Annales de Genetique 43 (2000) 125–130© 2000 Editions scientifiques et medicales Elsevier SAS. All rights reserved Case reportS0003-3995(00)01017-0/FLA

Unbalanced translocation t(15;22) in ‘severe’ Prader–Willi syndrome

Arabella Smith a*, Anna Jauch b, Luke St. Heaps a, Lisa Robson a, Brian Kearney c

a Department of Cytogenetics, Royal Alexandra Hospital for Children, P.O. Box 3515, Westmead, Parramatta, NSW 2124, Australiab Institute of Human Genetics, University of Heidelberg, Heidelberg, Germany

c Royal Alexandra Hospital for Children, Westmead, Westmead, Parramatta, NSW 2124, Australia

Abstract – A 13-year-old girl with an unbalanced karyotype 45,XX,-15,der(22)t(15;22)(q13;q13.3) de novo had Prader–Willi syndrome(PWS), (score 13.5), but with features of mental and physical retardation more severe than usually seen in PWS. The clinical diagnosis ofPWS was confirmed by methylation analysis that showed absence of the paternal band. With GTG banding, the cytogenetic breakpointon chromosome 15q13, with 15q14 intact, encompassed the PWS region, while the breakpoint on 22q was terminal. Investigations withFISH utilised ten different probes/combinations, namely SNRPN/PML, TUPLE1/22q13.3, TUPLE/ARSA, GABRB3, three YAC clonesand one cosmid for specific regions within chromosome 15q, painting probes for the long arm of chromosomes 15 and 22 and apantelomere probe. Deletion of SNRPN,TYAC 9 (at 15q11-12), TYAC19 (at 15q13) and GABRB3 (within the PWS locus), was evidenton the derivative (22) chromosome, while TYAC10 (at 15q22), cos15-5 (at 15q22) and PML (15q22) were not deleted. On the der(22),22q13.3 and ARSA were not deleted, but the most distal non specific pantelomeric probe was deleted. Thus, the severe phenotype couldbe attributable to deletion on chromosome 15q extending beyond q13 to q14, (further than the usual chromosome 15q deletion (q11-13)in PWS), or be related to loss of the very terminal 22q region (from ARSA to the pantelomere) or be due to genetic factors elsewhere inthe genome. © 2000 Editions scientifiques et medicales Elsevier SAS

SNRPN / FISH / probes / methylation / del(22(q13.3) / telomeres

1. Introduction

The Prader–Willi syndrome (PWS), first describedin 1956, comprises severe neonatal hypotonia, hy-pogonadism, gross obesity, short stature, smallhands and feet and mental handicap [17]. Despite theapparently distinctive phenotype, no feature is spe-cific to PWS and diagnosis is often delayed [2, 10].Clinical suspicion of PWS can be aided by use ofconsensus diagnostic criteria [11] which give a scoreto cardinal features, but these are not rated forseverity. Symptoms more (or less) severe than ‘usu-ally found in PWS’ are difficult to interpret and thisaspect of the clinical evaluation is not covered by thescoring system. Thus, some patients have been de-scribed with ‘atypical PWS’ [19], or ‘extended’ or‘expanded’ phenotypes [16], in others two diagnoses[23] or alternate diagnoses [9] were made and some

patients have a phenotype overlapping with PWS[1, 18].

The genetic mechanisms involved in the causationof PWS are complex, but well delineated and thecritical region for PWS on chromosome 15(q11-13)is subject to genomic imprinting [14]. The mostcommon mechanism, in �75 % of patients is pater-nal deletion, then maternal uniparental disomy(UPD) in �24 % of patients and about 1 % ofpatients have an imprinting defect [4]. The deletion isusually interstitial, within 15q11-13, but a structuralrearrangement may provide the framework for dele-tion and approximately 5 % of patients have astructural rearrangement [5, 12, 21, 22].

We present a patient with severe PWS and anunbalanced translocation, studied by FISH, to fur-ther characterise the breakpoints and elucidate thenature of the severity.

* Correspondence and reprints:E-mail address: ellies@nch.edu.au (A. Smith).

A. Smith et al. / Ann. Genet. 43 (2000) 125–130126

2. Case report

The patient was a 12-year-old girl, born to non-consanguinous healthy parents when the mother wasaged 31 and the father 40 years. There were noadverse prenatal influences. The family history wasunremarkable — the patient had an older sister andyounger brother (both normal) and there were nomiscarriages or stillbirths. She was born at term byvaginal delivery after induction of labour for lack ofmovements. Birth weight was 2.5 kg. Severe neonatalhypotonia was noted and initially tube feeding wasrequired, then she was very slow to feed and failed tothrive for over 6 months. At the age of 9 years, themain complaint was the insatiable appetite, constantexcessive hunger and quest for food, which contin-ued through the night and kept the family awake.

At 12 years, she had severe to profound retarda-tion and no speech at all, although hearing wasnormal and she had ongoing speech therapy. She didnot play or interact with her surroundings. She hadnever vomited, although she ate inedibles and for-aged for food. She did not indicate toiletting needsbut was toilet timed during the day (every 1 h 30min) and in nappies at night. Skin picking was aproblem and she had disrupted sleep pattern withsleep apnoea. Examination revealed a short obesegirl, with the facial features of PWS — narrowbitemporal diameter, almond shaped eyes with up-

slanting palpebral fissures, short tented upper lip(figure 1). She was markedly overweight — 75 kg (15kg over 97th centile), of short stature (height be-tween 140 and 143 cm (between 3 and 10th centiles)and head circumference 51.5 cm (0.5 cm above the3rd centile). Her foot length was 18 cm (3 cm below3rd centile) with small shoe size (size 1), whole handmeasurement was 15.5 cm (on 3rd centile) middlefinger length 6 cm (on 3rd centile). Reflexes weredecreased. The PWS score was maximum at 13.5[11].

The mother complained that the patient had hadsevere bowel problems for the past 4 years, withsignificant pain and constipation, requiring as manyas two suppositories per day. Extensive bowel inves-tigations (including X-ray and ultrasound scans)were normal and there were no markers for chronicabdominal disease, such as perianal lesions or club-bing or significant oedema. Full blood count, com-plete blood chemistry, liver function tests, serumamylase and renal function tests which were allnormal; abdominal ultrasound was normal. TheESR (30) was raised and she had a urinary tractinfection (UTI) detected on urine microscopy andculture. After antibiotic treatment of the UTI, theIVP was normal and catheter microurine showed nogrowth.

IQ tests at 5 years of age, with Vineland adaptivebehaviour scale and the Woodcock–Johnson scale ofindependent behaviour, suggested that compared tothe general population, she had a very low level ofcompetence in the three domains measured by theVinelands ABS. She had a ‘moderate deficit’ in thecommunication, socialisation and daily living skillsdomains and a ‘severe deficit’ in the motor skillsdomain. Overall, she had a severe deficit in the levelof functioning compared to other 5 year olds in thepopulation. Woodcock–Johnson SIB concurred withthose of the Vineland ABS in describing the patientas exhibiting a significant level of maladaptive be-haviour. Assessment at 12 years confirmed the earlierfindings.

3. Methods and results

3.1. Cytogenetic analysis

Cytogenetic analysis was performed by standardmethods on peripheral blood [22] and GTG bandingshowed that the patient had an unbalanced kary-otype –45,XX,–15,der(22)t(15;22)(q13;q13.3) (figure2). Both parents had normal karyotypes. The der(22)Figure 1. The patient at age 12 years.

A. Smith et al. / Ann. Genet. 43 (2000) 125–130 127

Figure 2. Partial GTG banded metaphase spread with the normal15 (bottom arrow), normal 22 (top left arrow), and der(22)chromosome (central arrow).

YACs (FITC labelled) within the PWS/AS region —YAC9 from 15q11-12 and TYAC19 from 15q13 —together with chromosome 22q arm painting probe(Cy3) showed both YACs deleted (figure 5a,b). Simi-lar procedures with two probes from outside thePWS/AS region — TYAC10 from 15q22 and cos-mid probe cos15-5 from 15q22 — showed that theYAC and cosmid were not deleted (figure 5c,d). Atwo colour FISH procedure using chromosome 22qarm painting probe (Cy3) and the all human telom-ere repeat probe (pantelomere, Oncor) showed sig-nals on the terminal ends of chromosome 15q and22q, but not within the der(22) (figure 5e), indicatingdeletion of the 22q telomeric region.

Figure 3. Metaphase spreads with FISH. (a) GABRB3 and PMLprobe showing deletion of GABRB3 (arrowhead); (b) probe Tu-ple1/22q13.3 — red signals, showing 22q13.3 intact within theder(22) chromosome (arrowhead).

chromosome was not a Robertsonian translocationbut an apparently tandem translocation. The break-point on chromosome 15q was within q11-13, withthe remainder of the chromosome intact and on 22qwas terminal at q13.3.

3.2. Methylation analysis

Methylation analysis was performed on a speci-men of peripheral blood, with the extracted DNAmodified by bisulphite treatment and SNRPN exon 1amplified by PCR [24]. An abnormal methylationpattern was present with only the maternal bandvisible (data not shown).

3.3. FISH

To more accurately define the derivative (22) chro-mosome, several hybridisations were performed onmetaphase spreads obtained from suspensions ofperipheral blood remaining after the cytogenetic har-vest. Use of probes GABRB3 and PML (Oncor)(both yellow signals) showed deletion for GABRB3,but not PML [n=25] (figure 3a). Dual hybridisationwith SNRPN/PML (Oncor) (green signals) and TU-PLE1/22q13.3 (Vysis) (red signals) showed deletionof SNRPN with PML intact, and 22q13.3 signalsintact on the normal chromosome 22 and within theder(22) [n=12] (figure 3b). Dual hybridisation withTUPLE/ARSA (Vysis) showed both to be intact[n=10] (figure 4). Two colour FISH, used with two

A. Smith et al. / Ann. Genet. 43 (2000) 125–130128

Figure 4. Metaphase spread with TUPLE/ARSA showing ARSA(green signal) intact on the der(22).

Figure 5. Two colour FISH with probes from within the PWS/ASregion (a) YAC9; (b) TYAC19. Both are deleted. Two colourFISH with probes from outside the PWS/AS region (c) TYAC10(15q21); (d) cos 15-5(15q22). Both non deleted. (e) pantelomere,deleted within the der(22) chromosome.

3.4. Interpretation of structure of der(22)

The unbalanced karyotype has arisen from loss ofthe small reciprocal product (15pter�15q13�22qtel) of the translocation. This most likely oc-curred as a 3:1 meiotic segregation error from abalanced translocation, paternal in origin. Sum-marising the cytogenetic and probe results, we couldnot precisely define the extent of the deletion. Forthe region at the border of 15q13-q14, we estimate�1 Mb, based on previous knowledge of the PWScritical region — 15q11-13 — being 4.5 Mb [4, 5,14, 21]. For the region 22q13.3 to pantelomere,�300 kb (figure 6) could be deleted (based on thedistance of ARSA to 22qtel in the Vysis data sheet).It is virtually impossible to obtain accurate compari-son of distances for probe regions when the probesare obtained from different sources — particularlyfor the subtelomeric probes, which of necessity con-tain ‘some’ specific genes [7, 8, 13].

4. Discussion

The severe phenotype in our patient could be dueto a deletion of 15q larger than usual. PWS/ASdeletions are well defined and within 15q11-13, ex-tend distally from D15S541 to D15S12, a region of4.5 Mb [3, 4, 9, 14, 19]. A deletion greater than 4.5

A. Smith et al. / Ann. Genet. 43 (2000) 125–130 129

Figure 6. Ideogram showing possible areas of deletion on 15q and 22q (within box) on the der(22). Note GB3=GABRB3.

Mb of DNA on chromosome 15q may have been thecause of the additional features in a male infant withPWS with a large deletion (to 15q15) [16]. We couldnot exclude a larger than usual deletion in ourpatient with the probes tested, extending possiblyfrom within or at the distal border of 15q13 toproximal border or within 15q14 — a very smallregion, estimated at �1 Mb (figure 6). Patients withsuch a small putative deletion have not been re-ported and associated clinical features are notknown.

In cases with translocation, such as ours, imbal-ance of the reciprocal chromosome involved in thetranslocation could be another cause of the severephenotype. We examined the chromosome 22 termi-nal region and could not equivocally exclude a verysmall deletion of 22q from 22q13.3 (ARSA) to thenonspecific pantelomere. Telomere attachments arerare in man, but have been reported in 7/10 othercases of unbalanced translocations in PWS investi-gated with FISH [6, 12, 15]. In our case, there is notelomeric attachment but a loss of chromosomalmaterial, presumably the reciprocal product of thetranslocation (consisting of 15pter�15q13�22tel).The clinical features of distal 22q deletions, from22q13 to 22qter appear to be variable but consistentfeatures reported are the lack of speech and severeretardation [7, 8, 13, 20] — features present in ourpatient, additional to the PWS phenotype.

A mutation elsewhere could also cause additionalor very severe features. Our patient had a number ofphysical problems, e.g. bowel symptoms, UTI and itis difficult to know if they represent an incidentalillness in a patient with PWS or a rare manifestation.We could not arrive at a conclusion on this. It seemslikely that the terminal del(22q) microdeletion has

made a significant contribution to the severity of thephenotype in our patient.

We have given a detailed clinical account of ourpatient, to highlight our difficulty with adequatelydefining her status. She had all the salient features ofPWS, but she was more mentally and physicallyretarded than expected in PWS, with low IQ testresults, no speech, could not read or play, with lackof interest in her surroundings and poor progresswith toiletting. Patients with PWS usually have mildmental retardation, good communication skills, nor-mal toiletting ability and are interested in their sur-roundings. Thus, while she had the maximum scorefor PWS [11], this assessment misrepresented herfunctioning. When the score is adequate to diagnosePWS, extreme features may overlap with other con-ditions, e.g. severe mental retardation with thefragile X [18] or matUPD14 [10], tall stature withSotos syndrome [3] or lead to a second diagnosis,e.g. very short stature in Blooms syndrome [23]. Tocover unusual patients, who have been called ‘severePWS, atypical PWS, extended PWS or expandedPWS’, a definition of terms is required, which in-cludes severity of features and the presence of addi-tional features.

Acknowledgements

Suna Husyn for the cytogenetic analysis. BruceBennetts and J. Jackson for the methylation test; DrKamath for the abdominal investigations; HelenDonis-Keller (Washington University of St. Louis,MO) for providing the TYAC clones; BernhardHorsthemke (University of Essen, Germany) forproviding YAC9.

A. Smith et al. / Ann. Genet. 43 (2000) 125–130130

References

[1] Berends M.J.W., Hordijk R., Scheffer H., Oosterwijk J.C.,Halley D.J.J., Sorgedrager N., Two cases of maternal uniparentaldisomy 14 with phenotype overlapping with the Prader–Williphenotype, Am. J. Med. Genet. 84 (1999) 76–79.

[2] Bray G.A., Dahms W.T., Swerdloff R.S., Fiser R.H., Atkin-son R.L., Carrel R.E., The Prader–Willi syndrome: A study of 40patients and review of the literature, Medicine 62 (1983) 59–79.

[3] Buchholz T., Jackson J., Robson L., Smith A., Evaluationof methylation analysis for diagnostic testing in 258 referralssuspected of Prader–Willi or Angelman syndromes, Hum. Genet.103 (1998) 535–539.

[4] Buiting K., Saitoh S., Gross S., Dittrich B., Schwartz S.,Nicholls R.D., Horsthemke B., Inherited microdeletions in theAngelman and Prader–Willi syndromes define an imprinting cen-tre on human chromosome 15, Nat. Genet. 9 (1995) 395–400.

[5] Butler M.G., Prader–Willi syndrome: current understand-ing of cause and diagnosis, Am. J. Med. Genet. 35 (1990) 319–332.

[6] Devriendt K., Petit P., Matthijs G., Vermeesch J.R.,Holvoet M., De Muelenaere A., Marynen P., Cassiman J.-J.,Fryns J.-P., Trisomy 15 rescue with jumping translocation ofdistal 15q in Prader–Willi syndrome, J. Med. Genet. 34 (1997)395–399.

[7] Doheny K.F., McDermid H.E., Harum K., Thomas G.H.,Raymond G.V., Cryptic terminal rearrangement of chromosome22q13.32 detected by FISH in two unrelated patients, J. Med.Genet. 34 (1997) 640–644.

[8] Flint J., Wilkie A.O.M., Buckle V.J., Winter R.M., HollandA.J., McDermid H.E., The detection of subtelomeric chromoso-mal rearrangements in idiopathic mental retardation, Nat. Genet.9 (1995) 132–139.

[9] Gillessen-Kaesbach G., Gross S., Kaya-Westerloh S., Pas-sarge E., Horsthemke B., DNA methylation based testing of 450patients suspected of having Prader–Willi syndrome, J. Med.Genet. 32 (1995) 88–92.

[10] Greenswag L., Adults with Prader–Willi syndrome: Asurvey of 232 cases, Dev. Med. Child. Neurol. 29 (1987) 145–152.

[11] Holm V.A., Cassidy S.B., Butler M.G., Hanchett J.M.,Greenswag L.R., Whitman B.Y., Greenberg F., Prader–Willisyndrome: Consensus diagnostic criteria, Pediatrics 91 (1993)398–402.

[12] Jauch A., Robson L., Smith A., Investigations withfluorescence in situ hybridisation (FISH) demonstrate loss of thetelomeres on the reciprocal chromosome in three unbalancedtranslocations involving chromosome 15 in Prader–Willi andAngelman syndrome, Hum. Genet. 96 (1995) 345–349.

[13] Nesslinger N.J., Gorski J.L., Kurczynski T.W.,Shapira S.K., Siegel-Bartelt J., Dumanski J.P., Cullen R.F.,French B.N., McDermid H.E., Clinical, cytogenetic, andmolecular characterization of seven patients with deletionsof chromosome 22q13.3, Am. J. Hum. Genet. 54 (1994)464–472.

[14] Nicholls R.D., New insights reveal complex mechanismsinvolved in genomic imprinting, Am. J. Hum. Genet. 54 (1994)733–740.

[15] Park V.M., Gustashaw K.M., Wathen T.M., The presenceof interstitial telomeric sequences in constitutional chromosomeabnormalities, Am. J. Hum. Genet. 50 (1992) 914–923.

[16] Pauli R.M., Meisner L.F., Szmanda R.J., ExpandedPrader–Willi syndrome in a boy with an unusual 15q chromo-some deletion, Am. J. Dis. Child. 137 (1983) 1087–1089.

[17] Prader A., Labhart A., Willi H., Ein Syndrom vonAdipositas, Kleinwuchs, Kryptorchismus und Oligophrenie nachmyatonieartigem Zustand im Neugeborenalter, SchweizerischeMedizinische Wochenschrift 86 (1956) 1260–1261.

[18] Schrander-Stumpel C., Gerver W.J., Meyer H., Engelen J.,Mulder H., Fryns J.-P., Prader–Willi-like phenotype in fragile Xsyndrome, Clin. Genet. 45 (1994) 175–180.

[19] Schulze A., Hansen C., Skakkebaek N.E., Brondum-Nielsen K., Ledbetter D.H., Tommerup N., Exclusion of SNRPNas a major determinant of Prader–Willi syndrome by a transloca-tion breakpoint, Nat. Genet. 12 (1996) 452–454.

[20] Slavotinek A., Maher E., Gregory P., Rowlandson P.,Huson S.M., The phenotypic effects of chromosome rearrange-ment involving bands 7q21.3 and 22q13.3, J. Med. Genet. 34(1997) 857–861.

[21] Smith A., Den Dulk G., Lipson A., Suter M., ClassicalPrader–Willi Syndrome with trisomy 15 (pter(q12) plus de novovariant 15p11, Ann. Genet. 32 (1989) 39–42.

[22] Smith A., Robson L., Neumann A., Mulcahy M., ChabrosV., Deng Z.-M., Woodage T., Trent R.J., Fluorescence in-situhybridisation and molecular studies used in the characterisation ofa Robertsonian translocation (13q15q) in Prader–Willi syndrome,Clin. Genet. 43 (1993) 5–8.

[23] Woodage T., Deng Z.-M., Prasad M., Smart R., LindemanR., Christian S.L., Ledbetter D.H., Robson L., Smith A., TrentR.J., A variety of genetic mechanisms are associated with thePrader–Willi syndrome, Am. J. Med. Genet. (Neuropsychiat.Genet.) 54 (1994) 219–226.

[24] Zeschnigk M., Lich C., Buiting K., Doerfler W.,Horsthemke B., A single tube PCR test for the diagnosis ofAngelman and Prader–Willi syndrome based on allelic methyla-tion differences at the SNRPN locus, Eur. J. Hum. Genet. 5(1997) 94–98.