Osteoclastogenesis and arthritis
- First Online:
- Cite this article as:
- Maruotti, N., Grano, M., Colucci, S. et al. Clin Exp Med (2011) 11: 137. doi:10.1007/s10238-010-0117-2
- 332 Views
There is emerging interest for osteoclasts as key players in the erosive and inflammatory events leading to joint destruction in chronic arthritis. In fact, chronic inflammatory joint diseases such as psoriatic arthritis and rheumatoid arthritis are often characterized by destruction of juxta-articular bone and erosions due to the elevated activity of osteoclasts, which are involved in bone resorption. The main step in inflammatory bone erosion is an imbalance between bone resorption and bone formation: osteoclast formation is enhanced by proinflammatory cytokines such as TNF-α, IL-1β, and IL-17 and is not balanced by increased activity of bone-forming osteoblasts. T-cells, stromal cells, and synoviocytes enhance osteoclast formation via expression of RANKL and, under pathologic conditions, of proinflammatory cytokines. In rheumatoid arthritis, accumulation of osteoclasts in synovial tissues and their activation associated with osteoclastogenic cytokines and chemokines at cartilage erosion sites suggest that they could be usefully selected as therapeutic target. In particular, in consideration of the primary role of RANKL and TNF-α in osteoclastogenesis, the control of the production of RANKL and the inhibition of TNF-α represent important strategies for reducing bone damage in this disease.
Chronic inflammatory joint diseases such as psoriatic arthritis (PsA) and rheumatoid arthritis (RA) are often characterized by destruction of juxta-articular bone and erosions due to the elevated activity of multinucleated cells, called osteoclasts, which are involved in bone resorption. The main step in inflammatory bone erosion is an imbalance between bone resorption and bone formation: osteoclast formation is enhanced by proinflammatory cytokines such as tumor necrosis factor-α (TNF-α), interleukin-1β (IL-1β), and IL-17 and is not balanced by increased activity of bone-forming osteoblasts . T-cells, stromal cells, and synoviocytes enhance osteoclast formation via expression of nuclear factor (NF)-κB ligand (RANKL) and, under pathologic conditions, of proinflammatory cytokines . Osteoclast formation and differentiation have been described in the inflamed synovial membrane, where they derive from precursors of the monocyte/macrophage lineage. Although histology of periarticular trabecular bone showed that osteoclastic bone resorption is significantly stimulated in chronic arthritis, such as RA, the real mechanism of the joint damage is still unknown.
RANKL induces the formation of the activator protein 1 (AP-1)/nuclear factor of activated T-cells cytoplasmic 1 (NFATc1) complex via inducing the expression of the c-Fos family and favouring c-Jun signaling [10, 13]. RANKL stimulates NFATc1 via TRAF6, NF-κB, and c-Fos induction . Numerous osteoclast-specific genes are controlled by NFATc1, in addition to other transcription factors such as AP-1 and MITF . Phospholipase Cc (PLCc) is involved in NFATc1 activation, via Ca2+ release from intracellular supplies . Protein tyrosine kinase Syk and immunoreceptor tyrosine-based activation motif (ITAM)-bearing molecules such as DNAX-activating protein (DAP12) and the Fc receptor common gamma chain (FcRc) are necessary for the activation of PLCc by RANK .
The main negative regulator of RANKL activity is osteoprotegerin (OPG), which is a soluble decoy receptor for RANKL produced by osteoblasts . This decoy receptor competitively inhibits the binding of RANKL to RANK on the cell membrane of osteoclasts, thus preventing RANK activation and the consequent osteoclastogenesis.
The role of osteoclasts in chronic arthritis
The role of osteoclasts in chronic arthritis has emerged in the last 25 years. In 1984, Bromley and Woolley found osteoclast-like cells at the bone destruction site . Later Takayanagi et al.  demonstrated that synovial fibroblasts induced osteoclastogenesis in cultures obtained by RA patients. In the following years, evidences of osteoclast role in RA continued to accumulate. RANKL expression was demonstrated in synovial fibroblasts and activated T-cells derived from synovial tissues . By using tissue sections from the bone-pannus interface at sites of bone erosions in RA patients, Gravallese et al.  identified osteoclast precursors in bone resorption lacunae within synovial pannus, and mature osteoclasts in bone marrow and within synovial pannus, by the identification of messenger RNA (mRNA) for tartrate-resistant acid phosphatase (TRAP) and cathepsin K in mononuclear cells and in multinucleated cells, respectively. Moreover, through a reverse transcriptase–polymerase chain reaction (RT–PCR), Gravallese et al.  identified mRNA for RANKL in synovial fibroblasts and activated T lymphocytes obtained from patients with RA.
By using cocultures of rheumatoid synovial fibroblasts and peripheral blood mononuclear cells in the presence of macrophage colony stimulating factor and 1,25-dihydroxyvitamin D3, Takayanagi et al.  found that synovial fibroblasts did not induce osteoclastogenesis when cocultured with peripheral blood mononuclear cells. On the contrary, cultured rheumatoid synovial fibroblasts induced osteoclastogenesis in the presence of 1,25(OH)2D3, with an up-regulation of RANKL expression . RANKL knockout mice, which are deficient in osteoclasts, were protected from bone erosion in a serum transfer model of arthritis .
Osteoclastogenesis plays a role even in PsA. In fact, Colucci et al.  demonstrated that B and T-cells and fibroblasts support osteoclastogenesis in PsA patients via the production of osteoclastogenic cytokines responsible for the extensive bone resorption, such as RANKL, TNF-α, and IL-7, while fibroblasts from synovial fluid produce only RANKL.
While bone erosions represent the most debilitating and irreversible aspect in RA and PsA, arthritis in systemic lupus erythematosus (SLE) patients is often characterized by the lack of erosions, probably due to IFN-α-induced inhibition of bone marrow-derived myeloid precursors differentiation in osteoclasts .
Cytokines and chemokines involved in osteoclastogenesis
While RANK-RANKL-osteoprotegerin system is the principal regulator of osteoclastogenesis in physiological conditions, in different situations such as postmenopausal osteoporosis and chronic inflammatory joint diseases including PsA and RA, proinflammatory cytokines, together with the RANK-RANKL complex, have an important role in inducing osteoclast differentiation and activation [25, 26]. TNF-α and RANKL influence each other in osteoclast activation . Moreover, in vitro and murine studies have hypothesized that TNF-α may induce osteoclastogenesis even in the absence of RANK signaling [28, 29]. Ochi et al.  showed that TNF-α stimulates osteoclastogenesis via inducing paired Ig-like receptor-A (PIR-A), a costimulatory receptor for RANK.
Abu-Amer et al.  found that bone marrow derived from mice expressing only TNF receptor type 1 (p55r) produced a larger number of osteoclasts than mice expressing TNF receptor type 2 (p75r). In consideration of this evidence, it has been hypothesized that TNF receptor type 1 (p55r), through the binding to soluble TNF-α, mediates the osteoclastogenic effect of TNF-α, while TNF receptor type 2 (p75r), which is the favored receptor of the membrane-associated TNF-α, has a negligible role in osteoclastogenesis. CD40 is another member of the TNF receptor family which is involved in osteoclastogenesis as demonstrated in cultures of CD40-ligated RA synovial fibroblasts (RASFs) expressing RANKL mainly via NF-kB activation.
TNF-α induces osteoclast precursor and marrow stromal cells, which express TNF receptor type 1 (p55r), to produce osteoclastogenic cytokines, such as IL-1, RANKL, and macrophage colony-stimulating factor (M-CSF) .
IL-1 is responsible for osteoclastogenic effects only in the presence of adequate levels of RANKL and induces the activation of a p38 MAP-kinase in osteoclast precursor and marrow stromal cells which is involved in TNF-α-mediated osteoclastogenesis .
M-CSF, which is synthesized by marrow stromal cells, is capable of recruiting osteoclasts, as demonstrated in mice characterized by a mutation in the coding region of the M-CSF, which developed an osteoclast-deficient osteopetrosis . M-CSF induces RANK expression on the cell surface of pre-osteoclasts rendering them responsive to the osteoclastogenic effects of RANKL . M-CSF mediates its osteoclastogenic effects by binding to c-fms, a tyrosine kinase receptor that induces osteoclast differentiation via activation of ERK1/2 and PI3-K/AKT .
Transforming growth factor-β (TGF-β) affects both RANKL-dependent and TNF-α-induced osteoclast differentiation . Moreover, TGF-β reduces RANKL expression and is involved in osteoblast and bone marrow cell production of OPG . Hase et al.  recently reported that TGF-β reduces RA synovial fibroblast production of OPG by down-regulation of TGF-β/Smad2 signaling and may play a role in RANKL-mediated osteoclastogenesis.
Proinflammatory cytokines, including TNF-α, IL-1, IL-6 and IL-17 play a role in the RA-related bone damage [39, 40]. Among these cytokines, TNF-α received great considerations because numerous studies showed the importance of anti-TNF therapy in reducing both bone erosions and inflammation and inducing apoptosis in macrophages in RA joints [41–44]. Even if it is well accepted that TNF-α and RANKL influence each other in osteoclast activation, it remains controversial whether TNF-α may induce osteoclastogenesis even in the absence of RANK signaling [27–29]. IL-1, IL-6, IL-6 receptors (sIL-6R), and IL-17 levels were found significantly higher in the synovial fluids of RA patients than in synovial fluids of osteoarthritis patients [39, 45]. Furthermore, IL-17 stimulates macrophage production of TNF-α, IL-1, and IL-6 . IL-17 has a role in osteoclastogenesis through an interaction between osteoclast progenitors and osteoblasts . Since IL-17 was involved both in prostaglandin E2 (PGE2) synthesis and in RANKL mRNA expression in cultures of osteoblasts, Kotake et al.  hypothesized that IL-17 may induce osteoblast-related COX-2-dependent PGE2 synthesis, which is responsible for RANKL expression. Moreover, it has been demonstrated that IL-17 is involved in joint degradation in juxta-articular bone and synovium explants from RA patients . Independent from RANKL signaling, recombinant human IL-23 may stimulate osteoclastogenesis in macrophage colony-stimulating factor-differentiated human peripheral blood mononuclear cells, and this process is dependent on IL-17, TNF-α, and osteoprotegerin .
Other cytokines involved in osteoclastogenesis are IL-7 and IL-11. IL-7 induces osteoclastogenesis by up-regulating the T-cell production of osteoclastogenic factors, such as RANKL and TNF-α . IL-11 has been reported to sustain osteoclast formation through RANKL expression .
Chemokines are important mediators in inflammation and immune responses by stimulating leukocyte chemotaxis and increasing the number of newly formed vessels . The distinction between the CXC, CC, and CX3C chemokine families is based on the presence or absence of an amino acid, X, between a pair of cysteine residues near the amino terminus . Numerous chemokines, including macrophage inflammatory protein 1α (MIP-1α), MIP-1β, interferon-γ-inducible protein 10 (IP-10; also called CXCL10), and monokine induced by IFN-γ (MIG), are involved in osteoclast differentiation and survival [53–55]. Abe et al.  showed that MIP-1α and MIP-1β are involved in osteolytic lesions in patients with multiple myeloma by favouring osteoclast differentiation. IP-10 plays an immune regulatory role by inducing activation and chemoattraction of T-cells, monocytes, and eosinophils [56, 57]. To investigate the relation between IP-10 and chronic inflammation, levels of IP-10 were examined in RA patients. IP-10 was significantly amplified in synovial fluid of RA patients, suggesting therefore that IP-10 plays an important role in the recruitment of T-cells to sites of inflammation . A recent in vivo study using a mouse model of collagen-induced arthritis found that IP-10 is increased in inflamed joints . It would appear, therefore, that IP-10 plays an important role in the progression of RA and promotes RANKL and TNF-α expression in the inflamed joints.
Osteoclastogenic and anti-osteoclastogenic factors
RANK-RANKL complex 
Soluble OSCAR 
Negative regulators of osteoclastogenesis
Among the inhibitor of osteoclastogenesis, IFN-α and IFN-β have a primary role . The binding of IFN-β to its receptor causes a signal cascade through the JAK/STAT pathway, responsible for inhibition of c-fos protein production and osteoclast proliferation and differentiation . IFN-γ is involved in the inhibition of RANK-RANKL-induced osteoclastogenesis and constitutes a negative feedback regulation of RANKL signaling. In fact, on the one hand RANKL induces the IFN-β gene in osteoclast precursor cells, on the other hand IFN-γ interferes with the osteoclast differentiation induced by RANKL .
Among interleukins, IL-2, IL-10, and IL-27 have a role as negative regulators of osteoclastogenesis. IL-2 has been proposed to suppress bone erosion . IL-10 has an inhibitory effect on RANK-induced osteoclastogenesis via suppressing triggering receptor expressed on myeloid cells 2 (TREM-2) . IL-27 inhibits the expression of RANK, NFATc1, ERK, p38, and NF-kB in osteoclast precursors. IL-27 has been identified in synovium where it plays an anti-inflammatory role and has a homeostatic effect in preventing bone erosions. In chronic inflammation, such as RA, this homeostatic role is inhibited .
IL-4 and granulocyte–macrophage colony-stimulating factor (GM-CSF) cause monocytic differentiation into dendritic cells (DCs) and stop monocytic differentiation into osteoclasts. IL-4 and GM-CSF may up-regulate TNF-α-converting enzyme (TACE) in monocytes, causing the inhibition of osteoclastogenesis through M-CSF receptor shedding .
Interferon regulatory factor 8 (IRF8) is a transcription factor expressed in immune cells, which has an inhibitory effect on osteoclastogenesis by inhibiting the function and expression of NFATc1  (Table 1).
Toll-like receptors (TLRs) have a regulating role in osteoclastogenesis both in promoting osteoclast-mediated bone resorption related to inflammatory conditions and in suppressing osteoclast development and differentiation via inhibition of RANK expression in cooperation with IFN-γ .
Independent from cytokine stimulation, there is a “costimulation axis” of osteoclasts, characterized by the presence of the osteoclast-associated receptor (OSCAR), a member of the leukocyte receptor complex which is expressed on osteoclasts, monocytes, and DCs [68–70]. OSCAR expression is up-regulated by microphthalmia-inducing transcription factor (MITF) and down-regulated by MafB and inhibitor of differentiation 2 [71–73]. Moreover, it has been demonstrated that TNF-α stimulates OSCAR expression in monocytes . On the contrary, RANKL and M-CSF do not induce directly OSCAR expression. Even if its ligand has not been still isolated, it is known that OSCAR is responsible for inducing osteoclast differentiation through an additive action with primary stimulation by RANKL. In patients affected by RA, OSCAR expression has been demonstrated in monocytes within the inflamed synovium and in mature osteoclasts within the inflamed synovial tissue and bone interface, where OSCAR promotes RANKL stimulation . Moreover, Herman et al.  found that membrane-bound OSCAR expression is amplified in circulating monocytes of RA patients, in direct relation to the inflammatory disease activity. As consequence of this amplified OSCAR expression, monocytes in RA patients showed a higher tendency to differentiate into osteoclasts than cells with low OSCAR expression. On the contrary, circulating soluble OSCAR expression is low in RA patients and inversely related to inflammatory disease activity, probably because soluble OSCAR may be a decoy receptor for the unknown ligand of membrane-bound OSCAR .
Nevertheless, OSCAR pathway is not the only costimulatory pathway in osteoclastogenesis. Other costimulatory receptors are TREM-2, which is responsible for calcium-dependent signaling pathways required for RANK-mediated activation of CaMKII and downstream MEK and ERK MAPKs that are important for osteoclastogenesis, Fc receptor common gamma subunit (FcRγ) and DNAX-activating protein 12 (DAP12), which are immunoreceptor tyrosine-based activation motif (ITAM)-containing adaptor proteins involved in a costimulatory signal for RANKL-induced osteoclastogenesis, PIR-A (ILT7 in humans) and SIRPβ1 [75–77]. Even if ligands for TREM-2, SIRPβ1, OSCAR, and PIR-A are still unknown, it has been hypothesized that ligands for PIR-A are expressed on osteoblasts, while ligands for TREM-2 and SIRPβ1 are probably expressed on myeloid cells and osteoclast precursors [76, 78, 79].
The role of T-cells in osteoclastogenesis
The role of T-cells in osteoclastogenesis is still controversial. The net influence of T-cell on osteoclastogenesis is a consequence of a balance between positive and negative cytokines expressed by the T-cell. On the one hand, there are many evidences that T-cells may induce osteoclastogenesis via RANKL expression [80, 81]. On the other hand, T-cells produce INF-γ, which inhibits RANKL signaling via degradation of TRAF6, and IL-4, which is anti-osteoclastogenic [82, 83].
Miranda-Carus et al.  have demonstrated that peripheral blood T-cells from patients with early RA express RANKL and IL-15 on the cell surface and promote osteoclastogenesis in autologous monocytes.
In PsA, T-cells through the production of osteoclastogenic cytokines, including RANKL and TNF-α, support osteoclastogenesis . Mori et al.  demonstrated osteoclastogenesis induced by T-cells purified from unstimulated mononuclear cells both from peripheral blood and from synovial fluid obtained by PsA patients, which was not evident in the same cultures from the controls.
Takayanagi  has hypothesized the existence of a T helper cell subset involved in bone resorption, called osteoclastogenic T helper cells (ThOc cells). ThOc cells may induce local inflammation and the production of RANKL and inflammatory cytokines, such as TNF-α, which are responsible for RANKL expression on synovial fibroblasts. Moreover, ThOc cells are characterized by a reduced production of INF-γ . Takayanagi has also hypothesized that ThOc cell subset may be identified with the Th17 cell subset . In fact, Th17 cells produce IL-17, a proinflammatory cytokine which induces RANKL expression on mesenchymal cells .
The role of natural killer cells in osteoclastogenesis
Natural killer (NK) cells are the main lymphocyte subpopulation involved in the cell-contact-mediated production of TNF in monocytic cells. NK cells may participate indirectly in osteoclastogenesis by enhancing TNF-α synthesis in macrophages as well as directly via RANKL and M-CSF, molecules that are further up-regulated on NK cells upon IL-15 stimulation . This event might be involved in the pathogenesis of RA in which the synthesis of TNF is improved.
The role of B-cells in osteoclastogenesis
Like T-cells, the role of B-cells in osteoclastogenesis is today only partially known. There are evidences that interaction between B-cell and T-cell may regulate B-cell production of cytokines involved in bone metabolism. In fact, B-cells suppress osteoclastogenesis if stimulated with Th1 cytokines while induce osteoclastogenesis when activated by Th2 cytokines .
On the one hand, there is evidence that B-cells inhibits osteoclastogenesis through the secretion of TGF-β, as demonstrated in vitro by using peripheral blood B-cells, and through the production of OPG, as demonstrated in both animal model and humans in vivo [88, 89]. Moreover, generation of osteoclasts from B-cell precusors has been demonstrated in vitro .
On the other hand, it has been demonstrated a role of B-cells in inducing osteoclastogenesis in multiple myeloma via RANKL expression and/or as an indirect effect of IL-7 secretion, which is involved in bone resorption in vivo [91–95]. More recently, a role of B-cells in bone resorption has been seen also in periodontal disease first in a rat model and later in vitro on lymphocytes isolated from gingival tissues of patients affected by periodontal disease [50, 96]. Furthermore, Colucci et al.  described B-cells involvement in the T-cell-dependent osteoclastogenesis in periodontitis through the production of IL-6 and IL-7.
To our knowledge, no data about a role of B-cells in osteoclastogenesis related to chronic arthritis have been reported in literature. Nevertheless, since B-cells have a central role in the pathogenesis of RA, studies on the relationship between B-cells and osteoclastogenesis in RA could be useful.
The role of DCs in osteoclastogenesis
The role of DCs in osteoclastogenesis is a new emerging field of research. There are evidences that immature DCs may be indirectly or directly involved in osteoclastogenesis. While under normal conditions, DCs are infrequently present in the bone or adjacent tissues, and they do not appear to influence bone remodeling , in RA patients numerous different DCs subtypes have been isolated in joints where they have a role in osteoclastogenesis [99–101], probably through the activation of naïve T-cells, which then induce osteoclast differentiation via RANK-RANKL complex [102, 103]. Another possible mechanism of action may be the presentation of unknown self-antigens to autoreactive T-cells responsible for the production of proteolytic enzymes involved in the damage of connective tissues . Moreover, it has been demonstrated that M-CSF and RANKL induce DCs differentiation into osteoclasts in vitro, a process called “transdifferentiation” by Rivollier et al. , suggesting also a direct role in bone damage.
There is emerging interest for osteoclasts as key players in the erosive and inflammatory events leading to joint destruction in chronic arthritis with special regard to RA. Accumulation of osteoclasts in rheumatoid synovial tissues and their activation associated with osteoclastogenic cytokines and chemokines at cartilage erosion sites suggest that they could be usefully selected as therapeutic target. In particular, in consideration of the primary role of RANKL and TNF-α in osteoclastogenesis, the inhibition of TNF-α and the control of the production of RANKL represent important strategies for reducing bone damage in this disease. Several studies showed the importance of anti-TNF inhibitors, such as etanercept and infliximab, in reducing both bone erosions and inflammation in RA joints [41, 42]. Regarding the possibility of control of RANKL production, denosumab, an anti-receptor activator of RANKL human monoclonal antibody, directly inhibits RANK/RANKL signaling pathway. Six published phase 2 and three clinical studies have demonstrated the safety and antiresorptive effect of denosumab in the treatment of postmenopausal osteoporosis and bone destruction due to metastatic lesions or RA .
Conflicts of Interest
The authors state they have no conflicts of interest.