Analysis of signalling pathways in the developing mammalian tooth and palate

The head is a highly complex part of the body and its development requires tightly regulated interactions between the forming tissues. Disruption of these interactions or signalling pathways at any stage may cause craniofacial malformations. Pax9 encodes a transcription factor with important functions during embryogenesis and organogenesis and Pax9 deficient mice exhibit craniofacial anomalies including absence of all teeth and a cleft palate. Interestingly, heterozygous mutations in the PAX9 gene were associated with oligodontia and facial clefting also in humans. To elucidate the functions of Pax9 during craniofacial morphogenesis, this project was primarily aimed at the further characterisation of the role of Pax9 for regUlating signalling processes governing tooth development and palatogenesis. Prior to this study, Sip1 was identified as a potential Pax9 target gene during tooth development. During the bud stage of odontogenesis mesenchymal Bmp4, which is under the regUlation of Pax9 and Msx1, induces gene expression specifically in the dental epithelium. Among induced genes is p21, a mediator of cell cycle arrest in the future epithelial signalling centre, the enamel knot. Sip1 is known to interact with Tgf{3lBmp signalling and, in another context was demonstrated to act as an inhibitor of expression of p21. This lead to the hypothesis that during the bud stage Sip1 function could be crucial for restricting the expression of Bmp4-inducible genes specifically to the dental epithelium. At all stages analysed (E12.5-E14.5) Sip1 was found to be expressed in the mesenchymal tissue compartment of all teeth. Interestingly, significantly reduced levels of Sip1 expression were observed not only in Pax9-1- mutants but also in Msx1-deficient dental mesenchyme. While this indicates that Pax9 is required but not sufficient for the expression of Sip 1, preliminary results from organ culture experiments using bead implantation suggest that Bmp4 may not be crucial for the regulation of Sip1 expression during the bud stage of odontogenesis. Furthermore, expression of Sip1 and p21 was detected in complementary patterns even when normal expression domains were not maintained, e.g. in the developing lower incisors of Pax9IMsx1 double heteroyzgous embryos. These observations support the idea that at the bud stage Sip1 may act as an inhibitor of p21 and other Bmp4-inducible genes in the dental mesenchyme to achieve a predominantly epithelial response to Bmp4 signalling at a timepoint when the odontogenic potential resides in the dental mesenchyme. To allow further investigation of Sip1 functions during tooth formation in an in vivo model, conditional inactivation of the gene in dental tissue is necessary as Sip1-deficiency results in lethality during embryonic development prior to tooth formation. For this purpose, a second aim was to generate a Pax9cre mouse strain to provide a tool for tissue specific gene inactivation in the dental mesenchyme. A targeting vector was produced and successfully used for targeting of the Pax9 locus in ES cells. This was confirmed in a peR based approach and by Southern blot hybridisation. Several ES cell clones containing the Pax9cre allele were used for blastocyst injection to generate chimaeric offsprings. Unfortunately, germline transmission was not achieved during the course of this project. However, karyotype analysis of the obtained ES cell lines confirmed the suitability of two targeted cell lines which will be used for future experiments. While morphological defects during tooth development were previously characterised Pax9 functions during palate development were not analysed in detail prior to this project. Thus, a careful characterisation of cleft palate formation in Pax9-/- was carried out using scanning electron microscopy and histological techniques. In vitro culture assays and sUbsequent histological analysis showed that Pax9-deficient palatal shelves were able to fuse suggesting that impaired shelf elevation is the principle reason for the cleft palate phenotype in homozygous Pax9 mutants. As the failure of shelf elevation is thought to be caused by mechanical hindrance due to abnormally shaped palatal shelves Pax9 function appears crucial for the control of palatal shelf morphology. As absence of Fgf10 function causes morphological defects similar to those observed in the palatal shelves of homozygous Pax9 mutants it was decided to analyse expression of the gene and its downstream target Shh in Pax9-deficient palatal shelves. In a second approach, using microarray analysis, gene expression profiles of normal and Pax9- deficient palatal shelves at E12.5 were generated and compared. Amongst the most de-regulated genes 19 factors were previously associated with craniofacial development and selected candidates for follow up analyses were chosen from this group. In the secondary palate of Pax9-/- embryos expression of Fgf10, Shh as well as of Msx1 and Bmp4 which are both also required for Shh expression were found to be downregulated. It was suggested that Pax9 regulates expression of (Msx1)/Bmp4 and Fgf10 which are thought to independently control expression of Shh, a crucial factor for mesenchymal cell proliferation. In addition, reduced expression of Shox2 and Fgf18 was observed in Pax9-deficient palatal shelves. Altered patterning involving the factors identified during this study is likely to contribute to the morphological defects observed in the secondary palate of homozygous Pax9 mutants. Interestingly, this project uncovered a previously not described defect during primary palate formation in homozygous Pax9 mutants. In the absence of Pax9 expression of Msx1, Bmp4, Fgf10, Shh and Fgf18 was also affected in the primary palate. This suggests that although primary and secondary palate are considered distinct embryological entities similar molecular interactions are required for the development of both structures.EThOS - Electronic Theses Online ServiceGBUnited Kingdo