Demonstration of amelogenin in the enamel-free cusps of rat molar tooth germsImmunofluorescent and immunoelectron microscopic studies.код для вставкиСкачать
THE ANATOMICAL RECORD 233:588-596 (1992) Demonstration of Amelogenin in the Enamel-Free Cusps of Rat Molar Tooth Germs: lmmunof luorescent and lmmunoelectron Microscopic Studies TETSUICHIRO INAI, KENGO NAGATA, TOSHIO KUKITA, AND KOJIRO KURISU Second Department of Conservative Dentistry (T.I.) and Second Department of Anatomy (K.N., T.K., K.K.), Faculty of Dentistry, Kyushu University,Fukuoka, Japan ABSTRACT The enamel-free cusps of 1-4-day-old r a t mandibular first molars were investigated using the monoclonal antibody En3 against rat amelogenin a t light and electron microscopic levels in order to clarify whether the enamel-free cusp is virtually devoid of enamel. At 1 day after birth, there were presecretory ameloblast-like cells (PALCs), which were short and were not polarized, a t the cusp tips. They were close to the outer enamel epithelium. Hematoxylin positive enamel matrix was not distinctly observed in the enamel-free cusp by light microscopy, but almost continuous immunofluorescence for amelogenin was detected a t the interface between PALCs and dentin. The penetration of immunopositive material toward the dental pulp was also observed in the enamel-free cusp. At 4 day after birth, both in the frontal section and in the horizontal section, almost continuous immunof luorescence was recognized a t the interface between PALCs and dentin in the enamel-free cusp. The penetration of amelogenin toward the dental pulp was not seen in the enamel-free cusp. By immunoelectron microscopy, immunolabelling was recognized in the Golgi apparatus of PALCs, in a layer of amorphous material a t the interface between PALCs and dentin, and in stippled material-like substance in the intercellular space between PALCs. Although no basement membrane was observed beneath PALCs, they did not have Tomes’ processes. These investigations suggest that PALCs in the enamel-free cusp differentiate into the secretory cells and that they can synthesize and secrete the amorphous material containing amelogenin a t the interface between PALCs and dentin. The penetration of amelogenin toward the dental pulp might play a role in the interaction between PALCs and odontoblasts in the enamel-free cusp and/or the initiation of mineralization of predentin. o 1992 Wiley-Liss, ~ n c . Tooth crown formation results from reciprocal and interdependent epithelial-mesenchymal interactions (Kollar, 1972, 1983; Ruch and Karcher-Dijuricic, 1975; Slavkin et al., 1984; Ruch, 1985). The cells of the inner enamel epithelium of the enamel organ differentiate into secretory ameloblasts that produce enamel, whereas ectomesenchymal cells adjacent to the epithelium differentiate into odontoblasts that produce predentin and dentin. However, there are non-enamel bearing regions at the occlusal cusps of molar teeth of the rat (Addison and Appleton, 1921; Mellanby, 1939; Lefkowiz et al., 1953) and of the mouse (Gaunt, 1956; Cohn, 1957). It has been reported that although mesenchymal cells differentiate into odontoblasts and consequently normal dentinogenesis occurs in the enamelfree cusp, the cells of the inner enamel epithelium in the enamel-free cusp fail to differentiate into secretory ameloblasts and to secrete enamel matrix. However, some workers have reported that the cells of the inner enamel epithelium in the enamel-free cusp differentiate into secretory cells which can secrete enamel ma0 1992 WILEY-LISS. INC. trix (Johannessen, 1961; Sutcliffe and Owens, 1980, 1981; Diab and Zaki, 1985; Nakamura et al., 1987; Sakakura et al., 1989). Nakamura et al. (1986) observed discrete immunoreactivity for polyclonal antibodies against mouse amelogenin at the interface between the cells of the inner enamel epithelium and adjacent dentin in the enamel-free cusps of mouse molars using indirect immunofluorescent microscopy. However, a precise immunohistochemical study in the enamel-free cusp using monoclonal antibodies against distinct enamel proteins at light and electron microscopic levels has not yet been reported. In the present study, we examined the enamel-free cusps of rat molar tooth germs using the monoclonal Received April 16, 1991; accepted January 8, 1992. Kojiro Kurisu’s present address is First Department of Oral Anatomy, Faculty of Dentistry, Osaka University, 1-8,Yamadaoka, Suita, Osaka 565, Japan. Address reprint requests there. Figs. 1-3 (legend on following page). 590 T. INAI ET AL antibody En3 (Inai et al., 1991) directed against rat amelogenin a t light and electron microscopic levels. MATERIALS AND METHODS We investigated the enamel-free cusps of rat mandibular first molars. The immunohistochemical localization of amelogenin in the enamel-free cusp was examined a t light and electron microscopic levels using the monoclonal antibody En3 (Inai et al., 1991) which recognizes rat amelogenin. IrnrnunofluorescenceProcedure The 1-4-day-old rats after birth were perfused with periodate-lysine-paraformaldehyde fixative (McLean and Nakane, 1974) through the left ventricle and the dissected mandibles were immersed in the same fixative for about 12 hours at 4°C. After decalcification by 10% ethylenediamine tetraacetic acid (EDTA) for 1 week at 4"C, the specimens were embedded in paraffin. Sections mounted on glass slides were incubated with the monoclonal antibody En3 for 1 hour a t room temperature. Control sections were treated with either normal mouse serum (Zymed Lab., Inc., South San Francisco, CA) or PBA (phosphate-buffered saline (PBS) containing 1%chick albumin) instead of En3. After repeated rinsing with PBS, the tissue sections were incubated with fluorescein isothiocyanate (FITC) labelled goat anti-mouse p-chain antibody (Zymed Lab., Inc.) for 30 minutes at room temperature. The entire procedure was carried out in a moist chamber to prevent the samples from drying. Finally the sections were washed extensively with PBS and sealed with PBS-buffered glycerol to which p-phenylenediamine had been added to decrease fluorescent fading as described by Johnson and Araujo (1981). The specimens were then examined with a Zeiss fluorescence microscope. After recording of the immunofluorescent reac- Figs. 1-4. Frontal and horizontal sections of mandibular first molars of 1-4-day-old rats. After recording of the immunofluorescent reaction using the monoclonal antibody En3, the same sections were stained with hematoxylin and eosin (H + E) for histological observations. Fig. 1. A micrograph of the frontal section of 1-day-old rat mandibular first molar stained with H + E. Ameloblasts in the cusp are close to the outer enamel epithelium (arrows). x 75. Fig. 2. Higher magnification of a n area in a square shown in Figure 1, stained with En3 (a) and H + E (b). The nuclei of most of ameloblasts in the cusp are migrating to the proximal cytoplasm and some of them are secreting hernatoxylin positive enamel matrix (arrowhead in b). Some ameloblasts (arrow in b) at the cusp tip which are close to the outer enamel epithelium have not been polarized. Almost continuous immunofluorescence for En3 is observed at the interface between ameloblasts and dentin. Note the staining for En3 in dentin, predentin, and odontoblast layer (arrows in a). A ameloblasts; 0: odontoblasts. X 270. Fig. 3. Micrographs of the frontal section of 4-day-old rat mandibular first molar stained with En3 (a) and H + E (b). No distinct enamel matrix is observed a t the interface between presecretory ameloblast-like cells (PALCs),which are short and not polarized, and dentin (D), but there are a few discrete thin layers stained with hernatoxylin between PALCs and dentin (arrows in b). Almost continuous immunofluorescence is recognized at the interface between PALCs and dentin, and it continues to enamel matrix (E) with imrnunoreactivity. x 320. tion using En3, the same sections were stained with hematoxylin and eosin (H + E) for histological observations. lrnrnunoelectron Microscopy The 4-day-old rats were perfused through the left ventricle with a mixture of 1%glutaraldehyde and 3% paraformaldehyde in 0.1 M cacodylate buffer containing 0.05% CaCl,, pH 7.4. The mandibles were dissected and immersed in the same fixative for about 6 hours a t 4°C. After decalcification in 10% EDTA for 1 week at 4"C, the mandibular first molars were cut with a Microslicer (Dosaka EM Co., Kyoto, Japan) into thick sections of 200-300 pm which were postfixed in osmium tetroxide reduced with potassium ferrocyanide (Karnovsky, 1971) for 2 hours at room temperature in the dark. The thick sections were washed in a sodium cacodylate buffer, dehydrated in a graded series of ethanol, and embedded in Epon 812. Ultrathin sections were mounted on nickel grids and floated on drops of a saturated aqueous solution of sodium metaperiodate for 1 hour and on PBL (0.1 M PBS containing 1%bovine serum albumin [BSAI and 0.1 M lysine-HC1) for 30 minutes. They were then treated with En3 antibody for 1 hour at room temperature. Control sections were treated with either normal mouse serum or PBL instead of En3. After repeated rinsing with PBS, the sections were incubated with colloidal gold labelled goat anti-mouse p-chain antibody. Colloidal gold labelling of goat anti-mouse p-chain antibody (Zymed Lab., Inc.) was carried out as described by Slot and Geuze (1984). The sections were then washed three times for 5 minutes each with PBS and finally again in distilled water. Throughout the entire procedure, care was taken to prevent the sections from drying by keeping them in a moist chamber. The sections were finally stained with uranyl acetate and lead citrate and then were examined with a JEOL 200CX electron microscope. Fig. 4. Micrographs of the horizontal section of 4-day-old rat mandibular first molar stained with En3 (a) and H E (b).A thin layer of hematoxylin positive enamel-like matrix (arrows in b), which continues to enamel matrix (E), is detected beneath presecretory ameloblast-like cells (PALCs), but there is a region where no enamel-like matrix is detected below PALCs (arrowhead in b). Continuous immunofluorescence is observed clearly between PALCs and dentin in a. x 250. + Fig. 5. Micrographs of control section, incubated with PBA instead of En3 (a)and stained with H + E (b),of 1-day-old rat mandibular first molar. Faint background staining is observed in predentin. A: ameloblasts; D: dentin; PD: predentin; 0: odontoblasts. x 270. Figs. 6-1 0. Imrnunoelectron micrographs stained with the monoclonal antibody En3. Rat mandibular first molars of day 4 were prepared and immunostained with post-embedding method as described in Materials and Methods. Fig. 6. Control section of enamel matrix (EM) and Tomes' process (TP) of secretory ameloblast in the area other than the enamel-free cusp incubated with PBL in place of En3. No specific labelling is observed. x 46,000. Fig. 7. Almost the same area as shown in Figure 6 stained with En3. Enamel matrix (EM) and secretory granule (arrow) in Tomes' process ITP) are labelled. x 46,000. Figs. 4-7. 592 T. INAI ET AL. RESULTS lmmunofluorescent Observations Figure 1 shows the frontal section of 1-day-old rat mandibular first molar. Ameloblasts at the cusp tip were close to the outer enamel epithelium. Higher magnification of the cusp of Figure 1 is shown in Figure 2. Most of ameloblasts in the cusp were columnar and polarized, and some of them were secreting a thin layer of hematoxylin positive enamel matrix (Fig. 2b). However, ameloblasts a t the cusp tip, those that were close to the outer enamel epithelium, were not polarized and the morphological features of them resembled that of presecretory ameloblasts (Fig. 2b). In this article, we call them “presecretory ameloblast-like cells (PALCs).”Odontoblasts in the cusp were well differentiated and produced predentin and dentin (Fig. 2b). Almost continuous immunofluorescence for the monoclonal antibody En3 was detected at the interface between ameloblasts including PALCs and dentin (Fig. 2a). The penetration of immunopositive material from ameloblasts including PALCs into the dental pulp was also observed (Fig. 2a). At day 4 after birth, the enamel-free cusps were easily detected both in the frontal section (Fig. 3) and in the horizontal section (Fig. 4). Ameloblasts in the area other than the enamel-free cusp were well differentiated and thick enamel was formed below them. However, ameloblasts in the enamel-free cusp were very similar t o PALCs of 1-day-old rat; namely, they were short and not polarized (Figs. 3b, 4b). In the enamelfree cusp hematoxylin positive enamel matrix-like material was not clearly detected, but in the marginal region of the enamel-free cusp a thin layer of enamellike matrix was recognized and it continued into the typical thick enamel matrix (Figs. 3b, 4b). Although distinct enamel matrix was not observed in the enamel-free cusps by H + E staining, almost continuous immunostaining for amelogenin was distinctly detected at the interface between PALCs and dentin in the enamel-free cusps both in the frontal section (Fig. 3a) and in the horizontal section (Fig. 4a). However, the penetration of immunoreactive material toward the dental pulp was not observed (Figs. 3a, 4a). Control sections stained using PBA in place of En3 showed faint background staining in predentin (Fig. 5a). enamel bearing region. The width of the layer varied and it was often incomplete as shown in Figure 9. No basement membrane was observed beneath PALCs (Figs. 8,9).The Golgi apparatus of PALCs (Fig. 10) and stippled material-like substance in the intercellular space between PALCs (Fig. 10, inset) showed immunolabelling for amelogenin. DISCUSSION Non-enamel bearing regions of molar occlusal CUSPS had been reported in the rat (Addison and Appleton, 1921; Mellanby, 1939; Lefkowiz et al., 1953) and in the mouse (Gaunt, 1956; Cohn, 1957). These investigators concluded that the cells of the inner enamel epithelium of non-enamel bearing regions did not differentiate into secretory ameloblasts, so enamel was not deposited on the dentin surface in the enamel-free cusp. However, Johannessen (1961) showed the presence of enamel on the enamel-free cusps of rat molars by microradiography. And Sutcliffe and Owens (1980) observed in their ultrastructural investigations that some patches of enamel-like matrix were deposited on the dentin surface in the enamel-free cusps of rat molars. Sutcliffe and Owens (1980, 1981) also indicated that the cells of the inner enamel epithelium in the enamel-free cusp secreted enamel-like matrix on the dentin surface and subsequently they resorbed the enamel-like matrix as well as the dentin surface just prior to tooth eruption, so the enamel-free cusp actually lacked enamel. Similarly, Diab and Zaki (1985) and Sakakura et al. (1989) reported the presence of an afibrillar layer on the dentin surface in the enamel-free cusp of mouse molar in light and electron microscopic studies. However, it is not possible to conclude only on the basis of its morphological feature that the afibrillar extracellular matrix in the enamel-free cusp is enamel. In the present immunohistochemical study using the monoclonal antibody En3 against rat amelogenin, we demonstrated that the Golgi apparatus of presecretory ameloblast-like cells (PALCs) in the enamel-free cusp showed positive staining and that the amorphous material on the dentin surface beneath PALCs showed labelling by immunoelectron microscopy. These observations suggest that PALCs in the enamel-free cusp differentiate into the cells which secrete the amorphous material containing amelogenin on the dentin surface in the enamel-free cusp. This suggestion is consistent with the previous report by Nakamura et al. lmmunoelectron Microscope Observations (1986) at the light microscopic level that there was Control sections incubated with PBL in place of En3 discrete immunostaining for mouse amelogenin at the showed negligible labelling (Fig. 6). interface between PALCs and adjacent dentin in the In a large portion of the crown of the mandibular enamel-free cusps of mouse molars. The amorphous first molar at day 4, the secretory ameloblasts with Tomes’ processes and enamel matrix with typical demineralized enamel structure were observed as shown in Figure 7. Enamel matrix showed heavy labelling for the monoclonal antibody En3. On the other hand, in Fig. 8 . The interface between presecretory ameloblast-like cells the enamel-free cusp, PALCs without Tomes’ process (PALCs) and dentin (D) in the enamel-free cusp. In spite of lack of processes in PALCs, a layer of amorphous material (AM) were observed instead of secretory ameloblasts (Figs. 8, Tomes’ which shows positive labelling for En3 a t the interface between 9). At the interface between PALCs and dentin there PALCs and dentin is observed. There is no basement membrane bewas a layer of amorphous material showing positive neath these cells. X62,OOO. Inset shows some patches of AM with labelling for En3 instead of typical enamel matrix (Fig. labelling for En3 in dentin in the enamel-free cusp. x 46,000. 8). Some patches of labelled material with various size Fig. 9. The interface between presecretory ameloblast-like cells and shape were observed in dentin or predentin matrix (PALCs) and dentin (D) in the enamel-free cusp where there is no in the enamel-free cusp (Fig. 8, inset) as well as in the En3-positive material. x 34,000. AMELOGENIN I N T H E ENAMEL-FREE CUSP Figs. 8 and 9. 593 594 T. INAI ET AL. Fig. 10. Golgi region of presecretory ameloblast-like cell in the enamel-free cusp. Note the labelling of condensing vacuoles indicating the synthesis of amelogenin in these cells. X 48,000. The presence of stippled material-like substance with labelling in the intercellular space between presecretory ameloblast-like cells is shown in inset. x 40,000. dense material which resembled the amorphous material in the enamel-free cusp was described at the dentino-enamel junction in papers by Warshawsky (1971) and Warshawsky and Vugman (1977). Similar material was decribed on the root-analogue surface of rabbit incisors, formed by Hertwig’s sheath and identified as enamel-like by Schonfeld and Slavkin (1977). Antibodies to amelogenin were localized between presecretory ameloblasts and dentin by Slavkin et al. (1988a) and by Nanci et al. (1989). These authors suggested that presecretory ameloblasts in the areas other than the enamel-free cusp also had the ability to synthesize and secrete the amorphous material containing amelogenin on the dentin surface. These previous studies as well as data presented here support the view that the material at the surface of dentin in the enamel-free cusp is a natural secretion product that precedes normal amelogenesis and is formed by the presecretory ameloblasts in preparation for enamel formation. The same factors that prevent enamel from forming on the surface of dentin in the roots might also prevent enamel formation on the enamel-free cusps. Although Sakakura et al. (1989)reported the differences between the afibrillar matrix in the enamel-free cusp and enamel in normal amelogenesis, the amorphous material in the present study is similar to enamel matrix because it contains amelogenin, which is the enamel matrix specific major protein, and it is the product derived from presecretory ameloblast-like cells in the enamel-free cusp, but not from odontoblasts. It has been unclear whether the amorphous material contains enamelin. The molecular constituents of the amorphous material need to be investigated. Recently, Inai et al. (1991) demonstrated using an immunohistochemical technique that presecretory ameloblasts secreted stippled material-like substance containing amelogenin into the predentin and this substance penetrated into the dental pulp in rat molar tooth germs. They hypothesized that the penetration of amelogenin might play a role in the interaction between presecretory ameloblasts and odontoblasts. If this hypothesis is true, the penetration of amelogenin should occur also in the enamel-free cusp because ectomesenchymal cells in the enamel-free cusp differentiate into odontoblasts and sequential dentinogenesis occurs normally. It was very easy to distinguish the non-enamel bearing region in the enamel-free cusp from the area of normal amelogenesis in the cusp of 4-day-old rat mandibular first molar because distinct thick enamel existed in the latter but no clear enamel was detected in the former by light microscopy. However, it was difficult to detect the non-enamel bearing region in the cusp of the 1-day-oldrat molar because clear hematoxylin positive enamel matrix had not been produced even in the area other than the non-enamel bearing region. In the present study, we regard “PALCs at the cusp tip of 1-day-old rat mandibular first molar” as ameloblasts in the future enamel-free cusp by the following two criteria. 1) The nuclei of most of amelo- 595 AMELOGENIN I N THE ENAMEL-FREE CUSP ACKNOWLEDGMENTS blasts in the cusp migrated to the proximal cytoplasm and they were elongating, but PALCs at the cusp tip This work was supported in part by a Grant for Scihad not polarized. 2) PALCs a t the cusp tip were close entific Research from the Japanese Ministry of Educato the outer enamel epithelium. Sutcliffe and Owens tion, Science and Culture (project 63771448). (1980) also reported that this proximity was observed at the cusp tips of rat mandibular first molars until LITERATURE CITED about 3 to 4 days after birth. We showed using l-dayAddison, W.H.F., and J.L. Appleton 1921 On the development of the old rat mandibular first molars that the penetration of ameloblasts of the albino rat, with special reference to the amelogenin from ameloblasts including PALCs in the enamel-free area. Anat. Rec., 21:43 (Abstract). future enamel-free cusp to the dental pulp also oc- Cohn, S.A. 1957 Development of the molar teeth in the albino mouse. Am. J. Anat., 101:295-319. curred before enamel formation. A thin layer of dentin M.M., and A.E. Zaki 1985 Ultrastructure and lysosomal cytowas formed in the cusp of the 1-day-oldrat molar. This Diab,chemistry of mouse molar enamel-free areas. J. Dent. Res., 64: penetration was not observed in the enamel-free cusp 324 (Abstract). of the 4-day-old rat molar, where thick dentin had been Gaunt, W.A. 1956 The development of enamel and dentin on the mo. lars of the mouse, with an account of the enamel-free areas. Acta formed by odontoblasts. But some patches of amor28:lll-134. phous material with labelling for amelogenin were rec- Inai,Anat., T., T. Kukita, Y. Ohsaki, K. Nagata, A. Kukita, and K. Kurisu ognized in the dentin in the enamel-free cusp by im1991 Immunohistochemical demonstration of amelogenin penemunoelectron microscopy. These patches with labelling tration toward the dental pulp in the early stages of ameloblast develonment in rat molar tooth eerms. Anat. Rec.. 229:259-270. in the dentin in the enamel-free cusp may imply that L.B. 1961 Presence of enamel-covered cusps in rat mothe penetration of amelogenin had occurred in the Johannessen, lars. Archs Oral Biol., 5:61-62. enamel-free cusp, because odontoblasts do not show Johnson, G.D., and G.M.C.N. Araujo 1981 A simple method of reducamelogenin gene expression (Snead et al., 1988).These ing the fading of immunofluorescence during microscopy. J . Immunol. Methods, 43:349-350. investigations indicate that the penetration of ameloE. 1971 Electron microscopy of the differentiating rat genin occurs in the enamel-free cusp as well as in the Kallenbach, incisor ameloblast. J . Ultrastruct. Res., 35.508-531. area of normal amelogenesis (Inai et al., 1991), and Karnovsky, M.J. 1971 Use of ferrocyanide reduced osmium tetroxide support the hypothesis that the penetration of ameloin electron microscop . Proceedings of the American Society on Cell Biology, New Orreans, Louisiana, 281:146 (Abstract). genin might play a role in the interaction between preE.J. 1972 Histogenetic aspects of dermal-epidermal interacsecretory ameloblasts including PALCs and odonto- Kollar, tions. In: Developmental Aspects of Oral Biology. H.C. Slavkin blasts. and L.C. Bavetta, eds. Academic Press, New York, pp. 126-149. However, the exact role of the penetration of amelo- Kollar, E.J. 1983 Epithelial-mesenchymal interactions in the mammalian integument: Tooth development as a model for instructive genin as well as the amorphous material in predentid induction. In: Epithelial-Mesenchymal Interactions in Developdentin is unknown. The amorphous material may play ment. R.H. Sawyer and J.F. Fallon, eds. Praeger, New York, pp. a role in the epithelial-mesenchymal interactions oc27-50. curring during odontogenesis and/or may participate in Lefkowitz, W., C.F. Bodecker, and D.F. Mardfin 1953 Odontogenesis of the rat molar: Prenatal stage. J . Dent. Res., 32:749-772. the initial mineralization of the dentin extracellular I.W., and P.K. Nakane 1974 Periodate-lysine-paraformalmatrix and the formation of the dentino-enamel junc- McLean, dehyde fixative, a new fixative for immunoelectron microscopy. J. tion (Slavkin et al., 1988a; Warshawsky, 1988; Nanci Histochem. Cytochem., 22:1077-1083. et al., 1989). The penetration of amelogenin first ap- Mellanby, H. 1939 The development of teeth in the albino rat. Br. Dent. J., 66:76-86. peared in differentiation I1 of Kallenbach (19711, in which the basement membrane between presecretory Nakamura, M., P. Bringas, and H.C. Slavkin 1986 Immunohistochemical comparisons of region-specific dental epithelial-derived ameloblasts and odontoblasts was intact and smooth, ECM proteins. Anat. Rec., 214:90A (Abstract). and almost disappeared a t initial enamel formation, Nakamura, M., P. Bringas, and H.C. Slavkin 1987 Position characand that during this period, odontoblasts endocytosed a teristics of “Intercellular communication” during odontogenetic epithelial differentiation. Anat. Rec., 218t96A (Abstract). part of the stippled material-like substance containing A,, J.P. Ahluwalia, J.R. Pompura, and C.E. Smith 1989 Bioamelogenin which penetrated from presecretory amelo- Nanci, synthesis and secretion of enamel proteins in the rat incisor. blasts into the dental pulp (Inai et al., 1991). In differAnat. Rec., 224:277-291. entiation 11-111, odontoblasts first showed positive im- Ruch, J.V. 1985 Epithelial-mesenchymal interactions in formation of mineralized tissues. In: The Chemistry and Biology of Mineralmunostaining with anti-mouse dentin phosphoprotein ized Tissues. W.T. Butler, ed. Ebsco Media, Inc., Birmingham, (DPP) polyclonal antibodies in their cytoplasms and AL, pp. 54-61. cell processes, and then DPP was also detected a t the Ruch, J.V., and V. Karcher-Dijuricic 1975 On odontogenic tissue inmineralization front (Slavkin et al., 198813). These interactions. In: Extracellular Matrix Influences on Gene Expression. H.C. Slavkin and R. Gruelich, eds. Academic Press, New vestigations indicate that the time of the expression York, pp. 549-554. and secretion of DPP by odontoblasts coincides with the Y., N. Fujiwara, and T. Nawa 1989 Epithelial cytodifferperiod of the penetration of amelogenin. Therefore, the Sakakura, entiation and extracellular matrix formation in enamel-free arpenetration of amelogenin might be related to the exeas of the occlusal cusp during development of mouse molars: pression and/or secretion of DPP by odontoblasts. Light and electron microscopic studies. Am. J. Anat., 184t287297. In conclusion, it is suggested that PALCs in the S.E., and H.C. Slavkin 1977 Demonstration of enamel maenamel-free cusp differentiate into secretory cells that Schonfeld, trix proteins on root-analogue surfaces of rabbit permanent inciproduce enamel-like matrix containing amelogenin on sor teeth. Calcif. Tiss. Res., 24:223-229. the dentin surface, so the enamel-free cusp does not Slavkin, H.C., C. Bessem, P. Bringas, M. Zeichner-David, A. Nanci, and M.L. Snead 1988a Sequential expression and differential lack an ameloblastic secretion product. The penetrafunction of multiple enamel proteins during fetal, neonatal, and tion of amelogenin toward the dental pulp in the early postnatal stages of mouse molar organogenesis. Differentienamel-free cusp might play a role in the epithelialation, 37:26-39. mesenchymal interactions and/or the initiation of min- Slavkin, H.C., M. MacDougall, M. Zeichner-David, P. Oliver, M. Nakamura, and M.L. Snead 1988b Molecular determinants of craeralization of predentin. ~~~ - 596 T. INAI ET AL. nial neural crest-derived odontogenic ectomesenchyme during dentinogenesis. Am. J. Med. Genet. Suppl., 4t7-22. Slavkin, H.C., M.L. Snead, M. Zeichner-David, P. Bringas, and G.L. Greenberg, 1984 Amelogenin gene expression during epithelialmesenchymal interactions. In: The Role of Extracellular Matrix in Development. R.L. Trelstad, ed. Alan R. Liss, Inc., New York, pp. 221-253. Slot, J.W., and H.J. Geuze 1984 Gold markers for single and double immunolabelling of ultrathin cryosections. In: Immunolabelling for Electron Microscopy. J.M. Polak and I.M. Varndell, eds. Elsevier Science Publishers, Amsterdam, pp. 129-142. Snead, M.L., W. Luo, E.C. Lau, and H.C. Slavkin 1988 Spatial- and temporal-restricted pattern for amelogenin gene expression during mouse molar tooth organogenesis. Development, 104:77-85. Sutcliffe, J.E., and P.D.A. Owens 1980 A light and scanning electron microscopic study of the development of enamel-free areas on the molar teeth of the rat. Archs Oral Biol., 25:263-268. Sutcliffe, J.E., and P.D.A. Owens 1981 Microscopy of resorption of enamel and dentine from the cusp tips of the molar teeth of rats. Archs Oral Biol., 26t29-39. Warshawsky, H. 1971 A light and electron microscopic study of the nearly mature enamel of rat incisors. Anat. Rec., 169:559-584. Warshawsky, H. 1988 Formation of enamel and dentin: A critical review. CRC Critical Reviews in Anatomy and Cell Biology, 1: 425-460. Warshawsky, H., and I. Vugman 1977 A comparison of the protein synthetic activity of presecretory and secretory ameloblasts in rat incisors. Anat. Rec., 188:143-172.