| Sign In to gain access to subscriptions and/or personal tools. |
Amelogenesis Imperfecta in a New Animal Model—a Mutation in Chromosome 5 (human 4q21)
1 Department of Prosthetic Dentistry, University Hospital Hamburg-Eppendorf, Martinistr. 52, D-20246 Hamburg, Germany; Correspondence: * corresponding author, springer{at}mkg.uni-kiel.de
Candidate genes for amelogenesis imperfecta (AI) and dentinogenesis imperfecta (DI) are located on 4q21 in humans. We tested our hypothesis that mutations in the portion of mouse chromosome 5 corresponding to human chromosome 4q21 would cause enamel and dentin abnormalities. Male C3H mice were injected with ethylnitrosourea (ENU). Within a dominant ENU mutagenesis screen, a mouse mutant was isolated with an abnormal tooth enamel (ATE) phenotype. The structure and ultrastructure of teeth were studied. The mutation was located on mouse chromosome 5 in an interval of 9 cM between markers D5Mit18 and D5Mit10. Homozygotic mutants showed total enamel aplasia with exposed dentinal tubules, while heterozygotic mutants showed a significant reduction in enamel width. Dentin of mutant mice showed a reduced content of mature collagen cross-links. We were able to demonstrate that a mutation on chromosome 5 corresponding to human chromosome 4q21 can cause amelogenesis imperfecta and changes in dentin composition.
Key Words: amelogenesis imperfecta (AI) structure collagen cross-links pyridinoline hydroxyproline
AI is a genetic disorder with autosomal-dominant, autosomal-recessive, or X-linked variants (Aldred et al., 2003). The classification of different types of amelogenesis imperfecta (AI) used to be based on the phenotype as a primary discriminator with hypoplastic and hypomineralized subtypes (Winter and Brook, 1975; Sundell and Valentin, 1986; Gibson et al., 1991; Shore et al., 2002). Recent suggestions have centered around the classification of AI on the basis of the mode of inheritance until the molecular genetic basis was clarified (Aldred and Crawford, 1995; Aldred et al., 2003). Clinical and radiographic appearances serve as secondary discriminators (Aldred et al., 2003). Candidate genes for autosomal-dominant AI include tuftelin (1q), albumin (4q), ameloblastin (4q), and enamelin (4q) (Aldred and Crawford, 1997; Hu et al., 2001; Kida et al., 2002; Mardh et al., 2002). Candidate genes for dentinogenesis imperfecta, like dentin sialophosphoprotein (DSPP), map to the same region, 4q21 (MacDougall et al., 1999; Xiao et al., 2001; Malmgren et al., 2004). An animal model has been reported regarding X-linked AI in mice, and it has been suggested that a study of these mice could aid our understanding of the function of the protein amelogenin during enamel formation and for developing therapeutic approaches for treating this developmental defect (Gibson et al., 2001). We hypothesized that a mutation in chromosome 5 of mice could lead to enamel and/or dentin abnormalities. Within the dominant ENU mutagenesis screen, a mouse mutant was isolated with an abnormal tooth enamel (ATE) phenotype. In the course of the present study, we attempted to localize the mutation and to perform an analysis of the microscopic and ultramicroscopic structure of the teeth of mutant mice and to compare them with the teeth of wild-type mice.
Animals and DNA Preparation Male C3HeBFeJ mice were injected intraperitoneally with ethylnitrosourea (ENU, 3 x 90 mg/kg body weight; GSF National Research Center for Environment and Health GmbH, Helmholtz Association) at the age of 10 wks. Seven wks later, treated mice were mated with untreated female C3H mice at the GSF Research Center according to the German law governing the protection of animals. From each treated male, 100 F1 (first filial generation) offspring were produced and examined within the ENU mouse mutagenesis screen consortium (Hrabe de Angelis and Balling, 1998; Hrabe de Angelis et al., 2000). Animals found to have an altered phenotype were genetically confirmed by being back-crossed to wild-type C3HeBFeJ animals. The mutation was mapped according to an out-cross/back-cross breeding scheme on a C57BL/6J background. DNA was prepared from tail tips of offspring showing an abnormal tooth phenotype according to standard procedures. For the genome-wide linkage analysis, several markers were used for each chromosome indicating linkage to chromosome 5 between the markers D5Mit18 and D5Mit10. In the present experiment, we examined 2 wild-type mice, 2 heterozygous ATE mutants, and 2 homozygous ATE mutants. All mice were male and 22 wks of age. Mice were formalin-fixed under general anesthesia by vascular perfusion with 4% phosphate-buffered formalin. The jaws were post-fixed at room temperature in the same fixative for an additional 4 days. ATE mutants are reported here for the first time. This animal experiment was approved by German government authorities and was carried out according to the German law for the protection of animals, a protocol that ensures the best humane practices.
Histology
Preparation and Hydrolysis of Samples
Analysis of Hydroxyproline
Statistics
Scanning Electron Microscopy (SEM)
Within a dominant ENU mutagenesis screen, a mouse mutant was isolated with an abnormal tooth enamel phenotype. In a genome-wide linkage analysis, the mutation was mapped to mouse chromosome 5 in an interval of 9 cM between the markers D5Mit18 and D5Mit10. A list of genes located between these markers is provided in the DISCUSSION section of this paper. Heterozygous mutants showed easily identifiable lighter-colored upper incisors in contrast to the wild-type controls. In homozygous animals, the phenotype was obvious.
Microradiographs and toluidine-blue staining of non-decalcified sections showed complete loss of enamel in the front and molar teeth of homozygotic ATE mutants, cracked enamel of reduced width (approx. 50%) in incisor and molar teeth of heterozygotic ATE mutants, and regular structure of incisor and molar teeth of wild-type mice (Fig. 1
Scanning electron micrographs (SEMs) of teeth of homozygotic ATE mutants showed loss of enamel with exposed dentinal tubules (Fig. 2
Only slight differences were observed between heterozygous and homozygous ATE mutants with regard to dental concentrations of Hyp, HP, and LP. There was a 1.08- to 2.59-fold increase of concentrations of Hyp, HP, and LP in incisors of heterozygous and homozygous ATE mutants as compared with incisors of wild-type mice. There was a 1.37- to 5.63-fold increase of dental concentrations of Hyp, HP, and LP in molar teeth of heterozygous or homozygous ATE mutants as compared with molar teeth of wild-type mice. Dental concentrations of HP and LP showed a significant correlation (correlation coefficient, 0.776, P = 0.003). Also, dental concentrations of HP and Hyp (correlation coefficient, 0.930, P < 0.001) and concentrations of LP and Hyp (correlation coefficient, 0.755, P = 0.005) showed a significant correlation (Fig. 3
We hypothesized that mutations in chromosome 5 of mice can lead to significant alterations of the structure of enamel and dentin. Within a dominant ENU mutagenesis screen, a mouse mutant was isolated with an abnormal tooth enamel phenotype. In the present study, microradiography and toluidine-blue staining of non-decalcified sections as well as scanning electron microscopy were chosen to ensure a precise judgment of the structure (Donath and Breuner, 1982; Gibson et al., 2001; Springer et al., 2002). These techniques have been used previously by the authors who first introduced a mouse model for X-linked amelogenesis imperfecta (Gibson et al., 2001). In our study, microradiography and toluidine-blue-stained sections showed that the thickness of enamel was reduced by approximately 50% in heterozygous ATE mutants, which compares with human hypoplastic AI (Wright et al., 1991). Mapping the mutant line revealed an interval of 9 cM on mouse chromosome 5 between the markers D5Mit18 and D5Mit10. According to the Mouse Genome Informatics (MGI)© database of The Jackson Laboratory© (09.02.2004), the following genes are located on mouse chromosome 5 between markers D5Mit18 and D5Mit10 (Position 45–54 in cM): casein gene family (Csn, position 45.0 cM), modifier of curly bare (Mcub, position 45.0 cM), diabesity 2 (Dbsty2, position 45.0 cM), MMMTV LTR integration site 5 (Pad 5, position 45.0 cM), seizure susceptibility 6 (Szs 6, position QTL), zinc finger protein 469 (Zfp469, position 45.1 cM), casein kappa (Csnk, position 45.2 cM), ameloblastin (Ambn, position 46.0 cM), marcel (Mc, position 48.0 cM), mammary tumor virus locus 32 (Mtv32, position 48.0 cM), ornithine decarboxylase related sequence 21 (Odc-rs21, position 48.0 cM), alpha fetoprotein (Afp, position 50.0 cM), albumin 1 (Alb 1, position 50.0 cM), betacellulin (Btc, position 51.0 cM), amphiregulin (Areg, position 51.0 cM), ADP-ribosyltransferase 3 (Art3, position 52.0 cM), shroom (Shrm, position 52.0 cM), annexin A3 (Anxa 3, position 54.0 cM), and WD repeat and FYVE (Wdfy3, position 54.0 cM). Of these genes, Mcub, Dbsty2, and Szs6 can be excluded as candidate genes causing the phenotype observed in the present study, since their expression is known and differs from that of the phenotype noted. Ameloblastin, located on position 46 cM, appears to be the strongest candidate (MacDougall et al., 1997; Lee et al., 2003; Nagano et al., 2003). The potential of the rest of the genes listed above of inducing the phenotype observed in this study is considered low, although the impact on the phenotype has not yet been described. Our findings are supported by human studies showing that a nonsense mutation within the enamelin gene and the ameloblastin gene, both mapping to the corresponding region 4q21 in humans, causes hypoplastic autosomal-dominant AI (Rajpar et al., 2001; Kida et al., 2002; Mardh et al., 2002; Hart et al., 2003). The mice were exposed to ENU, which causes point mutations. The authors suggest that a deletion may be unlikely. A point mutation may lead to changes of the tertiary structure of the protein. Also, it might be possible that exchange of one single base causes alternate splicing, premature chain abortion, or changes of the promoter region. After Gibson and co-workers (2001) introduced a new mouse model for X-linked AI, this study introduces a new mutant for autosomal-dominant hypoplastic AI. We were able to show that the concentration of collagen cross-links hydroxylysylpyridinoline (HP) and lysylpyridinoline (LP) is markedly increased in teeth of heterozygotic and homozygotic ATE mutants compared with concentrations in wild-type mice. HP and LP are two non-reducible cross-links of mature collagen, which are formed by a sequence of post-translational modifications. HP is a derivative of 3 residues of hydroxylysine and is present in virtually all mature tissues (tendon, vessel wall, cartilage, dentin, and bone). LP is a derivative of 2 residues of hydroxylysine and 1 residue of lysine and is found principally in dentin and bone (Eyre, 1992; Açil and Müller, 1994; Açil et al., 2002; Springer et al., 2003). The measurement of the dentinal concentration of mature collagen cross-links HP and LP is a reliable method showing reproducible concentrations independent of the kind of tooth (incisor/molar, etc.) and age (Açil et al., 2002). Collagen is considered to be the predominant protein within the organic content of this mineralized tissue (Dai et al., 1991). Mouse teeth are extremely small (length approx. 2 mm). Based on pilot experiments, we decided not to separate enamel from dentin, since we believed that the small proportion of enamel would barely influence the concentrations of HP and LP measured. We suggest that alterations of the concentration of HP and LP may be due to alterations in the content of collagen in the dentin of mice after mutation in chromosome 5 between markers D5Mit18 and D5Mit10. Candidate genes for dentinogenesis imperfecta, like the dentin sialophosphoprotein (DSPP) gene, map to 4q21 in humans (MacDougall et al., 1999; Malmgren et al., 2004). To date, to the best of our knowledge, none of the genes located between D5Mit18 and D5Mit10 has been associated with dentinogenesis imperfecta. We cannot explain the differences in the concentrations of collagen cross-links between affected and non-affected mice. We suggest that one of the genes mentioned participates in dentinogenesis. On this basis, it must be assumed that, in the model presented, the presumed ameloblastin mutation must have effects on enamel and dentin. Alterations in the collagenous structure of teeth of patients suffering from dentinogenesis imperfecta might influence restorative techniques (Perdigão et al., 2000). We were able to verify our hypothesis. A mutation in chromosome 5 of mice between markers D5Mit18 and D5Mit10 results in certain features of AI. The identification of the mutated gene between these markers will be the aim of a further study. An animal model for AI with known genotype facilitates the development of standardized conditions for the study of the possible influence of nutritional habits on the phenotype. Knowledge concerning the clinical features and dental complications of each variant of AI assists in the diagnosis of the condition and possibly results in the implementation of early preventive measures (Perdigão et al., 2000). The authors suggest that the ATE1 mutant with a mutation in 5E provides an animal model for the study of the phenotype of autosomal-dominant hypoplastic AI and the investigation of possible preventive measures to influence the outcome of this genetic condition.
We express our gratitude to our laboratory technicians, under the leadership of Mrs. Gisela Otto, for their assistance with the performance of the analytical procedures. Financial support was provided by each of the institutions involved in the present publication. We are grateful for the assistance of M. Suhr, FRCS, FDSRCS in the revision of this manuscript. Received for publication August 28, 2003. Revision received May 28, 2004. Accepted for publication June 2, 2004.
Journal of Dental Research, Vol. 83, No. 8,
608-612 (2004)
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||


