, . 17:8: 780–783 (1997)

SHORT COMMUNICATION

PROSPECTIVE PRENATAL DIAGNOSIS OFPRADER–WILLI SYNDROME DUE TO

MATERNAL DISOMY FOR CHROMOSOME 15FOLLOWING TRISOMIC ZYGOTE RESCUE

1*, . 1, . 1, . . . 2 . . 1

1Regional Genetics Laboratories and Department of Clinical Genetics, Birmingham Women’s Hospital, Edgbaston,Birmingham B15 2TG, U.K.

2Royal Shrewsbury Hospitals NHS Trust, Shrewsbury, U.K.

Received 18 September 1996Revised 16 January 1997Accepted 22 January 1997

SUMMARY

We present a prenatal predictive diagnosis of Prader–Willi syndrome arising as a result of maternal heterodisomyfor chromosome 15. The diagnosis arose following chorionic villus sampling which showed a mosaic trisomy 15karyotype with a chromosomally normal follow-up amniocentesis. Molecular studies on DNA extracted fromcultured amniocytes showed no evidence of a paternal allele at two widely separated loci and this was taken asevidence of maternal disomy predictive of Prader–Willi syndrome in the fetus. ? 1997 by John Wiley & Sons, Ltd.

Prenat. Diagn. 17: 780–783, 1997No. of Figures: 2. No. of Tables: 0. No. of References: 9.

: maternal disomy; trisomy 15 CVS

INTRODUCTION

Complete trisomy 15 is a very rare finding whichis not compatible with survival beyond the neo-natal period but is associated with severe intra-uterine growth retardation (IUGR) (Coldwellet al., 1981). A fetus with mosaicism for a trisomy15 cell line may survive to term but is likely to haveIUGR, craniofacial dysmorphism, minor skeletalabnormalities, hypotonia, and congenital heartdisease (Milunsky et al., 1996). Prenatal diagnosisof trisomy 15 or trisomy 15 mosaicism is furthercomplicated by a number of recent reports of live-births with Prader–Willi syndrome resulting frommaternal uniparental disomy for chromosome 15,

where chorionic villus sampling in early preg-nancy has shown the presence of a trisomy 15 cellline, but with a normal follow-up amniocentesis(Cassidy et al., 1992; Morichon-Delvallez et al.,1993; Purvis-Smith et al., 1992). This case repre-sents a prospective prenatal diagnosis of Prader–Willi syndrome due to maternal heterodisomy.

CASE REPORT

Chorionic villus sampling (CVS) was per-formed at 11 weeks’ gestation on a gravida 3,para 2 patient because of a previous livebirthwith Down syndrome, and a maternal age of39 years. Ultrasound examination at 11 weeksshowed no evidence of growth retardation. Chro-mosome analysis of direct CVS preparationsshowed a 47,XX,+15 karyotype in all 30 cells

*Correspondence to: Eileen Roberts, Regional GeneticsLaboratories, Birmingham Women’s Hospital NHS Trust,Edgbaston, Birmingham B15 2TG, U.K.

CCC 0197–3851/97/080780–04 $17.50? 1997 by John Wiley & Sons, Ltd.

examined. Cultured CVS material showed amosaic 46,XX/47,XX,+15 karyotype (45 cells:25cells). A follow-up amniocentesis sample at 15weeks showed an apparently normal karyotype in229 cells from three different cultures.DNA was extracted from cultured amniocytes

and parental bloods using a Puregene kit (Flow-gen). Molecular analysis using LS6-1 (D15S113) [ahighly polymorphic (CA)n repeat motif locatedwithin the Prader–Willi/Angelman syndrome criti-cal region on chromosome 15] and pMS-620 (aVNTR probe located on telomeric 15q) showedthe fetus to have inherited both maternal alleles atthese two loci, with no evidence of a paternal band(Fig. 1, data for pMS-620 not shown). Othermarkers from chromosome 15 proved uninforma-tive. False paternity was excluded using probesfrom chromosome 4 (D4S43 and D4S126) andSTR probes from the DMD region of the Xchromosome.These results were taken as indicative of mater-

nal heterodisomy for chromosome 15, and conse-quently a Prader–Willi phenotype, and thepregnancy was terminated at 20 weeks’ gestation.A detailed post-mortem examination showed no

evidence of any gross internal or external abnor-malities, in particular no evidence to support atrisomy 15 cell line within the fetus. Follow-upcytogenetic, molecular, and fluorescent in situhybridization (FISH) studies (using D15Z1,Oncor) were performed on a range of fetal andplacental tissues. In all 210 cells examined fromfetal tissues (blood, skin, muscle, liver, lung, andbrain), the karyotype was 46,XX. FISH studies onfetal tissues showed 9 per cent of cells to have three

signals for D15Z1, compared with 6 per cent in acontrol. Placental tissue from six different sitesshowed the karyotype in 41 cells to be 46,XX andin 139 cells to be 47,XX,+15; FISH studies showed54 per cent of placental cells with two signals forD15Z1 and 46 per cent with three signals. Thesestudies suggest that the trisomy 15 cell line wasconfined to the placenta. Molecular follow-upstudies confirmed the previous results with LS6-1(D15S113) and pMS-620 (Fig. 2) and therebyconfirmed the finding of maternal heterodisomyfor chromosome 15.

DISCUSSION

Maternal disomy for chromosome 15 as a causeof Prader–Willi syndrome is well established(Nicholls et al., 1989). This case represents only thethird reported case of a prospective prenatal diag-nosis of maternal heterodisomy for chromosome15 predictive of Prader–Willi syndrome, withoutthe apparent presence in the fetus of a trisomy 15cell line (Surh et al., 1994; Christian et al., 1996).Recent reports of similar cases of a trisomy 15

cell line detected on direct or cultured CVS whichwent on to have apparently normal karyotypes atamniocentesis but resulted in livebirths withPrader–Willi syndrome illustrate the importance ofprenatal molecular genetic investigations in suchcases (Cassidy et al., 1992; Morichon-Delvallezet al., 1993; Purvis-Smith et al., 1992).Follow-up investigations are essential to con-

firm the diagnosis and also to discriminate be-tween cases of this nature and genuine trisomy 15

Fig. 1—Results of molecular analysis using LS6-1 (D15S113). Lane 2=maternal DNAshowing alleles 1 and 4; lane 3=DNA extracted from cultured amniocytes showingalleles 1 and 4; lane 4=paternal DNA showing alleles 2 and 3

781 15

? 1997 by John Wiley & Sons, Ltd. . ., . 17: 780–783

mosaics, for which a characteristic phenotypeappears to be emerging (Milunsky et al., 1996).Interestingly, had this patient opted for prenataldiagnosis by amniocentesis rather than CVS, thematernal heterodisomy 15 would have goneundetected and a livebirth with Prader–Willisyndrome would have resulted.Although the recurrence risk of Prader–Willi

syndrome is predicted to be low, this patient hashad two trisomic conceptions and therefore therecurrence risk for any trisomy is likely to besignificant. The risk of an abnormal outcomeresulting from disomy for any chromosome fol-lowing trisomic zygote rescue is difficult to predictsince (i) although this is a theoretical possibility in1/3 of trisomic zygote ‘rescues’, there are verylimited prospective data to determine whether theobserved risk is as high as this; and (ii) not alltrisomic zygote rescues leading to disomy willresult in a clinical phenotype, but only in thosechromosomes with imprinting effects. However, itis none the less essential to perform molecularinvestigations for uniparental disomy when thechromosome involved has a known imprintingeffect (chromosomes 7, 11, 14, and 15) and may bewarranted in all cases of trisomy mosaicismdetected at CVS (Ledbetter and Engel, 1995;Christian et al., 1996).

We thank Anne Jay, Paul Miller, Sean Rose andClive Felix for technical assistance.

REFERENCES

Cassidy, S.B., Lai, L.W., Erickson, R.P., Magnuson, L.,Thomas, E., Gendron, R., Herrman, J. (1992). Tri-somy 15 with loss of the paternal 15 as a cause ofPrader–Willi syndrome due to maternal disomy, Am.J. Hum. Genet., 51, 701–708.

Christian, S.L., Smith, A.C.M., Macha, M., Black, S.H.,Elder, F.F.B., Johnson, J.M.-P., Resta, R.G., Surti,U., Suslak, L., Verp, M.S., Ledbetter, D.H. (1996).Prenatal diagnosis of uniparental disomy 15 followingtrisomy 15 mosaicism, Prenat. Diagn., 16, 323–332.

Coldwell, S., Fitzgerald, B., Semmens, J.M., Ede, R.,Bateman, C. (1981). A case of trisomy of chromosome15, J. Med. Genet., 18, 146–148.

Ledbetter, D.H., Engel, E. (1995). Uniparental disomyin humans: development of an imprinting map and itsimplications for prenatal diagnosis, Hum. Mol.Genet., 4, 1757–1764.

Milunsky, J.M., Wyandt, H.E., Huang, X.L., Kang,X.Z., Elias, E.R., Milunsky, A. (1996). Trisomy 15mosaicism and uniparental disomy (UPD) in a live-born infant, Am. J. Med. Genet., 61, 269–273.

Morichon-Delvallez, N., Mussat, P., Dumez, Y.,Vekemans, M. (1993). Trisomy 15 in chorionic villi

Fig. 2—Results from DNA extracted post-TOP using the probe pMS-620/Pst1. Lane 1=maternal DNA showingalleles 2 and 4. Lanes 2–8: fetal tissues all showing alleles 2 and 4; lane 2=fetal muscle tissue; lane 3=culturedamniocytes; lane 4=fetal skin; lane 5=fetal brain; lane 6=fetal lung; lane 7=fetal liver; lane 8=cultured amniocytes;lane 9=paternal DNA showing alleles 1 and 3

782 . .

. ., . 17: 780–783 (1997) ? 1997 by John Wiley & Sons, Ltd.

and Prader–Willi syndrome at birth (letter), Prenat.Diagn., 13, 307–308.

Nicholls, R.D., Knoll, J.H.M., Butler, M.G., Karam, S.,Lalande, M. (1989). Genomic imprinting suggested bymaternal heterodisomy in non-deletion Prader–Willisyndrome, Nature, 342, 281–285.

Purvis-Smith, S.G., Saville, T., Manass, S., Yip, M.Y.,Lam-Po-Tang, P.R.L., Duffy, B., Johnston, H.,

Leigh, D., McDonald, B. (1992). Uniparental disomy15 resulting from ‘correction’ of an initial trisomy 15,Am. J. Hum. Genet., 50, 1348–1350.

Surh, L.C., Wang, H., Hunter, A.G.W. (1994). Deletionand uniparental disomy involving the same maternalchromosome 15, N. Engl. J. Med., 330, 572–573.

783 15

? 1997 by John Wiley & Sons, Ltd. . ., . 17: 780–783