Tooth development begins with two different tissue components of an enamel organ and dental papilla. From the morphology of the enamel organ, the developmental processes of tooth germs can be subdivided into bud, cap, bell, and crown stages, which are followed by root formation stages with tooth eruption. The inner enamel epithelia in the enamel organ and cells in the dental papilla further cytodifferentiate into enamel-forming ameloblasts and dentin-forming odontoblasts, respectively. After the dentin matrices are secreted by functionally differentiated odontoblasts, the inner enamel epithelium becomes elongated and functionally polarized to become ameloblasts.
The expression of biological molecules is regulated strictly during the histogenesis of an organ. The tooth is not the exception and studies have been directed to identify the signaling molecules involved in its development (Pispa and Thesleff,2003). Several molecules including epidermal growth factor, TGF-beta, fibroblast growth factor, platelet-derived growth factor, Notch, dentin matrix protein-4, and endothelin receptors have been reported (Tanikawa and Bawden,1994; Harada et al.,1999; Thesleff and Mikkola,2002; Hao et al.,2007; Klopcic et al.,2007; Neuhaus and Byers,2007). However, most studies examined the early tooth developmental events from initiation to the early bell stage, which precedes the formation of dental hard tissue. Accordingly, there are relatively few reports on the molecular events for terminal cytodifferentiation and maturation of dental hard tissues as well as the preparation of an eruption pathway (Tompkins,2006). Moreover, in vitro tooth culture systems have been used in most studies on the differentiation of dental hard tissue cells and therefore, cannot represent precisely the events that take place in vivo (Unda et al.,2000).
Considering that molecules are expressed differentially between the cap stage molar germs (before the differentiation of odontogenic cells and eruptive tooth movement) and root formation stage molar germs (after the differentiation and eruptive movement), the aim of this study was to identify the signaling molecules involved in dental hard tissue formation and subsequent eruptive movement (Kim et al.,2008b,c,2009). In this study, cyclophilin A (Cyp-A) was expressed differentially in molar germs at the cap and root formation stages. The secreted form of Cyp-A is a potent chemoattractant for monocytes and macrophages (Pan et al.,2008; Payeli et al.,2008; Damsker,2009), which become fused to form osteoclasts that are essential for preparing an eruptive pathway of the tooth germ. Moreover, Cyp-A is internalized by binding to a cell surface receptor, known as an extracellular matrix metalloproteinase inducer (EMMPRIN; Yurchenko et al.,2001). This receptor is involved in the expression of matrix metalloproteinase proteins (MMPs), which are essential for matrix maturation in dental hard tissue (Sameshima et al.,2000; Li et al.,2001; Xie et al.,2010). In this study, we have clarified the functional significance of CypA during dental hard tissue formation and tooth eruption by examining its spatiotemporal expression and comparing its expression pattern with EMMPRIN as well as MCP-1 and CSF-1, which are two important chemoattractants for osteoclastogenesis.
MATERIALS AND METHODS
Animals and Tissue Preparation
Rat pups (Sprague-Dawley) were housed in laboratory animal care-approved facilities. All procedures were carried out in accordance with the ethical standards formulated by the animal care and use committee in Chonnam National University. Portions of the maxilla containing the molar germs were isolated, immersion-fixed overnight in a 4% paraformaldehyde solution and decalcified with ethylene diamine tetraacetic acid (pH 7.4). The samples were then dehydrated and embedded in paraffin. Five-μm-thick sagittal sections were cut serially for H-E and immunofluorescence staining.
RNA Preparation and DD-PCR
Twenty rat pups each at postnatal days 3, 6, and 9 were used. After carefully removing the covering gingiva and alveolar bone using a fine pincette, the maxillary second molar germs at postnatal days 3, 6, and 9 and the maxillary third molar germs at postnatal day 9 were extracted from their crypts along with their follicular tissues. The total RNA was extracted from the extracted molar germs using a Trizol® Reagent (Gibco BRL, MD) and DNase I-treated (Invitrogen, Carlsbad, CA).
A differential display-polymerase chain reaction (DD-PCR) was carried out using a GeneFishing™ DEG kit 101 (Seegene, CA). Combinations of dT-ACP2 with each of 20 annealing control primers (ACP) were made arbitrarily. Briefly, the reverse transcription reaction was initiated by incubating a mixture containing the total RNA and dT-ACP1 at 80°C for 3 min. The reactants were added to the mixture of dNTP, RNase inhibitor, and M-MLV reverse transcriptase, incubated at 42°C for 90 min, and then heated to 94°C for 2 min. DD-PCR was performed using an ABI 2400 thermocycler (Applied Biosystems/Perkin Elmer, CA). The products were resolved on 1.5% agarose gel.
Gel Elution, Subcloning, and Sequencing
The differentially expressed bands were eluted using a QIAquick gel extraction kit (Qiagen, CA). The eluted DNA was ligated with the pGEM-T® Easy Vector using T4 DNA ligase (Promega, WI) at 16°C overnight. The ligated vector was then transformed into DH5α competent cells, followed by inoculation in an ampicillin-containing LB broth. The positive insert on the agarose gel was confirmed using an E.Z.N.A plasmid miniprep kit (Omega Bio-tek, GA) and subsequent enzyme cutting with EcoR1. Finally, the insert was sequenced using a T7 promoter primer.
The Cyp-A primers were custom-designed to confirm the DD-PCR result. The levels of monocyte chemoattractant protein-1 (MCP-1), colony stimulating factor-1 (CSF-1), and EMMPRIN mRNAs were also measured. Table 1 lists the sequences and expected sizes. The PCR cycles were carried out in a GeneAmp PCR system 2400 (Applied Biosystems/Perkin Elmer, CA) using the following profile: denaturation for 1 min at 95°C, annealing for 1 min and an extension step at 72°C for 1 min. The PCR products were resolved on 1.2% agarose gel, and their size was checked using a 1 kb DNA ladder (Gibco BRL, MD). Either GAPDH or β-actin was used as a reference gene. For the negative control, DNA was omitted in the PCR reaction and the results were confirmed from the agarose gel. RT-PCR was also repeated to verify the products in the gel.
Table 1. Sequences of the oligonucleotide primers for RT-PCR
Amplicon Size (bp)
GenBank accession No.
5′ AGACAAAGTTCCAAAGACAG 3′
5′ GAGAGCAGAGATTACAGGG 3′
5′ ATGCAGGTCTCTGTCACGCTTCTG 3′
5′ ACCCATTCATCTCTATACAT 3′
5′ CGGGCATCATCCTAGTCTTGCTGA CTGTT 3′
5′ AAATAGTGGCAGTATGTGGGGGGGCATCCT 3′
5′ ACAGCAGTGGCATTGACATC 3′
5′ TGTTCCGATTTCTTTCCCAC 3′
5′ GAATCCTGTGGCATCCATGA 3′
5′ TCAGCAATGCCTGGGTACAT 3′
5′ CCATGGAGAAGGCTGGGG 3′
5′ CAAAGTTGTCATGGATGACC 3′
The differential expression of Cyp-A, EMMPRIN, CSF-1, and MCP-1 was also determined by real-time RT-PCR. cDNA was synthesized from the total RNA of the second molar germs and third molar germs using a SuperScript™ First-strand Synthesis System for RT-PCR (Invitrogen™ Life Technologies, CA). Equal amounts of cDNA were used for real-time amplification of the target genes using a Rotor-Gene RG-3000 (Corbett Research, Morklake, Australia). Amplified cDNA was detected using the SYBR Green PCR Master Mix Reagent kit (Qiagen, CA). The PCR conditions were as follows: incubation for 10 min at 95°C, followed by 45 cycles of 10 sec denaturation at 95°C, annealing for 15 sec at 60°C and 20 sec extension at 72°C. The target and reference genes were amplified in separate wells. The reaction mixture lacking cDNA was used as a negative control in each run. The data were analyzed using the Corbett Robotics Rotorgene software (Rotorgene 6 version 6.1, Build 90 software). β-actin was used as an internal control. Melting curve analyses were performed on each primer set to verify single products of each reaction. Melting curve data are included as a Supporting Information. To ensure experimental accuracy, real-time RT-PCR assays were conducted in triplicate for each sample. The mean fold changes in expression in the maxillary second molar germs compared to the third molar germs were calculated using the 2−ΔΔCt method and the range of the fold changes was calculated from the standard error (SE) of the ΔΔCt values.
Immunofluorescence staining was carried out using a TSA™ kit (Invitrogen, CA). Briefly, after blocking the endogenous peroxidase with 1% H2O2, the deparaffinized sections were reacted overnight with purified rabbit polyclonal anti-Cyp-A (Calbiochem, San Diego, CA) and purified goat monoclonal anti-EMMPRIN (Santa Cruz biotechnology), respectively, followed by a reaction with the horseradish peroxidase-conjugated secondary antibody. The sections were incubated in a Tyramide working solution and counterstained with propidium iodide for nuclear staining. The reactants were observed by LSM confocal microscopy (Carl Zeiss, Germany). The primary antibodies were substituted with normal serum for the negative control.
Real-time RT-PCR data were analyzed to determine the significance of differences in multiple comparisons of gene expression by using analysis of variance with Student t test. Values were expressed as mean ± SE. P values of <0.05 were considered significant.
At postnatal day 9, the third molar germs were at the cap or early bell stages of development. In contrast, on the same day the second molar germs were at the root formation stage, which was characterized morphologically by the completion of a crown outline and the presence of a Hertwig epithelial root sheath, and functionally through osteoclastic bone resorption for the preparation of an eruptive pathway. At postnatal days 3 and 6, the second molar germs were at the bell and crown stages, respectively. At all stages of tooth development, the molar germs were confined within the connective tissue of the dental follicle and bony crypts outside (Fig. 1). The occlusal region of the cap stage molar germs was overlaid by relatively thick alveolar bone. The surface of the developing bone was smooth and covered by many bone-forming osteoblasts, but few multinucleated cells were present (Fig. 1a1).
In contrast to the cap stage, many multinucleated cells were observed at the surface of the alveolar bone overlaying the occlusal region of the molar germs of the root formation stage (Fig. 1a2). The alveolar bone was thin and irregular, indicative of osteoclastic bone resorption having occurred. Many blood vessels and mononuclear cells were observed in the follicular tissue.
DD-PCR and Identification of Differentially Expressed Genes
When the specific primer ACP4 was used, a differentially expressed band of ∼200 bp was observed at the root formation stage only (Fig. 2). The band in Fig. 2 was subcloned and transformed into DH5α competent cells. The DNA was purified, sequenced, and identified as a part (213 bp) of Cyp-A (GenBank accession number XM_344509.1).
Cyp-A mRNA During Tooth Development
RT-PCR was performed using the specific primers for Cyp-A to compare the expression levels between the cap and root formation stage molar germs. Amplicons of the expected size (521 bp) were generated from both stage germs. However, the expression level appeared to be much higher at the root formation stage than at the cap stage. Real-time RT-PCR showed that the expression level was approximately four times higher at the root formation stage (Fig. 3a; P < 0.05). Cyp-A mRNA expression from the second molar germs was also analyzed at postnatal days 3 (bell stage), 6 (crown stage), and 9 (root formation stage). Although Cyp-A mRNA expression was up-regulated in a time-dependent manner, the level was significantly higher at the root formation stage (Fig. 3b; P < 0.05).
Downstream Molecules mRNA During Tooth Development
The mRNA level of chemotactic MCP-1 and CSF-1 was measured by RT-PCR to compare the Cyp-A expression pattern with the well-known chemotactic factors for osteoclast formation. The expression levels of both molecules were significantly higher at the root formation stage (the second molar germs) than at the cap stage (the third molar germs; P < 0.05). In addition, the real-time RT-PCR results showed that the fold changes in MCP-1 and CSF-1 were ∼4 and 12, respectively (Fig. 4a). The levels of CSF-1 and MCP-1 expression of the second molar germs increased significantly in a time-dependent manner from the early bell stage (postnatal day 3), the crown stage (postnatal day 6), and through the root formation stage (postnatal day 9; Fig. 4b; P < 0.05). Overall, these MCP-1 and CSF-1 expression patterns in the developing tooth germs occurred in parallel to those of Cyp-A.
The expression of EMMPRIN was also measured during tooth development by RT-PCR to compare the expression of Cyp-A with that of EMMPRIN, which is known to be its receptor. Its expression patterns were similar to those of Cyp-A. The level of EMMPRIN mRNA appeared to be significantly higher at the root formation stage (second molar germs) than at the cap stage (third molar germs). The real-time RT-PCR result showed that the fold change was approximately seven times (Fig. 5a; P < 0.05). In addition, the level from the second molar germs at postnatal day 3 (bell stage), 6 (crown stage) and 9 (root formation stage) increased in a time-dependent manner (P < 0.05) and in parallel to those of Cyp-A (Fig. 5b).
Strong immunoreactivity against Cyp-A was observed in the follicular tissues and odontogenic cells of the root formation stage molar germs at postnatal day 9 (Fig. 6a). In particular, the immunoreactivity against Cyp-A was region-specific. The reactivity was detected mainly in the cells and intercellular matrix of the dental follicle overlaying the occlusal region of the molar germs of the stage rather than the proximal and cervical region. However, the immunoreactivity was weak in the alveolar bone matrix and the lamina propria between the surface epithelium and developing alveolar bone (Fig. 6a1,a2). Strong reactivity against Cyp-A was also observed in the cytoplasm of both columnar enamel-forming ameloblasts and dentin-forming odontoblasts. However, both preameloblasts and preodontoblasts seldom showed immunoreactivity (Fig. 6a2). In contrast to the root stage molar germs, immunoreactivity against Cyp-A was weak in the inner enamel epithelium and dental papilla of the cap stage molar germs. Although reactivity was noted in the follicular tissues, it was weak and generalized rather than being region-specific (Fig. 6b).
To compare the localization of EMMPRIN with Cyp-A, immunofluorescence staining was performed from the immediately adjacent section to the section stained for Cyp-A. Similar to Cyp-A, strong immunoreactivity against EMMPRIN was observed in the follicular cells and intercellular matrix of the dental follicle in the root formation stage molar germs at postnatal day 9. The strong reactivity was observed in the occlusal region rather than in the proximal and cervical regions (Fig. 7). Strong reactivity against EMMPRIN was also observed in the cytoplasm of both ameloblasts and odontoblasts. However, preameloblasts and preodontoblasts seldom showed immunoreactivity (Fig. 7a). Similar to Cyp-A, there was virtually no immunoreactivity against EMMPRIN in the inner enamel epithelium and dental papilla of the cap stage molar germs. The reactivity against EMMPRIN was weak in the follicular tissues and was not specific to the occlusal region (Fig. 7b). The negative control did not show any staining (Fig. 8).
One means of communication between cells during tooth development is paracrine-mediated signaling molecules. Accordingly, many studies have focused on identifying these molecules and their receptors. Tooth germs at the root formation stage differ from those at the cap stage in that they have already begun eruptive movement and are in the middle of dental hard tissue formation. From this rationale, this study was carried out primarily to identify unknown molecules in tooth development by comparing gene expression between two different stage tooth germs using DD-PCR. Using the same cap and root formation stage models of the third and second molar germs, we recently reported that myelin basic protein (Kim et al.,2008b) and leukocyte antigen receptor tyrosine phosphatase (Kim et al.,2008c) are expressed differentially during tooth development. In this study, Cyp-A, as a ligand for an EMMPRIN, was found to be a true positive. Cyp-A was originally discovered as an intracellular ligand for cyclosporin A, an immunosuppressant (Handschumacher et al.,1984), but other functions have also been suggested. The secreted form of Cyp-A is a potent chemoattractant for macrophage colony-stimulating factor (MCSF)-dependant macrophages (Comalada et al.,2003; Sànchez-Tilló et al.,2006), monocytes (Galat,1993; Payeli et al.,2008), neutrophils (Sherry et al.,1992), eosinophils (Xu et al.,1992), and T cells (Kasinrerk,1999). This molecule is involved in the adhesion of monocytes/macrophages to the extracellular matrix (Yang et al.,2008) and the migration of murine bone marrow cells (Khromykh et al.,2007). In addition, the secreted form modulates osteoblastic activity (Andersen et al.,2003) as well as the functions of endothelial cells via paracrine and autocrine modes (Kim et al.,2004).
Eruptive movement of the tooth germs, which is regulated delicately in timing and sequence, cannot be the exception (Cerri et al.,2010; Liu et al.,2010). The eruptive movement of tooth germs begins by the formation of the root (at the root formation stage) and accompanies the localized resorption of the alveolar bone overlaying the occlusal region of tooth germs. This study showed that osteoclasts at the alveolar bone surface overlaying the occlusal region of the molar germs appeared to increase at the root formation stage.
In addition, immunofluorescence results showed that Cyp-A and EMMPRIN as its receptor were expressed strongly in follicular cells and the intercellular matrix overlaying the occlusal region of the root formation stage molars. The Cyp-A mRNA expression level was significantly higher at the root formation stage than at the cap stage. Moreover, the expression level increased in a stage-dependent manner up to the root formation stage. These patterns of mRNA expression were in parallel to those of MCP-1 and CSF-1, which cause monocytes to fuse to form osteoclasts. Considering the roles of Cyp-A in the chemoattraction of mononuclear cells, this study implicates extracellular Cyp-A in the bone matrix changes during eruptive movement of the tooth germ. Cyp-A secreted from follicular cells might induce the chemoattraction of monocytes to the follicular tissue and adjacent alveolar bone in a paracrine mode. In turn, the attracted monocytes may fuse to form osteoclasts, which will be involved in alveolar bone resorption during the eruptive process of tooth germs.
In the immunofluorescence investigation, Cyp-A was expressed in not only the dental follicular tissues but also in fully differentiated odontogenic cells, suggesting that Cyp-A may play dual roles. An attempt was made to microdissect the follicular tissues to define further its roles, but it was not technically possible to carry this out for the third molar germs because they were very small and the demarcation between the dental papilla and dental follicle was obscure. Although Cyp-A was expressed at the cap stage, its mRNA level was much lower than that at the root formation stage. In addition, the immunoreactivity at the cap stage was quite weak and observed only in the entire region of follicular tissues rather than the localized region overlaying the occlusal surface of the germs. These findings suggest that Cyp-A at the cap stage might be related to the preeruptive movement to accommodate the changing shapes of the molar germ rather than the eruptive movement toward the occlusal direction.
In addition to the destruction of the alveolar bone by osteoclasts, resolution of the organic matrix of the bone and surrounding connective tissue is essential for the formation of an eruption pathway. The resorption process of organic matrix is undertaken by MMPs (Kim et al.,2008a). Cyp-A is internalized by binding to the glycosylated cell surface receptor known as an EMMPRIN or basigin, neurothelin, or CD147 (Yurchenko et al.,2001). This receptor is involved in MMPs expression (Sameshima et al.,2000; Li et al.,2001) and regulates MMP production via the MAPK pathway (Lai et al.,2003). In rheumatoid arthritis, Cyp-A up-regulates MMP-9 expression via the EMMPRIN signaling pathway through direct binding to EMMPRIN (Yang et al.2008). This sequential event can also be postulated in tooth development. Therefore, further studies will be needed to determine the combined actions of Cyp-A, EMMPRIN as its receptor, and MMPs as the final effector molecules in the formation of an eruption pathway.
However, organic matrices of the developing dentin and enamel need to be resorbed until they become mature calcified matrices. In this study, Cyp-A was barely detected in either preodontoblasts and preameloblasts or dental papilla cells and inner enamel epithelium; however, it was strongly expressed in fully differentiated ameloblasts and odontoblasts, which are involved in compositional changes of the enamel and dentin matrices, respectively. In addition, the level of EMMPRIN expression was higher at the hard tissue formation stage than at the cap or early bell stage. Furthermore, the localization of EMMPRIN was in parallel to that of Cyp-A. The involvement of EMMPRIN in tooth development, possibly by regulating the expression of MMP genes, was recently evidenced by EMMPRIN siRNA-treated explants of tooth germs (Xie et al.,2010). Therefore, it is believed that the binding of Cyp-A to EMMPRIN, the cell surface receptor, may up-regulate the expression of MMPs that participate in the processing of secreted matrix proteins prior to mineralization. In addition, the low level of Cyp-A and EMMPRIN at the cap stage was due to the limited expression in follicular tissues, not in the dental hard tissue forming cells (Figs. 6b, 7b). Therefore, expression at the cap stage might be implicated in the limited remodeling of follicular tissues rather than the mineralization of dental hard tissue substances.
Overall, this study showed that Cyp-A, a potent chemoattractant as well as a ligand for EMMPRIN, was expressed differentially during tooth development, including tooth eruption, and was regulated in a spatiotemporal manner (Figs. 3, 6). This suggests that Cyp-A might be implicated in matrix changes for the preparation of an eruptive pathway as well as dental hard tissue formation. However, because this study focused primarily on identifying candidate molecules involved in these changes, Cyp-A and EMMPRIN expression in the later stages of root formation and tooth eruption were not examined. Therefore, further studies will be required.