THE ANATOMICAL RECORD 250:268–280 (1998) Crown Morphology, Enamel Distribution, and Enamel Structure in Mouse Molars S. PETTER LYNGSTADAAS,1,2* CHRISTINA B. MØINICHEN,1 AND STEINAR RISNES1 1Department of Oral Biology, Faculty of Dentistry, University of Oslo, Oslo, Norway 2Department of Pathology, Faculty of Dentistry, University of Oslo, Oslo, Norway ABSTRACT Background: Biomolecular research and genetic manipulations have stressed the importance of thorough knowledge of normal organ morphology. Mouse molar teeth are convenient models for studying basic interactions in organ development and morphogenesis. The aim of the present study was to provide basic information on their morphology. Methods: Intact and sectioned/ground molars of mice of various ages were observed with SEM. Results: Enamel-free areas (EFA) were present on cusp tips at time of eruption. The dominating structural configuration in enamel was prism decussation in inner enamel and parallel prisms in outer enamel. Prism decussation tended to be absent at cusp ridges and in the bottom of grooves. In the former location, the distinction between prisms and interprism was often obscured in the middle enamel zone due to decreased difference in orientation of their crystals. A thin layer of enamel, often aprismatic, covered the distal aspect of cusps in maxillary molars and the mesial aspect of cusps in mandibular molars. The enamel abutting on EFA was often aprismatic. Aprismatic enamel exhibited incremental lines with a periodicity of about 1 mm and was often traversed by cracks. The enamel surface was porous in the bottom of grooves. Parts of mouse molar enamel were incompletely mineralized at the time of eruption. Conclusions: SEM is a convenient method for combined studies of crown morphology and enamel structure. Based on morphological criteria, a modification of the cusp nomenclature is proposed. Enamel thickness and structure in mouse molars show regional variations. Fundamental similarities exist between mouse molar cusps and mouse incisors. Mouse molar enamel undergoes posteruptive maturation. Anat. Rec. 250:268–280, 1998. r 1998 Wiley-Liss, Inc. Key words: tooth; molar; crown morphology; dental enamel; scanning electron microscopy; mouse Over the last decade, methods for generating transgenic mouse lineages and gene knockouts have established the mouse as the most widely used animal model for the study of vertebrate genetics, embryogenesis, organogenesis, and tumorgenesis. An intimate knowledge of the normal development and anatomy of the mouse is important in order to evaluate and interpret observations from such research. Recent advances in biomolecular research (e.g., Chen et al., 1994; Brookes et al., 1995; Lyngstadaas et al., 1995; Matzuk et al., 1995; Thesleff et al., 1995a) have shed new light on the molecular biology of tooth formation. There is an increasing interest in using mouse molars as a model system for studying organ development and morphogenesis and biomineralization both in vitro and in vivo (Slavkin, 1991; Thesleff et al., 1995b; Lyngstadaas et al., 1995). Mouse molar teeth are paired mineralized organs formed at the interface between the ectoderm lining the r 1998 WILEY-LISS, INC. oral cavity and the underlying mesenchyme, and their development is controlled by basic molecular mechanisms common to most organs (Thesleff et al., 1995b). Teeth consist of the mineralized tissues dentine, cementum, and enamel. Dentine and cementum are calcified connective tissues closely related to bone (Linde and Goldberg, 1993). Enamel constitutes the outermost covering of adult amphibian, reptilian, and mammalian teeth. Biomineralization comes to an extreme in this tissue, the hardest and most durable in the vertebrate Contract grant sponsor: Research Council of Norway; Norwegian Cancer Society. *Correspondence to: S. Petter Lyngstadaas, Laboratory for Molecular Biology, Department of Pathology, Dental Faculty, University of Oslo, PO Box 1109, Blindern, N-0317 Oslo, Norway. E-mail: firstname.lastname@example.org Received 22 April 1996; Accepted 14 October 1997 269 MOUSE MOLAR ENAMEL body. Mature enamel contains 95% mineral by weight. The mineral component is mainly hydroxyapatite in the form of closely packed ultramicroscopic crystals, larger than those in other mineralized tissues (Sasaki et al., 1990). The crystals are organized by differential orientation into a basic pattern of prisms (rods) and interprism (interrod) which is common to all mammals but which shows intergeneric variations in design. The morphology of mouse molars has been described at the light microscopy level (Gaunt, 1955, 1956; Cohn, 1957; Lumsden, 1979). To our knowledge, no systematic examination of these teeth by scanning electron microscopy (SEM) has previously been reported. One reason for this may be the small size of these teeth, making sectioning and preparation for scanning electron microscopy technically demanding. Another reason may be that the ultrastructure of mouse molar teeth is assumed to be similar to the rat (Skobe, 1978; Risnes, 1979a). We report here a scanning electron microscope study of the mouse molar dentition, with emphasis on crown morphology, enamel distribution, and enamel structure/ultrastructure. MATERIALS AND METHODS Balb/C albino mice were kept in a temperature (20–22°C) and humidity (45–55%) controlled environment and maintained on a 14 h light/10 h dark cycle with ad libitum access to food and water. Husbandry was in accordance with Norwegian regulations, meeting the NIH Guidelines for the Care and Use of Laboratory Animals. A total of 17 mice aged 14, 17, or 35 days were obtained from different litters and sacrificed by cervical dislocation. Their upper and lower jaws were dissected out and fixed in 70% ethanol. After fixation, all soft tissue was carefully removed by dissection, light brushing, and ultrasonication. Jaw segments containing all three upper or lower molar teeth were cut out with a rotating diamond wheel (Risnes, 1981), air-dried, and mounted on brass cylinders with a cyanoacrylate glue. The specimens were sputter-coated with 50 nm gold-palladium and observed in a Philips SEM 515 operated at 15 kV, using a specially designed holder allowing multiangular viewing of the specimens (Risnes, 1982). The right maxillary and mandibular molar segments from a 35-day-old mouse were subjected to repeated cycles of transverse grinding, cleaning, etching, drying, and sputter-coating, allowing observation of seven serially ground planes through each molar segment. In addition, some molar segments were ground longitudinally in a mesiodistal direction, some after first having been embedded in Epont. Grinding was performed under a stereomicroscope using grits 800 and 1200 of 3Mt waterproof silicone carbide paper in a specially designed apparatus (Risnes, 1985). After grinding, the teeth were cleaned by careful brushing under running tap water and etched three times for 10 s in 0.1% nitric acid. After air-drying overnight, the specimens were sputter-coated with 30 nm gold-palladium and observed in the SEM. After acid-etching, immature enamel can be distinguished from mature enamel in the SEM by a difference in texture, especially by a difference in the distinctness of crystals (Risnes, 1990; Risnes et al., 1996). In the present study, we used a modified version of Gaunt’s (1955) cusp terminology. RESULTS Crown Morphology At an age of 17 days (Fig. 1a–d), the first and second molars had just erupted but had not reached occlusion. The cusps were virtually unworn. The third molars had not yet erupted. At an age of 35 days, all molars had reached occlusion, and their cusps were somewhat worn (Fig. 1e–h). The maxillary molars, M1, M2, and M3, exhibited three rows of cusps, lingual (L), buccal (B), and central, all markedly tilted in a distal direction and separated by two shallow grooves oriented mesiodistally and by deep, mesially convex grooves traversing buccolingually (Fig. 1a,c–e,g,h). In the mandibular molars, M1, M2, and M3, the cusps formed two rows, lingual (L) and buccal (B), separated by a shallow groove oriented mesiodistally and by deep, mesially convex grooves traversing buccolingually (Fig. 1b–d,f–h). On M1 cusps, B1, B2, L1, and L2 formed a cuspal complex surrounding a common dentin surface. The cusps of the mandibular molars, except cusps B1 and L1 on M1, had a pronounced mesial tilt. A rudimentary central cusp, cusp 4, was located at the most distal part of the mandibular molars. The three maxillary and the three mandibular molars showed a monotonic form gradient in which the greatest complexity was anteriorly; morphologically, the second molars closely resembled the posterior twothirds of the first molars, whereas the third molars resembled the posterior half of the second molars (Fig. 1a,b,e,f). M2 had lost the cusps corresponding to B1 and 1 of M1, but a rudimentary B1 was appreciable (Fig. 1c). M2 had lost the cusps corresponding to B1 and L1 of M1, but also here a rudimentary B1 was discernable (Fig. 1c). Enamel Distribution The distribution of enamel was observed both in unsectioned (Figs. 1, 2) and in sectioned teeth (Figs. 3, 4). Enamel-free areas (EFA) at the tip of the cusps were obliquely oriented, facing distally in the upper jaw and mesially in the lower jaw (Fig. 1a–h), except on cusps B1 and L1 of M1, where the EFA faced distally and was continuous with the EFA of the adjacent cusps B2 and L2 (Fig. 1b,f). At the time of eruption, before occlusal attrition had started, the EFA were partly surrounded by a rampart of enamel protruding somewhat beyond the dentin (Figs. 1a–d, 2a,b). Distinct pits could occupy variable portions of the EFA (Figs. 1a,b and 2b). Generally, the EFA reached to the entrance of the main transverse grooves (Fig. 2a,b). However, the EFA could extend for short distances into the transverse groove, especially in relation to the buccal and lingual cusps, where the EFA often showed tongue-like extensions (Fig. 2a,b). At an age of 35 days, attrition had increased the size of the EFA, reduced their obliquity, and reduced the height of the cusps (Figs. 1e–h). Serial transverse grinding (Fig. 3a–e, g–k) and selected longitudinal sections (Fig. 3f,l) demonstrated the distribution of enamel in more detail. The distal aspect of the cusps in the maxillary molars and the mesial aspect of the cusps in the mandibular molars were covered with a rather thin layer of enamel (Figs. 270 S.P. LYNGSTADAAS ET AL. Fig. 1. Crown morphology of mouse molars. Arrows relate to details shown in subsequent figures. a–d: Age 14 days. 324. a: Occlusal view of right maxillary molars. For cusp designation, see Fig. 1e. b: Occlusal view of right mandibular molars. For cusp designation, see Fig. 1f. c: Buccal view of right molars. Bullets mark rudimentary cusps B1 on M2 and M2. d: Lingual view of right molars. e–h. Age 35 days. 3 24. e: Occlusal view of right maxillary molars. Buccal cusps (B1, B2, B3), lingual cusps (L1, L2), and central cusps (1, 2, 3) are counted from mesial to distal. f: Occlusal view of right mandibular molars. Buccal cusps (B1, B2, B3) and lingual cusps (L1, L2, L3) are counted from mesial to distal. Distally, there is one central cusp (4). g: Buccal view of right molars (not the same as those shown in Fig. 1e,f). h: Lingual view of right molars (same as those shown in Fig. 1g). 3d, e, f, i, j, k, l, 4). All other aspects of the crown were more abundantly covered, but the thickness of the enamel varied somewhat from region to region, reaching about 110 µm at the summit of the convex aspect of the cusps (Fig. 3f). In maxillary molars, the central cusps were connected by low ridges elevating the bottom of the transverse grooves halfway between the buccal and lingual aspects (Fig. 3d,e). Adjacent to these ridges, the MOUSE MOLAR ENAMEL 271 Fig. 2. Enamel-free areas (EFA). a: Cusp L2 of M1 shown in Fig. 1a. EFA extends tongue-like (arrows) into groove aspect of cusp. E, enamel. Age 14 days. 3312. b: Cusp L3 of left M1 (corresponding cusp is seen in right M1 in Fig. 1b). The dentin surface of EFA is scalloped with a centrally placed pit (Pi). EFA extends tongue-like (arrows) into groove aspect of cusp. E, enamel. Age 17 days. 3300. thin enamel covering the distal aspects of the cusps was somewhat thickened (Fig. 4). A similar ridge existed between cusps L1 and 2 in M1 and M2 (Fig. 3d,e). Enamel Structure Considering the whole thickness of enamel, we observed three different configurations in mouse molars: 1) decussating prisms in the inner enamel and parallel prisms in the outer enamel (Fig. 5a), 2) parallel prisms only (Fig. 5a), and 3) aprismatic enamel (Fig. 5b). Of these, the first was the most common configuration. Figure 6 demonstrates schematically the distribution of prism decussation in mouse molars as seen in midcoronal, mesiodistal longitudinal planes and in midcuspal transverse planes. Prism decussation tended to be absent at the summit of the convex aspect of the cusps, in the bottom of grooves/invaginations, and in the thin enamel covering the distal aspect of the cusps in the maxillary molars and the mesial aspect of the cusps in the mandibular molars. Cervical to the level of the cusps, prism decussation was present around the whole circumference of the crown on all molars. Enamel With Prism Decussation The expression of prism decussation varied considerably, from fully developed over large areas (Figs. 5a, 7a–d) through fully developed in isolated areas (Fig. 7f) to vague attempts in isolated areas (Fig. 7g). Generally, prism decussation occupied the inner one-half to twothirds of the enamel thickness, while parallel prisms occupied the outer one-third to one-half (Figs. 5a, 7a,c). Toward the cemento-enamel junction, the latter component decreased (Fig. 7b). Prism decussation was of the uniserial lamellar type (Figs. 5a, 7a–f). The single-layered lamellae of prisms were generally arranged transversely to the long axis of the tooth, evidenced by the fact that the lamellae were cut transversely in longitudinal tooth sections (Figs. 5a, 7a,b) and ran roughly parallel with the enamel-dentin junction in transverse tooth sections (Fig. 7d). On the side aspects of the cusps, as seen in the transverse plane, the prism lamellae deviated away from the enamel-dentin junction in a direction away from the cusp summit (Fig. 8a,b), indicating that the prism lamellae were oriented obliquely relative to the ground transverse plane but transversely relative to the slope of the cusp. Figure 9 is a schematic representation of the orientation of the prism lamellae in mouse molars deduced from observations of serially ground transverse planes. The lamellae were generally inclined occlusally at an angle of about 55–65° relative to the enamel-dentin junction (Figs. 5a, 7a), an angle which tended to increase to 80° or more toward the cemento-enamel junction (Fig. 7b). The variability in the inclination of prism lamellae was reflected by the variable appearance of prisms in transverse tooth sections (Fig. 7c,d). The occlusal inclination of the prisms showed a distinct increase as they passed from the inner zone with decussation to the outer zone without decussation (Figs. 5a, 7a); in the latter, the angle between prisms and enamel-dentin junction varied between 10 and 40°. Transverse tooth sections through regions with occlusally inclined decussating lamellae showed that the interprism in these regions was arranged in radial sheets delimiting radial lines of prisms containing prisms belonging to different prism lamellae (Fig. 7d,f). 272 S.P. LYNGSTADAAS ET AL. Fig. 3. Sectioned and/or ground planes through molars of mice, age 35 days. Arrows relate to details shown in subsequent figures. Inserts show position of sectioned/ground plane. For cusp designation, see Fig. 1e,f. 324. a–e: Serial transverse grinding of right maxillary molars shown in Fig. 1e. f: Midcoronal mesiodistal longitudinal section through maxillary right molars seen from lingual side. g–k: Serial transverse grinding of right mandibular molars shown in Fig. 1f. l: Midcoronal mesiodistal longitudinal section through mandibular right molars seen from lingual side. This was most conspicuous close to the enamel-dentin junction where the transverse inclination of prisms was less pronounced. Little or no interprism was observed between prisms belonging to adjacent lamellae (Fig. 7e). However, in areas where the enamel was somewhat hypomineralized, a substance occupying the position of the interprism surrounded the prisms completely (Fig. 7a). Generally, the crystals constituting the interprism were oriented roughly at right angles to the prisms and their constituent crystals (Fig. 7e). Thus, in longitudi- 273 MOUSE MOLAR ENAMEL prisms coursed toward the enamel surface (Figs. 5a, 10a). Concomitantly, the orientation of the interprism crystals shifted from a distinct cervical inclination to an orientation approximately perpendicular to the enamel surface (Figs. 5a, 10a). In this way, the distinction between prisms and interprism became less obvious, due to a diminished difference in orientation of their constituent crystals (Fig. 10b). In transverse tooth sections, the prisms in the inner enamel tended to be arranged in radial lines delimited by radially directed, rather coarse sheets of interprism (Figs. 7f, 10c). Crosscut prism profiles varied greatly in size and shape but were mostly ovoid. Toward the middle or outer part of the enamel, the distinction between prisms and interprism became less evident (Fig. 10c). This was apparently due to a diminished difference in orientation between their constituent crystals (Fig. 10d), in agreement with the observations in the longitudinal plane. This blurred configuration could persist to the enamel surface, but sometimes a distinct prism-interprism pattern was reestablished, either in the form of a honeycomb pattern similar to that found in the outer part of the enamel with prism decussation or in the form of rounded prisms embedded in abundant interprism (Fig. 10e). The superficial 3–5 µm of the enamel was prism-free (Figs. 5a, 10a). Aprismatic Enamel Fig. 4. Groove region between cusps 1 and 2 of M1 from fourth transversely ground plane (see location in Fig. 3d). Thin enamel (E), which covers distal aspect of cusp 1, is thickened at the ridge connected to cusp 2 and shows a break in its continuity (arrow) corresponding to a tongue-like extension of EFA similar to that shown in Fig. 2a. D, dentin. 3212. nal sections the crystals of the interprism exhibited a distinct cervical inclination (Fig. 7a). Toward the outer enamel without prism decussation, the orientation of the interprism crystals changed, becoming less cervically inclined and almost perpendicular to the enamel surface (Figs. 5a, 7a). In transverse tooth sections the interprism of the outer enamel formed a distinct honeycomb pattern in which the ovoid ‘‘cells’’ were occupied by prisms (Figs. 7c, 8b). The superficial 3–5 µm of the enamel was devoid of prisms (Figs. 5a, 7a–c). Enamel With Parallel Prisms Only Regions with parallel prisms throughout the whole thickness of enamel were typically, but with some exceptions, found at the summit of the convex aspect of the cusps (Figs. 5a, 6, 8a, 10a,c) and also at the bottom of grooves/invaginations (Fig. 6). Generally, the prisms were inclined occlusally (Figs. 5a, 10a). Their angle relative to the enamel-dentin junction varied considerably from region to region and could increase from about 20° at the tip of a cusp to about 90° at the bottom of a groove. An increase in occlusal inclination of prisms in the outer enamel as found in regions with prism decussation was not observed. On the contrary, the angle between prisms and the enamel-dentin junction tended to increase as the The thin enamel covering the distal aspect of the cusps in the maxillary molars and the mesial aspect of the cusps in the mandibular molars was partly aprismatic, prisms being especially absent in its thinnest parts (Figs. 5b, 11a). Also, the enamel situated adjacent to the enamel-free areas tended to be aprismatic (Fig. 11b,c). The crystals were generally oriented perpendicular to the enamel-dentin junction and the enamel surface. Cracks (Fig. 11b,c) were a more frequent finding in aprismatic than in prismatic enamel and were visible also on the enamel surface (Fig. 11d). In transition zones between prismatic and aprismatic enamel, scattered prisms were embedded in abundant interprism (Fig. 11c). In the aprismatic enamel, lines resembling incremental lines, with a periodicity of about 1 µm, were frequently observed (Fig. 11a,b). Enamel Maturation Generally, the molar enamel of 14–17-day-old mice appeared fully matured with distinct crystals after acid-etching. Locally, however, areas of hypomineralized enamel could be found. Although areas of hypomineralized enamel could persist also in 35-day-old mice, especially in the bottom of the grooves (Fig. 7g), there was less of it than in younger mice. DISCUSSION Crown Morphology The crown morphology of mouse molars presented here relates to the BALB/c strain. Grüneberg (1965) found minor variations in normal crown morphology among various inbred strains and also among wild mice, while aberrations of a more major character were observed in certain mutants. There are fundamental similarities but also minor differences between the 274 S.P. LYNGSTADAAS ET AL. Fig. 5. Types and typical locations of enamel structure configurations. Longitudinal plane, age 35 days. a: Occlusally to contact area of 1 M and M2 (see location in Fig. 3f). M1 shows prism decussation in inner enamel (IE) and parallel prism in outer enamel (OE), while M2 shows parallel prisms throughout whole thickness of enamel. Superficial enamel (SE) is prism-free. DEJ, dentinoenamel junction. R, resin. 3680. b: Thin aprismatic enamel (E) covers distal aspect of cusp 1 in M1 (see location in Fig. 3f). D, dentin; R, resin. 31,250. Fig. 6. Distribution of prism decussation (hatched areas) in mouse molars. Dentin is stippled. The left part of figure shows midcoronal mesiodistal longitudinal planes through maxillary (MAX.) and mandibular (MAN.) right molars seen from lingual side. The right part of figure shows midcuspal transverse planes through maxillary and mandibular right molars seen from occlusal aspect. The plane through maxillary molars corresponds to the fourth transversely ground plane shown in Fig. 3d. The plane through mandibular molars represents parts of several transversely ground planes (i.e., those of Fig. 3h–k). Arrows indicate direction of force during mastication. crown morphology of mouse and rat molars (Kurahashi et al., 1968). The nomenclature of mouse molar cusps proposed by Gaunt (1955) involves an aspect of cusp homology based on interpretations of the sequence of development of cusps during odontogenesis (Gaunt, 1955, 1956, 1961b). From reconstructions of the molar tooth germs based on serial-sectioning of mouse jaws at various ages, Gaunt concluded that second molars were homologous with the distal part of first molars and that third molars were homologous with the mesial part of second molars. From a morphological point of view, however, we found that third molars bore a stronger resemblance to the distal than to the mesial part of second molars. Especially the three cusps in the mandibular third molar were more readily comparable with the three distal cusps (L3, B3, and 4) than the two mesial (L2, B2) and the distolingual (L3) cusps on the second molar. We therefore propose that the cusp nomenclature should be modified for the third molars, from L2, B2, and L3 to L3, B3, and 4 in the lower jaw (Fig. 1f) and from 2, L2, and B2 to 3, L2, and B3 in the upper jaw (Fig. 1e). The available odontogenetic evidence is not sufficiently conclusive and consistent to reject the possibility that each MOUSE MOLAR ENAMEL of the two posterior molars are formed through a reduction of the mesial part of the preceding tooth. From Gaunt’s (1955) reconstructions of the molar germs, it appears that the middle and distal parts of the first molars develop at about the same time and only slightly before the mesial part, both with respect to initial cusp height and onset of dentin and enamel formation. The identification by Gaunt (1955) of two central cusps (2 and 3) on the maxillary third molar is questionable. It seems unlikely that the maxillary and not the mandibular third molar should retain all the major components, although rudimentary, of the second molar. Enamel Distribution Enamel does not cover murine molar tooth crowns completely. It is well known that enamel-free areas (EFA) are present at the tip of the cusps both in the rat (Addison and Appleton, 1921; Mellanby, 1939) and in the mouse (Mahn, 1890; Gaunt, 1956, 1961a; Cohn, 1957). Prior to occlusal attrition, the cusp tips were concave, sometimes accentuated by a distinct pit in the EFA, the apical pit of Gaunt (1956, 1961a). The elevated rim of dentin and enamel is rapidly worn away during attrition. The formation of the enamel-free areas on the cusps of mouse and rat molars have been addressed by several investigators. Addison and Appleton (1921), Mellanby (1939), Gaunt (1956), and Cohn (1957) concluded that the cells of the inner enamel epithelium adjacent to enamel-free regions did not differentiate into secretory ameloblasts and consequently failed to produce enamel. However, more recent investigations have demonstrated that the epithelial cells over the enamel-free areas elaborate a thin, noncontinuous layer of organic matrix which contains amelogenin (Nakamura et al., 1991; Inai et al., 1992), which becomes mineralized by deposition of fine crystals (Sakakura et al., 1989) and which on demineralization leaves an organic residue devoid of collagen (Sakakura et al., 1989). Contrary to Sutcliffe and Owens (1980), who with the SEM were able to observe what seemed to be remnants of such an enamel layer on the EFA of rat molars, we could not find it on molars of 14-day-old mice. It has been suggested that this rudimentary enamel layer, together with the adjacent and subjacent dentin, is subjected to resorption prior to eruption (Sutcliffe and Owens, 1980, 1981), which is supported by the scalloped appearance of the dentin surface in the EFA observed in the present study. It is possible that the tendency toward enamel production on EFA is greater and/or that the tendency toward its resorption is slighter in the rat than in the mouse. This is corroborated by Johannessen’s (1961) finding that enamel was often present on the supposed EFA of unworn rat molars. We have demonstrated the presence of a thin layer of enamel covering the aspect of the cusps overhanging the transverse grooves, facing distally in the maxillary molars and mesially in the mandibular molars. These aspects have previously been described as enamel-free (Gaunt, 1956). Enamel Structure The structure of mouse molar enamel has many features in common with rat molar enamel (Skobe and 275 Stern, 1978; Risnes, 1979a). The characteristic uniserial lamellar prism pattern (Korvenkontio, 1934) present in the incisor enamel (Møinichen et al., 1996) dominates also in the molars. As in the rat (Risnes, 1979a), there are features which distinguish the structure of molar enamel from that of the incisor, including the presence of extensive regions without prism decussation and the variability in distinctness and distribution of prism decussation. The similarity of molar enamel to incisor enamel becomes especially evident when cusps are sectioned transversely. A transversely sectioned molar cusp (e.g., Fig. 8b) resembles a transversely sectioned incisor (Møinichen et al., 1996) where a layer of enamel with prism decussation in the inner enamel and parallel prisms in the outer enamel covers a markedly convex dentin surface. Contributing to the similarity is the fact that the prism rows tend to be oriented transversely to the cusp axis just as they are oriented transversely to the long axis of the incisor (Risnes, 1979b; Møinichen et al., 1996). Furthermore, the reconstructions of mouse molar development by Gaunt (1956) showed that the front of enamel formation on a cusp at a certain stage is C-shaped, with the arms of the C pointing occlusally. This is in accordance with the finding of Smith and Warshawsky (1976) that the front of enamel formation in rat incisors is correspondingly C-shaped, with the arms of the C pointing incisally. Finally, the eccentric distribution of enamel on the cusps, with abundant enamel on the convex aspect and lack of a prism-free area and its tongue-like extension or only very thin enamel on the opposite aspect, resemble the distribution of enamel on the incisors. Thus, it may be argued that murine incisors and molar cusps are analogous. Also, functional considerations support the incisor-cusp analogy. With a mandibular power stroke directed mesially and somewhat lingually as in the rat (Hiiemä, 1967; Weijs, 1975), the cusps of both upper and lower molars will meet the food and antagonist in a direction similar to that of the incisors with respect to enamel distribution. In the mouse, the distribution of prism decussation in molar enamel relative to the direction of the mandibular power stroke is different and in fact rather the opposite of the corresponding relationship in Arvicolidae (Koenigswald, 1982) (Fig. 6). In Arvicolidae, prism decussation is absent, while in the mouse it is partly present in the trailing edge enamel. In Arvicolidae, the leading edge enamel contains prism decussation, while in the mouse the leading edge enamel facing the transverse grooves is thin and lacks prism decussation. These differences may be related to the different cuspal arrangement on the molars of mouse and Arvicolidae, causing different distribution of stress zones in the enamel. In the present study, it was observed that the distinction between prisms and interprism is often obscured in the middle and outer enamel in regions without prism decussation, due to a diminished difference in the orientation of their crystals. It appears that this is mainly due to a change in the orientation of the interprism crystals from a cervical one to one perpendicular to the enamel surface but enhanced by a concomitant erection of the prisms. A similar change in interprism crystal orientation occurs at the transition from inner to outer enamel in regions with prism 276 S.P. LYNGSTADAAS ET AL. Fig. 7. MOUSE MOLAR ENAMEL 277 Fig. 8. Transversely ground planes through molar cusps. Age 35 days. a: First ground plane through cusp 1 of M1 (see location in Fig. 3a). Prism decussation in inner enamel (IE) is only vaguely expressed at summit of convex aspect of cusp. On side aspects of cusp, prism lamellae deviate away from dentin (D) in direction away from summit of convex aspect of cusp. EFA, enamel-free area; OE, outer enamel. 3252. b: First ground plane through cusp L1 of M1 (see location in Fig. 3g). Prism decussation is present in inner enamel (IE) over whole convex aspect of cusp. D, dentin; EFA, enamel-free area; OE, outer enamel. 3252. Fig. 7. Enamel with prism decussation. Age 35 days. a,b,g: Longitudinal plane. c–f: Transverse plane. a: Enamel on mesial aspect of left M3 (see corresponding location in Fig. 3p) shows prism decussation in inner enamel (IE) and parallel prisms in outer enamel (OE). Superficial enamel (SE) is prism-free. There is a great difference in crystal orientation between interprism (white arrows) and prisms (black arrows) both in inner and outer enamel. To the right is a region with hypomineralized enamel. Here, prism domains (p) are completely surrounded by interprism (ip). DEJ, dentinoenamel junction. Occlusal direction toward left. 31,930. b: Cervical enamel mesially on left M1 (see corresponding location in Fig. 3l). Inner enamel (IE) and outer enamel (OE) fades away; only prism-free superficial enamel (SE) remains at end of enamel attenuation. DEJ, dentinoenamel junction. R, resin. 31,050. c: Contact region between M1 and M2 in fourth transversely ground plane (see location in Figure 3j). Both molars show prism decussation in inner enamel (IE). In outer enamel (OE), prisms tend to be arranged in honeycomb pattern, especially evident in M1. Superficial enamel (SE) is prism-free. DEJ, dentinoenamel junction. 3745. d: Inner enamel with prism decussation in fifth transversely ground plane of M1 (mesiobuccal location in Fig. 3e). Transverse inclination of prisms increase from dentinoenamel junction (DEJ) and outwards. Radial sheets of interprism (arrows) are especially evident close to dentin. Various prism pattern aberrations are present. 31,200. e: Decussating prisms (p) with abundant interprism (ip) between prisms of the same lamella and little or no interprism between prisms of adjacent lamellae. DEJ, dentinoenamel junction. 35,000. f: Isolated area of prism decussation in fifth transversely ground plane of M1 (see location on mesial aspect of cusp B2 in Fig. 3e). Radial sheets of interprism (arrows) are more readily visible in areas without than with prism decussation. DEJ, dentinoenamel junction. 3680. g: Bottom of groove between cusps 1 and 2 of M1 (see location in Fig. 3f). Enamel (E) exhibits tendencies of decussating prisms (arrows) and immature enamel. D, dentin; R, resin. 3810. decussation, but here the distinction between prisms and interprism is retained due to a concomitant, distinctly increased incisal inclination of the prisms. The formation of enamel lacking prisms is generally related to ameloblasts lacking Tomes’ processes (Risnes, in press). There are two possible explanations for the convergence of prism and interprism crystals observed in enamel without decussation in the present study: 1) there may be a regression of Tomes’ processes, and 2) Tomes’ processes remain intact, but the direction of ameloblast movement, and hence orientation of prism crystals, approaches the orientation of interprism crystals. If the second explanation is correct, it means that there are no inherent physicochemical properties of the matrix or the mineral that enforce a certain angle between prism and interprism crystals. The reason prism decussation is largely absent from the summit of the convex aspect of the cusps and from the deepest part of grooves/invaginations remains to be clarified. The same phenomenon was observed in rat molars (Risnes, 1979a). It may be due to the spatial conditions in an ameloblastema which covers the complexly folded molar surface or to regional differences in ameloblast characteristics. The thickness of mouse molar enamel of about 100 µm is produced in about 15 days (Cohn, 1957), giving a production rate of about 7 µm per day, very close to the 278 S.P. LYNGSTADAAS ET AL. Fig. 9. Schematic representation of orientation of prism lamellae in regions with prism decussation. Based on observations of serial transverse grindings. The left part of the figure shows right maxillary (MAX.) and mandibular (MAN.) molars viewed from the lingual side, while the right part of the figure shows same from the buccal side. Fig. 10. Enamel without prism decussation. Age 35 days. a: Longitudinal plane through cusp B3 of M1 (see location in Fig. 1f). Orientation of interprism crystals (white arrows) and prisms/prism crystals (black arrows) changes from dentinoenamel junction (DEJ) toward surface, diminishing the difference in crystal orientation. Occlusal direction toward left. 31,150. b: Third transversely ground plane through cusp B3 of M1 (see location in Fig. 3i) showing approximately same enamel region as in Fig. 10a. Prisms are distinct in inner enamel, are hardly distinguishable in middle enamel, and are again visible in outer enamel. Arrows point at radial sheets of interprism. DEJ, dentinoenamel junction. 31,010. production rate calculated for mouse incisor enamel (Møinichen et al., 1996) but considerably slower than the 12–13 µm per day apposition rate calculated for rat incisor enamel (Risnes, 1979c; Smith and Nanci, 1989). The thin enamel covering the distal aspect of the cusps in the maxillary molars and the mesial aspect of the cusps in the mandibular molars may represent an enamel layer produced at a normal rate for only a short period of time, an enamel layer produced at a slow rate for a normal period of time, or an enamel layer pro- 279 MOUSE MOLAR ENAMEL Fig. 11. Aprismatic enamel. a: Transverse plane. Thin aprismatic enamel (E) on distal aspect of cusps 1/L1 of M1 (see location in Fig. 3e). Incremental lines (arrows) with periodicity of about 1 µm are oriented parallel with dentin surface. D, dentin. Age 35 days. 32,500. b: Longitudinal plane. Aprismatic enamel (E) adjacent to EFA on M3 (see location in Fig. 3f). Incremental lines (arrows) spaced about 1 µm run parallel with dentin surface. Cracks (C) traverse whole enamel thickness. D, dentin; R, resin. Age 35 days. 32,400. c: Transverse plane. Aprismatic enamel (E) adjacent to EFA on cusp B3 of M1 (see location in Fig. 3h). At transition from prismatic enamel (to the right), occasional prisms (p) are embedded in abundant interprism (ip). Cracks (C) may traverse whole thickness of enamel. D, dentin. Age 35 days. 31,550. d: Enamel (E) abutting on EFA between cusps L1 and L2 on M1 (see location in Fig. 1b). Dentin surface (D) exhibits globular appearance. Enamel (E) surface adjacent to EFA shows many cracks (C). Age 14 days. 33,540. duced at a slow rate for a short period of time. The presence of closely spaced lines with a periodicity of about 1 µm, resembling incremental lines, indicates a slow rate. of rat (Francis and Briner, 1966) and hamster (Tsjujimoto et al., 1994). Enamel Maturation Areas of hypomineralized enamel are normally present in mouse molar enamel at the time of eruption (14–17 days). Since at 35 days the areas of immature enamel are mostly eliminated, it appears that mouse molar enamel is subjected to posteruptive maturation. This phenomenon has also been described in the molars CONCLUSIONS Based on morphological considerations, modifications in the mouse molar cusp nomenclature of Gaunt (1955) are suggested. The distribution of enamel on the various aspects of the molars has been clarified together with the extent of the enamel-free areas. Considerable variations in its thickness and structure have been demonstrated, which should be taken into consideration when developmen- 280 S.P. LYNGSTADAAS ET AL. tal stages of mouse molars are studied. Such knowledge is crucial for a correct localization and interpretation of thin sections from the complexly folded molars. A great advantage of scanning electron microscopy in this connection is that it is possible to combine detailed structural information with an exact localization within the tooth. Tooth development is regulated by molecular mechanisms common to most organs (Thesleff et al., 1995b). The mouse molar is a convenient experimental model for the study of the basic mechanisms of organ development, including cell differentiation, cell interaction, morphogenesis, and production and mineralization of extracellular matrices. The present study provides a part of the morphological foundation which such studies should be based on and related to. ACKNOWLEDGMENTS This work was supported by grants from the Research Council of Norway and the Norwegian Cancer Society. We are grateful to D. Sørensen and his staff at The Laboratory Animal Unit, The National Hospital, University of Oslo, for the use of the animal facilities, to S. Stølen for his technical assistance with the SEM, and to P. Gran for the photographic work. LITERATURE CITED Addison, W.H.F., and J.L. Appleton, Jr. 1921 On the development of the ameloblasts of the molars of the albino rat, with special reference to the enamel-free areas. Anat. Rec., 21:43. Brookes, S.J., C. Robinson, J. Kirkham, and W.A. Bonass 1995 Biochemistry and molecular biology of amelogenin proteins of developing dental enamel. Arch. Oral Biol., 40:1–14. Chen, E., R. Piddington, S. Decker, J. Park, Z.-A. Yuan, W.R. Abrams, J. Rosenbloom, G. Feldman, and C.W. Gibson 1994 Regulation of amelogenin gene expression during tooth development. Dev. Dyn., 199:189–198. Cohn, S.A. 1957 Development of the molar teeth of the albino mouse. Am. J. Anat., 101:295–320. Francis, M.D., and W.W. Briner 1966 The development and regression of hypomineralized areas of rat molars. Arch. Oral Biol., 11:349– 354. Gaunt, W.A. 1955 The development of the molar pattern of the mouse. Acta Anat., 24:249–268. Gaunt, W.A. 1956 The development of enamel and dentine on the molars of the mouse, with account of the enamel-free areas. Acta Anat., 28:111–134. Gaunt, W.A. 1961a The presence of apical pits on the lower cheek teeth of the mouse. Acta Anat., 44:146–158. Gaunt, W.A. 1961b The development of the molar pattern of the golden hamster (Mesocricetus auratus W.), together with a re-assessment of the molar pattern of the mouse (Mus musculus) Acta Anat., 45:219–251. Grüneberg, H. 1965 Genes and genotypes affecting the teeth of the mouse. J. Embryol. Exp. Morphol., 14:137–159. Hiiemäe, K.M. 1967 Masticatory function in the mammals. J. Dent. Res., 46:883–893. Inai, T., K. Nagata, T. Kukita, and K. Kurisu 1992 Demonstration of amelogenin in the enamel-free cusps of rat molar tooth germs: Immunofluorescent and immunoelectron microscopic studies. Anat. Rec., 233:588–596. Johannessen, L.B. 1961 Presence of enamel-covered cusps in rat molars. Arch. Oral Biol., 5:61–62. Koenigswald, W. von 1982 Enamel structure in the molars of Arvicolidae (Rodentia, Mammalia), a key to functional morphology and phylogeny. In: Teeth: Form, Function and Evolution. B. Kurtén ed. Columbia University Press, New York, pp. 109–122. Korvenkontio, V.A. 1934 Mikroskopische Untersuchungen an Nagerincisiven unter Hinweis auf die Schmelzstruktur der Backenzähne. Histologisch-phyletische Studie. Ann. Zool. Soc. Zool.-Bot. Fenn. Vanamo, 2:1–274. Kurahashi, Y., N. Nagai, K. Watanabe, H. Watanabe, and K. Yama 1968 Chronological observation of the odontogenesis of rat molars. Bull. Tokyo Dent. Coll., 9:147–159. Linde, A., and M. Goldberg 1993 Dentinogenesis. Crit. Rev. Oral Biol. Med., 4:679–728. Lumsden, A.G.S. 1979 Pattern formation in the molar dentition of the mouse. J. Biol. Buccale, 7:77–103. Lyngstadaas, S.P., S. Risnes, B.S. Sproat, P.S. Thrane, and H.P. Prydz 1995 A synthetic, chemically modified ribozyme eliminates amelogenin, the major translation product in developing mouse enamel in vivo. EMBO J., 14:101–106. Mahn, R. 1890 Bau und Entwicklung der Molaren bei Mus und Arvicola. Morphol. Jahrb., 16:652–685. Matzuk, M.M., N. Lu, H. Vogel, K. Sellheyer, D.R. Roop, and A. Bradley 1995 Multiple defects and perinatal death in mice deficient in follistatin. Nature, 374:360–363. Mellanby, H. 1939 The development of teeth in the albino rat. Br. Dent. J., 66:76–86. Møinichen, C.B., S.P. Lyngstadaas, and S. Risnes 1996 Morphological characteristics of mouse incisor enamel. J. Anat., 189:325–333. Nakamura, M., P. Bringas, Jr., and H.C. Slavkin 1991 Inner enamel epithelia synthesize and secrete enamel proteins during mouse molar occlusal ‘‘enamel-free area’’ development. J. Craniofac. Genet. Dev. Biol., 11:96–104. Risnes, S. 1979a The prism pattern of rat molar enamel: A scanning electron microscope study. Am. J. Anat., 155:245–258. Risnes, S. 1979b A scanning electron microscope study of aberrations in the prism pattern of rat incisor inner enamel. Am. J. Anat., 154:419–436. Risnes, S. 1979c A method for calculating the speed of movement of ameloblasts during rat incisor amelogenesis. Arch. Oral. Biol., 24:299–306. Risnes, S. 1981 A rotating specimen holder for hard tissue sectioning. Stain Technol., 56:265–266. Risnes, S. 1982 Multiangular viewing of dental enamel in the SEM: A simple specimen holder system. Scand. J. Dent. Res., 90:80–82. Risnes, S. 1985 Multiangular viewing of dental enamel in the SEM: An apparatus for controlled mechanical specimen preparation. Scand. J. Dent. Res., 93:135–138. Risnes, S. 1990 Shark tooth morphogenesis. An SEM and EDX analysis of enameloid and dentin development in various shark species. J. Biol. Buccale, 18:237–248. Risnes, S. in press. Growth tracks in dental enamel. J. Hum. Evol. Risnes, S., C.B. Møinichen, D. Septier, and M. Goldberg. 1996 Effects of accelerated eruption on the enamel of rat lower incisor. Adv. Dent. Res., 10:261–269. Sakakura, Y., N. Fujiwara, and T. Nawa 1989 Epithelial cytodifferentiation and extracellular matrix formation in enamel-free areas of the occlusal cusp during development of mouse molars: Light and electron microscope studies. Am. J. Anat., 184:287–297. Sasaki, T., M. Goldberg, S. Takuma, and P. R. Garant 1990 Cell biology of tooth formation. Monogr. Oral Sci., 14:1–204. Skobe, Z., and D. Stern 1978 Scanning and transmission electron microscopy of enamel formation and structure in the molar teeth of rats. Arch. Oral Biol., 23:307–316. Slavkin, H. 1991 Molecular determinants during dental morphogenesis and cytodifferentiation: A review. J. Craniofac. Genet. Dev. Biol., 11:338–349. Smith, C.E., and Nanci, A. 1989 Secretory activity as a function of the development and maturation of ameloblasts. Connect. Tissue Res., 22:147–156. Smith, C.E., and H. Warshawsky 1976 Movement of entire cell populations during renewal of the rat incisor as shown by autoradiography after labeling with 3H-thymidine. The concept of a continuously differentiating cross-sectional segment. Am. J Anat., 145:225–259. Sutcliffe, J.E., and P.D.A. Owens 1980 A light and scanning electron microscopic study of the development of enamel-free areas on the molar teeth of the rat. Arch. Oral Biol., 25:263–268. Sutcliffe, J.E., and P.D.A. Owens 1981 Microscopy of resorption of enamel and dentine from the cusp tips of the molar teeth of rats. Arch. Oral Biol., 26:29–39. Thesleff, I., A. Vaahtokari, P. Kettunen, and T. Åberg 1995a Epithelialmesenchymal signaling during tooth development. Connect. Tissue Res., 32:9–15. Thesleff, I., A. Vaahtokari, and A.-M. Partanen 1995b Regulation of organogenesis. Common molecular mechanisms regulating the development of teeth and other tissues. Int. J. Dev. Biol., 39: 35–50. Tsjujimoto, Y., H. Ishibashi, and M. Yamazaki 1994 Study on hypomineralized areas in hamster molars. J. Nihon Univ. Sch. Dent., 36:122–126. Weijs, W.A. 1975 Mandibular movements of the albino rat during feeding. J Morphol., 145:107–124.