• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. Jul 20, 2010; 107(29): 13028–13033.
Published online Jul 6, 2009. doi:  10.1073/pnas.1000546107
PMCID: PMC2919936

Natural killer cells trigger osteoclastogenesis and bone destruction in arthritis


Osteoclasts are bone-eroding cells that develop from monocytic precursor cells in the presence of receptor activator of NF-κB ligand (RANKL) and macrophage colony-stimulating factor (M-CSF). Osteoclasts are essential for physiological bone remodeling, but localized excessive osteoclast activity is responsible for the periarticular bone destruction that characteristically occurs in patients with rheumatoid arthritis (RA). The origin of osteoclasts at sites of bone erosion in RA is unknown. Natural killer (NK) cells, as well as monocytes, are abundant in the inflamed joints of patients with RA. We show here that such NK cells express both RANKL and M-CSF and are frequently associated with CD14+ monocytes in the RA synovium. Moreover, when synovial NK cells are cocultured with monocytes in vitro, they trigger their differentiation into osteoclasts, a process dependent on RANKL and M-CSF. As in RA, NK cells in the joints of mice with collagen-induced arthritis (CIA) express RANKL. Depletion of NK cells from mice before the induction of CIA reduces the severity of subsequent arthritis and almost completely prevents bone erosion. These results suggest that NK cells may play an important role in the destruction of bone associated with inflammatory arthritis.

Keywords: macrophage colony-stimulating factor, receptor activator of NF-κB ligand, rheumatoid arthritis

An early advance in the understanding of osteoclastogenesis was the finding that coculture of osteoclast precursors and stromal cells yielded osteoclasts (1). This ultimately led to the discovery of two factors produced by stromal cells that are necessary and sufficient for osteoclast formation, namely macrophage colony-stimulating factor (M-CSF) and receptor activator of NF-κB ligand (RANKL), which act on osteoclast precursor cells to induce their differentiation (2). The importance of these factors is further supported by studies in animal models showing enhanced bone erosion on administration of M-CSF (3) or RANKL (4), and in animals rendered deficient for osteoprotegerin (OPG) (5), a decoy receptor that binds RANKL, preventing its binding to (RANK). Conversely, mutation or gene ablation of M-CSF (6), RANKL (7), or RANK (8) or overexpression of OPG (9) leads to increased bone thickness and decreased capacity to generate osteoclasts. In addition, an antibody to RANKL is effective in the treatment of osteoporosis in humans (10).

Pathological bone resorption occurs in patients with immune-mediated arthritis and is particularly prominent at the interface of inflamed synovial tissue and bone (11). There is evidence that osteoclasts are responsible for such periarticular erosions, but the origin of osteoclasts at this site is incompletely understood (11). It is known, however, that ostecoclasts can be derived from circulating CD14+ monocytes, which are recruited to inflamed joints, and that osteoclastogenesis is likely influenced by the relatively high levels of RANKL and M-CSF and low levels of OPG in the inflammatory mileu (11, 12). Indeed, the recent demonstration that a monoclonal antibody against RANKL inhibits the development and progression of bone erosion in patients with rheumatoid arthritis (RA) confirms an important role for RANKL in this disorder (13).

Natural killer (NK) cells are known to participate in the clearance of virus-infected, aberrant, or transformed cells (14). Moreover, NK cells are poised for a rapid release of cytokines and growth factors that might influence the initiation and development of immune responses mediated by T and B cells (1517). NK cells can be detected in the inflamed synovial tissue at an early stage of the disease, and they constitute up to 20% of all lymphocytes in the synovial fluid (SF) of patients with established RA (18, 19). Such SF NK cells express CD56 and CD94/NKG2A at bright levels, but essentially lack expression of other common NK cell–associated markers, such as CD16 and killer cell Ig-like receptors (18). Thus, the SF NK cells are phenotypically similar to a minor population of NK cells found in the blood of patients with RA and normal controls. Recent evidence shows that this CD56bright NK cell subset produces high levels of cytokines and might play a role in immunoregulation (15). Moreover, this NK cell subset has an up-regulated expression of several chemokine receptors and adhesion molecules that may participate in its preferential recruitment into the inflamed synovium (20), and enable the cells to engage and subsequently activate monocytes through a variety of receptor–ligand interactions (18, 21, 22). Here we asked whether the interaction between NK cells and monocytes results in the differentiation of monocytes into osteoclasts.


Synovial NK Cells Induce Differentiation of Monocytes Into Osteoclasts In Vitro.

We analyzed RA synovial tissue sections for the presence of NK cells and monocytes. Compared with monocytes, NK cell numbers were relatively low, and a fraction of NK cells were juxtaposed with monocytes (Fig. 1A and Fig. S1), enabling interactions that might result in reciprocal cellular activation. To assess this possibility, we isolated SF NK cells and CD14+ SF monocytes and cocultured them in vitro at a NK cell:monocyte ratio of 1:10. Cocultures were performed in medium alone or in medium supplemented with IL-15, a cytokine produced in the inflamed synovium (23) that is known to activate NK cells (24). In the presence of IL-15, such cocultures yielded many large, round, adherent cells that appeared only when NK cells and monocytes were placed in direct contact with each other (Fig. 1B). As shown in Fig. 1C, these large cells are multinucleated and express tartrate-resistant acid phosphatase (TRAP), a marker associated with osteoclasts. Moreover, as shown in Fig. 1D, the multinucleated cells express receptors for calcitonin (CTR) and F-actin, markers associated with functional resorbing osteoclasts.

Fig. 1.
NK cells are juxtaposed with monocytes in RA synovial tissue and induce their differentiation into osteoclasts in vitro. (A) Representative immunohistochemical staining of a patient with RA (n = 3) shows NK cells (blue) and CD14+ monocytes (brown) in ...

We tested whether the osteoclasts formed from monocytes in coculture with NK cells are capable of degrading bone minerals by culturing the cells on calcium phosphate‒coated discs. When synovial monocytes were cocultured with SF NK cells in the presence of IL-15, many large eroded areas could be distinguished on these discs; a representative result is shown in Fig. 2A. In addition, bone erosion was observed on dentine slices where resorption zones formed in the presence, but not in the absence, of IL-15–activated NK cells (Fig. 2B). Taken together, these results provide evidence that human monocytes can differentiate into functional osteoclasts in the presence of SF NK cells.

Fig. 2.
Osteoclasts that form on coculture between SF NK cells and monocytes erode bone substrates. (A) (Upper) Phase-contrast micrographs of osteologic discs that have been exposed to (left to right): freshly isolated SF monocytes from an RA patient cultured ...

We next induced the formation of osteoclasts from CD14+ peripheral blood (PB) monocytes with soluble M-CSF and RANKL, a standard method for generating osteoclasts in vitro, and compared the osteoclast-forming capacity with that of cocultures with SF NK cells. Interestingly, SF NK cells appeared more efficient in triggering osteoclast differentiation compared with exogenous M-CSF and RANKL, as indicated by both the greater number and larger size of TRAP+ multinucleated osteoclasts (Fig. S2 A and B).

RANKL and M-CSF Mediate Osteoclastogenesis Induced by NK Cells.

Because M-CSF and RANKL are required and sufficient for osteoclastogenesis (2), we tested whether SF NK cells express these molecules. As shown in Fig. 3A, a large fraction of CD3CD56+ SF NK cells (mean ± SEM, 54.9% ± 9.6%; n = 3), but not of CD3+CD56 SF T cells (2.5% ± 1.8%), express cell-surface RANKL. Notably, a small proportion of the SF NK cells appears to express low levels of M-CSF on the cell surface (Fig. 3A), and on prolonged culture of SF NK cells in IL-15, cell surface expression of both M-CSF and RANKL are further up-regulated (Fig. 3B).

Fig. 3.
RANKL and M-CSF expression by NK cells. (A) Histograms showing membrane expression of RANKL (Top) and M-CSF (Bottom) on freshly isolated SF NK cells (gated on CD56+CD3 cells), CD56CD3+ SF T cells (Middle), or CD56+CD3+ SF T cells (Right ...

We next analyzed RANKL expression on NK cells derived from healthy PB. In line with a previous report (25), we found that RANKL is associated mainly with the minor CD56bright subset (Fig. 3C and Fig. S3). Moreover, freshly isolated CD56bright NK cells from healthy PB appear to secrete similar amounts of M-CSF as SF NK cells when stimulated with IL-15 (Fig. 3D). Low levels of soluble RANKL (100 pg/mL) could be detected only when SF NK cells and CD56bright PB NK cells were stimulated with the highest dose of IL-15 (i.e., 100 ng/mL), suggesting that RANKL is not efficiently shed from the NK cell surface. In line with these findings, we also observed that isolated CD56bright PB NK cells from healthy donors are more potent than the CD56dim PB NK subset in promoting the formation of TRAP+ multinucleated cells from autologous CD14+ PB monocytes (Fig. 3E).

To determine whether M-CSF and RANKL mediate NK cell–induced osteoclastogenesis, we evaluated the effect of inhibiting these molecules with soluble M-CSF receptor (sM-CSFR) and OPG, respectively. SF NK cells were cultured in the presence of one or the other of these inhibitors together with autologous CD14+ SF monocytes. As shown in Fig. 4 A and B, reduced osteoclastogenesis was observed on blocking of M-CSF or RANKL. IFN-γ is known to inhibit osteoclastogenesis (26), but blocking of IFN-γ in the cocultures did not further enhance the capacity of SF NK cells to trigger osteoclastogenesis, a finding that may be explained by the relatively poor capacity of NK cells to produce IFN-γ in response to exogenous IL-15 (Fig. S4) (27).

Fig. 4.
Neutralization of either M-CSF or RANKL suppresses osteoclastogenesis induced by SF NK cells cocultured with autologous SF monocytes. (A) Autologous SF monocytes from patients with RA (n = 3) were cultured in the presence or absence of autologous SF NK ...

The finding that a small number of erosion zones form when RA synovial monocytes are cultured in the absence of NK cells (Figs. 2A and and4B)4B) suggests that osteoclasts can appear spontaneously, an observation in line with previous studies (28, 29).

NK Cells in the Joints of Mice with Collagen-Induced Arthritis Express RANKL, and NK Cell Depletion Reduces Signs of Arthritis and Bone Erosion.

To further evaluate the role of NK cells in osteoclastogenesis in arthritis, we studied their effect on bone erosion in collagen-induced arthritis (CIA) in DBA/1 mice, a model of human RA. Consistent with an important role for NK cells as mediators of osteoclast formation, many RANKL+ cells within the inflamed tissue of CIA mice appear to be NK cells, and such NK cells can be seen adjacent to eroded bone surfaces (Fig. 5A). To directly study the effect of NK cells in CIA, we used anti-asialo GM1 antibodies to deplete these cells from mice immunized with type II collagen, before the onset of arthritis. Two days after treatment with anti-asialo GM1 antibodies, the frequency of DX5+CD3 NK cells in spleen was 0.4 ± 0.2% (mean ± SD) compared with 2.4 ± 0.6% in IgG-treated mice, whereas the frequencies of DX5+CD3+ and DX5CD3 T cells remained unchanged, indicating that NK cells, but not the T or NK T cell populations, were specifically depleted, confirming previous reports (16, 30). Although the incidence of disease was 100% in all groups, the resultant NK cell–depleted mice showed significantly reduced signs of arthritis (Fig. 5B) and greatly reduced bone erosion, synovitis, and pannus formation (Fig. 5 C and D) compared with animals treated with rabbit Ig or PBS. These results indicate that NK cells not only play a pathogenic role in CIA-associated bone destruction, but also contribute to other manifestations of the disease.

Fig. 5.
Synovial tissue NK cells of mice with CIA express RANKL and NK cell–depletion ameliorates disease. (A) Adjacent sections of an ankle joint stained with antibody against NK cells (NKp46; Left), RANKL (Center), or control (Right) show RANKL+ NK ...


The major underlying immunologic events responsible for enhanced bone erosion in RA are still largely unknown. In this study, we have shown that NK cells in the SF of RA patients efficiently trigger formation of osteoclasts from monocytes. Such NK cells express both M-CSF and RANKL, which are responsible for osteoclastogenesis, and both molecules are further up-regulated on NK cells by IL-15, a cytokine abundantly expressed in the RA joint (23).

Emerging data indicate that IL-15 may play an important pathological role in the initiation of inflammatory arthritis. For example, a recent report has linked elevated levels of IL-15 in the SF of patients with early synovitis and the subsequent development of RA (31), and neutralization of IL-15 in mice with CIA was found to profoundly suppress the development of disease and prevented bone erosion (32). Although IL-15 can be produced by several cell types, its expression in RA synovial tissue tends to be associated with activated endothelial cells in the vicinity of inflammatory infiltrates (33). Interestingly, the CD56bright NK cells bear high-affinity receptors for IL-15 (24) and express a chemokine receptor pattern similar to that of monocytes (20). This subset predominates in the RA joint, as well as in the joints of patients with early synovitis (18), and its accumulation may depend on IL-15 associated with inflamed vascular endothelial cells, because this cytokine helps mediate increased NK cell adhesion, activation, and subsequent migration across endothelium (34). It is noteworthy that NK cells within the RA tissue can be found in close proximity to CD14+ monocytes (22) (Fig. 1A), supporting the view that NK cell interaction with monocytes in the RA synovium may result in the triggering of osteoclast differentiation. We hypothesize that the enhancing effect of IL-15 on inflammation and bone erosion might be due to its stimulation of M-CSF and RANKL expression by NK cells. IL-15 has been reported to stimulate the differentiation of rat osteoclast progenitors into preosteoclasts in vitro (35). In this study, however, IL-15 alone failed to increase osteoclast formation from isolated monocytes, whereas this formation was dramatically enhanced in the presence of NK cells and IL-15, indicating that the effect of IL-15 is mediated predominantly through NK cells in our system.

Immature dendritic cells (DCs) are capable of transdifferentiation into osteoclasts in vitro on subsequent culture with M-CSF and RANKL, a process that is further enhanced by TNF-α (36). Because SF NK cells can trigger differentiation of monocytes into immature DCs (22), we cannot exclude the possibility that osteoclasts generated from monocytes in our coculture system might be derived from immature DC-like cells undergoing transdifferentiation, particularly because SF NK cells are potent inducers of monocyte TNF-α synthesis (21). Further experiments using specific TNF-α inhibitors are needed to evaluate TNF-α‘s contribution to NK cell–mediated osteoclastogenesis.

IFN-γ inhibits osteoclastogenesis through a counterbalancing effect on RANKL (26). However, blockade of IFN-γ had only a minor effect on osteoclast formation in our assays, which is consistent with the poor capacity of IL-15 to stimulate IFN-γ release by NK cells (27). Notably, low levels of IFN-γ are present in the RA synovium (37), and clinical studies have failed to demonstrate that IFN-γ has an effect on bone resorption (38), suggesting that this cytokine plays a minor role in counterbalancing the effect of RANKL in vivo.

Evidence of a direct role for T cells in mediating osteoclastogenesis is mixed (3941). RANKL+ CD4+ PB T cells isolated from patients with early RA, if cultured with monocytes for 2 wk at a high T cell:monocyte ratio (4:1), can induce osteoclast differentiation (41). Other studies have shown that activated PB T cells induce RANKL-dependent osteoclastogenesis from adherent PBMCs if these cultures are pretreated with M-CSF (40, 42). Although a direct role for T cells in inducing osteoclastogenesis remains controversial, recent studies have established an indirect role for the Th17 cell subset that up-regulates RANKL expression by osteoblasts (43).

We detected little RANKL on human SF T cells. Similarly, we detected little RANKL on human PB T cells or on CD56dim PB NK cells compared with CD56bright PB NK cells. In line with the crucial role for RANKL in osteoclast formation, the CD56bright subset is more potent in this regard than both CD56dim NK cells (Fig. 4) and PB T cells derived from healthy donors (Fig. S5). However, a critical evaluation of whether SF NK cells are main contributors to processes leading to osteoclast formation requires a side-by-side comparison of SF NK cells and other SF cell subsets, including activated T cell subsets, such as SF-derived Th17 cells.

As another approach to evaluate the relative importance of NK cells in arthritis and bone erosion, we established CIA in DBA/1 mice, a well-characterized model for human RA. The destruction of bone in CIA, as in RA, appears to be largely regulated by RANKL (13, 44). Our results show that a large fraction of synovial NK cells in CIA, as well as in RA, express RANKL, suggesting that NK cells are important in the bone erosive processes associated with CIA. Indeed, we show that depletion of NK cells before the induction of disease almost completely prevented bone erosion in CIA. Interestingly, such mice also displayed significantly reduced inflammation, pannus formation, and synovitis. It should be noted that previous studies have shown that neutralization of RANKL in CIA, as well as in RA, reduces bone loss but affects inflammation only minimally (13, 44, 45). Given our finding that NK cell depletion reduces both inflammation and bone erosion in CIA, it is possible that inflammation and bone erosion might be affected by distinct NK cell subsets, where RANKL+ NK cells may be primarily involved in the latter pathway. Another interpretation is that the NK cell-dependent inflammatory process is independent of the expression of RANKL, and that the reduction in bone erosion is an indirect consequence of the reduced inflammation necessary to set the stage for erosive processes.

It should be noted that although subsets of T cells express asialo GM1, in vivo administration of anti-asialo GM1 antibody specifically reduced the number of NK cells without depleting conventional T cells or NKT cells, confirming previous reports (16, 30). This effect is likely due to the relatively high levels of asialo GM1 on NK cells compared with T cells (46). We have shown that reducing the number of NK cells in DBA/1 mice using this antibody prevented bone erosion and ameliorated CIA, which is in sharp contrast to a recent report by Lo et al. (30) that anti-asialo GM1 treatment exacerbated CIA. Based on these divergent results, future studies aimed at further defining the role of NK cells in arthritis probably should use animal models in which NK cells can be depleted with a different method. Transgenic mice expressing the diphtheria toxin (DT) receptor in NK cells as described by Walzer et al. (47), in which DT injection leads to complete and selective NK cell ablation, might prove useful for this purpose. Moreover, generation of mice that specifically lack RANKL expression on NK cells may shed further light on the specific role for RANKL versus RANKL+ NK cells in inflammation and bone erosion.


Cell Isolation and Culture.

Mononuclear cells from RA SF and healthy PB were obtained by Ficoll density gradient centrifugation, and CD14+ monocytes and CD56+CD3 NK cells were isolated and cultured as described previously (22). See SI Methods for details.

Analysis of in Vitro Osteoclast Differentiation and Function.

In vitro osteoclast differentiation and osteoclast function were analyzed according to standard methods with minor modifications. See SI Methods for details.


Cells were preincubated with FcR-blocking reagent (Miltenyi) according to the manufacturer's instructions, and staining was performed using FITC-, phycoerythrin (PE)-, or allophycocyanin (APC)-conjugated mAbs against CD3, CD14, and CD56 (all from BD Biosciences); PE-conjugated anti-RANKL (mouse IgG2b, clone MIH24) and mouse IgG2b control (eBioscience); APC-conjugated anti–M-CSF (R&D Systems), and isotype-matched controls (BD Biosciences) and analyzed on a FACSCalibur flow cytometer (BD Biosciences).


Secretion of RANKL (Peprotech) and M-CSF (R&D Systems) was measured by ELISA according to the manufacturer's instructions. In brief, NK cells were cultured in 96-well plates (Corning) at 105 cells in 200 μL, and cell-free supernatant was harvested and assayed after 72 h.


Frozen synovial tissue sections were derived from two female RA patients (age 58 and 72 y), and one male (age 68 y) treated with conventional DMARDs at the time of knee joint replacement surgery. The patients met the American College of Rheumatology criteria and the tissues were collected after informed consent under local Institutional Review Board-approved protocols. The tissue sections were incubated with an NK cell-specific mAb and mouse anti-human CD14 mAb (Dako Cytomation) as described previously (22).

CIA Studies.

Collagen induced arthritis (CIA) was induced in DBA/1 mice. See SI Methods.

Supplementary Material

Supporting Information:


We thank Claudia Benike (Stanford University) for a critical review of the manuscript, Lorna Torentino (Stanford University) for assistance with cell sorting, and Elizabeth A. Chlipala (Premier Laboratory, Boulder, CO) for assistance with immunohistochemistry. This work was supported by National Institutes of Health Grants HL57443, AR051748, and AI055468.


The authors declare no conflict of interest.

*This Direct Submission article had a prearranged editor.

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1000546107/-/DCSupplemental.


1. Udagawa N, et al. Origin of osteoclasts: Mature monocytes and macrophages are capable of differentiating into osteoclasts under a suitable microenvironment prepared by bone marrow–derived stromal cells. Proc Natl Acad Sci USA. 1990;87:7260–7264. [PMC free article] [PubMed]
2. Teitelbaum SL. Bone resorption by osteoclasts. Science. 2000;289:1504–1508. [PubMed]
3. Kodama H, et al. Congenital osteoclast deficiency in osteopetrotic (op/op) mice is cured by injections of macrophage colony-stimulating factor. J Exp Med. 1991;173:269–272. [PMC free article] [PubMed]
4. Lacey DL, et al. Osteoprotegerin ligand is a cytokine that regulates osteoclast differentiation and activation. Cell. 1998;93:165–176. [PubMed]
5. Bucay N, et al. Osteoprotegerin-deficient mice develop early-onset osteoporosis and arterial calcification. Genes Dev. 1998;12:1260–1268. [PMC free article] [PubMed]
6. Yoshida H, et al. The murine mutation osteopetrosis is in the coding region of the macrophage colony-stimulating factor gene. Nature. 1990;345:442–444. [PubMed]
7. Kong YY, et al. OPGL is a key regulator of osteoclastogenesis, lymphocyte development and lymph node organogenesis. Nature. 1999;397:315–323. [PubMed]
8. Dougall WC, et al. RANK is essential for osteoclast and lymph node development. Genes Dev. 1999;13:2412–2424. [PMC free article] [PubMed]
9. Simonet WS, et al. Osteoprotegerin: A novel secreted protein involved in the regulation of bone density. Cell. 1997;89:309–319. [PubMed]
10. Bone HG, et al. Effects of denosumab on bone mineral density and bone turnover in postmenopausal women. J Clin Endocrinol Metab. 2008;93:2149–2157. [PubMed]
11. Walsh NC, Crotti TN, Goldring SR, Gravallese EM. Rheumatic diseases: The effects of inflammation on bone. Immunol Rev. 2005;208:228–251. [PubMed]
12. Geusens PP, et al. The ratio of circulating osteoprotegerin to RANKL in early rheumatoid arthritis predicts later joint destruction. Arthritis Rheum. 2006;54:1772–1777. [PubMed]
13. Cohen SB, et al. Denosumab Rheumatoid Arthritis Study Group Denosumab treatment effects on structural damage, bone mineral density, and bone turnover in rheumatoid arthritis: A twelve-month, multicenter, randomized, double-blind, placebo-controlled, phase II clinical trial. Arthritis Rheum. 2008;58:1299–1309. [PubMed]
14. Lanier LL. NK cell recognition. Annu Rev Immunol. 2005;23:225–274. [PubMed]
15. Cooper MA, et al. Human natural killer cells: A unique innate immunoregulatory role for the CD56(bright) subset. Blood. 2001;97:3146–3151. [PubMed]
16. Martín-Fontecha A, et al. Induced recruitment of NK cells to lymph nodes provides IFN--gamma for T(H)1 priming. Nat Immunol. 2004;5:1260–1265. [PubMed]
17. Wilder JA, Koh CY, Yuan D. The role of NK cells during in vivo antigen-specific antibody responses. J Immunol. 1996;156:146–152. [PubMed]
18. de Matos CT, et al. Activating and inhibitory receptors on synovial fluid natural killer cells of arthritis patients: Role of CD94/NKG2A in control of cytokine secretion. Immunology. 2007;122:291–301. [PMC free article] [PubMed]
19. Tak PP, et al. Granzyme-positive cytotoxic cells are specifically increased in early rheumatoid synovial tissue. Arthritis Rheum. 1994;37:1735–1743. [PubMed]
20. Campbell JJ, et al. Unique subpopulations of CD56+ NK and NK-T peripheral blood lymphocytes identified by chemokine receptor expression repertoire. J Immunol. 2001;166:6477–6482. [PubMed]
21. Dalbeth N, et al. CD56bright NK cells are enriched at inflammatory sites and can engage with monocytes in a reciprocal program of activation. J Immunol. 2004;173:6418–6426. [PubMed]
22. Zhang AL, et al. Natural killer cells trigger differentiation of monocytes into dendritic cells. Blood. 2007;110:2484–2493. [PMC free article] [PubMed]
23. McInnes IB, et al. The role of interleukin-15 in T-cell migration and activation in rheumatoid arthritis. Nat Med. 1996;2:175–182. [PubMed]
24. Carson WE, et al. Interleukin (IL) 15 is a novel cytokine that activates human natural killer cells via components of the IL-2 receptor. J Exp Med. 1994;180:1395–1403. [PMC free article] [PubMed]
25. Wendt K, et al. Gene and protein characteristics reflect functional diversity of CD56dim and CD56bright NK cells. J Leukoc Biol. 2006;80:1529–1541. [PubMed]
26. Takayanagi H, et al. T-cell–mediated regulation of osteoclastogenesis by signaling cross-talk between RANKL and IFN--gamma. Nature. 2000;408:600–605. [PubMed]
27. Fehniger TA, et al. Differential cytokine and chemokine gene expression by human NK cells following activation with IL-18 or IL-15 in combination with IL-12: implications for the innate immune response. J Immunol. 1999;162:4511–4520. [PubMed]
28. Ritchlin CT, Haas-Smith SA, Li P, Hicks DG, Schwarz EM. Mechanisms of TNF-alpha- and RANKL-mediated osteoclastogenesis and bone resorption in psoriatic arthritis. J Clin Invest. 2003;111:821–831. [PMC free article] [PubMed]
29. Vandooren B, Melis L, Veys EM, Tak PP, Baeten D. In vitro spontaneous osteoclastogenesis of human peripheral blood mononuclear cells is not crucially dependent on T lymphocytes. Arthritis Rheum. 2009;60:1020–1025. [PubMed]
30. Lo CK, et al. Natural killer cell degeneration exacerbates experimental arthritis in mice via enhanced interleukin-17 production. Arthritis Rheum. 2008;58:2700–2711. [PubMed]
31. Raza K, et al. Early rheumatoid arthritis is characterized by a distinct and transient synovial fluid cytokine profile of T cell and stromal cell origin. Arthritis Res Ther. 2005;7:R784–R795. [PMC free article] [PubMed]
32. Ruchatz H, Leung BP, Wei XQ, McInnes IB, Liew FY. Soluble IL-15 receptor alpha-chain administration prevents murine collagen-induced arthritis: A role for IL-15 in development of antigen-induced immunopathology. J Immunol. 1998;160:5654–5660. [PubMed]
33. Oppenheimer-Marks N, Brezinschek RI, Mohamadzadeh M, Vita R, Lipsky PE. Interleukin 15 is produced by endothelial cells and increases the transendothelial migration of T cells in vitro and in the SCID mouse-human rheumatoid arthritis model in vivo. J Clin Invest. 1998;101:1261–1272. [PMC free article] [PubMed]
34. Allavena P, Giardina G, Bianchi G, Mantovani A. IL-15 is chemotactic for natural killer cells and stimulates their adhesion to vascular endothelium. J Leukoc Biol. 1997;61:729–735. [PubMed]
35. Ogata Y, et al. A novel role of IL-15 in the development of osteoclasts: Inability to replace its activity with IL-2. J Immunol. 1999;162:2754–2760. [PubMed]
36. Rivollier A, et al. Immature dendritic cell transdifferentiation into osteoclasts: A novel pathway sustained by the rheumatoid arthritis microenvironment. Blood. 2004;104:4029–4037. [PubMed]
37. Firestein GS, Zvaifler NJ. Peripheral blood and synovial fluid monocyte activation in inflammatory arthritis, II: Low levels of synovial fluid and synovial tissue interferon suggest that gamma-interferon is not the primary macrophage activating factor. Arthritis Rheum. 1987;30:864–871. [PubMed]
38. Veys EM, Menkes CJ, Emery P. A randomized, double-blind study comparing twenty-four-week treatment with recombinant interferon-gamma versus placebo in the treatment of rheumatoid arthritis. Arthritis Rheum. 1997;40:62–68. [PubMed]
39. Grcević D, et al. Activated T lymphocytes suppress osteoclastogenesis by diverting early monocyte/macrophage progenitor lineage commitment towards dendritic cell differentiation through down-regulation of receptor activator of nuclear factor-kappaB and c-Fos. Clin Exp Immunol. 2006;146:146–158. [PMC free article] [PubMed]
40. Kotake S, et al. Activated human T cells directly induce osteoclastogenesis from human monocytes: Possible role of T cells in bone destruction in rheumatoid arthritis patients. Arthritis Rheum. 2001;44:1003–1012. [PubMed]
41. Miranda-Carús ME, et al. Peripheral blood T lymphocytes from patients with early rheumatoid arthritis express RANKL and interleukin-15 on the cell surface and promote osteoclastogenesis in autologous monocytes. Arthritis Rheum. 2006;54:1151–1164. [PubMed]
42. Kotake S, et al. IFN–gamma–producing human T cells directly induce osteoclastogenesis from human monocytes via the expression of RANKL. Eur J Immunol. 2005;35:3353–3363. [PubMed]
43. Sato K, et al. Th17 functions as an osteoclastogenic helper T cell subset that links T cell activation and bone destruction. J Exp Med. 2006;203:2673–2682. [PMC free article] [PubMed]
44. Romas E, et al. Osteoprotegerin reduces osteoclast numbers and prevents bone erosion in collagen-induced arthritis. Am J Pathol. 2002;161:1419–1427. [PMC free article] [PubMed]
45. Kamijo S, et al. Amelioration of bone loss in collagen-induced arthritis by neutralizing anti-RANKL monoclonal antibody. Biochem Biophys Res Commun. 2006;347:124–132. [PubMed]
46. Smyth MJ, Crowe NY, Godfrey DI. NK cells and NKT cells collaborate in host protection from methylcholanthrene-induced fibrosarcoma. Int Immunol. 2001;13:459–463. [PubMed]
47. Walzer T, et al. Identification, activation, and selective in vivo ablation of mouse NK cells via NKp46. Proc Natl Acad Sci USA. 2007;104:3384–3389. [PMC free article] [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • MedGen
    Related information in MedGen
  • PubMed
    PubMed citations for these articles
  • Substance
    PubChem Substance links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...