Translocation of enamel proteins from inner enamel epithelia to odontoblasts during mouse tooth development.код для вставкиСкачать
THE ANATOMICAL RECORD 238:383-396 (1994) Translocation of Enamel Proteins From Inner Enamel Epithelia to Odontoblasts During Mouse Tooth Development MASANORI NAKAMURA, PABLO BRINGAS, JR.,ANTONIO NANCI, MARGARITA ZEICHNER-DAVID, BRIAN ASHDOWN, AND HAROLD C. SLAVKIN Center For Craniofacial Molecular Biology, School of Dentistry, University of Southern California, Los Angeles, California, USA (P.B., M.Z.-D.,B A . , H.C.S.), Department of Anatomy, School of Dentistry, Tohoku University, Sendai, Japan (M.N.) and Department of Stomatology, Faculty of Dentistry and Medicine, University of Montreal, Canada (A.N.) ABSTRACT The developmental problem of how dental epithelia and/or dental papilla ectomesenchyme induce and/or up- or down-regulate tooth formation are as yet unresolved issues. We have designed studies to map the synthesis and fate pathways of secreted amelogenin proteins from Kallenbach differentiation zones 11-IV during in vivo and in vitro mouse mandibular first molar tooth development (MI). Tooth organs from cap, bell, and crown stages were processed for reverse transcriptaselpolymerase chain reaction (RT-PCR)and high resolution Protein A immunocytochemistry using anti-amelogenin and anti-peptide antibodies. Cap stage MI were cultured for periods ranging from 10-21 days in vitro using either serumless, or 15% fetal calf sera-supplemented, chemically-defined medium. Amelogenin transcripts are expressed in the mouse embryonic molar from El5 through early postnatal development. Amelogenin antigens were first detected in Kallenbach’s differentiation zone 11. Amelogenin proteins secreted from preameloblasts were identified along cell processes and cell surfaces of odontoblasts adjacent to forming mantle dentine extracellular matrix (ECM) prior to biomineralization. Amelogenin proteins were restricted to forming endocytotic vesicles, clathrin-coatedvesicles, and lysozomes within odontoblasts. At later stages (e.g. 2 days postnatal development), enamel proteins were not identified in odontoblasts or predentine matrix following mineralization. Comparable observations for stages of development were noted for in vitro cultured tooth explants. Preameloblasts synthesize and secrete amelogenin proteins which bind to odontoblast cell surfaces possibly through the process of receptor-mediated endocytosis. We conclude that amelogenin proteins secreted from preameloblasts, prior to the initiation of biomineralization, were translocated to odontoblasts to serve as yet unknown biological functions. o 1994 Wiey-Liss, Inc. Key words: Tooth development, Mouse, Protein translocation, Amelogenin, Epithelial-mesenchymalinteractions, Intercellular communication, Immunocytochemistry, Differential gene expression, In vitro organ culture It has clearly been established that reciprocal epithelial-mesenchymal interactions mediate a number of epidermal organ systems including tooth development (Koch, 1967; Kollar, 1983; Lumsden, 1987; Lumsden and Buchanan, 1986; Mina and Kollar, 1987; Ruch, 1985, 1988; Ruch et al., 1982; Saxen et al., 1978; Slavkin, 1972, 1974, 1988; Thesleff, 1976, 1977; Thesleff and Hurmerinta, 1981). Nucleopore transfilter studies suggested that ectomesenchymal cell processes mediate the transfer of putative instructive materials (Saxen et al., 1978; Thesleff et al., 1978; Thesleff and Hurmerinta, 1981). Direct mesenchymal-epithelial cell-cell contacts have been described during Kallenbach differentiation zone IV and V in vivo (Kallenbach, 0 1994 WILEY-LISS, INC. 1971; Kallenbach and Piesco, 1979; Pannese, 1962; Silva and Kailis, 1972; Slavkin and Bringas, 1976; Slavkin et al., 1976), prior to the expression of amelogenin mRNA transcripts (Slavkin et al., 1984; Snead et al., 1984, 1988). However, the physical-chemical characterization of these putative instructive signals are not known. Therefore, when, where, and how hetero- Received April 22, 1993; accepted September 20, 1993. Address reprint requests to Professor Harold C. Slavkin, Center For Craniofacial Molecular Biology, 2250 Alcazar Street, CSA 1st Floor, University of Southern California, Los Angeles, California 90033. 384 M. NAKAMURA ET AL. typic tissues communicate with one another remains a n intriguing and significant problem area in developmental biology. Most extracellular matrix (ECM) proteins are synthesized as larger precursors which are post-translationally processed (e.g. glycosylation, phosphorylation, sulfation) and proteolytically cleaved on their way to the plasma membrane where the mature product is released from the cell by exocytosis (Farquhar, 1985). Typical merocrine-like epithelial secretory cells synthesize approximately 103-104 different polypeptides, the vast majority of which are directed for intracellular functions whereas a few “luxury” proteins are secreted for unique extracellular functions (Farquhar, 1985). Previous studies reported that inner enamel epithelia synthesize and secrete enamel proteins prior to initial extracellular biomineralization (Inai et al., 1991; Slavkin, 1988; Slavkin et al., 1988a, b, c; Snead et al., 1984). The precise molecular characterization and fate of these ectodermally-derived ECM molecules is not known. We hypothesize that one fate could be to participate in the formation of the putative dentine/ enamel junction. A second fate might be translocation from the epithelial cell surface to the adjacent ectomesenchyme-derived odontoblast cells. Such a n exocytosis of newly synthesized secretory enamel protein and subsequent receptor-mediated endocytosis by adjacent ectomesenchyme-derived cells (Brown e t al., 1983; Goldstein et al., 1985; Helenius et al., 1983; Wileman et al., 1985) may describe a route for the exchange of molecules which participate in regulatory processes required for developing tooth organogenesis. This paper reports the detection of amelogenin transcripts from E l 5 (15 days gestation) through early postnatal development and provides immunocytochemical evidence (based upon antigenic determinants which are cross-reactive with anti-amelogenin and anti-peptide antibodies) for the translocation of amelogenin proteins from preameloblasts to preodontoblasts during mouse molar tooth morphogenesis as visualized by immunohistology and high resolution immunocytochemistry. The evidence supports the hypothesis that epithelial-derived proteins are translocated to adjacent ectomesenchymal cells during instructive stages of tooth development. MATERIALS AND METHODS Tooth Development In Vivo Mandibular first molars (M,) and incisor tooth organs were dissected from timed-pregnant Swiss-Webster mice (a vaginal plug indicating day 0), staged according to Theiler (1972), and samples were taken each day ( 2 12 hrs) from cap (E15) through late crown stages of tooth formation (10 days postnatal development). The tooth organs were rinsed in cold PBS (phosphate-buffered saline, pH 7.4) and processed for subsequent light and transmission electron microscopy as previously described (Evans et al., 1988; Nanci et al., 1984, 1985, 1987; Slavkin et al., 1976, 1988a). Amelogenin Transcript Detection Tooth organs in vivo ( E l 4 through El91 and E l 5 explants cultured for periods up to 28 days were collected, rinsed with phosphate buffered saline (PBS) and then frozen with liquid nitrogen and stored a t -80°C. Total RNA was extracted according to Evans and Kamdar (1990). RT-PCR (reverse transcriptase/polymerase chain reaction) was performed according to Rappolee et al. (1989). Briefly, 3-7 tooth organs from each sample group were homogenized with sterile pestle in buffer containing 8 M guanidine hydrochloride, 3 M sodium acetate at pH 5.2, and 10% sodium sarcosyl(8:l:l v/v/v) on ice. The total RNA was extracted and precipitated with absolute ethanol at -20°C to remove any contaminating DNA. The precipitated RNA pellet was then dissolved in diethylpyrocarbonate (DEPC) water containing RNase inhibitor (Promega, Madison, WI). Reverse-transcription was started by adding 0.1 M methylmercury (11)hydroxide (Amresko, Solon, OH) to the sample for 7 min. at room temperature, followed by 0.7 M 2-mercaptoethanol (Sigma, St. Louis, MO) for 5 min. A mixture of random primers, RNase inhibitor (Pharmacia, Alameda, CA), M-MLV reverse transcriptase (GIBCO BRL, Gaithersburg, MD), dNTP (Pharmacia, Alameda, CA), and RT buffer (GIBCO BRL, Gaithersburg, MD) were added to the RNA extract. The samples were heated a t 42°C for one hour, boiled at 95°C for 5 min. and immediately cooled on ice. The reaction was repeated three times. PCR reaction solution was prepared by adding PCR buffer, 10 mM dNTP, and Thermus aquaticus (Taq) polymerase (Perkin-Elmer Cetus, Norwalk, CT) in DEPC water. Amelogenin (AMEL) primers were designed and synthesized based on the published sequence of the canonical mouse amelogenin (Snead et al., 1985; Lau et al., 1992). The sequences of AMEL primers used were as follows: 5‘ end GGT CTA GAA TGC CGA AAT GGG GAC C’M! G, and 3’ G GTA CTG AGG TI’G GGT TGT GGT. Beta-actin was used as a n internal positive control with primers as follows: 5‘ CAT CGT GGG CCG CTC TAG GCA CCA, and 3’ G GTA CTG AGG TTG GGT TGT GGT. A mixture containing 43 p1 PCR reaction solution, 2 pl of 3’ primer, 2 pl of 5’ primer, and 3 p1 RT reaction product was then loaded in small Eppendorf tubes, sealed with mineral oil and placed in a thermal cycler machine to be processed at 95°C for 5 min then 94°C for 1min, 55°C for 45 sec, and 72°C for 1 min in 60 cycles. Six microliters of the 50 p1 PCR reaction product was fractionated by electrophoresis in a 3% Nusieve (FMC, Rockland, ME) and 1% agarose gel (GIBCO BRL, Gaithersburg, MD). Gels were run at 85 volt for 45-50 min, until the bromophenol blue dye front migrated 6 cm. Molecular weight standard used was a mixture of ficoll, trypan blue, and +xl74/HaeIII (GIBCO BRL, Gaithersburg, MD). The gel was stained for 30 min in ethidium bromide, destained 10 min in water, viewed under UV light box, and photographed using Polaroid film. The positive control for AMEL production was 4 day postnatal mandibular molar tooth extracted mRNA from Swiss Webster mice. The negative control used no tissue-derived RNA but all the reagents. Subsequent sequences for the amplified AMEL confirmed that these PCR products were in fact amelogenin. Tooth Development In Vitro Cap stage molars were isolated and maintained in organ culture for 28 days using either serum-supplemented or serumless, chemically-defined media as previously described (Bringas et al., 1987; Evans et al., 385 TRANSLOCATION OF ENAMEL PROTEINS 1988; Yamada et al., 1980). The medium was FittonJackson’s modified BGJB (GIBCO, Staten Island, NY) with or without 15% FCS (fetal calf sera), and supplemented with 50 mM ascorbic acid and 50 unitdm1 streptomycin and penicillin. Cultures were maintained at 37°C in atmospheric conditions of 95% air and 5% CO,. Medium was pH 7.4 at the start of each culture and the medium was changed every other day. Electroblotting and lmmunostaining The electrophoretic transfer of proteins was based on Burnette’s modifications (Burnette, 1981) of Towbin’s method (Towbin et al., 1979). The transfers were conducted a t 4°C for 75 min at 0.6 A (60-90 V) after equilibration of the gels in the transfer buffer (0.192 M glycine; 0.025 M Tris, pH 8.3; 20% v/v methanol). It was necessary to use at least 200 pg protein per sample for Western blotting and immunodetection following proAntibodies Directed Against Amelogenin tein separation of IEF-PAGE (Slavkin et al., 1988a). Proteins of interest were determined to transfer only The sensitivity and specificity for the rabbit antibody toward the anode. After transfer, the membranes were to the major mouse amelogenin has been previously removed and washed for 10 min in TBS (20 mM Tris, reported (Slavkin et al, 1982; Slavkin et al., 1988a). pH 7.5; 0.5 M NaC1) and then immersed in TBS containing 3% gelatin for 30 min. The membranes were Synthetic Arnelogenin Anti-Peptide Antibodies then incubated in a solution of TBS with 1% gelatin Oligopeptides corresponding to the deduced amino (Hawkes et al., 1982) containing rabbit IgG antibodies acid sequence for the major mouse amelogenin (Lau et to either mouse amelogenins or antipeptide antibodies, al., 1992; Snead et al., 1985) were synthesized by Ap- in 1:2,000 dilution for 16 h at 30°C with gentle agitaplied Protein Technology (Cambridge, MA): (i) LP- tion. The incubated membranes were rinsed once with PHPGHPGYIC (residues 3-13, designated peptide distilled water and twice for 10 min with TBS, and then #l); (ii) IRQPYPSYGYERC (residues 30-40, desig- incubated with goat anti-rabbit horseradish-peroxinated peptide #2); (iii) SQQHPPSHTLQC (residues dase-conjugated (GAR/HRP) IgG at a 1:2,000 dilution 55-65, designated peptide #3); and (iv) HHQPNIPP- in 1% gelatidTBS for 1h. The filters were washed with SAQQPFQC (residues 99-113, designated peptide #4). water and TBS as described above, and the peroxidase The synthetic peptides were coupled to keyhole limpet color reaction was completed by placing membranes in hemocyanin (KLH) (Kagan and Glick, 1979) and used freshly prepared TBS solution containing 16% methaas immunogens to immunize young female New nol, 0.05% 4-chloro-l-naphthol, and 0.05% hydrogen Zealand White rabbits by multi-site subcutaneous in- peroxide for 15-30 min. The colorimetric reaction was jections. Boosting was performed every 7 days, and im- terminated by washing the membranes twice for 10 mune sera collected after three boosts. The IgG fraction min in distilled water. was prepared by DEAE-cellulose chromatography and lmmunohistochemistry used in enzyme-linked immunoabsorbent assays Tooth samples from E l 5 through two days postnatal (ELISA) to determine the titer and specificity of the antibody. A titer of 1:2,000 dilution was empirically as well a s from E l 5 molars cultured for periods up to 28 determined to be optimal for the immunocytochemistry days were fixed in Carnoy’s solution for 4 h and proand immunohistochemistry applications. The specific- cessed for paraffin embedding as recently described ity and sensitivity of the anti-peptide amelogenin an- (Hu et al., 1992). Five micron serial sections were prepared from tissue blocks. Immunohistochemistry was tibodies has been published (Slavkin et al., 1988a). used to identify the cellular and extracellular distribution of amelogenin protein antigens with the Zymed High-Resolution Two-DimensionalGel Electrophoresis streptavidin biotin system (Zymed, San Francisco, CA). Isoelectric focusing-polyacrylamide gel electrophore- Sections were mounted on Histostik coated slides, desis (IEF-PAGE) was conducted as described (O’Farrell, paraffinized and rehydrated. Endogenous peroxidase 1975). The first-dimension gel used a mixture of am- activity was blocked by 30% hydrogen peroxide and pholites (1.6% pH 5-8 and 0.4% pH 3-10) in a 10.5- absolute methanol (1:9, v/v), and slides were washed in cm-long gube gel with a diameter of 2 mm. The sample phosphate-buffered saline (PBS). Slides were treated buffer contained 9.5 M urea, 5% (v/v) 2-mercaptoetha- with a blocking solution, primary antibody, biotinynol, 2% (v/v) Nonidet P-40, and 2% (v/v) ampholites. lated secondary antibody, enzyme conjugate, substrate Extracted proteins (100-200 pg) from mouse teeth chromogen mixture, and counter stained with diluted were mixed with a n equal volume of lysis buffer after hematoxylin according to the specifications from the the addition of solid urea. Isoelectric focusing was con- vendor. ducted for 16 h at 400 V and 1 h at 800 V. Primary antibody raised against mouse amelogenin The second dimension was run according to the or synthetic amelogenin anti-peptide antibodies were method of O’Farrell (1975)) employing the Laemmli used for indirect immunostaining. The optimal dilution PAGE-system (Laemmli, 1970). Briefly, the IEF gels and incubation time for the amelogenin primary antiwere incubated in SDS equilibration buffer for 1-2 h body was determined to be 1:2,000 with 2 h or incubaand anchored into place with 1% agarose solution in tion at room temperature. Specificity controls included the same buffer. After electrophoresis through the pre-immune sera instead of primary antibody, secondstacking gel (4% acrylamide) at 10 mA per gel, gels ary antibody only, and amelogenin or synthetic peptide were run overnight through the resolving gel (10% antigen immunoadsorption of the primary antibody. acrylamide) at 3 mA per gel. It was observed that slow lmmunogold Cytochemistry runs eliminated streaking and imDroved seDaration of The protein A-gold technique was used to detect proteins. Samples were silver-stained aftkr electroenamel Drotein antigens as Dreviouslv described phoresis (Slavkin et al, 1988a). u v 386 M. NAKAMURA ET AL. M I 2 3 4 M 5 6 7 8 bP 369246- 123the detection of two amelogenin transcripts at 364 and 256 base pairs. The amelogenin transcript amplimer is barely detected a t E l 5 but is readily apparent by E l 7 (in vivo data not shown) or E l 5 explants cultured for 7 days in vitro (lane 6). Sequence data for RT-PCR amplified amelogenins confirmed that the detected amplimers were in fact amelogenins and the pattern of transcription reflected the process of alternative splicing as suggested by Gibson et al. (1991),and Lau et al. (1992). Fig. 1. Amelogenin expression during E l 5 mouse molar tooth development in serumless, chemically-defined medium. Total RNA samples extracted from E l 5 molar explants cultured in serumless medium were amplified for amelogenin utilizing reverse transcriptase/ polymerase chain reactions (RT-PCR). Beta-actin mRNA was used as an internal positive control. Lanes M designate molecular weight markers 369,246, and 123 base pairs. Lanes 1 4 represent 0, 7, 12, and 15 days in culture, respectively, and the presence of beta-actin. Lanes 5-8 represent 0 , 7 , 1 2 , and 15 days in culture, respectively, and 7.5 I kDa 6.5 I 5.5 7.5 1 6.5 I IEF PH I 5.5 7.5 t 6.5 5.5 S 68- D S 43- 26- Fig. 2. Comparison of anti-amelogenin and anti-peptide antibody specificity to identify mouse enamel proteins separated and fractionated by two-dimensional gel electrophoresis, electroblotted, and immunostained. A Two days postnatal amelogenin proteins immunostained with anti-amelogenin antibody. B Two days postnatal amelogenin proteins immunostained with anti-peptide #1 antibody. C: Acetic acid-soluble proteins extracted from E l 5 mandibular first molar explants cultured for 21 days in serumless, chemically-defined medium and immunostained with anti-amelogenin antibody. (Nanci et al., 1984, 1985, 1987, 1989). The antibody used was polyclonal rabbit anti-mouse amelogenin IgG antibody (Slavkin et al., 1982,1988a). As a n additional control for tissue preservation and the localization of enamel antigens, replicate samples were processed using anhydrous fixation (Diekwisch et al., 1993; Landis, 1983) and examined with transmission electron microscopy as previously described (Evans et al., 1988). RESULTS AMEL Transcription During Cap Stage Molar Tooth Development Amelogenin transcripts were detected from E 15 through early postnatal stages of mouse mandibular first molar tooth development in vivo and during tooth culture in vitro using RT-PCR (Fig. 1).Whereas previ- TRANSLOCATION O F ENAMEL PROTEINS 387 Fig. 3. Immunohistochemical localization of amelogenin proteins using either anti-amelogenin or anti-peptide polyclonal antibodies during translocation from inner enamel epithelium to dental papilla ectomesenchyme. A At birth ameloblasts (Am) express and secrete amelogenins (arrow) in juxtaposition to odontoblasts (Od) cell processes associated with forming dentin extracellular matrix. Bar line = 50 pm in A, B & C. B: Comparable developmental stage of enamel formation as shown in A but stained with anti-peptide #1 antibody (arrows). C: A control using primary anti-peptide #I antibody immunoabsorbed with anti-peptide #1 immunogen. D Survey showing patterns of immunostaining with anti-peptide #I antibody in forming cusps of E l 9 molar tooth organs. Positive immunolocalization is evident along the odontoblast cell bodies along the periphery of the dental papilla ectomesenchyme (Dpm). Bar line = 50 pm. E: Immuno- localization of amelogenins within preameloblasts, ameloblasts and adjacent odontoblast cell bodies (arrows) in E l 9 molars stained with anti-peptide #1 antibody. Bar line = 50 pm in E-I. F: E l 5 molar explants cultured for 12 days in vitro in serumless medium and stained with anti-amelogenin antibody showing amelogenin along odontoblast (Od) cell bodies (arrows). G: E l 5 molar explant cultured for 12 days in vitro with medium supplemented with 15%fetal calf serum and stained with anti-peptide #1 antibody. H Immunolocalization of amelogenins within preameloblasts, ameloblasts (Am), odontoblast cell processes (arrows), and adjacent odontoblast (Od) cell bodies during E l 8 molar tooth development. I: A control using primary anti-amelogenin antibody immunoabsorbed with amelogenin immunogen during E l 8 molar tooth development. ous in situ hybridization studies demonstrated initial, amelogenin transcription during newborn mouse molar development (Snead et al., 19881,the present studies report that amelogenin m ~ are detected ~ ~ during s 15-day gestation molar tooth development, E l 5 molar explants cultured for periods up to 15 days (Fig. 1, lanes 5-8), and throughout fetal and early postnatal molar tooth formation (data not shown). Alternatively spliced amelogenin products were detected a s 364 base pairs and 256 base pairs, respectively (Fig. 1, lanes 7-8). Two-Dimensional lrnrnunoblot Analysis of Enamel Protein Expression Using Anti-Arnelogenin and Anti-Peptide Antibodies A comparision of acetic acid soluble enamel proteins extracted from either in vivo or in vitro molar tooth organs and separated by two-dimensional gel electro- 388 M. NAKAMURA ET AL. Fig. 4. Odontoblast (Od) and ameloblast (Am) cytodifferentiation, and dentine (D) and enamel (En) ECM formation in viva and in vitro. A: Secretory amelogenesis and enamel formation in vivo a t 2 days postnatal development. B: Cap stage molar cultured for 12 days in BGJb medium supplemented with 15% FCS. Toluidine blue stained granules accumulate in predentine (pD) adjacent to odontoblast cell surfaces (arrows). Bar line = 40 Fm. FCS, cap stage tooth explants expressed crown morphogenesis and inner enamel epithelia differentiated into ameloblasts (Fig. 5B). Surveys of inner enamel epithelial differentiation in vivo and in vitro (with or without serum supplementation), demonstrated the deposition of electron-dense granular materials in differentiation zones 11-VI (Fig. 5A). During serumless cultures, both odontoblast and preameloblasts extended cell processes (Fig. 5A). Electron-dense granular materials were deposited along epithelial cell surfaces, epithelial and odontoblast cell processes, and also along the surfaces of odontoblasts lmrnunohistochemical Localization of Enamel Proteins (Fig. 5A,B, arrows). This electron-dense granular maDuring Epithelial Differentiation Into Ameloblasts terial was also observed prior to and during initial Mouse molar tooth development demonstrated ini- ECM biomineralization during in vivo development tial amelogenin localization in preameloblasts and (Fig. 6). The material persisted during epithelial difameloblasts followed by localization with odontoblasts ferentiation zones 11-V, thereafter, the material was (Fig. 3A-I). Immunohistochemistry was used to map not observed during in vivo or in vitro development. Immunocytochemical localization of enamel proteins the timing and positional patterns of amelogenin proteins. Both anti-amelogenin and anti-peptide antibod- using anti-mouse amelogenin antibodies identified the ies demonstrated amelogenin immunostaining associ- electron-dense granular materials to contain enamel ated with preameloblasts (Fig. 3A-B, arrows) in antigenic determinants (Fig. 6A-E). During tooth culdifferentiation zone I1 (Kallenbach, 19711, and associ- ture in serumless medium, enamel antigens were loated with odontoblasts (Fig. 3E-H, arrows). Ameloge- calized along both preameloblast and odontoblast cell nin translation products were first localized within the surfaces (Fig. 6A, arrows). Enamel antigenic domains inner enamel epithelium of E l 7 embryonic mouse were clustered along odontoblast cell surfaces and were internalized by coated pits (Fig. 6B, arrows). During mandibular first molars (data not shown). mouse mandibular first molar development differentiation zones 11-V, enamel protein antigens were locallmmunocyfochemical Localization of Extracellular Enamel ized in electron-dense granular materials localized Proteins on Odontoblast Cell Processes along cell processes (Fig. 6C, arrows). One control for Cap stage tooth explants produced enamel matrix antibody specificity used immunogen (amelogenin) adduring organ culture using medium either supple- sorbed primary antibody as shown in Figure 7. mented with 15% FCS or in serumless, chemically-defined medium. By 12 days, initial biomineralization Enamel Protein Antigen Localizations in was detected in explants cultured in serum suppleOdontoblast Lysozornes mented medium, and by 21 days enamel and dentine In the later differentiation zones, enamel antigens biomineralization was comparable to that observed during postnatal development (Fig. 4A,B). Using 15% were localized within lysozomes in odontoblasts (Fig. phoresis, electroblotted and immunostained with either anti-amelogenin or anti-synthetic peptide # 1 antibodies is shown in Figure 2. At 2 days postnatal development, anti-amelogenin antibody identified a larger number of epitopes (Fig. 2A) than the anti-peptide #1 antibody; both antibodies were used at 1:2,000 dilution (Fig. 2B). The sensitivity and specificity of the anti-peptide antibody to identify specific amelogenin epitopes during in vivo tooth development was also observed in E l 5 molar explants cultured for 21 days in serumless medium (Fig. 2C). Fig. 5.During epithelial-mesenchymal interactions, at Kallenbach differentiation zone V, electron-dense materials were deposited along the surfaces of extended odontohlast cell processes and adjacent to the odontoblast cell body (Od). A Numerous odontoblast cell processes approximate the undersurface of preameloblasts (pAm); electrondense granular materials were identified along epithelial and odon- toblast cell surfaces (arrows). M, cultured in serumless medium after 12 days in uitro. Bar line = 2 pm. B: M, development a t 1day postnatal tooth formation. Basal lamina is discontinuous. Electron-dense granular materials associated with cell processes were identified (arrows). Bar line = 1 pm. Fig. 6 TRANSLOCATION OF ENAMEL PROTEINS 391 6D,E). No enamel antigen was detected in rER, Golgi or mitochondria. Antigen was identified along odontoblast surfaces, aggregated in coated pits, and appeared to be localized within intracellular vesicles and lysozomes (Fig. 6D,E). Analyses of serial sections (silver) indicated that these antigens were localized within lysozomal bodies. Odontoblasts showed intercellular antigen labeling (Fig. 8A,B, arrows), continued accumulation along cell surfaces (Fig. 8B, arrows), and localization in lysozomes (Fig. 8B). No enamel antigen was detected in odontoblasts after differentiation zone VI . DISCUSSION We report the transcription of amelogenin during fetal tooth development. This finding provides additional evidence that inner enamel epithelium expresses amelogenin well in advance of evidence for tissue-specific enamel or dentine biomineralization. Further, we report a series of observations which suggest translocation of enamel polypeptides or peptides from preameloblasts to odontoblasts prior to dentine biomineralization in uivo and in uitro. Analysis of the patterns of enamel antigenic determinant immunolocalization led us to suggest that ectodermally-derived inner enamel epithelia secrete and transfer tissue-specific proteins to adjacent odontoblasts prior to the initiation of dentine biomineralization. Whereas, the epithelial-derived enamel protein(s) appeared to bind to the cell surface suggestive of ligand binding to receptors, possibly endocytosed by odontoblasts and translocated to endosomal compartments (Goldstein et al., 1985; Helenius et al., 1983; Lazarovits and Roth, 1988; Verner and Schatz, 1988; Wileman e t al., 19851, extreme caution should be exercised since fluid phase endocytosis would be undistinguished with the methods used in this study. Previous studies with autoradiographic or immunocytochemical techniques have described the compartments within ameloblasts during enamel protein synthesis and secretion (Greulich and Slavkin, 1965; Nanci et al., 1984, 1985, 1987; Slavkin et al., 1976; Warshawsky, 1968; Weinstock, 1972; Weinstock and Leblond, 1971). Our present studies confirm the immunohistochemical demonstration of amelogenin penetration from preameloblasts toward the dental pulp during early stages of rat molar tooth development (Inai et al., 1991). However, whereas Inai and colleagues Fig. 6. Immunocytochemical localization and suggested translocation of enamel protein antigenic determinants from inner enamel epithelia (preameloblasts, differentiation zone IV), along odontoblast cell processes, and internalization by coated pits and localization in lysozomes within odontoblasts. A: Differentiation zone IV during tooth organ culture after 12 days. Odontoblast (Od) cell processes extend towards preameloblasts (pAm). Arrows indicate enamel antigens along cell surface and associated with coated pits during internalization. Bar line = 1 pm. B Enamel antigens accumulate in regions (arrows) related to endocytosis. C: Suggested translocation of enamel antigens (arrows) from preameloblasts to odontoblasts mediated by cell processes (cp) extended through the forming dentine matrix (D) at 1-day postnatal development. D Enamel protein antigens were localized in odontoblast lysozomes. E: Odontoblast cell surface with enamel antigens (arrow) and intracellular lysozome a t differentiation zone VI. Bar line for B-E = 200 pm. Fig. 7. Isolated mouse 26 kDa amelogenin (immunogen) adsorbed primary antibody served a s one of the immunocytochemical controls to illustrate antibody specificity. Odontoblast cell process (cp) with electron-dense granular materials and forming dentine matrix. Bar line = 200 pm. (1991) described the localization of amelogenin protein antigen in 0-5 day-old rat molar development, our studies report for the first time, evidence for amelogenin expression and translocation during embryonic mouse molar development in vivo and in vitro. Epithelial-Ectomesenchyme Interactions During Tooth Cytodifferentiationand Morphogenesis The timing and position of instructive signal(s) which regulate either epithelia or mesenchyme during epidermal morphogenesis (e.g. thyroid gland, mammary gland, salivary gland, feather formation, hair for- 392 M. NAKAMURA ET AL. Fig. 8. Enamel proteins translocated from epithelia to odontoblasts were localized either between odontoblasts and/or within odontoblast lysozomes. A: Intercellular localization of enamel proteins (arrows) between odontoblasts. B: Differentiation zone VI showing surface antigens and odontoblast lysozomal localizations. (Arrows = enamel proteins) Bar line = 200 wm. mation, tooth formation) are not known. What is evident is t h a t ectodermally-derived first branchial arch epithelium appears to induce molar tooth formation between E9-Ell (Lumsden, 1987; Lumsden and Buchanan, 1986; Mina and Kollar, 1987). Thereafter, ectomesenchyme-derived signals appear to mediate the shapetform of tooth morphogenesis and perhaps serve to induce enamel production (Koch, 1967; Kollar, 1983; Lumsden, 1987; Lumsden and Buchanan, 1986; Ruch, 1985, 1988; Ruch et al., 1982; Saxen et al., 1976; Slavkin, 1972, 1974, 1988; Slavkin et al., 1984, 1988a, b, c; Snead et al., 1984,1988; Thesleff and Hurmerinta, 1981; Thesleff et al., 1978). Ultrastructural and transfilter studies have provided results which suggest that preodontoblast cell processes and/or ECM molecules “transfer” inductive signals from mesenchyme to epithelia (Kallenbach, 1971; Kallenbach and Piesco, 1979; Koch, 1967; Pannese, 1962; Saxen e t al., 1976; Silva and Kailis, 1972; Slavkin et al., 1976; Slavkin and Bringas, 1976; Thesleff et al.; 1978). Evidence from transfilter Millipore and Nucleopore filters (pore sizes ranging from 0.1-0.8 pm and thicknesses of 25-100 pm) demonstrate that dental papilla ectomesenchyme cell processes and/or epithelial microvilli protrude through 0.2 pm pore size and 25 pm thick Nucleopore filters to mediate instructive signals for epithelial cytodifferentiation (Koch, 1967; Thesleff and Hurmerinta, 1981; Thesleff et al., 1978); assayed according to epithelial cell elongation and enamel ECM production. Tritiated proline labeled enamel organ epithelia have been demonstrated to synthesize and secrete 3H-proline-labeled proteins which diffuse from epithelia to mesenchyme through Millipore transfilter tissue recombinations (Wolters, 1978). Despite the numerous suggested molecular processes which might control this heterotypic cell-cell communication (e.g. cell sur- 393 TRANSLOCATION OF ENAMEL PROTEINS face proteoglycans, galactosyltransferases, differentiation cell surface antigens, short-range diffusible molecules, ion transport, transfer of nucleic acids, etc.) (McMahon and West, 1976; Ruch, 1985; Saxen et al., 1976; Slavkin, 1972; Slavkin et al., 1988a, b, c; Thesleff, 1977; Thesleff and Hurmerinta, 1981; Wolters, 19781, the molecular process for inductive epithelial-mesenchymal interactions during tooth formation is not known. Translocation of Enamel Proteins: A Model for Macromolecular Transfer From Epithelia to Odontoblasts Am el o bl a st s Preameloblast Synthesis Enamel Proteins , Future DE+t I m,ii, Translocation l l ,cell The scheme shown in Figure 9 illustrates the sugprocess gested route of enamel protein translocation from preameloblasts to odontoblasts during tooth development. Our studies confirm previous studies demonstrating the synthesis and secretion of enamel proteins Elongation by mouse preameloblasts (Nanci et al., 1984, 1985, Coated 1987; Slavkin e t al., 1976, 1988a, b, c). The localization Preodontoblast of enamel antigenic determinants within the ECM is clearly demonstrated by several types of evidence. High resolution autoradiographic studies demonstrate isotopically-labeled tryptophan-containing enamel proteins secreted by preameloblasts and associated ? with extended odontoblast cell processes (Slavkin et al., 1976). Nanci and colleagues have shown the exOdontoblasts pression of enamel proteins in vivo (Nanci et al., 1985, 1987) and during in vitro mouse molar tooth developFig. 9. Scheme illustrating suggested translocation of enamel proment in serumless as well a s serum-supplemented me- teins from preameloblasts to odontoblasts. The suggested pathway is dium (Nanci et al., 1984). More recently, Inai and col- as follows: (i) synthesis and secretion of enamel proteins from inner epithelia in differentiation zones 11-IV; (ii) enamel protein(s)/ leagues (1991) demonstrated that the presecretory enamel type I collagen interactions associated with the formation of the puameloblasts in their early stages of differentiation both tative dentinoenamel junction (DEJ); (iii) translocation of enamel prosynthesized amelogenin and secreted this protein tein along odontoblast cell processes; (iv) suggested binding of enamel through a classical merocrine secretory pathway; im- protein(s1 to odontoblast cell surfaces and subsequent internalization coated vesicles; (v) coated vesicles form endosomes and lysozomes munostaining was also localized in the predentin a s by within odontoblasts; and (vi) enamel protein antigen is also localized well a s the intercellular spaces of odontoblasts. Fi- between odontoblasts and between odontoblasts and adjacent pulpal nally, a number of ultrastructural studies have re- fibroblast cells. Possible biological responses to the translocation inported that the enamel precursor-like material re- clude odontoblast up- or down-regulation of differential gene expression (e.g. dentine phosphoprotein, dentin sialoglycoprotein, growth ferred to as stippled material was observed in factors, growth factor receptors, fibronectin, etc.). predentin a s well as in dentin and along odontoblast cell surfaces (Reith, 1967; Suga, 1960; Watson, 1960; Yamamoto et al., 1980). Curiously, Yamamoto et al. (1980) were the first to suggest that enamel-like stip- translocated to adjacent odontoblasts during tooth depled material found in predentin originated from velopment. ameloblasts. Epithelial-derived enamel proteins participate in the Isolated enamel organ epithelia labeled with 3H- determination of the dentine-enamel junction. It is evproline for one hour an> recombined in transfilter stud- ident that odontoblast cell processes extend to the denies with isolated dental papilla ectomesenchyme, dem- tine-enamel interface (Kallenbach, 1971; Kallenbach onstrated the synthesis, secretion, and diffusion of and Piesco, 1979; Kelley et al., 1981; Ruch, 1988; Sigal isotopically labeled molecules from epithelia to mesen- et al., 1984; Thesleff and Hurmerinta, 1981; Thesleff et chyme (Wolters, 1978); proline is a major amino acid al., 1978). Both inner enamel epithelia and dental paconstituent of amelogenins and enamelins (Deutsch, pilla mesenchyme synthesize and secrete collagens 1989; Deutsch et al., 1991; Ogata et al., 1988; Termine (Trelstad and Slavkin, 1974).The dentine-enameljuncet al., 1980). One possibility is that the enamel organ tion may represent the coprecipitation of the collageepithelia synthesized and secreted types I and IV col- nous and non-collagenous extracellular matrix prolagens as previously demonstrated by Trelstad and teins (Arsenault and Robinson, 1989). The timing and Slavkin (Trelstad and Slavkin, 1974). However, in position for this supramolecular assembly may be Wolter’s study (Wolters, 1978), approximately 40% of controlled by the initially secreted enamel proteins the proline-labeled material which accumulated be- (Deutsch, 1989; Diekwisch et al., 1993; Silva and Kaltween previously labeled epithelia and non-labeled lis, 1972; Slavkin, 1988), and/or molecular interactions mesenchyme was digested by type I collagenase; the between anionic enamel proteins and extended odontoremaining material being epithelial-derived non-col- blast cell processes. lagenous extracellular matrix molecules. We suggest The present studies have defined the time- and posithat the material described in the present study repre- tion-restricted expression of enamel antigenic determisents epithelial-derived enamel proteins which were nants according to the convention of Kallenbach and I 441 I I 394 M. NAKAMURA ET AL. recently applied to mouse tooth development (Slavkin Research, National Institutes of Health, United States et al., 1988a, b, c). From Kallenbach’s differentiation Public Health Service, DE-02848 and DE-06425. zones 11-V we observed the fate and spatial pathway of LITERATURE CITED enamel proteins. In all cases we observed enamel anArsenault, A.L., and B.W. Robinson 1989 The dentino-enamel junctigenic determinants in electron-dense granular matetion: A structural and microanalytical study of early mineralizarial along the surfaces of cell processes, in coated vestion. Calcif. Tissue Int., 45t111-121. icles, and within lysozomes within the odontoblasts Bringas, Jr., P., M. Nakamura, E. Nakamura, J. Evans, and H.C. (Figs. 6, 8). We also observed material with enamel Slavkin 1987 Ultrastructural analysis of enamel formation during in vitro development using chemically-defined medium. Scanantigenic determinants between odontoblast cells. ning Microscopy Intl., lt1103-1108. These results support recent observations from Inai Brown, M.S., R.G.W. Anderson, and J.L. Goldstein 1983 Recycling and colleagues (1991) who reported the presence of receptors: the round-trip itinerary of migrant membrane procoated pits at the plasma membrane of odontoblasts in teins. Cell, 32.563-667. close proximity to immunolabeled amelogenin along Burnette, W.N. 1981 “Western blotting”: Electrophoretic transfer of proteins from sodium dodecyl sulfate-polyacrylamide gels to unwith immunolabeling of lysozomes within the odontomodified nitrocellulose and radiographic detection with antibody blasts, suggesting endocytosis of enamel protein by the and radioiodinated protein-A. Anal. Biochem., 112t195-203. odontoblasts prior to mineralization. Deutsch, D. 1989 Structure and function of enamel proteins. Anat. Rec., 224r189-210. Several key issues are essential in order to optimally D., A. Palmon, L.W. Fisher, N. Kolodnv, J.D. Termine. and interpret translocation of enamel proteins from Deutsch, M.F. Young 1991 Sequencing of bovine enamelin (“Tuftelin”) a preameloblasts to odontoblasts: (i) what is the percentnovel acidic enamel protein. J . Biol. Chemistry, 63r98-105. age of enamel protein presumed to be translocated from Diekwisch, T., S. David, P. Bringas, Jr., V. Santos, and H.C. Slavkin 1993 Antisense inhibition of AMEL translation demonstrates suepithelia to odontoblasts? (ii) the identification of cell pramolecular controls for enamel HAP crystal growth during emsurface binding sites (e.g. receptors), and (iii) the prebryonic mouse molar development. Development, 117.471-482. cise molecular characteristics of the transferred Evans, R., and S.J. Kamdar 1990 Stability of RNA isolated from macenamel polypeptides (see discussion of alternative rophages depends on the removal of an RNA-degrading activity early in the extraction procedure. Biotechniques, 8.857-859. splicing Lau et al, 1992). Further, it is not known how J., P. Bringas, Jr., M. Nakamura, E. Nakamura, V. Santos, the enamel antigens bind to the odontoblast cell sur- Evans, and H.C. Slavkin 1988 Metabolic expression of intrinsic developfaces, and/or if the subsequent endocytosis mediated by mental programs for dentine and enamel biomineralization in coated pits is specific for epithelial-derived proteins. serumless, chemically-defined, organotypic culture. Calcif. Tiss. Int., 42.220-230, The observation of immunolabeling in coated pits and M.G. 1985 Progress in unraveling pathways of golgi traffic. endosomes of odontoblasts is not sufficient to demon- Farquhar, Ann. Rev. Cell Biol., lt447-488. strate specific, saturable binding indicating receptor- Gibson, C.W., E. Golub, W. Ding, H. Shimokawa, M. Young, J . Termediated endocytosis. The function(s) of presumably mine, and J. Rosenbloom 1991 Identification of the leucrine rich arnelogenin peptide (LRAP) as the translation product of a n altransferred enamel proteins to odontoblasts remains ternatively spliced transcript. Biochem. Biophys. Res. Comm., a n enigma. 174r1306-1312. The precise mechanism by which enamel protein be- Goldstein, J.L., M.S. Brown, R.G.W. Anderson, D.W. Russell, and W.J. comes internalized by odontoblasts is not known. Schneider 1985 Receptor-mediated endocytosis: Concepts emerging from the LDL receptor system. Ann. Rev. Cell Biol., Irl-39. Enamel protein(s) may bind opportunistically to recepR.C., and H.C. Slavkin 1965 Amino acid utilization in the tors on the cell surface of odontoblasts and slide later- Greulich, synthesis of enamel and dentin matrices as visualized by autoraally into clathrin-coated pits. The coated pits possibly diography. In: The Use of Radioautography in Investigating Proinvaginate and form coated endocytic vesicles which tein Synthesis. C.P. Leblond, and K.B. Warren, eds. Academic Press, New York, pp. 199-214. could fuse to form endosomes or eventual lysozomes R., E. Niday, J. Gordon 1982 A dot-immunobinding assay for (Goldstein e t al., 1985). The contents of these or- Hawkes, monoclonal and other antibodies. Anal. Biochem., 119t142-147. ganelles are acidified by ATP-driven proton pumps Helenius, A., I. Mellman, D. Wall, and A. Hubbard 1983 Endosomes. (Pastan and Willingham, 1983), or are degraded by TIBS, 8t245-250. proteolytic enzymes within lysozomes (Brown et al., Hu, C.C., Y. Sakakura, Y. Sasano, L. Shum, P. Bringas, Jr., 2. Werb, and H.C. Slavkin 1992 Endogenous epidermal growth factor reg1983). Recent evidence indicates that the cytoplasmic ulates the timing and pattern of embryonic mouse molar tooth domains of cell surface receptors possess a unique feamorphogenesis. Internat. J. Develop. Biol., 36.505-516. ture apparently required for endocytosis. A correctly Inage, T., H. Shimokawa, Y. Teranishi, T. Iwase, Y. Toda, and I. Moro 1989 Immunocytochemical demonstration of amelogenins and positioned tyrosine presumably forms part of a small enamelins secreted by ameloblasts during the secretory and mat“epitope” of non-contiguous amino acids that becomes uration stages. Arch. Histol. Cytol., 52:213-229. recognized by a receptor and is required for receptor- Inai, T., T. Kukita, Y. Ohsaki, K. Nagata, A. Kukita, and K. Kurisu mediated internalization (Lazarovits and Roth, 1988). 1991 Immunohistochemical demonstration of amelogenin penetration toward the dental pulp in the early stages of ameloblast Immunoperturbation studies using FAB’ fragments didevelopment in rat molar tooth germs. Anat. Rec., 229t259-270. rected against specific enamel protein epitopes might Kagan, A., M. Glick 1979 In: Methods of Human Radiommunoassay. be useful towards a further dissection of the suggested B.M. Taffe, and H.A. Behram, eds. Academic Press, New York, translocation pathway during inner enamel epithelial pp. 328-329. and adjacent odontoblast cytodifferentiation. Eventu- Kallenbach, E. 1971 Electron microscopy of the differentiating rat incisor ameloblast. J . Ultrastruct. Res., 35.508-53. ally, mechanisms of chemical transduction might conE., and N.P. Piesco 1979 The changing morphology of the tribute towards a n enhanced understanding of differ- Kallenbach, epithelium-mesenchyme interface in the differentiation zone of ential gene expression in odontoblast cells. the growing teeth of selected vertebrates and its relationship to ACKNOWLEDGMENTS We thank Mrs. Helen Fludd for preparing this manuscript. These studies were supported in part by research grants from the National Institute for Dental possible mechanisms of differentiation. In: Tooth Morphogenesis I. J.V. Ruch, ed. Inserm, Paris, pp. 315-332. Kelley, K.W., G. Bergenholtz, and C.F. Cox 1981 The extent of the odontoblast process in rhesus monkeys (maca mulatta) as observed by scanning electron microscopy. Arch. Oral Biol., 26: 893-897. TRANSLOCATION OF ENAMEL PROTEINS Koch, W.E. 1967 In vitro differentiation of tooth rudiments of embryonic incisor tissues. J. Exp. Zool., 165r155-170. Kollar, E.J. 1983 Epithelial-mesenchymal interaction in the mammalian integument. In: Epithelial-Mesenchymal Interaction in Development. R.H. Sawyer, and J.F. Fallon, eds. Praeger Publishers, New York, pp. 27-50. Laemmli, U.K. 1970 Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature, 227r680-695. Landis, WYJ. 1983 Special techniques for problem tissues. In: Electron Microscopy. B.R. Jones, ed. Academic Press, New York, pp. 459- 476. Lau, E.C., J.P. Simmer, P. Bringas, Jr., D.D.J. Hsu, C.C. Hu, M. Zeichner-David, F. Thiemann, M.L. Snead, H.C. Slavkin, and A.F. Fincham 1992 Alternative splicing of the mouse amelogenin primary RNA transcript contributes to amelogenin heterogeneity. Biochem. Biophys. Res. Comm., 188r1253-1260. Lazarovits, J., and M. Roth 1988 A single amino acid change in the cytoplasmic domain allows the influenza virus hemagglutinin to be endocytosed through coated pits. Cell, 53r743-752. Lumsden, A.G.S. 1987 The neural crest contribution to tooth development. In: Developmental and Evolutionary Biology of the Neural Crest. P.F.A. Maderson, ed. John Wiley and Sons, New York, pp. 3-43. Lumsden, A.G.S., and J.A.G. Buchanan 1986 An experimental study of timing and topography of early tooth development in the mouse embryo with a n anlysis of the role of innervation. Archs. Oral Biol., 31:301-311. McMahon, D., and C. West 1976 Transduction of positional information during development. In: The Cell Surface in Animal Embryogenesis and Development. G. Poste, and G.L. Nicolson, eds. ElseviedNorth Holland Biomedical, New York, pp. 449-493. Mina, M., and E.J. Kollar 1987 The induction of odontogenesis in non-dental mesenchyme combined with early murine mandibular arch epithelium. Arch. Oral Biol., 32t123-127. Nanci, A., M. Bendayan, P. Bringas, Jr., and H.C. Slavkin 1984 High resolution immunocytochemical localization of enamel proteins in mouse ameloblasts in situ and in culture. In: Tooth Morphogenesis and Differentiation. A.B. Belcourt, and J.V. Ruch, ed. INSERM, Paris, Vol. 125,pp. 333-340. Nanci, A., M. Bendayan, and H.C. Slavkin 1985 Enamel protein biosynthesis and secretion in mouse incisor secretory ameloblasts as revealed by high resolution immunocytochemistry. J. Histochem. Cytochem., 33tll53-1160. Nanci, A., H.C. Slavkin, C.E. Smith 1987 Immunocytochemical and radioautographic evidence for secretion and intracellular degradation of enamel proteins by ameloblasts during the maturation stage of amelogenesis in rat incisors. Anat. Rec., 21 7t107-123. Nanci, A., J.P. Ahluwalia, J.R. Pompura, and C.E. Smith 1989 Biosynthesis and secretion of enamel proteins in the rat incisor. Anat. Rec., 224:227-291. O'Farrell, P.H. 1975 High resolution two-dimensional electrophoresis of proteins. J. Biol. Chem., 250r4007-4021. Ogata, Y., H. Shimokawa, and S. Sasaki 1988 Molecular and cellular biology. Purfication, characterization, and biosynthesis of bovine enamelins. Calcif. Tissue Int., 43~389-399. Pannese, E. 1962 Observations on the ultrastructure of the enamel organ. I11 Internal and external enamel epithelia. J . Ultrastruct. Res., 17r503-526. Pastan, I., and M.C. Willingham 1983 Receptor-mediated endocytosis: coated pits, receptosomes and the Golgi. Trends Biochem. Sci., 8r250-254. Rappolee, D.A., A. Wang, and D. Mark 1989 Novel method for studying mRNA phenotypes in single or small numbers of cells. J. Cell Biochem., 39rl-11. Reith, E.J. 1967 The early stages of amelogenesis as observed in molar teeth of young rats. J. Ultrastruct. Res., 17.503-526. Ruch, J.V. 1985 Epithelial-mesenchymal interactions. In: The Chemistry and Biology of Mineralized Connective Tissues. W. Butler, ed. EBSCO Media, Birmingham, Alabama, pp. 54-62. Ruch, J.V. 1988 Determinism of odontogenesis. Cell Biology Reviews, 14:l-86. Ruch, J.V., H. Lesot, V. Karcher-Djuricic, J.M. Meyer, and M. Olive 1982 Facts and hypotheses concerning the control of odontoblast differentiation. Differentiation, 21t7-12. Saxen, L., E. Lehtonen, M. Karkinen-Jaaskelainen, S. Nordling, and J. Wartiovaara 1976 Morphogenetic tissue interactions: Mediation by transmissible signal substances or through cell contacts? Nature, 259r662-663. Sigal, M.J., J.E. Aubin, A.R. Ten Cate, and S. Pitaru 1984 The odontoblast Drocess extends to the dentinoenamel iunction. J. Histochem.'& Cytochem., 32r872-877. 395 Silva, D.G., and D.G. Kailis 1972 Ultrastructural studies on the cervical loop and the development of the amelo-dentinal junction in the cat. Arch. Oral Biol., 17t279-289. Slavkin, H.C. 1972 Intercellular communication during odontogenesis. In: Developmental Aspects of Oral Biology. H.C. Slavkin, and L.A. Bavetta, eds. Academic Press, New York, pp. 165-201. Slavkin, H.C. 1974 Embryonic tooth formation. Oral Sci. Rev., 4tl- 139. Slavkin, H.C. 1988 Molecular biology of dental development: A review. In: Biological Mechanisms of Tooth Eruption and Root Resorption. 2. Davidovitch, ed. EBSCO Media, Birmingham, Alabama, pp. 107-116. Slavkin, H.C., and P. Bringas, Jr., 1976 Epithelial-mesenchymal interactions during odontogenesis IV: Morphological evidence for direct heterotypic cell-cell contacts. Dev. Biol., 50r428-442. Slavkin, H.C., C. Bessem, P. Bringas, Jr., M. Zeichner-David, A. Nanci, and M.L. Snead 1988a Sequential expression and differential function of multiple enamel proteins during fetal, neonatal and early postnatal stages of mouse molar organogenesis. Differentiation, 37t26-39. Slavkin, H.C., M. MacDougall, M. Zeichner-David, P. Oliver, M. Nakamura, and M.L. Snead 1988b Molecular determinants of cranial neural crest-derived odontogenic ectomesenchyme during dentinogenesis. Am. J. Med. Genet. (Suppl), 4r7-22. Slavkin, H.C., W. Mino, P. Bringas, Jr. 1976 The biosynthesis and secretion of precursor enamel protein by ameloblasts as visualized by autoradiography after tryptophan administration. Anat. Rec., 185r289-312. Slavkin, H.C., M.L. Snead, M. Zeichner-David, P. Bringas, Jr., and G.L. Greenberg 1984 Amelogenin gene expression during epithelial-mesenchymal interactions. In: The Role of Extracellular Matrix in Development. R.L. Trelstead, ed. Alan R. Liss, New York, pp. 221-253. Slavkin, H.C., M.L. Snead, M. Zeichner-David, M. MacDougall, A.G. Fincham, E.C. Lau, W. Luo, M. Nakamura, P. Oliver, and J. Evans 1988c Factors influencing the expression of dental extracellular matrix biomineralization. In: Cell and Molecular Biology of Vertebrate Hard Tissues. D. Evered, and S. Harnett, eds. Ciba Foundation Symposium 136. John Wiley and Sons, Chichester, pp. 22-41. Slavkin, H.C., M. Zeichner-David, M. MacDougall, P. Bringas, Jr., C. Bessem, and L. Honig 1982 Antibodies to murine amelogenins: Localization of enamel proteins during tooth organ development in vitro. Differentiation, 23r73-82. Snead, M.L., P. Bringas, Jr., C. Bessem, and H.C. Slavkin 1984 De novo gene expression detected by amelogenin gene transcript analysis. Dev. Biol., 104:255-258. Snead, M.L., E. Lau, M. Zeichner-David, A. Fincham, S.L.C. Woo, and H.C. Slavkin 1985 DNA sequence for cloned cDNA for murine amelogenin reveals the amino acid sequence for enamel-specific protein. Biochem. Biophys. Res. Comm., 12t812-818. Snead, M.L., W. Luo, E.C. Lau, and H.C. Slavkin 1988 Spatial- and temporal-restricted pattern of amelogenin gene expression during mouse molar tooth organogenesis. Development, 104:77-85. Suga, S. 1960 On the penetration of the enamel matrix substances to the dentine during amelogenesis, with special reference to its functional significance. Arch. Histol. Jpn., 20t477-498. Termine, J.D., A.B. Belcourt, P.J. Christner, K.M. Conn, and M.U. Nylen 1980 Properties of dissociatively extracted fetal tooth matrix proteins. I. Principal molecular species in developing bovine enamel. J. Biol. Chem., 255:9,760-9,768. Theiler, K. 1972 The House Mouse. Springer-Verlag, Berlin, pp. 109- 143. Thesleff, I. 1976 Differentiation of odontogenic tissues in organ culture. Scand. J . Dent. Res., 84t353-356. Thesleff, I. 1977 Tissue interactions in tooth development in vitro. In: Cell Interactions in Differentiation. M. Karkinen-Jaaskelainen, and L. Saxen, eds. Academic Press, London, pp. 191-207. Thesleff, I., and K. Hurmerinta 1981 Tissue interactions in tooth development. Differentiation, 18t75-88. Thesleff I., E. Lehtonen, and L. Saxen 1978 Basement membrane formation in transfilter tooth culture and its relation to odontoblast differentiation. Differentiation, 10t71-79. Towbin, H., T. Staehelin, J . Gordon 1979 Electrophoretic transfer of proteins from polyacrylamide gels to nitro-cellulose sheets: Procedures and some applications. Proc. Natl. Acad. Sci. USA, 76: 4,350-4,354. Trelstad, R.L., and H.C. Slavkin 1974 Collagen synthesis by the epithelial enamel organ of the embryonic tooth. Biochem. Biophys. - . Res. Comm., 59:43-47. 396 M. NAKAMURA ET AL. Verner, K., and G. Schatz 1988 Protein translocation across membranes. Science, 241 :1307-1313. Warshawsky, H. 1968 The fine structure of secretory ameloblasts in rat incisors. Anat. Rec., 161t211-230. Watson, M.L. 1960 The extracellular nature of enamel in the rat. J. Biophys. Biochem. Cytol., 7t489-497. Weinstock, A. 1972 Matrix development in mineralizing tissues as shown by radioautography: formation of enamel and dentin. In: Developmental Aspects of Oral Biology. H.C. Slavkin, and L.A. Bavetta, eds. Academic Press, New York, pp. 201-242. Weinstock, A,, and C.P. Leblond 1971 Elaboration of the matrix glycoprotein of enamel by the secretory ameloblasts of the rat incisor as revealed by radioautography after galactose-3H injection. J. Cell Biol., 51t26-43. Wileman, T., C. Harding, and P. Stahl 1985 Receptor-mediated endocytosis. Biochem. J., 232:l-14. Wolters, J.M.L. 1978 The transfilter transmission of [3H]-proline labelled materials in cultured rat tooth germs. Archs. Oral Biol., 235-55. Yamada, M., P. Bringas, Jr., M. Grodin, M. MacDougall, and H.C. Slavkin 1980 Developmental comparisons of murine secretory amelogenesis in vitro. Calcif. Tiss. Int., 31:161-171. Yamamoto, K., T. Nakata, Y . Tsuji, E. Kitano, T. Nishimoto, and M. Akai 1980 On the penetration of stippled material-like substances into the dentine. An electron microscopic study. J . Osaka Univ. Dent. Sch., 20:119-128.