| Sign In to gain access to subscriptions and/or personal tools. |
MSX1, PAX9, and TGFA Contribute to Tooth Agenesis in Humans
1 Departments of Pediatrics, Correspondence: * corresponding author, alexandre-vieira{at}uiowa.edu
In this study, we sought to determine the association between tooth agenesis and DNA sequence variation in the genes MSX1 and PAX9 in an ethnically diverse human population. Since cleft lip/palate is also associated with both tooth agenesis and the gene TGFA, we included TGFA in the analysis as well. Cheek swab samples were obtained for DNA analysis from 116 case/parent trios. Probands had at least one developmentally missing tooth, excluding third molars. Genotyping was performed by single-strand conformational polymorphism or kinetic polymerase chain-reaction assays. Transmission distortion of the marker alleles and DNA sequence analysis was performed. Results showed that tooth agenesis is associated with markers of the genes MSX1 and TGFA. No mutations were found in MSX1 or PAX9 coding regions. There were statistically significant data suggesting that MSX1 interacts with PAX9. These findings suggest that MSX1, PAX9, and TGFA play a role in isolated dental agenesis.
Key Words: anodontia • hypodontia • homeobox • transcription factors • transforming growth factor alpha
Human tooth agenesis can contribute to masticatory dysfunction, speech alterations, and esthetic problems, in addition to malocclusion. Mutations in MSX1 and PAX9 have been shown in forms of human tooth agenesis. MSX1 is strongly expressed in the dental mesenchyme and is eliminated from the dental epithelia during the bud, cap, and bell stages of tooth development (Mackenzie et al., 1991). Msx1 homozygous knockout transgenic mice have craniofacial anomalies that include absent teeth (Satokata and Maas, 1994). In humans, mutations in MSX1 coding regions cause tooth agenesis of various types of teeth, preferentially premolars (reviewed by Vieira, 2003). PAX9 is expressed in the neural-crest-derived mesenchyme of the maxillary and mandibular arches, and contributes to palate and tooth formation (Peters and Balling, 1999). Pax9-deficient mice present with cleft palate, agenesis of all teeth, and other craniofacial anomalies (Peters et al., 1998). In humans, mutations in PAX9 coding regions (Stockton et al., 2000; Nieminen et al., 2001; Frazier-Bowers et al., 2002; Das et al., 2003; Lammi et al., 2003; Jumlongras et al., 2004) or a PAX9 deletion (Das et al., 2002) cause preferential tooth agenesis of molars. The only exception is a mutation described in a Polish family presenting with a less severe molar phenotype (Mostowska et al., 2003). No studies have attempted to address human tooth agenesis by means of a genetic epidemiological approach in an ethnically diverse population. We measured the association of MSX1, PAX9, and a third gene expressed during craniofacial development, TGFA, with non-syndromic tooth agenesis, using a population from Rio de Janeiro, Brazil.
From March, 2001, to April, 2003, through dental records from the archives of the Department of Pediatric Dentistry and Orthodontics, School of Dentistry, Federal University of Rio de Janeiro, and from private practitioners located in Rio de Janeiro city, Brazil, we selected 116 nuclear families (father-mother-affected child) whose proband presented with at least one permanent tooth congenitally absent, with the exception of third molars. We used radiographs to confirm the diagnosis. Information regarding family history for tooth agenesis was obtained, in most cases, from the mothers and was collected during the dental visit. Positive family history was defined as any proband’s relative with reported congenital tooth agenesis (excepting third molars). Cases with additional non-dental abnormalities or cases with uncertain diagnosis were excluded from study. Written informed consent was obtained from each family member above the age of 14 yrs. Children from 7 to 13 yrs of age signed an assent document and also had to have their parents’ written consent. Both parents gave written consent to children age 6 or younger. This study had Federal University of Rio de Janeiro Hospital Institutional Review Board approval. Cheek swab DNA was obtained from 116 family-trios and extracted by modifications of published protocols. Polymerase chain-reactions were performed, and allelic variants of MSX1 (Padanilam et al., 1992), PAX9 (rs11847165), and TGFA (Basart et al., 1994; Machida et al., 1999) were tested. Subjects were genotyped by means of single-strand conformational polymorphism analysis, mutation-detection enhancement, and polyacrylamide gel electrophoresis techniques (markers MSX1-CA and TGFA Taq1) and by kinetic polymerase chain-reaction assays (markers PAX9 C1843T, TGFA C3296T, and TGFA C3827T, as described by Shi, 2002). Samples with known genotypes were included for each gel and polymerase chain-reaction. We used a silver-stain protocol to visualize the bands for single-strand conformational polymorphism analysis (Lidral and Reising, 2002). Samples were analyzed as a total group and subgrouped in cases with at least one missing incisor, cases with at least one missing premolar, and cases with a positive family history for tooth agenesis. There were insufficient cases with missing molars for this subgroup to have adequate power for analysis. We compared father, mother, and proband genotypes to determine the transmitted alleles vs. the non-transmitted alleles. Affected based-family controls and transmission disequilibrium tests were performed (Thomson, 1988; Spielman et al., 1993). Significance was established for alpha lower than 0.05.
Mutation searches were performed bidirectionally in the coding regions and intron-exon boundaries of MSX1 and PAX9 for all affected individuals. Primers and polymerase chain-rection conditions for MSX1 were described by Lidral et al.(1998). Primers and polymerase chain-reaction conditions for PAX9 are listed in Table 1
The distribution of cases by type of tooth missing and by selected characteristics of the probands was calculated. The proportion of female cases was 64% (p = 0.003 for males vs. females). Only 14% of the cases had an accompanying tooth defect, such as hypoplastic enamel, peg-shaped upper lateral incisors, and microdontia (Table 2  ).
The affected based-family controls analysis shows that the TGFA C3827T C allele is associated with all trios (p = 0.01). For the transmission disequilibrium test analysis, the MSX1-CA 169-base-pair allele is associated with all trios (p = 0.01), as well as cases with positive family history for tooth agenesis (p = 0.04). Also, the TGFA C3827T C allele is associated with all trios (p = 0.02), and the TGFA haplotype 117-base-pair-C-C is associated with all trios (p = 0.02). Borderline association can be seen for the trios with at least one missing incisor and TGFA markers C3296T (p = 0.06) and C3827T (p = 0.051). (Table 3 summarizes these results; the Web appendix presents complete data.)
The transmission disequilibrium test provides a measure of the statistical significance of a disease marker association, but does not provide a measure of the magnitude of this association. Therefore, we calculated the magnitude of the marker associations we detected by the transmission disequilibrium test by estimating the binomial proportion noted by T (± confidence interval) and described by Mitchell (2000). The MSX1-CA 169-base-pair allele was overtransmitted (the expected proportion is 50%) in a proportion of 61.9% (± 4.7%) to all cases and 65.8% (± 7.4%) to the cases with positive family history. The TGFA C3827T C allele was overtransmitted 73% (± 8.7%) to all cases, and the TGFA haplotype 117-base-pair-C-C was overtransmitted 66.7% (± 6.5%) to all cases.
We tested for possible MSX1-TGFA, MSX1-PAX9, and PAX9-TGFA interactions by observing the transmission of the marker alleles from parents heterozygous for both of the markers. (Table 4
Finally, direct sequence analysis of MSX1 and PAX9 coding regions did not disclose any nonsense, missense, or splice-site mutations.
We report here a genetic epidemiological approach to identifying genetic factors contributing to human tooth agenesis. The affected based-family controls and transmission disequilibrium test approaches allowed us to include cases of different ethnic origins and avoided the complications of mixed ancestry that can arise in case-control study designs. South America is, for the most part, a trihybrid of Native Indians, Caucasians, and Africans, but these three groups are not equally distributed across the continent. In Rio de Janeiro, individuals with African ancestry comprise about 40% of the population, with those of Native Indian ancestry not more than 1 to 2%.
Mutations in MSX1 have previously been demonstrated in non-syndromic tooth agenesis in humans (Vastardis et al., 1996; Lidral and Reising, 2002). The family reported by Vastardis et al.(1996) presented with oligodontia involving several types of missing teeth. However, premolars were always missing, which implicates MSX1 in premolar development. In the present study, only six families presented with oligodontia (6 or more teeth missing), so there may be insufficient power to detect association between MSX1 and oligodontia. This may also explain why MSX1 is not associated with cases with 1 or more missing premolars but without true oligodontia, since MSX1 may play a more substantial role in cases of missing premolars that include oligodontia and are likely to be familial cases. Our results suggest that MSX1 plays a substantial role in familial cases of tooth agenesis. The sample size studied here is not large enough to allow for analysis by the same phenotype (i.e., missing second premolars), and future studies can examine the roles of specific genes in the development of specific types of teeth (Vieira, 2003). PAX9 has been linked to molar agenesis (Vieira, 2003), but in the present report, only 14 families presented missing molars (Table 1 This is the first report to investigate the possible association between human tooth agenesis and TGFA. The analysis combining information from the three markers provides more power for the transmission disequilibrium test and for testing for MSX1-TGFA interaction. There is evidence that the TGFA locus is associated with isolated tooth agenesis in the studied population. A borderline association could be seen between cases with at least one missing incisor and the TGFA markers C3296T and C3827T. This suggests that TGFA might play a role in cases that have tooth agenesis that include incisors. TGFA is expressed during craniofacial development (Dixon et al., 1991), and mice that are Tgfa-deficient present eye anomalies and abnormal hair but no dental anomalies (Mann et al., 1993). MSX1 and TGFA have been associated with a related developmental craniofacial defect, cleft lip and palate (reviewed by Vieira and Orioli, 2001). The same MSX1-CA 169-base-pair allele that shows association with tooth agenesis in the present study has also been associated with oral clefts in other studies. Cases that present both cleft lip or cleft lip/palate and tooth agenesis outside the cleft region have also been associated with MSX1 variants (Slayton et al., 2003). Evidence of interaction between MSX1 and TGFA in oral clefts has also been reported (Jugessur et al., 2003). The present work did not find statistically significant evidence that MSX1 and TGFA interact in human tooth agenesis. However, there was statistically significant evidence of an interaction between MSX1 and PAX9. At the bud stage, PAX9 and MSX1 are co-expressed in the mesenchyme, and the function of both genes is required for the expression of BMP4. BMP4 signaling is involved in the induction of the enamel knot, a transient signaling center of the epithelium that directs the next phase of tooth development (Peters and Balling, 1999). The failure of the maintenance of BMP4 could be implicated in the arrest of tooth development at the bud stage, similar to that which occurs in homozygous Pax9 and Msx1 mutant embryos (Satokata and Maas, 1994; Peters et al., 1998). No mutations were found in the MSX1 or PAX9 coding region, which suggests that mutations in these genes may be found in regulatory regions that still need to be characterized. Analysis of the association data does suggest that a common variant(s) contributing to tooth agenesis can be found and may require large sample sizes and additional sequencing. In addition, deletions of these genes would have been missed by the direct sequencing approach used here. Our results do not reach formal significance when corrected for multiple testing, but the relevance of these genes to tooth development and the rare mutations found in inherited forms of tooth agenesis support the findings. In conclusion, this is the first report to suggest a role for TGFA in human tooth agenesis. MSX1 is further implicated with tooth agenesis in the studied population, in agreement with findings from previous studies. An MSX1 and PAX9 interaction appears to play a role in tooth agenesis in humans. Future studies should focus on mutation identification, in coding regions as well as in regulatory sequences and functional analysis.
We are very grateful to the families who enthusiastically participated in this study. We are also indebted to the colleagues who recommended their patients and helped with the sample collection: Marcelo de Castro Costa, Laura Primo, Ivete Souza, Rogerio Gleiser, Maria Encarnação Requejo, Aline Neves, Anna Renata Barbosa, Mariana Passos, Paula Alcântara, Juliana Abdenur, Rodolfo Castro, Bianca Santiago, Lívia Soares, Cristina Lepecka, Marta Lua, Luciana Pomarico, Renata Simões, Maria Bárbara Guimarães, Daniela Della Valle, Osvaldo Costa, Roberta Barcelos, Cláudia Tavares, Silvia Malaquias, Valéria Teixeira, Rui Meira, Patrícia Mendes, and Liana Amado. Anna Renata Barbosa also provided technical support for genotyping some families. Carla Nishimura did preliminary work on PAX9. This work is supported by NIH grants 5 D43 TW05503 and DE 08559-12. This paper is based on a thesis submitted to the graduate faculty, Federal University of Rio de Janeiro, in partial fulfillment of the requirements for a Master’s degree (R. Meira).
A supplemental appendix to this article is published electronically only at http://www.dentalresearch.org. Received for publication September 18, 2003. Revision received June 23, 2004. Accepted for publication June 28, 2004.
Journal of Dental Research, Vol. 83, No. 9,
723-727 (2004) This article has been cited by other articles:
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
ABSTRACT
INTRODUCTION
) during murine palatogenesis in vivo and in vitro. Anat Embryol 184:83–91.
