- Email: [email protected]
Interleukin-4 directly inhibits tumor necrosis factor-α-mediated osteoclast formation in mouse bone marrow macrophages
Immunology Letters 88 (2003) 193 /198 www.elsevier.com/locate/
Interleukin-4 directly inhibits tumor necrosis factor-a-mediated osteoclast formation in mouse bone marrow macrophages Hideki Kitaura a, Noriko Nagata a,b, Yuji Fujimura a,b, Hitoshi Hotokezaka a, Mutsuhito Tatamiya a, Noriko Nakao a, Noriaki Yoshida a, Koji Nakayama b,* a
Division of Orthodontic and Biomedical Engineering, Department of Developmental and Reconstructive Medicine, Nagasaki University Graduate School of Biomedical Sciences, 1-7-1 Sakamoto, Nagasaki 852-8588, Japan b Division of Microbiology and Oral Infection, Department of Developmental and Reconstructive Medicine, Nagasaki University Graduate School of Biomedical Sciences, 1-7-1 Sakamoto, Nagasaki 852-8588, Japan Received 10 January 2002; accepted 9 April 2003
Abstract Recently it has been found that osteoclast differentiation is induced by tumor necrosis factor (TNF)-a. Interleukin (IL)-4 was reported to suppress osteoclast differentiation and bone resorption. However, no study has investigated the effect of IL-4 on TNF-ainduced osteoclast formation. In this study, we investigated whether IL-4 inhibits TNF-a-mediated osteoclast formation in mouse bone marrow derived macrophages (BMM). First, IL-4 suppresses RANKL-induced osteoclast formation and bone resorption. Next, when BMM were cultured with TNF-a, osteoclast-like cells were formed. When they were cultured with both TNF-a and IL4, osteoclast formation and bone resorption was suppressed by IL-4 in a dose-dependent manner. It has been recently found that TNF-a and RANKL synergistically promote osteoclastogenesis. Finally, we investigated whether IL-4 had the ability to inhibit synergistic TNF-a and RANKL-induced osteoclastogenesis, with the result that it effectively inhibited the synergistic osteoclast formation in a dose-dependent manner. We conclude that IL-4 can strongly inhibit osteoclast formation that is related to both physiological bone resorption induced by RANKL and pathological bone resorption induced by TNF-a. # 2003 Elsevier Science B.V. All rights reserved. Keywords: TNF-a; IL-4; Osteoclast; RANKL
1. Introduction Osteoclasts, derived from hematopoietic stem cells, control bone resorption [1]. Two factors that induce mature osteoclast formation have been identified. One factor is receptor activator of NF-kB ligand (RANKL) [2], which is also called as osteoclast differentiation factor [3], osteoprotegerin ligand [4], or TNF-related
Abbreviations: RANKL, receptor activator of necrosis factor kBligand; BMM, bone marrow derived macrophage; TRAP, tartrateresistant acid phosphatase; TNF, tumor necrosis factor; TGF, transforming growth factor; IL, interleukin; IFN, interferon; MCSF, macrophage colony stimulating factor; a-MEM, alpha minimal essential medium; FBS, fetal bovine serum; PBS, phosphate buffered saline. * Corresponding author. Tel.: /81-95-849-7648; fax: /81-95-8497650. E-mail address: [email protected] (K. Nakayama).
activation-induced cytokine [5]. OPGL-deficient mice completely lacked osteoclasts and showed severe osteopetrosis [6]. The other factor, macrophage colony stimulating factor (M-CSF), is indispensable for proliferation and differentiation of osteoclast precursors [7]. Osteopetrotic op/op mice which lack M-CSF show deficiency of osteoclast development [8]. It has been reported that TNF-a induces osteoclastlike cells from M-CSF-dependent bone marrow derived macrophages (BMM) in vitro [9 /11]. TNF-a is known to play a major role in host defense, and exerts proinflammatory activities through various cells such as mononuclear phagocytes, in which it is responsible for the activation of bactericidal and cytocidal systems. TNF-a is also involved in differentiation of hematopoietic stem cells into both osteoclasts and macrophages, although their biological roles seem quite different. TNF-a-induced osteoclast recruitment is
0165-2478/03/$ - see front matter # 2003 Elsevier Science B.V. All rights reserved. doi:10.1016/S0165-2478(03)00082-8
194
H. Kitaura et al. / Immunology Letters 88 (2003) 193 /198
probably central to the pathogenesis of disorders with inflammatory osteolysis such as periprosthetic bone loss [12] and periodontal disease [13]. In actuality, TNF-a has been shown to be involved in the causes of postmenopausal osteoporosis [14]. The process of osteoclastogenesis can be controlled by local factors including cytokines, prostaglandins and systemic factors such as sex steroids. Among cytokines, IL-6 [15], IL-17 [16], transforming growth factor (TGF)b [17], TNF-a [9 /11,18] and TNF-b [19] increase osteoclast formation and bone-resorbing activity of osteoclasts. On the other hand, IL-4 [20,21], IL-10 [22], IL-13 [23], IL-18 [24] and interferon (IFN)-g [17] inhibit osteoclast formation and decrease various activities of osteoclasts. Recently, we reported that IL-12 induced apoptosis in TNF-a mediated osteoclast formation in bone marrow cells, resulting in decreased osteoclasts [25]. IL-4 is a pleiotropic immune cytokine secreted from activated Th2 cells that regulates the growth, activity, and survival of certain cells of the lymphoid lineage. IL-4 has been reported to inhibit bone resorption in vivo [26]. Furthermore, in vitro, IL-4 suppressed RANKL-induced osteoclast differentiation through direct action on osteoclast precursors that was independent of supportive cells [20,27]. A direct action by IL-4 on osteoclast precursors has also been hypothesized [20,27]. However, there is no study that investigated the effect of IL-4 on TNF-a-induced osteoclastogenesis. In recent years, it has been reported that TNF-a strongly induced osteoclasts in synergy with RANKL [11,18]. In this study, we showed that IL-4 suppressed TNF-a mediated osteoclast differentiation through direct action on osteoclast precursors that was independent of supportive cells, and that IL-4 also suppressed the synergistic effect of TNF-a on RANKL-induced osteoclastogenesis.
2. Materials and methods
2.1. Animals and reagents Five-week-old male ddY mice were purchased from Seac Yoshitomi, Ltd. (Fukuoka, Japan). Recombinant human M-CSF was purchased from Yoshitomi Pharmaceutical Co., Ltd. (Tokyo, Japan), recombinant mouse TNF-a from R & D systems (Minneapolis, MN, U.S.A.) and recombinant human soluble RANKL from PeproTech EC, Ltd. (London, U.K.). Recombinant mouse IL-4 was obtained from Genetic/Tech, Wako Pure Chemical Industrial, Ltd. (Osaka, Japan).
2.2. Preparation of bone marrow cells The femora and tibiae of mice were aseptically removed and dissected free of adhering tissues. The bone ends were cut off by scissors and the marrow cavity was flushed out by slow injection of alpha minimal essential medium (a-MEM) (SIGMA, Tokyo, Japan) at one end of the bone using a sterile needle to collect bone marrow cells. After washing with a-MEM, cells were incubated in culture medium [a-MEM containing 10% fetal bovine serum (FBS), 100 IU/ml penicillin G (Meiji Seika, Tokyo, Japan), and 100 mg/ml streptomycin (Meiji Seika)]. 2.3. Bone marrow derived macrophages Bone marrow cells were incubated in culture medium supplemented with M-CSF (100 ng/ml) at 1 /107 cells per 10 ml in a 10 cm culture dish. After 3 days culture, cells were washed vigorously with phosphate buffered saline (PBS) twice to remove non-adherent cells, harvested by pipetting with 0.02% EDTA in PBS, and seeded at 1 /106 cells per 10 ml in a 10 cm dish. After 3 days further culture, cells were harvested. We used these cells as BMM in this study [28]. 2.4. Culture for osteoclast formation BMM (1 /105) were cultured in 1 ml of the culture medium with M-CSF (50 ng/ml) and RANKL (50 ng/ ml) or TNF-a (50 ng/ml) using a 48-well plate (Becton Dickinson and Company, NJ). Cultures were maintained at 37 8C in a humidified atmosphere of 5% CO2 in air. 2.5. Tartrate-resistant acid phosphatase (TRAP) staining Cultured cells were fixed with 4% paraformaldehyde for 30 min and then 0.2% Triton X-100 for 5 min at room temperature and incubated in acetate buffer (pH 5.0) containing naphthol AS-MX phosphate (Sigma Chemical Co., St. Louis, MO, USA), fast red /violet LB salt (Sigma Chemical Co.), and 50 mM sodium tartrate. The number of TRAP positive multinuclear cells was counted under light microscopy. 2.6. Resorption pit assay BMM were seeded into each well of calcium phosphate apatite-coated plates (BioCoatTMOsteologicTMBone Cell Culture System, Nippon BD, Tokyo, Japan) in culture medium and incubated in the presence of MCSF (50 ng/ml) and RANKL (50 ng/ml) or TNF-a (50 ng/ml) with or without IL-4 (50 ng/ml) for 3 or 7 days. After incubation, all remaining cells were lysed using 1N
H. Kitaura et al. / Immunology Letters 88 (2003) 193 /198
195
NaOH. The images of resorbed pits were obtained under light microscopy. 2.7. Statistical analysis Differences between data were analyzed with the Student’s t test.
3. Results 3.1. Inhibition of RANKL-induced osteoclast formation and pit formation in BMM culture by IL-4 Mouse BMM were cultured for 3 days in the presence of M-CSF and RANKL. After culturing for 3 days, many osteoclasts were observed in the well (Fig. 1A). To examine the effects of IL-4 on RANKL induced osteoclast formation, we added IL-4 to this culture of BMM at a final concentration of 50 ng/ml. IL-4 markedly inhibited osteoclast formation (Fig. 1A), indicating that IL-4 had an inhibitory effect in BMM culture. We determined the dose-dependent effects of IL-4 on osteoclast formation using various concentrations of IL-4. The inhibitory effect of IL-4 on formation of multinuclear TRAP-positive cells was observed in a dose-dependent manner, even at a concentration of 0.01 ng/ml (Fig. 1B). When BMM were cultured on a calcium phosphate apatite-coated well in culture medium containing M-CSF and RANKL, a number of pits were formed on the surface of the calcium phosphate apatitecoated well at day 3. However, no pit formation was observed in the culture with IL-4 (50 ng/ml) at 3 days culture (Fig. 1C). 3.2. Inhibition of TNF-a induced osteoclast formation and pit formation in BMM culture by IL-4 We investigated if TNF-a could induce osteoclast formation in an in vitro culture system of BMM in the presence of M-CSF and TNF-a. After culturing for 7 days, osteoclasts appeared in the well (Fig. 2A). We added IL-4 to this culture of BMM at a final concentration of 50 ng/ml to examine the effects of IL-4 on TNF-a induced osteoclast formation. IL-4 markedly inhibited osteoclast formation (Fig. 2A). The results indicated that IL-4 had an inhibitory effect on osteoclast formation in BMM culture. To determine the dose-dependent effects of IL-4 on osteoclast formation, we used various concentrations of IL-4. The inhibitory effect of IL-4 on osteoclast formation was observed in a dose-dependent manner, even at a concentration of 0.01 ng/ml (Fig. 2B). We examined if TNF-a induced osteoclasts could form pits from BMM cultured on a calcium phosphate apatite-coated well in culture medium containing MCSF and TNF-a. A number of pits were formed on the
Fig. 1. Inhibition of RANKL induced osteoclast formation and pit formation in BMM culture by IL-4. A: BMM were cultured in the presence of both M-CSF (50 ng/ml) and RANKL (50 ng/ml) in 48-well plates with IL-4 (50 ng/ml) or without IL-4. TRAP staining was performed on day 3. B: BMM were cultured in the presence of both MCSF (50 ng/ml) and RANKL (50 ng/ml) in 48-well plates with various concentration of IL-4. TRAP staining was performed on day 3. C: Effect of IL-4 on RANKL induced pit formation. BMM were cultured in medium containing M-CSF (50 ng/ml) and RANKL (50 ng/ml) with IL-4 (50 ng/ml) or without IL-4 on calcium phosphate apatite-coated plates. After 3-day incubation, the plates were cleaned, and examined under light microscopy. Results were expressed as mean9/SEM of three cultures. *, P B/0.05 and **, P B/0.01 related to the activity of the culture without IL-4.
H. Kitaura et al. / Immunology Letters 88 (2003) 193 /198
196
(1ng/ml) for 3 days, respectively, osteoclast differentiation was not observed (Fig. 3A). However, when we exposed osteoclast precursors to low concentrations of both TNF-a (1ng/ml) and RANKL (1ng/ml) together for 3 days, osteoclasts were induced (Fig. 3A). TNF-a augments RANKL-induced osteoclast differentiation. Mouse BMM were cultured for 3 days in the presence of M-CSF, M-CSF and RANKL, M-CSF, RANKL and IL-4, and M-CSF, RANKL, IL-4 and TNF-a respectively. After culturing for 3 days, osteoclasts were induced in the well with RANKL, while IL-4 inhibited RANKL-induced osteoclast formation. When BMM were exposed to M-CSF, RANKL, IL-4 and TNF-a, osteoclasts were not observed (Fig. 3B, C). Analyzing the data, we conclude that IL-4 also inhibits synergistic TNF-a and RANKL-induced osteoclast formation.
4. Discussion In recent years, it has been reported that TNF-a induces formation of osteoclast in BMM without osteoblasts and stromal cells [9 /11]. The study suggested that even if a small number of stromal cells were present in the BMM preparation, they would not support osteoclast formation. In contrast, another group indicated that TNF-a fails to induce the differentiation of murine osteoclast precursors [18]. In this study, we also examined the effect of TNF-a on osteoclastogenesis and bone resorption. In cultures of BMM with TNF-a and M-CSF, TNF-a induced TRAPpositive cells and bone resorption, confirming that TNF-a was able to induce differentiation into osteoclasts under these conditions. However, we cannot exclude the possibility of contamination with very small amounts of RANKL that could be produced from remaining osteoblasts and stromal cells in this culture. Further studies are necessary to clarify this aspect.
Fig. 2.
surface of a calcium phosphate apatite-coated well at day 7. However, no pit formation was observed in the culture containing IL-4 (50 ng/ml) at 7 days culture (Fig. 2C). 3.3. Inhibitory effect of IL-4 on the synergistic effect of TNF-a on RANKL induced osteoclast formation. When we exposed osteoclast precursors to a low dose concentration of either TNF-a (1ng/ml) or RANKL
Fig. 2. Inhibition of TNF-a induced osteoclast formation and pit formation in BMM culture by IL-4. A: BMM were cultured in the presence of both M-CSF (50 ng/ml) and TNF-a (50 ng/ml) in 48-well plates with IL-4 (50 ng/ml) or without IL-4. TRAP staining was performed on day 7. B: Effect of IL-4 dose on TNF-a induced formation of TRAP-positive cells in BMM. BMM were cultured in the presence of both M-CSF (50 ng/ml) and TNF-a (50 ng/ml) in 48-well plates with various concentrations of IL-4. TRAP staining was performed on day 7. C: Effect of IL-4 on TNF-a induced pit formation. BMM were cultured in medium containing M-CSF (50 ng/ml) and RANKL (50 ng/ml) with IL-4 (50 ng/ml) or without IL-4 on calcium phosphate apatite-coated plates. After 7-day incubation, the plates were cleaned, and examined under light microscopy. Results were expressed as mean9/SEM of three cultures. *, P B/0.05 and **, P B/0.01 related to the activity of the culture without IL-4.
H. Kitaura et al. / Immunology Letters 88 (2003) 193 /198
Fig. 3. Analysis of cancellation of inhibitory effect of IL-4 on the RANKL-induced osteoclast formation by TNF-a. A: TNF-a potently synergizes with RANKL for induction of formation of TRAP-positive cells in BMM. BMM were cultured in the presence of M-CSF (50 ng/ ml) and TNF-a (1 ng/ml) and/or RANKL (1 ng/ml) in 48-well plates. TRAP staining was performed on day 3. B: Microscopic observation. BMM were cultured in the presence of M-CSF (50 ng/ml) and TNF-a (50 ng/ml) and/or RANKL (50 ng/ml) in 48-well plates with IL-4 (50 ng/ml). TRAP staining was performed on day 3. C: The number of TRAP-positive multinuclear cells. Results were expressed as mean9/ SEM of three cultures.
197
IL-4 is known to influence both osteoclasts and osteoblasts. Ovariectomy or parathyroid hormone-related protein-stimulated bone resorption is inhibited by IL-4 in vivo [29,30]. Furthermore, mice overexpressing IL-4 develop systemic bone loss akin to type 2 osteoporosis characterized by attenuated remodeling activity in which osteoblast and osteoclast recruitment is suppressed [31]. IL-4 also decreases release of isotopelabeled calcium from mouse long bone exposed to osteoclastogenic factors [32]. This cytokine also decreases formation of osteoclasts from cultured marrow [27,33]. We initially examined whether IL-4 can suppress RANKL induced osteoclast differentiation through direct action on osteoclast precursors, which is independent of supportive cells. IL-4 markedly inhibited osteoclast formation, indicating that IL-4 had an inhibitory effect on RANKL induced osteoclast formation in BMM culture. Furthermore, our results indicated that inhibition of RANKL-induced osteoclastogenesis by IL4 resulted in inhibition of bone resorption. Our results also supported those reports in which IL-4 inhibited RANKL mediated osteoclast formation. Next, we examined whether IL-4 can suppress TNF-a induced osteoclast differentiation through direct action on osteoclast precursors that is independent of supportive cells. IL-4 also markedly inhibited osteoclast formation. The results indicate that IL-4 had an inhibitory effect on TNF-a induced osteoclast formation in BMM culture. This inhibitory effect on TNF-amediated osteoclast formation by IL-4 resulted in inhibition of bone resorption. However, though TNFa is omnipresent in inflammatory sites, yet osteoclasts are essentially seen only in bone. Probably, TNF-a coexists with other inflammatory cytokines such as IL-4. Because TNF-a is proinflammatory, wherever TNF-a is present, so will be such other inflammatory cytokines. The factors that enable osteoclasts to form inflammatory sites in bone but not elsewhere remain uncertain. TNF-a and RANKL synergistically promote osteoclast formation [18]. In this study, however, there is no cancellation of inhibitory effect of IL-4 by the synergistic effect of TNF-a on RANKL-induced osteoclast formation. Lam et al. suggested that TNF-a and RANKL markedly potentiate NF-kB and SAPK/JNK, two signal pathways essential for osteoclastogenesis [18]. Those signaling pathways may be strongly inhibited by IL-4. In this study, IL-4 was confirmed to inhibit osteoclast formation in mouse BMM treated with M-CSF and RANKL, and was found to inhibit TNF-a-mediated osteoclast formation of mouse BMM. These results suggest that IL-4 can inhibit osteoclast formation that is related to both physiological bone resorption induced by RANKL and pathological bone resorption induced by TNF-a.
198
H. Kitaura et al. / Immunology Letters 88 (2003) 193 /198
Acknowledgements This work was supported in part by a Grant-in-Aid (No. 13771268) for Scientific Research from the Ministry of Education, Science, Sports and Culture, Japan.
References [1] T. Suda, N. Takahashi, T.J. Martin, Endocr. Rev. 13 (1992) 66 / 80. [2] D.M. Anderson, E. Maraskovsky, W.L. Billingsley, W.C. Dougall, M.E. Tometsko, E.R. Roux, M.C. Teepe, R.F. DuBose, D. Cosman, L. Galibert, Nature 390 (1997) 175 /179. [3] H. Yasuda, N. Shima, N. Nakagawa, K. Yamaguchi, M. Kinosaki, S. Mochizuki, A. Tomoyasu, K. Yano, M. Goto, A. Murakami, E. Tsuda, T. Morinaga, K. Higashio, N. Udagawa, N. Takahashi, T. Suda, Proc. Natl. Acad. Sci. USA 95 (1998) 3597 / 3602. [4] D.L. Lacey, E. Timms, H.L. Tan, M.J. Kelley, C.R. Dunstan, T. Burgess, R. Elliott, A. Colombero, G. Elliott, S. Scully, H. Hsu, J. Sullivan, N. Hawkins, E. Davy, C. Capparelli, A. Eli, Y.X. Qian, S. Kaufman, I. Sarosi, V. Shalhoub, G. Senaldi, J. Guo, J. Delaney, W.J. Boyle, Cell 93 (1998) 165 /176. [5] B.R. Wong, J. Rho, J. Arron, E. Robinson, J. Orlinick, M. Chao, S. Kalachikov, E. Cayani, F.3. Bartlett, W.N. Frankel, S.Y. Lee, Y. Choi, J. Biol. Chem. 272 (1997) 25190 /25204. [6] Y.Y. Kong, H. Yoshida, I. Sarosi, H.L. Tan, E. Timms, C. Capparelli, S. Morony, A.J. Oliveira-dos-Santos, G. Van, A. Itie, W. Khoo, A. Wakeham, C.R. Dunstan, D.L. Lacey, T.W. Mak, W.J. Boyle, J.M. Penninger, Nature 397 (1999) 315 /323. [7] H. Kodama, M. Nose, S. Niida, A. Yamasaki, J. Exp. Med. 173 (1991) 1291 /1304. [8] W. Wiktor-Jedrzejczak, A. Bartocci, A.J. Ferrante, A. AhmedAmsari, K.W. Sell, J.W. Pollard, E.R. Stanley, Proc. Natl. Acad. Sci. USA 87 (1990) 4828 /4832. [9] K. Kobayashi, N. Takahashi, E. Jimi, N. Udagawa, M. Takami, S. Kotake, N. Nakagawa, M. Kinosaki, K. Yamaguchi, N. Shima, H. Yasuda, T. Morinaga, K. Higashio, T.J. Martin, T. Suda, J. Exp. Med. 191 (2000) 275 /286. [10] Y. Azuma, K. Kaji, R. Katogi, S. Takeshita, A. Kudo, J. Biol. Chem. 275 (2000) 4858 /4864. [11] K. Fuller, C. Murphy, B. Kirstein, S.W. Fox, T.J. Chambers, Endocrinology 143 (2002) 1108 /1118. [12] K.D. Merkel, J.M. Erdmann, K.P. McHugh, A.Y. Abu, F.P. Ross, S.L. Teitelbaum, Am. J. Pathol. 154 (1999) 203 /210. [13] A.Y. Abu, F.P. Ross, J. Edwards, S.L. Teitelbaum, J. Clin. Invest. 100 (1997) 1557 /1565.
[14] R.B. Kimble, S. Srivastava, F.P. Ross, A. Matayoshi, R. Pacifici, J. Biol. Chem. 271 (1996) 28890 /28897. [15] Y. Gao, I. Morita, N. Maruo, T. Kubota, S. Murota, T. Aso, Bone 22 (1998) 487 /493. [16] S. Kotake, N. Udagawa, N. Takahashi, K. Matsuzaki, K. Itoh, S. Ishiyama, S. Saito, K. Inoue, N. Kamatani, M.T. Gillespie, T.J. Martin, T. Suda, J. Clin. Invest. 103 (1999) 1345 /1352. [17] S.W. Fox, K. Fuller, K.E. Bayley, J.M. Lean, T.J. Chambers, J. Immunol. 165 (2000) 4957 /4963. [18] J. Lam, S. Takeshita, J.E. Barker, O. Kanagawa, F.P. Ross, S.L. Teitelbaum, J. Clin. Invest. 106 (2000) 1481 /1488. [19] J. Pfeilschifter, C. Chenu, A. Bird, G.R. Mundy, G.D. Roodman, J. Bone Miner. Res. 4 (1989) 113 /118. [20] S. Wei, M.W. Wang, S.L. Teitelbaum, F.P. Ross, J. Biol. Chem. 277 (2002) 6622 /6630. [21] J.A. Riancho, M.T. Zarrabeitia, G.R. Mundy, T. Yoneda, M.J. Gonzalez, J. Bone Miner. Res. 8 (1993) 1337 /1344. [22] J.M. Owens, A.C. Gallagher, T.J. Chambers, J. Immunol. 157 (1996) 936 /940. [23] Y. Onoe, C. Miyaura, T. Kaminakayashiki, Y. Nagai, K. Noguchi, Q.R. Chen, H. Seo, H. Ohta, S. Nozawa, I. Kudo, T. Suda, J. Immunol. 156 (1996) 758 /764. [24] N. Udagawa, N.J. Horwood, J. Elliott, A. Mackay, J. Owens, H. Okamura, M. Kurimoto, T.J. Chambers, T.J. Martin, M.T. Gillespie, J. Exp. Med. 185 (1997) 1005 /1012. [25] H. Kitaura, N. Nagata, Y. Fujimura, H. Hotokezaka, N. Yoshida, K. Nakayama, J. Immunol. 169 (2002) 4732 /4738. [26] E. Lubberts, L.A. Joosten, M. Chabaud, L. van den Bersselaar, B. Oppers, C.J.J. Coenen-De Roo, C.D. Richards, P. Miossec, W.B. van den Berg, J. Clin. Invest. 105 (2001) 1697 /1710. [27] D.L. Lacey, J.M. Erdmann, S.L. Teitelbaum, H.L. Tan, J. Ohara, A. Shioi, Endocrinol. 136 (1995) 2367 /2376. [28] S. Takeshita, K. Kaji, A. Kudo, J. Bone Miner. Res. 15 (2000) 1477 /1488. [29] Y. Nakano, K. Watanabe, I. Morimoto, Y. Okada, K. Ura, K. Sato, K. Kasono, T. Nakamura, S. Eto, J. Bone Miner. Res. 9 (1994) 1533 /1539. [30] Y. Okada, I. Morimoto, K. Ura, Y. Nakano, Y. Tanaka, S. Nishida, T. Nakamura, S. Eto, Bone 22 (1998) 361 /365. [31] D.B. Lewis, H.D. Liggitt, E.L. Effmann, S.T. Motley, S.L. Teitelbaum, K.J. Jepsen, S.A. Goldstein, J. Bonadio, J. Carpenter, R.M. Perlmutter, Proc. Natl. Acad. Sci. USA 90 (1993) 11618 /11622. [32] K. Watanabe, Y. Tanaka, I. Morimoto, K. Yahata, K. Zeki, T. Fujihira, U. Yamashita, S. Eto, Biochem. Biophys. Res. Commun. 172 (1990) 1035 /1041. [33] A. Shioi, S.L. Teitelbaum, F.P. Ross, H.G. Welgus, H. Suzuki, J. Ohara, D.L. Lacey, J. Cell Biochem. 47 (1991) 177 /272.
Interleukin-4 directly inhibits tumor necrosis factor-a-mediated osteoclast formation in mouse bone marrow macrophages Hideki Kitaura a, Noriko Nagata a,b, Yuji Fujimura a,b, Hitoshi Hotokezaka a, Mutsuhito Tatamiya a, Noriko Nakao a, Noriaki Yoshida a, Koji Nakayama b,* a
Division of Orthodontic and Biomedical Engineering, Department of Developmental and Reconstructive Medicine, Nagasaki University Graduate School of Biomedical Sciences, 1-7-1 Sakamoto, Nagasaki 852-8588, Japan b Division of Microbiology and Oral Infection, Department of Developmental and Reconstructive Medicine, Nagasaki University Graduate School of Biomedical Sciences, 1-7-1 Sakamoto, Nagasaki 852-8588, Japan Received 10 January 2002; accepted 9 April 2003
Abstract Recently it has been found that osteoclast differentiation is induced by tumor necrosis factor (TNF)-a. Interleukin (IL)-4 was reported to suppress osteoclast differentiation and bone resorption. However, no study has investigated the effect of IL-4 on TNF-ainduced osteoclast formation. In this study, we investigated whether IL-4 inhibits TNF-a-mediated osteoclast formation in mouse bone marrow derived macrophages (BMM). First, IL-4 suppresses RANKL-induced osteoclast formation and bone resorption. Next, when BMM were cultured with TNF-a, osteoclast-like cells were formed. When they were cultured with both TNF-a and IL4, osteoclast formation and bone resorption was suppressed by IL-4 in a dose-dependent manner. It has been recently found that TNF-a and RANKL synergistically promote osteoclastogenesis. Finally, we investigated whether IL-4 had the ability to inhibit synergistic TNF-a and RANKL-induced osteoclastogenesis, with the result that it effectively inhibited the synergistic osteoclast formation in a dose-dependent manner. We conclude that IL-4 can strongly inhibit osteoclast formation that is related to both physiological bone resorption induced by RANKL and pathological bone resorption induced by TNF-a. # 2003 Elsevier Science B.V. All rights reserved. Keywords: TNF-a; IL-4; Osteoclast; RANKL
1. Introduction Osteoclasts, derived from hematopoietic stem cells, control bone resorption [1]. Two factors that induce mature osteoclast formation have been identified. One factor is receptor activator of NF-kB ligand (RANKL) [2], which is also called as osteoclast differentiation factor [3], osteoprotegerin ligand [4], or TNF-related
Abbreviations: RANKL, receptor activator of necrosis factor kBligand; BMM, bone marrow derived macrophage; TRAP, tartrateresistant acid phosphatase; TNF, tumor necrosis factor; TGF, transforming growth factor; IL, interleukin; IFN, interferon; MCSF, macrophage colony stimulating factor; a-MEM, alpha minimal essential medium; FBS, fetal bovine serum; PBS, phosphate buffered saline. * Corresponding author. Tel.: /81-95-849-7648; fax: /81-95-8497650. E-mail address: [email protected] (K. Nakayama).
activation-induced cytokine [5]. OPGL-deficient mice completely lacked osteoclasts and showed severe osteopetrosis [6]. The other factor, macrophage colony stimulating factor (M-CSF), is indispensable for proliferation and differentiation of osteoclast precursors [7]. Osteopetrotic op/op mice which lack M-CSF show deficiency of osteoclast development [8]. It has been reported that TNF-a induces osteoclastlike cells from M-CSF-dependent bone marrow derived macrophages (BMM) in vitro [9 /11]. TNF-a is known to play a major role in host defense, and exerts proinflammatory activities through various cells such as mononuclear phagocytes, in which it is responsible for the activation of bactericidal and cytocidal systems. TNF-a is also involved in differentiation of hematopoietic stem cells into both osteoclasts and macrophages, although their biological roles seem quite different. TNF-a-induced osteoclast recruitment is
0165-2478/03/$ - see front matter # 2003 Elsevier Science B.V. All rights reserved. doi:10.1016/S0165-2478(03)00082-8
194
H. Kitaura et al. / Immunology Letters 88 (2003) 193 /198
probably central to the pathogenesis of disorders with inflammatory osteolysis such as periprosthetic bone loss [12] and periodontal disease [13]. In actuality, TNF-a has been shown to be involved in the causes of postmenopausal osteoporosis [14]. The process of osteoclastogenesis can be controlled by local factors including cytokines, prostaglandins and systemic factors such as sex steroids. Among cytokines, IL-6 [15], IL-17 [16], transforming growth factor (TGF)b [17], TNF-a [9 /11,18] and TNF-b [19] increase osteoclast formation and bone-resorbing activity of osteoclasts. On the other hand, IL-4 [20,21], IL-10 [22], IL-13 [23], IL-18 [24] and interferon (IFN)-g [17] inhibit osteoclast formation and decrease various activities of osteoclasts. Recently, we reported that IL-12 induced apoptosis in TNF-a mediated osteoclast formation in bone marrow cells, resulting in decreased osteoclasts [25]. IL-4 is a pleiotropic immune cytokine secreted from activated Th2 cells that regulates the growth, activity, and survival of certain cells of the lymphoid lineage. IL-4 has been reported to inhibit bone resorption in vivo [26]. Furthermore, in vitro, IL-4 suppressed RANKL-induced osteoclast differentiation through direct action on osteoclast precursors that was independent of supportive cells [20,27]. A direct action by IL-4 on osteoclast precursors has also been hypothesized [20,27]. However, there is no study that investigated the effect of IL-4 on TNF-a-induced osteoclastogenesis. In recent years, it has been reported that TNF-a strongly induced osteoclasts in synergy with RANKL [11,18]. In this study, we showed that IL-4 suppressed TNF-a mediated osteoclast differentiation through direct action on osteoclast precursors that was independent of supportive cells, and that IL-4 also suppressed the synergistic effect of TNF-a on RANKL-induced osteoclastogenesis.
2. Materials and methods
2.1. Animals and reagents Five-week-old male ddY mice were purchased from Seac Yoshitomi, Ltd. (Fukuoka, Japan). Recombinant human M-CSF was purchased from Yoshitomi Pharmaceutical Co., Ltd. (Tokyo, Japan), recombinant mouse TNF-a from R & D systems (Minneapolis, MN, U.S.A.) and recombinant human soluble RANKL from PeproTech EC, Ltd. (London, U.K.). Recombinant mouse IL-4 was obtained from Genetic/Tech, Wako Pure Chemical Industrial, Ltd. (Osaka, Japan).
2.2. Preparation of bone marrow cells The femora and tibiae of mice were aseptically removed and dissected free of adhering tissues. The bone ends were cut off by scissors and the marrow cavity was flushed out by slow injection of alpha minimal essential medium (a-MEM) (SIGMA, Tokyo, Japan) at one end of the bone using a sterile needle to collect bone marrow cells. After washing with a-MEM, cells were incubated in culture medium [a-MEM containing 10% fetal bovine serum (FBS), 100 IU/ml penicillin G (Meiji Seika, Tokyo, Japan), and 100 mg/ml streptomycin (Meiji Seika)]. 2.3. Bone marrow derived macrophages Bone marrow cells were incubated in culture medium supplemented with M-CSF (100 ng/ml) at 1 /107 cells per 10 ml in a 10 cm culture dish. After 3 days culture, cells were washed vigorously with phosphate buffered saline (PBS) twice to remove non-adherent cells, harvested by pipetting with 0.02% EDTA in PBS, and seeded at 1 /106 cells per 10 ml in a 10 cm dish. After 3 days further culture, cells were harvested. We used these cells as BMM in this study [28]. 2.4. Culture for osteoclast formation BMM (1 /105) were cultured in 1 ml of the culture medium with M-CSF (50 ng/ml) and RANKL (50 ng/ ml) or TNF-a (50 ng/ml) using a 48-well plate (Becton Dickinson and Company, NJ). Cultures were maintained at 37 8C in a humidified atmosphere of 5% CO2 in air. 2.5. Tartrate-resistant acid phosphatase (TRAP) staining Cultured cells were fixed with 4% paraformaldehyde for 30 min and then 0.2% Triton X-100 for 5 min at room temperature and incubated in acetate buffer (pH 5.0) containing naphthol AS-MX phosphate (Sigma Chemical Co., St. Louis, MO, USA), fast red /violet LB salt (Sigma Chemical Co.), and 50 mM sodium tartrate. The number of TRAP positive multinuclear cells was counted under light microscopy. 2.6. Resorption pit assay BMM were seeded into each well of calcium phosphate apatite-coated plates (BioCoatTMOsteologicTMBone Cell Culture System, Nippon BD, Tokyo, Japan) in culture medium and incubated in the presence of MCSF (50 ng/ml) and RANKL (50 ng/ml) or TNF-a (50 ng/ml) with or without IL-4 (50 ng/ml) for 3 or 7 days. After incubation, all remaining cells were lysed using 1N
H. Kitaura et al. / Immunology Letters 88 (2003) 193 /198
195
NaOH. The images of resorbed pits were obtained under light microscopy. 2.7. Statistical analysis Differences between data were analyzed with the Student’s t test.
3. Results 3.1. Inhibition of RANKL-induced osteoclast formation and pit formation in BMM culture by IL-4 Mouse BMM were cultured for 3 days in the presence of M-CSF and RANKL. After culturing for 3 days, many osteoclasts were observed in the well (Fig. 1A). To examine the effects of IL-4 on RANKL induced osteoclast formation, we added IL-4 to this culture of BMM at a final concentration of 50 ng/ml. IL-4 markedly inhibited osteoclast formation (Fig. 1A), indicating that IL-4 had an inhibitory effect in BMM culture. We determined the dose-dependent effects of IL-4 on osteoclast formation using various concentrations of IL-4. The inhibitory effect of IL-4 on formation of multinuclear TRAP-positive cells was observed in a dose-dependent manner, even at a concentration of 0.01 ng/ml (Fig. 1B). When BMM were cultured on a calcium phosphate apatite-coated well in culture medium containing M-CSF and RANKL, a number of pits were formed on the surface of the calcium phosphate apatitecoated well at day 3. However, no pit formation was observed in the culture with IL-4 (50 ng/ml) at 3 days culture (Fig. 1C). 3.2. Inhibition of TNF-a induced osteoclast formation and pit formation in BMM culture by IL-4 We investigated if TNF-a could induce osteoclast formation in an in vitro culture system of BMM in the presence of M-CSF and TNF-a. After culturing for 7 days, osteoclasts appeared in the well (Fig. 2A). We added IL-4 to this culture of BMM at a final concentration of 50 ng/ml to examine the effects of IL-4 on TNF-a induced osteoclast formation. IL-4 markedly inhibited osteoclast formation (Fig. 2A). The results indicated that IL-4 had an inhibitory effect on osteoclast formation in BMM culture. To determine the dose-dependent effects of IL-4 on osteoclast formation, we used various concentrations of IL-4. The inhibitory effect of IL-4 on osteoclast formation was observed in a dose-dependent manner, even at a concentration of 0.01 ng/ml (Fig. 2B). We examined if TNF-a induced osteoclasts could form pits from BMM cultured on a calcium phosphate apatite-coated well in culture medium containing MCSF and TNF-a. A number of pits were formed on the
Fig. 1. Inhibition of RANKL induced osteoclast formation and pit formation in BMM culture by IL-4. A: BMM were cultured in the presence of both M-CSF (50 ng/ml) and RANKL (50 ng/ml) in 48-well plates with IL-4 (50 ng/ml) or without IL-4. TRAP staining was performed on day 3. B: BMM were cultured in the presence of both MCSF (50 ng/ml) and RANKL (50 ng/ml) in 48-well plates with various concentration of IL-4. TRAP staining was performed on day 3. C: Effect of IL-4 on RANKL induced pit formation. BMM were cultured in medium containing M-CSF (50 ng/ml) and RANKL (50 ng/ml) with IL-4 (50 ng/ml) or without IL-4 on calcium phosphate apatite-coated plates. After 3-day incubation, the plates were cleaned, and examined under light microscopy. Results were expressed as mean9/SEM of three cultures. *, P B/0.05 and **, P B/0.01 related to the activity of the culture without IL-4.
H. Kitaura et al. / Immunology Letters 88 (2003) 193 /198
196
(1ng/ml) for 3 days, respectively, osteoclast differentiation was not observed (Fig. 3A). However, when we exposed osteoclast precursors to low concentrations of both TNF-a (1ng/ml) and RANKL (1ng/ml) together for 3 days, osteoclasts were induced (Fig. 3A). TNF-a augments RANKL-induced osteoclast differentiation. Mouse BMM were cultured for 3 days in the presence of M-CSF, M-CSF and RANKL, M-CSF, RANKL and IL-4, and M-CSF, RANKL, IL-4 and TNF-a respectively. After culturing for 3 days, osteoclasts were induced in the well with RANKL, while IL-4 inhibited RANKL-induced osteoclast formation. When BMM were exposed to M-CSF, RANKL, IL-4 and TNF-a, osteoclasts were not observed (Fig. 3B, C). Analyzing the data, we conclude that IL-4 also inhibits synergistic TNF-a and RANKL-induced osteoclast formation.
4. Discussion In recent years, it has been reported that TNF-a induces formation of osteoclast in BMM without osteoblasts and stromal cells [9 /11]. The study suggested that even if a small number of stromal cells were present in the BMM preparation, they would not support osteoclast formation. In contrast, another group indicated that TNF-a fails to induce the differentiation of murine osteoclast precursors [18]. In this study, we also examined the effect of TNF-a on osteoclastogenesis and bone resorption. In cultures of BMM with TNF-a and M-CSF, TNF-a induced TRAPpositive cells and bone resorption, confirming that TNF-a was able to induce differentiation into osteoclasts under these conditions. However, we cannot exclude the possibility of contamination with very small amounts of RANKL that could be produced from remaining osteoblasts and stromal cells in this culture. Further studies are necessary to clarify this aspect.
Fig. 2.
surface of a calcium phosphate apatite-coated well at day 7. However, no pit formation was observed in the culture containing IL-4 (50 ng/ml) at 7 days culture (Fig. 2C). 3.3. Inhibitory effect of IL-4 on the synergistic effect of TNF-a on RANKL induced osteoclast formation. When we exposed osteoclast precursors to a low dose concentration of either TNF-a (1ng/ml) or RANKL
Fig. 2. Inhibition of TNF-a induced osteoclast formation and pit formation in BMM culture by IL-4. A: BMM were cultured in the presence of both M-CSF (50 ng/ml) and TNF-a (50 ng/ml) in 48-well plates with IL-4 (50 ng/ml) or without IL-4. TRAP staining was performed on day 7. B: Effect of IL-4 dose on TNF-a induced formation of TRAP-positive cells in BMM. BMM were cultured in the presence of both M-CSF (50 ng/ml) and TNF-a (50 ng/ml) in 48-well plates with various concentrations of IL-4. TRAP staining was performed on day 7. C: Effect of IL-4 on TNF-a induced pit formation. BMM were cultured in medium containing M-CSF (50 ng/ml) and RANKL (50 ng/ml) with IL-4 (50 ng/ml) or without IL-4 on calcium phosphate apatite-coated plates. After 7-day incubation, the plates were cleaned, and examined under light microscopy. Results were expressed as mean9/SEM of three cultures. *, P B/0.05 and **, P B/0.01 related to the activity of the culture without IL-4.
H. Kitaura et al. / Immunology Letters 88 (2003) 193 /198
Fig. 3. Analysis of cancellation of inhibitory effect of IL-4 on the RANKL-induced osteoclast formation by TNF-a. A: TNF-a potently synergizes with RANKL for induction of formation of TRAP-positive cells in BMM. BMM were cultured in the presence of M-CSF (50 ng/ ml) and TNF-a (1 ng/ml) and/or RANKL (1 ng/ml) in 48-well plates. TRAP staining was performed on day 3. B: Microscopic observation. BMM were cultured in the presence of M-CSF (50 ng/ml) and TNF-a (50 ng/ml) and/or RANKL (50 ng/ml) in 48-well plates with IL-4 (50 ng/ml). TRAP staining was performed on day 3. C: The number of TRAP-positive multinuclear cells. Results were expressed as mean9/ SEM of three cultures.
197
IL-4 is known to influence both osteoclasts and osteoblasts. Ovariectomy or parathyroid hormone-related protein-stimulated bone resorption is inhibited by IL-4 in vivo [29,30]. Furthermore, mice overexpressing IL-4 develop systemic bone loss akin to type 2 osteoporosis characterized by attenuated remodeling activity in which osteoblast and osteoclast recruitment is suppressed [31]. IL-4 also decreases release of isotopelabeled calcium from mouse long bone exposed to osteoclastogenic factors [32]. This cytokine also decreases formation of osteoclasts from cultured marrow [27,33]. We initially examined whether IL-4 can suppress RANKL induced osteoclast differentiation through direct action on osteoclast precursors, which is independent of supportive cells. IL-4 markedly inhibited osteoclast formation, indicating that IL-4 had an inhibitory effect on RANKL induced osteoclast formation in BMM culture. Furthermore, our results indicated that inhibition of RANKL-induced osteoclastogenesis by IL4 resulted in inhibition of bone resorption. Our results also supported those reports in which IL-4 inhibited RANKL mediated osteoclast formation. Next, we examined whether IL-4 can suppress TNF-a induced osteoclast differentiation through direct action on osteoclast precursors that is independent of supportive cells. IL-4 also markedly inhibited osteoclast formation. The results indicate that IL-4 had an inhibitory effect on TNF-a induced osteoclast formation in BMM culture. This inhibitory effect on TNF-amediated osteoclast formation by IL-4 resulted in inhibition of bone resorption. However, though TNFa is omnipresent in inflammatory sites, yet osteoclasts are essentially seen only in bone. Probably, TNF-a coexists with other inflammatory cytokines such as IL-4. Because TNF-a is proinflammatory, wherever TNF-a is present, so will be such other inflammatory cytokines. The factors that enable osteoclasts to form inflammatory sites in bone but not elsewhere remain uncertain. TNF-a and RANKL synergistically promote osteoclast formation [18]. In this study, however, there is no cancellation of inhibitory effect of IL-4 by the synergistic effect of TNF-a on RANKL-induced osteoclast formation. Lam et al. suggested that TNF-a and RANKL markedly potentiate NF-kB and SAPK/JNK, two signal pathways essential for osteoclastogenesis [18]. Those signaling pathways may be strongly inhibited by IL-4. In this study, IL-4 was confirmed to inhibit osteoclast formation in mouse BMM treated with M-CSF and RANKL, and was found to inhibit TNF-a-mediated osteoclast formation of mouse BMM. These results suggest that IL-4 can inhibit osteoclast formation that is related to both physiological bone resorption induced by RANKL and pathological bone resorption induced by TNF-a.
198
H. Kitaura et al. / Immunology Letters 88 (2003) 193 /198
Acknowledgements This work was supported in part by a Grant-in-Aid (No. 13771268) for Scientific Research from the Ministry of Education, Science, Sports and Culture, Japan.
References [1] T. Suda, N. Takahashi, T.J. Martin, Endocr. Rev. 13 (1992) 66 / 80. [2] D.M. Anderson, E. Maraskovsky, W.L. Billingsley, W.C. Dougall, M.E. Tometsko, E.R. Roux, M.C. Teepe, R.F. DuBose, D. Cosman, L. Galibert, Nature 390 (1997) 175 /179. [3] H. Yasuda, N. Shima, N. Nakagawa, K. Yamaguchi, M. Kinosaki, S. Mochizuki, A. Tomoyasu, K. Yano, M. Goto, A. Murakami, E. Tsuda, T. Morinaga, K. Higashio, N. Udagawa, N. Takahashi, T. Suda, Proc. Natl. Acad. Sci. USA 95 (1998) 3597 / 3602. [4] D.L. Lacey, E. Timms, H.L. Tan, M.J. Kelley, C.R. Dunstan, T. Burgess, R. Elliott, A. Colombero, G. Elliott, S. Scully, H. Hsu, J. Sullivan, N. Hawkins, E. Davy, C. Capparelli, A. Eli, Y.X. Qian, S. Kaufman, I. Sarosi, V. Shalhoub, G. Senaldi, J. Guo, J. Delaney, W.J. Boyle, Cell 93 (1998) 165 /176. [5] B.R. Wong, J. Rho, J. Arron, E. Robinson, J. Orlinick, M. Chao, S. Kalachikov, E. Cayani, F.3. Bartlett, W.N. Frankel, S.Y. Lee, Y. Choi, J. Biol. Chem. 272 (1997) 25190 /25204. [6] Y.Y. Kong, H. Yoshida, I. Sarosi, H.L. Tan, E. Timms, C. Capparelli, S. Morony, A.J. Oliveira-dos-Santos, G. Van, A. Itie, W. Khoo, A. Wakeham, C.R. Dunstan, D.L. Lacey, T.W. Mak, W.J. Boyle, J.M. Penninger, Nature 397 (1999) 315 /323. [7] H. Kodama, M. Nose, S. Niida, A. Yamasaki, J. Exp. Med. 173 (1991) 1291 /1304. [8] W. Wiktor-Jedrzejczak, A. Bartocci, A.J. Ferrante, A. AhmedAmsari, K.W. Sell, J.W. Pollard, E.R. Stanley, Proc. Natl. Acad. Sci. USA 87 (1990) 4828 /4832. [9] K. Kobayashi, N. Takahashi, E. Jimi, N. Udagawa, M. Takami, S. Kotake, N. Nakagawa, M. Kinosaki, K. Yamaguchi, N. Shima, H. Yasuda, T. Morinaga, K. Higashio, T.J. Martin, T. Suda, J. Exp. Med. 191 (2000) 275 /286. [10] Y. Azuma, K. Kaji, R. Katogi, S. Takeshita, A. Kudo, J. Biol. Chem. 275 (2000) 4858 /4864. [11] K. Fuller, C. Murphy, B. Kirstein, S.W. Fox, T.J. Chambers, Endocrinology 143 (2002) 1108 /1118. [12] K.D. Merkel, J.M. Erdmann, K.P. McHugh, A.Y. Abu, F.P. Ross, S.L. Teitelbaum, Am. J. Pathol. 154 (1999) 203 /210. [13] A.Y. Abu, F.P. Ross, J. Edwards, S.L. Teitelbaum, J. Clin. Invest. 100 (1997) 1557 /1565.
[14] R.B. Kimble, S. Srivastava, F.P. Ross, A. Matayoshi, R. Pacifici, J. Biol. Chem. 271 (1996) 28890 /28897. [15] Y. Gao, I. Morita, N. Maruo, T. Kubota, S. Murota, T. Aso, Bone 22 (1998) 487 /493. [16] S. Kotake, N. Udagawa, N. Takahashi, K. Matsuzaki, K. Itoh, S. Ishiyama, S. Saito, K. Inoue, N. Kamatani, M.T. Gillespie, T.J. Martin, T. Suda, J. Clin. Invest. 103 (1999) 1345 /1352. [17] S.W. Fox, K. Fuller, K.E. Bayley, J.M. Lean, T.J. Chambers, J. Immunol. 165 (2000) 4957 /4963. [18] J. Lam, S. Takeshita, J.E. Barker, O. Kanagawa, F.P. Ross, S.L. Teitelbaum, J. Clin. Invest. 106 (2000) 1481 /1488. [19] J. Pfeilschifter, C. Chenu, A. Bird, G.R. Mundy, G.D. Roodman, J. Bone Miner. Res. 4 (1989) 113 /118. [20] S. Wei, M.W. Wang, S.L. Teitelbaum, F.P. Ross, J. Biol. Chem. 277 (2002) 6622 /6630. [21] J.A. Riancho, M.T. Zarrabeitia, G.R. Mundy, T. Yoneda, M.J. Gonzalez, J. Bone Miner. Res. 8 (1993) 1337 /1344. [22] J.M. Owens, A.C. Gallagher, T.J. Chambers, J. Immunol. 157 (1996) 936 /940. [23] Y. Onoe, C. Miyaura, T. Kaminakayashiki, Y. Nagai, K. Noguchi, Q.R. Chen, H. Seo, H. Ohta, S. Nozawa, I. Kudo, T. Suda, J. Immunol. 156 (1996) 758 /764. [24] N. Udagawa, N.J. Horwood, J. Elliott, A. Mackay, J. Owens, H. Okamura, M. Kurimoto, T.J. Chambers, T.J. Martin, M.T. Gillespie, J. Exp. Med. 185 (1997) 1005 /1012. [25] H. Kitaura, N. Nagata, Y. Fujimura, H. Hotokezaka, N. Yoshida, K. Nakayama, J. Immunol. 169 (2002) 4732 /4738. [26] E. Lubberts, L.A. Joosten, M. Chabaud, L. van den Bersselaar, B. Oppers, C.J.J. Coenen-De Roo, C.D. Richards, P. Miossec, W.B. van den Berg, J. Clin. Invest. 105 (2001) 1697 /1710. [27] D.L. Lacey, J.M. Erdmann, S.L. Teitelbaum, H.L. Tan, J. Ohara, A. Shioi, Endocrinol. 136 (1995) 2367 /2376. [28] S. Takeshita, K. Kaji, A. Kudo, J. Bone Miner. Res. 15 (2000) 1477 /1488. [29] Y. Nakano, K. Watanabe, I. Morimoto, Y. Okada, K. Ura, K. Sato, K. Kasono, T. Nakamura, S. Eto, J. Bone Miner. Res. 9 (1994) 1533 /1539. [30] Y. Okada, I. Morimoto, K. Ura, Y. Nakano, Y. Tanaka, S. Nishida, T. Nakamura, S. Eto, Bone 22 (1998) 361 /365. [31] D.B. Lewis, H.D. Liggitt, E.L. Effmann, S.T. Motley, S.L. Teitelbaum, K.J. Jepsen, S.A. Goldstein, J. Bonadio, J. Carpenter, R.M. Perlmutter, Proc. Natl. Acad. Sci. USA 90 (1993) 11618 /11622. [32] K. Watanabe, Y. Tanaka, I. Morimoto, K. Yahata, K. Zeki, T. Fujihira, U. Yamashita, S. Eto, Biochem. Biophys. Res. Commun. 172 (1990) 1035 /1041. [33] A. Shioi, S.L. Teitelbaum, F.P. Ross, H.G. Welgus, H. Suzuki, J. Ohara, D.L. Lacey, J. Cell Biochem. 47 (1991) 177 /272.