Archives ofDisease in Childhood 1995; 72: 285-286 285 Molecular biology of Turner's syndrome Arch Dis Child: first published as 10.1136/adc.72.4.285 on 1 April 1995. Downloaded from

Turner's syndrome was first described in 19381 and results complicated by the possibility of undetected mosaicism for in a clinical picture most frequently comprising short a second sex chromosome. stature and gonadal dysgenesis. Other features may include coarctation of the aorta, renal anomalies, neck webbing, and lymphoedema. The clinical phenotype can Sex chromosome mosaicism in Turner's syndrome be extremely variable even in patients with the same Approximately one in 50 of all conceptuses are associated karyotype. The chromosomal basis of the condition was with a 45,X genotype but there is high intrauterine lethal- established in 1959 when a patient was described who ity such that only 1% of such conceptuses survive to term. exhibited loss of one sex chromosome.2 Approximately There is a higher percentage of mosaic karyotypes than 50% of cases have a 45,X karyotype with the remainder monosomy X in liveborns compared with fetuses which having mosaic karyotypes with one 45,X cell line and has led to the speculation that all liveborn infants with another cell line often containing a structurally abnormal Turner's syndrome are mosaic in a cell line critical for fetal X chromosome such as a ring X. survival.10 This hypothesis has not been supported by Turner's syndrome is one of the commonest chromo- studies on blood using Southern blotting with hypervari- somal anomalies in man with an incidence of at least one in able probes from the X chromosome11 or on cytogenetic 1850 live female births.3 Recently more interest has been studies examining multiple tissues'2 where the authors focused on this condition because of the advent of recom- found that 20% of patients still appear to have monosomy binant growth hormone treatment,4 and patients with X. It is possible that with more sensitive techniques low Turner's syndrome now comprise a large proportion of the level mosaicism may be detected. As mentioned pre- cases attending paediatric growth clinics. In spite of the viously, approximately 70-80% of cases of Turner's fact that it has been recognised as an entity for a consider- syndrome retain the maternal X. As the missing sex able length of time, the molecular biology of the condition chromosome may be either an X or a Y theoretically has not been fully elucidated. Recent research using 35-40% of cases may have occult Y mosaicism.13 This has molecular genetic techniques has improved our under- clinical relevance because if Y material is present there is a standing of the condition but more work is still required in risk of up to 30% of gonadoblastoma developing in the this field. dysgenetic gonads.14 At present if a Y chromosome is identified on cytogenetic analysis, gonadectomy is recom- mended but counselling is more difficult if there is an Parental origin ofthe retained X chromosome abnormal Y or a fragment of the Y. It would be helpful to The loss ofone ofthe sex chromosomes that is the basis for define an 'at risk' locus, the presence of which would indi- Turner's syndrome probably occurs after the zygote has cate a need for gonadectomy and there is a postulated formed or just after the fusion of the gametes. This being 'gonadoblastoma locus' on the Y chromosome which is the case one would expect an equal chance of either believed to be situated on the long arm of the Y just below http://adc.bmj.com/ parental X being retained. In practice, however, in the centromere.15 16 Although there is debate about the 70-80% of cases the retained X is maternal in origin.5 This need to screen patients with Turner's syndrome for occult observation has led to the speculation that there may be Y mosaicism, if such screening were to be implemented it genes present on the X chromosome which are expressed would be logical to include analysis for this particular area differently depending upon whether they are maternally or of the Y chromosome in the screening strategy, and recent paternally derived. This process is called 'genomic publication of polymerase chain reaction (PCR) primers imprinting' and has been described in a number of condi- covering the whole of the Y chromosome makes this a on September 30, 2021 by guest. Protected copyright. tions including Prader-Willi and Angelman's syndromes. fairly straightforward procedure.'7 It has been postulated that imprinting may play a part in Although theoretically 35-40% of patients with the high intrauterine mortality and variable phenotype Turner's syndrome may have Y mosaicism, in practice this seen in Turner's syndrome. Evidence for this has been has not been demonstrated. Studies to date have used contradictory. One study showed that the morphological either Southern blotting with Y specific probes or PCR appearances of fetuses retaining the paternal X differed with one or two Y specific primers'3 18 and have shown from those retaining the maternal X in a small sample, between 4-8% of cases of Turner's syndrome have Y perhaps indicating that fetuses retaining the paternal X did material present. Recently a new technique using Southern 'less well' than those retaining the maternal X.6 This led to blotting of DNA previously amplified by PCR has conjecture that loss of the maternal sex chromosome may purported to demonstrate a very high incidence of unsus- be less compatible with intrauterine survival than loss of pected Y mosaicism in patients with Turner's syndrome.'9 the paternal sex chromosome. However, the pathological This particular technique is extremely sensitive but is also descriptions of 45,X abortuses do not indicate any unusual very prone to contamination from external sources so that degree of developmental abnormality7 and the percentage such results should be interpreted with care.16 An added of aborted fetuses retaining the maternal X chromosome is difficulty in interpreting these findings is the uncertainty of the same as in liveborns.7 8 Therefore imprinting does not the effects of very low level mosaicism found in blood. seem to play an important part in the high fetal loss in Turner's syndrome conceptuses. Imprinting may affect the phenotype and recent work X inactivation of structurally abnormal X has indicated that patients retaining the maternal X have a chromosomes greater incidence of cardiovascular anomalies and neck Another factor which may affect the phenotype in Turner's webbing than those retaining the paternal X and also that syndrome is X inactivation of structurally abnormal the pretreatment height of those retaining the maternal X X chromosomes. Inactivation of most of one of the two X correlates very strongly with maternal height but not with chromosomes occurs in normal females at an early stage of paternal height.9 The interpretation of such findings are embryogenesis so that there is equal gene dosage between 286 Chu, Connor

males and females. Thus only one X chromosome should help explain the variability and make counselling patients be active in each cell. Inactivation of the X chromosome is easier.

controlled from the X inactivation centre situated on the X C E CHU Arch Dis Child: first published as 10.1136/adc.72.4.285 on 1 April 1995. Downloaded from long arm. Although X inactivation is initially thought to be J M CONNOR Duncan Guthrie Institute ofMedical Genetics, random, in cases with a structurally abnormal X chromo- Yorkhill, some the cell line in which the normal X is active gradually Glasgow G3 8SJ takes over.20 CEC was funded by Kabi-Pharmacia but is now in receipt of a grant from Anomalous X inactivation has been postulated to be SHERT. implicated in the phenotype found in a subgroup of patients with Tumer's syndrome and small ring X 1 Turner HH. A syndrome of infantilism, congenital webbed neck, and chromosomes who have dysmorphic facies, syndactyly of cubitus valgus. Endocrinology 1938; 23: 566-74. 2 Ford CE, Jones KW, Polani PE, de Almeida JC, Briggs JH. A sex chromo- the hands and feet, and severe mental retardation.2' A pro- some anomaly in a case of gonadal dysgenesis (Turner's syndrome). portion of these small rings remain active in the same cells Lancet 1959; i: 711-3. 3 Nielsen J, Wohlert M. Chromosome abnormalities found among 34 910 as the normal X and these cells therefore have functional newborn children: results from a 13-year incidence study in Arhus, duplication of areas of the X. The mechanism for these Denmark. Hum Genet 1991; 87: 81-3. 4 Rosenfeld RG, Frane J, Attie KM, et al. Six-year results of a randomised, rings remaining active was initially thought to be simple prospective trial of growth hormone and oxandrolone in Turner syn- loss of the X inactivation centre situated on the X long arm drome.7 Pediatr 1992; 121: 49-55. 5 Lorda-Sanchez I, Binkert F, Maechler M, Schinzel A. Molecular study of (Xq13) due to the small size of the rings. However, recent 45,X conceptuses: correlation with clinical findings. Am 7 Med Genet work looking at expression of a gene that is only expressed 1992; 42: 487-90. 6 Hassold T, Benham F, Leppert M. Cytogenetic and molecular analysis of from the inactive X-XIST (X inactive specific transcripts) sex-chromosome monosomy. Am J Hum Genet 1988; 42: 534-41. has shown that the inactivation centre is often present in 7 Epstein CJ. Mechanisms leading to the phenotype of Turner syndrome. In: Rosenfeld RG, Grumbach MM, eds. Turner syndrome. New York: Marcel these small rings22 but its expression may not be normal, Dekker, 1990: 13-28. possibly due to mutations in the gene.23 8 Jacobs PA, Betts PR, Cockwell AE, et al. A cytogenetic and molecular reap- praisal of a series of patients with Turner's syndrome. Ann Hum Genet 1990; 54: 209-23. 9 Chu CE, Donaldson MDC, Kelnar CJH, et al. Possible role ofimprinting in the Turner phenotype. JMed Genet 1994; 31: 840-2. The search for Turner 'genes' 10 Hook EB, Warburton D. The distribution of chromosomal genotypes The particular phenotype seen in Turner's syndrome can associated with Turner's syndrome: livebirth prevalence rates and evi- dence for diminished fetal mortality and severity in genotypes associated occur with a number of different karyotypic pictures and with structural X abnormalities or mosaicism. Hum Genet 1983; 64: 24-7. certain features have been tentatively mapped to areas of 11 Gicquel C, Cabrol S, Schneid H, Girard F, Le Bouc Y. Molecular diagnosis in Turner's syndrome. J Med Genet 1992; 29: 547-51. the X chromosome, for instance short stature to the X 12 Held KR, Kerber S, Kaminsky E, et al. Mosaicism in 45,X Turner syn- short arm, ovarian function to both the long and short drome: does survival depend on the presence of two sex chromosomes? Hum Genet 1992; 88: 288-94. arms.24 The phenotype in Turner's syndrome is thus not 13 Ostrer H, Clayton CM. Y chromosome mosaicism in 45,X Turner syn- thought to be due to loss of the entire X chromosome but drome [Letter]. Am J Med Genet 1989; 34: 294-6. 14 Verp MS, Simpson JL. Abnormal sexual differentiation and neoplasia. rather to haploid dosage of a gene or genes. If there is a Cancer Genet Cytogenet 1987; 25: 191-218. Turner's syndrome gene it must have the following charac- 15 Page DC. Hypothesis: a Y chromosomal gene causes gonadoblastoma in dysgenetic gonads. Development 1987; 101 (suppl): 151-5. teristics: it must escape inactivation and it must have a 16 Page DC. Y chromosomal sequences in Turner's syndrome and risk of homologue on the Y chromosome or all XY individuals gonadoblastoma or virilisation [Letter]. Lancet 1994; 343: 240. 17 Vollrath D, Foote S, Hilton A, et al. The human Y chromosome: a 48- http://adc.bmj.com/ would have the Turner phenotype. One gene fulfilling interval map based on naturally occurring deletions. Science 1992; 258: these criteria, RPS4X/RPS4Y, was put forward as a likely 52-9. 18 Yankowitz J, Neely EK, Hajdu K, Norton ME, Rosenfeld RG, Golbus MS. candidate but has the disadvantage of being located on the Screening of Turner subjects by PCR for occult Y chromosomal DNA. X long arm and gene expression studies in patients with X Am JHum Genet 1993; 53: 1766A. 19 Kocova M, Siegal SF, Wenger SL, Lee PA, Trucco M. Detection of Y isochromosomes have shown that it is expressed from the chromosomal sequences in Turner's syndrome by Southern blot analysis abnormal X.25 Several other genes also fulfil these criteria: of amplified DNA. Lancet 1993; 342: 140-3. 20 Therman E, Denniston C, Sarto GE, Ulber M. X chromosome constitution XE7 whose function is unknown, MIC2 which encodes a and the female phenotype. Hum Genet 1980; 54: 133-43. glycoprotein involved in T cell adhesion, and ZFX/ZFY 21 Kushnick T, Irons TG, Wiley JE, Gettig EA, Rao KW, Bowyer S. on September 30, 2021 by guest. Protected copyright. 45,X/46,Xr(X) with syndactyly and severe mental retardation. Am Y Med which encodes a zinc finger protein. It is probable that Genet 1987; 28: 567-74. more genes with these characteristics have yet to be dis- 22 Dennis NR, Collins AL, Crolla JA, Cockwell AE, Fisher AM, Jacobs PA. Three patients with ring (X) chromosomes and a severe phenotype. J Med covered. Genet 1993; 30: 482-6. In conclusion, Turner's syndrome is a relatively 23 Migeon BR, Luo, S, Stasiowski BA, et al. Deficient transcription of XIST from tiny ring X chromosomes in females with severe phenotypes. Proc common condition and its variable phenotype causes dif- NadAcad Sci USA 1993; 90: 1205-9. ficulties in diagnosis and counselling. The various factors 24 Therman E, Susman B. The similarity of phenotypic effects caused by Xp and Xq deletions in the human female: a hypothesis. Hum Genet 1990; 85: affecting the phenotype have not yet been fully elucidated 175-83. but include occult mosaicism, imprinting, and anomalous 25 Fisher EMC, Beer-Romero P, Brown LG, et al. Homologous ribosomal protein genes on the human X and Y chromosomes: escape from X- X inactivation. Thus far investigations have mainly con- inactivation and possible implications for Turner syndrome. Cell 1990; 63: centrated on blood but analysis of other tissues may also 1205-18.