PTEN Regulates RANKL- and Osteopontin-stimulated Signal Transduction during Osteoclast Differentiation and Cell Motility*

PTEN (also known as MMAC-1 or TEP-1) is a frequently mutated tumor suppressor gene in human cancer. PTEN functions have been identified in the regulation of cell survival, growth, adhesion, migration, and invasiveness. Here, we characterize the diverse signaling networks modulated by PTEN in osteoclast precursors stimulated by RANKL and osteopontin (OPN). RANKL dose-dependently stimulated transient activation of Akt before activation of PTEN, consistent with a role for PTEN in decreasing Akt activity. PTEN overexpression blocked RANKL-activated Akt stimulated survival and osteopontin-stimulated cell migration while a dominant-negative PTEN increased the actions of RANKL and OPN. PTEN overexpression suppressed RANKL-mediated osteoclast differentiation and OPN-stimulated cell migration. The PTEN dominant-negative constitutively induced osteoclast differentiation and cell migration. Our data demonstrate multiple roles for PTEN in RANKL-induced osteoclast differentiation and OPN-stimulated cell migration in RAW 264.7 osteoclast precursors.

The major function of PTEN appears to be down-regulation of the PI3K product PtdIns(3,4,5)P 3 , which regulates Akt and complex downstream pathways affecting cell growth, survival, and migration. In addition, PTEN has weak protein tyrosine phosphatase activity, which may target focal adhesion kinase (FAK) and Shc, and thereby modulate other complex pathways (1,7). In this report, we show that PTEN regulates the RANKL-activated Akt survival signaling pathway and the OPN-stimulated cell migration in RAW 264.7 osteoclast precursors. Moreover, we found that PTEN also regulates RANKL-induced osteoclast differentiation from RAW 264.7 osteoclast precursors. In addition, we suggest that RANKL may regulate balance of activated Akt and activated PTEN and have influence osteoclast differentiation. Thus, it is likely that PTEN plays multiple roles involving osteoclast formation, survival, and migration. Mouse bone marrow macrophages (BMMs) were prepared from the femur and tibia of 4 -6-week-old C57BL/6 mice and incubated in tissue culture dishes (100-mm dishes) in the presence of recombinant mouse macrophage-colony-stimulating factor (20 ng/ml). After 24 h in culture, the non-adherent cells were collected and layered on Histopaque gradient, and the cells at the gradient interface were collected. The cells were replated (60-mm dishes) at 65,000/cm 2 in ␣MEM, supplement with 10% heat-inactivated FBS in the presence of M-CSF (100 ng/ml). After 3 days in culture, cells were harvested for immunoblotting.
Preparation of Cell Lysates and Immunoblotting-24 -36 h after transfection, medium was removed, and cells were washed two times with phosphate-buffered saline (PBS) and then cultured in DMEM serum-free medium for 24 h. For RANKL (100 ng/ml) or OPN (25 g/ml) stimulation experiments, RANKL or OPN were added to the culture medium and incubated for 5, 10, 15, 30, and 60 min. The cells were then washed once with ice-cold PBS and lysed in a cell lysis buffer (New England Biolabs) to prepare whole cell lysates. Lysates were clarified by centrifugation at 14,000 ϫ g for 10 min, and protein concentrations in the supernatants were measured using the Bio-Rad protein assay reagent kit (Bio-Rad). Proteins were resolved by SDS-PAGE, electroblotted to polyvinylidene difluoride membrane (Millipore, Bedford, MA), blocked in 5% skim milk, 1ϫ PBS, 0.05% Tween 20, and probed with primary antibodies. For the detection of NFB/p50, nuclear extracts were used instead of whole cell lysates. Following incubation with horseradish peroxidase-conjugated goat anti-rabbit antibody (New England Biolabs), bound immunoglobulins were detected using enhanced chemiluminescence (Pierce).
Apoptosis Assay-After cotransfected, cells were treated with RANKL (100 ng/ml). Then, whole cell lysates were prepared as above. Lysates were clarified by centrifugation at 14,000 ϫ g for 10 min, and the supernatant fractions were harvested. Caspase-3 activity assay of cell extracts were measured using a kit (CaspACE TM assay system; Promega) according to the manufacturer's instructions.
Osteoclast Formation Assay-Cells were cultured in a 60-mm dish (40 ϫ 10 4 cells/5-ml dish) in DMEM containing 10% FBS overnight. Cells were then transfected with five expression vectors, respectively. After 24 h, media was removed, and cells were washed two times with PBS and then cultured in the above-mentioned medium with RANKL (100 ng/ml). After culturing for 2 days, cells were added to fresh medium and RANKL. Then, after culturing for 2 days, cells were fixed and stained for TRAP (30). TRAP-positive multinucleated cells (MNCs) containing more than three nuclei were counted as osteoclasts under microscopic examination (31).
Affinity Precipitation of Cellular Rho-The cells (RAW cells or RAW cells with GFP-PTEN⅐WT) were washed once with ice-cold PBS and lysed in Rho-binding lysis buffer (Upstate Biotechnology) to prepare whole cell lysates. Lysates were clarified by centrifugation at 14,000 ϫ g for 10 min, and equal volumes of lysates were incubated with the Rhotekin Rho binding domain (20 g; Upstate Biotechnology) beads at 4°C for 45 min. The beads were washed three times with wash buffer (Tris buffer containing 1% Triton X, 150 mM NaCl, 10 mM MgCl 2 , 10 g/ml each of leupeptin and aprotinin, and 0.1 mM phenylmethylsulfonyl fluoride). Bound Rho proteins were detected by immunoblotting using polyclonal anti-Rho.
Protein Purification-Constitutively active Rho (V14Rho), Rac (L61Rac), and cdc42 (V12cdc42) were cloned in-frame into a bacterial expression vector, pTAT-HA, to produce TAT fusion proteins. The vector pTAT-HA has an N-terminal His 6 leader followed by the 11-amino acid TAT protein transduction domain flanked by glycine residues, a hemagglutinin (HA) tag, and polylinker. The cDNAs encoding V14Rho, L61Rac, and V12cdc42 were cloned into the Tat-HA plasmid. High copy number plasmids were obtained by transformation of the pTAT-HA with V14Rho, L61Rac, and V12cdc42 in BL21. The purification protocol was adapted from the published procedure using a Ni-NTA column (32,33). Briefly, bacterial pellets were resuspended in a buffer containing 100 mM NaCl, 20 mM Hepes (pH 8.0), and 8 M urea and sonicated and centrifuged at 12,000 rpm for 10 min at 4°C. Imidazole was added to the supernatant to a final concentration of 10 -20 mM and purified in the Ni-NTA column as described (32,33). Addition of 8 M urea to the sonication buffer allows for the isolation of insoluble protein in bacterial inclusion bodies and efficient transduction into cells. Bound proteins were eluted with stepwise addition of 5-10 ml each of 100, 250, and 500 mM and 1 M imidazole in the above buffer. Urea was removed by rapid display by using the Slide-A-Lyzer cassette (Pierce) or by the use of desalting PD-10 columns (Sephadex G-25; Amersham Biosciences).
Migration Assay-Cell migration assays were performed using transwell migration chambers (Corning Inc., Corning, NY) (10). Membranes with a pore size of 8 m (Corning Inc.) were coated with OPN (25 ng/ml) at 4°C overnight (haptotaxis) and dried under air. Approximately 5 ϫ 10 4 cells (RAW cells with or without GFP, GFP-PTEN WT, GFP-PTEN C124A, Myc, or Myc-Akt K179M) were added to the upper chamber in DMEM containing 1% FBS and 2% bovine serum albumin (100 l) and allowed to adhere for 1-2 h. After cells with GFP-PTEN (WT) attached to the membrane, TAT fusion protein was added to a final concentration of 100 nM in the upper chamber in the above medium (100 l). Substrates such as OPN (25 g/ml) were added to the lower chamber in DMEM containing 1% FBS and 2% bovine serum albumin (600 l; chemotaxis). The cells were allowed to migrate for 12-14 h at 37°C in a tissue culture incubator with 5% CO 2 . After the incubation period, nonmigrated cells on the upper side on the membrane were removed with a cotton swab. Wells were fixed with an alcohol/formaldehyde/ acetic acid mixture (20:2:1) for 15 min. Filters were stained with hematoxylin stain (Sigma), rinsed well with water, and dried. Dried filters were cut out and mounted with permount solution (Thomas Scientific, Swedesboro, NJ) on a glass slide. Cells were viewed under a ϫ40 objective in an inverted microscope and counted (Zeiss microscope). Data are presented as the number of cell-migrated fields (mean Ϯ S.D.), and all assays were performed in triplicate.
PTEN Activity Assay-For the generation of whole cell lysates prepared from cells with or without RANKL (100 ng/ml) treatment, cells were washed once with ice-cold PBS and lysed in a cell lysis buffer (New England Biolabs) to prepare whole cell lysates. The lysates were centrifuged and supernatants analyzed using PTEN malachite green assay kit (Upstate Biotechnology) according to the manufacturer's instructions.

Expression of Akt Dominant-negative in RAW 264.7 Osteoclast Precursors Induces Apoptosis, but Does Not Influence
Osteoclast Differentiation-Recent studies have demonstrated that Akt regulates apoptosis at multiple sites and identified direct Akt targets including Bad, caspase 9, the forkhead family of transcription factors, and the NFB regulator IKK, each of which plays a critical role in mediating cell death (8,9,39,40). We showed that Myc-Akt (K179M) dominant-negative delayed nuclear translocation of NFB (Fig. 1E) compared with Myc-expressing cells and induced apoptosis (Fig. 2). In contrast, TRAP-positive MNC numbers were not influenced by Myc-Akt (K179M) dominant-negative-expressing cells (Fig. 3). These data indicate that the suppression of osteoclast differentiation may be attributed to a decreased number of RAW 264.7 osteoclast precursors by GFP-PTEN (WT)-induced apoptosis and that delayed nuclear translocation of NFB did not affect osteoclast production. Indeed, GFP-PTEN (WT) more strongly induced apoptosis, compared with the Myc-Akt (K179M) dominant-negative (Fig. 2).
Expression of PTEN and Akt in RAW 264.7 Osteoclast Precursors Regulates OPN-stimulated Migration-PTEN suppresses migration of a variety of cell types, including primary human fibroblasts, non-transformed mouse fibroblasts, and tumor cells (47,49). PTEN-null mouse fibroblasts also show enhanced rates of migration, which are reduced by reintroduction of PTEN (48). Since the role of PTEN in OPN-activated signaling pathway has not been studied, we investigated whether PTEN and Akt regulate OPN-stimulated migration in RAW 264.7 osteoclast precursors using migration assays. We have previously demonstrated that OPN stimulates osteoclast migration (10). RAW 264.7 osteoclast precursor migration was also strongly stimulated by OPN (Fig. 5A). GFP-PTEN (WT) and Myc-Akt (K189M) dominant-negative suppressed OPN-stimulated cell migration in haptotaxis and chemotaxis assays (Fig. 5,  B and C). In contrast, GFP-PTEN (C124A) mutant stimulated OPN-stimulated cell migration in chemotaxis assays, but not significantly in haptotaxis assays (Fig. 5, B and C). These data support the hypothesis that OPN-activated Rac and cell migration were suppressed by GFP-PTEN (WT) and Myc-Akt (K189M) dominant-negative in RAW 264.7 osteoclast precursors.

Rac Rescues the Suppression of Migration by GFP-PTEN (WT) in RAW 264.7 Osteoclast
Precursors-Rho and Rac are essential for osteoclast migration (10,41,44,45). Rho acts upstream of PI3K in osteoclasts (10). Rac also acts upstream of PI3K in breast carcinoma epithelial cells (43), whereas Rac acts downstream of Akt in endothelial cells (50). Therefore, the Rac signaling pathway in osteoclast migration is not clear. We found that Rac is downstream of Akt in RAW 264.7 osteoclast precursors since GFP-PTEN (WT) and Myc-Akt (K179M) dominant-negative blocked Rac phosphorylation (Fig. 4C). To further examine this issue, we performed migration assays following transduction of constitutive active Rho, Rac, and cdc42 in GFP-PTEN (WT)-transfected Raw cells. Constitutively active Rac, but not Rho and cdc42, rescued the suppression of migration in GFP-PTEN (WT)-expressing cells (Fig. 5D). We conclude that Rac acts downstream of Akt, and PTEN has influence on cell migration by Rac, not Rho, in RAW 264.7 osteoclast precursors.
RANKL May Regulate PTEN Activity in RAW 264.7 Osteoclast Precursors and BMMs-Little is known about modes of PTEN regulation. Recent studies have reported that PTEN transcription is regulated by p53 in immortalized mouse embryonic fibroblasts (51). However, it is not clear how PTEN activity is regulated in osteoclast precursors. To determine whether RANKL regulates PTEN activity, we examined expression of PTEN phosphorylation and its activity in RANKLtreated RAW 264.7 osteoclast precursors and BMMs. RANKL activated Akt (RAW: 5 min, BMMs: 5 and 10 min) and suppressed PTEN activity (RAW: 5 min, BMMs: 5 and 10 min) (Fig. 6). After activation by RANKL, activated Akt gradually decreased (Fig. 6, A and C). In contrast, PTEN activity was stimulated by RANKL at 15 and 30 min (Fig. 6, B and D). PTEN also activated by RANKL at 10 min (peak 15 min) in RAW 264.7 osteoclast precursors and 30 min in BMMs (Fig. 6,  A and C). These data indicate that RANKL directly or indirectly may regulate PTEN activity in RAW 264.7 osteoclast precursors and BMMs. DISCUSSION RANKL produced by osteoblastic lineage cells and activated T lymphocytes is an essential factor for osteoclast differentiation, fusion, activation, and survival, thus resulting in bone resorption and bone loss (13)(14)(15)(16)(17)(18). We demonstrate that RANKL activates Akt, which is essential for cell survival in osteoclasts in agreement with others (17, 19 -21). Although RANKL is known to stimulate phosphorylation of a serine residue at position 473 in Akt (Ser-473), we have shown stimulation of threonine phosphorylation at position 308 in Akt (Thr-308) (data not shown) in agreement with Wong et al. (19). However, RANKL did not stimulate Akt (Ser-473) phosphorylation (data not shown) in our hands except in GFP-PTEN (C124A) mutantexpressing cells (data not shown). Akt requires phosphorylation at positions Ser-473 and Thr-308 for full activation (8,9,39). Therefore, the mechanism of RANKL-stimulated activation of Akt demonstrated here is not yet clear.
Bad is a Bcl-2 family member regulated through phosphorylation at Ser-136 by activated Akt resulting in its inactivation and cell survival (9). We demonstrated that RANKL phosphorylates Bad, and the GFP-PTEN (C124A) mutant enhances the effect of RANKL, whereas the RANKL effect is negatively regulated by GFP-PTEN (WT). These results indicate that PTEN activity down-regulates both RANKL-activated Akt and Bad phosphorylation in RAW 264.7 osteoclast precursors, regulating RANKL-stimulated survival signals.
We demonstrated that GFP-PTEN (WT) negatively regulates RANKL-activated Akt survival signaling and suppresses osteoclast differentiation of RAW 264.7 osteoclast precursors. Mice deficient in NFB develop severe osteoporosis because of failed osteoclastogenesis (37). Therefore, NFB is essential for osteoclasts differentiation. GFP-PTEN (WT) and Myc-Akt (K179M) dominant-negative delayed nuclear translocation of NFB compared with GFP-expressing cells. However, the Myc-Akt (K179M) dominant-negative failed to suppress osteoclast differentiation in RAW 264.7 osteoclast precursors. We suggest that the number of RAW 264.7 osteoclast precursors were decreased by GFP-PTEN (WT) because of its strong induction of apoptosis compared with Myc-Akt (K179M) dominant-negative-expressing cells. Moreover, GFP-PTEN (WT) may directly suppress RANKL-activated TRAF 6/NFB signaling pathway, but how it does this is unclear (Fig. 7). In contrast, the GFP-PTEN (C124A) mutant enhanced RANKL-activated Akt survival signaling pathway and markedly induced nuclear translocation of NFB. It also stimulated osteoclast differentiation. Our data suggest that PTEN regulates the RANKLactivated Akt survival signaling pathway and RANKLinduced osteoclast differentiation in RAW 264.7 osteoclast precursors (Fig. 7).
OPN has been shown to stimulate macrophage and osteoclast migration (10,24 mediates cell migration by the PI3K/Akt pathway activation. However, adhesion to OPN did not support ␣ v ␤ 3 -mediated cell migration or activate the PI3K/Akt pathway (11). We have demonstrated that OPN stimulates production of phosphoinositides, including phosphatidylinositol trisphosphate in osteoclasts and activates gelsolin-associated PI3K (10,26). Here, we have demonstrated that OPN stimulates cell migration via the Akt pathway activation including Rac, and the effect is regulated by PTEN in RAW 264.7 osteoclast precursors (Fig. 7). Moreover, we demonstrated that Rac acts downstream of Akt. Because, GFP-PTEN (WT) negatively regulated OPN-activated Rac, but not activated Rho. In addition, L61Rac, but not V14Rho, rescued cell migration in GFP-PTEN-expressing cells.
It has been reported that PTEN inhibits cell migration (1)(2)(3)(4)(5)7). One recently identified mechanism is through its effects on PtdIns (3, 4, 5) P3 levels, which have downstream effects on Rac and cdc42 signaling (48). However, tyrosine-phosphorylated FAK can bind to and activate PI3K (52), and a contributing pathway affecting PtdIns (3, 4, 5) P3 through FAK and PI3K has also been identified (53). This is a second mechanism by which PTEN may inhibit phosphotyrosine-based signaling pathways that has proven useful for dissecting the signaling pathways that regulate cell migration, although they do not prove that PTEN normally regulates these pathways. In the studies reported here, we found that PTEN inhibition of OPNstimulated cell migration by PTEN through negative regulation of OPN-activated Akt and Rac, but not Rho.
As shown in Fig. 7, it is likely that PTEN plays multiple roles in RAW 264.7 osteoclast precursors. One question that remains is how PTEN activity is regulated. There is very little information on the regulation of PTEN expression, localization, or activity. In this study, we demonstrate that one possibility is RANKL regulation of PTEN activity. RANKL activates the PI3K/Akt survival signaling pathway and osteoclast differentiation, which PTEN also regulates (Fig. 7). Our data support that RANKL may regulate the balance between activated Akt and PTEN, which influences osteoclast differentiation.
In summary, we provide the first evidence that PTEN regulates the RANKL-and OPN-activated signaling pathways. Furthermore, PTEN activity influences osteoclast differentiation, survival, and migration in osteoclast precursors. The molecular target of PTEN is PtdIns(3,4,5)P 3 (28), but whether PTEN directly or indirectly regulates the RANKL-activated TRAF 6/NFB signaling pathway is not yet clear (Fig. 7).