JP2004154132A - Method for screening compound inhibiting osteoclastgenesis - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method for controlling osteoclastgenesis. <P>SOLUTION: The method for accelerating or inhibiting the osteoclastgenesis includes a process for accelerating or inhibiting Ca<SP>2+</SP>signal transmission in osteoclast precursor cells, respectively. NFATcl which is a Ca<SP>2+</SP>signal transmission molecule is proved to be a transcription factor essential to the osteoblast formation. The method inhibits the osteoclastgenesis by using a compound which inhibits the signal transmission through NFATcl. A method for screening a compound for regulating the osteoclastgenesis is provided. A system for forming osteoclast cells from ES cells in vitro is also provided. The compound is useful for developing a therapeutic agent for bone reduction diseases including rheumatoid arthritis. <P>COPYRIGHT: (C)2004,JPO

Description

  The present invention relates to a method for controlling an osteoclast-forming signal. More specifically, it relates to a method of controlling osteoclast formation.

  In cell differentiation, signaling induced by cell surface receptors activates transcription factors, which promote transcription of downstream effector molecules or other transcription factor (s), which To determine the fate of the cell line. Identifying the sequence of events in cell type specification involving multiple transcription factors is important for understanding the molecular mechanisms of cell differentiation and skeletal morphogenesis (Karsenty, G., and Wagner, EF (2002) Dev Cell 2, 389-406).

  Osteoclasts are multinucleated cells that are exclusively responsible for both physiological and pathological bone resorption (Manolagas, SC (2000) Endocr Rev 21, 115-137; Teitelbaum, SL (2000) Science 289, 1504 -1508). Bone resorption by osteoclasts must be coordinated with osteoblastic bone formation, and when this balanced regulation fails, osteoporosis, autoimmune arthritis, Paget's disease, and bone cancer Into various osteopenic disease states (Rodan, GA, and Martin, TJ (2000) Science 289, 1508-1514; Takayanagi, H. et al. (2000a) Arthritis Rheum 43, 259- 269; Takayanagi, H. et al., J. Clin.Invest. 104, 137-46 (1999); Kong, YY et al., Nature 402, 304-9 (1999); Pettit, AR et al., Am J Pathol 159, 1689-99 (2001)). Therefore, research on the regulation mechanism of osteoclast formation has been a central issue in bone cell biology.

  Monocyte / macrophage progenitor cells derived from hematopoietic stem cells develop into osteoclasts through intercellular signaling with mesenchymal cells including osteoblasts (Chambers, TJ (2000) J Pathol 192, 4-13; Suda, T. et al. (1999) Endocrine Rev 20, 345-357). The essential molecules for this intercellular signaling are RANKL and M-CSF (macrophage-colony stimulating factor). RANKL (receptor activator of nuclear factor κB ligand), also known as tnfs11, TRANCE, OPGL, ODF, etc., is one of the TNF family cytokines and a powerful and exclusive inducer of osteoclastogenesis (Lacey, DL et al. (1998) Cell 93, 165-176; Theill, LE et al. (2002) Annu Rev Immunol 20, 795-823; Yasuda, H. et al. (1998) Proc Natl Acad Sci USA 95, 3597- 3602; Kong, YY et al., Nature 397, 315-323 (1999); Suda, T. et al., Endocrine Rev. 20, 345-357 (1999)). RANKL expressed in osteoblasts causes signal transduction essential for osteoclast differentiation, whereas M-CSF secreted from osteoblasts mainly activates the anti-apoptotic factor Bcl-2. Produces a signal for progenitor cell survival (Lagasse, E., and Weissman, IL (1997) Cell 89, 1021-1031).

  Mice lacking RANKL, or op / op mice lacking a functional M-CSF gene, are responsible for bone sclerosis, one of the osteosclerotic diseases caused by impaired activation or differentiation of osteoclasts It causes marble disease (osteopetrosis) (Kong, YY et al. (1999b) Nature 397, 315-323; Yoshida, H. et al. (1990) Nature 345, 442-444). Osteopetrotic mice lacking osteoclasts due to impaired osteoclast development have significantly contributed to the identification of transcription factors essential for osteoclastogenesis. Targeting destruction of PU.1 results in complete loss of osteoclasts and macrophages, indicating that PU.1 is important for differentiation of early myeloid cells, including osteoclast precursors (Tondravi, MM et al. (1997) Nature 386, 81-84). In contrast, mice that lack c-Fos, a component of the activator protein (AP) -1 complex, or both p50 / 52, a component of NF (nuclear factor) -κB In mice, which develop severe osteopetrosis but do not impair macrophage development, these transcription factors play a role in specific functions during the monocyte / macrophage progenitor to mature osteoclast development phase (Franzoso, G. et al. (1997) Genes Dev 11, 3482-3496; Grigoriadis, AE et al. (1994) Science 266, 443-448; Johnson, RS et al. 1992) Cell 71, 577-586; Wang, ZQ et al. (1992) Nature 360, 741-745). MITF, which encodes the microphthalmia gene, is also essential for osteoclastogenesis (Moore, KJ (1995) Trends Genet 11, 442-448), and recent studies show that MITF is activated by M-CSF (Weilbaecher, KN et al. (2001) Mol Cell 8, 749-758) and have been suggested to be important genes for Bcl-2 induction and survival of this cell line (McGill, GG et al. (2002) Cell 109, 707-718).

Much work has been done to date on how RANKL transmits signals into cells. When RANKL binds to the receptor RANK, TRAF6 (tumor necrosis factor receptor-associated factor 6) and other TRAF family proteins are recruited, which activate the NF-κB and JNK pathways (Darnay, BG et al. 1998) J Biol Chem 273, 20551-20555; Galibert, L. et al. (1998) J Biol Chem 273, 34120-34127; Wong, BR et al. (1998) J Biol Chem 273, 28355-28359). The TRAF6 pathway has an essential role in osteoclast differentiation, as demonstrated by gene disruption studies and subsequent functional analysis of each TRAF6 domain during this process (Kobayashi, N. et al. (2001) Embo J 20, 1271-1280; Lomaga, MA et al. (1999) Genes Dev 13, 1015-1024; Naito, A. et al. (1999) Genes Cells 4, 353-362). However, introduction of a TRAF6 mutant exclusively rescuing NF-κB activity into TRAF6 ^{− / −} cells did not rescue osteoclastogenesis (Kobayashi, N. et al. (2001) Embo J 20 , 1271-1280). These results suggest that gene targeting studies reveal that NF-κB has an essential role, but that TRAF6 function is not mediated solely by NF-κB activity. I have.

RANKL also induces c-Fos expression, but c-Fos has been genetically shown to be a transcription factor essential for osteoclastogenesis (Matsuo, K. et al. (2000). ) Nat Genet 24, 184-187; Wagner, EF, and Karsenty, G. (2001) Curr Opin Genet Dev 11, 527-532). Among the target genes for c-Fos, Fra-1 has been shown to rescue osteoporosis in mice lacking c-Fos (Matsuo, K. et al. (2000) Nat Genet 24, 184-187), conditional targeting of this gene suggested that Fra-1 was not essential for osteoclastogenesis (Karsenty, G., and Wagner, EF (2002) Dev Cell 2, 389-406). Wagner, EF, and Karsenty, G. (2001) Curr Opin Genet Dev 11, 527-532). RANKL also activates other molecules such as ERK, p38, and Akt, but it is not genetically established whether these molecules have important functions in osteoclastogenesis (Matsumoto, M. et al. (2000) J Biol Chem 275, 31155-31161; Theill, LE et al. (2002) Annu Rev Immunol 20, 795-823; Wei, S. et al. (2002) J Biol Chem 277, 6622- 6630; Wong, BR et al. (1999) Mol Cell 4, 1041-1049). Thus, although it is well established that RANKL and downstream signaling pathways such as the TRAF6 and c-Fos pathways are essential for osteoclastogenesis, they are downstream of the TRAF6 and c-Fos pathways. The important genes that act are not known.
Karsenty, G., and Wagner, EF (2002) Dev Cell 2, 389-406 Manolagas, SC (2000) Endocr Rev 21, 115-137 Teitelbaum, SL (2000) Science 289, 1504-1508 Rodan, GA, and Martin, TJ (2000) Science 289, 1508-1514 Takayanagi, H. et al. (2000a) Arthritis Rheum 43, 259-269 Takayanagi, H. et al., J. Clin. Invest. 104, 137-46 (1999) Kong, YY et al., Nature 402, 304-9 (1999) Pettit, AR et al., Am J Pathol 159, 1689-99 (2001) Chambers, TJ (2000) J Pathol 192, 4-13 Suda, T. et al. (1999) Endocrine Rev 20, 345-357 Lacey, DL et al. (1998) Cell 93, 165-176 Theill, LE et al. (2002) Annu Rev Immunol 20, 795-823 Yasuda, H. et al. (1998) Proc Natl Acad Sci USA 95, 3597-3602 Kong, YY et al. (1999b) Nature 397, 315-323 Lagasse, E., and Weissman, IL (1997) Cell 89, 1021-1031 Yoshida, H. et al. (1990) Nature 345, 442-444 Tondravi, MM et al. (1997) Nature 386, 81-84 Franzoso, G. et al. (1997) Genes Dev 11, 3482-3496 Grigoriadis, AE et al. (1994) Science 266, 443-448 Johnson, RS et al. (1992) Cell 71, 577-586 Wang, ZQ et al. (1992) Nature 360, 741-745 Moore, KJ (1995) Trends Genet 11, 442-448 Weilbaecher, KN et al. (2001) Mol Cell 8, 749-758 McGill, GG et al. (2002) Cell 109, 707-718 Darnay, BG et al. (1998) J Biol Chem 273, 20551-20555 Galibert, L. et al. (1998) J Biol Chem 273, 34120-34127 Wong, BR et al. (1998) J Biol Chem 273, 28355-28359 Kobayashi, N. et al. (2001) Embo J 20, 1271-1280 Lomaga, MA et al. (1999) Genes Dev 13, 1015-1024 Naito, A. et al. (1999) Genes Cells 4, 353-362 Matsuo, K. et al. (2000) Nat Genet 24, 184-187 Wagner, EF, and Karsenty, G. (2001) Curr Opin Genet Dev 11, 527-532 Matsumoto, M. et al. (2000) J Biol Chem 275, 31155-31161 Wei, S. et al. (2002) J Biol Chem 277, 6622-6630 Wong, BR et al. (1999) Mol Cell 4, 1041-1049

An object of the present invention is to provide a method for controlling osteoclast formation. Specifically, the present invention provides a method for controlling osteoclast formation by promoting or inhibiting Ca ²⁺ signaling involved in osteoclast formation. The present invention further provides a method for screening for a compound that controls osteoclast formation, using Ca ²⁺ signaling as an index. Further, the present invention provides a system for forming osteoclasts in vitro from ES cells. According to the present invention, it is possible to screen for a new drug that controls bone resorption and bone destruction.

  Although the TRAF6 and c-Fos pathways are activated in RANKL-induced osteoclastogenesis, these pathways are also activated by various other cytokines such as IL-1 (Cao, Z. et al. (1996) Nature 383, 443-446; Ninomiya-Tsuji, J. et al. (1999) Nature 398, 252-256). This leads to the interesting possibility that RANKL signaling specifically induces osteoclast differentiation by activating unknown pathway (s). To identify the specific pathway (s) responsible for RANKL-dependent osteoclast differentiation, we used a genome-wide screening approach to specifically induce RANKL to induce IL- Genes that were not induced by 1 were identified. As a result, the present inventors have found that among the transcription factors, NFATc1, a member of the NFAT (nuclear factor of activated T cells) family, is the transcription factor that is most highly induced after RANKL stimulation, and that both the TRAF6 pathway and the Fos pathway are involved. Downstream of the pathway, it was found to be a molecular switch that determines osteoclast differentiation.

The NFAT transcription factor is involved in a wide range of cell development in addition to lymphocyte activation, and is well known to be regulated by calcineurin, a phosphatase regulated by Ca2 ⁺ (Crabtree, GR, and Olson, EN (2002) Cell 109 Suppl, S67-79; Rao, A. et al. (1997) Annu Rev Immunol 15, 707-747). In the present invention, it was found that RANKL strongly induces and activates NFATc1, and NFATc1 translocates to the nucleus and cooperates with c-Fos to directly regulate osteoclast genes including TRAP and calcitonin receptor. . The calcineurin inhibitors FK506 and cyclosporin A strongly inhibited RANKL-induced osteoclastogenesis in a dose-dependent manner. Ca ²⁺ chelators also inhibited NFATc1 expression elevation and inhibited osteoclast formation. Thus, we have shown that activation of NFATc1 by RANKL is also via the Ca ²⁺ -calcineurin pathway, which is the third pathway of RANKL signaling. Overexpression of NFATc1 by a retrovirus induces osteoclastogenesis without stimulation by RANKL, and embryonic stem (ES) cells lacking NFATc1 cannot differentiate into osteoclasts. These findings indicate that NFATc1 is a transcription factor essential for the development of osteoclasts and an integrator of RANKL-induced cell signaling. We further suggest that NFATc1 acts in conjunction with c-Fos to act on osteoclast gene promoters, including tartrate-resistant acid phosphatase (TRAP) and calcitonin receptor promoters. Revealed. Taken together, these suggest that NFATc1 plays an integral role in the RANKL-induced transcription machinery in terminal differentiation of osteoclasts.

  By the way, leflunomide is known as an isoxazole-based disease-modifying antirheumatic drug (DMARD), and acts on dihydroorotate dehydrogenase to inhibit de novo pyrimidine biosynthesis. Leflunomide has a protective effect on bone destruction in animal models and humans (Prakash, A. & Jarvis, B., Drugs 58, 1137-64 (1999); Sharp, JT, Arthritis Rheum 43, 495-505 (2000); Cohen, S. et al., Arthritis Rheum 44, 1984-92 (2001)), recently introduced into the treatment of rheumatoid arthritis (Kremer, JM, Ann Intern Med 134, 695-706 ( 2001); Prakash, A. & Jarvis, B., Drugs 58, 1137-64 (1999)). Leflunomide's anti-rheumatic effect is mainly exerted through suppression of T cell proliferation or activation (Breedveld, FC & Dayer, JM, Ann Rheum Dis 59, 841-9 (2000)), and immunosuppression in allografts It has also been incorporated as a drug (Williams, JW et al., Transplantation 73, 358-66 (2002)).

  Leflunomide (Prakash, A. & Jarvis, B., Drugs 58, 1137-64 (1999); Breedveld, FC & Dayer, JM, Ann Rheum Dis 59, 841-9 (2000)) and other clinical antirheumatic drugs Despite the above beneficial effects, little information is available on the molecular basis of the protective effects of these drugs on bone destruction. For example, it was not clear whether the osteoprotective effects of these drugs were indirect due to immunosuppressive effects or directly by inhibiting bone resorbable osteoclasts, which are effector cells of bone destruction. However, when the present inventors examined the effect of leflunomide using the osteoclast differentiation system, it was found that leflunomide had an effect of strongly directly inhibiting RANKL-mediated osteoclast differentiation. Leflunomide blocked RANKL-induced calcium signaling in osteoclast precursor cells in vitro and strongly inhibited the induction of NFATc1, an essential regulator of the transcriptional program of osteoclast differentiation. This inhibition was rescued by ectopic expression of NFATc1, indicating that NFATc1 is indeed an important target for leflunomide. We have also presented in vivo evidence that this effect is important for leflunomide as an osteoprotective agent. That is, leflunomide exhibited an inhibitory effect on inflammatory bone destruction induced by endotoxin in Rag-2 gene-deficient mice without T cell activation. The findings presented herein demonstrate that this anti-rheumatic drug has a direct inhibitory effect on osteoclastogenesis, and that an anti-rheumatic drug that exerts a direct inhibitory effect on RANKL-stimulated osteoclastogenesis is expressed in bone. Provide new strategies for effective use to improve related diseases.

The present invention relates to a method for controlling osteoclastogenesis by regulating Ca ²⁺ signaling, a method for screening a compound that regulates osteoclastogenesis, a pharmaceutical composition for suppressing osteoclastogenesis, and the like. In
(1) a method for promoting or suppressing osteoclast formation, which comprises the step of promoting or suppressing Ca ²⁺ signaling in osteoclast precursor cells, respectively;
(2) The method according to (1), wherein the step of promoting or suppressing Ca ²⁺ signal transduction in osteoclast precursor cells is a step of promoting or suppressing calcineurin expression or activity in the osteoclast precursor cells. ,
(3) The method according to (1), wherein the step of promoting or suppressing Ca ²⁺ signaling in osteoclast precursor cells is a step of promoting or suppressing expression or activity of NFATc1 in the osteoclast precursor cells. ,
(4) The method according to (1), further comprising a step of promoting or suppressing the expression or activity of one or more components of AP-1 in the osteoclast precursor cells.
(5) The method according to (3), wherein the activity of NFATc1 is an interaction with NFATc1 and an AP-1 component.
(6) the method according to (3), wherein the activity of NFATc1 is an interaction with NFATc1 and DNA;
(7) a method for suppressing osteoclast formation, which comprises the step of contacting an osteoclast precursor cell with a calcineurin inhibitor and / or leflunomide;
(8) A method for screening a compound that suppresses osteoclast formation,
(A) detecting Ca ²⁺ signaling in osteoclast precursor cells in the presence of a sample containing a test compound; and (b) selecting a compound that suppresses the Ca ²⁺ signaling. Method,
(9) Step (a) is a step of detecting calcineurin expression or activity in osteoclast precursor cells in the presence of a sample containing a test compound, and step (b) is expression level or activity of the calcineurin The method according to (8), which is a step of selecting a compound that suppresses
(10) Step (a) is a step of detecting the expression or activity of NFATc1 in osteoclast precursor cells in the presence of a sample containing a test compound, and Step (b) is the expression level or activity of the NFATc1 The method according to (8), which is a step of selecting a compound that suppresses
(11) The method according to (10), wherein the activity of NFATc1 is an interaction with NFATc1 and an AP-1 component.
(12) The method according to (10), wherein the activity of NFATc1 is an interaction with NFATc1 and DNA.
(13) A method for screening for a compound that suppresses osteoclast formation,
(A) in the presence of a sample containing the test compound, a step of detecting osteoclast formation in the osteoclast precursor cells that exogenously express NFATc1, and (b) a compound that suppresses the osteoclast formation. Selecting, a method comprising:
(14) A pharmaceutical composition for treating osteopenic disease, comprising at least one compound that inhibits Ca ²⁺ signaling, and one or more pharmaceutically acceptable carriers. ,
(15) The composition according to (14), wherein the compound is a compound that suppresses the expression or activity of calcineurin.
(16) The composition according to (14), wherein the compound is a compound that suppresses the expression or activity of NFATc1.
(17) The composition according to (16), wherein the compound is a compound that inhibits interaction with the NFATc1 and AP-1 components.
(18) The composition according to (16), wherein the compound is a compound that inhibits interaction with NFATc1 and DNA.
(19) The composition according to (14), wherein the osteopenic disease is a disease selected from the group consisting of osteoporosis, autoimmune arthritis, Paget's disease, and bone cancer.
(20) A method for producing osteoclasts,
(A) culturing embryonic stem cells to form a three-dimensional cell mass,
(B) culturing the cells obtained in step (a) in the presence of M-CSF,
(C) culturing the cells obtained in step (b) in the presence of M-CSF and RANKL,
A method comprising:

Although RANKL signaling is essential for osteoclast differentiation, it has not been previously known how it specifically determines the fate of this cell lineage. Gene disruption studies have identified molecules essential for osteoclastogenesis, such as TRAF6, NF-κB, and c-Fos, all of which have the ability of RANKL to induce osteoclast-specific genes. I couldn't explain. The present invention demonstrates that another essential transcription factor, NFATc1, cooperates with c-Fos to regulate osteoclast genes, and that cell-type-specific transcriptional effects are achieved through coordinated coordination of transcription factors. Clarified how this will be achieved. Since NFATc1-deficient mice are embryonic lethal, we used NFATc1-deficient ES cells to demonstrate the essential role of NFATc1 in osteoclast differentiation. In NFATc1 ^{− / −} cells, RANKL-induced osteoclast differentiation is blocked at the stage of monocyte / macrophage progenitors, and TRAP is not expressed, but c-Fms and RANK are normally expressed. Maturation, ie, the formation of TRAP ⁺ multinucleated cells. The inhibitory effects of FK506 and CyA on osteoclastogenesis and NFATc1 activation and induction revealed that calcineurin dephosphorylation activity is important for NFATc1 activation in osteoclastogenesis. Inhibition of osteoclastogenesis by Ca2 ⁺ chelators suggests that Ca2 ⁺ mobilization is important for RANKL-induced osteoclastogenesis. Furthermore, since NFATc1 induction and Ca ²⁺ oscillation were observed in RANKL-stimulated cells and not observed in IL-1-stimulated cells, it was clarified why RANKL induced osteoclasts and IL-1 did not induce Ca ^2+. It is suggested that the NFATc1 pathway activated by signal transduction explains. Thus, the NFATc1 pathway regulated by calcineurin is the third essential pathway of RANKL signaling to induce osteoclast differentiation. The addition of FK506 simultaneously with RANKL abolishes NFATc1 induction, suggesting that calcineurin is important for autonomous amplification of NFATc1 observed early in osteoclast differentiation (FIG. 8). Furthermore, when FK506 was added to BMM 48 hours after RANKL stimulation, osteoclastogenesis was completely inhibited despite the presence of NFATc1 (data not shown), indicating that calcineurin-dependent NFATc1 activation was It is suggested that it is essential even later in osteoclast differentiation.

  There have been many reports that the RANKL signal activates transcription factors such as NF-κB and AP-1. Meanwhile, it has been reported that the TRAP gene promoter is regulated by transcription factors such as PU.1, MITF, and IRF-4 (Matsumono et al., J. Biol. Chem. (2001) 276, 33086-92; Luchin et al., J. Bone Miner. Res. (2000) 15, 451-60). However, it is not known how RANKL transmits osteoclastogenic signals to transcription factors directly involved in osteoclast gene transcription, and it activates RANKL signaling and terminal differentiation of osteoclasts. The link between the transcription mechanisms that cause them was not fully understood. In the present invention, it has been shown that NFATc1 cooperates with AP-1 to activate regulatory lines of osteoclast genes such as TRAP and calcitonin receptor, and NFATc1: AP-1 complex is expressed in mature osteoclasts. It has been shown to directly regulate the expressed genes. Both promoters were activated by NFATc1 alone, which was amplified by the simultaneous addition of c-Fos. In addition, NFATc1 mutants that cannot cooperate with AP-1 have revealed that coordination of NFATc1 and AP-1 is important for maximal activation of these promoters. The FK506-treated BMM and c-Fos-deficient BMM both lost NFATc1 induction, indicating that the co-operation of NFATc1 and c-Fos is also important for autonomous amplification of NFATc1. Consistent with these results, a synergistic effect of NFATc1 and c-Fos carrying retroviral vectors on osteoclastogenesis was observed (data not shown). However, it was suggested that it is important for cell differentiation. Cooperation with NFATc1 may provide new insights into how the ubiquitous transcription factor AP-1 regulates cell type-specific transcription.

TRAF6 ^-/- BMM and c-Fos ^-/- BMM both lack a marked increase in RANAT-induced NFATc1 protein, suggesting that this induction is dependent on both pathways (Figure 15A). In contrast, the expression of TRAF6 and c-Fos in NFATc1 ^-/- ES cells stimulated with RANKL was observed normally (data not shown). Thus, the NFATc1 pathway acts downstream of both of these pathways. Taken together, the current knowledge of the RANKL signaling network that induces and activates NFATc1 can be summarized as follows. When RANKL binds to its receptor, RANK, it activates three essential pathways, including the TRAF6, c-Fos / AP-1 and Ca-calcineurin pathways. In a calcineurin-dependent manner, NFATc1 cooperates with AP-1 to increase NFATc1 mRNA. TRAF6 is also involved in this autonomous amplification of NFATc1 mRNA, probably through activation of NF-κB. TRAF6 also contributes to regulating NFATc1 protein levels through unknown mechanisms, such as regulating protein stabilization. Thus, significant up-regulation of NFATc1 protein levels by RANKL is achieved downstream of both the TRAF6 and c-Fos / AP-1 pathways. The up-regulated NFATc1 protein translocates to the nucleus in a calcineurin-dependent manner and cooperates with c-Fos / AP-1 to form a transcription mechanism essential for the induction of osteoclast genes. As shown in the Examples, NFATc1 mRNA induction was completely lost in c-Fos ^{− / −} cells but was still partially observed in TRAF6 ^{− / −} cells, indicating that the TRAF6 and c-Fos pathways Regulates downstream target genes differently.

  Inhibition of osteoclast differentiation or activity is important for the treatment of bone disease and there is increasing evidence of the effectiveness of that strategy (Kong, YY et al. (1999a) Nature 402, 304-309; Takayanagi, H et al. (1999) J Clin Invest 104, 137-146). Immunosuppressants FK506 and CyA have been widely used in human transplant surgery. As shown in this study, these calcineurin inhibitors have a potent inhibitory effect on osteoclast formation in vitro, suggesting a beneficial effect on bone mass. Contrary to expectations, osteoporosis is one of the side effects observed in patients treated with these drugs (Cvetkovic, M. et al. (1994) Transplantation 57, 1231-1237; Leidig-Bruckner, G. et al. (2001) Lancet 357, 342-347). This can be caused by the effects of these drugs on other cell types, such as osteoblasts. Therefore, when using these calcineurin inhibitors for the treatment of bone diseases, it is preferable to use an administration method (local administration or the like) that acts on osteoclasts as selectively as possible. Alternatively, a method of binding to a bisphosphonate-based compound having a tendency to be distributed in the vicinity of osteoclasts in a living body may be considered. Alternatively, in view of the specific molecular mechanism of osteoclast differentiation shown in the present invention, targeting NFATc1 activity by suppressing the interaction between NFATc1 and AP-1 is an osteoclast-specific drug. Is a good strategy to develop.

Leflunomide has an effective therapeutic effect on animal models of autoimmune arthritis induced by adjuvants, collagen, or proteoglycans (Prakash, A. & Jarvis, B., Drugs 58, 1137-64 (1999)) . In addition, studies of randomly controlled human rheumatoid arthritis have shown that leflunomide is one of the antirheumatic drugs that slows the progression of bone destruction (Sharp, JT, Arthritis Rheum 43, 495- 505 (2000); Cohen, S. et al., Arthritis Rheum 44, 1984-92 (2001)). In the present invention, it has been clarified that the effect of leflunomide directly targets osteoclast formation beyond the suppressive effect on T cells. Leflunomide also acted directly on monocyte / macrophage lineage osteoclast precursor cells and inhibited RANKL signaling by interfering with the Ca ²⁺ -activated NFATc1 pathway. We have also demonstrated that NFATc1 expression is indeed enhanced in inflammatory tissues adjacent to the site of bone destruction in animal models (FIG. 24). That is, leflunomide is expected to be applied to various bone destructive diseases and bone metabolic diseases in which the expression or activity of NFATc1 is enhanced. In this regard, it is interesting to note that selective and very strong expression of NFATc1 was observed in multinucleated osteoclasts at the bone / synovial interface in rheumatoid arthritis patients (FIG. 26). Taken together, suppression of the NFATc1 pathway may be a valuable strategy for bone destruction, such as arthritis, characterized by increased osteoclastogenesis. Inhibition of the NFATc1 pathway is also beneficial for suppressing steroid-induced osteoporosis, often found in RA patients. In addition, since it is considered that the increase in the expression of NFATc1 is correlated with the severity of the bone destruction disease, it is considered that the severity of the disease can be examined based on the expression level of NFATc1. In addition, as described above, the present inventors have shown that a calcineurin inhibitor such as FK506 has a strong inhibitory effect on osteoclast formation, and furthermore that leflunomide and FK506 are synergistic on osteoclast formation. It was found to show an inhibitory effect. Thus, combining multiple anti-osteoclast-forming agents is a promising approach in the treatment of bone destruction such as rheumatoid arthritis, on the one hand, increasing the efficacy of the treatment and, on the other hand, reducing the potential for side effects. right.

The present invention provides a method for controlling osteoclast formation, which targets a Ca ²⁺ signaling molecule, particularly a calcineurin signaling molecule including NFATc1. In addition, a novel screening method for a compound that suppresses osteoclast formation, which targets a Ca ²⁺ signal molecule, particularly a calcineurin signaling molecule including NFATc1, is provided. The present invention allows for the development of agents that selectively inhibit osteoclast formation. Such a drug is useful as a drug with few side effects that effectively suppresses bone destruction by osteoclasts. Since the enhancement of osteoclast formation is a cause of various osteopenic diseases including rheumatoid arthritis, the drug obtained by the present invention can be applied to the treatment of these bone diseases.

The present invention relates to a method for promoting or suppressing osteoclast formation, respectively, including a step of promoting or suppressing Ca ²⁺ signaling in osteoclast precursor cells. The present inventors have found that Ca ²⁺ signaling is an essential signal for osteoclast formation along with the TRAF6 and c-Fos pathways. By promoting Ca ²⁺ signaling in osteoclast precursor cells, differentiation of the cells into osteoclasts can be promoted, and conversely, by blocking Ca ²⁺ signaling, the destruction of the cells can be promoted. Differentiation into bone cells can be inhibited.

As used herein, an osteoclast is a multinucleated meganucleate formed by fusing mononuclear osteoclast precursor cells differentiated from macrophage / monocyte-derived cells (CFU-M) originating from hematopoietic stem cells. Cells (Udagawa, N. et al., Proc. Natl. Acad. Sci. USA 87: 7260, 1990). Generally, mature osteoclasts express tartrate-resistant acid phosphatase (TRAP), calcitonin receptor, vitronectin receptor, and M-CSF receptor (c-Fms), and exhibit bone resorption ability. Osteoclast formation can be determined based on at least one of these indicators. Preferably, the cells are TRAP-positive and have bone resorption activity. In an in vitro culture system, bone marrow monocyte / macrophage precursor cells (BMMs) are destroyed by the stimulation of RANKL in the presence of macrophage-colony stimulating factor (M-CSF). Differentiates into osteocytes.
Further, in the present invention, "osteoclast precursor cells" refers to macrophage monocyte cells capable of differentiating into osteoclasts, and more specifically, promote proliferation in response to M-CSF. Say cells to be. The osteoclast precursor cells may be cells contained in bone marrow or spleen, or may be cell lines. Osteoclast precursor cells can also be differentiated in vitro from hematopoietic stem cells (Example 8).

  Osteoclast formation can be observed, for example, according to the literature (Yasuda, H. et al., Proc. Natl. Acad. Sci. USA 95, 3597-602 (1998)). Specifically, for example, in the case of an individual, a bone slice can be prepared and bone resorption can be observed. In addition, osteoclastogenesis can be identified both in vivo and in vitro by known methods such as identifying multinucleated giant cells by microscopic observation, and by detecting TRAP staining and calcitonin receptor or vitronectin receptor expression. Can be detected. In the case of bone resorption, for example, the bone resorption activity on an ivory section can be detected by a known method.

Ca2 ⁺ signaling refers to signaling activated by Ca2 ⁺ release from the endoplasmic reticulum and subsequent Ca2 ⁺ entry into cells through the plasma membrane (Crabtree, GR, and Olson, EN (2002) Cell 109 Suppl, S67-79). Calcineurin is a phosphatase that mediates Ca ²⁺ signaling and regulates the induction and activation of transcription factors belonging to the NFAT family (Crabtree, GR, and Olson, EN (2002) Cell 109 Suppl, S67-79 Rao, A. et al. (1997) Annu Rev Immunol 15, 707-747). Among them, NFATc1 is one of the signal molecules involved in Ca ²⁺ signaling, activated by dephosphorylation by calcineurin (Cn), translocated from the cytoplasm to the nucleus, and found in the NFAT binding sequence of chromosomal DNA. Binds and regulates transcription of downstream genes (Graef IA, Chen F, Crabtree GR .: NFAT signaling in vertebrate development. Curr Opin Genet Dev 2001, 11: 505-12). NFATc1 interacts with AP-1 to form a NFATc1 / AP-1 / DNA complex. Examples of genes whose transcription is controlled by the binding of NFATc1 include the NFATc1 gene itself and genes such as TRAP, calcitonin receptor, cathepsin K, carbonic anhydrase II (CAII), and matrix metalloprotease (MMP) -9.

In the present invention, promoting Ca ²⁺ signal means activating Ca ²⁺ signaling, that is, increasing or activating the expression of a signal molecule, or increasing the expression of a target gene for Ca ²⁺ signaling. And through which osteoclast formation is promoted. Also to suppress the Ca ²⁺ signaling it may be to inhibit one or more of locations of Ca ²⁺ signaling. Specifically, for example, suppression of Ca ²⁺ signaling includes Ca ²⁺ mobilization, calcineurin activity or expression, NFATc1 expression or dephosphorylation, and the like. In the present invention, promotion or suppression of Ca ²⁺ signaling includes, more specifically, promotion or suppression of calcineurin expression or activity, respectively.

Calcineurin is a serine-threonine phosphatase regulated by calcium and calmodulin and plays an important role in the control of nerves, muscles, and the immune system (Shibasaki, F. et al., J. Biochem. 131, 1-15 (2002)). Calcineurin A, the catalytic subunit of calcineurin, contains a regulatory domain (Klee, CB et al. (1979) Proc. Natl. Acad. Sci. USA 76, 6270-6273), α, β and γ Are known and are highly conserved among mammals (Klee, CB et al. (1988) Adv. Enzymol. Relat. Areas Mol. Biol. 61: 149-200; Cohen , PT et al. (1996) Adv. Pharmacol. 36: 67-89; Goto, S. et al. (1986) Acta Neuropathol. (Berl) 72 (2): 150-6; Kuno, T. et al. (1992) J. Neurochem. 58 (5): 1643-51). Calcineurin B, a Ca2 ⁺ binding subunit, is a protein with an EF-hand (Aitken, A. et al. (1984) Eur.J. Biochem. 139 (3): 663-71), B1 and B2. Isoforms are known. In the present invention, calcineurin is a heterodimer in which any subunit of calcineurin A and any subunit of B are bound.

  Promotion of calcineurin expression includes promotion of transcription and / or translation of endogenous calcineurin gene, promotion of calcineurin mRNA stabilization, suppression of calcineurin protein degradation, expression of exogenous calcineurin, and the like. Conversely, suppression of calcineurin expression includes suppression of transcription and / or translation of endogenous calcineurin gene, promotion of calcineurin mRNA instability, promotion of calcineurin protein degradation, exogenous antisense nucleic acid to calcineurin mRNA or Including expression inhibition by expressing a ribozyme that cleaves calcineurin mRNA and the like. Examples of promotion and suppression of calcineurin activity include promotion and suppression of calcineurin dephosphorylation activity. For example, as calcineurin inhibitors, FK506, cyclosporin A (CyA), and analogs thereof are well known. With these calcineurin inhibitors, calcineurin activity can be suppressed.

  Cyclosporin A typically has the chemical name cyclo [3-hydroxy-4-methyl-2- (methylamino) -6-octenoyl] -2-aminobutyryl-methylglycyl-methyl-leucyl-valyl-methyl-leucyl-alanyl- Alanyl-methyl-leucyl-methyl-leucyl-methyl-valyl represented by, for example, the compound described as S7481 / F-1 in US 4,117,118 and its pharmaceutically acceptable salts, isomers, or hydrates Etc. are included. A plurality of metabolites and derivatives are known for cyclosporin A, and among these, a desired compound that significantly inhibits the expression of NFATc1 can be used as cyclosporin A in the present invention. The general chemical structure of cyclosporin A is shown in formula (I).

式(I)

Formula (I)

  FK506 is also called tacrolimus and typically has the chemical name 17-allyl-1,14-dihydroxy-12- [2- (4-hydroxy-3-methoxycyclohexyl) -1-methylvinyl] -23,25-dimethoxy Macrolide represented by -13,19,21,27-tetramethyl-11,28-dioxa-4-azatricyclo [22.3.1.0 4,9] octakos-18-ene-2,3,10,16-tetron Compound. FK506 includes, for example, compounds described as the general formula (I) in EP 184,162 and pharmacologically acceptable salts, isomers or hydrates thereof. The general chemical structure of FK506 is shown in formula (II). In the present specification, FK506 includes FR-900506, FR-900520, FR-900523, FR-900525, and derivatives thereof isolated from microorganisms of the genus Streptomyces such as Sreptomyces tsukulbensis No.9993. Is included. Many related compounds are known as FK506 (EP184162, EP315973, EP323042, EP423714, EP427680, EP465426, EP474126, WO91 / 13889, WO91 / 19495, EP484936, EP532088, EP532089, WO93 / 5059), Among them, a desired compound that significantly inhibits the expression of NFATc1 can be used as FK506 in the present invention.

式(II)

Formula (II)

In the present invention, promotion or suppression of the Ca ²⁺ signal is further specified, for example, promotion or suppression of the expression or activity of NFATc1. NFATc1 is one of the transcription factors that dephosphorylates in the Ca-calcineurin-dependent manner, translocates to the nucleus, and activates transcription.It was discovered as a factor that regulates cytokine production in T cells. Plays an important role in differentiation (Crabtree, GR and Olson, EN (2002) Cell 109: S67-S79). The nucleotide sequence of human NFATc1 mRNA is shown in Accession number NM_006162, and the amino acid sequence of the protein is shown in NP_006153. The nucleotide sequences of mouse NFATc1 mRNA are shown in Accession numbers AF239169, AF087434 (isoform a), and AF049606 (isoform b), and the amino acid sequences of the NFATc1 protein encoded by them are shown in AAC36725, AAC36725, and AAC05505, respectively. .

  Promotion of NFATc1 expression includes promotion of transcription and / or translation of endogenous NFATc1 gene, promotion of NFATc1 mRNA stabilization, suppression of NFATc1 protein degradation, exogenous NFATc1 expression, and the like. Conversely, to suppress the expression of NFATc1, transcription of endogenous NFATc1 gene and / or suppression of translation, promotion of NFATc1 mRNA destabilization, promotion of NFATc1 protein degradation, exogenously antisense nucleic acid against NFATc1 mRNA or Including expression inhibition by expressing a ribozyme that cleaves NFATc1 mRNA and the like is included.

  For example, in order to exogenously express NFATc1, NFATc1 may be incorporated into a desired expression vector and introduced into osteoclast precursor cells. The vector that can be used is not limited as long as it expresses NFATc1, and may be in the form of a polynucleotide containing DNA and RNA itself, or a complex containing the polynucleotide. For example, examples of the virus vector include a retrovirus vector, an adenovirus vector, and an adeno-associated virus vector. Adenovirus can be concentrated at a high concentration and is highly safe because the transgene is not integrated into the host genome. In fact, it has been reported that the gene transfer efficiency of an adenovirus vector to synovial cells is high, and it can be used for gene transfer to joints. In addition, gene introduction into a cell or a living body using a non-viral vector can be performed by administration of naked DNA, a method using Sendai virus liposome, or the like. For in vivo administration, the vector can be administered locally, for example, to the osteochondral site.

Promotion of NFATc1 activation includes promotion of dephosphorylation of NFATc1. Since NFATc1 is dephosphorylated and activated by calcineurin as described above, the promotion of NFATc1 activation can be achieved, for example, by increasing the expression of calcineurin or increasing the activity of calcineurin. In addition, activation of a Ca ²⁺ signal such as Ca ²⁺ mobilization may be used. Further, it may be suppression of phosphorylation of NAFTc1. As the NFATc1 phosphorylating enzyme, for example, GSK3 is known. By inhibiting these kinases, activation of NFATc1 can be promoted. Conversely, suppression of NFATc1 activation includes suppression of NFATc1 dephosphorylation, such as suppression of calcineurin expression or inhibition of calcineurin activity. It may also be a suppression or blocking of a Ca ²⁺ signal. It may also be the promotion of phosphorylation of NAFTc1. As described above, as a calcineurin inhibitor, FK506, cyclosporin A (CyA), analogs thereof, and the like can be used. Activation of NFATc1 can be suppressed by these calcineurin inhibitors. The present invention relates to a method of inhibiting osteoclast formation by FK506, CyA, or other calcineurin inhibitors. For example, bone resorption suppression can be suppressed by drug delivery of a calcineurin inhibitor to osteoclast precursor cells.

  The activity of NFATc1 includes an interaction with AP-1 and an interaction with DNA. NFATc1 forms a complex with AP-1 and binds to DNA to regulate target gene transcription. Mutant NFATc1 in which the interaction with AP-1 is inhibited cannot induce the expression of the target gene of NFATc1, and does not induce osteoclast formation. Therefore, by promoting or inhibiting the interaction between NFATc1 and the AP-1 component or the interaction between NFATc1: AP-1 and DNA, osteoclast formation can be promoted or suppressed, respectively.

  The AP-1 component refers to a member of the Fos family or the Jun family constituting AP-1 (Jochum, W. et al. (2001) Oncogene 20 (19): 2401-12; Transcription factor ”, Experimental Medicine Separate Volume, Yodosha, pp204-205). Members of the Fos family include c-Fos, FosB, Fra1, and Fra2. Members of the Jun family include c-Jun, JunB, and JunD. Among them, c-Jun and c-Fos are examples of AP-1 members that strongly promote osteoclast formation in cooperation with NFATc1. A particularly important AP-1 component is c-Fos (Abate, C. et al., Science 7, 1157-1161, 1990; Deng, T. and Karin, M, Nature 371, 171-175, 1994; Ofir , R. et al., Nature 348, 80-82, 1990; Bakin, AV et al., Science 283, 387-390, 1999). The nucleotide sequences of c-fos, fosB, fra1, and fra2 mRNA are shown in Accession numbers X06769, X14897, X16707, and X16706, respectively, and the amino acid sequences of the respective proteins are shown in CAA29937, CAA33026, CAA34679, and CAA34678.

Molecules that activate the phospholipase C (PLC) gamma-calcium pathway, such as DAP12 and FcRgamma, and their associated receptors play an important role as upstream molecules that activate the calcium-NFAT pathway. DAP12 (human: Accession number NM_003332 (protein ID NP_003323), mouse: NM_011662 (protein ID NP_035792)) associates with triggering receptor expressed on myeloid cells (TREM) -2 and activates the PLC-calcium pathway (Snyder , MR et al., J. Exp.Med. 197, 437-449 (2003); Colonna, M., J. Clin. Invest. 111, 313-314 (2003); Lucas, M. et al., Eur. J. Immunol. 32, 2653-2663 (2002); Paloneva, J. et al., Am. J. Hum. Genet. 71, 656-662 (2002); Paloneva, J. et al., Nat. Genet. 25, 357-361 (2000); Pekkarinen, P. et al., Genomics 54, 307-315 (1998); Lanier, LL et al., Nature 391, 703-707 (1998); Pekkarinen, P. et. al., Am. J. Hum. Genet. 62 (2), 362-372 (1998); Kaifu, T. et al., J. Clin. Invest. 111, 323-332 (2003); Colonna, M. , J. Clin. Invest. 111, 313-314 (2003); Diefenbach, A. et al., Nat. Immunol. 3, 1142-1149 (2002); McVicar, DW et al., J. Immunol. 169, 1721-1728 (2002); Sjolin, H. et al., J. Exp.Med. 195, 825-834 (2002); Tomasello, E. et al., J. Biol. Chem. 273, 34115-34119 ( 1 998); Lanier, LL et al., Nature 391, 703-707 (1998)). FcRgamma (gamma subunit of FcR) (human: Accession number NM_004106 (protein ID NP_004097), mouse: NM_010185 (protein ID NP_034315)) includes paired immunoglobulin-like receptor (PIR), osteoclast-associated receptor (OSCAR), etc. And activates the PLC-calcium pathway (Wu, J. et al., Lupus 11, 42-45 (2002); Kuster, H. et al., J. Biol. Chem. 267, 12782- 12787 (1992); Le Coniat, M. et al., Immunogenetics 32, 183-186 (1990); Kuster, H. et al., J. Biol. Chem. 265, 6448-6452 (1990); Craig, AW and Greer, PA, Mol.Cell. Biol. 22, 6363-6374 (2002); Nagaoka, Y. et al., J. Invest.Dermatol. 119, 130-136 (2002); Strzelecka-Kiliszek, A. and Sobota, A., Folia Histochem. Cytobiol. 40, 131-132 (2002); Gray, CA and Lawrence, RA, Eur. J. Immunol. 32, 1114-1120 (2002); Ra, C. et al., J. Biol. Chem. 264, 15323-15327 (1989)). For example, if the activation of signal transduction via these molecules is inhibited by, for example, inhibiting the binding of DAP12 or FcRgamma to its associated receptor, it is expected that Ca ²⁺ signaling can be inhibited. Also, this signal can be suppressed by using an inhibitory antibody against a receptor associated with DAP12 or FcRgamma or a substance that suppresses binding to a ligand. In the present invention, the Ca2 ⁺ signal includes a molecule that activates the PLCgamma-calcium pathway such as DAP12 or FcRgamma, or a signal transmitted via its associated receptor. DAP12 and FcRgamma have a signaling domain called immunoreceptor tyrosine-based activation motif (ITAM) and have tyrosine phosphorylation activity. It is known that the ITAM signal activates PLCgamma via Syk or ZAP70. In the present invention, the Ca ²⁺ signal includes an ITAM signal and a signal via Syk or ZAP70 activated downstream thereof. This signal can be suppressed by a protein or a small molecule drug that binds to the ITAM motif, or a protein or a small molecule drug that suppresses the tyrosine kinase activity of ITAM, or a drug that suppresses the activity such as Syk or ZAP70. In addition, by screening for compounds that inhibit these signal transductions, it is possible to isolate new drugs that inhibit osteoclast formation.

  The present invention also relates to a method for inhibiting osteoclast formation, comprising the step of contacting leflunomide with osteoclast precursor cells. The present inventors have found that leflunomide efficiently inhibits osteoclast formation through suppression of NFATc1 expression. Thus, leflunomide is useful for selectively inhibiting osteoclast formation. Since leflunomide effectively inhibits osteoclast formation, leflunomide is useful not only for the treatment of autoimmune arthritis such as rheumatoid arthritis, but also for the treatment of other osteopenic diseases. As used herein, leflunomide includes the active metabolite of leflunomide. For example, leflunomide typically includes the compound HWA 486 represented by the following formula (III), and the chemical name is 5-methyl-N- [4- (trifluoromethyl) phenyl] -4-isoxazolecarboxamide. (HWA 486: RR Bartlett et al. In: ALLewis and DE Furst (Eds.), Nonsteroidal Anti-inflammatory Drugs, Mechanisms and Clinical Uses; CCA Kuechle et al., Transplant Proc. 1991, 23: 1083-6; T Zielinski et al., Agents Action 1993, 38: C80-2). Leflunomide also includes active metabolite A77 1726 of HWA 486 (formula (IV)), and pharmaceutically acceptable salts, isomers or hydrates of HWA 486 and A77 1726. Examples of leflunomide include a compound represented by the formula (I) in EP13,376 (JP-A-62-72614 or US Pat. No. 4,284,786) and pharmacologically acceptable salts thereof.

式(III)

Formula (III)

式(IV)

Formula (IV)

  Furthermore, the present inventors have found that a combination of leflunomide and a calcineurin inhibitor such as FK506 and CyA exerts a synergistic inhibitory effect on osteoclast formation. This result indicates that administration of both a calcineurin inhibitor and leflunomide can more effectively suppress osteoclast formation. The combination of a calcineurin inhibitor and leflunomide is extremely effective in treating autoimmune arthritis and other osteopenic diseases.

In screening for compounds that suppress osteoclast formation in the present invention, compounds that suppress Ca ²⁺ signaling are selected. When Ca ²⁺ signaling is suppressed by such a compound, NFATc1 expression induction or activity level is suppressed, and formation of osteoclasts can be suppressed. In the screening for a compound that suppresses osteoclast formation in the present invention, any stage of the Ca ²⁺ signaling pathway leading to NFATc1 can be used as an index. In particular, compounds that directly act on the suppression of NFATc1 expression or activity are preferable because they are considered to show high specificity for suppressing osteoclast formation. The details of the suppression of NFATc1 expression and the suppression of the activity are as described above. For example, the screening of the present invention includes the following screening.
(1) Ca ²⁺ influx or Ca ²⁺ mobilization is detected in the presence of a test compound, and a compound that inhibits it is selected. For example, Ca ²⁺ chelator is a candidate compound.
(2) The activity of calcineurin is detected in the presence of the test compound, and a compound that inhibits the activity is selected.
(3) Expression of NFATc1 is detected in the presence of a test compound, and a compound that suppresses this is selected. Here, the compound that suppresses the expression of NFATc1 is determined by using (a) inhibition of induction of mRNA of NAFTc1, (b) inhibition of synthesis of NFATc1 protein, and (c) inhibition of stability of NFATc1 protein (promotion of degradation) as indices. Can be screened.
(4) NFATc1 activity is detected in the presence of a test compound, and a compound that suppresses the activity is selected. Compounds that inhibit NFATc1 activity include (a) inhibition of NFATc1 dephosphorylation, (b) promotion of NFATc1 phosphorylation, (c) inhibition of NFATc1 nuclear translocation, and (d) binding of NFATc1 to DNA. Screening can be performed using inhibition, (e) inhibition of binding between NFATc1 and the AP-1 component (preferably, binding to c-Fos and / or c-Jun) and the like as indices. For example, the binding between NFATc1 and c-Fos can be detected by immunoprecipitation-Western blotting using NFATc1 antibody and c-Fos antibody. Further, for example, the binding between NFATc1 and DNA can be measured by an electrophoretic mobility shift assay (EMSA) using an oligonucleotide containing a NFATc1 binding sequence. Based on the information of the interaction site of each component in the NFAT / AP-1 / DNA complex, by designing a low-molecular compound that binds to this site, it is possible to effectively treat the NFATc1 and AP-1 components or NFATc1 and DNA It would be possible to screen for compounds that inhibit the interaction with. The essential amino acids of the interaction between NFATc1 and the AP-1 component are exemplified herein.

In addition, the NFATc1 activity can also be based on NFATc1-dependent transcription induction. For this purpose, a sequence containing the MFATc1 binding site in the promoter region of a gene such as TRAP, calcitonin receptor, cathepsin K, carbonic anhydrase II (CAII), or matrix metalloprotease (MMP) -9 can be suitably used. These genes are induced in terminal differentiation of osteoclasts and contain multiple NFATc1 and AP-1 binding sites (Anusaksathien, O. et al. (2001) J Biol Chem 276, 22663-22674; David, JP et al. (2001) J Cell Physiol 188, 89-97; Motyckova, G. et al. (2001) Proc Natl Acad Sci USA 98, 5798-5803; Reddy, SV et al. (1995) J Bone Miner Res 10, 601 -606). Also, the promoter region of the NFATc1 gene can be used. NFATc1 binding sequence in the promoters of these genes (GGAAAA and similar sequences (eg, GGAAAAN, where N is A, T, C, or G), Macian, F. et al. (2001) Oncogene 20, 2476-2489) DNA, preferably further AP-1 binding sequence (TGA [G / C] TCA and its similar sequence (eg TGNTCTA, N is A, T, C or G), Bioscience new term library "transcription factor" , Experimental Medicine Separate Volume, Yodosha, pp204-205) to bind the desired reporter gene downstream of the DNA, and detect the reporter gene expression via reporter activity in the presence of NFATc1 and the test compound, A compound that reduces this is selected. NFATc1 may be supplied to the screening system endogenously or exogenously. Specific examples of the promoter sequence include a TRAP promoter as used in Examples and a calcitonin promoter (P3 promoter). For example, using an osteoclastic progenitor cell exogenously expressing NFATc1, in the presence of a sample containing a test compound, NFATc1, TRAP, or the activity of a promoter such as a calcitonin receptor gene in the cell is detected. By selecting a compound that suppresses the promoter activity, a candidate compound that inhibits the transcription activity by NFATc1 can be isolated.
The above screening can be performed by those skilled in the art using standard techniques (for example, Edited by J. Sambrook et al., Molecular Cloning: A Laboratory Manual, 2001, 3rd ed., Cold). Spring Harbor Laboratory).

  Further, the present invention provides a step of detecting osteoclast formation in osteoclast precursor cells exogenously expressing NFATc1 in the presence of a sample containing a test compound, and selecting a compound that suppresses the osteoclast formation. A method of screening for a compound that inhibits osteoclast formation. Osteoclast precursor cells overexpressing NFATc1 form osteoclasts even in the absence of RANKL. By screening for a compound that inhibits this osteoclast formation, a compound that inhibits NFATc1-induced osteoclast formation can be isolated.

The screening of the present invention can be performed using, for example, an osteoclast differentiation induction system. That is, in a system in which osteoclast precursor cells are stimulated with RANKL to form osteoclasts, Ca ²⁺ signaling is measured in the presence of a test compound. The osteoclast precursor cells are not limited as long as they can differentiate into osteoclasts. For example, bone marrow macrophage cells are used. Specifically, BMM derived from bone marrow can be used, or RAW 264.7 cells and ES cells (described later) can be used. RAW 264.7 cells, which differentiate into osteoclasts in response to RANKL in an M-CSF-independent manner, do not require M-CSF in osteoclast formation, but are required to form osteoclasts from BMM and ES cells. Performs RANKL stimulation in the presence of M-CSF. The concentrations of M-CSF and RANKL can be appropriately adjusted. For example, it is preferable that M-CSF is about 10 ng / ml and RANKL is about 100 ng / ml. The sample containing the test compound is not particularly limited, and examples thereof include a library of synthetic low-molecular compounds, a purified protein, an expression product of a gene library, a library of synthetic peptides, a cell extract, and a cell culture supernatant. Can be

Ca ²⁺ signal transduction is measured in the presence of a sample containing a test compound, and a compound that promotes or suppresses the signal transduction is selected as compared to a control. Control conditions include the case where the test compound is not contained or the test compound is contained at a lower dose.Typically, in the absence of the test compound (for example, only the carrier used for adding the test compound is used). Addition). A compound that regulates osteoclast formation can be obtained by selecting a compound that promotes or suppresses the signal transmission, as compared to the case where no sample containing the test compound is present. In addition, when a condition containing a test compound at a lower dose is used as a control condition, the dose dependency of the compound becomes apparent. Control conditions may also include cases in the presence of other compounds that can regulate osteoclast formation. In this case, a compound that promotes or suppresses the Ca ²⁺ signaling is selected as compared with the case where the compound used as the control is present. In such a screening, a compound having a higher effect than a certain compound can be obtained. Detection of Ca ²⁺ signaling may be performed as described above.

The screening method of the present invention can be performed using, for example, an in vitro assay system for osteoclast formation as described in the Examples of the present specification. Specifically, for example, non-contact bone marrow cells (5 × 10 ⁵ cells per well of a 24-well plate) are cultured for 2 days in α-minimal essential medium (α-MEM) containing 10 ng / ml of M-CSF. A system for forming osteoclasts by culturing in the presence of 100 ng / ml of soluble RANKL and 10 ng / ml of M-CSF for another 3 days can be used. Here, a sample containing the test compound is added and cultured. Further, as an in vivo system, for example, a system using an animal model of endotoxin-induced bone resorption as used in the examples can be mentioned. Furthermore, an osteoclast-forming system from ES cells described below can be used. In a screening system using ES cells, cells deficient in a desired gene involved in osteoclast formation can be used. Use of the osteoclast-forming system from gene-deficient ES cells can provide useful information on which signal pathway of the osteoclast-forming signal the test compound is involved in.

As a result of this detection, if the addition of the sample containing the test compound significantly promotes or suppresses Ca ²⁺ signaling as compared to the control condition, the test sample used regulates osteoclast formation. Of the compound to be used. The compound isolated by this screening method is useful for controlling osteoclast formation and is also useful for developing preventive and therapeutic drugs for bone metabolic diseases.

The present invention also relates to a method for examining the effect of a compound on osteoclast formation, comprising: (a) detecting Ca ²⁺ signaling in osteoclast precursor cells in the presence of a sample containing a test compound; And (b) associating the inhibition of Ca ²⁺ signaling with the inhibition of osteoclastogenesis and the promotion of the Ca ²⁺ signaling as an enhancement of osteoclast formation. Ca ²⁺ signaling is preferably performed by detecting calcineurin expression or activity, or NFATc1 expression or activity. By such a test method, a compound that suppresses osteoclast formation can be identified.

  The compound obtained by the screening of the present invention can be used as an inhibitor of osteoclast formation. Conditions for administering the compound for clinical application can be determined using the osteoclastogenic system described herein. That is, by using the above-described test method, administration conditions including administration amount, administration interval, and administration route can be examined, and conditions for obtaining appropriate preventive or therapeutic effects can be determined. Such a compound becomes a medicament for prevention or treatment of osteopenic diseases. The compound can be combined with the desired pharmaceutically acceptable carrier into a composition. Carriers include, for example, sterile water, physiological saline, buffers, vegetable oils, emulsifiers, suspensions, salts, stabilizers, preservatives, surfactants, sustained release agents, other proteins (such as BSA), transfection reagents (Including lipofection reagents, liposomes, etc.). Further, usable carriers include glucose, lactose, gum arabic, gelatin, mannitol, starch paste, magnesium trisilicate, talc, corn starch, keratin, colloidal silica, potato starch, urea and the like. The dosage form for formulation is not limited, and may be, for example, a solution (injection), powder, microcapsule, tablet or the like. Osteopenic diseases include, for example, osteoporosis, autoimmune arthritis, Paget's disease, and bone cancer. Administration can be performed systemically or locally, but if side effects or effects are reduced by systemic administration, local administration is preferred. For example, it is considered that the present compound can be effectively treated by combining the compound with a sustained-release agent by drug delivery targeting a disease site exhibiting bone loss.

  Administration to patients depends on the nature of the active ingredient, e.g., transdermal, intranasal, transbronchial, intramuscular, intraperitoneal, intravenous, intraarticular, subcutaneous, intraspinal, intraventricular, or It can be performed orally, but not limited to. Administration can be systemic or local, but when side effects due to systemic administration are problematic, local administration to the lesion site is preferred. The dose and administration method vary depending on the tissue transferability of the active ingredient of the pharmaceutical composition, the purpose of treatment, the weight and age of the patient, symptoms, and the like, but can be appropriately selected by those skilled in the art. As shown in the examples, 0.1 μg / ml of FK506 or CyA dramatically inhibited osteoclast formation in an in vitro system, and the inhibitory effect increased in a dose-dependent manner at concentrations up to 10 μg / ml. In addition, leflunomide at 10 μg / ml dramatically inhibited osteoclast formation, and the inhibitory effect increased in a dose-dependent manner at concentrations up to 100 μg / ml. For example, even in vivo, the dosage of the drug can be determined so as to maintain such an effective concentration around the target osteoclast precursor cell.

  Pharmaceutical compositions of the invention generally contain from 0.1 to 90% of the therapeutic drug by weight of the total composition. For parenteral administration, the dose is 0.0001 mg to 1000 mg, preferably 0.001 mg to 300 mg, more preferably 0.01 to 100 mg per kg of body weight per day. However, depending on the disease state, body weight, and the individual response of the patient to the treatment, the type of composition to which the therapeutic drug is administered, and the mode of administration, the course of the disease or the interval between administrations, these rates of administration may be adjusted accordingly I do. Therefore, it is sometimes appropriate to use less than the above-cited minimum dose, while in other cases it will be necessary to exceed the upper limit in order to obtain a therapeutic effect. Administration can be carried out once to several times, and can be administered 1 to 5 times a day. Target individuals include, for example, humans and non-human mammals such as mice, rats, rabbits, monkeys, and other vertebrates. The application to non-human mammals is also useful as a model for developing preventive or therapeutic methods for human bone destructive disease. This allows the development of new therapeutic protocols to prevent bone destruction, for example in autoimmune arthritis.

  The present invention also relates to a kit for treating an osteopenic disease, the kit comprising a calcineurin inhibitor and leflunomide. By using a calcineurin inhibitor and leflunomide together, a synergistic inhibitory effect on osteoclast formation can be obtained. Calcineurin inhibitors include, in particular, cyclosporin A and FK506.

  The pharmaceutical composition of the present invention is useful for treating and preventing a desired osteopenic disease involving bone resorption by osteoclasts. Diseases of primary therapeutic interest include, among others, autoimmune arthritis, including rheumatoid arthritis (RA). Rheumatoid arthritis is a chronic, autoimmune, inflammatory disease in which the proliferating synovium actively invades osteochondral, resulting in multiple joint destruction. The pharmaceutical composition of the present invention is suitably used for preventing this joint bone destruction.

  Other osteopenic diseases to which the pharmaceutical composition of the present invention can be applied include periodontal disease. In addition, the pharmaceutical composition of the present invention provides various diseases caused by enhanced bone resorption by osteoclasts, including giant cell tumor, cancer bone metastasis, and pigmented villonodular synovitis (PVS). May also be applied. It is also expected to be applied to the treatment of osteoporosis and hypercalcemia of cancer. In addition, it can be applied to the prevention and treatment of bone loss associated with Paget's disease, hepatitis and AIDS, bone resorption and destruction associated with leukemia and multiple myeloma, and bone resorption (losing) around artificial joints (Rodan, GA & Martin, TJ, Science 289, 1508-1514 (2000); Kong, YY et al., Nature 402, 304-309 (1999); Takayanagi, H. et al., Arthritis Rheum. 43, 259-269 ( 2000); Honore, P. et al., Nat. Med. 6, 521-528 (2000)). The present invention allows new therapeutic interventions by controlling osteoclast formation and function.

The present invention also provides a method for forming osteoclasts from ES (embryonic stem) cells. This method
(A) culturing ES cells to form a three-dimensional cell mass,
(B) culturing the cells obtained in step (a) in the presence of M-CSF,
(C) culturing the cells obtained in step (b) in the presence of M-CSF and RANKL,
It is a method including.

The above three-dimensional cell mass may be an embryonic body more specifically. A method for forming an embryonic body from ES cells is known, and can be prepared, for example, by culturing ES cells in hanging drops so that the ES cells are not brought into direct contact with a solid support. The culture period is, for example, two days. In step (b), the concentration of M-CSF is, for example, about 100 ng / ml. Culture may be performed in a petri dish. The culture conditions are, for example, 37 ° C., 5% CO ₂ , and the culture period is, for example, 2 days. In step (c), for example, the cells obtained in step (b) are dispersed and cultured by trypsin treatment or the like. The concentrations of RANKL and M-CSF are, for example, about 100 ng / ml and about 10 ng / ml, respectively. After about 4 days in culture, TRAP-positive osteoclasts are formed.

  The above-described method of forming osteoclasts from ES cells is superior to the conventional method in that feeder cells are not required. The osteoclast culture obtained by the above method is suitable for clinical applications such as cell transplantation, since there is no contamination of heterologous cells such as feeder cells. The osteoclasts thus formed can be used not only in the screening of the present invention but also locally for the treatment of diseases associated with abnormal ossification such as osteoarthritis and ligament ossification. Can be transplanted.

  Various modifications of the invention described in this specification will be readily apparent to those skilled in the art, but those embodiments are included in the scope of the present invention. The documents cited in this specification are incorporated as a part of this specification. Hereinafter, the present invention will be described in more detail with reference to examples, but the present invention is not limited to these examples.

[Example 1] In vitro osteoclast formation In vitro osteoclast formation used in Examples 2 to 11 was performed as follows. Non-adherent bone marrow cells from C57BL / 6 mice were seeded (5 × 10 ⁴ per well in a 24-well plate, 1.5 × 10 ⁷ per 10 cm dish) and contained 10 ng / ml M-CSF (Genzyme). The cells were cultured in α-MEM (Gibco BRL) containing% FBS (Sigma) for 2 days, and non-adherent cells including lymphocytes were washed away, and the adherent cells were used as BMM. Monocyte / macrophage progenitor cells from Fos ^{− / −} and TRAF6 ^{− / −} mice were harvested from the spleen and similarly cultured with 10 ng / ml M-CSF for 2 days. These osteoclast precursor cells were further cultured in the presence of 100 ng / ml soluble RANKL (Peprotech) and 10 ng / ml M-CSF to generate osteoclasts. Unless otherwise stated, these reagents were used at these concentrations. The calcineurin inhibitor FK506 (Calbiochem) or cyclosporin A (Sigma) was added simultaneously with RANKL. Three days later, TRAP ⁺ multinucleated (> 3 nuclei) cells were counted. All data were expressed as mean ± se (n = 6). TRAP ⁺ MNC was tested for bone resorption activity on dentin sections and calcitonin receptor expression as previously described (Takayanagi, H. et al. (2000b) Nature 408, 600-605).

[Example 2] Northern blot analysis and immunofluorescence staining
BMM was stimulated in the presence of M-CSF for the time indicated by RANKL. 5 mg of total RNA isolated with Sapasol was loaded in each lane. The blot was hybridized with a 230 bp EcoRI and SmaI fragment of mouse NFATc1 cDNA (Sherman, MA et al. (1999) J Immunol 162, 2820-2828) labeled with [α- ³² P] dCTP. For immunostaining, cells were fixed with 4% paraformaldehyde for 20 minutes and treated with 0.2% triton X for 5 minutes. Cells were sequentially incubated with 5% BSA / PBS for 30 minutes, 2 μg / ml mouse anti-NFATc1 monoclonal antibody in PBST (7A6, Santa Cruz) for 60 minutes, and then 4 μg / ml Alexa 488-labeled anti-mouse IgG antibody (Molecular Probe ) For 60 minutes.

[Example 3] Induction of NFATc1 by RANKL
To identify unknown pathway (s) that are selectively activated by RANKL, we used an oligonucleotide array (Affymetrix GeneChip) to perform genome-wide screening of mRNA induced by RANKL. Was performed. Bone marrow-derived monocyte / macrophage precursor cells (BMMs) obtained by culturing bone marrow cells in the presence of M-CSF for 48 hours, and further recombinant RANKL in the presence of M-CSF For 72 hours to differentiate into osteoclasts (Takayanagi, H. et al. (2000b) Nature 408, 600-605). In this process, total RNA was extracted at 2, 24, and 72 hours after RANKL addition, and the mRNA expression profile was compared to that of unstimulated BMM. To exclude pathways that are not specific for RANKL, M-CSF alone or BMM stimulated with IL-1 and M-CSF were used as negative controls. No osteoclasts were formed in these BMMs.

  The specific procedure of gene chip analysis is shown below. BMM was stimulated with M-CSF alone, RANKL and M-CSF, or IL-1 (10 ng / ml) and M-CSF for 2, 24, and 72 hours. Total RNA was extracted from BMM using Sepasol RNA extraction kit (Nakarai). Using total RNA (15 μg), cDNA synthesis by reverse transcription was followed by biotinylated cRNA synthesis by in vitro transcription. After fragmenting the cRNA, hybridization was performed on a mouse U74A GeneChip (Affiymetirix) displaying 11,000 mouse gene / EST probes according to the manufacturer's protocol. After washing the chip, it was stained with SA-PE and read using an Affymetrix Genechip scanner and the attached gene expression software.

  As expected, genes known to be abundantly expressed in osteoclasts, such as TRAP, calcitonin receptor, cathepsin K, carbonic anhydrase II (CAII), and matrix metalloprotease (MMP) -9 Was specifically induced at high levels by RANKL (FIG. 1A). The present inventors have noted the induction of transcription factors likely to include key regulators of RANKL signaling. FIG. 1B shows GeneChip data for highly induced transcription factors as well as those known to act downstream of RANKL signaling. Remarkably, NFATc1 outperformed other transcription factors in both mRNA expression levels and fold increase in expression. Consistent with this result, Northern blot analysis showed that RANKL induced BMM NFATc1 mRNA more than 12-fold in osteoclastogenesis (FIG. 2).

[Example 4] Expression of NFATc1 in osteoclasts
The expression and cellular localization of NFATc1 protein in BMM stimulated with RANKL and developing into multinucleated osteoclasts was examined. Immunofluorescence staining showed that expression of the NFATc1 protein increased steadily during RANKL-mediated osteoclastogenesis, and that the protein gradually translocated to the nucleus (FIG. 3). This protein is initially induced primarily in the cytoplasm and begins to translocate to the nucleus after 24 hours (FIG. 3). Expression of NFATc1 is earlier than the appearance of TRAP ⁺ cells observed after 48 hours. At 72 hours after the addition of RANKL, almost all nuclei of TRAP ⁺ multinucleated cells (MNCs) were NFATc1 positive (FIG. 3). This protein was also expressed in the cytoplasm. Transcription factors such as NF-κB were activated immediately after RANKL stimulation, and expression levels did not change significantly during osteoclastogenesis (FIG. 1B and (Takayanagi, H. et al. (2002) Nature 416). , 744-749; Takayanagi, H. et al. (2000b) Nature 408, 600-605) Therefore, the function of these transcription factors is not caused by the It was suggested that it was regulated by nuclear translocation induced by activation at an early stage of cell differentiation. In this context, RANKL-induced elevation of NFATc1 and concomitant nuclear translocation may be the most characteristic of cellular events in osteoclast differentiation. Further, according to histological examination of bone tissue, expression of NFATc1 was confirmed to be caused by TRAP-positive giant cells (Fig. 4), which are osteoclast markers, and cells expressing cathepsin K or MMP-9 (not shown). Data), and it was confirmed that NFATc1 was also expressed in osteoclasts in vivo.

[Example 5] Overexpression of NFATc1
To investigate the effect of NFATc1 ectopic expression, retrovirus vector (pMX-NFATc1-IRES-EGFP) carrying NFATc1 cDNA and internal ribosomal entry site-enhanced green fluorescent protein (EGFP) Was built. The retroviral vector pMX-NFATc1-IRES-EGFP is the same as pMX-IRES-EGFP (Nosaka, T. et al., EMBO J. 18, 4754-65 (1999)) using a 2.7 kb EcoRI fragment of mouse full-length NFATc1 cDNA. It was constructed by inserting into the site. Other retroviral vectors (pMX-c-Fos-IRES-EGFP, pMX-c-Jun-IRES-EGFP) have been described previously (Takayanagi, H. et al. (2002) Nature 416, 744-749). . Retrovirus packaging was performed by co-lipofection of 293T cells with these pMX vectors and pPAMpsi2 (Sato, M. et al., FEBS Lett. 441: 106-110 (1998)). Two days after inoculation, the efficiency of infection of BMM was assayed and further cultured with M-CSF in the presence or absence of RANKL for the osteoclast formation assay. Three days later, osteoclast formation was assessed by TRAP staining and bone resorption assays.

  Retains pMX-NFATc1-IRES-EGFP, or c-Fos (pMX-c-fos-IRES-EGFP), c-Jun (pMX-c-jun-IRES-EGFP), or EGFP only (pMX-EGFP) Other control viruses infected BMM. Surprisingly, overexpression of NFATc1 in BMM efficiently induced osteoclast differentiation even without exogenous addition of RANKL (FIGS. 5A and B). The retrovirus carrying c-Fos also induced osteoclastogenesis without RANKL, but its induction activity was much lower than that of the NFATc1 virus. In fact, virus-mediated osteoclastic activity of NFATc1 as much as RANKL itself (FIG. 5A) suggests that increased NFATc1 expression is sufficient to induce osteoclastogenesis. As expected, the addition of NFATc1 virus and RANKL had a synergistic effect on osteoclastogenesis (FIG. 5A).

[Example 6] Effect of calcineurin inhibitor In lymphocytes, it is well known that NFAT translocation to the nucleus is regulated by calcineurin-dependent dephosphorylation (Crabtree, GR, and Olson, EN (2002). ) Cell 109 Suppl, S67-79; Rao, A. et al. (1997) Annu Rev Immunol 15, 707-747). Immunosuppressants FK506 or cyclosporin A (CyA) specifically inhibit calcineurin dephosphorylation activity, thereby inactivating NFAT (Liu, J. et al. (1991) Cell 66, 807-815) . To investigate whether calcineurin-dependent activation of NFAT contributes to osteoclastogenesis, these calcineurin inhibitors were added to the osteoclastogenic system stimulated by RANKL and M-CSF. Both FK506 and CyA strongly inhibited RANKL-induced osteoclast formation from BMM in a dose-dependent manner (FIG. 6). The same inhibitory effect of both drugs on osteoclastogenesis was also observed in RAW 264.7 cells, a macrophage cell line that differentiates into osteoclasts in response to RANKL alone (data not shown), indicating that the calcineurin-NFAT pathway is It is suggested that it is important for RANKL-induced signaling for osteoclast formation. Interestingly, FK506 not only inhibited NFATc1 nuclear translocation, but also suppressed a characteristic increase in NFATc1 protein levels in osteoclastogenesis (FIG. 7).
Northern blot analysis showed that NFATc1 induction was suppressed at the mRNA level by FK506 (FIG. 8). NFAT has been reported to have a calcineurin-dependent positive feedback regulatory loop through binding to its own promoter (Zhou, B. et al. (2002) J Biol Chem 277, 10704-10711). Our results suggest that calcineurin-dependent autonomous amplification of the NFATc1 gene drives downstream of RANKL signaling.

Example 7 Ca2 ⁺ Signaling Induced by RANKL Calcineurin is activated by IP3-mediated Ca2 ⁺ release from the endoplasmic reticulum and subsequent Ca2 ⁺ entry into cells through the plasma membrane. (Crabtree, GR, and Olson, EN (2002) Cell 109 Suppl, S67-79). To obtain clues to calcineurin activation mechanism RANKL-induced blocked the mobilization of Ca ²⁺ by treating the cells with BAPTA-AM is a selective chelator Ca ^2+.

BAPTA treatment was performed as follows. Cells were incubated for 60 minutes in serum-free α-MEM in the presence of 15 μM BAPTA AM and 0.04% pluronic F127 to chelate intracellular Ca ²⁺ . The cells were then washed twice with α-MEM and incubated with α-MEM / FBS containing RANKL / M-CSF. This treatment was performed once a day during RANKL-induced osteoclastogenesis.

The Ca ²⁺ measurement was performed as follows. The cells were first incubated with serum-free DMEM for 30 minutes in the presence of 5 μM fluo-4 AM, 5 μM Fura Red AM, and 0.05% pluronic F127. The cells were then washed twice with DMEM and post-incubated with DMEM containing 10% FBS and 10 ng / ml M-CSF for 20 minutes. Cells loaded with these dyes were washed three times with Hank's balanced salt solution and mounted on an inverted table of a confocal microscope (Leica). Cells were excited at 488 nm, emission of fluo-4 at 505-530 nm and emission of FuraRed at 600-680 nm were measured simultaneously at 5-second intervals. To measure the intracellular Ca ²⁺ concentration of a single cell, the ratio of the fluorescence intensity of fluo-4 to FuraRed was calculated. The increase in ratio value from basal level was divided by the maximum ratio increase obtained with the addition of 10 μM ionomycin and expressed as% max ratio increase.

As expected, inhibition of Ca ²⁺ signaling by BAPTA-AM is with strongly inhibits osteoclast formation induced RANKL, since elevated expression of NFATc1 also inhibited (Fig. 9), Ca ²⁺ dependent Signaling was suggested to be important for RANKL-induced calcineurin activation leading to NFATc1 induction and activation. Consistent with these results, persistent Ca ²⁺ oscillations were observed in BMM stimulated with RANKL even after 24 h after stimulation (FIG. 10), which was observed in cells stimulated with IL-1. Did not. However, no rapid increase in Ca ²⁺ concentration was observed immediately after stimulation with either RANKL or IL-1. Addition of Ca2 ⁺ ionophores such as PMA and ionomycine did not significantly up-regulate RANKL-induced osteoclastogenesis (data not shown), indicating that Ca is important for NFATc1 activation in osteoclastogenesis. ²⁺ signaling may be a continuous Ca ²⁺ oscillation rather than a Ca ²⁺ spike. It is not clear whether RANKL directly activates Ca ²⁺ mobilization or activates it through induction of another ligand-receptor system.

Example 8 Osteoclastogenesis from ES Cells Gene targeting strategies in mice have limitations in elucidating multifunctional genes whose deletion would result in an embryonic lethal phenotype. For example, NFATc1-deficient mice fail to develop normal heart valves and septum and die of circulatory disturbance prior to day 14.5 of development, so the details of NFATc1 involvement in bone cell development cannot be assessed (de la Pompa, JL et al. (1998) Nature 392, 182-186; Ranger, AM et al. (1998) Nature 392, 186-190). Furthermore, it is almost impossible to obtain hematopoietic cells from fetal liver of mice lacking NFATc1. Therefore, the present inventors have constructed a new ES cell culture system in which ES cell-derived monocyte / macrophage precursor cells can efficiently differentiate into osteoclasts. In order to maintain a microenvironment suitable for the generation of hematopoietic cells, three-dimensional culture was achieved by forming an ES cell mass (embryoid body; EB).

Osteoclast formation from ES cells was performed as follows. 15% of feeder cells independent ES cell lines on dishes coated with gelatin ^{FBS, 1x10 -4 M 2-mercaptoethanol} , and 1000 U / ml leukemia inhibitory factor ( LIF), 50 U / ml streptomycin and 50 mg / ml penicillin And maintained in DMEM (Gibco). A wild-type ES cell line E14K [NFATc1 ^{+ / +} ], a heterozygous mutant strain NFDK34 [NFATc1 ^+/- ], and a homozygous mutant strain [NFATc1 ^{− / −} ], NFDK10 or NFDK17 were used. To differentiate the ES cells into osteoclasts, the ES cells were trypsinized and plated in an inverted bacterial Petri dish with a hanging drop to give ³ × 10 ³ ES cells in a 25 μl drop. On Day 2, the embryoid body in the hanging drop was collected and transferred to a culture dish. These EBs were cultured in Petri dishes for 2 days in α-MEM / 10% FBS containing a high concentration of M-CSF (100 ng / ml). Thereafter, EBs were treated with 0.25% trypsin-EDTA-PBS to collect single cells, and the expression of CD11b, c-Fms and RANK was examined by flow cytometry (day 4). The collected single cells were replated and cultured in α-MEM / 10% FBS containing RANKL (100 ng / ml) and M-CSF (10 ng / ml). Four days later, TRAP ⁺ MNC was efficiently formed during the culture (day 8). These TRAP ⁺ MNCs were identified as meeting osteoclast criteria, including bone resorption activity on dentin sections.

As described above, when EBs are stimulated with M-CSF for 2 days, the generation of monocyte / macrophage progenitor cells is induced, and these are enzymatically dispersed and replated on a new dish, followed by RANKL in the presence of M-CSF for 4 days. Upon stimulation, TRAP ⁺ multinucleated cells having bone resorption activity were differentiated from the dentin section (FIG. 11). Thus, this culture system makes it possible to examine the function of genes that induce embryonic lethality in mice by deletion in osteoclastogenesis.

Example 9 Inactivation of NFATc1 Gene The present inventors generated NFATc1 ^{− / −} ES cell lines from NFATc1 ^+/− ES cells by selection with increasing G418 concentration (Yoshida, H. et al. (1998) Immunity 8, 115-124). NFATc1 ^{− / −} ES cells completely lacked mature osteoclast formation in the presence of RANKL and M-CSF (FIG. 12A). In addition, no TRAPL ⁺ cells were found in RANKL-stimulated NFATc1 ^{− / −} ES cells. However, the generation of CD11b (Mac-1) or nonspecific esterase (NSE) positive monocyte / macrophage progenitor cells from NFATc1 ^{− / −} ES cells was not achieved from parental NFATc1 ^+/− ES cells. (Fig. 12B). In addition, the expression of c-Fms (M-CSF receptor) and RANK in NFATc1 ^{− / −} progenitors was comparable to those in NFATc1 ^+/− progenitors (data not shown). Taken together, these results demonstrate that NFATc1 plays an important role in TRAP gene induction and differentiation of monocyte / macrophage progenitor cells into mature multinucleated osteoclasts.

Example 10 Transcriptional Targets and Partners of NFATc1 We present genes specifically up-regulated in terminal differentiation of osteoclasts, including TRAP, calcitonin receptor, cathepsin K, CAII, and MMP-9 (Anusaksathien, O. et al. (2001) J Biol Chem 276, 22663-22674; David, JP et al. (2001) J Cell Physiol 188, 89-97; Motyckova, G. et al. (2001) ) Proc Natl Acad Sci USA 98, 5798-5803; Reddy, SV et al. (1995) J Bone Miner Res 10, 601-606). Thus, all of these promoters contained multiple NFATc1 sites, suggesting that NFATc1 directly regulates these osteoclast genes. To investigate this possibility, the TRAP promoter or calcitonin receptor promoter (P3 promoter) known to respond to RANKL stimulation (Anusaksathien, O. et al. (2001) J Biol Chem 276, 22663-22674; Matsumoto (2001) J Biol Chem 276, 33086-33092; Reddy, SV et al. (1995) J Bone Miner Res 10, 601-606). A luciferase reporter plasmid was constructed.

  Construction of the reporter plasmid for the luciferase assay was performed as follows. Insert about 1.8 kb KpnI-BglII fragment of mouse TRAP promoter (Reddy, SV et al. (1995) J Bone Miner Res 10, 601-606) into the same site of PicaGene luciferase reporter plasmid, basic vector 2 (TOYO INK) Thus, a reporter plasmid pTRAP-luc was constructed. The osteoclast-specific mouse calcitonin receptor promoter region (P3 promoter) was replaced with a sense 5'-AAGATCTCAACACAACTCTTAACTGGACC-3 '(SEQ ID NO: 1) and an antisense 5'-AAGCTTAATGTAAATGTAGATATAGC-3' (SEQ ID NO: 2). And amplified by PCR. Another reporter plasmid, pCTR-luc, was constructed by inserting a 461 bp PCR fragment into the BglII-HindIII site of basic vector 2. NFATc1 mutants T552G and R485A / L486A were generated by site-directed mutagenesis (Sawano, A., and Miyawaki, A. (2000) Nucleic Acids Res 28, E78). Primers for T552G and R485A / L486A are 5'-TGAGGAAAGGGGAGGGAGATATCGGGAGGAAGAACACCAGGG-3 '(SEQ ID NO: 3) and 5'-TTGGGACGGCTGACGACGCCGCGCTGAGGCCTCACGCCTTCTACCAGG-3' (SEQ ID NO: 4), respectively. 293T cells were transfected using FuGene 6 (Roche) and luciferase activity was assayed 24 hours after transfection. All data were expressed as mean ± s.e. (N = 6).

  TRAP has no strict specificity for osteoclasts, but is the most widely used osteoclast marker and is one of the most up-regulated genes by RANKL (FIG. 1A). Calcitonin receptors, on the other hand, are rather specific to osteoclasts and are not expressed on their precursor cells. As shown in FIG. 13, both promoters were activated only by overexpression of NFATc1, but this activation was not as strong as expected. It is well known that NFAT cooperates with AP-1 on the IL-2 promoter in T cell activation (Chen, L. et al. (1998) Nature 392, 42-48; Macian, F. et al. (2000) Embo J 19, 4783-4795; Macian, F. et al. (2001) Oncogene 20, 2476-2489). Since c-Fos is a component of the AP-1 complex essential for osteoclast formation, this c-Fos was co-transfected with NFATc1. Interestingly, co-transfection of NFATc1 and c-Fos induced significant activation of both the TRAP promoter and the calcitonin receptor promoter, whereas c-Fos alone activated these promoters only slightly. (FIG. 13). These results strongly suggest that in these osteoclast genes, NFATc1 cooperates with the AP-1 complex, including c-Fos. NFAT mutants with impaired molecular interaction with AP-1 have been reported based on the crystal structure of the NFAT: AP-1: DNA complex (Chen, L. et al. (1998) Nature 392, 42-48; Macian, F. et al. (2000) Embo J 19, 4783-4795). Similarly, the present inventors constructed NFATc1 mutants NFATc1T552G and NFATc1R485A / L486A, which cannot cooperate with AP-1, by introducing mutations at the binding sites for c-Jun and c-Fos, respectively. (FIG. 14). As expected, these mutant NFATc1 failed to activate these two promoters by synergistic co-operation with c-Fos (FIG. 13). It has been suggested that it is an essential partner of NFATc1 in the transcription machinery to be activated.

Example 11 Relationship of TRAF6, c-Fos, and NFATc1 To gain further clues about the relationship between the calcineurin-NFATc1 pathway, and other known essential signaling pathways such as the TRAF6 and c-Fos pathways, RANKL-induced NFATc1 expression in BMM from TRAF6 or c-Fos deficient mice was evaluated. Induction of NFATc1 protein was completely inhibited in c-Fos-deficient cells, and decreased, though not completely, in TRAF6 cells. Furthermore, Northern blot analysis demonstrated that both TRAF6 and c-Fos were essential for normal induction of NFATc1 mRNA (FIG. 15A). As expected, induction of NFATc1 mRNA was entirely dependent on c-Fos, suggesting that NFATc1 is a transcription target of AP-1. However, unlike NFATc1 protein expression, NFATc1 mRNA induction was also partially observed in TRAF6-deficient BMM, indicating that TRAF6 regulates NFATc1 expression at both the transcriptional and post-transcriptional levels. Suggested (FIG. 15B). Thus, despite the differences in regulatory mechanisms, NFATc1 plays a role in integrating these two pathways essential for osteoclastogenesis, since NFATc1 is induced downstream of both the TRAF6 and c-Fos pathways It was shown to fulfill.

[Example 12] In vitro osteoclast formation In vitro osteoclast formation used in Examples 13 to 18 was performed as follows. Non-adherent bone marrow cells from C57BL / 6 mice (5 x 10 ⁴ cells per well in a 24-well plate, 1.5-2.0 10 ⁷ cells per 10-cm plate) were seeded, and 10 ng / ml M-CSF (Genzyme) was added. The cells were cultured in α-MEM (Gibco BRL) containing 10% FBS (Sigma). Two days later, non-adherent cells including lymphocytes were washed away, and the adherent cells were used as BMM. BMM was further cultured in the presence of 100 ng / ml soluble RANKL (Peprotech) and 10 ng / ml M-CSF to form osteoclasts. To evaluate the effect of leflunomide, the active metabolite of leflunomide, A77 1726 (Aventis Pharma) was added at the same time as RANKL unless otherwise noted. Three days later, TRAP ⁺ multinuclear (> 3 nuclei) cells were counted. All data were expressed as mean ± se (n = 6). TRAP ⁺ MNC tested bone resorption activity on dentin sections and calcitonin receptor expression as previously described (Takayanagi, H. et al., Nature 408, 600-5 (2000)).

Example 13 Effect of Leflunomide on Osteoclast Differentiation To investigate whether leflunomide directly acts on osteoclast differentiation and if so how, RANKL stimulated mice in the presence of M-CSF first The effect of leflunomide on in vitro osteoclastogenesis from bone marrow monocytes / macrophages (BMM) was examined. As shown in FIG. 16A, leflunomide strongly inhibited the formation of tartrate-resistant acid phosphatase-positive polynuclear cells (TRAP ⁺ MNC) in a dose-dependent (1-100 μM) manner. At these concentrations, leflunomide had no significant effect on BMM survival and proliferation (see FIGS. 16A, 17; data not shown). A similar inhibitory effect was also observed in RANKL-induced osteoclastogenesis of RAW 264.7, a macrophage that also differentiates into osteoclasts in response to RANKL (data not shown). The inhibitory effect of leflunomide on osteoclast differentiation was restored by the addition of an excess of uridine, suggesting that this inhibition was due to de novo (new) blockade of pyrimidine biosynthesis (FIGS. 16A, 17). ). The stage at which the process of osteoclast formation was inhibited by the presence of leflunomide during RANKL stimulation was examined. As shown in FIG. 16B, when leflunomide was added at the early stage, i.e., between 0 and 24 hours after RANKL stimulation, TRAP ⁺ MNC formation was almost completely inhibited, but the drug was treated for 48 hours after RANKL stimulation. No such inhibition was observed when added later (FIG. 16B).

[Example 14] Effect of leflunomide on downstream signal pathway activated by RANKL Based on the above results, the mechanism by which leflunomide interferes with the intracellular signaling pathway activated by RANKL was examined. Specifically, RANKL transmits signals to cells through binding to its receptor, RANK, and activates TNF receptor-related factor (TRAF) family proteins such as TRAF6, which is the NF-κB pathway, mitogen-activated Activates the protein kinase (MAPK) pathway and the Akt pathway (Theill, LE, Annu Rev Immunol 20, 795-823 (2002); Takayanagi, H. et al., Nature 408, 600-5 (2000); Wong , BR et al., Mol Cell 4, 1041-9 (1999); Takayanagi, H. et al., Nature 416, 744-49 (2002)). RANKL also induces c-Fos that can function in the AP-1 transcription factor complex (Takayanagi, H. et al., Nature 416, 744-49 (2002); Matsuo, K. et al., Nat Genet 24, 184-7. (2000); Karsenty, G. & Wagner, EF, Dev Cell 2, 389-406. (2002)). RANKL selectively induces the NFATc1 gene, a member of the nuclear factor of activated T cells (NFAT) family, in a TRAF6 and c-Fos dependent manner. RANKL also causes Ca ²⁺ signaling and causes calcineurin-mediated activation of NFATc1. This activation is necessary and sufficient for osteoclast differentiation (see Examples 1-11).

  To elucidate the molecular basis of leflunomide's inhibition of osteoclastogenesis, the effect of leflunomide on RANKL-induced activation of these downstream signaling molecules was investigated. Specifically, BMM pre-incubated for 24 hours in the presence or absence of leflunomide (30 μM) was stimulated with RANKL, and an anti-phosphorylated MAPK antibody, an anti-MAPK antibody, an anti-phosphorylated Akt antibody, or an anti-Akt antibody ( New England Biomeds) as described previously (Takayanagi, H. et al., Nature 408, 600-5 (2000); Takayanagi, H. et al., Nature 416, 744-49 (2002)). Was analyzed by immunoblotting. BMM was preincubated for 24 hours in the presence or absence of leflunomide (30 μM) for the electrophoretic mobility shift assay (EMSA). After RANKL stimulation, cell extracts were prepared and described previously using an oligonucleotide probe containing the NF-κB binding site of the IFN-β promoter (Takayanagi, H. et al., Nature 408, 600-5 ( 2000); Takayanagi, H. et al., Nature 416, 744-49 (2002)). Anti-Rel-A antibody (Santa Cruz) was used for supershift.

  As shown in FIG. 18A, no inhibitory effect was observed on RANKL-induced activation of MAPK, including ERK, p38, and JNK, in leflunomide-treated BMM. In addition, these molecules have no or little effect on RANKL-induced activation of NF-κB (FIG. 18B) and Akt (FIG. 18C), indicating that It was suggested that leflunomide may not be the target of the action in transmission.

[Example 15] Genome-wide screening of target gene for the action of leflunomide In order to identify the molecular target of leflunomide, a genome-wide screening of RANKL-induced genes in BMM was performed in the presence or absence of leflunomide. First, BMM was stimulated with RANKL (100 ng / ml) and M-CSF (10 ng / ml) in the presence or absence of leflunomide (100 μM) .After 24 hours, BMM was stimulated using Sepasol RNA extraction kit (Nakarai). Total RNA was extracted from. Using total RNA (15 μg), cDNA synthesis by reverse transcription was followed by biotinylated cRNA synthesis by in vitro transcription. After fragmenting the cRNA, hybridization was performed on a mouse U74Av2 GeneChip (Affiymetirix) displaying 12,000 mouse gene / EST probes according to the manufacturer's protocol. After washing the chip, the chip was stained with SA-PE and analyzed using an Affymetrix Genechip scanner and attached gene expression software.

  Analysis of mRNA expression of transcription factors and effector molecules involved in RANKL signaling revealed that leflunomide exhibited the strongest inhibitory effect on NFATc1 transcription. As shown in FIG. 19, mRNA expression of NFATc1 was more than 20-fold induced by RANKL, while leflunomide reduced this expression by 1/8. Leflunomide selectively reduced NFATc1 mRNA levels rather than nonspecifically inhibiting transcription, as mRNA induction of the other genes shown in FIG. 19 did not have the dramatic effect of NFATc1. That is clear. These results suggested that leflunomide targets include a RANKL-induced NFATc1-induced pathway.

[Example 16] NFATc1 is an important target of leflunomide action on RANKL signaling
The protein expression of essential molecules in osteoclastogenesis including NFATc1, TRAF6, and c-Fos was further analyzed. Leflunomide strongly inhibited the expression of NFATc1 (FIG. 20). Induction levels of TRAF6 and c-Fos were also reduced, especially in the late stage (3 days after RANKL stimulation), but induction was still observable. Immunostaining also showed little expression of NFATc1 in BMM stimulated with RANKL in the presence of leflunomide, but under the same conditions, expression of c-Fos or TRAF6 was still observed, albeit at lower levels (FIG. 21). .

The mechanism by which NFATc1 expression was down-regulated by leflunomide was further investigated. RANKL-induced Ca ²⁺ oscillation is essential for autonomous amplification of NFATc1 gene expression, and this autonomous amplification has been shown to be important for sustained NFATc1-dependent transcriptional programs (see Examples 1 to 11). ). Therefore, the effect of leflunomide on Ca ²⁺ signaling in BMM stimulated with RANKL was examined. Interestingly, as shown in FIG. 22, since the Ca2 ⁺ oscillation induced by RANKL was inhibited by leflunomide, the inhibition of the Ca2 ⁺ signal induced by RANKL was at least due to the inhibition of NFATc1 expression by leflunomide. It was suggested to be partially explained. Furthermore, down-regulation of TRAF6 and c-Fos, transcription factors important for NFATc1 induction by RANKL, may additionally contribute to potent inhibition of NFATc1 by leflunomide. While cytokines such as TNF and IL-1 functionally related to RANKL do not induce oscillation (see Examples 1 to 11), it is still unknown how RANKL induces oscillation. The detailed mechanism of leflunomide on induced Ca2 ⁺ oscillations needs to be further studied.

  In addition, the immunoblot analysis and the immunofluorescence staining were performed as follows. To assess the expression of downstream molecules of RANKL signaling in osteoclastogenesis, BMM was stimulated with RANKL / M-CSF in the presence or absence of leflunomide (30 μM). Every 24 hours after the addition of RANKL, the whole cell extract of BMM was assayed by immunoblotting using the specific antibodies NFATc1 (7A6), TRAF6 (H-274), and c-Fos (H-125) (Santa Cruz). . For immunostaining, cells were cultured with RANKL / M-CSF for 3 days in the presence or absence of leflunomide (30 μM), fixed with 4% paraformaldehyde for 20 minutes, and then treated with 0.2% Triton X for 5 minutes. Cells are sequentially incubated with 5% BSA / PBS for 30 minutes, PBST containing 2 μg / ml anti-NFATc1 and 4 μg / ml anti-c-Fos or anti-TRAF6 antibodies for 60 minutes, then 4 μg / ml Alexa 488 or 8 μg / ml The cells were incubated with a secondary antibody (Molecular Probe) labeled with Alexa 546 for 60 minutes.

The Ca ²⁺ measurement was performed as follows. BMM was incubated with RANKL / M-CSF for 48 hours in the presence or absence of leflunomide (30 μM). Cells were loaded with calcium indicators (fluo-4 AM and Fura Red) as described above and observed with a confocal microscope (C1, Nikon). The increase in the ratio of the fluorescence intensity from the basal level was divided by the maximum increase in the ratio obtained by adding 10 μM ionomycin, and expressed as a% max ratio increase.

Example 17 Effect of Ectopic Expression of TRAF6, c-Fos, or NFATc1 on Inhibition of Osteoclastogenesis by Leflunomide Blockade of de novo pyrimidine biosynthesis by leflunomide thus results in multiple levels of RANKL signaling And caused a marked down-regulation of NFATc1, a protein important in osteoclastogenesis. Therefore, the effect of ectopic expression of TRAF6, c-Fos, or NFATc1 on the suppression of RANKL-induced osteoclastogenesis by leflunomide was investigated by retroviral gene transfer as in Example 5.

  The retroviral vectors pMX-NFATc1-IRES-EGFP (pMX-NFATc1), pMX-c-fos-IRES-EGFP (pMX-c-fos), and pMX-TRAF6-IRES-EGFP (pMX-TRAF6) are described above and in the literature (Takayanagi, H. et al., Nature 408, 600-5 (2000); Takayanagi, H. et al., Nature 416, 744-49 (2002)). Retrovirus packaging was performed by colipofecting these pMX vectors and pPAMpsi2 into 293T cells. Two days after inoculation, BMM was stimulated with RANKL / M-CSF and leflunomide (30 μM). Three days after the addition of RANKL, osteoclast formation was assessed by TRAP staining and bone resorption assays as previously described (Takayanagi, H. et al., Nature 408, 600-5 (2000)).

  Interestingly, ectopic expression of NFATc1 led to efficient formation of osteoclasts in the presence of leflunomide, but not c-Fos or TRAF6, indicating that RANKL-related NFATc1-induced pathways It became even clearer that it was an important target for action (FIG. 23).

Example 18 T cell-independent effects of leflunomide on inflammatory bone destruction In arthritis, T cells are involved in several pathways causing bone destruction (Choy, EH & Panayi, GS, N Engl J Med 344, 907-16 (2001)). In the arthritic synovium, abnormal activation of the immune system induces macrophage-derived inflammatory cytokines such as TNF-α and IL-1, which strongly induce RANKL in synovial fibroblasts (Takayanagi, H. et al. al., Arthritis Rheum. 43, 259-69 (2000); Takayanagi, H. et al., Nature 408, 600-5 (2000); Takayanagi, H. et al., Biochem Biophys Res Commun 240, 279-86. (1997)). In inducing bone-related disease states, T cells have been implicated in the initiation and exacerbation of this synovitis. The importance of this pathway is evident from the observed significant therapeutic effect on arthritic bone destruction in neutralizing these cytokines (Feldmann, M. & Maini, RN, Annu Rev Immunol 19 , 163-96 (2001); Choy, EH & Panayi, GS, N Engl J Med 344, 907-16 (2001)). In addition, activated T cells expressing RANKL may directly contribute to osteoclastogenesis (Kong, YY et al., Nature 402, 304-9 (1999)). As a result, RANKL levels induced by both synovial fibroblasts and T cells are too high to be counterbalanced by inhibitors such as IFN-γ or osteoprotegerin (OPG) (Kong, YY et al., Nature 402, 304-9 (1999); Takayanagi, H. et al., Nature 408, 600-5 (2000)). Some antirheumatic drugs have osteoprotective effects (Kremer, JM, Ann Intern Med 134, 695-706 (2001); Jones, G. et al., Rheumatology (Oxford) 42, 6-13 (2003) ), It is difficult to distinguish between T cell dependent effects and T cell independent effects due to the complex mechanism of bone destruction. Therefore, the present inventors have developed a model for inducing inflammatory bone destruction without T cells.

A lipopolysaccharide (Rag-2 ^-/- mouse) lacking both T cells and B cells (Rag-2 ^-/- mice) (Shinkai, Y. et al., Cell 68, 855-67 (1992)) LPS) was injected. This endotoxin-induced bone resorption was performed as follows. Rag-2 ^{− / −} mice (Taconic) have been described previously (Shinkai, Y. et al., Cell 68, 855-67 (1992)). As described in 7-week old mice (n = 10) (Takayanagi, H. et al., Nature 416, 744-49 (2002); Chiang, CY, Infect. Immun. 67, 4231-6 (1999)) Lipopolysaccharide (LPS) (Sigma) was administered at 25 mg / kg body weight by local injection into the calvaria. Mice received a single injection of leflunomide (10 mg / kg body weight) or saline at the LPS injection site. Five days later, calvarial tissues were embedded in 5% carboxymethyl cellulose sodium salt (CMC), frozen, and serial sections were described for TRAP (and hematoxylin) and NFATc1 (Takayanagi, H. et al., Nature 416, 744-49 (2002)). Parameters such as the number of osteoclasts and the number of NFATc1 ⁺ cells per trabecular surface millimeter were determined. Statistical analysis was performed by Student's t-test. These experiments have been approved by the Institute's Committee on Animal Experiments.

Low dose LPS infusion did not induce bone destruction in T cell deficient nude mice (Ukai, T., J. Periodontal. Res. 31, 414-22 (1996)), but we We succeeded in inducing bone destruction in Rag-2 ^{− / −} mice using LPS. This indicated that lymphocytes were not essential in this model system (FIG. 24). Interestingly, topical administration of leflunomide to these mice significantly improved bone destruction and excessive osteoclastogenesis (FIGS. 24, 25A). It is noteworthy that many of the inflammatory lesion TRAP ⁺ MNCs stained positive for NFATc1 (FIG. 24) and were significantly reduced by leflunomide treatment (FIG. 25A). In contrast, little expression of other NFAT family members such as NFATc2 was detected (data not shown). Thus, leflunomide has an anti-osteoclastogenic effect in vivo, which is not targeted to T cells. The possibility that leflunomide may also affect other cell types cannot be strictly ruled out, but the results show that leflunomide directly inhibits osteoclast precursor cells by interfering with NFATc1 expression This is consistent with the above in vitro results. Consistent with this, leflunomide had only a minimal effect on the inflammatory response of Rag-2 ^{− / −} mice (FIG. 25B, and data not shown).

The present invention provides a method for controlling osteoclast formation, which targets a Ca ²⁺ signaling molecule, particularly a calcineurin signaling molecule including NFATc1. Further, a novel screening method for a compound that suppresses osteoclast formation, which targets a Ca ²⁺ signal molecule, particularly a calcineurin signaling molecule including NFATc1, has been provided. The present invention allows for the development of agents that selectively inhibit osteoclast formation. Such a drug is useful as a drug with few side effects that effectively suppresses bone destruction by osteoclasts. Since the enhancement of osteoclast formation is a cause of various osteopenic diseases including rheumatoid arthritis, the drug obtained by the present invention is expected to be applied to the treatment of these bone diseases.

It is a figure which shows the result of the genome-wide screening by GeneChip of the gene induced by RANKL. In the mRNA expression measurement results shown on the left, the four columns for each gene show data at 0 h, 2 h, 24 h, and 72 h in order from the top after RANKL addition. In the Fold increase measurement results shown on the right, the three columns for each gene show data for 2 h, 24 h, and 72 h (vs. 0 h) in order from the top after RANKL addition. (A) An osteoclast gene known to be highly expressed in osteoclasts. As expected, these genes were specifically and highly induced by RANKL. The left panel shows mRNA expression levels (expressed as average difference), and the right panel shows the fold increase in mRNA expression compared to unstimulated cells. GeneChip analysis was repeated several times with similar results. A representative data set is shown. (B) List of transcription factors that are highly induced by RANKL or have been suggested to act downstream of RANKL signaling. The expression level and fold induction (fold induction) of NFATc1 were the most remarkable. The expression levels of most of the other transcription factors are stable, suggesting that these factors are regulated by activation status rather than by quantity. FIG. 2 shows the expression of NFATc1 in osteoclasts in vitro. Fig. 4 shows Northern blot analysis of NFATc1 mRNA. Total RNA was extracted from BMM stimulated with RANKL / M-CSF for the times indicated. 5 μg of total RNA was applied to each lane. FIG. 2 shows the expression of NFATc1 in osteoclasts in vitro. FIG. 4 shows immunostaining of NFATc1 protein in BMM stimulated with RANKL / M-CSF. NFATc1 is first induced in the cytoplasm and gradually translocates to the nucleus. Note that it is significantly expressed in multinucleated osteoclasts. Nuclei were stained with DAPI in the lower panel. FIG. 2 is a view showing the expression of NFATc1 in osteoclasts in vivo. Figure 4 shows NFATc1 expression in osteoclasts in vivo. Serial sections of proximal tibia (roximal tibia) from 8-week-old C57 / BL6 mice were stained with TRAP and hematoxylin, or immunostained with anti-NFATc1 antibody followed by Alexa 488-conjugated secondary antibody. FIG. 4 is a diagram showing effective induction of osteoclast formation by retrovirus expression of NFATc1. (A) NFATc1 retrovirus induces TRAP ⁺ MNC formation without RANKL. This osteoclastogenic activity was much stronger than c-Fos virus. Both viruses showed a synergistic effect with RANKL. (B) Co-localization of GFP and TRAP in NFATc1 virus infected cells. Most of the virus infected cells detected by GFP were positive for TRAP staining. The transfection efficiency was estimated to be about 35% for each virus. FIG. 3 shows the regulation of NFATc1 by calcineurin and Ca ²⁺ signaling. 4 shows the inhibitory effects of the calcineurin inhibitors FK506 and cyclosporin A on RANKL-induced osteoclastogenesis in BMM. FIG. 3 shows the regulation of NFATc1 by calcineurin and Ca ²⁺ signaling. FK506 dose-dependently inhibited RANKL-induced NFATc1 nuclear translocation and increased expression. FIG. 3 shows the regulation of NFATc1 by calcineurin and Ca ²⁺ signaling. Calcium chelator BAPTA-AM suppressed RANKL-induced osteoclast formation in BMM. NFATc1 induction was also greatly suppressed. FIG. 3 shows the regulation of NFATc1 by calcineurin and Ca ²⁺ signaling. FK506 inhibited the induction of NFATc1 mRNA by RANKL. FIG. 3 shows the regulation of NFATc1 by calcineurin and Ca ²⁺ signaling. FIG. 9 shows Ca ²⁺ oscillations observed in RANKL-treated BMM. Not observed in IL-1 treated BMM. The Ca ²⁺ concentration was calculated from the ratio of the fluorescence intensity of fluo-4 to FuraRed. The increase in ratio value from basal level was divided by the maximum ratio increase obtained with the addition of 10 μM ionomycin. The addition of Ionomycin is indicated by the arrow. FIG. 2 is a schematic diagram of a novel ES cell culture system that generates osteoclasts. Under LIF-deficient conditions, an embryoid body was formed by the hanging drop method. These EBs were stimulated with M-CSF (100 ng / ml) for 2 days. Thereafter, EBs were trypsinized to obtain single cells, which were replated and cultured in the presence of RANKL (100 ng / ml) and M-CSF (10 ng / ml) for 4 days. FIG. 3 is a view showing that NFATc1-deficient ES cells cannot differentiate into osteoclasts. (A) NFATc1 ^{− / −} ES cells impair osteoclast formation. TRAP ⁺ MNC is efficiently formed in NFATc1 ^+/- ES cell culture (day 8), but TRAP ⁺ MNC is not found in NFATc1 ^-/- ES cell culture. (B) Monocyte / macrophage progenitor cells differentiate normally from NFATc1 ^{− / −} ES cells. When EBs are stimulated by M-CSF, CD11b is equally expressed on progenitor cells (left, day 4). Nonspecific esterase (NSE) ⁺ colonies also form normally in both cultures (right, day 8). FIG. 4 shows that NFATc1: AP-1 activates the TRAP promoter and the calcitonin receptor promoter. NFATc1 activates the TRAP promoter and calcitonin receptor promoter in cooperation with c-Fos. For these promoters, NFATc1 has a strong synergistic effect with c-Fos, but NFATc1 mutants that cannot cooperate with AP-1 have no synergistic effect with c-Fos. FIG. 2 shows the amino acid sequence of the NFATc1 mutant used in the previous figure. NFATc1 mutants were designed based on sequence homology with NFATc2 mutants that have little co-operation with AP-1. The sequences in the figure are SEQ ID NOs: 5 (NFATc2), 6 (NFATc1), 7 (NFATc1T552G), and 8 (NFATc1R485A / L486A) in order from the top. FIG. 2 shows that NFATc1 acts downstream of the TRAF6 pathway and c-Fos pathway in RANKL signaling. RANKL signaling is hierarchically regulated. When RANKL binds to RANK, the TRAF6, c-Fos, and Ca pathways are activated. The c-Fos-AP-1 pathway is essential for NFATc1 mRNA induction and cooperates with calcineurin-dependent activation of NFATc1 itself to form an autonomous replication mechanism. The TRAF6 pathway is important not only for the induction of NFATc1 mRNA, but also for the post-transcriptional regulation of NFATc1 such as protein stability. The amplified NFATc1 protein translocates to the nucleus in a calcineurin-dependent manner and cooperates with the AP-1 complex to induce the osteoclast gene. (A) Expression of NFATc1 protein in c-Fos ^{− / −} cells and TRAF6 ^{− / −} cells stimulated with RANKL and M-CSF. Splenocytes treated with M-CSF were further cultured with RANKL / M-CSF for 3 days and immunostained with anti-NFATc1 antibody. NFATc1 protein expression was either deleted or greatly reduced in c-Fos ^{− / −} cells or TRAF6 ^{− / −} cells, respectively. (B) Expression of NFATc1 mRNA in c-Fos ^{− / −} cells and TRAF6 ^{− / −} cells stimulated with RANKL and M-CSF. NFATc1 mRNA induction was not observed in c-Fos ^{− / −} cells, but weak induction was observed in TRAF6 ^{− / −} cells. FIG. 2 shows the effect of leflunomide on RANKL-induced osteoclastogenesis. (A) Leflunomide inhibits the formation of TRAP ⁺ MNC from BMM stimulated by RANKL in the presence of M-CSF. This inhibition was rescued by the addition of uridine (50 μM). (B) The effect of leflunomide depends on the stage of differentiation. The addition of leflunomide (30 μM) at the early stage (0-24 h) completely inhibits osteoclast formation, whereas the addition at the late stage (48-72 h) has no effect. FIG. 2 shows the effect of leflunomide on RANKL-induced osteoclastogenesis. 1 shows a microscope image (TRAP staining) of an in vitro osteoclast differentiation system. The inhibitory effect of leflunomide is rescued by uridine. Note that leflunomide has no effect on the proliferation of osteoclast precursor cells. It is a figure which shows the effect of leflunomide on the signal transduction pathway which RANKL activates. (A) Leflunomide has no significant effect on RANKL-induced activation of p38, ERK, or JNK in BMM. (B) Effect of leflunomide on RANKL-induced activation of NF-κB in BMM. (C) Activation of Akt by RANKL is not affected by leflunomide. FIG. 4 shows the results of genome-wide screening of leflunomide targets in RANKL-induced genes. The white column shows the mRNA expression (average difference) after 0 h of RANKL addition and the black column shows 24 hours after addition of RANKL. FIG. 4 shows the inhibition of RANKL signaling by leflunomide through the downstream of the NFATc1 pathway. Figure 4 shows the effect of leflunomide on expression of essential mediators of RANKL signal. NFATc1 is significantly induced by RANKL and is dephosphorylated (indicated by an asterisk) during differentiation. Leflunomide has a strong inhibitory effect on NFATc1 expression. FIG. 4 shows the inhibition of RANKL signaling by leflunomide through the downstream of the NFATc1 pathway. FIG. 4 shows c-Fos, TRAF6, and NFATc1 expression in BMM stimulated with RANKL in the presence or absence of leflunomide. FIG. 4 shows the inhibition of RANKL signaling by leflunomide through the downstream of the NFATc1 pathway. FIG. 4 shows the effect of leflunomide on RANKL-induced Ca ²⁺ signaling of BMM. The Ca ²⁺ indicator was loaded to detect changes in [Ca ²⁺ ] i in single cells. Each line represents a different cell in the same field (see text for details). FIG. 4 shows the inhibition of RANKL signaling by leflunomide through the downstream of the NFATc1 pathway. FIG. 4 shows the effect of retroviral overexpression of c-Fos, TRAF6, and NFATc1 on the inhibitory effect of leflunomide. FIG. 2 shows the effect of leflunomide in a T cell-independent model of inflammatory bone destruction in Rag-2 ^{− / −} mice. LPS induces increased osteoclastogenesis and inflammation-related bone resorption in Rag-2 ^{− / −} mice. Note that bone destruction is accompanied by an increase in the number of TRAP ⁺ and NFATc1 ⁺ cells. Leflunomide significantly improves this bone resorption. FIG. 2 shows the effect of leflunomide in a T cell-independent model of inflammatory bone destruction in Rag-2 ^{− / −} mice. (A) Effect of leflunomide on bone resorption parameters such as osteoclast number, NFATc1 ⁺ cell number, and increase in bone marrow cavity area. LEF stands for leflunomide. (* p <0.05 vs saline). (B) Effect of leflunomide on inflammatory cell layer formation. FIG. 2 shows the expression of NFATc1 at the bone / synovium boundary in rheumatoid arthritis patients. TRAP ⁺ MNC at the bone / synovial border in patients with rheumatoid arthritis is NFATc1 positive. Similar results were obtained for all samples. The right panel is a magnified view of bone resorbing osteoclasts that express NFATc1 at high levels in arthritic joints. Rheumatoid arthritis tissue was obtained from the bone / synovium interface of rheumatoid arthritis patients (n = 4) undergoing knee replacement due to severe bone destruction under informed consent. Samples were embedded in CMC, frozen and analyzed for NFATc1 expression as described herein.

Claims (20)

A method for promoting or suppressing osteoclast formation, comprising the step of promoting or suppressing Ca ²⁺ signaling in osteoclast precursor cells, respectively. The method according to claim 1, wherein the step of promoting or suppressing Ca ²⁺ signaling in the osteoclast precursor cell is a step of promoting or suppressing the expression or activity of calcineurin in the osteoclast precursor cell. The method according to claim 1, wherein the step of promoting or suppressing Ca ²⁺ signaling in the osteoclast precursor cell is a step of promoting or suppressing the expression or activity of NFATc1 in the osteoclast precursor cell.   2. The method of claim 1, further comprising promoting or suppressing expression or activity of one or more components of AP-1 in said osteoclast precursor cells.   4. The method of claim 3, wherein the activity of NFATc1 is an interaction with NFATc1 and an AP-1 component.   4. The method according to claim 3, wherein the activity of NFATc1 is an interaction with NFATc1 and DNA.   A method for suppressing osteoclast formation, which comprises the step of contacting an osteoclast precursor cell with a calcineurin inhibitor and / or leflunomide. A method of screening for a compound that suppresses osteoclast formation,
(A) detecting Ca ²⁺ signaling in osteoclast precursor cells in the presence of a sample containing a test compound; and (b) selecting a compound that suppresses the Ca ²⁺ signaling. Method.   Step (a) is a step of detecting calcineurin expression or activity in osteoclast precursor cells in the presence of a sample containing a test compound, and step (b) is suppressing the calcineurin expression level or activity. The method according to claim 8, which is a step of selecting a compound.   Step (a) is a step of detecting the expression or activity of NFATc1 in osteoclast precursor cells in the presence of a sample containing the test compound, and step (b) suppresses the expression level or activity of the NFATc1 The method according to claim 8, which is a step of selecting a compound.   11. The method of claim 10, wherein the activity of NFATc1 is an interaction with NFATc1 and an AP-1 component.   The method according to claim 10, wherein the activity of NFATc1 is an interaction between NFATc1 and DNA. A method of screening for a compound that suppresses osteoclast formation,
(A) in the presence of a sample containing the test compound, a step of detecting osteoclast formation in the osteoclast precursor cells exogenously expressing NFATc1, and (b) a compound that suppresses the osteoclast formation. Selecting. A pharmaceutical composition for the treatment of an osteopenic disease, comprising at least one compound that inhibits Ca ²⁺ signaling, and one or more pharmaceutically acceptable carriers.   15. The composition according to claim 14, wherein the compound is a compound that suppresses the expression or activity of calcineurin.   The composition according to claim 14, wherein the compound is a compound that suppresses the expression or activity of NFATc1.   17. The composition of claim 16, wherein said compound is a compound that inhibits interaction with NFATc1 and AP-1 components.   17. The composition of claim 16, wherein said compound is a compound that inhibits interaction with NFATc1 and DNA.   15. The composition of claim 14, wherein the osteopenic disease is a disease selected from the group consisting of osteoporosis, autoimmune arthritis, Paget's disease, and bone cancer. A method for producing osteoclasts,
(A) culturing embryonic stem cells to form a three-dimensional cell mass,
(B) culturing the cells obtained in step (a) in the presence of M-CSF,
(C) culturing the cells obtained in step (b) in the presence of M-CSF and RANKL,
A method that includes

Images