DC-STAMP: A Key Regulator in Osteoclast Differentiation

Abstract

Osteoimmunology research is a new emerging research field that investigates the links between the bone and immune responses. Results from osteoimmunology studies suggest that bone is not only an essential component of the musculoskeletal system, but is also actively involved in immune regulation. Many important factors involved in immune regulation also participate in bone homeostasis. Bone homeostasis is achieved by a coordinated action between bone synthesizing osteoblasts and bone degrading osteoclasts. An imbalanced action between osteoblasts and osteoclasts often results in pathological bone diseases: osteoporosis is caused by an excessive osteoclast activity, whereas osteopetrosis results from an increased osteoblast activity. This review focuses on dendritic cell specific transmembrane protein (DC STAMP), an important protein currently considered as a master regulator of osteoclastogenesis. Of clinical relevance, the frequency of circulating DC STAMPþ cells is elevated during the pathogenesis of psoriatic diseases. Intriguingly, recent results suggest that DC STAMP also plays an imperative role in bone homeostasis by regulating the differentiation of both osteoclasts and osteoblasts. This article summarizes our current knowledge on DC STAMP by focusing on its interacting proteins, its regulation on osteoclastogenesis related genes, its possible involvement in immunoreceptor tyrosine based inhibitory motif (ITIM) mediated signaling cascade, and its potential of developing therapeutics for clinical applications.

Bone is a dynamic tissue that continuously undergoes remodeling through a concerted action between bone-resorbing osteoclasts (OC), bone-forming osteoblasts (OB), and osteocytes, the long-lived osteoblast-derived cells that reside within the bone matrix to monitor and orchestrate OC::OB-mediated bone homeostasis (Boyce, 2013a,b; Florencio-Silva et al., 2015). Imbalanced bone homeostasis results in pathological inflammatory bone diseases. Osteoporosis results from excessive bone erosion activity of OC, whereas osteopetrosis arises from increased bone synthetic activity of OB. Therefore, modulating the relative balance between OB and OC remains an effective therapeutic strategy for treating bone diseases such as osteoporosis and rheumatoid arthritis (RA).

OC are multinucleated giant cells derived from myeloid precursors. To date, they are the only cell type known to form ruffled membranes with bone-degrading activities (Boyce, 2013a). In response to macrophage colony-stimulating factor (M-CSF) and receptor activator of NF-kB ligand (RANKL) stimulation, single-nucleated osteoclast precursors (OCPs) differentiate into mature multinucleated OC through several rounds of cell–cell fusion. In addition to resorbing bone, intriguingly, OC can communicate with OB and other cells through bone remodeling to maintain skeletal integrity. Thus, the interaction between OC, OB and extracellular mediators regulate bone remodeling in health and disease (Teti, 2013; Charles and Aliprantis, 2014).

The differentiation of OC, a process termed osteoclastogenesis (OCgenesis), is a complex but orderly event. It involves aggregation of cell surface proteins required for cell–cell fusion, initiation of signaling cascades after cytokine stimulation, and activation of gene transcription programs that direct the transition from a single-nucleated monocyte to a functional bone-degrading multinucleated polykaryon (Boyce, 2013a). Dendritic cell-specific transmembrane protein(DC-STAMP) is currently considered the master regulator of osteoclastogenesis (Courtial et al., 2012; Islam et al., 2014; Zhang et al., 2014). It is a 470-amino acid protein encoded by a single open reading frame, has a signal sequence, seven putative transmembrane regions, three potential N-linked glycosylation sites, a protein kinase C phosphorylation site, and a 72-residue cytoplasmic tail containing multiple serines, two of which are likely to be the targets of phosphorylation. Knock-down of DC-STAMP completely abrogates cell–cell fusion during OCgenesis. OC isolated from DC-STAMP knockout (KO) mice are single nucleated OC due to deficiency in cell-cell fusion and exhibit mild osteopetrosis (Yagi et al., 2005). Identification of an immunoreceptor tyrosine-based inhibitory motif (ITIM) on the cytoplasmic tail of DC-STAMP suggests its role in signaling (Chiu et al., 2012b). In this review, we discussed the function of DC-STAMP beyond cell–cell fusion. First, we summarize research progress on DC-STAMP and its potential for clinical application. Second, we describe cellular events of osteoclastogenesis likely to be regulated by DC-STAMP.

Clinical Potentials of DC-STAMP

Biomarkers

Ps to PsA transition and TNFi medication response. Psoriatic arthritis (PsA) is an inflammatory joint disease that affects over 600,000 Americans (Gelfand et al., 2005). Bone damage develops in half of these patients within the first 2 years of disease, often leaving them with impaired function (Kane et al., 2003). The advent of anti-tumor necrosis factor therapies (TNFi) greatly reduced bone damage in PsA patients, however, only 50–60% of patients respond to these agents. To improve treatment outcomes, two hurdles need to be addressed: a limited understanding of key events that underlie pathologic bone destruction and the absence of biomarkers to predict TNFi response for therapy optimization. In most patients, psoriasis precedes the onset of arthritis by 10 years on average. The interval provides an ideal opportunity to identify cellular and/or serum biomarkers of arthritis in a susceptible population.

Our preliminary data showed that DC-STAMP is not only a valid biomarker whose expression is elevated after Ps to PsA transition, but also a response biomarker which declines rapidly in TNFi responders. We are currently recruiting and analyzing a bigger cohort of PsA patients (n > 200), to further confirm these findings (Chiu et al., 2012a).

Targeted drug delivery

We recently showed that DC-STAMP+ cells were recruited to the bone fracture sites and concentrated in OB proximity (submitted manuscript). An elevated DC-STAMP+ cell frequency in human blood was detected at a certain stage of bone healing. Using video capture techniques, the trafficking of DC-STAMP+ cells to fracture sites occurs in an organized order. Collectively, our data suggest that DC-STAMP, together with MMP-2 and MMP-9 (Galliera et al., 2010; Rocha et al., 2014), can be a useful tool for monitoring, in parallel with other bone turnover markers, to evaluate bone remodeling and tissue healing. Given that DC-STAMP+ cells are homing to bone fracture sites, a step forward approach is to molecular engineer a chimeric DC-STAMP protein by which DC-STAMP is able to bring small peptides/proteins such as bone morphogenetic protein 2 (BMP-2) or MMP-9 (Kim et al., 2015; Matsumoto et al., 2015; Miller et al., 2015; Saito et al., 2015; Wang et al., 2015a) with healing-promoting effects to bone fracture sites.

Updates on DC-STAMP Research

OC-STAMP versus DC-STAMP

In addition to DC-STAMP, OC-STAMP is another critical RANKL-induced, multipass transmembrane protein that promotes the formation and differentiation of osteoclass (Yang et al., 2008; Witwicka et al., 2015). In addition to their structure similarity, DC-STAMP and OC-STAMP share many common features. They both promote cell–cell fusion between osteoclast precursors when overexpressed; the presence of antibody or siRNA will block early events in osteoclastogenesis such as fusion, but leave late stage osteoclast differentiation unaffected; they are both induced by RANKL. Despite the similarities between DC-STAMP and OC-STAMP, several different properties are also noted as summarized in Table 1. Of note, although both DC-STAMP and OC-STAMP are essential for osteoclastogenesis, they are distinct proteins and are not interchangeable. DC-STAMP deficiency cannot be complemented by overexpression of OC-STAMP, and vice versa.

TABLE 1.

Comparison between OC-STAMP versus DC-STAMP

	OC-STAMP	DC-STAMP
Transmemrane	6-Pass	7-Pass (?)
Motif	No ITIM	ITIM
Bone phenotype	No	Increased BV/TV
		Increased trabecular bone volume
NFATc1	No effect	No effect on mRNA
		Has effect on protein

How DC-STAMP and OC-STAMP regulate osteoclastogenesis and whether they interact with each other remains to be further defined. The STAT6-STAT1 signaling axis was shown to be required for DC-STAMP/OC-STAMP-mediated cell fusion in macrophages (Miyamoto et al., 2012). Based on current data, three possibilities were proposed by Yang et al., for the relationship between DC-STAMP and OC-STAMP: (i) they are ligands of each other; (ii) they have their own distinct ligands. Once engaged, they provide positive reciprocal signals to each other to allow the fusion to proceed; (iii) they form a dimer as a receptor complex on the cell surface. Experiments such as immunoprecipitation and reciprocal cultures between DC-STAMP−/− and OC-STAMP−/− cells will help to determine these possibilities.

Whether OC-STAMP glycosylation will affect DC-STAMP surface expression or function remains to be investigated by using, the available mutated glycosylation-deficient (N162D) OC-STAMP (Witwicka et al., 2015). Based on the structural similarity between DC-STAMP and OC-STAMP and current topology analysis by the TMHMM algorithm (Witwicka et al., 2015), it will be interesting to investigate whether DC-STAMP is a 7-pass transmembrane protein as originally considered (Yagi et al., 2005). According to a recent review and two studies on OC-STAMP (Yang et al., 2008; Zhang et al., 2014; Witwicka et al., 2015), the topology of DC-STAMP (whether it is a 5, 6, or 7-pass transmembrane protein) remains an open question which needs further investigation to determine its similarity to OC-STAMP, a 6-pass transmembrane protein. To date, these two STAMPs are the only cell-type specific proteins essential for cell–cell fusion and the generation of osteoclasts and giant cells, making their role in osteoclastogenesis and macrophage fusion (Miyamoto et al., 2012) of particular interest and importance. Elucidating the involvement of DC-STAMP and OC-STAMP on the split and re-seal of the lipid bilayers of membranes during cell–cell fusion, and their possible interplay in the signaling cascade of osteoclastogenesis remain intriguing questions.

Regulators that Affect DC-STAMP Expression

Given that DC-STAMP ligand remains unknown, the functions DC-STAMP-interacting protein will provide indirect clues to the functions of DC-STAMP in cell–cell fusion and downstream signaling. To date, CCN2, OS9, and Pin-1 are three proteins that were shown to have physical interaction with DC-STAMP (Jansen et al., 2009; Nishida et al., 2011; Islam et al., 2014).

DC-STAMP-interacting proteins

CCN2

CCN2, one of the connective tissue growth factors, promotes endochondral ossification and significantly enhances the tartrate-resistant acid phosphatase (TRAP)+ multinucleated osteoclast formation in the presence of RANKL. The interaction between CCN2 and DC-STAMP is supported by real-time PCR, coimmunoprecipitation analysis, solid-phase binding assays, and successful complementation of osteoclastogenesis deficiency of CCN2−/− cells by overexpression of DC-STAMP.

OS9

OS9 was identified by screening a yeast 2-hybrid DNA library using the C terminus of DCSTAMP as bait. DC-STAMP and OS9 colocalized in endoplasmic reticulum (ER). TLR2-or TLR4-induced maturation of DCs led to translocation of DCSTAMP from the ER to the Golgi in a manner dependent on interaction with OS9; however, OS9 localization was unaffected. Based on these data, OS9 is considered to be involved in the modulation of ER-to-Golgi transport of DCSTAMP in response to TLR triggering.

Molecules that affect DC-STAMP expression

In addition to OS9, CCN2, and Pin1, the activity of DC-STAMP is either enhanced or suppressed, by a few modulators including proteins and miRNA involved in osteoclastogenesis (Table 2).

TABLE 2.

Summary of mediators that can regulate the expression of DC-STAMP at transcriptional or translational levels

	Protein	Effect	Reference
Enhance	Sbno-2	Sbno-2 regulates osteoclast fusion by enhancing the expression of DC- STAMP	Maruyama et al. (2013)
	CCN2	CCN2 promotes osteoclastogenesis via induction of and interaction with DC-STAMP	Nishida et al. (2011)
	PU.1**	Pu.1 activates the transcription of DC-STAMP	Courtial et al. (2012)
	MITF**	Pu.1 activates the transcription of DC-STAMP	Courtial et al. (2012)
Suppress
	Pin1	Pin1 regulates osteoclast fusion through suppression of the master regulator of  cell fusion DC-STAMP	Islam et al. (2014)
	MiR-7b	MiR-7b directly targets DC-STAMP causing suppression of NFATc1 and c-Fos signaling  during osteoclast fusion and differentiation	Dou et al. (2014)
	Tal1**	Tal1 regulates osteoclast differentiation through suppression of the master  regulator of cell fusion DC-STAMP	Courtial et al. (2012)

Tal1 counteracts the activating function of the transcription factors PU.1 and MITF.

Sbno2-null mice were osteopetrotic, with increases in trabecular bone volume and number and impaired osteoblastogenesis. Fusion of osteoclasts was also impaired in Sbno2-null mice, as was the formation of multinucleated osteoclasts in response to RANKL. Stimulation of Sbno2+/+, but not Sbno2−/− MCSF-derived macrophages with RANKL increased expression of DC-STAMP. Sbno2 bound to Tal1 and attenuated its inhibition of DC-STAMP expression, leading to activation of the DC-STAMP promoter by Mitf.

^**Tal1, PU.1, and MITF are all transcription factors.

Sbno2

Sbno2 strawberry notch homolog 2 regulates osteoclast fusion by enhancing the expression of DC-STAMP (Maruyama et al., 2013).

Pin1

Pin1 (Islam et al., 2014) binds, isomerizes DC-STAMP, and affects the expression levels and localization of DC-STAMP at the plasma membrane. Pin1−/− osteoclasts are larger than wild-type osteoclasts and have higher nuclei numbers, indicating greater extent of fusion. RT-PCR analysis showed that DC-STAMP signal is significantly increased in Pin1(−/−) osteoclasts. Immunohistochemistry revealed that DC-STAMP expression is significantly increased in the tibias of Pin1 KO mice. Collectively, these results indicate that Pin1 regulates osteoclast fusion via suppressing DC-STAMP, supporting the concept that DC-STAMP is involved in the regulation of osteoclast volume.

Tal-1

Tal-1 is a transcription factor that is expressed in osteoclasts. Deletion of Tal1 in osteoclast progenitors resulted in altered expression on more than 1,200 genes. DC-STAMP is a direct target gene of Tal1. Tal1 represses DC-STAMP expression by counteracting the activating function of the transcription factors PU.1 and MITF (Courtial et al., 2012).

MiR-7b

A panel of microRNAs (miRNA) were identified to play crucial roles in bone metabolism and osteoclast differentiation. Among them, miRNA miR-7b is directly related to OCPs fusion and specifically targets DC-STAMP to inhibit osteoclastogenesis. Overexpression of miR-7b in RAW264.7 cells attenuated the number of TRAP-positive cells number and the formation of multinucleated cells, whereas the inhibition of miR-7b enhanced osteoclastogenesis (Dou et al., 2014).

The identification of ITIM on the cytoplasmic tail of DC-STAMP suggests its involvement in the ITAM-ITIM network (Chiu et al., 2012b). This notion was supported by the changes of several key regulatory fusogenic genes including NFATc1, c-Fos, Akt, Irf8, Mapk1, and Traf6 after miR-7b targeting to DC-STAMP (Dou et al., 2014). Thus, identification of additional miRNAs, which block osteoclastogenesis via DC-STAMP, might serve as a potential therapeutic approach to treat osteoclast-related bone disorders. To date, in addition to miR-7b, many new miRNA that specifically target DC-STAMP have been reported (http://www.informatics.jax.org/interaction/explorer?markerIDs=MGI:1923016). It remains to be identified whether these miRNAs affect DC-STAMP-mediated osteoclastogenesis to a different extent at distinct regulation levels.

RNAi

The expression of DC-STAMP at both mRNA and protein levels can be significantly and specifically inhibited by RNAi (Zeng et al., 2013). Of note, DC-STAMP inhibition by RNAi consequently suppressed fusion and bone resorption of human osteoclasts, suggesting that inhibition of DC-STAMP by RNAi is an efficient and effective method of regulating osteoclast functions. Data from Zeng et al. (2013) demonstrated that the lentivirus-mediated RNAi was capable of efficiently suppressing DC-STAMP expression in primary human osteoclasts and inhibiting osteoclastogenesis.

Possible Involvement of DC-STAMP in Biological Pathways

Cytoskeleton rearrangement

An active rearrangement of the cytoskeleton is taking place during each cell-to-cell fusion, which involves the actin-dependent cytoskeleton signaling network (Takito et al., 2014). In order to form multi-nucleated osteoclasts, single nucleated osteoclast precursors need to go through many rounds of cell–cell fusion. As cell–cell fusion proceeds, the volume of cells expands and zipper-like actin superstructures appear transiently (Takito et al., 2012). The molecular mechanisms underlying the control of osteoclast cell size remain largely unknown. By using various inhibitors that specifically block the actin cytoskeleton signaling network (Takito et al., 2015), it has been shown that the size of osteoclasts is regulated by the actin-mediated cytoskeleton signaling network. Osteoclast polarization is accompanied by extensive reorganization of the actin cytoskeleton.

Several sensor proteins including Src, Rho, and Rac1 (Soriano et al., 1991; Izawa et al., 2012; Georgess et al., 2014) are involved in actin-mediated cytoskeleton reorganization signaling. Current data demonstrated that actin-mediated positive regulators of podosome formation favor the generation of large osteoclasts (Takito et al., 2015). Given that tailed-deleted DC-STAMP cannot complement OC-forming deficiency of DC-STAMP−/− cells and only generates cells with 3 or less nuclei without expanded cell volume whereas WT DC-STAMP can fully complement OC-forming deficiency of DC-STAMP−/− cells associated with a larger volume of osteoclasts (unpublished data, manuscript submitted), DC-STAMP implicated in controlling the size of osteoclasts in a direct or indirect manner.

In addition to actin, integrin is essential for the attachment of osteoclasts to bone matrix to initiate bone resorption and degradation. Once osteoclasts move to skeletal tissue to initiate bone resorption, a privileged microenvironment space is established between the osteoclast and the bone surface. Formation of this space is mediated by signals emanating from integrin and transits to its active high-affinity conformation by growth factor-initiated intracellular events targeting the matrix receptor’s cytoplasmic domain. Integrin-mediated OC attachment to matrix and activation is well reviewed by Boyce (2013a). Briefly, an active cytoskeleton remodeling occurs during two stages of osteoclast differentiation: cell–cell fusion and bone matrix attachment through a complex cascade, which connects the extracellular integrin-mediated activation signal to the intracellular actin-mediated cytoskeleton rearrangement. c-Src links a RANK/ab3 integrin complex to the osteoclast cytoskeleton (Izawa et al., 2012). In addition to c-Src, two proteins in the FAK kinase family, FAK and Pyk2, are also known to be involved in the regulation of cytoskeleton. The role of DC-STAMP in cytoskeleton remodeling of osteoclasts is suggested by four lines of evidence as follows. First, deletion of ITIM on DC-STAMP results in a failure of cytoskeleton remodeling and expanding of cell volume; second, there is a tyrosine residue in the ITIM motif of the DC-STAMP cytoplasmic tail, which is likely to be the target of SHIP-1 (Chiu et al., 2012b) or FAK-kinase family (Kim et al., 2007; Ray et al., 2012); third, similar to FAK kinases (FAK and Pyk2), DC-STAMP also regulates both OB and OC differentiation, suggesting that DC-STAMP and FAK kinases are in the same biological pathway to mediate cytoskeleton remodeling; fourth, like integrins, DC-STAMP is also a transmembrane protein. Given that c-Src links a RANK/ab3 integrin complex to the osteoclast cytoskeleton (Izawa et al., 2012), one possibility remains to be investigated whether DC-STAMP is also one component of the RANK/ab3 integrin complex. Of note, integrin was shown to regulate spleen tyrosine kinase (Syk) through an ITAM-independent pathway (Hughes et al., 2014). Thus, DC-STAMP and integrin are likely to regulate cytoskeleton remodeling, especially at each cell–cell fusion, through a complex signaling interplay between integrin-Syk-mediated signaling and ITAM- and ITIM-bearing receptors during osteoclastogenesis.

Fusion frequency and osteoclast volume

In contrast to WT DC-STAMP infection DC-STAMP−/− cells, which fully complements the OC-forming deficiency, deletion of ITIM on DC-STAMP caused a premature termination of cell–cell fusion, resulting in multinucleated cells with no more than four nuclei (manuscript submitted). A recent study suggested that DC-STAMP is involved in the regulation of cell–cell fusion and control of osteoclast volume via Pin1 (Islam et al., 2014). The interaction between Pin1 and DC-STAMP is summarized as follows: (i) Pin1−/− osteoclasts are larger than wild-type osteoclasts and have higher nuclei numbers, indicating a greater extent of fusion; (ii) RT-PCR analysis showed that the DC-STAMP signal is significantly increased in Pin1(−/−) osteoclasts; (iii) immunohistochemistry revealed that DC-STAMP expression is significantly increased in the tibias of Pin1 KO mice; (iv) Pin1 binds, isomerizes DC-STAMP, and affects the expression levels and localization of DC-STAMP at the plasma membrane. Pin1 regulates osteoclast fusion through suppression of the master regulator of cell fusion DC-STAMP.

Selection of fusion partner

Expression of DC-STAMP protein on the cell membrane is required for cell–cell fusion (Vignery, 2005). RANKL induces DC-STAMP expression on osteoclast precursors of which the DC-STAMP(lo) precursors are the master fusogens (Mensah et al., 2010). This finding suggests that DC-STAMP heterogeneity determines the fusion potential of osteoclast precursors. This concept is supported by several recent findings. First, Soe et al. (2015) showed that cell fusion occurs preferentially between fusion partners with a higher level of heterogeneity. This heterogeneity includes the number of nuclei, the level of maturity, and cell mobility. Intriguingly, fusion between a mobile and an immobile partner were most frequent (62%), while fusion between two mobile (26%) or two immobile partners (12%) was less frequent. In addition, a more mature osteoclast will preferentially fuse with a less mature pre-osteoclast. Interestingly, osteoclasts most often gain nuclei by the addition of one nucleus at a time, and a moving cell is usually the nucleus donor to an immobile cell (Soe et al., 2015). Secondly, it has been recently shown that the relative cell surface localization of DC-STAMP, CD47, and syncytin-1 determines the occurrence of cell–cell fusion and the site of fusion (Hobolt-Pedersen et al., 2014). Collectively, these findings suggest that the expression level as well as the cellular localization of DC-STAMP determine the selection of the fusion partner and the frequency of cell–cell fusion.

RANKL-ITAM-ITIM signaling

Costimulatory signals, one from RANK and one from immunoreceptor tyrosine-based activation motif-containing (ITAM-containing) receptors/adaptors, are required for the activation of osteoclastogenesis in osteoclast precursors (Humphrey et al., 2005; Nakashima and Takayanagi, 2009; Li et al., 2014). However, it is not well understood how ITAM costimulatory signals are integrated into RANK signaling. As an essential protein of osteoclastogenesis, identification of an ITIM on the cytoplasmic tail of DC-STAMP suggests its role in signaling (Chiu et al., 2012b). Based on a common counter action between the ITAM- and ITIM-bearing receptors in immune regulation (Nimmerjahn and Ravetch, 2007, 2008), DC-STAMP is likely to participate in the network of RANK- ITAM costimulation. The collective interplay between RANK, ITAM- and ITIM-signals determines the consequent outcome of osteoclastogenesis activation in osteoclast precursors (Chiu et al., 2012b). Several studies support the role of DC-STAMP in RANKL-ITAM-ITIM signaling through the NFATc1 axis. Targeting of DC-STAMP by MiR-7b caused a suppression of NFATc1 and c-Fos signaling during osteoclast fusion and differentiation (Dou et al., 2014). Suppression of DC-STAMP results in altered downstream signals and changed fusogenic genes and key regulating genes including NFATc1, c-Fos, Akt, Irf8, Mapk1, and TRAF6 (Matsuo et al., 2004; Yagi et al., 2007; Kim et al., 2008; Zhao et al., 2010). Up to date, there is no direct evidence demonstrating the involvement of DC-STAMP in the RANKL-ITAM-ITIM network. Future studies are required to decipher the interplay, if any.

OB::OC coupling

Although DC-STAMP is previously known to be essential for OC differentiation, our recent studies suggested that DC-STAMP also mediates OB development. This observation, together with the presence of DC-STAMP+ cells at fracture sites and delayed bone healing in DC-STAMP KO mice (manuscript submitted) suggested a role of DC-STAMP in OB::OC coupling. The communication between OC and OB is mutual and bi-directional as reviewed by Charles and Aliprantis (2014). OC communicate with OB by secreting growth factors (TGFb and IGF1) and clastokines, or physical interaction through the engagement of ephrins and Eph receptors (Charles and Aliprantis, 2014). Conversely, OB can induce apoptosis of OC through the FAS ligand/FAS pathway, a previously unrecognized mechanism that has an important role in the maintenance of bone mass in both physiological conditions and osteoporosis induced by ovariectomy (Wang et al., 2015b).

In addition to mutual regulations between OB and OC, the OB::OC coupling can be affected by commonly shared factors which are involved in the regulation of both OB and OC, including IL-17A, NFATc1, and Runx2. IL-17A induces cathepsin K and MMP-9 expression in osteoclasts via celecoxib-blocked prostaglandin E2 in osteoblasts (Zhang et al.,2011). NFATc1 is a transcription factor that regulates many biological pathways including OB and OC differentiation (Stern, 2006; Winslow et al., 2006; Choo et al., 2009; Penolazzi et al., 2011; Sesler and Zayzafoon, 2013). In addition, Pin-1 is the other factor that regulates not only DC-STAMP for OC development, but also Runx2 for OB differentiation (Islam et al., 2014). We are currently examining whether DC-STAMP regulates OB differentiation through its regulation on these molecules.

It will be of interest to investigate whether RANKL-induced DC-STAMP(Hi) and DC-STAMP(low) OC employ distinct communication tools, either growth factors, clastokines, or ephrins/Eph receptors, to communicate with OB in murine and human cells. Of note, OC were found surrounding the cancer cells (Kawano et al., 2011). This finding is in line with our recent finding showing that DC-STAMP+ TRAP+ cells are recruited to the OB proximity at the bone fracture site (unpublished data, manuscript in preparation). The co-localization of OB and OC is not observed in healthy bone without fractures. Our preliminary results suggest that DC-STAMP+ cells respond to unidentified factors by direct communication with OB and are attracted to the OB proximity. Additional work is necessary to further determine the molecular mechanism underlying this regulation.

Fracture repair

Laser capture microdissection (LCM) showed that the expression of DC-STAMP was decreased in osteoclasts from diabetic mice whose bone healing is delayed (Kasahara et al., 2010). This observation, together with our recent finding showing the presence of DC-STAMP+ cells at fresh bone fracture sites (unpublished data, manuscript submitted). As a quick review of OB::OC coupling, DC-STAMP KO mice have delayed bone healing compared to wild-type mice (unpublished data, manuscript in preparation), and elevated circulating DC-STAMP+ cells in human bone fractured patients during the bone healing stage (unpublished data, manuscript in preparation), suggest DC-STAMP has an important function in bone repair DC-STAMP in fracture repair.

Tooth development: Eruption and remodeling

Teeth are part of the skeletal system where active bone remodeling constantly occurs. Osteoclasts are involved in tooth development especially during the stage of tooth eruption (Bradaschia-Correa et al., 2013). Together with other cell subsets and cytokine profiles present at local tooth areas, osteoclasts play an important role in the pathogenesis of periodontal diseases (Garlet et al., 2004; Jager et al., 2005; Bradaschia-Correa et al., 2013; Kondo et al., 2013). A possible involvement of DC-STAMP in tooth development is suggested by an increased rate of tooth malocclusion in DC-STAMP KO mice, a phenotype which is closely linked to the homozygous DC-STAMP−/− loci (unpublished data, manuscript submitted). The malocclusion phenotype of DC-STAMP KO was discovered unexpectedly due to a frequent diet-uptake problem reported from our animal facility. More studies will be necessary to investigate the role of DC-STAMP in teeth development in the context of local cytokine profile including RANKL (Bradaschia-Correa et al., 2013) receptors, and physiological inhibitors and growth factors during peridontal disease progression and tooth eruption.

Conclusions

It has been 15 years since DC-STAMP was initially identified as a novel transmembrane protein preferentially expressed by dendritic cells (Hartgers et al., 2000). Knocking-down DC-STAMP completely abrogated the differentiation of osteoclasts and foreign body giant cells (Vignery, 2005; Yagi et al., 2005), and thus DC-STAMP is currently considered as the master regulator in osteoclastogenesis (Courtial et al., 2012; Islam et al., 2014; Zhang et al., 2014). Despite many significant advances within the past 15 years, the molecular mechanism underlying DC-STAMP-mediated regulation remains largely unknown due to the absence of a DC-STAMP ligand. If DC-STAMP is involved in signaling as suggested by the presence of ITIM on its cytoplasmic tail (Chiu et al., 2012b), dissecting its downstream signaling cascade after ligand engagement and activation is necessary to elucidate DC-STAMP-mediated osteoclastogenesis. The inability to identify the DC-STAMP ligand is analagous to the time before the identification of RANKL ligand. Once RANKL was identified, it open up new avenues of research in osteoclastogenesis (Wong et al., 1997; Yasuda et al., 1998). Thus, identification of the DC-STAMP ligand is an urgent priority, which may lead to a breakthrough in our current knowledge on OC differentiation and activation with similar impacts for OC research as the identification of RANKL (Wong et al., 1997; Yasuda et al., 1998).

In summary, current data indicates that DC-STAMP is not only a master regulator of cell fusion, but it may have other central functions that regulate bone homeostasis. Recent animal models show that DC-STAMP is a valid biomarker of psoriatic arthritis, is a potential marker of bone repair response, and may also play an unknown role in the pathogenesis of Paget’s disease. Additional studies will be required to elucidate the structure of DC-STAMP and to identify its ligand(s). These new discoveries will accelerate the validation of DC-STAMP as a biomarker and therapeutic agent in bone disorders. As several pieces of research data on DC-STAMP have been increasingly accumulated for the past 15 years, we anticipate important but currently unknown aspects of DC-STAMP will be understood in the next 15 years. These new findings will assemble all pieces of puzzles together to provide great promise for treating bone diseases via regulating DC-STAMP-mediated OC activity.