American J o u r n a l of Medical Genetics 35:132-135 (1990)
Prenatal Diagnosis of X-Linked Hypohidrotic Ectodermal Dysplasia by Linkage Analysis Jonathan Zonana, Albert Schinzel, Meena Upadhyaya, Nicholas S.T. Thomas, Ingrun Anton-Lamprecht, and Peter S. Harper Institute of Medical Genetics, University of Wales College of Medicine, Cardiff, Wales (J.Z., M.U., N.S.T.T., P.S.H.); Department of Medical Genetics, and Child Development and Rehabilitation Center, Oregon Health Sciences University, Portland (J.Z.); Institut fur Medizinische Genetik, University of Zurich, Switzerland (A.S.), Institut fur Ultrastrukturforschung der Haut, Hautktinik der Ruprecht-Karls-Universittit,Heidelberg, Federal Republic of Germany (I.A.-L.)
Prenatal diagnosis of X-linked hypohidrotic ectodermal dysplasia was previously performed by the direct histological analysis of fetal skin obtained by late second trimester fetoscopy. The recent gene mapping of the locus for the disorder to the region of Xqll-21.1 now permits the indirect prenatal diagnosis of the disorder by the method of linkage analysis, based on closely linked marker loci, during the first trimester of pregnancy. We report the prenatal diagnosis of a male fetus with a high probability of the disorder by a linkage analysis utilizing restriction fragment length polymorphisms at the DXS159, PGK1, and DXS72 loci, from a DNA sample obtained by a chorionic villus biopsy at 9 weeks gestation. After further counseling, the pregnancy was terminated but the diagnosis could not be confirmed by histological analysis, even though analysis of skin samples by light and electron microscopy showed lack of hair germs, primary dermal ridges, and sweat gland primordia, due to the early developmental stage of the fetus. The use of DNA-based linkage analysis now offers the opportunity for an earlier diagnosis of X-linked hypohidrotic ectodermal dysplasia by a method other than fetal skin sampling. However, families must also fully understand the present limitations of the method prior to undertaking the procedure.
KEY WORDS: DNA polymorphism, chorionic villus sampling, epidermal appendages
Received for publication May 26,1989; revision received July 28, 1989. Address reprint requests to Jonathan Zonana, M.D., Dept. of Medical Genetics L-103, Oregon Health Sciences University, 3181 SW Sam Jackson Park Rd., Portland, OR 97201-3098.
0 1990 Wiley-Liss, Inc.
INTRODUCTION X-linked hypohidrotic (anhidrotic) ectodermal dysplasia (HED),a developmental disorder affecting structures of ectodermal origin, including eccrine sweat glands, teeth, and hair, is associated with early morbidity and mortality [Clarke et al., 19871. Affected males may have problems of psychosocial adjustment due to their appearance, and restricted activities secondary t o heat intolerance. Despite recent improvements in the early diagnosis and treatment of the condition, a number of obligate or at-risk carriers have inquired about the availability of prenatal diagnosis for the disorder. The gene locus for X-linked HED was recently localized by linkage analysis, using restriction fragment length polymorphisms (RFLP), t o chromosome region Xqll-21.1 [Zonana et al., 1988al. The use of RFLP markers has resulted in improved carrier detection, and has opened the possibility offirst trimester prenatal and neonatal diagnosis LZonana et al., 19891. Here we report the results of a prenatal diagnosis of HED by the use of linkage analysis. MATERIALS AND METHODS Clinical Report The consultand (11-Zj, a 34-year-oldgravida 4, para 0 woman, who had 3 therapeutic abortions, was a relative of one of the patients investigated previously as part of a gene mapping study ofthe HED locus (Fig. 1) [Zonana et al., 1988al. She had 2 brothers (11-3,11-4)with classical manifestations of HED, and she and her mother (1-2) were classified as manifesting carriers on the basis of patchy vellus hair on the trunk and lower limbs, and upon the lack of epidermal appendages on histological examination of their skin [Arnold et al., 19841. She had terminated her first 2 pregnancies (III-1&2)because of the 50%risk of HED, after amniocentesis demonstrated male fetuses. Her third pregnancy, another male fetus (111-3), was terminated subsequent to the analysis of multiple skin biopsies obtained by fetoscopy at 20 weeks gestation, Histological examination showed a lack of
Prenatal Diagnosis of X-Linked HED
I
mined from our previous study [Zonana et al., 1988aJ (Table I). Risk calculations were performed twice using different values for the genetic distances; in method 1 the maximum estimate of the recombination fraction was used, and in method 2 the recombination fraction which represents the upper value of the approximate 95% confidence interval. The former method would represent the most likely risk figure, but the latter method would be a more conservative approach, “erring” on the side of the fetus’s being unaffected.
II
111
133
1
2
3
RESULTS
Fig. 1. Pedigree of family with hypohidrotic ectodermal dysplasia with consultand individual 11-2 and her at-risk fetus 111-4.0, obligate carrier; 0, affected male; 0 , terminations involving male fetus; 3, affected male abortus.
skin appendages a t multiple biopsy sites, and these findings were subsequently confirmed by an analysis of the abortus [Anton-Lamprecht et al., 1982; Arnold et al., 19841. Her father (1-1)was deceased, and there were no other affected males or known female carriers in the family. Cytogenetic and DNA Analyses A chorionic villus sample (CVS) was obtained from the current pregnancy at 9 weeks of gestation, and the sample was split for cytogenetic examination and for DNA analysis. DNA extraction from blood and CVS samples, digestion with restriction enzymes, Southern blot analyses, and probe hybridizations were performed by previously described methods [Zonana et al., 1988a; Upadhyaya et al., 19841. Relevant relatives were typed with probes from 7 marker loci known to be linked to the HED locus, which spanned the region Xpl.1 to Xq22. The source of the probes, restriction enzymes, allele sizes, and the order of the loci have been described previously [Zonana et al., 1988al. The gene order of the closest markers and the disease locus (DXS159-HEDPGK1-DXS72) was based on linkage analysis and on the analysis of somatic cell hybrids derived from a female with severe manifestations of HED and a n X-autosome translocation [Zonana et al., 1988133. Details of the probes which were informative and utilized in this family are listed in Table I.
Linkage Analysis Linkage analyses were performed using the genetic distances between HED and the marker loci as deter-
Both the consultand and her mother were heterozygous at several marker loci (Table I). The morphological analyses of the fetal skin samples obtained from the consultand’s previous pregnancy indicated that the fetus was affected, and therefore she was a n obligate carrier. In addition, we also calculated the prior probability of the consultand being a carrier for HED based on a n analysis at the DXS159 locus, a proximal marker tightly linked to the HED locus [Zonana et al., 1988a; Hanauer et al., 19881.Her probability of being a carrier utilizing a genetic distance between DXS159 and the HED locus of 1cm (9 max.) was 98.9%; a t a distance of 7 cm (upper 95% CI) it was 92.6%. The results of the above linkage analysis, combined with the prior physical and histological findings in both the consultand and the male fetus from her third pregnancy, indicated that she was indeed a carrier. The consultand was informative at 2 loci (PGK1 and DXS72) which are tightly linked to HED, but both loci appear to be located distal to the disease locus. Phase was unknown, since her mother was also heterozygous at these loci, and the consultand’s father was deceased. The initial step was to calculate the probability that the abnormal HED gene was coupled in the consultand (11-2) with either allele 1 or 2 a t each of these loci (Table 11). The probability of phase was established in her mother (I-Z), based on typing of the consultand‘s 2 affected brothers (11-3, 11-4). Using both methods 1 and 2, there was a greater than 99% probability that the abnormal gene was coupled with the 2 allele for both markers in the consultand’s mother. The probability of phase in the consultand was subsequently calculated, and the gene frequency of the alleles in the general population was utilized to determine the paternal contribution. There was a 99% probability that the abnormal HED gene was coupled with allele 2 at the PGKl locus in the consultand using method 1, and a 96.6% with method 2. Table I1 lists the results for the DXS72 locus.
TABLE I. Results of Informative Marker Loci
I~OCUS
DXS159
Distance from HED (in cM) 0 max Upper C.I. .01 .07
cpX289
Restriction enzyme PstI
Probe
PGKl
.02
.07
pSPTiPGK
BdlI
DXS72
.02
.08
pX65H7
Hind111
Allele size (kb) 1 (5.5) 2 (1.6) l(12.0) 2 (5.0) 1 (7.2) 2 (0.7)
1-2 12
Alleles of individuals 11-2 11-3 11-4 2 2 2
111-4
1,2
1,2
2
2
2
1,2
1,2
2
2
2
-
134
Zonana et al.
TABLE 11. Probability of Coupling Between HED and Marker Alleles in Relatives
Method 1
Method 2
Allele
(6 m a . )
(6 of upper C.I.)
Individual 1-2 PGKl DXS 72
1 2 1 2
,006 ,994
Individual 11-2 PGKl DXS 72
1
,000 .999 ,000 .999 .009 ,991 .017 .983
.994 ,034 .966 .073 .926
.029 ,971 .037 .963
.I00 .goo .I42 ,858
2 1 2 Probability of an affected fetus PGKl 1 2 DXS 72 1 2
,006
The chromosome constitution of the fetus was 46,XY, and allele 2 was present a t both the PGKl and DXS72 loci. The probability of the fetus’s being affected was the probability of the disease allele’s being coupled with marker allele 2 in the consultand, multiplied by (1- 01, plus the probability of its being coupled with allele 1in the consultand multiplied by (0). Thus, the fetus had a 97% probability of being affected using method 1with a n analysis a t the PGKl locus, and a probability of 96% employing the DXS72 locus. With method 2, the probabilities were 90% and 86%, respectively (Table 11). The couple, after extensive counseling, decided to terminate the pregnancy, which was done a t 11 weeks gestation by prostaglandin induction of labor to allow a complete morphological examination of the abortus. Fetal skin was examined by both light and electron microscopy, and samples were obtained from the scalp, the axilla, and the plantar surface. The epithelial development corresponded to a fetal age of 11weeks with lack of skin appendages. Fetal skin of the scalp and axilla was covered by a 3 layer epithelium, consisting of basal, intermediate, and periderm cells. Hemidesmosomes at the dermo-epidermal junction were not yet developed. No adnexal primordia were demonstrable, without hair germs in the scalp or axilla. The plantar epithelium was clearly three-layered and lacked the beginning differentiation of primary dermal ridges and sweat gland primordia. Hemidesmosomes and anchoring fibrils were present in the plantar skin in small amounts. The samples were not clearly different from age-matched controls, and no ultrastructural abnormalities were seen. No tooth germs were identified at autopsy.
DISCUSSION Prior t o the availability of the new molecular genetic technology, prenatal diagnosis of single gene disorders of abnormal morphogenesis had been possible only by the direct visualization of the involved structures. Direct diagnosis of HED has been accomplished by demonstration of either complete lack of, or reduction in, the number of pilosebaceous follicles and by the lack of sweat gland primordia in multiple skin biopsies obtained by fetoscopy a t around 20-21 weeks gestation [Anton-Lamprecht et al., 1982; Arnold et al., 1984;
Blanchet-Bardon and Nazzaro, 19871. The interpretation of the biopsies can be difficult if one does not appreciate the normal regional variability of the distribution of skin appendages of fetal skin, and that sweat gland primordia only begin to develop a t around 20 weeks gestation. An indirect approach to the prenatal diagnosis of structural defects by linkage analysis utilizing DNA polymorphisms has become increasingly available, as more single gene disorders of morphogenesis are mapped on the human genome. Such is now the case with HED. This new method of prenatal diagnosis has some major advantages, as well as disadvantages, which should be fully explained to the families before the procedure is undertaken. It permits a diagnosis to be made by CVS in the first trimester of pregnancy prior to the development of the affected structures, thereby allowing a n earlier termination of a n affected pregnancy. CVS is technically simpler and may present a lower risk to the pregnancy than does fetoscopy and multiple skin biopsies. Disadvantages to a linkage based indirect analysis include the need for the sampling of previously affected individuals, and frequently of additional relatives, and a n at-risk consultand who is informative at the appropriate marker loci. The counseling of families is more complex since one is dealing with the probabilities of an affected fetus, rather than a more definitive diagnosis based on direct observation. One can present the at-risk consultand with both a maximum likelihood of having a n affected fetus: and with probabilities based on the approximate 95% confidence interval. However, these statistical concepts are difficult for many families to comprehend fully. The use of close informative markers in the linkage analysis which flank the disease locus, not available in this case, can significantly reduce the uncertainty of the counseling situation. Clinicians must also be sure they are dealing with the X-linked form of HED, and be aware of the possibility of still undetected genetic heterogeneity for the disorder before undertaking a linkage based approach t o the prenatal diagnosis of the disorder, even though prior analysis of 36 affected families has shown no evidence of significant genetic heterogeneity in the X-linked form [Anton-Lamprecht et al., 1988, Zonana et al., 1988al. Finally, our present inability to confirm the diagnosis of HED by histological examination of the first trimester abortus is unsatisfactory from both a scientific and counseling perspective. The absence of epidermal appendages in the present abortus was not thought to be diagnostic of HED, as the findings were similar to the sequence of normal skin development in age-matched controls LHolbrook, 1979,19881.The inability to confirm a linkage based diagnosis in first trimester fetuses with disorders of morphogenesis which do not have apparent morphological changes until later gestational ages, will continue to be problematic until the primary genetic defects in these disorders are delineated and can be assayed directly.
ACKNOWLEDGMENTS We would like to thank Drs. P. Pearson, A. Riggs, and B. Schmeckpeper for use of the DNA probes.
Prenatal Diagnosis of X-Linked HED
REFERENCES Anton-Lamprecht I, Arnold ML, Rauskolb R, Schinzel A, Schmid W, Schnyder UW (1982): Prenatal diagnosis of anhidrotic ectodermal dysplasia. Hum Genet 62:180. Anton-Lamprecht I, Schleiermacher E, Wolf M 11988):Autosomal recessive anhidrotic ectodermal dyplasia: Report of a case and discrimination of diagnostic features. In Salinas CF, Opitz JM, Paul NW (eds):“Recent Advances in Ectodermal Dysplasias.”New York: Alan R. Liss, Inc., for the National Foundation-March of Dimes. BD:OAS 24(2):183-195. Arnold ML, Rauskolb R, Anton-Lamprecht I, Schinzel A, Schmid W 11984):Prenatal diagnosis of anhidrotic ectodermal dysplasia. Prenat IXagn 4:85--98. Blanchet-Bardon C, Nazzaro V (1987):Use of morphological markers in carriers as an aid in genetic counselling and prenatal diagnosis. Curr Probl Dermatol 16:109-119. Clarke A, Phillips DIM, Brown R, Harper PS (1987):Clinical aspects of X-linked hypohidrotic ectodermal dysplasia. Arch Dis Child 62:989-996. Hanauer A, Alembik Y, Arveiler B, Formiga L, Gilgenkrantz S, Mandel J L (1988): Genetic mapping of anhidrotic ectodermal dysplasia: DXS159, a closely linked proximal marker. Hum Genet 80:177180.
135
Holbrook K (1979):Human epidermal embryogenesis. Int J Dermatol 18:329-356. Holbrook K (1988): Structural abnormalities of the epidermally derived appendages in skin from patients with ectodermal dysplasia: Insight into developmental errors. In Salinas CF, Opitz JM, Paul NW (eds):“Recent Advances in Ectodermal Dysplasias.” New York: Alan R. Liss, Inc., for the National Foundation-March of Dimes. BD:OAS 2412):15-44. Upadhyaya M, Archer IM, Harper PS, Jasani B, Roberts A, Shaw DJ, Thomas NS, Williams H (1984): DNA and enzyme studies on chorionic villi for use in antenatal diagnosis. Clin Chim Acta 140:39-46. Zonana J, Clarke A, Sarfarazi M, Thomas NST, Roberts K, Marymee K, Harper PS (1988a): X-linked hypohidrotic ectodermal dysplasia: Localization within the region Xqll-21.1 by linkage analysis and implications for carrier detection and prenatal diagnosis. Am J Hum Genet 43:75-85. Zonana J, Roberts SH, Thomas NST, Harper PS (1988b): Recognition and reanalysis of a cell line from a manifesting female with X linked hypohidrotic ectodermal dysplasia and an X;autosome balanced translocation. J Med Genet 25383-386. Zonana J , Sarfarazi M, Thomas NST, Clarke A, Marymee K, Harper PS (1989): Improved definition of carrier status in X-linked hypohidrotic ectodermal dysplasia by use of restriction fragment length polymorphism-based linkage analysis. J Pediatr 114:392-399.
Prenatal Diagnosis of X-Linked Hypohidrotic Ectodermal Dysplasia by Linkage Analysis Jonathan Zonana, Albert Schinzel, Meena Upadhyaya, Nicholas S.T. Thomas, Ingrun Anton-Lamprecht, and Peter S. Harper Institute of Medical Genetics, University of Wales College of Medicine, Cardiff, Wales (J.Z., M.U., N.S.T.T., P.S.H.); Department of Medical Genetics, and Child Development and Rehabilitation Center, Oregon Health Sciences University, Portland (J.Z.); Institut fur Medizinische Genetik, University of Zurich, Switzerland (A.S.), Institut fur Ultrastrukturforschung der Haut, Hautktinik der Ruprecht-Karls-Universittit,Heidelberg, Federal Republic of Germany (I.A.-L.)
Prenatal diagnosis of X-linked hypohidrotic ectodermal dysplasia was previously performed by the direct histological analysis of fetal skin obtained by late second trimester fetoscopy. The recent gene mapping of the locus for the disorder to the region of Xqll-21.1 now permits the indirect prenatal diagnosis of the disorder by the method of linkage analysis, based on closely linked marker loci, during the first trimester of pregnancy. We report the prenatal diagnosis of a male fetus with a high probability of the disorder by a linkage analysis utilizing restriction fragment length polymorphisms at the DXS159, PGK1, and DXS72 loci, from a DNA sample obtained by a chorionic villus biopsy at 9 weeks gestation. After further counseling, the pregnancy was terminated but the diagnosis could not be confirmed by histological analysis, even though analysis of skin samples by light and electron microscopy showed lack of hair germs, primary dermal ridges, and sweat gland primordia, due to the early developmental stage of the fetus. The use of DNA-based linkage analysis now offers the opportunity for an earlier diagnosis of X-linked hypohidrotic ectodermal dysplasia by a method other than fetal skin sampling. However, families must also fully understand the present limitations of the method prior to undertaking the procedure.
KEY WORDS: DNA polymorphism, chorionic villus sampling, epidermal appendages
Received for publication May 26,1989; revision received July 28, 1989. Address reprint requests to Jonathan Zonana, M.D., Dept. of Medical Genetics L-103, Oregon Health Sciences University, 3181 SW Sam Jackson Park Rd., Portland, OR 97201-3098.
0 1990 Wiley-Liss, Inc.
INTRODUCTION X-linked hypohidrotic (anhidrotic) ectodermal dysplasia (HED),a developmental disorder affecting structures of ectodermal origin, including eccrine sweat glands, teeth, and hair, is associated with early morbidity and mortality [Clarke et al., 19871. Affected males may have problems of psychosocial adjustment due to their appearance, and restricted activities secondary t o heat intolerance. Despite recent improvements in the early diagnosis and treatment of the condition, a number of obligate or at-risk carriers have inquired about the availability of prenatal diagnosis for the disorder. The gene locus for X-linked HED was recently localized by linkage analysis, using restriction fragment length polymorphisms (RFLP), t o chromosome region Xqll-21.1 [Zonana et al., 1988al. The use of RFLP markers has resulted in improved carrier detection, and has opened the possibility offirst trimester prenatal and neonatal diagnosis LZonana et al., 19891. Here we report the results of a prenatal diagnosis of HED by the use of linkage analysis. MATERIALS AND METHODS Clinical Report The consultand (11-Zj, a 34-year-oldgravida 4, para 0 woman, who had 3 therapeutic abortions, was a relative of one of the patients investigated previously as part of a gene mapping study ofthe HED locus (Fig. 1) [Zonana et al., 1988al. She had 2 brothers (11-3,11-4)with classical manifestations of HED, and she and her mother (1-2) were classified as manifesting carriers on the basis of patchy vellus hair on the trunk and lower limbs, and upon the lack of epidermal appendages on histological examination of their skin [Arnold et al., 19841. She had terminated her first 2 pregnancies (III-1&2)because of the 50%risk of HED, after amniocentesis demonstrated male fetuses. Her third pregnancy, another male fetus (111-3), was terminated subsequent to the analysis of multiple skin biopsies obtained by fetoscopy at 20 weeks gestation, Histological examination showed a lack of
Prenatal Diagnosis of X-Linked HED
I
mined from our previous study [Zonana et al., 1988aJ (Table I). Risk calculations were performed twice using different values for the genetic distances; in method 1 the maximum estimate of the recombination fraction was used, and in method 2 the recombination fraction which represents the upper value of the approximate 95% confidence interval. The former method would represent the most likely risk figure, but the latter method would be a more conservative approach, “erring” on the side of the fetus’s being unaffected.
II
111
133
1
2
3
RESULTS
Fig. 1. Pedigree of family with hypohidrotic ectodermal dysplasia with consultand individual 11-2 and her at-risk fetus 111-4.0, obligate carrier; 0, affected male; 0 , terminations involving male fetus; 3, affected male abortus.
skin appendages a t multiple biopsy sites, and these findings were subsequently confirmed by an analysis of the abortus [Anton-Lamprecht et al., 1982; Arnold et al., 19841. Her father (1-1)was deceased, and there were no other affected males or known female carriers in the family. Cytogenetic and DNA Analyses A chorionic villus sample (CVS) was obtained from the current pregnancy at 9 weeks of gestation, and the sample was split for cytogenetic examination and for DNA analysis. DNA extraction from blood and CVS samples, digestion with restriction enzymes, Southern blot analyses, and probe hybridizations were performed by previously described methods [Zonana et al., 1988a; Upadhyaya et al., 19841. Relevant relatives were typed with probes from 7 marker loci known to be linked to the HED locus, which spanned the region Xpl.1 to Xq22. The source of the probes, restriction enzymes, allele sizes, and the order of the loci have been described previously [Zonana et al., 1988al. The gene order of the closest markers and the disease locus (DXS159-HEDPGK1-DXS72) was based on linkage analysis and on the analysis of somatic cell hybrids derived from a female with severe manifestations of HED and a n X-autosome translocation [Zonana et al., 1988133. Details of the probes which were informative and utilized in this family are listed in Table I.
Linkage Analysis Linkage analyses were performed using the genetic distances between HED and the marker loci as deter-
Both the consultand and her mother were heterozygous at several marker loci (Table I). The morphological analyses of the fetal skin samples obtained from the consultand’s previous pregnancy indicated that the fetus was affected, and therefore she was a n obligate carrier. In addition, we also calculated the prior probability of the consultand being a carrier for HED based on a n analysis at the DXS159 locus, a proximal marker tightly linked to the HED locus [Zonana et al., 1988a; Hanauer et al., 19881.Her probability of being a carrier utilizing a genetic distance between DXS159 and the HED locus of 1cm (9 max.) was 98.9%; a t a distance of 7 cm (upper 95% CI) it was 92.6%. The results of the above linkage analysis, combined with the prior physical and histological findings in both the consultand and the male fetus from her third pregnancy, indicated that she was indeed a carrier. The consultand was informative at 2 loci (PGK1 and DXS72) which are tightly linked to HED, but both loci appear to be located distal to the disease locus. Phase was unknown, since her mother was also heterozygous at these loci, and the consultand’s father was deceased. The initial step was to calculate the probability that the abnormal HED gene was coupled in the consultand (11-2) with either allele 1 or 2 a t each of these loci (Table 11). The probability of phase was established in her mother (I-Z), based on typing of the consultand‘s 2 affected brothers (11-3, 11-4). Using both methods 1 and 2, there was a greater than 99% probability that the abnormal gene was coupled with the 2 allele for both markers in the consultand’s mother. The probability of phase in the consultand was subsequently calculated, and the gene frequency of the alleles in the general population was utilized to determine the paternal contribution. There was a 99% probability that the abnormal HED gene was coupled with allele 2 at the PGKl locus in the consultand using method 1, and a 96.6% with method 2. Table I1 lists the results for the DXS72 locus.
TABLE I. Results of Informative Marker Loci
I~OCUS
DXS159
Distance from HED (in cM) 0 max Upper C.I. .01 .07
cpX289
Restriction enzyme PstI
Probe
PGKl
.02
.07
pSPTiPGK
BdlI
DXS72
.02
.08
pX65H7
Hind111
Allele size (kb) 1 (5.5) 2 (1.6) l(12.0) 2 (5.0) 1 (7.2) 2 (0.7)
1-2 12
Alleles of individuals 11-2 11-3 11-4 2 2 2
111-4
1,2
1,2
2
2
2
1,2
1,2
2
2
2
-
134
Zonana et al.
TABLE 11. Probability of Coupling Between HED and Marker Alleles in Relatives
Method 1
Method 2
Allele
(6 m a . )
(6 of upper C.I.)
Individual 1-2 PGKl DXS 72
1 2 1 2
,006 ,994
Individual 11-2 PGKl DXS 72
1
,000 .999 ,000 .999 .009 ,991 .017 .983
.994 ,034 .966 .073 .926
.029 ,971 .037 .963
.I00 .goo .I42 ,858
2 1 2 Probability of an affected fetus PGKl 1 2 DXS 72 1 2
,006
The chromosome constitution of the fetus was 46,XY, and allele 2 was present a t both the PGKl and DXS72 loci. The probability of the fetus’s being affected was the probability of the disease allele’s being coupled with marker allele 2 in the consultand, multiplied by (1- 01, plus the probability of its being coupled with allele 1in the consultand multiplied by (0). Thus, the fetus had a 97% probability of being affected using method 1with a n analysis a t the PGKl locus, and a probability of 96% employing the DXS72 locus. With method 2, the probabilities were 90% and 86%, respectively (Table 11). The couple, after extensive counseling, decided to terminate the pregnancy, which was done a t 11 weeks gestation by prostaglandin induction of labor to allow a complete morphological examination of the abortus. Fetal skin was examined by both light and electron microscopy, and samples were obtained from the scalp, the axilla, and the plantar surface. The epithelial development corresponded to a fetal age of 11weeks with lack of skin appendages. Fetal skin of the scalp and axilla was covered by a 3 layer epithelium, consisting of basal, intermediate, and periderm cells. Hemidesmosomes at the dermo-epidermal junction were not yet developed. No adnexal primordia were demonstrable, without hair germs in the scalp or axilla. The plantar epithelium was clearly three-layered and lacked the beginning differentiation of primary dermal ridges and sweat gland primordia. Hemidesmosomes and anchoring fibrils were present in the plantar skin in small amounts. The samples were not clearly different from age-matched controls, and no ultrastructural abnormalities were seen. No tooth germs were identified at autopsy.
DISCUSSION Prior t o the availability of the new molecular genetic technology, prenatal diagnosis of single gene disorders of abnormal morphogenesis had been possible only by the direct visualization of the involved structures. Direct diagnosis of HED has been accomplished by demonstration of either complete lack of, or reduction in, the number of pilosebaceous follicles and by the lack of sweat gland primordia in multiple skin biopsies obtained by fetoscopy a t around 20-21 weeks gestation [Anton-Lamprecht et al., 1982; Arnold et al., 1984;
Blanchet-Bardon and Nazzaro, 19871. The interpretation of the biopsies can be difficult if one does not appreciate the normal regional variability of the distribution of skin appendages of fetal skin, and that sweat gland primordia only begin to develop a t around 20 weeks gestation. An indirect approach to the prenatal diagnosis of structural defects by linkage analysis utilizing DNA polymorphisms has become increasingly available, as more single gene disorders of morphogenesis are mapped on the human genome. Such is now the case with HED. This new method of prenatal diagnosis has some major advantages, as well as disadvantages, which should be fully explained to the families before the procedure is undertaken. It permits a diagnosis to be made by CVS in the first trimester of pregnancy prior to the development of the affected structures, thereby allowing a n earlier termination of a n affected pregnancy. CVS is technically simpler and may present a lower risk to the pregnancy than does fetoscopy and multiple skin biopsies. Disadvantages to a linkage based indirect analysis include the need for the sampling of previously affected individuals, and frequently of additional relatives, and a n at-risk consultand who is informative at the appropriate marker loci. The counseling of families is more complex since one is dealing with the probabilities of an affected fetus, rather than a more definitive diagnosis based on direct observation. One can present the at-risk consultand with both a maximum likelihood of having a n affected fetus: and with probabilities based on the approximate 95% confidence interval. However, these statistical concepts are difficult for many families to comprehend fully. The use of close informative markers in the linkage analysis which flank the disease locus, not available in this case, can significantly reduce the uncertainty of the counseling situation. Clinicians must also be sure they are dealing with the X-linked form of HED, and be aware of the possibility of still undetected genetic heterogeneity for the disorder before undertaking a linkage based approach t o the prenatal diagnosis of the disorder, even though prior analysis of 36 affected families has shown no evidence of significant genetic heterogeneity in the X-linked form [Anton-Lamprecht et al., 1988, Zonana et al., 1988al. Finally, our present inability to confirm the diagnosis of HED by histological examination of the first trimester abortus is unsatisfactory from both a scientific and counseling perspective. The absence of epidermal appendages in the present abortus was not thought to be diagnostic of HED, as the findings were similar to the sequence of normal skin development in age-matched controls LHolbrook, 1979,19881.The inability to confirm a linkage based diagnosis in first trimester fetuses with disorders of morphogenesis which do not have apparent morphological changes until later gestational ages, will continue to be problematic until the primary genetic defects in these disorders are delineated and can be assayed directly.
ACKNOWLEDGMENTS We would like to thank Drs. P. Pearson, A. Riggs, and B. Schmeckpeper for use of the DNA probes.
Prenatal Diagnosis of X-Linked HED
REFERENCES Anton-Lamprecht I, Arnold ML, Rauskolb R, Schinzel A, Schmid W, Schnyder UW (1982): Prenatal diagnosis of anhidrotic ectodermal dysplasia. Hum Genet 62:180. Anton-Lamprecht I, Schleiermacher E, Wolf M 11988):Autosomal recessive anhidrotic ectodermal dyplasia: Report of a case and discrimination of diagnostic features. In Salinas CF, Opitz JM, Paul NW (eds):“Recent Advances in Ectodermal Dysplasias.”New York: Alan R. Liss, Inc., for the National Foundation-March of Dimes. BD:OAS 24(2):183-195. Arnold ML, Rauskolb R, Anton-Lamprecht I, Schinzel A, Schmid W 11984):Prenatal diagnosis of anhidrotic ectodermal dysplasia. Prenat IXagn 4:85--98. Blanchet-Bardon C, Nazzaro V (1987):Use of morphological markers in carriers as an aid in genetic counselling and prenatal diagnosis. Curr Probl Dermatol 16:109-119. Clarke A, Phillips DIM, Brown R, Harper PS (1987):Clinical aspects of X-linked hypohidrotic ectodermal dysplasia. Arch Dis Child 62:989-996. Hanauer A, Alembik Y, Arveiler B, Formiga L, Gilgenkrantz S, Mandel J L (1988): Genetic mapping of anhidrotic ectodermal dysplasia: DXS159, a closely linked proximal marker. Hum Genet 80:177180.
135
Holbrook K (1979):Human epidermal embryogenesis. Int J Dermatol 18:329-356. Holbrook K (1988): Structural abnormalities of the epidermally derived appendages in skin from patients with ectodermal dysplasia: Insight into developmental errors. In Salinas CF, Opitz JM, Paul NW (eds):“Recent Advances in Ectodermal Dysplasias.” New York: Alan R. Liss, Inc., for the National Foundation-March of Dimes. BD:OAS 2412):15-44. Upadhyaya M, Archer IM, Harper PS, Jasani B, Roberts A, Shaw DJ, Thomas NS, Williams H (1984): DNA and enzyme studies on chorionic villi for use in antenatal diagnosis. Clin Chim Acta 140:39-46. Zonana J, Clarke A, Sarfarazi M, Thomas NST, Roberts K, Marymee K, Harper PS (1988a): X-linked hypohidrotic ectodermal dysplasia: Localization within the region Xqll-21.1 by linkage analysis and implications for carrier detection and prenatal diagnosis. Am J Hum Genet 43:75-85. Zonana J, Roberts SH, Thomas NST, Harper PS (1988b): Recognition and reanalysis of a cell line from a manifesting female with X linked hypohidrotic ectodermal dysplasia and an X;autosome balanced translocation. J Med Genet 25383-386. Zonana J , Sarfarazi M, Thomas NST, Clarke A, Marymee K, Harper PS (1989): Improved definition of carrier status in X-linked hypohidrotic ectodermal dysplasia by use of restriction fragment length polymorphism-based linkage analysis. J Pediatr 114:392-399.
