Monitoring of canonical BMP and Wnt activities during postnatal stages of mouse first molar root formation

Abstract Objective This study aimed to explore the precise temporospatial distributions of bone morphogenetic protein (BMP) and Wnt signaling pathways during postnatal development of mammalian tooth roots after the termination of crown morphogenesis. Methodology A total of two transgenic mouse lines, BRE-LacZ mice and BAT-gal mice, were undertaken. The mice were sacrificed on every postnatal (PN) day from PN 3d up to PN 21d. Then, the first lower molars were extracted, and the dissected mandibles were stained with 5-bromo-4-chloro-3-indolyl-β-d-galactopyranoside (X-gal) and fixed. Serial sections at 10 µm were prepared after decalcification, dehydration, and embedding in paraffin. Results We observed BMP/Smads and Wnt/β-catenin signaling activities in the dental sac, dental pulp, and apical papilla with a certain degree of variation. The position of activation of the BMP/Smad signaling pathway was located more coronally in the early stage, which then gradually expanded as root elongation proceeded and was associated with blood vessels in the pulp and developing complex apical tissues in the later stage. However, Wnt/β-catenin signaling was highly concentrated in the mesenchyme below the cusps in the early stage, gradually expanded to regions around the root in the transition/root stage, and then disappeared entirely in the later stage. Conclusions These results further confirmed the participation of both BMP and Wnt canonical signaling pathways in tooth root development, as well as formed the basis for future studies on how precisely integrated signaling pathways regulate root morphogenesis and regeneration.


Introduction
Analogous to the ectodermal organogenesis, a series of reciprocal and sequential epithelialmesenchymal interactions are essential for tooth root development, which occurs after crown formation. 1 Multiple signaling cascades have implicated in these processes to guide root development in different stages. 2 The outer enamel epithelium is confluent with the inner enamel epithelium at the cervical loop, as well as grows in the apical direction forming the bilayer of an epithelial extension called Hertwig's epithelial root sheath (HERS), a transient structure that occurs before the complete formation of the tooth root. 3 The formed HERS extends apically and envelops the dental papilla to form a barrier separating from the dental sac. The HERS is known to induce the differentiation of odontoblasts and cementoblasts, to guide root growth, and to determine the number of tooth roots.
HERS disintegration orchestrates the deposition of root dentin, which subsequently initiates dental sac cell cementogenesis onto the dentin surface. 4,5 Moreover, it may also participate in cementum formation via epithelial-mesenchymal transition. 6,7 Thus, HERS plays a vital role in root development.
BMP and Wnt signaling pathways are crucial for root formation, including cell fate, growth, and patterning. The BMP-Smad4-Shh-Gli1-Sox2 signaling cascade worked in root development, where the disruption of BMP signaling in the dental epithelium causes delays in HERS formation and affects the stem cell niche environment. 8 Moreover, the ablation of Smad4, the key intracellular mediator of the TGF-β/ BMP signaling pathway, in odontoblasts disrupts odontoblast differentiation and causes abnormal root odontogenesis. 9,10 Previous studies also revealed the essential role of Wnt signaling during root formation; we detected the canonical Wnt activity in the mesenchyme and odontoblasts in BAT-gal/ TOPgal reporter mice. 11,12 Furthermore, recent reports showed that the disruption of β-catenin in immature odontoblasts generates molars without roots, whereas the constitutive stabilization of β-catenin can also lead to aberrant short roots with excessive cementum. 13,14 The cervical loop structure of mouse molars disappears after crown formation, and then HERS is formed to guide root development postnatally, which is similar to human tooth organogenesis. Thus, the first lower mouse molar (M1) is an ideal model for studying tooth root development. In the past decades, the molecular regulatory network of early tooth morphogenesis was studied extensively, with research mainly concentrating on prenatal stages. 15   involved CD-1 background mice. The mice were kept in a 12-hour light/dark period with a temperature of 22 ± 2 0 C and humidity of 45% ± 15% of the rooms.

Tissue preparation
Pups from mice of both sexes were sacrificed on every PN day from day 3 (PN3) up to PN21 (n: 5 pups).

Results
The results showed that M1 crown development harvested the M1 with the whole dental sac. Since the newly formed cusp was fragile and soft, it was difficult to separate the dental sac or epithelium from the crown until PN14 ( Figures 1A-1C). The whole-mount X-gal staining results detected BRE-LacZ activity in the dental sac around the tooth, the pulp, and the apical papilla (around the apical foramen) in the initial stages.
During the following stages of M1 morphogenesis (PN6-PN8), the teeth rapidly increased in size and length. BRE-LacZ activity became lower in the crown pulp but intensified in the root pulp, dental sac, and apical papilla ( Figures 1D-1F). Between PN9 and PN14, significant root growth and further tooth mineralization occurred with crown calcification (Figures 1G-1L).
Meanwhile, we established the eruption path for M1 and the tooth was ready to erupt into the oral cavity, to reach its occlusal position, and to perform its function after PN14 (data not shown). We detected strong X-gal staining at the dental sac around the root and the apical region ( Figures 1G-1L). Furthermore, the We traced Wnt/β-catenin signaling in the developing molar using the BAT-gal allele, a reporter mouse line that expressed β-galactosidase in the presence of activated β-catenin. On PN3, we established the enamel production. Different from BMP/Smad signaling activity, active Wnt/β-catenin signaling was restricted in the crown pulp region beneath the cusps and did not appear in the dental sac, root pulp, and apical papilla ( Figures. 2A-A 2 ). Wnt/β-catenin signaling activity was highly concentrated under the tip of the molar cusps during PN3-PN6, a stage before dentinogenesis in the root region (Figures 2A and 2B). We found some activated canonical Wnt signaling in the apical papilla and dental sac after rapid tooth growth and elongation on PN7 ( Figures 2C-C 2 ). Then, X-gal staining intensity reduced in the crown but increased in the apical area and dental sac ( Figures 2C-2F). When M1 was ready to erupt on PN14, the intensity of X-gal staining was weak at the zones below the cusps ( Figure 2F  crown and apical papilla, and the activity was barely detectable in the pulp (Figures 2G-2J).
We chose several typical postnatal tooth development stages (PN3, PN7, and PN10) to assess the histomorphology. Figure 3A showed that we primarily detected BMP/Smad-dependent signaling in the dental epithelium on PN3. Then, BMP signaling activity moved downward to the cervical loop, confining to the HERS and apical papilla. On PN7, we observed BRE activity in the odontoblasts in the crown pulp and blood vessels ( Figure 3B). Developing apical complex (DAC) comprises the apical papilla, dental sac, and HERS, regarded as inseparable integrity.
On PN10, the level of BMP/Smad signaling activity was higher in the odontoblasts of the dental pulp, vascular tissues, and DAC, which was associated with significant root elongation and advanced tooth mineralization ( Figure 3C). We observed Wnt/βcatenin signaling activity in a different manner. On PN3, we detected a high β-galactosidase expression level in the mesenchymal cells under the cusps where the odontoblast or pre-odontoblast existed ( Figure   3D). We associated the expression with the tip of the molar cusps but excluded from the ameloblasts.
With continuing root development, the number of X-gal-stained cells beneath the cusps decreased. On   A, B, and C, ×40) and Wnt/β-catenin signaling (D, E, and F, ×40) in M1 is shown. Panels A1, B1, C1, D1, E1, and F1 are high-magnification views (×100) of the red rectangular block as shown in panels A, B, C, D, E, and F, respectively. Panels A2, B2, C2, D2, E2, and F2 are high-magnification views (×100) of the blue rectangular block as shown in panels A, B, C, D, E, and F, respectively. On PN3, BMP/Smad-dependent signaling was primarily detected in the dental epithelium (A, A1, and A2); whereas active Wnt/β-catenin signaling was mainly expressed in the odontoblasts or pre-odontoblasts below the cusps and in the dental epithelium on the cusps; some expression was found in the apical region (D, D1, and D2). On PN7, BRE activity was detected in the pre-odontoblasts/odontoblasts in the crown pulp, blood vessels, and DAC (B, B1, and B2); Wnt/β-catenin activity was observed below the cusps, but none in the dental papilla and sac, nor in the vessels in the pulp (E, E1, and E2). On PN10, the same expression pattern was identified, associated with significant root elongation and advanced tooth mineralization (C, C1, and C2). Meanwhile, in BAT-gal mice, the number of X-gal-positive cells beneath the cusps decreased on PN10 (F, F1, and F2). (EE, enamel epithelium; OE, oral epithelium; DS, dental sac; CS, cusp; C, crown; R, root; DP, dental papilla; AP, apical papilla; P, pulp; CL, cervical loop; E, enamel; D, dentin; OD, odontoblast; V, vessel; HERS, Hertwig's epithelial root sheath) J Appl Oral Sci. 2021;29:e20210281 7/9 root growth certified that postnatal epithelial BMP signaling was dispensable in the regulation of root development. 22 Similar to many other developmental processes, epithelial−mesenchymal interactions were previously demonstrated to be essential for tooth root development. 4 Here, we showed that activated BMP signaling (positive X-gal staining) was specifically present in dental follicles and HERS at the early postnatal stage; then, the HERS became perforated, contributing to cementum regeneration and root formation (Figures 1 and 3). When M1 root formation was almost completed, BMP/Smad activity appeared restricted in the apical part of the root and in the pulp.
During molar root formation, BMP signaling seemed to influence odontogenesis in the light of dynamic spatial and temporal sites of the canonical BMP signaling activity. In developing M1, differentiated odontoblasts are parallel to the basement membrane after birth and then ready to secrete dentin matrix.
Next, the rapid tooth formation and root growth originate at P6. 23 Figure 3 showed that activated BMP signaling (positive X-gal staining) was localized more coronally, correlating the BMP signaling with dental mesenchyme cells committed into odontoblasts. probably generated primary odontoblasts that were responsible for root dentin production. 26 The results suggested that active BMP signaling was essential for the apical growth of the molar root.
Meanwhile, Wnt/β-catenin signaling is also crucial for root odontogenesis, as evidenced by the conditional inactivation of β-catenin in immature odontoblasts, which induces molars without roots. 13 Moreover, we observed the phenotype of ablation of Wnt ligands or overexpression of the inhibitor Dkk-1 in odontoblasts in the mandibular molars, with aberrantly short roots and dentin defects. 27,28 Thus, Wnt signaling in root formation needs to be tightly controlled. 4,15 In this study, we also observed regions with active In mouse M1, rapid tooth growth and root elongation begin on PN6. 23 During the transition/root stage of root formation in mice, we detected highly active Wnt/β-catenin signaling in odontoblast-lineage cells, HERS cells, and periodontal ligament cells. 31 We observed activated Wnt/β-catenin signals in the apical papilla and dental sac from PN7. Subsequently, we noted a decline of signaling activity in these cells when they received terminal differentiation from PN15 onward and eventually disappeared the signaling in M1. Axin2, known as a component of the β-catenin degradation complex, was also tightly linked with the developing roots. We mainly localized Axin2 expression surrounding the root sheath and dental papilla on PN10. 11 However, unlike the high Axin2 levels in the dental papilla/pulp (as this part proceeded to develop and differentiate), the Axin2 expression was

Conclusions
In short, the results of this study showed that BMP and Wnt signaling activities exhibited different and dynamic patterns during mouse M1 root development.
The position of activation of the BMP/Smad signal pathway was located more coronally in the early stage, which then gradually expanded as root elongation proceeded and was associated with blood vessels in the pulp and DAC tissues in the later stage. However, Wnt/β-catenin signaling was highly concentrated in the mesenchyme below the cusps in the early stage, gradually expanded to regions around the root at the transition/root stage, and then disappeared nearly the later stage. These findings emphasized the importance of spatial and temporal epithelial-mesenchymal signaling, such as BMP and Wnt signaling pathways, for postnatal dentinogenesis, as well as provided some clues for future studies to precisely explore the cellular and molecular mechanisms that regulate tooth root development.

Conflicts of Interest statement
The authors declare no competing or financial interests.