Phenotypic diversity and revision of the nomenclature for autosomal recessive amelogenesis imperfecta

Phenotypic diversity and revision of the nomenclature for autosomal recessive amelogenesis imperfecta Mohamad Nusier, MD, PhD,a Othman Yassin, DDS, MsC,b Thomas C. Hart, DDS, PhD,c Afsaneh Samimi,d and J. Timothy Wright, DDS, MS,e Irbid, Jordan, Amman, Jordan, Pittsburgh, Pa, and Chapel Hill, NC JORDAN UNIVERSITY OF SCIENCE & TECHNOLOGY, KING HUSSEIN MEDICAL CENTER, UNIVERSITY OF PITTSBURGH, AND UNIVERSITY OF NORTH CAROLINA

Purpose. The purpose of this study was to characterize the phenotype in 9 families with autosomal recessive amelogenesis imperfecta (ARAI), and to propose a classification system allowing inclusion and delineation of diverse ARAI phenotypes. Study design. Nine families with ARAI were evaluated clinically and radiographically. Exfoliated and extracted teeth were examined via light and scanning electron microscopy, with the enamel in one case evaluated by amino acid analysis. Results. The 9 families demonstrated diverse ARAI phenotypes including localized hypoplastic, generalized thin hypoplastic, hypocalcified and hypomaturation AI types. Conclusions. Some ARAI phenotypes observed in this study and reported in the literature cannot be classified using currently accepted ARAI nomenclature. Therefore, we propose a revised nomenclature permitting both classification of all ARAI clinical forms and inclusion of anticipated molecular-based nomenclature, such as now exists for some X-linked and autosomal dominant AI subtypes. (Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2004;97:220-30)

Amelogenesis imperfecta (AI) is a genetically and clinically heterogeneous group of heritable disorders primarily affecting tooth enamel formation. The prevalence of AI varies in different populations, ranging from 1 in 700 in Sweden to 1 in 14,000 in the USA.1 The AI trait can be transmitted by either autosomal dominant, autosomal recessive, or X-linked modes of inheritance. One large epidemiological study of 51 Swedish families segregating for AI showed approximately 6% of cases to be X-linked, 63% autosomal dominant, and 12% autosomal recessive forms of AI.2 The remaining 19% of AI cases were sporadic with neither family history nor discernable mode of transmission. The distribution of AI types also is known to vary in different populations. For example, autosomal recessive AI was the most prevalent AI type in a study a

Associate Professor, Department of Biochemistry & Molecular Biology, Jordan University of Science & Technology, School of Medicine, Irbid, Jordan. b Pediatric Dentistry Consultant, Department of Pediatric Dentistry, King Hussein Medical Center, Amman, Jordan. c Associate Professor, Department of Human Genetics, Department of Oral Pathology and Medicine, School of Dental Medicine, University of Pittsburgh. d Student, Department of Pediatric Dentistry, School of Dentistry, University of North Carolina. e Professor, Department of Pediatric Dentistry, School of Dentistry, University of North Carolina. 1079-2104/$ - see front matter © 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.tripleo.2003.008.007

220

evaluating 70,000 Israeli school children.3 Enamel findings in AI are highly variable, ranging from deficient enamel formation (hypoplastic AI) to defects in the mineral and protein content (hypocalcified and hypomaturation AI). Over the past decade, molecular defects causing at least some X-linked and autosomal dominant forms of AI have been identified. Twelve distinct mutations in the amelogenin gene (AMELX) have been associated with different X-linked AI phenotypes (OMIM 301200).4,5 Recently, enamelin gene (ENAM) mutations have been identified in autosomal dominant forms of hypoplastic AI (OMIM 104500).6-8 The genes responsible for the autosomal recessive AI (ARAI) types have not been identified, and it is not known whether their diverse phenotypes are the result of either allelic gene mutations or mutations in multiple genes. A family segregating for a syndromic form of ARAI with cone-rod dystrophy was evaluated and reported to be linked to a 5-MB interval on chromosome 2q11 (Lod score 7.03).9 However, linkage studies of 4 other families segregating for nonsyndromic ARAI showed a lack of linkage to this genetic interval, suggesting that this is not a common locus associated with ARAI. There is tremendous phenotypic diversity in the enamel findings of individuals with ARAI. Although several nomenclatures have been used to classify the different types of ARAI, none has proven sufficiently comprehensive to include all clinical phenotypes described in the literature, and none is based primarily on

ORAL SURGERY ORAL MEDICINE ORAL PATHOLOGY Volume 97, Number 2

mode of inheritance. Classification systems have been based on enamel phenotype, mode of inheritance, fundamental biological mechanism thought to produce the defect, and general phenotype. For example, Sundell proposed a classification that recognized 12 AI subtypes based on differences in the type and distribution of enamel defects.10 This system was later revised to include the mode of inheritance.11 The most commonly used AI nomenclature was developed and later revised by Witkop, using a system categorizing AI types first by major phenotype (i.e. hypoplastic, hypocalcified, hypomaturation), and then subdividing each AI type based on additional phenotypic features and mode of inheritance.1,12 Witkop recognized 4 distinct ARAI types.1 The Witkop classification of AI delineates 2 autosomal recessive hypoplastic forms including a local hypoplastic and an enamel agenesis type.1 This classification system recognizes 2 additional autosomal recessive forms of AI, including the pigmented hypomaturation type and AR hypocalcified AI. While many ARAI cases can be classified using the Witkop system, not all the hypoplastic AI phenotypes fit into the local hypoplastic or enamel agenesis subtypes. Others have reported ARAI as being rough hypoplastic, rough hypoplastic with follicular hyperplasia, and dysplastic dentin with follicular imperfecta.13-15 Still others recognize that current classification systems are not adequately robust to accommodate all phenotypically diverse ARAI cases, and have elected not to provide a traditional classification.16 There have been and continue to be differences of opinion as to the best approach for classifying the AI subtypes (whether to lump or split certain phenotypes and inheritance patterns), and what conventions to use to delineate the subtypes (e.g., enamel features, inheritance, presumed biological dysfunction). The use of different nomenclatures often renders valid comparison of distinct ARAI types difficult at best. Development of a comprehensive classification system is important, as this conceptual framework would likely help advance genetic studies to determine the molecular etiology of ARAI, enhance communication between clinicians and researchers, and facilitate patient counseling. It has been suggested that the nomenclature for AI be revised to include mode of inheritance, molecular basis, biochemical outcome, and phenotype of the condition.17 This type of nomenclature now exists for X-linked AMELX gene– and autosomal dominant ENAM gene–associated AI, using standard conventions proposed for reporting mutations in the medical literature.8,18 Unfortunately, the molecular etiology of all ARAI types remains unknown, precluding a molecular-based classification at this time.

Nusier et al 221

However, in the near future it is highly likely that this information will be available. Thus any ARAI classification system developed should be designed to facilitate incorporation of emerging information pertaining to specific gene mutations. There have been few studies attempting to characterize the different autosomal recessive forms of AI and their phenotypes, with much of our knowledge derived from individual case reports. Therefore the purposes of this study were to evaluate multiple families segregating for ARAI, to characterize the associated phenotypes, and to propose a robust classification system allowing inclusion of all known phenotypes and ultimately genetic information as it becomes known. METHODS This study was approved by the Institutional Review Board, with all study participants providing informed consent prior to entry in the study. Nine families were identified as having children with clinical manifestations consistent with a diagnosis of AI. The parents in all families showed neither clinical nor radiographic evidence of AI. None of the affected children showed clinical abnormalities involving the skin, hair, fingernails, or other systems that would suggest a syndromeassociated dental malformation. Family histories were evaluated and pedigrees constructed for each kindred. The families were evaluated clinically and when possible radiographically. Enamel structure of exfoliated and therapeutically extracted teeth was analyzed using light and scanning electron microscopy. Teeth were sectioned with a low-speed saw and diamond blade producing mineralized sections of approximately 125 ␮m in thickness. Teeth were fractured and enamel samples mounted on aluminum stubs and coated with Au-Pd and examined in a JOEL Scanning Electron Microscope.19 The enamel of 1 tooth was subjected to amino acid analysis to determine the quantity and quality of enamel proteins. The tooth was sectioned as described above, the enamel was dissected from the thin section, and care was taken to prevent dentin contamination. The enamel fragments were hydrolyzed in 6 N HCl at 104°C for 12 hours. The hydrolyzed amino acids were evaluated via high performance liquid chromatography (HPLC), with findings expressed as residues per thousand.19 Normal enamel was used as a control for microscopic and biochemical analyses. RESULTS Nine Jordanian families were ascertained through affected probands with AI. Detailed dental evaluation of the probands, their parents and siblings revealed multiple affected children in 7 of the families. None of

222 Nusier et al

ORAL SURGERY ORAL MEDICINE ORAL PATHOLOGY February 2004

Fig 1. Family pedigrees of the nine ARAI families, with probands identified by arrows.

the parents showed evidence of AI (Fig 1). There were 20 affected and 41 unaffected offspring in the 9 families. After correction for proband ascertainment bias, the ratio of affected to unaffected siblings was found to be 1:4, consistent with autosomal recessive inheritance. Parental consanguinity was identified in 3 of the 9 families. The ratio of affected to unaffected siblings, the presence of multiple affected children, the finding of unaffected parents, and the presence of consanguinity in some families were all observations consistent with inheritance of an autosomal recessive trait. A summary of enamel phenotypes and dental open bite findings in affected individuals is offered in Table I. All three major AI types including hypoplastic, hypomaturation, and hypocalcified AI occurred in this

population. Anterior open bite was observed in 6 of the 9 families, affecting individuals with hypoplastic, hypomaturation, and hypocalcified AI types. While the most common AI phenotype in these families was hypoplastic AI (5 families), there was extensive diversity in the character of enamel hypoplasia between the families. Autosomal recessive hypoplastic AI Four of the families with hypoplastic AI had generalized thinning of the enamel (AIJ3, AIJ5, AIJ7, and AIJ9). The affected children in family AIJ3 showed teeth that lacked contacts and were generally yellow in color (Fig 2). The teeth were decorated with small islands of enamel which, when compared with the

Nusier et al 223

ORAL SURGERY ORAL MEDICINE ORAL PATHOLOGY Volume 97, Number 2

Table I. Autosomal recessive AI phenotypes in 9 families Family #

AI type

Enamel

Open Bite

AIJ1 AIJ3 AIJ4 AIJ5 AIJ6 AIJ7 AIJ8 AIJ9 AIJ10

Hypocalcified Generalized thin hypoplastic Local hypoplastic Generalized thin hypoplastic Hypomaturation Local hypoplastic Hypocalcified Local hypoplastic Hypomaturation

Chalky white, extensive fracturing, normal thickness Normal to yellow color, rough surface, thin Normal to stained color, extensive pitting Yellow color, smooth surface, thin Yellow to orange brown color, normal thickness Normal to stained color, large enamel voids in crown mid 1/3 Chalk-white color, extensive fracturing, normal thickness Normal to stained color, coronal 2/3 enamel extremely thin to missing Yellow to orange brown color, normal thickness

⫹ – ⫹– – ⫹ ⫹– ⫹ ⫹– –

⫹, All affected individuals displayed dental open bite; ⫹–, dental open bite not present in all affected individuals.

Fig 2. The teeth appear yellow and rough in this case of AR generalized thin hypoplastic AI.

yellow areas, demonstrated a more normal color and increased thickness. Affected members of family AIJ5 had marked generalized thinning of the enamel, with little to no enamel apparent radiographically (Fig 3). However, microscopic evaluation of teeth from 2 of the affected individuals in family AIJ5 showed areas of thin prismatic enamel with small fine surface pits or grooves (Fig 4). Affected individuals in family AIJ7 also demonstrated generalized thinning of the enamel. However, the teeth had markedly hypoplastic to absent enamel—a feature especially pronounced in the middle third of the tooth crown (Fig 5). Histopathologically, the teeth from two affected individuals in family AIJ7 showed the enamel surface to be decorated with variable-sized pits up to 100 ␮m in diameter (Fig 6, A). The enamel had a normal prismatic structure (Fig 6, B). Amino acid composition of enamel in family AIJ7 was found to be generally similar to normal permanent molar enamel.20 Compared with normal enamel, AI

enamel contained nearly half the levels of aspartic acid, leucine and phenylalanine, and nearly 3 times the amount of glycine. There was no increase in proline that would suggest retention of amelogenin, and the total protein content was only slightly increased compared with the normal enamel (Fig 7). While enamel of affected individuals in family AIJ9 appeared to show some generalized thinning, the most prominent feature was an almost complete lack of enamel in the coronal two-thirds of the crown (Fig 8). The cervical enamel displayed normal color and translucency. The affected child had a severe anterior dental open bite. Affected individuals in family AIJ4 had hypoplastic enamel that reached normal thickness, but showed severe pitting of the entire enamel surface (Fig 9). The pits were stained black and in some areas appeared arranged in horizontal rows. The enamel around the pits appeared generally of normal color and translucency.

224 Nusier et al

ORAL SURGERY ORAL MEDICINE ORAL PATHOLOGY February 2004

Fig 3. Radiographically there is little to no enamel visible on the erupted or unerupted teeth of this AR generalized thin hypoplastic AI case.

Fig 4. Areas of thin prismatic enamel show increased pigmentation in this AR generalized thin hypoplastic AI case (ground section, unstained, original magnification ⫻200).

One of the 2 affected children showed a severe skeletal open bite. Autosomal recessive hypocalcified AI Two families (AIJ1, AIJ8) segregated for autosomal recessive hypocalcified AI. The phenotype was similar in both families, with the teeth of affected individuals having enamel of normal thickness on unerupted teeth. Radiographically the affected enamel had only a slightly greater radiopacity when compared with dentin. The only enamel observed to remain on the erupted teeth was located in the cervical areas (Fig 10). Microscopic evaluation showed prismatic enamel lacking normal translucency, indicating the enamel was abnormally mineralized (Fig 11).

Autosomal recessive hypomaturation AI Two families (AIJ6, AIJ10) segregated for autosomal recessive hypomaturation AI. In both families the enamel appeared of normal thickness with a yellow to orange-brown color (Fig 12). Radiographic analysis showed the enamel had only a slightly increased radiodensity compared with the dentin. Extensive calculus formation was seen on the mandibular incisors of affected individuals of family AIJ6 but not AIJ10. Dental open bites were present in both affected children of family AIJ6 but were not encountered in AIJ10. Interestingly, 1 affected child in family AIJ10 had a hypomaturation phenotype, while the other affected child had a generalized thin hypoplastic phenotype. This observation may be inter-

ORAL SURGERY ORAL MEDICINE ORAL PATHOLOGY Volume 97, Number 2

Fig 5. Severe enamel hypoplasia is observed radiographically as being primarily localized to the middle third of the crowns (arrows) of these unerupted teeth in this AR local hypoplastic AI case.

preted as evidence either suggesting the presence of 2 different ARAI-associated genes in the parents, or indicating highly variable expression of the same mutation. DISCUSSION The detailed characterization of ARAI in 9 families illustrates the inadequacy of current classification systems predicated on delineating AI cases based first on phenotype and then by mode of inheritance. With the identification of genetic alterations responsible for dominant forms of AI, there is increased interest in identifying those genes responsible for recessive forms of the disease. The lack of a comprehensive classification system is problematic for studies of ARAI. Molecular-based studies of ARAI will be facilitated by a comprehensive and robust classification system. Given the overlapping and diverse phenotypes seen in ARAI families, the findings from this and other studies support the use

Nusier et al 225

of mode of inheritance as the primary delineator of AI subtypes. The AI case can then be further classified based on molecular and phenotypic information as is now the practice in classification of X-linked and autosomal dominant AI types.4,8,18 While the AR hypocalcified and hypomaturation AI families presented in this report had enamel phenotypes similar to those described in the literature, there was tremendous diversity in the AR hypoplastic AI phenotypes. The marked phenotypic diversity of ARAI cases with a hypoplastic phenotype did not permit straightforward classification according to a currently accepted nosology recognizing two autosomal recessive hypoplastic types.1,21,22 Local hypoplastic ARAI has been used to classify dental phenotypes characterized either by generalized pitting as seen in family AIJ4, or by hypoplastic areas often localized in the coronal middle third similar to family AIJ7.1,23 Chosack et al3 felt this latter clinical pattern represented a new AI phenotype; however, Wikop’s revised classification included the Chosack phenotype in AR local hypoplastic AI.1 While the phenotypes in families AIJ4 (multiple small pits), AIJ7 (severe hypoplasia in middle third and large hypoplastic pits), and AIJ9 (severe hypoplasia in coronal two-thirds) are distinctly different, the term “local” hypoplastic is appropriate to characterize the phenotype. Definitive molecular studies will determine if these diverse local hypoplastic phenotypes result from variable expression of the same mutation, allelic mutations, or mutations in different genes. The fine pitting seen in family AIJ4 is remarkably similar to that seen in autosomal dominant local hypoplastic AI that is highly prevalent in Sweden and is associated with a mutation in the ENAM gene.7 The pits were stained black and often appeared arranged in horizontal rows and columns similar to the distribution seen in autosomal dominant local hypoplastic AI.1,11 Biochemical analysis of local hypoplastic enamel from family AIJ7 showed a slight increase in total enamel protein content and some changes in the amino acid content. The local hypoplastic ARAI enamel protein and amino acid composition determined in this study was very similar to that described in generalized thin “smooth” hypoplastic AI.20 There was no increase in proline, which would be indicative of retention of amelogenin and enamel hypomaturation.24,25 Taken together, the microscopic and biochemical analyses of local hypoplastic ARAI indicate the enamel generally to be of normal structure and composition. Therefore, the molecular bases for these conditions likely only affects the secretory function of ameloblasts, and not those processes

226 Nusier et al

ORAL SURGERY ORAL MEDICINE ORAL PATHOLOGY February 2004

Fig 6. A, Large hypoplastic pits are seen decorating the cervical enamel in this AR local hypoplastic AI case (SEM, AuPd coated, original magnification ⫻ 150). B, The enamel between the pits and surrounding the large hypoplastic areas in the middle third of the crown have a normal prismatic structure (SEM, AuPd coated, original magnification ⫻4000).

Fig 7. The enamel amino acid composition are presented diagrammatically showing the increased glycine, and decreased aspartic acid, leucine, and phenylalanine levels in the AI enamel compared with the normal enamel.

involved in the mineralization and maturation of enamel. Generalized thinning of the enamel is the predominant feature of the hypoplastic ARAI type that Witkop originally classified as rough hypoplastic (enamel agenesis) and then later shortened to “enamel agenesis.”1,12 This AI type is characterized by teeth having a rough granular light yellow surface, and showing no evidence of enamel on radiographs.1 Two of the families in the

present investigation had a hypoplastic phenotype predominantly characterized by severe generalized thinning of the enamel (families AJJ3 and AIJ5). In contrast to previous reports of ARAI enamel agenesis, teeth from individuals in family AIJ5 had thin enamel that histopathologically showed a prismatic structure with no evidence of coronal cementum.1,13 Previous microscopic studies of AR rough hypoplastic or “enamel agenesis” AI have shown a variety of enamel

ORAL SURGERY ORAL MEDICINE ORAL PATHOLOGY Volume 97, Number 2

Nusier et al 227

Fig 8. The coronal two-thirds of enamel is severely hypoplastic, whereas the cervical enamel appears to have a relatively normal color and translucency. The rough hypoplastic areas are stained orange and black.

Fig 9. Severe generalized pitting with black staining of some teeth and a dental open bite are present in this child with AR local hypoplastic AI.

defects including hypomineralization, nonprismatic enamel, and irregular surface globules.16,26 The findings of the present study and in published investigations describing the presence of enamel in cases of ARAI with generalized thin enamel do not support classification as “enamel agenesis” as proposed by Witkop.1,12 Further delineation of rough and smooth phenotypes is highly subjective and likely problematic due to variable expression of these features. Therefore we propose classification of this AI type as autosomal recessive “generalized thin” hypoplastic AI (Table II).

This new nosology first identifies the mode of inheritance as autosomal recessive, making it consistent with the most commonly used system for cataloging human hereditary conditions (OMIM5), and then subdivides the ARAI hypoplastic types as either local or generalized thin. Interestingly, autosomal dominant local and generalized thin hypoplastic AI are caused by allelic mutations of the ENAM gene.6,7 Once the molecular basis of the ARAI local and generalized thin hypoplastic conditions are identified, this information will be added to the classification scheme. This new nomen-

228 Nusier et al

ORAL SURGERY ORAL MEDICINE ORAL PATHOLOGY February 2004

Fig 10. There is a dental open bite, the coronal enamel has fractured from the crowns, and the remaining cervical enamel has a chalky white color in this child with AR hypocalcified AI.

Fig 11. The hypocalcified enamel was of normal thickness, had a prismatic structure, and was markedly more opaque than normal enamel (ground section, unstained, original magnification ⫻100).

clature allows all ARAI phenotypes to be classified without expanding the already complex nomenclature. Previous studies describe families with an autosomal recessive mode of inheritance and generalized thin phenotype that in some instances show failure of tooth eruption.16,26 None of the families in the present study had abnormal tooth eruption. It is unclear whether ARAI sub-

types with generalized thin enamel distinguished by presence or absence of disturbances in tooth eruption are caused by different molecular defects or represent variable expression of the same genetic mutation. Indeed, a similar enigma exists when evaluating the presence of dental open bite and AI. It is known that AI is associated with a greatly increased prevalence of dental open bite; yet the

Nusier et al 229

ORAL SURGERY ORAL MEDICINE ORAL PATHOLOGY Volume 97, Number 2

Fig 12. The teeth of this female with AR hypomaturation AI display the characteristic yellow to orange-brown coloration characteristic of this AI type.

Table II. Nomenclature for autosomal recessive AI Witkop subtype* IC–local hypoplastic IG—enamel agenesis IIA—pigmented hypomaturation IIIB—hypocalcified

Revised Classification AR AR AR AR

local hypoplastic generalized thin hypoplastic pigmented hypomaturation hypocalcified

OMIM # 204650 Not listed 204700 Not listed

*Shows AI types as classified by Witkop1; they relate to proposed nomenclature and OMIM classification number.

expression of this trait is variable even within AI families. Some affected family members exhibit an open bite and others do not.27-29 Indeed, variable expression of the dental open bite trait was observed both within and between the families of the present study. CONCLUSION Detailed phenotypic characterization of multiple families segregating for ARAI show tremendous diversity in clinical presentation of this condition. This diversity extends beyond the clinical presentation to the microscopic and biochemical alterations of affected enamel. While the classification of ARAI recognizes 4 distinct subtypes based on phenotype, only 2 are currently recognized by OMIM.5 Refined phenotype characterization will enhance future molecular studies directed at identifying the genetic mutations associated with each of the different ARAI types, ultimately allowing classification that includes the molecular basis of these AI disorders. This research was supported by the National Institute for Dental and Craniofacial Research Grant RO1 DE12879.

REFERENCES 1. Witkop CJJ. Amelogenesis imperfecta, dentinogenesis imperfecta and dentin dysplasia revisited, problems in classification. J Oral Pathol 1989;17:547-53. 2. Backman B, Holmgren G. Amelogenesis imperfecta: a genetic study. Hum Hered 1988;38:189-206. 3. Chosack A, Eidelman E, Wisotski I, Cohen T. Amelogenesis imperfecta among Israeli Jews and the description of a new type of local hypoplastic autosomal recessive amelogenesis imperfecta. Oral Surg 1979;47:148-56. 4. Wright JT, Hart PS, Aldred MJ, et al. Relationship of phenotype and genotype in X-linked amelogenesis imperfecta. Con Tis Res 2003;44(Suppl):72-8. 5. OMIM. Online Mendelian Inheritance in Man. In: National library of medicine. Baltimore (Md): Center for Medical Genetics, Johns Hopkins University and National Center for Biotechnology Information; 2003. 6. Rajpar MH, Harley K, Laing C, Davies DM, Dixon MJ. Mutation of the gene encoding the enamel-specific protein, enamelin, causes autosomal-dominant amelogenesis imperfecta. Hum Mol Genet 2001;10:1673-7. 7. Mardh KC, Backman B, Holgren G, Hu JC-C, Simmer JP, Forsman-Semb K. A nonsense mutation in the enamelin gene causes local hypoplastic autosomal dominant amelogenesis imperfecta (AIH2). Hum Mol Genet 2002;11:1069-74. 8. Hart PS, Michalec MD, Seow WK, Hart TC, Wright JT. Identification of the enamelin (g.8344delG) mutation in a new kindred and presentation of a standardized ENAM nomenclature. Arch Oral Biol 2003;48:589-96. 9. Downey LM, Keen TJ, Jalili IK, et al. Identification of a locus on chromosome 2q11 at which recessive amelogenesis imperfecta

230 Nusier et al

10. 11. 12. 13. 14. 15. 16. 17. 18. 19.

20. 21.

and cone-rod dystrophy cosegregate. Eur J Hum Gene 2002;10: 865-9. Sundell S, GK. Hereditary amelogenesis imperfecta: epidemiology and clinical classification in a Swedish child population. Swed Dent J 1985;9:157-69. Sundell S, Valentin J. Hereditary aspects and classification of hereditary amelogenesis imperfecta. Community Den Oral Epidemiol 1986;14:211-6. Witkop CJ, Sauk JJ. Heritable defects of enamel. In: Stewart R, Prescott G, editors. Oral facial genetics. St. Louis (Mo): Mosby; 1976. p. 151-226. Ooya K, Nalbandian J, Norkura T. Autosomal recessive rough hypoplastic amelogenesis imperfecta. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1988;65:449-58. Peters E, Cohen M, Altini M. Rough hypoplastic amelogenesis imperfecta with follicular hyperplasia. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1992;74:87-92. Nakata M, Kimura O, Bixler D. Interradicular dentin dysplasia associated with amelogenesis imperfecta. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1985;60:182-7. Williams SA, Ogden AR. Failure of eruption associated with anomalies of the dentition in siblings. Pediatr Dent 1988;10: 130-5. Aldred MJ, Savarirayan R, Crawford PJM. Amelogenesis imperfecta: a classification and catalogue for the 21st century. Oral Dis 2003;9:19-23. Hart PS, Hart TC, Simmer JP, Wright JT. Establishment of a nomenclature for X-linked amelogenesis imperfecta. Arch Oral Biol 2002;47:255-60. Wright JT, Aldred MJ, Crawford PJM, Kirkham J, Robinson C. Enamel ultrastructure and protein content in X-linked amelogenesis imperfecta. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1993;76:192-200.. Wright JT, Robinson C, Kirkham J. Enamel protein in smooth hypoplastic amelogenesis imperfecta. Pediatr Dent 1992;14:331-7. Witkop C, Kuhlamnn W, Sauk J. Autosomal recessive pig-

ORAL SURGERY ORAL MEDICINE ORAL PATHOLOGY February 2004

22. 23. 24.

25. 26.

27.

28.

29.

mented hypomaturation amelogenesis imperfecta. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1973;36:367-82. Wright JT. Analysis of a kindred with amelogenesis imperfecta. J Oral Pathol 1985;14:366-76. Seow WK. Clinical diagnosis and management strategies of amelogenesis imperfecta variants. Pediatr Dent 1993;15:384-93. Wright JT, Butler WT. Alteration of enamel proteins in hypomaturation amelogenesis imperfecta. J Dent Res 1989;68:132830. Wright JT, Hall KI, Yamauchi M. The enamel proteins in human amelogenesis imperfecta. Arch Oral Biol 1997;42:149-59. Ooya K, Nalbandian J, Noikura T. Autosomal recessive rough hypoplastic amelogenesis imperfecta. a case report with clinical, light microscopic, radiographic, and electron microscopic observations. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1988;65:449-58.. Persson M, Sundell S. Facial morphology and open bite deformity in amelogenesis imperfecta. Acta Odontol Scand 1982;40: 135-44. Rowley R, Hill FJ, Winter GB. An investigation of the association between anterior open-bite and amelogenesis imperfecta. Am J Orthodont 1982;81:229-35. Cartwright AR, Kula K, Wright JT. Craniofacial features associated with amelogenesis imperfecta. J Craniofac Genet Dev Biol 1999;19:148-56.

Reprint requests: Tim Wright Department of Pediatric Dentistry School of Dentistry Brauer Hall #7450 University of North Carolina Chapel Hill, NC 27599 USA [email protected]