosphorylation (Figure 4, top panel), and increased FAK phosphorylation (middle panel). Similar results were obtained with the Src⫺/⫺ OCLs, indicating CT-induced dephosphorylation of Pyk2 and phosphorylation of FAK do not require c-Src. Erk1/2, the MAP kinases that are involved in OC survival,20 have been reported to be activated downstream of FAK and Pyk2 following integrin ligation.5,27 Our laboratory showed previously that CT stimulates Erk1/2 phosphorylation in C1a-HEK cells.6 We therefore examined the effect of CT treatment on Erk1/2 phosphorylation in the Src⫹/? and Src⫺/⫺ OCLs (Figure 4, lower panel). CT induced the phosphorylation of Erk1/2, and the response was independent of the presence of Src.
Discussion CT is a potent inhibitor of osteoclastic bone resorption, although much remains to be learned about its exact mode of action. Treatment with CT induces retraction and inhibits mobility of isolated OCs,33,38 responses that are considered to contribute to the hormone’s inhibition of bone resorption. The loss of motility following the exposure of OCs to CT suggests that CT might regulate the function of the podosomes, which mediate the attachment of the OC to the bone matrix during locomotion and bone resorption. Pyk2 and FAK are two related tyrosine kinases that are known to play critical roles in integrin-mediated cell attachment in a variety of cells (reviewed in references 2 and 16). Pyk2 is strongly expressed in OCs8 and appears to play a key role in integrin-dependent signaling in these cells.8,9,25 OCs from Pyk2 knockout mice exhibit reduced bone resorbing activity and abnormally distributed podosomes in vitro and the mice exhibit a mild form of osteopetrosis, indicating that in vivo OC function is also impaired.31 In contrast to Pyk2, there has been little study of FAK in OCs. Although some investigators have failed to detect FAK in OCs,8 our confocal immunofluorescence data clearly show that FAK is expressed in this cell, although at much lower levels than the mesenchymal cells that were also present in these preparations. Pyk2, FAK, and paxillin, which binds to both FAK and Pyk2, had similar patterns of localization in isolated OCs plated on glass. The two kinases were associated with podosome-like
Bone Vol. 31, No. 3 September 2002:359 –365
Z. Zhang et al. Calcitonin regulates Pyk2 and FAK phosphorylation in osteoclasts
363
Figure 3. Increased calcium mimics the effect of calcitonin on Pyk2 tyrosine phosphorylation in osteoclasts and C1a-HEK cells. (A) (Left panels) Serum-starved rabbit OCLs were treated with vehicle, CT (1 nmol/L for 3 min), PMA (100 nmol/L for 10 min), or ionomycin (100 nmol/L for 3 min). Pyk2 was immunoprecipitated, and the immune complexes were immunoblotted with anti-phosphotyrosine antibody (top). The membrane was then stripped and reblotted with anti-Pyk2 antibody (bottom). (Center and right panels) Serum-starved rabbit OCLs were treated with vehicle, CT (1 nmol/L for 3 min), PMA (100 nmol/L for 10 min), calphostin C (CC; 1 mol/L for 30 min) prior to CT (1 nmol/L for 3 min) or ionomycin (100 nmol/L for 3 min) and then lysed. FAK was immunoprecipitated, and the immune complexes were blotted with anti-phosphotyrosine antibody (top). The membrane was then stripped and reprobed with anti-FAK antibody (bottom). Phosphotyrosine was quantified as described for Figure 1. The blots are representative of three or four replicate experiments. (B) C1a-HEK cells were transiently transfected with Pyk2 cDNA. After serum starvation for 18 h, the cells were treated with vehicle, CT (1 nmol/L for 3 min), PMA (100 nmol/L for 10 min), or ionomycin (100 nmol/L for 3 min), then lysed. Pyk2 was immunoprecipitated, and the immune complexes were blotted with anti-phosphotyrosine antibody. The membrane was then stripped and reblotted with anti-Pyk2 antibody. The blots are representative of three replicate experiments. (C) Authentic rabbit OCs were isolated and plated on coverslips as described in Materials and Methods. Cells were treated with 1 mol/L ionomycin or vehicle as indicated for 5 min, fixed, and stained for F-actin, then examined by confocal immunofluorescence microscopy as described for Figure 2. Ionomycin induced the disappearance of the actin ring in a manner similar to the effect of CT treatment.
punctate structures that occur most densely in a peripheral ring, but are also seen throughout the cell to a lesser extent. It may be significant, given the different effects of CT on the tyrosine
Figure 4. Calcitonin-induced changes in Pyk2, FAK, and Erk1/2 phosphorylation in osteoclasts are independent of c-Src. Murine Src⫹/? and Src⫺/⫺ OCLs were serum starved for 18 h, then treated with 1 nmol/L CT for 3 min, purified, and lysed. Pyk2 (upper panels) and FAK (middle panels) were immunoprecipitated, and the immune complexes were blotted with antiphosphotyrosine antibody. The membranes were then stripped and reblotted with anti-Pyk2 antibody or anti-FAK antibody, respectively. The total cell lysates were immunoblotted with anti-phospho-Erk1/2 antibody and antiactin antibody as a loading control (bottom panels).
phosphorylation of the two kinases, that most of the podosomes stain for either Pyk2 or FAK, not both. CT induced the loss of the prominent peripheral staining of Pyk2, FAK, and paxillin in parallel with the loss of the actin ring. CT induced the transient dephosphorylation of Pyk2, which correlated with the decreased phosphorylation of Tyr402, the autophosphorylation site of Pyk2. Phosphorylation of Tyr402 is required for full Pyk2 kinase activity, suggesting that CT might inhibit osteoclastic bone resorption in part by inhibiting Pyk2 activity. The inhibition of Pyk2 would be expected in turn to prevent the recruitment of Src and Cbl to the Pyk2·Src·Cbl complex that this laboratory has shown to be formed as a consequence of the activation of the vitronectin receptor.25 This possibility is perhaps consistent with the fact that osteoclasts that lack either Src or Cbl have reduced motility,25 similar to CTtreated OCs. In contrast to the decreased phosphorylation of Pyk2, CT treatment increased the tyrosine phosphorylation of FAK. Although the few contaminating stromal cells do contribute to the FAK content in the cell lysates, the CT-induced changes are occurring in the OCs because these are the only cells in the coculture preparations that express the CT receptor. The time courses of CT-induced changes in Pyk2 and FAK phosphorylation were also different, with FAK phosphorylation occurring earlier and reverting to the basal level in a shorter time than Pyk2 dephosphorylation. The decrease in Pyk2 phosphorylation following the activation of the CT receptor in OCLs was unexpected. Treatment of OCs with CT activates PKC and increases cytosolic calci-
364
Z. Zhang et al. Calcitonin regulates Pyk2 and FAK phosphorylation in osteoclasts
um.19,29,33,36,38 In other cell types, PKC activation or increased cytosolic calcium induce Pyk2 activation and tyrosine phosphorylation.12,18 Indeed, the phosphotyrosine content of transiently expressed Pyk2 in the CT receptor-expressing C1a-HEK cells increased in response to CT, PMA, or ionomycin. In the OCs, activating PKC with PMA stimulated the tyrosine phosphorylation of both FAK and Pyk2, whereas the PKC inhibitor, calphostin C, prevented the CT-induced increase in FAK phosphorylation, indicating that PKC mediates the CT-induced FAK phosphorylation but may not contribute to the changes in Pyk2 phosphorylation. In contrast, increasing cytosolic calcium with ionomycin decreased the phosphorylation of Pyk2, but had no effect on FAK phosphorylation. It thus appears that the regulation of Pyk2 function in OCs differs significantly from the regulatory mechanisms in many other cells, with increased cytosolic calcium possibly inducing some activity in OCs that dephosphorylates Pyk2 and thereby antagonizes PKC-induced Pyk2 phosphorylation. In this context, it is of interest to note that the phosphotyrosine levels of FAK and paxillin also decrease at about the same time as the Pyk2 dephosphorylation is occurring, and that this behavior contrasts with the sustained tyrosine phosphorylation of FAK and paxillin in CT-treated C1a-HEK cells.39 There are previous reports of a correlation between increased cytosolic calcium and protein tyrosine dephosphorylation.10,11,15,23 The identification of a putative calcium-dependent tyrosine phosphatase in OCs will be the subject of future research. Another apparent difference between the signaling mechanisms of OCs and HEK cells is the role of Src in coupling the CT receptor to FAK. Unlike the HEK cells, where CT-induced FAK phosphorylation is enhanced by overexpressing wild-type Src and inhibited by dominant-negative Src, the CT-induced FAK phosphorylation in OCs was observed in the Src⫺/⫺ OCLs. The coupling of the CT receptor to Erk1/2 was also independent of Src in the OCLs. As in C1a-HEK cells,6 treatment of OCLs with CT stimulated Erk1/2 phosphorylation. Miyazaki and colleagues have reported that inhibiting Erk activity induces OC apoptosis, whereas activation of Erk supports OC survival.20 Treatment with CT protects OCs from apoptosis and promotes OC survival,28 and this effect may be mediated, at least in part, by the enhanced Erk1/2 phosphorylation that we observed. Because FAK has been reported to promote adhesion-dependent cell survival in epithelial and endothelial cells,13 and overexpression of Pyk2 induces apoptosis in other cell types,37 the decreased Pyk2 phosphorylation and increased FAK phosphorylation may also be related to CT’s promotion of OC survival. The fact that CT, which reduces OC motility, induces a reduction in the level of Pyk2 tyrosine phosphorylation provides further evidence that the phosphorylation of Pyk2 Y402 and the consequent binding of Src may be critical for podosome function and normal OC motility. Src is not required for the dephosphorylation of Pyk2. In fact, the dephosphorylation is more nearly complete in the Src⫺ OCLs, consistent with the binding of Src to Pyk2402 reported earlier,25 which would tend to protect phosphotyrosine402 from being hydrolyzed by phosphatases. According to our hypothesized role of the Pyk2·Src·Cbl complex in OC motility, dephosphorylation of Pyk2Y402 would be expected to prevent the formation of the complex and therefore inhibit OC motility in much the same way as deleting either Src or Pyk2. It is not clear what the function of FAK might be in the podosome, but whatever the function it is clearly different from the function of Pyk2, because the presence of the two kinases in individual podosomes appears to be largely mutually exclusive and CT induces an increase in FAK phosphotyrosine level, not a decrease. Perhaps FAK plays a role in the formation of podo-
Bone Vol. 31, No. 3 September 2002:359 –365
somes or stabilizes podosomes once they are formed, whereas Pyk2 promotes the relatively rapid disassembly of existing podosomes by recruiting Src and Cbl, as described elsewhere.25 In conclusion, we have shown that OCs express both Pyk2 and FAK and that the two adhesion kinases are associated with different populations of similar actin-rich punctate structures, probably podosomes. Treatment of the OCs with CT disrupted the actin ring and simultaneously induced the relocation of the Pyk2, FAK, and paxillin away from the OC periphery. CT also induced increased FAK and paxillin phosphorylation and decreased Pyk2 phosphorylation in OCLs. The change in Pyk2 phosphorylation was mimicked by increasing free cytosolic calcium, possibly implicating a calcium-dependent protein tyrosine phosphatase. Our results offer new insights into the signaling mechanisms by which CT affects osteoclast function.
Acknowledgments: We are grateful to our colleagues, Emilia DiDomenico and Karen Ford, for their technical support, and to Dr. Adam Houghton and Dr. Archana Sanjay for their helpful comments. This work is supported by National Institutes of Health Grants DE-04724 and AR-42927 (to R.B.) and by Yale Core Center for Musculoskeletal Disorders Grant AR-46032 (to W.C.H.).
References 1. Astier, A., Manie, S. N., Avraham, H., Hirai, H., Law, S. F., Zhang, Y., Golemis, E. A., Fu, Y., Druker, B. J., Haghayeghi, N., Freedman, A. S., and Avraham, S. The related adhesion focal tyrosine kinase differentially phosphorylates p130Cas and the Cas-like protein, p105HEF1. J Biol Chem 272: 19719 –19724; 1997. 2. Avraham, H., Park, S.-Y., Schinkmann, K., and Avraham, S. RAFTK/Pyk2mediated cellular signalling. Cell Signal 12:123–133; 2000. 3. Baron, R. Molecular mechanisms of bone resorption.An update. Acta Orthop Scand 266:66 –70; 1995 (Suppl.). 4. Brinson, A. E., Harding, T., Diliberto, P. A., He, Y., Li, X., Hunter, D., Herman, B., Earp, H. S., and Graves, L. M. Regulation of a calcium-dependent tyrosine kinase in vascular smooth muscle cells by angiotensin II and plateletderived growth factor. Dependence on calcium and the actin cytoskeleton. J Biol Chem 273:1711–1718; 1998. 5. Chen, Q., Kinch, M. S., Lin, T. H., Burridge, K., and Juliano, R. L. Integrinmediated cell adhesion activates mitogen-activated protein kinases. J Biol Chem 269:26602–26605; 1994. 6. Chen, Y., Shyu, J.-F., Santhanagopal, A., Inoue, D., David, J. -P., Dixon, S. J., Horne, W. C., and Baron, R. The calcitonin receptor stimulates Shc tyrosine phosphorylation and Erk1/2 activation. Involvement of GI, protein kinase C, and calcium. J Biol Chem 273:19809 –19816; 1998. 7. David, J.-P., Neff, L., Chen, Y., Rincon, M., Horne, W. C., and Baron, R. A new method to isolate large numbers of rabbit osteoclasts and osteoclast-like cells: Application to the characterization of serum response element binding proteins during osteoclast differentiation. J Bone Miner Res 13:1730 –1738; 1998. 8. Duong, L. T., Lakkakorpi, P. T., Nakamura, I., Machwate, M., Nagy, R. M., and Rodan, G. A. PYK2 in osteoclasts is an adhesion kinase, localized in the sealing zone, activated by ligation of ␣v3 integrin, and phosphorylated by Src kinase. J Clin Invest 102:881–892; 1998. 9. Duong, L. T., Nakamura, I., Lakkakorpi, P. T., Lipfert, L., Bett, A. J., and Rodan, G. A. Inhibition of osteoclast function by adenovirus expressing antisense protein-tyrosine kinase 2. J Biol Chem 276:7484 –7492; 2001. 10. Errasfa, M. and Stern, A. Melittin inhibits epidermal growth factor-induced protein tyrosine phosphorylation: Comparison with phorbol myristate acetate and calcium ionophore A23187. Biochim Biophys Acta 1222:471–476; 1994. 11. Falet, H., Pain, S., and Rendu, F. Tyrosine unphosphorylated platelet SHP-1 is a substrate for calpain. Biochem Biophys Res Commun 252:51–55; 1998. 12. Felsch, J. S., Cachero, T. G., and Peralta, E. G. Activation of protein tyrosine kinase PYK2 by the m1 muscarinic acetylcholine receptor. Proc Natl Acad Sci USA 95:5051–5056; 1998. 13. Frisch, S. M., Vuori, K., Ruoslahti, E., and Chan-Hui, P. Y. Control of adhesion-dependent cell survival by focal adhesion kinase. J Cell Biol 134: 793–799; 1996.
Bone Vol. 31, No. 3 September 2002:359 –365 14. Hanks, S. K. and Polte, T. R. Signaling through focal adhesion kinase. Bioessays 19:137–145; 1997. 15. Harp, J. B., Sayeski, P. P., Scanlon, M., Bernstein, K. E., and Marrero, M. B. Role of intracellular calcium in the angiotensin II-mediated tyrosine phosphorylation and dephosphorylation of PLC-␥1. Biochem Biophys Res Commun 232:540 –544; 1997. 16. Hauck, C. R., Klingbeil, C. K., and Schlaepfer, D. D. Focal adhesion kinase functions as a receptor-proximal signaling component required for directed cell migration. Immunol Res 21:293–303; 2000. 17. Hunter, S. J., Schraer, H., and Gay, C. V. Characterization of the cytoskeleton of isolated chick osteoclasts: Effect of calcitonin. J Histochem Cytochem 37:1529 –1537; 1989. 18. Lev, S., Moreno, H., Martinez, R., Canoll, P., Peles, E., Musacchio, J. M., Plowman, G. D., Rudy, B., and Schlessinger, J. Protein tyrosine kinase PYK2 involved in Ca2⫹-induced regulation of ion channel and MAP kinase functions. Nature 376:737–745; 1995. 19. Malgaroli, A., Meldolesi, J., Zallone, A. Z., and Teti, A. Control of cytosolic free calcium in rat and chicken osteoclasts. The role of extracellular calcium and calcitonin. J Biol Chem 264:14342–14347; 1989. 20. Miyazaki, T., Katagiri, H., Kanegae, Y., Takayanagi, H., Sawada, Y., Yamamoto, A., Pando, M. P., Asano, T., Verma, I. M., Oda, H., Nakamura, K., and Tanaka, S. Reciprocal role of ERK and NF-B pathways in survival and activation of osteoclasts. J Cell Biol 148:333–342; 2000. 21. Ohmori, T., Yatomi, Y., Asazuma, N., Satoh, K., and Ozaki, Y. Involvement of proline-rich tyrosine kinase 2 in platelet activation: Tyrosine phosphorylation mostly dependent on ␣IIb3 integrin and protein kinase C, translocation to the cytoskeleton and association with Shc through Grb2. Biochem J 347:561–569; 2000. 22. Raja, S., Avraham, S., and Avraham, H. Tyrosine phosphorylation of the novel protein-tyrosine kinase RAFTK during an early phase of platelet activation by an integrin glycoprotein IIb–IIIa-independent mechanism. J Biol Chem 272: 10941–10947; 1997. 23. Rock, M. T., Dix, A. R., Brooks, W. H., and Roszman, T. L. 1 integrinmediated T cell adhesion and cell spreading are regulated by calpain. Exp Cell Res 261:260 –270; 2000. 24. Rodriguez-Fernandez, J. L. and Rozengurt, E. Bombesin, vasopressin, lysophosphatidic acid, and sphingosylphosphorylcholine induce focal adhesion kinase activation in intact Swiss 3T3 cells. J Biol Chem 273:19321–19328; 1998. 25. Sanjay, A., Houghton, A., Neff, L., Didomenico, E., Bardelay, C., Antoine, E., Levy, J., Gailit, J., Bowtell, D., Horne, W. C., and Baron, R. Cbl associates with Pyk2 and Src to regulate Src kinase activity, ␣v3 integrin-mediated signaling, cell adhesion, and osteoclast motility. J Cell Biol 152:181–195; 2001. 26. Schaller, M. D. and Sasaki, T. Differential signaling by the focal adhesion kinase and cell adhesion kinase . J Biol Chem 272:25319 –25325; 1997. 27. Schlaepfer, D.D., Hanks, S.K., and van der Geer, P. Integrin-mediated signal transduction linked to Ras pathway by GRB2 binding to facial adhesion kinase. Nature 372:786 –791; 1994. 28. Selander, K. S., Harkonen, P. L., Valve, E., Monkkonen, J., Hannuniemi, R., and Vaananen, H. K. Calcitonin promotes osteoclast survival in vitro. Mol Cell Endocrinol 122:119 –129; 1996.
Z. Zhang et al. Calcitonin regulates Pyk2 and FAK phosphorylation in osteoclasts
365
29. Shyu, J.-F., Zhang, Z., Hernandez-Lagunas, L., Camerino, C., Chen, Y., Inoue, D., Baron, R., and Horne, W. C. Protein kinase C antagonizes pertussis-toxinsensitive coupling of the calcitonin receptor to adenylyl cyclase. Eur J Biochem 262:95–101; 1999. 30. Sieg, D. J., Ilic, D., Jones, K. C., Damsky, C. H., Hunter, T., and Schlaepfer, D. D. Pyk2 and Src-family protein-tyrosine kinases compensate for the loss of FAK in fibronectin-stimulated signaling events but Pyk2 does not fully function to enhance FAK- cell migration. EMBO J 17:5933–5947; 1998. 31. Sims, N. A., Aoki, K., Bogdanovich, Z., Maragh, M., Okigaki, M., Logan, S., Neff, L., DiDomenico, E., Sanjay, A., Schlessinger, J., and Baron, R. Impaired osteoclast function in Pyk2 knockout mice and cumulative effects in Pyk2/Src double knockout. J Bone Miner Res 14(Suppl. 1):S183; 1999. 32. Soriano, P., Montgomery, C., Geske, R., and Bradley, A. Targeted disruption of the c-src proto-oncogene leads to osteopetrosis in mice. Cell 64:693–702; 1991. 33. Su, Y., Chakraborty, M., Nathanson, M. H., and Baron, R. Differential effects of the 3⬘,5⬘-cyclic adenosine monophosphate and protein kinase C pathways on the response of isolated rat osteoclasts to calcitonin. Endocrinology 131:1497– 1502; 1992. 34. Suzuki, H., Nakamura, I., Takahashi, N., Ikuhara, T., Matsuzaki, K., Isogai, Y., Hori, M., and Suda, T. Calcitonin-induced changes in the cytoskeleton are mediated by a signal pathway associated with protein kinase A in osteoclasts. Endocrinology 137:4685–4690; 1996. 35. Tanaka, S., Amling, M., Neff, L., Peyman, A., Uhlmann, E., Levy, J. B., and Baron, R. c-Cbl is downstream of c-Src in a signalling pathway necessary for bone resorption. Nature 383:528 –531; 1996. 36. Teti, A., Paniccia, R., and Goldring, S. R. Calcitonin increases cytosolic free calcium concentration via capacitative calcium influx. J Biol Chem 270: 16666 –16670; 1995. 37. Xiong, W.-C. and Parsons, J. T. Induction of apoptosis after expression of PYK2, a tyrosine kinase structurally related to focal adhesion kinase. J Cell Biol 139:529 –539; 1997. 38. Zaidi, M., Datta, H. K., Moonga, B. S., and MacIntyre, I. Evidence that the action of calcitonin on rat osteoclasts is mediated by two G proteins acting via separate post-receptor pathways. J Endocrinol 126:473–481; 1990. 39. Zhang, Z., Hernandez-Lagunas, L., Horne, W. C., and Baron, R. Cytoskeletondependent tyrosine phosphorylation of the p130Cas family member HEF1 downstream of the G protein-coupled calcitonin receptor. Calcitonin induces the association of HEF1, paxillin, and focal adhesion kinase. J Biol Chem 274:25093–250938; 1999. 40. Zheng, C., Xing, Z., Bian, Z. C., Guo, C., Akbay, A., Warner, L., and Guan, J-L. Differential regulation of Pyk2 and focal adhesion kinase (FAK). The C-terminal domain of FAK confers response to cell adhesion. J Biol Chem 273:2384 –2389; 1998.
Date Received: March 22, 2002 Date Revised: May 13, 2002 Date Accepted: May 15, 2002
Calcitonin Induces Dephosphorylation of Pyk2 and Phosphorylation of Focal Adhesion Kinase in Osteoclasts Z. ZHANG,1,2 L. NEFF,1 A. L. M. BOTHWELL,2 R. BARON,1,3,4 and W. C. HORNE1,4 1
Departments of Cell Biology and Orthopaedics, 2Section of Immunobiology, 3Yale Cancer Center, and 4Yale Core Center for Musculoskeletal Disorders, Yale University School of Medicine, New Haven, CT, USA
Introduction Calcitonin induces the association and tyrosine phosphorylation of focal adhesion kinase (FAK), paxillin, and HEF1 in HEK-293 cells that overexpress the calcitonin receptor (C1a-HEK), but the hormone’s effect on these adhesion-related proteins in osteoclasts is not known. We therefore studied the effect of calcitonin on the tyrosine phosphorylation and subcellular distribution of paxillin, HEF1, FAK, and Pyk2, a FAK-related tyrosine kinase, in osteoclasts. Osteoclasts expressed both Pyk2 and FAK, with Pyk2 much more highly expressed. The two tyrosine kinases and paxillin were prominently associated with small punctate structures that were most densely clustered in the region of the peripheral F-actin-rich ring. Some of the punctate structures stained either for Pyk2 alone or FAK alone. Treatment with calcitonin disrupted the actin ring and induced the loss of the peripheral staining of paxillin, Pyk2, and FAK. In calcitonin-treated osteoclast-like cells, the tyrosine phosphorylation of paxillin and FAK increased, whereas the tyrosine phosphorylation of Pyk2 decreased. Calcitonin also induced increased phosphorylation of Erk1 and Erk2 in osteoclasts, as it did in the C1a-HEK cells. The unexpected dephosphorylation of Pyk2 correlated with decreased phosphorylation of Tyr402, the autophosphorylation site of Pyk2. The calcitonin-induced dephosphorylation of Pyk2 was not observed in C1a-HEK cells transfected with Pyk2, suggesting that the reduced phosphorylation seen in osteoclasts may be specific to these cells. Treatment of osteoclast-like cells with 12-phorbol 13-myristate acetate increased the tyrosine phosphorylation of both Pyk2 and FAK, and calphostin C, an inhibitor of protein kinase C, blocked calcitonin-stimulated FAK phosphorylation. Increasing intracellular calcium with ionomycin caused a decrease in the tyrosine phosphorylation of Pyk2 and the loss of the actin ring in a manner similar to the effect of calcitonin. Ionomycin had no effect on FAK tyrosine phosphorylation. Calcitonin (CT)-induced changes in Pyk2, FAK, and Erk1/2 phosphorylation were independent of c-Src. (Bone 31:359 –365; 2002) © 2002 by Elsevier Science Inc. All rights reserved.
Osteoclasts (OCs) attach to extracellular matrix or the bone surface via specialized attachment structures called podosomes, which form a prominent F-actin-rich ring that is thought to correspond to the sealing zone of resorbing OCs.3 The dynamic assembly and disassembly of podosomes are important for OC function, as seen in Src⫺/⫺ OCs, which exhibit decreased cell motility and adhesion structures that are abnormal in terms of both intracellular location and turnover.25 These defects may contribute to the reduced resorbing activity of Src⫺/⫺ OCs. We and others have suggested that Pyk2, a member of the focal adhesion kinase (FAK) family that is highly expressed in OCs, plays a key role in the Src-dependent regulation of the adhesion and motility of OCs. Pyk2 localizes in the podosomes of OCs and binds to c-Src upon activation of ␣v3 integrin, the main integrin expressed in OCs.8,25 Pyk2-deficient OCs and OCs expressing antisense Pyk2 exhibit abnormal adhesion structures and decreased bone resorption in vitro and in vivo, whereas Pyk2/Src double knockout mice develop more severe osteopetrosis and deficient bone remodeling than mice with only the Pyk2 or Src gene deletion.9,31 Pyk2 and FAK have similar domain structures, with a central kinase (catalytic) domain, multiple phosphorylated tyrosines, two proline-rich regions in the C-terminus, and a focal adhesion targeting domain. For full kinase activity, both proteins require the autophosphorylation of homologous sites (Pyk2 Tyr402, FAK Tyr397) that serve as binding sites for Src family kinases. Although the two proteins are 61% identical overall, the N- and C-terminal regions outside of the kinase domain are only 42% and 36% identical, respectively,18 suggesting that there are likely to be both similarities and differences in the signaling functions of the two proteins. Consistent with this, Pyk2 compensates for FAK’s role in fibronectin-induced cell spreading, Grb2-Shc association, and Src-dependent Erk2 activation in FAK⫺/⫺ cells,30 and both proteins can be activated by integrin ligation and activation of G-protein-coupled receptors in a variety of cells.1,12,14,24 However, Pyk2 and FAK activation have been reported to differ in activated platelets in terms of time course and dependence on integrin engagement,22 and the two proteins respond differently to the treatment of rat aortic smooth muscle cells, transfected Chinese hamster ovary cells, and chicken embryo cells with various soluble factors.26,40 Calcitonin (CT) inhibits OC motility, induces OC retraction, and disrupts the actin-ring structure of OCs,17,33,34,38 suggesting that it might regulate the activity of proteins such as FAK and Pyk2 that are involved in the regulation of osteoclast adhesion.
Key Words: calcitonin, osteoclast, Pyk2, focal adhesion kinase, paxillin.
Dr. Zhang is the recipient of a fellowship from the Arthritis Foundation. Address for correspondence and reprints: Dr. William C. Horne, Department of Orthopaedics, Yale University School of Medicine, P.O. Box 208044, New Haven CT 06520-8044. E-mail: [email protected] © 2002 by Elsevier Science Inc. All rights reserved.
359
8756-3282/02/$22.00 PII S8756-3282(02)00834-7
360
Z. Zhang et al. Calcitonin regulates Pyk2 and FAK phosphorylation in osteoclasts
Our laboratory previously reported that CT stimulates tyrosine phosphorylation and association of FAK, paxillin, and HEF1 in HEK-293 cells that overexpress the C1a isoform of the CT receptor (C1a-HEK).39 Moreover, induction of these responses involves activation of protein kinase C (PKC) and increased cytosolic calcium, two signaling effectors known to activate Pyk2 and increase Pyk2 phosphorylation.4,18,21,25 We therefore examined the effect of CT on Pyk2, FAK, paxillin, and HEF1 in OCs. Paxillin, Pyk2, and FAK were all relatively enriched in the region of the F-actin-rich podosome ring in authentic OCs and OC-like cells (OCLs) derived from rabbit bone marrow cultures, and treatment with CT induced their dispersion. The tyrosine phosphorylation of paxillin and FAK increased transiently following CT treatment, whereas Pyk2 phosphorylation decreased transiently, suggesting that Pyk2 and FAK, the two related kinases involved in cellular adhesion structures, play different roles downstream of CT receptor activation on OCs. Materials and Methods Reagents and Antibodies Salmon calcitonin was purchased from Peninsula Laboratories, Inc. (Belmont, CA). Calphostin C and ionomycin were from Calbiochem-Novabiochem International (San Diego, CA). Phorbol 12-myristate 13-acetate (PMA), prostaglandin E2, and fetal bovine serum (FBS) were from Sigma. The anti-FAK polyclonal antibody (Cat. No. 06-543) used for immunoprecipitation and immunostaining and the anti-phosphotyrosine antibody (Clone 4G10) were from Upstate Biotechnology, Inc. (Lake Placid, NY). The anti-FAK antibody used for western blotting, the monoclonal anti-paxillin antibody and the anti-Pyk2 antibody were from Transduction Laboratories. The antibody against phospho-Pyk2 was from Biosource International (Camarillo, CA). Anti-HEF1 antibody was from Santa Cruz Biotechnology (Santa Cruz, CA). Antibodies against phosphorylated Erk1/2 and total Erk1/2 were purchased from New England Biolabs (Beverly, MA). Stock solutions of calphostin C, ionomycin, and PMA were prepared in dimethylsulfoxide (DMSO). Isolation of Osteoclasts and Osteoclast-like Cells Rabbit authentic osteoclasts and osteoclast-like cells. Bone cells were isolated and cultured as described elsewhere,7 with modifications. Briefly, the long bones and scapulae of neonatal New Zealand white rabbits (60 –90 g) were dissected free of soft tissue and minced in ␣-modified minimal essential medium with 0.55 g/L sodium bicarbonate and 10 mmol/L HEPES (pH 7.10). The bone fragments were then allowed to settle, the cell suspension was collected, and the cells were pelleted by centrifugation at 1500g for 5 min at 4°C. To isolate mature OCs, the cell pellets were resuspended in medium supplemented with 10% FBS, aliquoted (100 L) onto coverslips in six well dishes, and cultured for 1 h at 37°C in humidified air with 5% CO2. Then, 2 mL of medium supplemented with 5% FBS was added to each well. After an additional 3 h incubation, the cultures were washed with fresh medium to remove nonadherent cells, and the remaining cells were cultured for another 2–3 h. The cells were then changed to fresh medium with 1% FBS and cultured for another 0.5 h before stimulation. To produce rabbit OCLs, the resuspended cells were cultured in medium supplemented with 10% FBS and 10⫺8 mol/L 1,25dihydroxyvitamin D3 (Roussel-UCLAF, Romainville, France) for 9 or 10 days. The medium was changed every other day. Cells were cultured in medium with 1% FBS for 18 h before stimulation.
Bone Vol. 31, No. 3 September 2002:359 –365
Murine osteoclast-like cells. Murine OCLs were produced by coculturing calvarial osteoblasts with spleen and bone marrow cells, as described elsewhere.25 Transient Transfection C1a-HEK cells were cultured and transiently transfected as described elsewhere.39 Immunoprecipitation and Immunoblotting Prior to treatment, rabbit and murine cocultures were incubated for 18 h in medium containing 1% FBS. The serum-starved OCLs were stimulated with CT or other agents in the presence of the cocultured mesenchymal cells at 37°C. The dishes were then transferred onto ice and the OCLs purified at 0°C by peeling away the layer of less adherent, mainly stromal cells. The purified OCLs were processed for immunoprecipitation and western blotting, as previously described,39 with slight modifications. The purified OCLs and C1a-HEK cells were lysed in 1% IGEPAL CA-630, 150 mmol/L NaCl, 50 mmol/L Tris-HCl (pH 8.0), 5 mmol/L ethylene-diamine tetraacetic acid (EDTA), 1 mmol/L phenylmethylsulfonylfluoride, 5 mmol/L iodoacetamide, 10 mmol/L NaF, 0.4 mmol/L Na2VO4, and 10 g/mL aprotinin. The lysates were incubated with specific antibodies and protein G-Sepharose, and the immune complexes were washed with PBS containing 1 mmol/L phenylmethylsulfonylfluoride, 5 mmol/L iodoacetamide, 10 mmol/L NaF, 0.4 mmol/L Na2VO4, and 10 g/mL aprotinin. The samples were then processed for immunoblotting as described elsewhere.39 Immunostaining and Confocal Microscopy Authentic OCs on coverslips were fixed in PBS containing 3.7% formaldehyde for 10 min at room temperature, and washed in PBS. Coverslips for actin labeling were extracted in ice-cold acetone for 3–5 min, and returned to PBS. All other coverslips were permeabilized in 0.05% saponin for 30 min. The coverslips were blocked in 5% normal goat serum for 30 min, then incubated in the appropriate primary antibody for 2 h, washed in PBS, incubated for 1 h in the appropriate fluorescein- or rhodamine-conjugated secondary antibody, and washed. Those coverslips used for actin labeling were incubated in a 1:40 dilution (in PBS) of rhodamine phalloidin (Molecular Probes) for 20 min and washed. All coverslips were mounted in FluorSave. Cells were examined using a scanning laser confocal imaging system (MRC-600; Bio-Rad Laboratories). Images were recorded, composite images were compiled, and image enhancements performed using Adobe Photoshop 4.0. Results Calcitonin Modulates Tyrosine Phosphorylation of Podosome-associated Proteins in Osteoclasts We initially sought to determine if CT induced changes in the tyrosine phosphorylation of FAK family members, paxillin and HEF1, in OCLs as it does in the HEK-293 cells stably transfected with the CT receptor (CTR). Unlike the HEK cells, the predominant FAK family member in OCs is Pyk2,8,25 whereas the expression of FAK in OCs is more controversial, with some investigators reporting its presence and others failing to detect it.8,35 We therefore first examined the expression of Pyk2 and FAK in OCs by western blot (Figure 1A) and confocal immunofluorescence microscopy (Figure 2A,B). Both FAK and Pyk2 were detected in lysates of rabbit (Figure 1 and Figure 3) and
Bone Vol. 31, No. 3 September 2002:359 –365
Z. Zhang et al. Calcitonin regulates Pyk2 and FAK phosphorylation in osteoclasts
361
Double staining of FAK and Pyk2 (Figure 2B) showed that, although the locations of the two proteins were similar in the untreated OCs, at least as many of the individual punctate structures stained only for Pyk2 or FAK as stained for both. Treatment with 1 nmol/L CT induced the retraction of OCs, which could be detected by 3 min, and was maintained until at least 30 min (Figure 2C). CT also disrupted the actin ring, as previously reported,34 and dispersed the peripherally enriched Pyk2, FAK, and paxillin (Figure 2A, lower images), which became diffusely distributed throughout the OCs. To examine the effect of CT treatment on the tyrosine phosphorylation of Pyk2, FAK, paxillin, and HEF1, OCLs were treated with 1 nmol/L CT for time durations as long as 30 min, then washed and lysed. The individual proteins were immunoprecipitated and analyzed for phosphotyrosine content by western blot. In contrast to what we previously found in the C1a-HEK cells, OCLs expressed relatively low amounts of HEF1 and there was little change in the tyrosine phosphorylation after CT treatment (Figure 1B). The tyrosine phosphorylation of both paxillin (Figure 1C) and FAK (Figure 1D) increased transiently following CT addition, with the phosphotyrosine content of both proteins peaking at 1–3 min and returning to near-basal levels by about 10 min. Unexpectedly, CT induced a decrease in the tyrosine phosphorylation of Pyk2, which reached a minimum by 3–10 min and returned to the basal level by 30 min (Figure 1E). This decrease in Pyk2 tyrosine phosphorylation was correlated with decreased phosphorylation of Tyr402 (Figure 1F), the autophosphorylation site of Pyk2.
Calcium, But Not PKC, Induces Decreased Tyrosine Phosphorylation of Pyk2 in Osteoclasts Figure 1. Calcitonin regulates the tyrosine phosphorylation of paxillin, FAK, and Pyk2 in osteoclasts. (A) To examine the expression of FAK in OCs, rabbit OCLs were generated and lysed, and the total cell lysates processed for western blotting with anti-Pyk2 (top left) and anti-FAK (top right) antibodies, as described in Materials and Methods. The membranes were then stripped and reblotted with anti-actin antibody to demonstrate equal loading (bottom). (B)–(F) Rabbit OCLs were cultured overnight in low serum, then treated with 1 nmol/L CT for the indicated times and purified and lysed, as described in Materials and Methods. HEF1 (B), paxillin (C), FAK (D), and Pyk2 (E) were immunoprecipitated from the lysates and the immune complexes were western blotted with anti-phosphotyrosine antibody [(B–E) top]. The membranes were then stripped and reblotted with antibody to the appropriate antigen (bottom). The amounts of phosphotyrosine and protein in the blots were quantified by densitometry, and the phosphotyrosine was corrected for the sample load based on the amount of protein and expressed as fold increase or decrease relative to the control lane (the numbers below each pair of blots). (F) Total cell lysates were immunoblotted with anti-Pyk2 phospho-Tyr402 antibody (top). The membranes were then stripped and reblotted with anti-Pyk2 (bottom). In (B) and (F), the OCLs were treated with CT for 3 min. Each blot is representative of three or four replicate experiments.
murine (Figure 4) OCLs by western blot, although the level of Pyk2 was much higher, consistent with other reports. In untreated authentic OCs, Pyk2-positive punctate structures were densely present in the peripheral F-actin-rich ring and similar structures were also more sparsely distributed throughout the rest of the cell (Figure 2A, upper right image). The pattern of the FAK staining (Figure 2A, upper center image) was similar to that of Pyk2 (and to that of paxillin, Figure 2A, upper left image), but the intensity of FAK staining was much weaker than that of Pyk2, consistent with the relative levels seen in the western blots.
The CT-induced decrease in Pyk2 tyrosine phosphorylation in OCLs was unexpected, because we and others have shown that signaling from the CTR activates PKC and increases cytosolic calcium,6,19,29,33,36,38 and it has been generally observed that stimuli that activate PKC or increase cytosolic calcium, including the activation of some G-protein-coupled receptors, stimulate Pyk2 phosphorylation.12,18 We therefore examined the roles of PKC and calcium in the regulation of Pyk2 phosphorylation in OCLs and C1a-HEK cells. In the OCLs (Figure 3A), PMA caused the tyrosine phosphorylation of both Pyk2 (left) and FAK (middle) to increase moderately, in contrast to the effect of CT on Pyk2 phosphorylation. Ionomycin, on the other hand, induced a decrease in Pyk2 tyrosine phosphorylation (left), similar to the effect of CT and in contrast to what has been reported in the literature for other cell types. Ionomycin did not change FAK phosphorylation (right). Thus, the CT-induced decrease in Pyk2 phosphorylation in OCLs appears to be mediated by increased cytosolic calcium, suggesting that calcium activates different signaling effectors in OCs than in many other cells. C1a-HEK cells were transiently transfected with Pyk2 cDNA, because Pyk2 was not detectable by western blot (data not shown). As predicted from the published reports and our earlier findings, the tyrosine phosphorylation of the transiently expressed Pyk2 was strongly induced in the C1a-HEK cells by CT, PMA, and ionomycin (Figure 3B). We also examined the effect of ionomycin on the actin ring. Treatment of authentic mature osteoclasts with 1 mol/L ionomycin for 5 min caused a loss of the actin ring, similar to the effect of CT (Figure 3C).
362
Z. Zhang et al. Calcitonin regulates Pyk2 and FAK phosphorylation in osteoclasts
Bone Vol. 31, No. 3 September 2002:359 –365
Figure 2. Effect of calcitonin on osteoclast morphology and the distribution of paxillin, Pyk2, and FAK in osteoclasts. Authentic rabbit OCs were isolated and plated on coverslips as described in Materials and Methods. (A) Cells treated with 1 nmol/L CT for 3 min (lower images), or not (upper images), were fixed and stained for F-actin (not shown) and either paxillin, FAK, or Pyk2, then examined by confocal immunofluorescence microscopy. The optical slice in these and the other images was taken at the level of cell attachment to the substrate, in order to visualize attachment structures. In untreated OCs, all three proteins were enriched in the peripheral actin-rich ring. Treatment with 1 nmol/L CT for 3 min disrupted the peripheral actin ring (not shown) and induced the redistribution of all three proteins to a more diffuse pattern throughout the cell interior. (B) Double labeling of untreated OCs with monoclonal anti-FAK (red) and polyclonal anti-Pyk2 (green). In an enlarged portion of the merged image (lower panel) with the red channel enhanced to increase FAK intensity, numerous punctate structures that stain only for FAK can be seen (arrowheads), in addition to similar structures that stain only for Pyk2 (green) or both proteins (yellow). (C) The CT-induced cell retraction and loss of the actin-rich peripheral podosome ring in authentic rabbit OCs. Cells were treated with 1 nmol/L CT for the indicated times, fixed, and labeled with rhodamine-conjugated phalloidin to visualize F-actin distribution.
Calcitonin-induced Changes in Pyk2, FAK, and Erk1/2 Are Independent of c-Src c-Src is required for normal OC function,32 in part because of the role that it plays in modulating OC attachment and motility.25 c-Src also plays a role in modulating FAK and Pyk2 activity.14 We therefore examined the effect of deleting c-Src on the CT-induced changes in the tyrosine phosphorylation of FAK and Pyk2 in OCs (Figure 4). For these experiments, OCLs were prepared by coculturing osteoblastic cells derived from wild-type mouse calvariae with bone marrow and spleen cells isolated from Src⫺/⫺ mice or their normal littermates. OCLs were treated with CT, lysed, and the phosphotyrosine content of FAK and Pyk2 analyzed. As in the rabbit OCLs, treatment of wild-type murine OCLs with CT decreased Pyk2 phosphorylation (Figure 4, top panel), and increased FAK phosphorylation (middle panel). Similar results were obtained with the Src⫺/⫺ OCLs, indicating CT-induced dephosphorylation of Pyk2 and phosphorylation of FAK do not require c-Src. Erk1/2, the MAP kinases that are involved in OC survival,20 have been reported to be activated downstream of FAK and Pyk2 following integrin ligation.5,27 Our laboratory showed previously that CT stimulates Erk1/2 phosphorylation in C1a-HEK cells.6 We therefore examined the effect of CT treatment on Erk1/2 phosphorylation in the Src⫹/? and Src⫺/⫺ OCLs (Figure 4, lower panel). CT induced the phosphorylation of Erk1/2, and the response was independent of the presence of Src.
Discussion CT is a potent inhibitor of osteoclastic bone resorption, although much remains to be learned about its exact mode of action. Treatment with CT induces retraction and inhibits mobility of isolated OCs,33,38 responses that are considered to contribute to the hormone’s inhibition of bone resorption. The loss of motility following the exposure of OCs to CT suggests that CT might regulate the function of the podosomes, which mediate the attachment of the OC to the bone matrix during locomotion and bone resorption. Pyk2 and FAK are two related tyrosine kinases that are known to play critical roles in integrin-mediated cell attachment in a variety of cells (reviewed in references 2 and 16). Pyk2 is strongly expressed in OCs8 and appears to play a key role in integrin-dependent signaling in these cells.8,9,25 OCs from Pyk2 knockout mice exhibit reduced bone resorbing activity and abnormally distributed podosomes in vitro and the mice exhibit a mild form of osteopetrosis, indicating that in vivo OC function is also impaired.31 In contrast to Pyk2, there has been little study of FAK in OCs. Although some investigators have failed to detect FAK in OCs,8 our confocal immunofluorescence data clearly show that FAK is expressed in this cell, although at much lower levels than the mesenchymal cells that were also present in these preparations. Pyk2, FAK, and paxillin, which binds to both FAK and Pyk2, had similar patterns of localization in isolated OCs plated on glass. The two kinases were associated with podosome-like
Bone Vol. 31, No. 3 September 2002:359 –365
Z. Zhang et al. Calcitonin regulates Pyk2 and FAK phosphorylation in osteoclasts
363
Figure 3. Increased calcium mimics the effect of calcitonin on Pyk2 tyrosine phosphorylation in osteoclasts and C1a-HEK cells. (A) (Left panels) Serum-starved rabbit OCLs were treated with vehicle, CT (1 nmol/L for 3 min), PMA (100 nmol/L for 10 min), or ionomycin (100 nmol/L for 3 min). Pyk2 was immunoprecipitated, and the immune complexes were immunoblotted with anti-phosphotyrosine antibody (top). The membrane was then stripped and reblotted with anti-Pyk2 antibody (bottom). (Center and right panels) Serum-starved rabbit OCLs were treated with vehicle, CT (1 nmol/L for 3 min), PMA (100 nmol/L for 10 min), calphostin C (CC; 1 mol/L for 30 min) prior to CT (1 nmol/L for 3 min) or ionomycin (100 nmol/L for 3 min) and then lysed. FAK was immunoprecipitated, and the immune complexes were blotted with anti-phosphotyrosine antibody (top). The membrane was then stripped and reprobed with anti-FAK antibody (bottom). Phosphotyrosine was quantified as described for Figure 1. The blots are representative of three or four replicate experiments. (B) C1a-HEK cells were transiently transfected with Pyk2 cDNA. After serum starvation for 18 h, the cells were treated with vehicle, CT (1 nmol/L for 3 min), PMA (100 nmol/L for 10 min), or ionomycin (100 nmol/L for 3 min), then lysed. Pyk2 was immunoprecipitated, and the immune complexes were blotted with anti-phosphotyrosine antibody. The membrane was then stripped and reblotted with anti-Pyk2 antibody. The blots are representative of three replicate experiments. (C) Authentic rabbit OCs were isolated and plated on coverslips as described in Materials and Methods. Cells were treated with 1 mol/L ionomycin or vehicle as indicated for 5 min, fixed, and stained for F-actin, then examined by confocal immunofluorescence microscopy as described for Figure 2. Ionomycin induced the disappearance of the actin ring in a manner similar to the effect of CT treatment.
punctate structures that occur most densely in a peripheral ring, but are also seen throughout the cell to a lesser extent. It may be significant, given the different effects of CT on the tyrosine
Figure 4. Calcitonin-induced changes in Pyk2, FAK, and Erk1/2 phosphorylation in osteoclasts are independent of c-Src. Murine Src⫹/? and Src⫺/⫺ OCLs were serum starved for 18 h, then treated with 1 nmol/L CT for 3 min, purified, and lysed. Pyk2 (upper panels) and FAK (middle panels) were immunoprecipitated, and the immune complexes were blotted with antiphosphotyrosine antibody. The membranes were then stripped and reblotted with anti-Pyk2 antibody or anti-FAK antibody, respectively. The total cell lysates were immunoblotted with anti-phospho-Erk1/2 antibody and antiactin antibody as a loading control (bottom panels).
phosphorylation of the two kinases, that most of the podosomes stain for either Pyk2 or FAK, not both. CT induced the loss of the prominent peripheral staining of Pyk2, FAK, and paxillin in parallel with the loss of the actin ring. CT induced the transient dephosphorylation of Pyk2, which correlated with the decreased phosphorylation of Tyr402, the autophosphorylation site of Pyk2. Phosphorylation of Tyr402 is required for full Pyk2 kinase activity, suggesting that CT might inhibit osteoclastic bone resorption in part by inhibiting Pyk2 activity. The inhibition of Pyk2 would be expected in turn to prevent the recruitment of Src and Cbl to the Pyk2·Src·Cbl complex that this laboratory has shown to be formed as a consequence of the activation of the vitronectin receptor.25 This possibility is perhaps consistent with the fact that osteoclasts that lack either Src or Cbl have reduced motility,25 similar to CTtreated OCs. In contrast to the decreased phosphorylation of Pyk2, CT treatment increased the tyrosine phosphorylation of FAK. Although the few contaminating stromal cells do contribute to the FAK content in the cell lysates, the CT-induced changes are occurring in the OCs because these are the only cells in the coculture preparations that express the CT receptor. The time courses of CT-induced changes in Pyk2 and FAK phosphorylation were also different, with FAK phosphorylation occurring earlier and reverting to the basal level in a shorter time than Pyk2 dephosphorylation. The decrease in Pyk2 phosphorylation following the activation of the CT receptor in OCLs was unexpected. Treatment of OCs with CT activates PKC and increases cytosolic calci-
364
Z. Zhang et al. Calcitonin regulates Pyk2 and FAK phosphorylation in osteoclasts
um.19,29,33,36,38 In other cell types, PKC activation or increased cytosolic calcium induce Pyk2 activation and tyrosine phosphorylation.12,18 Indeed, the phosphotyrosine content of transiently expressed Pyk2 in the CT receptor-expressing C1a-HEK cells increased in response to CT, PMA, or ionomycin. In the OCs, activating PKC with PMA stimulated the tyrosine phosphorylation of both FAK and Pyk2, whereas the PKC inhibitor, calphostin C, prevented the CT-induced increase in FAK phosphorylation, indicating that PKC mediates the CT-induced FAK phosphorylation but may not contribute to the changes in Pyk2 phosphorylation. In contrast, increasing cytosolic calcium with ionomycin decreased the phosphorylation of Pyk2, but had no effect on FAK phosphorylation. It thus appears that the regulation of Pyk2 function in OCs differs significantly from the regulatory mechanisms in many other cells, with increased cytosolic calcium possibly inducing some activity in OCs that dephosphorylates Pyk2 and thereby antagonizes PKC-induced Pyk2 phosphorylation. In this context, it is of interest to note that the phosphotyrosine levels of FAK and paxillin also decrease at about the same time as the Pyk2 dephosphorylation is occurring, and that this behavior contrasts with the sustained tyrosine phosphorylation of FAK and paxillin in CT-treated C1a-HEK cells.39 There are previous reports of a correlation between increased cytosolic calcium and protein tyrosine dephosphorylation.10,11,15,23 The identification of a putative calcium-dependent tyrosine phosphatase in OCs will be the subject of future research. Another apparent difference between the signaling mechanisms of OCs and HEK cells is the role of Src in coupling the CT receptor to FAK. Unlike the HEK cells, where CT-induced FAK phosphorylation is enhanced by overexpressing wild-type Src and inhibited by dominant-negative Src, the CT-induced FAK phosphorylation in OCs was observed in the Src⫺/⫺ OCLs. The coupling of the CT receptor to Erk1/2 was also independent of Src in the OCLs. As in C1a-HEK cells,6 treatment of OCLs with CT stimulated Erk1/2 phosphorylation. Miyazaki and colleagues have reported that inhibiting Erk activity induces OC apoptosis, whereas activation of Erk supports OC survival.20 Treatment with CT protects OCs from apoptosis and promotes OC survival,28 and this effect may be mediated, at least in part, by the enhanced Erk1/2 phosphorylation that we observed. Because FAK has been reported to promote adhesion-dependent cell survival in epithelial and endothelial cells,13 and overexpression of Pyk2 induces apoptosis in other cell types,37 the decreased Pyk2 phosphorylation and increased FAK phosphorylation may also be related to CT’s promotion of OC survival. The fact that CT, which reduces OC motility, induces a reduction in the level of Pyk2 tyrosine phosphorylation provides further evidence that the phosphorylation of Pyk2 Y402 and the consequent binding of Src may be critical for podosome function and normal OC motility. Src is not required for the dephosphorylation of Pyk2. In fact, the dephosphorylation is more nearly complete in the Src⫺ OCLs, consistent with the binding of Src to Pyk2402 reported earlier,25 which would tend to protect phosphotyrosine402 from being hydrolyzed by phosphatases. According to our hypothesized role of the Pyk2·Src·Cbl complex in OC motility, dephosphorylation of Pyk2Y402 would be expected to prevent the formation of the complex and therefore inhibit OC motility in much the same way as deleting either Src or Pyk2. It is not clear what the function of FAK might be in the podosome, but whatever the function it is clearly different from the function of Pyk2, because the presence of the two kinases in individual podosomes appears to be largely mutually exclusive and CT induces an increase in FAK phosphotyrosine level, not a decrease. Perhaps FAK plays a role in the formation of podo-
Bone Vol. 31, No. 3 September 2002:359 –365
somes or stabilizes podosomes once they are formed, whereas Pyk2 promotes the relatively rapid disassembly of existing podosomes by recruiting Src and Cbl, as described elsewhere.25 In conclusion, we have shown that OCs express both Pyk2 and FAK and that the two adhesion kinases are associated with different populations of similar actin-rich punctate structures, probably podosomes. Treatment of the OCs with CT disrupted the actin ring and simultaneously induced the relocation of the Pyk2, FAK, and paxillin away from the OC periphery. CT also induced increased FAK and paxillin phosphorylation and decreased Pyk2 phosphorylation in OCLs. The change in Pyk2 phosphorylation was mimicked by increasing free cytosolic calcium, possibly implicating a calcium-dependent protein tyrosine phosphatase. Our results offer new insights into the signaling mechanisms by which CT affects osteoclast function.
Acknowledgments: We are grateful to our colleagues, Emilia DiDomenico and Karen Ford, for their technical support, and to Dr. Adam Houghton and Dr. Archana Sanjay for their helpful comments. This work is supported by National Institutes of Health Grants DE-04724 and AR-42927 (to R.B.) and by Yale Core Center for Musculoskeletal Disorders Grant AR-46032 (to W.C.H.).
References 1. Astier, A., Manie, S. N., Avraham, H., Hirai, H., Law, S. F., Zhang, Y., Golemis, E. A., Fu, Y., Druker, B. J., Haghayeghi, N., Freedman, A. S., and Avraham, S. The related adhesion focal tyrosine kinase differentially phosphorylates p130Cas and the Cas-like protein, p105HEF1. J Biol Chem 272: 19719 –19724; 1997. 2. Avraham, H., Park, S.-Y., Schinkmann, K., and Avraham, S. RAFTK/Pyk2mediated cellular signalling. Cell Signal 12:123–133; 2000. 3. Baron, R. Molecular mechanisms of bone resorption.An update. Acta Orthop Scand 266:66 –70; 1995 (Suppl.). 4. Brinson, A. E., Harding, T., Diliberto, P. A., He, Y., Li, X., Hunter, D., Herman, B., Earp, H. S., and Graves, L. M. Regulation of a calcium-dependent tyrosine kinase in vascular smooth muscle cells by angiotensin II and plateletderived growth factor. Dependence on calcium and the actin cytoskeleton. J Biol Chem 273:1711–1718; 1998. 5. Chen, Q., Kinch, M. S., Lin, T. H., Burridge, K., and Juliano, R. L. Integrinmediated cell adhesion activates mitogen-activated protein kinases. J Biol Chem 269:26602–26605; 1994. 6. Chen, Y., Shyu, J.-F., Santhanagopal, A., Inoue, D., David, J. -P., Dixon, S. J., Horne, W. C., and Baron, R. The calcitonin receptor stimulates Shc tyrosine phosphorylation and Erk1/2 activation. Involvement of GI, protein kinase C, and calcium. J Biol Chem 273:19809 –19816; 1998. 7. David, J.-P., Neff, L., Chen, Y., Rincon, M., Horne, W. C., and Baron, R. A new method to isolate large numbers of rabbit osteoclasts and osteoclast-like cells: Application to the characterization of serum response element binding proteins during osteoclast differentiation. J Bone Miner Res 13:1730 –1738; 1998. 8. Duong, L. T., Lakkakorpi, P. T., Nakamura, I., Machwate, M., Nagy, R. M., and Rodan, G. A. PYK2 in osteoclasts is an adhesion kinase, localized in the sealing zone, activated by ligation of ␣v3 integrin, and phosphorylated by Src kinase. J Clin Invest 102:881–892; 1998. 9. Duong, L. T., Nakamura, I., Lakkakorpi, P. T., Lipfert, L., Bett, A. J., and Rodan, G. A. Inhibition of osteoclast function by adenovirus expressing antisense protein-tyrosine kinase 2. J Biol Chem 276:7484 –7492; 2001. 10. Errasfa, M. and Stern, A. Melittin inhibits epidermal growth factor-induced protein tyrosine phosphorylation: Comparison with phorbol myristate acetate and calcium ionophore A23187. Biochim Biophys Acta 1222:471–476; 1994. 11. Falet, H., Pain, S., and Rendu, F. Tyrosine unphosphorylated platelet SHP-1 is a substrate for calpain. Biochem Biophys Res Commun 252:51–55; 1998. 12. Felsch, J. S., Cachero, T. G., and Peralta, E. G. Activation of protein tyrosine kinase PYK2 by the m1 muscarinic acetylcholine receptor. Proc Natl Acad Sci USA 95:5051–5056; 1998. 13. Frisch, S. M., Vuori, K., Ruoslahti, E., and Chan-Hui, P. Y. Control of adhesion-dependent cell survival by focal adhesion kinase. J Cell Biol 134: 793–799; 1996.
Bone Vol. 31, No. 3 September 2002:359 –365 14. Hanks, S. K. and Polte, T. R. Signaling through focal adhesion kinase. Bioessays 19:137–145; 1997. 15. Harp, J. B., Sayeski, P. P., Scanlon, M., Bernstein, K. E., and Marrero, M. B. Role of intracellular calcium in the angiotensin II-mediated tyrosine phosphorylation and dephosphorylation of PLC-␥1. Biochem Biophys Res Commun 232:540 –544; 1997. 16. Hauck, C. R., Klingbeil, C. K., and Schlaepfer, D. D. Focal adhesion kinase functions as a receptor-proximal signaling component required for directed cell migration. Immunol Res 21:293–303; 2000. 17. Hunter, S. J., Schraer, H., and Gay, C. V. Characterization of the cytoskeleton of isolated chick osteoclasts: Effect of calcitonin. J Histochem Cytochem 37:1529 –1537; 1989. 18. Lev, S., Moreno, H., Martinez, R., Canoll, P., Peles, E., Musacchio, J. M., Plowman, G. D., Rudy, B., and Schlessinger, J. Protein tyrosine kinase PYK2 involved in Ca2⫹-induced regulation of ion channel and MAP kinase functions. Nature 376:737–745; 1995. 19. Malgaroli, A., Meldolesi, J., Zallone, A. Z., and Teti, A. Control of cytosolic free calcium in rat and chicken osteoclasts. The role of extracellular calcium and calcitonin. J Biol Chem 264:14342–14347; 1989. 20. Miyazaki, T., Katagiri, H., Kanegae, Y., Takayanagi, H., Sawada, Y., Yamamoto, A., Pando, M. P., Asano, T., Verma, I. M., Oda, H., Nakamura, K., and Tanaka, S. Reciprocal role of ERK and NF-B pathways in survival and activation of osteoclasts. J Cell Biol 148:333–342; 2000. 21. Ohmori, T., Yatomi, Y., Asazuma, N., Satoh, K., and Ozaki, Y. Involvement of proline-rich tyrosine kinase 2 in platelet activation: Tyrosine phosphorylation mostly dependent on ␣IIb3 integrin and protein kinase C, translocation to the cytoskeleton and association with Shc through Grb2. Biochem J 347:561–569; 2000. 22. Raja, S., Avraham, S., and Avraham, H. Tyrosine phosphorylation of the novel protein-tyrosine kinase RAFTK during an early phase of platelet activation by an integrin glycoprotein IIb–IIIa-independent mechanism. J Biol Chem 272: 10941–10947; 1997. 23. Rock, M. T., Dix, A. R., Brooks, W. H., and Roszman, T. L. 1 integrinmediated T cell adhesion and cell spreading are regulated by calpain. Exp Cell Res 261:260 –270; 2000. 24. Rodriguez-Fernandez, J. L. and Rozengurt, E. Bombesin, vasopressin, lysophosphatidic acid, and sphingosylphosphorylcholine induce focal adhesion kinase activation in intact Swiss 3T3 cells. J Biol Chem 273:19321–19328; 1998. 25. Sanjay, A., Houghton, A., Neff, L., Didomenico, E., Bardelay, C., Antoine, E., Levy, J., Gailit, J., Bowtell, D., Horne, W. C., and Baron, R. Cbl associates with Pyk2 and Src to regulate Src kinase activity, ␣v3 integrin-mediated signaling, cell adhesion, and osteoclast motility. J Cell Biol 152:181–195; 2001. 26. Schaller, M. D. and Sasaki, T. Differential signaling by the focal adhesion kinase and cell adhesion kinase . J Biol Chem 272:25319 –25325; 1997. 27. Schlaepfer, D.D., Hanks, S.K., and van der Geer, P. Integrin-mediated signal transduction linked to Ras pathway by GRB2 binding to facial adhesion kinase. Nature 372:786 –791; 1994. 28. Selander, K. S., Harkonen, P. L., Valve, E., Monkkonen, J., Hannuniemi, R., and Vaananen, H. K. Calcitonin promotes osteoclast survival in vitro. Mol Cell Endocrinol 122:119 –129; 1996.
Z. Zhang et al. Calcitonin regulates Pyk2 and FAK phosphorylation in osteoclasts
365
29. Shyu, J.-F., Zhang, Z., Hernandez-Lagunas, L., Camerino, C., Chen, Y., Inoue, D., Baron, R., and Horne, W. C. Protein kinase C antagonizes pertussis-toxinsensitive coupling of the calcitonin receptor to adenylyl cyclase. Eur J Biochem 262:95–101; 1999. 30. Sieg, D. J., Ilic, D., Jones, K. C., Damsky, C. H., Hunter, T., and Schlaepfer, D. D. Pyk2 and Src-family protein-tyrosine kinases compensate for the loss of FAK in fibronectin-stimulated signaling events but Pyk2 does not fully function to enhance FAK- cell migration. EMBO J 17:5933–5947; 1998. 31. Sims, N. A., Aoki, K., Bogdanovich, Z., Maragh, M., Okigaki, M., Logan, S., Neff, L., DiDomenico, E., Sanjay, A., Schlessinger, J., and Baron, R. Impaired osteoclast function in Pyk2 knockout mice and cumulative effects in Pyk2/Src double knockout. J Bone Miner Res 14(Suppl. 1):S183; 1999. 32. Soriano, P., Montgomery, C., Geske, R., and Bradley, A. Targeted disruption of the c-src proto-oncogene leads to osteopetrosis in mice. Cell 64:693–702; 1991. 33. Su, Y., Chakraborty, M., Nathanson, M. H., and Baron, R. Differential effects of the 3⬘,5⬘-cyclic adenosine monophosphate and protein kinase C pathways on the response of isolated rat osteoclasts to calcitonin. Endocrinology 131:1497– 1502; 1992. 34. Suzuki, H., Nakamura, I., Takahashi, N., Ikuhara, T., Matsuzaki, K., Isogai, Y., Hori, M., and Suda, T. Calcitonin-induced changes in the cytoskeleton are mediated by a signal pathway associated with protein kinase A in osteoclasts. Endocrinology 137:4685–4690; 1996. 35. Tanaka, S., Amling, M., Neff, L., Peyman, A., Uhlmann, E., Levy, J. B., and Baron, R. c-Cbl is downstream of c-Src in a signalling pathway necessary for bone resorption. Nature 383:528 –531; 1996. 36. Teti, A., Paniccia, R., and Goldring, S. R. Calcitonin increases cytosolic free calcium concentration via capacitative calcium influx. J Biol Chem 270: 16666 –16670; 1995. 37. Xiong, W.-C. and Parsons, J. T. Induction of apoptosis after expression of PYK2, a tyrosine kinase structurally related to focal adhesion kinase. J Cell Biol 139:529 –539; 1997. 38. Zaidi, M., Datta, H. K., Moonga, B. S., and MacIntyre, I. Evidence that the action of calcitonin on rat osteoclasts is mediated by two G proteins acting via separate post-receptor pathways. J Endocrinol 126:473–481; 1990. 39. Zhang, Z., Hernandez-Lagunas, L., Horne, W. C., and Baron, R. Cytoskeletondependent tyrosine phosphorylation of the p130Cas family member HEF1 downstream of the G protein-coupled calcitonin receptor. Calcitonin induces the association of HEF1, paxillin, and focal adhesion kinase. J Biol Chem 274:25093–250938; 1999. 40. Zheng, C., Xing, Z., Bian, Z. C., Guo, C., Akbay, A., Warner, L., and Guan, J-L. Differential regulation of Pyk2 and focal adhesion kinase (FAK). The C-terminal domain of FAK confers response to cell adhesion. J Biol Chem 273:2384 –2389; 1998.
Date Received: March 22, 2002 Date Revised: May 13, 2002 Date Accepted: May 15, 2002