Deletion of ameloblastin exon 6 is associated with amelogenesis imperfecta

Abstract

Amelogenesis imperfecta (AI) describes a heterogeneous group of inherited dental enamel defects reflecting failure of normal amelogenesis. Ameloblastin (AMBN) is the second most abundant enamel matrix protein expressed during amelogenesis. The pivotal role of AMBN in amelogenesis has been confirmed experimentally using mouse models. However, no AMBN mutations have been associated with human AI. Using autozygosity mapping and exome sequencing, we identified genomic deletion of AMBN exon 6 in a second cousin consanguineous family with three of the six children having hypoplastic AI. The genomic deletion corresponds to an in-frame deletion of 79 amino acids, shortening the protein from 447 to 368 residues. Exfoliated primary teeth (unmatched to genotype) were available from family members. The most severely affected had thin, aprismatic enamel (similar to that reported in mice homozygous for Ambn lacking exons 5 and 6). Other teeth exhibited thicker but largely aprismatic enamel. One tooth had apparently normal enamel. It has been suggested that AMBN may function in bone development. No clinically obvious bone or other co-segregating health problems were identified in the family investigated. This study confirms for the first time that AMBN mutations cause non-syndromic human AI and that mouse models with disrupted Ambn function are valid.

Figure 1.

Family pedigree and dental phenotype. (A) Pedigree of the Costa Rican family investigated. Affected family members are shaded. (B) The permanent teeth of affected individuals (VI:3 presented) were characterized by generalized hypoplastic AI involving all teeth. (C) The enamel surface was rough, but hard (VI:3 and VI:4 presented in Ci and Cii, respectively). Note: some teeth have been restored with metal or tooth-coloured restorations (examples marked*).

Figure 2.

Genotyping of the mutation and identification of the deletion breakpoints. (A) PCR analysis of AMBN exon 6 in members of the family investigated. Amplification of exon 6 alone was only observed in individuals with clinically normal teeth and a normal AMBN genotype, whereas PCR with primers spanning from exons 5 to 7 amplified a product of 794 bp in affected and carrier individuals only, reflecting loss of exon 6. Bands obtained with primers amplifying P53 are shown as a positive control for each sample. (B) Sequence electropherograms of mutant and wild-type alleles to identify the deletion breakpoints. Comparison of the mutant sequence with a control sequence revealed a deletion of 2347 bp spanning exon 6 and the majority of the surrounding introns either side. The deletion begins 114 bp from the end of exon 5. (C) Wild-type amino acid sequence of AMBN exon 6. Important domains, which are likely to play a key role in the function of the mature protein, are labelled. Schematics of the exon-intron layout of wild-type and the mutant genomic DNA sequences are also shown, showing the lack of exon 6 and the approximate break-points with respect to the local exons.

Figure 3.

High resolution X-ray CT analysis of exfoliated teeth 1–5. Typical CT sections through the teeth are presented using false colour calibrated with respect to mineral density to generate mineral density maps. Mean enamel mineral density for each tooth is also shown graphically. Tooth 1 exhibits an enamel layer apparently normal in structure and density. Teeth 2 and 3 exhibit an obvious enamel covering though it is thinner in many areas, chipped, absent in places and is significantly reduced in mineral density compared with tooth 1 (P < 0.0001). Teeth 4 and 5 exhibit an even thinner covering of enamel with enamel absent in some areas. Again enamel mineral density is significantly lower than in tooth 1 (P < 0.0001). The only non-significant differences in enamel mineral density occur between teeth 4 and 5. Videos of 3D rendered CT data showing surface detail and internal structure of all teeth are available as Supplementary Material.

Figure 4.

SEM of representative exfoliated teeth. Tooth 1 exhibits normal enamel architecture comprising prisms (rods) of individual enamel crystallites classically described in numerous research publications and text books. The SEM representative of the thicker enamel found covering tooth 2 is lacking the obvious long range order and structure that characterizes normal enamel. However, very occasional areas exhibiting prism-like structures are present (indicated by arrows on inset micrograph). Where present, the enamel covering teeth 4 and 5 is thin and exhibits no structural similarities to normal enamel. The white horizontal lines show where separate micrographs have been collaged to create a wider field of view.