Paternally Derived De Novo Interstitial Duplicationof Proximal 15q in a Patient WithDevelopmental Delay

T.K. Mohandas,1 Jonathan P. Park,1 Richard A. Spellman,1 James J. Filiano,2Alexander C. Mamourian,3 Arnold B. Hawk,1 Dorothy R. Belloni,1 Walter W. Noll,1 andJohn B. Moeschler1,4*1Department of Pathology, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire2Department of Pediatrics, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire3Department of Radiology, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire4Center for Genetics and Child Development, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire

Interstitial duplications of proximal 15qcontaining the Prader-Willi syndrome/Angelman syndrome (PWS/AS) region havebeen found in patients with autism or atypi-cal autism. In these cases with an abnormalphenotype, the duplications were mater-nally derived. Paternal origin of the dupli-cation has been associated with a normalphenotype. We report on a patient who pre-sented with nonspecific developmental de-lay and partial agenesis of the rostral cor-pus callosum. Fluorescence in situ hybrid-ization (FISH) studies using probes specificfor the PWS/AS region demonstrated adouble signal on one chromosome 15, indi-cating the presence of an interstitial dupli-cation of proximal 15q involving the PWS/AS region in the patient. Parental chromo-somes were normal with FISH studies.Methylation analysis at exon alpha of theSNRPN locus showed a maternal band at 4.2kb and a paternal band of apparent doubleintensity at 0.9 kb, suggestive of one copy ofthe maternal allele and two copies of the pa-ternal allele in the patient. Microsatelliteanalysis was informative at the GABRB3 lo-cus in the family, which showed the inheri-tance of two different paternal alleles and amaternal allele in the patient consistentwith the origin of this duplication from anunequal crossing over between the twochromosome 15 homologs in the father. Thisis the first report of an abnormal phenotypeassociated with a paternally derived dupli-

cation of proximal 15q shown to contain thePWS/AS region by molecular techniques.Am. J. Med. Genet. 82:294–300, 1999.© 1999 Wiley-Liss, Inc.

KEY WORDS: agencies of the corpus callo-sum; autism; mental retarda-tion; communication disor-der; chromosome 15; duplica-tion 15q

INTRODUCTION

Recent studies have shown that proximal 15q is achromosomal region predisposed to structural rear-rangements. About 70% of patients with the Prader-Willi (PWS) or Angelman (AS) syndrome result from ade novo interstitial deletion in 15q11-q13 that spansapproximately 4 Mb in the paternally or maternallyderived chromosome 15, respectively [Lalande, 1996].The proximal and distal breakpoints of these deletionsare clustered, suggesting that they originate from un-equal exchanges between sister or nonsister chroma-tids [Robinson et al., 1998; Carrozzo et al., 1997]. Inaddition, 15q11-q13 region is also associated with thegeneration of inverted duplications, seen as bisatellitedsupernumerary marker chromosomes. The breakpointsleading to these inv dup(15) chromosomes as well asPWS/AS deletions are clustered in the same intervals,indicating the presence of ‘‘hotspots’’ for chromosomebreakage in proximal 15q [Huang et al., 1997]. An un-equal exchange resulting in an interstitial deletion isexpected to produce a reciprocal interstitial duplica-tion. Given the estimated incidence of the common de-letion involving the PWS/AS region at around 1:15,000and assuming no survival disadvantage, there shouldbe an equal number of individuals with a duplicationinvolving the PWS/AS region.

The phenotype associated with the duplications,however, has not been appreciated until recently. Ear-

*Correspondence to: John B. Moeschler, M.D., Center forGenetics and Child Development, Dartmouth-Hitchcock Medi-cal Center, Lebanon, NH 03767. E-mail: john.moeschler@dartmouth.edu

Received 11 May 1998; Accepted 29 October 1998

American Journal of Medical Genetics 82:294–300 (1999)

© 1999 Wiley-Liss, Inc.

lier cytogenetic investigations noted the presence ofpossible interstitial duplications in proximal 15q; how-ever, lack of molecular characterization of these dupli-cations made karyotype-phenotype correlation diffi-cult. Using molecular reagents for fluorescence in situhybridization (FISH) and microsatellite analysis, re-cent studies have shown that familial and de novo in-terstitial duplication of 15q encompassing the PWS/ASregion are associated with developmental delay andautistic behavior [Bundey et al., 1994; Cook et al.,1997; Browne et al., 1997; Schroer et al., 1998]. More-over, duplications associated with an abnormal pheno-type were maternally derived. In five of the families,the propositus’ mothers or grandmothers who were car-riers of the duplication and who inherited it from theirfathers were normal [Cook et al., 1997; Browne et al.,1997; Schroer et al., 1998].

We report on a patient who presented with nonspe-cific developmental delay and partial agenesis of therostral corpus callosum and was found to have an in-terstitial duplication of proximal 15q involving thePWS/AS critical region. The finding of abnormal callo-sal development has not been reported, heretofore, inpatients with proximal 15q duplication. Further, wedemonstrate that the duplication in this patient is pa-ternal in origin.

SUBJECT AND METHODSCase Report

Patient H.M., a boy, first presented to GeneticsClinic at age 2 years and 4 months with long-standingdevelopmental delay but acutely worsening behaviorssuggesting ‘‘neurodegeneration’’ to the primary carepediatrician. Specifically, the patient was noted toseem ‘‘fearful with any new experience,’’ ‘‘afraid to go tobed,’’ and to exhibit screaming and aggressive out-bursts. H.M. was first noted to have developmental de-lays by age six months when he was not vocalizing.According to his parents, his smile, while present atage 1 month, was ‘‘not normal’’ and he ‘‘never laughed.’’By age 15 months he was babbling and repeatingsounds in imitation, which he lost shortly thereafter.By age 20 months, he used simple gestures to expressneeds and desires and seemed to understand one-stepdirections. By 24 months, he used simple hand ges-tures for communication, but no verbalizations orwords. By age 28 months, he had learned to use 30signs for words, but still no speech. His motor develop-ment had been entirely normal.

H.M’s past medical history was otherwise unremark-able. This was the first pregnancy for his mother andfather who were 22 and 31 years old, respectively, atthe time of his birth. His mother took terbutaline forpremature labor from 30 weeks forward. Birth was anuncomplicated vaginal vertex delivery with vacuum ex-traction. Birth weight was 8 lbs., 4 oz.; length was 22.5inches; head circumference was 36 cm. Apgar scoreswere 8 and 9 at 1 and 5 minutes, respectively. Familyhistory was notable for mental illness. The patient’smother was on antidepressant medication, althoughshe had not had a psychiatric consultation. Her fatherand sister were known to have a diagnosis of depres-sion and were under psychiatric care.

H.M’s growth was entirely normal with height,weight, and head circumference following the growthcurves at approximately the 90th centile. He did have arapid weight gain between ages 12 and 24 months. Hisphysical findings including morphological and neuro-logical status, were entirely normal except for his se-verely delayed vocal output (Fig. 1). Formal develop-mental evaluation could not be completed because ofhis refusal to interact with the examiner. The patientwas noted to imitate actions but not sounds or words.We concluded that H.M. presented clinically with a‘‘nonspecific mental retardation syndrome’’ withtroublesome behaviors and with no signs of ‘‘neurode-generation.’’ He did not present as a child with autismor a pervasive developmental disorder. Clinically, hedid not have Angelman syndrome or Prader-Willi syn-drome. The patient was treated with Prozact and Risp-eradolt with a resultant diminution of the behavioralproblems.

Cytogenetics and Fluorescence inSitu Hybridization

Chromosome preparations from peripheral blood cul-tures and cytogenetic analyses were performed usingstandard techniques. Digoxigenin-labeled probes spe-cific for the PWS/AS critical regions [D15S10, D15S11,GABRB3, and SNRPN (Oncor, Gaithersburg, MD)]were hybridized to chromosome preparations fromH.M. and his parents. FISH hybridization and detec-tion were performed under conditions described previ-ously [Park, 1996].

Methylation Analysis

Analysis of differential methylation at exon alpha ofSNRPN [Sutcliffe et al., 1994] was carried out as de-scribed by Kubota et al. [1996]. Briefly, 5 mg of DNA

Fig. 1. The propositus at age 28 months. Note the lack of significantminor anomalies.

dup(15q) 295

isolated from the lymphocytes of H.M. and his parentswere digested with XbaI (New England Biolabs, Bev-erly, MA) and the methylation-sensitive enzyme NotI(New England Biolabs), electrophoresed through a0.8% agarose gel, and transferred to a Zetabind (Cuno)membrane. Hybridizations were performed with a 0.9-kb NotI fragment (phagemid clone pNO.9 (ATCC #95678)) specifying exon alpha of SNRPN. This probedetects a 4.2-kb band from the methylated maternalallele and a 0.9-kb band from the unmethylated pater-nal allele [Kubota et al., 1996].

Microsatellite Analysis

Four polymorphic microsatellite repeats associatedwith the loci, D15S11, D15S113, GABRB3, andGABRA5 from within the PWS/AS critical region [Mu-tirangura et al., 1993] were used to further examinethe duplication in H.M.. Primers were obtained forthese loci (Research Genetics, Huntsville, AL), andpolymerase chain reaction (PCR) assays were carriedout using published methods [Mutirangura et al.,1993]. Specifically, the forward primer (10 pmol) wasend-labeled for 20 min at 37°C with 10 U T4 poly-nucleotide kinase (Boehringer Mannheim, Indianapo-lis, IN) and 10 mCi gamma 32pATP (3,000 Ci/mmol)(NEN). The PCR reactions contained 50 ng of genomicDNA, 10 mM Tris, pH 8.3, 50 mM KCl-PE Buffer 2(PE/Applied Biosystems), 100 mM each dNTP (Boe-hringer Mannheim), 1.5 mM MgCl2 (PE/Applied Bio-systems), 10 pmol each of forward (labeled) and reverseprimers, and 1 U Amplitaq (PE/Applied Biosystems,Forster City, CA). Initial denaturation was at 94°C for2 min followed by 35 cycles of denaturation at 94°C for30 sec, annealing at 55°C for 30 sec, extension at 72°Cfor 30 sec, with a final extension at 72°C for 2 min. Tenml of product was mixed with 5 ml formamide-loadingbuffer and heated at 95°C for 5 min. Five ml of thismixture was loaded on a 6% acrylamide/7 M urea geland ran at 55 watts. The dried gel was exposed to Fujimedical X-ray film for 2 hr at −80°C with two intensi-fying screens.

RESULTSCytogenetic and FISH Studies

Patient’s weight gain prompted chromosome andFISH studies to rule out PWS, although clinically hedid not have the characteristic features of the syn-drome. Four probes specific for the PWS region(D15S10, D15S11, GABRB3, and SNRPN) were usedfor FISH studies on H.M., whereas SNRPN alone wasused to study his parents. All four probes showed eithera significantly enlarged signal or a merged doublesignal on the proximal long arm of one chromosome 15homolog in each cell analyzed on H.M. Results ofGABRB3 hybridization are shown in Fig. 2A. TheSNRPN signal on metaphases of H.M.’s parents ap-peared normal (data not shown). High-resolution G-banded karyotype of H.M. was consistent with thepresence of a duplication on one chromosome 15 (Fig.2B), and the parental karyotypes were within normallimits.

Methylation Analysis

Methylation in exon alpha of SNRPN was examinedin the M family. Both a paternal (0.9-kb) and a mater-nal (4.2-kb) allele were observed in H.M.’s parents (Fig.3). However DNA from H.M. showed both a maternalband at 4.2 kb as well as a paternal band of apparentdouble intensity at 0.9 kb, suggestive of one copy of thematernal allele and two copies of the paternal allele.

Microsatellite Analysis

Of the four loci tested from within the commonly de-leted region in PWS/AS, only the microsatellite repeatassociated with the GABRB3 locus was found to beinformative in this family. Results of this analysis areshown in Fig. 4. H.M. inherited three alleles desig-nated ‘a,’ ‘c,’ and ‘d.’ Allele ‘d’ is of maternal origin,whereas both alleles ‘a’ and ‘c’ could only have beencontributed by the father, consistent with a duplicationof the PWS/AS critical region observed in FISH studiesand the presence of a heavier paternal band in themethylation analysis. Inheritance of two different pa-ternal alleles in H.M. is most consistent with the originof this duplication from an unequal crossing over be-tween the two chromosome 15 homologs in the father.

Additional Studies

In addition to the molecular and cytogenetic studiesdescribed above, H.M. was evaluated for the presenceof Fragile-X by DNA analysis and was found to be nor-mal. Serum amino acid and urinary quantitative or-ganic acid screens and electron spray tandem massspectrometry of blood (NeoGen Screening, Inc, Pitts-burgh, PA) were all negative.

Brain MR was abnormal (Figs. 5 and 6). Although aportion of the body of the corpus callosum was present,the genu, rostrum, and much of the splenium were ab-sent. The MR also demonstrated an area of heterotopicgray matter in the medial right hemisphere with masseffect indenting the superior aspect of the right lateralventricle.

DISCUSSION

Unequal exchanges leading to the formation of dele-tion and duplication chromosomes associated with dis-ease states have been observed in several regions of thehuman genome [Purandare and Patel, 1997]. Proximalchromosome 15q appears to be one such unstable re-gion. Deletions involving this region are known tocause Prader-Willi syndrome or Angelman syndrome,depending on the parent-of-origin of the deleted chro-mosome [Lalande, 1996]. Duplications of this same re-gion (15q11-q13) have been reported in apparently nor-mal individuals, in those with developmental delays, inpatients with autism, and on occasion, in individualswith PWS or AS [De France et al., 1984; Ludowese etal., 1991; Clayton-Smith et al., 1993; Baker et al., 1994;Jalal et al., 1994]. Because many patients inheritedsuch duplications from apparently normal parents, thefinding of a duplication was often interpreted as a nor-mal euchromatic variation. The duplications were also

296 Mohandas et al.

not characterized using molecular methods in theseearly investigations.

Recently, there have been reports of patients withtypical or ‘‘atypical’’ autism who have an interstitialduplication of 15q characterized using molecular tech-niques [Bundey et al., 1994; Cook et al., 1997]. Morespecifically, only those patients who had duplicationsthat contain the PWS/AS critical region manifestedthis clinical phenotype [Bundey et al., 1994; Cook et al.,1997; Browne et al., 1997; Woods et al., 1997; Repettoet al., 1998; Schroer et al., 1998], whereas duplicationsof proximal 15q that do not contain the PWS/AS criticalregion appear to be euchromatic variants [Browne etal., 1997]. In the most extensive study published todate, Browne et al. [1997] identified proximal 15q du-plications in 15 patients among those referred for cy-togenetic analysis over a 14-year period. Molecular andFISH studies showed that patients with interstitial du-

plications containing the PWS/AS critical region haddevelopmental delay as well as learning and/or speechdifficulties in the absence of minor anomalies. Thesepatients were not described as autistic. Individualswith duplications that did not include the PWS/AS re-gion, however, had no consistent clinical phenotype.Parents and other family members of this latter groupof patients who also had the duplication were unaf-fected, and the patients inherited the duplications fromtheir mother or father.

These molecular studies also showed that there is aparent of origin effect for the 15q duplication contain-ing the PWS/AS critical region. Patients who had theclinical phenotype either inherited the duplicationfrom a carrier mother or the duplication was derived denovo from maternal chromosomes [Bundey et al., 1994;Cook et al., 1997; Browne et al., 1997; Woods et al.,1997; Repetto et al., 1998; Schroer et al., 1998]. All

Fig. 2. A: Metaphases showing hybridization of GABRB3 and control probe PML on chromosome 15 homologs. Note the double signal or merged doublesignal (arrow heads) in the proximal long arm of one homolog in each cell. B: G-banded chromosome 15 homologs from H.M. showing duplication (arrow)of the PWS/AS region.

dup(15q) 297

individuals with a maternally derived duplication con-taining the PWS/AS critical region manifested theclinical phenotype. Five of the mothers or grandmoth-ers of propositus in these families who were carriers ofthe duplication and who, in turn, inherited the dupli-cation from their fathers were normal. Mothers of pro-positus who inherited the duplication from their moth-ers, however, were affected [Browne et al., 1997; Cooket al., 1997; Schroer et al., 1998].

The patient reported here, H.M., presented with de-velopmental delays and a specific communication dis-order (lack of speech, but the ability to sign). He alsopresented with problematic behaviors characterized byfrequent and extreme outbursts of crying and refusals.However, he did not present as a child with a pervasivedevelopmental disorder. Of additional interest, the MRexam demonstrated an abnormal corpus callosum. Us-ing the mamillary body-anterior commissure line (the‘‘MAC line’’) as an anatomic marker for the division ofgenu and body, the genu of the corpus callosum is com-pletely absent. Although much of the existing litera-ture has suggested that the corpus callosum formsfrom anterior to posterior, Kier and Truwit [1996] pro-vided evidence that the callosum develops both anteri-orly and posteriorly from the anterior body, ie, bidirec-tionally from the MAC line. This case would provideadditional support for that contention with the caudal-to-rostral development of the genu having failed to oc-

cur properly. Abnormal callosal development has notbeen reported in patients with proximal 15q duplica-tion. Whether this developmental anomaly is a markerfor 15q duplications remains to be clarified by addi-tional delineation of this syndrome.

The clinical findings in H.M., who has a de novo in-terstitial duplication on a paternal chromosome 15, arenot in keeping with the findings reported previously.As noted above, five females who had a paternally de-rived duplication containing the PWS/AS critical re-gion were found to be normal. It is possible that thebreakpoints generating the 15q duplications, althoughclustered in proximal 15q, are variable at the molecu-lar level resulting in dosage changes for a different setof genes in the case reported here. The presence of mul-tiple repeat sequences in proximal 15q could accountfor breakpoint variability in these rearrangements.Robinson et al. [1998] have reported variability in thebreakpoints at the molecular level leading to 15q du-plications in the cases they studied. Alternatively, theclinical findings in H.M. may not be causally related tothe duplication; however, these finding raise the ques-tion whether all paternal duplications are benign withrespect to phenotype.

It is of interest that the phenotypic effects of theduplication 15q are somewhat similar to those of

Fig. 4. Microsatellite analysis at the GABRB3 locus in H.M. and hisparents. H.M. has inherited alleles ‘a’ and ‘c’ from his father and allele ‘d’from his mother.

Fig. 3. Methylation analysis at exon alpha of the SNRPN locus in pa-tient H.M. and his parents. The same amount of DNA (5 mg) was digestedand electrophoresed in each case. Note the increased intensity of the pa-ternal chromosome 15 specific band in H.M.

298 Mohandas et al.

Prader-Willi syndrome and Angelman syndrome, con-sisting of developmental and behavioral anomalieswith speech development much more affected thannonverbal intelligence. Additionally, the phenotype as-sociated with the presence of inverted duplications of15q that involve the PWS/AS critical region includesdevelopmental delay, abnormal speech, hypotonia, sei-zures, and behavior disturbances described in somecases as autistic [Schroer et al., 1998]. These invertedduplications seen as bisatellited supernumerarymarker chromosomes result in tetrasomy for proximal15q and are also maternal in origin [Cheng et al., 1994;Mignon et al., 1996].

The phenotype of 15q duplication is still emergingand, as delineated thus far, is one of subtle anomalies.The presentation appears to be one of the ‘‘nonspecificdevelopmental delay’’ conditions [Curry et al., 1997],

manifesting significant expressive language disabilityor autism (both typical and atypical) without a patternof major or minor anomalies and with normal growthparameters. Patients presenting with such findings, inwhom other studies (eg, standard karyotype, Fragile-Xstudies) are negative, are candidates for evaluation of15q for duplication involving the PWS/AS region.Based on the findings reported here, it appears thatadditional delineation of the 15q duplication pheno-type, including parent of origin differences, is needed.The availability of FISH and molecular methods shouldfacilitate the delineation of the phenotype associatedwith this chromosome anomaly as well as its preva-lence.

REFERENCES

Baker P, Piven J, Schwartz S, Patil S. 1994. Brief report: duplication ofchromosome 15q11-13 in two individuals with autistic disorder. J Au-tism Dev Disord 24:529–535.

Browne CE, Dennis NR, Maher E, Long FL, Nicholson JC, Sillibourne J,Barber JCK. 1997. Inherited interstitial duplications of proximal 15q:genotype-phenotype correlations. Am J Hum Genet 61:1342–1352.

Bundey S, Hardy C, Vickers S, Kilpatrick MW, Corbett JA. 1994. Dupli-cation of the 15q11-13 region in a patient with autism, epilepsy andataxia. Dev Med Child Nerol 36:736–742.

Carrozzo R, Rossi E, Christian SL, Kittikamron K, Livieri C, Corrias A,Pucci L, Fois A, Simi P, Bosio L, Beccaria L, Zuffardi O, Ledbetter DH.1997. Inter- and intrachromosomal rearrangements are both involvedin the origin of 15q11-q13 deletions in Prader-Willi syndrome. Am JHum Genet 61:228–231.

Cheng SD, Spinner NB, Zackai EH, Knoll JHM. 1994. Cytogenetic andmolecular characterization of inverted duplicated chromosomes 15from 11 patients. Am J Hum Genet 55:753–759.

Clayton-Smith J, Webb T, Cheng XJ, Pembrey ME, Malcolm S. 1993. Du-plication of chromosome 15 in the region 15q11-13 in a patient withdevelopmental delay and ataxia with similarities to Angelman syn-drome. J Med Genet 30:529–531.

Cook EH, Lindgren V, Leventhal BL, Courchesne R, Lincoln A, Shulman C,Lord C, Courchesne E. 1997. Autism or atypical autism in maternallybut not paternally derived proximal 15q duplication. Am J Hum Genet60:928–934.

Curry CJ, Stevenson RE, Aughton D, Byrne J, Carey JC, Cassidy S, CunnifC, Graham JM Jr, Jones MC, Kaback MM, Moeschler JB, Schaefer GB,Schwartz S, Tarleton J, Opitz J. 1997. Evaluation of mental retarda-tion: Recommendations of a consensus conference: American College ofMedical Genetics. Am J Med Genet 72:468–477.

De France HF, Beemer FA, Ippel PF. 1984. Duplication of chromosome 15qin a boy with the Prader-Willi syndrome: further cytogenetic confu-sion. Clin Genet 26:379–382.

Huang B, Crolla JA, Christian SL, Wolf-Ledbetter ME, Macha ME, Pap-enhausen PN, Ledbetter DH. 1997. Refined molecular characterizationof the breakpoints in small inv dup(15) chromosomes. Hum Genet 99:11–17.

Jalal SM, Persons DL, Dewald GW, Lindor NM. 1994. Form of 15q proxi-mal duplication appears to be a normal euchromatic variant. Am J MedGenet 52:495–497.

Kier EL, Truwit CL. 1996. The normal and abnormal genu of the corpuscallosum: an evolutionary, embryologic, anatomic, and MR analysis.Am J Neroradiol 17:1631–1641.

Kubota T, Sutcliffe JS, Aradhya S, Gillessen-Kaesbach G, Christian SL,Horsthemke B, Beaudet AL, Ledbetter DH. 1996. Validation studies ofSNRPN methylation as a diagnostic test for Prader-Willi syndrome.Am J Med Genet 66:77–80.

Lalande M. 1996. Parental imprinting and human disease. Ann Rev Genet30:173–195.

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

Mignon C, Malzac P, Moncla A, Depetris D, Roeckel N, Croquette MF,Mattei MG. 1996. Clinical heterogeneity in 16 patients with inv dup 15

Fig. 5. Sagittal T1 weighted scan (TR500TE17 slice thickness 5 mm)demonstrates the abnormal corpus callosum. Using the MAC line drawnbetween the mamillary bodies (solid arrow) and anterior commissure (openarrow), the entire genu is absent.

Fig. 6. Coronal T2 weighted scan (TR 4500TE120 slice thickness 4 mm)demonstrates the mass-like heterotopic gray matter of the right medialhemisphere (solid arrow) indenting the right lateral ventricle (open arrow).

dup(15q) 299

chromosome: Cytogenetic and molecular studies, search for an imprint-ing effect. Eur J Hum Genet 4:88–100.

Mutirangura A, Greenberg F, Butler MG, Malcolm S, Nicholls RD,Chakravarti A, Ledbetter DH. 1993. Multiplex PCR of three dinucleo-tide repeats in the Prader-Willi/Angelman critical region (15q11-q13):Molecular diagnosis and mechanism of uniparental disomy. HumMolec Genet 2:143–151.

Park JP. 1996. Shared synteny of human chromosome 17 loci in Canids.Cytogenet Cell Genet 74:133–137.

Purandare SM, Patel PI. 1997. Recombination hotspots and human dis-ease. Genome Res 7:773–786.

Repetto GM, White LM, Bader PJ, Johnson D, Knoll JHM. 1998. Intersti-tial duplications of chromosome region 15q11q13: clinical and molecu-lar characterization. Am J Med Genet 79:82–89.

Robinson WP, Dutly F, Nicholls RD, Bernasconi F, Penaherrera M, Micha-

elis RC, Abeliovich, Schinzel AA. 1998. The mechanisms involved information of deletions and duplications of 15q11-q13. J Med Genet11:130–136.

Schroer RJ, Phelan MC, Michaelis RC, Crawford EC, Skinner SA, CuccaroM, Simensen RJ, Bishop J, Skinner C, Fender D, Stevenson RE. 1998.Autism and maternally derived aberrations of chromosome 15q. Am JMed Genet 76:327–336.

Sutcliffe JS, Nakao M, Christian S, Orstavik KH, Tommerup N, LedbetterDH, Beaudet AL. 1994. Deletions of a differentially methylated CpGisland at the SNRPN gene define a putative imprinting control region.Nat Genetics 8:52–58.

Woods CG, Robinson M, Gardiner C, Roussounis T. 1997. Four cases of thecondition described by Bundey with autism and a maternal 15q11-13duplication. Am J Hum Genet 61:A117.

300 Mohandas et al.