Skip Navigation


Human Molecular Genetics Advance Access originally published online on October 19, 2005
Human Molecular Genetics 2005 14(23):3605-3617; doi:10.1093/hmg/ddi388
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
14/23/3605    most recent
ddi388v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (11)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Kist, R.
Right arrow Articles by Peters, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kist, R.
Right arrow Articles by Peters, H.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2005. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Reduction of Pax9 gene dosage in an allelic series of mouse mutants causes hypodontia and oligodontia

Ralf Kist1, Michelle Watson1, Xiaomeng Wang1, Paul Cairns1, Colin Miles1, Donald J. Reid2 and Heiko Peters1,*

1Institute of Human Genetics, University of Newcastle upon Tyne, International Centre for Life, Central Parkway, Newcastle upon Tyne NE1 3BZ, UK and 2Department of Oral Biology, School of Dental Sciences, University of Newcastle upon Tyne, Framlington Place, Newcastle upon Tyne NE2 4BW, UK

* To whom correspondence should be addressed. Tel: +44 1912418622; Fax: +44 1912418666; Email: heiko.peters{at}ncl.ac.uk

Received August 24, 2005; Accepted October 12, 2005


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Missing teeth (hypodontia and oligodontia) are a common developmental abnormality in humans and heterozygous mutations of PAX9 have recently been shown to underlie a number of familial, non-syndromic cases. Whereas PAX9 haploinsufficiency has been suggested as the underlying genetic mechanism, it is not known how this affects tooth development. Here we describe a novel, hypomorphic Pax9 mutant allele (Pax9neo) producing decreased levels of Pax9 wild-type mRNA and show that this causes oligodontia in mice. Homozygous Pax9neo mutants (Pax9neo/neo) exhibit hypoplastic or missing lower incisors and third molars, and when combined with the null allele Pax9lacZ, the compound mutants (Pax9neo/lacZ) develop severe forms of oligodontia. The missing molars are arrested at different developmental stages and posterior molars are consistently arrested at an earlier stage, suggesting that a reduction of Pax9 gene dosage affects the dental field as a whole. In addition, hypomorphic Pax9 mutants show defects in enamel formation of the continuously growing incisors, whereas molars exhibit increased attrition and reparative dentin formation. Together, we conclude that changes of Pax9 expression levels have a direct consequence for mammalian dental patterning and that a minimal Pax9 gene dosage is required for normal morphogenesis and differentiation throughout tooth development.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Congenital absence of teeth is a common developmental anomaly in humans and is estimated to have a prevalence ranging between 2.3 and 10% of the population worldwide, not counting third molars (wisdom teeth) (1Go). Oligodontia is commonly referred to the absence of more than six permanent teeth (excluding third molars), whereas in hypodontia up to six teeth are missing. Several recent studies indicate that heterozygous mutations in the gene encoding the paired domain- containing transcription factor PAX9 causes oligodontia in a non-syndromic, autosomal-dominant manner (2Go). A consistent manifestation is the absence of most permanent molars, but missing second premolars and incisors have been reported as well (3Go–7Go). Moreover, size reduction of both primary and permanent teeth has been associated with mutations in PAX9 (8Go,9Go). Affected patients may have a normal primary dentition but absence of primary teeth has been reported in about half of the families analysed (3Go,5Go,8Go,10Go,11Go). Whereas the heterogeneity of mutations and the genetic background may explain the inter-familial and intra-familial variability of the phenotype, respectively, the available data strongly suggest that PAX9 is haploinsufficient for tooth development in humans.

The dentition of rodents is evolutionarily highly adapted and these characteristics have to be considered when using the mouse as a model system to study mammalian tooth development. Mice form three molars and one incisor within each side of upper and lower jaws and a tooth-less region, called diastema, separates these two dental fields. Humans have a more basic dental formula comprising central incisors, lateral incisors, canines, premolars and molars. In addition, mice develop only one set of teeth, whereas humans form two generations, deciduous and permanent teeth. Finally, rodent incisors grow continuously to compensate for the wear of dentin and enamel at the incisor tips. Despite these developmental and anatomical differences, morphological and molecular evidence indicates that the genetic control of tooth formation is highly conserved among vertebrates. This is strongly supported by recent identification of the same genes being responsible for tooth abnormalities in mice and humans (12Go,13Go).

Tooth development is regulated by a series of reciprocal interactions between the dental epithelium and the underlying, neural crest-derived dental mesenchyme (14Go,15Go). In the mouse embryo, tooth formation is initiated at embryonic day 10.5 (E10.5) and subsequently progresses through a series of morphologically distinguishable stages known as bud stage, cap stage, bell stage, secretory stage and finally eruption. During initiation, the dental epithelium secretes growth factors that induce the expression of specific genes in the dental mesenchyme in a highly co-ordinated pattern. At E12.5, the mesenchyme has condensed to form the so-called dental papilla, which at this stage acquires the potential to control further tooth development (16Go,17Go). At the late bud stage, a new signalling centre, the enamel knot, is induced in the epithelium. The enamel knot is fully established at E14.5 and together with the secondary enamel knots formed at later stages control subsequent tooth morphogenesis (18Go). The development of the human primary dentition essentially follows the same morphological stages and starts at the sixth prenatal week with the formation of 20 tooth buds representing the precursors of the primary (deciduous) teeth. On the lingual side of these tooth buds, successional tooth buds develop from the fifth month in utero to form rudiments of the permanent dentition. In contrast, the 12 permanent molars are located posterior to the deciduous molars and are thus not formed via successional tooth buds, but, as in rodents, develop directly from the dental lamina.

A Pax9 null allele in mice was generated previously and homozygous mutants were shown to exhibit a wide range of developmental abnormalities including cleft secondary palate, tooth agenesis, anterior digit duplication and missing thymus, parathyroid glands and ultimobranchial bodies (19Go). During early tooth development, Pax9 expression is induced in the prospective dental mesenchyme at E10.5 through Fgf8 secreted by the oral epithelium (20Go) and Pax9 was shown to be essential for the developmental progression of the first molars beyond the bud stage (19Go). At this stage, Pax9 is required to maintain the mesenchymal expression of a positive feedback loop involving the homeobox gene Msx1 and a gene encoding a member of the Tgf-ß growth factor family, Bmp4. Similar to Pax9-deficient mice, tooth development is also arrested at the bud stage in mice lacking Msx1 (21Go) and the importance of the Msx1-Bmp4 feedback loop at the bud stage was demonstrated in organ culture and transgenic rescue experiments (18Go,22Go,23Go). These observations suggest that the Pax9-regulated expression of the Msx1-Bmp4 feedback loop defines a key step for the molecular control of early tooth morphogenesis.

A delay in the formation of the remaining teeth in children with hypodontia has been noted decades ago (24Go) but it is still not known how the morphogenesis of different teeth is coordinated during development. Moreover, a model system to study the pathogenesis of non-syndromic oligodontia is currently not available, as mice heterozygous for the null allele Pax9lacZ have a normal dentition, and homozygous mutants exhibit an early arrest of odontogenesis affecting all teeth. To address these limitations, we have generated a novel, hypomorphic Pax9 allele (Pax9neo) and investigated the developmental consequences of a more drastic Pax9 gene dosage reduction. We produced an allelic series by compounding the Pax9neo mutation with the Pax9lacZ null allele and analyses of these mice revealed that, similar to humans, tooth formation is the most sensitive developmental process affected by reduction of Pax9 gene dosage. Interestingly, the dentition is not uniformly affected and the minimal Pax9 gene dosage required for the formation of individual teeth varies among different tooth types. In addition, we identified structural defects in the remaining teeth of Pax9neo/neo and Pax9neo/lacZ mutants, demonstrating a hitherto unknown role of Pax9 for the formation of enamel and dentin.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Generation of a hypomorphic Pax9 allele in mice
Previous gene targeting experiments in mice have shown that insertion of a neo-cassette into an intron can interfere with normal splicing, thereby reducing the amount of wild-type mRNA of the targeted gene (25Go,26Go). We applied a similar strategy and cloned a targeting construct in which a removable neomycin resistance cassette (neo) flanked by FRT sites was inserted into the intron between exons 2 and 3 of Pax9 (Fig. 1A). In addition, to allow future, tissue-specific inactivation of Pax9 using appropriate Cre mice, loxP sites flanking exons 1 and 2 were inserted. This novel Pax9 allele, containing the neo-cassette in intron 2, is referred to as Pax9neo. Gene targeting in mouse embryonic stem (ES) cells produced two correctly targeted ES cell clones (Fig. 1B), which were used for blastocyst injection to establish two independent Pax9neo mutant mouse lines. Like heterozygous Pax9lacZ (Pax9+/lacZ) mice (19Go), heterozygous Pax9neo (Pax9+/neo) mutant mice of both lines are fertile and phenotypically normal.



View larger version (22K):
[in this window]
[in a new window]
  		Figure 1. Generation of the Pax9neo allele. (A) Schematic overview of the gene targeting strategy. A 16.4 kb XbaI-fragment of the wild-type Pax9 locus is shown. The targeting construct contains an FRT-flanked neomycin resistance cassette (neo) inserted into the second intron of Pax9 and a viral thymidine kinase cassette (tk). The Pax9 exons 1 and 2 (black boxes), including the start codon and the paired domain, respectively, are flanked by loxP sites. After homologous recombination, the targeted locus is termed Pax9neo. (B) Probes A and B were used for Southern blot analysis and detected the expected fragment sizes for the wild-type and the mutant allele in two correctly targeted ES cell clones (1C9 and 3C7). X, XbaI.

 
The phosphoglycerate kinase 1 (Pgk1) promoter of the neo-cassette is known to contain cryptic splice sites (25Go,26Go), and alternatively spliced mRNAs generated from the Pax9neo locus were identified by RT–PCR using RNA isolated from a Pax9+/neo embryo. DNA sequencing of the predominant RT–PCR product revealed that a hybrid Pax9neo mRNA is made by splicing the paired box-containing exon 2 of Pax9 to the 3'-region of the Pgk1 promoter (Fig. 2A and B). As the coding region of neo is in frame with exon 2 of Pax9, translation of this mRNA predicts an aberrant Pax9neo fusion protein containing the paired domain and the full-length neomycin phosphotransferase (Fig. 2B) but lacking the majority of the C-terminal region of Pax9 which is thought to contain the transactivation domain (27Go).



View larger version (33K):
[in this window]
[in a new window]
  		Figure 2. The Pax9neo mutation causes alternative Pax9 mRNA splicing and results in reduced levels of wild-type Pax9 mRNA. (A) Genomic structure of the Pax9neo locus and generation of the wild-type Pax9 mRNA (black line) and mutant Pax9neo mRNA (dotted line) by alternative splicing. The Pgk1 promoter (shaded box) of the neo-cassette and the locations of primers used for RT–PCR analysis are indicated. (B) Sequence of the Pax9neo locus around the alternative splice sites. Exon 2 is alternatively spliced to the 3' end of the Pgk1 promoter. The wild-type splice donor site and the predominantly used mutant splice acceptor site (25Go,26Go) are boxed. The alternative splicing creates a hybrid Pax9neo mRNA containing Pax9 exons 1 and 2 and two novel in-frame codons (GCC AAT), followed by the complete coding region of neo. The neo stop codon and the bovine growth hormone (bGH) polyadenylation signal are underlined and the deduced, partial amino acid sequence of a hypothetical Pax9neo protein is shown in bold. (C) Semi-quantitative RT–PCR analysis of RNA isolated from E12.5 mouse embryo heads. Wild-type Pax9 mRNA levels gradually decrease in Pax9+/neo and Pax9neo/neo embryos, whereas mutant Pax9neo mRNA levels are increased accordingly. Two weak Pax9neo bands indicate the presence of additional minor splice forms. Similar Gapdh expression levels illustrate that equal amounts of cDNA have been used for all genotypes tested. (D) Quantitative real-time RT–PCR analysis of RNA isolated from whole E10.5 mouse embryos. Normalized wild-type Pax9 mRNA levels from two experiments, each performed in triplicate for each genotype are shown for the allelic series of Pax9 mutant mice. A gradual reduction of wild-type Pax9 mRNA levels was observed, which correlates with the increasing severity of the phenotype (Fig. 3).

 
Quantification of Pax9 wild-type transcript levels in an allelic series of Pax9 mutant mice
Semi-quantitative RT–PCR experiments indicated that the Pax9neo mRNA was produced at a significantly higher level in Pax9neo/neo mutants when compared with that of Pax9+/neo mutants, whereas the level of Pax9 wild-type mRNA was reduced accordingly (Fig. 2C). To quantify the relative abundance of wild-type Pax9 mRNA in mouse embryos that carry the available Pax9 mutant alleles in all possible combinations, we performed real-time PCR assays. As the first morphological defects of Pax9-deficient mouse embryos occur in the pharyngeal pouch endoderm at E11.5 (19Go), RNA was extracted from E10.5 embryos to minimize variations of Pax9 mRNA levels caused by a loss of Pax9 expressing cells or by an increase of Pax9 mRNA levels owing to compensatory activation of Pax9 transcription. The real-time RT–PCR analysis revealed that 44% and 32% of Pax9 wild-type mRNA was present in Pax9+/neo and Pax9+/lacZ embryos, respectively. The amount of Pax9 wild-type transcripts was further reduced to 20% in Pax9neo/neo mutants and to 7% in Pax9neo/lacZ compound mutants (Fig. 2D). The drastic reduction of wild-type mRNA produced by the Pax9neo allele demonstrates the efficient interference of the neo-cassette with normal splicing. However, the Pax9 mRNA levels in Pax9+/neo and Pax9+/lacZ embryos, which both contain one intact copy of the wild-type Pax9 locus, are unexpectedly low. Whereas the molecular mechanisms resulting in <50% of Pax9 mRNA in these mutants remain to be investigated, the allelic Pax9 mutant series provides a basis for analysing the effects of a gradual decrease of functional Pax9 gene dosage on development and organogenesis.

Pax9neo/neo and Pax9neo/lacZ mutants exhibit hypodontia and oligodontia with variable severities
Pax9neo/lacZ mutants, but not Pax9neo/neo mutants, were under-represented at weaning (3 weeks after birth) and subsequently remained considerably smaller than their wild- type littermates. Besides dental abnormalities, skeletal staining and inspection of inner organs by autopsy revealed no additional defects in Pax9neo/neo and Pax9neo/lacZ mutant mice. The secondary palate was consistently intact and digit duplication was never observed. In addition, the size of the thymus was normal and clinical symptoms related to an abnormal calcium or phosphate homeostasis caused by defects in the parathyroid glands and ultimobranchial bodies were not noticed. Thus, the severe dental abnormalities of Pax9neo/lacZ mutants are likely to represent the principal cause for the observed growth retardation but minor deficiencies in other, Pax9 expressing organs might contribute to this phenotype.

In adult Pax9neo/neo mice, the lower incisors and all third molars were consistently affected. In most cases, these teeth were absent but in a significant number of mutants, we also observed hypoplastic lower incisors as well as smaller third molars (Figs 3 and 4A). In approximation to the definition distinguishing hypodontia from oligodontia in humans, we found that 35% (8/23) of Pax9neo/neo mutants were affected by hypodontia (absence of up to three teeth out of 16), whereas 52% (12/23) lacked four or more teeth, which we classified as oligodontia. The remaining 13% (3/23) developed all teeth but lower incisors and third molars were hypoplastic. In contrast, Pax9neo/lacZ mutants consistently manifest oligodontia. The lower dentition was always affected stronger and in severe forms no teeth were found in the lower jaws of these mutants (Figs 3 and 4B). Pax9neo/neo mutants and Pax9neo/lacZ compound mutants used for scoring the dental phenotypes were maintained on a mixed genetic background and this may account for the variability of the dental phenotypes. However, 35% (8/23) of Pax9neo/neo mutants showed tooth defects that were not identical in the left and right halves of the jaws (Fig. 4), suggesting that stochastic effects may also play a significant role for these variations.



View larger version (128K):
[in this window]
[in a new window]
  		Figure 3. Pax9neo/neo and Pax9neo/lacZ mice display hypodontia or oligodontia with different severities. The upper incisors (A), lower incisors (D), upper molars (G) and lower molars (J) of a 10-week-old wild-type mouse are shown for comparison. The upper incisors of Pax9neo/neo mice (B) form normally, but need regular clipping to prevent overgrowth because of hypoplasia of the lower incisors (E). The chalky-white colour of the lower incisors indicates an enamel defect. The upper third molars (H) and lower third molars (K) are unilaterally or bilaterally missing in Pax9neo/neo mice. Pax9neo/lacZ mice are more severely affected: the upper incisors are hypoplastic and lack enamel (C), whereas lower incisors are absent (F). The upper third molars (I) and the lower second and third molars (L) are unilaterally or bilaterally missing in Pax9neo/lacZ mice. Note also the marked attrition of the lower first molar (L).

 


View larger version (54K):
[in this window]
[in a new window]
  		Figure 4. Frequency of hypodontia and oligodontia in Pax9neo/neo and Pax9neo/lacZ mice. Schematic representation of the dentition of adult Pax9neo/neo(A) and Pax9neo/lacZ (B) mice. Unaffected teeth are indicated by a white star, hypoplastic or unerupted teeth by a grey star and missing teeth by a black star. The increasing severity of dental defects is shown from top to bottom and the occurrence of a specific dental phenoytype (N) is noted. Pax9neo/neo mice are generally less affected and display a broader range of variation than Pax9neo/lacZ mice. The assignment of the phenotype to represent hypodontia or oligodontia was adopted from the definitions usually applied for missing teeth in humans.

 
Enamel defects and reparative dentin formation
The hypoplastic and non-erupted teeth of Pax9neo/neo and Pax9neo/lacZ mice indicate that Pax9 is not only required for early tooth development but is also involved in differentiation and tooth growth. A defect in the enamel of incisors is indicated by the lack of a brownish pigmentation (Fig. 3C and E), which is typically present in the continuously growing incisors of mice (Fig. 3A and D). Histological analyses of 5-month-old Pax9neo/neo mutants show that ameloblasts and enamel are missing and that the dentin is connected to the alveolar bone by a periodontal ligament (Fig. 5I and L). A small population of ameloblasts undergoing differentiation could be identified in 2-month-old mutants, indicating that the lack of enamel is associated with a gradual loss of ameloblasts. These ameloblasts do not elongate and produce only a thin layer of pre-enamel (Fig. 5H and K). In contrast to the incisors, ground sections show a normal thickness of enamel in the molars of both Pax9neo/neo and Pax9neo/lacZ mutants (Fig. 5B and C).



View larger version (129K):
[in this window]
[in a new window]
  		Figure 5. Structural and cellular defects in Pax9neo/neo and Pax9neo/lacZ teeth. (A–C) Sagittal ground sections of lower first molars of 4-month-old mice. (B) Mild attrition is observed in Pax9neo/neo molars and is usually more pronounced on the anterior cusps (arrow). (C) Severe attrition is seen in Pax9neo/lacZ molars (arrows), whereas the thickness of the enamel is not affected. Irregular dentin formation is indicated by accentuated incremental lines (arrowheads in inset). (D–F) Sagittal sections of lower first molars of 2-month-old mice. (E) Localized reactionary dentin formation (asterisk) in a Pax9neo/neo molar. The calciotraumatic line (arrowheads) separates reactionary dentin from normal dentin. (F) Massive reactionary and reparative dentin formation (asterisk) in a Pax9neo/lacZ molar. Note also the lack of a distinct predentin layer in Pax9neo/neo and Pax9neo/lacZ molars (arrows in E and F). (G–I) Sagittal sections of lower incisors of adult mice. (G) Wild-type lower incisors show a thick layer of dentin and enamel, the latter indicated by the enamel space after decalcification. (H) The pulp chamber of Pax9neo/neo incisors is enlarged and a thin layer of dentin and pre-enamel is formed (arrow). (I) The enamel layer is missing (arrowhead) in older Pax9neo/neo incisors. (J–L) Higher magnification of the labial side of lower incisors. (J) Polarized ameloblasts and a thick layer of unmineralized pre-enamel are present in a wild-type incisor. (K) In young Pax9neo/neo incisors, the ameloblasts are cuboidal in shape and have secreted only a thin layer of pre-enamel. (L) In older Pax9neo/neo incisors, ameloblasts are completely missing and an ectopic periodontal ligament connects dentin with the mandibular bone. am, ameloblasts; bo, bone; de, dentin; en, enamel; es, enamel space; od, odontoblasts; pd, predentin; pe, pre-enamel; pu, pulp; pl, periodontal ligament. Scale bars: (A) 500 µm; (A, inset) 20 µm; (D and G) 100 µm; (J) 20 µm.

 
Although the molars of the hypomorphic Pax9 mutants apparently form normal enamel, inspection of the first lower molars in situ as well as ground sections consistently revealed a significant attrition (N=22 for Pax9neo/neo and N=4 for Pax9neo/lacZ), resulting in the secondary absence of a distinct cusp pattern (Figs 3L and 5C). The attrition is consistently more pronounced in older mutant mice, and Pax9neo/lacZ mutants are more severely affected than Pax9neo/neo mutants. Histological analysis suggests that the increased attrition initiates the formation of reparative dentin, which was frequently observed near the occlusal surface of the first molar of Pax9neo/lacZ mutants (Fig. 5F). Irregular dentin formation in Pax9neo/lacZ mutants is also demonstrated by the presence of ‘contour lines of Owen’, which represent exaggerated incremental lines indicative of a pathologic condition (inset in Fig. 5C). A mixture of reactionary dentin (containing dentinal tubules) and reparative dentin (missing dentinal tubules) was found in molars of Pax9neo/neo mutants (Fig. 5E), whereas reactionary dentin only was occasionally seen in 5-month-old control mice (data not shown).

Reduction of Pax9 gene dosage affects tooth development at multiple stages
Histological analysis of Pax9neo/lacZ mutants at newborn stage (P0) revealed that the rudiments of all third molars are missing, indicating that the development of these teeth is arrested prior to the bud stage (Fig. 6H) (data not shown). The lower second molars of these mutants are arrested at the bud stage (Fig. 6G), whereas the lower first molars have proceeded to the bell stage but lack the polarization and alignment of ameloblasts and odontoblasts (Fig. 6F). In contrast, upper first and second molars of Pax9neo/lacZ mutants are mainly affected by growth retardation at P0 (data not shown). This is in agreement with the dental scoring of adult mutant mice showing that the upper dentition is consistently less affected (Fig. 4B).



View larger version (123K):
[in this window]
[in a new window]
  		Figure 6. The development of individual molars is affected at different stages in Pax9neo/lacZ mice. (A–H) Frontal sections of developing teeth in wild-type (A–D) and Pax9neo/lacZ (E–H) newborn mice. (E) The lower incisors do not develop in Pax9neo/lacZ mice and ectopic bone is formed instead. (B) In wild-type mice, the lower first molar has progressed to the early secretory stage, at which odontoblasts and ameloblasts show typical elongated cell shape and polarized cell alignment and start to secrete extracellular matrix (inset). (F) In Pax9neo/lacZ mice, the first molar has developed to the bell stage and matrix deposition is missing (arrows in inset). (C) The second molar has reached the bell stage in the wild-type. (G) The mutant second molar is arrested at the bud stage (arrow). (D) The third molar has developed to the late bud stage in the wild-type. (H) The third molar bud is absent in the mutant. am, ameloblasts; bo, bone; mc, Meckel's cartilage; od, odontoblasts; pd, predentin; pu, pulp. Scale bars: (A and D) 500 µm; (B, inset) 50 µm.

 
The requirement for a higher Pax9 gene dosage during the development of the lower second and third molars is associated with early morphological changes in the posterior part of the molar field in Pax9neo/neo mutants and Pax9neo/lacZ mutants (Fig. 7). Whereas the dental papilla of the first molar has formed in both mutants at E13.5 (Fig. 7B and C), the size of mesenchymal cell condensations underlying the posterior part of the dental lamina is clearly reduced (Fig. 7H and I). In wild-type embryos, the mesenchymal cell condensations forming the future dental papillae stain strongly for Pax9 protein (Fig. 7J), and the staining intensity is gradually reduced in Pax9neo/neo and Pax9neo/lacZ mutants (Fig. 7K and L). Interestingly, we found that in both mutants, some cells in the near vicinity of the dental epithelium express high levels of Pax9 protein (Fig. 7K and L). Although we have not been able to directly quantify the Pax9 protein levels, the data suggest that within single cells the level of Pax9 protein reduction does not strictly follow the levels of Pax9 mRNA reduction initially observed in E10.5 embryos. This indicates the presence of a regulatory, post-transcriptional feedback mechanism, which may compensate for Pax9 gene dosage reduction.



View larger version (166K):
[in this window]
[in a new window]
  		Figure 7. Decreased Pax9 protein levels correlate with reduced mesenchymal cell condensation in the posterior dental field. (A–C) Frontal sections of lower first molar bud stage in mouse embryos at E13.5. The size of the dental papilla (dp) is outlined and is similar in all three genotypes. (D–F) Pax9 immunohistochemistry of adjacent sections to those shown in (A–C). Pax9 protein levels are consistently decreased in dental mesenchymal cells of both mutants. Note that the first molar is the least affected tooth in the lower dentition of both mutants. (G–I) Frontal sections of the posterior part of the dental lamina (dl) and underlying dental mesenchyme in mouse embryos at E13.5. The area of condensing mesenchymal cells is outlined and is reduced in both mutants. This reduction is more pronounced in Pax9neo/lacZ mutants and the dental lamina is less thickened. (J–L) Pax9 immunohistochemistry of adjacent sections to those shown in (G–I). The decrease in Pax9 protein levels correlates with the reduced number of condensing mesenchymal cells.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 REFERENCES
 
Genetic features of the Pax9neo allele
The hypomorphic Pax9neo allele provides several useful genetic characteristics similar to those described in other ‘allelogenic’ mouse strains (25Go,26Go). First, the insertion of a neo-cassette into intron 2 of Pax9 introduces an artificial splice acceptor site that is utilized alternatively to the natural splice site, thereby efficiently interfering with normal splicing of the Pax9 gene. Secondly, the neo-cassette is removable in vivo by Flp-mediated recombination of the FRT sites (28Go) flanking the neo-cassette, resulting in the reversion to a Pax9 allele (Pax9flox) with wild-type characteristics (data not shown). The presence of loxP sites allows tissue-specific inactivation of Pax9 and successful excision of exons 1 and 2 in vivo was demonstrated by crossing Pax9flox mice with mice expressing Cre recombinase under the control of the Wnt1 promoter (data not shown). Finally, the combination of the hypomorphic allele Pax9neo with the null allele Pax9lacZ results in an allelic series that allows studying the effects of a gradual reduction of Pax9 gene dosage in vivo. However, the strong reduction of wild-type Pax9 mRNA to levels <50% in phenotypically normal Pax9+/neo and Pax9+/lacZ E10.5 embryos (Fig. 2D) was unexpected and necessitates further investigations about the molecular feedback mechanisms that regulate the level of Pax9 transcription during development.

The hybrid Pax9neo mRNA predicts a protein that contains the DNA-binding paired domain fused to the full-length neomycin phosphotransferase encoded by the neo gene and thus lacks the C-terminal, transactivating domain of Pax9. A similar mutation, generating an aberrant C-terminal region of the human PAX9 protein, has been previously described (5Go) and the resulting dental phenotype in these patients was essentially indistinguishable to that caused by presumed null alleles of PAX9. Thus, we consider the Pax9neo allele to generate a functionally inactive protein, although at present it cannot be entirely ruled out that the Pax9neo fusion protein has a residual, transactivating function or may compete for Pax9 binding sites in Pax9 target genes.

Reduction of Pax9 gene dosage in mice results in a broad spectrum of dental defects
Inter- and intra-familial variations in the number of missing teeth caused by mutations in the human PAX9 gene are consistently observed (2Go) and the different severities of the dental phenotypes between, and within, Pax9neo/neo and Pax9neo/lacZ mutants confirm these findings. The lack of a clear genotype–phenotype correlation strongly suggests that the genetic background plays a major role in determining the phenotypic severity of both humans and mice, and the novel mouse models described in this study provide a tool to identify relevant genetic modifiers that interact with Pax9 and thereby exacerbate, or attenuate, the clinical outcome of oligodontia.

In contrast to humans, the developing lower incisor of mice appears to be extremely susceptible to reduced Pax9 gene dosage. This may be explained by the fact that lower incisors in rodents grow much faster when compared with all other teeth, eventually spanning the whole anterior–posterior axis of the lower jaw. Permanent production of dentin and enamel in the incisors of adult mice is believed to depend on continuous mesenchymal–epithelial interactions at the apical pole, the cervical loop. Mesenchymal expression of Fgf3 and Fgf10 was shown to activate epithelial proliferation in stem cells and transit-amplifying epithelial cells of the cervical loop (29Go–31Go). Preliminary analysis revealed an overlapping Pax9 expression pattern with those of Fgf3 and Fgf10, suggesting that Pax9 function could be involved in this stimulation (data not shown).

The delayed onset of dentin deposition, absence of a distinct pre-dentin layer and the frequently observed attrition followed by reparative dentin formation in the molars of Pax9neo/neo and Pax9neo/lacZ mutants demonstrates a role of Pax9 for normal dentin formation. Whereas an altered dentin structure has not been reported in oligodontia patients caused by PAX9 mutations, individual dietary and oral hygiene practices as well as fluoride exposure may complicate a systematic scoring. Nevertheless, we suggest the allelic series of Pax9 mouse mutants to represent a useful in vivo model for analyses of molecular pathways involved in reparative dentin formation. In addition, further investigations are required to identify the molecular downstream events of Pax9 gene dosage reduction on tooth development, on maintaining a stem cell compartment in the continuously growing incisors and on tooth mineralization. The structural defects of the molars could be related to abnormal dentin, abnormal enamel or to a combination of both. Because at later stages of tooth formation Pax9 protein is detectable in odontoblasts and ameloblasts (data not shown), a cell type-specific, conditional inactivation of Pax9 is required to distinguish between these possibilities.

Oligodontia and the role of Pax9 in dental fields
The human dentition is divided into four dental fields (incisors, canine, premolars and molars) within each quadrant of the jaws and when a tooth is missing, all posterior teeth within this dental field are also often absent. Accordingly, if only a single tooth is missing, it is typically the most posterior tooth within its dental field. These patterns are observed for missing molars and premolars, but not for incisors, in oligodontia caused by mutations in PAX9 (32Go,33Go). Our analysis showed that the absence of molars in the allelic series of Pax9 mutant mice follows the same pattern (Fig. 4), but it is important to note that the missing molars are not arrested at the same stage during development: in Pax9neo/lacZ mutants, the development of the third molar is arrested prior to the bud stage, whereas the second and first molars develop to the bud stage and bell stage, respectively (Fig. 6). In contrast, in Pax9-deficient mice, the development of the first molar is arrested at the bud stage and that of the second and third molar is not initiated (19Go). The arrest at different stages demonstrates that Pax9 is required at multiple stages during odontogenesis. Moreover, in Pax9neo/lacZ mutants the frequency of a specific molar to be missing correlates inversely with the developmental stage at which this molar arrests (Figs 4 and 8), suggesting that the reduction of Pax9 gene dosage affects the development of the dental field as a whole. Normally, the developmental stage of a posterior tooth is associated with a more advanced developmental progression of the preceding tooth of the same dental field. Interestingly, our data show that this relation is also strictly followed after a drastic reduction of Pax9 gene dosage (Fig. 8), resulting in disturbed tooth development and, eventually, in oligodontia. A different pathologic mechanism is apparently responsible for oligodontia caused by mutations in the homeobox gene MSX1, which can result in the absence of the first molars, whereas the second molars were formed in the same individuals (34Go–36Go). Formally, our data are compatible with the hypothesis that the requirement for a minimal Pax9 gene dosage increases along the anterior–posterior axis of a dental field. Developmentally, however, the posterior teeth within a dental field may be affected secondarily. In such a scenario, the initial level of Pax9 expression shortly after tooth induction would be involved in controlling the number of odontogenic, mesenchymal cells and a reduced cell number would be sufficient for the formation of a normal dental papilla of the first molar only (Fig. 7). Therefore, an initially restricted pool of odontogenic, mesenchymal cells could limit the number of teeth that eventually complete their development and such a restriction would predominantly affect the later developing, posterior molars.



View larger version (19K):
[in this window]
[in a new window]
  		Figure 8. Relation between the frequency of absent individual molars and their developmental stages. In normal mice (equivalent to the normal human population), the risk for missing a specific molar is conversely related to its developmental progression. In Pax9neo/lacZ mutant mice (suggested to correspond to oligodontia patients), the first and second molars have progressed to the bell and bud stage, respectively, which are characteristic for the second and third molars in wild-type mice. Thus, a reduction of Pax9 gene dosage results in a developmental delay affecting the dental field as a whole and the risk for individual molars to be absent increases correspondingly (Fig. 4). In its extreme form (Pax9lacZ/lacZ mutant mice), the absence of Pax9 causes an early arrest of developmental progression and the ‘risk’ of tooth agenesis increases to 100%.

 
In addition to providing an in vivo model system for oligodontia, our findings may also have implications for explaining evolutionary changes in patterning the dentition. Fossil records indicate that reductions of teeth within a dental field tend to occur in the reverse order of how teeth are formed during development (32Go) and our data show that a gradual decrease of Pax9 gene dosage affects molar development in the same pattern (Figs 4 and 8). While additional genes implicated in dental patterning clearly have been involved in parallel, it appears plausible that variations of Pax9 gene dosage have contributed to the establishment of different dental formulas, which is considered to be a driving force for the diversification of mammalian species.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
REFERENCES
 
Generation of Pax9neo mice
The Pax9 gene-targeting construct (Fig. 1A) was cloned by inserting a short 3'-arm and a long 5'-arm into the targeting vector pPNT4 (37Go). The 1.3 kb 3'-arm was generated by long-range PCR using genomic DNA isolated from TBV-2 mouse ES cells (38Go) as template, Platinum Pfx DNA polymerase (Invitrogen, UK) and the primers SA-F and SA-R (Table 1). These primers introduced BamHI and EcoRI restriction sites, which were used to insert the 3'-arm into pPNT4 to produce pPNT4-SA. For generation of the 5'-arm, a 5.6 kb SalI/NotI fragment from a Pax9 subclone previously used to create Pax9lacZ mice (19Go) was modified by inserting an oligonucleotide linker containing a loxP site and a diagnostic XbaI site into an EcoRI site located 1.1 kb upstream of Pax9 exon 1. This modified SalI/NotI fragment was then inserted into the compatible restriction sites of pPNT4-SA to generate pPNT4-SA/LA. To extend the long-arm, a 4.6 kb fragment upstream of the 5.6 kb SalI/NotI fragment was generated by long-range PCR using primers LA-F and LA-R (Table 1), cut with SalI and then introduced into the SalI site of pPNT4-SA/LA to generate the final Pax9 targeting construct. The Pax9 coding sequence and the loxP and FRT sites were confirmed by DNA sequencing. The targeting construct was linearized with SfiI and electroporated into TBV-2 ES cells (38Go). ES cells were cultured on a layer of SNLP (clone 76/7-4) feeder cells (39Go). Culture conditions, electroporation, G418 and ganciclovir selection, and isolation, propagation and analysis of resistant ES cell colonies were essentially as described (40Go). DNA of resistant ES cell clones was analysed by nested PCR using external primers P9ex-F and P9ex-R in the first PCR and internal primers P9in-F and P9in-R in the second PCR reaction (Table 1). Candidate clones were confirmed by Southern blot analysis (41Go) using the external probes A and B (Fig. 1A and B) and an internal neo probe (data not shown). Details are available upon request.


View this table:
[in this window]
[in a new window]
  		Table 1. PCR primers used for cloning, genotyping and RT–PCR
 
The targeting frequency was 0.5% and two correctly targeted ES cell clones (1C9 and 3C7; Fig. 1B) were microinjected into C57BL/6 blastocysts (42Go). Resulting chimeric males were mated to C57BL/6 females and offspring with agouti coat colour, indicating germline transmission, were genotyped by Southern blot analysis as described earlier. Two mouse lines were established and backcrossed to C57BL/6 or CD-1 mice (Harlan Tekland, UK and Charles River, UK).

Mouse husbandry and genotyping
Mice were housed under pathogen-free conditions and received a pelleted diet (2019S, Harlan Tekland) and water ad libitum. Toothless mice were fed a soaked diet (RM3, Special Diet Services, UK) and the continuously growing upper incisors of Pax9neo/neo mice were regularly clipped to prevent overgrowth because of hypoplasia of the lower incisors. Embryos were staged by taking mid-day on the day of vaginal plug detection as embryonic day 0.5 (E0.5). Pax9neo mice were genotyped by PCR analysis of ear punch tissue (42Go) using primers P9lox1-F and P9lox1-R (Table 1), which span the loxP site upstream of exon 1. The presence of the neo-cassette was additionally confirmed using primers neo2-F and neo2-R (Table 1). Previously generated Pax9lacZ mice (19Go) were maintained on an inbred C57BL/6 genetic background and genotyped by PCR using the primers P9lacZ-F, P9lacZ-R1 and P9lacZ-R2 (Table 1). Mice analysed in this study (generations N1F1 and N2F1) were on a mixed C57BL/6x129S2/SvPas genetic background (Pax9neo/neo) or on a mixed C57BL/6x129S2/SvPasxCD-1 genetic background (Pax9neo/lacZ). All procedures were carried out under personal and project licences issued by the Home Office, UK and were approved by the Local Ethics Committee.

Semi-quantitative RT–PCR and quantitative real-time RT–PCR analysis
Tissue was homogenized and total RNA was isolated from E12.5 heads or whole E10.5 mouse embryos using TRIZOL reagent (Invitrogen). Two micrograms of RNA was reverse transcribed into cDNA using SuperScript II Reverse Transcriptase (Invitrogen), according to manufacturer's instructions. For semi-quantitative RT–PCR analysis (Fig. 2C), the following PCR reactions were performed: Pax9 (wild-type) with primers E2-F(a) and E4-R; Pax9neo with primers E2-F(b) and neo-R; Glyceraldehyde-3-phosphate dehydrogenase (Gapdh) with primers Gapdh-F and Gapdh-R (Table 1). DNA sequence of the RT–PCR products was analysed using BioEdit v7.0.3 software (43Go).

Quantitative real-time RT–PCR (‘TaqMan’) analysis was performed in an ABI PRISM 7700 Sequence Detector (Applied Biosystems, USA) using commercial TaqMan Gene Expression Assays (P/N: 4331182; Applied Biosystems) for Pax9 (Mm00440629_m1) and the endogenous control Gapdh (Mm99999915_g1). Singleplex 50 µl reactions contained 1x TaqMan Universal PCR Master Mix (P/N: 4304437), 1x ‘TaqMan Assay Mix’ (unlabelled PCR primers and FAM dye-labelled TaqMan probe) and diluted cDNA. Amplification conditions were as described by the manufacturer. Raw data were collected and analysed using Sequence Detection System v1.7 software (Applied Biosystems). Two experiments were performed and each was done in triplicate for each genotype of the Pax9 mutant allelic series (Fig. 2D). Pax9 expression levels were normalized to Gapdh expression levels and the standard deviation was calculated using the Comparative CT Method as described (44Go).

Ground sections of mouse teeth
Jaws of adult mice were dissected in phosphate-buffered saline (PBS) and fixed for at least 3 days in 4% neutral-buffered formalin, washed in running water to remove fixative, radiographed and photographed for record purposes. Ground sections were prepared following the technique described in (45Go). The jaws were embedded in polyester resin and 150–180 µm thick longitudinal sections were taken that bisected the three molars using a Microslice 2 annular saw. The ground section was then attached to a microscope slide with dental sticky wax and lapped with 3 µm calcined alumina using a Logitech PM2A precision lapping machine so that the plane of section passed through the dentine horns. This surface was then cleaned in an ultrasonic bath, polished with 0.1 µm diamond paste, cleaned and air-dried after removal from the slide with gentle heat. The cleaned polished surface of the ground section was then zero bonded to a fresh microscope slide with UV384 resin and lapped to a final thickness of 60–80 µm. After surface debris was removed, the sections were dehydrated through a graded series of alcohol baths and mounted in DPX mounting medium. Sections prepared from at least two adult mice per genotype (wild-type, Pax9neo/neo and Pax9neo/lacZ) were then studied using polarized, incident and transmitted light microscopy.

Histology and Pax9 immunohistochemistry
Specimens were prepared in PBS and fixed overnight in 4% paraformaldehyde/PBS. The heads of newborn mice and jaws of adult mice were decalcified (0.15 M EDTA/4% formalin, pH 7) for 2 and 6 weeks, respectively. Specimens were dehydrated in a graded series of ethanol, processed in a Shandon Pathcentre tissue processor (Thermo Electron Corporation, UK) and embedded in paraffin. Tissue sections of 5 µm thickness were cut using a Leica RM 2135 microtome (Leica Microsystems, Germany) and mounted on Superfrost plus microscopic slides (BDH, UK). Haematoxilin and eosin staining was performed on at least four jaws per genotype (wild-type, Pax9neo/neo and Pax9neo/lacZ) and per stage using standard procedures and sections were mounted in DPX mounting medium (RA Lamb, UK). Pax9 immunohistochemistry was carried out on paraffin sections using a monoclonal anti-Pax9 antibody as described (46Go). Alkaline phosphatase activity was visualized using Fast Red (Sigma, UK) as a substrate. All slides were processed simultaneously. Pictures were taken with an Axiocam HRc camera on an Axioplan 2 microscope using AxioVision Software v.4.3 (Carl Zeiss, Germany) and processed using Adobe Photoshop v.7.0 (Adobe Systems Inc.).


   ACKNOWLEDGEMENTS
 
We thank Pamela Walton for preparing ground sections and John Whitworth for help and discussion during the initial phase of scoring the dental phenotypes in mice. We also thank Marcus Conrad for the pPNT4 vector, Allan Bradley for SNLP feeder cells, Veronique Blanquet and Wolfgang Wurst for the ES cell line TBV-2 and Mikhajlo Zubko and David Lydall for introduction to the ABI PRISM 7700 Sequence Detector. This work was funded by the Wellcome Trust (UK).

Conflict of Interest statement. None declared.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 

  1. Rose, J. (1966) A survey of congenitally missing teeth, excluding third molars, in 6000 orthodontic patients. Dent. Pract. Dent. Rec., 17, 289–295.

  2. Mostowska, A., Kobielak, A. and Trzeciak, W.H. (2003) Molecular basis of non-syndromic tooth agenesis: mutations of MSX1 and PAX9 reflect their role in patterning human dentition. Eur. J. Oral Sci., 111, 365–370.[CrossRef][ISI][Medline]

  3. Stockton, D.W., Das, P., Goldenberg, M., D'Souza, R.N. and Patel, P.I. (2000) Mutation of PAX9 is associated with oligodontia. Nat. Genet., 24, 18–19.[CrossRef][ISI][Medline]

  4. Kobielak, A., Kobielak, K., Wisniewski, A.S., Mostowska, A., Biedziak, B. and Trzeciak, W.H. (2001) The novel polymorphic variants within the paired box of the PAX9 gene are associated with selective tooth agenesis. Folia Histochem. Cytobiol., 39, 111–112.[Medline]

  5. Frazier-Bowers, S.A., Guo, D.C., Cavender, A., Xue, L., Evans, B., King, T., Milewicz, D. and D'Souza, R.N. (2002) A novel mutation in human PAX9 causes molar oligodontia. J. Dent. Res., 81, 129–133.[Abstract/Free Full Text]

  6. Das, P., Hai, M., Elcock, C., Leal, S.M., Brown, D.T., Brook, A.H. and Patel, P.I. (2003) Novel missense mutations and a 288-bp exonic insertion in PAX9 in families with autosomal dominant hypodontia. Am. J. Med. Genet., 118A, 35–42.

  7. Jumlongras, D., Lin, J.Y., Chapra, A., Seidman, C.E., Seidman, J.G., Maas, R.L. and Olsen, B.R. (2004) A novel missense mutation in the paired domain of PAX9 causes non-syndromic oligodontia. Hum. Genet., 114, 242–249.[CrossRef][ISI][Medline]

  8. Nieminen, P., Arte, S., Tanner, D., Paulin, L., Alaluusua, S., Thesleff, I. and Pirinen, S. (2001) Identification of a nonsense mutation in the PAX9 gene in molar oligodontia. Eur. J. Hum. Genet., 9, 743–746.[CrossRef][ISI][Medline]

  9. Lammi, L., Halonen, K., Pirinen, S., Thesleff, I., Arte, S. and Nieminen, P. (2003) A missense mutation in PAX9 in a family with distinct phenotype of oligodontia. Eur. J. Hum. Genet., 11, 866–871.[CrossRef][ISI][Medline]

  10. Das, P., Stockton, D.W., Bauer, C., Shaffer, L.G., D'Souza, R.N., Wright, T. and Patel, P.I. (2002) Haploinsufficiency of PAX9 is associated with autosomal dominant hypodontia. Hum. Genet., 110, 371–376.[CrossRef][ISI][Medline]

  11. Klein, M.L., Nieminen, P., Lammi, L., Niebuhr, E. and Kreiborg, S. (2005) Novel mutation of the initiation codon of PAX9 causes oligodontia. J. Dent. Res., 84, 43–47.[Abstract/Free Full Text]

  12. Thesleff, I. (2003) Developmental biology and building a tooth. Quintessence Int., 34, 613–620.[ISI][Medline]

  13. Miletich, I. and Sharpe, P.T. (2003) Normal and abnormal dental development. Hum. Mol. Genet., 12, R69–R73.[Abstract/Free Full Text]

  14. Pispa, J. and Thesleff, I. (2003) Mechanisms of ectodermal organogenesis. Dev. Biol., 262, 195–205.[CrossRef][ISI][Medline]

  15. Tucker, A. and Sharpe, P. (2004) The cutting-edge of mammalian development; how the embryo makes teeth. Nat. Rev. Genet., 5, 499–508.[ISI][Medline]

  16. Mina, M. and Kollar, E.J. (1987) The induction of odontogenesis in non-dental mesenchyme combined with early murine mandibular arch epithelium. Arch. Oral Biol., 32, 123–127.[CrossRef][ISI][Medline]

  17. Lumsden, A.G. (1988) Spatial organization of the epithelium and the role of neural crest cells in the initiation of the mammalian tooth germ. Development, 103 (suppl), 155–169.

  18. Jernvall, J., Aberg, T., Kettunen, P., Keranen, S. and Thesleff, I. (1998) The life history of an embryonic signaling center: BMP-4 induces p21 and is associated with apoptosis in the mouse tooth enamel knot. Development, 125, 161–169.[Abstract]

  19. Peters, H., Neubuser, A., Kratochwil, K. and Balling, R. (1998) Pax9-deficient mice lack pharyngeal pouch derivatives and teeth and exhibit craniofacial and limb abnormalities. Genes Dev., 12, 2735–2747.[Abstract/Free Full Text]

  20. Neubuser, A., Peters, H., Balling, R. and Martin, G.R. (1997) Antagonistic interactions between FGF and BMP signaling pathways: a mechanism for positioning the sites of tooth formation. Cell, 90, 247–255.[CrossRef][ISI][Medline]

  21. Satokata, I. and Maas, R. (1994) Msx1 deficient mice exhibit cleft palate and abnormalities of craniofacial and tooth development. Nat. Genet., 6, 348–356.[CrossRef][ISI][Medline]

  22. Chen, Y., Bei, M., Woo, I., Satokata, I. and Maas, R. (1996) Msx1 controls inductive signaling in mammalian tooth morphogenesis. Development, 122, 3035–3044.[Abstract]

  23. Zhao, X., Zhang, Z., Song, Y., Zhang, X., Zhang, Y., Hu, Y., Fromm, S.H. and Chen, Y. (2000) Transgenically ectopic expression of Bmp4 to the Msx1 mutant dental mesenchyme restores downstream gene expression but represses Shh and Bmp2 in the enamel knot of wild-type tooth germ. Mech. Dev., 99, 29–38.[CrossRef][ISI][Medline]

  24. Garn, S., Lewis, A. and Bonne, B. (1961) Third molar polymorphism and the timing of tooth formation. Nature, 192, 989.[Medline]

  25. Meyers, E.N., Lewandoski, M. and Martin, G.R. (1998) An Fgf8 mutant allelic series generated by Cre- and Flp-mediated recombination. Nat. Genet., 18, 136–141.[CrossRef][ISI][Medline]

  26. Nagy, A., Moens, C., Ivanyi, E., Pawling, J., Gertsenstein, M., Hadjantonakis, A.K., Pirity, M. and Rossant, J. (1998) Dissecting the role of N-myc in development using a single targeting vector to generate a series of alleles. Curr. Biol., 8, 661–664.[CrossRef][ISI][Medline]

  27. Sheng, G., Harris, E., Bertuccioli, C. and Desplan, C. (1997) Modular organization of Pax/homeodomain proteins in transcriptional regulation. Biol. Chem., 378, 863–872.[Medline]

  28. Rodriguez, C.I., Buchholz, F., Galloway, J., Sequerra, R., Kasper, J., Ayala, R., Stewart, A.F. and Dymecki, S.M. (2000) High-efficiency deleter mice show that FLPe is an alternative to Cre-loxP. Nat. Genet., 25, 139–140.[CrossRef][ISI][Medline]

  29. Harada, H., Kettunen, P., Jung, H.S., Mustonen, T., Wang, Y.A. and Thesleff, I. (1999) Localization of putative stem cells in dental epithelium and their association with Notch and FGF signaling. J. Cell Biol., 147, 105–120.[Abstract/Free Full Text]

  30. Kettunen, P., Laurikkala, J., Itaranta, P., Vainio, S., Itoh, N. and Thesleff, I. (2000) Associations of FGF-3 and FGF-10 with signaling networks regulating tooth morphogenesis. Dev. Dyn., 219, 322–332.[CrossRef][ISI][Medline]

  31. Harada, H., Toyono, T., Toyoshima, K., Yamasaki, M., Itoh, N., Kato, S., Sekine, K. and Ohuchi, H. (2002) FGF10 maintains stem cell compartment in developing mouse incisors. Development, 129, 1533–1541.

  32. Line, S.R. (2003) Variation of tooth number in mammalian dentition: connecting genetics, development, and evolution. Evol. Dev., 5, 295–304.[CrossRef][ISI][Medline]

  33. Pemberton, T.J., Das, P. and Patel, P.I. (2005) Hypodontia: genetics and future perspectives. Braz. J. Oral Sci., 4, 695–706.

  34. Vastardis, H., Karimbux, N., Guthua, S.W., Seidman, J.G. and Seidman, C.E. (1996) A human MSX1 homeodomain missense mutation causes selective tooth agenesis. Nat. Genet., 13, 417–421.[CrossRef][ISI][Medline]

  35. van den Boogaard, M.J., Dorland, M., Beemer, F.A. and van Amstel, H.K. (2000) MSX1 mutation is associated with orofacial clefting and tooth agenesis in humans. Nat. Genet., 24, 342–343.[CrossRef][ISI][Medline]

  36. Jumlongras, D., Bei, M., Stimson, J.M., Wang, W.F., DePalma, S.R., Seidman, C.E., Felbor, U., Maas, R., Seidman, J.G. and Olsen, B.R. (2001) A nonsense mutation in MSX1 causes Witkop syndrome. Am. J. Hum. Genet., 69, 67–74.[CrossRef][ISI][Medline]

  37. Conrad, M., Brielmeier, M., Wurst, W. and Bornkamm, G.W. (2003) Optimized vector for conditional gene targeting in mouse embryonic stem cells. Biotechniques, 34, 1136–1138.[Medline]

  38. Wiles, M.V., Vauti, F., Otte, J., Fuchtbauer, E.M., Ruiz, P., Fuchtbauer, A., Arnold, H.H., Lehrach, H., Metz, T., von Melchner, H. et al. (2000) Establishment of a gene-trap sequence tag library to generate mutant mice from embryonic stem cells. Nat. Genet., 24, 13–14.[CrossRef][ISI][Medline]

  39. McMahon, A.P. and Bradley, A. (1990) The Wnt-1 (int-1) proto-oncogene is required for development of a large region of the mouse brain. Cell, 62, 1073–1085.[CrossRef][ISI][Medline]

  40. Matise, M.P., Auerbach, W. and Joyner, A.L. (2000) Production of targeted embryonic stem cell clones. In Joyner, A.L. (ed.), Gene Targeting: A Practical Approach, 2nd edn. Oxford University Press, Oxford, UK, pp. 101–132.

  41. Church, G.M. and Gilbert, W. (1984) Genomic sequencing. Proc. Natl Acad. Sci. USA, 81, 1991–1995.[Abstract/Free Full Text]

  42. Nagy, A., Gertsenstein, M., Vintersten, K. and Behringer, R. (2003) Manipulating the Mouse Embryo: A Laboratory Manual, 3rd edn. Cold Spring Harbor Laboratory Press, New York, USA.

  43. Hall, T.A. (1999) BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucl. Acids Symp. Ser., 41, 95–98.

  44. Livak, K.J. and Schmittgen, T.D. (2001) Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) method. Methods, 25, 402–408.[CrossRef][ISI][Medline]

  45. Reid, D.J., Schwartz, G.T., Dean, C. and Chandrasekera, M.S. (1998) A histological reconstruction of dental development in the common chimpanzee, Pan troglodytes. J. Hum. Evol., 35, 427–448.[CrossRef][ISI][Medline]

  46. Gerber, J.K., Richter, T., Kremmer, E., Adamski, J., Hofler, H., Balling, R. and Peters, H. (2002) Progressive loss of PAX9 expression correlates with increasing malignancy of dysplastic and cancerous epithelium of the human oesophagus. J. Pathol., 197, 293–297.[Medline]

  47. Rozen, S. and Skaletsky, H. (2000) Primer3 on the WWW for general users and for biologist programmers. Methods Mol. Biol., 132, 365–386.[Medline]


Add to CiteULike CiteULike