GENOMICS
l&971-975
(1991)
A Deletion in the Amelogenin Gene (AMG) Causes X-Linked Amelogenesis Imperfecta (AIHI ) MARIA *Department
NIKLAS DAHL,* YUTAKA NAKAHORI, tn2 YASUO NAKAGOME, LAGERSTR~M,*~’ BIRGITTA BACKMAN,* ULF LANDEGREN,* AND ULF PETTERSSON*
of Medical Abnormalities
t
Genetics, University of Uppsala, Biomedical Center, Uppsala, Sweden; tDepartment of Congenital Research, National Children‘s Medical Research Center, Taishido, Tokyo 754, Japan; and *Department of Pedodontics, University of Urned, Urned, Sweden Received
February
12, 1991;
revised
April 8, 1991
genesis imperfecta have been described (Witkop, 1989). Two different clinical forms of AIHl have been observed (Backman, 1988, Witkop, 1989), characterized predominantly by either hypoplasia of the enamel or hypomineralization. In the hypoplastic form the enamel is correctly mineralized but exhibits macroscopic defects, while in the hypomineralization form the enamel is soft and insufficiently mineralized (Weinman et al., 1945). There is no apparent correlation between the mode of inheritance and the clinical form of amelogenesis imperfecta. Amelogenin and enamelin are the predominant proteins in enamel (Termine et aZ., 1980). The amelogenin gene was recently mapped to the p22 region of the human X chromosome using a mouse amelogenin cDNA and rodent-human somatic cell hybrids (Lau et aZ., 1989). The gene for AIHl has been assigned to the ~22.3 region of the X chromosome by linkage analysis (Lagerstrom et al., 1990). The amelogenin gene thus represents a candidate gene for AIHl. To implicate the amelogenin gene in the causation of AIHl it is necessary to demonstrate structural alterations correlating with the disease. In the present study we describe a 5-kb deletion, including at least two exons of the amelogenin gene (Nakahori et aZ., 1991), that segregates with the disease in a family with AIHl.
Amelogenesis imperfecta is characterized by the defective formation of tooth enamel. Here we present evidence that the X-linked form of this disorder (AIHl) is caused by a structural alteration in one of the predominant proteins in enamel, amelogenin. Southern blot analysis revealed a deletion extending over 6 kb of the amelogenin gene in males with the hypomineralization form of the AIHl. Carrier females were heterozygous for the molecular defect. The deletion appears to include at least two exons of the amelogenin gene and the extent of the deletion was verified by PCR analysis. The mutation was shown to segregate with the disease among 15 analyzed individuals belonging to the same kindred. Our results link a defect in the amelogenin gene to the abnormal formation of enamel. We thus conclude that the amelogenin protein has a role in biominerabzation of tooth enamel. o isei AC&I& PWS, I~C.
INTRODUCTION
Amelogenesis imperfecta represents a collection of genetic disorders affecting the formation of enamel in the primary as well as the permanent dentition. The reported prevalence of the disease varies from 7:100,000 in the United States (Witkop, 1958) to 140:100,000 in the northern part of Sweden (Backman and Holm, 1986). Amelogenesis imperfecta is genetically heterogeneous since families demonstrating autosomal dominant, autosomal recessive, and X-linked (AIHl) inheritance have been reported (Witkop and Sauk, 1976; Biickman and Holmgren, 1988). A variety of clinical manifestations of amelo-
MATERIALS
AND
METHODS
Subjects The study comprises a Swedish family characterized by the hypomineralized form of AIHl. This family has previously been studied by linkage analysis (Family B in Lagerstrom et al., 1990). The main clinical feature of the affected males is an enamel that is softer than normal (Witkop and Rae, 1971; Backman and Holmgren, 1988). Females, heterozygous for the
1 To whom correspondence and reprint requests should be addressed at Department of Medical Genetics, Biomedical Center, Box 589, S-761 23 Uppsala, Sweden. 2 Present address: Molecular Genetics Group, Institute of Molecular Medicine, John Radcliffe Hospital, Headington, Oxford OX3 9DU, U.K. 971
All
o&33-7543/91 $3.00 Copyright 0 1991 by Academic Press, Inc. rights of reproduction in any form reserved.
972
LAGERSTROM
ET
AL.
20-~1 sample of each amplification rated on a 1.5% agarose gel.
reaction
was sepa-
RESULTS
-
Ak
1 Kb
Ai!
FIG. 1. Schematic drawing of a region from the human amelogenin gene (Ref. (9)). Selected restriction enzyme cleavage sites are indicated as well as fragments A, B, and C which were used as probes for Southern blot hybridization. Three identified exons of the amelogenin gene (Ref. (9)) are shown as black boxes. The precise extent of the deletion (shown as an open box) is not known. The deletion begins 5’ of the BamHI site and the 3’ breakpoint has been mapped between the second and third XbaI site. It encompasses 5 kb and thus includes two or three exons of the amelogenin gene. The locations of the oligonucleotides used for PCR analysis are represented by arrows.
mutant gene, exhibit varying degrees of vertically arranged ridges of abnormal and normal appearing enamel. This distribution has been attributed to the random inactivation of the X chromosome (Berkman and Singer, 1971).
Southern Blot Analysis Total genomic DNA was prepared from peripheral blood collected from AIHl patients, their relatives, and normal controls. Ten-microgram aliquots of DNA were digested with different restriction endonucleases. The samples were subjected to electrophoresis through 0.9% agarose gels and analyzed by Southern blot hybridization (Southern, 1975) using nylon membranes (Pall Biodyne, BioSupport Division, N.Y.). The amelogenin genomic segment and the gel purified subfragments were radiolabeled by random primed synthesis (Pharmacia oligolabeling kit).
Detection of a Deletion
in the Amelogenin
A genomic clone containing an 11-kb EcoRI fragment of the human amelogenin gene (Nakahori et al., 1991) was used as a probe in a Southern blot analysis of members of the AIHl family. The structure of the genomic fragment is outlined in Fig. 1. DNA samples from affected males, carrier females, unaffected family members, and normal controls were digested with EcoRI and subjected to Southern blot analysis. The results of the analysis revealed the expected 11-kb fragment in all the normal subjects. In the affected males this fragment was replaced by a 6-kb fragment, suggesting that approximately 5 kb had been deleted from the amelogenin gene in these patients (Fig. 2A). Carrier females exhibited both the ll- and the 6-kb fragments (Fig. 2A). Figures 2A and 2B also demonstrate that additional weak bands were observed in both normal and affected male subjects. These are likely to represent Y-chromosome-specific sequences which cross-hybridize with the amelogenin gene (Nakahori et al., 1991) and they did not differ when DNA samples from affected and normal individuals were compared.
Mapping
of the Amelogenin
Deletion
Three subcloned fragments of the 11-kb amelogenin segment, designated A, B, and C (Fig. l), were used as probes to define the extent of the deletion.
(B)
Polymerase
Chain Reaction (PCR)
Three primer pairs mapping within the sequenced part of the amelogenin gene were constructed. A fourth primer pair amplifying the seventh exon of the human cw-galactosidase A gene was used as a control (Table 1, Fig. 1). Samples of 0.5 pg genomic DNA were amplified in 50 miU KCl; 10 n&f Tris-HCl, pH 8.3; 1.5 n&f MgCl,; 125 mg/ml BSA; 0.035 unit/p1 of AmpliTaq (Cetus); 1 PM of each primer; and 200 PM each of dATP, dGTP, dCTP, and dTTP in a volume of 50 ~1 overlaid with l-2 drops of mineral oil (Sigma). The reactions were incubated at 94°C for 1 min, 55°C for 1 min, and 72’C for 3 min for 30 cycles. A final incubation at 72°C for 7 min followed the last cycle. A
Gene
1
2
3
4
(A) 11 kb * 6kb
a9
l
2kb
FIG. 2. Southern blot analysis of DNA samples from a normal female (lane l), a carrier female (lane 2), a normal male (lane 3), and an affected male (lane 4). The asterisks indicate Y-chromosome-specific bands seen in all male subjects. (A) DNA samples digested with EcoRI and probed with an 11-kb fragment of the amelogenin gene. Normal individuals exhibit an 11-kb fragment, whereas a 6-kb band is present exclusively in the affected male and the carrier female. (B) DNA samples digested with X&I and probed with fragment B. In affected males and carrier females a 2-kb fragment is observed instead of the normal 4-kb fragment.
AMELOGENIN TABLE
AM1 AM2 AM3 AM4 AM5 AM6 g-41 g-42
IN AMELOGENESIS
1
Primer Pairs Used for PCR Analysis genin Gene (AMI-AM6) and a Primer ing a Segment of the cw-Galactosidase gA2), Used as an Internal Control Primer size
DELETION
of the AmeloPair AmplifyA Gene (gAl-
Sequence (5’-3’) AGGAGCTCCAGCCATAAGGCTATAACC GGCTCGAGGTTGAGGAGAACCTCAAAC ATACCCGGGTTTGAGGTTCTCCTCAAC ATCCCGGGTACTGGTGAGAAACAGAGA ACAGCTGGTTGGAGTCACCTGAGCCAAT TCGACTACCTTTGTAGCCTGTTCAG ACTATCAGTAATAAAGCTTCTTGC ATGAAGCTTTTAAAGTGAATGGAGAA
Product (base pairs) 840 1230 935 535
With fragments A and C as probes (Fig. l), identical hybridization patterns were obtained (data not shown) when DNA samples from normal and affected individuals were analyzed after cleavage with X&I, which cuts the 11-kb fragment in three places. This result suggests that the deletion is located between the two external XbaI cleavage sites (Fig. 1). Probing XbaI blots with the B fragment revealed an approximateIy 4-kb fragment in normal males and females. In affected males this fragment was replaced by a weakly hybridizing 2-kb fragment (Fig. 2B), suggesting that the deletion covers most of the region corresponding to fragment B. Further studies demonstrated that the KpnI site and the internal XbaI site were missing, whereas both the BamHI site and an AZuI site located 200 bp 3’ to the BamHI site were present (data not shown). The most likely localization of the deletion, on the basis of these findings, is indicated in Fig. 1. From the partial sequence that has been established for the human amelogenin gene (Nakahori et al., 1991) it can be concluded that the deleted region encompasses at least two exons of the amelogenin gene (Fig. 1).
PCR Analysis of the Deletion
in Affected Males
Sequence information from a 4-kb region surrounding the KpnI site in the amelogenin gene is available (Nakahori et al., 1991). Based on this sequence three different primer pairs were constructed (Table 1, Fig. 1) and PCR analyses were performed. No PCR products were seen when the DNA samples from affected males were amplified with any of the three primer pairs, whereas DNA samples from carrier females, unaffected family members, and normal con-
IMPERFECTA
973
trols gave PCR products of the expected sizes (Fig. 3). PCR analysis using a primer pair from the seventh exon of the a-galactosidase A gene was used as a control. All DNA samples gave PCR products of the expected size with this primer pair (Fig. 3). The deletion thus includes at least one of the primer sequences for each of the three PCR products from the amelogenin gene.
The Amelogenin Disease
Gene Deletion
Segregates with the
To ascertain that the observed deletion is the cause of AIHl in the affected individuals it was necessary to demonstrate that the deletion is confined to individuals diagnosed as affected or as carriers. By PCR analysis the deletion could be demonstrated in all affected males and the absence of the deletion was shown in all healthy males analyzed (Fig. 3). In addition, 15 individuals belonging to the same kindred were analyzed by Southern blotting (not shown). Fragments diagnostic for the deletion were demonstrated in each of the 4 affected males. The 8 carrier females were heterozygous with one abnormal and one normal band. Two unaffected males and one noncarrier female from the same kindred all exhibited normal hybridization patterns. The observed deletion in the amelogenin gene thus correlates with the expression of the dental abnormality in this family. We also investigated another family (Family A in Lagerstrom et al, 1990) characterized by hypoplasia of the enamel and mapping to the same genomic region as the above family. Southern blot and PCR analyses failed to reveal any abnormalities in affected members of this family (data not shown).
DISCUSSION
Tooth enamel is the hardest tissue in the vertebrate body and includes two main protein classes,enamelin and amelogenin (Termine et al., 1980). The amelogenin protein has been suggested to regulate the formation of enamel crystallites (Eastoe, 1965; Termine et al., 1980), and highly conserved amino acid sequences for amelogenin have been found in all mammalian species studied (Fincham et al., 1983; Slavkin et al., 1981; Snead et al., 1985). AIHl is an inherited disorder of enamel formation characterized by clinical heterogeneity. The disease locus has been assigned to the ~22.3 region of the X chromosome in a linkage analysis of two families exhibiting AIHl (Lagerstrom et al., 1990). The amelogenin gene has
974
LAGERSTRCM
ET
AL.
Affected
FIG. 3. sequence
PCR analysis of DNA samples from members of a family with AIHl. Primers AMl-AM6 are specific for the amelogenin gene and are described in Table 1. The primer set gAl-gA2 is used as control, amplifying a segment from the a-galactosidase A gene.
been independently mapped to the same region of the human X chromosome (Lau et al., 1989), suggesting that mutations in the amelogenin gene might be the cause of AIHl. In the present study a 5-kb deletion in the amelogenin gene was demonstrated in all affected males in a family characterized by hypomineralization AIHl. All carrier females in the same family were heterozygous for the deletion. The aberration was not observed in unaffected family members or normal controls. The deletion thus segregates with the disease among 15 individuals in the family pedigree. Our results show, moreover, that the deletion extends over a region in the amelogenin gene that contains at least two of the exons. We conclude from the present study that the hypomineralization form of AIHl is caused by a mutation in the amelogenin gene. In a previous study the hypoplastic and the hypomineralization forms of AIHl were shown to map to the same region of the X chromosome (Lagerstrom et al., 1990). It thus seems likely that mutations in the amelogenin gene cause both forms of AIHl and the difference might be related to the nature of the defect in the amelogenin gene. At present, we are unable to determine which part of the amelogenin protein is affected by the deletion or if the mutant gene remains in reading frame. In this regard it will be of interest to compare the mutations in the two families with AIHl mapping to the same region of the X chromosome, but exhibiting strikingly different effects on the formation of tooth enamel (Lagerstrom et al., 1990). By demonstrating that a defect in the amelogenin gene can cause AIHl, we are able to confirm the func-
tional role of the amelogenin formation of tooth enamel.
protein in regulating
the
ACKNOWLEDGMENTS Mrs. Elsy Johnsen is gratefully acknowledged for skillful technical assistance. Financial support for this project was provided by Swedish Medical Research Council Grants B33-19X-07511-03A and by the Swedish National Bank for Technical Development, the Bank of Sweden Tercentenary Foundation, the Marcus Borgstrbm Foundation, Pharmacia, and CRC Grant 9672-17.
REFERENCES 1.
BACKMAN, B., AND fecta: Prevalence county. Community
2. BACKMAN, perfecta:
Amelogenesis a northern
imperSwedish
Dent. Oral Epidemiol. 14: 43-47.
B., AND HOLMGREN, G. (1963). Amelogenesis A genetic study. Hum. Hered. 38: 189-206.
3. BACKMAN,
B.
manifestations
Stand. J. Dent. 4.
Horn, A-K. (1936). and incidence in
im-
(1936). Amelogenesis imperfecta-Clinical in 51 families in a northern Swedish county. Res. 96: 505416.
BERKMAN, M. D., AND SLNGER, A. (1971). Demonstration of the lyon hypothesis in X-linked dominant hypoplastic amelogenesis imperfecta. Birth Defects, Orig. Artic. Ser. 7: 204-209.
5. EASTOE, J. E. (1965). 6.
The chemical composition of bone and tooth. In “Advances in Flourine Research and Dental Caries Prevention,” Vol. 3, pp. 5-15, Pergamon Press, Oxford. FINCHAM, A. G., BELCOURT, A. B., -, J. D., BUTLER, W. T., m COTHRAN, W. C. (1983). Amelogenins: Sequence homologies in enamel-matrix proteins from three mammalian species. Biochm. J. 211: 149-154.
7. LAGERST~~M,
M., DML,
N., Isis,
L., BACKMAN,
B., AND
AMELOGENIN
8.
9.
10.
11.
12.
DELETION
IN AMELOGENESIS
PETTERSSON, U. (1990). Mapping of the gene for X-linked amelogenesis imperfecta by linkage analysis. Am. J. Hum. Gem%. 48: 120-125. LAU, E. C., MOHANDRAS, T. K., SHAPIRO, L. J., SLAVKIN, H. C., AND SNEAD, M. L. (1989). Human and mouse amelogenin gene loci are on the sex chromosomes. Genomics 4: 162168. NAKAHORI, Y., TAKENAKA, O., AND NAKAGOME, Y. (1991). A human X-Y homologous region encodes “amelogenin.” Genomics 9: 264-269. SLAVKIN, H. C., ZEICHNER-DAVID, M., AND SIDDIQUI, M. A. Q. (1981). Molecular aspects of tooth morphogenesis and differentiation. Mol. Aspects Med. 4: 125-188. SNEAD, M. L., LAU, E. C., ZEICHNER-DAVID, M., FINCHAM, A. G., WOO, S. L. C., AND SLAVKIN, H. C. (1985). DNA sequence for cloned cDNA for murine amelogenin reveals the amino-acid sequence for enamel-specific protein. Biochem. Biophys. Res. Commun. 129: 812-818. SOUTHERN, E. M. (1975). Detection of specific sequences
13.
14. 15. l6 17. 18.
IMPERFECTA
975
among DNA fragments separated by gel electrophoresis. J. Mol. Biol. 98: 503-517. TERMIN?& J. D., BELCOURT, A. B., CHRISTNER, P. J., CONN, K. M., AND NYLEN, M. U. (1980). Properties of dissociatively extracted fetal tooth matrix proteins. 1. Principal molecular species in developing bovine enamel. J. Biol. Chem. 255: 9760-9768. WEINMAN, J. P., SVOBODA, J. F., AND WOODS, R. W. (1945). Hereditary disturbances of enamel formation and calcification. J. Am. Dent. Assoc. 32: 397-418. WITKOP, C. J. (1958). Genetics and dentistry. Eugenet. Q. 5: 15-21. WITKOP, C. J., AND RAO, S. (1971). Inherited defect in tooth structure. Birth Defects, Orig. Artic. Ser. 7: 153-184. WITKOP, C. J., AND SAUK, J. J. (1976). Heritable defects of enamel. In “Oral Facial Genetics” (F. E. Stewart and J. Prescott, Eds.), pp. 151-226, Mosby, St. Louis. WITKOP, C. J., JR. (1989). Amelogenesis imperfecta, dentinogenesis imperfecta and dentin dysplasia revisited: Problems in classification. J. Oral Pathol. 17: 547-553.