Tooth development begins with the invagination of the oral epithelium, deriving from the ectoderm, into the underlying ectomesenchyme, deriving from neural crest cells. This epithelial invagination initiates a complex cascade of genetic and subsequently cellular reactions which are the basis of dentin and enamel biomineralization (Berkowitz et al.,2002; Arnold,2006). In general, the biomineralization process always starts with the production of an extracellular organic matrix, consisting mainly of proteins, which function as scaffolding for the subsequent mineralization of hydroxyapatite crystallites (Höhling et al.,1990; Plate et al.,1992; Wiesmann et al.,2005). As biomineralization processes are one of the oldest cellular achievements in evolution, it can be assumed that the basic mechanisms have been conserved and are similar throughout all vertebrates (Donoghue et al.,2000). However, the cell biology of the formation of teeth is still not fully understood, and the assembly process for the construction of a whole tooth is mainly unknown. Although attempts have been made in bioengineering whole teeth (Young et al.,2002,2005), many open questions still remain, and suitable animal models for studying biomineralization are still needed.
Zebrafish (Danio rerio) has become one of the most investigated laboratory animals during the past decade. The short generation time, the transparency of its embryos, and the ease of mutagenesis have made zebrafish ideal for experimental studies (Gerhard and Cheng,2002; Nüsslein-Vollhard and Dahm,2002; Yelick and Schilling,2002; Gerhard,2003). Zebrafish form essentially the same skeletal and muscle tissue types as their higher vertebrate counterparts. Bones of zebrafish develop similar to that of human bones, either from direct ossification as dermal bones (cranial bones), deriving from connective tissue cells, or as chondroidal ossification deriving from cartilaginous precursors (axial skeleton; Verreijdt et al.,2002; Bird and Mabee,2003). The zebrafish belongs to the cyprinid family, which lacks oral dentition (i.e., teeth in the oral cavity) but has pharyngeal teeth associated with the posterior branchial arches (pharyngeal jaws). Therefore, the tooth epithelium of pharyngeal teeth may be of endodermal origin. These pharyngeal teeth are replaced continuously following a distinct pattern of replacement (Van der heyden et al.,2001). On each side of both pharyngeal jaws, they are arranged in three rows of which the ventral row consists of five teeth, the mid-dorsal row consists of four teeth, and the dorsal row contains two teeth (Van der heyden et al.,2001; Wautier et al.,2001). The basic morphology of these pharyngeal teeth is similar to the morphology of most vertebrate teeth (Huysseune et al.,1998; Neues et al.,2006). The crown is covered by a thin layer of enameloid, whereas the tooth consists of tubular dentin, which surrounds the pulp chamber with odontoblasts and pulp cells (Neues et al.,2006). Enameloid is different from enamel of vertebrate teeth due to its different content of organic components. The root is attached to the underlying branchial bone by a specialized bone of attachment. These pharyngeal teeth are continuously replaced within 11–13 days during the whole lifespan, which makes the zebrafish a model organism to study tooth development and biomineralization of enamel and dentin (Van der heyden et al.,2001; Huysseune and Sire,2004; Huysseune and Thesleff,2004; Bartlett et al.,2005). Tooth eruption and ankylosis of the pharyngeal teeth is completed before completion of dentin mineralization (Van der heyden et al.,2000). It was the aim of the study to compare human odontogenesis with that of zebrafish to assess microanalytic similarities and differences of tooth development for the optimal interpretation of experimental biomineralization outcomes of zebrafish pharyngeal dentition.
MATERIAL AND METHODS
The mandible of a 16-week-old human fetus, containing three tooth buds on each side, was cut into two halves and embedded in Technovit 9100 (Kulzer, Germany). The fetus was obtained from Marienhospital in Witten, Germany, in accordance with BestG NRW Section 14 Abs. 2 dated 17.06.2003 and with the consent of the parents. In addition, five complete heads of zebrafish (wild-type strain AB-2 stock # 1091 from Zebrafish International Resource Center, Eugene, OR) were used for histological examination. After the approval of the local authorities according to the German Animal Protection Law § 6 Seq. 1, the fish were anesthetized and killed by decapitation. Two heads were decalcified in 5% HNO3 and embedded in paraffin. Histological sections of 5-μm thickness were stained with Azo–Karmin (Azan). Three heads were embedded in Technovit 9100. Of the Technovit-embedded specimens, serial sections of 80-μm thickness were cut with a saw microtome (Leica 1600) and the sections were investigated with polarized light microscopy (PLM). All sections containing tooth buds (two sections of each zebrafish head and three sections of each human tooth bud) were used for scanning electron microscopy. The sections were coated with carbon and investigated with a Philips XL 30 FEG scanning electron microscope (Philips, The Netherlands) at 20 kV using the backscattered electron detector. Quantitative element analysis of Ca, P, and C was carried out with energy dispersive X-ray analysis (EDX) with an S-UTW detector (EDAX, USA) with a count rate of between 1,800 and 2,000 counts/sec and a dead time of 30%. Measuring time was 180 sec (live seconds) with a resolution of 135.8 eV and an amplification time of 100 μsec. On each section, five spot measurements (spot size 2 nm) were made in predentin, mineralizing dentin adjacent to predentin, mineralized dentin, mineralized enamel, and mineralizing enamel surface. Line scans through the tooth buds were made at 512 points with a dwell time of 1,000 msec and amplification time of 10 μsec. To ensure that the pharyngeal teeth of zebrafish were in a similar developmental stage to those of the human tooth buds, only teeth that were still covered by enamel epithelium and, therefore, were not yet erupted were selected for EDX analysis.
Histology of the zebrafish developing teeth showed enamel epithelium covering the outer side of the tooth anlage, a layer of dentin, odontoblasts underneath the dentin, and mesenchyme cells within the pulp chamber (Fig. 1). PLM of the human mandible revealed three tooth buds (two incisors and one canine) on each side of the mandible in the bell-stage of their development. Mineralizing dentin, dentin, and enamel could be well distinguished with PLM (Fig. 2a). Mineralizing dentin was clearly demarcated from the pulp and dentin by changing birefringency. Predentin and the pulp tissue could not be differentiated with PLM. Enamel was characterized by negative birefringency.
PLM of the sectioned zebrafish teeth showed a distinct enameloid layer on the crown of the teeth and a tubular dentin body surrounding the pulp chamber (Fig. 2b). The teeth are attached to the underlying branchial bone by acrodontal fixation.
EDX element analysis showed different Ca, P, and C content in predentin, mineralizing dentin and mineralized dentin in both the human tooth anlage and the zebrafish developing tooth. The Ca and P content increased from predentin toward mineralized dentin (Fig. 3), whereas the C content decreased accordingly (Table 1). Overall, the Ca and P content of the small layer of predentin of the zebrafish teeth was higher than in the human teeth (Table 2). Also the Ca/P ratio in predentin was significantly different from that of mineralized dentin in the human as well as in the zebrafish teeth. Line scans of developing dentin of human tooth buds and zebrafish developing teeth showed a steady increase of the Ca and P content within predentin toward mineralizing dentin (Fig. 3). These results could be verified with EDX element mapping which showed in human and zebrafish developing teeth an increasing signal for calcium (Fig. 4a,c) and phosphorus and decreasing signal for carbon (Fig. 4b,d) from predentin toward mineralized dentin.
Table 1. Element content in wt% (± SD) in predentin, mineralizing dentin, and mature dentin of human teeth
76.9 ± 11.7
7.8 ± 10.05
2.9 ± 2.8
2.5 ± 0.85
35.6 ± 20.1
30.7 ± 12.4
14.1 ± 5.1
2.1 ± 0.33
21.5 ± 6.4
36.6 ± 7.2
17.3 ± 2.0
2.1 ± 0.21
Table 2. Element content in wt% (±SD) in predentin, mineralizing dentin and mature dentin of zebrafish teeth
71.5 ± 20.8
18.1 ± 15.0
8.6 ± 5.7
1.9 ± 0.4
51.9 ± 19.9
30.8 ± 13.9
17.2 ± 5.7
1.9 ± 0.2
23.3 ± 6.7
50.7 ± 5.4
23.6 ± 1.8
2.1 ± 0.2
Biomineralization of dentin begins with the secretion of predentin, which is not yet mineralized and contains various matrix proteins (Beniash et al.,2000; Arnold,2006). These matrix proteins function as scaffolding for the subsequent mineralization of predentin. The composition of the first mineral that is formed at the mineralization front in predentin may vary from different calcium phosphates (Boskey,2003) to amorphous calcium phosphate precursors (Olszta et al.,2003a,b). The minerals first formed at the mineralization front in predentin may also vary during dentin mineralization of different species.
These differences in the Ca/P ratio are well represented, indicating differences in the mineral composition of predentin and mineralizing dentin in human tooth buds and zebrafish developing teeth. However, the end product of the biomineralization process in both cases is hydroxyapatite as indicated by the identical Ca/P ratio for mature dentin.
The rapid increase of the Ca and P content at the mineralization front in human and zebrafish teeth indicates a high affinity of calcium and phosphorus to matrix proteins. These findings support other studies that demonstrated calcium uptake within a narrow area at the mineralization front (Rabie and Veis,1995; Goldberg and Septier,1996; Beniash et al.,2000). After this rapid uptake of calcium and phosphorus, dentin is almost completely mineralized and there is only a small zone of still mineralizing dentin left. These findings support the theory that the proteins of the predentin matrix themselves may serve as nucleators for hydroxyapatite crystallization (Boskey et al.,1989; Boskey,1996,2003).
The main difference in the pattern of mineralization between the human and zebrafish teeth is the early and continuous increase in the calcium and phosphorus content in zebrafish teeth in contrast to human predentin, exhibiting a rather late and steep increase of the mineral content. It is, therefore, hypothesized that the rapid biomineralization of zebrafish teeth parallel to their short life span of development, function and exfoliation within 11–13 days is characterized by continuous active attraction of calcium and phosphate ions to the collagen matrix. In contrast to the lifelong human predentin mineralization for years in deciduous teeth and for many decades in permanent teeth, the same process with identical mineral formulation takes place in zebrafish pharyngeal teeth but within a few days. Therefore, from an experimental point of view, the zebrafish dentition can simulate in very short intervals long-lasting influences to and challenges of the tooth biomineralization in man and other mammalians.
In conclusion, structural elements of human tooth buds and zebrafish developing teeth concerning mineralizing tissues of the pulp–dentin complex divided into predentin, mineralizing dentin, and mature dentin are similar, but with completely different developmental periods of a few days versus years and decades. Therefore, the zebrafish pharyngeal dentition may serve as a model for biomineralization research, developmental aberrations, and environmental influences.
For the first time, in this study we compared mineralization of human dentin and dentin of zebrafish pharyngeal teeth with micoanalytic and micromorphological methods. It could be shown that there are certain differences in the accumulation of Ca at the mineralization front of predentin between both species, but the main mineralization pattern may be the same.