HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 276, Issue 50, 47434-47444, December 14, 2001

This Article Abstract Full Text (PDF) All Versions of this Article: 276/50/47434    most recent M107494200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Chellaiah, M. A. Articles by Hruska, K. A. Search for Related Content PubMed PubMed Citation Articles by Chellaiah, M. A. Articles by Hruska, K. A. Social Bookmarking                      What's this?

Phosphatidylinositol 3,4,5-Trisphosphate Directs Association of Src Homology 2-containing Signaling Proteins with Gelsolin^*

Meenakshi A. Chellaiah^§, Rajat S. Biswas, David Yuen, Ulises M. Alvarez^¶, and Keith A. Hruska^¶

From the  Department of Oral and Craniofacial Biological Sciences, University of Maryland, Baltimore, Maryland 21201 and ^¶ Department of Pediatrics, Washington University School of Medicine, St. Louis, Missouri 63110

Received for publication, August 6, 2001, and in revised form, September 24, 2001

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Podosomes are adhesion structures in osteoclasts and are structurally related to focal adhesions mediating cell motility during bone resorption. Here we show that gelsolin coprecipitates some of the focal adhesion-associated proteins such as c-Src, phosphoinositide 3-kinase (PI3K), p130^Cas, focal adhesion kinase, integrin _v3, vinculin, talin, and paxillin. These proteins were inducibly tyrosine-phosphorylated in response to integrin activation by osteopontin. Previous studies have defined unique biochemical properties of gelsolin related to phosphatidylinositol 3,4,5-trisphosphate in osteoclast podosomes, and here we demonstrate phosphatidylinositol 3,4,5-trisphosphate/gelsolin function in mediating organization of the podosome signaling complex. Overlay and GST pull-down assays demonstrated strong phosphatidylinositol 3,4,5-trisphosphate-PI3K interactions based on the Src homology 2 domains of PI3K. Furthermore, lipid extraction of lysates from activated osteoclasts eliminated interaction between gelsolin, c-Src, PI3K, and focal adhesion kinase despite equal amounts of gelsolin in both the lipid-extracted and unextracted experiment. The cytoplasmic protein tyrosine phosphatase (PTP)-proline-glutamic acid-serine-threonine amino acid sequences (PEST) was also found to be associated with gelsolin in osteoclast podosomes and with stimulation of _v3-regulated phosphorylation of PTP-PEST. We conclude that gelsolin plays a key role in recruitment of signaling proteins to the plasma membrane through phospholipid-protein interactions and by regulation of their phosphorylation status through its association with PTP-PEST. Because both gelsolin deficiency and PI3K inhibition impair bone resorption, we conclude that phosphatidylinositol 3,4,5-trisphosphate-based protein interactions are critical for osteoclast function.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Osteoclasts are multinucleated giant cells with bone-resorbing activity. As osteoclasts crawl over bone surfaces, they require rapid attachment and release from the extracellular matrix. Adhesion structures called podosomes present in highly motile cells are also found in osteoclasts. Osteoclasts are unique because they use the speed of podosome assembly and disassembly to generate high rates of motility. Podosome formation stabilizes the bone matrix-cell interface and forms an isolated compartment between the ruffled border and the bone surface (1, 2). Consistent with their function as adhesion sites, podosomes contain many of the same proteins found in focal adhesions, such as F-actin, vinculin, talin, gelsolin, fimbrin, and -actinin (3-7).

We have shown previously that osteopontin (OPN)¹ binding to integrin _v3 in osteoclasts stimulates gelsolin-associated PI3K. This leads to increased levels of gelsolin-associated polyphosphoinositides, such as phosphatidylinositol 4,5-bisphosphate (PtdIns 4,5-P₂), phosphatidylinositol 3,4-bisphosphate, and phosphatidylinositol 3,4,5-trisphosphate, uncapping of actin barbed ends, and actin filament formation (8). Moreover, OPN stimulates gelsolin-associated c-Src, leading to increased gelsolin-associated PI3K activity (9). Elimination of gelsolin resulted in osteoclasts devoid of podosomes and actin rings but still capable of organization for bone resorption. These osteoclasts failed to respond to OPN with stimulation of motility and bone resorption (10). Also, these observations in vitro suggest an important role for gelsolin in osteoclast-actin dynamics in vivo. They also demonstrate the functional role and significance of gelsolin as well as the associated signaling molecules during OPN-induced osteoclast cytoskeletal organization. However, the molecular mechanisms involved in the recruitment of signaling molecules with gelsolin have not been elucidated.

Integrin activation following matrix-integrin binding results in an increased tyrosine phosphorylation of FAK (11, 12). Activation of FAK has been shown to induce its association with signal-transducing molecules such as c-Src, PI3K, paxillin, and p130^Cas (13-20). Paxillin acts as a scaffolding adaptor protein in integrin signaling by binding to several other integrin assembly proteins including vinculin, integrin ₁, FAK, and c-Src (21, 22). Binding of OPN- or RGD-containing peptides to _v3 stimulated the formation of signal-generating complexes consisting of FAK, c-Src, and PI3K associated with _v3 (23) and gelsolin (10). Integrin receptor engagement to the ligand or clustering of integrin receptors on the surface leads to the formation of focal adhesions. Because gelsolin deficiency blocks podosome assembly and _v3-stimulated signaling related to motility and bone resorption (10), we hypothesized that gelsolin may be an adaptor protein that could influence actin reorganization by recruiting the signaling proteins in the osteoclast podosome. Moreover, integrin-mediated tyrosine phosphorylation of proteins may modulate the protein-protein interactions between integrin and the downstream effectors.

A universal feature of the process of signal transduction and transmission in eucaryotic cells is the transient but specific association between protein molecules. A large body of evidence supports the view that these interactions are mediated by distinct functional modules within these proteins (24, 25). Many of the proteins involved in signal transduction contain recapitulations of polypeptide segments called the Src homology (SH) domains: SH1, SH2, or SH3 (26, 27). SH2-containing proteins are often downstream targets of protein tyrosine kinases (24, 28, 29) and SH3 domains, which bind polyproline sequences (30-32). Another new class of functional domain is pleckstrin homology domains. Pleckstrin homology domains have been suggested to bind to inositol phosphates (27, 33). However, gelsolin does not contain any of the domains mentioned above but does contain actin and phosphoinositide-binding domains (34-39). Functions of phosphoinositide-binding domains include targeting of proteins to sites of signal transduction and/or direct regulation of protein activity by phosphoinositide binding (40). Recent evidence suggests that the biochemical and physical organization of lipid molecules in the plasma membrane can affect integrin-mediated cellular functions (41-47). The indirect effects of lipid molecules on integrin function are manifested mainly through cytoskeletal or signaling molecules. Cytoskeleton-lipid interactions are involved in mediating the linkage of the cytoskeleton with membrane bilayer and also in defining the membrane architecture in specific areas (such as focal adhesions). Talin, vinculin, and -actinin are among the main integrin-associated cytoskeletal molecules that bind to or are modulated by membrane lipids (48-51).

An important insight into the role of phosphoinositides results from the finding that PtdIns P₃ associates directly with the SH2 domain of kinase and that PtdIns P₃ also disrupts the association of PI3K with tyrosine-phosphorylated proteins by binding to the SH2 domains of the p85 subunit (52). PtdIns P₃ displayed a discriminative affinity to 17-mer peptide, corresponding to the protein kinase C phosphorylation and calmodulin-binding domain of a brain-specific protein (53). In addition to protein kinase C activation, PtdIns P₃ provides an alternative mechanism for regulating protein kinase C activity in vivo by recruiting and concentrating its target proteins at the interface to facilitate the subsequent protein kinase C phosphorylation (54). A by-product of these studies has been the discovery that not only do SH2 and SH3 domains mediate interactions of proteins, but they may also bind to the lipid products of PI3K. As a result of this interaction, other SH2-containing proteins may be recruited to the plasma membrane to initiate downstream signaling (52, 55). From these observations, we hypothesized that the signaling molecules are associated with gelsolin through a unique phospholipid-based association.

This report describes the initial characterization of the signaling proteins that are associated with gelsolin as well as the role of PtdIns P₃ in the interaction of proteins with gelsolin. We focused on defining the interaction of PtdIns P₃ with the SH2 or SH3 domains of kinases, including c-Src, lck, and p85 (PI3K). A unique mechanism of PtdIns P₃-SH2 domain-mediated interaction of proteins with gelsolin is described.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Materials-- The antibodies anti-gelsolin (monoclonal antibody), talin (monoclonal antibody), vinculin (monoclonal antibody), PI3K (polyclonal antibody), nonimmune mouse IgG, and rabbit IgG were purchased from Sigma. Monoclonal FAK and paxillin antibodies were obtained from Transduction Laboratories (Lexington, KY). Polyclonal antibodies such as anti-phosphotyrosine (PY20), anti-FAK, anti-p130^Cas, anti-c-Cbl, and anti-PYK2 were bought from Santa Cruz Biotechnology Inc. (Santa Cruz, CA). Monoclonal antibodies such as anti-PI3K and PY20 were obtained from Upstate Biotechnology (Lake Placid, NY). Monoclonal Src antibody was purchased from Oncogene (Uniondale, NY). PtdIns P₂ and PtdIns P₃ antisera were purchased from Advanced Magnetics (Cambridge, MA) and Echelon Research Laboratories Inc. (Salt Lake City UT), respectively. Rainbow molecular weight marker and [³²P]orthophosphate were obtained from Amersham Pharmacia Biotech. Protein estimation reagent, molecular weight standards for proteins, and PAGE reagents were bought from Bio-Rad (Richmond, CA). Protein A-Sepharose, phospholipid standards, and all the other chemicals were from Sigma.

pGEX vectors containing the cDNA sequences encoding the SH2 and SH3 domains of p85, full-length p85, and the SH2 domain of lck were kindly provided by Dr. L. C. Cantley (Department of Cell Biology, Harvard Medical School, Boston, MA) (52); pGEX vector containing the SH2 domain of Src was kindly provided by Dr. W. J. Muller (McMaster University, Hamilton, Canada) (56). Polyclonal gelsolin antibody and full-length gelsolin cDNA were provided by Dr. D. J. Kwiatkowski (Division of Experimental Medicine, Brigham and Women's Hospital, Harvard Medical School). Monoclonal _v (LM142) as well as _v3 (LM609) antibodies and anti-PTP-PEST antibody was a gift from Dr. D. A. Cheresh (Scripps Clinic and Research Foundation, La Jolla, CA) and Dr. M. L. Tremblay (McGill University, Montreal, Canada), respectively. The monoclonal antibody to leupaxin was developed by ICOS, Inc. (Bothel, WA) and provided by Dr. Anandarup Gupta (Dental School, University of Maryland, Baltimore, MD).

Preparation of Osteoclast Precursors-- Avian osteoclast precursors were prepared as described previously (8). Mouse osteoclasts were generated in vitro using mouse bone marrow cells. Cells isolated from five mice were cultured in 100-mm dishes with 20 ml of -minimum Eagle's medium supplemented with 10% fetal bovine serum (-10). After culture for 24 h, nonadherent cells were layered on Histopaque 1077 (Sigma) and centrifuged at 300 × g for 15 min at room temperature. The cell layer between the Histopaque and the media was removed and washed with -10 media at 2000 rpm for 7 min at room temperature. Cells were resuspended in -10 media and cultured with the appropriate concentrations of macrophage colony-stimulating factor 1 (10 ng/ml) and osteoprotegrin ligand (55-75 ng/ml). After 3 days in culture, media were replaced with fresh cytokines. The multinucleated osteoclasts were seen from day 4.

Lysate Preparation and Western Analysis-- After 4 or 5 days in culture, cells were washed three times with ice-cold PBS, and lysates were made from avian or mouse osteoclasts. Protein contents were measured using the Bio-Rad protein assay reagent kit. Equal amounts of lysate proteins were immunoprecipitated with antibodies or nonimmune serum as mentioned under "Results." Immunoprecipitations and Western analysis were carried out as described previously (8, 9). Immunoprecipitations were also performed in lipid-extracted and unextracted lysates. Lipid extraction was performed in about 10 mg of avian osteoclast lysate protein. An equal volume of chloroform:methanol (4:1) mixture was added to osteoclast lysate, and the aqueous layer containing the proteins was separated. The organic layer containing chloroform was extracted once more with lysis buffer and pooled. Lipid-extracted and unextracted lysates were dialyzed against 4-5 liters of PBS with two to three changes for 24 h at 4 °C. Protein content was measured by using the Bio-Rad protein assay kit. There was about 25-30% loss in protein content during lipid extraction.

Immunostaining-- Osteoclasts cultured on whale dentine slices or glass coverslips were immunostained with different antibodies as shown (Figs. 2-4) (57, 58). Briefly, cells were fixed with 3% paraformaldehyde and permeabilized with 10 mM Tris-HCl, pH 7.4, 150 mM NaCl, and 1 mM CaCl₂ containing 0.1% Triton X-100 for 1 min. The cells were washed and incubated with primary antibodies (1:100 dilution) for 2 h, washed, and then counterstained with Cy2-or Cy3-conjugated anti-mouse or anti-rabbit IgG for 2 h. The cells were washed and mounted on a slide in a mounting solution (Vector Laboratories, Inc., Burlingame, CA). The cells were viewed on a Zeiss LSM 410 confocal laser-scanning microscope (Thornwood, NY), and images were recorded as described previously (10).

Purification of Lipids for Binding Studies-- Avian osteoclast precursor cells, after 4 days in culture, were kept in serum-free (PO[stack][low]4[high][/stack]) media for 2 h. The cells were labeled as described previously (8, 59) with carrier-free [³²P]orthophosphate for 2 h at 37 °C. After labeling, they were washed twice with serum-free media and incubated with OPN (25 µg/ml) for 15 min at 37 °C. Lipids were extracted from ³²P-labeled and unlabeled lysates as described previously (8) and dried under N₂. The dried lipids were reconstituted in 500 µl of chloroform:methanol (1:1) and spotted on silica gel TLC plates pretreated with 1.2% potassium oxalate in methanol and water (2:3). The plates were developed as described previously (8), and bands were visualized by autoradiography. The PtdIns P₃, PtdIns P₂, PtdIns P, and PtdIns spots were scraped and extracted as described previously with a minor modification (23). The ³²P-labeled phospholipid spots present in the cellulose slurry were incubated overnight with 1 N HCL. The next day, the cellulose slurry was extracted as described previously (23) and counted (PtdIns P₃ = 1 × 10⁶ cpm, PtdInsP₂ = 2 × 10⁶ cpm; PtdIns P = 3.5 × 10⁶ cpm, and PtdIns = 1 × 10⁷ cpm). The incorporation into PtdIns P₃ is 50% of the incorporation into PtdIns P₂ or 33% of the incorporation into PtdIns P. The purity of the isolated phosphoinositides was analyzed by TLC and high pressure liquid chromatography. For the unlabeled PtdIns P₃, labeled PtdIns P₃ was used as marker. After purification, a known amount of PtdIns P₂ (Sigma) was used to quantitate the unlabeled PtdIns P₃ by exposing the TLC to iodine vapor.

Lipid Binding Assay-- GST fusions proteins coupled to Sepharose beads were allowed to bind to 100 µM phosphatidyl serine vesicle containing the following ³²P-labeled phosphoinositide: 12,000 dpm of PtdIns 4,5-P₂ and 6000 dpm of PtdIns P₃. The binding was performed in 10 mM Hepes (pH 7.0), 1 mM EDTA buffer containing 0.02% Nonidet P-40. Binding was carried out essentially as described previously (52). Binding of each phospholipid to the indicated proteins was analyzed. After binding, Sepharose beads were washed with buffer containing 0.5% Nonidet P-40, and the bound lipids were extracted with chloroform:methanol mixture and analyzed by TLC (8).

Overlay Assay Using Lipid-spotted Membrane-- To spot the purified phosphoinositides onto a nitrocellulose membrane, lipid solutions containing PtdIns P₃ (purified), PtdIns P₂, and PtdIns P (Sigma) were used. About 2 pmol of purified PtdIns P₃ and 3-5 pmol of PtdIns P₂ or PtdIns P (Sigma) phospholipids in 1-2 µl of a chloroform:methanol:water (1:2:0.8; by volume) mixture were spotted and dried. The blots were blocked overnight at 4 °C with 5% bovine serum albumin in PBS and 0.1% Tween 20 (PBS-T). The membrane was washed three times with PBS-T and incubated overnight at 4 °C with the indicated proteins (Fig. 9B) at a concentration of 10 µg/ml in PBS-T containing 0.5% bovine serum albumin and 12 mM 2-mercaptoethanol. The blots were washed and immunoblotted with anti-GST monoclonal antibody (Santa Cruz Biotechnology Inc.) in 1:1000 dilutions as described previously (8).

GST Fusion Proteins-- pGEX vectors containing cDNA sequences encoding the SH2 and SH3 domains of p85, full-length p85, and the SH2 domain of c-Src and lck were expressed in Escherichia coli as GST fusion proteins and purified as described previously (52).

Overlay Assay Using GST Fusion Protein-containing Membrane-- To analyze the binding of lipid-associated gelsolin to the indicated GST fusion proteins (Fig. 7), 3-5 µg of GST fusion proteins were subjected to 12% SDS-PAGE and blotted onto nitrocellulose membrane. The membrane was blocked with 5% milk in PBS-0.5% Tween 20 for 3 h and subsequently incubated overnight with lipid-extracted or unextracted avian osteoclast lysates (about 3-5 mg of protein) in PBS-T. 2-Mercaptoethanol and bovine serum albumin were added to a final concentration of 12 mM and 0.1%, respectively. The blot was washed twice in PBS-T for 30 min each time and immunoblotted with anti-gelsolin antibody (1:1000 dilution) as described previously (8). Lysates (unextracted and extracted) were made from two different osteoclast preparations, and binding was performed in duplicate filters for each lysate preparation.

GST Pull-down Assay-- 5 µg of GST fusion proteins noncovalently coupled to Sepharose beads were incubated with 500 µg of lipid-extracted and unextracted lysates for 2 h at 4 °C. After binding, the Sepharose beads were washed four to five times with lysis buffer and washed three times with cold PBS. Bound proteins were boiled with SDS-PAGE sample buffer (60) and subjected to 7.5% SDS-PAGE. The proteins were transferred onto a polyvinylidene difluoride membrane and immunoblotted with anti-gelsolin antibody as described previously (8).

Data Analysis-- All comparisons were made as a percentage of the control, which refers to vehicle-treated cells. The other treatment groups in each experiment were normalized to each control value. Data presented are the mean ± S.E. of experiments done at different times normalized to intraexperimental control values. For statistical comparisons, analysis of variance was used with the Bonferroni corrections (Instat for IBM, version 2.0; Graphpad Software).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Immunofluorescent Staining of Various Proteins in Osteoclasts-- Osteoclasts plated on a glass coverslip demonstrate multiple actin-containing dot-like structures in the area corresponding to the clear zone of osteoclasts (3-5, 10, 61-63), whereas the resorbing osteoclasts demonstrated a band of F-actin-containing podosomes (61, 64). We have shown that OPN binding to the _v3 integrin of osteoclast podosomes stimulated cytoskeletal reorganization and bone resorption by activating a heteromultimeric signaling complex that includes gelsolin, FAK, c-Src, and PI3K (8, 9). The activity of these enzymes was focused in a signaling complex associated with gelsolin (8, 9). To further investigate whether gelsolin interacts with the signaling molecules associated with integrin _v3 of osteoclast podosome (23) and focal adhesion proteins, osteoclasts from chicken cultured on bone slices were immunostained with various antibodies and anti-gelsolin as shown in Fig. 1. Colocalization of gelsolin with _v3 (Fig. 1A) and actin (Fig. 1I) is shown. Actin distribution is seen at the periphery as a ring-like structure at the resorbing area. The formation of actin ring is a prerequisite for the initiation of bone resorption. Osteoclasts stained for gelsolin/actin (Fig. 1I) and actin/_v3 (Fig. 1J) demonstrate actin ring at the resorbing area. Punctate staining of actin and _v3 is seen throughout the cell, showing their colocalization (Fig. 1J). Also, at the resorbing area, colocalization of _v3 (Fig. 1A), FAK (Fig. 1B), PI3K (Fig. 1D), paxillin (Fig. 1E), c-Src (Fig. 1G), and talin (Fig. 1K) with gelsolin is seen. Punctate staining is shown at the basolateral membrane of the resorbing osteoclast stained for gelsolin and _v3 (Fig. 1A). Colocalization of FAK, PI3K, c-Src, and paxillin with gelsolin is also seen at the periphery of the cell. Very little or no colocalization of C-Cbl (data not shown), leupaxin (Fig. 1F), and PYK2 (Fig. 1C) with gelsolin is observable.

View larger version (41K): [in this window] [in a new window]   Fig. 1.   Immunofluorescence localization with antibodies against the indicated proteins and gelsolin in osteoclasts plated on bone. Colocalization of the proteins is seen in yellow. Arrows in the figures point to the position of the pits underneath the osteoclast. Green staining represents gelsolin distribution, except in F, where red staining represents gelsolin distribution. Bar, 25 µm.

Association of Gelsolin with _v3-associated Proteins-- Because immunostaining of osteoclasts showed colocalization of gelsolin with various proteins, we further confirmed the association of these proteins with gelsolin by immunoprecipitation and Western analyses (Fig. 2, A and B). Avian osteoclast lysates were immunoprecipitated with anti-gelsolin antibody and analyzed by Western blot with various antibodies as shown in Fig. 2. In Fig. 2A, Src (lane 1), PI3K (lane 3), FAK (lane 7), vinculin (lane 11), _v, (lane 16) and paxillin (lane 22) are detected in gelsolin immunoprecipitates. Consistent with the immunostaining data (Fig. 1), c-Cbl (data not shown), leupaxin (lane 6), and PYK2 (lane 19) are not detected in the analysis. Coimmunoprecipitation of PYK2 is seen in the PI3K immunoprecipitates (lane 18) but not in gelsolin immunoprecipitates (lane 19). Western analyses with anti-PY20 antibody have demonstrated association of tyrosine-phosphorylated proteins with 125-, 85-, 60-, and 42-44 kDa proteins (lane 13). p130^Cas has been shown to associate with FAK, paxillin, and vinculin (22, 65). Therefore, anti-gelsolin immunoprecipitates made from lysates of osteoclasts treated with PBS or OPN were analyzed by Western blot with a p130^Cas antibody. Whereas the level of gelsolin remained the same (Fig. 2B, lanes 4 and 5), OPN stimulated association of p130^Cas with gelsolin (lane 2) as compared with vehicle-treated controls (lane 1).

View larger version (45K): [in this window] [in a new window]   Fig. 2.   Western analysis of focal adhesion components associated with gelsolin. A, gelsolin was immunoprecipitated from equal amounts of proteins from OPN-treated lysates and analyzed by Western blot with antibodies against the indicated proteins. Proteins that are coprecipitated with gelsolin are indicated to the left of the blot. Immunoprecipitations with antibodies against leupaxin (lane 5), FAK (lane 8), vinculin (lane 11), phosphotyrosine (lane 14), v (lane 15), and PYK2 (lane 20) are used as markers or control. IgG heavy chain (IgGHC) is indicated to the left of the blot. IgG heavy chain is not seen in the immunoblotting of membranes with polyclonal antibodies anti-PI3K (lane 4) or anti-PYK2 (lane 19). B, equal amounts of lysate proteins (200 µg) made from PBS-treated (lanes 1 and 4) or OPN-treated (lanes 2, 3, 5, and 6) osteoclasts were immunoprecipitated with monoclonal gelsolin antibody (lanes 1, 2, 4, and 5) or with nonimmune serum (NI; lanes 3 and 6). The immunoprecipitated protein was divided into two equal aliquots, which were subjected to SDS-PAGE and immunoblotted with either rabbit polyclonal p130Cas (lanes 1-3) or monoclonal gelsolin (lanes 4-6) antibody. The arrows point to the p130Cas (left) and gelsolin (right) proteins.

To further analyze whether the ability of gelsolin to associate with focal adhesion molecules is altered after OPN treatment, immunoprecipitates made with various antibodies as mentioned in Fig. 3A from lysates of osteoclasts treated with PBS or OPN were analyzed by Western blot with an anti-gelsolin antibody. Western analysis of immunoprecipitates made with anti-PI3K (lanes 1 and 2), PY20 (lanes 5 and 6), anti-_v (lanes 11 and 12), anti-FAK (lanes 13 and 14), and anti-PtdIns 4,5-P₂ (lanes 15 and 16) have demonstrated coimmunoprecipitation of gelsolin. The increase in the levels of gelsolin indicates an increase in association of those proteins with gelsolin in OPN-treated cells. The increase in the levels of gelsolin in the PY20 immunoprecipitates also indicates that gelsolin-associated signaling molecules are phosphorylated more upon OPN stimulation (10). Therefore, anti-phosphotyrosine coprecipitates more gelsolin in OPN-treated cells. To confirm the observation of the effect of OPN on the association of signaling proteins with gelsolin and to define the interaction in more detail, osteoclasts cultured on glass coverslips were treated with PBS or OPN and immunostained with various antibodies and gelsolin. We have previously demonstrated that when osteoclasts were treated with OPN, the entire cell subsurface was occupied with podosomes within 15 min. Immunostaining of osteoclasts with anti-gelsolin antibody and rhodamine phalloidin for actin showed that gelsolin was colocalized with actin in podosomes and along the inner edge of actin ring-like structures (10). Here we show that PI3K (red) and gelsolin (green) staining (Fig. 3B) is seen in the multiple rows of podosomes localized in the area corresponding to the clear zone in osteoclasts at the periphery in vehicle-treated cells (Fig. 3B, PBS). Osteopontin treatment caused immunofluorescent staining of gelsolin and PI3K in podosomes on the entire subsurface of the cell (Fig. 3B, OP) similar to the gelsolin/actin staining shown previously (10).

View larger version (71K): [in this window] [in a new window]   Fig. 3.   The effect of osteopontin stimulation on the association of proteins with gelsolin. A, lysates made from PBS ()- or OPN (+)-treated osteoclasts were immunoprecipitated with the antibodies indicated below each lane, subjected to 8% SDS-PAGE, and immunoblotted with a monoclonal gelsolin antibody. The arrow points to the position of gelsolin (left). Although antibodies to leupaxin, FAK, and PtdIns P2 are monoclonal, the affinity to IgG heavy chain varies when lysates are immunoblotted with gelsolin monoclonal antibody. Immunoprecipitation and immunoblotting with gelsolin antibody were used as control (lanes 7 and 8). The results shown are representative of four independent experiments. B, effect of OPN on PI3K distribution in osteoclasts isolated from mice. Osteoclasts plated on glass coverslips were fixed and immunostained with anti-gelsolin (green) or PI3K antibodies after treatment with PBS or OPN. Cells were viewed by confocal microscopy. Colocalization of the proteins is seen in yellow. Bar, 25 µm.

Protein tyrosine phosphatases PTEN and PTP-PEST have been shown to modulate tyrosine phosphorylation of proteins associated with FAK and FAK functions (66-71). Therefore, we sought to determine whether PTEN or PTP-PEST associates with gelsolin. Anti-gelsolin immunoprecipitates were analyzed for the association with PTEN by Western analysis. We failed to show PTEN association with gelsolin.² We then examined gelsolin association with PTP-PEST by Western blot (Fig. 4A) and immunostaining (Fig. 4B) analyses. In Fig. 4A, lysates made from vehicle- and OPN-treated osteoclasts were immunoprecipitated with antibodies to PTP-PEST (lanes 3 and 4) or gelsolin (lanes 1 and 2) and analyzed by Western blot with anti-phosphoserine antibody (lanes 1-4). OPN stimulates the tyrosine phosphorylation of proteins associated with _v3 (23) and gelsolin (10) (Fig. 3, lanes 3 and 4). In Fig. 4A, we demonstrate an increase in serine phosphorylation of PTP-PEST (lanes 2 and 4) in OPN-treated osteoclasts as compared with vehicle-treated cells (lanes 1 and 3). Lanes 5 and 6 demonstrate the levels of PTP-PEST immunoprecipitated. Western analysis of gelsolin immunoprecipitates demonstrates the levels of PTP-PEST coimmunoprecipitated with gelsolin (lanes 7 and 8). These results demonstrate that PTP-PEST is coprecipitated with gelsolin and that its phosphorylation on phosphoserine residue is increased upon OPN treatment (lane 2). Immunostaining analysis of osteoclasts plated on glass coverslips demonstrates colocalization of gelsolin (green) and PTP-PEST (red) in the podosome and membrane of osteoclast (Fig. 4B).

View larger version (62K): [in this window] [in a new window]   Fig. 4.   Analysis of PTP-PEST association with gelsolin. A, equal amounts of lysate proteins (200 µg) made from PBS-treated (lanes 1, 3, 5, and 7) or OPN-treated osteoclasts (lanes 2, 4, 6, and 8) were immunoprecipitated with an antibody to gelsolin (lanes 1, 2, 7, and 8) or PTP-PEST (lanes 3-6). The immunoprecipitated protein was divided into two equal aliquots, which were subjected to SDS-PAGE. One half of the anti-gelsolin (lanes 1 and 2) and anti-PTP-PEST (lanes 3 and 4) immunoprecipitates was immunoblotted with a polyclonal phosphoserine antibody (lanes 1-4), and the other half of the immunoprecipitates was immunoblotted with a polyclonal PTP-PEST antibody (lanes 5-8). The arrows point to the position of PTP-PEST. Treatments are as indicated below the figure. The results represent one of three experiments performed from three separate osteoclast preparations. B, immunolocalization of gelsolin and PTP-PEST in osteoclasts isolated from mice. Osteoclasts plated on glass coverslips were fixed and immunostained with anti-gelsolin (green) or PTP-PEST (red) antibodies. Cells were viewed by confocal microscopy. Colocalization of the proteins is seen in yellow. The arrow indicates the areas shown at higher magnification (right). Left bar, 25 µm; right bar, 10 µm.

Role of Lipids in Interaction of Proteins-- Our results demonstrate that gelsolin is associated with c-Src, PI3K, FAK, p130^Cas, paxillin, talin, and vinculin. The association of these signaling molecules with gelsolin is increased upon OPN stimulation in both avian and mouse osteoclasts (Fig. 3) (8-10). The question of how these proteins interact with gelsolin remains. The primary structure of gelsolin reveals actin and phosphoinositide binding sites (34, 36, 39). We tested the possibility that phosphoinositides associated with gelsolin may mediate the interaction of SH2 domain-containing proteins (52). We used equal amount of proteins from lipid-extracted (Fig. 5, lanes 6-9) and unextracted lysates for immunoprecipitation with various antibodies. Immunoprecipitation of lipid-extracted lysates with anti-PtdIns 4,5-P₂ (lane 8), anti-PI3K (lane 7), and anti-Src (lane 9) fails to show association of gelsolin. There was no change in the levels of gelsolin in lipid-extracted (lane 6) or unextracted (lane 2) lysates. The absence of gelsolin in the anti-PtdIns P₂ antibody immunoprecipitate (lane 8) indicates complete extraction of phosphoinositides associated with gelsolin. These results indicate that the interaction of proteins with gelsolin is mediated through phospholipids associated with gelsolin.

View larger version (50K): [in this window] [in a new window]   Fig. 5.   Analysis of the role of phospholipids in interactions of proteins with gelsolin. Equal amounts of lysate proteins (200 µg) made from OPN-treated osteoclasts were immunoprecipitated with the indicated antibodies before (lanes 1-5) and after lipid extraction (lanes 6-9). The immunoprecipitates were immunoblotted with a monoclonal gelsolin antibody. Gelsolin, IgG heavy chain (IgGHC), and IgG light chain (IgGLC) are marked by arrows. The results represent one of three experiments performed from three separate osteoclast preparations.

Role of Phosphoinositides in Protein Interactions-- The binding of phosphoinositides with SH2 domains of proteins was further defined by using in vitro binding of labeled phosphoinositides with GST fusion proteins containing SH2 and SH3 domains of PI3K or c-Src (52). GST fusion proteins of SH2 and SH3 domains of PI3K, c-Src, and lck (Fig. 7C) and ³²P-labeled phospholipids were purified as described under "Experimental Procedures." 500-1000 cpm of PtdIns P₃ and 1000-2000 cpm of PtdIns P₂, phosphatidylinositol 3,4-bisphosphate, PtdIns 3P, and PtdIns were used for binding studies. We used less PtdIns P₃ because of the relatively low endogenous level of this lipid and the low yield during the extraction procedure. The lipids were dried, resuspended in 10 mM Hepes, pH 7.0, 1 mM EDTA (HE buffer), sonicated, and allowed to bind to the indicated GST fusion proteins coupled to GST-Sepharose for 1.5 h at room temperature in a solution of 0.02% Nonidet P-40-HE buffer containing 100 mM NaCl (HNE-Nonidet P-40 buffer) (52). The Sepharose beads were then washed with 1 ml of HNE-Nonidet P-40 buffer, and the lipids that remained associated with the beads were extracted with chloroform:methanol and resolved by TLC (Fig. 6). As shown in Fig. 6A, PtdIns P₃ binds to GST-gelsolin (lane 1), SH2 domains of p85 (lanes 3 and 4), and full-length p85 (lane 7) proteins with greater affinity than it does to srcSH2 domain protein (lane 5). LckSH2 (lane 6) or p85SH3 (lane 8) domain proteins do not show any binding of PtdIns P₃. Phospholipid spots were scraped from the TLC plates, and radioactivity was determined by liquid scintillation counting. The amount of radioactivity from PtdIns P₃ () and PtdIns P₂ () bound to GST fusion proteins is shown (Fig. 6B). Radioactivity bound to GST protein was used as control. Data from three experiments are shown, and values are expressed as a percentage of the control. The specificity of PtdIns P₃ binding to the p85 NH₂ or COOH SH2 domains (***, p < 0001) is greater than that to the c-Src SH2 domain (*, p < 01). PtdIns P₂ binding was seen with Gst fusion proteins of full length p85 and gelsolin. No such binding was observed with other Gst fusion proteins such as NH₂, COOH terminal SH₂ domains, SH₃ domain of p85, or SH₂ domains of src, and lck. Basal level binding of PtdIns P₂ with p85 SH3, p85, c-Src, and lck SH2 domains was observed. A very low level binding of other phosphoinositides (PtdIns P or PtdIns) with GST fusion proteins was observed.

View larger version (30K): [in this window] [in a new window]   Fig. 6.   Analysis of binding of purified 32P-labeled PtdIns P2 and PtdIns P3 with GST fusion proteins. 32P-labeled PtdIns P2 and PtdIns P3 were allowed to bind to the indicated GST fusion proteins immobilized to Sepharose beads as described under "Experimental Procedures" (52). The bound phosphoinositides were extracted from the GST fusion proteins and analyzed by TLC. The arrow to the left of the panel indicates the migration of the PtdIns P3. The experiment is representative of one of three experiments. B, phospholipid spots were scraped from the TLC plates, and radioactivity was determined by liquid scintillation counting. The binding is expressed as a percentage of control. Data shown are the mean +S.E. from three different experiments. ***, p < 0001; **, p < 001; *, p < 01 (compared with GST control). p85NHSH2, PI3K NH2 SH2 domain; p85CSH2, PI3K COOH SH2 domain.

Overlay Assay-- The lipid-protein interaction was further confirmed by an overlay assay in which a blot of GST fusion proteins was probed with lysates made from OPN-treated osteoclasts for 24-48 h at 4 °C. The bound protein was then detected with anti-gelsolin antibody (Fig. 7). Binding of gelsolin with GST-p85 (full length) (lane 1), GST-p85N-SH2 domain (lane 2), and GST-p85C-SH2 (lane 3) is shown in Fig. 7A. Weak binding to the GST-Src SH2 domain (lane 5) was seen, and no detectable binding was observed with GST alone (lane 7), SH3 domain of p85 (lane 6), and SH2 domain of lck (lane 5). In addition to the above-mentioned observations, the role of phosphoinositides in the protein interaction was further verified by doing the overlay assay with lipid-extracted lysates from OPN-treated osteoclasts (Fig. 7B). Lysates were incubated with immobilized GST/p85 SH2 domains to nitrocellulose membrane. Binding of gelsolin from the unextracted osteoclast lysates (lanes 1 and 2) to GST-p85N-SH2 (lane 2) and GST-p85C-SH2 domains (lane 1) is seen, whereas lipid-extracted lysates failed to show interaction of gelsolin with fusion proteins, although gelsolin was still present in the lipid-extracted lysates (Fig. 6A, lane 6). Coomassie Blue-stained gel containing GST fusion proteins is shown in Fig. 7C. Equal amounts of proteins were used for the overlay and GST pull-down assays after quantitating the protein concentration by scanning the gel and by protein estimation using the Bio-Rad kit.

View larger version (24K): [in this window] [in a new window]   Fig. 7.   Demonstration of binding of lipids associated with gelsolin in osteoclast lysates to the SH2 domains of PI3K by overlay assay. A, GST fusion proteins indicated below the lanes were immobilized to a nitrocellulose membrane and incubated with osteoclast lysates as described under "Experimental Procedures." B, the indicated GST fusion proteins immobilized to a membrane were incubated with unextracted (lanes 1 and 2) or lipid-extracted (lanes 3-5) osteoclast lysates. C, SDS-PAGE and Coomassie Blue staining of the purified GST fusion proteins. D, GST pull-down assay. GST fusion proteins coupled to Sepharose beads were allowed to bind to lipid-extracted or unextracted (lanes 1-4) lysates. After binding, the beads were washed, and the eluted proteins were subjected to SDS-PAGE. The potential role of lipids in the binding of gelsolin with GST fusion proteins is analyzed by immunoblotting the membranes with anti-gelsolin antibody (A, B, and D). Binding of gelsolin from the lipid-extracted lysate to GST fusion proteins was not seen (B, lanes 3-5, and D). Lysates (unextracted and extracted) were made from two different osteoclast preparations, and binding was done in duplicate filters for each lysate preparation. These results are representative of four separate binding experiments.

GST Pull-down Assay-- In addition to the above-mentioned experiments, lipid-extracted and unextracted lysates were incubated with the indicated immobilized GST fusion proteins coupled to Sepharose beads as shown in Fig. 7B for 2 h or overnight at 4 °C. The Sepharose beads were washed, and binding of gelsolin to the GST fusion protein was assessed by immunoblotting with anti-gelsolin antibody. As shown in Fig. 7D, binding of gelsolin from unextracted lysates (as such) with GST-p85 (full length) (lane 1), GST-p85COOH-SH2 domain (lane 2), and GST-p85NH-SH2 domain (lane 3) is seen. Very low-level binding was seen with GST-Src SH2 domain (lane 4). No detectable binding of gelsolin to GST-p85, GST-p85COOH-SH2 domain, or GST-p85NH-SH2 domain was seen from the lipid-extracted osteoclast lysates. Also, no binding was seen with the lckSH2 or p85SH3 domain from unextracted or lipid-extracted lysates. Our results indicate that the NH₂ and COOH SH2 domains of p85 have more affinity to PtdIns P₃ than the c-Src SH2 domain.

Analysis of Binding of Proteins to PtdIns P3-- To investigate whether anti-PtdIns P₃ could coprecipitate the proteins associated with gelsolin as shown in Fig. 2A, lysates made from OPN-treated osteoclasts were immunoprecipitated with anti-PtdIns P₃ and analyzed by Western blot with the indicated antibodies (Fig. 8). Coprecipitation of PI3K (lane 1), c-Src (lane 3), p130^Cas (lane 4), and gelsolin (lane 5) was seen. Western analysis with anti-phosphotyrosine antibody demonstrates coprecipitation of a number of tyrosine-phosphorylated cellular proteins, such as p130^Cas, p125^Fak, p90-100 (an unknown protein), p85^PI3K, p68^paxillin, and p60^c-src. Lane 8 shows a shorter exposure of the blot shown in lane 7. Lanes 2 and 6 are immunoprecipitations with nonimmune serum.

View larger version (20K): [in this window] [in a new window]   Fig. 8.   Analysis of proteins that are coprecipitated by anti-PtdIns P3. Lysates from OPN-treated osteoclasts were immunoprecipitated with anti-PtdIns P3 and analyzed by Western blot with the antibodies indicated below each lane. Arrows to the left of each blot mark proteins that are coprecipitated with anti-PtdIns P3. Lanes 2 and 6 are immunoprecipitation with nonimmune serum. Gsn, gelsolin protein. Lane 8 is a short exposure of the blot shown in lane 7.

The ability of proteins to bind to PtdIns P₃ was further assessed by an overlay assay (Fig. 9). For this assay, PtdIns P₃ (spot 1 in Fig. 9), PtdIns P₂ (spot 2), and PtdIns P (spot 3) were spotted on a nitrocellulose filter as described previously (72, 73). Immunoblotting with anti-PtdIns P₂ and anti-PtdIns P₃ recognizes PtdIns P₂ and PtdIns P₃ spots, respectively (Fig. 9A). To do the overlay assay, the blot was probed with GST fusion proteins as indicated in Fig. 9B after blocking as described under "Experimental Procedures." Strong binding of p85 SH2 fusion proteins to PtdIns P₃ is shown. Low-level binding of PtdIns P₃ is shown with c-Src and lck SH2 domains. Similarly, low-level binding of Src and p85 SH2 domains to PtdIns P₂ was observed.

View larger version (40K): [in this window] [in a new window]   Fig. 9.   Analysis of phospholipid binding properties of different GST fusion proteins. The ability of binding of fusion proteins with lipids was analyzed by lipid-protein overlay assay. Lipids were spotted and overlaid with protein as described under "Experimental Procedures." The numbers at the top represent the following phosphoinositides: 1, PtdIns P2; 2, PtdIns P3; and 3, PtdIns 3P. A, spotted lipids were immunoblotted (IB) with the antibodies indicated to the left of the blot. B, proteins used for the overlay assay are indicated to the left of the blot. After binding overnight, the blots were immunoblotted with an anti-GST antibody to detect the interaction of proteins with the lipids immobilized to the membrane. The results are representative of three separate experiments.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Gelsolin, an actin-binding protein, controls the length of actin filaments in vitro and controls cell shape and motility in vivo by a variety of mechanisms (74). Gelsolin severs assembled actin filaments, caps the fast-growing plus end, and promotes the growth of actin filament by creating nucleation sites (75-77). Binding of gelsolin to phosphoinositides causes the release of gelsolin from the filament end, providing a site for rapid monomer addition (77-79). Thus far, we have demonstrated the following sequence of events to explain how OPN/_v3-mediated signaling stimulates actin filament formation and actin reorganization in osteoclasts: 1) OPN binding to integrin _v3 in osteoclasts stimulates gelsolin-associated phosphoinositides, uncaps barbed end actin, and results in actin filament formation (8), supporting the concept described by Stossel (75) and others (76, 79). Phosphoinositides phosphorylated at 3-hydroxyl position by PI3K associated with gelsolin also participate in actin filament reorganization and podosome formation (8). 2) Elimination of c-Src by antisense oligodeoxynucleotides to c-Src prevents OPN/_v3 activation of gelsolin-associated PI3K (9). 3) PI3K activity is also disrupted by the treatment of cells with C3 transferase, whereas gelsolin-associated c-Src activity is not affected by the same treatment. These results demonstrate that OPN stimulation of c-Src and rho are upstream of PI3K (80). 4) Osteoclasts isolated from Gsn/ mice lack podosomes and fail to respond to OPN with a change in F-actin (10). The other actin capping or bundling proteins cannot substitute gelsolin function in podosome assembly. Our analyses demonstrate the functional role and significance of gelsolin and the associated signaling molecules during OPN-induced osteoclast cytoskeleton organization.

Involvement of signaling molecules in the regulation of cytoskeletal organization in osteoclast function has been widely studied. Targeted disruption of c-Src in mice leads to only one major phenotype: severe osteopetrosis caused by an intrinsic defect in osteoclasts. Although osteoclasts are present in src/ mice in normal or increased numbers, they fail to form ruffled borders and consequently do not resorb bone (81). As mentioned earlier, normal osteoclasts adhered to plastic through podosomes, which are represented by dot-like aggregations of actin, clustering in a ring around the periphery of the cells. Osteoclasts from src/ mice have a markedly different distribution of actin, with no peripheral podosome arrangement (82). PYK2 and p130^Cas were reported to be the key effectors in _v3 integrin-mediated signaling pathways, and their activation requires c-Src in osteoclasts (83, 84). Inhibition of PI3K activity, ruffled border formation, and bone resorption by wortmannin has been reported in osteoclasts by several studies (85-87). Involvement of rho p21 protein in osteoclastic bone resorption has also been demonstrated by using the C3 exoenzyme (88). These observations demonstrate that signaling molecules such as c-Src, PI3K, PYK2, p130^Cas, and rho A are essential for osteoclast function.

Our present observations have demonstrated the interaction of focal adhesion-associated proteins such as FAK, PI3K, c-Src, p130^Cas, paxillin, vinculin, and talin with gelsolin (Figs. 1 and 2). Gelsolin coprecipitates the signaling molecules with different affinity. Although osteoclast gelsolin appears to be central to the localization of tyrosine kinases involved in regulation of the actin cytoskeleton, the exact mechanism of interaction of proteins with gelsolin in the podosome of osteoclast is unknown. Focal adhesions contain many structural proteins such as talin, tensin, and -actinin, as well as tyrosine kinases such as FAK, c-Src, and Csk, that are activated upon extracellular matrix binding. Multimeric protein complexes are then formed, which contain many different adapter proteins, including p130^Cas, Shc, Grb2, Crk, and Nck. These interactions confer an important role to focal adhesions in signal transduction pathways (89, 90). Paxillin has been demonstrated to interact with FAK and vinculin as well as with numerous SH2-containing proteins (91-93). Paxillin binds to many proteins involved in effecting changes in the organization of actin cytoskeleton. Its primary function is as a molecular adapter or scaffold protein that provides multiple docking sites at the plasma membrane for an array of signaling and structural proteins (22). The FAK functions in regulating tyrosine phosphorylation of several proteins, including paxillin, p130^Cas, talin, and tensin (94). FAK is also an integrin-linked tyrosine kinase that associates with Src family kinases and paxillin, and it is thought to regulate focal adhesion assembly and integrin-mediated signaling (19).

We have also shown here that gelsolin is associated with PTP-PEST and that OPN induced a transient phosphorylation of PTP-PEST on serine. PTP-PEST has been shown to play a role in cytoskeleton reorganization by promoting the turnover of focal adhesion required for cell migration (70). Also, a remarkable degree of selectivity for p130^Cas as a substrate both in vitro and in intact cells (68, 96) has been displayed by PTP-PEST. Phosphorylation of PTP-PEST at two major sites (Ser-29 and -439) has been documented in 12-O-tetradecanoylphorbol-13-acetate-treated HeLa cells, and the increase in phosphorylation lowers PTP-PEST affinity for substrates, thereby lowering the phosphatase activity (97). We have previously shown that OPN modulates tyrosine phosphorylation of signaling proteins such as FAK, PI3K, and c-Src associated with _v3 (23) and gelsolin (10). We presume that the increase in serine phosphorylation of PTP-PEST may decrease its affinity to the other signaling proteins associated with _v3 or gelsolin and therefore lead to an increase in tyrosine phosphorylation of proteins. Although there is biochemical evidence suggesting that two focal adhesion proteins, FAK and p130^Cas, are substrates for the other tyrosine phosphatase PTEN (MMAC1) (66), we have failed to detect PTEN in this complex. Based on our observations, we speculate that PTP-PEST may be the potential regulator of tyrosine phosphorylation status of other signaling proteins in the podosomes of osteoclasts.

Although proline-rich tyrosine kinase (PYK2), and leupaxin share similar structural organization with FAK and paxillin, respectively (98-100), they are not found to be colocalized or coprecipitated with gelsolin (Figs. 1 and 2). Western analysis of the gelsolin immunoprecipitates with phosphotyrosine antibody demonstrates 42-44-kDa proteins (Fig. 2A, lane 13), which are closer to the molecular mass of leupaxin. Therefore, we tested for the presence of leupaxin in gelsolin immunoprecipitates, and we failed to detect leupaxin association with gelsolin. PYK2 has been reported to interact with p130^Cas and paxillin in a manner similar to that of FAK. Although FAK and PYK2 are structurally similar, each has the capacity to mediate distinctly different signaling responses (101, 102). FAK activity has been shown to be necessary for focal adhesion turnover (103), and targeting of active c-Src to focal contacts has been shown to promote the turnover of these structures during cell motility (104). FAK appears to be one of the important integration sites of various intracellular signals, including those via integrin receptors providing docking sites for signaling proteins containing SH3 or SH2 domains such as c-Src, PI3K, and p130^Cas (16, 105-107). Recently, it has been shown in c-Src/ osteoclasts that macrophage colony-stimulating factor causes recruitment of ₃ integrin and phospholipase C and induces stable association of ₃ integrin with phospholipase C, PI3K, and PYK2 (108).

Treatment of osteoclasts with antisense oligodeoxynucleotides to c-Src reduced bone resorption as well as OPN/_v3 activation of gelsolin-associated PI3K. We observed only a basal level association of PI3K with gelsolin. However, OPN-induced translocation of PI3K from the Triton-insoluble to the Triton-soluble fraction was not seen in antisense oligodeoxynucleotides to Src-pretreated cells (9). The signaling mechanism in response to various factors is not unique. Selective interactions of proteins are required to mediate the regulatory mechanisms and signaling cascades mediated by integrin _v3 within the osteoclast in response to different factors. Although PTEN,² leupaxin (100), PYK2 (83), and c-Cbl (110, 111) are not found in gelsolin immunoprecipitates, these proteins have been demonstrated in osteoclast function in _v3-mediated signaling mechanisms.

Experiments to understand the mechanism involved in the association of signaling molecules with gelsolin were subsequently pursued. Rameh et al. (52) have demonstrated that the amount of PI3K associated with the tyrosine-phosphorylated proteins inversely correlates with the amount of PI3K lipid products present in the cell. Phosphatidylinositol 3,4,5-trisphosphate was shown to disrupt the association of PI3K with tyrosine-phosphorylated proteins by binding to the SH2 domain of the p85 subunit of PI3K (52). Because SH2 domains of signaling proteins have been shown to interact with phosphoinositides, we attempted to study the biological role of phosphoinositides in the interaction of proteins with gelsolin. The role of lipids in the interaction of proteins was evident from the experiments done with lipid-extracted and unextracted lysates. Overlay and GST pull-down assays confirm the specific interaction of phosphoinositides with the SH2 domains of c-Src and PI3K. More specifically, the interaction of PI3K (p85) SH2 domains with the PI3K lipid product PtdIns P₃ was evident from the overlay assay. We have shown here that the interaction of SH2-containing proteins with gelsolin in the podosome of osteoclast is mediated through phosphoinositide-protein interactions. PI3K SH2 domain binding with PtdIns P₃ is more than the binding of c-Src SH₂. SH2 domains of PI3K and c-Src also bind to PtdIns P₂, albeit more weakly than their binding to PtdIns P₃. This raises the possibility that the initial phospholipid-SH2 domain-mediated interaction of gelsolin with PI3K and the subsequent protein-protein interactions among focal adhesion proteins lead to architectures of complexes formed at the podosomes of osteoclast. This complex may be involved in connecting integrin to actin because colocalization and coprecipitation of integrin _v3 are seen with gelsolin. Overall, these analyses demonstrate that phosphoinositides, through their binding to the modular domains of kinase(s), mediate the targeting of signaling molecules to podosomes of osteoclasts. These observations lead us to propose the model shown in Fig. 10 describing signaling complex formation in osteoclast in response to integrin activation.

View larger version (31K): [in this window] [in a new window]   Fig. 10.   Model of gelsolin and the associated signaling complex. Liganding of integrin v3 with osteopontin stimulates association of signaling molecules with v3 in the podosomes of osteoclasts. One of the actin-capping proteins that was coprecipitated with v3 is gelsolin. OPN stimulated association of PtdIns 4,5-P2 and PtdIns P3 with gelsolin. Lipid binding studies with GST fusion proteins demonstrated binding of PtdIns P2 with gelsolin and PI3K. Gelsolin-associated PtdIns 4,5-P2 is the substrate for PI3K, and recognition of this phosphoinositide by PI3K leads to its translocation from actin to gelsolin in the Triton-soluble fraction. Activation of lipid second messenger pathway triggered by integrin has been shown to modulate the interaction of cytoskeletal proteins. Talin, vinculin, and -actinin are among the main integrin-associated cytoskeletal molecules that bind to or are modulated by phosphoinositides (48-51). OPN/v3-mediated signaling is mediated through the recruitment of focal adhesion proteins, such as FAK, p130Cas, talin, vinculin, paxillin, c-Src, and PI3K, at the cytoplasmic domains of the integrins. Gelsolin is one of the proteins in the integrin-associated complex in podosomes of osteoclast. Also, our observations demonstrate here that gelsolin interaction with the signaling protein(s) is initially mediated through the novel phosphoinositide-protein interaction. The association of PTP-PEST in this complex and regulation of its phosphorylation status by OPN may help to regulate the phosphorylation and recruitment of the other signaling proteins with integin in osteoclasts.

The association of SH2-containing proteins with phosphotyrosine target molecules is phosphorylation-dependent. This interaction occurs in a sequence-specific manner (24, 28). Several SH2-containing proteins are recruited to the receptor complex in response to stimulation. Among the proteins that bind to the platelet-derived growth factor receptor are c-Src, PI3K, Nck, GAP, Syp, and phospholipase C (24). A number of signaling proteins were identified to associate with gelsolin. Antibody to PtdIns P₃ coprecipitates several signaling proteins, most of which contain SH2 domain- and SH3 domain-containing proteins. Songyang et al. (112, 113) introduced the use of phosphotyrosine-containing peptides and chromatographically identified peptide families that bind to particular SH2 domains. Tyrosine phosphorylation of paxillin on a YXXP motif results in the formation of a high-affinity binding site for the SH2-SH3-containing linker proteins (114, 115). Synthetic phosphopeptide has been shown to inhibit SH2 binding to phosphoinositides (52) or receptor binding (116).

A variety of chemical, physical, and biological approaches to the intracellular delivery of peptides into living cells have been investigated (reviewed in Ref. 117). Hall et al. (109) introduced phospholipase C-SH2 domain-binding phosphopeptides into cells, which were able to inhibit inositol phosphate accumulation and neurite outgrowth stimulated by fibroblast growth factor. Similarly, Rojas et al. (95) inhibited epidermal growth factor-induced Ras activation with signal sequence-based internalization of a Grb2 binding site phosphopeptide at Tyr-1068 of the epidermal growth factor receptor. In our competition assays, we observed that phosphotyrosine peptide YMPM (which binds specifically to p85 SH₂ domains) competed with phosphoinositide binding. We did not observe any competition with the unphosphorylated YMPM peptide or peptide specific to Src SH2 domains (pYEEI).³

After binding of osteoclasts through _v3 to proteins such as OPN, podosomes are assembled in a band at the periphery (Fig. 1, I and J) in an area referred to as the clear zone. Cellular organization for bone resorption is associated with a transition of the osteoclast expressing the band of podosomes to one expressing dense actin ring. When osteoclasts are stimulated to move by, possibly through secretion of the autocrine OPN, the actin filaments of the podosomes and actin rings are severed. Uncapping of barbed end actin by phosphoinsoitides promotes new actin filament formation. Changes in the cytoskeletal organization lead to the development of cellular processes as the cell migrates. Osteoclasts isolated from gelsolin-deficient mice lack the adhesion structures called podosomes. Integrin _v3-associated signaling molecules, other actin-capping/binding proteins, albeit compromised adhesion but motility and bone resorption only to a lesser extent in osteoclasts from gelsolin-null mice. Osteoclast podosome assembly/disassembly is crucial for migration, and the formation of podosome stabilizes the bone matrix-cell interface and forms an isolated compartment between the ruffled border and the bone surface. Because osteoclasts from gelsolin-null mice are devoid of podosomes and failed to respond to OPN with a change in F-actin or functions of osteoclast, we conclude that remodeling of the cytoskeleton and assembly/disassembly of podosomes are dependent on gelsolin and the associated signaling molecules. We have shown here that PI3K is present in the podosomes of osteoclasts and colocalized with gelsolin. The product of PI3K, PtdIns P₃, is crucial for the protein interactions at the podosome area. OPN/_v3 mediated increased phosphorylation of signaling, and cytoskeletal proteins regulate the ability of these proteins to physically interact with each other through specific domains (Fig. 10), thereby modulating the cytoskeleton and assembly/disassembly of podosomes, the critical processes for osteoclast functions. Studies using specific phosphotyrosine peptide strongly suggest that phosphoinositides recruit SH2-containing proteins and that the binding requires pTyr-pocket. Additional experiments on the delivery of phosphopeptides specific for SH2 domains of various proteins into osteoclasts will unravel the question of which specific protein directly interacts with gelsolin to bring about podosome turnover.

The findings of novel PtdIns P₃-based mechanisms of protein complex formation around the actin-capping/severing protein gelsolin are relatively unique to the osteoclasts and cells expressing podosomes. Podosomes are unique cell adhesion structures associated with high rates of motility and rapid rates of assembly/disassembly. The findings of PtdIns P₃-based scaffolding and enzymatic regulation of tyrosine phosphorylation reported herein provide molecular insights into the specific nature of podosome operation. Cells expressing focal adhesions are characterized by lower rates of motility due to different and slower mechanisms of structural assembly/disassembly. Furthermore, the osteoclast appears to be uniquely dependent on gelsolin as the actin-severing and podosome actin-capping protein (10).

	ACKNOWLEDGEMENTS

We thank Dr. David Cheresh (Scripps Clinic and Research Foundation, La Jolla, CA) for valuable discussions, Dr. John Freeman (Department of Orthopedics, Barnes-Jewish Hospital, St. Louis, MO) for assistance with the confocal microscopy, Dr. Arasu Chellaiah for proofreading the manuscript, and Suzanne Swanson for technical assistance.

	FOOTNOTES

^* This work was supported by National Institutes of Health Grants AR46292 (to M. A. C.), AR41677 (to K. A. H.), DK09976 (to K. A. H.) and by a grant (DRIF) from the University of Maryland Dental School, Baltimore, MD.The costs of publication of this article were defrayed in part by the payment of page charges. The 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: Dept. of Oral and Craniofacial Biological Sciences, University of Maryland, 666 W. Baltimore St., Baltimore, MD 21201-1586. Tel.: 410-706-2083; Fax: 410-706-0193; E-mail: mac001@dental.umaryland.edu.

Published, JBC Papers in Press, September 27, 2001, DOI 10.1074/jbc.M107494200

² Soga, N., Chellaiah, M. and Hruska, K. A., unpublished observations.

³ M. A. Chellaiah and K. A. Hruska, unpublished data.

	ABBREVIATIONS

The abbreviations used are: OPN, osteopontin; FAK, focal adhesion kinase; SH, Src homology; GST, glutathione S-transferase; PtdIns 4, 5-P₂, phosphatidylinositol 4,5-bisphosphate; PTP, protein tyrosine phosphatase; PI3K, phosphoinositide 3-kinase; PtdIns P₂, phosphatidylinositol 3,4,5-trisphosphate; PtdIns P, phosphatidylinositol phosphate; PEST, proline glutamic acid-serine-threonine amino acid sequences; RGD, Arg-Gly-Asp cell adhesion sequences. PTEN, phosphatase and tensin homolog.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				G. H. Li, Y. Shi, Y. Chen, M. Sun, S. Sader, Y. Maekawa, S. Arab, F. Dawood, M. Chen, G. De Couto, et al. Gelsolin Regulates Cardiac Remodeling After Myocardial Infarction Through DNase I-Mediated Apoptosis Circ. Res., April 10, 2009; 104(7): 896 - 904. [Abstract] [Full Text] [PDF]

				M. A. Chellaiah, D. Kuppuswamy, L. Lasky, and S. Linder Phosphorylation of a Wiscott-Aldrich Syndrome Protein-associated Signal Complex Is Critical in Osteoclast Bone Resorption J. Biol. Chem., March 30, 2007; 282(13): 10104 - 10116. [Abstract] [Full Text] [PDF]

				M. Halle, Y.-C. Liu, S. Hardy, J.-F. Theberge, C. Blanchetot, A. Bourdeau, T.-C. Meng, and M. L. Tremblay Caspase-3 Regulates Catalytic Activity and Scaffolding Functions of the Protein Tyrosine Phosphatase PEST, a Novel Modulator of the Apoptotic Response Mol. Cell. Biol., February 1, 2007; 27(3): 1172 - 1190. [Abstract] [Full Text] [PDF]

				K. Yogo, M. Mizutamari, K. Mishima, H. Takenouchi, N. Ishida-Kitagawa, T. Sasaki, and T. Takeya Src Homology 2 (SH2)-Containing 5'-Inositol Phosphatase Localizes to Podosomes, and the SH2 Domain Is Implicated in the Attenuation of Bone Resorption in Osteoclasts Endocrinology, July 1, 2006; 147(7): 3307 - 3317. [Abstract] [Full Text] [PDF]

				C. Varon, F. Tatin, V. Moreau, E. Van Obberghen-Schilling, S. Fernandez-Sauze, E. Reuzeau, I. Kramer, and E. Genot Transforming Growth Factor {beta} Induces Rosettes of Podosomes in Primary Aortic Endothelial Cells. Mol. Cell. Biol., May 1, 2006; 26(9): 3582 - 3594. [Abstract] [Full Text] [PDF]

				A. P. Hall, F. R. Westwood, and P. F. Wadsworth Review of the Effects of Anti-Angiogenic Compounds on the Epiphyseal Growth Plate Toxicol Pathol, February 1, 2006; 34(2): 131 - 147. [Abstract] [Full Text] [PDF]

				M. A. Chellaiah Regulation of Actin Ring Formation by Rho GTPases in Osteoclasts J. Biol. Chem., September 23, 2005; 280(38): 32930 - 32943. [Abstract] [Full Text] [PDF]

				S. Linder and P. Kopp Podosomes at a glance J. Cell Sci., May 15, 2005; 118(10): 2079 - 2082. [Full Text] [PDF]

				C. S. Moreno, S. Ramachandran, D. G. Ashby, N. Laycock, C. A. Plattner, W. Chen, W. C. Hahn, and D. C. Pallas Signaling and Transcriptional Changes Critical for Transformation of Human Cells by Simian Virus 40 Small Tumor Antigen or Protein Phosphatase 2A B56{gamma} Knockdown Cancer Res., October 1, 2004; 64(19): 6978 - 6988. [Abstract] [Full Text] [PDF]

				M. K.Y. Siu and C. Y. Cheng Extracellular Matrix: Recent Advances on Its Role in Junction Dynamics in the Seminiferous Epithelium During Spermatogenesis Biol Reprod, August 1, 2004; 71(2): 375 - 391. [Abstract] [Full Text] [PDF]

				Y. Calle, G. E. Jones, C. Jagger, K. Fuller, M. P. Blundell, J. Chow, T. Chambers, and A. J. Thrasher WASp deficiency in mice results in failure to form osteoclast sealing zones and defects in bone resorption Blood, May 1, 2004; 103(9): 3552 - 3561. [Abstract] [Full Text] [PDF]

				S.-H. Chen, M. R. Bubb, E. G. Yarmola, J. Zuo, J. Jiang, B. S. Lee, M. Lu, S. L. Gluck, I. R. Hurst, and L. S. Holliday Vacuolar H+-ATPase Binding to Microfilaments: REGULATION IN RESPONSE TO PHOSPHATIDYLINOSITOL 3-KINASE ACTIVITY AND DETAILED CHARACTERIZATION OF THE ACTIN-BINDING SITE IN SUBUNIT B J. Biol. Chem., February 27, 2004; 279(9): 7988 - 7998. [Abstract] [Full Text] [PDF]

				P. D. Arora, M. Glogauer, A. Kapus, D. J. Kwiatkowski, and C. A. McCulloch Gelsolin Mediates Collagen Phagocytosis through a Rac-dependent Step Mol. Biol. Cell, February 1, 2004; 15(2): 588 - 599. [Abstract] [Full Text] [PDF]

				I. Kaverina, T. E. B. Stradal, and M. Gimona Podosome formation in cultured A7r5 vascular smooth muscle cells requires Arp2/3-dependent de-novo actin polymerization at discrete microdomains J. Cell Sci., December 15, 2003; 116(24): 4915 - 4924. [Abstract] [Full Text] [PDF]

				M. A. Chellaiah, R. S. Biswas, S. R. Rittling, D. T. Denhardt, and K. A. Hruska Rho-dependent Rho Kinase Activation Increases CD44 Surface Expression and Bone Resorption in Osteoclasts J. Biol. Chem., August 1, 2003; 278(31): 29086 - 29097. [Abstract] [Full Text] [PDF]

				M. A. Khadeer, Z. Tang, H. S. Tenenhouse, M. V. Eiden, H. Murer, N. Hernando, E. J. Weinman, M. A. Chellaiah, and A. Gupta Na+-dependent phosphate transporters in the murine osteoclast: cellular distribution and protein interactions Am J Physiol Cell Physiol, June 1, 2003; 284(6): C1633 - C1644. [Abstract] [Full Text] [PDF]

				P. M. Becker, A. A. Kazi, R. Wadgaonkar, D. B. Pearse, D. Kwiatkowski, and J. G. N. Garcia Pulmonary Vascular Permeability and Ischemic Injury in Gelsolin-Deficient Mice Am. J. Respir. Cell Mol. Biol., April 1, 2003; 28(4): 478 - 484. [Abstract] [Full Text] [PDF]

				K. D. Niswender, B. Gallis, J. E. Blevins, M. A. Corson, M. W. Schwartz, and D. G. Baskin Immunocytochemical Detection of Phosphatidylinositol 3-kinase Activation by Insulin and Leptin J. Histochem. Cytochem., March 1, 2003; 51(3): 275 - 283. [Abstract] [Full Text] [PDF]

				T. Sugatani, U. Alvarez, and K. A. Hruska PTEN Regulates RANKL- and Osteopontin-stimulated Signal Transduction during Osteoclast Differentiation and Cell Motility J. Biol. Chem., February 7, 2003; 278(7): 5001 - 5008. [Abstract] [Full Text] [PDF]

				M. A. Chellaiah, N. Kizer, R. Biswas, U. Alvarez, J. Strauss-Schoenberger, L. Rifas, S. R. Rittling, D. T. Denhardt, and K. A. Hruska Osteopontin Deficiency Produces Osteoclast Dysfunction Due to Reduced CD44 Surface Expression Mol. Biol. Cell, January 1, 2003; 14(1): 173 - 189. [Abstract] [Full Text]

This Article Abstract Full Text (PDF) All Versions of this Article: 276/50/47434    most recent M107494200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Chellaiah, M. A. Articles by Hruska, K. A. Search for Related Content PubMed PubMed Citation Articles by Chellaiah, M. A. Articles by Hruska, K. A. Social Bookmarking                      What's this?