IBMS BoneKEy | Article

Mysteries in Ca2+ signaling during osteoclast differentiation

Takako Negishi-Koga
Hiroshi Takayanagi

DOI:10.1138/20090393

Nuclear factor of activated T cells c1 (NFATc1) activation by Ca2+ signaling is an essential part of the signaling axis in osteoclast differentiation. Although the molecular signaling cascade downstream of receptor activator of nuclear factor-κB ligand (RANKL) has been extensively studied, the origin of the intracellular Ca2+ increase and the dynamics required for efficient NFATc1 activation have been unclear. Using mice lacking inositol 1,4,5-triphosphate (IP3) receptor (IP3R) type 2 (IP3R2) and IP3R3, a study from last year by Kuroda and coauthors found that RANKL-induced Ca2+ oscillation and osteoclast differentiation were impaired in vitro, but that there was no defect in osteoclastogenesis in vivo (). To understand this discrepancy, the authors focused on the role of osteoblasts, which have the ability to induce osteoclastogenesis in IP3R2/3-deficient cells. Surprisingly, Ca2+ oscillation was not detected in the coculture system, but NFATc1 induction and osteoclastogenesis were observed. Their findings suggest the existence of an as yet unidentified (possibly Ca2+ oscillation-independent) regulatory mechanism for NFATc1 activation mediated by osteoblasts, and may open up new directions for research on Ca2+ signaling during osteoclastogenesis.

RANKL signaling during osteoclastogenesis

Osteoclasts are cells of monocyte-macrophage origin that decalcify and degrade the bone matrix. The differentiation of osteoclasts is regulated primarily by three signaling pathways that are activated by RANKL, macrophage colony-stimulating factor (M-CSF), and immunoreceptor tyrosine-based activation motif (ITAM) (). Whereas M-CSF promotes the proliferation and survival of bone marrow monocyte/macrophage precursor cells (BMMs) (), RANKL activates the differentiation process by inducing the master transcription factor for osteoclastogenesis, NFATc1.

The robust induction of NFATc1 is dependent on Ca2+ signaling, which is mediated by the activation of ITAM in adaptor molecules such as DNAX-activating protein 12 (DAP12) and Fc receptor common γ subunit (FcRγ) (). DAP12 and FcRγ associate with costimulatory receptors of the immunoglobulin superfamily, including osteoclast-associated receptor (OSCAR), triggering receptor expressed in myeloid cells-2 (TREM-2), signal-regulatory protein β1 (SIRPβ1), and paired immunoglobulin-like receptor-A (PIR-A) (). Although the ligands of the immunoglobulin-like receptors remain to be elucidated, it has been shown that osteoblasts supply putative ligands for costimulatory receptors associated with FcRγ, suggesting an important role of osteoblasts in osteoclast costimulation ().

The phosphorylation of ITAM results in the recruitment of the nonreceptor tyrosine kinase Syk. Meanwhile, RANKL stimulates Tec tyrosine kinases to form a complex with Syk, which activates phospholipase Cγ (PLCγ) (). It is well-established that PLCγ generates IP3, which evokes Ca2+ release from the endoplasmic reticulum (ER) through IP3Rs, with subsequent Ca2+ entry from the extracellular milieu taking place through plasma membrane channels (). Although this mechanism has not been confirmed in osteoclasts, Ca2+ oscillation induced by RANKL has been thought to be important for the efficient activation of NFATc1 via the Ca2+-dependent phosphatase calcineurin (). Consistent with this, in most previous reports, osteoclastogenesis was impaired when Ca2+ oscillation was not observed (). As osteoclastogenesis is also impaired in PLCγ2-deficient mice, it is expected that IP3-mediated Ca2+ signaling plays an important role in osteoclastogenesis (Fig. 1).

In vivo and in vitro analysis of IP3R2/3-deficient mice

To address the physiological role of IP3-mediated signals in the osteoclast lineage, Kuroda and colleagues () investigated the bone phenotype of IP3R2/3-deficient mice, because they had found IP3R2 and IP3R3 to be the major IP3Rs in osteoclasts among the 3 known types of IP3Rs. However, bone mass and osteoclast number were normal in these mice, suggesting that IP3-mediated Ca2+ signaling is dispensable for osteoclastogenesis, or that IP3R2/3 are dispensable for transmitting IP3-mediated Ca2+ signals. IP3R2/3 are reportedly essential for transmitting IP3-mediated Ca2+ signaling under certain conditions (), but there is little information regarding the osteoclast lineage.

The authors then focused on the observation that Ca2+ oscillation during in vitro osteoclastogenesis was impaired in IP3R2/3-deficient osteoclasts 48 hours after RANKL stimulation (). They concluded that the combined deficiency of IP3R2 and IP3R3 results in impaired osteoclastogenesis in vitro due to a lack of Ca2+ oscillation and suggested that IP3R2 and IP3R3 are essential for Ca2+ oscillation in osteoclasts. However, it was also observed that impairment of osteoclast differentiation from IP3R2/3-deficient BMMs was partially rescued by coculturing with osteoblasts. They proposed that the result in this coculture condition was consistent with the in vivo observation that osteoclastogenesis was normal in IP3R2/3-deficient mice ().

Surprisingly, they could not detect Ca2+ oscillation in IP3R2/3-deficient multinuclear osteoclasts, although NFATc1 induction and osteoclast differentiation were observed. Furthermore, FK506 did not affect osteoclastogenesis in IP3R2/3-deficient BMMs in coculture with osteoblasts (Table 1). Thus, the authors proposed the interesting hypotheses that 1) Ca2+ oscillation is not required for NFATc1 activation and osteoclast formation; 2) calcineurin is not essential for NFATc1 activation and osteoclast formation; and 3) oscillation -independent calcineurin-independent osteoclastogenesis is supported by osteoblasts.

Is NFATc1 activation dependent on Ca2+ oscillation?

NFAT is activated by a low, but sustained Ca2+ plateau level (), and Ca2+ oscillation is thought to be beneficial in keeping NFATc1 in the nucleus and ensuring the long-lasting transcriptional activation of NFATc1 that is required for terminal differentiation into osteoclasts (). However, the genetic evidence for the Ca2+ signaling pattern required for osteoclastogenesis is lacking. The rare example is RGS10-deficient mice, which exhibit high bone mass and impaired osteoclastogenesis (). RGS10 is important for RANKL-induced Ca2+ oscillation, suggesting the importance of Ca2+ oscillation in osteoclastogenesis. However, it is difficult to rule out the possibility that disruption of RGS10 affects other types of Ca2+ signaling.

Usually Ca2+ oscillation becomes detectable approximately 24 hours after RANKL stimulation, when NFATc1 induction becomes evident (). Therefore, it is likely that Ca2+ signaling patterns other than oscillation are important for NFATc1 activation in the earlier phase. It is unclear up until which stage osteoclastogenesis is dependent on oscillation (). Thus, it remains to be determined which type of Ca2+ signal is absolutely necessary for osteoclastogenesis in each differentiation stage. In this study (), Ca2+ oscillation was evaluated at a limited number of time points, and it can be difficult to detect Ca2+ signaling in mixed cell culture systems like cocultures of osteoblasts and osteoclast precursor cells. It will first be necessary to determine the Ca2+ signaling pattern required for osteoclastogenesis at multiple stages and then to analyze the differences in Ca2+ signaling events in knockout mice, which together will provide the requisite fundamental information on the role of the distinct Ca2+ signaling pattern.

Is NFATc1 activation dependent on calcineurin?

It is widely accepted that calcineurin is a central regulator of NFAT and that immunosuppressants such as cyclosporine A and FK506 inhibit the activation of calcineurin, thereby blocking the nuclear import and transcriptional activation of NFAT. However, the authors demonstrated that FK506 only partially inhibited osteoblast-mediated osteoclastogenesis and did not inhibit osteoblast-mediated osteoclastogenesis in IP3R2/3-deficient cells, suggesting that NFATc1 activation is not completely dependent on calcineurin. Recent reports have suggested calcineurin-independent NFAT activation mechanisms (), but it is important to be careful about reaching conclusions based only on inhibitor or overexpression experiments. Since calcineurin-deficient mice are available, it will be important to analyze NFATc1 activation and osteoclastogenesis in these mice.

What compensates for the loss of IP3R2/3?

Since there is no obvious bone phenotype in IP3R2/3-deficient mice, certain compensatory mechanism(s) are obviously at work in these animals. The authors () propose that osteoblasts mediate this mechanism in a Ca2+ oscillation-independent manner, but is this compensatory mechanism Ca2+-independent? Considering the increased bone mass in PLCγ2-deficient mice, it is highly likely that PLCγ-mediated IP3 production and subsequent activation of ER Ca2+ release play a substantial role. Therefore, it is plausible that IP3-mediated Ca2+ signaling would have to be achieved by the compensatory mechanism. One possibility is that IP3Rs other than IP3R2/3 (possibly IP3R1) may be functional.

In non-excitable cells, IP3Rs function as major Ca2+ channels that release Ca2+ from the ER, but it is also reported that Ca2+ release from the ER is mediated in an IP3-independent manner via ryanodine receptors (RyRs) and/or nicotinic acid adenine dinucleotide phosphate (NAADP) receptors (). Another possibility is that IP3-independent Ca2+ release (possibly mediated by RyRs, or NAADP receptors, etc.) is compensatorily upregulated by an as yet unknown mechanism (Fig. 1).

The remaining issue is why these complex compensatory mechanisms are activated only in vivo or in the presence of osteoblasts. It is of crucial importance to identify the compensatory mechanism, which will provide molecular insight into bone remodeling based on the functional coupling between osteoblasts and osteoclasts.

New directions for research into osteoclastogenesis

It is often the case that genetically-modified mice exhibit no obvious phenotype, despite a defect observed in suggestive in vitro experiments. The study from Kuroda et al. () provided an interesting example in which osteoblast-mediated signals are compensatory for the loss of certain types of Ca2+ signaling. These findings have shed light on an unexpected mechanism underlying osteoblast-mediated NFATc1 activation, which will lead to new directions in research into NFATc1 regulation and Ca2+ signaling.

This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 United States License.