Immunocytochemical and radioautographic evidence for secretion and intracellular degradation of enamel proteins by ameloblasts during the maturation stage of amelogenesis in rat incisors.код для вставкиСкачать
THE ANATOMICAL RECORD 217:107-123 (1987) Immunocytochemical and Radioautographic Evidence for Secretion and lntracellular Degradation of Enamel Proteins by Ameloblasts During the Maturation Stage of Amelogenesis in Rat Incisors ANTONIO NANCI, HAROLD C. SLAVKIN, AND CHARLES E. SMITH Dkpartements de stomatologie et d’anatomie, Universitt de Montreal, Montrkal, Quebec, Canada H3C 3J7 (A.N.); Laboratory for Developmental Biology, Graduate Program in Craniofacial Biology, School of Dentistry, University of Southern California, Los Angeles, C A 90089 (H. C.S.); and Departments of Anatomy and Oral Biology, McGill University, Montreal, Quebec, Canada H3A 2B2 (C.E.S.) ABSTRACT In the continuously erupting rat incisor the ameloblasts progress through distinct stages associated with the secretion and maturation of enamel. We have examined the possibility that the so-called “postsecretory” ameloblasts of the maturation stage of amelogenesis remain biosynthetically active and are engaged in the synthesis, secretion, and degradation of enamel gene products. The ultrastructural distribution of antigenic sites for enamel proteins was studied within enamel organ cells during the early maturation stage of amelogenesis in rat incisors by using the protein A-gold immunocytochemical technique and rabbit polyclonal antibodies developed against mouse amelogenins. All regions of amelogenesis from late secretion through the first complete modulation from ruffle-ended to smooth-ended ameloblasts were examined. Specific immunolabelling was found within the rough endoplasmic reticulum, Golgi saccules, secretory granules, and lysosomes of ameloblasts throughout these regions. The heaviest intracellular immunolabelling was found within secretory granules and lysosomes (multivesicular type). Quantitative analyses showed that the Golgi saccules and the multivesicular lysosomes of modulating ameloblasts were generally less immunoreactive compared to similar organelles in ameloblasts secreting the inner enamel layer. Radioautographic studies confirmed that ameloblasts of the maturation stage incorporated 3H-leucineand 3Hmethionine and secreted labelled proteins into the enamel layer. Grain counts indicated that ameloblasts from the first ruffle-ended band incorporated about twofold less 3H-methionine and secreted about tenfold less labelled proteins into the enamel compared to ameloblasts secreting the inner enamel layer. The results of this study confirm that ameloblasts do not terminate biosynthesis and secretion of enamel proteins once the final layer has been deposited on the surface of the developing enamel. They continue to form and release new proteins during the maturation stage which intermix with older proteins laid down initially during the secretory stage of amelogenesis. Secretory activity for enamel proteins has been detected in ameloblasts up to at least the second ruffle-ended phase of maturation, at which point the enamel matrix is partially soluble in EDTA. Several authors have suggested that certain epithelia1 cells have the capacity to down regulate the production of specific gene products using intracellular degradative mechanisms (reviewed in Bienkowski, 1983). One model for critically examining these possibilities can be found in developing enamel in which ectodermally derived ameloblasts produce an extracellular matrix of finite thickness (reviewed in Smith, 1979). The organic matrix of enamel is believed to contain two main subfamilies of proteins termed amelogenins and enamelins (reviewed in Eastoe, 1979; Termine et al., 0 1987 ALAN R. LISS, INC. 1980; Robinson et al., 1983; Shimizu and Fukae, 1983; Fincham and Belcourt, 1985; Robinson and Kirkham, 1985). Amelogenins are found in highest concentrations within newly formed and partially mineralized “young” enamel. Enamelins comprise the major protein associated with the mineral in fully “mature” enamel (references above). Current biochemical evidence suggests that small amounts of enamelins are present along with Received June 2, 1986; accepted August 12, 1986. 108 A. NANCI, H.C. SLAVKIN. AND C.E. SMITH the amelogenins in newly formed enamel (Termine et al., 1980; Robinson et al., 1983; Shimokawa et al., 1984; Zeichner-David et al., 1985). Hence, enamel maturation is viewed as a process which results in the breakdown and selective loss of amelogenins with the preferential retention of the enamelins (reviewed in Robinson and Kirkham, 1985). Radioautographic studies by several workers have indicated that ameloblasts incorporate the greatest amounts of labelled amino acids during the secretory stage of amelogenesis (Slavkin et al., 1976; Warshawsky, 1979; Blumen and Merzel, 1982; Takano and Ozawa, 1984; see also Robinson et al., 1982). Such findings are consistent with the idea that these cells actively synthesize and secrete large quantities of matrix proteins (amelogenins and enamelins) in order to create a n enamel layer by appositional growth. There is additional evidence suggesting that ameloblasts also form exportable proteins after the enamel layer has developed to its full thickness (Weinstock, 1972; Warshawsky, 1979; Warshawsky and Josephsen, 1981). The exact chemical nature of these proteins remains obscure, but it is believed that they may represent components of a basal-lamina-like structure which appears on the surface of the enamel near the beginning of the maturation stage of amelogenesis (Weinstock, 1972; Takano, 1979). It is also possible that a portion of these proteins could constitute newly formed enzymes such as proteases (“amelogenases”) which some workers believe are secreted extracellularly to cleave the amelogenins into small molecular weight fragments thereby promoting enamel maturation (reviewed in Carter et al., 1984; Crenshaw and Bawden, 1984). No evidence presently exists that part, or all, of this material is newly synthesized enamel proteins secreted by the ameloblasts during the maturation stage of amelogenesis. We originally undertook this investigation to probe the distribution of antigenic sites to enamel proteins with the lysosomal system of the ruffle-ended ameloblasts of the maturation stage of amelogenesis. These cells are believed to internalize fragments of partially degraded amelogenins and digest them via the lysosomes (Reith and Cotty, 1967; Kallenbach, 1974; Ozawa et al., 1983; Sasaki, 1984b). We assumed that if these cells resorb enamel proteins, then there should be a subpopulation of lysosomes containing material with sflicient antigenicity to be detected by high-resolution immunocytochemistry. Over the course of this investigation it became apparent that ameloblasts associated with maturing enamel showed not only immunoreactive lysosomes but also a pattern in the intracellular distribution of antigenic sites to enamel proteins reminiscent of the one reported recently for ameloblasts from secretory stage of amelogenesis (Nanci et al., 1985). Hence, it became necessary also to reevaluate the synthetic potential of the modulating ameloblasts by conventional light microscopic radioautography. barbital (M.T.C. Pharmaceuticals, Hamilton, Ontario). The left ventricle was cannulated and the vascular system was flushed for about 30 seconds with lactated Ringer’s solution (Abbott Laboratories, Montreal). Four animals were perfused for 10 minutes with 1% glutaraldehyde in 0.1 M or 0.08 M sodium cacodylate buffer, pH 7.3, containing 0.05% CaC12.Two animals were perfused with 4% paraformaldehyde in 0.1 M sodium cacodylate buffer, pH 7.4. The mandibles were removed and immersed for 1 hour in fresh fixative at 4°C. They were then washed briefly in cold 0.1 M sodium cacodylate buffer, pH 7.4, containing 4% sucrose (CS buffer) and decalcified in 4.13% disodium EDTA (BDH Chemicals, Montreal) for 14 days at 4°C (Warshawsky and Moore, 1967).The mandibles were washed for 2 days in frequent changes of cold CS buffer. Excess bone and molars were removed, and most incisors with surrounding alveolar bone were subdivided into four segments about 5 mm in length. Each segment was then split in half by passing a razor blade down the midline of the segment (Smith, 1974). Some of the hemisegments were postfixed for 1 hour a t 4°C in 1.5%potassium ferrocyanide/l% osmium tetroxide (reduced osmium; Karnovsky, 1971). The hemisegments were dehydrated in graded acetone, infiltrated, and flat embedded with Epon 812 substitute (MECA Laboratories, Montreal) so as to produce longitudinal sections of the enamel organ. Other unosmicated tooth specimens were dehydrated in graded methanol, infiltrated, and embedded in Lowicryl K4M (Bendayan, 1984)for cross sections of the enamel organ. Thin sections of selected areas were cut with a diamond knife on a Reichert Ultracut E microtome, mounted on 200-mesh nickel grids having a carbon-coated Formvar film, and processed for immunocytochemistry. Radioautographicstudies Twelve male Wistar rats weighing 109 5 g were anesthetized with sodium pentobarbital and injected via the external jugular vein with 0.1 ml of a normal saline solution containing 1mCi of either [meth~l-~HI-L-methionine (s.a. 80 Ci/mM) or [3,4,5, -3H (NI]-L-leucine (s.a. 147 Ci/mM) (Dupont NEN, Boston). The radioactive amino acids had been concentrated in a Speed Vac (Savant Instrument Inc., Farmingdale, NY) and resuspended in normal salinejust prior to use. After 6 minutes the rats were injected via the opposite external jugular vein with 0.1 ml of a normal saline solution containing a n excess of %old” L-methionine or L-leucine (10 mM) (Sigma Chemical Co., St. Louis, MO). The rats were killed in pairs at 20 minutes, 1hour, and 4 hours after the initial injection of radioactive tracer by first perfusing the vascular system via the left ventricle with lactated Ringer’s for about 30 seconds and then with fixative for 20 minutes. Rats from the 20-minute group were perfused for 4 minutes with 4% paraformaldehyde in 0.08 M sodium cacodylate buffer, pH 7.4., containing 0.05% CaC12 (done to avoid nonspecific binding of amino acids to protein; see Josephsen and Warshawsky, 1982). MATERIALS AND METHODS This was followed by 5% glutaraldehyde in 0.05 M soTissue Preparation dium cacodylate buffer, pH 7.3, containing 0.05% CaC12 for 16 minutes. The remaining rats were perfused dilmrnunocytochemicalstudies rectly for 20 minutes with the glutaraldehyde fixative Six male Wistar rats weighing 100 f 10 g body weight solution after vascular prewashing. The mandibles were (Charles Rivers Canada, St.-Constant, Quebec) were an- removed and immersed for a total of 4 hours in fresh esthetized by intraperitoneal injection of sodium pento- fixative a t 4°C. The mandibles were washed overnight SECRETORY ACTIVITY IN MODULATING AMELOBLASTS at 4°C in CS buffer on a rotator, decalcified in 4.13% disodium EDTA for 16 days, washed, and split into hemisegments as described above. All hemisegments were postfixed for 2 hours at 4°C in reduced osmium, dehydrated in graded alcohol and propylene oxide, and flat embedded in Epon 812 substitute. A series of l-pmthick sections was cut with glass knives on a Reichert Omu4 or Ultracut E microtome. The first sections from each block were placed on separate glass slides and stained with toluidine blue. The remaining sections were mounted on alcohol-cleaned glass slides and stained with Regaud’s hematoxylin. Three slides were prepared from each block with 3 to 4 sections per slide. The slides were dipped in Kodak NTB2 liquid emulsion, exposed for 4 days (two slides, for quantitation) or 14 days (one slide, for photography), and developed (Kopriwa and Leblond, 1962). Cytochemical studies Eleven male Sprague-Dawley or Wistar rats weighing 100-150 g were anesthetized with chloral hydrate (0.4 mg/g body weight) and perfused for 8-10 minutes via the left ventricle with 2% or 2.5% glutaraldehyde in 0.1 M or 0.08 M sodium cacodylate buffer, pH 7.3, containing 0.05% CaC12. In some experiments 2% dextran (40,000 MW, Pharmacia, Montreal) was included in the fixative solution, and the vascular system was flushed with lactated Ringer’s prior to perfusion with fixative. The mandibles were removed and immersed in fresh fixative for 1 hour a t 4°C. They were washed with cold CS buffer for about 2 hours and then decalcified in 4.13% EGTA (Eastman Kodak Co., Rochester) for 5 or 14 days a t 4°C with constant agitation. The mandibles were washed in frequent changes of CS buffer for 2 days a t 4°C on a rotator. In some cases enamel organs were dissected from the labial surfaces of incisors in three parts corresponding to the secretory stage and the early and the late maturation stages of amelogenesis. In other cases whole incisors were cut grossly into three segments containing one of these same three stages. The location of cross cuts demarcating these stages was established by measurements taken with a ruler through the eyepieces of a dissecting microscope. All dissections were done in a petri dish filled with ice-cold CS buffer. The segments were embedded in 7% agar and cut at 5070 pm thickness on a Sorvall TC-2 tissue chopper, and the sections were washed in CS buffer and processed for cytochemistry. lmmunocytochemical Labelling The protein A-gold technique was used for detecting antigenic sites in thin sections from Lowicryl- or Eponembedded material (Bendayan, 1984). Thin sections cut from osmicated and Epon-embedded material were treated with sodium metaperiodate for 1 hour at room temperature, washed with distilled water, and incubated with rabbit polyclonal antibody against SDS-denatured mouse amelogenins (U10 dilution) followed by protein A-gold as described previously (Nanci et al., 1985). The sections then were grid stained with uranyl acetate and lead citrate and examined with a Philips 300 or 410 electron microscope. Controls consisted of incubating sections with 1)antibody that had been exposed to a n excess of amelogenin, 2) pre-immune serum, or 3) protein A-gold alone (see Nanci et al., 1985). The specificity of the antibody used in this study has been 109 documented elsewhere (Slavkin et al., 1982, 1984). It has been found to cross-react with amelogenins and enamelins derived from mouse as well as a variety of other mammalian species (Slavkin et al., 1984). Only results from glutaraldehyde-fixed, osmicated, and Epon-embedded material are presented in this report because this procedure gave the best balance between labelling intensity, background, and resolution. Cytochemistry Tissue chopper sections were washed and incubated for 15, 30, 60, or 90 minutes at 37°C and pH 5.0 in conventional acid phosphosphatase media by using pglycerophosphate (BGP) (Gomori, 1950) or cytidine 5‘monophosphate (CMP) (Novikoff, 1963) as substrates. Some sections were also incubated for 20 or 40 minutes a t 37°C and pH 3.9 in the medium described by Doty et al. (19771, with inorganic trimetaphosphate (TMP) as substrate. Controls for all experiments consisted of incubating separate sections in duplicate media prepared with all components except substrate. Following incubations the sections were washed, osmicated, and processed for embedding in Epon as described previously (Smith, 1981). Thin sections were cut on a Reichert Omu4 ultramicrotome, grid stained with lead citrate, and examined in a Philips 400 electron microscope. Some thin sections were mounted on nickel grids, treated with sodium metaperiodate, and incubated with antibody as described above. Quantitative Analyses Distribution of gold particles (immunocytochernistry) Nonoverlapping and randomly selected micrographs were taken in the supranuclear compartment and along the apical surfaces of ameloblasts situated near the junction between a n initial smooth-ended band (SEo) and the first true ruffle-ended band of maturation (RE1) (see Fig. 2 and Results sections for details). Similar micrographs were also taken from ameloblasts engaged in early and mid-inner enamel secretion (IES). Prints were made and a Zeiss MOP-3 manual image analyzer was then used to compute profile areas within which gold particles were located (Tables 1, 2). Data were expressed directly in micrometers squared by using the internal correction capability of the MOP, and data for organelles other than lysosomes were pooled across individual prints. Density of labelling was computed separately for each set of individual (lysosomes) or pooled (other organelles) area measurements and particle counts. Statistical tests were carried out on resultant means by using H P Statistical Library routines run on a n HP series 9000 model 216 microcomputer (Hewlett-Packard Desktop Computer Division, Fort Collins, CO). It was estimated that a total of at least 100 ameloblasts were sampled at each site quantitated for this study (Tables 1, 2; IES, SEo, RE1). Distribution of silver grains (radioautography) Silver grains were counted with the aid of a Whipple ocular grid mounted in a self-focusing eyepiece on Wild light microscope. The counts were done by using a n oilimmersion objective lens within rectangular areas of the grid measuring 1 vertical x 3 horizontal squares or 8.6 pm x 25.8 pm a t the magnification used. The location for sampling within the secretory zone of amelogenesis 110 A. NANCI, H.C. SLAVKIN, AND C.E. SMITH TABLE 1. Immunocytochemistry: Density of gold labelling over enamel and organelles in ameloblasts following incubation with antiamelogenin antibody No. of Compartment Secretion stage Inner Enamel Secretion (IESI3 En ame1 Ameloblast Nucleus Mitochondria Endoplasmic reticulum Golgi saccules5 Secretion granules6 Lysosomes Totals Maturation stage Smooth-ended (SE01 Enamel4 Ameloblast Nucleus Mitochondria Endoplasmic reticulum Golgi saccules5 Secretion granules6 Lysosomes Totals Ruffle-ended (REJ Enamel4 Ameloblast Nucleus Mitochondria Endoplasmic reticulum Golgi saccules5 Secretion granules6 Lysosomes Totals individual or summed fields’ Total No. of particles counted (No.) Total enclosed area Density of labellins2 (pm2) (No./pm ) 31 1,024 9.7 121.4 f 53.7 8 21 28 50 24 408 440 233 452 2,769 408 2,240 7,566 186.3 59.1 53.5 206.4 3.1 116.1 634.2 2.6 f 1.4 3.8 f 2.7 8.4 2.3 15.1 f 7.1 137.6 & 40.7 17.4 f 40.7 15 620 6.1 101.1 f 22.4 14 51 22 46 15 458 278 620 314 973 183 1,638 4,626 67.1 166.5 36.7 100.1 1.5 169.1 547.1 4.0 f 1.9 3.8 f 2.3 8.5 f 2.5 8.9 f 5.7 135.5 f 63.1 10.4 & 17.4 15 463 8.1 33 55 36 58 9 599 607 712 743 1,305 81 3,707 7,618 133.8 156.1 76.9 136.7 4.1 230.8 746.5 Background Background 58.8 = 3.4 f 2.47 = 3.8 f 2.Z7 = 4.8 f 3.37 20.8 5.3 f 4.5 4.4 f 2.2 9.6 f 2.9 10.6 5.7 140.2 f 83.5 19.0 f 25.8 Background ‘Samples were pooled in all cases except for lysosomes where the number of fields equals the number of individual lysosomal profiles measured. ‘Mean k standard deviation. 3Data pooled from early and mid-inner enamel secretion, 4Data from small areas of matrix sampled just distal (deep)to the Tomes’ processes (IES)or the basal lamina covering the surface of the enamel layer (SEo,RE,). 5Area measurements included the cytoplasmic spaces between the saccules. ‘Immature and/or mature secretory granules encountered near the Golgi apparatus. 7Average of nuclear and mitochondrial labelling; other values were significantly higher compared to this background as evaluated by the Student’s t-test (P< ,051. was selected arbitrarily a t a point where the thickness of enamel from the base of interrod prongs to the dentino-enamel junction (DEJ) occupied seven vertical rectangles. This location corresponded roughly to the thickness of enamel through which newly formed proteins seemed to diffuse by 4 hours after injection of 3Hmethionine. Ameloblasts at this position were slightly beyond the midpoint in the secretory zone, and they were forming inner enamel. For animals at 20 minutes after injection the horizontal lines of the grid were aligned parallel to DEJ and the edges of the prongs of interrod material a t the apex of the ameloblasts, and the number of silver grains over the Tomes’ processes andor matrix in each rectangle was counted. The grid was then moved and superimposed over the cells adjacent to the area of enamel just counted. Care was taken to align the top of the grid at the same position corresponding to the base of the preceding counts. The num- ber of silver grains over cytoplasm andor nuclei of cells was scored across all rectangles extending vertically from the apex of ameloblasts to the outer basement membrane of the enamel organ. This process was repeated for two sections per slide, and for each of two slides per tooth yielding a total of 16 samples per position per time interval from which means and standard deviations were computed after corrections for background labelling (estimated at 0.75 & 1.2 grains per rectangle based on counts done over areas of plastic devoid of tissue; therefore, the correction factor for background = 0.75 x [the number of rectangles occupying a compartment of interest]). For animals a t 1and 4 hours after injection the grid lattice was advanced incisally from the seven-rectangle-thick position a n amount equal to the estimated average migration occurring relative to the animals at 20 minutes after injection (baseline) (18 pm and 99 pm, respectively; estimated from data in 111 SECRETORY ACTIVITY IN MODULATING AMELOBLASTS TABLE 2. Immunocytochemistry: Distribution of reactive and unreactive lysosomes in ameloblasts following incubations with antiamelogenin antibody Overall No. Lysosomal t y p e No. labelled No. No. Density of labelling3 weakly heavily labelled4 labelled5 observed Frequency density of labelling’ 137 159 112 34 39 27 8.5 f 17.7 11.1 & 18.5 37.0 & 68.3 52 (38) 75 (47) 69 (62) 22.5 f 22.8 22.6 f 21.8 59.5 f 79.2 26 (50) 39 (52) 27 (39) l(2) l(1) 13 (19) 206 85 167 45 19 36 4.5 & 7.1 9.8 k 21.5 18.1 20.9 + 70 (34) 37 (44) 124 (74) 12.3 f 7.3 21.6 28.7 23.9 k 21.4 48 (69) 23 (62) 56 (45) 0 (0) l(3) 2 (2) 198 137 264 33 23 44 7.4 & 13.1 9.8 k 14.3 32.4 f 31.0 82 (41) 70 (51) 214 (81) 17.2 & 15.9 18.0 5 16.2 39.5 f 30.3 49 (60) 40 (57) 46 (21) 0 (0) 0 (0) 11 (5) above background’ Secretion stage Inner enamel secretion (IES)~ Dark Pale Multivesicular Maturation state Smooth-ended (SE01 Dark Pale Multivesicular Ruffle-ended (RE1) Dark Pale Multivesicular ‘Mean number of gold particles per pm2 (+ standard deviation) for all lysosomes in each class. %ee Table 1 for definition of background; values in parentheses indicate percent labelled lysosomes in each class. 3Mean number of gold particles per pm2 (* standard deviation) for labelled lysosomes in each class. 4Lysosome~showing ten + background or fewer gold particles per pm2; values in parentheses indicate the percent of labelled lysosomes that were weakly labelled. 5Lysosome~showing 100+background or more gold particles per pm2; values in parentheses indicate the percent of labelled lysosomes that were heavily labelled. ‘Data pooled from early and mid-inner enamel secretion. a BL cv dl IES inv isg mvb n Pl S sg tP Abbreviations Apical surface of ameloblast Basal lamina Coated vesicle Dense lysosome Inner enamel secretion (secretory stage) Apical invagination Immature secretory granule Multivesicular body Nuclear compartment of ameloblast Pale lysosome Supranuclear compartment of ameloblast Secretory granule Tomes’ process Regional classification (from most apical to most incisal) NTP ASH PST RE SE LPG NPG RE0 GM No Tomes’ process Ameloblasts shrink in height Postsecretory transition (combined NTP ASH) RuMe-ended (subscript indicates sequence beginning with the initial or ‘0’band) Smooth-ended (as above) Large pigment granules No large pigment granules Reduced enamel organ Gingival margin + Smith and Warshawsky, 1975). The sampling position for counts within the maturation stage of amelogenesis was located at a point 1mm incisally from where Tomes’ processes were no longer evident at the apices of the tall columnar ameloblasts (see Figs. 1, 3). Ameloblasts at this site in all teeth counted were engaged in the first ruffle-ended phase of maturation (RE,; see Fig. 3). Counts were made vertically in rectangular areas across enamel and adjacent cells of the enamel organ as described above. The density of silver grains was estimated by dividing the total number of grains over enamel or cells by the total number of rectangular areas within which grains were distributed (Table 3). Al- Fig. l . Mesial side of the left mandible from a male Wistar rat weighing about 100 g body weight. The apical (NTP) and incisal (GM) limits of regions defined relative to the maturation stage of amelogenesis in the incisor are indicated by the white arrows. The site marked “NTP’ corresponds to the origin, or zero point, for the length measurements presented in Figures 2-4. X 3 . though only the data from animals injected with 3Hmethionine are presented in this report the qualitative findings from animals injected with dH-leucine were identical to those for methionine. Length measurements Lengths of various regions within the maturation zone of amelogenesis (Figs. 2,3) were estimated directly from a light microscope by using 1-pm-thick toluidine-bluestained sections. The microscope, equipped with a drawing tube, was mounted over the magnetic tablet of the MOP, and a light-emitting cursor was traced along the apices of ameloblasts to measure the apicalhncisal 112 A. NANCI, H.C. SLAVKIN. AND C.E. SMITH Mandibular Incisor Mandibulor Incisor -3 U lOOg rots - ) . u m +’ 1600- m 0) W 0 -E, soo- -2 I 1200- b. .+ I s - 1 w E t - 400- .- 6 2 0 z Q 0 f -0 0 .A- O - Enomel o Enomel w NO Enamel w - - Timemes I 2 3 4 5 7 6 10 9 S ii 12 Giniivol morgin Distance ( rnm 1 process Fig. 2. Bar graph summarizing data from length measurements in various sequential regions defined by the appearance of the enamel matrix in EDTA-decalcifiedincisors. The height of each bar represents the mean length standard deviation (SD). The right ordinate shows a time scale based on the cell migration studies by Smith and Warshawsky (1975). At the bottom of the figure the length of each region (fSD) is represented to scale along a n abcissa beginning at the point where the Tomes’ processes were no longer visible at the apices of ameloblasts and extending incisally to the gingival margin. O 3200- i5 Mondibular Incisor I n 2800+m C 2400- --E, I 2000 - I 1600. 0 C ‘5 1200- - V E .A- 0 800- f w . NTP ASH RE, PST !E4 SE, R E 5 L f f i NPG R E 0 I M I i w ‘ I N o Tomes process ... ... . .. . . . .... ... .... ... .. . . .. . .. .. .. .. .. . . .. . SEO ........ ................... . hY \ 2 3 4 5 6 7 8 ... ... . .. : hW N 1 1 cc--(w 9 Distance ( m m I 1 0 1 1 . ..... . :. I ] w 1 1 2 Gingival rnarqm Fig. 3. Bar graph summarizing data from length measurements in various sequential regions defined by the morphology of ameloblasts. The height of each bar represents the mean length standard deviation (SD). The right ordinate shows a time scale based on the cell migration studies by Smith and Warshawsky (1975). At the bottom of the figure the lengths of each region are represented to scale along a n abcissa beginning at the point where the Tomes’ processes were no longer visible at the apices of ameloblasts and extending incisally to the gingival margin. The lines under the long bar indicate the mean position (k SD) for points of transition from smooth-ended ( S E p 4 ) (shaded) to ruffle-ended (REo-5) bands, and for the location of the gingival margin. The region marked “REo/SEo” (stippled bar third from left) was variable in its composition and contained ameloblasts displaying either an incompletely ruffled apical surface or a true smooth-ended apical surface or a combination of both (with smooth-end cells usuallv abuttine RE,). The uresence of distinct accumulations of mitochondria at the Gase df apicaf imaginations provided one criterion by which ameloblasts of the first true ruffle-ended band (RE,) could be distinguished from those forming a partially ruffle-ended band (REo) when smooth-ended cells (SE,) were not present in this region. + I 2 3 4 5 6 7 8 9 1 0 1 1 1 2 Distance (rnrn ) Fig. 4. Probability maps showing the location of boundaries between sequential regional compartments measured relative to the enamel matrix (top panel) or the ameloblasts (bottom panel). The line bisecting each map represents the average or mean location of a boundary in all incisors (50% probability), as is illustrated for the cellular compartments by the horizontal bar running across the middle panel. In these maps a vertical boundary line would mean that the compartment began (or terminated) at the same location in all teeth whereas a tilted line means the reference point was distributed over a wider range of values between incisors. The map in the lower panel further demonstrates that certain compartments were not observed on all teeth (e.g., an SE4 band was present on only about 75% of all incisors used in this study). Note these maps suggest that there is a 100% probability the first complete modulation (RE, to SE1) will occur relative to matrix which is insoluble in EDTA and in which the enamel rods usually are clearly visible. There is a 70% probability the second modulation (RE2 to SE2)will occur before the matrix is soluble in EDTA. length of various regions. Fields were enlarged at x 200 magnification and data were converted into micrometers by using the internal correction capability of the MOP. Measurements were taken on one section per segment per animal, and the data from 21 teeth were used to obtain the means and standard deviations shown in Figures 2 and 3. The probability lines in Figure 4 were computed by first mapping the position of the regions for each incisor on graph paper and then determining the cumulative number of teeth containing the region in sequential 250-pm steps from the apical reference point (no Tomes’ process) incisally to the gingival margin. In Figures 3 and 4 the transitional areas from ruffle-endedto smooth-endedameloblasts (Josephsen and Fejerskov, 1977) were included with the ruffle-ended bands. RESULTS Location and Morphological Features of Regions Sampled Relative to Early Maturation The classifications Of Warshawsky and Smith (1974)9 Kalknbach (197419 Josephsen and FeJerskov (19771, Takano (1979), and Inage and Toda (1984) were used to SECRETORY ACTIVITY IN MODULATING AMELOBLASTS divide the maturation zone of amelogenesis into regions arrayed sequentially from the point marking the site where Tomes’ processes were no longer visible at the apices of arneloblasts’ to the gingival margin (Figs. 14). Across these regions characteristic changes in the appearance of the enamel matrix and in the morphology of ameloblasts were evident at the level of the light microscope in EDTA-decalcified incisors. Compartments defined relative to the enamel matrix (Fig. 2) included regions within which the matrix 1)was deeply stained and the outlines of enamel rods were invisible or indistinct, 2) contained clearly visible enamel rods, 3) shrank in thickness and the rods became pale staining, and 4) was absent and represented by a clear space. Compartments defined relative to ameloblasts (Fig. 3) included regions within which the cells 1)were tall in height but lacked Tomes’ process (NTP- a list of abbreviations and their meanings precedes Fig. 11, 2) shrank in height (ASH), 3) displayed variable apical morphology consisting either entirely of a poorly defined ruffle-end (mostfrequent) or a short segment of incompletely ruffled cells followed by true smooth-ended ameloblasts (REdSEo) 4) possessed a true ruffled apex (RE1, RE2, etc.), 5) were smooth-ended with lateral intercellular spaces (SE1, SE2, etc.), 6) contained large pigment granules in both the supranuclear and infranuclear compartments (LPG), 7) were devoid of these granules (NPG), or 8)were cuboidal in shape (REO). Measurements taken from 1-pm-thick plastic sections showed that all regions could be identified fairly consistently within incisors from different animals despite wide variations in the exact location of certain transition points such as the start of EDTAsoluble enamel or the location of smooth-ended bands (Figs. 2-4). These data also indicated that the length of a n incisor over which detectable quantities of enamel proteins were present in maturing enamel (regions of indistinct rods + visible rods partially soluble enamel) on the average was almost identical to the exact length of the secretory zone of amelogenesis (4,688*520 pm vs. 4,618 f 296 pm, respectively, as measured in Wistar rats weighing 1OOg). Furthermore, ameloblasts generally completed only two modulations, or they were engaged in the third ruffle-ended phase, relative to the change from EDTA-insoluble to EDTA-soluble enamel (Fig. 4).Regions of interest to this study included those representing transition and reorganization from secretion to maturation (NTP, ASH, REdSEo in Fig. 3) and initial modulation (RE1, SE1, RE2 in Figs. 3 and 4). These regions occupied about 2.5 mm at the apical aspect of the maturation zone of amelogensis in rat incisors. + lrnrnunocytochernicalLabelling Distribution of gold particles Enamel organ sections from regions of early maturation incubated with antiamelogenin antibodies showed specific labelling with gold particles. Gold particles were found in the heaviest concentration over the enamel layer, over granular material situated in the extracellular spaces along the lateral sides or within apical invaginations of ruffle-ended ameloblasts and over the protein synthetic (rough ER, Golgi saccules, secretory granules) and degradative (lysosomes) organelles of ameloblasts (Table 1,Figs. 5-12). Nuclei and mitochondria (Fig. 6) rarely contained labelling above a n amount 113 attributed to background (Table 1; see also Nanci et al., 1985). Occasional degenerating ameloblasts were observed which showed accumulations of immunoreactive material in the cytoplasm, and small focal patches of immunoreactive granular material were sometimes observed between papillary layer cells near the bases of ameloblasts (not shown). No significant labelling was found over coated vesicles a t the apical surfaces of ameloblasts (Fig. 6) or over other intracellular and extracellular sites related to cells of the papillary layer including macrophages. Gold particles were almost completely absent over enamel and over ameloblasts within control sections incubated under conditions where antibody was preabsorbed with excess antigen (Fig. 9). Distribution of irnmunoreactiveameloblasts Ameloblasts with obvious immunoreactive Golgi saccules, secretory granules, and lysosomes were observed continuously across all regions up to the beginning of the second ruffle-ended phase of maturation (NTP, ASH, REdSEo, REl,SEz,REZ, Fig. 3; Figs. 10, 11).Although no major differences in the pattern of intracellular labelling were obvious in these morphologically different ameloblasts, a gradual decrease in the intensity of immunolabelling was evident incisally across these regions especially in relation to Golgi saccules, small secretory granules, and the invaginations at the apical surfaces of ruffle-ended ameloblasts (e.g., Figs. 7, 8). Ameloblasts which contained ferritin pigment in the cytoplasm or the lysosomes (RE2 and beyond) showed little immunolabelling except over lysosomes (Fig. 16) and occasional residual granular material in the apical invaginations of ruffle-ended cells. Distribution of irnrnunoreactive lysosornes in arneloblasts Lysosomes were identified on the basis of cytochemical localizations by using acid phosphatase (AcPase) and inorganic trimetaphosphatase (TMPase) as enzyme markers (Novikoff, 1963; Doty et al., 1977). Based on these experiments the lysosomes in ameloblasts were divided into three morphologically distinct groups-the consistently TMPase-positive dark lysosomes and the variably TMPase-positive and AcPase-positive pale lysosomes and multivesicular bodies (Figs. 13-15). All types were variably immunoreactive, but the multivesicular bodies showed the greatest frequency and intensity in labelling with gold particles (Table 2). Immunoreactive lysosomes were found in ameloblasts throughout all regions where these cells abutted EDTAinsoluble enamel (Fig. 4). Ameloblasts sampled from areas of EDTA-soluble enamel (probably late RE2 or RE3) still contained some immunoreactive lysosomes. Dual immunocytochemical and cytochemical localizations ‘It is a n oversimplification to equate the disappearance of Tomes’ processes with the exact beginning of the maturation stage of amelogenesis. For example, the final enamel layer is rod-free and secreted by ameloblasts which have no Tomes’ process. The definition of postsecretory transition given by Warshawsky and Smith (1974) (used in this study) takes into account the overlap of events between cessation in appositional growth of the enamel layer and the completion of morphological changes associated with maturation of the enamel. Other classifications, such as the one given by Kallenbach (1974), define postsecretory transition exclusively in the context of the physical shortening of ameloblasts. The difference between these two classifications constitutes 255 f 87 pm linear distance or about 9.4 hours of cell time in the rat incisor. 114 A. NANCI, H.C. SLAVKIN, AND C.E. SMITH SECRETORY ACTIVITY IN MODULATING AMELOBLASTS 115 Fig. 9. Immunocytochemical preparations showing oblique cuts through the apical portions of ruffle-ended ameloblasts @El) incubated with antibody (A) and with antibody that was preabsorbed with excess antigen (B).The reaction present over the enamel and granular material within the infoldings of apical invaginations (inv) is abolished under these control conditions thereby demonstrating the specificity of labelling with gold. Note in Figure 9A that some structures (arrows) contain immunoreactive material which lack the halos normally seen in apical invaginations (inv). ~26,000. suggested that heavily labelled lysosomes were usually TMPase negative (Figs. 13, 14). Analysis of lysosomal types In early MIlooth-ended(SEO) and IWEIe-ended (RE11 ameloblasts indicated that the smooth-ended cells contained more dense type and fewer mult~Vesiculartype compared to ruffle-ended ameloblasts (Table 2). The 1YSOsomes in ruffle-endedcells also appeared more heavily labelled than in smooth-ended ameloblasts (Table 2). Radioautographic Studies Sites of incorporation of tracer at 20 minutes after injection At 20 minutes after injection of SH-methionine(or 3Hleucine) silver grains were present Over all cells of the enamel organ throughout the entire maturation zone. The heaviest reaction was seen over the supranuclear compartments and apical surfaces of ameloblasts abutting EDTA-insolubleenamel esDeciallv in reeions at the begnning of the maturation zine (NYP, ASH, SE&Ed RE1) (Figs. 3, 4, 17). Ameloblasts from smooth-ended bands (SEo,SE1),and those from the second and subsequent ruffle-ended bands (RE& showed less intense labelling at their apical surfaces compared to ameloblasts Figs. 5-8. Immunocytochemial preparations illustrating the distribution of gold labelling at the apical surfaces of ameloblasts from the from the most apically positioned ruffle-endedband (RE0 initial, partially ruffle-ended (RE1, Fig. 5) and/or smooth-ended (SEo, andor RE11 (Fig. 17). Fig. 6) band, and near the beginning (RE1 <) or the end (RE1>) of the first true ruflle-ended band of maturation. Numerous gold particles Distribution of silver grains at 4 hours after injection are present over the enamel layer which is separated from the cells by By 4 hours fewer silver grains were evident over the a basal lamina-like structure (BL). Small immunoreactive secretory granules (sg) are found near the apical surface or among the invagi- supranuclear regions and apical surfaces of ameloblasts nations (inv) of the plasma membrane in ruffleended cells. Granular (Table 3, Fig. 18). Silver grains were seen over previmaterial between the infoldings of the apical invaginations is heavily ously unlabelled areas of the enamel matrix both adjalabelled with gold. Granular material appears heaviest in the apical invaginations of ameloblasts at the beginning of the first true ruffle- cent to, and at a distance from, the apical surfaces of ended band (compare Figs. 5 and 6 ; a difference in staining intensity ameloblasts. The intensity of labelling over the matrix along the apical surface can be seen in light micrographs such as those did not always appear uniform along the same region illustrated in Fig. 18 [E,F,H]). Immunoreactive multivesicular bodies ('hot' and 'cold' spots as illustrated for regions ASH and (mvb)are found near the apices of the ruffle-ended cells (e.g., Figs. 7,8). Coated vesicles (cv) and mitochondria (m) show little, or no, immuno- RE1 in Fig. 18). The intensity of the radioautographic reaction also seemed to decline in an incisal direction labelling (Figs. 5,6). Figures 5-7: ~26,000.Figure 8: x 17,750. 116 A. NANCI, H.C. SLAVKIN, AND C.E. SMITH 117 SECRETORY ACTIVITY IN MODULATING AMELOBLASTS TABLE 3. Radioautography: Density of silver grains over enamel and enamel organ cells following a single injection of 3H-Methionine Compartment Secretion stage3 Enamel Cells Ameloblasts Tomes' process Distal part Proximal part Supranuclear Nuclearhnfranuclear Papillary layer cells Maturation stage4 Enamel Cells Ameloblasts Apex Supranuclear Nuclearhnfranuclear Papillarylayer cells No. of silver grains' (No.) 20 minutes Percent Area total counted2 grains (R units) (%) 61 f 29 389 f 83 361 f 79 119 f 44 85 f 35 34 f 15 181 f 38 62 f 23 7 14 10.5 3 2 1 4 3.5 14 86 80 26 19 7 40 14 29 f 14 3.5 6 7 f 6 145 & 36 108 f 25 36 k 11 43 f 10 28 f 13 7 10 5 1 2 2 5 95 71 24 28 18 37 f 14 5 24 Density of labelling' (No./R) No. of silver grains' (No.) 8.8 f 4.2 485 f 110 28.1 f 6.0 187 f 49 36.9 k 8.4 156 f 41 39.6 f 14.6 85 f 28 40.6 k 18.2 75 f 30 33.8 f 15.1 10 f 5 44.1 f 11.6 40 f 16 17.7 f 6.7 31 10 8.7 f 4.2 31 f 14 4 hours Percent Area total counted2 grains (R units) (%) Density of labelling' (No./R) Change in density of labelling 7.3 14 10.5 3 2 1 4 3.5 72 28 23 13 11 1 6 5 66.4 f 15.1 13.0 f 3.6 14.7 f 3.8 28.2 f 9.4 37.6 f 15.2 9.5 f 5.4 10.0 f 4.1 8.7 f 2.9 7.5 .5 .4 .7 .9 .3 .2 .5 3.5 5 8.9 k 3.9 1.0 0.8 f 0.7 45 f 16 14.5 f 3.1 120 f 27 21.1 k 4.9 75 f 20 35.6 f 11.2 24 f 9 26 f 6 21.7 5.0 13.3 f 5.4 24 f 11 7 10 5 1 2 2 27 73 45 14 16 15 6.5 f 2.3 11.9 f 2.3 15.2 f 3.0 23.5 f 9.1 13.2 f 3.2 12.7 f 3.9 8.1 45 f 11 5 27 8.8 f 2.7 1.1 7.8 f 2.5 .8 .7 .7 .6 1.0 'Mean f standard deviation; all values were corrected for a n average background labelling of 7.5 f 3.0 grains per 10R. 'One R unit = 221.9 pm' (see Materials and Methods). 3Samples were taken from mid inner enamel secretion (see Materials and Methods). *Samples were taken from the apical part of first ruffle-ended band (RE1) (see Materials and Methods). across the various regions related to EDTA-insoluble enamel (compare panels B-H in Fig. 18). Silver grains were evident in the matrix above ameloblasts from both smooth-ended and ruffle-ended bands (Fig. 18 D-H). DISCUSSION Interpretations of results from biochemical andor morphological studies of the maturation stage of amelogenesis have been based to a large extent on the assumption that the ameloblasts do not secrete additional enamel proteins into the matrix once the final layer of enamel has been deposited. Hence, a phenomenon such as the maturation-related shift from predominantly intermediate components to higher proportions of low and high molecular weight components in enamel has been Figs. 10-12. Immunocytochemical preparations of the supranuclear compartment of ruffle-ended (Figs. 10, 12) and smooth-ended (Fig. 11) ameloblasts. The protein synthetic organelles (endoplasmic reticulum, Golgi saccules, and secretory granules [isg, sg]) and degradative organelles (dark lysosomes [dl], pale lysosomes [pl], and multivesicular bodies [mvb]) of ameloblasts are immunoreactive throughout the initial ruffle-/smooth-ended band (REdSEo) and first true ruffle-ended band (RE1). Gold particles associated with the endoplasmic reticulum are sometimes seen within the cisternal lumen, but more often along the cisternal membrane or near ribosomes (inset). The lysosomes show variable labelling. Multivesicular bodies are frequently and intensely labelled with gold. The arrows in Figure 11 indicate occasional dense structures, presumably elements of the lysosomal system, that were difficult to classify by the method used in Table 2 (not included in quantitative analyses). Figure 10: ~ 3 5 , 0 0 (inset 0 X36,750). Figures 11 and 12: ~ 2 6 , 0 0 0 . attributed exclusively to the breakdown of preexisting proteins in the matrix (amelogenins)(reviewed in Robinson and Kirkham, 1985).Similarly, the presence of granular material at the surfaces or within invaginations and granules at the apices of ruffle-ended ameloblasts has been postulated to be indicative of resorptive activity by these cells CReith, 1970; Kallenbach, 1974; Takano, 1979). The data presented in this study suggest that the maturation process is more complicated, and a portion of the changes believed to result solely from degradation may derive from continued secretory activity by the ameloblasts (Kallenbach, 1974). The strongest evidence for continued biosynthesis of enamel proteins by modulating ameloblasts from the maturation stage is based on our observations that these cells contain significant immunoreactivity (above background) in the endoplasmic reticulum, Golgi saccules, and small secretory granules. Evidence that this material is secreted is provided by the radioautographic studies which document similiarities in the pattern of uptake of labelled amino acid precursors and subsequent passage and diffusion of labelled proteins within the matrix when comparing modulating ameloblasts to those from the secretory stage (Table 3). We assume that the lower immunoreactivity within modulating ameloblasts is due to less overall biosynthetic activity as opposed t o alterations in the antigenic properties of enamel proteins being formed by these cells. This conclusion is suggested by data from the radioautographic studies which show that these ameloblasts incorporate about two times less radioactive tracer and release about ten times less labelled 118 A. NANCI, H.C. SLAVKIN, AND C.E. SMITH SECRETORY ACTIVITY IN MODULATING AMELOBLASTS Fig. 16. Immunocytochemical preparation of the supranuclear compartment of a n ameloblast from the second smooth-ended band. (SE2; see Fig. 3). The intensity of the immunocytochemical reaction over Golgi saccules declines to background levels as ameloblasts progress farther into the maturation stage of amelogenesis. These cells (SE2) show ferritin both throughout the cytoplasm and accumulated within numerous lysosomes that appear unlabelled with gold (arrows). However, some multivesicular bodies (mvb) remain intensely immunoreactive. ~24,875. Figs. 13, 14. Dual cytochemical and immunocytochemical preparations illustrating the distribution of TMPase-positive and immunoreactive lysosomes in the supranuclear compartment of ruffle-ended ameloblasts (RE1). Dark lysosomes (dl) often show TMPase activity (+) but little immunolabelling with gold. This is in contrast to the pale lysosomes (pl) and multivesicular bodies (mvb) which show spotty, or no (-), TMPase activity and moderate or heavy labelling with gold. Focal areas of smooth endoplasmic reticulum, or SER (Ozawa et al., 1983)show TMPase activity but no immunoreactivity. ~49,000. Fig. 15. Cytochemical preparation of the supranuclear compartment of ameloblasts illustrating the distribution of AcPase-positive lysosomes using P-glycerophosphate as substrate. Multivesicular bodies react unpredictably for AcPase activity but many are positive thereby confirming their lysosomal nature. Within the multivesicular bodies the vesicles (arrowhead) sometimes show heavy deposits of reaction product. ~38,000. 119 material into the matrix over 4 hours compared to ameloblasts forming the inner enamel layer (based on density of labelling; Table 3). Presently we are unsure where modulating ameloblasts terminate biosynthesis and secretion of the enamel proteins. Data in this study have indicated only that the immunoreactivity and intensity of radioautographic reactions over ameloblasts decline gradually in a n incisal direction, and both are evident at least up to the second ruffle-ended phase, at which point the cells are associated with partially EDTA-soluble enamel. Similarly, the molecular characteristics and the longterm fate of the enamel proteins released by modulating ameloblasts into maturing enamel are unclear at this time. A portion, or all, of these newly formed proteins presumably would undergo degradation as occurs for most proteins added to the matrix during its original appositional growth phase (secretory stage). While our classification system for lysosomes was crude, it was capable of revealing similarities and differences for various components of the lysosomal system in ameloblasts derived from the secretory and maturation stages of amelogenesis. It is evident from Table 2 that the lysosomes with dark contents are the leaset immunoreactive in all cases. These are the same population of lysosomes which generally show strong and consistent TMPase activity in cytochemical localizations (Smith, 1979; Sasaki et al., 1984; Figs. 13, 14). This suggests that dark lysosomes likely correspond to typical secondary lysosomes where degradation of proteins has been ongoing for some time (Bainton, 1981). We believe that the more immunoreactive pale lysosomes and intensely immunoreactive multivesicular bodies represent components of the lysosomal system positioned functionally closer to the endosomal compartment than the dense lysosomes (deDuve, 1983).The high content of antigenic sites and high frequency of labelling observed in these structures are consistent with a prelysosomal compartment in which there is limited degradation. The cytochemical findings that a portion of these structures shows spotty AcPase activity (Smith, 1979; Takano and Ozawa, 1980; Ozawa et al., 1983; Fig. 15) and TMPase activity (Smith, 1979; Figs. 13, 14) are also consistent with their functional transition toward secondary lysosomes (reviewed in Bainton, 1981; deDuve, 1983; Bergeron et al., 1985). Studies by several workers have shown that the lysosomes in ruffle-ended ameloblasts will accumulate exogeneous protein tracers from 10 minutes to 1 hour or more after injection (Kallenbach, 1980; Takano and Ozawa, 1980; Sasaki, 1984b; Sasaki et al., 1984). Based on this rationale, the data in Tables 1and 2 add more direct support to the hypothesis that modulating ameloblasts endocytose enamel proteins during the maturation stage (Reith and Cotty, 1967).These data further support the contention of Takano and Ozawa (1980) that smooth-ended ameloblasts may be lessendocytotically active thanruffle-ended ameloblasts, albeit only marginally according to our data (Table 2). An observation in Table 2 that cannot be explained easily at this time concerns why the multivesicular bodies, and to a lesser extent the pale and dark lysosomes, show more intense immunoreactivity on a per lysosome basis in ameloblasts during the secretory stage compared to the maturation stage. This clearly implies that there is likely substantial endocytotic (re- 120 A. NANCI, H.C. SLAVKIN, AND C.E. SMITH Fig. 17. Light microscope radioautographs showing the distribution of silver grains across contiguous regions of the maturation stage (BH) up to the second ruffle-ended band (H) at 20 minutes after an intravenous injection of meth~l-~H-rnethionine. Panel A shows ameloblasts from the secretory stage of amelogenesis (IES) for comparison. An intense radioaugograpbic reaction is evident in the supranuclear compartment@ and along the apical portions (a) of ameloblasts undergoing postsecretory transition (B, C) and modulation (D-H). Fewer silver grains appear to be present over these compartments in ameloblasts positioned more incisally (compare panels D and H). Reactions along the apical surface appear less intense in smooth-ended ameloblasts versus ruffle-ended ameloblasts (compare panels F-H). In panel D the apical morphology of ameloblasts which are illustrated was incompletely r m e d (REo). The enlarged space laterally between ameloblasts are artifacts due to initial perfusion fixation with paraformaldehyde. All radioautographs were dipped in the same emulsion and exposed equally (14days). ~ 7 0 0 . SECRETORY ACTIVITY IN MODULATING AMELOBLASTS Fig. 18. Light microscope radioautographs showing the distribution of silver grains across contiguous regions of the maturation stage (BH) up to the second rume-ended band (H)at 4 hours after a n intravenous injection of methyl-3H-methionine. Panel A shows ameloblasts from the secretory stage of amelogenesis (IES) for comparsion. Radioactive proteins have been secreted by the ameloblasts and these have diffused deep into the maturing enamel layer. Less intense, and more patchy, radioautographic reactions are evident over the matrix where enamel rods are clearly visible (panels G and H) and where amelob- 121 lasts are changing in height prior to initial modulation (panel C). Silver grains are present over the matrix near both smooth-ended and ruffle-ended ameloblasts (panels G and H). In panel D the apical morphology of ameloblasts which are illustrated was incompletely rumed (RE,$ The data in Table 3 were obtained from cells positioned similarly to those illustrated in panel E (same for Fig. 17). All radioautographs were dipped in the same emulsion and exposed equally (14 days). ~ 7 0 0 . 122 A. NANCI, H.C. SLAVKIN, AND C.E. SMITH sorptive) activity during appositional growth of the enamel layer (reviewed in Smith, 1979; Blumen and Merzel, 1982; Ozawa et al., 1983; Sasaki, 1984a; Robinson and Kirkham, 1985). The major correlation evident from data in Tables 1 and 2 suggests that when the biosynthetic organelles show high immunoreactivity then the lysosomes also show substantial immunoreactivity. Although the focus of this study was not on banding (modulation) per se, there are three points about this phenomenon that pertain to the data presented herein. First, over the course of this investigation easily the most recurring theme from incisor to incisor was that of variability. This included 1)variability in length of ruffle bands (Josephsen and Fejerskov, 1977) and in the amount and density of immonoreactive material present in the apical invaginations of ruffle-ended ameloblasts (most dramatic across RE,); 2) variability in the number and exact location of smooth-ended bands in the maturation zone (at least three to as many as seven including SEo; the best match in number and position was usually between the incisors from the same animal; see Josephsen and Fejerskov, 1977); 3) variability in the existence of true smooth-ended ameloblasts in the initial region of modulation (about 20% of incisors showed a SEo region, and these were not always present on both mandibular incisors from the same animal; Josephsen and Fejerskov, 1977; Crenshaw and Takano, 1982); and 4) variability in immunoreactivity and intensity of radiautographic reactions (declining gradient incisally). Second immunocytochemical, cytochemical , or radioautographic comparisons of rufne-ended ameloblasts to smooth-ended ameloblasts were inconclusive. Cells with either apical morphology showed immunoreactive Golgi saccules, secretory granules, and lysosomes as well as the ability to incorporate radioactive amino acids and secrete labelled proteins into the enamel martrix. With the exception of data given in Table 2 and the radioautographic finding of less intense apical reactions over smooth-ended ameloblasts at early times after injection of radioactive amino acids, any differences that were noted seemed more closely related to incisal gradients than to profound functional differences between these cells. Lastly, on the average the ameloblasts undergo only two of the several characteristic modulations over the inverval of time when the majority of the organic material disappears from the maturing matrix as visualized in EDTA-decalcified incisors. A smooth-ended band is often found near the point where the matrix changes appearance from indistinct to distinct enamel rods (SE11 and where the matrix changes from EDTAinsoluble to EDTA-soluble (SE?). The reason(s1for continued biosynthesis and secretion of enamel proteins by modulating ameloblasts is unclear at this time. On the one hand, it could be argued that this activity is unimportant and merely reflects the inability of the cells to completely shut down the genes coding for enamel proteins (residual secretion). On the other hand, this activity could be a n expression of something more subtle such as a change in the type or molecular characteristics of the newly formed proteins (to high or low molecular weights or both). Irrespective of the reasons, the identification of such secretory activity in ameloblasts during the early phases of the maturation stage introduces additional factors that must be taken into account when defining global events related to the process of amelogenesis. ACKNOWLEDGMENTS Several people made important contributions to this investigation. We thank Annie BBlanger and Micheline Fortin for their technical assistance in animal preparation and sectioning, Dr. Beatrix Kopriwa and Fernando Evaristo for their assistance in dipping and processing of radioautographs, Denis Kay for instruction on the use of the Speed Vac, Margo Oeltzschner for making the line drawings, and Tony Graham and Gaston Lambert for their photomicrography and photocopy work. We also thank Dr. MoYse Bendayan for his advise concerning immunocytochemical procedures. Portions of this investigation were supported by grants from the Medical Research Council of Canada (A.N. and C.E.S.), by grant DE-02848 from the National Institutes of Health (H.C.S.), and by the E.D. Broughton Research Fund (Faculty of Dentistry, McGill). LITERATURE CITED Bainton, D.F. (1981) The discovery of lysosomes. J. Cell Biol., 91:66s76s. Bendayan, M. (1984) Protein A-gold electron microscopic immunocytochemistry: Methods, applications, and limitations. J. Electron Microsc. Tech., 1:243-270. Bergeron, J.J.M., J. Cruz, M.N. Khan, and B.I. Posner (1985) Uptake of insulin and other ligands into receptor-rich endocytic components of target cells: The endosomal apparatus. Annu. Rev. Physiol., 47:383-403. Bienkowski, R.S. (1983)Intracellular degradation of newly synthesized secretory proteins. Biochem. J., 214:l-10. Blumen, G., and J. Merzel (1982) New evidences for the role of secretory ameloblasts in the removal of proline labelled proteins from young enamel as visualized by autoradiography. J. Biol. Buccale, 10:73-83. Carter, J., A.C. Smillie, and M.G. Shepherd (1984)Proteolytic enzyme in developing porcine enamel. In: Tooth Enamel IV. R.W. Fearnhead and S. Suga, eds. Elsevier Science Publishers, Amsterdam, pp. 229-233. Crenshaw, M.A., and Y. Takano (1982) Mechanisms by which the enamel organ controls calcium entry into developing enamel. J. Dent. Res., 61:1574-1579. Crenshaw, M.A., and J.W. Bawden (1984) Proteolytic activity in embryonic bovine secretory enamel. In: Tooth Enamel IV.R.W. Fearnhead and S. Suga, eds. Elsevier Science Publishers, Amsterdam, pp. 109-113. deDuve, C. (1983) Lysosome revisited. Eur. J. Biochem., 137:391-397. Doty, S.B., C.E. Smith, A.R. Hand, and C. Oliver (1977) Inorganic trimetaphosphatase as a histochemical marker for lysosomes in light and electron microscopy. J. Histochem. Cytochem., 2513811384. Eastoe, J.E. (1979)Enamel protein chemistry: Past, present and future. J. Dent. Res., 58:753-763. Fincham, A.G., and A.B. Belcourt (1985) Amelogenin biochemistry: Current concepts. In: The Chemistry and Biology of Mineralized Tissues. W.T. Butler, ed. University of Alabama, Birmingham, pp. 240-247. Gomori, G. (1950) An improved histochemical technique for acid phosphatase. Stain Technol., 25%-85. Inage, T., and Y. Toda (1984) An electron microscopic mapping of the maturation ameloblasts of a rat incisor. In: Tooth Enamel IV. R.W. Fearnhead and S. Suga, eds. Elsevier Science Publishers, Amsterdam, pp. 245-250. Josephsen, K., and 0. Fejerskov (1977) Ameloblast modulation in the maturation zone of the rat incisor enamel organ: A light and electron microscopic study. J. Anat., 124:45-70. Josephsen, K., and H. Warshawsky (1982) Radioautography of rat incisor dentin as a continuous record of the incorporation of a single dose of 3H-labelled proline and tyrosine. Am. J. Anat., 164:45-56. Kallenbach, E (1974)Fine Structure of rat incisor ameloblasts in transition between enamel secretion and maturation stages. Tissue Cell 6:173-190. SECRETORY ACTNITY IN MODULATING AMELOBLASTS Kallenbach, E. (1980) Access of horseradish peroxidase (HRP) to the extracellular spaces of the maturation zone of the rat incisor enamel organ. Tissue Cell, 12:165-174. Karnovsky, M.J. (1971) Use of ferrocyanide-reduced osmium tetroxide in electron microscopy. Proc. Am. Soc. Cell Biol., 11:146a. Kopriwa, B.M., and C.P. Leblond (1962) Improvement in the coating technique of radioautography. J. Histochem. Cytochem., 10:269284. 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. 33:1153-1160. Novikoff, A.B. (1963) Lysosomes in the physiology and pathology of cells: Contributions of staining methods. In: Ciba Foundation Symposium on Lysosomes. A.V.S. deReuck and M.P. Cameron, eds. Little Brown and Company, Boston, pp. 36-73. Ozawa, H., M. Yamada, T. Uchida, T. Yamamoto, and Y. Takano (1983) Fine structural and cytochemical studies on the Golgi-SER system of ameloblasts with special reference to its resorptive function. In: Mechanisms of Tooth Enamel Formation. S. Suga, ed. Quintessence Publishing Company, Tokyo, pp. 17-48. Reith, E.J. (1970) The stages of amelogenesis as observed in molar teeth of young rats. J. Ultrastruct. Res., 3Ot111-151. Reith, E.J., and V.F. Cotty (1967)The absorptive activity of ameloblasts during the maturation of enamel. Anat. Rec. 157.577-588. Robinson, C., J. Kirkham, H.D. Briggs, and P.J. Atkinson (1982) Enamel proteins: From secretion to maturation. J. Dent. Res., 61r1490-1495. Robinson, C., J. Kirkham, H.D. Briggs, and P.J. Atkinson (1983) Changes in the protein components of rat incisor enamel during tooth development. Arch. Oral Biol., 28:993-1000. Robinson, C., and J. Kirkham (1985) Dynamics of amelogenesis as revealed by protein compositional studies. In: The Chemistry and Biology of Mineralized Tissues. W.T. Butler, ed. University of Alabama, Birmingham, pp. 248-263. Sasaki, T. (1984a) Tracer, cytochemical, and freeze-fracture study on the mechanisms whereby secretory ameloblasts absorb exogeneous proteins. Acta AnatLBasel), 118:23-33. Sasaki, T. (1984b) Endocytic pathways at the ruffled borders of rat maturation ameloblasts. Histochemistry,80:263-268. Sasaki, T., S. Higashi, T. Tachikawa, and S. Yoshiki (1984) Absorptive and digestive functions of maturation ameloblasts in rat incisors. In: Tooth Enamel IV.R.W. Fearnhead and S. Suga, eds. Elsevier Science Publishers, Amsterdam, pp. 266-270. Shimizu, M., and M. Fukae (1983)Enamel proteins. In: Mechanisms of Tooth Enamel Formation. S. Suga, ed. Quintessence Publishing Company, Tokyo, pp. 125-141. Shimokawa, H., P. Wassmer, M.E. Sobel, J.D. Termine (1984) Characterization of cell-free translation products of mRNA from bovine ameloblasts by monoclonal and polyclonal antibodies. In: Tooth Enamel IV. R.W. Fearnhead and S. Suga, eds. Elsevier Science Publishers, Amsterdam, pp. 161-166. Slavkin, H.C., W. Mino, and P. Bringas (1976) The biosynthesis and secretion of precursor enamel protein by ameloblasts as visualized by autoradiography after tryptophan administration. Anat. Rec., 185t289-312. 123 Slavkin, H.C., M. Zeichner-David,M. MacDougall, P. Bringas, C. Bessem, and L.S. Honig (1982) Antibodies to murine amelogenins: Localization of enamel proteins during tooth organ development in vitro. Differentiation, 23:73-82. Slavkin, H.C., M. Zeichner-David, M.L. Snead, P. Bringas, C. Bessem, and V. Santos (1984) Epithelial-mesenchymal interactions and enamel gene inductionevolutionary implications. In: Tooth Enamel N . R.W. Fearnhead and S. Suga, eds. Elsevier Science Publishers, Amsterdam, pp. 167-171. Smith, C.E. (1974)A method for preparing longitudinal semi-thin Epon sections of entire at incisors. Arch. Oral Biol., 19:1045-1048. Smith, C.E. (1979)Ameloblasts: Secretory and resorptive functions. J. Dent. Res., 58:695-706. Smith, C.E. (1981) Correlated biochemical and cytochemical studies of nicotinamide adenine dinucleotide phosphatase (NADPase) activity in ameloblasts using structural analogues of NADP. J. Histochem. Cytochem., 292322-836. Smith, C.E., and H. Warshawsky (1975)Cellular renewal in the enamel organ and the odontoblast layer of the rat incisor as followed by radioautography using 3H-thymidine. Anat. Rec. 183:523-562. Takano, Y. (1979) Cytochemical studies of ameloblasts and the surface layer of enamel of the rat incisor at the maturation stage. Arch. Histol. Jpn., 42:ll-32. Takano, Y., and H. Ozawa (1980) Ultrastructural and cytochemical observations on the alternating morphologic changes of the ameloblasts at the stage of enamel maturation. Arch. Histol. Jpn., 43:385-399. Takano, Y., and H. Ozawa (1984)Autoradiographic and tracer experiments on the exit route for resorbed organic matrix of the enamel a t the stage of maturation. In: Tooth Enamel N.R.W. Fearnhead and S. Suga, eds. Elsevier Science Publishers, Amsterdam, pp. 271-275. Termine, J.D., A.B. Beicourt, 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., 255:9760-9768. Warshawsky, H. (1979) Radioautographic studies on amelogenesis. J. Biol. Buccale, 7t10.5-126. Warshawsky, H., and G. Moore (1967)A technique for the fixation and decalcification of rat incisors for electron microscopy. J. Histochem. Cytochem., 15542-549. Warshawsky, H., and C.E. Smith (1974)Morphological classification of rat incisor ameloblasts. Anat. Rec., 179t423-466. Warshawsky, H., and K. Josephsen (1981) The behavior of substances labelled with 3H-proline and 3H-fucose in the cellular processes of odontoblasts and ameloblasts. Anat. Rec., 2OOt1-10. Weinstock, A. (1972) Secretory function of ‘postsecretory’ ameloblasts as shown by electron microscope radioautography. IADR Absracts #136, pp. 82. Zeichner-David,M., J. Vides, M. Snead, and H. Slavkin (1985)Enamel protein characterization by mRNA translation in a cell-free system. In: The Chemistry and Biology of Mineralized Tissues. W.T. Butler, ed. University of Alabama, Birmingham, pp. 264-269.