|Home | About | Journals | Submit | Contact Us | Français|
Late tooth morphogenesis is characterized by a series of events that determine crown morphogenesis and the histodifferentiation of epithelial cells into enamel-secreting ameloblasts and of mesenchymal cells into dentin-secreting odontoblasts. Functional ameloblasts are tall, columnar, polarized cells that synthesize and secrete a number of enamel-specific proteins. After depositing the full thickness of enamel matrix, ameloblasts shrink in size and regulate enamel maturation. Amelogenesis imperfecta (AI) is a heterogeneous group of inherited defects in enamel formation. Clinically, AI presents as a spectrum of enamel malformations that are categorized as hypoplastic, hypocalcified, or hypomaturation types, based upon the thickness and hardness of the enamel. The different types of AI are inherited, either as X-linked, autosomal dominant or autosomal recessive traits. Recently, several gene mutations have identified to cause the subtypes of AI. How these genes, however, coordinate their function to control amelogenesis is not understood.
In this review, we discuss the role of genes that play definitive role on the determination of ameloblast cell fate and life cycle, based on studies in transgenic animals.
The development of teeth is a complex process that involves sequential and reciprocal interactions between dental epithelium and mesenchyme. It is characterized by events that determine the histodifferentiation of mesenchymal cells into dentin-secreting odontoblasts and of epithelial cells into enamel-secreting ameloblasts (Kollar and Lumsden, 1979; Thesleff and Nieminen, 2005).
The differentiation process of epithelial cells into functional ameloblasts consists of several morphologic and functional changes that occur in time and involve considerable growth, elongation of the cytoplasm, polarization and the appearance of processes that secrete matrix (Fig. 1). These morphologic changes are known as: (i) the inductive stage (pre-ameloblasts), where the cells of inner enamel epithelium begin to differentiate into ameloblasts, elongate, their nuclei shift distally (away from the dental papilla), and their cytoplasm becomes filled with organelles needed for synthesis and secretion of enamel proteins; (ii) the initial-secretory stage, where the proximal end of the newly formed ameloblasts (near the dental papilla) is flat and the matrix secreted is called rodless enamel matrix; (iii) the secretory stage, where the ameloblasts lengthen, polarize, form conical projections, known as Tomes' processes, and deposit enamel in the form of rods and prism; (iv) the maturation stage where the ameloblasts function to resorb much of the water and organic matrix from enamel in order to provide space for the growing enamel crystals (Nanci A and Ten Cate, 2007; Fig. 1).
Animal and human studies that employ the tools of contemporary molecular genetics have identified a number of genes that act at specific stages of the ameloblast life cycle and regulate its patterning and differentiation process. The purpose of this review is to discuss recent findings regarding genes that definitively control the ameloblast life cycle and function, based on animal studies.
The presumptive ameloblast begins its life cycle as a proliferative cell, low columnar in shape, separated from adjacent mesenchyme by a basement membrane. Once stimulated, the differentiating ameloblast ceases proliferation and begins to grow in height. Tissue recombination experiments using dental and non-dental tissues have shown that ameloblast cytodifferentiation is regulated by a series of reciprocal interactions between epithelium and mesenchyme (Kollar and Baird, 1970; Thesleff and Hurmerinta, 1981). Signals from the dental epithelium first induce the differentiation of underlying mesenchymal cells into odontoblasts. The odontoblasts deposit dentin matrix and signal back to the epithelium, inducing differentiation of epithelial cells into functional ameloblasts (Karcher-Djuricic et al., 1985).
The molecules mediating this induction are members of the transforming growth factor (TGFβ) superfamily. BMP2 and TGFβ1 are secreted by odontoblasts and induce ameloblast differentiation in vitro (Coin et al., 1999). Recent studies show that BMP4, another TGFβ signal, is secreted by the odontoblasts to induce ameloblast differentiation in vivo. It is shown that BMP4 releasing beads induce strong ameloblastin expression, an ameloblast-specific gene, in cultured teeth and that BMP4 inhibitor, noggin, dramatically inhibits induction of ameloblast differentiation (Wang et al., 2004).
Although, these studies indicate the importance of odontoblasts in the induction of epithelium into ameloblast cell fate, an epithelial-dependent, cell autonomous control is also needed for ameloblasts to fully differentiate, mature and deposit enamel matrix. Pre-secretory, secretory and mature ameloblasts express several (i) secreted proteins, such as ameloblastin, amelogenin, enamelin, tuftelin, dentin sialoprotein, apin, amelotin (ii) enzymes such as kallikrein 4 and enamel proteinases, such as matrix metalloproteinase 20, (iii) signaling molecules like BMPs, TGFβ1, SHH and WNTs and (iv) transcription factors like Msx2, Sp3, Sp6 and Dlxs (Zeichner-David et al., 1995; Krebsbach et al., 1996; Aberg et al., 1997; Begue-Kirn et al., 1998; Dassule et al., 2000; Hu et al., 2002; Gritli-Linde, 2002; Mustonen, et al., 2004; Wang et al., 2004; Hart et al., 2004; Bei et al., 2004; Wright, 2006; Moffatt et al., 2006a, 2006b; Wang et al., 2007; Dubrowolski et al., 2008; Lezot et al., 2008; http//:bite-it.helsinki.fi). Studies using transgenic animals provide functional data showing that disruption of the ameloblast signaling and its mediators result in aberrations of ameloblast differentiation and enamel deposition (Table 1). Below, we discuss these studies.
Shh is strongly expressed in pre-ameloblasts and secretory ameloblasts (Dassule et al., 2000; Gritli-Linde, et al., 2002). Conditional inactivation of Shh in the epithelium results in defective ameloblasts that exhibit little elongation, organization, polarization and some enamel matrix deposition (Dassule et al., 2000). When Smoothened a transmembrane protein essential for all Hh signaling, is conditionally removed from the dental epithelium, results in a complete failure of epithelial cells to proliferate, to grow, to polarize and to become secretory ameloblasts (Gritli-Linde, et al., 2002). These data suggest that although the dental mesenchyme is necessary for ameloblast fate induction, it is not sufficient, that cell-autonomous epithelial-epithelial interactions play an important role and that Shh may be an endogenous epithelial factor regulating ameloblast cytodifferentiation.
Shh, although important, is not the only cell-autonomous factor that determines ameloblast cytodifferentiaon and/or function. Several members of the TGFβ superfamily are also expressed in ameloblasts during the pre-secretory and secretory stages. Overexpression of TGFβ1 in the dental epithelium under the Dspp promoter results in hypoplastic enamel and detached pre-secretory ameloblasts from dentin (Haruyama et al., 2006). In contrast to TGFβ1 transgenic mice where some enamel matrix is deposited, overexpression of follistatin, a BMP inhibitor, in the epithelium abrogates ameloblast differentiation. The K14-follistatin mice lack enamel, the ameloblasts fail to form and do not express enamel specific markers like ameloblastin, MMP20 and DSPP (Wang et al., 2004). The critical role of follistatin in ameloblast differentiation is further exemplified by the fact that in follistatin knockout mice, functional ameloblasts differentiated on the normally enamel-free surface (Wang et al., 2004). The latter indicates that follistatin is essential for enamel-free area formation by preventing ameloblast differentiation (Wang et al., 2004). Experiments on cultured tooth explants suggest that the mechanism by which follistatin prevents ameloblast differentiation is by inhibiting the ameloblast-inducing activity of BMP4 from the underlying odontoblasts. They also show that follistatin expression is induced by activin from the surrounding dental follicle. Thus, follistatin controls ameloblast differentiation in a cell-autonomous manner by integrating the opposing effects of two non cell-autonomous signals that of activin and BMP4 from dental follicle and odontoblasts, respectively (Wang et al., 2004).
Lack of enamel formation and loss of ameloblast differentiation is also observed in mice overexpressing other signaling factors, such as Wnt3 or Ectodysplasin (Eda-A1) under the K14 promoter (Millar et al., 2003; Mustonen et al., 2004). Ectopic expression of Wnt3 in the tooth epithelium causes progressive loss of ameloblasts from postnatal lower incisor teeth while overexpression of Eda-A1 causes disruption of ameloblast differentiation and enamel formation. The striking similarity of phenotypes observed in transgenic mice over-expressing Follistatin, Wnt and Eda-A1 suggests that these genes may share or reside in the same pathway controling ameloblast differentiation.
At the secretory stage, the ameloblasts are characterized by strong cell-cell adhesion formation and different molecules of the cell-cell adhesion apparatus, including the E-cadherin, catenins, the integrins ανβ5, α6β4, connexin 43 and laminins are strongly expressed by pre-ameloblasts and secretory ameloblasts (Pasqualini et al., 1993; Meyer et al., 1995; Salmivirta et al., 1996; Garrod et al., 1996; Green et al., 1996; Fausser et al., 1998). Mouse and human mutations provide further evidence for the role of some cell-cell adhesion molecules on ameloblast differentiation and function.
Laminin 5 isoforms, for example, are highly expressed in secretory ameloblasts (Yoshiba et al., 1998). Mutations in human LAMININ-5 isoforms result in variants of Herlitz junctional epidermolysis bullosa (EB) or dystrophic EB, groups of recessive inherited disorders characterized by dermal-epidermal separation (Aberdam et al., 1994; Kivirikko et al., 1995, McGrath et al., 1995; Christiano et al., 1997). In some cases of EB caused by mutations in LAMININ-5, its receptor Integrin α6β4 or type VII collagen, enamel hypoplasia is observed (Vidal et al., 1995; Christiano et al., 1996; 1997). When laminin 5a3 is mutated in mice, among other epithelial abnormalities, ameloblast differentiation is affected resulting in reduced enamel deposition. The ameloblast reach the secretory stage and secrete little enamel matrix (Ryan et al., 1999). A potential mechanism by which Laminin 5 might control ameloblast differentiation is by interaction with its receptor, integrin β4. It is shown that this interaction is important for stabilizing the architecture of epithelial cells by means of their hemidesmosome assembly (Baker et al., 1996). Absence or reduction in laminin5α3 expression, therefore, would be expected to result in profound alterations in ameloblast structure and function, since positional information and strong cell-cell adhesions become critical prerequisites for the structural integrity of ameloblasts (Meyer et al., 1995). Consistent with this, in the Msx2 mutant teeth where laminin 5a3 is absent in ameloblasts, profound changes in the structural integrity of ameloblasts and in the enamel deposition process is observed (Bei et al., 2004 and Fig. 2).
Mutations in the human connexin 43 gene, a member of gap junction-specific family of genes, the connexins, result in oculodentodigital dysplasia (ODDD). ODDD is a dominant negative inherited disorder affecting different organs including the teeth that exhibit hypoplastic enamel. When the human Cx43G138R point mutation is inserted in to the mouse Cx43 gene, the transgenic mice exhibit all ODDD phenotypes and hypoplastic enamel (Dobrowolski et al., 2008). Additional studies suggest that this mutation is associated with gap junctional dysfunction and increased cellular ATP release in cardiomyocytes that result in abnormal function and arrhythmia, providing thus a potential mechanism by which Cx43 may operate in ameloblasts, as well.
Several human mutations in ameloblast-specific genes like amelogenin, ameloblastin, enamelin; proteolytic enzymes enamelysin (MMP20), Kallikrein 4 and the non-extracellular matrix protein FAM83H have identified to cause the different subtypes of Amelogenesis Imperfecta (AI) indicating their essential role for the correct deposition and maturation of dental enamel. (Aldred et al., 2002; Wright, 2006; Lee et al., 2008; Kim et al., 2008). Amelogenesis imperfecta (AI) is a heterogeneous group of inherited defects in enamel formation. Clinically, AI presents as a spectrum of enamel malformations that are categorized as hypoplastic, hypocalcified, or hypomaturation types, based primarily upon the thickness and hardness of the enamel (Witkop, 1989). The different types of AI reflect differences in the timing during amelogenesis, when the disruption occurs and are inherited, either as an X-linked, autosomal dominant or autosomal recessive traits.
Mouse mutations exist for four ameloblast-specific genes, the Amelx, Enamelin, Ameloblastin and Mmp20 genes. When Amelx and Mmp20 genes are deleted from the mouse genome result in the development of enamel defects (Gibson et al., 2001; Caterina et al., 2002). The amelogenin mutant incisors exhibit disorganized hypoplastic enamel containing a rough knobby surface, while the (MMP20, enamelysin) mutant enamel is hypoplastic with improperly processed amelogenin. Consistent with the mouse studies, 14 different AMELX human mutations are known to cause the enamel malformation disease, amelogenesis imperfecta (AI) (Lench et al., 1995; Collier et al., 1997; Hart et al., 2002; Kim et al., 2004; Wright et al., 2003). In addition, the human MMP20 is located to chromosome 11 and an autosomal-recessive form of AI was recently discovered in a family that had a mutation in the intron 6 splice acceptor (AG to TG) (Kim et al., 2005). The ameloblastin null mice develop severe enamel hypoplasia (Fukumoto et al., 2004). The dental epithelium initially differentiates into enamel-secreting ameloblasts. Subsequently, the cells are detached from the matrix, loose polarity, resume proliferation and, thus, their status is reversed from differentiated to undifferentiated one, suggesting that ameloblastin, a cell adhesion molecule, maintains the differentiation state of ameloblasts at the secretory stage, by binding to ameloblasts and by inhibiting their proliferation (Fukumoto et al., 2004, 2005). In contrast to ameloblastin null mice, the enamelin knock out/LacZ knock in mice exhibit a later defect in amelogenesis. The mice secrete enamel matrix but its organization and mineralization process are severely disturbed resulting in the formation of an irregular mineralized crust arising from the fusion of small mineralization foci in the deeper part of the accumulated enamel matrix (Hu et al., 2008).
Transcription factors are critical for early tooth development, but recent data shows that several of them are essential for late stages of tooth morphogenesis including amelogenesis. Specificity protein 3 (Sp3) is a ubiquitously expressed protein that belongs to the Sp-family of transcription factors. This family consists of several members, Sp1-6, that are characterized by their ability to bind G-rich elements, such as GC-boxes and related motifs found widely distributed among promoters and enhancers of several genes. Sp3 homozygous null mice exhibit a hypoplastic phenotype in both, dentine and enamel matrices (Bouwman et al., 2000). Sp6 null mice, also known as Epfn-/- mice, among different phenotypes exhibit enamel deficiency. In these mice rapid proliferation and differentiation of the inner dental epithelium are inhibited and, thus, the dental epithelium retains the progenitor phenotype (Nakamura et al., 2008). Transfection of Epfn promoted dental epithelial cell differentiation into ameloblasts and activated promoter activity of the enamel matrix ameloblastin gene, suggesting that Sp6 promotes ameloblast differentiation by controlling the rate of proliferation of inner enamel epithelium (Nakamura et al., 2008; Ruspita et al., 2008).
Mice lacking the homeobox gene Msx2 exhibit defects in amelogenesis. In the Msx2 knock out mice the ameloblasts reach the secretory stage of their differentiation process and sparse amounts of enamel matrix are deposited (Satokata et al., 2000; Bei et al., 2004; Fig. 2). In a case of amelogenesis imperfecta, cleft lip and palate and polycystic kidney disease sequence analysis of the human homolog of MSX2 gene identified a missense mutation of T447C that resulted in the conversion of methionine to threonine at 129, further indicating the important role of Msx2 during amelogenesis (Suda et al., 2006). Interestingly, Msx2 is selectively required for the expression of the extracellular matrix gene, Laminin 5 a 3, which, as mentioned above, play an essential role during ameloblast differentiation, suggesting that Laminin 5 a 3 and Msx2 genes may function within the same genetic pathway (Bei et al., 2004).
In addition, the similarity in the odontogenic phenotype between Msx2 and Sp3/Sp6 and their overlapping expression in ameloblasts raises the possibility that these transcription factors interact closely within a molecular cascade that operates in these cells, regulating amelogenesis. Consistent with this idea, Msx2 and Sp3 transcription factors interact in vitro and in vivo (Bei, unpublished). The tooth phenotype and the enamel defects in Msx2 mutants are also similar to those observed in transgenic mice expressing urokinase-type plasminogen activator in the enamel organ (Zhou et al., 1999). Of interest, in the urokinase transgenic mice, a reduction of laminin5 expression is observed in the basal basement membrane and in the apical pole of ameloblasts. Since the urokinase-type plasminogen activator gene is expressed, like Msx2, in both secretory ameloblasts and stratum intermedium cells, suggests that a potential genetic relationship between these two genes exist, as well.
Mutations in, both, mouse and humans reveal the involvement of several classes of genes in ameloblast life cycle and its function, a process that emerges as a model for elucidating the molecular regulatory cascades that operate in other developmental systems, as well. Most of these genes appear to function in the context of evolutionarily conserved genetic hierarchies. As such, the availability of model organisms for the systematic investigation of the conserved genetic and molecular cascades, that regulate enamel formation, could lead to treatments of human diseases affecting amelogenesis (amelogenesis imperfecta), could further our knowledge in the field of tooth regeneration and help us extrapolate our findings to other developmental systems that form via similar mechanisms.
This work was supported by grants from the NIH (RO3 DE 018415), Shiseido Inc. funds and Harvard Medical School (Milton Fund Award) to MB.