v)
.-c0 c)
0
4
824
glioma.’ Pediatric Neurology, 3, 29-32. 22. Tarbell, N. J . (1990) ‘Radiation therapy in the management of patients with optic chiasm gliomas.’ In Optic Gliomas in Neurofibromatosis. New York: National Neurofibromatosis Foundation. pp. 149-158. 23. Easle, J. D., Scharf, L., Chou, J . L., Riccardi, V. M. (1988) ‘Controversy in the management of optic pathway gliomas-29 patients treated with radiation therapy at Baylor College of Medicine from 1967 through 1987.’ Neurofibromatosis, 1, 248-251.
24. Edwards, M. S. B. (1990) ‘Management and treatment of chiasmatic gliomas.’ In Optic Cliomas in Neurofibromatosis. New York: National Neurofibromatosis Foundation. pp. 123-126. 25. Packer, R. J . , Savino, P . J., Bilaniuk, L. T., Zimmerman, R. A , , Schatz, N. J., Rosenstock, J. G.,Nelson, D. S., Jarrett, P. D., Bruce, D. A , , Schut, L. (1983) ‘Chiasmastic gliomas of childhood: a reappraisal of natural history and effectiveness of cranial irradiation.’ Child’s Brain, 10, 393-403.
Old Syndromes and New Cytogenet i cs
DiGeorge, Alagille. Two are duplications: Beckwith-Wiedemann and PallisterKillian. Not every individual diagnosed clinically with one of the above has a microscopically visible chromosome abnormality. Molecular studies have detected deletions or duplications when cytogenetically none was apparent. This is not surprising, since to be microscopically visible an alteration must comprise at least 2 to 5 per cent of a chromosome’s length, the equivalent of approximately two million base pairs or 40 to 50 genes. For each of the above syndromes we will discuss the chromosomal relationship to the clinical phenotype and the r61e of molecular cytogenetics in elucidating etiology.
THEnew cytogenetics-the application of molecular genetics to high-resolution karyotyping-has changed the face of syndromology. For 20 years, from the time that the karyotypic abnormalities in Down, Turner and Klinefelter syndromes were identified to the advent of the new cytogenetics, no other previously described clinical syndromes were recognized as chromosomal in origin. Now, with these new, powerful tools, we can recognize the r81e of very small deletions or duplications in syndromes of uncertain etiology. The cytogenetic abnormalities are detectable, for the most part, with high-resolution banding of synchronized lymphocyte cultures capable of resolving up to 900 bands per haploid complement. Molecular analysis has confirmed and defined the chromosomal origins using gene dosage measurements, in situ DNA hybridization and restriction fragmentlength polymorphisms, while also providing some unexpected observations. Together, these methods have demonstrated that the phenotypes of many multisystem disorders are due t o the involvement of genes related t o each other by their physical proximity on a chromosome, rather than by their functionthe ‘contiguous gene syndromes’ of SCHMICKEL’.The majority are microdeletions: Miller-Dieker, Langer-Giedion, AGR triad, Prader-Willi, Angelman,
Miller-Dieker syndrome Miller-Dieker syndrome is characterized by type I lissencephaly (smooth brain) and distinctive facies, including bitemporal hollowing with prominent forehead, often with vertical furrows, short nose with upturned nares, protuberant upper lip with thin vermillion border and small jaw. Apneic or tonic seizures usually begin between three and six months and life expectancy is about two years. Deletions, including the critical region of 17~13.3, have been observed cytogenetically in some patients, while in others with normal chromosomes, molecular studies have established the existence of submicroscopic deletions. Where parental origin of de novo deletions could be determined by DNA analysis of families, using chromosome 17 specific polymorphic probes, five (of six) were in the paternally derived deleted chromosome2, Based on pedigree analysis of segre-
’.
gation in the few familial cases, MillerDieker syndrome had previously been considered to be inherited as an autosoma1 recessive disorder. It is now clear that familial occurrence was due to a balanced translocation or inversion in a carrier parent, with the affected offspring inheriting a deleted 17 as a result of meiotic segregation.
Langer-Giedion syndrome This syndrome combines the earlier described trichorhinophalangeal (TRP) syndrome of Giedion with additional anomalies described by Langer. TRP, inherited as an autosomal dominant disorder, consists of hair anomalies (sparse scalp but bushy eyebrows), pearshaped nose and brachyphalangy, with cone-shaped epiphyses and Perthes-like changes. Langer described an entity incorporating TRP with mental retardation, microcephaly, multiple exostoses and redundant skin. Approximately half of the reported individuals with the syndrome have an interstitial deletion on the long arm of chromosome 8, the critical segment being 8q244. An elegant technique, the microdissection of banded chromosomes from normal metaphases followed by universal enzymatic amplification of the dissected DNA, identified sequences that were used to probe the involved region of 8q. Of the two patients studied with this technique, one had a deletion of band q23 and subband q24.11, while the other was missing the distal half of band q23 and the proximal half of sub-band q24.1’. The individual genes within the region have not yet been identified, although TRP with mental retardation has occurred in association with a deletion at 8q23. As in other contiguous gene syndromes, the degree of mental retardation is variable and individuals have been described with apparent 8q23 deletions but lacking the Langer-Giedion phenotype.
Wilms tumor (WAGR). Nearly 25 years after the association between aniridia and Wilms tumor was first reported, the deletion of llp13 was identified as the etiological factor6. The size of the deletion may correlate with the degree of mental retardation: a few of the reported cases have normal or borderline normal intelligence and very small deletions. The genito-urinary abnormalities in males vary also, from normal development to male pseudohermaphroditism. Tumor development is apparently independent of deletion size, since half-brothers with identical deletions were discordant for tumor’. The short arm of chromosome 11 has been subjected to intensive analysis, using probes for Mendelian genes and anonymous DNA segments assigned to 1lp13 in the search for the gene(s) involved in Wilms tumor. These studies have excluded the locus for familial Wilms tumor from chromosome 11, but have provided strong evidence for a discrete aniridia gene on llp13, based on cytogenetic studies of cotransmission with translocations involving 1lp13* and linkage analysis of families with aniridia and normal karyotypes’. Previously a locus for isolated, dominantly inherited aniridia (designated A N I ) was linked to the enzyme acid phosphatase-1 locus on the short arm of chromosome 2, based on one large pedigree exhibiting variable expression, lack of visual impairment and n o male-to-male transmission”. Reanalysis of that family using llp13 DNA markers should determine whether there are two (or more) different discrete genes for autosomal dominant aniridia. Like Miller-Dieker syndrome, paternal inheritance of the de novo deletion is the rule. Of eight AGR patients studied, most with Wilms tumor, seven inherited a paternally derived deleted chromosome 11. The eighth, with a maternally derived deletion, did not differ phenotypically from the others”
AGR triad This is the only microdeletion syndrome not to have an eponym. Its acronym derives from the key features: aniridia, genito-urinary malformations and mental retardation (AGR). One-third t o one-half of children with the deletion also develop
Prader- Willi syndrome Prader-Willi syndrome (PWS), first described in 1956, consists of the clinical findings of obesity, hypotonic musculature, mental retardation, hypogonadism, short stature and small hands and feet. Feeding problems in infancy (poor suck
-
m
3
hl W
hl-
m
2
6
2 2
s
2 8 %
B .-. QJ
4
9 c
P
3
8
2
2
4
825
.-2
.-
d u
0
4
826
and failure to thrive) are succeeded by hyperphagia and obesity, usually in the second year and continuing into adulthood. Since the first report of an apparently balanced D:D translocation in PWS in 1976, numerous structural rearrangements have been described, including unbalanced translocations, additional small markers and microdeletions, often with mosaicism. Most involved the pericentromeric region of the long arm of chromosome 15. In 1981, LEDBETTER et al. 12, using high-resolution banding, detected deletions in about half of 40 patients studied, and suggested that a minute deletion at 15qll-12 might be the cause of the syndrome, an observation quickly confirmed by others. PWS was the first syndrome for which exclusively paternal origin for the de novo deleted 15 was found, based on cytogenetic polymorphisms of 1 5 ~ ’ Molecular ~. markers confirmed the observation. At the time, this surprising inheritance pattern was attributed to the greater vulnerability of male gametogenesis to environmental insults occurring through an effect on DNA replication. Molecular studies using DNA probes mapping to 15ql1 showed DNA sequences t o be deleted in some, but not all, PWS patients. Various theories were advanced to explain how deletions, translocations, duplications, or even apparently normal chromosomes, could produce the same phenotype. Molecular analysis of two PWS patients lacking a visible deletion ended the speculations. One carried a familial Robertsonian 13: 15 translocation; the other had two apparently normal chromosomes 15. Cloned DNA markers for the 15qll-13 subregion showed, surprisingly, that both probands lacked any contribution of paternal alleles at 15qll-13, having inherited only maternal chromosomes 15. The transmission of the two normal homologues of chromosome 15 from their mothers-uniparental (maternal) heterodisomy-indicated that it was not the deletion of 15ql l itself which caused PWS, but rather the absence of a paternal contribution of 15ql 114.
Angelman syndrome The biological significance of band 15qll became even more intriguing with the
recognition of the identical deletion in approximately half of the studied patients with the Angelman (‘happy puppet’) syndrome (AS) of jerky movements, unprovoked laughter and mental retardation’j. To reconcile the differences in PWS and AS phenotype with identical chromosomal findings, it was first suggested that a larger deletion in Angelman syndrome not only caused the more severe mental retardation and lack of speech but also ‘suppressed’ or altered the presumed hypothalamic abnormality causing the uncontrolled appetite and obesity in PWS. This seemed unlikely because their additive effects have allowed deletions of increasing size to be used for gene mapping for half a century. The relationship between PWS and AS has now been elucidated at the molecular level. Using cloned segments of 15qll-13 as probes, the deletions in PWS and AS appear to be identical. But in contrast to the inheritance of the paternal deleted 15 in PWS, the AS patients studied each inherited a maternally deleted 1516. These differences in phenotypes, despite identical chromosome deletions, are attributed to imprinting, the differential expression of genetic information, depending on the parent from whom it is inherited. Imprinting is discussed in more detail in the concluding section and was recently reviewed by HALL”.
Alagille syndrome Alagille syndrome includes chronic cholestasis caused by intrahepatic biliary hypoplasia, pulmonary stenosis, vertebral arch defects and embryotoxon posterior. Although most cases are sporadic, pedigree studies have suggested autosomal dominant inheritance, with reduced penetrance and variable expressivity. The first report of deletion 20~11.2-12.1was confirmed in three additional patients, although other reports of 20p deletions with the typical facies, cardiac and vertebral defects, and mental retardation made no mention of cholestasis. In situ hybridization and densitometric scanning have confirmed hemizygosity for 20p probes in two probands”, 19. The deletion 20p- may parallel the l l p - deletion, i.e. not all individuals with the deletion manifest the associated developmental
defect, paucity of interlobular bile ducts or Wilms tumor, respectively, either because the deletions d o not include the critical locus or because other factors, as yet unidentified, interact with the deletion to produce the visceral abnormalities. Combined high-resolution and molecular analysis of a series of patients undoubtedly will be forthcoming, since this is the second most frequent form of infantile intrahepatic cholestasis.
DiGeorge syndrome The DiGeorge malformation sequence is an embryonic developmental defect involving the foregut in the region of the third and fourth pharyngeal pouches and the fourth branchial arch. It is manifested as absence or ectopia of part or all of the thymus, absent or hypoplastic parathyroids and great vessel malformations. Abnormalities involving chromosome 22 range from complete monosomy t o loss of 22ql1, resulting from unbalanced translocation (in approximately 20 per cent of cases studied), to interstitial deletions of 22ql 120. Unlike other contiguous gene syndromes, the DiGeorge sequence has also been reported in association with various chromosome abnormalities, deletion lop13 (or 1 0 ~ 1 4being ) the most common2’, and including 17p13, the same region deleted in Miller-Dieker syndrome. Although the DiGeorge sequence has not been reported in infants with MillerDieker, lissencephaly has been noted in a patient with DiGeorge (chromosomes unknown). The wide spectrum of chromosome anomalies may reflect the multiplicity of genes needed for the normal progression of neural crest cells into the branchial arches. Southern blotting and DNA dosage analysis of cell lines from four patients with DiGeorge syndrome and monosomy of 22pter-qll due t o unbalanced translocations confirmed the deletion of an anonymous marker (D22S9) mapping to 22q1122923.The locus was not deleted in three patients with normal chromosomes, nor in two with interstital deletions of 22qll. Detection of a different deleted locus, BCRL2, in one of the latter helps to define the critical region of 22q for this syndrome23.No molecular studies of other chromosomes associated with
DiGeorge syndrome have been reported as yet.
Beckwith- Wiedemann syndrome The main features of BeckwithWiedemann syndrome are exomphalos (omphalocele or umbilical hernia), macroglossia and visceral gigantism, with hypoglycemia in the newborn. Neoplasms, especially nephroblastoma and adrenal carcinoma, occur with increased frequency. Mental retardation has been observed in about 10 per cent of cases. A small proportion of reported invidividuals have had duplications of 1lp15, either occurring de novo or resulting from adjacent segregation of a parental translocation24. There are also familial cases with normal chromosomes. D N A analysis of affected individuals with normal chromosomes has not yet provided evidence of increased gene dosage for several loci at llp15, as would be expected if a duplication were present2’, while patients with constitutional chromosomal duplications have had gene duplications, as expected. Because of the 1 lp15 association, three karyotypically normal families segregating for Beckwith-Wiedemann syndrome were analyzed for DNA markers mapped to 1 1 ~In~all~three, . evidence was found for tight linkage to markers at llp15. In the largest pedigree it was assumed that four unaffected sisters, the children of unaffected parents, were carriers, since each had one or more affected offspring. Once the syndrome was expressed it was inherited in a dominant manner, with affected grandchildren in the pedigree. Noteworthy in this family was the unexplained paucity of transmitting males, either affected or presumed carriers with affected offspring. Yet in other cases in which the origin of the duplication could be determined, it was paternal in origin. The linkage study was uninformative regarding duplications in the 1lp15 region. Another curious discrepancy in the inheritance of Beckwith-Wiedemann syndrome is the fact that all monozygotic twin pairs have been female and discordant. Molecular anslysis of one twin pair using llp15 markers for linkage and dosage found no detectable differences”.
N”
m
b‘
82 7
2 .-* c
C
4
828
Pallister-Killian syndrome The Pallister-Killian syndrome is characterized by severe mental retardation, unusual facial appearance-coarse features, prominent forehead, ptosis, short anteverted nostrils-and abnormal scalp hair distribution (bitemporal alopecia). The associated chromosome abnormality is a tissue-limited mosaicism consisting of tetrasomy of the short arm of chromosome 12, manifested as a small marker chromosome, isochromosome 12p, found almost exclusively in fibroblasts but not peripheral lymphocytes28. When diagnosed in the newborn period, the i(12p) has been observed in 100 per cent of metaphases from direct bone-marrow analysis, but only in 23 per cent of phytohemagglutinin (PHA)-stimulated bonemarrow cells. Mosaicism for the isochromosome has been found in PHAstimulated peripheral lymphocytes in the newborn period, but the abnormal mitoses rapidly disappear from the circulation. The identification of the isochromosome has been confirmed by dosage analysis and in situ hybridization using a cDNA probe localized on 12p, as well as quantitative measurements of lactate dehydrogenase B and superoxide dismutase, enzymes mapped to 1 2 ~ No ~ studies ~ . of parental origin of the i(12p) have been reported. Summary This annotation has been confined t o well-established clinical syndromes with recently discovered chromosomal etiologies. It deliberately omits retinoblastoma, the oft-cited paradigm of a contiguous gene syndrome, since it is usually inherited as a Mendelian single gene disorder. However, it was recognition of both the deletion of band q14 of chromosome 13 in mentally retarded children with retinoblastoma, and the linkage of retinoblastoma with the genetic marker esterase D, which resulted in the eventual cloning of the gene. Also omitted are microdeletions of the x chromosome. These disorders are seen primarily in males, who manifest the phenotypic effects of the deletion of the loci of various combinations of adjacent genes: Duchenne muscular dystrophy, glycerol kinase deficiency, adrenal hypoplasia,
optic albinism, hypogonadatropic hypogonadism and anosmia (Kallman syndrome), chondrodysplasia punctata and icthyosis. Many are also mentally retarded. The third group omitted are Mendelian disorders occurring with atypical mental retardation (not usually part of the disorder), the presumption being that they include small deletions. It is expected that other contiguous gene syndromes will be recognized eventually; Rubinstein-Taybi and Cornelia de Lange syndromes are prime candidates. Why do deletions have such dramatic consequences when a normal homologue of the region is present? If their effects were due to the uncovering of recessive genes, we would expect to see greater variations in phenotype among carriers, including normal individuals whose deletions were masked by the protective effects of dominant alleles in the homologous regions. Imprinting-the ‘stamping’ of a gene as it passes through the germ line-provides a more satisfactory explanation. If the intact regions homologous to the deletions are imprinted, as suggested by the preferential parental inheritance documented for many of the syndromes, the intact chromosome need not carry recessive mutations. s.APIENZA30 suggests that imprinting results in a process of allelic inactivation, which affects the phenotype like a classical mutation. Imprinted genes would behave like recessives under these conditions, producing similar phenotypes, i.e. clinically recognizable syndromes. We do not know yet whether all of the syndromeassociated contiguous regions are imprintable, but where it is found, gamete-specific inheritance provides strong evidence. Much of the evidence for imprinting comes from the mouse. Complementation studies of disomic mice, produced by intercrossing heterozygotes for Robertsonian translocations, identified regions of chromosomes with functions determined by their parental origin3’. Some regions must be of paternal origin and others of maternal origin in order for development to proceed normally. Using mice with complementary duplication/deficient chromosomes from adjacent-1 disjunctions in both reciprocal translocation heterozygote parents, some chromosomal regions were
found to influence phenotypic characters such as shape or size of neonates, depending on parental origin. It has been suggested that methylation of genes provides the mechanism of imprinting. In transgenic mice, the methylation state of certain imprintable genes has been shown to be present in the germ line, demethylation occurring in the male, de novo methylation in the female. The offspring reproduce this pattern, resulting in phenotypes (for imprintable genes) which differ according to the parental origin. P w s and AS demonstrate that, in man as well as the mouse, not only do identical regions become functionally different, depending on the germ line through which they pass, but also that the diploid state must be maintained for normal development. Clearly, both maternal and paternal genes of the 15qll region are needed. We do not know yet how the different phenotypes result from the same deletion; how in the presence of its homologue the inheritance of a paternal 15qll ‘protects’ against PWS, whereas the maternal 15qll in similar circumstances prevents AS. Thus far the situation for chromosome 15 is unique. No other paired contiguous gene syndromes resulting from imprinting by the male and female parent, respectively, have been documented, although cases of similar deletions but differing phenotypes suggest the possibility. Uniparental disomy can result from a nondisjunction in either the first reductional (heterodisomy) or the second equational meiotic division (isodisomy). In the case of the exceptional PWS probands, each egg with two different chromosomes 15 was probably fertilized by a normal sperm, followed by loss of a paternal 15 from the zygote. Evidence for post-zygotic loss of trisomic chromosomes comes from the finding of placental mocaicism, limited to the cytotrophoblast, in viable gestations of trisomy 13 and 18, but not 213’. By analogy with PWS, we might expect the complementary situation of paternal disomy, the double contribution of paternal 15s and lack of maternal contribution, to occur in AS. It has yet to be described. The previous cases of uniparental disomy were in individuals with cystic fibrosis, who inherited two cystic
fibrosis alleles from their heterozygous carrier mothers. Contiguous gene syndromes raise intriguing questions. If imprinting in these disorders causes mental retardation in addition to the recessive phenotype, what proportion of cases of isolated mental retardation of unknown etiology have similar etiology? The breakpoints of balanced translocations have frequently provided clues to gene loci. In the absence of any gene markers, how do we search for microdeletions in mental retardation? It is intriguing to speculate also on the possible contribution of uniparental disomy to abnormal fetal development. The hydatidiform mole is a zygote containing two sets of paternally derived chromosomes. If lack of the maternal complement results in a molar pregnancy, what proportion of karyotypically normal spontaneous abortions are uniparental for an entire chromosome or imprinted region? How many imprintable loci are there in man? Cases of discordant parent/child phenotypes with identical translocations were attributed in the past to ‘aneusomie de recombination’. These should be reinvestigated to determine what proportion, if any, represents uniparental disomy. Although imprinting may best be studied in transgenic mice, we need answers regarding the human situation. Advances in our knowledge often ha;e begun with the clinician who recognized unusual manifestations of a gene. Even in the era of molecular genetics, the astute diagnostician remains an important element in the discovery process. HOPE H. PUNNETT, PhD ELAINEH . ZAKAI,M.D.* Department of Pediatrics, Temple University School of Medicine; Director, Cytogenetics Laboratory, St. Christopher’s Hospital for Children, 3601 A Street, Philadelphia, PA 19134-1095. *Department of Pediatrics, University of Pennsylvania School of Medicine; Division of Human Genetics and Molecular Genetics, The Children’s Hospital of Philadelphia, 34th and Civic Center Boulevard, Philadelphia, PA 19104. Acknowledgement This work was supported in part by National Institutes of Health Grant 1F33 CA086659 to H.H.P.
-
z m
r4 00
n i m 0”
5
5 e
2
8
2 is
2 P,
.s 2
-95
:
2 L
4
829
2 .-
c)
c
0 C
d
830
References 1. Schmickel, R. D. (1986) ‘Contiguous gene syndrome: a component of recognizable syndromes.’ Journal of Pediatrics, 109, 23 1-24], 2. van Tuinen, B., Dobyns, W. B., Rich, D. C., Summers, K. M., Robinson, T. J., Nakamura, Y., Ledbetter, D. H. (1988) ‘Molecular detection of microscopic and submicroscopic deletions associated with Miller-Dieker syndrome.’ American Journal of Human Genetics, 43, 587-596. 3. Schwartz, C . E., Johnson, J . P., Holycross, B., Mandeville, T. M., Sears, T. S., Graul, E. A., Carey, J. C., Schroer, R. J., Phelen, M. C., Szollar, J., Flannery, D. B., Stevenson, R. E. (1988) ‘Detection of submicroscopic deletions in band 1 7 ~ 1 3in ,patients with the MillerDieker syndrome. American Journal of Human Genetics, 43, 597-604. 4. Langer, L. O., Krassikoff, N., Laxova, R., Scheer-Williams, M., Lutter, L. D., Gorlin, R. J., Jennings, C. J., Day, D. W. (1984) ‘The tricho-rhino-phalangeal syndrome with exostoses (or Langer-Giedion syndrome): four additional patients without mental retardation and review of the literature.’ American Journal of Medical Genetics, 19, 81-1 11. 5. Ludecke, H.-J., Senger, G., Claussen, U., Horsthemke, B. (1989) ‘Cloning defined regions of the human genome by microdissection of banded chromosomes and enzymatic amplification.’ Nature, 338, 348-350. 6. Riccardi, V. M., Sujansky, E., Smith, A. C., Francke, U. (1978) ‘Chromosomal imbalance in the aniridia-Wilms tumor association: 1 l p interstitial deletion.’ Pediatrics, 61, 604-610. 7. Lavedan, C., Barichard, F., Azoulay, M., Couillin, P., Molina Gomez, D., Nicolas, H., Quack, B., Rethore, M.-0.. Noel, B., Junien, C. (1989) ‘Molecular definition of de novo and genetically transmitted WAGR-associated rearrangements of llp13.’ Cytogenetics and Cell Genetics, 50, 70-74.. 8. Simola, K. 0. J., Knuutila, S., Kaitila, I., Pirkola, A., Pohja, P . (1983) ‘Familial aniridia and translocation t(4: 11) (q22:p13) without Wilms’ tumor.’ Human Genetics, 63, 158-161. 9. Mannens, M., Bleeker-Wagemakers, E. M., Bliek, J., Hoovers, J., Mandjes, I., van Tol, S., Frants, R. R., Heyting, C., Westerveld, A., Slater, R. M. (1989) ‘Autosomal dominant aniridia linked to the chromosome l l p 1 3 markers catalase and D1 IS151 in a large Dutch family.’ Cytogenetics and Cell Genetics, 52, 32-36. 10. Ferrell, R. E., Chakravarti, A., Hittner, H . M., Riccardi, V. M. (1980) ‘Autosomal dominant aniridia: probable linkage to acid phosphatase-1 locus o n chromosome 2.’ Proceedings of the National Academy of Sciences, USA, 77, 1580-1582. 1 1 . Huff, V., Meadows, A., Riccardi, V. M., Strong, L. C., Saunders, G . F. (1990) ‘Parental origin of de novo constitutiona! deletions of chromosome band llp13. American Journal of Human Genetics, 47, 155-160. 12. Ledbetter, D. H., Riccardi, V. M., Airhart, S. D., Strobel, R. J., Keenan, B. S., Crawford, J. D. (1981) ‘Deletions of chromosome !5 as a cause of the Prader-Willi syndrome. New England Journal of Medicine, 304, 325-329. 13. Butler, M. G., Palmer, C. G. (1983) ‘Parental origin of chromosome 15 deletion in Prader-
Willi syndrome.’ Lancet, 1, 1285-1286. 14. Nicholls, R. D., Knoll, J. H. M., Butler, M. G . , Karam, S., Lalande, M. (1989) ‘Genetic imprinting suggested by maternal heterodisomy in non-deletion Prader-Willi syndrome.’ Nature, 342, 281-285. 15. Kaplan, L. C., Wharton, R., Elias, E., Mandell, F., Donlon, T., Latt, S. A. (1987) ‘Clinical heterogeneity associated with deletions in the long arm of chromosome 15: report of 3 new cases and their possible genetic significance.’ American Journal of Medical Genetics, 28, 45-53. 16. Knoll, J. H. M., Nicholls, R. D., Magenis, R. E., Graham, J. M., Jr., Lalande, M., Latt, S. A. (1989) ‘Angelman and Prader-Willi syndromes share a common chromosome 15 deletion but differ in Darental orinin of the deletion.’ American -Journal Medical Genetics, 32, 285-290. 17. Hall, J . G. (1990) ‘Genornic imprinting: review and relevance t o human diseases.’ American Journal of Human Genetics, 46, 857-873. 18. Byrne, J. L. B., Harrod, M. J. E., Friedman, J . M., Howard-Peebles, P . N. (1986) ‘del(20p) with manifestations of arteriohepatic dysplasia.’ American Journal of Medical Genetics, 24, 673-678. 19. Schnittger, S., Hofers, C., Heidemann, P., Beermann, F., Hansmann, I . (1989) ‘Molecular and cytogenetic analysis of an interstitial 20p deletion associated with syndromic intrahepatic ductular hypoplasia (Alagille syndrome).’ Human Genetics, 83, 239-244. 20. Zhang, F., Deleuze, J.-F., Aurias, A., Dutrillaux, A,-M., Hugan, R.-N., Alagille, D., Thomas, G., Hadchouel, M. (1990) ‘Interstitial deletion of the short arm of chromosome 20 in arteriohepatic dysplasia. Journal of Pediatrics, 116, 73-77. 21. de la Chapelle, A,, Herva, R., Koivisto, M., A d a , P. (1981) ‘A deletion in chroTosome 22 can cause DiGeorge syndrome. Human Genetics, 57, 253-256. 22. Carey, A. H., Roach, S., Williamson, R., Dumanski, J. P., Nordenskjold, M., Collins, V. P., Rouleau, G., Blin, N., Jalbert, P., Scambler, P . J. (1990) ‘Localization of 27 DNA markers t o the region of human chromosome 22qll-pter deleted in patients with DiGeorge syndrome and duplicated in the der22 syndrome.’ Genomics, 7, 299-306. 23. Fibison, W. J., Budarf, M., McDermid, H., Greenberg, F., Emanuel, B. S. (1990) ‘Molecular studies of DiGeorge syndrome.’ American Journal of Human Genetics, 46, 888-895. 24. Waziri, M., Patil, S. R., Hanson, J. W., Bartley, J. A. (1983) ‘Abnormality of chromosome 11 in patients with features of BeckwithWiedemann syndrome.’ Journal of Pediatrics, 102, 873-876. 25. Henry, I . , Jeanpierre, M., Couillin, P., Barichard, F., Serre, J.-L., Journel, H., Lamouroux, A., Turleau, C., De Grouchy, J., Junien, C . (1989) ‘Molecular definition of the llp15.5 region involved in BeckwithWiedemann syndrome and probably i; predisposition t o adrenocortical carcinoma. Human Genetics, 81, 273-277. 26. Schofield, P . N., Lindham, S., Engstrom, W. (1989) ‘Analysis of gene dosage on chromosome 11 in children suffering from BeckwithWiedemann syndrome.’ European Journal of Pediatrics, 148, 320-324. 27. Litz, C . E., Taylor, K. A., Qiu, J. S., Pescovitz,
07
0. H., de Martinville, B. (1988) ‘Absence of
detectable chromosomal and molecular abnormalities in monozygotic twins discordant for the Wiedemann-Beckwith syndrome.’ American Journal of Medical Genetics, 30, 821-833. 28. Hersh, J. H., Graham, J. M., Jr., Destrempes, B. S., Greenstein, R. M. (1983) ‘TeschlerNicola/Killian syndrome: a case report.’ Journal of Clinical Dysmorphology, 1, 20-24. 29. Peltomaki, P., Knuutila, S., Ritvanen, A,, Kaitila, I., de la Chapelle, A. (1987) ‘PallisterKillian syndrome: cytogenetic and molecular
studies.’ Clinical Genetics, 31, 399405. 30. Sapienza C. S. (1989) ‘Genome imprinting and dominance modification.’ Annals of the New York Academy of Sciences, 564, 24-37. 31. Cattanach, B. M., Kirk, M. (1985) ‘Differential activity of maternally and paternally derived chromosome regions in mice.’ Nature, 315, 496498. 32. Kalousek, D. K., Barrett, I. J., McGillivray, B. C. (1989) ‘Placentak mosaicism and intrauterine survival of trisomies 13 and 18.’ American Journal of Human Genetics, 44, 338-343.
-
m m
d
2 0-
m
2
NOTICES
Children-the Forgotten Mourners
10th Margaret Lowenfeld Day Conference
London, 17th September 1990
Cambridge, Saturday 10th November 1990
This symposium, with discussion groups, will be held at Queen Charlotte’s and Chelsea Hospital on Monday, 17th September 1990. Fee f45. Further information from Symposium Secretary, RPMS Institute of Obstetrics and Gynaecology, Queen Charlotte’s and Chelsea Hospital, Goldhawk Road, London W6 OXG. Tel.: 081-740 3904.
Neurofibromatosis Clinical Care Sympoisum Cincinnati, Ohio, October 1990 This anmal symposium, sponsored by The National Neurofibromatosis Foundation, will be held at the Cincinnati Convention Center, Cincinnati, on Sunday, 21st October 1990 from 8.30am to 12.30pm. Further information from The National Neurofibromatosis Foundation, Inc., 141 Fifth Avenue, Suite 7 3 , New York, NY 10010.
1990 American Epilepsy Society Annual Meeting San Diego, California, 10th to 15th November 1990 The four main symposia at this annual meeting, to be held at the Sheraton Harbor Island Hotel, San Diego, will be ‘Spread of seizure discharge’; ‘Controversies in epileptology’ (‘Is topographic mapping of EEG useful?’ and ‘Are current methodologies of anti-epileptic drug trials satisfactory?’); ‘Mesial temporal sclerosis: the neuropathology of epilepsy’ and ‘Seizures in the first year of life’. Further information from The American Epilepsy Society, 638 Prospect Avenue, Hartford, CT 06105-4240. Tel.: (203) 232 4825.
Preverbal communication was central to Margaret Lowenfeld’s thinking. To celebrate the centenary of her birth, this seminar will relate some of her ideas to current thinking, with particular reference to play, nursery schools and autism. There will be a demonstration of Lowenfeld techniques. Speakers include John Davis, Mollie Dundas, Dorothy Einon, Nina Farhi, Ian Goodyer, Elizabeth Newson, Robert Royeton, Alison Van Dyk, Margarita Wood and Beric Wright. This one-day conference will be of interest to child psychotherapists and psychiatrists, psychologists, social workers, health visitors and .others interested in the emotional health of children in its social context. Details from Child Care and Development Group, Free School Lane, Cambridge CB2 3RF.
Frontiers of Research in Developmental Neurology Groningen, The Netherlands, 2lst to 23rd March 1991 This international workshop is being organized by the Department of Developmental Neurology, University Hospital, Groningen. Topics will be ‘Results of brain imaging in relation to their functional neurological correlates in infants and children’; ‘Qualitative vs. quantitative neurological assessment of infants and children’; and ‘Instrumental measurements of complex normal and abnormal movements in children’. Further information from Professor H. F. R. Prechtl, Department of Developmental Neurology, University Hospital, Oostersingel59, 9713 EZ Groningen, The Netherlands.
831
.-c0 c)
0
4
824
glioma.’ Pediatric Neurology, 3, 29-32. 22. Tarbell, N. J . (1990) ‘Radiation therapy in the management of patients with optic chiasm gliomas.’ In Optic Gliomas in Neurofibromatosis. New York: National Neurofibromatosis Foundation. pp. 149-158. 23. Easle, J. D., Scharf, L., Chou, J . L., Riccardi, V. M. (1988) ‘Controversy in the management of optic pathway gliomas-29 patients treated with radiation therapy at Baylor College of Medicine from 1967 through 1987.’ Neurofibromatosis, 1, 248-251.
24. Edwards, M. S. B. (1990) ‘Management and treatment of chiasmatic gliomas.’ In Optic Cliomas in Neurofibromatosis. New York: National Neurofibromatosis Foundation. pp. 123-126. 25. Packer, R. J . , Savino, P . J., Bilaniuk, L. T., Zimmerman, R. A , , Schatz, N. J., Rosenstock, J. G.,Nelson, D. S., Jarrett, P. D., Bruce, D. A , , Schut, L. (1983) ‘Chiasmastic gliomas of childhood: a reappraisal of natural history and effectiveness of cranial irradiation.’ Child’s Brain, 10, 393-403.
Old Syndromes and New Cytogenet i cs
DiGeorge, Alagille. Two are duplications: Beckwith-Wiedemann and PallisterKillian. Not every individual diagnosed clinically with one of the above has a microscopically visible chromosome abnormality. Molecular studies have detected deletions or duplications when cytogenetically none was apparent. This is not surprising, since to be microscopically visible an alteration must comprise at least 2 to 5 per cent of a chromosome’s length, the equivalent of approximately two million base pairs or 40 to 50 genes. For each of the above syndromes we will discuss the chromosomal relationship to the clinical phenotype and the r61e of molecular cytogenetics in elucidating etiology.
THEnew cytogenetics-the application of molecular genetics to high-resolution karyotyping-has changed the face of syndromology. For 20 years, from the time that the karyotypic abnormalities in Down, Turner and Klinefelter syndromes were identified to the advent of the new cytogenetics, no other previously described clinical syndromes were recognized as chromosomal in origin. Now, with these new, powerful tools, we can recognize the r81e of very small deletions or duplications in syndromes of uncertain etiology. The cytogenetic abnormalities are detectable, for the most part, with high-resolution banding of synchronized lymphocyte cultures capable of resolving up to 900 bands per haploid complement. Molecular analysis has confirmed and defined the chromosomal origins using gene dosage measurements, in situ DNA hybridization and restriction fragmentlength polymorphisms, while also providing some unexpected observations. Together, these methods have demonstrated that the phenotypes of many multisystem disorders are due t o the involvement of genes related t o each other by their physical proximity on a chromosome, rather than by their functionthe ‘contiguous gene syndromes’ of SCHMICKEL’.The majority are microdeletions: Miller-Dieker, Langer-Giedion, AGR triad, Prader-Willi, Angelman,
Miller-Dieker syndrome Miller-Dieker syndrome is characterized by type I lissencephaly (smooth brain) and distinctive facies, including bitemporal hollowing with prominent forehead, often with vertical furrows, short nose with upturned nares, protuberant upper lip with thin vermillion border and small jaw. Apneic or tonic seizures usually begin between three and six months and life expectancy is about two years. Deletions, including the critical region of 17~13.3, have been observed cytogenetically in some patients, while in others with normal chromosomes, molecular studies have established the existence of submicroscopic deletions. Where parental origin of de novo deletions could be determined by DNA analysis of families, using chromosome 17 specific polymorphic probes, five (of six) were in the paternally derived deleted chromosome2, Based on pedigree analysis of segre-
’.
gation in the few familial cases, MillerDieker syndrome had previously been considered to be inherited as an autosoma1 recessive disorder. It is now clear that familial occurrence was due to a balanced translocation or inversion in a carrier parent, with the affected offspring inheriting a deleted 17 as a result of meiotic segregation.
Langer-Giedion syndrome This syndrome combines the earlier described trichorhinophalangeal (TRP) syndrome of Giedion with additional anomalies described by Langer. TRP, inherited as an autosomal dominant disorder, consists of hair anomalies (sparse scalp but bushy eyebrows), pearshaped nose and brachyphalangy, with cone-shaped epiphyses and Perthes-like changes. Langer described an entity incorporating TRP with mental retardation, microcephaly, multiple exostoses and redundant skin. Approximately half of the reported individuals with the syndrome have an interstitial deletion on the long arm of chromosome 8, the critical segment being 8q244. An elegant technique, the microdissection of banded chromosomes from normal metaphases followed by universal enzymatic amplification of the dissected DNA, identified sequences that were used to probe the involved region of 8q. Of the two patients studied with this technique, one had a deletion of band q23 and subband q24.11, while the other was missing the distal half of band q23 and the proximal half of sub-band q24.1’. The individual genes within the region have not yet been identified, although TRP with mental retardation has occurred in association with a deletion at 8q23. As in other contiguous gene syndromes, the degree of mental retardation is variable and individuals have been described with apparent 8q23 deletions but lacking the Langer-Giedion phenotype.
Wilms tumor (WAGR). Nearly 25 years after the association between aniridia and Wilms tumor was first reported, the deletion of llp13 was identified as the etiological factor6. The size of the deletion may correlate with the degree of mental retardation: a few of the reported cases have normal or borderline normal intelligence and very small deletions. The genito-urinary abnormalities in males vary also, from normal development to male pseudohermaphroditism. Tumor development is apparently independent of deletion size, since half-brothers with identical deletions were discordant for tumor’. The short arm of chromosome 11 has been subjected to intensive analysis, using probes for Mendelian genes and anonymous DNA segments assigned to 1lp13 in the search for the gene(s) involved in Wilms tumor. These studies have excluded the locus for familial Wilms tumor from chromosome 11, but have provided strong evidence for a discrete aniridia gene on llp13, based on cytogenetic studies of cotransmission with translocations involving 1lp13* and linkage analysis of families with aniridia and normal karyotypes’. Previously a locus for isolated, dominantly inherited aniridia (designated A N I ) was linked to the enzyme acid phosphatase-1 locus on the short arm of chromosome 2, based on one large pedigree exhibiting variable expression, lack of visual impairment and n o male-to-male transmission”. Reanalysis of that family using llp13 DNA markers should determine whether there are two (or more) different discrete genes for autosomal dominant aniridia. Like Miller-Dieker syndrome, paternal inheritance of the de novo deletion is the rule. Of eight AGR patients studied, most with Wilms tumor, seven inherited a paternally derived deleted chromosome 11. The eighth, with a maternally derived deletion, did not differ phenotypically from the others”
AGR triad This is the only microdeletion syndrome not to have an eponym. Its acronym derives from the key features: aniridia, genito-urinary malformations and mental retardation (AGR). One-third t o one-half of children with the deletion also develop
Prader- Willi syndrome Prader-Willi syndrome (PWS), first described in 1956, consists of the clinical findings of obesity, hypotonic musculature, mental retardation, hypogonadism, short stature and small hands and feet. Feeding problems in infancy (poor suck
-
m
3
hl W
hl-
m
2
6
2 2
s
2 8 %
B .-. QJ
4
9 c
P
3
8
2
2
4
825
.-2
.-
d u
0
4
826
and failure to thrive) are succeeded by hyperphagia and obesity, usually in the second year and continuing into adulthood. Since the first report of an apparently balanced D:D translocation in PWS in 1976, numerous structural rearrangements have been described, including unbalanced translocations, additional small markers and microdeletions, often with mosaicism. Most involved the pericentromeric region of the long arm of chromosome 15. In 1981, LEDBETTER et al. 12, using high-resolution banding, detected deletions in about half of 40 patients studied, and suggested that a minute deletion at 15qll-12 might be the cause of the syndrome, an observation quickly confirmed by others. PWS was the first syndrome for which exclusively paternal origin for the de novo deleted 15 was found, based on cytogenetic polymorphisms of 1 5 ~ ’ Molecular ~. markers confirmed the observation. At the time, this surprising inheritance pattern was attributed to the greater vulnerability of male gametogenesis to environmental insults occurring through an effect on DNA replication. Molecular studies using DNA probes mapping to 15ql1 showed DNA sequences t o be deleted in some, but not all, PWS patients. Various theories were advanced to explain how deletions, translocations, duplications, or even apparently normal chromosomes, could produce the same phenotype. Molecular analysis of two PWS patients lacking a visible deletion ended the speculations. One carried a familial Robertsonian 13: 15 translocation; the other had two apparently normal chromosomes 15. Cloned DNA markers for the 15qll-13 subregion showed, surprisingly, that both probands lacked any contribution of paternal alleles at 15qll-13, having inherited only maternal chromosomes 15. The transmission of the two normal homologues of chromosome 15 from their mothers-uniparental (maternal) heterodisomy-indicated that it was not the deletion of 15ql l itself which caused PWS, but rather the absence of a paternal contribution of 15ql 114.
Angelman syndrome The biological significance of band 15qll became even more intriguing with the
recognition of the identical deletion in approximately half of the studied patients with the Angelman (‘happy puppet’) syndrome (AS) of jerky movements, unprovoked laughter and mental retardation’j. To reconcile the differences in PWS and AS phenotype with identical chromosomal findings, it was first suggested that a larger deletion in Angelman syndrome not only caused the more severe mental retardation and lack of speech but also ‘suppressed’ or altered the presumed hypothalamic abnormality causing the uncontrolled appetite and obesity in PWS. This seemed unlikely because their additive effects have allowed deletions of increasing size to be used for gene mapping for half a century. The relationship between PWS and AS has now been elucidated at the molecular level. Using cloned segments of 15qll-13 as probes, the deletions in PWS and AS appear to be identical. But in contrast to the inheritance of the paternal deleted 15 in PWS, the AS patients studied each inherited a maternally deleted 1516. These differences in phenotypes, despite identical chromosome deletions, are attributed to imprinting, the differential expression of genetic information, depending on the parent from whom it is inherited. Imprinting is discussed in more detail in the concluding section and was recently reviewed by HALL”.
Alagille syndrome Alagille syndrome includes chronic cholestasis caused by intrahepatic biliary hypoplasia, pulmonary stenosis, vertebral arch defects and embryotoxon posterior. Although most cases are sporadic, pedigree studies have suggested autosomal dominant inheritance, with reduced penetrance and variable expressivity. The first report of deletion 20~11.2-12.1was confirmed in three additional patients, although other reports of 20p deletions with the typical facies, cardiac and vertebral defects, and mental retardation made no mention of cholestasis. In situ hybridization and densitometric scanning have confirmed hemizygosity for 20p probes in two probands”, 19. The deletion 20p- may parallel the l l p - deletion, i.e. not all individuals with the deletion manifest the associated developmental
defect, paucity of interlobular bile ducts or Wilms tumor, respectively, either because the deletions d o not include the critical locus or because other factors, as yet unidentified, interact with the deletion to produce the visceral abnormalities. Combined high-resolution and molecular analysis of a series of patients undoubtedly will be forthcoming, since this is the second most frequent form of infantile intrahepatic cholestasis.
DiGeorge syndrome The DiGeorge malformation sequence is an embryonic developmental defect involving the foregut in the region of the third and fourth pharyngeal pouches and the fourth branchial arch. It is manifested as absence or ectopia of part or all of the thymus, absent or hypoplastic parathyroids and great vessel malformations. Abnormalities involving chromosome 22 range from complete monosomy t o loss of 22ql1, resulting from unbalanced translocation (in approximately 20 per cent of cases studied), to interstitial deletions of 22ql 120. Unlike other contiguous gene syndromes, the DiGeorge sequence has also been reported in association with various chromosome abnormalities, deletion lop13 (or 1 0 ~ 1 4being ) the most common2’, and including 17p13, the same region deleted in Miller-Dieker syndrome. Although the DiGeorge sequence has not been reported in infants with MillerDieker, lissencephaly has been noted in a patient with DiGeorge (chromosomes unknown). The wide spectrum of chromosome anomalies may reflect the multiplicity of genes needed for the normal progression of neural crest cells into the branchial arches. Southern blotting and DNA dosage analysis of cell lines from four patients with DiGeorge syndrome and monosomy of 22pter-qll due t o unbalanced translocations confirmed the deletion of an anonymous marker (D22S9) mapping to 22q1122923.The locus was not deleted in three patients with normal chromosomes, nor in two with interstital deletions of 22qll. Detection of a different deleted locus, BCRL2, in one of the latter helps to define the critical region of 22q for this syndrome23.No molecular studies of other chromosomes associated with
DiGeorge syndrome have been reported as yet.
Beckwith- Wiedemann syndrome The main features of BeckwithWiedemann syndrome are exomphalos (omphalocele or umbilical hernia), macroglossia and visceral gigantism, with hypoglycemia in the newborn. Neoplasms, especially nephroblastoma and adrenal carcinoma, occur with increased frequency. Mental retardation has been observed in about 10 per cent of cases. A small proportion of reported invidividuals have had duplications of 1lp15, either occurring de novo or resulting from adjacent segregation of a parental translocation24. There are also familial cases with normal chromosomes. D N A analysis of affected individuals with normal chromosomes has not yet provided evidence of increased gene dosage for several loci at llp15, as would be expected if a duplication were present2’, while patients with constitutional chromosomal duplications have had gene duplications, as expected. Because of the 1 lp15 association, three karyotypically normal families segregating for Beckwith-Wiedemann syndrome were analyzed for DNA markers mapped to 1 1 ~In~all~three, . evidence was found for tight linkage to markers at llp15. In the largest pedigree it was assumed that four unaffected sisters, the children of unaffected parents, were carriers, since each had one or more affected offspring. Once the syndrome was expressed it was inherited in a dominant manner, with affected grandchildren in the pedigree. Noteworthy in this family was the unexplained paucity of transmitting males, either affected or presumed carriers with affected offspring. Yet in other cases in which the origin of the duplication could be determined, it was paternal in origin. The linkage study was uninformative regarding duplications in the 1lp15 region. Another curious discrepancy in the inheritance of Beckwith-Wiedemann syndrome is the fact that all monozygotic twin pairs have been female and discordant. Molecular anslysis of one twin pair using llp15 markers for linkage and dosage found no detectable differences”.
N”
m
b‘
82 7
2 .-* c
C
4
828
Pallister-Killian syndrome The Pallister-Killian syndrome is characterized by severe mental retardation, unusual facial appearance-coarse features, prominent forehead, ptosis, short anteverted nostrils-and abnormal scalp hair distribution (bitemporal alopecia). The associated chromosome abnormality is a tissue-limited mosaicism consisting of tetrasomy of the short arm of chromosome 12, manifested as a small marker chromosome, isochromosome 12p, found almost exclusively in fibroblasts but not peripheral lymphocytes28. When diagnosed in the newborn period, the i(12p) has been observed in 100 per cent of metaphases from direct bone-marrow analysis, but only in 23 per cent of phytohemagglutinin (PHA)-stimulated bonemarrow cells. Mosaicism for the isochromosome has been found in PHAstimulated peripheral lymphocytes in the newborn period, but the abnormal mitoses rapidly disappear from the circulation. The identification of the isochromosome has been confirmed by dosage analysis and in situ hybridization using a cDNA probe localized on 12p, as well as quantitative measurements of lactate dehydrogenase B and superoxide dismutase, enzymes mapped to 1 2 ~ No ~ studies ~ . of parental origin of the i(12p) have been reported. Summary This annotation has been confined t o well-established clinical syndromes with recently discovered chromosomal etiologies. It deliberately omits retinoblastoma, the oft-cited paradigm of a contiguous gene syndrome, since it is usually inherited as a Mendelian single gene disorder. However, it was recognition of both the deletion of band q14 of chromosome 13 in mentally retarded children with retinoblastoma, and the linkage of retinoblastoma with the genetic marker esterase D, which resulted in the eventual cloning of the gene. Also omitted are microdeletions of the x chromosome. These disorders are seen primarily in males, who manifest the phenotypic effects of the deletion of the loci of various combinations of adjacent genes: Duchenne muscular dystrophy, glycerol kinase deficiency, adrenal hypoplasia,
optic albinism, hypogonadatropic hypogonadism and anosmia (Kallman syndrome), chondrodysplasia punctata and icthyosis. Many are also mentally retarded. The third group omitted are Mendelian disorders occurring with atypical mental retardation (not usually part of the disorder), the presumption being that they include small deletions. It is expected that other contiguous gene syndromes will be recognized eventually; Rubinstein-Taybi and Cornelia de Lange syndromes are prime candidates. Why do deletions have such dramatic consequences when a normal homologue of the region is present? If their effects were due to the uncovering of recessive genes, we would expect to see greater variations in phenotype among carriers, including normal individuals whose deletions were masked by the protective effects of dominant alleles in the homologous regions. Imprinting-the ‘stamping’ of a gene as it passes through the germ line-provides a more satisfactory explanation. If the intact regions homologous to the deletions are imprinted, as suggested by the preferential parental inheritance documented for many of the syndromes, the intact chromosome need not carry recessive mutations. s.APIENZA30 suggests that imprinting results in a process of allelic inactivation, which affects the phenotype like a classical mutation. Imprinted genes would behave like recessives under these conditions, producing similar phenotypes, i.e. clinically recognizable syndromes. We do not know yet whether all of the syndromeassociated contiguous regions are imprintable, but where it is found, gamete-specific inheritance provides strong evidence. Much of the evidence for imprinting comes from the mouse. Complementation studies of disomic mice, produced by intercrossing heterozygotes for Robertsonian translocations, identified regions of chromosomes with functions determined by their parental origin3’. Some regions must be of paternal origin and others of maternal origin in order for development to proceed normally. Using mice with complementary duplication/deficient chromosomes from adjacent-1 disjunctions in both reciprocal translocation heterozygote parents, some chromosomal regions were
found to influence phenotypic characters such as shape or size of neonates, depending on parental origin. It has been suggested that methylation of genes provides the mechanism of imprinting. In transgenic mice, the methylation state of certain imprintable genes has been shown to be present in the germ line, demethylation occurring in the male, de novo methylation in the female. The offspring reproduce this pattern, resulting in phenotypes (for imprintable genes) which differ according to the parental origin. P w s and AS demonstrate that, in man as well as the mouse, not only do identical regions become functionally different, depending on the germ line through which they pass, but also that the diploid state must be maintained for normal development. Clearly, both maternal and paternal genes of the 15qll region are needed. We do not know yet how the different phenotypes result from the same deletion; how in the presence of its homologue the inheritance of a paternal 15qll ‘protects’ against PWS, whereas the maternal 15qll in similar circumstances prevents AS. Thus far the situation for chromosome 15 is unique. No other paired contiguous gene syndromes resulting from imprinting by the male and female parent, respectively, have been documented, although cases of similar deletions but differing phenotypes suggest the possibility. Uniparental disomy can result from a nondisjunction in either the first reductional (heterodisomy) or the second equational meiotic division (isodisomy). In the case of the exceptional PWS probands, each egg with two different chromosomes 15 was probably fertilized by a normal sperm, followed by loss of a paternal 15 from the zygote. Evidence for post-zygotic loss of trisomic chromosomes comes from the finding of placental mocaicism, limited to the cytotrophoblast, in viable gestations of trisomy 13 and 18, but not 213’. By analogy with PWS, we might expect the complementary situation of paternal disomy, the double contribution of paternal 15s and lack of maternal contribution, to occur in AS. It has yet to be described. The previous cases of uniparental disomy were in individuals with cystic fibrosis, who inherited two cystic
fibrosis alleles from their heterozygous carrier mothers. Contiguous gene syndromes raise intriguing questions. If imprinting in these disorders causes mental retardation in addition to the recessive phenotype, what proportion of cases of isolated mental retardation of unknown etiology have similar etiology? The breakpoints of balanced translocations have frequently provided clues to gene loci. In the absence of any gene markers, how do we search for microdeletions in mental retardation? It is intriguing to speculate also on the possible contribution of uniparental disomy to abnormal fetal development. The hydatidiform mole is a zygote containing two sets of paternally derived chromosomes. If lack of the maternal complement results in a molar pregnancy, what proportion of karyotypically normal spontaneous abortions are uniparental for an entire chromosome or imprinted region? How many imprintable loci are there in man? Cases of discordant parent/child phenotypes with identical translocations were attributed in the past to ‘aneusomie de recombination’. These should be reinvestigated to determine what proportion, if any, represents uniparental disomy. Although imprinting may best be studied in transgenic mice, we need answers regarding the human situation. Advances in our knowledge often ha;e begun with the clinician who recognized unusual manifestations of a gene. Even in the era of molecular genetics, the astute diagnostician remains an important element in the discovery process. HOPE H. PUNNETT, PhD ELAINEH . ZAKAI,M.D.* Department of Pediatrics, Temple University School of Medicine; Director, Cytogenetics Laboratory, St. Christopher’s Hospital for Children, 3601 A Street, Philadelphia, PA 19134-1095. *Department of Pediatrics, University of Pennsylvania School of Medicine; Division of Human Genetics and Molecular Genetics, The Children’s Hospital of Philadelphia, 34th and Civic Center Boulevard, Philadelphia, PA 19104. Acknowledgement This work was supported in part by National Institutes of Health Grant 1F33 CA086659 to H.H.P.
-
z m
r4 00
n i m 0”
5
5 e
2
8
2 is
2 P,
.s 2
-95
:
2 L
4
829
2 .-
c)
c
0 C
d
830
References 1. Schmickel, R. D. (1986) ‘Contiguous gene syndrome: a component of recognizable syndromes.’ Journal of Pediatrics, 109, 23 1-24], 2. van Tuinen, B., Dobyns, W. B., Rich, D. C., Summers, K. M., Robinson, T. J., Nakamura, Y., Ledbetter, D. H. (1988) ‘Molecular detection of microscopic and submicroscopic deletions associated with Miller-Dieker syndrome.’ American Journal of Human Genetics, 43, 587-596. 3. Schwartz, C . E., Johnson, J . P., Holycross, B., Mandeville, T. M., Sears, T. S., Graul, E. A., Carey, J. C., Schroer, R. J., Phelen, M. C., Szollar, J., Flannery, D. B., Stevenson, R. E. (1988) ‘Detection of submicroscopic deletions in band 1 7 ~ 1 3in ,patients with the MillerDieker syndrome. American Journal of Human Genetics, 43, 597-604. 4. Langer, L. O., Krassikoff, N., Laxova, R., Scheer-Williams, M., Lutter, L. D., Gorlin, R. J., Jennings, C. J., Day, D. W. (1984) ‘The tricho-rhino-phalangeal syndrome with exostoses (or Langer-Giedion syndrome): four additional patients without mental retardation and review of the literature.’ American Journal of Medical Genetics, 19, 81-1 11. 5. Ludecke, H.-J., Senger, G., Claussen, U., Horsthemke, B. (1989) ‘Cloning defined regions of the human genome by microdissection of banded chromosomes and enzymatic amplification.’ Nature, 338, 348-350. 6. Riccardi, V. M., Sujansky, E., Smith, A. C., Francke, U. (1978) ‘Chromosomal imbalance in the aniridia-Wilms tumor association: 1 l p interstitial deletion.’ Pediatrics, 61, 604-610. 7. Lavedan, C., Barichard, F., Azoulay, M., Couillin, P., Molina Gomez, D., Nicolas, H., Quack, B., Rethore, M.-0.. Noel, B., Junien, C. (1989) ‘Molecular definition of de novo and genetically transmitted WAGR-associated rearrangements of llp13.’ Cytogenetics and Cell Genetics, 50, 70-74.. 8. Simola, K. 0. J., Knuutila, S., Kaitila, I., Pirkola, A., Pohja, P . (1983) ‘Familial aniridia and translocation t(4: 11) (q22:p13) without Wilms’ tumor.’ Human Genetics, 63, 158-161. 9. Mannens, M., Bleeker-Wagemakers, E. M., Bliek, J., Hoovers, J., Mandjes, I., van Tol, S., Frants, R. R., Heyting, C., Westerveld, A., Slater, R. M. (1989) ‘Autosomal dominant aniridia linked to the chromosome l l p 1 3 markers catalase and D1 IS151 in a large Dutch family.’ Cytogenetics and Cell Genetics, 52, 32-36. 10. Ferrell, R. E., Chakravarti, A., Hittner, H . M., Riccardi, V. M. (1980) ‘Autosomal dominant aniridia: probable linkage to acid phosphatase-1 locus o n chromosome 2.’ Proceedings of the National Academy of Sciences, USA, 77, 1580-1582. 1 1 . Huff, V., Meadows, A., Riccardi, V. M., Strong, L. C., Saunders, G . F. (1990) ‘Parental origin of de novo constitutiona! deletions of chromosome band llp13. American Journal of Human Genetics, 47, 155-160. 12. Ledbetter, D. H., Riccardi, V. M., Airhart, S. D., Strobel, R. J., Keenan, B. S., Crawford, J. D. (1981) ‘Deletions of chromosome !5 as a cause of the Prader-Willi syndrome. New England Journal of Medicine, 304, 325-329. 13. Butler, M. G., Palmer, C. G. (1983) ‘Parental origin of chromosome 15 deletion in Prader-
Willi syndrome.’ Lancet, 1, 1285-1286. 14. Nicholls, R. D., Knoll, J. H. M., Butler, M. G . , Karam, S., Lalande, M. (1989) ‘Genetic imprinting suggested by maternal heterodisomy in non-deletion Prader-Willi syndrome.’ Nature, 342, 281-285. 15. Kaplan, L. C., Wharton, R., Elias, E., Mandell, F., Donlon, T., Latt, S. A. (1987) ‘Clinical heterogeneity associated with deletions in the long arm of chromosome 15: report of 3 new cases and their possible genetic significance.’ American Journal of Medical Genetics, 28, 45-53. 16. Knoll, J. H. M., Nicholls, R. D., Magenis, R. E., Graham, J. M., Jr., Lalande, M., Latt, S. A. (1989) ‘Angelman and Prader-Willi syndromes share a common chromosome 15 deletion but differ in Darental orinin of the deletion.’ American -Journal Medical Genetics, 32, 285-290. 17. Hall, J . G. (1990) ‘Genornic imprinting: review and relevance t o human diseases.’ American Journal of Human Genetics, 46, 857-873. 18. Byrne, J. L. B., Harrod, M. J. E., Friedman, J . M., Howard-Peebles, P . N. (1986) ‘del(20p) with manifestations of arteriohepatic dysplasia.’ American Journal of Medical Genetics, 24, 673-678. 19. Schnittger, S., Hofers, C., Heidemann, P., Beermann, F., Hansmann, I . (1989) ‘Molecular and cytogenetic analysis of an interstitial 20p deletion associated with syndromic intrahepatic ductular hypoplasia (Alagille syndrome).’ Human Genetics, 83, 239-244. 20. Zhang, F., Deleuze, J.-F., Aurias, A., Dutrillaux, A,-M., Hugan, R.-N., Alagille, D., Thomas, G., Hadchouel, M. (1990) ‘Interstitial deletion of the short arm of chromosome 20 in arteriohepatic dysplasia. Journal of Pediatrics, 116, 73-77. 21. de la Chapelle, A,, Herva, R., Koivisto, M., A d a , P. (1981) ‘A deletion in chroTosome 22 can cause DiGeorge syndrome. Human Genetics, 57, 253-256. 22. Carey, A. H., Roach, S., Williamson, R., Dumanski, J. P., Nordenskjold, M., Collins, V. P., Rouleau, G., Blin, N., Jalbert, P., Scambler, P . J. (1990) ‘Localization of 27 DNA markers t o the region of human chromosome 22qll-pter deleted in patients with DiGeorge syndrome and duplicated in the der22 syndrome.’ Genomics, 7, 299-306. 23. Fibison, W. J., Budarf, M., McDermid, H., Greenberg, F., Emanuel, B. S. (1990) ‘Molecular studies of DiGeorge syndrome.’ American Journal of Human Genetics, 46, 888-895. 24. Waziri, M., Patil, S. R., Hanson, J. W., Bartley, J. A. (1983) ‘Abnormality of chromosome 11 in patients with features of BeckwithWiedemann syndrome.’ Journal of Pediatrics, 102, 873-876. 25. Henry, I . , Jeanpierre, M., Couillin, P., Barichard, F., Serre, J.-L., Journel, H., Lamouroux, A., Turleau, C., De Grouchy, J., Junien, C . (1989) ‘Molecular definition of the llp15.5 region involved in BeckwithWiedemann syndrome and probably i; predisposition t o adrenocortical carcinoma. Human Genetics, 81, 273-277. 26. Schofield, P . N., Lindham, S., Engstrom, W. (1989) ‘Analysis of gene dosage on chromosome 11 in children suffering from BeckwithWiedemann syndrome.’ European Journal of Pediatrics, 148, 320-324. 27. Litz, C . E., Taylor, K. A., Qiu, J. S., Pescovitz,
07
0. H., de Martinville, B. (1988) ‘Absence of
detectable chromosomal and molecular abnormalities in monozygotic twins discordant for the Wiedemann-Beckwith syndrome.’ American Journal of Medical Genetics, 30, 821-833. 28. Hersh, J. H., Graham, J. M., Jr., Destrempes, B. S., Greenstein, R. M. (1983) ‘TeschlerNicola/Killian syndrome: a case report.’ Journal of Clinical Dysmorphology, 1, 20-24. 29. Peltomaki, P., Knuutila, S., Ritvanen, A,, Kaitila, I., de la Chapelle, A. (1987) ‘PallisterKillian syndrome: cytogenetic and molecular
studies.’ Clinical Genetics, 31, 399405. 30. Sapienza C. S. (1989) ‘Genome imprinting and dominance modification.’ Annals of the New York Academy of Sciences, 564, 24-37. 31. Cattanach, B. M., Kirk, M. (1985) ‘Differential activity of maternally and paternally derived chromosome regions in mice.’ Nature, 315, 496498. 32. Kalousek, D. K., Barrett, I. J., McGillivray, B. C. (1989) ‘Placentak mosaicism and intrauterine survival of trisomies 13 and 18.’ American Journal of Human Genetics, 44, 338-343.
-
m m
d
2 0-
m
2
NOTICES
Children-the Forgotten Mourners
10th Margaret Lowenfeld Day Conference
London, 17th September 1990
Cambridge, Saturday 10th November 1990
This symposium, with discussion groups, will be held at Queen Charlotte’s and Chelsea Hospital on Monday, 17th September 1990. Fee f45. Further information from Symposium Secretary, RPMS Institute of Obstetrics and Gynaecology, Queen Charlotte’s and Chelsea Hospital, Goldhawk Road, London W6 OXG. Tel.: 081-740 3904.
Neurofibromatosis Clinical Care Sympoisum Cincinnati, Ohio, October 1990 This anmal symposium, sponsored by The National Neurofibromatosis Foundation, will be held at the Cincinnati Convention Center, Cincinnati, on Sunday, 21st October 1990 from 8.30am to 12.30pm. Further information from The National Neurofibromatosis Foundation, Inc., 141 Fifth Avenue, Suite 7 3 , New York, NY 10010.
1990 American Epilepsy Society Annual Meeting San Diego, California, 10th to 15th November 1990 The four main symposia at this annual meeting, to be held at the Sheraton Harbor Island Hotel, San Diego, will be ‘Spread of seizure discharge’; ‘Controversies in epileptology’ (‘Is topographic mapping of EEG useful?’ and ‘Are current methodologies of anti-epileptic drug trials satisfactory?’); ‘Mesial temporal sclerosis: the neuropathology of epilepsy’ and ‘Seizures in the first year of life’. Further information from The American Epilepsy Society, 638 Prospect Avenue, Hartford, CT 06105-4240. Tel.: (203) 232 4825.
Preverbal communication was central to Margaret Lowenfeld’s thinking. To celebrate the centenary of her birth, this seminar will relate some of her ideas to current thinking, with particular reference to play, nursery schools and autism. There will be a demonstration of Lowenfeld techniques. Speakers include John Davis, Mollie Dundas, Dorothy Einon, Nina Farhi, Ian Goodyer, Elizabeth Newson, Robert Royeton, Alison Van Dyk, Margarita Wood and Beric Wright. This one-day conference will be of interest to child psychotherapists and psychiatrists, psychologists, social workers, health visitors and .others interested in the emotional health of children in its social context. Details from Child Care and Development Group, Free School Lane, Cambridge CB2 3RF.
Frontiers of Research in Developmental Neurology Groningen, The Netherlands, 2lst to 23rd March 1991 This international workshop is being organized by the Department of Developmental Neurology, University Hospital, Groningen. Topics will be ‘Results of brain imaging in relation to their functional neurological correlates in infants and children’; ‘Qualitative vs. quantitative neurological assessment of infants and children’; and ‘Instrumental measurements of complex normal and abnormal movements in children’. Further information from Professor H. F. R. Prechtl, Department of Developmental Neurology, University Hospital, Oostersingel59, 9713 EZ Groningen, The Netherlands.
831
