The effects of high dose of parathyroid hormone on fetal osteoclasts and their precursors in vivoAn ultrastructural-cytochemical study.код для вставкиСкачать
THE ANATOMICAL RECORD 243:421-429 (1995) The Effects of High Dose of Parathyroid Hormone on Fetal Osteoclasts and Their Precursors In Vivo: An Ultrastructural-Cytochemical Study HISAYUKI ISAKI AND HIDEYA HANAOKA Department of Orthopedic Surgery, Keio University School of Medicine, Tokyo, Japan ABSTRACT Background: It is not well known how the immediate precursors of osteoclast develop into osteoclasts in the fetus. This ultrastructural-cytochemical study was designed to clarify the formation process of the osteoclasts and their increased activities in the fetal mouse limb buds after administration of high dose parathyroid hormone (PTH). Methods: Twenty-four or forty-eight hours after the high doses of PTH were injected into amniotic fluid of the pregnant C,H mice, the femoral limb buds of embryos were dissected out. Tartrate-resistant acid phosphatase (TRAP) reactions were performed while preparing specimens for electron microscopy. Results: Both control and PTH-given preosteoclasts and osteoclasts exhibited TRAP-positivities in dense bodies and vesicles. As effects of PTH, a binucleated preosteoclast of tandem fashion was observed. More osteoclastic hyperactivities were observed in the diaphyseal bone marrow. An osteoclast with a large cytoplasm exhibited two sets of clear zones and ruffled borders. Some osteoclasts demonstrated prominent amoeboid figures, while other osteoclasts developed large cytoplasmic vacuoles, which contained pieces of calcified chondroid bars. Conclusions: Our results revealed the progression of maturation from young preosteoclasts to osteoclasts. An existence of a peculiar binucleated preosteoclasts suggested one of the processes for multinucleation of the osteoclast. Quite remarkable osteoclastic hyperactivities were obviously the effects of high dose PTH. Our results also indicated the endophagocytic ability of the osteoclast. How PTH affected the osteoclasts and their precursors in the diaphyseal bone marrow can be speculated. 0 1995 Wiley-Liss, Inc. Key words: C,H mouse, Enchondral ossification, Fetus, Limb bud, Osteoclast, Parathyroid hormone (PTH), Tartrate-resistant acid phosphatase (TRAP),Ultrastructure The hypothesis of osteoclast derivation from hematopoietic stem cell is predominant (Burger et al., 1987; Mundy and Roodman, 1987; Helfrich et al., 1989; Sasaki et al., 1989; Nijweide et al., 1990). However, it is not well known how the immediate precursors of osteoclasts develop into osteoclasts in periosteum of the uninvaded fetal limb buds where the osteoclasts are first formed (Hanaoka et al., 1989), although the process of their differentiation in the periosteum of fetal rat parietal bone was examined by Ejiri (1983).Several studies on the ultrastructural changes in osteoclasts after administration of parathyroid hormone (PTH) have revealed that PTH causes an increase in the number of osteoclasts and that it activates osteoclasts to resulting in an increase in various cytoplasmic organelles, as well as in the extent of the clear zone and ruffled border regions (Lucht and Maunbach, 1973; 0 1995 WILEY-LISS. INC Holtrop et al., 1974; King et al., 1978; Miller et al., 1984; Klaushofer et al., 1989; Barengolts et al., 1990). The following ultrastructural-cytochemical study was carried out on the formation process of the osteoclasts and their increased activities in the fetal limb buds after administration of high dose parathyroid hormone. MATERIALS AND METHODS Under ether drop anesthesia, laparotomy was performed on eight C,H mice that were in their 14th day Received May 19, 1995; accepted August 21, 1995. Address reprint requests to Dr. Hideya Hanaoka, Department of Orthopedic Surgery, Keio University School of Medicine, 35 Shinanomachi, Shinjuku-ku, Tokyo, 160, Japan. 422 H. ISAKI AND H. HANAOKA Fig, 1, A multinucleated osteoclast of control without administration of PTH is invading into the limb bud through the bone collar (B). This cell has standard osteoclastic features including a TRAP-positive vesicle (arrow). No vascular buds have invaded, yet. Note the hypertrophic chondrocyte (C) adjacent to the calcified chondroid bar (CB). Also, note the primitive cells (P) of oval shape, which have invaded into the limb bud. It remains uncertain whether these cells are the macrophages, the very young preosteoclasts, or any other cells (~3,400). of pregnancy. Ten units of PTH (Asahi Kasei Co., Tokyo) were injected through the wall of the uterus into the amniotic fluid to increase the number and activity of osteoclasts and preosteoclasts. Either 24 or 48 hours later, 60 femoral limb buds of 15- or 16-day-old embryos were dissected out together with their periosteum. Similarly, ten control femoral limb buds of the same age without administration of PTH were dissected out. Tartrate-resistant acid phosphatase (TRAP) reactions were performed while preparing specimens for electron microscopy. Femoral limb buds were fixed in 0.5% glutaraldehyde and 4% paraformaldehyde in 0.1 M cacodylate buffer at 4°C for 90 minutes. They were washed with 8% sucrose in 0.1 M cacodylate buffer a t 4°C overnight. After they were embedded in 5% agarose (Sigma Chemical Co., St. Louis, MO), sections of 50 to 80 bm were cut on a vibratome (Sorvall, TC 2 Type). To demonstrate TRAP activity, preincubation was performed with 50 mM potassium sodium tartrate in 10 ml of 0.2 M acetate buffer (pH 5.5) a t 37°C for 2.5 hours, followed by incubation in a reaction buffer of 50 ml of 0.05 M acetate buffer (pH 5.0), 4 g of sucrose, 50 mg of lead nitrate, and 5 ml of 3% p glyc- erophosphate at 37°C for 30 minutes. Specimens were then postfixed in 1% osmium tetroxide in 0.1 M cacodylate buffer a t 4°C for 1 hour, dehydrated in a graded series of ethanol, and embedded in EPON 812. Thin sections were cut on a Porter-Blum 11-B ultratome and stained with lead citrate to increase contrast. They were examined on a Hitachi HU-12AS electron microscope. RESULTS Individual femoral limb buds appeared to differ slightly in the developmental stage, despite having the same fetal age. The bone collars of the limb buds were usually ossified, but the amount of ossification varied to some degree from limb bud to limb bud. In addition, osteoclasts and preosteoclasts of various developmental stages were observed even in serial sections from the same limb bud. The femoral limb buds of 15-day-old embryos were just beginning to be invaded by osteoclasts, but were not yet invaded by vascular buds, so that no bone marrow had been formed. Bone marrow had formed in the diaphysis of the femoral limb buds of the 16-day-old embryos, and osteoclasts, endothelial EFFECTS OF HIGH DOSE PTH ON OSTEOCLASTS 423 Fig. 2. An apparently mononuclear control osteoclast, identified by the ruffled border facing the bone collar, is of round shape, and abundant of mitochondria ( x 5,200). Fig. 3. A control preosteoclast in the cambium layer shows TRAP-positivedense bodies and vesicles (arrows), despite less osteoclastic features ( x 6,200). cells, erythrocytes, monocytic cells, some unidentifiable primitive cells, and a few possible preosteoblasts, but no mature osteoblasts were observed there a t this stage. Figure 1demonstrated a multinucleated osteoclast of control without administration of PTH, which was just invading into the limb bud. An apparently-mononuclear control osteoclast, identified by the ruffled border 424 H. ISAKI AND H. HANAOKA Fig. 4. A binucleated osteoclast situated outside the bone collar. Note the abundant mitochondria, TRAP-positive (arrows) dense bodies, and vesicles and RER with narrow canaliculi in the cytoplasm ( x 5,000). facing the bone collar, was of round shape, and abundant of mitochondria (Fig. 2). A control preosteoclast in the cambium layer showed TRAP-positive dense bodies and vesicles, despite less osteoclastic features (Fig. 3). The following were the observations in the specimens of PTH administration. Number of preosteoclasts and osteoclasts increased and various unusual types of these cells were observed. There were binucleated osteoclasts which were TRAP-positive in both dense bodies and vesicles (Fig. 4). In addition to standard osteoclastic features including mitochondria and dense bodies, some osteoclasts demonstrated relatively abundant rough endoplasmic reticulum (RER), whose cisternae showed narrower canaliculi than usual seen in the postnatal osteoclasts (Fig. 4). Similarly, preosteoclasts were identified by TRAP-positive reactions in both dense bodies and vesicles, and by their round or spindle shape and a n abundance of mitochondria (Fig. 5). Some also showed RER with narrower sternal canaliculi than usual, like those of the osteoclasts (Fig. 5). We also observed a peculiar binucleated preosteoclast, which appeared to have been formed by the fusion of two preosteoclasts in a tandem fashion (Fig. 6). A series of maturation was observed among these preosteoclasts. The younger they were, the lighter their nuclei and cytoplasm, the greater the nucleus/ cytoplasm (N/C) ratio, and the fewer the organelle in the cytoplasm. The youngest osteoclast precursor we were able to identify was a preosteoclast with a high NIC ratio, and a light nucleus and cytoplasm (Fig. 7). Although it showed few organelles in its cytoplasm, this cell showed TRAP-positive dense bodies (Fig. 7, Inset). Such preosteoclasts (Figs. 5-7) were scattered among the osteoblasts and preosteoblasts of the cambium layer, and some of them were adjacent to the bone collar. Osteoclasts in the diaphyseal bone marrow were more hyperactive than those in the periosteum. One apparently mononuclear osteoclast had developed a large cytoplasm, so that it had a goldfish appearance (Fig. 8). And this cell had a clear zone with polar distribution. Another osteoclast with large cytoplasm exhibited two sets of clear zones and ruffled borders which were resorbing the calcified chondroid bars (Fig. 9). Some osteoclasts demonstrated prominent amoeboid figures (Fig. 10). Other osteoclasts exhibited many large cytoplasmic vacuoles, possibly d.erived from lysosomes, which contained pieces of calcified chondroid bars and other piece of possible autophagozomes (Fig. 9). In many instances, the mitochondria were enlarged and vacuolated. No mature osteoblasts were observed in the diaphyseal bone marrow as described before. TRAP-positivities were less occasionally observed in the osteoclastic cells in the diaphyseal bone marrow than in the periosteum. DISCUSSION In our experimental study, high dose PTH caused the appearance of more abundant preosteoclasts, including very young preosteoclasts and a n unusual type of binucleated preosteoclast as well a s apparently mononu- EFFECTS OF HIGH DOSE PTH ON OSTEOCLASTS 425 Fig. 5. A mononuclear preosteoclast situated outside the bone collar. Note the mitochondria, TRAP-positive (arrow) dense bodies and vesicles, and RER with narrow canaliculi ( x 4,800). clear osteoclasts, compared with those in the controls. These preosteoclasts and osteoclasts exhibited not only changes in the cytoplasmic organelles, but also far more remarkable morphologic changes than those in our controls or in any other prior report (Lucht and Maunbach, 1973; Holtrop et al., 1974; King et al., 1978; Klaushofer et al., 1989; Barengolts et al., 1990). Since tartrate-resistant acid phosphatase was found to be more specific to osteoclasts than acid phosphatase (Minkin, 1982), ultrastructural histochemistry using TRAP reaction have been performed on osteoclasts by a few researchers (Clark et al., 1989; Sasaki et al., 1989). Using ultrastructural staining by the TRAP reaction and other ultrastructural features of osteoclasts and preosteoclasts, we sought osteoclast precursors in the cambium layer of uninvaded limb buds where osteoclasts are first formed, and where no hematopoiesis occurs. In such limb buds we observed a gradual preosteoclast maturation series, with the youngest osteoclast precursor cell showing a high N/C ratio, a light nucleus and cytoplasm, and TRAP-positive dense bodies, most of which were lysosomes. However, we were unable to identify any younger precursors of osteoclasts than this stage. The appearance of a binucleated preosteoclast may indicate the strong effect of PTH to cause cell fusion. The preosteoclasts apparently fused without first developing into mature osteoclasts. Hanaoka (1979) reported that a n osteoclast fused with a preosteoblastic cell and concluded that the osteoclasts increased the number of their nuclei by the fusion of osteoblastic and osteoclastic cells or by the fusion between osteoclastic cells. Ejiri (1983) observed a flattened binucleated cell, one half of which was rich in mitochondria, a n osteoclastic feature, and another half of which had welldeveloped RER, a n osteoblastic feature. He presumed that this cell had been formed by this type of cell fusion. These findings, including ours, indicate the possible steps in the process of multinucleation of the preosteoclasts and osteoclasts, although the main process of multinucleation should be the cell fusion between osteoclastic cells, as reported by Fukushima et al. (1991). The osteoclasts in the diaphyseal bone marrow became more hyperactive. We observed that a n osteoclast exhibited two sets of clear zones and ruffled borders in the same cytoplasm. This phenomenon has been suggested by a scanning electron microscopic study of Kanehisa and Heersche (19881,but has never been confirmed by transmission electron microscopy. It is, in truth, necessary to prove by serial sections whether or not i t might be a transverse section of a single set of clear zone and ruffled border. However, as seen in Figure 11, pieces of calcified chondroid bars are loosely scattered in the diaphyseal bone marrow. Therefore, it is more likely that the osteoclast has developed two sets of clear zone and ruffled borders to absorb the separate pieces of calcified chondroid bars which are situated far away from each other. Quite bizzare amoeboid figures of osteoclasts were reported in a cultured osteoclast (Zambonin-Zallone and Teti, 19911, but have never been reported in vivo. The existence of pieces of calcified chondroid bar in large vacuoles indicates that they have been endocy- 426 H. ISAKI AND H. HANAOKA Fig. 6. A peculiar binucleated preosteoclast appears to have been formed by the fusion of two preosteoclasts in a tandem fashion. It has TRAP-positive vesicles (arrow). The majority of the cells adjacent to bone collar are either osteoblasts or preosteoblasts, having well-developed RER. Note the fibroblasts in the fibrous layer. A mononuclear cell in the left-hand side, having features similar to those of the binucleated preosteoclast, might be a preosteoclast, however, there is no definite evidence to prove it, because it shows no TRAP-positivities ( x 4,000). Fig. 7. The youngest preosteoclast type, with a high NIC ratio and a light nucleus ( x 6,800). Inset: Higher view of a TRAP-positive dense body (arrow) ( x 20,000). EFFECTS OF HIGH DOSE PTH ON OSTEOCLASTS 427 Fig, 8.An apparently mononuclear osteoclast with a goldfish appearance. Note TRAP-positive dense bodies (arrow). This cell has a clear zone with polar distribution. Monocytic cells are also seen ( X 4,000). Fig. 9. An osteoclast with a large cytoplasm exhibiting two sets of clear zones and ruffled borders is resorbing the calcified chondroid bars. No TRAP-positivities are observed in this cell ( X 4,000). Fig. 10. An osteoclast demonstrating a prominent amoeboid shape. The cytoplasm is extended quite remarkable, and a thick clear zone and ruffled border facing the calcified chondroid bar are observed at the end of the cytoplasm ( X 2,600). Fig. 11. An osteoclast demonstrating many large cytoplasmic vacuoles which contain a piece of calcified chondroid bar (arrow) and other small piece of possible autophagozomes. Unidentifiable primitive cells are also seen ( x 4,000). EFFECTS OF HIGH DOSE PTH ON OSTEOCLASTS tosed prior to phagocytosis, a s noted for osteocytes in the literature (Soskolne, 1978; Elmardi et al., 1990). In the diaphyseal bone marrow of the limb buds of 16-day-old fetal mice, calcified chondroid bars were being resorbed to form the bone marrow space and no bone formation took place. This is the reason why no mature osteoblasts were observed there in our study, while remarkable hyperactivity of the osteoclasts affected by PTH were observed. It is believed that PTH affects the osteoclasts indirectly through the osteoblasts, since the former have been believed not to express the PTH receptor while the latter have been shown to express i t (McSheehy and Chambers, 1986). How PTH affected the osteoclasts in the diaphyseal bone marrow where no mature osteoblasts were present in our study is not certain. One possibility is that the PTH affected the osteoclasts and their precursors through the osteoblasts while the former had been in the periosteum before their migration into the bone marrow. Another is that PTH affected the osteoclasts and their precursors in the bone marrow directly, a s the osteoclasts have been reported to have the PTH receptor (Teti et al., 1991; Agarwala and Gay, 1992). The third is that PTH affected the osteoclasts and their precursors either through the possible preosteoblasts in the bone marrow or through the parathyroid target (PT) cells termed by Rouleau et al. (1990). Such PT cells could exist in the diaphyseal bone marrow in our study as well as in the metaphyseal bone marrow in their study, although we could not identify if they were present in our study. ACKNOWLEDGMENTS This work was supported by a grant-in-aid from J a pan Orthopaedics and Traumatology Foundation, Inc. (JOTF), grant no. 0039 LITERATURE CITED Agarwala, N. and C.V. Gay 1992 Specific binding of parathyroid hormone to living osteoclasts. J . Bone Min. Res., 7.531-539. Barengolts, E., R. Buschmann, D.H. Shevrin, E.C. Abramson, and S.C. Kukreja 1990 Effects of hypercalcemia-producing tumor extract and parathyroid hormone on osteoclast ultrastructure. Acta Anat., 137:160-164. Burger, E.H.. E.P. van de Wijngaert, M.C. Tas, and J.W.M. van der Meer 1987 Fetal bone condl'tioned medium stimulates osteoclast precursor cell growth but has no effect on macrophages. In: Calcium Regulation and Bone Metabolism: Basic and Clinical Aspects, Vol. 9. Excerpta Medica, Amsterdam, pp. 308-313. Clark, S.A., W.W. Ambrose, T.R. Anderson, R.S. Terrell, and S.A. Toverud 1989 Ultrastructural localization of tartrate-resistant, purple acid phosphatase in rat osteoclasts by histochemistry and immunocytochemistry. J. Bone Min. Res., 4:399-405. Ejiri, S. 1983 The pre-osteoclast and its cytodifferentiation into the osteoclast: Ultrastructural and histochemical studies of fetal parietal bone. Arch. Histol. Jpn., 46r533-557. 429 Elmardi, A.S., M.V. Katchburian, and E. Katchburian 1990 Electron microscopy of developing calvaria reveals images that suggest that osteoclasts engulf and destroy osteocytes during bone resorption. Calcif. Tissue Int., 46r239-245. Fukushima, O., P.J. Bekker, and C.V. Gay 1991 Characterization of the functional stages of osteoclasts by enzyme histochemistry and electron microscopy. Anat. Rec., 231:298-315. Hanaoka, H. 1979 The origin of the osteoclast. Clin. Orthop., 145: 232-236. Hanaoka, H., H. Yabe, and H. Bun 1989 The origin of the osteoclast. Clin. Orthop., 239t286-298. Helfrich, M.H., R.H.P. Mieremet. and C.W. Thesingh 1989 Osteoclast formation in vitro from progenitor cells present in the adult mouse circulation. J . Bone Min. Res., 4t325-334. Holtrop, M.E., L.G. Raisz, and H.A. Simmons 1974 The effect of parathyroid hormone, colchicine and calcitonin on the ultrastructure and the activity of osteoclasts in organ culture. J. Cell Biol., 60: 346-355. Kanehisa, J. and J.N.M. Heersche 1988 Osteoclast bone resorption: In vitro analysis of the rate of resorption and migration of individual osteoclast. Bone, 9t73-79. King, G.J., M.E. Holtrop, and L.G. Raisz 1978 The relationship of ultrastructural changes in osteoclasts to resorption in bone cultures stimulated with parathyroid hormone. Metab. Bone Dis. Relat. Res., 1:67-70. Klaushofer, K., H. Horander, 0. Hoffmann, E. Czenvenka, U. Koning, K. Koller, and M. Peterlik 1989 Interferon y and calcitonin induce differential changes in cellular kinetics and morphology of osteoclasts in cultured neonatal mouse calvaria. J. Bone Min. Res., 4.585-606. Lucht, U. and A.B. Maunbach 1973 Effect of parathyroid hormone on osteoclast in vivo: An ultrastructural and histochemical study. Z. Zellforsch., 141:529-544. McSheehy, P.M. and T.J. Chambers 1986 Osteoclastic cells mediate osteoclastic responsiveness to parathyroid hormone. Endocrinology, 118t824-828. Miller, S.C., B.M. Bowman, and R.L. Myers 1984 Morphological and ultrastructural aspects of the activation of avian medullary bone osteoclasts by parathyroid hormone. Anat. Rec., 208:223-231. Minkin, C. 1982 Bone acid phosphatase: Tartrate-resistant acid phosphatase as a marker of osteoclast function. Calcif. Tissue Int., 34:285-290. Mundy, G.R. and D. Roodman 1987 Osteoclast ontogeny and function. In: Bone and Mineral Research, Vol. 5. Elsevier, Amsterdam, pp. 209-278. Nijweide, P.J., C.E. Hagenaars, W.E. Modderman, and R.J.P. Mulder 1990 The cell lineage of the osteoclast. In: Calcium Regulation and Bone Metabolism: Basic and Clinical Aspects, Vol. 10. Excerpta Medica, Amsterdam, pp. 415-424. Rouleau, M.F., J . Mitchell, and D. Goltzman 1990 Characterization of the major parathyroid hormone target cell in the endosteal metaphysis of rat long bones. J. Bone Min. Res., 5t1043-1053. Sasaki, T., N. Takahashi, S. Higashi, and T. Suda 1989 Multinucleated cells formed on calcified dentine from mouse bone marrow cells treated with lu,25-dihydroxyvitamin D, have ruffled borders and resorb dentine. Anat. Rec., 224t379-391. Soskolne, W.A. 1978 Phagocytosis of osteocytes by osteoclasts in femoral of two week-old rabbits. Cell Tissue Res., 195t557-564. Teti, A,, R. Rizzoli, and A. Zambonin-Zallone 1991 Parathyroid hormone binding to cultured avian osteoclasts. Biochem. Biophy. Res. Commun., 174:1217-1222. Zambonin-Zallone, A,, and A. Teti 1991 Isolation and behaviour of cultured osteoclast. In: Bone, Vol. 2. CRC Press, Boca Raton, pp. 87-118.