Fc Gamma Receptors as Regulators of Bone Destruction in Inflammatory Arthritis

Bone erosion is one of the primary features of inflammatory arthritis and is caused by excessive differentiation and activation of osteoclasts. Fc gamma receptors (FcγRs) have been implicated in osteoclastogenesis. Our recent studies demonstrate that joint-deposited lupus IgG inhibited RANKL-induced osteoclastogenesis. FcγRI is required for RANKL-induced osteoclastogenesis and lupus IgG-induced signaling transduction. We reviewed the results of studies that analyzed the association between FcγRs and bone erosion in inflammatory arthritis. The analysis revealed the dual roles of FcγRs in bone destruction in inflammatory arthritis. Thus, IgG/FcγR signaling molecules may serve as potential therapeutic targets against bone erosion.


INTRODUCTION
Inflammatory arthritis is a group of diseases characterized by joint inflammation and bone damage. About 0.1% of adults develop inflammatory arthritis annually (1). Rheumatoid arthritis (RA) is a chronic autoimmune disease characterized by progressive synovitis and bone destruction, causing irreversible joint damage and disability (1)(2)(3)(4). Bone erosion is the central hallmark of RA in ultrasonography identification (5,6). Anti-citrullinated protein antibodies (ACPAs) are considered to be among the leading risk factors for bone destruction in RA (7). Ankylosing spondylitis (AS) and psoriatic arthritis (PsA) are other common inflammatory arthritis diseases with bone destruction (8,9).
Systemic lupus erythematosus (SLE) is a chronic autoimmune disease characterized by multiorgan tissue damage and high levels of autoantibodies in the serum (10). Arthritis is a common clinical manifestation with a prevalence of 69 to 95% in patients with SLE (11). However, only 4 to 6% of patients with SLE arthritis display bone erosion on plain radiographs (12)(13)(14). As to ACPA positive SLE patients, which are also called rhupus patients, they often overlapped clinical features and fulfilled American College of Rheumatology (ACR) criteria for RA classification (15,16). It is still unclear why lupus arthritis without ACPA lacks bone destruction. Recently, FcgRs have been reported to exert a regulatory effect on osteoclastogenesis (17)(18)(19)(20)(21)(22)(23)(24)(25). Our recent study demonstrated that joint-deposited lupus IgG triggered arthritis without bone erosion in mice and lupus IgG inhibited osteoclastogenesis induced by receptor activator of nuclear factor kappa-B ligand (RANKL). FcgRI exerted an inhibitory effect of lupus IgG on RANKL-induced osteoclastogenesis (26). Our study suggests that FcgR could function as critical regulators of inflammatory arthritis.
Here, we review the published studies and demonstrate the association between the FcgR and bone erosion in inflammatory arthritis.
FcgRs are divided into activating and inhibitory receptors and coexpressed on the same cell (38). Activating FcgRs, including FcgRI and FcgRIII, contain an immunoreceptor tyrosine-based activation (ITAM) in intracellular structure and transmit their signals via the ITAM, which recruits spleen tyrosine kinase (Syk) (39). FcgRIIB is the only known inhibitory FcgR with an immunoreceptor tyrosine-based inhibitory motif (ITIM) in its intracytoplasmic domain (40). The phosphorylation of ITIM counteracts the signals mediated by activating FcgRs (41)(42)(43). FcgRIIB is expressed widely on B cells, macrophages, and mast cells and downregulates several cellular functions, such as B-cell activation and mast cell degranulation (44). The activating-toinhibitory (A/I) ratio on the same cell acts as the specific checkpoint for the arrest or progression of an immune response. Surprisingly, when monomeric or low-affinity immune complexes bind to activating FcgRs, the normally activating ITAM domain cannot induce co-aggregation of activating receptors, thereby partially phosphorylating the ITAM domain. Thus, partial tyrosine phosphorylation of ITAM by Src family kinases may result in the recruitment of inhibitory SHIP. This is called inhibitory ITAM (ITAMi) signal and is important in maintaining immune homeostasis (45)(46)(47).
Unlike other activating FcgRs, FcgRII proteins do not require the common FcR g-chain for stable expression or function. They all have signaling motifs in their intracellular cytoplasmic domains (48). All the above FcgRs are transmembrane glycoproteins, except for human FcgRIIIB, which is expressed on neutrophils and is a glycophosphatidylinositol (GPI)-anchored protein (49,50). The mechanisms by which FcgRIIIB transduces signals are still unknown (51).

FcgRs AND ARTHRITIS
During autoimmune diseases, such as RA and SLE, the autoantibodies and immune complexes cause inflammation via FcR aggregation (52). The altered expression of FcgRs on immune cells in the circulation and synovium of RA patients is the first indication of their involvement in inflammation (53)(54)(55)(56)(57)(58)(59)(60). The absence of all FcgRs does not affect the number of osteoclast precursors or their osteoclastogenic potential. However, it reduces joint inflammation and bone erosion during inflammatory arthritis (61). FcgRIIB is particularly critical for maintaining the balance of an efficient inflammatory response or countering unwanted autoimmunity attacks. Multiple clinical studies have shown that FcgRIIB is a reliable biomarker for SLE susceptibility in different ethnic groups. FcgRIIB and its signaling pathways represent a vital checkpoint in peripheral and central tolerance and in controlling the development of autoreactive antibodies (62).
In addition to the altered expression of FcgRs, genetic variants associated with related single-nucleotide polymorphisms (SNPs) in populations with RA and lupus arthritis have been reported. Several genes encoding FcgRs that alter the affinity of FcgRs for IgGs have been described in several RA populations. In particular, some of these, such as the hFcgRIIa-R131 variant, which is related to an increased risk of developing RA, even influence the susceptibility to RA development and the response to treatment (63)(64)(65)(66)(67)(68)(69)(70). In addition, an association between lupus arthritis and the FCGR2A as well as FCGR3A low copy number genotypes has been observed in Taiwan patients with SLE. The FCGR3A low copy number genotype was significantly enriched in patients with SLE having arthritis (71)(72)(73). Moreover, a metaanalysis revealed the association of the FcgRIIa-R131 allele with SLE, especially in African Americans, whereas the FcgRIIIa-F176 allele was associated with SLE in Caucasians and other groups (74). Furthermore, Tsang et al. demonstrated the association between low-affinity FcgR polymorphisms and susceptibility to SLE (75).
Autoantibodies and their immune complexes play a central role in shaping a pro-inflammatory environment. Indeed, complexes of ACPA and rheumatoid factor (RF) induce the production of potent inflammatory cytokines (93)(94)(95)(96). This effect is predominantly mediated by FcgR signaling on macrophages (51,97). Tumor necrosis factor (TNF)-a, in combination with cytokines interleukin (IL)-4 and IL-13, downregulates FcgRmediated function by decreasing the expression of activating FcgRs, suggesting that downregulated activating FcgRs might have an anti-inflammatory effect (98).
The Fc receptors on white blood cells are essential for effective phagocytosis of immune complexes and bacteria. Moreover, FcgRI is upregulated during infection. FcgRI (CD64) has previously been reported to distinguish systemic infections from inflammatory autoimmune diseases and viral infections. Patients without inflammatory and infectious conditions, such as osteoarthritis, have a lower level of neutrophil FcgRI than those with infections (99-104). Oppegaard et al. investigated the use of FcgRI in discerning septic arthritis from inflammatory joint disease and found that FcgRI is highly specific for infectious diseases, including septic arthritis. However, its sensitivity is poor in local infections (104). Although distinct meta-analyses have confirmed this, more large prospective studies need to be conducted to verify several cut-off values reported in the neutrophil FcgRI test in the clinical setting (105,106).
Human and murine activating FcgRs are not functionally equivalent. A few studies performed in transgenic mice expressing human FcgRs examined their involvement in inflammatory arthritis (107). The results confirmed that the expression of the human FcgRIIA is associated with spontaneous autoimmune inflammation, with a crucial role in autoimmune diseases (92).

FcgR ROLE IN BONE EROSION Osteoclast Activation and Differentiation
Bone balance depends on a dynamic regulation of bone formation and resorption, which are predominantly mediated by osteoblasts and osteoclasts, respectively (108,109). Enhanced osteoclast activity could result in severe bone destruction as exemplified in autoimmune inflammatory diseases such as RA, whereas defective osteoblast differentiation causes diseases with a high bone mass, including osteopetrosis. Osteoclasts are the only bone-resorbing cells and play a central role in bone erosion. Osteoclasts are derived from multinucleated progenitors of the monocyte/macrophage family and are the link between immune and bone systems. RANK and RANKL are critical factors that together regulate osteoclast functions. In addition, macrophage colony-stimulating factor (M-CSF) is an essential cytokine in osteoclastogenesis (109)(110)(111). RANKL is majorly secreted by osteoblasts, osteocytes, T cells, and endothelial cells. And osteocytes express a much higher amount of RANKL required for osteoclastogenesis than osteoblasts (112,113). The most important negative regulator of RANKL is the decoy receptor osteoprotegerin (OPG), which inhibits osteoclastogenesis by preventing RANKL-RANK interaction. The RANKL-RANK-OPG system modulates bone homeostasis by regulating osteoclasts (114). Osteoblasts and osteocytes also produce OPG to suppress osteoclastogenesis by masking RANKL signaling (115,116). RANKL initiates osteoclastogenesis by inducing nuclear factor of activated T-cells, cytoplasmic 1 (NFATc1), via TNF receptor-associated factor 6 (TRAF6) and c-Fos pathways (117) (Figure 2). NFATc1 is the master transcription factor for pro-osteoclastogenic genes. In addition, several proinflammatory cytokines produced by innate immune cells and T cells, such as TNFa, IL-17, IL-1, and IL-6, stimulate osteoclastogenesis directly or indirectly (118).

FcgRs and Osteoclastogenesis
Apart from the M-CSF and RANKL signaling, an ITAM costimulatory signal provided by the accessory protein for RANKL-RANK is required for osteoclastogenesis (119). Takayanagi et al. first reported that the activation of NFATc1 was insufficient for terminal differentiation of monocytes/ macrophages into osteoclasts; calcium signals and calcineurin activation are essential for this process (117,120). Calcium signals in myeloid cells are provided by the ITAM-bearing proteins, Fc receptor g subunit, and its functional analog DNAX activation protein of 12 kDa (DAP12). Both the accessory proteins are intracellular adaptor molecules and play a crucial function in transducing the costimulatory signals for RANKL (121). Mice lacking the accessory proteins display a severe osteopetrotic phenotype with deficient osteoclast function (122).
FcRg-chain is associated with immunoglobulin (Ig)-like receptors, such as osteoclast-associated receptor (OSCAR) and paired Ig-like receptor-A (PIR-A) (Figure 2). DAP12 is associated with its signaling counterpart, triggering receptor expressed on myeloid cell-2 (TREM-2), and signal-regulatory protein b1 (SIRPb1), which are expressed on the cell membrane of osteoclast precursors and are essential for the communication between osteoclast precursors (33,123). Activation of RANKL-RANK rapidly phosphorylates the ITAM motifs and recruits the protein kinase Syk, subsequently activating multiple downstream signaling cascades, such as phospholipase Cg (PLCg) and Bruton's tyrosine kinase (BTK) as well as Tec kinases. They all enhance the effects of RANKL-signaling by augmenting the calcium influx required for the activation of NFATc1. NFATc1 subsequently migrates to the nucleus, where it binds to its gene promoter and triggers an auto-amplifying feedback loop (124,125).
Osteoclasts and their precursors express FcgR (126), whereas FcgRI, FcgRIIB, and FcgRIIIA are significantly upregulated during human ex vivo osteoclastogenesis (127). Blocking of the FcR and deleting the FcgR gene reduce osteoclastogenesis FIGURE 2 | Overview of the osteoclast signaling network. A schematic representation of ITAM-mediated costimulatory signal in the RANKL-induced TRAF6 signaling pathway of osteoclast differentiation. In osteoclast precursors, phosphorylation of ITAM stimulated by immunoreceptors and RANKL-RANK interaction recruits the Syk family kinases, thus activating phospholipase Cg (PLCg), Bruton's tyrosine kinase (BTK), as well as Tec kinases. They augment the calcium influx required for the activation of NFATc1. NFATc1 subsequently migrates to the nucleus, where it binds to its gene promoter and triggers an auto-amplifying feedback loop. Calcium signals in osteoclast precursors are provided by the ITAM-bearing proteins, Fc receptor g subunit, and its functional analog DNAX activation protein of 12 kDa (DAP12). FcRg-chain is associated with the immunoglobulin (Ig)-like receptors, such as osteoclast-associated receptor (OSCAR) and paired Ig-like receptor-A (PIR-A). DAP12 is associated with its signaling counterpart, triggering receptor expressed on myeloid cell-2 (TREM-2), and signal-regulatory protein b1 (SIRPb1).

Zuo and Deng
FcgRs Frontiers in Immunology | www.frontiersin.org June 2021 | Volume 12 | Article 688201 stimulated by IgG complexes on osteoclast precursor cells (25). Although FcgRs are required for osteoclastogenesis and bone resorption in inflammatory disorders, their specific role in bone homeostasis is not completely understood.

FcgRs and Bone Erosion
Costimulatory signals mediated by the ITAM motif cooperate with RANKL for bone homeostasis, suggesting the link between ITAM-harboring adaptors FcgRs and bone erosion (117). As an important link binding the bone system and immune system, FcgRs are not only receptors for the Fc portion of IgG but are also costimulatory molecules for RANKL-induced osteoclastogenesis (119,121). Bone damage has been reported in seropositive RA patients before clinical disease onset, highlighting that osteoclastogenesis is independent of joint inflammation (128). This finding challenged the concept that inflammation is the primary trigger for bone erosion in inflammatory arthritis and indicated that bone loss might precede inflammation (129). The expression of Fcg receptors on osteoclast precursor cells and mature osteoclasts has been measured. FcgRI and FcgRIII are primarily expressed on human preosteoclasts, whereas the inhibitory FcgRIIB is majorly expressed on mature osteoclasts (127). Under physiological conditions, activating FcgRI and FcgRIV in mice does not have a major role in bone characteristics and osteoclast development (22). Bone homeostasis is not significantly different in mice with FcgRI or IV deficiency compared with wild mice (17). In addition, the deficiency of FcgRIIB does not affect osteoclastogenesis (23). Activating FcgRs transmit the positive signal. In contrast, FcgRIII functions as an inhibitory receptor in the differentiation of osteoclast precursor cells under physiological conditions. FcgRIII deprives the FcRg subunit's availability for other Ig-like receptors activating receptors, such as PIR-A and OSCAR, thus transmitting an ITAM-mediated inhibitory signal for osteoclastogenesis (130). Naive FcgRIII -/mice have increased osteoclast numbers and an osteoporotic phenotype (22).
The relative importance of various FcgRs in osteoclastogenesis changes in the inflammatory arthritis microenvironment. Studies demonstrated that the stimulation of FcgRI and FcgRIV increases both osteoclast differentiation and function both in vitro and in vivo (22,30). FcgRIII levels are increased, and FcgRIIB levels are decreased on bone marrow cells from mice with collagen-induced arthritis (CIA), indicating that FcgRIII induces osteoclastogenesis under inflammatory conditions (22). Furthermore, human RA patients with the FcgRIIIa-158V allele endure severe bone erosion compared with patients with the FcgRIIIa-158F allele (131,132). Similarly, artificial crosslinking of FcgRI and FcgRIV leads to increased osteoclast differentiation without affecting their resorbing function in vitro (17). Osteoclast numbers and bone erosion were decreased in FcgRIV -/mice compared with wild mice in a serum transfer model (17). FcgRIIB -/mice spontaneously developed osteoporosis, which was reversed by an additional knockout of activating FcgRs (22).
De-sialylated IgGs binding to FcgRs with strong affinity have substantially high stimulatory effects on both murine and human osteoclasts (127,133). In addition, IgGs were less sialylated during inflammation (22). Harre et al. confirmed that the interactions between immune complexes and osteoclasts were related to the degree of IgG sialylation, and only non-sialylated or low-sialylated immune complexes drive osteoclastogenesis. RA patients with low Fc sialylation levels of IgGs have significantly higher bone loss. The pro-osteoclastogenic effect of non-sialylated immune complexes is a common feature of all IgG antibodies (127). A recent study showed that in induced pluripotent stem cell derived mesenchymal stem cell (iMSCs), the sialylation degree of IgG determines the antibodies directed osteogenic potential by regulating immune responses and osteoclastogenesis (24),but desialylated IgG complexes do not affect arthritis-mediated bone loss (134).
Although the signaling of activating FcgRs mediated by immune complex increases osteoclast differentiation, different results exist for immune complex/FcgR on osteoclastogenesis and osteoclast function ( Table 1). Previous studies demonstrated the immune complex-induced inhibition of osteoclastogenesis, which possibly acts via activating FcgRs (23,139). This suggests that FcgRs may have dual roles in bone destruction in inflammatory arthritis. High levels of autoantibodies are a characteristic feature of SLE compared with other inflammatory arthritis (140,141). The deposition of autoantibodies or immune complexes causes lupus nephritis (142), skin damage (143), splenomegaly (144), and damage to other organs. Lupus IgG can promote the differentiation of monocytes into DCs (145). These indicate that lupus autoantibodies may also play a protective role in bone destruction in inflammatory arthritis. Recently, our research results (26) demonstrated that jointdeposited lupus IgG induced arthritis without bone erosion by intraarticular injection of lupus IgG in mice. Monocytes/ macrophages and their product TNFa are required for the development of lupus IgG-induced arthritis. To understand the mechanism of lupus IgG-induced arthritis with deficiency of bone erosion, we determined whether lupus IgG inhibited RANKLinduced osteoclastogenesis. We found that lupus IgG directly suppressed RANKL-induced osteoclastogenesis in a dosedependent manner in vitro. The inhibitory effect of lupus IgG on osteoclastogenesis is related to timepoint in lupus IgG and RANKL treated macrophages. Deficiency of FcgRII and FcgRIII did not affect the inhibitory effect of lupus IgG on osteoclastogenesis, indicating that the inhibitory effect of lupus IgG on osteoclastogenesis is dependent on FcgRI. Lupus IgG and RANKL can downregulate the surface expression of FcgRI on bone marrow macrophages (20). Research results suggest that lupus IgG inhibits osteoclastogenesis by competitively occupying FcgRI on monocytes/macrophages and reducing RANKL signaling. The effect of activation or repression of RANKL-induced osteoclastogenesis depends on the extent of FcgRI occupancy by IgG. This protective mechanism explains non-destructive arthritis in SLE. In addition, it implies that FcgRI could be a therapeutic target for bone erosion in inflammatory arthritis.
The deposition of ACPA is important for osteoclastogenesis in RA (146). Different studies have identified that ACPA prevalence is significantly increased in SLE patients with erosive arthritis (16). Recent studies have explored the direct effect of ACPA-mediated bone erosion. ACPA IgG together with their citrullinated antigens forms immune complexes that stimulate immune cells via their interaction with FcgRs (93,147). By using polyclonal ACPAs purified from ACPA-containing serum of RA patients, Harre et al. provided the first validation that ACPAs can directly promote osteoclast differentiation and activation (7). ACPA IgG might affect osteoclastogenesis by the activation of Fc receptors on osteoclasts directly. IgG Fc sialylation is crucial for immune complex-osteoclast interactions (127). Besides, ACPA IgG is shown less sialylated than random IgG (148). There are other published papers regarding the detailed mechanisms of ACPA's direct regulation, but the exact mechanism of ACPA's direct effect on erosion remains unclear (149-152).

FcgR IMMUNOTHERAPY
The crucial role of FcgRs in both inflammatory arthritis and bone erosion may offer a promising therapeutic target for bone destruction in inflammatory arthritis. One indirect mechanism involves the neutralization of autoimmune IgG Fc by soluble FcgRs, These drugs include the recombinant soluble FcgIIB receptor SM101 (NCT03851341) and monoclonal antibody targeted the receptors. For example, antagonistic monoclonal antibody against the hFcgRIIIA has been shown to be effective in a patient with immune thrombocytopenia (ITP) refractory to all conventional therapies (153). And human recombinant soluble FcgRIIB treatment could ameliorate collagen-induced arthritis by reducing immune complexes-stimulated inflammation and joint swelling (154). Besides, recombinant human soluble FcgRII was evaluated as an effective therapeutic strategy in inhibiting chronic murine lupus pathology (155).
Another mechanism involves the direct blocking of the IgGbinding site on FcgRs. Recombinant multimeric Fc fragments with a high affinity for FcgRs have been shown to be efficacious in animal models of RA, ITP, and graft-versus-host disease (GVHD) (156). These include PF-06755347 (NCT03275740), CSL730 (NCT04446000) and CSL777 (Preclinical) (157). However, nonspecific crosslinking of activating FcgRs could lead to undesired clinical adverse events, and monovalent antibody derivatives, such as Fab, may reduce severe clinical adverse events (158). Up to now, results of above molecules from clinical trials have been promising in autoimmune diseases, but further long-term data are needed (159,160).
Intravenous immunoglobulin (IVIG) treatment is efficient in several different immune disorders (161,162). IVIG consists predominantly of IgG and a small fraction of immune complexes. It exerts anti-inflammatory effects in both humans and animal models by its Fc but not Fab fragments (163). Besides, previous studies confirmed that IVIG could directly inhibits human osteoclastogenesis by suppressing the RANK signaling, the suppressive effect is partly mediated by IgG immune complexes contained within IVIG preparations (138). Our study showed that lupus IgG induced synovial inflammation but inhibited RANKL-induced osteoclastogenesis. The suppressive effect is mediated by the competitive occupation of FcgRI on monocytes/macrophages (26).

CONCLUSIONS AND FUTURE PERSPECTIVES
Bone erosions are remarkable features in inflammatory arthritis, such as RA, but not in lupus arthritis. Osteoclasts are major cells for bone erosions. Activating FcgR containing ITAM motifs is required for RANKL-induced osteoclastogenesis. FcgR can effectively regulate inflammatory arthritis and bone erosions. Based on published studies, we conclude that FcgR may have dual roles in osteoclastogenesis. The effect of activating and inhibiting osteoclastogenesis depends on the extent of FcgRI occupancy by IgG and RANKL, respectively. Specific IgG molecules or Fc fragments with a high affinity for FcgRI designed to occupy FcgRI may exert the inhibitory effect on bone erosion. The sialylation level of IgG Fc binding to FcgRs needs to be taken into account as well. A deeper understanding of FcgRs involved in physiological and pathological osteoclastogenesis will be valuable in identifying new targets and developing potential therapeutic strategies for inflammatory arthritis.

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

AUTHOR CONTRIBUTIONS
YZ and G-MD designed the manuscript and figures. YZ and G-MD drafted the manuscript and approved the final version of the manuscript. G-MD revised the final version of the manuscript critically. All authors contributed to the article and approved the submitted version.