| Sign In to gain access to subscriptions and/or personal tools. |
Organized Tooth-specific Cellular Differentiation Stimulated by BMP4Department of Craniofacial Development, Dental Institute, Kings College London, Floor 28 Guy’s Hospital, London Bridge, London SE1 9RT, UK; Correspondence: * corresponding author, paul.sharpe{at}kcl.ac.uk
Mammalian teeth develop on the oral surface of the first pharyngeal arch by a series of reciprocal interactions between epithelial and mesenchymal cells. The embryonic first pharyngeal arch oral epithelium is able to induce tooth formation when combined with mesenchymal cells from the second pharyngeal arch, a region devoid of tooth development. Second pharyngeal arch mesenchyme is thus competent to form teeth if provided with the correct signals. First-arch oral epithelium expresses several signaling molecules that could be potential inducers of tooth development, including BMP4. The addition of BMP4 to intact second-arch explants resulted in the development of organized structures containing layers of cells that express marker genes of tooth-specific cells, odontoblasts and ameloblasts. Thus, although overt tooth development did not occur, BMP4 has the ability to stimulate organized differentiation of epithelial- and mesenchymal-derived dental-specific cells from non-dental primordia.
Key Words: 2nd pharyngeal arch • tooth development • epithelium • mesenchyme • tissue engineering
The ability to induce cellular differentiation in adults to replace lost or damaged cells, or to tissue-engineer organs in vitro, is the current goal of regenerative medicine. An understanding of the nature of the signals and tissue interactions that regulate these processes in the embryo is essential to the development of such approaches. To investigate such signals involved in tooth development, we set out to identify molecules whose expression correlated with the location of tooth initiation, and to assay the effects of the addition of such molecules to tissues that do not form teeth. Teeth are an example of organs that develop from reciprocal interactions between embryonic epithelium (oral epithelium) and mesenchyme. In common with most organs, teeth contain specialized differentiated cells—in this case, odontoblasts that produce dentin, and ameloblasts that secrete enamel proteins. In mammals, teeth do not develop on the second pharyngeal arch, a structure that consists of the same cell types (neural-crest-derived ectomesenchyme, epithelium, and endoderm) as the first pharyngeal arch, where teeth do develop. Furthermore, the expression of Hox genes in second-pharyngeal-arch ectomesenchyme does not provide an explanation for why teeth do not develop, since teeth can form from Hox-positive ectomesenchymal cells (James et al., 2002). Recombination experiments have shown that early oral epithelium (E9-E11) can induce tooth development in second-arch mesenchyme and aboral first-arch mesenchyme (Mina and Kollar, 1987; Tucker et al., 1999). In addition, such epithelium can induce tooth development in the cranial neural crest and trunk neural crest if recombined prior to migration (Lumsden, 1988). Limb mesenchyme, in contrast, is unable to form teeth (Lumsden, 1988). Second-arch epithelium, aboral first-arch epithelium, or limb epithelium is unable to induce this response (Mina and Kollar, 1987; Lumsden, 1988; Tucker et al., 1999). Oral epithelium after E11 is also no longer able to induce this effect, the capacity to stimulate tooth development having transferred to the mesenchyme. Thus, tooth development is observed in recombinations between dental mesenchyme from E11 onward, when combined with non-dental epithelium (Ruch, 1984; Mina and Kollar, 1987). These recombination experiments indicate that the odontogenic potential resides initially in the early oral epithelium, and suggests that this epithelium must provide signals to the underlying mesenchyme that are not present in second-arch epithelium. In this paper, we have attempted, using protein-loaded beads implanted into the second pharyngeal arch, to identify the signals which are unique to the oral epithelium of the first pharyngeal arch.
Tissue Recombination Mice of the CD1 strain were used. First and second pharyngeal arches were dissected in DMEM containing glutamax-1. The epithelium and mesenchyme were isolated following incubation in Dispase (Gibco BRL, Paisley, UK) made up in calcium- and magnesium-free PBS at 2 U/mL for 10–15 min at 37°C. After incubation, the tissues were washed in DMEM with 10% FBS, and mechanically separated with the use of fine tungsten needles. First-arch epithelium was replaced on the second-arch mesenchyme, and second-arch epithelium was replaced on the second-arch mesenchyme as a control. The explants were cultured on transparent Nucleopore membrane filters (0.1-µm pore diameter; Whatman VWR Ltd., Lutterworth, UK), supported by a metal grid following the Trowell (1959) technique as modified by Saxén (1966). After one day in culture, the explants were transferred to renal capsules for 12 days. Resulting tissue was dissected out of the kidney and fixed in Bouin’s, decalified (0.5 M EDTA), and wax-embedded for sectioning. Slides were stained with a trichrome stain.
Explants Cultured with BMP4 Protein
In situ Hybridization All experiments involving animals were carried out under license from the UK Home Office.
To confirm that we could generate teeth from second-arch mesenchyme, as has been previously observed (Mina and Kollar, 1987), we carried out heterotypic tissue recombinations where oral epithelium from E10 embryos was recombined with mesenchyme from the second pharyngeal arch, a tissue that does not form teeth. Transplants of these recombinations into renal capsules of adult mice resulted in the formation of teeth (N = 8/9) (Fig. 1B  ), whereas cultures of intact second pharyngeal arches did not produce any teeth (N = 7/7) (Fig. 1A). These results agree with those from earlier studies, that the oral epithelium contains the instructive information to initiate tooth formation. Moreover, the ability of second-arch mesenchyme to respond to tooth-inducing signals and participate in tooth development establishes that there is no specific odontogenic population of cranial neural crest cells (Ruch, 1984; Mina and Kollar, 1987; Lumsden, 1988).
Experiments in avian embryos, where cranial neural crest cells are ablated prior to migration, have shown that signaling molecules expressed in oral epithelium are maintained in the absence of ectomesenchyme (Veitch et al., 1999). It appears, therefore, that spatial expression of tooth-inductive signals in oral epithelium is an intrinsic property of the epithelium. The early (E10) embryonic oral epithelium produces FGFs, BMPs, and Shh signaling proteins, and all of these have been implicated in tooth initiation. Of these, only BMP4 is not expressed in the epithelium of the second arch (Fig. 2A ) (Veitch et al., 1999). Expression of BMP4 shifts from the oral epithelium to the mesenchyme at E11 (Fig. 2B) (Vainio et al., 1993; Tucker et al., 1998). This change in expression coincides with the transition time, when the ability to induce tooth development switches from the epithelium to the mesenchyme (Mina and Kollar, 1987). The timing of expression of BMP4 in the oral epithelium, therefore, makes it a strong candidate for the tooth-induction signal. To determine if the absence of BMP4 expression in the second arch might account for the lack of tooth development in this arch in mammals, we added BMP4 ectopically to cultures of second pharyngeal arches. Second pharyngeal arches of E11 embryos were cultured intact as explants, with BMP4 or BSA control beads implanted just below the epithelium. After overnight culture, the explants were transplanted under the renal capsules of adult mice and left for 12 days to develop. Ectopic growths from the renal capsules were fixed and sectioned. Control explants developed into fibrous cysts, with little or no obvious cellular organization or differentiation (n = 4/4) (Figs. 3I–3K). Explants treated with BMP4 did not form morphologically obvious teeth, and, superficially, appeared to be similar to the cysts observed in the controls (n = 5/5). However, when sectioned, these cysts appeared to contain organized cellular structure, with evidence of cellular differentiation (Fig. 3). To identify possible odontogenic cells, we hybridized sections with probes for dentosialophosphoprotein (Dspp), a gene whose predominant site of expression is developing odontoblasts, and amelogenin, which is expressed only in dental cells, predominantly ameloblasts. Expression of these tooth-cell marker genes was absent in control cysts but was clearly evident in those structures obtained from BMP4-treated explants (Fig. 3). Dspp and amelogenin expression was evident in adjacent layers of cells surrounding a fibrous core. The Dspp-positive cells had a slightly elongated, almost cuboidal, shape and were inside the layer of amelogenin-positive cells (Fig. 3D). The amelogenin-positive cells did not show any signs of polarization, but had a distinct rounded morphology that was obviously different from that of the more fibroblastic-like cells found adjacent to the chondrocytes (Fig. 3H).
In mammals, teeth develop only in the oral cavity on the jaw margins, whereas in other vertebrates—such as fish, for example—teeth develop in other regions, including the tongue, palate, and pharynx (Cassin and Capuron, 1979; Barlow and Northcutt, 1995; Graveson et al., 1997). Accumulating evidence supports the concept that the early pre-odontogenic epithelium provides the source of signals that initiate tooth development, and that the responding ectomesenchyme cells are plastic until they receive these signals (Mina and Kollar, 1987; Lumsden, 1988; Ferguson et al., 2000). In mammals, this property is restricted to oral epithelium, whereas in fish, other epithelia, including endoderm, can produce teeth. Recombinations of mouse embryonic pharyngeal endoderm with first-arch ectomesenchyme does not produce teeth, suggesting that either the mouse endoderm is inhibitory for odontogenesis, or that the oral epithelium produces signals that the endoderm does not (unpublished). To determine if the mouse first-arch epithelium was able to induce tooth formation from non-odontogenic ectomesenchyme, we recombined oral epithelium with ectomesenchyme from the second pharyngeal arch. These recombinations reproducibly formed complete tooth crowns following renal transfer. This result clearly demonstrates that the oral epithelium produces odontogenic initiation signals, and that non-odontogenic ectomesenchyme can be directed to form teeth. Members of the TGFβ superfamily have been implicated as playing important roles at all stages of tooth development and in repair (Thesleff and Sharpe, 1997; Heikinheimo et al., 1998; Magloire et al., 2001). The earliest expression of BMP4 in pre-odontogenic oral epithelium has an essential role in regulating the spatial expression of homeobox genes in the ectomesenchyme (Vainio et al., 1993; Neubüser et al., 1997; Tucker et al., 1998). BMP4 subsequently becomes localized to the dental placodes, and, throughout the rest of tooth development, expression switches between epithelial- and mesenchymal-derived cells. During early cytodifferentiation into pre-odontoblasts and pre-ameloblasts, BMPs are suggested to play roles in inducing ameloblast differentiation (Coin et al., 1999; Wang et al., 2004). Since BMP4 is not expressed in the epithelium of the pharyngeal arches caudal to the first arch, we reasoned that it was a good candidate for an odontogenic initiation signal. By adding exogenous BMP4 to intact cultures of second pharyngeal arches, we were able to investigate the potential of BMP4 to induce odontogenesis in non-dental cells. Although no recognizable teeth or hard tissues (enamel and dentin) were obtained, we did observe organized cell layers not evident in controls. These cells differentially expressed Dspp and amelogenin in adjacent layers in the same relative orientation as that in teeth. These cells thus have properties of pre-odontoblasts and pre-ameloblasts. Cells derived from second-arch cultures not treated with BMP4 did not produce cells expressing these genes. It thus seems that the addition of one signaling protein early in development can stimulate odontogenic cell differentiation. Although complete teeth were not formed from second-arch cultures, these experiments do highlight the potential of BMP4 to induce tooth cell differentiation. BMPs have previously been shown to be capable of redirecting adult dental pulp cells to differentiate into odontoblasts, and it is possible that, in our second-arch cultures, BMP4 is similarly acting to stimulate the differentiation of pre-odontoblasts from ectomesenchyme cells, which in turn may stimulate adjacent epithelial cells to differentiate into pre-ameloblasts. Clearly, the formation of a complete tooth from non-dental cells requires other factors in addition to BMP4, but this experimental system can provide a simple assay to screen for the combination of molecules required to induce tooth formation.
We thank Malcolm Snead for critically reading the manuscript and for the amelogenin probe, and Mary MacDougall for the Dspp probe. The work was funded by the MRC and Wellcome Trust. Received for publication February 25, 2005. Revision received April 6, 2005. Accepted for publication April 21, 2005.
Journal of Dental Research, Vol. 84, No. 7,
603-606 (2005) This article has been cited by other articles:
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
ABSTRACT
INTRODUCTION
