RANK and RANKL
RANK and RANKL Family Members
RANK; receptor activator of nuclear factor-κB, TNFRSF11A; tumor necrosis factor receptor superfamily, member 11a, CD265. RANKL; receptor activator of nuclear factor-κB ligand, TNFSF11; tumor necrosis factor (ligand) superfamily, member 11, OPGL; osteoprotegerin ligand, ODF; osteoclast differentiation factor, TRANCE; TNF-related activation-induced cytokine, CD254.
In the late 1980s, an in vitro coculture system for osteoclast formation was established. This system was shown to require cell-to-cell contact between calvarial cells and bone marrow cells for osteoclast differentiation (Suda et al. 1999). Based on this finding, it was proposed that osteoclastogenesis-supporting mesenchymal lineage cells express an osteoclast differentiation factor (ODF) in the form of a membrane-associated protein (Suda et al. 1999). In the late 1990s, the potential inhibitor of osteoclastogenesis osteoprotegerin (OPG) was cloned. OPG is a decoy receptor that associates with a transmembrane protein of the tumor necrosis factor (TNF) superfamily, OPGL, which turned out to be the long-sought ODF (Takayanagi 2007; Theill et al. 2002). Interestingly, immunologists cloned the same molecule as a stimulator of dendritic cells expressed by T cells, and named it receptor activator of nuclear factor-κB ligand (RANKL), or TNF-related activation-induced cytokine (TRANCE) (Lorenzo et al. 2008). The receptor for RANKL is RANK, a type I transmembrane protein, which assembles into a functional trimer upon ligand binding, which is similar to other members of the TNF receptor family (Nakashima and Takayanagi 2009). The RANK and RANKL system currently provides a paradigm that enables the molecular understanding of the linkage among bone metabolism, the organization of lymphoid tissues, the regulation of body temperature, mammary gland development, and tumorigenesis.
The Role of RANKL in the Bone and the Immune Systems
Mice with a disruption of Rank or Rankl exhibit severe osteopetrosis accompanied by a defect in tooth eruption owing to a complete lack of osteoclasts. These genetic findings clearly demonstrate that RANK and RANKL are essential for osteoclastogenesis in vivo. In contrast, mice lacking Opg exhibit severe osteoporosis due to both an increased number and enhanced activity of osteoclasts (Takayanagi 2007; Theill et al. 2002). In humans, mutations in RANK, RANKL, and OPG have been identified in patients with bone disorders, including familial expansile osteolysis, autosomal recessive osteopetrosis, and juvenile Paget’s disease of bone (Nakashima and Takayanagi 2009).
RANKL functions as a membrane-anchored molecule and is released from the cell surface as a soluble molecule following proteolytic cleavage by matrix metalloproteinases (MMPs) (Nakashima et al. 2000). Both the soluble and membrane-bound forms of RANKL function as agonistic ligands for RANK. However, previous reports have suggested that membrane-bound RANKL is more efficient than soluble RANKL (Nakashima and Takayanagi 2009). In addition, previous studies have indicated that RANKL serves as both a chemotactic and survival factor for osteoclasts, and that RANKL is mainly expressed in cells of mesenchymal lineage such as osteoblasts, bone marrow stromal cells, and synovial cells. RANKL expression can be upregulated by certain osteoclastogenic factors such as vitamin D3, prostaglandin E2, parathyroid hormone, interleukin (IL)-1, IL-6, IL-11, IL-17, and TNF-a (Nakashima et al. 2000; Theill et al. 2002). However, the major source of RANKL in vivo remains unclear, since RANKL is expressed by several different cell types in both the bone and bone marrow, including osteoblasts, osteocytes, bone marrow stromal cells, and lymphocytes. A recent report demonstrated that osteocytes embedded within the bone matrix both express a much higher amount of RANKL and have a much greater capacity to support osteoclastogenesis than osteoblasts or bone marrow stromal cells. Furthermore, the crucial role of RANKL expressed by osteocytes was confirmed by the severe osteopetrotic phenotype observed in mice specifically lacking RANKL in osteocytes. These results clearly indicate that the osteocytes are the major source of RANKL in bone remodeling in vivo (Nakashima et al. 2011).
Intriguingly, in addition to the defect in osteoclasts, both RANK- and RANKL-deficient mice are defective in the development and organization of secondary lymphoid tissue (Takayanagi 2007; Theill et al. 2002). However, RANKL-deficient mice also have a reduced thymus size and impaired thymocyte differentiation. Although the mRNA of RANK is present in the thymus of RANKL-deficient mice, RANK-deficient mice do not display any obvious defects in thymocytes. This phenotypic difference in the thymus is the only evident distinction between RANK- and RANKL-deficient mice (Lorenzo et al. 2008; Takayanagi 2007; Theill et al. 2002). This observation suggests that RANKL has the potential to act on another receptor during the course of thymocyte development, a subject which remains to be investigated further. Severe immunodeficiency is not observed in RANKL-deficient mice, nor are there any obvious adverse effects in the immune system due to the administration of anti-RANKL antibody in humans (McClung et al. 2006). The loss of RANKL in T cells seems to be compensated by CD40L in mice (Lorenzo et al. 2008). These observations initially suggested that the immunological function of RANKL is of lesser importance, but recent studies have revealed a crucial role for RANKL in the immune system. RANKL has been shown to play a critical role in a pathological model of inflammatory bowel disease by stimulating dendritic cells (Nakashima and Takayanagi 2009; Takayanagi 2007), suggesting that RANKL is distinctly involved in the activation of dendritic cells under certain autoimmune conditions. On the other hand, keratinocytes express RANKL in response to ultraviolet stimulation of the skin, which appears to activate Langerhans cells and trigger the expansion of regulatory T (Treg) cells in draining lymph nodes (Nakashima and Takayanagi 2009). Vitamin D3, which is produced in the skin in response to sun exposure, has long been known to have immunosuppressive functions and to induce RANKL on osteoclastogenesis-supporting mesenchymal cells in bone. Thus, the suggested role for RANK/RANKL might be the missing link which mediates sunlight-induced immunosuppression. In addition, recent reports suggest that RANK is a key molecule in the development of autoimmune regulator (Aire)-expressing medullary thymic epithelial cells (mTECs), and cooperation between RANK and CD40 also promotes mTEC development, thereby establishing self-tolerance (Nakashima and Takayanagi 2009). Although the functions of RANKL/RANK in the immune system need to be elucidated in greater detail, the discovery and subsequent functional analysis of RANKL has become the driving force behind advances in the understanding of the osteoimmune axis.
The Intracellular Signal Transduction of RANKL
The essential role of NF-κB in osteoclastogenesis has been demonstrated genetically (Takayanagi 2007). NF-κB p50 and p52 double-deficient mice develop severe osteopetrosis because of a defect in osteoclastogenesis. The upstream kinase complex that mediates the phosphorylation and degradation of inhibitor of NF-κB (IκB) comprising the catalytic subunits IκB kinase a (IKKα), IKKβ, and the non-catalytic subunit IKKκ (also known as NEMO) are also important for RANK signaling and osteoclastogenesis. In mice, IKKβ is required for RANKL-induced osteoclastogenesis both in vitro and in vivo, whereas IKKγ appears to be required only in vitro, not in vivo. Importantly, patients with X-linked osteopetrosis, lymphedema, anhidrotic ectodermal dysplasia, and immunodeficiency (OL-EDA-ID syndrome) bear a X420W point mutation in IKKγ (Nakashima and Takayanagi 2009).
The activator protein 1 (AP-1) transcription factor complex is also essential for osteoclastogenesis (Wagner and Eferl 2005). RANK activates AP-1 through an induction of c-Fos. Induction of c-Fos is dependent on the activation of calcium/calmodulin-dependent protein kinase type IV (CaMKIV) and cyclic AMP-responsive element-binding protein ( CREB) (Sato et al. 2006), but there are several reports that suggest NF-κB is involved in the c-Fos induction (Nakashima and Takayanagi 2009). In addition, c-Fos expression is induced by treatment with macrophage colony-stimulating factor (M-CSF). Recent investigation has revealed that peroxisome proliferator-activated receptor-γ (PPAR-γ) plays an unexpected role in osteoclastogenesis by directly regulating c-Fos expression (Nakashima and Takayanagi 2009). Thus, it appears that the induction of c-Fos is not regulated by a single pathway.
Importantly, RANKL specifically and potently induces nuclear factor of activated T cells, cytoplasmic 1 (NFATc1), the master regulator of osteoclast differentiation, and this induction is dependent on both the TRAF6-NF-κB and c-Fos pathways (Takayanagi et al. 2002). The NFAT family of transcription factors was originally discovered in T cells, but its members are involved in the regulation of a variety of biological systems (Crabtree and Olson 2002). The activation of NFAT is mediated by a specific phosphatase, calcineurin, which is activated by calcium-calmodulin signaling. The essential and sufficient role of the Nfatc1 gene in osteoclastogenesis has been shown both in vitro and in vivo. The Nfatc1 promoter contains NFAT binding sites and NFATc1 specifically autoregulates its own promoter during osteoclastogenesis, thus enabling the robust induction of NFATc1 (Takayanagi 2007). AP-1 containing c-Fos, together with continuous activation of calcium signaling, is crucial for this autoamplification (Takayanagi et al. 2002). NFATc1 regulates a number of osteoclast-specific genes in cooperation with other transcription factors such as AP-1, PU.1, and MITF (Takayanagi 2007). Osteoclasts mature into multinuclear giant cells by the fusion of numerous mononuclear osteoclasts. The expression of fusion-mediating molecules such as the d2 isoform of the vacuolar ATPase Vo domain (Atp6v0d2) and the dendritic cell-specific transmembrane protein (DC-STAMP) is directly regulated by NFATc1 (Nakashima and Takayanagi 2009). A previous study indicated that CREB, activated by CaMKIV, also cooperates with NFATc1 in the activation of osteoclast-specific genes (Fig. 1). On the other hand, NFATc1 activity is negatively regulated during osteoclastogenesis by other transcription factors, such as interferon regulatory factor-8 (IRF-8), B cell lymphoma 6 (Bcl6), and v-maf musculoaponeurotic fibrosarcoma oncogene family protein B (MafB) (Miyauchi et al. 2010; Nishikawa et al. 2010; Zhao et al. 2009). The expression of such negative regulators was observed to be repressed through osteoclastogenesis (Fig. 1). This repression is consistent with the notion that high NFATc1 activity is a prerequisite for efficient osteoclastogenesis, but the mechanism by which the expression of these anti-osteoclastogenic regulators is repressed during RANKL-induced osteoclastogenesis has remained obscure. Recent data indicate that B lymphocyte-induced maturation protein-1 (Blimp1), which is induced by RANKL through NFATc1 during osteoclastogenesis, functions as a transcriptional repressor of anti-osteoclastogenic genes such as IRF-8, Bcl6, and MafB (Nishikawa et al. 2010). Therefore, NFATc1 choreographs the determination of cell fate in the osteoclast lineage by inducing the repression of negative regulators as well as through its effect on positive regulators. However, compared with the wealth of information on RANK signaling in osteoclasts, it is as yet unclear whether RANK uses the same signaling mechanisms in the immune system and other systems.
Phospholipase Cγ (PLCγ), which mediates Ca2+ release from intracellular stores, is crucial for the activation of the key transcription factor NFATc1 via calcineurin (Takayanagi et al. 2002). However, despite the evident importance of the calcium-NFAT pathway, it had long been unclear how RANKL activates calcium signals. RANK belongs to the TNF receptor family, which has yet to be directly connected to calcium signaling. The activation of PLCγ by RANK requires the protein tyrosine kinase Syk, along with immunoreceptor tyrosine-based activation motif (ITAM)-bearing molecules, such as DNAX-activating protein (DAP12) and the Fc receptor common gamma chain (FcRγ) (Koga et al. 2004). In the osteoclast lineage, the immunoglobulin-like receptors (IgLR) associated with DAP12 include triggering receptor expressed in myeloid cells 2 (TREM-2) and signal-regulatory protein ß1 (SIRPß1) while those associated with FcRγ include osteoclast-associated receptor (OSCAR) and paired immunoglobulin-like receptor A (PIR-A). As ITAM signals are essential for osteoclastogenesis, but by themselves cannot induce osteoclastogenesis, these signals are most accurately described as costimulatory signals for RANK. The binding of M-CSF to its receptor c-Fms also generates a signaling complex comprised of phosphorylated DAP12 and the nonreceptor tyrosine kinase Syk (Nakashima and Takayanagi 2009). In addition, mutation of TREM-2 or DAP12 in humans leads to Nasu-Hakola disease, which is characterized by bone cysts (Koga et al. 2004; Takayanagi 2007). Thus, RANKL and M-CSF signals appear to converge on the ITAM signaling pathway (Fig. 1).
It is also conceivable that RANK activates an as yet unknown pathway that specifically synergizes with or upregulates ITAM signaling. Tec family tyrosine kinases such as Btk and Tec are activated by RANK and are involved in the phosphorylation of PLCγ, which leads to the release of calcium from endoplasmic reticulum (ER) through the generation of IP3 (Shinohara et al. 2008). An osteopetrotic phenotype in Tec and Btk double-deficient mice revealed these two kinases play an essential role in the regulation of osteoclastogenesis. Tec and Btk had already been reported to play a key role in proximal BCR signaling, but this study established their crucial role in linking the RANK and ITAM signaling pathways (Fig. 1). This study also identified an osteoclastogenic signaling complex, composed of Tec kinases and scaffold proteins, which affords a new paradigm for understanding the signal transduction mechanisms involved in osteoclast differentiation.
Bone Destruction with Arthritis as a RANKL Disease
In rheumatoid arthritis (RA), a long-standing question is how abnormal T cell activation (characterized by the infiltration of CD4+ T cells) mechanistically induces bone damage. The identification of osteoclast-like giant cells at the interface between synovium and bone in rheumatoid joints dates back to the early 1980s (Takayanagi 2009). These pathological findings led us to hypothesize that osteoclasts play an important role in the bone resorption that occurs in arthritis and that the osteoclasts are formed in the synovium (Takayanagi 2009). Can osteoclasts be generated from synovial cells alone? This question was answered in the affirmative by generating osteoclasts in synovial cell culture without adding any other cells, thus demonstrating that rheumatoid synovial cells contain both osteoclast precursor and osteoclastogenesis-supporting cells (Takayanagi 2009). Further studies indicated that synovial fibroblasts express membrane-bound factor(s) that stimulate osteoclastogenesis and induce the differentiation of synovial macrophages into osteoclasts, but it was not until RANKL was cloned that the membrane-bound factor in synovial cells was brought to light (Takayanagi et al. 2000a).
Importantly, inflammatory cytokines such as IL-1, IL-6, and TNF-a, which are abundant in both the synovial fluid and synovium of RA patients, have a potent capacity to induce RANKL on synovial fibroblasts/osteoblasts and to accelerate RANKL signaling, thus directly contributing to the bone destruction process. Several groups have demonstrated the high expression of RANKL in the synovium of RA patients (Takayanagi 2009). RANKL was shown to be expressed by synovial cells and T cells, both of which are found in the inflamed synovium(Takayanagi 2007, 2009). As RANKL is expressed in activated T cells, T cells may have the capacity to induce osteoclast differentiation by directly acting on osteoclast precursor cells under pathological conditions (Kong et al. 1999). However, interferon-γ (IFN-γ), which is produced by T cells, potently suppresses RANKL signaling through a rapid degradation of TRAF6 (Takayanagi et al. 2000b). To fully understand the effects of T cells on osteoclastogenesis, it is absolutely necessary to elucidate the specific effects of the various cytokines which T cells produce. It has been shown that IL-17-producing Th17 cells are the exclusive osteoclastogenic T cell subset among the known Th subsets (Takayanagi 2007, 2009). Since even Th17 cells stimulate osteoclastogenesis mainly through RANKL induction on synovilal fibroblasts, it is as yet still unclear how T cell RANKL contributes to bone destruction in the face of synovial fibroblasts expressing RANKL to a higher extent.
Nevertheless, a series of reports has established that the bone damage associated with inflammation is the fundamental pathological condition caused by an abnormal expression of RANKL. Additionally, osteoclast-deficient mice and osteopetorosis patients are protected from bone erosion in arthritis (Kadono et al. 2009; Takayanagi 2009). In the absence of osteoclasts, bone destruction did not occur, despite a similar level of inflammation, indicating that RANKL and osteoclasts are indispensable for the bone loss associated with inflammation. Blocking RANKL by OPG treatment significantly prevented bone destruction in adjuvant arthritis (Kong et al. 1999). Consistent with this, anti-RANKL and anti-osteoclast therapies have been shown in clinical trials as well as in the treatment of an animal model of arthritis to be beneficial for the inhibition of bone loss without affecting the immune system (Takayanagi 2007).
RANK and RANKL in Mammary Gland Development and Tumorigenesis
During pregnancy, increased ductal side branching and the development of lobuloalveolar structures are the result of an expansion and proliferation of ductal and alveolar epithelium (Hennighausen and Robinson 2005). Previously, genetic findings demonstrated that mice with a disruption of Rank or Rankl fail to develop mammary glands during pregnancy, resulting in the death of newborns (Fata et al. 2000). These mice exhibit normal mammary development and normal ductal elongation and side-branching of the mammary epithelial tree into the mammary fat pad during puberty. However, their mammary epithelium fails to proliferate and form lobuloalveolar structures during pregnancy (Fata et al. 2000). The mammary gland defect in female RANKL-deficient mice can be reversed by recombinant RANKL treatment. These data clearly indicate that RANKL is an essential regulator of alveolar epithelial cell proliferation. Although RANK is constitutively expressed on mammary epithelial cells, the expression of RANKL is absent in virgin glands, but gradually increases during pregnancy. RANKL expression in mammary epithelial cells is induced by pregnancy hormones such as prolactin, progesterone, and PTHrP (Fata et al. 2000). A previous report showed that kinase-dead IKKa mutant mice display a severe lactation defect due to the impaired proliferation of mammary epithelial cells (Cao et al. 2001). The phenotype can be rescued by mammary-specific overexpression of cyclin D1. These data suggest that IKKa activity in response to RANKL is required for NF-κB activation and cyclin D1 induction in mammary epithelial cells during pregnancy. However, it is reported that cyclin D1 is normally expressed in mammary epithelial cells of RANK-deficient mice, but in these animals there is a defect in nuclear translocation of the basic helix-loop-helix transcriptional regulator, inhibitor of DNA binding 2 (Id2) (Kim et al. 2006). Genetic deletion of Id2 results in a similar phenotype having a lactation defect. Id2 regulates the proliferation of mammary epithelial cells through a suppression of the cell cycle inhibitor p21 in response to RANKL. Thus, RANK/RANKL plays an essential role in mammary gland development, but further study is required to completely elucidate the signaling pathways and transcriptional regulators.
The mammary gland in the period from puberty to menopause develops through tightly choreographed stages of cell proliferation (Hennighausen and Robinson 2005). Steroid hormones such as estrogen and progesterone have a prominent role in both the healthy and diseased states of breast tissue. Reproductive history is the strongest risk factor for breast cancer, and increased risk of breast cancer is correlated with a greater number of ovarian hormone-dependent reproductive cycles (Beral et al. 2005). There is also an increased risk of breast cancer associated with pregnancy in the short term. Although a proliferative role for the steroid hormones in this gland is well accepted, it is still unclear how the mammary gland translates hormonal signals into cell proliferation. Recent studies have implied that RANK/RANKL functions in mammary stem cell (MaSC) biology. MaSC is defined as a cell that can both self-renew and propagate the full spectrum of cell types that make up the mammary gland (Shackleton et al. 2006). The MaSC activity that was increased in mice treated with steroid hormones and pregnancy led to a dramatically increased number of MaSC in mice (Asselin-Labat et al. 2010; Joshi et al. 2010). In contrast, ovariectomy or aromatase inhibitor treatment markedly reduced the MaSC number and outgrowth potential in vivo (Asselin-Labat et al. 2010). In aged mice, MaSCs also display stasis upon cessation of the reproductive cycle (Joshi et al. 2010). MaSCs carry no known receptors for estrogen or progesterone, but these stem cells are highly responsive to steroid hormone signaling. Studies have shown that neutralization of RANKL in pregnant mice reduces the capacity of the MaSC-enriched basal cell population to form colonies. These data suggest that RANKL, a known progesterone target, may act as a crucial molecule that links progesterone-responsive mammary cells to MaSCs.
Hormone replacement therapy (HRT) is associated with an increased risk of breast cancer (Beral et al. 2005). In particular, progesterones or their synthetic derivatives (progestins) such as medroxyprogesterone acetate (MPA) markedly increase the risk of an abnormal mammogram and breast cancer. Recently, it was revealed that MPA treatment triggers the induction of RANKL expression in progesterone receptor (PR)-positive luminal mammary epithelial cells, resulting in autocrine or/and paracrine stimulation of RANK signaling in the mammary epithelium (Schramek et al. 2010). Importantly, specific deletion of RANK in mammary epithelium cells prevents both the onset and progression of MPA-driven mammary cancer and impairs self-renewal of breast cancer stem cells. Tumorigenesis in an MPA-driven tumor model as well as in a spontaneous tumor model was also inhibited by a neutralizing antibody to RANKL (Gonzalez-Suarez et al. 2010). In contrast, mammary-specific overexpression of RANK results in the acceleration of preneoplasias of the mammary glands and an increase in mammary tumor formation after either multiparity or treatment with a carcinogen and progestin (Gonzalez-Suarez et al. 2010). These findings show that the RANK/RANKL system is crucial for tumorigenesis.
Bone is the most common site for the distal spread of breast and prostate cancer (Mundy 2002). Bone metastases result in serious morbidity, including skeletal-related events such as pain, fractures, and hypercalcemia, increasing the mortality risk. Therefore, the clinical priority is to prevent metastases and bone loss owing to excessive osteoclastic bone resorption. Indeed, many clinical trials have evaluated the potential activity of anti-osteoclastic agents in cancer having bone metastases. The representative anti-osteoclastic agents include bisphosphonates and a RANKL neutralizing antibody (Denosumab) (Fornier 2010). A recent clinical study showed that Denosumab is superior to bisphosphonates such as Zoledronic acid for the delay or prevention of skeletal-related events in patients with advanced breast cancer with bone metastases (Stopeck et al. 2010). Interestingly, RANK is highly expressed in several human breast cancer cell lines and primary human breast tumors (Jones et al. 2006). Functionally, it has been shown that RANKL can stimulate the directed migration of mammary epithelial cells as well as prostate cancer and melanoma cells toward a source of RANKL. Furthermore, in an in vivo metastasis model, OPG reduced the tumor burden in bones and ameliorated clinical paralysis, but did not affect the frequency of the spread of metastases into other tissues (Jones et al. 2006). A recent clinical study reported that the level of RANK expression in primary breast cancer positively correlates with the development of bone metastases and may be a predictive marker of bone metastasis risk (Santini et al. 2011). Similar to breast cancer, prostate cancer metastasizes to bone through RANK signaling. Previous reports showed that IKKa activation by RANKL inhibits the expression of Maspin, a metastasis suppressor in prostate epithelial cells. Maspin expression reduced metastatic activity, whereas Maspin ablation restored this activity (Luo et al. 2007). These findings suggest that RANK-expressing tumor cells might sense RANKL as a chemoattractant and migrate in a coordinated fashion to a source of RANKL produced in the bone.
Bone-related diseases such as osteoporosis and RA afflict a great number of patients. Women taking progesterone derivatives for contraception or HRT have been shown epidemiologically to have an increased risk of breast cancer. These diseases are presenting a tremendous burden to the health care costs. Genetic approaches have established that the RANKL/RANK system is the central regulator of osteoclastogenesis, lymph node organogenesis, mammary gland development, and thymic epithelial cell development. In addition, recent data have revealed an entirely novel and unexpected function for RANKL/RANK in female thermoregulation and the central fever response (Hanada et al. 2009). RANKL has attracted the attention of scientists and pharmaceutical companies, since it plays a pivotal role in the pathogenesis of osteoporosis, RA, tumorigenesis, and metastasis. Novel drugs specifically targeting RANK/RANKL and their signaling pathways provide a potential means to revolutionize the treatment of various diseases associated with this pathway (Kearns et al. 2008).
- Nakashima T, Kobayashi Y, Yamasaki S, Kawakami A, Eguchi K, Sasaki H, et al. Protein expression and functional difference of membrane-bound and soluble receptor activator of NF-κB ligand: modulation of the expression by osteotropic factors and cytokines. Biochem Biophys Res Commun. 2000;275:768–75.PubMedCrossRefGoogle Scholar