Expression of Axin2 Indicates a Role for Canonical Wnt Signaling in Development of the Crown and Root During Pre- and Postnatal Tooth Development

Transcription

1 DEVELOPMENTAL DYNAMICS 239: , 2010 SPECIAL ISSUE RESEARCH ARTICLE Expression of Axin2 Indicates a Role for Canonical Wnt Signaling in Development of the Crown and Root During Pre- and Postnatal Tooth Development Mahtab Lohi, Abigail S. Tucker, and Paul T. Sharpe* Previously two reporter mice, TOPgal and BATgal, have been used to uncover the spatial patterns of canonical Wnt activity up to the bell stage of tooth development. To further understand the function of this pathway, not only at the early developmental stages of odontogenesis but also in postnatal teeth, we have used Axin2-lacZ mice a direct reporter of canonical Wnt activity. As tooth development progresses, Axin2 expression becomes localized to the primary and secondary enamel knots, and the underlying mesenchyme. In postnatal teeth, Axin2 expression is observed in developing odontoblasts, in the dental pulp and concentrated around the developing roots. Expression is excluded from the ameloblasts and associated with the enamel-free zones at the tip of the molar cusps. This expression identifies new roles for Wnt signaling in defining the regions where enamel will form, and controlling root development at late stages of tooth development. Developmental Dynamics 239: , Wiley-Liss, Inc. Key words: tooth; Wnt; Axin2; cusp; root Accepted 25 June 2009 INTRODUCTION During murine embryogenesis teeth develop by a series of epithelial/mesenchymal interactions. On the eleventh day of embryonic development (E11), the first sign of tooth development is seen as thickenings of the oral epithelium. These thickenings grow, and the epithelium starts to invaginate into the underlying neural crestderived mesenchyme to form buds. At the bud stage (E13), the mesenchyme condenses around the invaginating epithelium. The cap stage (E14) follows when the epithelium extends further into the mesenchyme, wrapping itself around the condensing mesenchyme to form a cap. The continued growth of the tooth germ leads to the bell stage. A signaling center develops at the tip of the late bud, known as the enamel knot (Jernvall et al., 1994). Multicuspid molars go on to form secondary enamel knots that sit at the sites of the future cusps. In certain species, such as the mouse, enamelfree areas form at the tips of the cusps during postnatal molar development, that have been referred to as tertiary enamel knots (Gaunt, 1956; Luukko et al., 2003). Eventually the condensing mesenchyme becomes surrounded by the invaginating epithelium by the late-bell stage (E18). From the cap and bell stages, the epithelium goes through histodifferentiation where the epithelial cells are converted into morphologically distinct components and the shape of the crown is determined by folding morphogenesis. The epithelial structures are referred to as the enamel or dental organ, which is divided into the inner enamel epithelium, central stellate reticulum, stratum intermedium, and the outer enamel epithelium. The inner epithelium is responsible for providing the cells for the ameloblast cell lineage. The stellate reticulum makes up the central bulk of the early enamel organ and is thought to pro- Department of Craniofacial Development, GKT Dental Institute, Guy s Hospital, London, United Kingdom Grant sponsor: MRC. *Correspondence to: Paul Sharpe, Department of Craniofacial Development, GKT Dental Institute, Guy s Hospital, London SE1 9RT, United Kingdom. paul.sharpe@kcl.ac.uk DOI /dvdy Published online 3 August 2009 in Wiley InterScience ( Wiley-Liss, Inc.

2 AXIN2 AND WNT SIGNALING IN TOOTH DEVELOPMENT 161 vide nutrients and support for the amelobast layer. The outer enamel epithelial cells provide nutritional substances and oxygen to the ameloblasts and other enamel organ cells by associating with a capillary plexus. The late bell stage is characterized by the cytodifferentiation of the enamelforming ameloblasts and the dentineforming odontoblasts, derived from the dental papilla cells lying adjacent to the inner enamel epithelium. The inner epithelium induces neighboring cells in the dental papilla to differentiate into odontoblasts. The odontoblasts then produce a layer of predentine, which instructs the inner enamel epithelial cells to differentiate into ameloblasts and start secreting enamel matrix (Cox et al., 1992; Linde and Goldberg, 1993; Meikle, 2000; Kim and Simmer, 2007; Bei, 2008). Several Wnts have been shown to be expressed in the developing tooth. Sarkar and Sharpe (1999) carried out a comparative in situ hybridization analysis of six Wnt genes Wnts-3, -4, -5a, -6, -7b, and 10b, the Wnt receptor MFz6 and receptor agonist/antagonists MFrzb1 and Mfrp2 during tooth development. These were mapped to mouse embryonic heads between E11.5 and E15.5, which encompasses the earliest formation of the epithelial thickening to the early bell stage. To identify the spatial and temporal pattern of Wnt/ -catenin activity in tooth development Wnt reporter mice, such as TOPGAL or BAT-gal, which are transgenes that are randomly integrated into the mouse genome, have been utilized, in addition to with immunofluorescence for nuclear -catenin (Liu et al., 2008). These reporter lines are widely used to map Wnt responsiveness, but their expression has been shown to vary from each other, and from the localization of nuclear -catenin (Brugmann et al., 2007; Liu et al., 2008). In the tooth, Wnt activity has been observed in the dental epithelial placodes of the molars at E11.5 and E12.5 (Liu et al., 2008). Using nuclear localization of -catenin, however, Wnt activity was evident not only in the tooth bud epithelium but also the underlying mesenchyme. By E14.5, reporter gene expression was identified in the epithelial cells of the primary enamel knot, but nuclear -catenin was also observed in the mesenchymal cells underlying the enamel knot. This discrepancy could be due to several factors. Detection of nuclear -catenin may be more sensitive in the mesenchyme. The reporter gene constructs may not have been able to detect mesenchymal Wnt signaling, or cofactors which are needed for activating transcription may be missing from the mesenchyme. By the early bell stage, reporter gene expression was detected in the developing secondary enamel knots in the molars, overlapping with the previously reported expression of Wnt10b (Sarkar and Sharpe, 1999; Liu et al., 2008). There have been several studies identifying the importance of Wnt signaling in tooth development. The emphasis has been on the role of Wnt signaling in shaping the tooth, through its signaling by means of enamel knots, and on its role in the formation of supernumerary teeth (Jarvinen et al., 2006; Liu et al., 2008). Transgenic mice where -catenin expression has been stabilized in the epithelium of the oral cavity result in the formation of multiple teeth. From E13 onward, the tooth buds were irregular with small epithelial buds forming at their tips. As tooth development progressed, the epithelial structures extended further into the mesenchyme, expressing enamel knot markers and resulting in abnormal morphogenesis. When cultured under the kidneys of host mice, the molar tooth germs formed multiple teeth, compared with the three molars formed from controls (Jarvinen et al., 2006). Ectopic teeth have also been observed in epiprofin knockout mice, where Wnt signaling is up-regulated, as shown by the enhanced expression of the Wnt target gene Lef-1 (Nakamura et al., 2007), and in the Lef1 overexpressing mice (Zhou et al., 1995). Enhanced Wnt signaling can, therefore, lead to the formation of supernumerary teeth. In keeping with this important role of Wnt signaling in tooth development, loss of Wnt signaling leads to arrest of tooth development. If the Wnt antagonist Dickkopf (Dkk1) is overexpressed in the oral epithelium tooth development arrests at the placode/ early bud stage (Andl et al., 2002; Liu et al., 2008). Tooth arrest at the bud stage also occurs in the Lef-1 knockout mouse (Sasaki et al., 2005). The role of Wnt signaling in the primary and secondary enamel knots in molars has previously been analyzed using Dickkof (DKK) overexpression in epithelial cells to inhibit Wnt activity. When the inhibition is targeted to the bell stage of tooth development, signaling in the secondary enamel knots is disturbed and cusp formation is disrupted (Liu et al., 2008). Wnt signaling, therefore, appears crucial at multiple stages of tooth development. To investigate Wnt signaling further we have utilized the Axin2-lacZ reporter line. Axin2-lacZ is currently considered an accurate reporter mouse for canonical Wnt activity, as the lacz reporter gene is under control of the Axin2 promoter, and Axin2 is a direct target of canonical Wnt signaling (Yu et al., 2005; Kim and Simmer, 2007). Axin2, also known as Conductin/Axil, is a homolog of Axin1 and shares 45% amino acid identity. Axin1 and Axin2 have similar biochemical and cell biological properties but may differ in their in vivo functions (Lustig et al., 2002; Lammi et al., 2004; Yu et al., 2005). Axin1 is homogenously distributed in the embryo, whereas Axin2 is more selectively expressed in specific tissues (Lammi et al., 2004). Axin2 has two functions in the canonical Wnt signaling pathway: first, similar to Axin1, it is a cytoplasmic component that functions as a negative regulator of canonical Wnt signaling by inducing degradation of -catenin. However, it is also a direct target for the canonical Wnt signaling pathway and, hence, forms a negative feedback loop (Jho et al., 2001). In humans, mutations in Axin2 lead to two different types of tooth agenesis: hypodontia, which is a congenital lack of one or more permanent teeth, and oligodontia, a congenital lack of six or more permanent teeth (Lammi et al., 2004). Previous studies using Wnt reporters have revealed discrepancies between nuclear -catenin and Wnt reporter expression during tooth development, and so we wished to clarify Wnt responsiveness using the Axin2 reporter during early stages of tooth development. Previous studies have also concentrated on molar development and so expression in the incisors was also investigated. In addition, because canonical Wnt activity has so far only been investigated up to the

3 162 LOHI ET AL. bell stage of tooth development, we investigated expression at later stages of tooth development to gain an insight into the possible later roles of canonical Wnt signaling in tooth development. RESULTS Axin2 Expression Indicates a Role for Wnt Signaling in the Mesenchyme and Epithelium of the Developing Tooth To identify Wnt activity in tooth development we used reporter mice expressing Axin2-lacZ where the Axin2 locus is interrupted by a lacz insert (Yu et al., 2005). At E10.5 and E11.5, Axin2 expression (blue stain), was observed in both the epithelium and mesenchyme of the developing mandible, maxilla and lateral nasal processes, but excluded from the central frontonasal prominence (Fig. 1A D), agreeing with the expression patterns previously reported for TOPgal and BATgal (Brugmann et al., 2007). At E13.5, the developing molar and incisor tooth germs have reached the bud stage. In the incisors, Axin2 expression became more defined with blue cells forming a spherical ball at the anterior part of the tooth bud, where the enamel knot forms. Lower levels of expression were also noted in the condensing mesenchyme (Fig. 1 E,F). In the molar tooth buds, Axin2 expression was also observed in the developing enamel knots and the surrounding mesenchyme (Fig. 1G,H). At E14.5, the developing tooth germs have reached the cap stage. At this stage, Axin2 was expressed strongly in the primary enamel knots of both incisors and molars, with weak staining in the underlying mesenchyme (Fig. 1I,J). The late cap stage is reached at E15.5. At this stage, the primary enamel knot undergoes apoptosis, and Axin2 expression was observed over a more diffuse area of the enamel organ and the underlying mesenchyme of the dental papilla (Fig. 1K). The expression of Axin2, therefore, more closely resembles the localization of nuclear -catenin, rather than the other Wnt reporter mouse lines. By E18.5, tooth development has reached the bell stage. It is at this stage that the shape of the tooth crown is determined and the cells (ameloblasts and odontoblasts) start to differentiate and begin orchestrating the formation of enamel and dentine respectively. In the molars, the secondary enamel knots appear under the future cusps. In the molars, Axin2 was expressed in the secondary enamel knots, in the stellate reticulum, and in the underlying odontoblasts (Fig. 1L). Axin2 Expression Indicates New Roles for Wnt Signaling During Postnatal Tooth Development All previous studies have investigated canonical Wnt signaling up to the bell stage. It is important to understand the role of canonical Wnt signaling in postnatal teeth to completely appreciate its involvement in tooth organogenesis. A range of postnatal (P) molar teeth were investigated: P7, P10, P12, P15, and P21. Molar teeth were analyzed from P7, before tooth eruption. At this stage, epithelial Axin2 expression was observed at the tips of the cusps in the region of the enamel-free zone (Fig. 2A). In close up, the expression was tightly restricted to the cells sitting in the trough on the top of the cusps, with expression absent from the ameloblast cells surrounding the enamel on either side (Fig. 2B). A similar restricted expression to these small areas of epithelium was observed at P10 (Fig. 2C). In the mesenchyme, Axin2 expression was observed in the developing odontoblasts, as previously noted at E18 (Figs. 1L and 2D). Expression in the odontoblasts remained strong at P12 (Fig. 2E), as the tooth erupted, but by P15 expression started to diminish from cells nearest to the cusps (Fig. 2F). Axin2 expression was also strongly associated with the developing roots. At P10, high levels of expression were observed in the developing roots, concentrated around the root sheath and dental papilla (Fig. 3A,B). At P12, Axin2 expression was observed in the region of the developing Hertwig s epithelial root sheath (HERS) and dental papilla (Fig. 3C,D). While expression in the odontoblasts was reduced at P15, expression in the root of the tooth remained at high levels, as this region of the tooth continued to develop and differentiate (Fig. 3E). Between the roots, in the inter-radicular region, Axin2 was also highly expressed, forming a V-shape (Fig. 3F). Small numbers of positive cells were also observed throughout the molar dental papilla/pulp. To investigate these expressions further, pulp Fig. 1. Expression of Axin2 in Axin2-lacZ embryos at embryonic day (E) 10.5 E18.5. A F,H,J L: Embryos were stained with X-gal and sectioned in a frontal plane, 10 m thick. G,I: Alternatively with the aid of a tissue chopper, slices of tissue with tooth germs visible were sectioned at 250 m and stained with X-gal. A D: At E10.5 (A,B) and E11.5 (C,D), Axin2 expression is observed in the epithelium and the mesenchyme. E I: At E13.5, Axin2 expression could be seen in the developing enamel knot of the upper and lower the incisors and the surrounding mesenchyme (E,F). E: Note expression was also observed in developing vibrissae. G,H: The same expression pattern was identified in the molars. I,J: By E14.5, Axin2 was expressed in the enamel knot and the mesenchyme directly beneath it. L: By E18.5, Axin2 expression was identified in the secondary enamel knots and in the odontoblasts. De, developing enamel knot; Dp, dental pulp; En, enamel knot; Ep, epithelium; Ma, maxilla: Me, mesenchyme; Mn, mandible; Od, odontoblast; Sen, sencondary enamel knot; Ui, upper incisor. Fig. 2. Axin2 expression in cups and odontoblast of postnatal molars. Axin2-lacZ postnatal teeth ranging from postnatal day (P) 7 P15 were extracted, stained with X-gal, and sectioned in a frontal plane at 10 m thickness. A,B: In P7 molars, Axin2 expression was identified in the enamel-free areas. C,D: At P10, the expression of Axin2 could clearly be observed in the enamel-free area (C) and in the odontoblasts (D). E: Axin2 expression could still be identified in odontoblast cell at P12. F: In P15 molars, the expression of Axin2 was diminished in the odontoblast cells toward the cusp of the tooth. De, dentin; Dp, dental pulp; Efa, enamelfree area; En, enamel; Od, odontoblast. Fig. 3. Axin2 expression in root forming regions of postnatal molars. Axin2-lacZ postnatal teeth ranging from postnatal day (P) 10 P15 were extracted stained with X-gal and sectioned in a frontal plane at 10 m thickness. A,B: At P10, strong Axin2 expression could be identified in the root forming regions (A), more specifically in the dental papilla and root sheath (B). C,D: At P12, the expression of Axin2 was observed in the dental papilla, root sheath, and the odontoblast cells. E,F: At P15, the same expression pattern was identified (E) and Axin2 expression could be seen between the roots, forming a V-shape (F). Dr, developing root; Od, odontoblast; Rs, root sheath.

4 AXIN2 AND WNT SIGNALING IN TOOTH DEVELOPMENT 163 Fig. 1. Fig. 2. Fig. 3.

5 164 LOHI ET AL. cells were extracted from reporter mice and cultured. These pulp cultures showed strong blue staining, indicating that Wnt signaling is active in these cells. After passaging (which happened after 6 weeks), however, the signal was rapidly lost, indicating that the cells are no longer exposed to a Wnt signal (Fig. 4A,B). Axin2 Expression in Postnatal Incisors To follow the changing patterns of canonical Wnt signaling during cytodifferentiation, we turned to the continuously growing mouse incisor, where two different postnatal stages were investigated: P3 and P15. In the incisor, ameloblasts and, therefore, enamel only form on the labial side of the developing tooth. At P3 and onward, no expression was detected in the ameloblast layer of the tooth (Fig. 5A,B). Strong expression, however, was associated with the epithelial cells at the very tip of the incisor, where enamel does not form (Fig. 5C). The expression here is, therefore, similar to the enamel-free areas located at the tips of the molar cusps. Along the enamelfree lingual side of the incisor high levels of Axin2 expression were associated with the nonameloblast epithe- Fig. 4. Fig. 4. Axin2 expression in dental pulp cells. Dental pulp cells extracted from postnatal day (P) 7 Axin2-lacZ teeth were cultured for 2 weeks and then stained with X-gal. A: Axin2 expression was identified in the nucleus of the cells. The dental pulp cells were passaged and then stained with X-gal. B: No Axin2 expression could be identified in the cells after a further 6 weeks of culture. Fig. 5. Axin2 expression in postnatal incisor teeth. Postnatal Axin2-lacZ incisors at postnatal day (P) 3 and P15 were extracted stained with X-gal and sectioned in a frontal plane at 10 m thickness. A D: At P3, Axin2 expression was evident in the mesenchymal tissue surrounding the incisor and faintly in the odontoblasts, but no expression was observed in the ameloblasts (A,B). E,F: At P15, Axin2 expression could be identified in the odonoblasts (E) at the cervical loop; however, at the tip Axin2, expression was no longer evident in the odonotoblasts (F). G,H: Axin2 expression was also identified in the bone surrounding the tooth (G), but no expression was observed in the ameloblasts (G,H). Am, ameloblasts; Bo, bone; De, dentin; En, enamel; Me, mesenchymal tissue, Od, odonotoblasts. Fig. 5.

6 AXIN2 AND WNT SIGNALING IN TOOTH DEVELOPMENT 165 lium adjacent to the dentine (Fig. 5A,C,D). In early postnatal incisor teeth (P3) before eruption, faint Axin2 expression was evident in all the odontoblast cells and the dental follicle that surrounds the tooth (Fig. 5A,C,D). As the postnatal incisor erupted and started to mature (P15), strong Axin2 expression was observed in the odontoblasts at the cervical loop region of the incisor (Fig. 5E), however, toward the tip of the tooth Axin2 expression was no longer observed in the odontoblast cells (Fig. 5F). Thus as the odontoblasts mature Wnt activity is reduced, agreeing with the findings from molar development. In these postnatal teeth, Axin2 could also be observed in the surrounding bone, and mesenchymal tissue (Fig. 5G,H). DISCUSSION Fig. 6. Summary diagram of Axin2 expression at different stages of tooth development. Diagrammatical representation of Axin2 expression (blue) at three different stages of tooth development and in postnatal teeth. A summary of the Wnt genes identified in the epithelium observed throughout odonotogenesis is shown here in red (Sarakar and Sharpe, 1999; Yamashiro et al., 2007). In the mesenchyme, the expression of two Wnt genes could be identified Wnt5a, a noncanonical Wnt at the bud and cap stage of tooth development (Sarakar and Sharpe, 1999) and from the late bell stage Wnt10a was observed, all shown in yellow. Canonical Wnt activity in tooth development has previously been documented (Liu et al., 2008) with the use of TOPGAL/BAT-gal reporter mice and immunofluorescence for nuclear -catenin. However, inconsistencies between these approaches led us to analyze this using Axin2-lacZ reporter mice. At early stages of development, E10.5 and E11.5, the expression of Axin2 is widespread in the developing maxilla and mandible in both the epithelium and mesenchyme but excluded from the frontonasal process. As tooth development progresses, Axin2 expression becomes more specific, with clear expression in the developing primary enamel knots of incisors and molars at E13.5 and E14.5. Importantly, in addition to the epithelial expression, staining was observed in the underlying mesenchyme at these stages, expression not reported by the TOPgal or BATgal reporter mice, but agreeing with nuclear -catenin staining. Using the Axin2 reporter we also show that at E18.5, canonical Wnt activity is present in the developing odontoblasts, as well as the secondary enamel knots, expression that has not been shown previously. The Axin2 reporter mouse, therefore, represents a more accurate indicator of canonical Wnt activity compared with BATgal or TOPgal. Canonical Wnt activity in odontoblasts has previously been inferred from the expression of Wnt10a, which shifts from the secondary enamel knots to the odontoblasts during the bell stage of tooth development (Yamashiro et al., 2007). A diagrammatical representation of Axin2 expression in the different stages of prenatal tooth development is presented in Figure 6. These results were compared with those described of Sarkar and Sharpe (1999); Millar et al. (2003), Luukko et al. (2003), and Yamashiro et al. (2007) and comparisons drawn between where the members of the Wnt family and Axin2 are expressed in odontogenesis. In an effort to understand the role of canonical Wnt activity in postnatal teeth, Axin2 expression was examined in postnatal incisors and molars. The range of postnatal teeth examined encompassed pre-eruption of teeth (P8) to almost complete maturation (P21). In early postnatal molars, Axin2 expression was observed at the tip of the cusps known as the enamel-free areas. These areas have been proposed to enable the wearing away of the molar teeth (Gaunt, 1956). Wnt5a, Wnt6, and Wnt10a have also been shown to be expressed above the enamel-free zone, although their expression is fairly broad and spread over other areas of the tooth (Luukko et al., 2003).

7 166 LOHI ET AL. Canonical Wnt activity is completely absent at all stages investigated in the developing ameloblasts in molars and incisors, while present in areas of the epithelium where ameloblasts do not form (molars-enamel free areas, Incisors lingual side and tip of tooth). This indicates that canonical Wnt activity does not play a role in the terminal differentiation of ameloblasts and may, in fact, act to keep epithelial cells in a proliferative state. This is an intriguing possibility and correlates with experiments where Wnt3 has been overexpressed in the oral epithelium (Millar et al., 2003). In this study, overexpression of Wnt3 in the tooth epithelium led to a loss of ameloblasts, suggesting that canonical Wnt activity needs to be lost in order for ameloblasts to differentiate. From E18 to P12, Axin2 was found to be expressed in the odontoblast lineage. In the molars, expression was lost in these cells as they underwent terminal differentiation from P15 onward, expression being lost in a wave from the cusp of the tooth toward the roots. In the continuously growing incisors, this loss of expression was also observed in a wave, with the more mature odontoblasts at the tip of the incisor losing expression, while those odontoblasts near the cervical loops retained high levels of signaling. Canonical Wnt activity may, therefore, play an important role in induction and development of odontoblasts, while loss of canonical Wnt activity may be required for terminal differentiation. In keeping with this, Wnt10a has been shown to be able to induce DSPP expression in odontoblasts (Yamashiro et al., 2007). Dickkopf family members, Dkk1 and Dkk2, that modulate canonical Wnt activity, have an equivalent expression pattern to what we have observed in the postnatal molars (Fjeld et al., 2005). Axin2 expression was also observed in the dental papilla/pulp during late stages of tooth development. The dental pulp is a source of mesenchymal stem cells (DPSCs) that have the ability to differentiate into odontoblastlike cells, pulpal fibroblasts, and adipocytes (Gronthos et al., 2002). Canonical Wnt activity has been shown to play a critical role in development and stem cell self-renewal (Reya and Clevers, 2005). In keeping with this, canonical Wnt activity has previously been shown to inhibit dental pulp stem cell differentiation (Scheller et al., 2008). The presence of Axin2-positive cells in the dental pulp may, therefore, play a role in preventing the DPSCs from differentiating. The observed loss in Axin2 expression in prolonged culture of these cells is, therefore, significant. High levels of canonical Wnt activity were also noted in the developing molar roots, and while Axin2 expression was lost in the crown odontoblasts at late stages of postnatal development, the root odontoblasts continued to express Axin2 at high levels. The identification of canonical Wnt activity during root development has previously been indicated by expression of Wnt10a in the root odontoblasts (Yamashiro et al., 2007), similar to our observed expression of Axin2. High Axin2 levels were also associated with the inter-radicular region between the developing roots. Members of the Shh signaling pathway have also been shown to be expressed in this region, along with Bmp3 (Khan et al., 2007; Yamashiro et al., 2007). It is tempting to speculate that, Wnt signals, in combination with Bmp and Shh signaling, may play a role in shaping the developing roots, perhaps in a similar manner to their role in the shaping of the tooth crown. In the incisor, the lingual side of the tooth has been suggested to function as a root analogue, and forms the epithelial rests of Malassez (ERM). The lingual side of the incisor expressed high levels of Axin2, in a similar manner to the high levels of Axin2 associated with the developing root during molar development. This thus supports the theory that the lingual side of the incisor represents a root (Tummers and Thesleff, 2009). Our results identify areas of activity of canonical Wnt signaling in tooth development and demonstrate that Axin2-lacZ reporter mice are currently the most accurate transgenic mice to detect canonical Wnt activity. These findings suggest a role for canonical Wnt signaling in cusp morphogenesis because activity is carried through from primary enamel knots to the enamel-free areas in molar teeth. For the first time, we have demonstrated canonical Wnt activity in a range of tissues in postnatal teeth and indicate that this activity may play a role in not only the formation of the cusps, but also in enamel and dentin and formation of the roots. We show that canonical Wnt activity is detectable in primary cultures of pulp cells, but this is rapidly lost with passaging of the cells. Such loss of activity has obvious consequences for the likely survival and maintenance of mesenchyme stem cells reported to be present in cultures of tooth pulp. EXPERIMENTAL PROCEDURES Axin2-lacZ Reporter Mouse The analysis of canonical Wnt activity was achieved by using Axin2-lacZ (Yu et al., 2005) reporter mice that were crossed with wild-type (wt) CD1 mice. A range of Axin2-lacZ wt embryos were collected from embryonic ages E10.5, E11.5, E13.5, E14.5, E15.5, and E18.5 and postnatal day (P) 3, 7,10, 12, 15, and 21. Processing of Embryos Heads from embryos were fixed, whole-mount stained with X-gal (Furth et al., 1994), paraffin-embedded, and sectioned at 10 m. The sections were counterstained with saffarin. To ensure full penetration of X-gal, some heads were sectioned using a tissue chopper (Matalova et al., 2004) at 250 m and sections of tissues were stained with X-gal post-fixed and photographed. Processing Postnatal Teeth Incisors and molars were extracted, stained with X-gal, fixed and decalcified in a decalcification solution consisting of sodium citrate (40%) and formic acid (100%) in phosphate buffered saline. The teeth were paraffinembedded, sectioned at 10 m and counterstained with saffarin. Tissue Culture Postnatal teeth at P7 were extracted from Axin2-lacZ reporter mice and the dental pulp mechanically removed. The pulp was directly placed into culture medium (MSCBM PS L-Glut).

8 AXIN2 AND WNT SIGNALING IN TOOTH DEVELOPMENT 167 To be able to visualize the individual dental pulp cells the pulp was dissociated with the aid of a blade. These dental pulp cells were then cultured for 2 weeks or passaged after 6 weeks and then stained with X-gal. ACKNOWLEDGMENTS We thank Dr. R. Schmidt-Ulrich providing the Axin2-lacZ mice. M.L. was supported by an MRC studentship. REFERENCES Andl T, Reddy ST, Gaddapara T, Millar SE WNT signals are required for the initiation of hair follicle development. Dev Cell 2: Bei M Molecular genetics of ameloblast cell lineage. J Exp Zoolog B Mol Dev Evol 310B: Brugmann SA, Goodnough LH, Gregorieff A, Leucht P, ten Berge D, Fuerer C, Clevers H, Nusse R, Helms JA Wnt signalling mediates regional specification in the vertebrate face. Development 134: Cox CF, White KC, Ramus DL, Farmer JB, Snuggs HM Reparative dentin: factors affecting its deposition. Quintessence Int 23: Fjeld K, Kettunen P, Furmanek T, Kvinnsland IH, Luukko K Dynamic expression of Wnt signalling-related Dickkopf1, -2 and -3 mrnas in the developing mouse tooth. Dev Dyn 233: Furth PA, Onge L, Boger H, Gruss P, Gossen M, Kistner A, Bujard H, Hennighausen L Temporal control of gene expression in transgenic mice by a tetracycline-responsive promoter. Proc Natl Acad Sci U S A 91: Gaunt WA The development of enamel and dentine on the molars of the mouse, with an account of the enamel-free areas. Acta Anat (Basel) 28: Gronthos S, Brahim J, Li W, Fisher LW, Cherman N, Boyde A, DenBesten P, Robey PG, Shi S Stem cell properties of human dental pulp stem cells. J Dent Res 81: Ivy YH, Jerchow B, Sheu T, Liu B, Costantini F, Puzas JE, Birchmeier W, Hsu W The role of Axin2 in calvarial morphogenesis and craniosynostosis. Dev Dis 132: Jarvinen E, Salazar-Ciudad I, Birchmeier W, Taketo MM, Jernavll J, Thesleff I Continuous tooth generation in mouse is induced by activated epithelia Wnt/beta-catenin signalling. Proc Natl Acad Sci U S A 103: Jernvall J, Kettunen P, Karavanova I, Martin LB, Thesleff I Evidence for the role of the enamel knot as a control center in mammalian tooth cusp formation: non-dividing cells express growth stimulating Fgf-4 gene. Int J Dev Biol 38: Jho E, Zhang T, Domon C, Joo CK, Freund JN, Costasntini F Wnt/beta-Catenin/Tcf signalling induces the transcription of Axin2, a negative regulator of the signalling pathway. Mol Cell Biol 22: Khan M, Seppala M, Zoupa M, Cobourne MT Hedgehog pathway gene expression during early development of the molar tooth root in the mouse. Gene Expr Patterns 7: Kim JW, Simmer JP Hereditary dentin defects. J Dent Res 86: Lammi L, Arte S, Somer M, Jarvien H, Lahermo P, Thesleff I, Pirinen S, Nieminen P Mutations in Axin2 cause familial tooth agenesis and predispose to colorectal cancer. Am J Hum Genet 74: Linde A, Goldberg M Dentinogenesis. Crit Rev Oral Biol Med 4: Liu F, Chu EY, Watt B, Zhang Y, Gallant NM, Andl T, Yang SH, Lu M-M, Piccolo S, Schmidt-Ulrich R, Taketo MM, Morrisey EE, Atit R, Dlugosz AA, Millar SE Wnt/beta-catenin signalling directs multiple stages of tooth morphogenesis. Dev Biol 313: Lustig B, Jerchow B, Sachs M, Sigrid W, Pietsch T, Karsten U, Wetering M, Clevers H, Schlag PM, Birchmeier W, Behrens J Negative feedback loop of Wnt signalling through upregulation of conductin/axin2 in colorectal and liver tumors. Mol Cell Biol 22: Luukko K, Løes S, Furmanek T, Fjeld K, Kvinnsland IH, Kettunen P Identification of a novel putative signalling center, the tertiary enamel knot in the postnatal mouse molar tooth. Mech Dev 120: Matalova E, Tucker AS, Misek I Apoptosis-related factors (Fas receptor, Fas ligand, FADD) in early tooth development of the field vole (Microtus agrestis). Arch Oral Biol 50: Meikle MC Craniofacial development, growth and evolution. Bressingham, Norfolk, England: Bateson Publishing. Millar SE, Kovama E, Reddv ST, Andl T, Gaddapara T, Piddington R, Gibson CW Over-and ectopic expression of Wnt3 causes progressive loss of ameloblasts in postnatal mouse incisor teeth. Connect Tissue Res 44: Nakamura T, Vega SD, Fukumoto S, Jimenez L, Unda F, Yamada Y Transcription factor epiprofin is essential for tooth morphogenesis by regulating epithelial cell fate and tooth number. J Biol Chem 283: Sasaki T, Ito Y, Xu X, Han J, Bringas P, Maeda T, Slavkin HC, Grosschedl R, Chai Y LEF1 is a critical epithelial survival factor tooth morphogenesis. Dev Biol 278: Sarkar L, Sharpe PT Expression of Wnt signalling pathway genes during tooth development. Mech Dev 85: Scheller EL, Chang J, Wang CY Wnt/beta-catenin inhibits dental pulp stem cell differentiation. J Dent Res 87: Reya T, Clevers H Wnt signalling in stem cells and cancer. Nature 14: Tummers M, Thesleff I The importance of signal pathway modulation in all aspects of tooth development. J Exp Zoolog Mol Dev Evol 312B: Yamashiro T, Zheng L, Shitaku Y, Saito M, Tsubaimoto T, Takada K, Takano- Yamamoto T, Thesleff I Wnt10a regulates dentin sialophosphoprotein mrna expression and possibly links odontoblast differentiation and tooth morphogenesis. Differentiation 75: Zhou P, Byrne C, Jacobs J, Fuchs E Lymphoid enhancer factor 1 directs hair follicle patterning and epithelial cell fate. Genes Dev 9: