Differentiation and function of osteoclasts. Takeshi Miyamoto and Toshio Suda

Transcription

1 REVIEW Differentiation and function of osteoclasts Takeshi Miyamoto and Toshio Suda The Sakaguchi Laboratory of Developmental Biology, School of Medicine, Keio University, Tokyo, Japan (Received for publication on September 9, 2002) Abstract. Osteoclasts, which are responsible for bone resorption, are rare cells with only 2 3 cells seen per 1 mm 3 of bone. However, the loss of function in osteoclasts, problems with their differentiation and decrease in their number lead to bone osteosclerosis/osteopetrosis. On the other hand, an increase in their number or function induces bone osteoporosis, indicating that osteoclasts play a pivotal role in bone homeostasis. It has been demonstrated that bone destruction and hypercalcemia induced by metastatic tumors are carried out by osteoclasts activated by the tumor cells, and the inhibition of osteoclast formation prevents the bone destruction and even bone metastasis. Abnormal osteoclast function is closely related to various diseases. Furthermore, osteoclasts are indispensable in forming bone marrow to produce blood cells, and the absence of osteoclasts causes osteopetrosis, resulting in extramedullary hematopoiesis. Although the physiological roles of osteoclasts are well described, the mechanisms of their differentiation remain to be elucidated. Recently, RANK (receptor activator of nuclear factor kappab) and its ligand (RANKL) have been identified and their essential roles in osteoclastogenesis have been demonstrated, which has provided new insights into the osteoclast differentiation pathway. We have established an in vitro osteoclast culture system by isolating osteoclast precursor cells and culturing them in the presence of macrophage colony stimulating factor (M-CSF) and soluble RANKL. This system has enabled us to analyze the regulation mechanisms in osteoclast formation. (Keio J Med 52 (1): 1 7, March 2003) Key words: osteoclast, differentiation, bone resorption Role of Osteoclasts In postnatal life, bone is continuously remodeled by osteoblasts and osteoclasts, which synthesize bone matrix and resorb bone, respectively. Osteoblasts are derived from mesenchymal stem cells with a potential to differentiate to chondrocytes, adipocytes and stromal cells (Fig. 1).1 On the other hand, osteoclasts are derived from hematopoietic stem cells (Fig. 1). These two kinds of cell, though of different origins, cooperate with each other to maintain homeostasis of bony tissue. Any imbalance between bone formation by osteoblasts and bone resorption by osteoclasts causes bone abnormalities including osteopetrosis (due to a defect in osteoclastogenesis or the absence of functional osteoclasts), osteosclerosis (due to increased bone formation) and osteoporosis (due to increased bone resorption or a relative decrease in bone formation compared with bone resorption). Therefore, osteoclasts are indispensable to maintain bone homeostasis. Hematopoiesis shifts to the bone marrow concomitantly with the formation of the bone marrow cavity from the fetal liver. The bone is formed through chondrocytes or the mesenchymal membrane. Proliferating chondrocytes prevent the vascular invasion mediated by angiogenesis-inhibitory factors such as chondromodulin in the early embryonic stage.2 In the later stage, hypertrophic chondrocytes differentiating from chondrocytes in the center of the bone induce vascular invasion, mediated by angiogenic factors such as vascular endothelial growth factors (VEGFs) (Fig. 2). The induced vascular system establishes the blood supply to, and circulation in the bony structures, and it supplies osteoclasts, which resorb the bone to create the bone marrow cavity filled with hypertrophic and mineralized chondrocytes. Hematopoietic stem cells Reprint requests to: Dr. Toshio Suda, The Sakaguchi Laboratory of Developmental Biology, School of Medicine, Keio University, 35 Shinanomachi, Shinjuku-ku, Tokyo, , Japan, sudato@sc.itc.keio.ac.jp 1

2 2 Miyamoto T and Suda T: Differentiation and function of osteoclasts Characterization of Osteoclasts Fig. 1 Differentiation pathway of mesenchymal stem cells and hematopoietic stem cells. Key players in bone formation, osteoblasts and osteoclasts belong to different cell lineages. Osteoclasts are derived from hematopoietic stem cells, whereas osteoblasts are from mesenchymal stem cells. (HSCs) lodge in the stromal cells of the newly formed bone marrow, whereafter medullary hematopoiesis begins. In the absence of osteoclasts, the bone marrow cavity is not formed and extramedullary hematopoiesis takes place in the spleen, liver and kidney. The definition of osteoclasts is bone resorbing cells. Osteoclast differentiation has various characteristic features, such as multinucleation induced by the cell fusion of mononuclear osteoclasts to cover a larger area, synthesis of the vacuolar proton pump and acid to dissolve the bone mineral, formation of ruffled borders to secret protons and acid, and formation of a sealing zone to prevent proton and acid leakage. Mononuclear osteoclasts tightly adhere to bone, and fuse with each other to form multinuclear osteoclasts. The fused cells are re-organized, polarized and construct osteoclast-specific structures. Transcytosis develops from the basal to the apical side of osteoclasts to enable the discharge of dissolved bone debris.3 Multinucleation tartrate-resistant acid phosphatase (TRAP) activity and the calcitonin receptor (CTR) are available markers to detect osteoclasts, other than their bone resorbing activity. Differentiation of Osteoclasts Before 1980, the origin of multinuclear osteoclasts was controversial, with arguments existing between osteogenic precursors and HSCs, but the latter were shown to be the origin of osteoclasts.4 Osteoclasts are Fig. 2 Bone marrow formation. Bone is formed through chondrocytes. The hypertrophic chondrocytes give rise to mineralized or apoptotic chondrocytes at the center of the bone, then vascular cells invade toward the mineralized/apoptotic chondrocytes. Subsequently, osteoclasts develop in the bone marrow and enlarge the cavity through resorption, and hematopoiesis begins.

3 Keio J Med 2003; 52 (1): Fig. 3 Development of osteoclasts from hematopoietic stem cells (HSCs). HSCs give rise to monocyte/macrophage progenitor cells, and then cells differentiate into mononuclear osteoclasts, which express various osteoclastic molecules such as vitronectin receptor (integrin a v b 3 ), TRAP, CTR and MMP9. Mononuclear osteoclasts already have bone resorbing activity, and they fuse with each other to form multinuclear osteoclasts. The cytoskeletal system is reorganized in multinuclear osteoclasts to generate osteoclast-specific structures such as the sealing zone, ruffled borders, and a transcytosis system develops to discharge the resorbed bone debris. derived from HSCs through monocyte lineage progenitor cells (Fig. 3).5,6 Mononuclear osteoclasts express enzymatic activity of TRAP,7 CTR, vitronectin receptor (integrin a v b 3 ) and matrix metalloproteinase (MMP) 9. Although mononuclear osteoclasts also resorb bone, multinucleation is brought about by cell fusion of mononuclear osteoclasts. This multinucleation is the most characteristic feature of osteoclasts. Cell fusion is not an osteoclast-specific phenomenon, but is observed in other tissues such as myotubules and placentas. However, amongst these cell types osteoclasts are the largest, with more than 100 nuclei seen in some osteoclasts. The mechanism involved in the cell fusion is not clarified, but a mechanism similar to viral fusion is postulated, since some inhibitors of viral fusion prevent osteoclast fusion (unpublished data). Multinucleated cells are highly polarized cells and well re-organize to form new cytoskeletal constructions such as a sealing zone and ruffled borders. The polarity is strictly established in osteoclasts. The sealing zone and ruffled border develop in the basal side and the bone debris resorbed from the basal area is transferred via the cytoplasm to the apical side. An in vitro study on osteoclastogenesis has been performed in a co-culture system between osteoclast precursor cells derived from bone marrow or spleen mononuclear cells with stroma cells in the presence of osteotropic factors.8 In this system, bone marrow derived stroma cells (ST2, MC3T3E1 or MC3T3-G2- PA6) or primary mouse calvariae-derived osteoblasts were cultured as osteoclastogenesis-supporting cells, in the presence of osteotropic factors such as 1,25- dihydroxy-vitamin D 3 (1,25(OH) 2 D 3 ), parathyroid hormone (PTH), dexamethasone (DEX), prostaglandin E 2, and gp130 family proteins such as interleukin (IL)- 6, IL-11, oncostatin M or IL-1. Cell-to-cell contact between osteoclast precursor cells and stroma cells is required in this co-culture system,9 and this evidence provides the speculation that the osteoclastogenesis inducible molecule might be expressed on the stroma cells in the presence of osteotropic factors. This molecule was isolated from the stroma cells by the expression cloning technique and was named RANKL (also called osteoclast differentiation factor (ODF)/ osteoprotegerin ligand (OPGL)/TNF-related activation induced cytokine (TRANCE) RANKL is a type 2 transmembrane protein and a member of TNFa superfamily. RANKL has two receptors, one is RANK initially isolated from the dendritic cell cdna library10 and the other is the osteoprotegerin (OPG)/ osteoclastogenesis inhibitory factor (OCIF). This soluble receptor15,16 inhibits osteoclastogenesis by preventing the RANK-RANKL interaction. The former is expressed in osteoclast progenitors and the latter is expressed in many tissues including osteoclast inducible stroma cells. 1,25(OH) 2 D 3 and DEX induce expression of RANKL whereas they inhibit the expression of OPG/OCIF in stroma cells. The balance of the expression level between RANKL and OPG/OCIF is critical for osteoclast formation. In order to analyze the differentiation of osteoclasts in detail, an in vitro culture system is crucial. Although RANKL is a transmembrane protein, the extracellular region has sufficient activity to induce osteoclast differentiation.11,12 Thus osteoclasts are formed in the presence of macrophage colony stimulating factor (M-CSF) and a soluble form of RANK (srankl) instead of stroma cells. Cell Density and Multinucleation We have improved and established a more sophisticated culture system by isolating osteoclast precursor cells from murine bone marrow.17 To isolate osteoclast precursor cells, we studied expressions of c-fms, a receptor of M-CSF and RANK, since M-CSF plus

4 4 Miyamoto T and Suda T: Differentiation and function of osteoclasts Fig. 4 Stroma-free culture system. Osteoclast precursor cells were isolated from bone marrow mononuclear cells on the basis of the expression of c-fms and RANK (A), and cultured with M-CSF alone or M-CSF and srankl for 6 days in a stroma-free culture system (B). Cultured cells were subjected to TRAP-solution assay (C) or TRAP-staining (D). Note that the TRAP-activity was greatest in c-fms þ RANK cells cultured with M-CSF and srankl (C). Large numbers of multinuclear TRAP-positive cells were observed among c-fms þ RANK cells (Da), while only mononuclear TRAP-positive cells developed from c-fms þ RANK þ cells (Db), and c-fms RANK þ cells did not proliferate in this culture (Dc). Bar ¼ 100 mm. RANKL induce osteoclast differentiation in a stroma cell-free culture in vitro. Bone marrow mononuclear cells were subdivided into three fractions by cell surface markers (c-fms þ RANK, c-fms þ RANK þ, c- Fms RANK þ ) (Fig. 4). Each fraction was sorted by FACS and cultured in the presence of M-CSF and RANKL for 6 days and TRAP staining was performed to analyze the osteoclastogenesis. Multinuclear TRAPpositive osteoclasts were contained in c-fms þ RANK cells, while only mononuclear TRAP-positive cells were detected in c-fms þ RANK þ cellsandnotrappositive cells were seen in c-fms RANK þ cells (Fig. 4). The difference in multinuclear formation between c- Fms þ RANK and c-fms þ RANK þ is caused by the cell density, and the difference of cell density results from the proliferating activity between these fractions.17 c- Fms þ RANK fraction showed colony-forming activity in the presence of M-CSF alone or M-CSF and srankl, while c-fms þ RANK þ did not. Multinuclear osteoclasts were formed from c-fms þ RANK þ cells when ten times more cells were cultured suggesting adequate cell density is required to fuse and form multinuclear osteoclasts.17 From the results above, c-fms þ RANK cells have the highest multinuclear osteoclast formation activity, however, these cells do not express RANK. We have shown that M-CSF induced RANK expression in the c-fms þ RANK cells first, and then RANKL induced the differentiation signal (Fig. 5).18 In the absence of RANKL, these cells differentiated into macrophages through a default pathway. Anchorage-dependent Osteoclastogenesis Osteoclast formation is restricted to the bone surface, suggesting that bone provided a suitable environment for osteoclastogenesis. Although the expression of RANKL is relatively limited in osteoblasts, it is also expressed in other cells and tissues, such as T cells and spleen cells. We therefore studied the effect of adherent or non-adherent culture conditions on osteoclast proliferation and differentiation. We prepared the adher-

5 Keio J Med 2003; 52 (1): Fig. 5 Differentiation model of osteoclasts, macrophages and dendritic cells from common precursor cells. A determination of osteoclast and dendritic cell differentiation from common precursor cells is shown. ent condition in plastic culture dishes with a liquid medium to permit cell adhesion, whereas the non-adherent condition was prepared using a methylcellulose semisolid culture medium to inhibit cell adhesion to the substrate. Under adherent condition, osteoclastogenesis with multinucleation was induced.17 However, under non-adherent condition, osteoclast formation was reduced to approximately 30% compared with the adherent condition and no multinuclear osteoclasts were formed.17 The reduced differentiation activity to osteoclasts seen under non-adherent condition was recovered, when the cells were transferred to the adherent condition, indicating that osteoclasts progenitors can differentiate into osteoclast through adhesion to the matrix. Alternatively, osteoclastogenesis was stimulated by oligomerized RANKL under non-adherent conditions. RANKL is a member of the TNFa superfamily known to act as a trimer form. Actually, the trimerized RANKL called leucin-zipper RANKL (LZ-RANKL) effectively induced osteoclastogenesis compared with the monomer form even under non-adherent conditions. Thus, adhesion might be required to transduce the efficient RANK signaling for osteoclastogenesis. On the other hand, multinucleation was not fully induced by LZ-RANKL. This suggests that the polarity of cell adhesion is important for cell fusion. Anchoragedependent osteoclastogenesis might be involved in the prevention of ectopic osteoclast formation. Dendritic Cell Differentiation from Osteoclast Precursor Cells Although the osteoclast differentiation is reduced under non-adherent conditions, it is not completely inhibited, suggesting that some other osteoclastogenesis inhibitory system might be involved in tissues other than bone. Lymphoid tissues including the spleen and lymph nodes express the molecules for osteoclastogenesis in vitro such as M-CSF and RANKL. However, no osteoclast formation is induced in these tissues. We analyzed the effect of GM-CSF expressed in lymphoid tissues on osteoclast differentiation.5 Osteoclastogenesis was completely inhibited by GM-CSF at an early stage of differentiation in a dose-dependent manner, and cells were committed to a dendritc cell lineage, judging from the fact that CD11c and DEC205 were expressed by GM-CSF. Therefore, osteoclast precursor cells have the potency to differentiate into osteoclasts, macrophages and dendritic cells (Fig. 5). GM-CSF did not inhibit expression of c-fms and RANK, suggesting that the signals of M-CSF and RANKL might be transduced even in the presence of

6 6 Miyamoto T and Suda T: Differentiation and function of osteoclasts GM-CSF. However, among the intracellular molecules critical for osteoclast differentiation, c-fos expression is strongly inhibited by GM-CSF. Overexpression of c-fos overcomes the inhibitory effect of GM-CSF on osteoclastogenesis. It is suggested that c-fos suppression is an important event for inhibition of osteoclast differentiation. In dendritic cell differentiation, GM-CSF induces RANK expression in common precursor cells, and RANKL induces the terminal differentiation. On the contrary to osteoclastogenesis, dendritic cell differentiation is inhibited by M-CSF or c-fos expression.17 Most recently, interferon-beta (IFN-b) was reported to inhibit osteoclastogenesis in a RANKL-induced c-fos dependent manner, which is a different mechanism from viral infection.19 RANKL induces the c-fos expression in osteoclast precursor cells, and c-fos induces IFN-b which in turn inhibits the c-fos expression in a negative-feedback regulation. Thus various molecules effectively affect the osteoclast/dendritic cell/macrophage common precursor cells and regulate their differentiation. Malignant Tumor Cells Induce Osteoclasts via RANKL In adult T-cell leukemia (ATL) patients, hypercalcemia is induced concomitantly with an increased number of osteoclasts and accelerated bone resorption. Several cytokines, such as PTHrP, IL-1 and TGFbeta are implicated in ATL associated hypercalcemia. These cytokines indirectly induce osteoclastogenesis via stroma cells. However, from the aspect that increased osteoclastogenesis is mainly observed in the local area of tumor invasion, we speculated that ATL cells might express a membrane associated protein and directly induce osteoclastogenesis. Among the molecules, we studied the expression of RANKL. ATL cells from patients with hypercalcemia induced osteoclast differentiation, whereas ATL cells from non-hypercalcemic patients did not.20 ATL cells from hypercalcemic patients expressed high levels of RANKL, whereas cells from normocalcemic patients did not. A soluble receptor for RANKL inhibited osteoclastogenesis, indicating that RANKL expressed by ATL directly induced osteoclast differentiation. As observed in ATL, other bone metastatic tumors such as breast cancer and multiple myeloma cells express RANKL, and induce osteoclastogenesis. These lines of evidence indicate that pathological bone fractures and hypercalcemia caused by bone metastasis are induced by osteoclasts activated by tumor cells. These fractures sometime cause paralysis, and the presence of hypercalcemia is associated with a poor prognosis. The regulation of osteoclastogenesis in the metastatic bone tumor is an important problem when trying to maintain the patient s activities of daily living. Recently, bisphosphonates have been used therapeutically to prevent osteoclast-induced bone fractures and hypercalcemia. Future Aspects There are several diseases which cause bone destruction such as osteoporosis, bone metastases and bone marrow invasion from malignant tumors. If we are able to regulate osteoclastogenesis, bone destruction might be prevented and the quality of life of the patient is improved. We intend to regulate bone diseases, and now we are trying to isolate osteoclast-specific molecules by the DNA subtraction method between osteoclasts and macrophages derived from common precursor cells. Since these two types of cells express quite similar molecules, osteoclast-specific molecules have been successfully identified. We have already cloned several molecules and investigated their function on osteoclastogenesis. These osteoclast-specific molecules are strong candidates to target the regulation of osteoclast function. In the near future, we will hopefully be able to regulate osteoclast activity to prevent bone destruction by osteoclasts. References 1. Pittenger MF, Mackay AM, Beck SC, Jaiswal RK, Douglas R, Mosca JD, Moorman MA, Simonetti DW, Craig S, Marshak DR: Multilineage potential of adult human mesenchymal stem cells. Science 1999; 284: Hiraki Y, Inoue H, Iyama K, Kamizono A, Ochiai M, Shukunami C, Iijima S, Suzuki F, Kondo J: Identification of chondromodulin I as a novel endothelial cell growth inhibitor. Purification and its localization in the avascular zone of epiphyseal cartilage. J Biol Chem 1997; 272: Nesbitt SA, Horton MA: Trafficking of matrix collagens through bone-resorbing osteoclasts. Science 1997; 276: Ash P, Loutit JF, Townsend KM: Osteoclasts derived from haematopoietic stem cells. Nature 1980; 283: Miyamoto T, Ohneda O, Arai F, Iwamoto K, Okada S, Takagi K, Anderson DM, Suda T: Bifurcation of osteoclasts and dendritic cells from common progenitors. Blood 2001; 98: Akagawa KS, Takasuka N, Nozaki Y, Komuro I, Azuma M, Ueda M, Naito M, Takahashi K: Generation of CD1þRelBþ dendritic cells and tartrate-resistant acid phosphatase-positive osteoclast-like multinucleated giant cells from human monocytes. Blood 1996; 88: Baron R, Neff L, Tran Van P, Nefussi JR, Vignery A: Kinetic and cytochemical identification of osteoclast precursors and their differentiation into multinucleated osteoclasts. Am J Pathol 1986; 122: Udagawa N, Takahashi N, Akatsu T, Sasaki T, Yamaguchi A, Kodama H, Martin TJ, Suda T: The bone marrow-derived stromal cell lines MC3T3-G2/PA6 and ST2 support osteoclast-like cell differentiation in cocultures with mouse spleen cells. Endocrinology 1989; 125: Takahashi N, Akatsu T, Udagawa N, Sasaki T, Yamaguchi A, Moseley JM, Martin TJ, Suda T: Osteoblastic cells are involved

7 Keio J Med 2003; 52 (1): in osteoclast formation. Endocrinology 1988; 123: Anderson DM, Maraskovsky E, Billingsley WL, Dougall WC, Tometsko ME, Roux ER, Teepe MC, DuBose RF, Cosman D, Galibert L: A homologue of the TNF receptor and its ligand enhance T-cell growth and dendritic-cell function. Nature 1997; 390: Yasuda H, Shima N, Nakagawa N, Yamaguchi K, Kinosaki M, Mochizuki S, Tomoyasu A, Yano K, Goto M, Murakami A, et al: Osteoclast differentiation factor is a ligand for osteoprotegerin/ osteoclastogenesis-inhibitory factor and is identical to TRANCE/RANKL. Proc Natl Acad Sci USA 1998; 95: Lacey DL, Timms E, Tan HL, Kelley MJ, Dunstan CR, Burgess T, Elliott R, Colombero A, Elliott G, Scully S, et al: Osteoprotegerin ligand is a cytokine that regulates osteoclast differentiation and activation. Cell 1998; 93: Wong BR, Rho J, Arron J, Robinson E, Orlinick J, Chao M, Kalachikov S, Cayani E, Bartlett FS 3rd, Frankel WN, et al: TRANCE is a novel ligand of the tumor necrosis factor receptor family that activates c-jun N-terminal kinase in T cells. J Biol Chem 1997; 272: Wong BR, Josien R, Lee SY, Sauter B, Li HL, Steinman RM, Choi Y: TRANCE (tumor necrosis factor [TNF]-related activation-induced cytokine), a new TNF family member predominantly expressed in T cells, is a dendritic cell-specific survival factor. J Exp Med 1997; 186: Simonet WS, Lacey DL, Dunstan CR, Kelley M, Chang MS, Luthy R, Nguyen HQ, Wooden S, Bennett L, Boone T, et al: Osteoprotegerin: a novel secreted protein involved in the regulation of bone density. Cell 1997; 89: Tsuda E, Goto M, Mochizuki S, Yano K, Kobayashi F, Morinaga T, Higashio K: Isolation of a novel cytokine from human fibroblasts that specifically inhibits osteoclastogenesis. Biochem Biophys Res Commun 1997; 234: Miyamoto T, Arai F, Ohneda O, Takagi K, Anderson DM, Suda T: An adherent condition is required for formation of multinuclear osteoclasts in the presence of macrophage colonystimulating factor and receptor activator of nuclear factor kappa B ligand. Blood 2000; 96: Arai F, Miyamoto T, Ohneda O, Inada T, Sudo T, Brasel K, Miyata T, Anderson DM, Suda T: Commitment and differentiation of osteoclast precursor cells by the sequential expression of c-fms and receptor activator of nuclear factor kappab (RANK) receptors. J Exp Med 1999; 190: Takayanagi H, Kim S, Matsuo K, Suzuki H, Suzuki T, Sato K, Yokochi T, Oda H, Nakamura K, Ida N, et al: RANKL maintains bone homeostasis through c-fos-dependent induction of interferon-beta. Nature 2002; 416: Nosaka K, Miyamoto T, Sakai T, Mitsuya H, Suda T, Matsuoka M: Mechanism of hypercalcemia in adult T-cell leukemia: overexpression of receptor activator of nuclear factor kappab ligand on adult T-cell leukemia cells. Blood 2002; 99: