P2Y6 Nucleotide Receptors Activate NF-κB and Increase Survival of Osteoclasts*

Nucleotides, released from cells during inflammation and by mechanical stimulation, act through the P2 family of nucleotide receptors. Previous studies have demonstrated the expression of P2Y1 and P2Y2 receptors in osteoclasts. The aim of this study was to determine whether osteoclast P2Y receptors signal through NF-κB, a key transcription factor regulating osteoclastogenesis. Immunofluorescence was used to detect the p65 subunit of NF-κB, which upon activation translocates from the cytosol to nuclei. Low levels of NF-κB activation were observed in untreated rabbit osteoclasts and in those exposed to 2-methylthio ADP (P2Y1 agonist) or ATP or UTP (P2Y2 agonists). In contrast, UDP or INS48823 (P2Y6 agonists) induced a significant increase in the number of cells exhibiting NF-κB activation, a process sensitive to the proteasome inhibitor lactacystin. In osteoclasts purified by micromanipulation, reverse transcription-PCR revealed the presence of P2Y1, P2Y2, and P2Y6 receptor transcripts, and application of agonists for these receptors induced the transient rise of cytosolic calcium. Treatment of rat osteoclasts with UDP or INS48823, but not 2-methylthio ADP or UTP, increased osteoclast survival. Osteoprotegerin (a decoy receptor for RANK ligand) did not significantly alter the effects of UDP on NF-κB localization or osteoclast survival, consistent with a direct action. Moreover, SN50 (cell-permeable peptide inhibitor of NF-κB) suppressed the enhancement of cell survival induced by UDP and INS48823. Our findings demonstrate the presence of functional P2Y6 receptors in osteoclasts. Thus, nucleotides, following their release at sites of inflammation and mechanical stimulation, can act through P2Y6 receptors to initiate NF-κB signaling and enhance osteoclast survival.

Nucleotides, such as ATP and UTP, are signaling molecules that mediate diverse biological effects following their release into the extracellular fluid. In trauma, nucleotides are released from damaged cells and activated platelets (1). ATP is also released from nerve terminals in the peripheral and central nervous systems (2,3). Mechanical stimulation or fluid shear stress induces the release of nucleotides including ATP and UTP in a variety of non-neuronal cells (4 -6). In the extracellular environment, ATP and UTP are degraded by a family of enzymes collectively called ecto-nucleotidases (7). ATP is degraded to ADP, AMP, and adenosine, whereas UTP is degraded to UDP, UMP, and uridine. ATP, UTP, as well as some of their degradation products are biologically active.
Nucleotides act through the P2 family of cell surface receptors. P2Y receptors are G protein-coupled receptors that, in many cases, signal through the release of Ca 2ϩ from intracellular stores, whereas P2X receptors are ligand-gated cation channels (8). To date, seven mammalian P2X receptors, P2X1-7, and at least eight P2Y receptors, P2Y1, 2, 4, 6, and 11-14, have been identified. P2X receptors are activated by adenine nucleotides, whereas P2Y receptors are activated by adenine and/or uridine nucleotides. P2Y1 receptors are activated selectively by ADP (9). The P2Y2 receptor is activated by ATP and UTP with equal potency, and the human P2Y4 receptor is activated selectively by UTP, whereas at the rat P2Y4 receptor ATP and UTP are equipotent. The P2Y6 receptor is activated selectively by UDP (8).
Osteoclasts are multinucleated cells with the unique ability to resorb bone. Together with bone forming osteoblasts, they regulate bone mass. A variety of endocrine and paracrine factors control osteoclast formation and life span. Until recently, it was believed that the major determinants of bone resorption were osteoclast proliferation and activation. However, growing evidence has led to the proposal that the control of the osteoclast life span is critical for the regulation of bone resorption (10,11). Receptor activator of nuclear factor B (RANK) 1 ligand, a protein expressed by osteoblasts and activated T cells, acts directly on osteoclasts to stimulate their formation and activity (12). RANK ligand signals by binding to its receptor RANK, a member of the tumor necrosis factor receptor superfamily. The soluble, decoy receptor osteoprotegerin (OPG) also binds RANK ligand, preventing signaling. Multiple downstream transcription factors are activated during RANK signaling, including activator protein 1 and NF-B (13). Inactive NF-B resides in the cytoplasm bound to the inhibitory protein IB. Phosphorylation of IB leads to its ubiquitination and degradation by the proteasome. IB degradation permits NF-B to translocate to the nucleus where it regulates the expression of a variety of genes involved in inflammation and immunity, cell proliferation, responses to stress, and apoptosis (14). The essential role of NF-B in osteoclast formation was demonstrated in mice deficient for both the p50 and p52 sub-units of NF-B (15). These mice showed a pronounced osteopetrotic phenotype as a result of the absence of osteoclasts.
Previous studies have shown that low concentrations of ATP stimulate the resorptive activity and formation of rodent osteoclasts (16). This effect was attributed to the P2Y1 receptor, because ADP and a selective P2Y1 agonist, 2MeSADP, stimulated osteoclast formation and resorption (17). Signaling through many P2Y receptors involves activation of phospholipase C (PLC) leading to the formation of inositol 1,4,5trisphosphate and release of Ca 2ϩ from intracellular stores (8). In osteoclasts, several P2Y receptor agonists were shown to induce elevation of cytosolic free calcium concentration ([Ca 2ϩ ] i ) through this pathway (18). In addition, immunocytochemistry and electrophysiological studies have shown the presence of P2Y2, P2X4, and P2X7 receptors in osteoclasts (19 -21). Our previous work demonstrated that P2X7 receptors couple to NF-B signaling in osteoclasts (22); however, it was unknown whether P2Y receptors activate NF-B as well. It is also unknown whether other P2Y receptors in addition to P2Y1 and P2Y2 are expressed in authentic osteoclasts. The aim of this study was to examine which P2Y receptors are expressed in osteoclasts and to determine whether they signal through NF-B. Our results show that in addition to previously demonstrated P2Y1 and P2Y2 receptors, osteoclasts express functional P2Y6 receptors. We also establish that P2Y6 receptors couple to activation of NF-B and promote osteoclast survival. 199 (containing 25 mM HEPES and 26 mM  HCO 3 Ϫ ), HCO 3 Ϫ -free Medium 199 (25 mM HEPES), heat-inactivated fetal bovine serum, and antibiotic solution (penicillin 10,000 units/ml; streptomycin 10,000 g/ml; amphotericin B 25 g/ml) were from Invitrogen. Nucleotides were purchased from Sigma. Mouse monoclonal antibody against the p65 subunit of NF-B was from Santa Cruz Biotechnology (Santa Cruz, CA; catalog number sc-8008). Mounting medium (Vecta-Shield) and biotinylated goat anti-mouse IgG were from Vector Laboratories (Burlingame, CA). Fluorescein-conjugated streptavidin and fluorescent probes TOTO-3 and fura-2-AM were from Molecular Probes (Eugene, OR). The P2Y6-selective agonist INS48823 was a generous gift from Inspire Pharmaceuticals (Durham, NC). The cell-permeable peptide inhibitor of NF-B, SN50, the inactive control peptide, SN50M, and the irreversible proteasome inhibitor, lactacystin, were from Calbiochem (La Jolla, CA). RANK ligand (recombinant protein consisting of human residues 151-316 fused at the N terminus to a linker peptide and a FLAG tag) was purchased from Alexis Corp. (San Diego, CA), and OPG (recombinant protein consisting of human residues 21-194) was from Research Diagnostics Inc. (Flanders, NJ).

Materials-Medium
Osteoclast Isolation and Culture-Osteoclasts were isolated from the long bones of neonatal New Zealand White rabbits and Wistar rats according to previously described procedures (21). Rabbit and rat long bones were dissected free of soft tissues and cut with a scalpel to release bone fragments into 2-3 ml of osteoclast culture medium, which consisted of Medium 199 buffered with HEPES and HCO 3 Ϫ supplemented with 15% fetal bovine serum and 1% antibiotic solution. The cells were suspended by repeated passage through a pipette and plated on glass coverslips. The rabbit bone cell preparations were maintained at 37°C, 5% CO 2 for 2 h after isolation, and then fresh culture medium was added. The cells were incubated at 37°C, 5% CO 2 for 2-7 days before use. The majority of nonosteoclastic cells were removed using Pronase E (0.001% in PBS with 0.5 mM EDTA) at room temperature (22-25°C) with intermittent agitation. When the majority of nonosteoclastic cells had been removed (1-5 min), the Pronase E solution was replaced with fresh medium. Samples partially purified with Pronase E were used for immunofluorescence studies and calcium measurements. To obtain homogeneous populations of osteoclasts for RNA isolation, some samples were further purified as follows. The bottom and edge of coverslips were cleaned with cotton swabs to remove adherent cells and were then placed in medium. Osteoclasts were identified by phase contrast microscopy as having three or more nuclei. Cells with fewer than three nuclei were removed using a micromanipulator and janitor pipette (20), leaving pure preparations of 20 -80 adherent multinucleated osteoclasts/coverslip.
Following isolation, rat osteoclasts were incubated at 37°C in 5% CO 2 for 1 h, then gently washed with PBS to remove nonadherent cells, and incubated in fresh culture medium for 1-3 h before use. To quantify survival, the numbers of rat osteoclasts on coverslips were determined using a phase contrast microscope at time 0 (time of addition of test substances) and at 18 h. Osteoclasts were identified as having three or more nuclei. The number of surviving osteoclasts was expressed as a percentage of the number of osteoclasts at time 0 (100% ϭ 98 Ϯ 3 osteoclasts/coverslip, mean Ϯ S.E., n ϭ 278). These procedures were approved by the Council on Animal Care of the University of Western Ontario.
Determination of NF-B Localization by Immunofluorescence-Following 5-6 days in culture, rabbit osteoclasts on glass coverslips were incubated with test agents in osteoclast culture medium at 37°C and, at the indicated times, washed in PBS, fixed with 4% paraformaldehyde (10 min), washed in PBS (2 ϫ 10 min), permeabilized with 0.1% Triton X-100 in PBS (10 min), washed in PBS (2 ϫ 5 min), and blocked with 1% normal goat serum in PBS (blocking solution) for 1-2 h at room temperature. Osteoclasts were incubated overnight at 4°C with primary antibody to the p65 subunit of NF-B (diluted 1:100 in blocking solution). The coverslips were then washed and incubated for 2 h at room temperature with biotinylated secondary antibody diluted 1:100 in blocking solution, followed by washing and incubation with fluoresceinconjugated streptavidin (1:100 in PBS) for 30 min at room temperature. The nuclei were stained with TOTO-3 (2 M), the coverslips were washed and mounted on slides, and the cells were observed using a Zeiss LSM 510 laser-scanning confocal microscope. All of the osteoclasts on each coverslip were examined (usually 150 -200 osteoclasts/coverslip). The osteoclasts were rated positive for nuclear localization of NF-B if the fluorescence intensity of three or more nuclei exceeded that of the cytoplasm.
Total RNA was isolated from coverslips containing purified rabbit osteoclasts or bone cells (containing stromal cells and cells of both osteoblast and osteoclast lineages) using the RNeasy mini kit (Qiagen). Genomic DNA was removed by incubation with 0.05-0.1 unit/ml DNase I (amplification grade; Invitrogen) at 25°C for 15 min followed by heat inactivation at 65°C for 10 min. Typically, the RNA from a single coverslip was divided into two equal fractions. One fraction was not subjected to reverse transcription to detect genomic DNA contamination, and the second fraction was reverse transcribed into singlestranded cDNA using Superscript II (Invitrogen) according to the manufacturer's instructions. A portion (20 -25%) of the resulting cDNA (or control sample that did not undergo reverse transcription) was used as template in each PCR performed with AdvanTaq Plus polymerase (Clontech Laboratories, Palo Alto, CA). Each 25-l reaction contained 1 ϫ AdvanTaq Plus PCR buffer, 0.2 mM of each dNTP, 0.2 mM primer, and 10% dimethyl sulfoxide (no dimethyl sulfoxide was used for PCR of P2Y1). PCR was performed with a hot start at 95°C for 2 min, followed by 35-36 cycles at 94°C for 30 s, 64°C for 30 s, and 68°C for 1 min, with a final extension at 68°C for 3 min. PCR using nested primers was carried out on first round PCR products using the protocol described above. PCR products were separated on a 1% agarose gel containing 0.5 g/ml ethidium bromide and detected under UV illumination. Identities of amplified products were verified by sequencing of selected samples.
Fluorescence Measurement of Cytosolic Free Ca 2ϩ Concentration-[Ca 2ϩ ] i of single osteoclasts loaded with fura-2 was monitored using microfluorimetric techniques. Cells on glass coverslips were incubated for 40 min at room temperature in HCO 3 Ϫ -free osteoclast culture medium containing 1.5 M fura-2-AM. The coverslips were then placed in a chamber mounted on the stage of a Nikon Diaphot inverted phase contrast microscope and bathed in HCO 3 Ϫ -free osteoclast culture me-dium supplemented with 15% fetal bovine serum and 1% antibiotic solution. The ratio of fluorescence emission at 510 nm with alternate excitation wavelengths of 345 and 380 nm was measured using a Deltascan illumination system (Photon Technology International, London, Canada) as described previously (18). Test substances were dissolved in the HCO 3 Ϫ -free osteoclast culture medium and applied locally to cells by pressure ejection from a micropipette.
Statistical Analyses-Data are expressed as means Ϯ S.E. The sample size (n) indicates the number of osteoclasts for Ca 2ϩ fluorescence determinations or the number of separate cell preparations for immunofluorescence and survival studies. Unless otherwise indicated, the data were analyzed by one-way analysis of variance followed by a Bonferroni's post test. The differences were accepted as statistically significant at p Ͻ 0.05.

UDP Induces NF-B Translocation in Rabbit Osteoclasts-
Activation of NF-B results in its translocation from the cytoplasm to the nuclei. Our previous work demonstrated that immunofluorescence is an effective tool for examining NF-B activation in rabbit osteoclasts (22). To determine whether P2Y receptors activate NF-B, rabbit osteoclasts were exposed to various P2Y receptor agonists. UDP (10 M), an agonist at P2Y6 receptors, increased the proportion of osteoclasts displaying nuclear localization of NF-B with a maximal effect at 3 h of exposure (Fig. 1, A and D). Control osteoclasts showed low levels of nuclear NF-B for up to 4 h (Fig. 1B). The extent of NF-B translocation was dependent on UDP concentration, with maximal effects observed at 10 -100 M UDP (Fig. 1E). In contrast to its action on osteoclasts, UDP did not induce NF-B translocation in rabbit bone marrow stromal cells. After 3 h of exposure, 15 Ϯ 7% of vehicle-treated and 16 Ϯ 8% of UDPtreated stromal cells ( Fig. 1C) showed nuclear localization of NF-B (n ϭ 3, p Ͼ 0.05). Thus, UDP induces activation of NF-B in rabbit osteoclasts but not bone marrow stromal cells.
To examine whether NF-B activation occurs through P2Y receptors other than P2Y6, osteoclasts were exposed to 2Me-SADP (P2Y1 receptor agonist) or UTP (P2Y2 receptor agonist) for 3 h, and the effect was compared with that of UDP (Fig. 2). In these experiments, osteoclasts exposed to 2MeSADP (10 M) or UTP (10 M) showed low levels of nuclear localization of NF-B. Similarly, our previous results showed predominantly cytoplasmic localization of NF-B in osteoclasts exposed to a low concentration of ATP (10 M, a concentration sufficient to activate multiple P2 receptors including P2Y2 and P2X4 but not P2X7) (22). In contrast, treatment with UDP (10 M) significantly increased the percentage of osteoclasts with nuclear localization of NF-B (Fig. 2, p Ͻ 0.05). Therefore, the P2Y6 receptor agonist UDP, but not agonists at other P2Y receptors, activates NF-B in rabbit osteoclasts.
The canonical pathway for NF-B activation involves proteasome-mediated IB degradation. To investigate the mechanism of UDP-induced NF-B activation, we examined the effects of the specific proteasome inhibitor, lactacystin (23). Lactacystin (10 M) abolished UDP-induced NF-B activation (Fig. 2B), consistent with involvement of the classical pathway for NF-B activation.
Expression of P2Y Receptors in Rabbit Osteoclasts-The presence of P2Y1 and P2Y2 receptors as well as P2X4 and P2X7 receptors has been previously shown in osteoclasts (17)(18)(19)(20)(21). To examine whether additional P2Y receptors are expressed on  osteoclasts, reverse transcription-PCR was performed on heterogeneous populations of bone marrow cells (bone cells) as well as on osteoclasts purified by micromanipulation (osteoclasts). Bone marrow cells, containing cells of both osteoblast and osteoclast lineages, possessed transcripts for P2Y1, P2Y2, P2Y6, and P2Y11 receptors and ALP (used as a marker for cells of the osteoblast lineage) (Fig. 3Ai). In purified osteoclasts, expression of P2Y1 and P2Y2 receptors was confirmed. Moreover, the presence of P2Y6 receptor transcripts and the absence of P2Y11 transcripts were shown for the first time (Fig. 3Aii).
P2Y1 and P2Y2 receptor agonists cause elevation of [Ca 2ϩ ] i in osteoclasts (18). We next examined whether P2Y6 receptors in osteoclasts also couple to Ca 2ϩ signaling. Application of 2MeSADP ( The Role of P2Y6 Receptors in Osteoclast Survival-The physiological roles of P2Y6 receptors are poorly understood. In 1321N1 human astrocytes, activation of P2Y6 receptors prevents apoptosis induced by tumor necrosis factor ␣ (25), suggesting that the P2Y6 receptor regulates cell survival. We assessed the effects of nucleotides on osteoclast survival using isolated rat osteoclasts, which spontaneously undergo apoptosis over a period of 24 h in culture. Osteoclasts treated with 2MeSADP (10 M) or UTP (10 M) had survival rates that were not significantly different from control (Fig. 4A). In contrast, RANK ligand (10 ng/ml) and UDP (10 M) promoted osteoclast survival (Fig. 4A). Significant enhancement of osteoclast survival was observed at UDP concentrations of at least 3 M (Fig. 4B).
Possible Role of RANK Ligand in P2Y6 Receptor-mediated NF-B Activation and Enhancement of Osteoclast Survival-Cells isolated from rabbit long bones include marrow stromal cells and cells of the osteoblast lineage. Osteoblasts express multiple P2Y receptors, including P2Y6 (26). Thus, extracellular nucleotides may act on osteoblasts to up-regulate expression of RANK ligand, which could then bind to its receptor RANK on osteoclasts activating NF-B and increasing survival. To examine this possibility, osteoclast preparations were treated with the decoy receptor for RANK ligand, OPG. OPG did not affect basal levels of NF-B activation. Moreover, treatment with OPG (100 ng/ml) for 20 min prior to and during 3 h of incubation with UDP (10 M) did not prevent NF-B activation (Fig. 5Ai). On the other hand, pretreatment with OPG significantly inhibited NF-B activation induced by exogenous RANK ligand (100 ng/ml, 30 min), demonstrating the effectiveness of OPG (Fig. 5Aii). In keeping with these findings, treatment with OPG (100 ng/ml) did not significantly alter osteoclast survival under basal conditions or in the presence of UDP (10 M, 18 h) (Fig. 5B). In contrast, OPG abolished the increase in survival induced by exogenous RANK ligand (10

FIG. 3. Osteoclasts express P2Y1, P2Y2, and P2Y6 nucleotide receptors, which induce transient elevation of [Ca 2؉ ] i . A, cells
were isolated from the long bones of neonatal rabbits and cultured for 2-4 days on glass coverslips. Some samples were then washed, and RNA was isolated (Bone cells). In other samples, osteoclasts were purified using Pronase E, followed by removal of nonosteoclasts using micropipettes prior to RNA isolation (Osteoclasts). The samples were divided into those that did (ϩ) or did not (Ϫ) undergo reverse transcription. Nested PCR was then carried out to detect mRNA transcripts for P2Y receptors or ALP, an osteoblast marker. P2Y1, P2Y2, and P2Y6 receptor transcripts were detected in preparations of both bone cells and purified osteoclasts. In contrast, expression of P2Y11 receptor was detected only in bone cell preparations. ng/ml) (Fig. 5B). Thus, UDP induces nuclear translocation of NF-B and enhances osteoclast survival independently of RANK ligand.
The Role of NF-B Activation in Osteoclast Survival-NF-B plays an important role in controlling survival of a number of cell types (14). To examine whether the effect of UDP on osteoclast survival is mediated through activation of NF-B, we used the cell-permeable peptide inhibitor of NF-B, SN50 (27). Treatment of rabbit osteoclasts with SN50 (20 M) 20 min prior to and during incubation with UDP (10 M) abolished the effect of UDP on NF-B translocation (Fig. 6, A and B), establishing the effectiveness of SN50 as an inhibitor of NF-B activation in osteoclasts. We next assessed the effects of SN50 on survival of rat osteoclasts. SN50 (20 M) significantly suppressed the enhancement of survival induced by UDP, whereas it had no effect on the survival of control osteoclasts not treated with nucleotide (Fig. 6C). Thus, the effect of UDP on osteoclast survival is mediated at least in part through the activation of NF-B.
Characterization of the Effects of the P2Y6-selective Agonist, INS48823-INS48823 is a stable analog of diuridine 5Јtriphosphate, a selective agonist at the P2Y6 receptor (28). The selectivity of INS48823 has been assessed previously based on calcium mobilization in 1321N1 astrocytoma cell lines infected with a retrovirus encoding different P2Y receptors. The EC 50 for INS48823 at the P2Y6 receptor was similar to that of UDP (ϳ125 nM). INS48823 had no appreciable activity at P2Y1, P2Y2, or P2Y4 receptors. 2 In the present study, INS48823 was found to induce transient rise of [Ca 2ϩ ] i in fura-2-loaded rabbit and rat osteoclasts (Fig. 7A). Moreover, treatment of rabbit osteoclasts for 3 h with INS48823 (1 M) induced significant nuclear translocation of NF-B (Fig. 7B). This effect was abol-2 J. G. Douglass, III and C. C. Redick, unpublished observations. The cells treated with UDP showed significant increase in nuclear translocation of NF-B that was not inhibited by OPG (*, p Ͻ 0.05 for the effect of UDP). Aii, osteoclasts were treated with OPG (100 ng/ml) or vehicle for 20 min prior to and during 30 min of exposure to RANK ligand (100 ng/ml). As expected, OPG significantly inhibited RANK ligand-induced nuclear translocation of NF-B (#, p Ͻ 0.05 for the effect of OPG). The data are the means Ϯ S.E., n ϭ three or four separate experiments. B, isolated rat osteoclasts were treated with OPG (100 ng/ml) or its vehicle for 20 min prior to and during 18 h of incubation with or without UDP (10 M) or RANK ligand (10 ng/ml). UDP and RANK ligand significantly enhanced rat osteoclast survival (*, p Ͻ 0.05 for the effect of UDP or RANKL compared with corresponding control). The effect of UDP was not significantly inhibited by OPG. As expected, OPG significantly inhibited the effect of RANK ligand on osteoclast survival (#, p Ͻ 0.05 for the effect of OPG, n ϭ four to seven separate experiments). ished by the NF-B inhibitor, SN50 (20 M). In contrast, the inactive control peptide, SN50M (20 M) (27), had no significant effect on the response to INS48823. INS48823 also enhanced osteoclast survival, even at concentrations as low as 1 M (Fig. 7C). Furthermore, the increase in osteoclast survival induced by INS48823 was abolished by SN50, but not by SN50M (Fig. 7D). These data indicate that P2Y6 receptors signal through NF-B to enhance osteoclast survival. DISCUSSION There is growing evidence that extracellular nucleotides play an important role in the regulation of bone remodeling (29 -32). However, little is known about the role of individual nucleotide receptors in osteoclasts and osteoblasts. This study showed the expression of functional P2Y6 receptors on osteoclasts that couple to NF-B signaling, leading to the enhancement of cell survival.
The presence of functional P2X4, P2X7, P2Y1, and P2Y2 receptors was previously demonstrated on osteoclasts (18 -21). We have confirmed expression of P2Y1 and P2Y2 receptors and shown that authentic osteoclasts express P2Y6 receptors. Using reverse transcription-PCR, it was demonstrated previously that osteoclast-like cells derived in vitro from human peripheral blood monocytes express a wide range of P2X (P2X1, 4, 5, 6, and 7) and P2Y (P2Y1, 2, 4, 6, and 11) receptors (33). Because these results were obtained from a mixed population of cells, it is unclear whether P2 receptor subtypes were expressed on osteoclast-like cells or other cell types present in the cultures. Nevertheless, our findings indicate the presence of P2Y6 and the absence of P2Y11 receptors in authentic rabbit osteoclasts. In addition, the P2Y6 receptor agonist UDP induced transient Ca 2ϩ elevation in rabbit and rat osteoclasts, consistent with the expression of functional P2Y6 receptors. Responses to INS48823 provided further evidence for the presence of functional P2Y6 receptors. Moreover, the dependence of NF-B translocation and osteoclast survival on UDP concentration observed in the present study is consistent with the concentration dependence reported for P2Y6 receptor-mediated responses in other systems (e.g. 34,35). In the present study, concentrations of INS48823 as low as 1 M stimulated osteoclast survival, perhaps reflecting its greater stability compared with UDP, which is hydrolyzed rapidly by ectonucleotidases (35).
ATP has been shown to stimulate bone resorption in vitro; however, the receptor mediating this effect was not initially identified (16). Subsequent studies demonstrated that the P2Y1 receptor mediates the effect of ATP and its breakdown product ADP on bone resorption (17). This effect involved increased resorptive activity as well as enhanced osteoclast formation. However, P2Y receptors are expressed on osteoclasts and on osteoblasts; thus it is not clear whether these actions are mediated by P2Y1 receptors on osteoclasts or on osteoblasts. Some have suggested that ATP stimulates osteoclast activity indirectly by inducing expression of RANK ligand by osteoblasts (33). Bone cell preparations used in our studies contain cells from the osteoclast and osteoblast lineages. Thus, UDP could act on osteoblasts to induce expression of RANK ligand, which could then bind to osteoclasts to activate NF-B and promote their survival (13). However, treatment of our cell preparations with the decoy receptor for RANK ligand, OPG, did not significantly inhibit the effect of UDP on either NF-B activation or osteoclast survival. These findings are consistent with UDP acting directly through stimulation of P2Y6 receptors on osteoclasts.
By demonstrating that P2Y6 receptors enhance osteoclast survival, we have identified a new role for nucleotide receptors in osteoclast physiology. Enhancement of cell survival through P2Y6 receptors is consistent with data from 1321N1 human astrocytes, in which activation of P2Y6 receptors prevented tumor necrosis factor ␣-induced apoptosis (25). Because the predominant form of cell death for osteoclasts in vitro and in vivo is apoptosis, it is likely that activation of P2Y6 receptors initiate anti-apoptotic signals that enhance the life span of osteoclasts. A change in the rate of osteoclast apoptosis is thought to be an important mechanism regulating bone resorption in vivo (10,11). Therefore, nucleotides released into the extracellular fluid might act on P2Y6 receptors to enhance osteoclast survival and thus increase bone resorption.
Little is known about signals activated by the P2Y6 receptor. Many P2Y receptors, including P2Y6, act through pertussis toxin-insensitive G q to activate PLC␤ leading to formation of inositol 1,4,5-trisphosphate and subsequent release of Ca 2ϩ from intracellular stores (8). In J774 macrophages, P2Y6 receptors potentiate lipopolysaccharide-induced NF-B activation (36). We have shown previously that P2X7 receptors activate NF-B in osteoclasts (22). In the present study, we show that P2Y6 receptors on osteoclasts induce transient increase of [Ca 2ϩ ] i and translocation of NF-B through a proteasome-dependent mechanism. Like P2Y6, other G protein-coupled receptors that activate NF-B (such as receptors for thrombin, platelet-activating factor, endothelin, and bradykinin) couple through G q to activation of PLC␤ (37). Cytosolic Ca 2ϩ , released through the PLC cascade, can regulate the activity of a variety of transcription factors including NF-B (24,38). Activation of P2Y1, P2Y2, and P2Y6 receptors in osteoclasts induces transient rise in [Ca 2ϩ ] i ; however, only the P2Y6 receptor couples to NF-B activation, indicating that a rise in [Ca 2ϩ ] i alone is not sufficient to activate NF-B in osteoclasts. In keeping with these findings, the anti-apoptotic effect of P2Y6 receptor activation in 1321N1 astrocytoma cells was not affected by chelation of intracellular Ca 2ϩ (39). In some systems, G proteincoupled receptors couple to NF-B through the sequential activation of conventional protein kinase C isoforms and IB kinase, leading to IB degradation by the proteasome (37). It is possible that P2Y6 receptor acts through protein kinase C or alternatively through phosphatidylinositol 3-kinase to activate NF-B in osteoclasts. However, determining the precise pathway through which P2Y6 activates NF-B will require use of other cell systems, because of difficulties in isolating osteoclasts in sufficient number and purity to carry out conventional biochemical assays.
NF-B is essential for osteoclast development (15,40). Recent evidence also suggests that NF-B is involved in the bone destruction observed in inflammatory arthritis (41). In addition to its role in inflammation, NF-B contributes to the regulation of immune responses, cell proliferation, and apoptosis. The importance of NF-B in the suppression of apoptosis was discovered in p65 knock-out mice, which die in utero from extensive liver apoptosis (42). Suppression of apoptosis by NF-B is dependent on several pathways. These include induction of cellular inhibitors of apoptosis, caspase-8-c-FLIP, and tumor necrosis factor receptor-associated factors 1 and 2, which are able to inhibit apoptosis initiated through death receptors as well as mitochondrial pathways (14). Our results show that activation of NF-B through P2Y6 receptor signaling enhances survival of osteoclasts.
In vivo, nucleotides such as ATP and UTP are released into the extracellular environment of bone during processes such as inflammation and mechanical stimulation. Upon release, ATP and UTP can activate P2Y2 receptors and subsequently break down to form ADP and UDP. ADP can then act on P2Y1 receptors to stimulate bone resorption, and UDP can activate P2Y6 receptors. Initiation of NF-B signaling through P2Y6 receptors would enhance osteoclast survival, further increasing bone resorption. In pathological states such as rheumatoid arthritis and periodontitis, activation of P2Y6 receptors could contribute to increased bone destruction. Therefore, P2Y6 receptor antagonists might be useful therapeutically for the treatment of inflammatory bone diseases.