Phenotypic variation in Waardenburg syndrome: mutational heterogeneity, modifier genes or polygenic background?
Phenotypic variation in Waardenburg syndrome: mutational heterogeneity, modifier genes or polygenic background? Arti Pandya1,*, Xia-Juan Xia1, Barbara L. Landa1, Kathleen S. Arnos2, Jamie Israel2, Joyce Lloyd1, Anthony L. James1, Scott R. Diehl3, Susan H. Blanton4 and Walter E. Nance1
1Department of Human Genetics, Medical College of Virginia, Richmond, VA 23298, USA, 2Gallaudet University, Washington D.C. 20002, USA, 3Molecular Epidemiology and Disease Indicators Branch, DEODP, NIDR, National Institutes of Health, Bethesda, Maryland 20892, USA and 4Department of Pediatrics, University of Virginia, Charlottesville, Virginia 22908, USA
Received November 21, 1995;Revised and Accepted January 19, 1996
We have identified 11 mutational changes in the PAX3 gene in patients with type 1 Waardenburg syndrome (WS1) including three in the paired domain, six within or immediately adjacent to the homeodomain and two previously described polymorphic variants in exons 2 and 6. The affected members of one family carried substitutions involving two base pairs separated by one unaltered codon. Two of the deleterious mutations were identical and three others were identical to previously reported mutations. A comparison of clinical findings in families carrying substitutions in the same codon failed to reveal conspicuous similarities. Although subtle mutation-specific effects may well exist, allelic heterogeneity clearly cannot account for within family variation. However, the striking concordance of a pair of monozygotic twins with Waardenburg syndrome (WS) and previous reports of similar pairs indicate that phenotypic variation in WS has a genetic basis. If the genetic effects are mediated by oligogenic epistasis, as studies in the mouse suggest, it may ultimately be possible to predict clinically relevant aspects of the Waardenburg phenotype.
Waardenburg syndrome is an autosomal dominant disorder of neural crest differentiation which was first described in 1948 by a Dutch ophthalmologist (1 ). The cardinal features include deafness, dystopia canthorum and pigmentary changes of the eyes, hair and skin. Less common but clinically relevant findings include spina bifida (2 ,3 ), Hirschsprung disease (4 ) and gastrointestinal dysmobility (5 ). The phenotype is quite variable, even within sibships. Clinically the syndrome has been classified into three types. The Waardenburg Syndrome Consortium developed a diagnostic algorithm for WS1 which includes consideration of family history, major and minor features, and a numerical estimate of the canthal index (6 ,7 ). WS type 2 (WS2) differs from type 1 in that dystopia canthorum is generally absent. WS type 3 (WS3) is a more severe form in which there are associated limb abnormalities. All or most WS1 cases are caused by mutations at the PAX3 locus on 2q35 while all or most WS2 families are unlinked to PAX3. WS3 can result from homozygosity for a PAX3 mutation (8 ) or reflect an unusual manifestation of the heterozygous phenotype (9 ,10 ). PAX3 belongs to a family of genes which code for structurally related transcription factors. The PAX3 protein contains two highly conserved DNA binding motifs, a paired domain and a paired type homeobox, as well as a highly conserved octapeptide and a 3' Ser-Thr-Pro rich region which is involved in transcriptional activation (11 ,12 ). The recessive splotch mutation in the murine homolog of PAX3 is associated with pigmentary, craniofacial and auditory abnormalities (13 ) as well as spina bifida (14 ). Developmental studies have shown that Pax-3 is expressed in early neurogenesis in the spinal cord and brain as well as in the embryonic precursors of neural crest cells, the craniofacial mesectoderm and limb mesenchyme (15 ). Approximately 20% of WS2 mutations map to the human homolog of the mouse microphthalmia locus (MITF) on 3p12 (16 ). Other families which include many features of WS2 along with Hirschsprung's disease are caused by mutations in the endothelin-B receptor locus (EDNRB) on 13q22 (17 ). Finally some families with piebaldism resulting from mutations in KIT on 4q12 can exhibit features which mimic WS.
To date, more than 40 mutations in the PAX3 gene have been reported (18 -20 ) including changes in the paired and homeo domains and in the 3' Ser-Thr-Pro rich region of the gene. Virtually all previously reported mutations have been `private' or unique to single families which has frustrated attempts to correlate phenotypic variation with specific mutations. Our goals are to further characterize the spectrum of PAX3 mutations seen in patients with WS1, to compare the phenotypic features of the syndrome in families with the same or similar mutations, and to show that even within family variation in WS is genetic in origin.
This report is based on 11 families with a clinical diagnosis of WS1 ascertained from the Genetics Clinic of the Medical College of Virginia and the Genetic Services Center at Gallaudet University in Washington D.C. A total of 89 available members in the 11 families were examined using the Waardenburg Consortium protocol. Table 1 shows the distribution of clinically defined or anamnestic features in the 61 affected members of eight families in which a pathologic mutation was detected. The W-Index was calculated using the formula proposed by Arias (6 ), and the mean for each family is given. A cut off of greater than 1.95 is considered diagnostic of WS1. Patients with either unilateral or bilateral hearing loss are classified as affected with deafness in the table. None of the affected individuals manifest gastrointestinal symptoms. Clinical data on the two families in whom polymorphic PAX3 variants were detected are not shown in the table. The eleventh family included an affected pair of 21 year old female twins in whom a PAX3 mutation has not yet been identified. The twins were reportedly dichorionic but typing for six polymorphic DNA markers yielded a probability of monozygosity of greater than 0.999. The twins were concordant for dystopia, blue irides, chronic constipation and profound sensorineural hearing loss, with remarkably similar audiograms.
All affected and normal family members were screened for single stranded conformational polymorphism (SSCP) changes in exon 2 through 8 of the PAX3 gene. The primers used for amplification of these exons are shown in Table 2 . These primers were designed to include the intron-exon boundaries based on the sequence structure reported by Burri et al. (21 ), and Tassabehji et al. (22 ). To assess the frequency of non pathologic base pair substitutions, 25 normal controls were also screened in addition to the unaffected family members. All of the controls yielded normal test results. A positive SSCP result was characterized by the presence of extra bands on the autoradiogram. Figure 1 a shows an example of such a mobility shift seen in exon 5 for family VCU 56.
Once a putative mutation was detected by SSCP analysis, confirmation was achieved by direct sequencing and by cloning and sequencing the double stranded product from one affected and one normal member from each family. The mutation was confirmed in other affected members by restriction enzyme digestion or by allele specific oligonucleotide hybridization. A summary of the mutations is given in Table 3 .
We describe nine exonic mutations in the PAX3 gene which cause WS1. Six of the mutational changes have not been described previously but two of these are identical, and the remaining three are the same as previously described changes. Three mutations involve the paired domain in exon 2 and include one amino acid replacement, one nonsense substitution, and a single base pair frameshift deletion. The Gly48 -> Arg substitution in family VCU 13 changes a residue conserved in all nine of the PAX proteins as well as in the paired domains of the Drosophila paired, gooseberry and gooseberry neuro genes (24 ). A substitution in the same amino acid (Gly -> Ser) in the Pax-1 protein results in the undulated mutation in the mouse (25 ). In the Drosophila paired protein, Gly48 contacts the sugar phosphate backbone and also makes base contacts in the minor groove (24 ). These data are all consistent with the deleterious effects observed with the polar Gly48 -> Arg substitution in family VCU 13. Six mutations occur within or near the homeodomain in exons 5 and 6, and include three nonsense substitutions and the replacement by nonpolar amino acids of adjacent arginine residues at positions 270 and 271 within the third helix of the homeodomain. This is a region known to be involved in DNA recognition. The Arg271 is a highly conserved residue that is directly involved in the binding of the homeodomain of the Drosophila paired protein to the phosphate backbone in the major groove of DNA (26 ). This substitution was accompanied by a second replacement of Arg273 -> Lys. The polar amino acid lysine is also found at this position in PAX6 (27 ) and several other homeodomain genes (26 ). In PAX3, a single substitution at this site could therefore represent a polymorphism or possibly result in a gain of function. Although the occurrence of a double substitution is unusual, it is not without precedent; similar events have been described at the cystic fibrosis locus (28 ). As noted below, the Arg271 replacement alone can cause WS1, but it is not clear what effect the isolated replacement of Arg273 by lysine would have, if any. Possibly, in the presence of one harmful mutation at Arg271, selection against further changes in the gene may be relaxed.
The two identical nonsense mutations in the homeodomain are Glu210 -> Term (G -> T)628 substitutions, while the third is identical to the Arg -> Term (C -> T)667 substitution reported previously by Baldwin et al. (18 ). These investigators noted that chain termination appears to occur with greater frequency among homeodomain mutations whereas amino acid substitutions are more frequent in the paired domain. While we have now observed a nonsense mutation in the paired domain as well (family VCU 2), our observation of two recurrent nonsense substitutions within the homeodomain is consistent with their findings. Tassabehji et al. (20 ) have characterized the region extending from amino acids 266-271 in the homeodomain as being a mutational `hot spot'. Our observation of two recurrent mutations in this region, including the fourth reported substitution involving the first two base pairs of the Arg271 codon, and a double mutation involving that codon and a nearby residue are consistent with their interpretation. However, it will be important to compare the haplotypes and ethnic origins of `recurrent' mutations in this region to exclude a founder effect, as in the case of some BRCA1 mutations (29 ). The general similarity in the phenotypic effects of WS1 mutations suggests haploinsufficiency as the causal mechanism. Our failure to observe substitutions of amino acids in the PAX3 homeodomain that are directly involved in DNA sequence recognition may indicate that some of these substitutions result in a gain of function that is either lethal or not recognizable as WS1. Baldwin et al. (18 ) found that deafness was more common in patients who carry a mutation in the paired domain. When we include subjects in whom historical data were available along with those who were examined, the two groups do not differ significantly in our data: deafness is present in eight of 22 subjects (36%) with a paired domain mutation and 17 of 39 (43%) with mutations elsewhere in the gene.
Our discoveries provide an opportunity to compare the effects of the same mutational changes in different families. In families VCU 3 and VCU 56 identical nonsense substitutions result in a premature chain termination, nine codons 5' to the start of the homeodomain. Phenotypic data was available on the proband and adopted mother in family VCU 56 and on six affected members of family VCU 3 (Table 1 ). Features in common include dystopia, deafness, white forelock, poliosis and skin hypopigmentation. The proband in family VCU 56 also had heterochromia. Since the two WS1 genes had the same repeat number at the polymorphic CA repeat in the 5' regulatory region of the gene, they may be descendants of a common ancestor. Further characterization of the haplotypes will evaluate the possibility.
In family VCU 27, we observed an Arg271 -> His (G -> A)812 and an Arg273 -> Lys (G -> A)818 substitution. Lalwani et al. (19 ) have described a different substitution of arginine 271 by glycine. In their family, all six affected members had dystopia canthorum. Hearing impairment and premature greying was noted in five of the six and two had brilliant blue eyes or heterochromia. In our family there is hearing loss and premature greying in four of eight affected family members and iris hypopigmentation in three of eight. Tassabehji et al. (20 ) also reported a family with an isolated Arg271 -> His (G -> A)812 substitution but did not describe the clinical findings.
The Arg223 -> Term (C -> T)667 nonsense substitution within the homeodomain in family VCU 29 is identical to a mutation described previously by Baldwin (18 ). Their family contained a high proportion of affected individuals with premature greying and heterochromia irides, but not white forelock. In our family, nine of 14 affected individuals had deafness, eight had a white forelock, and two had iris hypopigmentation.
In family VCU 50 the Arg270 -> Cys (C -> T)808 substitution is identical to the mutation recently described by Tassabehji et al. (20 ). Our family was of German extraction, but details about the ethnic origin and clinical features of the previously reported family were not provided.
Although the effects of these homologous substitutions are not conspicuously similar, as more data on recurrent mutations accumulate, it should be possible to conduct rigorous tests for variation in the effects of different WS1 mutations. In a multivariate factor analysis of the WS phenotype, Reynolds et al. (30 ) have shown that the clinical features cluster into a Facial Factor, comprised of dystopia, synophrys and hypoplasia of the nasal alae, and a Pigmentary Factor including deafness, white forelock, and hypopigmentation of the skin and irides. A comparison of the factor scores within and among families with the same mutations would permit the tests for effects specific to particular mutations.
Although subtle mutation specific effects may ultimately be demonstrable, it already seems clear that mutational heterogeneity can not account for the substantial phenotypic variation that exists between and within families carrying the same mutation. Our observations on the concordance of monozygotic (MZ) twins with WS strongly suggest that variation in the expression of WS is genetically determined. The pair we report are concordant for dystopia, bilateral blue eyes, chronic constipation and profound sensorineural deafness with remarkably similar audiograms. Waardenburg (31 ) refers to two similar MZ pairs who were also concordant for deafness (32 ,33 ). From these observations, we cannot infer whether the genetic effects are mediated by the `polygenic background' or by interactions with specific genetic modifiers. However, Puffenberger et al. (34 ) recently mapped a recessive gene for Hirschsprung disease to the endothelin B receptor (EDNRB) locus on 13q22. Many affected family members showed features of WS including deafness and pigmentary abnormalities. The authors used identity-by-descent and association mapping to detect an anonymous locus at 21q22 which appears to modify the phenotypic expression of the EDNRB gene. In the mouse, Ednrb mutations cause the Piebald phenotype which varies greatly in different mouse strains. Paven et al. (35 ) mapped six modifier loci which explained virtually all of these differences including three plausible candidates: Mgf, Kit and e. Thus, mutations at a cluster of genes that influence melanocyte function can produce effects in man or the mouse which are quite variable and exhibit considerable phenotypic overlap. In man, the `melanocyte cluster' may include PAX3, KIT, EDNRB and its ligand EDN3, MITF, MGF, RET, MSHR and its ligand MSH, as well as OCA1 and OCA2 which are known to exhibit epistatic interactions (36 ). In the presence of a major gene at one of these loci, dominant or recessive epistatic modifiers at other members of the cluster may account for phenotypic variation in the expression of the major gene. This postulated system of oligogenic epistases within gene clusters is very different from either multifactorial or polygenic transmission where the genetic effects are assumed to result from the small additive contributions of genes at many loci. If applicable to WS, knowledge of the genotype at relevant modifier loci may allow prediction and possibly the treatment of clinically relevant aspects of the phenotype, including deafness, gastrointestinal dysmobility, Hirschsprung disease or spina bifida. Finally, because of its associated pleotropism the deafness in WS can be recognized as being genetic even though its penetrance is only 20%. For other gene clusters, such as those affecting blood pressure, cancer, Alzheimers disease, cleft palate, or schizophrenia, the major locus may not exhibit pleiotropy and cases resulting from oligogenic epistasis may be falsely identified as sporadic or nongenetic. A more complete understanding of gene expression in the melanocyte cluster could thus provide insights that are relevant to many other common genetic diseases.
DNA was isolated from whole peripheral blood specimens or transformed lymphocytes using the inorganic method of extraction described by Bell et al. (37 ).
A quantity of 500 ng of genomic DNA was amplified in a 25 µl volume containing 2.5 mM MgCl2, 200 µM each of dATP, dGTP, dCTP and 20 µM dTTP, 10 pmol of each oligonucleotide primer, 0.1 µl [32P]dTTP (800 Ci/mmole) and 2.5 U Taq Polymerase. After initial denaturation for 4 min at 94oC, 30 cycles of amplification were performed as follows: denaturation (1 min at 94oC), annealing (1 min at 68oC), and extension (1.5 min at 72oC), followed by a final 10 min extension at 72oC. Five µl of PCR product were diluted in 10 µl quench buffer (0.1% sodium dodecyl sulfate in 10 mM ethylene diamine tetraacetic acid) and 12.5 µl 20 mM EDTA in 98% formamide containing 0.05% each of bromophenol blue and xylene cyanol. The samples were denatured for 8 min at 85oC and rapidly cooled on ice. Five µl aliquots were loaded onto a 8% polyacrylamide (acrylamide/bisacrylamide ratio of 49:1) with 5% glycerol 38 × 50 cm nondenaturing sequencing gel in 1* TBE buffer.
Electrophoresis was carried out in 0.5* TBE upper buffer and 1* TBE lower buffer, at room temperature at 5 W for 12.5 h and 20 W for 11 h. Cold room gels were run without glycerol at 50 W for 10-14 h. Gels were dried and exposed overnight to an autoradiogram with an intensifying screen at -70oC.
Genomic DNA from each individual was amplified using primers for exons 2 through 8 shown in Table 2 . Cycling was performed on a Perkin Elmer 9600 thermal cycler for 35 cycles with the optimal annealing temperature varying from 58oC to 68oC for each exon.
The PCR product was quantitated using a [lambda] DNA standard on a 3% NuSieve Gel. For direct sequencing the sequence PCR Product Sequencing Kit marketed by USB was used according to their protocol.
Cloning of DNA fragments for sequencing was performed using the Eukaryotic TA Cloning Kit marketed by Invitrogen. Subsequently plasmid DNA was isolated using the DNA miniprep kit (Bio-RAD) and 10 µg was used to sequence using the Sequenase version 2.0 DNA Sequencing Kit (USB).
The exon 5 mutation in family VCU 29 was confirmed in the rest of the affected members using AlwNI restriction enzyme digestion.
Exon 2 changes in family VCU 13, and exon 6 changes in families 27 and 50 (example Fig. 2 ), were confirmed in all the affected members by allele specific hybridization using a slight modification of the method described by Thein (38 ).
The authors gratefully acknowledge the technical help of Wanda Hunt and Razieh Javaheri. We also thank Drs J. P. Fryns and J. J. Cassiman from the University of Leuven, Leuven, Belgium, for referring family VCU 52 for study. This work was supported by research grant number 5 R01 DC 00038-05 from the National Institute on Deafness and other Communication Disorders, NIH.
1 Waardenburg, P.J. (1951) A new syndrome combining developmental anomalies of the eyelids, eyebrows and nose root with pigmentary defects of the iris and head hair and with congenital deafness. Am. J. Hum. Genet. 3: 195-253.
2 Arnos,K.S., Pandya,A., Markello,T., Israel,J., Landa,B., Blanton,S. and Nance,W.E. A kindred with unusual features of Waardenburg syndrome and familial tremor. [Abstract] (1994) Am. J. Hum. Genet.55: 426.
3 Chatkupt,S., Chatkupt,S., Johnson,W.G. (1993) Waardenburg syndrome and myelomeningocele in a family. J. Med. Genet. 30: 83-84.MEDLINE Abstract
4 Omenn,G.S. and McKusick,V.A. (1979) The association of Waardenburg syndrome and Hirschsprung megacolon. Am. J. Med. Genet.3: 271-275.
5 Valenzuela,G., Weber, F.H., Nance,W. and McCallum,R.W. (1995) Waardenburg syndrome and gastric stasis in adults. V.M.Q. 122: 279-280.
6 Arias,S. and Mota M. (1978) Apparent non-penetrance for dystopia in Waardenburg syndrome type 1, with some hints on the diagnosis of dystopia canthorum. J. Genet. Hum. 26: 103-131. MEDLINE Abstract
7 Farrer,L.A., Grundfast,K.M., Arnos,J., Arnos, K.S., Asher, J.H. Jr., Beighton, P., Diehl, S.R. and Fex, Jorgen. (1992) Waardenburg syndrome (WS) type 1 is caused by defects at multiple loci, one of which is near ALPP on chromosome 2: First report of the WS Consortium. Am. J. Hum. Genet. 50: 902-913.MEDLINE Abstract
8 Zlotogora, J., Lerer, I., Bar-David, S., Ergaz, Z. and Abeliovich, D. (1995) Homozygosity for Waardenburg syndrome. Am. J. Hum. Genet.56:1173-1178.MEDLINE Abstract
9 Hoth, C.F., Milunsky, A., Lipsky, N., Sheffer, R., Clarren, S.K. and Baldwin, C.T. (1993) Mutations in the paired domain of the human PAX3 gene cause Klein-Waardenburg syndrome (WS-III) as well as Waardenburg syndrome Type 1 (WS-1). Am. J. Hum. Genet.52:455-462.MEDLINE Abstract
10 Pasteris, N.G., Trask, B.J., Sheldon, S. and Gorski, J.L. (1993) Discordant phenotype of two overlapping deletions involving the PAX3 gene in chromosome 2q35. Hum. Mol. Genet.2:953-959.MEDLINE Abstract
11 Chalepakis,G., Jones,F.S., Edelman,G.M. and Gruss,P. (1994) Pax-3 contains domains for transcription activation and transcription inhibition. Proc. Natl Acad. Sci. USA91: 12745-12749.MEDLINE Abstract
12 Bennicelli,J.L., Fredericks,W.J., Wilson,R.B., Rauscher,F.J. and Barr,F.G. (1995) Wild type PAX3 protein and the PAX3-FKHR fusion protein of alveolar rhabdomyosarcoma contain potent, structurally distinct transcriptional activation domains. Oncogene11: 119-130.MEDLINE Abstract
13 Asher, J.H. Jr., Gillikin, L., Swenson,L.B., Dolan,D.F., Morell, R., and Friedman, T.B. (1995) Hearing and deaf splotch mice: A murine model for deafness in Waardenburg syndrome? In: The Molecular Biology of Hearing and Deafness, Bethesda, Maryland, USA, October 6-8 , 1995, p. 46.
14 Moase,C.E. and Trasler,D.G. (1992) Splotch locus mouse mutants: models for neural tube defects and Waardenburg syndrome type 1 in humans. J. Med. Genet. 29: 145-151.MEDLINE Abstract
15 Goulding,M.D., Chalepakis,G., Deutsch,U., Erselius,J.R. and Gruss,P. (1991) Pax-3, a novel murine DNA binding protein expressed during early neurogenesis. EMBO J. 10: 1135-1147.MEDLINE Abstract
16 Hughes, A.E., Newton, V.E., Liu, X.Z. and Read, A.P. (1994) A gene for Waardenburg syndrome type 2 maps close to the human homologue of the mouse microphthalmia gene at chromosomal location 3p12p14.1. Nature Genet.7:509-512.MEDLINE Abstract
17 Puffenberger, E.G., Hosoda, K. and Washington, S.S. (1994) A missence mutation of the endothelin-B receptor gene in multigenic Hirschsprung's disease. Cell79:1257-1266.MEDLINE Abstract
18 Baldwin, C.T., Hoth, C.F., Macina, R.A. and Milunsky, A. (1995) Mutations in PAX3 that cause Waardenburg syndrome type I: Ten new mutations and review of the literature. Am. J. Med. Genet. 58:115-122.MEDLINE Abstract
19 Lalwani, A.K., Brister, J.R., Fex, J., Grundfast, K.M., Ploplis, B., San Agustin, T.B. and Wilcox, E.R. (1995) Further elucidation of the genomic structure of PAX3, and identification of two different point mutations within the PAX3 homeobox that cause Waardenburg syndrome type 1 in two families. Am. J. Hum. Genet.56:75-83.MEDLINE Abstract
20 Tassabehji,M., Newton,V.E., Lice,X., Brady,A., Donnai,D., and Krajewska-Walasek,M. (1995) The mutational spectrum in Waardenburg syndrome. Hum. Mol. Genet. 4: 2131-2137.MEDLINE Abstract
21 Burri, M., Tromvoukis, Y., Bopp, D., Frigerio, G. and Noll, M. (1989) Conservation of the paired domain in metazoans and its structure in three isolated human genes. EMBO J. 8:1183-90.MEDLINE Abstract
22 Tassabehji, M., Newton, V.E., Turnbull, K., Seemanova, E., Kunze, J., Sperling, K., Strachan, T. and Read, A.P. (1994) PAX3 gene structure and mutations: close analogies between Waardenburg syndrome and the splotch mouse. Hum. Mol. Genet.3: 1069-1074.MEDLINE Abstract
23 Wilcox,E.R., Rivolta,M.N., Ploplis,B., Potteri,S.B. and Fex,J. (1992) The PAX3 gene is mapped to human chromosome 2 together with a highly informative CA dinucleotide repeat. Hum. Mol. Genet. 1: 215.MEDLINE Abstract
24 Xu,W., Rould,M.A., Jun,S., Desplan,C. and Pabo,C.O. (1995) Crystal structure of a paired doman-DNA complex at 2.5 resolution reveals structural basis for Pax developmental mutations. Cell80: 639-650.MEDLINE Abstract
25 Balling,R., Deutsch,U. and Gruss,P. (1988) undulated, a mutation affecting the development of the mouse skeleton, has a point mutation in the paired box of Pax-1. Cell 55: 531-535.MEDLINE Abstract
26 Wilson,D.S., Guenther,B., Desplan,C. and Kuriyan,J. (1995) High resolution crystal structure of a paired (Pax) class cooperative homeodomain dimer on DNA. Cell82: 709-719.MEDLINE Abstract
27 Martin, P., Carriere, C., Dozier, C., Quatannens, B., Mirahel, M.A., Vandenbunder, B., Stehelin, D. and Saule, S. (1992) Characterization of a paired box- and homeobox- containing quail gene (Pax-QNR) expressed in the neuroretina. Oncogene7,1721-1728.MEDLINE Abstract
28 Macek, M., Vavrova, V., Davis, C.L., Hamosh, A. and Cutting, G.R. (1993) Two F508 homozygote siblings with delayed onset of pancreatic insufficiency carry a second cystic fibrosis transmembrane conductance regulator gene mutation on one allele. [Abstract] Am. J. Hum. Genet.53: 83.
29 Struewing,J.P., Abeliovich,D., Peretz, T., Avishai,N., Kaback,M.M., Collins,F.S. and Brody,L.C. (1995) The carrier frequency of the BRCA1 185delAG mutation is approximately 1 percent in Ashkenazi Jewish individuals. Nature Genet.11: 198-200.MEDLINE Abstract
30 Reynolds, J.E., Meyer, J.M., Landa, B., Stevens, C.A., Arnos, K.S., Israel, J., Marazita, M.L., Bodurtha, J., Nance, W.E. and Diehl, S.R. (1995) Analysis of variability of clinical manifestations in Waardenburg syndrome. Am. J. Med. Genet.57:540-547.MEDLINE Abstract
31 Waardenburg, P.J., Franceschetti, A. and Klein, D. (1961) Lacrimal apparatus.In: Charles C. Thomas, IV, Genetics and Ophthalmology, pp. 291-300.
32 Hoeve, J.V.D. (1916) Abnorme Lange der Tranenrorhrchen mit Ankyloblepharon. Klin. Mbl. Augenheilk.56:232.
33 Lavergne, G. (1960) Problemes d'eugenisme poses par une famille atteinte de syndrome de Waardenburg-Klein. Soc. belge. Ophtal.193:181.
34 Puffenberger, E.G., Kauffman, E.R., Bolk, S., Matise, T.C., Washington, S.S., Augrist, M., Weissenbach, J., Garver, K., Mascari, M., Ladda, R., Slangenhaupt, S.A. and Chakravart, A. (1994) Identity by descent and association mapping of a recessive gene for Hirschsprung disease on chromosome 13q22. Hum. Mol. Genet.3:1217-1225.MEDLINE Abstract
35 Paven, W.J., Mac, S., Cheng, M. and Tilghman, S.M. (1995) Quantitative trait loci that modify the severity of spotting in piebald mice. Genome Res.5:29-41.
36 Tripathi, R.K., Droetto, S. and Spritz, R.A. (1992) Many patients with `tyrosinase positive' ocalocutaneous albinism have tyrosinase gene mutations. Am. J. Hum. Genet.51:179.
37 Bell, G.I., Karam, J.H. and Rutter, W.J. (1981) Polymorphic DNA region adjacent to the 5' end of the human insulin gene. Proc. Natl Acad. Sci. USA78:5759-5763.MEDLINE Abstract
38 Thein, S.L. (1995) The use of synthetic oligonucleotides as specific hybridization probes in the diagnosis of genetic disorders. In: Davies, K.E. ed. Human Genetic Disease Analysis: A practical approach,2nd ed. Oxford: IRL Press, pp. 21-33.
*To whom correspondence should be addressed
This page is maintained by OUP admin. Last updated Thu Oct 31 15:23:07 GMT 1996. Part of the OUP Journals World Wide Web service.Copyright Oxford University Press, 1996
