Review
A nomenclature for X-linked amelogenesis imperfecta

https://doi.org/10.1016/S0003-9969(02)00005-5Get rights and content

Abstract

Mutations of the X-chromosome amelogenin gene (AMELX) are associated with amelogenesis imperfecta (AI) phenotypes (OMIM no. 301200). Currently, 12 different AMELX mutations have been identified in individuals with abnormal enamel characteristic of AI. A notable feature of AI is the variable clinical phenotype, spurring interest in genotype–phenotype correlations. It is important that researchers and clinicians have an informative and reliable means of reporting and communicating these molecular defects. Therefore, the purpose here was to present a systematic nosology for reporting the genomic, cDNA and protein consequences of AMELX mutations associated with AI. The proposed nomenclature adheres to conventions proposed for other conditions and can be adopted for the autosomal forms of AI as the molecular basis of these conditions becomes known.

Introduction

The amelogenesis imperfectas (AIs) are a group of inherited conditions that adversely affect the development of dental enamel, causing anomalies in its amount, structure and composition (Witkop, 1989). The genetic basis of the X-linked form and at least one of the autosomal-dominant forms has been identified. The X-linked form, AIH1, results from mutations in the X-chromosome amelogenin gene (AMELX). To date, 12 allelic AMELX mutations have been reported (Lagerstrom et al., 1991, Aldred et al., 1992, Lagerstrom-Fermer et al., 1995, Lench and Winter, 1995, Collier et al., 1997, Kindelan et al., 2000; Sekiguchi et al., 2001a, Sekiguchi et al., 2001b; Greene et al., 2002, Hart et al., 2002). Mutational analyses and careful evaluations of the phenotype of affected individuals with the X-linked type have begun to reveal genotype–phenotype correlations, emphasizing the value of characterizing specific forms of AI at the DNA and protein levels (Hart et al., 2000, Ravassipour et al., 2000). Although the utility of molecular-based nomenclatures has been recognized, and their adoption suggested, a comprehensive nosology for the AIs does not exist (Aldred and Crawford, 1995). Indeed most of the molecular defects associated with AI have been discovered only within the past 4 years. With the identification of multiple AMELX mutations and the recent finding of a mutation in the enamelin gene associated with AI (Rajpar et al., 2001), the need for a comprehensive nosology to characterize the genetic and protein defects in AI has become apparent.

Previous nomenclatures have been inconsistent and differed depending upon whether numbering began with the mature or the signal peptide. Most AMELX mutations have been numbered from the mature peptide, but this does not describe comprehensively the diverse mutations occurring in the amelogenin gene. AMELX mutations involving the signal peptide, and alternative splicing, do not fit into this system (Lagerstrom-Fermer et al., 1995; Sekiguchi et al., 2001a, Sekiguchi et al., 2001b). To be of broad utility, a comprehensive nosology of AI should permit classification of all known forms, and be able logically and understandably to integrate novel forms as they are identified. We therefore now propose a standard nomenclature based upon internationally agreed upon guidelines (Antonarakis and Group, 1998, Den Dunnen and Antonarakis, 2000) using designated reference sequences. All known AMELX mutations are described using this new nosology. In the future, newly described mutations should be reported in all three formats as described below.

For genomic DNA, the reference sequence is GenBank (http://www.ncbi.nlm.nih.gov/Genbank/) accession no. AY040206 (Fig. 1a). The A of the initiator ATG is taken as +1. The nucleotide 5′ of the translation initiation codon is −1. The nucleotide 3′ of the translation termination codon is designated as 1. The genomic sequence is preceded by a lower case g, while cDNA is preceded by c and the protein designation by p.

For cDNA numbering, the reference sequence is GenBank accession no. AF436849 (Fig. 1b). The A of the initiator ATG is taken as +1; this cDNA contains exon 4. Exon 4 is alternatively spliced and is found in less than 1% of all human AMELX transcripts (Salido et al., 1992). Despite this low abundance of alternatively spliced exon 4 containing amelogenin, recent studies suggest that this product could be functionally significant (Veis et al., 2000). Because it is possible that AMELX mutations in exon 4 spliced products might be identified, and for completeness, the full-length amelogenin cDNA and protein that include exon 4 are used. The nucleotide 5′ of the translation initiation codon is −1. The nucleotide 3′ of the translation termination codon is designated 1. GenBank accession no. XM_010354 should not be used, as it contains an extra G in exon 5.

The human amelogenin protein has a 16-amino acid signal peptide and generally a 175-amino acid secreted mature protein. For splice variants containing exon 4, which represent less than 1% of all human transcripts (Salido et al., 1992), the mature protein is of 189-amino acids. The proposed nomenclature for mutant amelogenin proteins begins numbering from the initiator methionine in the signal peptide as the +1 position. This same nosology may also be used to report deduced amelogenin proteins. The methionine of the secreted amelogenin becomes no. 17 (Fig. 1b) and the C-terminal amino acid is no. 205. Frameshift mutations should indicate the first affected amino acid and include the length of the new reading frame. For example, the deletion noted by Lench and Winter (1995) would be designated as P158fsX187, indicating that this mutation produces a frameshift with Pro58 as the first amino acid affected and that the new reading frame is open for 30-amino acids. This frameshift introduces a premature stop codon that terminates the protein 16-amino acids shorter than the wild type.

All nucleotide and amino acid changes should conform to the internationally agreed guidelines detailed in Antonarakis and Group (1998) and Den Dunnen and Antonarakis (2000). Single-letter designations for amino acids are preferred. For example, the mutation that causes a conserved histidine at position 77 to become a leucine (Hart et al., 2002) would be designated H77L.

Twelve mutations have so far been described in the AMELX amelogenin gene. Of these mutations, five are 1 bp deletions that produce frameshifts, one is a 9 bp deletion that results in the deletion of three amino acids and the insertion of a new amino acid at the deletion junction, three are mis-sense substitutions, two are non-sense point mutations and one is a large deletion. Table 1 lists each mutation according to the new nomenclature. It is important to point out that these mutations have all been identified at the level of genomic DNA; protein and cDNA descriptions are hypothetical, based upon the genomic findings. Given the tissue-specific nature of amelogenin, it is normally not possible to analyse amelogenin protein and RNA in affected individuals.

Section snippets

Discussion

This proposed nosology permits a logical and understandable description of AI causing alterations at the genomic, cDNA and protein levels. The adoption of consistent genomic and cDNA references and an accepted convention for numbering peptides will greatly reduce the current confusion and enhance communication between investigators and clinicians. It is important to report newly described mutations in all three formats, as description at one level does not automatically identify the effect at

References (22)

  • Greene, S., Yuan, Z., Wright, J., Amjad, A., Abrams, W., Buchanan, J., Gibson, G., 2002. A gene deletion resulting in...
  • Cited by (70)

    • Deletion of epithelial cell-specific p130Cas impairs the maturation stage of amelogenesis

      2022, Bone
      Citation Excerpt :

      Enamel matrix proteins are composed of amelogenin, which accounts for up to 90%, enamelin, ameloblastin, and tuftelin [51]. Loss-of-function mutations in these proteins in humans and the deletion of these proteins in mice result in enamel hypoplasia [4,52,53]. Amelogenin-deficient mice exhibited a similar phenotype to human X-linked amelogenesis imperfecta, in which ameloblast differentiation was normal but irregular enamel crystal and thinner enamel were observed [52].

    • Identification and characterization of the cDNA sequence encoding amelogenin in rabbit (Oryctolagus cuniculus)

      2016, Gene
      Citation Excerpt :

      Amelogenins are one major type of proteins in enamel extracellular matrix (ECM) during tooth amelogenesis. They are essential for proper enamel formation as demonstrated by human genetic analyses (Hart et al., 2002; Wright et al., 2003; Kim et al., 2004) as well as mouse knock-out/knock-in studies (Lyngstadaas et al., 1995; Gibson et al., 2001; Fukumoto et al., 2004; Hu et al., 2008; Wright et al., 2011). However, their clear functions are still far to be understood, especially about how the proteins interact with other proteins in ECM, with the growing crystals, and with cells to form teeth structure.

    • Acquired and Developmental Disturbances of the Teeth and Associated Oral Structures

      2016, McDonald and Avery's Dentistry for the Child and Adolescent: Tenth Edition
    • Acquired and Developmental Disturbances of the Teeth and Associated Oral Structures

      2011, McDonald and Avery Dentistry for the Child and Adolescent
    • A study of polymorphism in human AMELX

      2007, Archives of Oral Biology
    View all citing articles on Scopus

    Adoption of a standardized nomenclature for AI mutations was completed after acceptance of this manuscript (see Hart et al., 2002, pp. 261–265). The new mutation designation is as follows Pro40Thr=P70T, His47Leu=H77L.

    View full text