Eva Matalová, Institute of Animal Physiology and Genetics, v.v.i., Academy of Sciences of the Czech Republic, Veveří 97, 602 00 Brno, Czech Republic. T: + 420 5 32290155; F: + 420 5 41212988; E:email@example.com
The first mouse molar (M1) is the most common model for odontogenesis, with research particularly focused on prenatal development. However, the functional dentition forms postnatally, when the histogenesis and morphogenesis of the tooth is completed, the roots form and the tooth physically anchors into the jaw. In this work, M1 was studied from birth to eruption, assessing morphogenesis, proliferation and apoptosis, and correlating these with remodeling of the surrounding bony tissue. The M1 completed crown formation between postnatal (P) days 0–2, and the development of the tooth root was initiated at P4. From P2 until P12, cell proliferation in the dental epithelium reduced and shifted downward to the apical region of the forming root. In contrast, proliferation was maintained or increased in the mesenchymal cells of the dental follicle. At later stages, before tooth eruption (P20), cell proliferation suddenly ceased. This withdrawal from the cell cycle correlated with tooth mineralization and mesenchymal differentiation. Apoptosis was observed during all stages of M1 postnatal morphogenesis, playing a role in the removal of cells such as osteoblasts in the mandibular region and working together with osteoclasts to remodel the bone around the developing tooth. At more advanced developmental stages, apoptotic cells and bodies accumulated in the cell layers above the tooth cusps, in the path of eruption. Three-dimensional reconstruction of the developing postnatal tooth and bone indicates that the alveolar crypts form by resorption underneath the primordia, whereas the ridges form by active bone growth between the teeth and roots to form a functional complex.
The first lower mouse molar (M1) has been used as a common model for odontogenesis research for a 100 years and thus the majority of our knowledge of tooth development has been acquired from experiments and studies in this tooth. The ability to manipulate mouse embryos moved the original examinations based particularly on morphology, histology and related methods in histological sections to valuable functional studies, connecting morphogenetic patterns with key genes. Moreover, knowledge obtained in the mouse M1 has been successfully extrapolated to human odontogenesis, and related disorders connected with altered tooth number, size, structure and shape found in human dentistry (Fleischmannova et al. 2008).
Most studies of the mouse M1 have been focused on prenatal development of the tooth, when the tooth crown development is initiated; gradually, step by step, the tooth gains and fixes its final shape by deposition of dental hard tissues (reviewed in Tucker & Sharpe, 2004). However, the journey towards the functional tooth begins postnatally.
In the postnatal period, the histo- and morphogenesis of the tooth is completed and, most importantly, the erupted M1 firmly anchors into the jaw. Most data about postnatal M1 development focus on key events, such as root formation and elongation, cementogenesis and mechanisms of tooth eruption. Root development starts with the formation of the Hertwig’s epithelial root sheath (HERS) – an epithelial bilayer that extends apically below the level of the crown cervical margin (Diekwisch, 2001; Zeichner-David et al. 2003). The formed HERS proliferates and tends to bend inward to establish a barrier between the dental papilla (pulp) and dental follicle (periodontium). The root sheath forms a continuous sheet structure at early postnatal stages that acts as a guide for root elongation and segregation and induces the differentiation of dental papilla cells into odontoblasts (reviewed by Shimazu et al. 2009). Another key phenomenon is HERS disintegration orchestrated with root dentin deposition, which, in turn, opens up a space for migration of mesenchymal cells – future cementoblasts – onto the dentin surface to initiate cementogenesis (Andujar et al. 1985; reviewed by Bosshardt & Schroeder, 1996; Diekwisch, 2001; Shimazu et al. 2009). HERS cells may have a diverse fate. They undergo apoptosis (Kaneko et al. 1999; Cerri et al. 2000; Cerri & Katchburian, 2005) or undergo epithelial–mesenchymal transformation (Thomas, 1995; Zeichner-David et al. 2003), they can be incorporated into the advancing cementum front (Lester, 1969; Luan et al. 2006), and/or migrate away from the roots surface into the periodontal ligament, where they form rests of Malassez (Wentz et al. 1950; Wesselink & Beertsen, 1993; Yokohama-Tamaki et al. 2006). Prior to tooth eruption, gradual cell death occurs in the ameloblasts between late secretory and late maturation stages at the culmination of enamel formation (reviewed in Matalova et al. 2004).
This paper aims to provide an overview of postnatal M1 morphogenesis up to eruption accomplished by proliferation and apoptosis distribution, matched with the surrounding bone remodeling and formation of the alveolar bone.
Materials and methods
Mice (CD-1 strain) were sacrificed at every postnatal (P) day up to P16 according to the experimental protocol approved by the Laboratory Animal Science Committee of the IAPG CAS, v.v.i. The mouse heads were immediately fixed in 4% buffered paraformaldehyde and decalcified in buffered ethylenediaminetetraacetic acid (EDTA). Mandible quadrants were dissected, dehydrated in gradient series of ethanol, treated with xylene and embedded in paraffin. Frontal serial sections of 4 μm were cut and split over four slides: hematoxylin-eosin (HE) staining (morphology), proliferating cell nuclear antigen (PCNA) immunohistochemistry (proliferation), TdT-mediated dUTP-biotin nick end labeling (TUNEL) assay (apoptosis) and tartrate-resistant acid phosphatase (TRAP) staining (osteoclast activity).
Cell proliferation was confirmed by immunohistochemical detection of PCNA along with mitotic figure evaluation in HE sections. PCNA immunohistochemical staining was processed after section rehydration and endogenous peroxidase activity elimination by 3% hydrogen peroxide in phosphate-buffered saline [PBS; 5 min at room temperature (RT)]. To visualize the primary antibody (sc-7907; Santa Cruz) applied at a concentration of 4 μg ml−1 overnight at 4 °C, the peroxidase-conjugated streptavidin–biotin system (Vectastain) and chromogen substrate diaminobenzidine (DAB, K3466; Dako, Denmark) reaction were used, making positive cell nuclei brown. Slides were counterstained by hematoxylin to clearly distinguish negative cell nuclei in blue.
TdT-mediated dUTP-biotin nick end labeling (TUNEL, S7100; Chemicon-Millipore, USA) was employed to detect apoptosis along with evaluation of apoptotic bodies in HE sections. After rehydration, endogenous peroxidase activity was eliminated by 3% hydrogen peroxide in PBS (5 min at RT) and samples pretreated by proteinase K (Chemicon-Millipore, 21627) 20 μg ml−1 at RT for 15 min. After equilibration buffer, the reaction mixture was prepared following the method manual and applied at 37 °C for 50 min. After anti-digoxigenin-peroxidase reaction at RT for 30 min, positive cell were visualized by chromogen substrate diaminobenzidine (DAB; K3466; Dako) and slides counterstained by hematoxylin.
The TRAP substrate reaction was used to detect osteoclastic activity along with morphological confirmation of the cell type in HE sections. After rehydration, slides were immersed in the reaction mixture prepared according manufacturer’s directions (387A-1KT; Sigma-Aldrich, AR-MED Ltd., UK) and kept at 37 °C for 2 h to achieve the color reaction with Fast Red substrate. Slides were counterstained by hematoxylin.
Three-dimensional (3D) reconstruction
Mouse heads, age ranging from stages P0 to P14, were fixed in Bouin’s solution and dehydrated in alcohol with increasing concentrations up to 100%, according to standard histological procedures. Depending on the size and gross preparation, the specimens were decalcified using EDTA. Paraffin embedding was carried out according to standard procedures, and the specimens were cut as 7-μm-thick serial sections (Leica, Reichert-Jung RM 2065, Nußloch, Germany) in frontal planes. Routine staining was performed with hematoxylin-eosin. In addition, staining methods according to Masson-Goldner, Giesson modified according to Domagk, trichrome elastica staining (Pressnell & Schreibman, 1997) and TRAP staining (Sigma, Deideshofen, Germany) were applied according to Jösten et al. (1993).
The sections were photographed using a standard microscope (Zeiss, Göttingen, Germany) equipped with a motorized stage (x, y and z axes motorized, Merzhäuser, Wetzlar, Germany) at magnifications of between 1.25× and 40×. The images were taken with a camera (Color View IIIu; Olympus-SIS, Münster, Germany) and analyzed using the software analysis doku (Olympus-SIS). The 3D module of the software analysis (Olympus-SIS) was used to generate 3D images from the serial sections. The single sections were brought into alignment according to the general rules of 3D reconstructions (Gaunt & Gaunt, 1978). In the orofacial region, structures such as the facial contour, the form of the tongue and other paired organs, Meckel’s cartilage and the course of nerves were used to check the alignment for accuracy. The contours of all the tissues and organs of relevance were diagnosed on the screen at higher magnifications, and traced using the 3D function of the software (Olympus-SIS). All structures were reconstructed as triangular meshes by means of the 3D function of the software. Surfaces were color-coded in the 3D reconstructions as explained in Fig. 1.
The early stages of postnatal development are associated with the completion of crown formation (P0 and P2) and the initiation of root development (P4). The alveolar bone surrounds almost the entire dental primordium and the buccal and lingual ridges even grow over the occlusal plane of the tooth. The enamel organ at these stages is clearly visible and shows a familiar arrangement of the four epithelial layers (OEE, outer enamel epithelium; SR, stellate reticulum; SI, stratum intermedium, and IEE, inner enamel epithelium). It tends to be reduced (due to the disappearance of SR) particularly in the regions where the cusps reach the alveolar crests. The apical part of the tooth crown terminates with the cervical loop (cl) at P0 and P2 or the HERS at P4.
At P0, ameloblasts at the tips of the tooth cusps trigger enamel secretion. This process is switched on after coronal predentin deposition, which starts to be formed during late embryonic stages and proceeds postnatally. At P0, differentiated odontoblasts are aligned perpendicularly to the basement membrane of the IEE-producing predentin (Fig. 2A–C). Ameloblast differentiation begins at the tips of the cusps (the lingual cusp first) and gradually spreads downwards buccally and lingually (Fig. 2A1). The enamel organ is joined to the surface epithelium via the dental stalk, which begins to disintegrate due to apoptosis of cells in the dental stalk or in the adjacent OEE (Fig. 2C2) and gradually becomes disconnected to the enamel organ. Nearly all cell layers of the enamel organ proliferate (except for the already differentiated ameloblasts and cells of SI, at the tips of the tooth cusps) (Fig. 2A1, B1 and C1). PCNA-positive cells were also found in mesenchymal cells of the dental follicle and in the dental pulp (Fig. 2B1) or in cells along the outer surface of the alveolar ridges (Fig. 2A1). The latter correlates with places of bone apposition. The resorption processes, caused either by apoptotic elimination of osteoblasts or by osteoclast activity, occur along the base of the tooth crypt (Fig. 2B2,B3) and in the inner margin of buccal and lingual ridges (Fig. 2A2,A3). Apoptotic cells and bodies were particularly located in the alveolar bone, where the SR is reduced (along the tooth cusps) (Fig. 2A2). The newly formed bone is primarily deposited on the outer surface of buccal and lingual ridges and in the interdental space forming the alveolar crest that separates M1 from M2 (Fig. 1A–C). A schematic illustration of the spatial distribution of proliferation, apoptosis and TRAP-positive osteoclasts in the lower M1 at P0 can be seen in Fig. 3A.
At P2, enamel production is established (Fig. 4A–C). Cell proliferation in the dental epithelium, therefore, reduces and moves downward to the cervical loop (Fig. 4B1,C1). However, mitosis persists in the mesenchymal cells of the dental sac at the tooth base near the apical foramen (Fig. 4B1,C1) and along the outer margins of the alveolar crests (Fig. 4A1). Bone turnover proceeds with apoptotic cells and TRAP-positive osteoclasts, removing bony matrix situated along the base of the tooth crypt and in the inner margin of the buccal and lingual ridges above the tooth cusps (Fig. 4A2–C2 and A3–C3). A schematic illustration of the spatial distribution of proliferation, apoptosis and TRAP-positive osteoclasts in the lower M1 at P2 can be seen in Fig. 3B.
The development of the tooth root commences at P4 with the emergence of the HERS, which is formed at the apical region of the enamel organ by proliferation of the inner and outer enamel epithelium cells from the cervical loop (Fig. 5A–C). The double cell layer of the most apical part of HERS, the diaphragm, tends to bend inward and narrow the apical foramen, thus separating the dental papilla from the dental follicle. The HERS cells proliferate (Fig. 5B1) together with the mesenchymal cells of the dental follicle, particularly adjacent to the lingual alveolar crests (Fig. 5A1), then in the dental pulp around the apical foramen (Fig. 5B1) and in the cells located along the outer surface of the alveolar ridges (Fig. 5A1,C1). The latter corresponds with the places of bone apposition (Fig. 1D–F). The distribution of apoptotic cells and TRAP-positive osteoclasts shows similar patterns to those at P0 or P2 (Fig. 5A2–C2 and A3–C3). However, the number of apoptotic cells and active osteoclasts in the inner margin of the alveolar crests located above the cusps increases, so the emphasis is now placed on their removal rather than their formation. This tendency is especially apparent during the following postnatal stages (P6 and P8). A schematic illustration of the spatial distribution of proliferation, apoptosis and TRAP-positive osteoclasts in the lower M1 at P4 can be seen in Fig. 3C.
During the following stages of M1 development (P6–P8) the tooth rapidly increases in size, followed by dynamic bone reshaping due to bone resorption. Simultaneously, odontoblasts in the dental root differentiate and trigger the deposition of predentin/dentin, which influences the continuity of the HERS.
Rapid tooth growth and elongation is initiated at P6 (Fig. 6A–C). The epithelial sheath bilayer proliferates and moves downwards into the surrounding bone (Fig. 6B1). Although bone regression accelerates around the whole dental primordium (Fig. 6B2,B3), it is particularly intensive along the inner margins of the alveolar crests located in the occlusal plane (Fig. 6A2,A3,C3). Clusters of multinuclear TRAP-positive osteoclasts and apoptotic cells situated in this region act to open the alveolus and prepare the eruption path. The soft tissues located near the bony ridges also begin to recede. Cells of the SI and some ameloblasts along the tooth cusps and in the intercuspal space become apoptotically eliminated (Fig. 6C2). A schematic illustration of the spatial distribution of proliferation, apoptosis and TRAP-positive osteoclasts in the lower M1 at P6 can be seen in Fig. 3D.
At P8, odontoblasts of the dental pulp in the root region initiate dentinogenesis (Fig. 7A–C), accompanied by the beginning of HERS fenestration. PCNA-positive cells can be observed in cells of the enamel organ (in the diaphragm of the epithelial root sheath, Fig. 7B1,C1, and in the SR in the intercuspal region, Fig. 7A1), and in mesenchymal cells near the apical foramen (Fig. 7B1). Proliferation steadily increases in the mesenchymal cells of the dental sac adjacent to the tooth root (Fig. 7B1,C1). Bone regression proceeds along the whole dental primordium (Fig. 7B3,C3) but primarily in the buccal and lingual ridges above the tooth cusps (Fig. 7A2,A3). At P8, a few apoptotic cells were also detected in the dental papilla and in the epithelium close to the cemento-enamel junction on both the buccal and the lingual sides of the tooth (Fig. 7B2,C2). This finding reflects the beginning of HERS fenestration at P8. Bone apposition is obvious along the outer margin of the alveolar crests with proliferation of osteoblasts (Fig. 7A1), particularly on the buccal side of the tooth. A schematic illustration of the spatial distribution of proliferation, apoptosis and TRAP-positive osteoclasts in the lower M1 at P8 can be seen in Fig. 3E.
Between P10 and P16, significant root elongation and advanced tooth mineralization occur and the alveolar crests surrounding the dental crown are removed. The alveolus opens widely so the tooth can move occlusally towards the surface epithelium. Also, the inferior part of the bony socket enlarges slightly, which is clearly apparent at P14 and P16. The newly formed space facilitates rapid development of the periodontal ligaments that fix the tooth to the alveolar bone. In this manner, the tooth gradually embeds and anchors into the bone during and after eruption through the bone and soft tissues.
The first stages of escape of the tooth crown from its surrounding bone are visible at P10, when, by bone regression in previous stages, the tooth gains enough space to move occlusally towards the oral epithelium (Fig. 8A–C). The growth of the root, however, continues. So cells of the epithelial root sheath still proliferate (particularly the most apical part of HERS, the diaphragm, where the dentin has not yet been deposited). Proliferation is also observed in the mesenchymal cells of the dental papilla (round the apical foramen) and the dental sac beside the tooth (Fig. 8A1,C1). Advancing root dentin deposition brings about further HERS disintegration, which opens up space for migration of mesenchymal cells (future cementoblasts) on the root dentin surface, with a few epithelial cells facing the root dentin undergoing cell death (Fig. 8C2). Apoptosis together with active osteoclasts was also apparent at the tips of alveolar septa and surrounding soft tissues, especially on the buccal side of the tooth (Fig. 8A2,A3,B2,B3). The bone formation is redirected primarily into the region between developing roots, into the interradicular space (Fig. 1G–I). This finding reflects the beginning of tooth root bifurcation. A schematic illustration of the spatial distribution of proliferation, apoptosis and TRAP-positive osteoclasts in the lower M1 at P10 can be seen in Fig. 3F.
The P12 stage is associated with the rapid growth of periodontal ligaments and the beginning of bony socket expansion (Fig. 9A–C). Root elongation proceeds, but only a few proliferating cells can be found in the most apical part of HERS, in the diaphragm (Fig. 9C1). Increasing numbers of PCNA-positive cells were detected in the mesenchyme of the dental follicle beside the tooth root, in the region of the periodontal ligament fibroblasts and future cementoblasts (Fig. 9C1). Increasing numbers of cells in the periodontium are balanced by cell death, so some cells of the periodontium undergo apoptosis (Fig. 9C2). The apoptotic elimination of cells located near the tooth crown proceeds as well, as does the bone resorption, which occurs particularly on the tops of the alveolar septa and beneath the dental primordium. In keeping with this, apoptotic cells and active osteoclasts were located along the base of the crypt (Fig. 9C2,C3) and on the top of the lingual and particularly buccal crests, Fig. 9A2–A3 and B2–B3). The loss of bone in the inferior part of the alveolar compartment brings about the enlargement of the socket, which is necessary for the proper growth of the periodontal ligaments. A schematic illustration of the spatial distribution of proliferation, apoptosis and TRAP-positive osteoclasts in the lower M1 at P12 can be seen in Fig. 3G.
During later postnatal stages the enamel epithelium surrounding the tooth crown transforms into the reduced enamel organ, which fuses with the oral epithelium. The appearance of this structure, composed of cell remnants of ameloblasts (IEE), SI, SR and OEE, implies that the crown mineralization and maturation of hard tissues have culminated and the tooth is about to erupt into the oral cavity. Cell proliferation in both the dental epithelium and mesenchyme of the dental sac decreases rapidly (at P14) and completely stops (at P16). The tooth elongation continues, but it is caused by the deposition of the hard substances particularly in the root region rather than by the active growth of the root.
At P14, the alveolus is widely open, and the eruption path is clear and prepared for the tooth to emerge in the oral cavity and gain its functional position. The base of the socket is expanded so that the periodontal ligaments can develop properly and firmly fix the tooth to the bone (Fig. 10A–C). Cell division in cells of the dental epithelium ceases (Fig. 10A1,B1), and it is also very low in the fibroblasts of the periodontal ligament. Only a few PCNA-positive cells were detected in the periodontium next to the tooth root, particularly close to the cemento-enamel junction (Fig. 10C1). On the other hand, the number of apoptotic cells increases. Apoptosis principally eliminates soft tissues located above the tooth cusps in the reduced enamel epithelium (ameloblasts, SI, oral epithelium) (Fig. 10A2,B2). A single TUNEL-positive cell was also found in the periodontium (Fig. 10C2). Bone turnover significantly slows down; however, a few active osteoclasts are still observed on the top of the alveolar crest on the buccal side of the tooth (Fig. 10A3), at the base of the bony crypt beneath each root or along the outer osseous surface of the alveolar ridges (Fig. 1J–L). Active osteoblasts deposit the bony material along the inner wall of the compartment (Fig. 1J–L). A schematic illustration of the spatial distribution of proliferation, apoptosis and TRAP-positive osteoclasts in lower M1 at P14 can be seen in Fig. 3H.
At P16, the M1 is fully developed and almost protrudes into the oral cavity (Fig. 11A–C). Cell proliferation can be seen neither in cells of the enamel organ nor in the periodontium (Fig. 11A1–C1). However, apoptosis in soft tissues located above the tooth cusps continues (Fig. 11A2). TUNEL-positive cells were also found in mesenchymal cells of the dental pulp (Fig. 11C2) and even in odontoblasts (Fig. 11B2). Bone resorption is generally low, with only a few active osteoclasts detected beneath the dental primordium (Fig. 11C3). A schematic illustration of the spatial distribution of proliferation, apoptosis and TRAP-positive osteoclasts in lower M1 at P16 can be seen in Fig. 3I.
In the present research, morphogenesis of the mouse lower M1 was followed from P0, when the development of the dental primordium was well in progress, to P16, when the functional tooth was about to erupt into the oral cavity. The spatial arrangement and timing of proliferation and cell death were monitored within the tooth organ and in surrounding tissues along with bone turnover (apposition and resorption) and formation of alveolar compartments in 3D models. Although cell proliferation was particularly involved in the growth of the tooth and its elongation and formation of periodontal structures, apoptosis played an essential role in shaping the surrounding tissues and controlling cell numbers. Apoptosis was reported to occur during all stages of molar morphogenesis (reviewed in Matalova et al. 2004). In postnatal mouse M1, apoptosis was seen particularly in the dental stalk and adjacent enamel epithelium. Programmed cell death also participated in removal of osteoblasts and, together with osteoclasts, exercised an influence on the remodeling of the jaw bone and formation of alveolar compartments. In more advanced developmental stages, before tooth eruption, apoptosis was involved in elimination of cells in the reduced enamel organ located in the eruption zone.
Our findings complete and refine the limited data reported in previous studies. We confirmed that whereas odontoblasts of the dental papilla differentiate during late prenatal stages (at P0 they have already aligned themselves perpendicularly to the basement membrane of the IEE and initiated predentin deposition), ameloblasts begin to differentiate and trigger enamel secretion postnatally. At P0, the majority of cells in the inner enamel epithelium (future ameloblasts) and SI are still proliferating, apart from ameloblasts at the tips of the prospective cusps, which are about to enter their secretory phase. At P2, as the crown mineralization advances and more ameloblasts become engaged in enamel production, cell proliferation in the dental epithelium is reduced and moves downward to the cervical loop and later (at P4) Hertwig’s epithelial root sheath. By that time, the dental crown morphogenesis is completed and odontogenic development moves to the region of developing roots. However, deposition of hard tissues in the coronal domain proceeds, so the size of the crown keeps on increasing in all directions (Diamond & Applebaum, 1942), exerting continuous pressure on the surrounding tissues, particularly on the adjacent alveolar ridges. The ridges are steadily removed either by apoptotic elimination of osteoblasts or by osteoclastic bone resorption. The bone turnover is accompanied by simultaneous elimination of soft tissues located in the occlusal plane. TUNEL-positive cells were particularly abundant at P6 in ameloblasts and cells of the stratum intermedium situated above the tooth cusps and in the interdental space. Following the gradual withdrawal of cells located in the eruption zone, the alveolus opens widely, and the dental crown moves occlusally and approaches the surface epithelium (starting at P10). Remaining ameloblasts along with other layers of epithelial cells persist and cover the enamel until the tooth erupts. Prior to tooth eruption, the enamel organ atrophies, epithelial cells of the reduced enamel organ change in shape and undergo apoptosis, or they are moved away as the tooth presses towards to the oral cavity.
Histomorphogenesis in the crown region is followed by histomorphogenesis in the root region. According to our results, the lower M1 roots start to develop at P4 with the formation of HERS as an extension of the inner and outer dental epithelia at the apical edge of the dental organ. This is somewhat earlier than that described in previous studies. According to Zeichner-David et al. (2003), the HERS appears at P7. Similarly Yen & Sharpe (2008) reported that IEE and OEE give rise to HERS at P6. These differences in timing might be associated with the strain of mice used, as the day they give birth varies; CD-1 mice, as used in this study, are usually born at 19.5 days.
The HERS structure goes beyond the enamel-dentin junction and moulds the dental papilla to form the tooth root (Bosshardt & Schroeder, 1996; Shimazu et al. 2009; Fleischmannova et al. 2010). It initiates formation of dentin in the root region as well as cementum deposition (Lester, 1969; Slavkin et al. 1989). The growth of HERS occurs by directed proliferation of epithelial cells of the root sheath (Cho & Garant, 2000; Harada et al. 2002). The most apical part of HERS, the diaphragm, bends inward and grows between the dental papilla and dental follicle, thus narrowing the apical foramen and acting as a barrier between the dental pulp and periodontium (Grant & Bernick, 1971; Thomas, 1995). From P1 to P7, the HERS layer is continuous; later, at P8–P23, immediately after root dentin is formed, the epithelial sheath disintegrates into a network of HERS cells (Luan et al. 2006). Based on our results, at P8, apoptosis eliminated a few HERS cells located near the cemento-enamel junction on both sides, buccal and lingual, of the tooth, which matches with the beginning of HERS fenestration. At P10, the HERS fragmentation continues, with more TUNEL-positive HERS cells detected along the tooth root. Disintegration of the epithelial sheath enables inner cells of the dental follicle (future cementoblasts) to come in contact with the root dentin and initiate cementum deposition (Bosshardt & Schroeder, 1996; Diekwisch, 2001). What actually happens to the HERS cells is still a matter of debate. Our results agree with previous reports (Kaneko et al. 1999) which have shown that apoptosis in HERS cells is relatively rare and occurs in cells located from the cervical area down to apical part of the root, where the root dentin deposition processes. Although we confirmed cell death in HERS at P8 and then at P10, the TUNEL reaction was positive only in a couple of cells located in the region mentioned above. Interestingly, at P20, there was an increase in apoptosis in HERS facing the root surface, which indicates that more HERS cells are eliminated later, when the tooth already protrudes into the oral cavity (data not shown). Those HERS cells, which remain vital, can be incorporated into the advancing cementum front, undergo epithelial–mesenchymal transformation, or migrate away from the root surface into the region of future periodontal ligaments to form the rests of Malassez. To determine their fate, lineage labeling would be necessary (Andujar et al. 1985; Gurling & Sampson, 1985; Thomas, 1995; Kaneko et al. 1999; Cerri et al. 2000; Zeichner-David et al. 2003; Cerri & Katchburian, 2005; Luan et al. 2006).
From P4 to P12, by intensive proliferation of the HERS, the root elongates and moves downward into the jawbone. However, at P14, cell division in the root sheath suddenly ceases. Thus the additional tooth elongation observed after this stage is caused by the deposition of the hard substances rather than by the active growth of the dental root. Simultaneously with the root development, adjacent mesenchymal cells of the dental follicle (particularly those cells situated in the apical region of the tooth germ), proliferate and generate cell populations that, consequently, form the developing periodontium and also cell populations that will contribute to the developing radicular pulp (Thomas, 1995). We detected mitoses in mesenchymal cells of the dental sac around the apical foramen already in a newborn mouse (P0) as well as in more advanced stages of odontogenesis. They were particularly intensive at P12, when cell proliferation culminated. Consequently, at P14, new populations of cells differentiated into fibroblasts producing periodontal ligaments and cementoblasts producing root cementum. Some cells situated near the apical foramen differentiated into mesenchymal cells of the dental pulp. The intensive proliferation of cells in the dental sac must be balanced by their apoptotic elimination during later postnatal stages as they become crowded within a space delimited by the size of an alveolar compartment. We detected apoptosis in fibroblast-like cells during stages of periodontium differentiation and initiation of cementum deposition (P12 and P14). These findings correlate with previously published data (Cerri et al. 2000). Moreover, during more advanced stages of dentinogenesis, apoptosis would be expected in mesenchymal cells and odontoblasts of the dental pulp which become crowded by the reduction in pulp space. Although we confirmed apoptosis in mesenchymal cells within the dental papilla at P8, P10 and P14, it was generally very low and did not occur in odontoblasts. However, at P16, there was a TUNEL-positive cell in a layer of odontoblasts on the lingual side of the dental root. This finding reflects the beginning of odontoblast elimination at P16 during tooth eruption.
The inferior elongation of the root and developing periodontium exert permanent pressure on the alveolar bone surrounding the dental primordium. Thus the inner margins of the bony socket are steadily resorbed along with apoptotic elimination of osteoblasts and by activity of osteoclasts, and simultaneously widened to create enough space for the proper development of the periodontal ligament. The dental primordium, therefore, gradually embeds into the socket and attaches to the surrounding bone.
Although much information can be generated from 2D histological data, some developmental aspects can only be followed using 3D reconstructions. Such an approach enables one to assess the sites of bone resorption or apposition in the whole mandible, provides a much more complex view of the integration of the tooth into the jaw, and explains the formation of the alveolar process. During both crown and root formation, surrounding bone remodeling must be synchronized. Based on our results, the M1 at P0 is nearly completely encapsulated by mineralized bone but as the tooth development proceeds, it sinks further down into the bone beneath, escapes from the bone surrounding the crown, and redirects the growth of the alveolar bone around its developing roots. Thus the sockets are modeled by bone resorption underneath the primordia, whereas alveolar partitions, or crests, are formed by active bone growth in-between teeth and roots at later stages. Nevertheless, the height of the alveolar crests needs to be adjusted to enable tooth eruption and is then further maintained and regulated. Prior to eruption, the newly formed bone is deposited on the bottom of the bony crypts, helping to push the tooth up to the oral cavity. Bone resorption and apposition are therefore an integral part of tooth eruption.
Our findings show the temporo-spatial correlation of M1 postnatal development with bone formation and remodeling. These processes are not very dynamic during late prenatal and earlier postnatal stages prior to root formation because tooth germs show only minor bodily movements in all directions within the bony compartments. The most dynamic changes in the shape of the alveolar process begin to occur together with root development and tooth eruption (Cho & Garant, 2000). Later, when the tooth has already protruded in the oral cavity, the bone turnover slows down, but it persists until the tooth is embedded in the bony socket. Due to the tooth–periodontium interactions and mechanical load on the tooth during mastication, the vertical level of the alveolar crests is steadily maintained. The preservation of marginal bone height is critical for tooth anchorage. Proper development of all tooth-periodontium-alveolar structures is achieved by well-coordinated signaling interactions (Thesleff & Sharpe, 1997; Sodek & McKee, 2000; Tucker & Sharpe, 2004; Wise & King, 2008; Fleischmannova et al. 2010), and mechanical forces between the tooth and its neighboring tissues (Benjamin & Hillen, 2003; Silver et al. 2003; Blechschmidt, 2004; Ingber, 2005; Radlanski & Renz, 2006). Mechanical interactions should be considered, particularly when these tissues grow at different speeds. Bone formation is triggered in regions where the mesenchymal tissue undergoes a shearing traction (Blechschmidt, 1948, 1978), whereas bone resorption occurs in places exposed to continuing pressure. In the case of the mouse first molar development, an expanding dental primordium exerts a shearing force as it slides against its adjacent mesenchymal cells located between M1 and M2, then on the outer osseous surface of the buccal and lingual ridges, and later, as the tooth roots develop, also in the interradicular space. As a result, detracted or stretched mesenchymal cells situated in these regions differentiate into osteoblasts that start to secrete the bony material. Thus the newly formed bone is deposited in the interdental space separating M1 from M2 and along the outer margins of alveolar crests and later (from P10 onward) also in the interradicular space, where it forms partitions between the roots. The inner margin of the alveolar crypt and the terminal parts of the septa are exposed to constant pressure as the tooth enlarges, elongates and presses towards the oral cavity. Thus the bone situated there is steadily removed until the tooth clears its eruption path, gains its final size, and erupts into the oral cavity.
The alveolar bone proper, together with the periodontal ligaments, cementum and gingival, originates in the dental follicle that is derived from neural crest (Diep et al. 2009). These structures develop along with the formation of the roots and tooth eruption. The size and shape of the tooth determine the morphology of the alveolar bone, which continues to undergo remodeling as the tooth progresses in development and later erupts (Cho & Garant, 2000). However, the pattern of the alveolar bone as it adjusts to dental development is not completely clear. In general, there are two possibilities as to how alveolar compartments are formed. Either the bone is already there, so the tooth merely sinks into it (such an approach assumes massive bone resorption around the whole dental primordium and almost no bone apposition) or the bone is formed just underneath the primordia and then actively grows into the space in-between teeth or roots, forming alveolar crests. In the latter case, active bone growth and almost no bone resorption would be expected. We confirmed that both processes, bone resorption and apposition, are running simultaneously and they are correlated with the tooth growth and elongation, so that the tooth gradually embeds itself into the socket. Crypts are formed by bone resorption underneath the primordia, whereas ridges are formed by active bone growth in the interdental and interradicular spaces in later stages.
We hope this overview of the first mouse molar–bone morphogenesis and related cellular events will create a basis for more detailed studies aiming to complete the jigsaw of physiological dynamics during formation and maintenance of the tooth–bone complex.
This work was supported by a joint project of the Grant Agency of the Czech Republic (524/08/J032) and the Deutsche Forschungsgemeinschaft (Ra 428/1-9). Applied tooth-bone research was supported by the Ministry of Health of the Czech Republic (project NT 11420) and CR-UK cooperation by the Royal Society Joint International Grant (JP080875).
We thank Barbara Danielowski and Irene Schwarz for their very skilful processing of the histological material and the computer work in the Berlin Laboratory, and Katarína Marečková for sample preparation in the Brno lab.