Receptor Activator of NF-κB (RANK) Cytoplasmic Motif, 369PFQEP373, Plays a Predominant Role in Osteoclast Survival in Part by Activating Akt/PKB and Its Downstream Effector AFX/FOXO4*

Receptor activator of NF-κB ligand (RANKL) plays a crucial role in osteoclast differentiation, function, and survival. RANKL exerts its effect by activating its receptor RANK (receptor activator of NF-κB), which recruits various intracellular signaling molecules via specific motifs in its cytoplasmic tail. Previously, we identified three RANK cytoplasmic motifs (Motif 1, 369PFQEP373; Motif 2, 559PVQEET564; and Motif 3, 604PVQEQG609) mediating osteoclast formation and function. Here, we investigated RANK cytoplasmic motifs involved in osteoclast survival. Motif 1, in contrast to its minimal role in osteoclast formation and function, plays a predominant role in promoting osteoclast survival. Moreover, whereas Motif 2 and Motif 3 are highly potent in osteoclast formation and function, they exert a moderate effect on osteoclast survival. We also investigated the role of these motifs in activating Akt/protein kinase B (PKB), which has been implicated in RANKL-induced osteoclast survival. Motif 1, but not Motif 2 or Motif 3, is able to stimulate Akt/PKB activation. Because Akt/PKB has been shown to utilize distinct downstream effectors (glycogen synthase kinase-3β, FKHR/FOXO1a, BAD, and AFX/FOXO4) to regulate cell survival, we next determined which downstream effector(s) is activated by Akt/PKB to promote osteoclast survival. Our data revealed that RANKL only stimulates AFX/FOXO4 phosphorylation, indicating that AFX/FOXO4 is a key downstream target activated by Akt/PKB to modulate osteoclast survival. Taken together, we conclude that Motif 1 plays a predominant role in mediating osteoclast survival in part by activating Akt/PKB and its downstream effector AFX/FOXO4.

ber of the tumor necrosis factor (TNF) superfamily and plays a pivotal role in the differentiation and function of osteoclasts, the principal bone-resorbing cells (1)(2)(3)(4)(5). RANKL regulates osteoclast formation and function by binding to its receptor RANK expressed on osteoclast precursors and mature osteoclasts (1,6,7). RANKL also has a decoy receptor, osteoprotegerin, which inhibits RANKL function by competing with RANK for binding RANKL (8,9).
Moreover, it has also been established that RANKL plays an important role in stimulating osteoclast survival (10 -12). Consistent with the notion, osteoprotegerin, the decoy receptor for RANKL, inhibits osteoclast survival (13,14). RANKL exerts its anti-apoptotic effects in part by activating Akt/PKB through a signaling complex involving TNF receptor-associated factor 6 (TRAF6) and c-Src (12). In this model, RANKL induces TRAF6, c-Src, and RANK to form a trimer. Within this trimeric complex, TRAF6 induces the activation of c-Src, which in turn activates phosphotidylinositol-3-kinase (PI3-kinase). The activated PI3kinase then leads to the activation of Akt/PKB. It has become clear that Akt/PKB may utilize distinct downstream pathways (GSK3␤, FKHR/ FOXO1a, BAD, and AFX/FOXO4) to regulate cell survival (15)(16)(17). However, the precise downstream signaling pathway activated by Akt/ PKB to regulate osteoclast survival has remained elusive.
RANK belongs to the TNF receptor (TNFR) family (3), and members of the TNFR family are characterized by lack of intrinsic enzymatic activity. As such, the TNFR family members transduce intracellular signals primarily by recruiting various adaptor proteins such as TRAF proteins through the specific motifs in their cytoplasmic domains (18 -20). Recently, using a chimeric receptor approach, we identified three RANK cytoplasmic motifs (Motif 1, 369 PFQEP 373 ; Motif 2, 559 PVQEET 564 ; and Motif 3, 604 PVQEQG 609 ) regulating osteoclast formation and function (21). In addition, we demonstrated that these three motifs activate different signaling pathways (21). Motif 1 activates NF-B, c-Jun N-terminal kinase, extracellular signal-regulated kinase, and p38, whereas Motif 2 triggers the activation of NF-B and p38. In contrast, Motif 3 only activates NF-B. However, the specific RANK cytoplasmic motifs and downstream signaling pathways involved in osteoclast survival have not been functionally identified.
In the present study, we sought to elucidate functional RANK motifs that play a role in osteoclast survival. We have demonstrated that the three RANK motifs (Motif 1, 369 PFQEP 373 ; Motif 2, 559 PVQEET 564 ; and Motif 3, 604 PVQEQG 609 ) mediating osteoclast formation and function are also involved in modulating osteoclast survival. Moreover, although Motif 1 plays a minimal role in osteoclast formation and function (21), it is highly potent in promoting osteoclast survival in part by activating Akt/PBK pathway. In addition, we investigated which downstream effector(s) is activated by Akt/PKB to promote osteoclast survival, and our data indicate that AFX/FOXO4 is the key downstream effector activated by Akt/PKB to modulate osteoclast survival.

EXPERIMENTAL PROCEDURES
Chemicals and Reagents-Chemicals were purchased from Sigma unless indicated otherwise. Blasticidin was from EMD Biosciences, Inc. Preparation of Retrovirus Encoding the Chimeric Receptors-The retrovirus vectors encoding various chimeric receptors were prepared in a previous study (21). Retrovirus was prepared using Plat-E packaging cells (22). Plat-E cells were cultured in Dulbecco's modified Eagle's medium with 10% heat-inactivated fetal bovine serum as described in Ref. 22 and transiently transfected using Lipofectamine Plus (Invitrogen) as described in Ref. 21. Virus supernatant was collected at 48, 72, and 96 h after transfection.
Culturing and Infection of Bone Marrow Macrophages (BMMs)-Bone marrow cells were isolated from long bones of 4 -8-week-old TNFR1 Ϫ/Ϫ R2 Ϫ/Ϫ double knock-out mice (The Jackson Laboratory, Bar Harbor, ME) or wild-type mice (Harlan Industries, Indianapolis, IN) as described (23). BMMs were prepared by culturing isolated bone marrow cells in ␣-minimal essential medium containing 10% heat-inactivated fetal bovine serum in the presence of 0.1 volume of culture supernatant of monocyte/macrophage colony-stimulating factor (M-CSF)-producing cells for 2 days as previously described (24). Cells were then infected with virus for 24 h in the presence of 0.1 volume of culture supernatant of M-CSF-producing cells and 8 g/ml polybrene. Cells were further cultured in the presence of M-CSF and 2 g/ml puromycin for selection and expansion of transduced cells. Selected cells were subsequently used for various studies.
In Vitro Osteoclast Survival Assay-Retrovirally infected BMMs were cultured in 24-well tissue culture plates (1 ϫ 10 5 cells/well) in ␣-minimal essential medium containing 10% heat-inactivated fetal bovine serum in the presence of 0.01 volume of culture supernatant of M-CSFproducing cells (final M-CSF concentration is 22 ng/ml) and 100 ng of glutathione S-transferase-RANKL (25) for 4 days to stimulate osteoclast formation. After osteoclasts were formed, the cultures were treated with different factors as indicated in individual assays and then continued for 14 more hours. The cultures were stained for tartrate-resistant acid phosphatase activity using a commercial kit (387-A; Sigma). Survived osteoclasts were determined as those cells with strong tartrate-resistant acid phosphatase activity, more than three nuclei, and intact plasma membrane. Immunofluorescence Assays-Osteoclast apoptosis was assessed by determining distribution of phosphatidylserine on the outer leaflets of the plasma membrane using an Annexin V-FITC fluorescence kit from BD Pharmingen (San Diego, CA). The assay was performed basically according to the instruction manual. The stained osteoclasts were viewed and documented using an Olympus IX70 inverted fluorescence microscope.
Western Analysis-BMMs infected with retrovirus were cultured in serum-free ␣-minimal essential medium in the absence of M-CSF for 16 h before treatment with TNF-␣ for various times as indicated in individual experiments. For assays involving osteoclasts, BMMs were treated with M-CSF (22 ng/ml) and RANKL (100 ng/ml) for 4 days to stimulate osteoclast formation. After osteoclasts were formed, the cultures were treated with PBS or RANKL for 2 or 5 h. Infected BMMs or mature osteoclasts were washed twice with ice-cold PBS and then lysed in buffer containing 20 mM Tris, pH 7.5, 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% Triton X-100, 2.5 mM sodium pyrophosphate, 1 mM ␤-glycerophosphate, 1 mM Na 3 VO 4 , 1 mM NaF, and 1ϫ protease inhibitor mixture 1 (P-2850; Sigma) and 1ϫ protease inhibitor mixture 2 (P-5726; Sigma). 40 g of cell lysates were boiled in the presence of SDS sample buffer (0.5 M Tris-HCl, pH 6.8, 10% (w/v) SDS, 10% glycerol, 0.05% (w/v) bromphenol blue) for 5 min and loaded for electrophoresis on 10% SDS-PAGE. Proteins were transferred to nitrocellulose membranes (catalogue number 162-0147) from Bio-Rad using a semi-dry blotter (Bio-Rad). Membranes were blocked in blocking solution (5% nonfat dry milk in TBS containing 0.1% Tween 20) for 1 h to prevent nonspecific binding and then washed three times with TBS-T (TBS containing 0.1% Tween 20). Membranes were incubated with primary antibodies in TBS-T containing 5% bovine albumin (A-7030; Sigma) overnight at 4°C. The next day, membranes were washed three times with TBS-T and incubated with secondary antibody in TBS-T containing 5% nonfat dry milk for 1 h. Membranes were washed extensively, and enhanced chemiluminescence (ECL) detection assay was performed using the SuperSignal West Dura kit from Pierce.

The Chimeric Receptor Approach Permits the Investigation of RANK
Signaling Involved in Osteoclast Survival-Previously we developed a chimeric receptor approach for delineating functional motifs in the RANK cytoplasmic domain mediating osteoclast differentiation and function (21). The chimeric receptor consists of mouse TNFR1 external domain linked to the transmembrane and intracellular domains of mouse RANK. The wild-type chimeric receptor in which no mutation is introduced in the RANK cytoplasmic domain was designated as WT (Fig. 1A). When expressed in osteoclast precursors (namely, BMMs) derived from TNFR1 and R2 double knock-out mice (TNFR1 Ϫ/Ϫ R2 Ϫ/Ϫ ), WT was capable of modulating osteoclast differentiation and function in response to TNF-␣ as a surrogate (21). In addition, numerous mutant chimeric receptors were prepared in the previous study (21) (Fig. 1A). The RANK cytoplasmic domain contains six putative TRAF binding motifs (PTM1-PTM6). In W1-W6, all PTMs except one are mutated. In P1-6, all PTMs are mutated (Fig. 1A). Using these mutants, we successfully identified three functional RANK motifs that regulate osteoclast formation and function (21). Thus, we reason that these mutants can also serve as useful tools for elucidating RANK motifs involved in osteoclast survival in the current study.
To investigate the role of the six PTMs in osteoclast survival, we need to obtain mature osteoclasts expressing the six mutants (W1-W6). However, the previous study revealed that whereas W3, W5, and W6 can partially promote osteoclast formation, W1, W2, and W4 are inca-pable of mediating osteoclast formation ( Fig. 1B) (21). Fortunately, the chimeric receptor approach possesses a unique "switch" feature that allows us to overcome the problem. As shown in Fig. 1C, infected BMMs not only express a chimeric receptor but also retain endogenous RANK. As a result, infected BMMs may be treated with RANKL and M-CSF to stimulate osteoclast formation. Once osteoclasts are formed, M-CSF and RANKL are removed and the cultures are treated with TNF-␣ alone to assess the impact of the expressed chimeric receptor on osteoclast survival (Fig. 1C).
To determine whether the chimeric receptor approach indeed has the switch feature and whether RANK motifs mediating osteoclast survival are among the six PTMs, uninfected BMMs and infected BMMs expressing WT or P1-6 were treated with M-CSF (22 ng/ml) and RANKL (100 ng/ml) to stimulate osteoclast formation ( Fig. 2A). Four days later, osteoclasts were formed. A representative osteoclast culture is shown in Fig. 2A, top panel. Then, M-CSF and RANKL were removed, and the cultures were treated with vehicle (PBS), M-CSF, RANKL, or TNF-␣ for 14 h (Fig. 2A). As expected, osteoclasts expressing wild-type chimera (WT) survived in response to M-CSF or RANKL; more importantly, these cells also survived in response to TNF-␣. As negative control, osteoclasts derived from uninfected BMMs failed to survive in response to TNF-␣. These data indicate that the chimeric receptor approach does possess the unique switch feature. In addition, although osteoclasts expressing WT survived with stimulation of TNF-␣, those expressing P1-6 failed to do so, indicating that RANK motifs mediating osteoclast survival are indeed among the six PTMs. Fig. 2B shows the quantification of the data in Fig. 2A.
Identification of Three Functional RANK Motifs Mediating Osteoclast Survival-To identify the functional RANK cytoplasmic motifs, we repeated the osteoclast survival assay shown in Fig. 2 with mutant chimeras W1-W6. We also used WT as positive control and P1-6 as negative control. In these assays, after the removal of M-CSF and RANKL, osteoclast cultures were treated with either PBS or TNF-␣ for 14 h. As shown in Fig. 3A, whereas W1, W2, and W4 failed to promote osteoclast survival, W3, W5, and W6 were able to mediate osteoclast survival. The data shown in Fig. 3A are quantified in Fig. 3B. The survival assays also showed that the potency of these motifs in promoting osteoclast survival is in the order of W3 ϾW5 ϾW6. As shown in Fig. 1A, in W3 all PTMs except PTM3 are mutated. Similarly, in W5 and W6 all PTMs except PTM3 and PTM5 are mutated, respectively. To further investigate the role of these mutants in osteoclast survival, we examined and compared the apoptosis rate of osteoclasts expressing the different chimeric receptor mutants by determining distribution of phosphatidylserine on the outer leaflets of the plasma membrane (Fig. 4). These assays indicate that osteoclasts expressing W1, W2, and W3 undergo extensive apoptosis, whereas those expressing W5 and W6 show some degree of apoptosis. In contrast, cells expressing W3 barely exhibit any distribution of phosphatidylserine on the outer leaflets of the plasma membrane. Taken together, we conclude that we have identified PTM3, PTM5, and PTM6 as functional motifs capable of promoting osteoclast survival. To be consistent with our previous study (21), we renamed PTM3, PTM5, and PTM6 as Motif To further demonstrate that these three RANK motifs play functional roles in osteoclast survival, we repeated the osteoclast survival assay with the mutant chimeric receptors called P3, P5, and P6 (21). In P3, PTM3 (Motif 1) is mutated. In P5, PTM5 (Motif 2) is mutated. In P6, PTM6 (Motif 3) is mutated. Fig. 5A shows osteoclast cultures obtained from the survival assays, and the data are quantified in Fig. 5B. Consistent with the notion that PTM3 (Motif 1) plays a predominant role in modulating osteoclast survival, P3 exhibited very low capacity in modulating osteoclast survival. In contrast, P5 and P6 still demonstrated considerable ability to mediate osteoclast survival, revealing that PMT5 (Motif 2) and PTM6 (Motif 3) are less potent than PTM3 (Motif 1) in mediating osteoclast survival.

PTM3 (Motif 1), but Not PTM5 (Motif 2) or PTM6 (Motif 3), Activates the Akt/PKB Pathway to Promote Osteoclast Survival-RANK has been
shown to be able to activate six major pathways: NF-B, c-Jun NH 2terminal kinase, extracellular signal-regulated kinase, p38, Akt/PKB, and NFATc1 (26). Among these distinct pathways, NF-B, extracellular signal-regulated kinase, and Akt/PKB are capable of modulating osteoclast survival (12,(27)(28)(29). In accord with their ability to promote osteoclast survival, PTM3 (Motif 1) has been shown to be able to activate NF-B and extracellular signal-regulated kinase, whereas PTM5 (Motif 2) and PTM6 (Motif 3) are able to activate NF-B (21). Because RANK also activates Akt/PKB pathway (12,28), the question is whether these three functional RANK motifs are able to activate the Akt/PKB pathway. To address this issue, we performed Western analysis with an antibody against phosphorylated Akt/PKB as previously described (21).
First, we infected TNFR1 Ϫ/Ϫ R2 Ϫ/Ϫ BMMs with virus encoding either the wild-type chimera or the mutant P1-6. Infected cells were then treated with TNF-␣ for various times, and the activation of the Akt/PKB signaling pathway was determined by Western analysis (Fig. 6A). Whereas the wild-type chimera induced the phosphorylation of Akt/ PKB with 15-and 30-min treatments (Fig. 6A, lanes 1-3), P1-6 mutant failed to do so (lanes 4 -6), indicating that some of the six PTMs are implicated in activation of the Akt/PKB pathway. Next, we determined which PTM activates Akt/PKB by repeating Western analysis with W3, W5, and W6 mutants. As shown in Fig. 6B, whereas W3 induced Akt/ PKB phosphorylation with 15-and 30-min treatments (lanes 1-3), W5 and W6 failed to do so (lanes 4 -9), revealing that only PTM3 (Motif 1) is able to activate the Akt pathway.
The assays above were performed with BMMs, which are osteoclast precursors. To further support a functional role for Akt/PKB in osteoclast survival, it is important to determine whether Akt/PKB is activated by RANKL in mature osteoclasts by repeating the above assays with mature osteoclasts. However, our initial attempts failed because most osteoclasts died after 16 or 8 h of starvation (in ␣-minimal essential medium without serum, M-CSF, or RANKL). To overcome this problem, we used a different experimental strategy, depicted in Fig. 7A. First, we treated BMMs with M-CSF and RANKL to promote osteoclast formation. Once osteoclasts were formed, cells were either lysed immediately or lysed after having been treated with vehicle (PBS) or RANKL for 2 or 5 h (Fig. 7A). The lysates were then used for Western analysis for the state of Akt/PKB activation (phosphorylation). As shown in Fig. 7B, Akt/PKB phosphorylation is decreased more dramatically in control cultures (lanes 2 and 4) than in the cells treated with RANKL (lanes 3 and 5), indicating that RANKL plays a role in activating Akt/PKB in mature osteoclasts.
Next, we investigated the activation of AFX/FOXO4 in mature osteoclasts. We treated BMMs with M-CSF and RANKL to promote oste-oclast formation as described in Fig. 7A above. Once osteoclasts were formed, cells were either lysed immediately or lysed after having been treated with vehicle (PBS) or RANKL for 2 or 5 h. The lysates were then used for Western analysis for the state of GSK3␤, BAD, FKHR/ FOXO1a, and AFX/FOXO4 activation (phosphorylation). As shown in Fig. 9A, AFX/FOXO4 phosphorylation is decreased more dramatically in control cultures (lanes 2 and 4, the upper band represents the phosphorylated form of AFX/FOXO4) than in the cells treated with RANKL (lanes 3 and 5), indicating that RANKL plays a role in activating AFX/ FOXO4 in mature osteoclasts. In contrast, FHKR/FOXO1a and Bad phosphorylation do not decline following RANKL withdrawal (Fig. 9, B and C), indicating that these two factors are not involved in RANKL-  PTM3 (Motif 1) is capable of activating the Akt/PKB pathway in BMMs. A, activation of the Akt/PKB pathway by the wild-type chimera (WT) and P1-6 in BMMs was assessed as phosphorylation of Akt/PKB using Western analysis with an antibody against phospho-Akt (p-Akt). Immunoblots were stripped and then reprobed with an antibody against Akt as loading controls. Min, minute. B, activation of the Akt/PKB pathway by W3, W5, and W6 in BMMs. The assays were performed as described in panel A.

FIGURE 7. Activation of Akt pathway by RANKL in osteoclasts.
A, diagram describing the experimental procedures for preparing cell lysates for Western analysis. Hrs, hours. B, activation of the Akt/PKB pathway by RANKL in osteoclasts was assessed as phosphorylation of Akt/PKB using Western analysis with an antibody against phospho-Akt (p-Akt). Immunoblots were stripped and then reprobed with an antibody against Akt/PKB as loading control. induced osteoclast survival. Finally, our data also demonstrate that GSK3␤ is not phosphorylated in mature osteoclasts (Fig. 9D), indicating that GSK3␤ is not involved in osteoclast survival either.

DISCUSSION
The control of osteoclast life span has been recognized as a critical regulatory factor in bone remodeling, and alteration in osteoclast life span is attributed to the pathogenesis of bone disorders, including postmenopausal osteoporosis (30). Since the discovery of RANKL in the late 1990s (1-4), RANKL has been shown to play pivotal roles not only in osteoclast formation and function (1, 6, 7) but also in osteoclast survival (10 -12). However, the molecular mechanism by which RANKL regulates osteoclast survival has not been fully understood. RANKL exerts its functions by binding to and activating its receptor, RANK, which recruits various TRAF proteins through specific motifs in the cytoplasmic domain upon activation. Recently, we identified three RANK cytoplasmic motifs (Motif 1, 369 PFQEP 373 ; Motif 2, 559 PVQEET 564 ; and Motif 3, 604 PVQEQG 609 ) regulating osteoclast formation and function (21). In the present study, we investigated the molecular mechanism by which RANKL regulates osteoclast survival by identifying RANK motifs modulating osteoclast survival and the downstream signaling pathways involved. The current study revealed that the same three motifs are also involved in osteoclast survival. However, it has also become clear that the three RANK motifs play different roles in osteoclast differentiation, function, and survival. Although Motif 1 plays a minimal role in osteoclast differentiation and function (21), it is highly potent in mediating osteoclast survival (Fig. 3). Moreover, although Motif 2 and Motif 3 are highly capable of mediating osteoclast formation and function (31), they are not as effective as Motif 1 in promoting osteoclast survival (Fig. 3).
Akt/PKB is a potent factor regulating osteoclast survival in response to a variety of stimuli, including RANKL, interleukin-1, and TNF-␣ (12,28,32,33). RANKL activates Akt/PKB through a signaling complex involving TRAF6 and c-Src (12). To investigate the molecular mechanism by which the three RANK motifs modulate osteoclast survival, we examined whether these motifs are capable of activating Akt/PKB. Our data indicate that Motif 1, but not Motif 2 or Motif 3, is able to activate Akt/PKB. This result is consistent with the previous finding that 369 PFQEP 373 is a TRAF6 binding motif (35). Given that Motif 2 and Motif 3 can also exert small effect on osteoclast survival, then the question is how these two motifs mediate osteoclast survival. Previously, we demonstrated that Motif 2 and Motif 3 can activate NF-B pathway (21), which has been shown to be able to promote osteoclast survival (27). Moreover, our previous study showed that Motif 1 can also activate NF-B and extracellular signal-regulated kinase. Both NF-B and extracellular signal-regulated kinase are mediators of osteoclast survival (27)(28)(29). These data explain why Motif 1 is much more potent than Motif 2 and Motif 3 in promoting osteoclast survival, as summarized in Fig 10. It has been well established that Akt/PKB may utilize distinct downstream effectors such as p21 CIP1 , MDM2, TSC2, endothelial nitric-oxide synthase, GSK3␤, FKHR/FOXO1a, BAD, and AFX/FOXO4 to regulate a variety of cellular processes such as cell proliferation, growth, and survival (15)(16)(17). Among these diverse effectors, GSK3␤, FKHR/ FOXO1a, BAD, and AFX/FOXO4 have been shown to play roles in cell survival (15)(16)(17). Although Akt/PKB has been shown to play a key role in regulating osteoclast survival (12,28,32,33), the precise downstream effector(s) activated by Akt/PKB to regulate osteoclast survival has still remained elusive. Thus, we extended our study to identify Akt/PKBactivated downstream effector(s) involved in osteoclast survival. Such study revealed that Akt/PKB regulates osteoclast survival by targeting  . Activation of GSK3␤, FKHR/FOXO1a, BAD, and AFX/FOXO4 pathways by RANKL in osteoclasts. Cell lysates were prepared as described in Fig. 7A. The Western assays were performed as described in Fig. 7. A, phosphorylation of AFX/FOXO4. B, phosphorylation of BAD. C, phosphorylation of FKHR/FOXO1a. D, phosphorylation of GSK3␤.
the downstream effector AFX/FOXO4. Notably, FHKR/FOXO1a and Bad are phosphorylated in osteoclasts, but their phosphorylation does not decline following RANKL withdrawal (Figs. 9, B and C). This result suggests that these factors may be involved in osteoclast survival but they are not involved in RANKL-induced osteoclast survival. As such, FHKR/FOXO1a and Bad phosphorylation may likely be regulated by other survival factors such as adhesion molecules on mature osteoclasts. Interestingly, our data also demonstrate that GSK3␤ is not phosphorylated in mature osteoclasts (Fig. 9D), indicating that GSK3␤ is not involved in osteoclast survival at all. AFX (ALL1 fused gene from chromosome X) is a transcription factor belonging to the Forkhead transcription factor superfamily (36) and was identified as a mammalian homolog of the Forkhead transcription factor dauer arrest phenotype-16 in Caenorhabditis elegans (37)(38)(39)(40). The Forkhead superfamily contains Ͼ100 members identified in different species, and ϳ40 mammalian homologs have been identified so far (36,41). Members of this superfamily are characterized by the presence of a highly conserved DNA-binding domain known as the Forkhead box, which was named based on its homology to the DNA-binding domain of the Drosophila homeotic Forkhead protein and the hepatic nuclear factor-3 transcription factors (42,43). Recently, the Forkhead transcription factors have been renamed Fox (Forkhead box) transcription factors (44). Moreover, given that members of the Forkhead superfamily are highly divergent outside of the Forkhead box, Forkhead transcription factors have been further categorized into 17 subfamilies ranging from FOXA to FOXQ (44).
AFX, newly named as FOXO4, belongs to the FOXO subfamily that comprises several other members including FOXO1a (FKHR) and FOXO3a (FKHRL1) (44). The FOXO factors have been shown to play important roles in cell proliferation and survival (44). Especially, activation of a FOXO factor is sufficient to induce the expression of proapoptotic genes and to trigger apoptosis in cells of hematopoietic origin (44). Phosphorylation of a FOXO factor by Akt/PKB causes its retention in cytoplasm and/or its translocation from the nucleus to cytoplasm, resulting in the inhibition of the transcriptional activation of the proapoptotic genes. Although FOXO1a (FKHR) and FOXO4 (AFX) are downstream targets of Akt/PKB, our data indicate that only AFX/ FOXO4 is involved in RANKL-mediated osteoclast survival. This result is consistent with a recent study demonstrating that AFX/FOXO4 plays an important role in apoptosis in cells of hematopoietic origin (45). Specifically, AFX/FOXO4 activates apoptosis by inducing the expression of the BCL-6 transcriptional repressor (45) (Fig. 10). In supporting this notion, the BCL-6 promoter contains multiple binding sites for AFX/FOXO4 (45). It was further shown that BCL-6 regulates apoptosis by regulating the expression of BCL-X L , an anti-apoptotic protein (45) (Fig. 10). More interestingly, it has also been demonstrated that macrophages isolated from BCL-6 Ϫ/Ϫ exhibited enhanced survival, supporting a potential role of BCL-6 in AFX-induced osteoclast precursor survival. In line with this finding, our data indicate that AFX/FOXO4 is phosphorylated in response to RANKL stimulation.
It is worthwhile to note that AFX/FOXO4 knock-out mice have been recently generated and reported (34). The AFX/FOXO4 Ϫ/Ϫ mice are viable, and no apparent abnormalities have been identified (34). However, it is not clear whether the skeletal development and, especially, osteoclast survival in these mice have been carefully examined. Given that AFX/FOXO4 pathway represents only one of numerous pathways implicated in regulating osteoclast survival (Fig. 10), AFX/FOXO4 Ϫ/Ϫ osteoclasts are likely to exhibit only subtle differences in osteoclast survival compared with wild-type controls. As a result, any potential in vivo phenotypes (bone mineral density, osteoclast numbers) may only be detectable in older mice. Thus, a careful examination of the AFX/ FOXO4 Ϫ/Ϫ mice is needed to determine whether any abnormality in osteoclast survival results from the deletion of the gene for AFX/FOXO4.