Fluorescence In Situ Hybridization DetectableMosaicism for Angelman Syndrome WithBiparental Methylation

Mustafa Tekin,1 Colleen Jackson-Cook,1 Arlene Buller,2 Andrea Ferreira-Gonzalez,2 Arti Pandya,1Carleton T. Garrett,2 and Joann Bodurtha1*1Department of Human Genetics, Virginia Commonwealth University/Medical College of Virginia,Richmond, Virginia

2Department of Pathology, Virginia Commonwealth University/Medical College of Virginia, Richmond, Virginia

We present a child with mild to moderateglobal developmental delay including se-vere speech impairment, inappropriatehappy demeanor, wide-based gait, frequentear infections with mild hearing loss, deep-set eyes, a wide mouth, widely-spaced teeth,normal head circumference, and no sei-zures. Results of peripheral blood lympho-cyte chromosomal analysis with GTG band-ing were normal. However, fluorescence insitu hybridization (FISH) studies showedmosaicism for a deletion of probes (D15S10and SNRPN) from the Angelman syndrome(AS) critical region with approximately 40%of peripheral lymphocytes having the dele-tion. The deleted chromosome 15 alsoshowed centromeric duplication, whichwas detected with a D15Z1 probe [46,XX,dic(15)(pter→q11.1::p11.2→q11.1::q13→qter)].The same duplication pattern was observedin 30% of the nuclei obtained from a buccalsmear. Methylation studies using polymer-ase chain reaction with sodium bisulfite-treated DNA demonstrated a normal bipa-rental methylation pattern. To the best ofour knowledge, this is the first case with ASand a FISH detectable deletion in a mosaicpattern. We recommend FISH studies forthe detection of mosaicism in the patientswith AS clinical findings even if results ofthe methylation studies are normal. Am. J.Med. Genet. 95:145–149, 2000.© 2000 Wiley-Liss, Inc.

KEY WORDS: Angelman syndrome; chro-mosome deletion; FISH; DNAmethylation; mosaicism

INTRODUCTION

In this report, we describe a child with clinical find-ings of Angelman syndrome (AS). The patient wasfound to have mosaicism for a deletion in Prader-Willisyndrome (PWS) critical region by fluorescence in situhybridization (FISH), whereas methylation studies re-vealed a normal biparental pattern.

CLINICAL REPORT

The patient is a 28-month-old Caucasian girl whowas born normally to a 35-year-old G4P2A1 mother at40 weeks of gestation. She has a healthy older sister.One brother died at age 11 days of congenital heartdisease reported to be coarctation of the aorta at au-topsy. Amniocentesis done in another laboratory be-cause of advanced maternal age showed normal fetalchromosomes. The patient’s birth weight, length, and

*Correspondence to: Joann Bodurtha, M.D., M.P.H., 1101 EastMarshall Street, Department of Human Genetics, Virginia Com-monwealth University, PO Box 980033, Richmond, VA 23298.

Received 3 February 2000; Accepted 5 June 2000 Fig. 1. Facial appearance of the patient.

American Journal of Medical Genetics 95:145–149 (2000)

© 2000 Wiley-Liss, Inc.

Fig

.2.

FIS

Hon

met

aph

ase

chro

mos

omes

from

lym

phoc

ytes

.T

wo

cell

lin

esw

ere

not

edin

the

lym

phoc

ytes

from

this

pati

ent.

Th

epr

omin

ent

lin

e(s

een

in60

%of

met

aph

ase

spre

ads

anal

yzed

)h

adsi

gnal

spr

esen

tfo

rS

NR

PN

(a)

and

D15

S10

(b)

(gre

ensi

gnal

sre

pres

ent

the

D15

Z1

prob

e;or

ange

sign

als

that

are

loca

ted

onth

epr

oxim

allo

ng

arm

ofch

rom

osom

e15

repr

esen

tth

eS

NR

PN

orD

15S

10pr

obes

;ora

nge

sign

als

that

are

loca

ted

toth

edi

stal

lon

gar

mof

chro

mos

ome

15re

pres

ent

the

con

trol

prob

eP

ML

).In

the

seco

nd

cell

lin

e,th

eS

NR

PN

(c)

and

D15

S10

(d)

loci

wer

ede

lete

d.A

lso,

two

sign

als

wer

epr

esen

tfo

rth

eD

15Z

1pr

obe.

head circumference (OFC) were normal. On the firstday of life she was noted to have three short periods ofrapid eye movements with unusual body posture. Dur-ing the examination she was found to have increasedmuscle tone, a prominent nose, and micrognathia.Electroencephalography (EEG), cranial ultrasonogra-phy, and echocardiography findings on the second dayof life were normal. Peripheral blood chromosomeswere normal using standard (GTG) banding. She didnot continue to have these unusual movements beyondthe first day of life. However, at age 6 months she wasconfirmed to have developmental delay. At age 28months, there was global delay worst in the areas ofexpressive (6–8 months) and receptive language (8–10months). Her gross motor and cognitive skills werefound to be at a 12- to 21-month level and 9- to 12-month level, respectively. She has had frequent middleear infections. A recent hearing evaluation with brain-stem auditory evoked responses and otoacoustic emis-sions demonstrated bilateral mild to moderate senso-rineural hearing loss. She was noted to be always in ahappy mood with frequent smiling but not exaggeratedlaughter episodes. Her current height, weight, andOFC are 91cm (75th centile), 13.5 kg (90th centile), and50 cm (50th centile), respectively. She has deep seteyes, mild ptosis of the left eyelid, wide mouth withwidely spaced teeth, prominent nose and nasal bridgewith round nasal tip, wide-based gait with the handsheld flexed, and mild ataxia (Fig. 1). Her hair, eye, andskin colors are slightly fairer than those observed inthe parents.

Since the patient had some findings suggestive of ASwe performed FISH studies to detect a possible dele-tion in the PWS/AS critical region. Peripheral lympho-cytes were obtained using standard culturing and har-vesting techniques. FISH was performed according tothe manufacturer’s protocol (Vysis Inc., DownersGrove, IL), using probes for D15S10 and SNRPN andtwo control probes: D15Z1 (localized to 15p11.2) andPML (localized to 15q22). A total of 21 and 22 meta-phases for D15S10 and SNRPN probes were scored,respectively. The results of both critical region probes(D15S10 and SNRPN) showed that the patient had mo-saicism for a deletion in approximately 40% of periph-eral blood lymphocytes (Fig. 2). The aberrant chromo-some 15 was also dicentric, having duplication of the15p11.2→15q11.1 region [46,XX, dic(15) (pter→q11.1::p11.2→q11.1::q13→qter)]. Buccal mucosa cellswere investigated for the presence of the same dupli-cation since confirmational studies with anothersample have been recommended in patients with mo-saicism for FISH detectable deletions [Nicholls, 2000].These studies demonstrated that 30% of the nuclei, ob-tained from buccal smears, had an aberrant FISH pat-tern (Fig. 3). The parents had normal chromosomes.

The patient was also evaluated for the presence of apaternal-only DNA methylation pattern. For methyl-ation-sensitive polymerase chain reaction (PCR) analy-sis the proposita’s DNA, obtained from peripheralblood cells, was modified by sodium bisulfite treatmentusing modifications to the protocol described by others[Hayatsu et al., 1982; Shapiro and Mohandas, 1983;

Fig. 3. FISH on interphase nuclei from buccal mucosa cells. Mosaicism for the aberrant chromosome 15 was scored by evaluating the number of nucleihaving two signals for D15Z1 (and four orange signals) compared with those having three signals or a large bipartite signal (and three orangesignals)(arrows).

Mosaicism for Angelman Syndrome 147

Clark et al., 1994; Herman et al., 1996]. Previouslydescribed primers for unmethylated (UNMET), meth-ylated (MET), and wild-type (WT) forms of the SNRPNgene were used for PCR reactions [Kosaki et al., 1997].Primers specific to methylated and unmethylated DNAwere multiplexed and the WT primers were processedin a separate reaction. The 50-mL reaction volume con-sisted of 0.1 vol of sodium bisulfite-treated DNA, 0.2mM dNTPs (Promega, Madison, WI), 2.5 units Ampli-Taq Gold (Perkin-Elmer, Foster City, CA), 3 mMMgCl2, 20 pmol each of WT-Forward and WT-Reverse,MET-Forward, MET-Reverse, UNMET-Forward, andUNMET-Reverse, 50 mM KCl, and 10 mM Tris-HCl(buffer II, pH 8.3; Perkin-Elmer). “Touch-down” PCRwas accomplished with the following conditions: 95°Cfor 10 min; 5 cycles of 67°C for 1 min and 72°C for 1min; 27 cycles of 65°C for 1 min and 72°C for 1 min witha final extension of 5 min at 72°C. PCR products wereseparated on a 3% agarose gel and visualized by ethid-ium bromide staining. The results revealed a normalbiparental methylation pattern (Fig. 4).

DISCUSSION

AS is caused by deficiency of expression of a generegion on the 15q11 to 15q13 bands of the maternalchromosome 15. The mechanisms leading to AS arevaried and include an interstitial deletion within theq11-q13 region in the maternally inherited chromo-some 15 in approximately 70% of affected individuals[Cassidy and Schwartz, 1998], paternal uniparental di-somy (UPD) with maternal deficiency for 15q11-15q13in 3 to 5% of individuals, imprinting mutations inwhich the maternal chromosome has a paternal meth-ylation pattern and imprinted gene expression for15q11-15q13 in 7 to 9% of patients, and loss-of-functionmutations in UBE3A (which maps within the criticalregion) in 4 to 8% of patients [Fang et al., 1999; Jianget al., 1999; Malzac et al., 1998a].

Most of the critical deletions can be detected by FISH

using only one probe, SNRPN [Cassidy and Schwartz,1998]. Almost all of the patients with a classical dele-tion, UPD, or imprinting mutations show a paternal-only DNA methylation pattern for chromosome 15q11-q13. Thus, the usage of methylation analysis as aninitial screening test for PWS or AS is well acceptedand efficient.

Several individuals with somatic mosaicism for aninterstitial chromosome deletion such as Smith-Magenis syndrome [Finucane et al., 1993], 22q11 dele-tion [Consevage et al., 1996], and PWS [Cassidy et al.,1984; Malzac et al., 1998b; Golden et al., 1999] havebeen reported. Mosaicism in PWS syndrome is rare,and only a few cases have been described. Cassidy et al.[1984] was the first to report mosaicism in PWS. Theydescribed two patients who were mosaic for a chromo-some 15 deletion, using high resolution banding tech-niques. The clinical manifestations of these two pa-tients did not differ from those of the rest of the patientpopulation they studied. Malzac et al. [1998b] reportedan additional case with typical PWS clinical findingsand 20% mosaicism for the deletion detected by FISHin peripheral blood lymphocytes. Recently, Golden etal. [1999] reported on another child with atypical PWSfindings and deletion of the SNRPN probe in 78% ofperipheral lymphocytes. To the best of our knowledge,our case is the first example of a patient with FISH-detectable mosaicism for a deletion of chromosome15q11-q13 and clinical findings of AS. There have beenfew examples of individuals with mosaicism for UBE3Agene mutations [Malzac et al., 1998a], and imprintingmutations [Saitoh et al., 1997]. However, these indi-

TABLE I. Comparison of the Clinical Findings Present inPatients With AS and Our Case’s Findings

Clinical findingsa Our patient

ConsistentSevere developmental delay +/−Severe speech impairment +Movement or balance disorder +Apparent happy demeanor and frequent

laughter/smiling +Frequent

Microcephaly −Seizures −Abnormal EEG −b

Less commonFlat occiput −Occipital groove −Protruding tongue −Tongue thrusting; suck/swallowing disorders −Feeding problems during infancy −Prognathia −Wide mouth, wide-spaced teeth +Frequent drooling +Excessive chewing/mouthing behaviors −Strabismus −Hypopigmented skin, light hair and eye color +/−Hyperactive lower limb deep tendon reflexes −Uplifted, flexed arm position +Increased sensitivity to heat −Sleep disturbance +Attraction to/fascination with water −

aAdapted from the diagnostic criteria for Angelman syndrome [Williams etal., 1995].bEEG was performed on the second day of life.

Fig. 4. The results of DNA methylation studies. Lane 1, 0X marker;lane 2, modified normal human placental DNA with MET/UNMET multi-plexed; lane 3, modified AS-positive cell line DNA (NIGMS, #GM11404)with MET/UNMET multiplexed; lane 4, modified patient DNA with MET/UNMET multiplexed. MET, 131 bp; UNMET, 164 bp.

148 Tekin et al.

viduals were unaffected mothers or grandfathers whotransmitted the mutation to the following generationswho were found to have AS. Although our patient hasseveral findings typical of AS, developmental delay ap-pears to be milder than that observed in typical cases.In classical AS, at least relative microcephaly and sei-zures commence by age 2 to 3 years, yet our patientcurrently does not have these findings. A comparison ofthe typical manifestations of AS and our patient’s find-ings is shown in Table I.

The use of traditional (GTG banding) cytogeneticand/or molecular genetic techniques did not detect themosaic deletion in this patient. Identification of a clini-cally important deletion by standard Giemsa bandingcytogenetic techniques has been shown to be compli-cated by several factors including chromosome length,deletion size, and proximal band heteromorphisms[Kaplan et al., 1987; Ludowese et al., 1991; Delach etal., 1994; Repetto et al., 1998; Riordan and Dawson,1998]. In this case, detection (using GTG banding) wasfurther complicated by mosaicism and the presence of aconcomitant duplication/deletion of pericentric chro-matin in the aberrant chromosome. The mechanism ofdeletion in our patient is different from the mechanismthat gives rise to typical nonmosaic AS deletions. Giventhe mosaicism in our case, the rearrangement mustoccurred during mitosis. Thus, one could speculate thatthe DNA sequences on proximal 15q that have beenrecognized as a “hotspot” for meiotic recombination andrearrangements also can misalign/exchange in mitosisto produce imbalances [Lichten and Goldman, 1995].

Methylation-sensitive PCR analysis documented anormal methylation pattern due to the presence of thenormal cell line. However, as depicted in Figure 4,there was reduced amplification of the methylated al-lele compared with the unmethylated allele for this pa-tient, which may have been due to the presence of mo-saicism. The pattern of unequal amplification ofmethylated and unmethylated alleles was reproduciblein further testing, whereas equal amplification was ob-served for normal controls (A. Butler, A. Ferreira-Gonzalez, and A. Pandya, unpublished data). Sincequantitative PCR with the appropriate internal con-trols was not used, the unusual band pattern was con-sidered a possible artifact and, therefore, the patientwas considered to have a biparental methylation pat-tern. Current methods to detect changes in the meth-ylation status of the PWS/AS region, including PCRand Southern hybridization-based methylation analy-ses, are not sufficiently quantitative to detect mosa-icism. However, further studies to rule out mosaicismmay be indicated if an unusual band pattern is ob-served using methylation-sensitive PCR.

Since methylation studies show a uniparental pat-tern in all individuals who have PWS or AS with aclassical deletion, UPD, or imprinting center muta-tions, generally FISH would be unnecessary in a pa-tient who has normal methylation pattern. However,on the basis of our findings and previous reports withmosaic PWS, FISH analyses should be performed in allcases with PWS or AS even if there is biparental meth-ylation when a patient has suggestive clinical findingsand/or unusual band patterns in methylation analysis.

REFERENCESCassidy SB, Schwartz S. 1998. Prader-Willi and Angelman syndromes:

disorders of genomic imprinting. Medicine (Baltimore) 77:140–151.Cassidy SB, Thuline HC, Holm VA. 1984. Deletion of chromosome 15 (q11-

13) in a Prader-Labhart-Willi syndrome clinic population. Am J MedGenet 17:485–495.

Clark SJ, Harrison J, Paul CL, Frommer M. 1994. High sensitivity map-ping of methylated cytosines. Nucleic Acids Res 22:2990–2997.

Consevage MW, Seip JR, Belchis DA, Davis AT, Baylen BG, Rogan PK.1996. Association of a mosaic chromosomal 22q11 deletion with hypo-plastic left heart syndrome. Am J Cardiol 77:1023–1025.

Delach JA, Rosengren SS, Kaplan L, Greenstein RM, Cassidy SB, BennPA. 1994. Comparison of high resolution chromosome banding andfluorescence in situ hybridization (FISH) for the laboratory evaluationof Prader-Willi syndrome and Angelman syndrome. Am J Med Genet52:85–91.

Fang P, Lev-Lehman E, Tsai TF, Matsuura T, Benton CS, Sutcliffe JS,Christian SL, Kubota T, Halley DJ, Meijers-Heijboer H, Langlois S,Graham JM Jr, Beuten J, Willems PJ, Ledbetter DH, Beaudet AL.1999. The spectrum of mutations in UBE3A causing Angelman syn-drome. Hum Mol Genet 8:129–135.

Finucane BM, Kurtz MB, Babu VR, Scott CI Jr. 1993. Mosaicism for de-letion 17p11.2 in a boy with the Smith-Magenis syndrome. Am J MedGenet 45:447–449.

Golden WL, Sudduth KW, Burnett SH, Kelly TE. 1999. Mosaicism inPrader-Willi syndrome: detection using fluorescent in situ hybridiza-tion. Am J Med Genet 85:424–425.

Hayatsu H, Yamashita Y, Yui S, Yamagata Y, Tomita K, Negishi K. 1982.N-Sulfomethylation of guanine, adenine and cytosine with formalde-hyde-bisulfite: a selective modification of guanine in DNA. Nucleic Ac-ids Res 10:6281–6293.

Herman JG, Graff JR, Myohanen S, Nelkin BD, Baylin SB. 1996. Methyl-ation-specific PCR: a novel PCR assay for methylation status of CpGislands. Proc Natl Acad Sci USA 93:9821–9826.

Jiang Yh, Lev-Lehman E, Bressler J, Tsai TF, Beaudet AL. 1999. Geneticsof Angelman Syndrome. Am J Hum Genet 65:1–6.

Kaplan LC, Wharton R, Elias E, Mandell F, Donlon T, Latt SA. 1987.Clinical heterogeneity associated with deletions in the long arm ofchromosome 15: report of 3 new cases and their possible genetic sig-nificance. Am J Med Genet 28:45–53.

Kosaki K, McGinniss MJ, Veraksa AN, McGinnis WJ, Jones KL. 1997.Prader-Willi and Angelman syndromes: diagnosis with a bisulfite-treated methylation-specific PCR method. Am J Med Genet 73:308–313.

Lichten M, Goldman ASH. 1995. Meiotic recombination hotspots. AnnuRev Genet 29:423–444.

Ludowese CJ, Thompson KJ, Sekhon GS, Pauli RM. 1991. Absence of pre-dictable phenotypic expression in proximal 15q duplications. ClinGenet 40:194–201.

Malzac P, Moncla A, Pedeillier K, Vo Van C, Girardot L, Voelckel MA.1998b. Atypical molecular findings identify limits of technical screen-ing tests for Prader-Willi and Angelman syndrome diagnoses. Am JMed Genet 78:242–244.

Malzac P, Webber H, Moncla A, Graham JM, Kukolich M, Williams C,Pagon RA, Ramsdell LA, Kishino T, Wagstaff J. 1998a. Mutationanalysis of UBE3A in Angelman syndrome patients. Am J Hum Genet62:1353–1360.

Nicholls RD. 2000. Mosaicism in Prader-Willi syndrome. Am J Med Genet90:175–176.

Repetto GM, White LM, Bader PJ, Johnson D, Knoll JH. 1998. Interstitialduplications of chromosome region 15q11q13: clinical and molecularcharacterization. Am J Med Genet 79:82–89.

Riordan D, Dawson AJ. 1998. The evaluation of 15q proximal duplicationsby FISH. Clin Genet 54:517–521.

Saitoh S, Buiting K, Cassidy SB, Conroy JM, Driscoll DJ, Gabriel JM,Gillessen-Kaesbach G, Glenn CC, Greenswag LR, Horsthemke B,Kondo I, Kuwajima K, Niikawa N, Rogan PK, Schwartz S, Seip J,Williams CA, Nicholls RD. 1997. Clinical spectrum and molecular di-agnosis of Angelman and Prader-Willi syndrome patients with an im-printing mutation. Am J Med Genet 68:195–206.

Shapiro LJ, Mohandas T. 1983. DNA methylation and the control of geneexpression on the human X chromosome. Cold Spring Harb SympQuant Biol 47:631–637.

Williams CA, Angelman H, Clayton-Smith J, Driscoll DJ, Hendrickson JE,Knoll JH, Magenis RE, Schinzel A, Wagstaff J, Whidden EM, Zori RT.1995. Angelman syndrome: consensus for diagnostic criteria: Angel-man Syndrome Foundation. Am J Med Genet 56:237–238.

Mosaicism for Angelman Syndrome 149