JOURNAL OF MORPHOLOGY 267:1048 –1065 (2006)

Morphological Variations in a Tooth Family Through Ontogeny in Pleurodeles waltl (Lissamphibia, Caudata) Tiphaine Davit-Be´al, Franc¸oise Allizard, and Jean-Yves Sire* CNRS UMR 7138, Equipe “Evolution & De´veloppement du Squelette,” Universite´ Paris 6, 75005 Paris, France ABSTRACT Most nonmammalian species replace their teeth continuously (so-called polyphyodonty), which allows morphological and structural modifications to occur during ontogeny. We have chosen Pleurodeles waltl, a salamander easy to rear in the laboratory, as a model species to establish the morphological foundations necessary for future molecular approaches aiming to understand not only molecular processes involved in tooth development and replacement, but also their changes, notably during metamorphosis, that might usefully inform studies of modifications of tooth morphology during evolution. In order to determine when (in which developmental stage) and how (progressively or suddenly) tooth modifications take place during ontogeny, we concentrated our observations on a single tooth family, located at position I, closest to the symphysis on the left lower jaw. We monitored the development and replacement of the six first teeth in a large growth series ranging from 10-day-old embryos (tooth I1) to adult specimens (tooth I6), using light (LM), scanning (SEM), and transmission electron (TEM) microscopy. A timetable of the developmental and functional period is provided for the six teeth, and tooth development is compared in larvae and young adults. In P. waltl the first functional tooth is not replaced when the second generation tooth forms, in contrast to what occurs for the later generation teeth, leading to the presence of two functional teeth in a single position during the first 2 months of life. Larval tooth I1 shows dramatically different features when compared to adult tooth I6: a dividing zone has appeared between the dentin cone and the pedicel; the pulp cavity has enlarged, allowing accommodation of large blood vessels; the odontoblasts are well organized along the dentin surface; tubules have appeared in the dentin; and teeth have become bicuspidate. Most of these modifications take place progressively from one tooth generation to the next, but the change from monocuspid to bicuspid tooth occurs during the tooth I3 to tooth I4 transition at metamorphosis. J. Morphol. 267:1048 –1065, 2006. © 2006 Wiley-Liss, Inc. KEY WORDS: Lissamphibia; Pleurodeles waltl; tooth; cusps; pedicel; development; SEM; TEM

Molecular biology has enabled a rapid increase of knowledge of the genetic control of tooth development in mammals, particularly in the mouse model. The expression pattern of hundreds of genes has been depicted (see website: [email protected]), but, although functional studies have demonstrated the role of some of these genes, numerous questions © 2006 WILEY-LISS, INC.

remain unanswered. Mammalian, and especially rodent dentition, is highly derived and specialized: reduction of tooth number, teeth restricted to the jaw bones, various tooth types, complex cusp patterns, and only one or two tooth generations. These features contrast with the condition found in the other vertebrate clades: high number of similar teeth, numerous dentigerous bones, simple cusp pattern, and repeated tooth replacement during life (polyphyodonty). Mammalian dentition derives from such a polyphyodont dentition. Comparative studies of tooth development in nonmammalian, polyphyodont models will advance our understanding of the molecular processes involved in tooth development and replacement (tooth patterning, tooth initiation, tooth type, cusp formation, etc.). A number of nonmammalian species, which have several tooth generations during the larval period, are easy to breed. Potentially relevant species are encountered either within Teleostei (trout, carp, tilapia, etc.) or within Caudata (salamanders and newts) (Sire et al., 2002). The other possible model species, chondrichthyans (sharks and rays), anurans (toads and frogs), and reptiles (crocodiles, lizards, and snakes) develop teeth late in ontogeny, at the end of a long embryonic period, and they lack the typical first-generation teeth described in taxa with a short embryonic period (Sire et al., 2002). In Teleostei and Caudata, the young larvae possess firstgeneration teeth, which are similarly conical in all species and show a simple structure (reduced pulp cavity and absence of dentin tubules) (Huysseune and Sire, 1997a,b; Huysseune and Sire, 1998; DavitBe´al et al., 2006). Such “minimalistic” teeth are considered to represent the ancestral condition for the first-generation teeth in vertebrates (Sire et al., 2002). Several teleost species, among which the zebrafish, Danio rerio, and the medaka, Oryzias latipes, both easy to breed and intensively used in de-

*Correspondence to: Dr. Jean-Yves Sire, Universite´ Paris 6, UMR 7138, Case courrier 5, 7 quai St-Bernard, 75251 Paris Cedex 05, France. E-mail: [email protected] Published online 24 May 2006 in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/jmor.10455

TOOTH MORPHOLOGY IN P. WALTL

velopmental genetics, are obvious candidates for evolutionary developmental studies on dentition. Indeed, numerous molecular tools and a large number of mutants are available in these species. However, there are some disadvantages of working with these teleost species. In the zebrafish, knowledge of tooth development has been acquired only recently (Huysseune et al., 1998; Van der heyden and Huysseune, 2000; Van der heyden et al., 2000) and there are virtually no data for the medaka. Moreover, it is difficult to study tooth development in the zebrafish (teeth are located in the pharyngeal region only) and the first molecular data on tooth development have been obtained only recently (Laurenti et al., 2004; Jackman et al., 2004; Borday-Birraux et al., 2006). In contrast to the zebrafish, numerous data on tooth development in caudates have accumulated during the last century (see review in Davit-Be´al et al., 2006). Among the several species in which teeth have been studied, Pleurodeles waltl, a salamander easy to breed in the laboratory, has been used in the past as an experimental model for odontogenetic studies (Chibon, 1966, 1967, 1970, 1972). Lissamphibians are phylogenetically closer to mammals (both are tetrapods) than teleosts, and in juveniles and adults the tooth structure is more similar to that in mammals than to that in teleosts. In P. waltl, several tooth generations succeed each other, starting from the first larval stage and continuing throughout life. Tooth size, shape, number, orientation, and structure are modified during ontogeny (Chibon, 1977). Therefore, P. waltl appears to be an excellent model species for investigations aiming to understand how such changes are achieved through ontogeny. Over the past years, several studies have been devoted to the description of tooth development in larval and/or adult Pleurodeles waltl and in other lissamphibians (review in Davit-Be´al et al., 2006). The morphological and structural changes that occur in teeth during ontogeny, from larvae to adults include: 1) a shift from monocuspid teeth in larvae to bicuspid teeth in adults (Kerr, 1960; Chibon, 1977; Clemen and Greven, 1977; Wistuba et al., 2002); 2) the presence, in juveniles and adults but not in the first larval stages, of teeth characterized by a dentin cone prolonged by a pedicel, both being separated by a dividing zone (Howes, 1978; Greven, 1989; Greven and Clemen, 1990; Wistuba et al., 2002); and 3) a shift from enameloid-covered larval teeth (Smith and Miles, 1971; Roux, 1973; Roux and Chibon, 1973; Bolte and Clemen, 1992; Kogaya, 1999) to enamel-covered juvenile and adult teeth (Smith and Miles, 1971). However, this knowledge was obtained from various species and from dispersed studies dealing with either specific aspects of tooth morphogenesis and cytodifferentiation, or morphological descriptions using light (LM), scanning (SEM), or transmission electron microscopy (TEM) (reviewed in Davit-Be´al et al., 2006). In particular, we need to

1049

know at which stage the typical bicuspid tooth is achieved and whether the enameloid/enamel transition, and the formation of the pedicel, occur progressively or suddenly through ontogeny. Answering these questions at the morphological level will permit us to lay solid foundations for further studies aiming to understand the underlying molecular processes. The present study was designed to answer some of these questions, using Pleurodeles waltl as a model. In contrast to previous studies that used a few teeth taken separately either in larvae, in juvenile, or in adult specimens (e.g., Smith and Miles, 1971; Wistuba et al., 2002; review in Davit-Be´al et al., 2006), we focused our attention on a single tooth family (i.e., the first-generation tooth at a given position in early larvae and all its successors until adulthood), using LM, SEM, and TEM. A large growth series enabled us to monitor step by step (and tooth after tooth) all morphological modifications of the teeth (both external and structural features) during P. waltl ontogeny.

MATERIALS AND METHODS Biological Material Pleurodeles waltl Michahelles, 1830, is easy to breed (Gallien, 1952) and adults (several pairs) have been kept in our laboratory since 1995. Males and females are sexually mature at about 2 years. The larvae used in the present study come from eggs obtained in April 2001 and from September 2003 to May 2004. The larvae were staged according to Gallien and Durocher (1957). In our breeding conditions (water temperature varying from 16 – 22°C during the year, 12-h light period), hatching occurred at Stage 36, i.e., 13 days postfertilization (dpf), and the larvae took their first prey (Artemia nauplii) at Stage 38, 1 week after hatching (17 dpf). The duration of the larval period was ⬃3.5 months. Metamorphosis ended at Stage 56 (110 dpf) and specimens were sexually mature at 18 months. Several specimens of each of the following ontogenetic stages were used: Embryos: Stages 33 and 34 (10 and 11 dpf, i.e., 3 and 2 days before hatching, respectively); Larvae: each stage sampled from Stage 36 (13 dpf) to Stage 56 (110 dpf, end of metamorphosis); Juveniles: 4-, 5- (75 mm), 6- (96 mm), 8- (100 mm), 10- (115 mm), 12- (128 mm), 14- (130 mm), 15- (135 mm), and 18-monthold (150 mm); Adults: 21, 24-month, and 6-year-old (160 –170 mm). All specimens (a total of 50) were measured, killed with an overdose of anesthetic (MS222), decapitated, and used for LM, SEM, and TEM observations. Some dissected jaws were dehydrated and photographed under the binocular microscope.

Scanning Electron Microscopy (SEM) The right lower jaw quadrant of 19 specimens (10 larval stages from Stages 40 –56; six juveniles from 4 –15-month old, and three adults), was immersed in 2.5% KOH for 2–30 min (depending on the jaw size) at room temperature. The soft tissues were carefully removed using forceps and thin brushes. The samples were then dehydrated in a graded series of ethanol, air-dried, glued on a copper support, coated with a thin layer of gold/palladium, and observed with a JEOL SEM 35 operating at 25 kV, either in mesial or in lingual view.

Journal of Morphology DOI 10.1002/jmor

1050

´ AL ET AL. T. DAVIT-BE

Light (LM) and Transmission Electron Microscopy (TEM) Depending on the size, the specimens were fixed either entirely (larvae until Stage 42), or heads only (larval Stages 43–52), or upper and lower jaws (specimens larger than 25 mm). The samples were immersed for 2 h, at room temperature, in a mixture containing 1.5% glutaraldehyde and 1.5% paraformaldehyde in 0.1 M cacodylate buffer (pH 7.4). Some samples were decalcified, either for 3 days (larvae until Stage 44), 7 days (larvae ⬎Stage 44), 15 days (juveniles), or 21 days (adults) at 4°C in the same fixative, to which 5% EDTA was added. This decalcifying mixture was changed every 2 days. After rinsing in the same buffer, the samples were postfixed in 1% osmium tetroxide in cacodylate buffer and rinsed rapidly. They were next dehydrated using a graded series of ethanol and embedded in Epon for transverse or longitudinal sectioning. Several samples were processed into uninterrupted series of transverse or longitudinal 1-␮m-thick sections using a diamond knife. The sections were stained with Toluidine blue to be examined with a light microscope. In a few samples, the series was interrupted at appropriate locations for ultrathin sectioning. Thin sections were contrasted with uranyl acetate and lead citrate and viewed with a Philips 201 transmission electron microscope operating at 80 kV.

Brief Overview of Tooth Development and Dentition in Pleurodeles waltl In Pleurodeles waltl the first tooth germs (first-generation teeth of Sire et al., 2002) are visible at Stage 33 (10 dpf) (Chibon, 1966, 1970, 1977). These teeth are conical, have a single cusp, and attach to the bone support at hatching (Stage 36, 13 dpf). They pierce the oral epithelium at Stage 37 (15 dpf), when the mouth opens, and become functional. In larvae, all bones of the oral cavity are toothed: a single tooth row is present on the upper (premaxillary bone) and the lower (dentary bone) jaw, several rows are located on the vomer and palatopterygoid (upper jaw), and the coronoid (lower jaw) is entirely covered by teeth; the maxillary ossifies late in ontogeny, a few days before metamorphosis (Signoret, 1959). At metamorphosis (Stage 56, 110 dpf) the palatopterygoid and the coronoid disappear along with the teeth they support, and the vomers expand (Signoret, 1959; Corsin, 1966). During the 3.5-month-long larval period, P. waltl develops 12 tooth positions (numbered I, II, III, etc.) on each lower jaw quadrant. At each position, three to four tooth generations (numbered 1, 2, 3, etc.) develop during the larval period, and tooth replacement continues after metamorphosis and throughout life (Chibon, 1966). The first-generation tooth and all its successors at a given position constitute a tooth family. After metamorphosis and during juvenile and adult life, new tooth positions (⫽families) are added posteriorly on the arcades: a 12-month-old specimen possesses 17–18 tooth positions on each lower jaw quadrant. In previous studies dealing with P. waltl odontogenesis, neither the position of a tooth examined nor its rank in the family were reported. We will see in the following that such data are important as tooth morphology changes through successive replacements.

Choice of the Tooth Family We follow the history of the tooth family located at position I on the right lower jaw, from the first-generation tooth developing in the late embryo, i.e., tooth I1, to its homolog in adulthood (2-yearold specimens), i.e., tooth I6. We chose the lower jaw because 2) in contrast to the upper jaw, it does not change much during metamorphosis; 3) the dentary bone ossifies early (Stage 37); and 4) it bears only a single tooth row in adults. We chose tooth family I because it presents several advantages: i) the first-generation tooth forms early in ontogeny, and therefore is present in all specimens examined from Stage 33 onwards; ii) being one of the most anteriorly located positions it reduces the amount of sec-

Journal of Morphology DOI 10.1002/jmor

tioning; iii) it is the position closest to the symphysis and, therefore, easily recognizable on sections; and iv) each tooth belonging to this family develops and shows the same morphological features as teeth in other positions.

RESULTS The formation of the first-generation teeth and of their successors in the anterior region of the lower jaw in Pleurodeles waltl was monitored by combining observations with the binocular microscope and SEM, LM, and TEM. By the end of the embryonic period (Stages 33–36), tooth II1 develops first, followed by teeth I1 and III1, which develop simultaneously, and then tooth IV1. These first developing teeth are called embryonic teeth, and they are functional (i.e., attached to the dentary bone and their tip piercing the oral epithelium) in early larvae at Stage 37 (15 dpf). Next, their successors (i.e., II2, followed by I2 and III2, then IV2) develop, along with the first-generation teeth of new tooth families (e.g., V1, VI1, etc.), and so on. These teeth are called larval teeth until postmetamorphic stages. It is noteworthy that tooth replacement does not follow exactly the same order as tooth formation. For instance, although I1 and III1 are formed simultaneously in all specimens examined, I2 and III2 appear with various delays, depending on the specimen. Chronology of Tooth Succession The developmental sequence of the six teeth succeeding each other at position I has been deduced from serial sections of late embryos and early larval stages, and from serial sections and SEM for older larval stages, juveniles, and adults. The chronology of tooth succession is summarized in Figure 1. Tooth I1. First visible morphologically 3 days before hatching (Stage 33: 10 dpf), it is well developed after hatching (Stage 36: 13 dpf), and becomes functional (i.e., attached to the dentary bone surface and its tip piercing the oral epithelium) 2 days later (Stage 37: 15 dpf), when the mouth is open. The first resorption lacunae provoked by odontoclasts (⫽osteoclasts) are observed at Stage 48 (50 dpf). The odontoclasts start to resorb the tooth from outside, at the distal-lingual side at midheight of the dentin shaft. Then they penetrate the pulp cavity. Here, they resorb the tooth base and the proximal-labial side of the dentin shaft. Finally, the tooth tip is entirely resorbed. The tooth is no longer present at Stage 52a (64 dpf). The total cycle of embryonic tooth I1 extends over 55 days (Fig. 1), with a developmental period of 6 days, a functional period of ⬃40 days, and a resorption period of 16 days. Because the cycle includes days in which teeth are both functional and resorbing, the total number of days in the cycle is lower than the total number of days listed for each phase of the tooth cycle. Indeed, teeth remain functional until an advanced stage of resorption.

TOOTH MORPHOLOGY IN P. WALTL

1051

Fig. 1. Monitoring of tooth succession at position I (the closest to the symphysis) on the right lower jaw of Pleurodeles waltl. This graphical representation summarizes the results obtained from transverse serial sections of the anterior region of the right lower jaw quadrant in a growth series from late embryos to 2-year-old adults (see Materials and Methods for further details). The first-generation tooth, I1, forms at the end of the embryonic period; teeth I2 and I3 develop during the larval period; tooth I4 development starts before metamorphosis; tooth I5 covers the entire juvenile period; and tooth I6 forms during the juvenile period and its functional life extends to the first months of the adult period. Development covers the period from initiation to attachment. The functional period includes most of the resorption period, as teeth remain functional until an advanced stage of resorption.

Tooth I2. The first replacement tooth at position I is morphologically visible 1 day after hatching (Stage 36: 13 dpf). It develops until Stage 42 (28 dpf), at which time it attaches to the dentary bone near to I1. It is noteworthy that development of I2 does not provoke resorption of I1. The first signs of resorption of I2 are observed at Stage 52a (64 dpf). As described for I1, the odontoclasts first resorb the tooth from outside, at the distal-lingual side, but more toward the tooth base, close to its junction with the bone support. Resorption starts in the region facing the second developing replacement tooth at this position, i.e., I3. The subsequent steps of resorption are similar to those described for I1, including entire resorption of the tooth tip by odontoclasts. I2 is no longer visible on sections at Stage 55c (105 dpf), before the end of the metamorphosis period. Therefore, the total life cycle of I2 is ⬃90 days, with a developmental period of 15 days, a functional period probably covering 50 days, and a resorption period of 40 days (Fig. 1). Tooth I3. This tooth is morphologically visible at Stage 44 (36 dpf). The tooth germ grows until Stage 55a (90 dpf), at which time the tooth attaches to the bone surface. Resorption of I3 starts during the fifth month (⬃135 dpf), i.e., after metamorphosis, and the tooth has disappeared in 6-month-old specimens (⬃180 dpf). The total life cycle of I3 is close to 140

days, with a developmental period of 43 days, a functional period close to 70 days, and a resorption period of 45 days. The ontogenetic period of I3 (from 36 –90 dpf) corresponds to the resorption period of I1 (43– 64 dpf) and to the beginning of I2 resorption (64 –105 dpf) (Fig. 1). Tooth I4. This tooth has started its morphogenesis phase at Stage 55a (90 dpf), i.e., shortly before metamorphosis, which starts at ⬃100 dpf. Therefore, I4 grows during the metamorphosis period. It erupts at 5.5 month pf, and resorption starts around the eighth month. I4 is no longer visible in 10month-old specimens. The total life cycle of I4 is close to 7 months, with a developmental period of 2.5 months, a functional period close to 4 months, and a resorption period of 2.5 months (Fig. 1). Tooth I5. This tooth starts to develop in the fifth month and it attaches to the dentary by the end of the eighth month. Resorption has started in 12month-old specimens, and I5 is no longer present in 15-month-old salamanders. I5 is the first tooth that accomplishes its complete cycle after metamorphosis. Its life cycle covers 10 months, with a developmental period of 4 months, a functional period of ⬃5 months, and a resorption period of ⬃3 months (Fig. 1). Tooth I6. This tooth is first visible at the beginning of the eighth month and attaches around the Journal of Morphology DOI 10.1002/jmor

1052

´ AL ET AL. T. DAVIT-BE

Fig. 2. Tooth orientation. Lingual view of the teeth on the right lower jaw in a growth series of Pleurodeles waltl with attention paid to the six successive teeth at position I. Symphysis on the left. A: General view of the dentition in a 2-year-old adult. Only a few positions lack teeth (arrows). The similarly shaped teeth are arranged in a single row on the arcade. B: Larva, Stage 41 (25 dpf). The first-generation teeth, I1 and II1, are attached to the dentary bone surface and they are two times smaller than the first replacement teeth. C: Larva, Stage 48 (50 dpf). D: Larva, Stage 55a (90 dpf). E: Juvenile, 6-month-old. F: Juvenile, 12-month-old. G: Adult. Except for the first-generation teeth, the teeth are strongly ankylosed to the bone support in a pleurodont mode. They are roughly perpendicular to the bone surface and their upper region slightly curved lingually. Half of their height protrudes out of the bone surface. At the lingual side, their base is pierced by a large foramen. The arrowheads point to the tips of “old” teeth, which were either broken or worn during feeding. Tooth replacement leads to complete resorption of the previous tooth (C,E,F) before attachment of their successors, which have not been conserved during preparation of the samples. Scale bars in A ⫽ 500 ␮m; B ⫽ 50 ␮m; C–G ⫽ 100 ␮m.

14th month. The first resorption features are seen in 18-month-old specimens and the tooth is no longer present in 22-month-old animals. Its total life cycle needs 14 months, with a developmental period of 6 months, a functional period of ⬃8 months, and a resorption period of ⬃4 months (Fig. 1). These six successive teeth are particular in that each is representative of a period of ontogeny: I1 is an embryonic tooth, which starts to be formed in late embryos, before hatching; I2 is a larval tooth, accomplishing its entire cycle during the larval period; I3 is also a larval tooth, but it is subjected to metamorphosis during its functional period; I4 is also subjected to metamorphosis, which occurs during its early phase of development; I5 is a juvenile tooth, its cycle covering the entire juvenile period; and I6 forms at the end of the juvenile period and its cycle extends into adulthood. In addition, it is noteworthy that in all larval stages examined growth and attachment of I2 is not concomitant to I1 resorption. Both teeth are, therefore, functional, and two rows of teeth are present when I3 starts to develop. The growth of this third tooth in the family is responsible for the resorption of I2, and, probably, for the accelerated resorption of I1. There is no indication of what mechanism could be responsible for the first stages of resorption observed for I1, at least at the morphological level, Journal of Morphology DOI 10.1002/jmor

although in all subsequent replacement events there is a clear relation between the growing tooth and the resorption of the previous tooth. Morphological Changes Through Ontogeny In the following, we selected various tooth characters, known to change from one replacement tooth to another during salamander ontogeny. Our descriptions are supported by SEM pictures for external features (Figs. 2–5), and by LM (Figs. 6, 7), and TEM (Fig. 8) data for structural aspects. For comparisons of external features, we focused our attention mostly on functional teeth, either recently attached or older. Changes of external features (Figs. 2–5) Orientation and shape. Two angles were considered to describe tooth orientation, either with regard to the bone support (i.e., labial or lingual angle) or to the oral cavity (i.e., mesial or distal inclination). For the former, pictures were taken from inside the buccal cavity (i.e., lingual views), perpendicular to the anteroposterior axis of the dentary bone (Fig. 2). For the latter, mesial views were taken with the symphysis at the foreground (Fig. 3). Height and width variation from the first-generation

TOOTH MORPHOLOGY IN P. WALTL

1053

Fig. 3. Tooth orientation. Proximal (mesial) view of the teeth on the anterior region of the right mandible in a growth series of Pleurodeles waltl, with particular attention paid to the successive teeth at position I. Symphysis at the foreground. A: Larva, Stage 41 (25 dpf). B: Larva, Stage 48 (50 dpf). C: Larva, Stage 55a (90 dpf). D: Juvenile, 6-month-old. E: Juvenile, 12-month-old. F: Adult. The tooth base is perpendicular to the bone surface, while the upper half of the dentin shaft is curved lingually (angle of 20°). The three first teeth (I1 to I3) have only one cusp, while the other (I4 to I6) have two (see also Fig. 4). Some tooth tips are worn or broken (arrowheads). Note the progressive appearance of the dividing zone through ontogeny (see also Fig. 5). Scale bars in A ⫽ 20 ␮m; B ⫽ 50 ␮m; C–F ⫽ 100 ␮m.

to the sixth tooth are indicated in Table 1. In adult Pleurodeles waltl, the teeth are juxtaposed with a narrow space between (Fig. 2A). Of the 42 tooth positions counted on the right lower jaw, only a few lacked teeth, suggesting a slow replacement process.

The dentition is homodont, i.e., all teeth have a similar bottle-like shape. The attachment is pleurodont, i.e., the tooth base is ankylosed by means of its labial wall to the dentary bone. The first tooth in the series, embryonic tooth I1, differs largely from its

Fig. 4. Cusps. Mesial view of the tip of the six successive teeth at position I on the right lower jaw in a growth series of Pleurodeles waltl. Each representative tooth was chosen as recently attached one to avoid wears and breaks. Lingual side on the right. A: Larva, Stage 40 (22 dpf), tooth I1. B: Larva, Stage 44 (36 dpf), tooth I2. C: Larva, Stage 55a (90 dpf), tooth I3. D: Juvenile, 6-month-old, tooth I4. E: Juvenile, 10-month-old, tooth I5. F: Juvenile, 15-month-old, tooth I6. The tip of teeth I1, I2, and I3 are conical (one cusp) without any morphological indication of another cusp. Two cusps, the large one being at the lingual side, are clearly visible in teeth I4, I5, and I6. Note the progressive development of a median ridge on the mesial side of the cusps (arrowheads). The arrows point to the limit between the enamel covering and the dentin shaft. Scale bars ⫽ 10 ␮m.

Journal of Morphology DOI 10.1002/jmor

1054

´ AL ET AL. T. DAVIT-BE

Fig. 5. Dividing zone. Changes occurring at the mid-region of the six successive teeth at position I of the right lower jaw in a growth series of Pleurodeles waltl. Lingual view, mesial on the left. A: Larva, Stage 41 (25 dpf), tooth I1. B: Larva, Stage 44 (36 dpf), tooth I2. C: Larva, Stage 55a (90 dpf), tooth I3. D: Juvenile, 6-month-old, tooth I4. E: Juvenile, 12-month-old, tooth I5. F: Adult, tooth I6. In the tooth I1, the dividing zone is not visible, but the tooth base is enlarged (arrowhead). In teeth I2 and I3 a clear curvature indicates the future region of the dividing zone (arrow). A narrow, but clear dividing zone is present in teeth I3 to I6 (arrows). Scale bars ⫽ 10 ␮m.

successors by the following characters (cf. Fig. 2B and 2C–G): its height is five to six times smaller (60 – 80 ␮m) than that of I2 (300 –350 ␮m); subsequent teeth at position I increase by ⬃50 ␮m per tooth generation (Table 1); its base is enlarged, the width of which represents half of the tooth height, while the other teeth have a conical base, the width of which does not increase much during ontogeny (Table 1); it is attached to the upper surface of the dentary bone, while the other teeth are attached to the labial wall; it is inclined anteriorly, while the other teeth are implanted perpendicular to the bone surface; and it is separated from its neighbor II1 by

a 150-␮m wide, interdental space, whereas the next generation teeth are located close to each other. Resorption features and absence of teeth are observed more frequently during the larval and juvenile period than in adults. This suggests that young animals replace their teeth more rapidly than adults, as reported in monitoring tooth replacement (Fig. 1). In larvae and juveniles, the tips of “old” functional teeth are often worn or broken off before being replaced (Figs. 2, 3). At the lingual side, the base of each tooth in the series is pierced by a single, large foramen, which communicates with the pulp cavity. Mesial views reveal that, although the tooth

Fig. 6. Transverse (A–I) and longitudinal (J) 1-␮m-thick sections of the anterior region of the right mandible, aiming to compare some developmental steps of larval teeth (A–D) and juvenile/adult teeth (E–H) at position I in Pleurodeles waltl. I: Relation of adjacent tooth positions with regard to the dental lamina. J: Relation of three successive teeth from the same family with regard to the dental lamina. A: Stage 36 (13 dpf), tooth I2; morphogenesis. The epithelial cells of the basal layer of the buccal epithelium have differentiated and form an enamel organ. B: Stage 39 (20 dpf), tooth I2, early cytodifferentiation. The cells of the dental organ have differentiated into an outer and an inner dental epithelium. Facing the latter the mesenchymal cells of the dental papilla have differentiated into odontoblasts. C: Stage 43 (33 dpf), tooth I2, late cytodifferentiation. Odontoblasts have deposited predentin, while ameloblasts have differentiated. D: Stage 44 (36 dpf), tooth I2, end of development. At the tooth tip, enameloid and dentin are maturing, while dental matrix is still deposited along the tooth shaft, the base of which is now close to the dentary bone surface. E: Replacement teeth. While a preceding tooth in the family (I5) is functional, the replacement tooth (I6) is in a late stage of cytodifferentiation. Its successor (tooth I7) is already initiated. An epithelial strand (arrow) extends into the mesenchyme and is connected to the upper part of the dental organ of I6 (arrowhead). F: Germ tooth I6 undergoes its late morphogenesis phase: the cells of the enamel organ and the facing mesenchymal cells are differentiating. G: Tooth I6 is now at the early cytodifferentiation phase. Note its particular orientation with regard to the outer dental epithelium of the preceding tooth I5. Ameloblasts and odontoblasts have differentiated and a thin layer of predentin (arrow) is deposited between both layers. Enamel and dentin of the preceding tooth in the family (I5) are maturing in the upper region, while dental matrix is still being deposited along the future pedicel. Arrowhead points to the connection of the I6 epithelial strand to I5 enamel organ. H: Ameloblasts and odontoblasts of tooth I6 are differentiated and the tooth germ has elongated. I: The dental lamina links four adjacent positions: tooth family I, the closest to the symphysis (cytodifferentiation of I3), family II (cytodifferentiation of II3), III (upper region of the dental organ of III3), and IV (base of IV2, functional). J: Three successive teeth from tooth family I (I4, functional tooth; I5, well-developed germ; I6, initiation stage), linked by a continuous dental lamina. Scale bars in A–D ⫽ 10 ␮m; E–H ⫽ 50 ␮m; I,J ⫽ 100 ␮m. am, ameloblasts; be, buccal epithelium; bv, blood vessel; db, dentary bone; de, dentine; dl, dental lamina; eo, enamel organ; ide, inner dental epithelium; od, odontoblasts.

Journal of Morphology DOI 10.1002/jmor

TOOTH MORPHOLOGY IN P. WALTL

1055

Figure 6

Journal of Morphology DOI 10.1002/jmor

Fig. 7. Transverse, 1-␮m-thick sections of the right mandible, showing the structure of the six successive teeth at position I in a growth series of Pleurodeles waltl. A–C: tooth I1. D: Tooth I2. E: Tooth I3. F: Tooth I4. G: Tooth I5. H–J: Tooth I6. EDTA decalcified samples: the enameloid/enamel covering appears as an empty space along the tooth tip. A,D–H: For convenient comparison all are recently attached teeth. A: Some days after hatching, Stage 36 (13 dpf). The dentin shaft is not entirely mineralized. The tooth base is attached to the thin, developing layer of bone located at the surface of Meckel’s cartilage. The pulp cavity contains only odontoblasts, which are close to osteoblasts in the deep region. The cells of the inner dental epithelium are well polarized along the whole shaft surface. B: Same tooth, Stage 39 (20 dpf). The tooth matrices are well mineralized and the tooth base is ankylosed (primary form of attachment) to the dentary bone that is now well developed. The odontoblasts are still active in the upper region of the pulp cavity, while some necrotic pictures are seen in the deep region. A capillary blood vessel has penetrated the pulp cavity through a foramen. At the shaft surface, the inner dental epithelium has retracted from the tooth base and the cells are no longer polarized. C: Same tooth, Stage 46 (43 dpf). Most odontoblasts have regressed except in the very upper region of the pulp cavity. Several blood vessels occupy the rest of the pulp and large vessels are seen on each sides of the tooth. There is no indication of the presence of a dividing zone. The section passes through the foramen (arrow). The dentary bone has enlarged. D: Stage 44 (36 dpf). The mode of attachment is a secondary one, i.e., the tooth base attaches by means of attachment bone to the already formed dentary bone. The limit between both is indicated by a cement line (white arrow). The pulp cavity is filled with odontoblasts facing a wide layer of predentin matrix. The inner dental epithelium cells are retracting from the tooth base. The tooth shaft is not straight as in the previous tooth and a curvature in the last third part of the tooth suggests the future location of the dividing region. E: Stage 55c (105 dpf). The tooth has attached to the dentary bone by means of recently deposited attachment bone (secondary mode of attachment). The frontier between the two bony matrices is indicated by a cement line (white arrow). The odontoblasts are located in the upper half region of the pulp cavity, where they are depositing predentin. This other part of the pulp cavity is occupied by capillary blood vessels. The mineralization of the dentin cone is interrupted in the region of the dividing zone, which delimits the dentin shaft proper from the so-called pedicel. The inner dental epithelium starts to retract from the tooth base. F,G: Juveniles, 6- and 10-month-old, respectively. These teeth show the same features as described for the previous one in E, except a higher number of odontoblasts. Note the large, unmineralized region along the pedicel and starting at the level of the dividing zone. H: In the teeth of late juveniles (here, 14-month-old specimen) the pulp cavity is large and occupied by numerous odontoblasts, most of them being polarized. The dividing zone is not clearly defined. I: Same tooth in a 15-month-old specimen. The number of odontoblasts has decreased but still active in the upper region of the pulp cavity where predentin matrix is deposited. Capillary blood vessels occupy most of the deep region of the pulp cavity. The dividing region is now visible. At the shaft surface, the inner dental epithelium has started to retract from the tooth base. J: Same tooth, in an 18-month-old specimen. Most odontoblasts have regressed except in the very upper region of the pulp cavity and in some places against the dentin wall. The dividing zone is more marked. Scale bars ⫽ 10 ␮m. ab, attachment bone; be, buccal epithelium; bv, blood vessel; cl, cement lines; db, dentary bone; de, dentine; ide, inner dental epithelium; od, odontoblasts; pc, pulp cavity.

TOOTH MORPHOLOGY IN P. WALTL

base is perpendicularly attached to the dentary bone surface, the upper half of all teeth is lingually curved, with an angle of ⬃20° to the anteroposterior axis of the dentary (Fig. 3). I1 (Fig. 3A) differs from its successors (Fig. 3B–F) in being less curved. Therefore, with the exception of I1, the orientation and the general shape of the teeth that develop successively at position I (I2 to I6) do not change through ontogeny except for tooth size (which is in relation to individual growth): between I2 and I3 we noted an increase of 15% in height and 20% in width, while the variation was reduced to 10% and 4% between I5 and I6, respectively (Table 1). However, two major changes are obvious: bicuspid suddenly succeed to monocuspid teeth (Fig. 4), and, in the mid-region of the tooth shaft, a dividing zone appears progressively, separating the dentin cone from the pedicel (Fig. 5). Cusps. Comparison of the tips of the six successive teeth at position I reveals a dramatic change occurring between I3 and I4 (Fig. 4). The conical tip of I1 to I3 (Fig. 4A–C) is replaced by a bicuspidate tip in I4, and this change is conserved in I5 and I6 (Fig. 4D–F). A major cusp is located at the lingual side and a minor cusp at the labial side (Fig. 3). The surface of the tip of I1 to I3 is smooth (enamel covering) and contrasts with the rough surface of the dentin cone. In the larval teeth the hypermineralized layer extends for ⬃15–20 ␮m from the tip, versus 60 – 80 ␮m in juvenile and adult teeth. The surface of the tip of I4 to I6 is also smooth, but the mesial and distal sides of the two cusps are underlined by a thin ridge, which is more distinguishable in adult than in juvenile teeth (Fig. 4D–F). Starting from I3, the tip is orange, a color due to the presence of ferritin in the enamel layer (not shown). The sudden change between I3 and I4, i.e., the last larval tooth with a single cusp and the first bicuspid juvenile tooth, appears to be correlated with metamorphosis. Pedicel. Adult teeth are characterized by the division of the dentin shaft into a dentin cone and a pedicel, the base of which is attached to the bone support by means of bone of attachment. The transition between dentin cone and pedicel is clearly marked externally by a narrow, rough area, the so-called dividing zone, located approximately at mid-height of the tooth shaft (Fig. 5). However, in embryonic tooth I1 the shaft surface is homogeneous, with no indication of the presence of such a dividing zone (Fig. 5A). Larval tooth I2 shows an enlargement of the mid-region of the tooth shaft, which could indicate the location of a dividing zone, but the latter is hardly distinguishable (Fig. 5B). In I3, the dividing zone is indicated by an irregular surface at the mid-height of the tooth shaft (Fig. 5C; see also Fig. 3C). In I4 to I6, the dividing zone is clearly visible as a narrow, 10 –20-␮m wide, granular region (Fig. 5D–F), marking the transition between the dentin cone and the pedicel (Fig. 3).

1057

Developmental and Structural Features Development of the successive teeth at position I. In contrast to the numerous morphological and structural differences described for successive teeth in the preceding and following sections, tooth development at position I (morphogenesis, differentiation, and attachment phases) showed similar features for all teeth, including embryonic tooth I1. Therefore, we chose to compare only development of the larval tooth (I2) with juvenile teeth (I5 to I6) (Fig. 6). The first phase of tooth development, morphogenesis, is divided into two steps: early morphogenesis, which morphologically starts by a thickening of the oral epithelium and ends when an epithelial bud has formed (Fig. 6A), and late morphogenesis, during which the bud develops into an epithelial bell, the dental organ, enveloping a mesenchymal cell population, the dental papilla (Fig. 6B). The first teeth (I1 and I2) start to develop at the deep surface of the buccal epithelium without a dental lamina being present. Early morphogenesis of replacement teeth is morphologically different from that of the first-generation tooth, in that replacement initiates as an epithelial invagination differentiating from the upper region of the dental epithelium of the predecessor tooth in the family. A bilayered epithelial strand extends from this region and penetrates deep into the mesenchyme. It runs along the lingual side of the preceding tooth down to the dentary bone surface (Fig. 6E). During the late morphogenesis step, the cells located at the extremity of the epithelial strand interact with a mesenchymal cell population located between the epithelial strand and the base of the preceding tooth and organize into an epithelial bell (Fig. 6F–H). The morphogenesis phase of replacement teeth is always initiated while the preceding tooth is still developing, generally in the cytodifferentiation phase (Fig. 6E–H). The second odontogenetic phase, cytodifferentiation, is again divided into two steps: an early step, during which the two opposite epithelial and mesenchymal cell populations differentiate into a dental organ and a dental papilla, respectively; and a late step, during which ameloblasts and odontoblasts being differentiated, respectively, from the inner dental epithelium and from the facing dental papilla cells, start to deposit tooth matrix, enamel, and dentin, respectively (Fig. 6C–E,G). In terms of cellular events the cytodifferentiation phase is morphologically similar for all teeth examined at position I during ontogeny, except for the first larval tooth, which develops closer to the deep surface of the buccal epithelium (Fig. 6C,D) than the subsequent teeth, all of which form at a distance from it (Fig. 6E–H). The odontoblasts and ameloblasts differentiate sequentially, during the late phase of cytodifferentiation, from the tip to the base of the tooth. This means that predentin can be in the process of Journal of Morphology DOI 10.1002/jmor

1058

´ AL ET AL. T. DAVIT-BE

deposition at the tooth base, while dental matrices are already well mineralized at the tooth tip (Fig. 6D,G). Tooth development is completed when the last phase, attachment, has been accomplished. At the end of this phase the tooth is ankylosed to the dentary bone. The tooth tip pierces the oral epithelium and becomes functional (see Fig. 7). It is noteworthy that in Pleurodeles waltl the enamel organ is composed only of two layers, the inner and outer dental epithelium, whatever the tooth generation. There is neither a stellate reticulum nor a stratum intermedium differentiating, not even in adult specimens. All tooth families are connected to a particular bilayered epithelial extension called the dental lamina, which extends, below the oral epithelium, rather vertically into the mesenchyme (Fig. 6I). All teeth forming in a family are also connected to this dental lamina by means of an epithelial strand running more or less horizontally (Fig. 6J). Structural modifications through successive replacement at position I. Most of the external modifications described at the SEM level in the previous section (Figs. 2–5) are concomitant with structural changes concerning the pulp cavity, the dividing zone, and the dental matrices. Except for the formation of the cusps, most of these changes take place at the end of the cytodifferentiation phase and during tooth attachment, and the structure continues to be modified during the functional life of the tooth. Therefore, for convenient comparisons, we mostly examined successive functional teeth, either recently attached or after several weeks of function (Fig. 7). The structural changes occurring at the tooth tip during the first steps of odontogenesis, i.e., the progressive replacement of enameloid by enamel through successive replacement, will be described in another article. The embryonic tooth I1 shows a primary type of attachment to the dentary bone: both matrices at the tooth base and at the dentary bone surface form simultaneously, and are not distinguishable from each other (Fig. 7A–C). All subsequent replacement teeth attach secondarily to already-formed bone matrix and the limit between attachment bone and dentary bone is clearly marked by so-called cementing lines (Fig. 7D,E). Changes concerning the pulp cavity. In welldeveloped I1 (Stages 36 –38, 13–17 dpf), the pulp cavity is entirely filled with a few (10 –12 cells), stacked, mostly crescent-shaped odontoblasts, each of them being in contact with the dentin cone surface (Figs. 7A, 8A). They have a large nucleus, with condensed patches of chromatin, and a cytoplasm rich in cisternae of the rough endoplasmic reticulum (RER), secretory vesicles filled with an electrondense, amorphous substance, and mitochondria (Fig. 8A). These cells are actively involved in predentin matrix deposition. Although large blood vessels are located around the tooth base, no capillary is seen inside the pulp cavity. In specimens at Stages Journal of Morphology DOI 10.1002/jmor

40 – 41 (22–25 dpf) the tooth is attached and the activity of most odontoblasts has decreased, as revealed by the small amount of cytoplasm. Some odontoblasts are still depositing predentin matrix along the tooth shaft (the so-called secondary dentin) (Fig. 7B). Intercellular spaces have appeared and small capillary blood vessels have penetrated the pulp cavity through the foramen located at the lingual side of the tooth base. These capillaries are connected to the large blood vessels surrounding the tooth base. Necrotic cells are also present in the basal region of the pulp cavity. During the following days (Stages 44 – 48), active odontoblasts are only observed in the upper region of the pulp cavity, while the basal region is mainly occupied by blood vessels (Fig. 7C). Intercellular spaces have enlarged and the cytoplasm is reduced, but RER cisternae and some mitochondria are still visible, indicative of a low synthetic activity (Fig. 8C). The pulp cavity of erupted I2 (Stage 42, 28 dpf) is entirely filled with active odontoblasts (20 cells at least), which are depositing predentin matrix against the tooth shaft (Fig. 7D). Later in I2 life, intercellular spaces appear in the central region of the pulp, the number of active cells decreases, and capillary blood vessels penetrate the basal region of the pulp cavity, as described for I1. In recently ankylosed I3, the pulp cavity contains numerous active odontoblasts (40 –50) (Fig. 7E). After some weeks of functional life, most of them are

Fig. 8. Structural modifications of teeth occurring through successive replacement at position I, in the morphology of the odontoblasts (A,B), in the dividing zone (C,D), and in the dentin layer (E–G). TEM. A: In well-developed but small larval teeth (here, tooth I3, Stage 51, 61 dpf), the odontoblasts located in the upper region of the pulp cavity are crescent-shaped and are involved in dentin matrix deposition on both sides. See also Figure 7A. B: In well-developed juvenile teeth (here tooth I5 at the end of the attachment process in a 9-month-old specimen), the odontoblasts are polarized, rectangular cells involved in the deposition of dentin matrix on a single side. See also Figure 7F,H. C: In the embryonic and first larval tooth (here tooth I1, Stage 39, 20 dpf), the tooth cone is not divided into a dentin shaft and a pedicel. The matrix of the zone of inflexion (see Fig. 7B–D) is not different from the neighboring regions, except for the presence of a denser background substance. D: In subsequent larval, juvenile and adult teeth (here tooth I3, Stage 55c, 105 dpf) the dentine cone shows a clear, unmineralized dividing zone (hatched line) separating the dentin shaft from the pedicel (see Fig. 7E–J). E: In developing teeth (here tooth I2, Stage 36, 13 dpf) numerous cytoplasmic prolongations of the odontoblasts are located within the whole thickness of the recently deposited predentin layer. Some of these extensions are close to the basement membrane separating the ameloblasts from the matrix. F: Upper region of the same tooth I2 when it is functional (here, Stage 48, 50 dpf). The cytoplasmic prolongations are still visible, and even close to the ameloblast surface (arrows). They are located within tubules embedded in the dentin matrix. G: In contrast to the upper region, the dentin matrix of the tooth shaft (here an adult tooth) does not house dentin tubules. Scale bars in A–D,F,G ⫽ 2 ␮m; E ⫽ 1 ␮m. am, ameloblasts; db, dentary bone; de, dentine; dz, dividing zone; od, odontoblasts; pc, pulp cavity; pd, predentin; pe, pedicel.

Figure 8

Journal of Morphology DOI 10.1002/jmor

´ AL ET AL. T. DAVIT-BE

1060

TABLE 1. Height and width at the base for the six successive functional teeth (i.e., recently attached to the dentary bone) at position I on the right lower jaw of Pleurodeles waltl, from 1-month-old larvae to 18-month-old young adults Tooth No Tooth height (␮m) Tooth base width (␮m) Age (month) Length (mm)

1

2

3

4

5

6

60–80 30–40 1 16–18

300–350 100 2 23

400 120 3 45

425–450 120 6 95

500–550 125 12 140

550–600 130 18 150

Measurements were taken in recently attached teeth to avoid errors due to wear.

located in the upper half of the pulp cavity. The basal region is occupied by large capillaries, which have entered the pulp cavity before the tooth was completely attached. A layer of active cells persists along the dentin cone surface, the pedicel and the surface of the attachment bone, where they continue to deposit matrix. Similar features were observed for I4 (Fig. 7F) and I5 (Fig. 7G). Three stages of the functional period of I6 were chosen (Fig. 7H–J) to be compared to similar stages of I1 (Fig. 7A–C). The pulp cavity of recently attached I6 contains a large number (100 –150) of odontoblasts, which are highly polarized and arranged into an uninterrupted layer along the tooth shaft (cf. Fig. 7H and Figs. 7A, 8A,B). Numerous RER cisternae, large secretory vesicles, and Golgi saccules, as well as numerous cell prolongations penetrating the predentin matrix, are indicative of high synthetic activity. Later in the life of I6 the odontoblast activity decreases. The odontoblasts become restricted to the upper half region of the pulp cavity, where secondary dentin continues to be deposited (Fig. 7I,J). Changes concerning the dividing zone. In embryonic tooth I1 and in larval tooth I2, the structure of the dentin shaft does not reveal features that could be interpreted as a separation between a dentin cone and a pedicel, neither at the LM (Fig. 7A–D) nor at the TEM (Fig. 8C) level. However, the dentin shaft shows an inflexion zone in its mid-region. This inflexion may foreshadow the location of the dividing zone in subsequent teeth (Fig. 7A–D). However, no change is detected in the organization of the dentin and predentin matrix when comparing this inflexion zone to adjacent regions (Fig. 8C). In contrast to I1 and I2, a dividing zone is visible in the mid-region of the dentin shaft of I3. At the LM level this zone appears less contrasted than the adjacent regions, which means that it is not mineralized (Fig. 7E). Observations at the ultrastructural level confirm that this zone is unmineralized (Fig. 8D). However, neither the aspect of the matrix nor the organization of the dividing zone differ when compared with that of the adjacent regions. The only difference consisted of the presence of an electrondense background substance between the collagen fibrils in the dividing zone, which is not visible in the mineralizing bordering regions (Fig. 8D). The unmineralized dividing zone separating the dentin cone from the pedicel is prolonged, on both sides of Journal of Morphology DOI 10.1002/jmor

the dentin cone, by large unmineralized collagen bundles resembling ligaments. The internal ligament extends over a larger distance toward the tooth base than the external one (Fig. 7E). The dividing zone, which forms in I4 to I6 after metamorphosis, is more distinct than in larval tooth I3 (Fig. 7F–J). During the entire life of a functional tooth, the dividing zone slowly mineralizes and it is sometimes hardly distinguishable in some old teeth, except at the pulp side, where the unmineralized ligament remains always visible (Fig. 7I,J). In all recently attached teeth the enamel organ extends over the whole surface of the tooth shaft and comes close to the bone support (Fig. 7A,E,H). This region of the enamel organ appears to be homologous to the so-called Hertwig’s sheath in mammals. During the functional period of each tooth, Hertwig’s sheath progressively retracts from the tooth base (Fig. 7B,F,G,I,J). In I1 and I2, retraction of the extremity of Hertwig’s sheath, called the cervical loop, stops at the level of the inflexion zone (Fig. 7C,D). From I3 onwards, the cervical loop retracts to the level of the dividing zone (Fig. 7H). It is noteworthy that the resorption of the tooth shaft by odontoclasts starts in the region that is no longer covered (protected?) by Hertwig’s sheath. Changes concerning the dentin matrix. At the LM level, the structure of the tooth shaft (dentin cone and pedicel) appears similar in all teeth examined (Fig. 7). Predentin matrix is deposited first from tip to base, and progressively mineralizes to become dentin. In all functional teeth the dentin shaft continues to thicken at the pulp side by deposition of secondary dentin, even in a well-advanced state of functionality. The odontoblasts depositing the matrix of predentin, dividing zone, pedicel, and even attachment bone show similar features and apparently belong to the same population of dental papilla cells (see, e.g., Fig. 7H). Ultrastructural observations reveal that a large number of cytoplasmic prolongations issued from the odontoblasts are located within the recently deposited predentin matrix in all forming teeth (Fig. 8E). However, in embryonic tooth I1 dentin tubules are not observed when the dentin layer is well mineralized, suggesting that the odontoblast extensions have retracted during the mineralization process. In contrast, dentin tubules are present in the dentin of larval (I2, I3), juvenile (I4, I5), and adult (I6) teeth (Fig. 8F), but they are only seen in the tip of the tooth, where the dentin

TOOTH MORPHOLOGY IN P. WALTL

layer is the thickest, and not in the dentin of the tooth shaft, dentin cone, and pedicel (Fig. 8G). For all teeth examined in the series, the first matrix deposited comes from the odontoblasts. This means that the ameloblasts deposit their contribution only when a layer of loose, predentin-like matrix has been secreted. In embryonic and larval teeth (i.e., I1, I2, and I3), this thick layer of loose, predentin-like matrix, is called enameloid. When recently deposited the enameloid matrix resembles predentin (it is collagen rich), but during the mineralization and maturation processes it mineralizes more fully than the dentin, probably under the influence of the ameloblasts. Indeed, when decalcified this layer only appears as an artifactual empty space located between the ameloblasts and the dentin surface (Fig. 7). In I1 the enameloid region is covered with a thin enamel layer. Through successive replacement cycles, the thickness of the enameloid layer reduces, while that of the enamel layer increases. This is obvious in postmetamorphic specimens in which, from I4 onwards, a layer of enameloid is hardly distinguishable between dentin and enamel. The enameloid– enamel transition from embryonic to adult teeth will be described in detail elsewhere. DISCUSSION Our study provides a detailed description of external morphology, structure, development, and replacement of six successive teeth belonging to a single family in a polyphyodont species. Several objectives were fulfilled: 1) we completed and detailed a number of data available in the literature for Pleurodeles waltl, and for Caudata in general; indeed, previous data were obtained from various species and from unidentified tooth positions and/or tooth generation (Chibon, 1966, 1967, 1970; Smith and Miles, 1971; Wistuba et al., 2002); and 2) we answered questions of how and when changes in tooth morphology (number of cusps, formation of the dividing zone, tooth matrix organization, modifications within the pulp cavity) took place during P. waltl ontogeny. We have now established solid bases of knowledge appropriate to answer some questions on tooth biology using P. waltl as a model species. In particular, we can now undertake investigations at the molecular level aiming to understand the genetic control of, e.g., tooth replacement, resorption, or the monoscuspid to bicuspid transition. These issues are discussed below. Life Cycle of Teeth and Replacement Process Monitoring the life cycle of successive teeth from the initiation stage to complete resorption was a long-lasting task that involved the study of several growth series composed of numerous specimens from various nests. In this respect Pleurodeles waltl

1061

proved to be a suitable model in that several hundreds of larvae were obtained during the study period. Such a longitudinal study was necessary for an accurate identification of each tooth in the series, i.e., its family and its rank in the family. Serial sectioning of growth series was of great help in this identification since it revealed the relation between the developing teeth and their tissue environments. The chronology of tooth succession in Pleurodeles waltl indicates that the life cycle of a tooth increased from 55 days for tooth I1 to 14 months (i.e., ⬃7 times longer) for tooth I6 in 2-year-old adults. Each odontogenic step, i.e., the developmental and the functional periods, increased accordingly. When combining these data with the knowledge of the duration of each larval stage (or of the age of the specimens), we now possess a powerful tool to link the developmental phases for each tooth in a family to the ontogenetic life stage of the animal. The relatively short life span of the first teeth in the family studied (from 2– 4 months for I1 to I3) compared to that in juveniles and adults could explain why there were more positions lacking teeth and more resorption features in larvae than in adults when observed with SEM: numerous teeth were being replaced and the newly developing teeth, not yet attached, were not conserved during sample preparation. In the past, authors have investigated the variations in tooth replacement in salamanders in relation to seasonal cycles using various techniques: Alizarin red staining, radiographs, and/or wax impression (Plethodon cinereus: Lawson et al., 1971; Necturus maculosus: Miller and Rowe, 1973). Chibon (1977) was the first to calculate the duration of the tooth cycle in Pleurodeles waltl, using tritiated proline: teeth became functional after 5 days of growth in early larvae, 8 days in older larvae, and 16 days in postmetamorphosed specimens. In the present study, we confirm the rapid growth of the first-generation tooth (6 days to become functional) but Chibon’s data represent an overestimation of tooth growth compared to our data on old larvae (15 days for tooth I2 and 43 days for tooth I3) and juveniles (2.5 and 4 months for tooth I4 and tooth I5, respectively). Also, for larval teeth the functional period of 20 days estimated by Chibon (1977) represents an underestimation: from 40 days for tooth I1 to 70 days for tooth I3 in our study. The low values obtained by Chibon are probably linked to 1) variations in growth speed (calculated indirectly from tritiated proline incorporation) during tooth morphogenesis and cytodifferentiation, and 2) data obtained from teeth developing at different loci on the arcade. This reinforces the importance of monitoring tooth succession at a given position throughout ontogeny. The growth of the replacement tooth provoking resorption of the previous, functional tooth was reported many times in the literature for reptiles (e.g., Edmund, 1960; Westergaard and Ferguson, 1986; Journal of Morphology DOI 10.1002/jmor

1062

´ AL ET AL. T. DAVIT-BE

Delgado et al., 2003) and teleost fish (e.g., Huysseune et al., 1998; Van der heyden et al., 2000). Probably, pressure of the tooth germ on the basal region of the preceding tooth in the series initiates odontoclast recruitment. Therefore, the finding in Pleurodeles waltl that the first-generation tooth was not replaced by its successor was rather unexpected. Indeed, resorption of tooth I1 starts long after tooth I2 has grown and attached to the dentary bone surface. Such a phenomenon was never commented on in the literature, probably because the authors considered tooth I2 as belonging to another tooth family growing between positions I and II (Roux and Chibon, 1974; Chibon, 1977). A similar case of the presence of two functional teeth from a same family was reported in the zebrafish pharyngeal dentition (Van der heyden et al., 2000). Our observations suggest that the growth of the third tooth in the family (I3) provokes the resorption of I2, as expected. The only explanation why growth of tooth germ I2 does not lead to I1 resorption is that the small size of embryonic tooth I1 leaves enough space on the dentary bone surface to accommodate I2, without provoking I1 resorption. A similar explanation was proposed for the zebrafish larval teeth (Van der heyden et al., 2000). This suggests that such a phenomenon could be more common than expected in polyphyodont species having toothed larvae. It does not occur in species having a long embryonic period, such as reptiles (Sire et al., 2002). The question of which factor could induce I1 resorption in P. waltl larvae should be raised. We can only speculate that I1 resorption could be linked either to the growth of tooth I3— although it develops rather far from tooth I1— or, rather, to the remodeling of the dentary bone surface during the larval period. Further observations are needed to understand this mechanism. The chronological table of tooth development in position I should also permit clarification of different steps of tooth resorption. Typically, multinucleated odontoclasts, similar to osteoclasts described in many vertebrate species, have already been reported in a frog and in the salamander Ambystoma mexicanum (see also Yaeger and Kraucunas, 1969; Wistuba et al., 2000). In the present study we found (but not shown) rather frequently odontoclasts resorbing teeth, as described by previous authors. However, a detailed study of the last steps of tooth resorption would be interesting because the question of the fate of the tooth tip is still debated. It appeared from some of our observations that the enameloid and/or enamel cap is entirely resorbed, both in larvae and at least in young adults. It is generally assumed that, in adult lissamphibians, the tooth tip is shed into the mouth during the resorption process, when the unmineralized dividing zone is destroyed (Lawson et al., 1971; Clemen and Greven, 1980; Casey and Lawson, 1981). However, several authors have reported that, at least in larvae, the tooth could be entirely resorbed (Chibon, Journal of Morphology DOI 10.1002/jmor

1977, working on Pleurodeles waltl; Shaw, 1985, 1986, 1989, working on Xenopus laevis), making minerals and organic matrix available for reuse. Another outcome of our descriptions relates to the possibility of studying the initiation of the replacement process. We know from the present study, and from studies in teleost fish, that tooth replacement is initiated from a quiescent cell population (stem cells?, but see Huysseune and Thesleff, 2004) located in the upper region of the outer dental epithelium at the lingual side of the previous tooth. Our timetable of tooth succession at position I provides, for each tooth, the developmental stage (in days postfertilization) at which tooth replacement is initiated. Except for tooth I2, our study indicates that in Pleurodeles waltl the initiation process starts precociously when the upper region of the previous tooth is completed, i.e., long before the tooth becomes attached and functional. This would mean that a correlation exists between a well-defined developmental stage of the preceding tooth and the initiation of its successor. This finding supports the hypothesis proposed by Gillette (1955) in Rana pipiens and by Lawson et al. (1971) in Plethodon cinereus that initiation of tooth replacement is not a random process, but is probably under the influence of a local mechanism (for instance, completion of the tooth tip of the preceding tooth) rather than being triggered by a stimulus traveling along the jaw, as proposed by Edmund (1960, 1962) in his Zahnreihe theory of tooth replacement in reptiles, in which new tooth germs are initiated by a single stimulus at any moment in time, resulting in replacement waves. The Edmund’s model is similar to the morphogenetic field model proposed by Osborn some years later (1970, 1971, 1973), i.e., the initiation of tooth replacement (primordia) is externally controlled by field molecules acting through a gradient of concentration (see review in Davit-Be´al et al., 2006). In P. waltl, the factor responsible for the initiation of tooth replacement seems to be more related to a particular developmental period of the preceding tooth than to a general stimulus controlling waves of tooth replacement. However, the spatial and temporal location of the activation of such a factor remain to be found. Development Our comparison of tooth development in larval and adult Pleurodeles waltl largely confirms previous descriptions of the odontogenesis in various lissamphibian species, although generally obtained from unidentified tooth position and/or tooth in the series (Smith and Miles, 1971; Roux, 1973; Roux and Chibon, 1973; Kogaya, 1999). Wistuba et al. (2002) studied the ultrastructure of various odontogenetic stages through ontogeny in Ambystoma mexicanum, e.g., early larvae, typical larvae, neotenic, and “true” juveniles and adults, in which metamorphosis was

TOOTH MORPHOLOGY IN P. WALTL

induced by thyroxine injection to obtain postmetamorphic specimens. In P. waltl the developmental steps and the cellular processes involved during tooth development were similar for each tooth examined in the series, when we exclude variations related to animal size. The only obvious difference concerns the initiation phase of tooth I1 compared to its successors. This is a primary type of initiation, morphologically identified as a thickening of the oral epithelium (placode stage), versus the secondary type of initiation observed for all subsequent teeth, which is characterized by a budding of the upper region of the dental organ of the preceding tooth in the series (see also Sire et al., 2002). Therefore, tooth development in Pleurodeles waltl can be used as a model for studying odontogenesis, and in particular to study the genetic control of tooth morphogenesis and cytodifferentiation, as well as the mechanisms controlling tooth replacement. Changes of External and Structural Features In Pleurodeles waltl tooth shape, size, and orientation change through ontogeny. Whereas the teeth in recently hatched larvae are small, widely spaced, and possess a generalized (i.e., conical) shape, a smooth surface, and a straight orientation, they acquire a specific shape in premature juveniles and adults, i.e., large, bicuspidate, posteriorly curved, and close-set. Such external modifications would not have been possible in the absence of a successive tooth replacement through ontogeny. Most of these changes were known in the lissamphibian literature but the present study adds a chronological table to these modifications through P. waltl ontogeny. We have found that most modifications take place progressively during the larval life, i.e., a little modification occurs from one tooth to its successor, except for the sudden transition between monocuspid and bicuspid teeth at metamorphosis. In Pleurodeles waltl the first-generation teeth have a generalized morphology and structure that is different from that found in adult specimens. This minimalistic tooth structure in a salamander larva is similar to that described in larval actinopterygian fishes (Huysseune and Sire, 1997a,b; Huysseune et al., 1998). It is related to the small size of the teeth (less than 100 ␮m), in contrast to the first-generation teeth in lizards and sharks, which look like miniatures of adult teeth (Sire et al., 2002). In juvenile and adult specimens, replacement teeth first complete their upper region, while the dentine cone and the tooth base are not deposited yet. Indeed, early in tooth formation the tooth cap (enamel covering) is identical (shape, size, and ornamentation) to that of the functional tooth, as observed in a lizard (Delgado et al., 2003). This

1063

indicates that tooth morphology does not change much from one tooth to its successor, and that the tooth cap is completed long before the tooth is attached. Therefore, the factors suspected to be responsible for morphological changes of the upper tooth region (i.e., enameloid– enamel transition and cusp formation) must be studied in early developmental stages of successive teeth. Our chronological table helps in the identification of appropriate stages to be studied with the aim of finding morphological characteristics (at the cellular level) indicating that the tooth cap is completed, and of checking whether or not the completion of the tooth tip is related to the initiation of the replacement tooth. A noteworthy event concerns the sudden transition from a single cusp (in teeth I1, I2, and I3) to two cusps (from tooth I4 onwards). This change is correlated with metamorphosis and this spectacular transition has attracted the attention of several authors (e.g., Kerr, 1960; Chibon, 1972; Clemen and Greven, 1974; Beneski and Larsen, 1989a,b; Greven and Clemen, 1990). We know that the formation of this region of the tooth is controlled by the ameloblasts. Several experiments have demonstrated that the sudden and important increase of the level of the thyroid hormone, thyroxin (T4 and its deionated form T3, which is more active), during this crucial period is responsible for the formation of bicuspid teeth (Chibon, 1972; Gabrion and Chibon, 1972; Greven and Clemen, 1990). When metamorphosis is blocked (hypophysectomy or thyroid dysfunction), the teeth which normally form two cusps remain monocuspid (Greven and Clemen, 1990). Conversely, when metamorphosis is induced in neotenic axolotls (thyroxin injection), normally monocuspid teeth change into bicuspid (Clemen, 1988). From our timetable and from the developmental stages identified, it is clear that the appropriate stage to study this transition is to be found at the beginning of metamorphosis, when tooth I4 has undergone its morphogenesis phase. During this phase the enamel organ differentiates.

CONCLUSION The present study highlights the advantages of the salamander Pleurodeles waltl as a model for study of several aspects of odontogenesis in depth. In particular, assuming that genetic tools will be easy to obtain in P. waltl, we will investigate the various genetic pathways that control initiation and development of first-generation teeth, continuous tooth replacement, resorption and odontoclast recruitment and behavior, mono- to bicuspid transition, and eventually, formation of the dividing zone and the pedicel. Journal of Morphology DOI 10.1002/jmor

1064

´ AL ET AL. T. DAVIT-BE

ACKNOWLEDGMENTS We thank Prof. Ann Huysseune (Ghent University, Belgium) and Prof. David Wake (University of California, Berkeley, CA) for critical reading and comments on the article. SEM and TEM work was carried out at the Service de Microscopie e´lectronique-Universite´ Paris 6 and CNRS.

LITERATURE CITED Beneski JTJ, Larsen JH. 1989a. Interspecific, ontogenetic and life history variation in the tooth morphology of mole salamanders (Amphibia, Urodela, Ambystomatidae). J Morphol 199:53– 69. Beneski JTJ, Larsen JH. 1989b. Ontogenetic alterations in the gross tooth morphology of Dicamptodon and Rhyacotriton (Amphibian, Urodela, Dicamptodontidae). J Morphol 199:165–174. Bolte M, Clemen G. 1992. The enamel of larval and adult teeth of Ambystoma mexicanum Shaw (Urodela: Ambystomatidae) — a SEM study. Zool Anz 228:167–173. Borday-Birraux V, Van der heyden C, Debiais-Thibaud M, Verreijdt L, Stock DW, Huysseune A, Sire J-Y. 2006. Expression of Dlx genes during the development of zebrafish pharyngeal dentition: evolutionary implications. Evol Dev 8:130 –141. Casey J, Lawson R. 1981. A histological and scanning electron microscope study of the teeth of caecilian amphibians. Arch Oral Biol 26:49 –58. Chibon P. 1966. Analyse expe´rimentale de la re´gionalisation et des capacite´s morphoge´ne´tiques de la creˆte neurale chez l’amphibien urode`le Pleurodeles waltlii Michah. Me´m Soc Zool 36:1–107. Chibon P. 1967. Etude expe´rimentale par ablations, greffes et autoradiographie, de l’origine des dents chez l’amphibien urode`le Pleurodeles waltlii Michah. Arch Oral Biol 12:745–753. Chibon P. 1970. L’origine de l’organe admantin des dents. Etude au moyen du marquage nucle´aire de l’ectoderme stomode´al. Ann Embryol Morphol 3:203–213. Chibon P. 1972. Etude ultrastructurale et autoradiographique des dents chez les amphibiens. Relations entre la morphoge´ne`se dentaire et l’activite´ thyroı¨dienne. Bull Soc Zool Fr 97:437– 448. Chibon P. 1977. Vitesse de croissance et renouvellement des dents chez les amphibiens. J Embryol Exp Morphol 42:43– 63. Clemen G. 1988. Competence and reactions of early- and latelarval dental laminae in original and not-original dental systems of Ambystoma mexicanum Shaw. Arch Biol (Bruxelles) 99:307–324. Clemen G, Greven H. 1974. Morphologische Untersuchungen an der Mundho¨hle von Urodelen. I. Die Gaumenza¨hne von Salamandra salamandra L. (Salamandridae : Amphibia). Forma Funct 7:249 – 280. Clemen G, Greven H. 1977. Morphologische Untersuchungen an der Mundho¨hle von Urodelen. III. Die Munddachbezahnung von Ambystoma mexicanum Cope (Ambystomatidae : Amphibia). Zool Jb Anat 98:95–136. Clemen G, Greven H. 1980. Morphologische Untersuchungen an der Mundho¨hle von Urodelen. VII. Die Munddachbezahnung von Amphiuma (Amphiumidae: Amphibia). Bonn Zool Beitr 31:357–362. Corsin J. 1966. The development of the osteocranium of Pleurodeles waltlii Michahelles. J Morphol 119:209 –216. Davit-Be´al T, Chisaka H, Delgado S, Sire J-Y. 2006. The amphibian teeth. Current knowledge, unanswered questions, and some directions for future research. Biol Rev (in press). Delgado S, Davit-Be´al T, Sire J-Y. 2003. The dentition and tooth replacement pattern in Chalcides (Squamata; Scincidae). J Morphol 25:146 –159.

Journal of Morphology DOI 10.1002/jmor

Edmund AG. 1960. Tooth replacement phenomena in lower vertebrates. Contrib R Ont Mus, Life Sci Div 52:1–190. Edmund AG. 1962. Sequence and rate of tooth replacement in the Crocodilia. Contrib R Ont Mus, Life Sci Div 56:1– 42. Gabrion J, Chibon P. 1972. Morphologie dentaire et disposition des dents chez les tritons ne´ote´niques (Triturus helveticus Raz.). C R Se´anc Soc Biol 167:50 –54. Gallien L. 1952. Elevage et comportement du pleurode`le au laboratoire. Se´ances Soc Biol 456 – 461. Gallien L, Durocher M. 1957. Table chronologique du de´veloppement chez Pleurodeles waltlii Michah. Bull Biol Fr Belg 91:97– 114. Gillette R. 1955. The dynamics of continuous succession of teeth in the frog (Rana pipiens). Am J Anat 96:1–36. Greven H. 1989. Teeth in extant amphibia: morphology and some implications. Fortschr Zool 35:451– 455. Greven H, Clemen G. 1990. Effect of hypophysectomy on the structure of normal and ectopically transplanted teeth in larval and adult urodeles. Acta Embryol Morphol Exp 11(n. sp):33– 43. Howes R. 1978. Regeneration of ankylosed teeth in the adult frog premaxilla. Acta Anat 101:179 –186. Huysseune A, Sire J-Y. 1997a. Structure and development of first-generation teeth in the cichlid Hemichromis bimaculatus (Teleostei, Cichlidae). Tissue Cell 29:679 – 697. Huysseune A, Sire J-Y. 1997b. Structure and development of teeth in three armoured catfish Corydoras arcuatus, C. aeneus and Hoplosternum littorale (Siluriformes, Callichthyidae). Acta Zool (Stockh) 78:69 – 84. Huysseune A, Sire J-Y. 1998. Evolution of patterns and processes in teeth and tooth-related tissues in non-mammalian vertebrates. Eur J Oral Sci 106:437– 481. Huysseune A, Thesleff I. 2004. Continuous tooth replacement: the possible involvement of epithelial stem cells. BioEssays 26:665– 671. Huysseune A, Van der heyden C, Sire J-Y. 1998. Early development of the zebrafish (Danio rerio) pharyngeal dentition (Teleostei, Cyprinidae). Anat Embryol 198:289 –305. Jackman WR, Draper BD, Stock DW. 2004. Fgf signaling is required for zebrafish tooth development. Dev Biol 274:139 – 157. Kerr T. 1960. Development and structure of some actinopterygian and urodele teeth. Proc Zool Soc Lond 133:401– 422. Kogaya Y. 1999. Immunohistochemical localisation of amelogenin-like proteins and type I collagen and histochemical demonstration of sulphated glycoconjugates in developing enameloid and enamel matrices of the larval urodele (Triturus pyrrhogaster) teeth. J Anat 195:455– 464. Laurenti P, Thaeron C, Allizard F, Huysseune A, Sire J-Y. 2004. The cellular expression of eve1 suggests its requirement for initial patterning of the dentition and differentiation of the ameloblasts in the zebrafish (Danio rerio). Dev Dyn 230:727– 733. Lawson R, Wake D, Beck N. 1971. Tooth replacement in the red-backed salamander, Plethodon cinereus. J Morphol 134: 259 –270. Miller W, Rowe D. 1973. Preliminary investigation of variations in tooth replacement in adult Necturus maculosus. J Morphol 140:63–76. Osborn JW. 1970. New approach to Zahnreihen. Nature 225:343– 346. Osborn JW. 1971. The ontogeny of tooth succession in Lacerta vivipara Jacquin (1787). Proc R Soc Lond B 179:261–289. Osborn JW. 1973. The evolution of dentitions. Am Sci 61:548 – 559. Roux J. 1973. Etude ultrastructurale de la dentinoge´ne`se chez la larve du triton Pleurodeles waltlii (Amphibien Urode`le). J Biol Buccale 1:21–32. Roux J, Chibon P. 1973. Etude ultrastructurale de l’ame´loge´ne`se chez la larve du triton Pleurodeles waltlii (Amphibien Urode`le). J Biol Buccale 1:33– 44.

TOOTH MORPHOLOGY IN P. WALTL Roux J, Chibon P. 1974. Pre´sence de cellules ne´crotiques et de phagocytes dans la pulpe de dents fonctionnelles d’Amphibiens Urode`les. Re´sultats pre´liminaires. C R Se´anc Soc Biol 278: 1369 –1372. Shaw JP. 1985. Tooth replacement in adult Xenopus laevis (Amphibia: Anura). J Zool Lond 207:171–179. Shaw JP. 1986. A longitudinal study of tooth resorption in newly metamorphosed Xenopus laevis with comments on tooth resorption in amphibians. J Zool 208:215–228. Shaw JP. 1989. A morphometric study of bone and tooth volumes in the pipid frog Xenopus laevis (Daudin), with comments on the importance of tooth resorption during normal tooth replacement. J Exp Zool 249:99 –104. Signoret J. 1959. Anatomie de la re´gion ce´phalique chez Pleurodeles waltlii Michah. Bull Soc Zool Fr 84:33–51. Sire JY, Huysseune A. 2003. Formation of skeletal and dental tissues in fish: a comparative and evolutionary approach. Biol Rev 78:219 –249. Sire JY, Davit-Be´al T, Delgado S, Van der heyden C, Huysseune A. 2002. The first generation teeth in non-mammalian lineages: ev-

1065

idence for a conserved ancestral character? Microsc Res Tech 59:408 – 434. Smith M, Miles A. 1971. The ultrastructure of odontogenesis in larval and adult Urodeles. Differentiation of the dental epithelial cells. Z Zellforsch 121:470 – 498. Van der heyden C, Huysseune A. 2000. Dynamics of tooth formation and replacement in the zebrafish (Danio rerio) (Teleostei, Cyprinidae). Dev Dyn 219:486 – 496. Van der heyden C, Huysseune A, Sire J-Y. 2000. Development and fine structure of pharyngeal teeth in juvenile zebrafish (Danio rerio) (Teleostei, Cyprinidae). Cell Tissue Res 302:205–219. Westergaard B, Ferguson MWJ. 1986. Development of the dentition in Alligator mississipiensis. Early embryonic development in the lower jaw. J Zool Lond 210:575–597. Wistuba J, Bolte M, Clemen G. 2000. The odontoclasts of Ambystoma mexicanum. Ann Anat 182:415– 422. Wistuba J, Greven H, Clemen G. 2002. Development of larval and transformed teeth in Ambystoma mexicanum (Urodela, Amphibia): an ultrastructural study. Tissue Cell 34:14 –27. Yaeger J, Kraucunas E. 1969. Fine sructure of the resorptive cells in the teeth of frogs. Anat Rec 164:1–14.

Journal of Morphology DOI 10.1002/jmor