Psychological studies in Huntington's disease: making up the balance

Letters to the Editor

Presymptomatic testing in myotonic dystrophy:genetic counselling approaches

Siv Fokstuen, Jenny Myring, Christine Evans, Peter S Harper

EDITOR—We report the genetic counsellingapproaches used in a series of 72 presympto-matic genetic tests for myotonic dystrophyundertaken in our centre over an eight yearperiod. The study has identified factors whichinfluenced the counsellor’s approach, andwhich can provide a basis for further, moresystematic research.

Genetic counselling in myotonic dystrophyhas always been diYcult and complex, owing tothe extreme variability of the disorder, in bothseverity and age at onset, with anticipationbetween generations and influence of the sex ofthe aVected parent.

The identification of a CTG repeat expansionwithin the 3' untranslated region of the myotonicdystrophy protein kinase gene on chromosome191 as the primary molecular defect hastransformed our understanding of the geneticaspects of this disorder and provides the basis foran accurate and specific diagnostic and pr-esymptomatic test. The broad correlation of thesize of the CTG expansion with age at onset andseverity of the phenotype allows a limited degreeof prognosis to be given to those found to havethe mutation, particularly for very large or mini-mal gene expansions. Subjects carrying aminimal expansion (less than 100 repeats) usu-ally show few or no muscle symptoms,2 but maydevelop cataract in later life; they contribute to apool of mutation carriers who may transmitclinically significant disease to their oVspring asa result of anticipation.

Presymptomatic genetic testing for late onsetdominantly inherited disorders first becamepossible for Huntington’s disease (HD),3 forwhich extensive experience has resulted inwidely accepted guidelines for genetic counsel-ling protocols; these comprise two pre-testsessions for information and preparation com-bined with post-test support.4 With appropriateadaptation, this has become a model for otherlate onset genetic disorders of the nervous

system and to some extent also for the familialcancers. Although direct presymptomatic testingfor myotonic dystrophy has now been availablefor eight years, we are not aware of studies so faron the counselling approach for this disorder inrelation to presymptomatic testing. This lack ofknowledge has prompted the present study, inconjunction with experience of direct moleculartesting as a service in our centre, reported sepa-rately.5

Patients and methodsSince the identification of the specific genemutation in 1992, up to June 2000, out of atotal of 287 molecular analyses for the disorderin subjects living in Wales (population 2.9 mil-lion), there were 78 presymptomatic tests formyotonic dystrophy; by comparison 205 diag-nostic tests on symptomatic subjects were per-formed and four prenatal tests. The laboratorymethods used and the overall composition andoutcomes of the series are described in theaccompanying paper.5 For the purpose of thepresent study, we only included the 72 subjectsseen for presymptomatic testing by 11 diVerentclinical geneticists providing the clinical genet-ics service for Wales over this eight year period,based in the Institute of Medical Genetics,CardiV, the remaining six samples having beenreceived from neurologists and paediatricians.Data about the counselling approach wereascertained through the clinical genetics serv-ice notes and correspondence, supplementedby further discussion with the relevant clinicalgeneticist. We did not recontact any patients.

No single or specific counselling approach topresymptomatic testing in myotonic dystrophyhas been advocated in Wales over this period,though one of us (PSH) has had a long stand-ing clinical and research interest in thedisorder. In general, subjects requesting pre-symptomatic testing were seen in their localmedical genetics clinic by the clinical geneticistand genetics nurse specialist with designatedresponsibility for the particular district, not in aspecialist clinic for the disorder.

ResultsWhen grouped according to the perspective ofthe subject tested, three diVerent presympto-matic test situations could be distinguished(table 1). The largest group of subjects (A,n=58) represented the classical presympto-matic test situation with persons at 50% risk.

Table 1 Presymptomatic testing for myotonic dystrophy: principal indications

Presymptomatic test situation Total

Test results

Normal Abnormal

(A) Direct request for presymptomatic testing 58 48 10(B) Presymptomatic testing considered in relation to possible

testing of ongoing pregnancy 5 5 0(C) Testing of asymptomatic parent in relation to diagnostic

investigation of child 9 3 6Total 72 56 16

J Med Genet 2001;38:846–893846

J Med Genet2001;38:846–850

Institute of MedicalGenetics, University ofWales College ofMedicine, Heath Park,CardiV, CF4 4XN, UKS Fokstuen*J MyringC EvansP S Harper

Correspondence to:Professor Harper,[email protected]

*Present address: Division ofMedical Genetics, CentreMedicale Universitaire, 1Rue Michel-Servet, 1211Geneva 4, Switzerland,[email protected]

www.jmedgenet.com

group.bmj.com on July 15, 2011 - Published by jmg.bmj.comDownloaded from

http://jmg.bmj.com/

http://group.bmj.com/

Ten proved to have an abnormal result. Two ofthem showed minimal clinical abnormalities atthe time of testing, but did not regard theirsymptoms as abnormal. Patients with an ongo-ing pregnancy formed a separate subgroup (B,n=5). All five (four females, one male) werehealthy subjects at 50% risk asking for apresymptomatic test during an ongoing preg-nancy. Three of them wished for a prenataldiagnosis in case of an abnormal presympto-matic test result. Results proved normal in allof them. The third category (C) concernedpatients with no known family history of myo-tonic dystrophy at the time of the testing. Forall of them, the issue of presymptomatic testingarose during the genetic counselling sessionalthough their original request had concernedanother issue. We included nine cases in thisgroup, six of them parents (five mothers, onefather) of newly diagnosed aVected children.All were referred as part of making the diagno-sis in their children. They felt healthy, althoughfive showed minor clinical features of myotonicdystrophy. Molecular testing was abnormal inall of them. The other three (one couple and amother) had lost a fetus with unexplainedarthrogryposis and/or neuromuscular abnor-malities and wanted to know the risk for a fur-ther pregnancy. All three were completelyhealthy at clinical examination and the testresult was normal in all of them.

As summarised in table 2 and discussedbelow, three diVerent counselling approachesto presymptomatic testing for myotonic dystro-phy could be identified. The most commonpractice (n=46) was to take the blood sample atthe end of the initial genetic counsellingsession. The factors which decided the choiceof the approach are listed in table 3.

DiscussionThere is a general consensus that presympto-matic testing for late onset, mendelian, geneticdisorders should not be considered as a purelylaboratory procedure, but that it should be

linked to appropriate genetic counselling inorder to achieve a proper foundation of infor-mation, preparation, and support.6 How this canbe ensured will vary according to the disorderand a universal model cannot be applied. Thereis, however, a range of issues common to mostgenetic disorders of late onset. These issuesinclude the implications of an abnormal resultfor reproductive decisions, the genetic implica-tions to oVspring and other relatives, conse-quences for future employment and insurance,strategies for coping with an abnormal result,and the availability of support from family,friends, and professionals. Most reportedexperience so far has come from Huntington’sdisease, which represents an extreme situationregarding its severity and current lack oftherapy.7 A two step approach has generally beenrecommended, both to allow the complex issuesto be discussed fully and to give space to thinkabout the testing process.4

For myotonic dystrophy, however, no guide-lines have been produced so far, and theapproaches used have been based on individualpatterns of clinical practice. One clear diVerencebetween the two disorders relates to thepenetrance of the gene, since whereas inHuntington’s disease a considerable risk of seri-ous disease exists for healthy subjects at risk whohave passed 50 years of age, for myotonicdystrophy the great majority of oVspring of anaVected person will show definite clinicalfeatures of the disorder by early adult life, whilemost late onset cases are clinically mild. Brunneret al8 showed in a total of 139 clinically normaloVspring of myotonic dystrophy patients thatthe residual risk of carrying the myotonicdystrophy gene mutation is approximately 8%between the ages 20 and 39 and a comparablerisk (8.6%) was shown in a linkage based series.9

In the present series, the proportion of abnormalpresymptomatic test results in clinically normalsubjects (seven out of 69, 10.1%) was compar-able. However, in only two such cases (malesaged 19 and 37 years) was the abnormal expan-sion in the range likely to be associated with sig-nificant neuromuscular symptoms (>100 re-peats). Thus, likelihood of a clinically normalmyotonic dystrophy relative carrying a mutationsignificant for their own health is small, espe-cially after the age of 40 years.

We can now consider in turn the diVerentapproaches to genetic counselling identified inthis series (table 2). The factors listed in table3 are among those particularly relevant to theapproach that may be most appropriate.

The two stage approach, corresponding tothat used for Huntington’s disease and compar-able late onset neurodegenerative disorders, wasused in only seven of the 72 cases, being oVeredbut rejected in two others. Factors favouring thisapproach include unfamiliarity with the disorderand its consequences, complexities of familyrelationships and dynamics, and young age ofthe person to be tested, giving a greaterlikelihood that an abnormal result might indi-cate relatively severe disease. In general, wewould advocate it in any situation where theperson is uncertain about a decision to be testedand where the amount or complexity of new

Table 2 DiVerent approaches for presymptomatic testing used in Wales

Approach TotalNormal/abnormalmolecular result

(1) Two stage approachA Blood taken at second genetic counselling session 7 7/–B OVer of a two step approach, rejected by the person at risk 2 1/1

(2) Intermediate approachA Blood taken by family doctor or genetic nurse specialist at

interval after first genetic counselling session 3 3/–B Blood taken at first genetic counselling session, stored until

confirmation to proceed from patient. 5 5/–(3) One stage approach

A Blood taken at first genetic counselling session 46 35/11B Blood taken by the genetic nurse specialist at home, genetic

counselling clinic session only in case of an abnormal result 9 5/4Total 72 56/16

Table 3 Relevant factors for counselling approach inpresymptomatic testing for myotonic dystrophy

Factors

AgeClinical assessmentFamiliarity with disorderFamily dynamicSubject’s perception of possible symptomsTime pressure (pregnancy)

Letters 847

www.jmedgenet.com


http://jmg.bmj.com/


information to be considered is too great to begiven satisfactorily in a single session or where aperiod of reflection is needed.

In contrast to Huntington’s disease, wherethis complex situation is almost invariablypresent, this is not always the case for myotonicdystrophy, so we see no need for a two stepapproach to be recommended in all presymp-tomatic testing situations for myotonic dystro-phy; equally, where it is suggested but declined,as occurred in two cases in our series, we see noreason why it should be insisted on.

In the great majority of presymptomatic testsin the present series (46 out of 72), a bloodsample for analysis was taken at the end of theinitial genetic counselling session. This ap-proach seems particularly appropriate wherethe subject is familiar with the disorder, wherethere are no family complexities, or where theirage (over 40 years) makes the likelihood of anabnormal result small and where even anabnormal result is unlikely to have major clini-cal significance. This last factor contrastsstrongly with presymptomatic testing for Hunt-ington’s disease, where the chance of an abnor-mal result remains high at a relatively advancedage and where the clinical consequences of thedisorder are serious regardless of age at onset.In myotonic dystrophy, older subjects detectedas having clinically insignificant expansionsshould perhaps be considered gene carriersrather than aVected patients.

In eight instances in this series, an interme-diate approach was used (2A and B in table 2),where opportunity was given for reflection fol-lowing the interview by either storing the bloodsample until the person had confirmed thatthey wished testing to proceed (five cases) or byarranging for the blood sample to be taken afteran interval by the family doctor or geneticsnurse specialist (three cases). This wouldappear to provide a useful option when the casedoes not clearly fall into either the two stage orsingle stage situations.

In five cases, the request for presymptomatictesting was made in the context of an ongoingpregnancy, with prenatal diagnosis beingwished for in three of these. There was thus asituation of time pressure, making a full twostep approach diYcult even if considereddesirable, a problem not infrequently met alsoin presymptomatic testing for Huntington’sdisease. In all three cases potentially requiringprenatal diagnosis, blood was taken at the endof the first interview; all had a normal result sothat no further action was required. In the twocases where prenatal diagnosis was not wishedfor, an intermediate approach as outlinedabove was used.

In nine cases in this series, blood was taken ata home visit by the genetics nurse specialistbefore the genetic counselling clinic attendance,this only being arranged if the result was abnor-mal. This was considered appropriate at the timein cases where no special complexities wereexpected and it is likely that some unanticipateddiYculties would have been detected at thehome visit, allowing the sampling to be post-poned. However, we would not now recommendthis approach for a number of reasons. First, it

gives no opportunity for clinical assessmentbefore testing is undertaken, a factor of consid-erable importance in view of the high frequencyof asymptomatic subjects at risk who show clini-cal abnormalities. Second, while some of the rel-evant issues would have been able to bediscussed at the home visit, it is unlikely that afull picture of the potential consequences oftesting would have been obtained in this way.Thirdly, by seeing only those subjects with anabnormal result in clinic, no opportunity wasgiven to those with a normal outcome to discussthe disorder and its consequences for aVectedfamily members, something that might berelevant to them as a relative, even though notaVecting themselves.

Two areas which require particular discussionare the role of clinical examination in thepresymptomatic testing process and the import-ance of the subject’s perception as to whethertesting was presymptomatic or diagnostic. Previ-ous family studies have clearly shown that aconsiderable proportion of asymptomatic familymembers show definite clinical abnormalities onexamination (17.6% in one early study).8 In thepresent study, some clinical abnormalities werepresent in nine of the 16 cases with an abnormalpresymptomatic test result, in the total series of78 cases (see accompanying paper5 for details).By contrast, the likelihood of an abnormalgenetic test result in a carefully examined andclinically normal adult is low (around 8% in thestudies discussed above). The figure of seven outof 78 cases in the present series is comparable. Itcan thus be argued that clinical assessment islikely to make a greater contribution to outcomethan laboratory analysis and should always beundertaken before genetic testing. On the otherhand, since some people request testing prima-rily for reproductive reasons and may not recog-nise the possibility of being themselves aVected,this possibility and the reasons for clinicalassessment will need careful explanation beforetesting is performed, with the timing of theassessment depending on the individual situa-tion.

A related important issue is the possiblediVerence in perception of the person beingtested and the clinician involved as to whetherthe testing is “presymptomatic” or “diagnostic”.We have considered as presymptomatic in ourseries all those cases who had no complaints,even though abnormal or suspicious clinical fea-tures might have been present at the time oftesting. When the referring clinician has recog-nised these features it is possible that the testsituation may be handled as a diagnostic confir-mation. A particular issue in myotonic dystro-phy exists for the mothers of children suspectedof having congenital myotonic dystrophy, whomay be tested as part of their child’s diagnosis,whereas for the mother, the situation is likely tobe presymptomatic. Example 1 illustrates this.

Example 1 illustrates the diYculty in disclos-ing the diagnosis in minimally aVected subjectswith no subjective complaints and who are notfamiliar with the disorder. The reaction of themother clearly shows that she was caught in adilemma and did not have enough space tounderstand what was going on or to think about

848 Letters

www.jmedgenet.com


http://jmg.bmj.com/


her own health. The diagnosis became shiftedtowards herself although the clinical request hadconcerned her son. For the clinician, the moth-er’s situation appeared diagnostic. The view ofthe authors, however, is that as such people con-sider themselves healthy, a presymptomaticrather than diagnostic approach seems moreappropriate. This means that there is a need forinformation and counselling before moleculartesting. It seems appropriate to answer first thequestion of the diagnosis in the child andpostpone the blood sampling in the mother untilshe is informed, prepared, and supported.

No instance of presymptomatic testing of ahealthy young child was recorded in our series,the youngest being aged 16 years. In general,the policy recommended for late onset disor-ders was followed6 with full discussion, clinicalassessment where required, but with postpon-ing of genetic testing until an age was reachedwhere the implications of testing could beunderstood and consent given.

In general, this policy was understood andsupported by parents, but in a small number ofcases a strong and persistent wish for testing ofa healthy child was expressed, as outlined inexample 2.

Example 2 shows how perceived barriers caninterfere with full appreciation of the issuesinvolved in a diYcult situation. One possibleexplanation of the outcome might be thatremoval of the perceived barrier allowed afuller understanding of the potential problems,even though these had been repeatedly ex-plained on previous occasions.

A final counselling issue requiring mentionrelates to the potential need for disclosure ofgenetic test results in relation to insurance, a

topic of considerable debate in the UK andelsewhere. Although most of the period cov-ered by our retrospective study predated theemergence of this issue, it is now a topic thatrequires discussion with people before testing.However, the small number of subjects in theseries (two out of 78) who were clinically nor-mal, yet had an abnormal test result of clinicalsignificance to themselves, suggests that thereis little need for insurers to have access to suchresults, and that a normal clinical assessment isa more relevant factor.

In conclusion, the counselling approach inrelation to presymptomatic testing for myotonicdystrophy needs to be flexible, so as to respondto the variable and complex issues that mayarise. Adherence to a two stage process does notseem required for the majority of situations, buteach instance requires careful consideration andat least one full interview giving opportunity fordetailed discussion of the issues would seemessential. While the retrospective nature of thepresent study makes it unwise to draw definitiveconclusions, the issues raised should provide thestarting point for further, more systematic study,as well as giving a general framework that may beuseful for those involved with presymptomatictesting for this important and exceptionally vari-able condition.

Example 1A 3 year old boy, accompanied by his mother,was referred by a paediatrician for evaluationof a possible genetic cause for his unex-plained motor diYculties. The motherwanted to know the diagnosis in her son andthe risk for further children. The familyhistory was unremarkable. Clinical examina-tion of the son showed reduced muscle toneand slight developmental delay. The motherhad grip myotonia on examination, but hadnot previously regarded it as abnormal. Shestated that she and her son both slept withoutfully closing their eyes. For the clinicalgeneticist involved, the diagnosis of myo-tonic dystrophy was clear. A possible clinicalconnection between the symptoms of bothwere discussed and blood was taken formolecular testing for myotonic dystrophyfrom the mother only. At the second appoint-ment, the test result, which confirmed thediagnosis in the mother, was discussed. Themother was very upset about the result and apause for reflection was necessary beforetesting the son. At the third appointment, allmedical and reproductive implications werediscussed in detail and blood was taken fromthe son for molecular analysis. The resultshowed an abnormal expansion.

Example 2Presymptomatic testing was requested for a 1year old child by the mother on account of afamily history of myotonic dystrophy in thechild’s father. At the time of the consultation,the parents had separated and it was not pos-sible to determine whether or not the fatherwas clinically aVected, though the diagnosisof myotonic dystrophy in other family mem-bers was clear. The child was healthy anddeveloping normally; clinical examinationwas oVered and was entirely normal. It wasexplained that in view of this, it would be wisefor presymptomatic testing to be postponeduntil the child was of an age to consent andannual review was oVered. Despite this, themother remained determined that the childshould be tested, although recognising theissues of the child being unable to consentand that there were no specific medical orother benefits from childhood testing. Themother and child were seen on three furtheroccasions, with the mother’s demand un-changed despite normal clinical assessments.Following discussion between the profession-als involved, it was considered that the issueof testing was itself becoming a significantproblem that might outweigh any disadvan-tages. Four years after the initial consultation,testing of the child was oVered, again in thecontext of full explanation of the potentialconsequences, and the mother was requestedto bring the child at a convenient time for ablood sample to be taken. She did not returnand has not requested testing of the child overthe subsequent two years.

Letters 849

www.jmedgenet.com


http://jmg.bmj.com/


We thank all the clinical staV of the Institute of Medical Genet-ics for permission to report information on their patients and, inparticular, Professor Angus Clarke and Dr Helen Hughes forhelpful discussions. Siv Fokstuen was supported by a grant fromthe Swiss Academy of Medical Sciences.

1 Brook JD, McCurrach ME, Harley HG, Buckler AJ, ChurchD, Aburatani H, Hunter K, Stanton VP, Thirion JP,Hudson T. Molecular basis of myotonic dystrophyexpansion of a trinucleotide (CTG) repeat at the 3' end ofa transcript encoding a protein-kinase family member. Cell1992;68:799-808.

2 Gennarelli M. Novelli G, Andreasi BF, Martorell L, CornetM, Menegazzo E, Mostacciuolo ML, Martinez JM,Angelini C, Pizzuti A, Baiget M, Dallapiccola B. Predictionof myotonic dystrophy clinical severity based on thenumber of intragenic (CTG)n trinucleotide repeats. Am JMed Genet 1996;65:342-7.

3 Huntington’s Disease Collaboration Research Group. Anovel gene containing a trinucleotide repeat that isexpanded and unstable on Huntington’s disease chromo-somes. Cell 1993;72:971-83.

4 International Huntington’s Association and World Federa-tion of Neurology. Guidelines for the molecular genetic

predictive test in Huntington’s disease. J Med Genet 1994;31:555-9.

5 Fokstuen S, Myring J, Meredith L, Ravine D, Harper PS.Eight years experience of direct molecular testing for myo-tonic dystrophy in Wales. J Med Genet 2001;38:e42.

6 Advisory Committee on Genetic Testing. Genetic testing forlate onset disorders. London: Department of Health, 1997.

7 Harper PS, Lim C, Craufurd D. Ten years of presympto-matic testing for Huntington’s disease: the experience ofthe UK Huntington’s Disease Prediction Consortium. JMed Genet 2000;37:567-71.

8 Brunner HG, Smeets HJM, Nillesen W, Van Oost BA, VanDen Biezenbos JBM, Joosten EDMG, Pinckers AJLG,Hamel BCJ, Theeuwes ADGM, Wieringa B, Ropers HH.Myotonic dystrophy predictive value of normal results onclinical examination. Brain 1991;144:2303-11.

9 Reardon W, Floyd JL, Myring J, Lazarou LP, Meredith AL,Harper PS. Five years experience of predictive testing formyotonic dystrophy using linked DNA markers. Am J MedGenet 1992;43:1006-11.

10 Harper PS. Myotonic dystrophy. 3rd ed. London: Saunders,2001.

Trinucleotide repeat contraction: a pitfall inprenatal diagnosis of myotonic dystrophy

Jeanne Amiel, Valérie Raclin, Jean-Marie Jouannic, Nicole Morichon,Hélène HoVman-Radvanyi, Marc Dommergues, Josué Feingold, Arnold Munnich,Jean-Paul Bonnefont

EDITOR—Myotonic dystrophy (DM) is a com-mon autosomal dominant disorder character-ised by myotonia, muscle weakness, ECGabnormalities, cataracts, hypogonadism, andfrontal balding in the typical adult form (MIM160900). The genetic defect consists of theamplification of an unstable CTG trinucleotiderepeat in the 3' untranslated region of thedystrophia myotonica protein kinase gene(DMPK), which maps to 19q13.3.1 2 Normalsubjects have five to 37 repeat copies whileaVected subjects have over 50 repeats.1 There issome correlation between repeat length andclinical symptoms, especially with respect to theage at onset.3–6 In the vast majority of cases, thenumber of repeats increases during parent-oVspring transmission of the mutant allele, thusproviding some molecular basis to the observa-tion of anticipation (increased severity of thedisease in successive generations).7 8 However, adecrease in repeat size is occasionally observedin the oVspring, mostly in the case of paternaltransmission of an expansion of over 600trinucleotide repeats,9 but contraction of aparental expanded repeat back to the normalrange when transmitted to oVspring seems to bean extremely rare phenomenon.10–12 Here wereport such a case and emphasise the directimpact of this situation on prenatal diagnosis(PND) of DM.

Material and methodsPATIENTS

DM was diagnosed in II.2 (fig 1), who presentedwith mild atrophy of the head and neck muscles,myotonia of the hands, and frontal balding at theage of 37 years, while his first wife (II.1) waspregnant. The sister of patient II.2 (II.4) wasmore severely aVected, with marked impairment

of walking from the age of 43 years. Their father,I.2, had no symptoms of DM at the age of 61years. III.1, the first oVspring of II.2, wasseverely aVected with muscular weakness andmental retardation. II.2 was first referred to ourunit during the pregnancy of his second wife(II.3) for first trimester PND.

METHODS

DNA was extracted from leucocytes of II.2 andII.3 and from CVS performed at 11 menstrualweeks. Study of the CTG repeat size was carriedout by Southern blotting (EcoRI/PM10M6 andPstI/PM10M6)13 and PCR amplification withprimers flanking the CTG repeat in parental andfetal DNA.1 Poly (CA) microsatellite markerswere used both for linkage analysis at the DMlocus and to rule out false paternity (D19S223,D19S412, D19S606, D19S596, D19S879,D20S194, D12S78, with heterozygosity of 81,80, 81, 53, 76, 91, and 91% respectively, dataavailable through Genebank).

ResultsThe healthy mother (II.3) had two alleles of 10CTG repeats, while the father (II.2) displayeda wild type allele of 13 CTG repeats and amutated allele of approximately 200 CTGrepeats (figs 2 and 3). The DM allele in I.2,II.4, and III.1 had previously been estimated asapproximately 60, 400, and 600 CTG repeatsin size, respectively (data not shown for II.4and III.1). PCR amplification of the CTGrepeat region from the CVS DNA showed twonormal alleles, a 10 trinucleotide repeat alleleinherited from the mother and a 30 trinucleo-tide repeat allele not found in the father. FetalDNA testing by Southern blotting failed todetect any expanded allele (fig 3). Results ofthe CVS DNA analysis were confirmed both by

websiteextraThe accompanyingpaper5 can be foundon our website

www.jmedgenet.com

850 Letters

J Med Genet2001;38:850–852

Département deGénétique et UnitéINSERM U-393,HôpitalNecker-EnfantsMalades, 149 rue deSèvres, 75743 Pariscedex 15, FranceJ AmielV RaclinN MorichonJ FeingoldA MunnichJ-P Bonnefont

Maternité, HôpitalNecker-EnfantsMalades, Paris, FranceJ-M JouannicM Dommergues

Service de Biochimie,Hôpital AmbroiseParé, Boulogne,FranceH HoVman-Radvanyi

Service de BiochimieB, HôpitalNecker-EnfantsMalades, Paris, FranceJ-P Bonnefont

Correspondence to:Dr Bonnefont,[email protected]

www.jmedgenet.com


http://jmg.bmj.com/


PCR amplification of the CTG repeat regionand Southern blotting of DNA extracted fromcultured amniocytes. Linkage analysis in I.2,II.2, II.3, and III.3, using poly (CA) microsat-ellite markers flanking the DMPK gene,showed that the fetus had inherited thepaternal DM allele (fig 1). The probability offalse paternity was assessed to be less than 10-3.The parents were informed about this unusualsituation implying some uncertainty regardingthe fetal status and subsequently decided tocontinue the pregnancy.

DiscussionHere we report on the contraction of a largeexpanded DM allele to the normal range duringa father-oVspring transmission. To our knowl-edge, this is the first case where contraction of aparental expanded allele back to the normalrange has been detected during the prenatalperiod. No somatic mosaicism could be identi-fied in either choriocytes or amniocytes. Arecombination event at the DM locus appearsunlikely, based on both haplotype analysis andPCR amplification of the CTG repeat, indicat-ing that the fetus did not inherit the paternalwild type DM allele (fig 1). Such a contractionof a DM allele back to the normal range has sel-dom been reported.10–12 In one case, a discon-tinuous gene conversion between the wild typeand the DM allele was shown.11 In the otherthree cases, as in the case reported here, thismechanism appears unlikely, because of the dif-ferent numbers of CTG repeats in the father/oVspring wild type alleles.10 12 A contraction ofthe DM allele or a double recombinationdisrupting the CTG repeats is a possibility.These events could be either meiotic or earlymitotic. It is worth mentioning that in all casesreported to date, the contraction event waspaternal in origin, making the hypothesis of ameiotic event more likely. In one case, long termfollow up established that the oVspring re-mained asymptomatic at or beyond the age ofdisease onset in the transmitting parent, alsofavouring the hypothesis of a meiotic event.9

Understanding when and how the contractionevent occurs would be of importance toappreciate both the risk for the children ofdeveloping the disease and their own risk oftransmission of DM to their oVspring.

A partial reduction in size of a trinucleotiderepeat above the normal range during parent-oVspring transmission seems to be far more fre-quent than a reduction back to the normalrange. In a large series of 1489 DM parent-oVspring pairs reported by Ashizawa et al,9 apartial reduction was noted in 6.4% of cases.Such a contraction was more frequently ob-served in paternal than in maternal transmis-sions (10% versus 3%, respectively).9 In thesecases, the size of the parental DM allele variedfrom approximately 500 to 1000-1500 CTGrepeats. Interestingly, the observed number ofsibships whose members had inherited a con-tracted CTG repeat was greater than expected.This could argue either for a predisposition toreduction during transmission of an expandedallele in some subjects or for negative selectionagainst sperm carrying the largest CTG expan-sions. This last hypothesis has been raised inFRAXA, where males carrying a full mutation intheir somatic cells transmit only premutatedalleles to their daughters (MIM 309550).14

While expansion of a mutated DM allele dur-ing parent-oVspring transmission is almostinvariably associated with clinical anticipation,the rare events of contraction raise diYcultissues with respect to PND and genetic counsel-ling in DM. In the large series of Ashizawa et al,9

Figure 1 Linkage analysis at the DMPK locus in I.2, II.2, II.3, and III.3. The poly(CA) microsatellite markers used and genetic distances are as follows: cen - D19S223 -(4.2 cM) -D19S217 - (2.1 cM) - D19S412 - (6.3 cM) - D19S606 - (1.4 cM) -D19S596 - (1.3 cM) -D19S879 - tel. The DMPK gene lies between D19S217 andD19S412. Note that III.3 inherited the paternal DM allele. The number of CTG repeatevaluated by PCR amplification is given in parentheses.

I 1 2 (20/60)

421112

2313

52

II

III

19q13.3D19S223

D19S217D19S412

D19S606

DMPK

D19S596D19S879

1 2 3 4 5

1 2 3

231243

3132

14

431213

2112

54

411111

2111

53

(13/–) (10/10)

(10/30)

1 2

C

3 I.2 II.2 II.3 III.3

Figure 2 PCR amplification of the CTG repeat region. DNA analysis from three controls(C) and patients I.2, II.2, II.3 (leucocytes) and III.3 (CVS) are shown. The PCRproducts were loaded onto a 2% agarose gel. The estimated number of CTG repeats for bothalleles for each subject is as follows: C1: 5/42, C2: 5/20, C3: 5/90, I.2: 20/60, II.2: 13/-,II.3: 10/10, III.3: 10/30. The minor PCR product, larger than the predominant product,seen in the controls, II.2, and III.3, is likely to be a non-specific PCR product.

Letters 851

www.jmedgenet.com


http://jmg.bmj.com/


approximately half of the oVspring who inher-ited a contracted allele showed clinical anticipa-tion despite the reduced CTG repeat size. In allthese cases, however, the size of the contractedallele remained above the normal range. More-over, Southern blot analysis showed some over-lap between the boundaries of the “smear” insome parent-oVspring pairs. Conversely, in thefour cases where the transmitted DM allelereverted to the normal range, the clinical pheno-type seemed to be normal, taking into accountthe absence of data regarding long term followup in these cases.9–12

Taken together, these data strengthen thewell known fact that direct analysis of fetalDNA should be used as the primary approachin PND, since a reliable prediction of the seri-ousness of the phenotype cannot be basedupon haplotyping using polymorphic markerslinked to the DMPK locus. Moreover, thedetection of a contraction event in a fetus bySouthern blotting warrants further molecularinvestigations in order to assess the size of theCTG repeat accurately. Indeed, while detec-tion of a DM allele remaining above the normal

range does not preclude clinical anticipation,the observation of a contracted allele back tonormality should allow reassurance of couplesat risk for transmitting DM.

1 Buxton J, Shelbourne P, Davies J, Jones C, Van Tongeren T,Aslanidis C, de Jong P, Jansen G, Anvret M, Riley B, Wil-liamson R, Johnson K. Detection of an unstable fragment ofDNA specific to individuals with myotonic dystrophy.Nature 1992;355:547-8.

2 Harley HG, Rundle SA, Reardon W, Myring J, Crow S,Brook JD, Harper PS, Shaw DJ. Unstable DNA sequencein myotonic dystrophy. Lancet 1992;339:1125-8.

3 Suthers GK, Huson SM, Davies KE. Instability versuspredictability: the molecular diagnosis of myotonic dystro-phy. J Med Genet 1992;29:761-5.

4 Tsilfidis C, MacKenzie, AE, Mettler G, Barcelo J, KornelukRG. Correlation between CTG trinucleotide repeat lengthand frequency of severe congenital myotonic dystrophy.Nat Genet 1992;1:192-5.

5 Harley HG, Rundle SA, MacMillan JC, Myring J, Brook JD,Crow S, Reardon W, Fenton I, Shaw DJ, Harper PS. Size ofthe unstable CTG repeat sequence in relation to phenotypeand parental transmission in myotonic dystrophy. Am JHum Genet 1993;52:1164-74.

6 Hamshere MG, Harley H, Harper P, Brook JD, Brookfield JF.Myotonic dystrophy: the correlation of (CTG) repeat lengthin leucocytes with age at onset is significant only for patientswith small expansions. J Med Genet 1999;36:59-61.

7 Harper PS, Harley HG, Reardon W, Shaw DJ. Anticipationin myotonic dystrophy: new light on an old problem. Am JHum Genet 1992;51:10-16.

8 Lavedan C, Hofmann-Radvanyi H, Shelbourne P, Rabes JP,Duros C, Savoy D, Dehaupas I, Luce S, Johnson K, JunienC. Myotonic dystrophy: size- and sex-dependent dynamicsof CTG meiotic instability, and somatic mosaicism. Am JHum Genet 1993;52:875-83.

9 Ashizawa T, Anvret M, Baiget M, Barcelo JM, Brunner H,Cobo AM, Dallapiccola B, Fenwick RG Jr, Grandell U,Harley H, Junien C, Koch ME, Korneluk RG, Lavedan C,Miki T, Mulley JC, López de Munain A, Novelli G, RosesAD, Seltzer WK, Shaw DJ, Smeets H, Sutherland GR,Yamagata H, Harper PS. Characteristics of intergenera-tional contractions of the CTG repeat in myotonic dystro-phy. Am J Hum Genet 1994;54:414-23.

10 Shelbourne P, Winqvist R, Kunert E, Davies J, Leisti J,Thiele H, Bachmann H, Buxton J, Williamson B, JohnsonK. Unstable DNA may be responsible for the incompletepenetrance of the myotonic dystrophy phenotype. Hum MolGenet 1992;1:467-73.

11 O’Hoy KL, Tsilfidis C, Mahadevan MS, Neville CE,Barcelo J, Hunter AG, Korneluk RG. Reduction in size ofthe myotonic dystrophy trinucleotide repeat mutation dur-ing transmission. Science 1993;259:809-12.

12 Brunner HG, Jansen G, Nillesen W, Nelen MR, de Die CE,Howeler CJ, van Oost BA, Wieringa B, Ropers HH, SmeetsHJ. Brief report: reverse mutation in myotonic dystrophy. NEngl J Med 1993;328:476-80.

13 Brook JD, McCurrach ME, Harley HG, Buckler AJ, ChurchD, Aburatani H, Hunter K, Stanton VP, Thirion JP,Hudson T, John R, Zemelman B, Snell RG, Crow S, DaviesJ, Shelbourne P, Buxton J, Jones C, Juvonen V, Johnson K,Harper PS, Shaw DJ, Housman DE. Molecular basis ofmyotonic dystrophy: expansion of a trinucleotide (CTG)repeat at the 3' end of a transcript encoding a protein kinasefamily member. Cell 1992;68:799-808.

14 Mandel JL. Questions of expansion. Nat Genet 1993;4:8-9.

Psychological studies in Huntington’s disease:making up the balance

Magdalena Duisterhof, Rutger W Trijsburg, Martinus F Niermeijer, Raymund A C Roos,Aad Tibben

EDITOR—Huntington’s disease (HD) is anincurable neurodegenerative disease, charac-terised by involuntary movements, changes inbehaviour and personality, and cognitive im-pairment, leading to death 15 to 20 years afterits onset.1 HD is an autosomal dominantlyinherited disorder, the gene for which is local-ised on the short arm of chromosome 4.2 Sub-jects carrying the gene will develop the diseasein the absence of other causes of death. The

mean age of onset is 40 years, by which timegene carriers may have passed on the gene totheir oVspring. The age of onset ranges from 2to 75 years3 so that those at risk (that is, riskcarriers at 50% or 25% genetic risk) can neverbe sure of having escaped HD.

Since 1986, presymptomatic DNA testingusing genetic linkage analysis has made itpossible for risk carriers to have their riskmodified to approximately 98% or 2%. After

Figure 3 Southern blotting (EcoRI/ PM10M6) in patients III.3, II.3, I.2, and II.2. Theexpanded allele in II.2 is not found in his oVspring III.3.

9.8 kb

II.2I.2II.3III.3

8.6 kb

852 Letters

www.jmedgenet.com


http://jmg.bmj.com/


identification of the HD gene mutation in1993, CAG repeat size analysis of the hunting-tin gene allowed complete certainty of eitherhaving or not having HD.2

Risk carriers, being raised in a family in whichHD played a major role, could be expected tohave specific adjustment problems. Yet, only onestudy addressed the psychological functioning ofpeople at risk for HD before presymptomatictesting was introduced. Most psychologicalstudies were started when clinicians and re-searchers became concerned about the eVects ofa presymptomatic test on people at risk.

The aim of this article is to review studiesaddressing psychological and psychiatric adjust-ment of people at risk for HD. The methodsused by the studies (that is, objectives, inclusionand exclusion criteria, recruitment, assessment,design, and statistical analyses) and their resultsare presented. General trends and limitations ofthe present work are described and a directionfor future research is presented.

The term “carriers” is used to designate allsubjects who underwent linkage or mutationanalysis and were found to have an increasedrisk or were identified with a pathologicalrepeat length of the IT15 huntingtin gene. Theterm “non-carriers” denotes those with adecreased risk result or those having a normalrepeat size of the IT15 gene.

MethodsA search of published reports was conducted inthe MEDLINE and PsycLIT databases usingthe keywords “Huntington’s disease”, “psycho-logical”, “psychiatric”, “predictive testing”, “ad-justment”, and “family”. Cross references inidentified papers were also used. Quantitativestudies on the psychological wellbeing of thoseat risk were included; this could be conducted by

questionnaire or by interview. Studies address-ing attitudes are included when they indirectlyrefer to wellbeing in the pre- and/or post-testperiod. Case descriptions or clinical impressionswere excluded from this analysis as well asneurological and pharmacological studies.

ResultsA total of 18 articles provided a quantitativeanalysis on the wellbeing of subjects at risk forHD.4–21 Characteristics of the studies are sum-marised in table 1.

STUDY OBJECTIVES

Before predictive testing became possible, Fol-stein et al11 12 analysed psychological character-istics and psychiatric disorders among the oV-spring of HD patients and other at risk people.The pre- to post-test adjustment of carriersand non-carriers was evaluated in sevenstudies.5 9 14 15 19–21 Attitudes, indirectly refer-ring to wellbeing, before test disclosure4 or inthe post-test period were addressed in fourstudies.4 6 16 18 A few studies also comparedadjustment in test applicants with adjustmentin their partners.15 16 18 20 In order to identifythose subjects who were at risk for poor adjust-ment after the test, predictive studies wereintroduced.7 8 10 13 17

INCLUSION AND EXCLUSION CRITERIA

In the study of Folstein,12 subjects at 50% and25% risk were included. Folstein et al11

included oVspring (aged 15 years and older) ofHD patients. Exclusion criteria were notreported in these studies.

Studies on adaptation after testing for HDhad comparable criteria for inclusion in pre-and post-test studies: people aged over 18 yearsand at risk for HD. Exclusion criteria were:having symptoms of HD, a severe depression or

Table 1 Studies of psychological functioning of HD risk carriers

Study Sample size

Carrier/non-carrier/uninformative* Mean age Measurement time Objective Statistical methods

BostonMeissen et al14 15 4/7/5 — Baseline, 3 mth, 9 mth Percentages

VancouverBloch et al4 51 NA 39.3 Before test Description PercentagesWiggins et al21† 135 37/58/40 37.5 Baseline, 1 wk, 6 mth, 12 mth Course ANOVA, Kruskal-Wallis testLawson et al13† 135 37/58/40 37.5 Baseline, 1 wk, 6 mth, 12 mth Prediction, description t tests or Mann-Whitney U tests,

chi-square testBaltimore

Folstein et al11 112 NA 26.7 NA Description Percentages, chi-square testBrandt et al5 55 12/30/13 35.4 Baseline, 3 mth, 6 mth, 9 mth,

12 mthDescription, baseline,course

Percentages

Folstein12 147/161 NA — NA Description Clinical impressionsCodori and Brandt6 68 17/51 37.7 8 visits in 3 y after test Description PercentagesCodori et al7 160 52/108/— 34.4 Baseline, 3 mth, 6 mth, 9 mth,

12 mthPrediction F tests

Rotterdam/LeidenTibben et al16 18 9/9 35.9 12 mth Description Percentages, clinical impressionsTibben et al17‡ 63 29/44 31.6 Baseline, 6 mth Prediction Backward regression analysisTibben et al18‡ 63 24/39 31.6 6 mth Description PercentagesTibben et al19‡ 73 29/44 32.1 Baseline, 1 wk, 6 mth Course MANOVATibben et al20‡ 49 20/29 32.2 Baseline, 1 wk, 6 mth, 3 y Course MANOVADudokdeWit et al9§ 25 9/16 39.5 Baseline, 1 wk, 6 mth Course MANOVADudokdeWit et al10§ 25 9/16 39.5 Baseline, 6 mth Prediction Multiple regression analysis

LeuvenDecruyenaere et al8 53 22/31 34 Baseline, 1 mth, 12 mth Prediction, course Multiple regression analysis, t tests

IndianapolisQuaid and Wesson15 19 5/14 36.9 Baseline, 3 mth, 6 mth, 9 mth,

12 mthComparison of groups Mann-Whitney U test

*Genetic status not assessable in linkage test.†‡§Same population.NA: not applicable.

Letters 853

J Med Genet2001;38:852–861

Department ofMedical Psychologyand Psychotherapy,Erasmus UniversityMedical CentreRotterdam, DrMolewaterplein 50,Room CF 226, 3015 GERotterdam, TheNetherlandsM DuisterhofR W TrijsburgA Tibben

Department of ClinicalGenetics, ErasmusUniversity MedicalCentre Rotterdam,Rotterdam, TheNetherlandsM F NiermeijerA Tibben

Department ofNeurology, LeidenUniversity MedicalCentre, Leiden, TheNetherlandsR A C RoosA Tibben

Correspondence to: DrDuisterhof,[email protected]

www.jmedgenet.com


http://jmg.bmj.com/


other major psychiatric illness, or, by history,being at risk for suicide (Baltimore Group,USA,23 Boston, USA,14 Indianapolis, USA,24

Leuven Group, Belgium,25 Rotterdam/LeidenGroup, The Netherlands,26 Vancouver Group,Canada27). In the study by Meissen et al,14 sec-ondary exclusion criteria were: a recently expe-rienced stressful event, moderate depression, asuicide attempt more than 10 years before test-ing, or a family history of suicide.14 The Leuvengroup included risk carriers with a psychiatrichistory, provided that social support was avail-able and that the risk carriers were receivingpsychiatric treatment (M Decruyenaere, 1999,personal communication).

Postponement or exclusion from testingwere reported for various reasons: because ofmanifest symptoms of HD (n=44 and 105),severe depression (n=3,6 1,14 and 323), andevaluation by a psychiatrist (n=214). Theseexclusion criteria were not applied in any of thestudies of Decruyenaere, DudokdeWit, andTibben et al (personal communications, 1999).

RECRUITMENT

In the studies unrelated to presymptomatic test-ing, oVspring of patients were asked to partici-pate in the study and were identified eitherthrough a survey of HD patients in Maryland11

or when they visited the Baltimore Huntington’sDisease Project Research Clinic with questionsconcerning their own and/or parents’ future.12

Psychological pre- and post-test follow upwas oVered on a research basis, by requesting

informed consent from presymptomatic testparticipants. Information on the availability ofthe presymptomatic testing reached risk carri-ers through the Newsletters of the HD Society,the general practitioner, neurologist, clinicalgenetics service, relatives, or the public net-works. Pre-test written information was pro-vided in several centres. General informationwas mailed to all 50% risk carriers in onegroup.15 The Vancouver Group mailed adescription of the research project to allfamilies on the HD registry, requesting them tocontact the researcher.27

ASSESSMENT OF PSYCHOLOGICAL ADAPTATION

Wellbeing was assessed through self-reportquestionnaires and by means of interviews.The questionnaires used are summarised intables 2 and 3. Folstein et al11 performedpsychiatric examination by means of a struc-tured interview, the Diagnostic InterviewSchedule (DIS), which yields diagnoses ac-cording to the Diagnostic and StatisticalManual of Mental Disorders (DSM-III); anindependent psychiatric interviewer validatedthese diagnoses. In other studies, the interviewsprovided additional information to the self-report questionnaires. Brandt et al5 adminis-tered the Schedule of AVective Disorders andSchizophrenia (SADS)-Change Interview28 toassess at least moderate to severe symptoms inone or more domains. Lawson et al13 askedcounsellors and clinicians to indicate partici-pants who had experienced adverse events, forexample, a suicide attempt or the formulationof a suicide plan, psychiatric hospitalisation,depression lasting longer than two months, amarked increase in substance abuse, or thebreakdown of important relationships. Tibbenet al16 evaluated feelings and cognition in carri-ers and non-carriers and their partners.

The design and statistical analyses of theinvestigated studies are summarised in table 1.

Results of reviewed studiesSTUDIES UNRELATED TO PRESYMPTOMATIC

TESTING

Children of HD patients had a high rate ofpsychiatric disorder (25% conduct disorder orantisocial personality disorder, 18% majordepression).11 Most conditions (anxiety anddepression) were mild or occurred only in ado-lescence (conduct disorder).12 Introversion/

Table 2 Descriptive/course studies: used questionnaires

Study Questionnaires

VancouverBloch et al4 SCL-90-R (GSI), BDI, GWS (MHI), Behaviour Survey, Reasons

for Living Scale, AQWiggins et al21 SCL-90-R (GSI), BDI, GWS

BaltimoreBrandt et al5 SCL-90-R, BDI, BHS, MCMI-2Codori and Brandt6 Nine items regarding eVects of test resultFolstein12 EPI, GHQ-30Folstein et al11 —

BostonMeissen et al14 BDI

Rotterdam/LeidenTibben et al16 AQ, BHSTibben et al18 AQTibben et al19 BHS, GHQ-60, IESTibben et al20 BHS, IESDudokdeWit et al9 IES

IndianapolisQuaid and Wesson15 SCL-90-R (GSI, PSDI, PST), BDI, BHS, GWS (MHI), Life

Satisfaction Index

Abbreviations are shown in the Appendix.

Table 3 Outcome variables and predictor variables in studies predicting psychological well being

Study Predictor variables Outcome variables

Tibben et al17 Intrusion (IES), avoidance (IES), hopelessness (BHS), psychopathologicalstates (GHQ), social support (SSQ)

Intrusion (IES), avoidance (IES), hopelessness (BHS)

Decruyenaere et al8 General anxiety (STAI trait), situational anxiety (STAI state), depression(BDI), ego strength (MMPI), coping styles (UCL)

General anxiety (STAI trait), situational anxiety (STAIstate), depression (BDI), ego strength (MMPI)

Codori et al7 Genetic status, gender, marital status, parenting status, risk perception,estimated years to onset of HD in those testing positive

Hopelessness (BHS), depression (BDI)

Lawson et al13 Global Severity Index (SCL-90), depression (BDI), social support (SSQ),adverse events questionnaire

Adverse events*

DudokdeWit et al10 Intrusion (IES), avoidance (IES), anxiety (HADS), depression (HADS),psychological problems (SCL-90), hopelessness (BHS), loneliness (lonelinessscale), social support (SSQ), family functioning (FACES), gender, age,religion, marital status, parenting status

Intrusion (IES), avoidance (IES)

*A suicide attempt or the formulation of a suicide plan, psychiatric hospitalisation, depression lasting longer than two months, a marked increase in substance abuse,the breakdown of important relationships, increases on BDI and or GSI (SCL-90).For abbreviations see table 2.

854 Letters

www.jmedgenet.com


http://jmg.bmj.com/


extraversion and neuroticism were similar tothose in the general population.12

DESCRIPTIVE STUDIES RELATED TO

PRESYMPTOMATIC TESTING

Psychological wellbeing of test applicantsbefore disclosure of test resultThe mean scores of psychological wellbeingand Huntington specific distress before disclo-sure of the test result (baseline level) fell withinthe normal range.4 5 7 9 19–21 In the Dutchstudies,19 29 the mean scores of risk carriersindicated mild signs of hopelessness; this couldnot be confirmed in other studies.7 15 Approxi-mately 20% of the risk carriers scored at mildlevels of depression and hopelessness, whereasvery few scored at the level of moderate orsevere depression.4 For about 20% of the riskcarriers, their scores on the GHQ-60 indicatedthe possible presence of psychiatric morbid-ity.19 30

Most test applicants had a normal psycho-logical profile.31 In comparison to the generalpopulation, they were more socially extra-verted, had higher ego strength, and reactedmore with active coping, palliative coping,social support seeking, and comforting ideas.

Later identified carriers and non-carriers didnot diVer in general wellbeing and Huntingtonspecific distress.5 8 9 15 19–21

Course of psychological wellbeing afterthe test resultGeneral measures of psychological wellbeing andHuntington specific distressAnalysis of distress in identified gene carriers atseven days post-test showed more depression,hopelessness, and a decrease in generalwellbeing.5 8 19–21 However, their mean scoresremained in the mild range. A return tobaseline levels of anxiety and depressionoccurred in the first month.8 Hopelessness,depression, and general wellbeing returned tobaseline level within six months5 19 21 andremained there one and three years post-test.8 20 21 Although not diVering from baseline,Wiggins et al21 found linear declines for distressand depression over a 12 month period. OnlyBrandt et al5 reported a slight increase ingeneral distress after one year. However,because the dropout rate in their sample wasextremely high (75%), this finding should beinterpreted with caution.

The non-carriers were more optimisticregarding their future at seven days post-test;however, this more positive view of the futuredisappeared after six months and threeyears.18 20 On the other hand, anxiety anddepression decreased from baseline one monthand one year after the test result.8 Also, generaldistress, assessed by means of the GSI index,decreased in the first year after the testresult.5 21

In comparison to carriers, non-carriersreported less general distress, less depression,less hopelessness, and a greater sense ofwellbeing one week after the test result.19 21

This diVerence disappeared in the first year. Atsix months follow up, only general wellbeing

was significantly greater for the non-carriers,but this diVerence disappeared at 12 monthsup to three years after the test result, returningto baseline level.8 21 32 Only Quaid and Wesson15

found a higher general wellbeing for carriersthan non-carriers after 12 months.

In comparison to a “no change” group, con-sisting of 23 subjects who did not want to takethe test and 17 subjects for whom the test wasuninformative, both carriers and non-carriersscored lower for depression and higher forwellbeing.21 However, it cannot be inferredfrom these findings that testing has benefits,since particularly an uninformative result canlead to an increase in distress, the wish for cer-tainty about carrier status being frustrated.

A subgroup of both carriers and non-carriershad diYculties adjusting to their new carrierstatus. About 10-20% of both carriers andnon-carriers showed psychological problems inthe post-test period.8 13 16 19 21 32 33 Interviewswith carriers, three months after the result,indicated that half of the carriers had periods ofsevere depression, whereas the other half hadsuVered moderate depression.14 Therapistsidentified a minority of carriers and non-carriers as having psychiatric symptoms in thefirst year after the test.5 However, very fewpeople committed or attempted suicide orneeded psychiatric hospitalisation after predic-tive testing.34

With regard to Huntington specific distress,carriers showed a slight increase of avoidancebehaviour in the first six months, whichreturned to baseline level after three years.20

Non-carriers showed a decrease in avoidanceduring the first six months post-test,9 19 whichreturned to baseline level at the three year fol-low up.20 For both groups, intrusive thoughtsdecreased in the first six months,9 19 whereasthese increased to baseline level at the threeyear follow up.20

The observations by Lawson et al13 underlinethe general impression that both carriers andnon-carriers have problems in adapting to thetest result, but at diVerent moments in time.The number of adverse events was similar forcarriers and non-carriers. For the carriers,adverse events took place within 10 days afterthe test result, whereas for non-carriers adverseevents occurred six months after the result orlater. Seventy percent of these events wereidentified by clinical criteria, that is, suicidalideation, depression lasting longer than twomonths, substance abuse, or a breakdown of animportant relationship, either alone or inconjunction with a raised score on one or morequestionnaires.13

Attitudinal studies regarding the test resultBefore a test result was given, risk carriersexpressed concern about the future and guiltabout the possibility of passing on the gene.4

After six months, half of the carriers stated thatthe results had not influenced their lives andhalf of them also rarely thought of the result,indicating that denial plays a role.18 Thenon-carriers expressed relief in the first weeksafter the test result was given, but after sixmonths half of the non-carriers appeared to

Letters 855

www.jmedgenet.com


http://jmg.bmj.com/


deny the impact of the test result, as wasreflected by absence of relief and emotionalnumbness.18 Some of them have expressed sur-vivor guilt.6 16

Comparison of at risk population and partnersAt baseline, spouses reported more depressionthan their at risk partners,15 whereas hopeless-ness was comparable for carriers and theirpartners.16 During the three year follow up,carriers and their partners showed similar pat-terns of avoidance, intrusion, and hopeless-ness, whereas non-carrier partners reportedless avoidance, intrusion, and hopelessnessthan the non-carriers.32 After three years, part-ners of carriers were still showing more avoid-ance than partners of non-carriers. In contrast,Quaid and Wesson15 found comparable distressfor carrier partners and non-carrier partners inthe first year after disclosure of the test result.

Whereas distress was similar for carriers andtheir partners, their attitudes towards the testresult diVered. Carriers did not report anincrease in problems after they received anunfavourable test result. Their partners didmention having problems, but expressed reluc-tance to seek help or to talk about it with theirspouse, owing to feelings of guilt and not want-ing to hurt them. This was especially the casefor those who became aware of the risk for HDat a later stage, for example, after marriage. Forthe non-carriers, most of them did not experi-ence relief, whereas their partners did.16 18

Having children proved to be an additionalstress factor for partners during and after thetest procedure.20 At baseline, partners withchildren were significantly more hopeless thanpartners without children. One week after thetest, carrier partners with children reportedsignificantly more hopelessness, avoidancethoughts, and intrusive feelings than carrierpartners without children. At six months andthree years after the test, this diVerence inavoidance thoughts and intrusive feelings wassustained.

Prediction of wellbeingFive studies aimed to identify pre-test variablesthat predict the way subjects adapt to their testresult7 8 10 13 17 (for the variables see table 3).

Test resultIn general, test outcome did not predictpsychological adjustment. Only Codori et al7

found carriers to be more likely to be pessimis-tic about their future than non-carriers.

General and Huntington specific distressThe level of psychological adaptation after thetest (anxiety, depression, hopelessness, intru-sion, and avoidance) was predicted by the samemeasures at baseline.7 8 10 17 The more depres-sive symptoms reported at baseline, the moredistress subjects reported at the one year followup, and the greater the chances that they wererated as having experienced an adverse event,as defined by Lawson et al.13 Pessimism, a lowavoidance, and dissatisfaction with availablesupport at the moment of testing predicted

pessimism at six months.17 Those severely anx-ious before the test were more likely to showlow intrusion six months after disclosure.

Biographical variablesHaving children predicted post-test intrusionand hopelessness.7 10 Women showed moreintrusion and avoidance than men six monthsafter disclosure of the test result.10 For carriers,being married or having children predictedhopelessness, as did the estimated years toonset of HD.7

Social supportSubjects who were satisfied with the perceivedquality of support of others felt less hopelessafter either test result.17 The more pre-testavoidance and the less satisfaction with avail-able support, the more avoidance behaviourwas reported six months post-test.17 However,the larger the number of persons perceived asbeing supportive before the test, the moreavoidance was reported post-test.10

Risk perceptionRisk perception refers to the expectations onehas about the test result. No support was foundfor the hypothesis that for identified carriers,those with a low perceived risk of being acarrier would have a less favourable adjustmentthan persons with a high perceived risk.7

Personality measures and coping strategiesEgo strength was associated with a lowergeneral anxiety and depression level one yearafter the test. Moreover, ego strength in combi-nation with the coping strategy “comfortingideas” predicted a lower general anxiety.8

DiscussionReviewing psychological research on wellbeingin Huntington patients and those at riskshowed only a few studies that were unrelatedto presymptomatic testing. HD in a parentproved to have a profound impact on adoles-cence. A high rate of psychiatric disorders inadolescents was seen, but not among adults.Most of the research was carried out as part ofthe presymptomatic testing protocol. In gen-eral, wellbeing of the group of test applicantswas normal before test disclosure. Both carri-ers and non-carriers had diYculties in adaptingto the test result, but at diVerent moments intime. Distress in carriers increased in the firstweeks post-test, which returned to baselinelevel within one year. The relief non-carriersexpressed in the first weeks disappearedafterwards; they experienced most distress atsix months. Within one year, non-carriersseemed to be somewhat less distressed thanthey were before test disclosure, but they hadnot developed more optimistic future expect-ancies.

A subgroup of both carriers and non-carriershad long lasting adaptation problems. Thosereporting to be distressed before test disclosuremost often had problems in adapting to the testresult. Although wellbeing seemed to beindependent of test outcome, wellbeing was

856 Letters

www.jmedgenet.com


http://jmg.bmj.com/


related to having children, certain personalitytraits (ego strength, coping), and the subjectiveestimation of the number of years before onsetof HD.

These findings have been shown to be help-ful in guidance and counselling of risk carriersin testing programmes. However, the researchstill has some serious limitations that need tobe overcome for progress to be made in thisresearch field. Limitations and a promising newdirection will be discussed below.

STUDY POPULATION

In the study of Folstein et al,11 60% were notwilling to participate in the study, leading to apossible underestimation of the problems ofrisk carriers. The study population in otherstudies consisted of risk carriers who visited agenetic centre and/or applied for a predictivetest. The percentage of those at risk whorequested testing when approached by regis-tries or testing centres varied from 9% inWales, 10% in Indiana, 16% in the Manchesterarea, to 20% in the Vancouver area.35 In TheNetherlands, 752 out of 1032 subjects at risk,applying for presymptomatic testing in theperiod 1987 to 1997, decided to be tested,which is 24% of the at risk persons registered inthe Leiden Roster for HD.36 It was suggested37

and confirmed4 31 that persons who participatein the studies on testing form a resourcefulself-selected group. Those who decided not tobe tested had more frequent expectations ofunfavourable emotional reactions and showedmore hopelessness than tested subjects.6 38 Onthe other hand, the level of anxiety, egostrength, and coping strategies were not diVer-ent between the tested and untested groups.39

Also, the untested participants form a self-selected resourceful group; both tested anduntested participants had a higher ego strengthin comparison to the general population.31

Little is known about the wellbeing of thosewho do not seek testing and who do notparticipate in psychological studies. Therefore,bias seems to be involved in the estimation ofadaptation in HD risk carriers.

Although we need to be careful to generalisethe findings to the whole HD population, weshould take into account that diVerences wereobserved within the group of test applicants.Some subjects acknowledged the burden ofHD, but saw themselves as being able to facethe truth. Others denied a burden of HD intheir lives and disagreed that the results had aprofound impact on their lives.4 40

Moreover, the dropout rates in most followup studies are high. Information from relativesabout the wellbeing of these dropouts suggestthat those who declined participation in followup research, both carriers and non-carriers,often have serious problems they do not wantto disclose, indicating that risk carriers apply-ing for the test may have more problems thanthe studies suggest.

SELF-REPORT MEASUREMENT

Adjustment was usually assessed by means ofself-report questionnaires. Interpretations offindings based on self-reporting may be

problematical because low scores on mentalhealth questionnaires may indicate that peopledeny problems, trying to maintain an illusion ofmental health.41 Since denial may arise in reac-tion to stressful or traumatic experiences,42 thiscan be expected in a testing procedure for HD.Tibben et al16 found that some carriers did notmention having had depressive periods in thepost-test period, whereas their partners re-ported the opposite about them.16 Moreover,test applicants were more defensive whenfilling out the MMPI than the general popula-tion. Also, female participants obtained ahigher lie score than women in the generalpopulation.31

DudokdeWit et al29 introduced the possibilityof assessing the manner in which participantsdiscuss the disease, the test, and its implica-tions in terms of coherence. Coherence refersto the ability to discuss and to reflect uponemotions, feelings, and ideas without becomingentangled in it or avoiding discussion of thesubject.43 DudokdeWit et al29 found that onethird of the participants spoke incoherentlyabout their possible inherited disease, themajority of them (two thirds) using anavoidance (dismissing) strategy, one thirdbeing entangled. It turned out that those show-ing avoidance reported fewer problems thanthose being entangled did. Dismissing subjectsgenerally have more psychological and psychi-atric problems than others do.44

These findings support the impression ofclinicians and counsellors that a group of HDrisk carriers who report themselves to be func-tioning well are in fact having diYculty withbeing aware of the impact of their experienceswith HD on their lives, reflected in sustainednumbness.10 45 This numbness may reflectwarding oV a variety of guilt feelings, anger,resentment, and hostility towards the familyand its HD history and the inability to createnew life perspectives. When the reality of asituation is avoided, it cannot be integrated intoone’s personal life, which might lead to adapta-tion problems. To gain more insight into thepsychological dynamics, denial and otherpsychological defences need to be studied;since defences are unconscious, more subtlemeasures are required.

GLOBAL MEASUREMENT

Another problem is that most measures usedare global ones. The IES is the only Hunting-ton specific questionnaire that provides insightinto the process of a person’s working throughthe situation. Research with more specific andsensitive measures is needed for assessment ofthe process of adjustment between and withintest applicants in the post-test period.

THEORETICAL FRAMEWORK

Tibben et al19 and DudokdeWit et al9 used thestress response theory of Horowitz et al46 toformulate their hypotheses. In other studies, itis unclear which underlying theoretical as-sumptions are used, the design and statisticsnot being guided by clear hypotheses. Atheoretical framework is needed to providemore insight into the observations.37 47 A

Letters 857

www.jmedgenet.com


http://jmg.bmj.com/


psychological model or theory will contributeto our understanding about the psychologicaldynamics that characterise this study popula-tion.

LACK OF FAMILY PERSPECTIVE ON WELLBEING OF

RISK CARRIERS

HD is a family disease.12 The initial onset ofsymptoms is usually between 30 and 50 years,a period when people are raising a family.People at risk were generally familiar with thedisease from early childhood, knowing thesymptoms in the parent and/or other familymembers. Clinicians have shown how the pres-ence of HD in a family can aVect the familydynamics.48–51 In some of the reviewed studies,the influence of HD on family dynamics can beinferred. Post-test studies indicated the diY-cult and diVerent processes test participantsand their partners go through. Marriage andcareer need to be reconsidered14 and the neces-sary social support may no longer be available.Having children is an additional stress factorfor both carriers and their partners.20

However, wellbeing in HD risk carriers hasrarely been related to their childhood experi-ences. Folstein et al11 investigated how child-hood experiences contribute to a more or lessfavourable adaptation in later life. They foundconduct disorder in adolescents and antisocialpersonality disorder in adults to be related toexperiences of having lived in a disorganisedhousehold. No relation was found betweenanxiety or depression and family factors.Recently, Decruyenaere et al52 found a low butsignificant correlation between the partici-pants’ age at which the parent showed the firstsymptoms and psychological functioning be-fore test disclosure. Psychological adjustmentto the test result was not correlated with the ageof the participant at onset of HD in the parent.

To identify adjustment problems in adultrisk carriers, childhood experiences and familydynamics need to be taken into account. In ouropinion, the attachment theory53 provides ameaningful theoretical framework for describ-ing childhood experiences in HD families andgenerating hypotheses concerning the influ-ence of childhood experiences on later adapta-tion.

ATTACHMENT THEORY

Attachment theory, developed by John Bowlby,postulates a universal human need to formclose aVectionate bonds. It is a normativetheory of how the “attachment system” func-tions in all humans.54 The attachment theoryconcerns the nature of early experiences ofchildren, and the impact of these experienceson aspects of later functioning. The centralassumption of attachment theory is thatindividual social behaviour may be understoodin terms of generic mental models of socialrelationships constructed by the person.55

These models, although constantly evolvingand subject to modification, are strongly influ-enced by the child’s experiences with theprimary caregivers. The attachment systemserves as a primary mechanism for the regula-tion of infant safety and survival and is highly

activated in times of danger.53 An infant is con-sidered securely attached if he or she regardsthe parents as people to rely on when facing afrightening situation. A responsive and sensi-tive way of parenting generally gives rise to asecure attachment pattern. Secure infants areable to explore new situations and to experi-ence proximity and comfort in times ofdistress, illness, or tiredness. Insecure attach-ment is often found in those who in childhoodhave experienced rejection or neglect by one orboth parents, or who were asked to take care ofthe parent instead of being taken care of.56

Insecure attachment in either infancy or adult-hood is related to the occurrence of psychopa-thology in adulthood.57

Three types of insecure attachment can bediscerned. An avoidant (dismissing) attachedinfant shifts attention away from rejectingcaregivers and minimises displays of distress.An ambivalent (preoccupied) attached infant ishighly focused on the caregiver and maximisesdistress through insistent demands for care andattention. A third group of infants appear toexhibit a range of seemingly undirected behav-ioural responses giving the impression of disor-ganisation and disorientation.58 These infantsmay display (momentarily) bizarre and contra-dictory behaviour. Frightening experienceswith caregivers who behaved in threatening,frightened, or dissociated ways and experiencesof loss and trauma may lead to a disorganisedattachment pattern.59 60 It is generally held thatfor such infants the caregiver has served both asa source of fear and as a source of reassurance,thus the arousal of the attachment behaviouralsystem produces strong conflicting motiva-tions. Not surprisingly, a history of severeneglect or physical or sexual abuse is oftenassociated with the manifestation of thispattern.61 62 It is generally held that the pattern-ing of attachment related behaviour is under-pinned by diVerent strategies adopted by chil-dren to regulate their emotional reactions.63 AsaVect regulation is acquired with the help of thechild’s primary caregiver, the child’s strategywill be inevitably a reflection of the caregiver’sbehaviour towards him/her. Established attach-ment patterns or working models guide aperson’s response to frightening situations andinterpretation of the caregiver’s response.

The stability of early childhood attachmentpatterns is well demonstrated. During develop-ment from infancy to childhood, attachmentworking models become diYcult to change.However, current experiences with attachmentfigures continue to influence the attachmentworking model.64–66

How can the attachment theory be relevantfor people dealing with Huntington’s disease?We speculate that the presence of HD in afamily involves specific stressors, which mightinfluence the attachment relationship betweenparents and their children for diVerent reasons.First, the aVected parent in the onset phase ofHD may become preoccupied with the diagno-sis, their own future, and the frightening recol-lections of his/her parent or other relativesgoing through the HD disease progression. As

858 Letters

www.jmedgenet.com


http://jmg.bmj.com/


the disease progresses, the patient is less recep-tive to the questions of the children and maybecome depressive or aggressive. These moodand personality changes, together with thechoreic movements, may frighten or alienatetheir oVspring.67 Second, the disease may leadto changes in the family system. The unaf-fected parent will experience a change inresponsibilities and dependency of the spousein the relationship; the aVected spouse be-comes a person who insidiously needs care.Some healthy partners may feel unable to takeup this task and will leave the household.Changes in the household may lead to neglectof the children. Some children may take up thecare of the ill parent. The unaVected parentmay seek one of the children as a substitutepartner.48. Third, the fact that the children areat risk for developing HD also puts stress onparent-child bonding. The parents may beconcerned about the carrier status of the childand may have feelings of guilt by having passedon the gene. Knowing that their children mayget the disease can also create an emotionaldistance.67 Some parents also have predictionsor even fantasies about their children, thinkingthat they may or may not develop HD.68 Thehealthy parent often has the diYcult task ofrearing these children and informing themabout their risk without the help of thepartner.69 70 With regard to testing, having chil-dren is an additional stress factor for thehealthy parent.20 To summarise, a familyburdened by a genetic disorder may have todeal with several types of loss: loss of the physi-cal capacity of the aVected person, loss of his orher own personality, loss of the old family sys-tem, and loss through death. This may beaccompanied by shame, secretiveness, andsocial isolation.

Mental representations of attachment inadults are assessed by means of the AdultAttachment Interview (AAI).71 The AAI askssubjects about childhood attachment relation-ships and the meaning which a personcurrently gives to attachment experiences. Theinstrument is rated according to the scoringsystem developed by Main and Goldwyn72

which classifies people into Secure/Autonomous, Insecure/Dismissing and super-imposed on these Insecure/Preoccupied,Unresolved/Disorganised with respect to lossor trauma, categories according to the struc-tural qualities of their reports of early experi-ences. The assessment of attachment by meansof coherence43 may help to overcome problemswith self-report measures in previous studies.Next to the attachment representation, theAdult Attachment Interview generates infor-mation about the psychodynamics and de-fences in a person.

Although Huntington’s disease is a highlydramatic disorder with an increased chance forchildren to become traumatised, we speculatethat the findings are also important for othergenetic disorders that have been passed on toconsecutive generations. This does not need tobe restricted to autosomal dominant, late onsetdiseases with full or partial penetrance, but alsofor X linked disorders and diseases with

recessive inheritance patterns. There is furtherimportant evidence that attachment relation-ships may play a key role in the transgenerationaltransmission of hardship and deprivation.People categorised as secure are three or fourtimes more likely to have children who aresecurely attached to them.60 This turns out to betrue even in prospective studies where parentalattachment is assessed before the birth of thechild.55 73–75 These findings also emphasise theimportance of quality of parenting in determin-ing the child’s attachment classification. Investi-gation of the attachment relationship in HDfamilies and its influence on adult functioningmay contribute to a greater understanding ofearlier research findings and serve to improvegenetic counselling and intervention.

1 Harper P. Huntington’s disease. London: WB Saunders,1996.

2 Huntington’s Disease Collaborative Research Group. Anovel gene containing a trinucleotide repeat that isexpanded and unstable on Huntington’s disease chromo-somes. The Huntington’s Disease Collaborative ResearchGroup. Cell 1993;72:971-83.

3 Roos RA, Vegter-van der Vlis M, Hermans J, Elshove HM,Moll AC, van de Kamp JJ, Bruyn GW. Age at onset inHuntington’s disease: eVect of line of inheritance andpatient’s sex. J Med Genet 1991;28:515-19.

4 Bloch M, Fahy M, Fox S, Hayden MR. Predictive testing forHuntington disease. II. Demographic characteristics,life-style patterns, attitudes, and psychosocial assessmentsof the first fifty-one test candidates. Am J Med Genet 1989;32:217-24.

5 Brandt J, Quaid KA, Folstein SE, Garber P, Maestri NE,Abbott MH, Slavney PR, Franz ML, Kasch L, KazazianHH. Presymptomatic diagnosis of delayed-onset diseasewith linked DNA markers. The experience in Huntington’sdisease. JAMA 1989;261:3108-14.

6 Codori AM, Brandt J. Psychological costs and benefits ofpredictive testing for Huntington’s disease. Am J MedGenet 1994;54:174-84.

7 Codori AM, Slavney PR, Young C, Miglioretti DL, BrandtJ. Predictors of psychological adjustment to genetic testingfor Huntington’s disease. Health Psychol 1997;16:36-50.

8 Decruyenaere M, Evers-Kiebooms G, Boogaerts A, Cassi-man JJ, Cloostermans T, Demyttenaere K, Dom R, FrynsJP, Van den Berghe H. Prediction of psychologicalfunctioning one year after the predictive test for Hunting-ton’s disease and impact of the test result on reproductivedecision making. J Med Genet 1996;33:737-43.

9 DudokdeWit AC, Duivenvoorden HJ, Passchier J, Niermei-jer MF, Tibben A. Course of distress experienced bypersons at risk for an autosomal dominant inheritable dis-order participating in a predictive testing program: anexplorative study. Rotterdam/Leiden Genetics Workgroup.Psychosom Med 1998;60:543-9.

10 DudokdeWit AC, Tibben A, Duivenvoorden HJ, NiermeijerMF, Passchier J. Predicting adaptation to presymptomaticDNA testing for late onset disorders: who will experiencedistress? Rotterdam Leiden Genetics Workgroup. J MedGenet 1998;35:745-54.

AppendixAQ, Attitude QuestionnaireBDI, Beck Depression Inventory (BDI)76

BHS, Beck Hopelessness Scale (BHS)77

EPI, Eysenck Personality InventoryFACES, Family Adaptability and Cohesion Evalua-tion ScalesGHQ, General Health Questionnaire30

GSI, Global Severity IndexGWS, General Wellbeing Scale78

HADS, Hospital Anxiety Depression Scale79

IES, Impact of Event Scale46

MCMI-2 , Millon Clinical Multiaxial Inventory II80

MHI, Mental Health IndexMMPI, Minneapolis Multiphasic Personality Inven-tory81

PSDI, Positive Symptom Distress Index (SCL-90)PST, Positive Symptom Total (SCL-90)SCL-90, Symptom Checklist 9082

SSQ, Social Support QuestionnaireSTAI, State Trait Anxiety Questionnaire83

UCL, Utrecht Coping List84

Letters 859

www.jmedgenet.com


http://jmg.bmj.com/


11 Folstein SE, Franz ML, Jensen BA, Chase GA, Folstein MF.Conduct disorder and aVective disorder among theoVspring of patients with Huntington’s disease. PsycholMed 1983;13:45-52.

12 Folstein SE. Huntington’s disease. A disorder of families.Baltimore: The Johns Hopkins University Press: 1991.

13 Lawson K, Wiggins S, Green T, Adam S, Bloch M, HaydenMR. Adverse psychological events occurring in the firstyear after predictive testing for Huntington’s disease. TheCanadian Collaborative Study Predictive Testing. J MedGenet 1996;33:856-62.

14 Meissen GJ, Myers RH, Mastromauro CA, Koroshetz WJ,Klinger KW, Farrer LA, Watkins PA, Gusella JF, Bird ED,Martin JB. Predictive testing for Huntington’s disease withuse of a linked DNA marker. N Engl J Med 1988;318:535-42.

15 Quaid KA, Wesson MK. Exploration of the eVects ofpredictive testing for Huntington disease on intimate rela-tionships. Am J Med Genet 1995;57:46-51.

16 Tibben A, Vegter-van der Vlis M, Skraastad MI, Frets PG,van der Kamp JJ, Niermeijer MF, van Ommen GJ, RoosRA, Rooijmans HG, Stronks D, Verhage F. DNA-testingfor Huntington’s disease in The Netherlands: a retrospec-tive study on psychosocial eVects. Am J Med Genet1992;44:94-9.

17 Tibben A, Duivenvoorden HJ, Vegter-van der Vlis M, Nier-meijer MF, Frets PG, van de Kamp JJ, Roos RA,Rooijmans HG, Verhage F. Presymptomatic DNA testingfor Huntington disease: identifying the need for psycho-logical intervention. Am J Med Genet 1993;48:137-44.

18 Tibben A, Frets PG, van de Kamp JJ, Niermeijer MF,Vegter-van der Vlis M, Roos RA, Rooymans HG, vanOmmen GJ, Verhage F. On attitudes and appreciation 6months after predictive DNA testing for Huntingtondisease in the Dutch program. Am J Med Genet1993;48:103-11.

19 Tibben A, Duivenvoorden HJ, Niermeijer MF, Vegter-vander Vlis M, Roos RA, Verhage F. Psychological eVects ofpresymptomatic DNA testing for Huntington’s disease inthe Dutch program. Psychosom Med 1994;56:526-32.

20 Tibben A, Timman R, Bannink EC, Duivenvoorden HJ.Three-year follow-up after presymptomatic testing forHuntington’s disease in tested individuals and partners.Health Psychol 1997;16:20-35.

21 Wiggins S, Whyte P, Huggins M, Adam S, Theilmann J,Bloch M, Sheps SB, Schechter MT, Hayden MR. The psy-chological consequences of predictive testing for Hunting-ton’s disease. N Engl J Med 1992;327:1401-5.

22 Tibben A, Frets PG, van de Kamp JJ, Niermeijer MF,Vegter-van der Vlis M, Roos RA, van Ommen GJ, Duiven-voorden HJ, Verhage F. Presymptomatic DNA-testing forHuntington disease: pretest attitudes and expectations ofapplicants and their partners in the Dutch program. Am JMed Genet 1993;48:10-16.

23 Brandt J, Quaid KA, Folstein SE. Presymptomatic DNAtesting for Huntington’s disease. J Neuropsychiatry ClinNeurosci 1989;1:195-7.

24 Quaid K. Presymptomatic testing for Huntington disease:recommendations for counseling. J Genet Couns 1992;1:277-302.

25 Evers-Kiebooms G. Predictive testing for Huntington’s dis-ease in Belgium. J Psychosom Obstet Gynecol 1990;11:61-72.

26 Tibben A. What is knowledge but grieving? On psychologicaleVects of presymptomatic DNA-testing for Huntington’s disease.PhD thesis. Erasmus University, Rotterdam, 1993.

27 Fox S, Bloch M, Fahy M, Hayden MR. Predictive testing forHuntington disease. I. Description of a pilot project inBritish Columbia. Am J Med Genet 1989;32:211-16.

28 Endicott J, Spitzer RL. A diagnostic interview: the schedulefor aVective disorders and schizophrenia. Arch GenPsychiatry 1978;35:837-44.

29 DudokdeWit AC, Tibben A, Duivenvoorden HJ, NiermeijerMF, Passchier J, Trijsburg RW. Distress in individuals fac-ing predictive DNA testing for autosomal dominantlate-onset disorders: comparing questionnaire results within-depth interviews. Am J Med Genet 1998;75:62-74.

30 Tarnopolsky A, Hand DJ, McLean EK, Roberts H, WigginsRD. Validity and uses of a screening questionnaire (GHQ)in the community. Br J Psychiatry 1979;134:508-15.

31 Decruyenaere M, Evers-Kiebooms G, Boogaerts A, Cassi-man JJ, Cloostermans T, Demyttenaere K, Dom R, FrynsJP, van den Berghe H. Predictive testing for Huntington’sdisease: risk perception, reasons for testing and psychologi-cal profile of test applicants. Genet Couns 1995;6:1-13.

32 Tibben A, Roos RA, Niermeijer MF. Psychologicalconsequences of presymptomatic testing for Huntington’sdisease. Lancet 1997;349:809.

33 Huggins M, Bloch M, Wiggins S, Adam S, Suchowersky O,Trew M, Klimek ML, Greenberg CR, EleV M, ThompsonLP, Knight J, MacLeod P, Girard K, Theilmann J, HedrickA, Hayden MR. Predictive testing for Huntington diseasein Canada: adverse eVects and unexpected results in thosereceiving a decreased risk. Am J Med Genet 1992;42:508-15.

34 Almqvist EW, Bloch M, Brinkman R, Craufurd D, HaydenMR. A worldwide assessment of the frequency of suicide,suicide attempts, or psychiatric hospitalization after predic-tive testing for Huntington disease. Am J Hum Genet 1999;64:1293-304.

35 Meiser B, Dunn S. Psychological impact of genetic testingfor Huntington’s disease: an update of the literature. J Neu-rol Neurosurg Psychiatry 2000;69:574-8.

36 Maat-Kievit A, Vegter-van der Vlis M, Zoeteweij M, Losek-oot M, van Haeringen A, Roos RA. Paradox of a better testfor Huntington’s disease. J Neurol Neurosurg Psychiatry2000;69:579-83.

37 Kessler S. Predictive testing for Huntington disease: apsychologist’s view. Am J Med Genet 1994;54:161-6.

38 van der Steenstraten IM, Tibben A, Roos RA, van de KampJJ, Niermeijer MF. Predictive testing for Huntingtondisease: nonparticipants compared with participants in theDutch program. Am J Hum Genet 1994;55:618-25.

39 Decruyenaere M, Evers-Kiebooms G, Boogaerts A, Cloost-ermans T, Cassiman JJ, Demyttenaere K, Dom R, Fryns JP,van den Berghe H. Non-participation in predictive testingfor Huntington’s disease: individual decision-making,personality and avoidant behaviour in the family. Eur JHum Genet 1997;5:351-63.

40 Tibben A, Niermeijer MF, Roos RA, Vegter van de Vlis M,Frets PG, van Ommen GJ, van de Kamp JJ, Verhage F.Understanding the low uptake of presymptomatic DNAtesting for Huntington’s disease. Lancet 1992;340:1416.

41 Shedler J, Mayman M, Manis M. The illusion of mentalhealth. Am Psychol 1993;48:1117-31.

42 Zilberg NJ, Weiss DS, Horowitz MJ. Impact of Event Scale:a cross-validation study and some empirical evidence sup-porting a conceptual model of stress response syndromes. JConsult Clin Psychol 1982;50:407-14.

43 Grice HP. Logic and conversation. In: Cole P, Moran JL,eds. Syntax and semantics. New York: Academic Press,1975:41-58.

44 Dozier M, Lee SW. Discrepancies between self- andother-report of psychiatric symptomatology: eVects of dis-missing attachment strategies. Dev Psychopathol 1995;7:217-26.

45 Tibben A, Vegter-vd Vlis M, vd Niermeijer MF, Kamp JJ,Roos RA, Rooijmans HG, Frets PG, Verhage F. Testing forHuntington’s disease with support for all parties. Lancet1990;335:553.

46 Horowitz M, Wilner N, Alvarez W. Impact of Event Scale: ameasure of subjective stress. Psychosom Med 1979;41:209-18.

47 Salkovskis PM, Rimes KA. Predictive genetic testing:psychological factors. J Psychosom Res 1997;43:477-87.

48 Hans MB, Koeppen AH. Huntington’s chorea. Its impacton the spouse. J Nerv Ment Dis 1980;168:209-14.

49 Kessler S, Bloch M. Social system responses to Huntingtondisease. Fam Process 1989;28:59-68.

50 Kessler S. Forgotten person in the Huntington disease fam-ily. Am J Med Genet 1993;48:145-50.

51 Tibben A, Vegter-van der Vlis M, Roos RAC. Presymp-tomatische DNA-diagnostiek bij de chorea vanHuntington: reacties op de zekerheid niet-gendrager te zijn.Ned Tijdschr Geneeskd 1990;134:701-4.

52 Decruyenaere M, Evers-Kiebooms G, Boogaerts A, Cassi-man JJ, Cloostermans T, Demyttenaere K, Dom R, FrynsJP. Psychological functioning before predictive testing forHuntington’s disease: the role of the parental disease, riskperception, and subjective proximity of the disease. J MedGenet 1999;36:897-905.

53 Bowlby J. Attachment. 2nd ed. London: Penguin Books,1989.

54 Bowlby J. A secure base: parent-child attachment and healthyhuman development. New York: Basicbooks, 1988.

55 Fonagy P, Steele M, Moran G, Steele M, Higgitt AC. Thecapacity for understanding mental states: the reflective selfin parent and child and its significance for security ofattachment. Infant Ment Health J 1991;13:200-16.

56 Bowlby J. Developmental psychiatry comes of age. Am JPsychiatry 1988;145:1-10.

57 Greenberg MT. Attachment and psychopathology in child-hood. In: Cassidy J, Shaver PR, eds. Handbook ofattachment. New York: The Guilford Press, 1999:469-96.

58 Main M, Solomon J. Procedures for identifying infants asdisorganized/disoriented during the Ainsworth StrangeSituation. In: Greenberg MT, Cicchetti D, Cummings EM,eds. Attachment in the preschool years: theory, research, andintervention. Chicago: University of Chicago Press, 1990:121-60.

59 Schuengel C, Bakermans-Kranenburg MJ, van IjzendoornMH. Frightening maternal behavior linking unresolved lossand disorganized infant attachment. J Consult Clin Psychol1999;67:54-63.

60 van Ijzendoorn M. Adult attachment representations,parental responsiveness, and infant attachment: a meta-analysis on the predictive validity of the Adult AttachmentInterview. Psychol Bull 1995;117:387-403.

61 Cicchetti D, Beeghly M. Symbolic development in mal-treated youngsters: an organizational perspective. In:Cicchetti D, Beeghly M, eds. Atypical symbolic development.New Directions for Child Development. San Francisco:Jossey-Bass, 1987:5-29.

62 Main M, Hesse E. Parents’ unresolved traumatic experi-ences are related to infant disorganized attachment status:is frightened and/or frightening parental behavior the link-ing mechanism? In: Greenberg DCM, Cummings EM, eds.Attachment in the preschool years: theory, research andintervention. Chicago: University of Chicago Press, 1990:161-82.

63 Kobak R. The emotional dynamics of disruptions in attach-ment relationships: implications for theory, research, andclinical intervention. In: Cassidy J, Shaver PR, eds.Handbook of attachment. New York: The Guilford Press,1999:21-43.

860 Letters

www.jmedgenet.com


http://jmg.bmj.com/


64 Waters E, Merrick S, Treboux D, Crowell J, Albersheim L.Attachment security in infancy and early adulthood: atwenty-year longitudinal study. Child Dev 2000;71:684-9.

65 Hamilton CE. Continuity and discontinuity of attachmentfrom infancy through adolescence. Child Dev 2000;71:690-4.

66 Weinfield NS, Sroufe A, Egeland B. Attachment frominfancy to early childhood in a high-risk sample: continuity,discontinuity, and their correlates. Child Dev 2000;71:695-702.

67 Fanos JH. Developmental tasks of childhood andadolescence: implications for genetic testing. Am J MedGenet 1997;71:22-8.

68 Kessler S. Invited essay on the psychological aspects ofgenetic counseling. V. Preselection: a family coping strategyin Huntington disease. Am J Med Genet 1988;31:617-21.

69 Evers-Kiebooms G, Swerts A, Van Den Berghe H. Partnersof Huntington patients: implications of the disease andopinions about predictive testing and prenatal diagnosis.Genet Couns 1990;1:151-9.

70 Martindale B, Bottomley V. The management of familieswith Huntington’s chorea: a case study to illustrate somerecommendations. J Child Psychol Psychiatry 1980;21:343-51.

71 George C, Kaplan N, Main M. Adult attachment interview.3rd ed. Department of Psychology, University of Califor-nia, Berkeley, 1996.

72 Main M, Goldwyn R. Adult attachment scoring andclassification systems. University of California at Berkeley:unpublished manuscript, 1998.

73 Benoit D, Parker KCH. Stability and transmission ofattachment across three generations. Child Dev 1994;65:1444-57.

74 Steele H, Steele M, Fonagy P. Associations amongattachment classifications of mothers, fathers, and theirinfants. Child Dev 1996;67:541-55.

75 Ward MJ, Carlson EA. Associations among adult attach-ment representations, maternal sensitivity, and infant-mother attachment in a sample of adolescent mothers.Child Dev 1995;66:69-79.

76 Beck AT, Steer RA. Beck Depression Inventory Manual. SanAntonio, Texas: The Psychological Corporation, 1987.

77 Beck AT, Weissman A, Lester D, Trexler L. Themeasurement of pessimism: the hopelessness scale. J Con-sult Clin Psychol 1974;42:861-5.

78 Ware JEJ, Johnston SA, Davies-Avery A, Brook RH. Concep-tualization and measurement of health for adults in the HealthInsurance Study. Santa Monica, California: Rand, 1979.

79 Zigmond AS, Snaith RP. The hospital anxiety anddepression scale. Acta Psychiatr Scand 1983;67:361-70.

80 Millon T. Millon Clinical Multiaxial Inventory-II Manual.Minneapolis: National Computer Scoring Systems, 1987.

81 Graham JR. The MMPI. A practical guide. New York: OxfordUniversity Press, 1987.

82 Derogatis L. SCL-90R Manual-II. Towson, MD: ClinicalPsychometric Research, 1983.

83 Van der Ploeg HM, Defares PB, Spielberger CD. Handlei-ding bij de Zelfbeoordelingsvragenlijst: een Nederlandstaligebewerking van de Spielberger State Trait Anxiety Inventory.New York: Swets & Zeitlinger, 1980.

84 Schreurs PJ, Van de Willige G, Tellegen B, Brosschot JF. DeUtrechtse Coping Lijst. Omgaan met problemen en gebeurtenis-sen. Lisse: Swets & Zeitlinger, 1988.

A novel 3' mutation in the APC gene in a familypresenting with a desmoid tumour

Diana Eccles, John Harvey, Adrian Bateman, Fiona Ross

EDITOR—Desmoid tumours, also known asinfiltrative fibromatoses, are rare benign tu-mours which often recur after local resectionand can cause death through local infiltrationof vital structures.1 The estimated incidence inthe general population of such tumours is 1-2per million but in familial adenomatouspolyposis (FAP) they occur in up to 15% ofcases.2 Likely precipitating factors includetrauma and female sex hormones, sincefemales are more often aVected than males.3

The majority of desmoid tumours in FAP (over90%) arise in the mesentery of the bowel or inthe abdominal wall musculature. In recentyears, several families have been describedwhere the predominant phenotype is ofdesmoid disease and where the colonic pheno-type is minimal.3–6 We describe another suchfamily with a novel protein truncating mutationin the 3' end of the APC gene.

Methods and resultsCLINICAL DETAILS

The index case presented at 29 years of agewith a firm, slightly tender swelling within theright rectus abdominus muscle. A 6 x 5 cmtumour was locally excised and conventionalhistological examination showed infiltrativefibromatosis. The tumour recurred after sixyears and was resected again along with 30 cmof adherent small bowel. A year later, a furtherabdominal wall recurrence was resected and onthis occasion fresh tissue was submitted forcytogenetic analysis. Full colonoscopy beforereferral to the genetics service showed no

evidence of colonic adenomas throughout thecolon. Repeat colonoscopy after the genemutation was identified still failed to show anycolonic pathology, although contrast dye spraywas not undertaken on either occasion. Theonly relevant family history was that her fatherhad a previous history of a sigmoid colectomycarried out at 56 years of age for a carcinoma ofthe colon. He had been discharged from followup after seven years during which endoscopyhad shown no further pathology and he hasremained well and symptom free since. He hadnever had any palpable lumps. Review of thehistopathology from the resection specimenshowed a Dukes B adenocarcinoma of thecolon with six adenomas in the surroundingcolonic mucosa.

CYTOGENETIC ANALYSIS

Fresh desmoid tumour was subjected tocytogenetic analysis which showed normalfibroblasts mixed with abnormal cells showingan interstitial deletion involving chromosome5q22.

MOLECULAR ANALYSIS

DNA from the index case was examined formutations in the APC gene using denaturinghigh performance liquid chromatography(DHPLC). A single base substitution G>A wasidentified at nucleotide position 7511, codon2504, which changes tryptophan (TGG) to astop codon (TAG). The mutation was presentin both the index case and her father.

Letters 861

J Med Genet2001;38:861–863

Wessex ClinicalGenetics Service,Princess Ann Hospital,SouthamptonSO16 5YA, UKD Eccles

Wessex RegionalGenetics Laboratory,Salisbury DistrictHospital, Wilts, UKJ HarveyF Ross

Department ofHistopathology,Southampton GeneralHospital,Southampton, UKA Bateman

Correspondence to: DrEccles, [email protected]

www.jmedgenet.com


http://jmg.bmj.com/


IMMUNOHISTOCHEMISTRY

The colonic tumour and the desmoid tumourwere examined for â-catenin expression usingimmunohistochemistry. This showed intensenuclear and cytoplasmic â-catenin expressionin the colonic tumour. A similar, although lessintense, pattern of nuclear expression withsome cytoplasmic expression was apparent inthe desmoid tumour from the index case.

DiscussionClassical familial adenomatous polyposis coli(FAP) is recognised by the presence of manyhundreds of colonic adenomas throughout thecolon. Since the identification of the generesponsible for FAP in 1991, it has been possi-ble to recognise certain genotype/phenotypecorrelations which may not only allow theclinician to direct the molecular geneticisttowards the most likely region where amutation may be located, but conversely maypredict the likely natural course of the diseaseassociated with a given mutation and directmanagement strategies accordingly.7 Wheremutations fall into the 5' and 3' extremes of thegene, an attenuated colonic phenotype isusually seen with fewer and later onset polyps.8

Whereas mutations at the 5' end of the gene(exons 1-4) are not associated with desmoidtumour development, mutations arising in the3' half of the gene are more often associatedwith desmoid tumours, and in rare families thisis the dominant phenotype with little else tosuggest the diagnosis of polyposis on examina-tion of the colon.3 5 Although cytogeneticanalysis of desmoid tumours is technically dif-ficult and requires fresh tumour tissue, it maybe a useful additional investigation where thephenotype for FAP is absent. Cytogeneticdeletion of 5q in desmoid tumours is more fre-quently associated with FAP than apparentlysporadic tumours,9 10 so, even in the absence oftypical FAP features, a cytogenetic deletion of5q in the tumour may direct a careful search forpossible underlying APC gene mutations.

APC is involved in many critical cellularfunctions including cell adhesion (viaE-cadherin) and in negatively regulatingâ-catenin levels. â-catenin is involved in thetranscriptional activation of several genesinvolved in cell cycle entry and progression.APC mutation leads to reduced degradation ofâ-catenin (reflected in overexpression ofâ-catenin shown using immunohistochemistryin the tumours described in this case).

The natural history of desmoid tumours isvariable but they can follow a highly aggressivecourse. However, even in families where thedesmoid phenotype is 100% penetrant,3 5 theclinical course is unpredictable. The explana-tion for the particular predilection of 3'mutations to be associated with desmoidtumour formation is still a matter of specula-tion. Low levels of truncated protein producthave been reported in similarly distal muta-tions, possibly owing to an unstable mRNA;the stability of the mRNA may vary accordingto tissue type.3 5 There are likely to be bothenvironmental and genetic modifiers exertingan eVect on APC gene mutations in view of the

variation in phenotype between subjects withthe same mutation. It is possible that the eVectof these putative genetic modifiers will diVeraccording to tissue type. Detection of suchmodifiers and determination of their mode ofaction could be extremely helpful in selectingpotential medical treatment for patients withdesmoid tumours, since surgery is best avoidedif the tumours are not life threatening. A recentpaper looking at second hits in adenomas anddesmoid tumours in FAP included data onlyfor patients with germline mutations 5' ofcodon 1462. In desmoid tumours from thesepatients, there was either loss of the wild typeallele or a somatic APC mutation. The somaticmutations were consistently 3' of codon 1426.In our patient, there was loss of heterozygosityindicated by a cytogenetic deletion but a 3'germline mutation. This suggests that thesiting of first and second hits may beinterchangeable with respect to the end result(a desmoid tumour), but if somatic mutationsin colonic epithelium are in some way relianton the germline mutation this may explain thevery diVerent clinical phenotype presented bymany patients with unusually 3' mutations.11

For families where the desmoid phenotype isthe dominant clinical feature and the risk ofcolorectal cancer is likely to be lower than whereadenomatous polyps are frequent, colonic resec-tion should be reserved for situations wherethere is significant colonic pathology. There maybe an argument for favouring subtotal colec-tomy and ileorectal anastomosis if colectomybecomes necessary. Giving clear information toat risk members of the family reported here isdiYcult, as the position of the mutation leads usto suspect a high risk for desmoid tumoursalthough the clinical picture is not of 100%desmoid tumour penetrance. Although thecolon cancer risk must also be increased, theposition of the mutation in the family described,the association of desmoid tumours with trauma(and surgery in particular), the late onset, andthe presence of only a few adenomas in the onlyidentified gene carrier with colon cancer and thenormal appearance of the colonic mucosa in theindex case in her late thirties all suggest that aconservative approach to surgery is appropriate.We have oVered genetic testing to at risk familymembers and suggested that a screening colon-oscopy every two years starting at 20 years, withdye spray or random mucosal biopsies to deter-mine whether any microadenomas are present,would be a reasonable course of action for a genecarrier or for a family member at 50% risk whochose not to have a predictive genetic test.

We know this disease is caused by a mutationin the APC gene and could therefore be calledfamilial adenomatous polyposis (FAP) orperhaps better attenuated adenomatous poly-posis coli (AAPC). This acknowledges thatthere is an increased colon cancer risk;however, this potentially underplays the signifi-cance of the desmoid tumour phenotype whichmay be a very important factor to take intoaccount in making decisions about surgery.

We would like to thank Cdr P Barker for referring the patientand Dr Jane Collins for providing the â-catenin antibody.

862 Letters

www.jmedgenet.com


http://jmg.bmj.com/


1 Smith AJ, Lewis JJ, Merchant NB, Leung DH, WoodruVJM, Brennan MF. Surgical management of intra-abdominaldesmoid tumours. Br J Surg 2000;87:608-13.

2 Klemmer S, Pascoe L, DeCosse J. Occurrence of desmoidsin patients with familial adenomatous polyposis of thecolon. Am J Med Genet 1987;28:385-92.

3 Eccles DM, Van der Luijt RB, Breukel C, Bullman H, Bun-yan DJ, Fisher A, Barber J, Du BC, Primrose J, Burn J,Fodde R. Hereditary desmoid disease is a phenotypic vari-ant of attenuated polyposis due to a 3' frameshift mutationat codon 1924 of the APC gene. Am J Hum Genet 1996;59:1193-201.

4 Scott RJ, Froggatt NJ, Trembath R, Evans DGR, HodgsonSV, Maher ER. Familial infiltrative fibromatosis (desmoid

tumours) (MIM135290) caused by a recurrent 3' APCgene mutation. Hum Mol Genet 1996;5:1921-4.

5 Couture J, Mitri A, Lagace R, Smits R, Berk T, BouchardHL, Fodde R, Alman B, Bapat B. A germline mutation atthe extreme 3' end of the APC gene results in a severedesmoid phenotype and is associated with overexpressionof beta-catenin in the desmoid tumor. Clin Genet 2000;57:205-12.

6 Halling KC, Lazzaro CR, Honchel R, Bufill JA, Powell SM,Arndt CA, Lindor NM. Hereditary desmoid disease in afamily with a germline Alu I repeat mutation of the APCgene. Hum Hered 1999;49:97-102.

7 Vasen H, Van der Luijt RB, Slors JFM, Buskens E, de RuiterP, Baeten CGM, Shouten WR, Oostvogel HJM, KuijpersJHC, Tops C, Meera Khan P. Molecular genetic tests as aguide to surgical management of familial adenomatouspolyposis. Lancet 1996;348:433-5.

8 Spirio L, Olschwang S, Groden J, Robertson M, SamowitzW, Joslyn G, Gelbert L, Thliveris A, Carlson M, Otterud B.Alleles of the APC gene: an attenuated form of familialpolyposis. Cell 1993;75:951-7.

9 Bridge JA, Meloni AM, NeV JR, Deboer J, Pickering D,Dalence C, JeVrey B, Sandberg AA. Deletion 5q indesmoid tumor and fluorescence in situ hybridization forchromosome 8 and/or 20 copy number. Cancer GenetCytogenet 1996;92:150-1.

10 Giarola M, Wells D, Mondini P, Pilotti S, Sala P, Azzarelli A,Bertario L, Pierotti MA, Delhanty JD, Radice P. Mutationsof adenomatous polyposis coli (APC) gene are uncommonin sporadic desmoid tumours. Br J Cancer 1998;78:582-7.desmoid tumor. Clin Genet 2000;57:205-12.

11 Lamlum H, Papadopoulou A, Ilyas M, Rowan A, Gillet C,Hanby A, Talbot I, Bodmer W, Tomlinson I. APCmutations are suYcient for the growth of early colorectaladenomas. Proc Natl Acad Sci USA 2000;97:2225–8.

A silent mutation in exon 14 of the APC gene isassociated with exon skipping in a FAP family

Mariapina Montera, Francesca Piaggio, Cristiana Marchese, Viviana Gismondi,Alessandro Stella, Nicoletta Resta, Liliana Varesco, Ginevra Guanti, Cristina Mareni

EDITOR—Familial adenomatous polyposis(FAP) is an autosomal dominantly inheriteddisorder characterised by the development ofhundreds to thousands of adenomatous polypsin the colon and rectum. If left untreated, thereis a very high risk of colorectal cancer. Adeno-matous polyps may also develop proximally inthe stomach and the distal part of theduodenum. FAP is also associated with a vari-ety of extracolonic benign and malignantmanifestations, including congenital hypertro-phy of the retinal pigment epithelium(CHRPE), dental abnormalities, desmoid tu-mours, osteomas, epidermoid cysts, hepatob-lastoma, and thyroid neoplasia.1 Germlinemutations of the APC gene localised onchromosome 5q21.22 are responsible forFAP.2 3 APC is a tumour suppressor geneencoding a 2843 amino acid protein, whichcontains multiple functional domains andwhich mediates growth regulatory signals by itsassociation with a variety of cytoplasmicproteins. More than 300 diVerent APC muta-tions have so far been identified distributedthroughout the whole gene, with a higher con-centration in the 5' part of exon 15 (codons713-1597).4 5 The majority of mutations arepredicted to introduce premature terminationsignals resulting from single nucleotide altera-tions, small insertions or deletions, or splice

site mutations that lead to truncation of thenormal protein product.4 Missense mutationshave rarely been reported and their functionalimplications are often unclear.6 7 Larger dele-tions and insertions have been described, aswell as genomic rearrangements resulting fromrecombinations mediated by Alu elementswhich cause inappropriate exon splicing.8–10

Isoforms of APC transcripts lacking exon 9,exon 10A, and exon 14 encoded sequenceshave been reported.2 11–13 Isoforms lacking exon9 or exon 14 owing to splice site mutationshave also been associated with a FAPphenotype.14–17

In this study, we describe a G→T transver-sion at nucleotide position 1869 in exon 14which gives rise to a silent mutation, since bothnormal and mutated alleles encode an arginineresidue at codon 623. This exonic mutationinduces complete skipping of exon 14, leadingto truncated APC protein and resulting in aFAP phenotype.

Materials and methodsThe index case of family GE08 (IV.1, fig 1,table 1) was referred to our institution forgenetic counselling. The patient underwenta total colectomy for diVuse polyposis. Adiagnosis of FAP was made on the basis offamily history and histopathological results.

+ A diagnosis of a desmoid tumour shouldlead to a careful clinical and molecularsearch for evidence of familial adenoma-tous polyposis.

+ Cytogenetic or molecular evidence of lossof an APC allele in the desmoid tumour isadditional evidence of the likelihood ofan underlying germline mutation.

+ Mutations in the 3' part of the APC genecan be associated with a very attenuatedcolonic phenotype not clinically diagnos-tic of FAP.

Letters 863

J Med Genet2001;38:863–867

Dipartimento diMedicina Interna,Università di Genova,Viale Benedetto XV/6,16132 Genova, ItalyM MonteraF PiaggioC Mareni

Ospedale MaurizianoUmberto I, Torino,ItalyC Marchese

Istituto Nazionale perla Ricerca sul Cancro,Genova, ItalyV GismondiL Varesco

Dipartimento diMedicina Interna e delLavoro, Sezione diGenetica Medica,Università di Bari,ItalyA StellaN RestaG Guanti

Correspondence to:Dr Mareni, [email protected]

www.jmedgenet.com


http://jmg.bmj.com/


After informed consent had been obtainedfrom numerous family members, medicalrecords were reviewed to confirm the diagnosisof polyposis, polyposis and cancer, and the ageof occurrence. The clinical information ob-tained is shown in table 1. The pedigree wasconstructed as shown in fig 1. Peripheral bloodsamples were obtained from the proband andfamily members III.5, IV.1, IV.2, IV.3, IV.4,

IV.5, IV.6, IV.15, IV.16, IV.18, IV.19, IV.20,IV.21, IV.22, IV.26, V.1, V.2, V.3, V.4, V.11, andV.12.

Mutation analysis was carried out on ge-nomic DNA prepared from isolated leucocytesusing SNAP Whole Blood DNA Isolation Kit(Invitrogen, Carlsbad, CA). DNA was alsoextracted from paraYn embedded normaltissue from subject III.5.18 The protein trunca-tion test (PTT) was performed to detect trun-cating mutations in exon 15.19 Mutationscreening of the entire APC coding sequencewas performed on genomic DNA using singlestrand conformation polymorphism (SSCP).Thirty-seven diVerent segments of the APCgene were amplified by PCR using the primerpairs reported elsewhere.2 SSCP was carriedout as described previously.18 The variant con-former was confirmed in at least three diVerentsamples from the same person and by directDNA sequence analysis using an ABI PRISMTM 377 DNA Sequencer (Perkin Elmer, Fos-ter City, CA).

mRNA was isolated from IL-2 transformedlymphoblastoid cell lines by using the Micro-Fast Track mRNA Isolation Kit (Invitrogen,Carlsbad, CA). First strand cDNA was synthe-sised with 2 µg of polyA+ RNA in a reactionmixture containing 1.5 mmol/l MgCl2, 10mmol/l Tris HCl, 50 mmol/l KCl (pH 8.3), 200µmol/l each dNTP, 200 U of M-MLV ReverseTranscriptase (Ambion, Austin, TX), 25 URNase OUT (Life Technologies, Gaithers-burg, MD), 50 µmol/l of primer RV7-A20 andprimer 15A-RP,2 and 0.25 U Taq polymerase ina volume of 50 µl. Primers RV7-A and 15A-RPenabled the APC region corresponding tocodon 493 to 759 to be amplified. The PCRproducts were visualised by gel electrophoresison a 1.5% agarose gel stained with ethidiumbromide. The band intensity was measured

Table 1 Characteristics and clinical data from FAP family GE08

SubjectsClinicalpresentation*‡

Age at diagnosispolyps/cancer§

Number ofadenomas

Exon 14mutation

II.1 †Symptomatic ? ? NTII.3 †Symptomatic ? ? NTIII.1 †Symptomatic ? ? NTIII.4 †Symptomatic ? ? NTIII.5 Symptomatic 61/61 >100 PresentIV.1 Symptomatic 46 >100 PresentIV.2 Call up 44 0 AbsentIV.3 Call up 42 0 AbsentIV.4 Symptomatic 40 6 PresentIV.5 Call up 36 0 AbsentIV.6 Call up 34 0 AbsentIV.9 NK 34 NK NTIV.10 Symptomatic 44 7 PresentIV.11 Call up 36 0 AbsentIV.12 Call up 40 0 AbsentIV.13 Call up 38 0 PresentIV.14 Symptomatic 34/34 >100 PresentIV.15 Call up 39 1 AbsentIV.16 Symptomatic 36/36 >100 PresentIV.17 NK 32 NK NTIV.18 NK 31 NK NTIV.19 NK 29 NK NTIV.20 Call up 25 0 AbsentV.1 Call up 15 0 PresentV.2 Call up 13 0 AbsentV.3 NK 11 NK AbsentV.4 NK 10 NK AbsentV.6 NK 11 NK PresentV.7 NK 10 NK AbsentV.8 NK 2 NK NT

*Symptomatic, patients showing symptoms of FAP and/or CRC.†Symptomatic, no data available.‡No extracolonic manifestations were present in aVected subjects.§UnaVected subjects and asymptomatic gene carriers, actual age, and age at DNA test are given.Call up, called for endoscopic examination as at risk subjects; NK, not known; NT, not tested.

Figure 1 Pedigree of FAP family GE08. Grey symbols, colon polyposis; half filled symbols, colon cancer; open symbols,clinically unaVected subjects; G, the wild type nucleotide; G→T, R623R mutation; NT, not tested for mutation; numberunder symbols, age; III.6 and III.7, disease phenotypic unknown.

1

I

II

III

IV

V

2

2 6 8

1

46G→T

44G

42G

40G→T

36G

34G

15G→T

13G

11G

10G

34NT

44G→T

36G

40G

38G→T

34G→T

39G

36G→T

32NT

31NT

29NT

25G

11G→T

10G

2NT

61G→T

2 4 6 11 12 15 16 17 197

1

58 66 45 36 54

60

43 5

1 2 3

7

3 5 9 10 14 18 20138

6 7 83 41 2 5

? ?

864 Letters

www.jmedgenet.com


http://jmg.bmj.com/


directly on the agarose gel by Gene Snap4.00.00 and Gene Tools Version 3.00.13 SynGene software (Synoptics Ltd, Cambridge,UK). Amounts of each mRNA were expressedboth as absolute values and as the ratiobetween the 793 band, which corresponds tocDNA encompassing the APC exon 11-15A,and the 578 band which corresponds to thealternatively spliced transcript lacking exon 14.GAPDH gene coamplification was carried outas an internal control. Moreover, the PCRproducts were separated by 6% silver stainedpolyacrylamide gel electrophoresis. The bandsof 793 and 578 nucleotides were cut oV andeluted in water overnight at 37°C. One µl ofeach mixture was then used to amplify the twobands separately with primers RV7-A and15A-RP. The identity of the PCR products wasascertained by direct sequencing. A stretch ofintronic 40 nucleotides upstream of exon 14was also sequenced for analysing the acceptorsite branch point.

Protein extraction and western blot analysiswere performed according to Gismondi et al21

using 10% SDS PAGE gel. EBV transformedcells of subjects IV.1, IV.16, a normal control,and a control carrying a nonsense mutation atcodon 563 were analysed.

ResultsThe index patient had a clinical diagnosis ofFAP and underwent colectomy at the age of 46.The pedigree of family GE08 (fig 1) showssome peculiarities regarding the age of occur-rence and the multiplicity of polyps. The age atwhich polyps and cancer appeared is later thanin classical polyposis, except for IV.14 andIV.16 (table 1). These patients belonged to abranch of the family in which a consanguine-ous marriage had taken place between firstcousins, both possible carriers of the disease.The number of adenomatous polyps rangedfrom six to more than 100 in aVected members.No upper gastrointestinal tract lesions,CHRPE, or other extracolonic manifestationsare present in the family.

During mutation screening of the APC cod-ing region, a G→T transversion at nucleotideposition 1869 was detected by SSCP and directsequencing of exon 14 in the index patientIV.1. This mutation changes codon 623 from

CGG to CGT, both encoding arginine(R623R).22 Since the entire open reading frameof the APC gene was analysed by SSCP andPTT several times without detecting any addi-tional sequence variation, this is the onlydetectable DNA alteration unique to thispatient.

The sequence at the branch point of the exon14 acceptor site also did not show anyalteration. Subsequent screening for the muta-tion in 21 additional relatives showed the pres-ence of the mutation in three successivegenerations, as shown in fig 1, and its segrega-tion with the disease. In patient III.5, themutation was detected on paraYn embeddedtissue. Direct sequencing of DNA from pa-tients IV.14 and IV.16 did not show homozy-gosity for the R623R mutation. This allelicvariant was not observed in any of 100 DNAsamples from normal controls.

Messenger RNA purified from transformedlymphoblastoid cell lines from three aVectedsubjects (IV.1, IV.4, IV.16) and from three nor-mal controls was examined. The analysis ofcDNA with primer sets RV7-A and 15A-RPshowed the two isoforms representing themRNA transcript containing exon 14 and thealternatively spliced transcript lacking exon 14leading to a stop codon in exon 15A.

The densitometric ratio between the twoisoforms of 793 and 578 bp amplicons wascomparable in all normal controls, being 4.2.By contrast, the band intensity of 578 bp wasgreater in the three patients examined, with adensitometric ratio of 0.6 (p<0.03) showing adrastic increase in the alternatively spliced iso-form of exon 14 (fig 2A).

Direct DNA sequencing of the two bandsshowed only the wild type allele in the 793 bpmRNA isoform containing the normal splicedRNA while the G→T transversion was notpresent. In the retrotranscripted, short 578 bpisoform, the mRNA showed a complete lack ofexon 14.

Western blot analysis was performed on pro-tein extracts from a normal control and fromsubjects IV.1 and IV.16. In addition, a FAPpatient with a nonsense mutation at codon 563was analysed as an APC protein truncatedcontrol (fig 2B). Using 10% SDS-PAGEanalysis, the APC wild type protein was not

Figure 2 (A) Agarose gel electrophoresis showing the expression of the two exon 14 isoforms of 793 and 578 bp; ÖX174,DNA molecular marker; C, normal controls; IV.4, IV.1, IV.16, aVected subjects. (B)Western blot analysis showing the 64kDa APC truncated protein in patients IV.1 and IV.16 and the 62 kDa APC protein in the truncated control in M; C isthe normal not truncated protein.

A

C C C

793 bp APC

φx 1

74

IV-4

IV-1

IV-1

6

GAPDH

578 bp

240 bp

B

C M

116 kDa

IV-1

IV-1

6

84 kDa

58 kDa

48 kDa

36 kDa

Letters 865

www.jmedgenet.com


http://jmg.bmj.com/


detectable; the only truncated protein wasobserved in subjects IV.1 and IV.16 and in theAPC mutated control indicating that themutant mRNA is translated into a stable trun-cated protein. No truncated APC protein wasdetected in the normal control.

DiscussionThe exonic silent mutation R623R reportedhere creates a complete skipping of exon 14 ofthe APC gene and gives rise to a stabletruncated APC protein. This mutation has notbeen described previously and does not repre-sent a polymorphic gene variant as it has beentested on more than 100 subjects. Thismutation occurs as the only DNA change in allaVected members of a large three generationFAP family. Clinically, the family is character-ised by an unaggressive form of the disease interms of age of onset and number of polyps(table 1, fig 1), except for the branch derivedfrom a consanguineous marriage, both parentsbeing possible carriers of FAP. This diVerencemight occur because of the intrafamilialvariability of FAP or because of some unknownrecessive modifier gene inherited from bothparents. Although alternative skipping of exon14 is a physiological event observed in normalsubjects, a splice acceptor site mutation leadingto increased expression of an APC mRNA iso-form without exon 14 has been previouslydescribed in a FAP family.16 In our family,splice sites do not show any modification and,furthermore, to rule out the possibility of anintron alteration in the tract near the acceptorsplice site, which might eliminate lariat forma-tion at the branch point, a stretch of 40 nucle-otides upstream of the exon 14 acceptor sitewas also sequenced.

Our data indicate that the mutation harbour-ing allele is expressed exclusively in thealternative isoform and that the DNA exonicsilent mutation might to be able to produce acomplete skipping of exon 14, suggesting theexistence of a mechanism of splicing regulationdistinct from splice junctions. Silent mutationsnot leading to amino acid change are generallyconsidered to be normal variants and arethought to have no role in disease. However, afew papers have described exonic silent muta-tions able to induce exon skipping in thefibrillin-1 gene, the MLH1 gene, and thehuman phenylalanine hydroxylase gene, allassociated with the corresponding diseases.23–25

Recent studies have indicated that sequenceelements that are distinct from the splice sitesare also needed for normal splicing. These ele-ments are required for eYcient splicing andmay aVect splice site recognition duringconstitutive and alternative splicing.26 27 Theyare found within coding exons and are calledexonic splicing enhancers (ESE). Since noconsensus sequence that describes ESE is rec-ognised, these elements are diYcult to identify,not least because they are not as purine rich aswas originally thought.28 Silent point mutationsthat aVect ESE and lead to inappropriate exonskipping have been described in human geneticdiseases. Frontotemporal dementia linked tochromosome 17 missense mutation aVects

splicing of tau gene exon 10 by acting on anESE.29 Carbohydrate deficient glycoproteinsyndrome is associated with a missense muta-tion that disrupts a splicing enhancer sequence,resulting in exon 5 skipping.30 Splicing diVer-ence between the two forms of spinal muscularatrophy, SMN1 and SMN2, is attributed to asilent mutation in an ESE located in the centreof SMN exon 7.31 Very recently, Liu et al32

reported aberrant exon skipping in the BRCA1gene resulting from nonsense, missense, ortranslationally silent mutations, which havedisrupted a critical ESE. The G→T silentgenomic DNA mutation is the only changefound in family GE08 and it is located in apurine rich DNA sequence (ACCG-GAGCCAG). This region in the middle ofexon 14 might represent an ESE sequencenecessary for the control of the alternativesplicing and the one base substitution mightdisrupt this, thus provoking total exon 14 skip-ping. Experiments are being carried out to testenhancer activity of the exon 14 APC se-quences by inserting them into another tran-script that could be assayed by in vitro splicingand/or by transfection,33 followed by mutagen-esis of the sequence, to identify the elementsthat compose the putative enhancer.

This work was supported by MURST, Ministero dell’Universitàe della Ricerca Scientifica e Tecnologica, Cofinanziamento1998, Ricerca Finalizzata-Ministero Sanità 1999,ICS030.1RF99.34. We also acknowledge Galliera GeneticBank, Progetto Telethon C42.

1 Jagelman DG. Extracolonic manifestations of familialadenomatous polyposis. Cancer Genet Cytogenet1987;27:319-25.

+ We describe a silent third base codonmutation in exon 14 of the APC gene, aG→T transversion at nucleotide 1869(R623R), associated with complete skip-ping of exon 14 and production of a stabletruncated APC protein. The R623R mu-tation segregates with the disease in a largeFAP family. This mutation is the sole vari-ation found in the entire APC codingsequence, is not a polymorphism, and hasnot been described before.

+ To evaluate whether the silent genomicDNA variation could aVect RNA tran-scription, RT-PCR analysis was carriedout in aVected subjects. Although exon 14is subject to alternative splicing, an unbal-anced expression of the two exon 14isoforms was found. Sequence analysis ofthe two isoforms showed that the mutantAPC allele expressed exclusively mRNAlacking exon 14. Western blot analysisshowed that the mutant mRNA was trans-lated into a stable truncated protein.

+ Our findings show a possible new modelof APC mutation causing disease andsuggest that exonic elements mightmodulate the splicing of APC gene exon14. They therefore underline the import-ance of investigating the significance ofsilent mutations.

866 Letters

www.jmedgenet.com


http://jmg.bmj.com/


2 Groden J, Thliveris A, Samowitz W, Carlson M, Gelbert L,Albertsen H, Joslyn G, Stevens J, Spirio L, Robertson M,Sargeant L, Krapcho K, Wollf E, Burt R, Hughes JP, War-rington J, McPherson J, Wasmuth J, Le Paslier D, Abderra-him H, Cohen D, Leppert M, White R. Identification andcharacterization of the familial adenomatous polyposis coligene. Cell 1991;66:589-600.

3 Nishisho I, Nakamura Y, Miyoshi Y, Miki Y, Ando H, HoriiA, Koyama K, Utsonomiya J, Baba S, Hedge P. Mutationsof chromosome 5q21 genes in FAP and colorectal cancerpatients. Science 1991;253:665-9.

4 Soussi T. APC mutation database, July 1999. Available athttp//perso.curie.fr/thierry.soussi/APC.htlm

5 Wallis YL, Morton DG, McKeown CM, Macdonald F.Molecular analysis of the APC gene in 205 families:extended genotype-phenotype correlations in FAP and evi-dence for the role of the APC amino acid changes in colo-rectal cancer predisposition. J Med Genet 1999;36:14-20.

6 Stella A, Montera M, Resta N, Marchese C, Susca F, Gen-tile M, Romio L, Pilia S, Prete F, Mareni C, Guanti G.Four novel mutations of the APC (adenomatous polyposiscoli) gene in FAP patients. Hum Mol Genet 1994;3:1687-8.

7 Ficari F, Cama A, Valanzano R, Curia MC, Palmirotta R,Aceto G, Esposito DL, Crognale S, Lombardi A, MesseriniL, Mariani-Costantini R, Tonelli F, Battista P. APC genemutations and colorectal adenomatosis in familial adeno-matous polyposis. Br J Cancer 2000;82:348-53.

8 Miyoshi Y, Ando H, Nagase H, Nishisho I, Horii A, Miki Y,Mori T, Utsunomiya J, Baba S, Petersen G, Hamilton SR,Kinzler, KW, Vogelstein B, Nakamura Y. Germlinemutations of the APC gene in 53 familial adenomatouspolyposis patients. Proc Natl Acad Sci USA 1992;89:4452-6.

9 DeRosa M, Scarano MI, Panariello L, Carlomagno N, RossiGB, Tempesta A, Borgheresi P, Renda A, Izzo P. Threesubmicroscopic deletions at the APC locus and their rapiddetection by quantitative-PCR analysis. Eur J Hum Genet1999;7:695-703.

10 Su L, Steinbach G, Sawyer JC, Hindi M, Ward PA, LynchM. Genomic rearrangements of the APC tumor-suppressorgene in familial adenomatous polyposis. Hum Genet 2000;106:101-7.

11 Joslyn G, Carlson M, Thliveris A, Albertsen H, Gelbert H,Samowitz W, Groden J, Stevens J, Spirio L, Robertson M,Sargeant L, Kraho K, Wollf E, Burt R, Hughes JP,Warrington J, McPherson J, Wasmuth J, Le Paslier D,Abderrahim H, Cohen D, Leppert M, White R. Identifica-tion of deletion mutations and three new genes at thefamilial polyposis locus. Cell 1991;66:601-13.

12 Sulekova Z, Ballhausen WG. A novel coding exon of thehuman adenomatous polyposis coli gene. Hum Genet 1995;96:469-71.

13 Samowitz WS, Thliveris A, Spirio LN, White R. Alterna-tively spliced adenomatous polyposis coli (APC) gene tran-scripts that delete exons mutated in attenuated APC. Can-cer Res 1995;55:3732-4.

14 Cama A, Esposito DL, Palmirotta R, Curia MC, Ranieri A,Ficari F, Valanzano R, Modesti A, Battista P, Tonelli F,Mariani-Costantini R. A novel mutation at the splice junc-tion of exon 9 of the APC gene in familial adenomatouspolyposis. Hum Genet 1994;3:305-8.

15 Varesco L, Gismondi V, Presciuttini S, Groden J, Spirio L,Sala P, Rossetti C, De Benedetti L, Bafico A, Heouaine A,Grammatico P, Del Porto G, White R, Bertario L, FerraraGB. Mutation in a splice-donor site of the APC gene in afamily with polyposis and late age of colonic cancer death.Hum Genet 1994;93:281-6.

16 Bala S, Sulekova Z, Ballhausen WG. Constitutive APC exon14 skipping in early-onset familial adenomatous polyposisreveals a dramatic quantitative distortion of APC gene-specific isoforms. Hum Mutat 1997;10:201-6.

17 Spirio L, Green J, Robertson J, Robertson M, Otterud B,Sheldon J, Howse E, Groden J, White R, Leppert M. Theidentical 5' splice-site acceptor mutation in five attenuatedAPC families from Newfoundland demonstrates a foundereVect. Hum Genet 1999;105:388-98.

18 Montera M, Resta N, Simone C, Guanti G, Marchese C,Civitelli S, Mancini A, Pozzi S, De Salvo L, Bruzzone D,Donadini A, Romio L, Mareni C. Mutational germlineanalysis of hMSH2 and hMLH1 genes in early onset colo-rectal cancer patients. J Med Genet 2000;37:e7.

19 van der Luijt R, Khan M, Vasen H, van Leeuwen C, Tops C,Roest P, den Dunnen C, Fodde R. Rapid detection oftranslation-terminating mutations at the adenomatouspolyposis coli (APC) gene by direct protein truncation test.Genomics 1994;20:1-4.

20 Spirio L, Olschwang S, Groden J, Robertson M, SamowitzW, Joslyn G, Gelbert L, Thliveris A, Carlson M, Otterud B,Lynch H, Watson P, Lynch P, Laurent-Puig P, Burt R,Hughes JP, Thomas G, Leppert M, White R. Alleles of theAPC gene: an attenuated form of familial polyposis. Cell1993;75:951-7.

21 Gismondi V, Stagnaro P, Pedemonte S, Biticchi R, Presciut-tini S, Grammatico P, Sala P, Bertario L, Groden J, VarescoL. Chain-terminating mutations in the APC gene lead toalterations in APC RNA and protein concentration. GenesChrom Cancer 1998;22:278-86.

22 Antonarakis SE. Recommendations for a nomenclature sys-tem for human gene mutations. Nomenclature WorkingGroup. Hum Mutat 1998;11:1-3.

23 Liu W, Qian C, Francke U. Silent mutation induces exonskipping of fibrillin-1 gene in Marfan syndrome. Nat Genet1997;16:328-9.

24 Nystrom-Lahti M, Holmberg M, Fildalgo P, Salovaaro R, dela Chapelle A, Jiricny J, Peltomaki P. Missense andnonsense mutations in codon 659 of MLH1 cause aberrantsplicing of messenger RNA in HNPCC kindreds. GenesChrom Cancer 1999;26:372-5.

25 Chao HK, Hsiao KJ, Su TS. A silent mutation induces exonskipping in the phenylalanine hydroxylase gene in phe-nylketonuria. Hum Genet 2001;108:14-19.

26 Coulter LR, Landree MA, Cooper TA. Identification of anew class of exonic splicing enhancers by in vivo selection.Mol Cell Biol 1997;17:2143-51.

27 Cooper TA, Mattox W. The regulation of splice-siteselection, and its role in human disease. Am J Hum Genet1997;61:259-69

28 Blencowe BJ. Exonic splicing enhancers: mechanism ofaction, diversity and role in human genetic diseases. TrendsGenet 2000;25:106-10.

29 D’Souza I, Poorkai P, Hong M, Nochlin D, Lee VM, BirdTD, Shellenberg GD. Missense and silent tau genemutations cause frontotemporal dementia withparkinsonism-chromosome 17 type, by aVecting multiplealternative RNA splicing regulatory elements. Proc NatlAcad Sci USA 1999;96:5598-603.

30 Villaumier-Barrot S, Barnier A, Cuer M, Durand G, Grand-champ B, Seta N. Characterization of the 415G>A(E139K) PMM2 mutation in carbohydrate-deficient glyco-protein syndrome type Ia disrupting a splicing enhancerresulting in exon 5 skipping. Hum Mutat 1999;14:543-4.

31 Lorson CL, Andronophy EJ. An exonic enhancer is requiredfor inclusion of an essential exon in the SMA-determininggene SMN. Hum Mol Genet 2000;2:259-65.

32 Liu HX, Cartegni L, Zhang MQ, Krainer A. A mechanismfor exon skipping caused by nonsense or missensemutations in BRCA1 and other genes. Nat Genet 2001;27:55-8.

33 Liu HX, Chew SL, Cartegni L, Zhang MQ, Krainer A.Exonic splicing enhancer motif recognized by human SC35under splicing conditions. Mol Cell Biol 2000;20:1063-71.

Evidence of somatic mosaicism for a MECP2mutation in females with Rett syndrome:diagnostic implications

Violaine Bourdon, Christophe Philippe, Thierry Bienvenu, Bernadette Koenig,Marc Tardieu, Jamel Chelly, Philippe Jonveaux

EDITOR—Rett syndrome (RTT) (MIM312750) is an X linked dominant neurodevel-opmental disorder that occurs almost exclu-sively in females. AVected girls are consideredto have a normal perinatal period followed by a

period of regression, loss of acquired purpose-ful manual and speech skills, hand wringing,gait disturbance, and growth retardation.1 2

A gene for RTT has been identified in theXq28 region which encodes the methyl-CpG

Letters 867

J Med Genet2001;38:867–870

Laboratoire deGénétique Médicale,CHU Nancy-Brabois,Avenue du Morvan,54511 Vandoeuvre lesNancy, FranceV BourdonC PhilippeP Jonveaux

Laboratoire deGénétique etPhysiopathologie desRetardsMentaux-ICGM,Faculté de MédecineCochin, Paris, FranceT BienvenuJ Chelly

Site du Neuhof, Centrede Ressources enMatière de DéficiencesSensorielles et dePolyhandicaps LouisBraille - RaoulClainchard - AugusteJacoutôt, Strasbourg,FranceB Koenig

Département dePédiatrie, Service deNeurologie, CHUBicêtre, Le KremlinBicêtre, FranceM Tardieu

Correspondence to:Dr Jonveaux,[email protected]

www.jmedgenet.com


http://jmg.bmj.com/


binding protein 2 (MeCP2) involved in tran-scriptional silencing.3 4 This disorder most fre-quently occurs sporadically and results from ade novo mutation, although a few familial caseshave been reported. Many studies5–16 haveshown that the MECP2 gene is mutated inapproximately 80% of patients with classicalRTT and the MECP2 mutation spectrum

includes missense, nonsense, and frameshiftmutations, as well as larger rearrangements likedeletions encompassing a few hundred bp.16

The failure to detect MECP2 mutations in theremaining 20% may indicate the presence ofmutations in unexplored regions of the MECP2gene, such as regulatory elements or non-coding regions, notably in the new first exon17

or in an additional RTT locus.Here, we report for the first time mosaicism

for a somatic MECP2 mutation found in twounrelated females aVected with RTT. Thesetwo girls were diagnosed according to theinternational criteria of the Rett SyndromeDiagnostic Criteria Work Group.18

Case reportsThe first patient (case 1) is 13 years old. ShesuVers from classical Rett syndrome with 7/9 ofthe necessary criteria, 4/8 of the supportive cri-teria, and none of the exclusion criteria.18 Morespecifically, she had a normal neonatal periodand head circumference at birth and a phase ofsocial withdrawal at the age of 12 months whenshe lost purposeful hand skills and developedstereotypic hand movements, ataxia, andapraxia. She suVered from breathing dysfunc-tion and peripheral vasomotor disturbances.She had severely impaired development butacquired independent walking at the age of 24months. However, she did not acquire micro-cephaly or develop epilepsy.

The second patient (case 2) was reported asan atypical case of RTT without any period ofregression. Both mental and motor develop-ment were very slow. At the age of 4 years, shehad acquired microcephaly (−2 SD) and hadvery limited ambulation, but her hand use wascorrect without hand wringing movements.She developed epilepsy and progressive scolio-sis. She is a placid girl without useful speechbut she communicates well by eye movements.

Methods and resultsFor case 1, an initial study on DNA extractedfrom a lymphoblastoid cell line by denaturinggradient gel electrophoresis (DGGE) andsequencing showed that she carried a 26 bpdeletion starting at position 1165. To confirmthis mutation, DNA was extracted from a freshblood sample and the deletion was assessed bydirect sequencing. Surprisingly and despite acareful examination of the sequence, we didnot find the 26 bp deletion with DNAextracted from leucocytes. This sample wasreanalysed by DGGE and heteroduplexes weredetected while the homoduplex correspondingto the deleted band was absent (fig 1A). Weconfirmed this result by conformation sensitivegel electrophoresis (CSGE) analysis, whichshowed the heteroduplexes but not the mutanthomoduplex (fig 1B). The results obtainedfrom peripheral blood lymphocytes suggestedmosaicism for a somatic mutation.

In order to determine the level of mosaicism,we used a semiquantitative approach based onfluorescent PCR. The MECP2 gene exon 3portion containing the deletion was PCRamplified, the reverse primer being conjugated

Figure 1 Detection of the two somatic mutations byheteroduplex analysis. (A) DGGE results for case 1. DNAextracted from a lymphoblastoid cell line (lane 1) and afresh blood sample (lane 2). The homoduplex bandcorresponding to the deleted allele is missing in DNAextracted from lymphocytes. Ht, heteroduplex; Ho,homoduplex; 40%-90% formamide gradient. (B) CSGEresults for case 1 (lanes 1-3) and case 2 (lane 5). Case 1:the somatic mutation is shown on DNA extracted from afresh blood sample (lane 1); the father (lane 2) and themother (lane 3) do not carry the 26 bp deletion. Case 2: asomatic deletion was detected on DNA extracted fromlymphocytes as indicated by the absence of the deletedhomoduplex band (lane 5). Lane 4 depicts a CSGEpattern of a 31 bp deletion localised in the deletion proneregion of the MECP2 gene; the two homoduplex bands areof equal intensity.

1 2 3 4 5

40%A

B

1 2

90%

+

–

Ht

Ho

Ho

Ht

868 Letters

www.jmedgenet.com


http://jmg.bmj.com/


to 6-FAM (6-carboxy-fluorescein). PCR prod-ucts were analysed on an ABI 310 sequencerand peak areas were generated by ABI Gene-scan and Genotyper software. The ratiobetween the deleted and normal peak areasshowed that only 36% of lymphocytes har-boured the deletion, that is, 18% of X chromo-somes bore the 26 bp deletion (fig 2A). Thissemiquantitative approach confirms that case 1does have somatic mosaicism for the MECP2deletion. The relatively low level of somaticmosaicism could explain the normal sequenc-ing result. Thus, mosaicism was quantified indiVerent tissues. DNA was extracted frombuccal mucosa cells19 and hair bulb cells. 20 Thelevel of mosaicism was about the same in buc-cal mucosa cells (30%) as in lymphocytes, butlower in hair bulbs cells (17.5%) (fig 2A).

DiscussionOn the basis of these results, we hypothesisedthat some patients with RTT may in fact carrya somatic mutation. Small deletions (from 7 to170 bp) within the region between bp 1096and 1165 of the MECP2 gene have been recur-rently identified.5 7 9 10 12 15 16 They do not aVect

the two functional domains but result in theloss of one fifth of the protein. Interestingly, ithas been shown that the deletion of thecarboxy-terminal 63 amino acids of theMeCP2 protein impairs binding with thenucleosomal DNA during the transcriptionregulation process.21 These recurrent deletionsmay be the result of palindromic and quasipal-indromic sequences within this region, whichare believed to form secondary structures thatrender the region vulnerable to deletions.Therefore, using our fluorescent PCR ap-proach, we reanalysed the 3' region of theMECP2 gene, between bp 1096 and 1165, in acohort of 29 patients diagnosed as typical oratypical RTT; for these patients, we failed todetect any mutation using a bidirectionalsequencing strategy of the entire MECP2 cod-ing region. A second somatic mosaicism for a27 bp deletion was identified in peripheralblood lymphocytes from case 2 with atypicalRTT; the mosaicism rate was quantified withour fluorescent approach to be about 37% (fig2B). We confirmed this result by CSGE analy-sis (fig 1B).

Figure 2 Semiquantitative fluorescent PCR of the somatic mosaicism rate. (A) Case 1. Genotyper traces of the fluorescentPCR products obtained with three diVerent tissues, blood (1), buccal mucosa cells (2), and hair bulb cells (3), shown withthe three respective ratios of peak areas (X1, X2, X3). For each peak, the fragment size in bp and the peak area calculatedby Genescan is indicated. We assumed that the mosaicism rate could be estimated by calculating the ratio between thedeleted and the normal peak areas. (B) Case 2. Genotyper trace of a fluorescent PCR product obtained from blood with theratio of peak areas (Y).

2000

BloodB

1000

335

13 198

362

35 786

2000 3 Hair bulb cells

15001000500

336

4242

362

24 124

2 Buccal mucosa cells

X3 = 17.5%

X2 = 30%

X1 = 36%

Y = 37%

1000200030004000

336

16 269

362

54 012

3000

BloodA

1000

1000

336

26 168

362

72 328

Letters 869

www.jmedgenet.com


http://jmg.bmj.com/


In both cases, numerical aberrations of the Xchromosomes as a cause for the uncommonfluorescent PCR patterns were excluded by thepresence of a normal 46,XX karyotype.

These two patients show a similar deletionwith an equivalent mosaicism rate in blood, buta distinct clinical presentation. X inactivationstudy on proband 1 with typical Rett syndromeshowed a random pattern of inactivation in theperipheral blood. Although the results have tobe extrapolated from the peripheral blood cells,it would suggest that in the brain the majorityof mutated X chromosomes may remain activein the girl with classical Rett syndrome. Ourresults illustrate clearly once again the diYcultyin establishing a correlation between genotypeand phenotype in RTT.

Recently, a boy with a mosaic mutation hasbeen described.22 To our knowledge, we showfor the first time that somatic mosaicism forMECP2 mutation in girls is not infrequent (twosomatic mutations on 102 putative RTT casesstudied) and may cause diVerent phenotypes.These clinical and molecular findings suggestthat multiple forms of mosaicism (X inactiva-tion mosaicism and somatic mosaicism) maybe present in a single patient with RTT. Mosai-cism has been documented for chromosomalabnormalities, mitochondrial mutations, tripletrepeats,23 and in a growing number of domi-nant and recessive X linked gene disorders,such as Duchenne muscular dystrophy,24 hae-mophilia B,25 Conradi-Hünermann-Happlesyndrome,26 and double cortex/lissencephalysyndrome.27 Because a proportion of cells carrythe mutation not only in blood but also in tis-sues deriving from other cell lineages, it mustbe assumed that the mutation occurred veryearly during embryogenesis.

Finally, the detection of mosaic mutationdepends mainly on the method used for theidentification of mutations within the MECP2gene. Nowadays, the method of choice foridentifying deleterious mutations relies ondirect DNA sequencing. The ability of thismethod to detect mosaic mutations is poor,which is particularly true when the mosaicismrate is low. Our findings underline the need forat least two complementary approaches, suchas methods based on heteroduplex analysis andsequencing, for an eYcient screening of theMECP2 gene.

We thank Dr Deblay for critical advice, Dr Florence Rousseletfor her technical contribution, and l’Association Française duSyndrome de Rett, l’Association Française contre les myopa-thies, and the Ministère de l’Education Nationale, de la Recher-che et de la Technologie for their financial support.

1 Rett A. On an unusual brain atrophy syndrome in hyperam-monemia in childhood. Wien Med Wochenschr 1966;11:723-6.

2 Hagberg B, Aicardi J, Dias K, Ramos O. A progressive syn-drome of autism, dementia, ataxia, and loss of purposefulhand use in girls: Rett’s syndrome: report of 35 cases. AnnNeurol 1983;14:471-9.

3 Amir RE, Van den Veyver IB, Wan M, Tran CQ, Francke U,Zoghbi HY. Rett syndrome is caused by mutations inX-linked MECP2, encoding methyl-CpG-binding protein2. Nat Genet 1999;23:185-8.

4 Nan X, Campoy FJ, Bird A. MeCP2 is a transcriptionalrepressor with abundant binding sites in genomic chroma-tin. Cell 1997;21:471-81.

5 Amir RE, Van den Veyver IB, Schultz R, Malicki DM, TranCQ, Dahle EJ, Philippi A, Timar L, Percy AK, Motil KJ,Lichtarge O, Smith EO, Glaze DG, Zoghbi HY. Influenceof mutation type and X chromosome inactivation on Rettsyndrome phenotypes. Ann Neurol 2000;47:670-9.

6 Amano K, Nomura Y, Segawa M, Yamakawa K. Mutationalanalysis of the MECP2 gene in Japanese patients with Rettsyndrome. J Hum Genet 2000;45:213-36.

7 Bienvenu T, Carrie A, de Roux N, Vinet MC, Jonveaux P,Couvert P, Villard L, Arzimanoglou A, Beldjord C, FontesM, Tardieu M, Chelly J. MECP2 mutations account formost cases of typical forms of Rett syndrome. Hum MolGenet 2000;9:1377-84.

8 Buyse IM, Fang P, Hoon KT, Amir RE, Zoghbi HY, RoaBB. Diagnostic testing for Rett syndrome by DHPLC anddirect sequence analysis of the MECP2 gene: identificationof several novel mutations and polymorphisms. Am J HumGenet 2000;67:1428-36.

9 Cheadle JP, Gill H, Fleming N, Maynard J, Kerr A, LeonardH, Krawczak M, Cooper DN, Lynch S, Thomas N, HughesH, Hulten M, Ravine D, Sampson JR, Clarke A. Long-readsequence analysis of the MECP2 gene in Rett syndromepatients: correlation of disease severity with mutation typeand location. Hum Mol Genet 2000;9:1119-29.

10 De Bona C, Zappella M, Hayek G, Meloni I, Vitelli F, Brut-tini M, Cusano R, LoVredo P, Longo I, Renieri A.Preserved speech variant is allelic of classic Rett syndrome.Eur J Hum Genet 2000;8:325-30.

11 Hampson K, Woods CG, Latif F, Webb T. Mutations in theMECP2 gene in a cohort of girls with Rett syndrome. JMed Genet 2000;37:610-12.

12 Huppke P, Laccone F, Kramer N, Engel W, Hanefeld F. Rettsyndrome: analysis of MECP2 and clinical characterizationof 31 patients. Hum Mol Genet 2000;9:1369-75.

13 Obata K, Matsuishi T, Yamashita Y, Fukuda T, KuwajimaK, Horiuchi I, Nagamitsu S, Iwanaga R, Kimura A, OmoriI, Endo S, Mori K, Kondo I. Mutation analysis of themethyl-CpG binding protein 2 gene (MECP2) in patientswith Rett syndrome. J Med Genet 2000;37:608-10.

14 Wan M, Lee SS, Zhang X, Houwink MI, Song HR, AmirRE, Budden S, Naidu S, Pereira JL, Lo IF, Zoghbi HY,Schanen NC, Francke U. Rett syndrome and beyond:recurrent spontaneous and familial MECP2 mutations atCpG hotspots. Am J Hum Genet 1999;65:1520-9.

15 Xiang F, Buervenich S, Nicolao P, Bailey ME, Zhang Z,Anvret M. Mutation screening in Rett syndrome patients. JMed Genet 2000;37:250-5.

16 Bourdon V, Philippe C, Labrune O, Amsallem D, ArnouldC, Jonveaux P. A detailed analysis of the MECP2 gene:prevalence of recurrent mutations and gross DNArearrangements in Rett syndrome patients. Hum Genet2001;108:43-50.

17 Reichwald K, Thiesen J, Wiehe T, Weitzel J, Stätling WH,Kioschis P, Poustka A, Rosenthal A, Platzer M. Compara-tive sequence analysis of the MECP2-locus in human andmouse reveals new transcribed regions. Mamm Genome2000;11:182-90.

18 Trevarthen E, Moser HW, and the Diagnostic CriteriaWorking Group. Diagnostic criteria for Rett syndrome.The Rett Syndrome Diagnostic Criteria Work Group. AnnNeurol 1988;23:425-8.

19 Philippe C, Porter DE, Emerton ME, Wells DE, SimpsonAHRW, Monaco AP. Mutation screening of the EXT1 andEXT2 genes in patients with hereditary multiple exostoses.Am J Hum Genet 1997;61:520-8.

20 Walsh PS, Metzger DA, Higuchi R. Chelex 100 as a mediumfor simple extraction of DNA for PCR-based typing fromforensic material. Biotechniques 1991;10:506-13.

21 Chandler SP, Guschin D, Landsberger N, WolVe AP. Themethyl-CpG binding transcriptional repressor MeCP2 sta-bly associates with nucleosomal DNA. Biochemistry 1999;38:7008-18.

22 Clayton-Smith J, Watson P, Ramsden S, Black GCM.Somatic mutation in MECP2 as a non-fatal neurodevelop-mental disorder in males. Lancet 2000;356:830-2.

23 Bernards A, Gusella JF. The importance of geneticmosaicism in human disease. N Engl J Med 1994;331:1447-9.

24 Bunyan DJ, Robinson DO, Collins AL, Cockwell AE, Bull-man HMS, Whittaker PA. Germline and somatic mosai-cism in a female carrier of Duchenne muscular dystrophy.Hum Genet 1994;93:541-4.

25 Costa J-M, Vidaud D, Laurendreau I, Vidaud M, Fressin-aud E, Moisan JP, David A, Meyer D, Lavergne JM.Somatic mosaicism and compound heterozygosity infemale hemophilia B. Blood 2000;96:1585-7.

26 Has C, Bruckner-Tuderman L, Müller D, Floeth M, FolkersE, Donnai D, Traupe H. The Conradi-Hünermann-Happle syndrome (CDPX2) and emopamil bindingprotein: novel mutations, and somatic and gonadal mosai-cism. Hum Mol Genet 2000;9:1951-5.

27 Gleeson JG, Minnerath S, Kuzniecky RI, Dobyns WB, YoungID, Ross ME, Walsh CA. Somatic and germline mosaicmutations in the doublecortin gene are associated with vari-able phenotypes Am J Hum Genet 2000;67:574-81.

870 Letters

www.jmedgenet.com


http://jmg.bmj.com/


Temperature sensitive acyl-CoA oxidase import ingroup A peroxisome biogenesis disorders

Atsushi Imamura, Nobuyuki Shimozawa, Yasuyuki Suzuki, Zhongyi Zhang,Toshiro Tsukamoto, Tadao Orii, Takashi Osumi, Naomi Kondo

EDITOR—Peroxisome biogenesis disorders(PBDs) are lethal genetic diseases character-ised by a number of peroxisomal metabolicabnormalities, including the oxidation of verylong chain fatty acids (VLCFAs), biosynthesisof bile acids and plasmalogen, and detoxifica-tion of H2O2. Peroxisomal matrix proteins aresynthesised on free polyribosomes and directedto the organelle by cis acting peroxisometargeting signals (PTSs). PTS1 is a C-terminaltripeptide Ser-Lys-Leu (SKL) sequence andlater the consensus sequence was broadened to(S/A/C/K/N)-(K/R/H/Q/N/S)-L, based on sub-sequent studies. Acyl-CoA oxidase (AOX) hasSKL and D bifunctional protein has AKL.1–4

PTS2 is an N-terminal cleavable peptide(-R/KLX5Q/HL) that resides in peroxisomal3-ketoacyl CoA thiolase (PT), alkyl-dihydroxyacetonephosphate synthase, andphytanoyl-CoA hydroxylase.5–9 PBDs are ge-netically classified into at least 12 complemen-tation groups (CGs) and each CG containsvarious clinical phenotypes, for example, Zell-weger syndrome (ZS), neonatal adrenoleuco-dystrophy (NALD), and infantile Refsumdisease (IRD).10 11 ZS patients have severeneurological defects, liver dysfunction, andrenal cysts and die before 1 year of age. NALDpatients have symptoms similar to ZS patients,but they survive a little longer, and IRDpatients show milder abnormalities in the cen-tral nervous system and survive even longer.We identified the restoration of peroxisomebiogenesis in a temperature sensitive (TS)manner in fibroblasts from milder forms ofPBDs, that is, all IRD patients and someNALD patients belonging to groups CG-A(CG8), CG-C (CG4), CG-E (CG1), CG-F(CG10), CG-H, and CG6.12–15 In these cells,peroxisomes were formed at 30°C and bio-chemical activities of peroxisomes, includingthe oxidation of VLCFAs and dihydroxy-acetonephosphate acyltransferase (DHAP-AT), and the import of peroxisomal enzymes,were also restored.16 However, virtually no per-oxisomes were formed in ZS cells at 30°C andimport of peroxisomal enzymes did not im-prove.16 Here, we elucidate temperature de-pendent import and processing of AOX at30°C which is unique to fibroblasts from ZSpatients belonging to CG-A. Correlation be-tween the import of peroxisomal enzymes andbiochemical functions of peroxisomes is alsodiscussed.

Materials and methodsCELL LINES

Skin fibroblasts from the patients belonging toCG-A (CG8) including four with ZS (A-02,

06, 10, and 14), two with NALD (A-05 and08), and one with IRD (A-04) were cultured at37°C or 30°C in an atmosphere of 5% CO2 inMEM supplemented with 10% fetal calfserum. A-02, A-06, and A-14 were Japanesebabies diagnosed as ZS with typical dysmor-phic features, who died at a few months of age.The clinical data of A-04 and A-08 have beenpreviously reported, whereas those of A-05 andA-10 have not.12 In addition, ZS fibroblastsbelonging to CG-C (C-08), CG-E (E-14), andCG-F (F-01) were cultured under the samecondition (the numbers and clinical data ofthese patients have been previously de-scribed12). All cell lines were classified by com-plementation analysis as previously de-scribed.10 17 18

IMMUNOFLUORESCENCE STUDY

For the detection of peroxisomes and theimport of PTSs, cells were fixed after 72 hours’incubation at either 37°C or 30°C, permeabi-lised with 0.1% Triton X-100, and processedfor indirect immunofluorescence staining.19

The first antibodies we used were rabbitantibodies to human catalase, AOX, D bifunc-tional protein, and PT, and in double immuno-fluorescence rabbit anti-rat PMP70 antibodywas used.

BIOCHEMICAL ASSAYS

Peroxisomal VLCFA oxidation in fibroblastswas assessed by the ratio of lignoceric acid(C24:0)/palmitic acid (C16:0) oxidation activ-ity.20 The activity of DHAP-AT, the firstenzyme in the pathway leading to plasmalogenbiosynthesis, was measured as described previ-ously21 using 14C labelled DHAP as substrate.Continuous cell labelling with 35S-methionineand immunoprecipitation of AOX with rabbitanti-human AOX antibody was performed asdescribed previously.19 22

ResultsIMMUNOFLUORESCENCE STUDY IN FIBROBLASTS

FROM CG-A PATIENTS

The fibroblasts from PBD patients belongingto the CG-A were examined by immuno-fluorescence microscopy to determine theimport of PTS1 and PTS2 containing peroxi-somal matrix proteins. The immunoreactivityof these proteins in control cells showed thesame punctate pattern as previous reports(data not shown).23 The most striking resultwas that in the fibroblasts from ZS patients, theimport of AOX was rescued apparently afterincubation at 30°C, whereas it was severelyreduced or absent at 37°C (fig 1C, D, table 1).

Letters 871

J Med Genet2001;38:871–874

Department ofPaediatrics, GifuUniversity School ofMedicine, 40Tsukasa-machi, Gifu500-8705, JapanA ImamuraN ShimozawaY SuzukiZ ZhangN Kondo

Department ofPaediatrics, OgakiMunicipal Hospital,Ogaki, Gifu 503-8502,JapanA Imamura

Department of LifeScience, HimejiInstitute ofTechnology, Kamigori,Hyogo 678-1297, JapanT TsukamotoT Osumi

Faculty of HumanWelfare, ChubuGakuin University,Seki, Gifu 501-3936,JapanT Orii

Correspondence to:Dr Imamura,[email protected]

www.jmedgenet.com


http://jmg.bmj.com/


The AOX and 70 kDa peroxisomal membraneprotein (PMP70) were co-localised in thesecells (data not shown). In other CGs, none ofthe ZS cell lines showed such TS characteris-tics. There was a little cross reactivity of AOXantibody with mitochondria (fig 1C) co-localised with anti-mitochondria antibody(data not shown). The import of catalase (fig1A, B), D bifunctional protein (fig 1E, F), andPT (fig 1G, H) was defective after incubationat 30°C (table 1). In cell lines from NALD andIRD patients, the import of catalase, AOX, Dbifunctional protein, and PT greatly improvedafter incubation at 30°C, whereas import ofthese enzymes was deficient at 37°C (table 1).

BIOCHEMICAL ANALYSIS OF FIBROBLASTS FROM

CG-A PATIENTS

The relative VLCFA oxidation capacity wasexpressed as a ratio of C24:0/C16:0 fatty acidoxidation activity of the cell line. Using thisassay, a lower ratio of C24:0/C16:0 fatty acidoxidation indicates lower peroxisomal VLCFAoxidation activity.20 In the same way, the loweractivity of DHAP-AT indicates lower plasm-alogen biosynthesis capacity of peroxisomes.21

The VLCFA oxidation ratio and the activity ofDHAP-AT in ZS patients belonging to CG-Ashowed a very low level after incubation at30°C, although a slight increase in activitieswas observed. Cell lines from NALD patients,showing TS import of PTSs, showed markedimprovement in activities after incubation at30°C (table 2). AOX, the first enzyme of theperoxisomal beta oxidation system, is proc-essed into two smaller polypeptides in peroxi-somes, and this processing was defective in thefibroblasts from patients with PBDs.22 On con-tinuous labelling of the cells from a ZS patient(A-06) at 30°C, the processed form of AOX(subunit B, 53 kDa) was identified in additionto the unprocessed form of AOX (subunit A,75 kDa), as was the case with those from thecontrol and NALD patient (A-05). Theseresults indicate that at least AOX is processedin a temperature dependent manner even in ZScells. In ZS cells belonging to CG-E and CG-F,subunit B was not detected at 30°C, indicatingthat these ZS cells were not temperaturedependent (fig 2).

DiscussionThe diagnosis in PBD patients is based onclinical features combined with a series of tests

Figure 1 Immunofluorescent staining for antibodies of peroxisome matrix proteins infibroblasts from a ZS patient with CG-A (A-06). Cells were cultured for 72 hours at both37°C (panels A, C, E, and G) and 30°C (panels B, D, F, and H), and stained withanti-human catalase (panels A and B), AOX (panels C and D), D bifunctional protein(panels E and F), and PT (panels G and H) rabbit antibody. Bar=10 µm.

D-bifunctionalprotein

HG

FE

DC

BA

AOX

Catalase

PT

30°C37°C

Table 1 The import of peroxisome matrix proteins of fibroblasts from complementationgroup A (CG-A) patients

No Type

Catalase AOXD bifunctionalprotein PT

37°C 30°C 37°C 30°C 37°C 30°C 37°C 30°C

A-06 ZS 0 10 10 100 0 0 0 10A-02 ZS 0 0 0 60 0 0 0 0A-14 ZS 0 0 0 50 0 0 0 0A-10 ZS 0 5 10 60 0 0 0 5A-05 NALD 0 90 0 100 5 60 5 80A-08 NALD 0 80 0 90 0 50 0 80A-04 IRD 0 60 0 80 0 50 0 70

Cells were cultured for 72 hours at either 37°C or 30°C and immunostained by antibodies of per-oxisome matrix proteins. Immunopositive cells were defined as over 100 immunoreactive particlesstained by each antibody in a cell. Immunopositive cells among 20 cells were counted in five viewfields at a magnification ×1000 and data are averages of these in percentages.

Table 2 Biochemical assay of fibroblasts fromcomplementation group A (CG-A) patients

No. Type

C24/C16 DHAP-AT

37°C 30°C 37°C 30°C

A-06 ZS 0.040 0.069 0.09 0.15A-14 ZS 0.011 0.042 0.18 0.30A-10 ZS 0.007 0.030 0.27 0.46A-05 NALD 0.077 0.199 0.39 1.65A-08 NALD 0.013 0.287 0.19 1.55Control 0.522 0.478 1.53 1.90

Fibroblasts were cultured for eight days at either 37°C or 30°C,and then assayed for beta oxidation activities of lignoceric (C24)and palmitic acids (C16) and the ratio of C24/C16 wascalculated. DHAP-AT activity was measured in each cell line.These results were the average of duplicates. DHAP-AT activitywas indicated as nmol/mg/120 min. The methods have beendescribed in Suzuki et al20 and Shimozawa et al.21

872 Letters

www.jmedgenet.com


http://jmg.bmj.com/


to assess peroxisomal function and structure,and the absence or reduction of both PTS1 andPTS2 protein imports are observed in the vastmajority of cells from patients with PBDs.24 Infibroblasts from milder forms of PBDs, we elu-cidated the TS restoration of peroxisomebiogenesis including the import of PTSs afterincubation at 30°C.12–16 The temperature de-pendent manner of peroxisome biogenesis hasbeen identified in cells from mildly aVectedpatients belonging to various diVerent CGsincluding CG-A in PBDs, whereas it has neverbeen seen in those from ZS patients with theseCGs.12–15 In this study, we clarified the restora-tion of AOX import and processing after incu-bation at 30°C in fibroblasts from patients withZS, NALD, and IRD belonging to CG-A. AOXis one of the PTS1 proteins, and is transportedto the peroxisome by binding with Pex5p, aPTS1 receptor.25 26 Pex5p is severely reduced inthe fibroblasts of patients with CG-A, C, andE. However, cells with those CGs share theability to import small amounts of both PTS1and PTS2 proteins.27 The genes responsible forCG-C and E, PEX6 and PEX1, encode Pex6pand Pex1p, respectively. These proteins belongto the AAA (ATPases associated with diversecellular activities) family of proteins and arerequired for stability of the PTS1 receptor inyeast as well as in human cells.28–32 The productof the CG-A gene, which is still unknown, mayalso physically function with Pex5p. Thebiochemical functions of peroxisome asmeasured by VLCFA oxidation andDHAP-AT activity are restored apparently inthe fibroblasts from milder NALD patients,whereas they showed lower levels in the ZScells even at 30°C as was the case with otherCGs. This might be caused by the defectiveimport of peroxisomal beta oxidation enzymesother than AOX. The impairment in fibroblastVLCFA beta oxidation and the clinical mani-festations were more severe in bifunctionalprotein deficient than AOX deficient patients.33

Therefore, ZS patients with CG-A showed assevere a clinical course and prognosis as thosewith other CGs, notwithstanding the AOXimport was rescued at 30°C in the cells frompatients with CG-A.12 34 TS import of AOX in

ZS fibroblasts is unique to CG-A. A peroxinencoded by the candidate gene for CG-A mayinteract with Pex5p independently or with thePex5p-AOX complex, and may lead to thecharacteristic import of AOX at 30°C. Isola-tion of the PEX gene for CG-A will help toclarify the function of peroxin for CG-A andmechanism of the selective TS import of AOXin CG-A patients.

We thank T Hashimoto for anti-human catalase, acyl-CoA oxi-dase, and D bifunctional protein antibodies. This work was sup-ported in part by a Grant in Aid for Scientific Research(10670718, 10670721) from the Ministry of Education,Science, Sports and Culture of Japan, by Health ScienceResearch Grants from the Ministry of Health and Welfare ofJapan, by the Mother and Child Health Foundation, and by theIchiro Kanehara Foundation.

1 Gould SJ, Keller GA, Hosken N, Wilkinson J, Subramani S.A conserved tripeptide sorts proteins to peroxisomes. J CellBiol 1989;108:1657-64.

2 Gould SJ, Keller GA, Subramani S. Identification of a per-oxisomal targeting signal at the carboxy terminus of fireflyluciferase. J Cell Biol 1987;105:2923-31.

3 Amery L, Brees C, Baes M, Setoyama C, Miura R,Mannaerts GP, Van Veldhoven PP. C-terminal tripeptideSer-Asn-Leu (SNL) of human D-aspartate oxidase is afunctional peroxisome-targeting signal. Biochem J 1998;336:367-71.

4 Jiang LL, Miyazawa S, Souri M, Hashimoto T. Structure ofD-3- hydroxyacyl-CoA dehydratase/D-3-hydroxyacyl-CoAdehydrogenase bifunctional protein. J Biochem (Tokyo)1997;121:364-9.

5 Swinkels BW, Gould SJ, Bodnar AG, Rachubinski RA, Sub-ramani S. A novel, cleavable peroxisomal targeting signal atthe amino-terminus of the rat 3-ketoacyl-CoA thiolase.EMBO J 1991;10:3255-62.

6 Osumi T, Tsukamoto T, Hata S, Yokota S, Miura S, FujikiY, Hijikata M, Miyazawa S, Hashimoto T. Amino-terminalpresequence of the precursor of peroxisomal 3-ketoacyl-CoA thiolase is a cleavable signal peptide for peroxisomaltargeting. Biochem Biophys Res Commun 1991;181:947-54.

7 Tsukamoto T, Hata S, Yokota S, Miura S, Fujiki Y, HijikataM, Miyazawa S, Hashimoto T, Osumi T. Characterizationof the signal peptide at the amino terminus of the rat per-oxisomal 3-ketoacyl-CoA thiolase precursor. J Biol Chem1994;269:6001-10.

8 Mihalik SJ, Morrell JC, Kim D, Sacksteder KA, Watkins PA,Gould SJ. Identification of PAHX, a Refsum disease gene.Nat Genet 1997;17:185-9.

9 Jansen GA, Ofman R, Ferdinandusse S, Ijlst L, MuijsersAO, Skjeldal OH, Stokke O, Jakobs C, Besley GT, WraithJE, Wanders RJ. Refsum disease is caused by mutations inthe phytanoyl-CoA hydroxylase gene. Nat Genet 1997;17:190-3.

10 Shimozawa N, Suzuki Y, Zhang Z, Imamura A, Kondo N,Kinoshita N, Fujiki Y, Tsukamoto T, Osumi T, Imanaka T,Orii T, Beemer F, Mooijer P, Dekker C, Wanders RJ.Genetic basis of peroxisome-assembly mutants of humans,Chinese hamster ovary cells, and yeast: identification of anew complementation group of peroxisome-biogenesis dis-orders apparently lacking peroxisomal-membrane ghosts.Am J Hum Genet 1998;63:1898-903.

11 Gould SJ, Valle D. Peroxisome biogenesis disorders:genetics and cell biology. Trends Genet 2000;16:340-5.

12 Imamura A, Tsukamoto T, Shimozawa N, Suzuki Y, ZhangZ, Imanaka T, Fujiki Y, Orii T, Kondo N, Osumi T.Temperature-sensitive phenotypes of peroxisome-assemblyprocesses represent the milder forms of humanperoxisome-biogenesis disorders. Am J Hum Genet 1998;62:1539-43.

13 Imamura A, Tamura S, Shimozawa N, Suzuki Y, Zhang Z,Tsukamoto T, Orii T, Kondo N, Osumi T, Fujiki Y.Temperature-sensitive mutation in PEX1 moderates thephenotypes of peroxisome deficiency disorders. Hum MolGenet 1998;7:2089-94.

14 Shimozawa N, Suzuki Y, Zhang Z, Imamura A, Toyama R,Mukai S, Fujiki Y, Tsukamoto T, Osumi T, Orii T,Wanders RJ, Kondo N. Nonsense and temperature-sensitive mutations in PEX13 are the cause of complemen-tation group H of peroxisome biogenesis disorders. HumMol Genet 1999;8:1077-83.

15 Imamura A, Shimozawa N, Suzuki Y, Zhang Z, TsukamotoT, Fujiki Y, Orii T, Osumi T, Wanders RJA, Kondo N.Temperature-sensitive mutation of PEX6 in peroxisomebiogenesis disorders in complementation group C (CG-C):comparative study of PEX6 and PEX1. Pediatr Res2000;48:541-5.

16 Imamura A, Shimozawa N, Suzuki Y, Zhang Z, TsukamotoT, Fujiki Y, Orii T, Osumi T, Kondo N. Restoration of bio-chemical function of the peroxisome in the temperature-sensitive mild forms of peroxisome biogenesis disorder inhumans. Brain Dev 2000;22:8-12.

17 Yajima S, Suzuki Y, Shimozawa N, Yamaguchi S, Orii T,Fujiki Y, Osumi T, Hashimoto T, Moser HW. Complemen-tation study of peroxisome-deficient disorders by immuno-fluorescence staining and characterization of fused cells.Hum Genet 1992;88:491-9.

Figure 2 Biosynthesis and processing of AOX. Fibroblasts were cultured for 72 hours andthen incubated for 24 hours in the presence of 35S-methionine at either 37°C (lanes 1, 3, 5,7, and 9) or 30°C (lanes 2, 4, 6, 8, and 10). The cell lysate, the same amount of protein,was immunoprecipitated with anti-human AOX antibody and subjected to SDS-PAGEand fluorography. Lanes 1 and 2, control fibroblasts; 3 and 4, ZS with CG-A (A-06)fibroblasts; 5 and 6, NALD with CG-A (A-05) fibroblasts; 7 and 8, ZS with CG-Efibroblasts; 9 and 10, ZS with CG-F fibroblasts.

75 kDa

53 kDa

Temperature (°C) 37

Control ZS(CG-A)

NALD(CG-A)

ZS(CG-E)

ZS(CG-F)

30 37 30 37 30 37 30 37 30

1 2 3 4 5 6 7 8 9 10

Letters 873

www.jmedgenet.com


http://jmg.bmj.com/


18 Shimozawa N, Suzuki Y, Orii T, Moser A, Moser HW, Wan-ders RJ. Standardization of complementation grouping ofperoxisome-deficient disorders and the second Zellwegerpatient with peroxisomal assembly factor-1 (PAF-1) defect.Am J Hum Genet 1993;52:843-4.

19 Tsukamoto T, Miura S, Nakai T, Yokota S, Shimozawa N,Suzuki Y, Orii T, Fujiki Y, Sakai F, Bogaki A, Yasumo H,Osumi T. Peroxisome assembly factor-2, a putative ATPasecloned by functional complementation on a peroxisome-deficient mammalian cell mutant. Nat Genet 1995;11:395-401.

20 Suzuki Y, Shimozawa N, Yajima S, Yamaguchi S, Orii T,Hashimoto T. EVects of sodium 2-[5-(4-chlorophenyl)pentyl]-oxirane-2-carboxylate (POCA) onfatty acid oxidation in fibroblasts from patients withperoxisomal diseases. Biochem Pharmacol 1991;41:453-6.

21 Shimozawa N, Suzuki Y, Orii T, Yokota S, Hashimoto T.Biochemical and morphologic aspects of peroxisomes inthe human rectal mucosa: diagnosis of Zellweger syndromesimplified by rectal biopsy. Pediatr Res 1988;24:723-7.

22 Suzuki Y, Shimozawa N, Orii T, Igarashi N, Kono N,Hashimoto T. Molecular analysis of peroxisomal beta-oxidation enzymes in infants with Zellweger syndrome andZellweger-like syndrome: further heterogeneity of theperoxisomal disorder. Clin Chim Acta 1988;172:65-76.

23 Shimozawa N, Zhang Z, Suzuki Y, Imamura A, TsukamotoT, Osumi T, Fujiki Y, Orii T, Barth PG, Wanders RJ,Kondo N. Functional heterogeneity of C-terminal peroxi-some targeting signal 1 in PEX5-defective patients.Biochem Biophys Res Commun 1999;262:504-8.

24 Slawecki ML, Dodt G, Steinberg S, Moser AB, Moser HW,Gould SJ. Identification of three distinct peroxisomeprotein import defects in patients with peroxisome biogen-esis disorders. J Cell Sci 1995;108:1817-29.

25 Dodt G, Braverman N, Wong C, Moser A, Moser H. W,Watkins P, Valle D, Gould SJ. Mutations in the PTS1receptor gene, PXR1, define complementation group 2 ofthe peroxisome biogenesis disorders. Nat Genet 1995;9:115-25.

26 Wiemer EA, Nuttley WM, Bertolaet BL, Li X, Francke U,Wheelock MJ, Anne UK, Johnson KR, Subramani S.

Human peroxisomal targeting signal-1 receptor restoresperoxisomal protein import in cells from patients with fatalperoxisomal disorders. J Cell Biol 1995;130:51-65.

27 Dodt G, Gould SJ. Multiple PEX genes are required forproper subcellular distribution and stability of Pex5p, thePTS1 receptor: evidence that PTS1 protein import ismediated by a cycling receptor. J Cell Biol 1996;135:1763-74.

28 Fukuda S, Shimozawa N, Suzuki Y, Zhang Z, Tomatsu S,Tsukamoto T, Hashiguchi N, Osumi T, Masuno M, Imai-zumi K, Kuroki Y, Fujiki Y, Orii T, Kondo N. Human per-oxisome assembly factor-2 (PAF-2): a gene responsible forgroup C peroxisome biogenesis disorder in humans. Am JHum Genet 1996;59:1210-20.

29 Yahraus T, Braverman N, Dodt G, Kalish JE, Morrell JC,Moser HW, Valle D, Gould SJ. The peroxisome biogenesisdisorder group 4 gene, PXAAA1, encodes a cytoplasmicATPase required for stability of the PTS1 receptor. EMBOJ 1996;15:2914-23.

30 Reuber BE, Germain-Lee E, Collins CS, Morrell JC,Ameritunga R, Moser HW, Valle D, Gould SJ. Mutations inPEX1 are the most common cause of peroxisomebiogenesis disorders. Nat Genet 1997;17:445-8.

31 PortsteVen H, Beyer A, Becker E, Epplen C, Pawlak A,Kunau WH, Dodt G. Human PEX1 is mutated in comple-mentation group 1 of the peroxisome biogenesis disorders.Nat Genet 1997;17:449-52.

32 Tamura S, Okumoto K, Toyama R, Shimozawa N,Tsukamoto T, Suzuki Y, Osumi T, Kondo N, Fujiki Y.Human PEX1 cloned by functional complementation on aCHO cell mutant is responsible for peroxisome-deficientZellweger syndrome of complementation group I. Proc NatlAcad Sci USA 1998;95:4350-5.

33 Watkins PA, McGuinness MC, Raymond GV, Hicks BA,Sisk JM, Moser AB, Moser HW. Distinction betweenperoxisomal bifunctional enzyme and acyl-CoA oxidasedeficiencies. Ann Neurol 1995;38:472-7.

34 Moser HW. Genotype-phenotype correlations in disordersof peroxisome biogenesis. Mol Genet Metab 1999;68:316-27.

Novel mutations of FOXP3 in two Japanesepatients with immune dysregulation,polyendocrinopathy, enteropathy, X linkedsyndrome (IPEX)

Ichiro Kobayashi, Reza Shiari, Masafumi Yamada, Nobuaki Kawamura, Motohiko Okano,Asao Yara, Akihiro Iguchi, Nobuyoshi Ishikawa, Tadashi Ariga, Yukio Sakiyama,Hans D Ochs, Kunihiko Kobayashi

EDITOR—Immune dysregulation, polyendocrin-opathy, enteropathy, X linked syndrome(IPEX), also known as X linked autoimmunity-allergic dysregulation syndrome (XLAAD), ischaracterised by enteropathy and involvement ofthe endocrine system, such as insulin dependentdiabetes mellitus (IDDM) and thyroiditis,which develop in association with autoantibod-ies in early infancy (MIM 304930, 304790).1 2

IPEX has been mapped to chromosomeXp11.23-Xq13.3.3 4 Recent studies have indi-cated that FOXP3, a member of forkhead/winged-helix proteins, is a causative gene forboth IPEX and an equivalent mouse, scurfy.5–8

Human FOXP3 consists of 11 exons andencodes 431 amino acids containing a zincfinger (Zn) domain, a leucine zipper (Zip) motif,and a forkhead domain.6 8 We have previouslyreported two unrelated Japanese patients with Xlinked autoimmune enteropathy associated withtubulonephropathy and endocrinopathy.2 9 10

We report here novel mutations in the FOXP3gene of these patients.

Patients and methodsClinical and laboratory findings of our patientshave been previously reported.2 9 10 Briefly,patient 1, a boy, now 11 years old, wasdiagnosed as having autoimmune thyroiditis,autoimmune haemolytic anaemia, and auto-immune enteropathy at the age of 2 weeks, 2months, and 5 months, respectively. Renaltubular dysfunction was also noted. His mater-nal uncle and his older brother had died of asimilar diarrhoeal disease complicated byIDDM, suggesting X linked inheritance of thedisease.2 He has been treated with a combina-tion of tacrolimus (0.3 mg/day) and beta-methasone (0.3-0.5 mg/day).9 Hypocalcaemiaand hypokalaemia often develop in spite ofsupplementation with calcium, potassium, andvitamin D, which suggests renal damage result-ing from either the underlying disease or a side

874 Letters

J Med Genet2001;38:874–876

Department ofPaediatrics, HokkaidoUniversity School ofMedicine, North-15West-7, Kita-ku,Sapporo 060-8638,JapanI KobayashiR ShiariM YamadaN KawamuraM OkanoK Kobayashi

Paediatric Clinic,Naha MunicipalHospital, Naha, JapanA Yara

Paediatric Clinic,Kitami Red CrossGeneral Hospital,Kitami, JapanA IguchiN Ishikawa

Gene Therapy,Hokkaido UniversitySchool of Medicine,Sapporo, JapanT ArigaY Sakiyama

Division ofImmunology,Infectious Disease andRheumatology,Department ofPediatrics, Universityof Washington, Seattle,USAH D Ochs

Correspondence to:Dr Kobayashi,[email protected]

www.jmedgenet.com


http://jmg.bmj.com/


eVect of tacrolimus. In addition, he is suVeringfrom osteoporosis associated with multiplecompression fractures of the vertebral bodiesand steroid induced cataract. He has failed togain weight and height, although his diarrhoeahas been fairly well controlled.

Patient 2, a boy, was diagnosed as havingautoimmune enteropathy, tubulonephropathy,IDDM, and thyroiditis in his early infancy. Hedied of sepsis before using immunosuppressantsat the age of 3 years. His maternal uncle died ofintractable diarrhoea.9 Both of the patientsshowed extremely high levels of serum IgE.10 11

Genomic DNA was isolated from Epstein-Barr virus transformed cell lines in patient 1and his family and from a necropsy specimen inpatient 2 using a standard method. Each exonincluding the exon-intron junction was ampli-fied by polymerase chain reaction (PCR)(94°C for five minutes, followed by 35 cycles of94°C for 30 seconds, 60°C for 30 seconds, and72°C for one minute) with primers listed intable 1. Additional primers were also used forsequence analyses.8 PCR products were sup-plied for direct sequencing. Both PCR amplifi-cation and sequencing were performed usingGeneAmp PCR System 2400 (Perkin Elmer).

Sequence analyses were performed by GeneAnalyzer 310 (ABI PRISM).

Results and discussionPatient 1 showed a deletion of a singlenucleotide T227 in exon 2, which results in aframeshift and generates a premature stop atcodon 128 (fig 1A). Accordingly, the truncatedprotein in patient 1 lacks all of the domains andis apparently non-functional. This mutationwas not observed in his healthy brother, sister,or father. His mother was found to beheterozygous for this mutation. Seven muta-tions of FOXP3 have been reported in IPEX.Three cases carry single amino acid substitu-tions in the forkhead domain, whereas anotherhas a single amino acid deletion in the Zipdomain.6–8 Deletion of the forkhead domainwas reported in one case, which resulted fromexon 9 skipping and a frameshift accompaniedby premature termination.8 The remaining twocases involve deletion of a stop codon whichresults in the addition of new residues.6 7

Patient 2 showed an A1087G substitution inexon 10, which results in an Ile363Valsubstitution (fig 1B). Wildin et al6 reported thatno sequence variations are found in exons 10

Table 1 Primers used in the analyses of the FOXP3 gene

Forward primer Reverse primer

Exon 1 5'-AGTATCTCATACCGCCCTAGCACACGTGTGA-3' 5'-ACAGTAAAGGTCGGCACCTGTAGGTCCAGGTA-3'Exons 2–3 5'-AGTGCAGAGTATTTGAATTAGACACAGAACAGTG-3' 5'-AGAATAGCCTACACTGCTCACAGCCAAGGATCTG-3'Exons 4–5 5'-AGAGGCCTTGTGGGCCAAGCTCCAGAGCCCA-3' 5'-GAGCTGAGATCTGCACCCTAGACCTCTCCCCA-3'Exon 6 5'-GCATGTGTTAAGGGAACGAGGGGTGT-3' 5'-GGTTTTGCGCACTATCCCTA-3'Exon 7 5'-GGGATAGTGCGCAAACC-3' 5'-CAGCAGTCTGAGTCTGCCACCA-3'Exon 8 5'-GGCGACAGAGCAAGACTCAGTA-3' 5'-CACCCAGAGCCTGTCAGGATTAGGA-3'Exon 9 5'-GACGGTGAGATCTCAGGCCTGTAGACTCACCTTG-3' 5'-AACCCACTCTGAGGGCACTCAGAGGGAGACA-3'Exons 10–11 5'-AGGTCATAGCCCCTCTAAACCCCAAGTTTG-3' 5'-ACCTCTGCCTCCCACCAGTTTGGCCCCTGTTCG-3'

Figure 1 Sequence analyses of exon 2 of patient 1 (A) and exon 10 of patient 2 (B). Arrowhead indicates a deletion of Tat nt 227 in patient 1. Underline indicates an A to G transition at nt 1087 in patient 2.

Patient 2Patient 1

Control

Exon 10Exon 2 BA

Control

Letters 875

www.jmedgenet.com


http://jmg.bmj.com/


and 11 in unaVected subjects, indicating thatthe Ile363Val substitution is not a polymor-phism but a missense mutation. Ile363 islocated in the middle of the second á-helix ofthe forkhead domain and is highly conservedamong FOX family members,13 14 suggestingthat Ile363 is critical for the function ofFOXP3.

IPEX patients, as well as scurfy mouse, showautoimmune and allergic features which areassociate with skewed expression of Th2 typecytokines.1–8 15–17 Our patients showed variousautoantibodies against the intestine, renaltubules, thyroid, or red blood cells.2 12 In addi-tion, overexpression of Th2 type cytokinessuch as interleukin-4 has also been found in theperipheral blood mononuclear cells from pa-tient 1.11 Thus, it is predicted that FOXP3 con-tributes to maintenance of immune toleranceand regulates cytokine expression. Clarificationof its precise function in normal immunesystems could provide new insights into themechanism of autoimmunity and allergy.

We thank family members for their participation in this study.This work was supported by Grant in Aid from the Ministry ofEducation, Science, Sports and Culture, Japan.

1 Powell BR, Buist NRM, Stenzel P. An X-linked syndrome ofdiarrhea, polyendocrinopathy, and fatal infection in in-fancy. J Pediatr 1982;100:731-7.

2 Satake N, Nakanishi M, Okano M, Tomizawa K, Ishizaka A,Kojima K, Onodera M, Ariga T, Satake A, Sakiyama Y,Matsumoto S. A Japanese family of X-linked auto-immuneenteropathy with haemolytic anaemia and polyendocrino-pathy. Eur J Pediatr 1993;152:313-15.

3 Ferguson PJ, Blanton SH, Saulsbury FT, McDuYe MJ,Lemahieu V, Gastier JM, Francke U, Borowitz SM,Sutphen JL, Kelly TE. Manifestations and linkage analysisin X-linked autoimmunity-immunodeficiency syndrome.Am J Med Genet 2000;90:390-7.

4 Bennett CL, Yoshioka R, Kiyosawa H, Barker DF, Fain PR,Shigeoka AO, Chance PF. X-linked syndrome of polyendo-crinopathy, immune dysfunction, and diarrhea maps toXp11.23-Xq13.3. Am J Hum Genet 2000;66:461-8.

5 Brunkow ME, JeVery EW, Hjerrild KA, Paeper B, Clark LB,Yasayko SA, Wilkinson JE, Galas D, Ziegler SF, RamsdellF. Disruption of a new forkhead/winged-helix protein,scurfin, results in the fatal lymphoproliferative disorder ofthe scurfy mouse. Nat Genet 2001;27:68-73.

6 Wildin RS, Ramsdell F, Peake J, Faravelli F, Casanova JL,Buist N, Levy-Lahad E, Mazzella M, Goulet O, Perroni L,Bricarelli FD, Byrne G, McEuen M, Proll S, Appleby M,Brunkow ME. X-linked neonatal diabetes mellitus, enter-opathy and endocrinopathy syndrome is the human equiv-alent of mouse scurfy. Nat Genet 2001;27:18-20.

7 Bennett CL, Christie J, Ramsdell F, Brunkow ME,Ferguson PJ, Whitesell L, Kelly TE, Saulsburry FT,Chance PF, Ochs HD. The immune dysregulation, polyen-docrinopathy, enteropathy, X-linked syndrome (IPEX) iscaused by mutation of FOXP3. Nat Genet 2001;27:20-1.

8 Chatila TA, Blaeser F, Ho N, Lederman HM, Voulgaropou-los C, Helms C, Bowcock AM. JM2, encoding a forkhead-related protein, is mutated in X-linked autoimmunity-allergic dysregulation syndrome. J Clin Invest 2001;106:R75-81.

9 Kobayashi I, Nakanishi M, Okano M, Sakiyama Y,Matsumoto S. Combination therapy with tacrolimus andbetamethasone for a patient with X-linked auto-immuneenteropathy. Eur J Pediatr 1995;154:594-5.

10 Kobayashi I, Imamura K, Yamada M, Okano M, Yara A,Ikema S, Ishikawa N, Kobayashi K, Sakiyama Y. A 75-kDautoantigen recognized by sera from patients with X-linkedautoimmune enteropathy associated with nephropathy.Clin Exp Immunol 1998;111:527-31.

11 Kawamura N, Furuta H, Tame A, Kobayashi I, Ariga T,Okano M, Sakiyama Y. Extremely high serum level of IgEduring immunosuppressive therapy: paradoxical eVect ofcyclosporine A and tacrolimus. Int Arch Allergy Immunol1997;112:422-4.

12 Kobayashi I, Imamura K, Kubota M, Ishikawa S, YamadaM, Tonoki H, Okano M, Storch WB, Moriuchi T,Sakiyama Y, Kobayashi K. Identification of an autoimmuneenteropathy-related 75-kilodalton antigen. Gastroenterology1999;117:823-30.

13 Crisponi L, Deiana M, Loi A, Chiappe F, Uda M, Amati P,Bisceglia L, Zelante L, Nagaraja R, Porcu S, Ristalde MS,Marzella R, Rocchi M, Nicolino M, Lienhardt-Roussie A,Nivelon A, Verloes A, Schelessinger D, Gasparini P,Bonneau D, Cao A, Pilia G. The putative forkheadtranscription factor FOXL2 is mutated inblepharophimosis/ptosis/epicanthus inversus syndrome.Nat Genet 2001;27:159-66.

14 Blixt Å, Mahlapuu M, Atiola M, Pelto-Huikko M, EnerbäckS, Carlsson P. A forkhead gene, FOXE3, is essential for lensepithelial proliferation and closure of the lens vesicle. GenesDev 2000;14:245-54.

15 Lyon MF, Peters J, Glenister PH, Ball S, Wright E. Thescurfy mouse mutant has previously unrecognized hemato-logical abnormalities and resembles Wiskott-Aldrich syn-drome. Proc Natl Acad Sci USA 1990;87:2433-7.

16 Godfrey VL, Wilkinson JE, Russel LB. X-linked lymphore-ticular disease in the scurfy (sf) mutant mouse. Am J Pathol1991;138:1379-87.

17 Kanangat S, Blair P, Reddy R, Deheshia M, Godfrey V,Rouse BT, Wilkinson E. Disease in the scurfy (sf) mouse isassociated with overexpression of cytokine genes. Eur JImmunol 1996;26:161-5.

Maternal uniparental isodisomy 11q13→qter in adysmorphic and mentally retarded female withpartial trisomy mosaicism 11q13→qter

Dieter Kotzot, Benno Röthlisberger, Mariluce Riegel, Albert Schinzel

EDITOR—Partial trisomy mosaicism describesthe presence of a normal cell line together withan unbalanced translocation in a second cellline. Its incidence is not known. Only a fewcases have been published,1 almost all withdevelopmental delay and a pattern of dysmor-phism. The presence of a normal cell linepoints towards postzygotic formation, but theorigin and mechanism of formation have so faronly been investigated in one case of partial tri-somy 16p mosaicism and in another case ofpartial trisomy 21q mosaicism.2 3 In the

former, a complex formation by trisomy first,translocation second, and uniparental disomyand partial trisomy third was inferred. In thelatter, paternal meiotic origin ofder(21;21)(q10;q10) mosaicism (46,XX/46,XX,der(21;21)(q10;q10),+21) in a girl withmild Down syndrome was described.

Here, we report on a 25 year old woman withmental retardation, dysmorphic features, par-tial trisomy 11q13→qter mosaicism (46,XX,der(19)t(11;19)(q13;p13.3)/46,XX), maternaluniparental isodisomy 11q13→qter in the

876 Letters

J Med Genet2001;38:876–881

Institut fürMedizinische Genetik,Universität Zürich,Rämistrasse 74,CH-8001 Zürich,SwitzerlandD KotzotB RöthlisbergerM RiegelA Schinzel

Correspondence to:Professor Schinzel,[email protected]

www.jmedgenet.com


http://jmg.bmj.com/


normal cell line, and two maternal and onepaternal segment(s) 11q13→qter in theabnrmal cell line.

Case reportThe female patient is the second child of ahealthy, unrelated, white couple. An olderbrother is healthy. At the proband’s birth, hermother was 38 years old and her father was 39years old. Following information, the parentsopted against prenatal cytogenetic diagnosis.Delivery by caesarean section took place at 42weeks of a normal gestation. Weight (2500 g)and length (48 cm) were below the 10thcentile. A right inguinal hernia, ipsilateral pesequinovarus, and left hip dysplasia were surgi-cally corrected. At the age of 6 years, height(1.7 m) was on the 10th centile, weight (22 kg)on the 75th centile, and occipitofrontal headcircumference (OFC) (51 cm) on the 50thcentile. At the last re-examination at the age of

25 years, height (1.50 m) was below the 3rdcentile and OFC (55 cm) was between the 50thand 75th centile. Dysmorphic features in-cluded deep set eyes, downward slantingpalpebral fissures, broad nasal root, large nares,flat and broad philtrum, diastema of the lowerincisors, small and low set ears, contractures atboth elbows with inability to supinate orpronate, ulnar deviation of both hands at thewrists, slender fingers, clinodactyly of fingersIV and V, and small feet with hypoplastic nailsand partial 2/3 syndactyly (fig 1). Severe devel-opmental delay was obvious.

MethodsChromosome analysis was performed on GTGbanded fibroblast and lymphocyte cell culturesat a level of about 550 bands according tostandard procedures. Fluorescence in situhybridisation (FISH) was done according tothe manufacturer’s instructions (Vysis Inc,Downers Grove, IL, USA).

Figure 1 Face of the proband aged 6 months (A) and 6 years (B, C).

Table 1 Results of molecular investigations of genomicDNA from lymphocytes and fibroblasts using microsatellitesmapping to chromosome 11

Marker Location (M/P blood/P fibroblasts/F)

D11S2071 p15.5 ab/a/a/acD11S922 p15.5 bd/ad/ad/acD11S1318 p15.5 ac/ab/ab/bcD11S1334 p15.4 a/ab/ab/bD11S860 p15 ac/ab/ab/bcD11S1324 p14 bc/ab/ab/acD11S905 p12-p13 bc/ab/ab/abD11S436 p12-p11.22 bc/ab/ab/abCentromereD11S4191 q13.1 ac/ab/ab/bcD11S1314 q13.1 ab/ac/ac/acD11S916 q13-q23 ac/ab/ab/abD11S527 q13.5 ? ac/a/ab/bdD11S2002 q21-q23 ab/b/bc/cD11S901 q14 ad/d/bd/bcD11S876 q14.1-q23.3 ab/b/ab/aD11S940 q22 ab/bc(?)/bc/cD11S2000 q22 b/b/b/abD11S927 q22 bc/c/bc/abD11S1356 q23 ab/b/bd/cdD11S528 q22.3 ac/a/ac/bcD11S934 q23.3-q24 a/a/ab/bcD11S912 q25 bd/d/cd/acD11S968 q24.1-q25 ac/c/cd/bd

Informative markers are underlined. F = father, P = patient, M= mother.

Figure 2 Molecular results of marker D11S940 with twodistinct maternal alleles (A) and two copies of the paternalallele (E) in genomic DNA from the parents, only onematernal allele in DNA from nucleated blood cells of thepatient (B), but an additional paternal allele in genomicDNA from the patient’s fibroblasts (C), and only thepaternal allele in the derivative chromosome 19microdissected from the patient’s fibroblasts (D). Theadditional weak signal in DNA from nucleated blood cellsof the patient (B) is either an artefact or represents low levelmosaicism for a paternal allele.

A B C

D11S940

D E

Letters 877

www.jmedgenet.com


http://jmg.bmj.com/


For molecular investigations, DNA wasextracted from blood and cultured fibroblastsof the patient and from blood of both parents.PCR amplification of highly polymorphic mic-rosatellites (Research Genetics®, Huntsville,AL, USA) was carried out under standardcondition. Bands were visualised by silverstaining.

For a more detailed investigation of the originof the normal chromosomes 11 and of thesegment 11q13→qter translocated onto onechromosome 19, a new technical approachrecently developed in our laboratory was ap-plied.4 Briefly, the derivative chromosome 19

was microdissected from 10 metaphases accord-ing to standard procedures, the dissectedchromosomes collected in a PCR tube contain-ing 2 µl collection drop solution (10 mmol/lTris/HCl, 10 mmol/l NaCl, 3 mg/ml proteinaseK PCR grade), and incubated for two hours at60°C. Subsequently, proteinase K was inacti-vated at 90°C for 10 minutes. Whole genomeamplification was performed according to aslightly modified (no gelatine, 500 µmol/l dNTP,1 mmol/l MgCl, 100 µmol/l totally degenerated,15 nucleotide long primer) primer-extension-preamplification polymerase chain reaction(PEP-PCR) protocol.5 Finally, multiple highlypolymorphic microsatellites were analysed bytime release PCR (0.1 mmol/l dNTP, 0.24µmol/l primers, 1.25 U AmpliTaq Gold®, 1.5-2.5mmol/l MgCl, using 5 µl aliquots of the pream-plified DNA in a final volume of 50 µl in aTechne Progene® thermocyler for 56 cycles:60-64°C for four minutes, 94°C for oneminute), run on a 6% polyacrylamide gel, andbands visualised by silver staining.

ResultsChromosome analysis from lymphocytes at theage of 6 years showed partial trisomy 11qmosaicism (46,XX,der(19)t(11;19)(q13;p13.3)[17]/46,XX[83]). Twenty years later, bloodchromosome analysis showed a 46,XX karyo-type in 40 metaphases. In fibroblasts, mosai-cism was also present (46,XX,der(19)t(11;19)(q13;p13.3)de novo[30]/46,XX[20]). Involve-ment of chromosomes 11 and 19 in therearrangement was confirmed by FISH withlibraries of chromosomes 11 and 19 (Vysis®

Inc, Downers Grove, IL, USA). FISH with a19p subtelomeric probe (Vysis® Inc) showedsignals on both chromosomes 19 in the normalcell line and on the normal and the derivativechromosome 19 in the abnormal cell line, andFISH with an “all telomeres” probe (Vysis®

Inc) failed to show a telomere at the transloca-tion breakpoint. Therefore, the breakpoint on19p must be distal to the subtelomeric locus ofthis probe, and the rearrangement led to dupli-cation of 11q13→qter without significant con-comitant 19p deletion in the abnormal cellline. The karyotypes of both parents were nor-mal.

Molecular investigations performed at theage of 25 years showed maternal uniparentalisodisomy 11q13→qter in DNA from nucle-ated blood cells (table 1). In DNA from thepatient’s fibroblasts, paternal bands of weakerintensity of markers mapping to 11q13→qterwere found (fig 2). The breakpoint wasdetermined between markers D11S916 andD11S527. In blood and fibroblasts, biparentalinheritance of chromosome 19 markers wasshown. The results of investigations withseveral microsatellites from various other chro-mosomes were in agreement with correctpaternity (data not shown).

Following microdissection of the derivativechromosome 19 from 10 metaphases, primer-extension-preamplification, and subsequentmicrosatellite analysis, only a paternal alleleand no maternal allele was found at markersD11S912 and D11S940 (fig 2). The additional

Zygote

Line 1

Mitosis

Line 2

Line 3

a b c

a b

11q

11q

or

11

mat pat

Break

Interphase

Line 5

Line 4Reduplication

Mechanism 1

Reduplication

Normal cellline with matUPD (11q)

Abnormalcell line

11q

a b

11q

11q

d e

c

11q

Break

Reduplication

Mechanism 2


Abnormalcell line

c d

11q

Figure 3 Diagram of the mitotic mechanisms 1 and 2 explaining the molecular resultsobtained in the patient. Mechanism 1. Postzygotic formation with segmental UPD(11q13→qter) in the normal cell line and in the abnormal cell line. The zygote is normalwith biparentally inherited chromosomes 11 (line 1). During mitosis both homologues areduplicated regularly (line 2). Then, both paternal sister chromatids break at 11q13 and onepaternal sister chromatid is lost (line 3a) resulting in a daughter cell with del(11q13→qter)(line 3b) and a second daughter cell with t(11;19)(q13pter) (line 3c). Finally, most likelyduring interphase in both daughter cells, somatic reduplication of the maternal segmentoccurs (line 4a and b) resulting in segmental maternal UPD (11q13→qter) (line 5a andb). Mechanism 2. Postzygotic formation with segmental UPD (11q13→qter) in the normalcell line only. Again, the zygote is normal with biparentally inherited chromosomes 11 (line1) and during mitosis both homologues are duplicated regularly (line 2). Then, the breakoccurs in only one paternal sister chromatid, which is translocated to one chromosome 19.Therefore, one daughter cell has del(11)(q13→qter) (line 3d), while the other has onenormal maternal chromosome 11, one normal paternal chromosome 11, and an additionalpaternal segment 11q13→qter translocated to 19p (line 3e). At least, in the normal cell lineduring interphase, somatic reduplication occurs (line 4c) resulting in segmental maternalUPD (11q13→qter) (line 5c).

878 Letters

www.jmedgenet.com


http://jmg.bmj.com/


weak signal in DNA from the patient’s blood atthe level of the paternal allele was interpretedas either an artefact or as representing low levelmosaicism for the der(19) chromosome.

DiscussionThe unexpected molecular results obtained inthe present case could theoretically be ex-plained by three diVerent mitotic and two mei-otic mechanisms of formation.

(1) Each homologue of chromosome 11 wasregularly duplicated in the late interphase of anormal mitosis (fig 3, left, line 2). Bothpaternal sister chromatids broke and simulta-neously one chromatid segment 11q13→qterwas translocated to one chromosome 19, whileher sister chromatid segment was lost (line 3a).In the anaphase, one normal maternal chroma-tid and one deleted paternal chromatid segre-gated to each of the two daughter cells, as wellas the derivative chromosome 19 to one of thedaughter cells (line 3c) and the normalchromosome 19 to the other (line 3b). Finally,mitotic reduplication of the maternal segment11q13→qter in both daughter cells (line 4a andb) resulted in maternal segmental uniparentalisodisomy 11q13→qter in the normal (line 5a)as well as in the abnormal cell line (line 5b).

(2) Again, in a normal 46,XX zygote, inher-itance of chromosome 11 was biparental andeach homologue of chromosome 11 was dupli-cated in the late interphase of a regular mitosis

(fig 3, right, line 2). Subsequently, after separa-tion of the sister chromatids, a break occurredin only one paternal chromatid at 11q13, andthe segment 11q13→qter was translocated tochromosome 19. The complete paternal chro-matid segregated together with the derivativechromosome 19 regularly to one daughter cell(line 3e), while the isolated part of the 11qchromatid segregated with the other daughtercell together with two normal chromosomes 19(line 3d). Finally, mitotic reduplication of thematernal segment 11q13→qter in the daughtercell containing the deleted 11 (line 4) resultedin maternal uniparental disomy (UPD)(11q13→qter) in the now (through correction)normal cell line (line 5c) and, in contrast to thefirst mechanism, in biparental inheritance ofthe whole chromosome 11 in the abnormal cellline (line 5d).

(3) Considering the third mechanism, twoindependent mitotic events must be postulated(fig 4). The first is characterised by anexchange between two non-sister chromatids(line 3a) resulting in two daughter cells withopposite segmental UPD (11q13→qter) (line6a and b). The cell line with maternal UPD(11q13→qter) would persist as shown in blood(line 6a), whereas the cell line with paternalsegmental UPD (11q13→qter) would not beviable (line 6b). In another cell with biparentalinheritance of chromosome 11 (line 2b), abreak in either one (line 3c) or in both (line 3b)

Figure 4 Diagram of the mitotic mechanism 3 explaining the molecular results obtained in the present case. After severalmitotic cell divisions of a normal zygote (line 1) with biparentally inherited chromosomes 11, in one cell (line 2a) anexchange between two non-sister chromatids occurs (line 3a). The results are two cells with opposite maternal and paternalsegmental UPD (11q13→qter) (line 6a and b). The cell with paternal UPD (11q13→qter) is not viable (line 6b). Inaddition, in a completely independent mitosis either a break in both (line 3b) or in only one chromatid (line 3c) occursfollowed by a scenario as described in fig 2, except the lethality of the cell lines with del(11)(q13qter) (line 6d and g). In thealso possible case of maternal UPD (11q13→qter) in the abnormal cell line, again the maternal segment 11q13→qter ismitotically reduplicated in interphase (lines 5 and 6c).

Zygote

Mitosis

Mitosis

Mechanism 3

Line 2

Line 1

Line 3

Line 4

11

mat pat

Interphase

Line 6

Line 5

19

a

X

a

b

a


b

Lethal

c

Reduplication

Abnormalcell line

d

Lethal

fe11q

11q

Abnormalcell line

g

Lethal

Break Breakb

11q

11q

11q

or

c

Letters 879

www.jmedgenet.com


http://jmg.bmj.com/


chromatids would have occurred. In the lattercase, one segment would have been translo-cated to one chromosome 19 (line 4), whereasthe other would have been lost (line 6e). Thecell line with del(11)(q13qter) would not havebeen viable (line 6d). In the other cell line,maternal segmental UPD (11q13→qter) (line6c) would have been formed by reduplicationduring interphase (line 5). In the alternativecase of breaking of only one chromatid (line3c), one cell line with del(11)(q13qter) (lethal)(line 6g) and a second cell line with biparen-tally inherited chromosomes 11 and an addi-tional translocation of the paternal segment11q13-qter would have arisen (line 6f). Thispartial mechanism is less likely, because themolecular investigations showed a weakerintensity of the paternal versus the maternalalleles.

(4) A fourth mitotic mechanism (fig 5)would be characterised by a normal zygote andsegmental mitotic reduplication on the mater-nal chromosome. Then, a crossing over be-tween one paternal chromatid and the redupli-cated segment as well as translocation of thepaternal segment to 19p occurred. Segregationresulted in the karyotype described.

(5) A fifth, in part meiotic mechanism wouldrequire a balanced 11;19 translocation alreadypresent in the paternal gamete (fig 6, left, line1a). Thereafter, again a mitotic reduplicationof the maternal segment 11q13→qter must

have happened (line 2) and, in addition, in thenormal cell line the loss of the translocatedpaternal segment 11q13→qter (line 5a) or eventhe loss of the entire derivative chromosome 19combined with mitotic reduplication of its nor-mal homologue must be postulated. The latterwas excluded by the biparental inheritance ofchromosome 19 in blood and fibroblasts (datanot shown).

(6) A sixth, again in part meiotic mechanismwould require a trisomic zygote resulting frommaternal non-disjunction (fig 6, right, line 1b).Mitotic crossing over between the paternalchromatid and one maternal chromatid (line4b) would result in two diVerent daughter cells(line 5e and f). One with biparental chromo-somes 11 and translocation of the paternal seg-ment 11q13→qter to one chromosome 19 (line5e), and a second with exclusively maternal11q13→qter material (line 5f). In addition, inboth cell lines, one chromosome must havebeen lost, the chromosome with del(11)(q13→qter) in one (line 5d), and one completematernal chromosome 11 in the other (line5g). This mechanism is supported by theadvanced maternal age of 38 years at delivery;on the other hand, complete isodisomy wouldbetter fit a postmeiotic formation.

In total, each mechanism requires a mini-mum of three subsequent or in part simultane-ous events. With the presently available mo-lecular investigations, it was not possible todefine one single possible mechanism offormation. Isodisomy in the normal cell linecould be considered to indicate mitotic forma-tion. Apart from paternal UPD (11p15) in upto 20% of cases with Beckwith-Wiedemannsyndrome,6 segmental UPD has rarely beenreported.7–11 Similarly, paternal UPD of thewhole chromosome 11 has been described onlytwice, in a fetus with severe intrauterine growthretardation, intestinal malrotation, and con-fined placental mosaicism,12 and in a mosaicstate in a girl with Beckwith-Wiedemannsyndrome.13 To the best of our knowledge,maternal segmental UPD of chromosome 11has not been reported so far. Full trisomy 11 isnot viable and non-mosaic duplication of11q13→qter has been reported only rarely.The phenotype of the latter is characterised bycardiac anomalies, restricted elbow movementsincluding supination and pronation, facial dys-morphism, small and low set ears, and mentalretardation.1

The clinical consequences of the complexrearrangement found in our patient are diYcultto evaluate. The phenotype strongly resemblesseveral cases with duplication of 11q13→qter,and thus could be caused only by this. On theother hand, an additional eVect of mosaicismfor the loss of the normal homologue of animprinted gene and/or of homozygosity for amutated allele of a recessive gene cannot yet befully excluded.

We are grateful to the family for their cooperation. The studywas supported by the Swiss National Foundation, grant Nos32-45604.95 and 32-56053.98.

Figure 5 Diagram for mechanism 4 with partial endoreduplication and subsequent(“jumping”) translocation onto chromosome 19 of the segment 11q13→qter.

Zygote

Mitosis

1119

mat pat

11q

Normal cell linewith mat UPD (11q)

Abnormal cell line

11q

Reduplication of 11q

X

880 Letters

www.jmedgenet.com


http://jmg.bmj.com/


1 Schinzel A. Human cytogenetics database. Oxford: OxfordMedical Databases Series, Oxford: Oxford UniversityPress, 1994.

2 Schinzel A, Kotzot D, Brecevic L, Robinson WP, Dutly F,Dauwerse H, Binkert F, Ausserer B. Trisomy first, translo-cation second, uniparental disomy and partial trisomythird: a new mechanism for complex chromosomalaneuploidy. Eur J Hum Genet 1997;5:308-14.

3 Kotzot D, Schinzel A. Paternal meiotic origin of der(21;21)(q10;q10) mosaicism (46,XX/46,XX, der(21;21)(q10;

q10),+21) in a girl with mild Down syndrome. Eur J HumGenet 2000;8:709-12.

4 Röthlisberger B, Schinzel A, Kotzot D. A new molecularapproach to investigate origin and formation of structuralchromosome aberrations. Chrom Res 2000;8:451-3.

5 Dietmaier W, Hartmann A, Wallinger S, Heinmöller E,Kerner T, Endl E, Jauch KW, Hofstädter F, RüschoV J.Multiple mutation analyses in single tumor cells withimproved whole genome amplification. Am J Pathol1999;154:83-95.

6 Henry I, Puech A, Riesewijk A, Ahnine L, Mannens M,Beldjord C, Bitoun P, Tournade MF, Landrieu P, Junien C.Somatic mosaicism for partial paternal isodisomy inWiedemann-Beckwith syndrome: a post-fertilization event.Eur J Hum Genet 1993;1:19-29.

7 Collier DA, Barrett T, Curtis D, Macleod A, Bundey S.DIDMOAD syndrome: confirmation of linkage to chromo-some 4p, evidence for locus heterogeneity and a patientwith uniparental isodisomy for chromosome 4p. Am J HumGenet Suppl 1995;57:1084.

8 Lopez-Gutierrez AU, Riba L, Ordonez-Sanchez ML,Ramirez-Jimenez, Cerrillo-Hinojosa M, Tusie-Luna MT.Uniparental disomy for chromosome 6 results in steroid21-hydroxylase deficiency: evidence of diVerent geneticmechanisms involved in the production of the disease. JMed Genet 1998;35:1014-19.

9 Martin RA, Sabol DW, Rogan PK. Maternal uniparentaldisomy of chromosome 14 confined to an interstitialsegment (14q23-14q24.2). J Med Genet 1999;36:633-6.

10 Mergenthaler S, Wollmann HA, Kloos P, Albrecht B,Spranger S, Eggermann K, Zerres K, Eggermann T. Clini-cal indications for uniparental disomy (UPD) testing ingrowth retarded patients: presentation of own results bysearching for UPDs 2, 6, 7, 14 and 20. Am J Hum Genet2000;67(suppl 2):599.

11 Yang XP, Inazu A, Yagi K, Kajinami K, Koizumi J, MabuchiH. Abetalipoproteinemia caused by maternal isodisomy ofchromosome 4q containing an intron 9 splice acceptormutation in the microsomal triglyceride transfer proteingene. Arterioscler Thromb Vasc Biol 1999;19:1950-5.

12 Webb A, Beard J, Wright C, Robson S, Wolstenholm J,Goodship J. A case of paternal uniparental disomy forchromosome 11. Prenat Diagn 1995;15:773-7.

13 Dutly F, Baumer A, Kayserili H, Yüksel-Apak M, Zerova T,Hebisch G, Schinzel A. Seven cases of Wiedemann-Beckwith syndrome, including the first reported case ofmosaic paternal isodisomy along the whole chromosome11. Am J Med Genet 1998;79:347-53.

Figure 6 Diagram of the more complex and therefore more unlikely meiotic mechanisms to explain the molecular resultsobtained in the present case. Mechanism 5. A paternal 11;19 translocation is already present in the zygote (line 1a).Subsequently, during mitosis, somatic reduplication of the maternal segment 11q13→qter occurs (line 2). In a secondmitosis the paternal segment 11q13→qter translocated to 19p must be lost (line 5a) to receive a cytogenetically normal cellline (line 5b). The last step seems particularly unlikely. Mechanism 6. Subsequent to a trisomic zygote owing to maternalnon-disjunction (line 1a) a crossing over between the paternal and one maternal allele occurred (line 4b) and two daughtercells arose, one with biparental inheritance of chromosome 11 and translocation of 11q13→qter to chromosome 19 (line 5e)and one with UPD 11q13→qter (line 5f). In addition, one chromosome 11 deleted for 11q13→qter (line 5d) and onecomplete maternal chromosome 11 (line 5g) were lost during mitosis.

ZygoteLine 1

Mitosis

Line 3

Line 2

Line 5Normal cell linewith mat UPD (11q)

Abnormalcell line

Mechanism 5

Line 4

11a

Reduplication

a

ba

mat pat

19

11q

11q

11q

11qc

11q

11q

Normal cell linewith mat UPD (11q)

Abnormalcell line

Mechanism 6

11b

b

ed g

mat pat

19

11q

11p

f

X

+ A 25 year old woman with mentalretardation, dysmorphic features, andpartial trisomy mosaicism 11q13→qter(karyotype 46,XX,der(19)t(11;19)(q13;p13.3)/46,XX) is reported.

+ Mosaicism for the abnormal karyotypewas found in blood in childhood, whereasonly the normal chromosomal comple-ment was detectable at the age of 25years, while in fibroblasts mosaicism waspresent.

+ Molecular investigations showed mater-nal uniparental isodisomy 11q13→qter inthe normal cell line in DNA from bloodcells and a weaker additional paternalallele by analysing DNA extracted fromcultivated fibroblasts.

+ Using a new technical approach combin-ing microdissection, primer-extension-preamplification PCR, and subsequentmicrosatellite analysis, we were able toshow that the der(19) chromosomecontains the paternal allele in the unbal-anced cell line.

Letters 881

www.jmedgenet.com


http://jmg.bmj.com/


High frequencies of ICF syndrome-likepericentromeric heterochromatin decondensationand breakage in chromosome 1 in a chorionicvillus sample

Melanie Ehrlich, Fern Tsien, Delma Herrera, Viola Blackman, Jennifer Roggenbuck,Cathy M Tuck-Muller

EDITOR—The immunodeficiency, centromericregion instability, and facial anomalies syn-drome (ICF) usually involves mutations aVect-ing the catalytic domain in DNMT3B, one ofthe three human genes known to encode DNAmethyltransferases.1–3 ICF always results indefective immunity, a high frequency ofchromosomal abnormalities in the vicinity ofthe centromere (pericentromeric region) ofchromosome 1 and/or chromosome 16 inmitogen stimulated lymphocytes, and hy-pomethylation of a small portion of thegenome.4–6 ICF symptoms are manifested oftenfrom infancy and this syndrome can cause earlychildhood death from infections. The DNAsequences targeted for undermethylation inICF include the heterochromatin adjacent tothe centromeres of chromosomes 1 and 16(1qh and 16qh), where a high incidence ofchromatin decondensation, chromosome andchromatid breaks, and rearrangements to formmultiradial chromosomes are characteristicallyseen in mitogen stimulated ICF blood culturesand in ICF lymphoblastoid cell lines.6–9 Theseaberrations are more common in chromosome1 than in chromosome 16 and only infre-quently observed in chromosome 9.5 8 9 Wedescribe an unusual primary culture from achorionic villus (CV) biopsy in which a highfrequency of ICF-like chromosomal abnor-malities was observed. However, follow upindicated that the infant did not have ICF.

A 30 year old, gravida 4, para 0, ab 3 whitefemale was referred for genetic counselling andCV sampling because of a previous pregnancywith trisomy 13. Both the patient and herhusband were phenotypically normal andhealthy apart from their reproductive history.The patient’s first two pregnancies ended inspontaneous abortion at 15 and 10 weeks. Nofetal studies had been performed; parental chro-mosomes were analysed and reported to be nor-mal. The third pregnancy was found to beaVected with trisomy 13 (47,XY,+13) and wasthen terminated. During the fourth pregnancy,the patient was oVered and accepted CVsampling for prenatal diagnosis of trisomyconditions.

In this CV sample, routine cytogeneticanalysis of 20 metaphases from an eight dayculture in Chang B medium with Chang Csupplement (Irvine Scientific) showed fourcells with pericentromeric breaks in chromo-some 1, one with a deletion of the long arm of

chromosome 1, and seven with decondensa-tion of 1qh (fig 1B, C). The high frequency ofchromosome 1 abnormalities led to the exam-ination of an additional 100 metaphases fromthe eight day culture. Forty-one percent of the120 cells examined had abnormalities in thepericentromeric region of chromosome 1.Decondensation of the pericentromeric het-erochromatin of chromosomes 1, 16, and 9was seen in 38, 5, and 0.8% (one cell) of themetaphases, respectively (fig 1). Eleven cells(9%) had chromosome breaks in the pericen-tromeric heterochromatin of chromosome 1such that both arms were present but widelyseparated, one cell (0.8%) had a similar breakat 16qh, and one (0.8%) had a pericentro-meric break of only one chromosome 1chromatid. Three cells (2.5%) had a deletionof 1q. In addition, a triradial(1)(p,q,q) wasobserved (fig 1D).

We passaged these CV cells three times at 1:5splits after the clinical analysis and examined100 additional metaphases. The frequency of16qh decondensation increased from 5% to26% and decondensation of 1qh increased from38% to 49% at passage 3 compared to the initialeight day primary culture. At passage 3, therewere 11 metaphases with a deletion of 1q, sixwith a chromosome break at 1qh, two with amultiradial chromosome (a triradial(1)(p,p,q) inone and a quadriradial(1)(p,p,q,q) in the other),one with an isochromosome composed of twopericentromerically fused 1q arms, and one witha deletion of 16q.

Because the ICF syndrome, in which thesespecific chromosomal anomalies are found, isalways accompanied by hypomethylation oflimited portions of the genome, including themajor DNA sequence in 1qh and 16qh,satellite 2 (Sat2), we examined methylation ofchromosome 1 Sat2 DNA sequences in the

Figure 1 Examples of the diVerent chromosome 1 and 16abnormalities found in the proband’s CV sample after aneight day culture. (A) Normal chromosome 1, (B)decondensation of chromosome 1 at the qh region, (C)whole arm deletion of 1q, (D) triradial(1)(p,q,q), (E)normal chromosome 16, (F) decondensation of 16qh.

A B C D E F

882 Letters

J Med Genet2001;38:882–884

Human GeneticsProgram/HaywardGenetics Center,Tulane MedicalSchool, 1430 TulaneAvenue, New Orleans,LA 70112, USAM EhrlichF TsienD HerreraV BlackmanC M Tuck-Muller

Department ofBiochemistry, TulaneMedical School, NewOrleans, LA 70112,USAM Ehrlich

Departments ofPediatrics andObstetrics andGynecology, OchsnerClinic, New Orleans,LA 70121, USAJ Roggenbuck

Department ofMedical Genetics,University of SouthAlabama, Mobile, AL36688, USAC M Tuck-Muller

Correspondence to:Dr Ehrlich,[email protected]

www.jmedgenet.com


http://jmg.bmj.com/


patient’s CV sample. This was done aspreviously described6 by Southern blot analysisof BstBI digests of DNA from these cellcultures at passage 3 with a Sat2 probe specificfor chromosome 1 (fig 2). We found that DNAfrom the patient’s CV sample was hypomethyl-ated in this 1qh sequence compared to normalpostnatal somatic tissues. However, the pa-tient’s CV DNA was no more hypomethylatedthan randomly chosen CV samples (fig 2). Theundermethylation of Sat2 DNA in these CVsamples was the same as that seen in term pla-centa.10 The diVerences in DNA methylationbetween normal postnatal tissues and CV sam-ples or placenta are consistent with their diVer-ent cell lineages. Term placenta (from whichthe embryonic membranes have been re-moved) and its chorionic villi are mostlyderived from extraembryonic cells (see below)as compared to postnatal somatic tissues,which are derived from epiblast cells of theembryoblast (inner cell mass).

One case of ICF had been diagnosed from anamniotic fluid culture in a family known to beat risk because of a previously aVected child.12

The other was diagnosed from a blood sampleof a 20 week old fetus who was the sib of anICF patient.4 There have been no reports ofprenatal diagnosis of ICF by CV sampling,although by linkage analysis ICF was excluded(90% probability) in a CV sample from a fam-ily with an aVected child.13

The present patient was counselled that thecytogenetic findings on the CV sample couldreflect a culture artefact, a fetus aVected withICF, or other unknown aetiology. Given thehigh frequency of ICF-like chromosomalanomalies in the patient’s CV cells, a follow upstudy of amniotic fluid at 16 weeks was doneand showed no cytogenetic anomalies. At 39weeks of gestation, the patient delivered ahealthy male infant, weighing 3856 g, with nofeatures of the ICF syndrome. The patientdeclined a peripheral blood study on the infantbut at the current age of 23 months, the child ishealthy and thriving.

Sat2 methylation analysis was done on thepatient’s amniocytes as described above for

DNA from the CV cultures. Amniocytes, likepostnatal somatic tissues, are derived from epi-blast cells of the embryoblast. A high level ofmethylation of this sequence was seen in DNAfrom the patient’s amniocyte culture as well asin random amniocyte samples (data notshown). These results are incompatible withthe patient harbouring ICF type DNMT3Bmutations because all tested ICF cell popula-tions from diverse tissues or cultured cell typesdisplay Sat2 hypomethylation.4–6 11

We compared the frequency of ICF-likepericentromeric chromosome 1 anomalies inthis patient’s CV sample to others. Retrospec-tive examination of 2250 metaphases from 26clinical CV samples (50-100 metaphases each)from eight day random CV cultures, which hadbeen interpreted as normal, showed that only47 of the metaphases (2%) had a pericentro-meric abnormality of chromosome 1 or 16.However, the majority of the CV samples(58%) displayed low levels of these anomalieswith 1-8% (median 2%) of their metaphasesshowing these chromosome 1 or chromosome16 aberrations. Ninety-four percent of theseanomalies were decondensation of 1qh or16qh, with 3.4-fold more 1qh than 16qhdecondensation. The only pericentromericrearrangements seen were two whole arm dele-tions of 1q and one break at 1qh resulting inseparated 1p and 1q arms. No clonal abnor-malities except for the chromosome 1 andchromosome 16 aberrations were observed inthese metaphases from randomly chosen CVsamples. These CV samples were obtained overseveral years and included samples tested at thesame time as the patient’s sample and all wereanalysed by the same method. Also, no changein the lot of medium or fetal calf serum canexplain the diVerent results obtained from thepatient’s sample and the random samples.

Anecdotal observations of these types ofpericentromeric chromosome 1 and 16anomalies in normal CV metaphases are com-mon although, to our knowledge, they havebeen described by only one group in detail14

and mentioned by another.13 In the latter case,Bjorck et al13 stated without elaboration that“heterochromatic decondensation can occa-sionally be seen in both amniocytes andcultured chorionic villi without any pathologi-cal significance.” The former group, Miguez etal,14 found in a study of 244 24 hour CV cul-tures that about 9% of the metaphasesdisplayed chromosomal lesions, usually breaksor gaps at various fragile sites. The most com-mon site was at 1qh (1q12) or 1q21.1although the quality of the chromosomesallowed only 36% of the preparations to be“successfully banded” and the frequency ofcases with this anomaly was not given. Inanother report, this group describeddecondensation in 1qh, 9qh, 16qh, or Yqh in2.4, 3.6, and 0.3, and 0.2%, respectively, of the5820 examined metaphases, with 47% of 33924 hour CV cultures displaying such conden-sation in at least one metaphase.15 In contrastto those investigators who observed deconden-sation in 9qh to be most frequent, we found

Figure 2 Southern blot analysis showing hypomethylationof Sat2 DNA from chromosome 1 in term placenta, CVsamples, and ICF cells but not in postnatal somatic tissues.BstBI digested DNA from the following samples washybridised to a chromosome 1 Sat2 specific probe andautoradiographed: term placenta; the patient’s CV(CV1)culture from passage 3; four random CV samples frompassage 3 (CV A, CV C, CV G, and CV H); lung andthymus from normal trauma victims; and a lymphoblastoidcell line (LCL) from an ICF patient.

Pla

cen

ta

CV

1

CV

A

CV

C

CV

G

CV

H

Lun

g

Th

ymu

s

ICF

LCL

Letters 883

www.jmedgenet.com


http://jmg.bmj.com/


only a single metaphase exhibiting deconden-sation of 9qh in this study and that was in theCV sample with the high frequency of 1qhdecondensation. This diVerence between theirresults and ours could be because of the use ofone day rather than eight day cultures. Thelonger culture time aVords better quality met-aphase chromosomes but selects for extraem-bryonic mesoderm cells (which are derivedfrom hypoblast cells of the embryoblast) in theCV samples as opposed to mostly cytotro-phoblast metaphases (which are trophoblastderivatives) in the one day CV cultures.16 Celltype specific diVerences in the frequencies ofchromosome anomalies could also explainwhy we did not observe chromosomal aberra-tions in the eight day amniocyte culture fromthe patient who gave such a high percentage ofchromosome 1 anomalies in the correspond-ing CV culture.

The chromosome 1 (and chromosome 16)pericentromeric aberrations observed in thisclinical CV sample may have formed duringculture in vitro. However, there was nothingremarkable about the CV tissue used for cellculture or in this culture’s hypomethylation inthe 1qh region that could explain why it yieldedsuch a high frequency of chromosome 1anomalies. Whatever the cause of the anoma-lies in this patient’s sample, our analysis of CVsamples indicates that ICF-like chromosomalabnormalities are part of the normal spectrumfor CV chromosomes and need not indicateany clinical condition. Furthermore, we con-clude that CV sampling should not beattempted to prenatally diagnose ICF by thechromosome abnormalities used to diagnosethis syndrome in mitogen treated bloodsamples from immunodeficient patients. Infamilies at risk for ICF, prenatal analysis forDNMT3B mutations13 would be preferable.

This work was partially supported by NIH grant CA 81506 (toME).

1 Hansen RS, Wijmenga C, Luo P, Stanek AM, Canfield TK,Weemaes CM, Gartler SM. The DNMT3B DNA methyl-transferase gene is mutated in the ICF immunodeficiencysyndrome. Proc Natl Acad Sci USA 1999;96:14412-17.

2 Xu G, Bestor TH, Bourc’his D, Hsieh C, Tommerup N,Hulten M, Qu S, Russo JJ, Viegas-Péquignot E. Chromo-some instability and immunodeficiency syndrome causedby mutations in a DNA methyltransferase gene. Nature1999;402:187-91.

3 Wijmenga C, Hansen RS, Gimelli G, Bjorck EJ, Davies EG,Valentine D, Belohradsky BH, van Dongen JJ, Smeets DF,van den Heuvel LP, Luyten JA, Strengman E, Weemaes C,Pearson PL. Genetic variation in ICF syndrome: evidencefor genetic heterogeneity. Hum Mutat 2000;16:509-17.

4 Jeanpierre M, Turleau C, Aurias A, Prieur M, Ledeist F,Fischer A, Viegas-Pequignot E. An embryonic-like meth-ylation pattern of classical satellite DNA is observed in ICFsyndrome. Hum Mol Genet 1993;2:731-5.

5 Smeets DFCM, Moog U, Weemaes CMR, Vaes-Peeters G,Merkx GFM, Niehof JP, Hamers G. ICF syndrome: a newcase and review of the literature. Hum Genet 1994;94:240-6.

6 Tuck-Muller CM, Narayan A, Tsien F, Smeets D, Sawyer J,Fiala ES, Sohn O, Ehrlich M. DNA hypomethylation andunusual chromosome instability in cell lines from ICF syn-drome patients. Cytogenet Cell Genet 2000;89:121-8.

7 Fryns JP, Azou M, Jacken J, Eggermont E, Pedersen JC, Vanden Berghe H. Centromeric instability of chromosomes 1,9, and 16 associated with combined immunodeficiency.Hum Genet 1981;57:108-10.

8 Tiepolo L, Maraschio P, Gimelli G, Cuoco C, Gargani GF,Romano C. Multibranched chromosomes 1, 9, and 16 in apatient with combined IgA and IgE deficiency. Hum Genet1979;51:127-37.

9 Turleau C, Cabanis MO, Girault D, Ledeist F, Mettey R,Puissant H, Marguerite P, de Grouchy J. Multibranchedchromosomes in the ICF syndrome: immunodeficiency,centromeric instability, and facial anomalies. Am J MedGenet 1989;32:420-4.

10 Narayan A, Ji W, Zhang XY, Marrogi A, GraV JR, Baylin SB,Ehrlich M. Hypomethylation of pericentromeric DNA inbreast adenocarcinomas. Int J Cancer 1998;77:833-8.

11 Miniou P, Jeanpierre M, Bourc’his D, Coutinho BarbosaAC, Blanquet V, Viegas-Pequignot E. Alpha-satellite DNAmethylation in normal individuals and in ICF patients:heterogeneous methylation of constitutive heterochromatinin adult and fetal tissues. Hum Genet 1997;99:738-45.

12 Fasth A, Forestier E, Holmberg E, Holmgren G, NordensonI, Soderstrom T, Wahlstrom J. Fragility of the centromericregion of chromosome 1 associated with combinedimmunodeficiency in siblings: a recessively inheritedentity? Acta Paediatr Scand 1990;79:605-12.

13 Bjorck EJ, Bui TH, Wijmenga C, Grandell U, NordenskjoldM. Early prenatal diagnosis of the ICF syndrome. PrenatDiagn 2000;20:828-31.

14 Miguez L, Fuster C, Perez MM, Miro R, Egozcue J. Spon-taneous chromosome fragility in chorionic villus cells. EarlyHum Dev 1991;26:93-9.

15 Perez MM, Miguez L, Fuster C, Miro R, Genesca G,Egozcue J. Heterochromatin decondensation in chromo-somes from chorionic villus samples. Prenat Diagn 1991;11:697-704.

16 Crane JP, Cheung SW. An embryogenic model to explaincytogenetic inconsistencies observed in chorionic villusversus fetal tissue. Prenat Diagn 1988;8:119-29.

884 Letters

www.jmedgenet.com


http://jmg.bmj.com/


Supernumerary marker chromosome (1) ofpaternal origin and maternal uniparental disomy 1in a developmentally delayed child

Benno Röthlisberger, Tatjana E Zerova, Dieter Kotzot, Tamara I Buzhievskaya, DaminaBalmer, Albert Schinzel

EDITOR—At least 168 cases with a supernu-merary marker chromosome (SMC) from allchromosomes not including chromosome 15have been documented.1 Birth prevalence isestimated at 0.14 to 0.72 per 1000.2 Subjectswith a SMC have a partial trisomy (duplica-tion) and in some cases a partial tetrasomy(triplication) of the genetic material containedin the SMC. The risk of an abnormalphenotype associated with a randomly ascer-tained de novo SMC derived from acrocentricautosomes (excluding chromosome 15) is esti-mated to be approximately 7% compared withapproximately 28% for SMCs derived fromnon-acrocentric autosomes.1 The great vari-ability of clinical findings in patients withSMCs originating from the same chromosomeis probably the result of variation in size andgenetic content, the degree of mosaicism, anduniparental disomy of the normal homologuesof the chromosome from which the SMCderived.

Evidence that subjects with SMCs mighthave an increased risk for UPD of the structur-ally normal homologues of the SMCs has beenreported by several authors. To the best of ourknowledge the coexistence of SMCs with UPDhas been described for chromosomes 6, 7, 15,20, and X.3–7 Here, we describe a furtherpatient with multiple congenital anomalies,developmental delay, and the unique finding ofcoexistence of SMC 1 mosaicism and maternaluniparental disomy 1.

Case reportThe female patient was born at term after anuneventful pregnancy. At her birth, her motherwas 33 years old and her father was 47 yearsold. Birth weight was 2500 g and length 49 cm.At the age of 6 years, she was investigatedbecause of mental retardation. Height (1.14 m)and weight (17 kg) were within the normal

range, but head circumference (44 cm) was farbelow the 3rd centile. Additional findings weretemporal narrowing, downward slanting palpe-bral fissures, long eyelashes, high palate,pointed chin, low set and dysplastic ears, hipdysplasia, and tapering fingers with clinodac-tyly of fingers 2, 4, and 5.

MethodsChromosome preparations from blood lym-phocyte cultures were performed according tostandard procedures. GTG banded chromo-somes were examined.

Chromosome microdissection was per-formed according to a slightly modified versionof the protocol described by Senger et al.8 FISHanalysis was performed according to Lichterand Ried9 using the specific chromosomelibrary generated by microdissection.

Genomic DNA from the patient and bothparents was amplified by standard PCR withcommercially available highly polymorphicmicrosatellites (Research Genetics®, Hunts-ville, AL, USA), loaded onto a 6%polyacrylamide/urea gel, and visualised bysilver staining.

Probably owing to the low mosaicism in leu-cocytes, it was not possible to determine theparental origin of the SMC unambiguously bystandard methods. We therefore applied a newapproach recently developed in our labora-tory.10 In a first step, 20 SMCs were dissectedand collected in a PCR tube containing 2 µlcollection drop solution (10 mmol/l Tris/HCl,10 mmol/l NaCl, 3 mg/ml proteinase K PCRgrade), and incubated for two hours at 60°C.Subsequently, proteinase K was inactivated at90°C for 10 minutes. Whole genome amplifica-tion was performed according to an improvedPEP protocol (I-PEP)11 with the followingmodifications: no gelatine, 500 µmol/l dNTP, 1mmol/l MgCl, and 100 µmol/l totally degener-ated 15 nucleotide long primer. Finally, multi-ple highly polymorphic microsatellites wereanalysed by time release PCR (0.1 mmol/ldNTP, 0.24 µmol/l primers, 1.25 U AmpliTaqGold® (PE Applied Biosystems, Branchburg,NJ, USA), 1.5-2.5 mmol/l MgCl, using 5 µlaliquots of the preamplified DNA in a finalvolume of 50 µl in a Techne Progene® thermo-cyler (Techne Incorporated, Princeton, NJ,USA) for 56 cycles, 60-64°C for four minutes,94°C for one minute), run on a 6% polyacryl-amide gel, and visualised by silver staining.

ResultsChromosome preparations from blood lym-phocyte cultures of the patient showed an

Table 1 Results of molecular genetic investigations

Primer Locus cM Alleles SMC Origin of alleles

D1S220 p31.1 82.50 bb,ab,ac NP Mat isoD1S216 p22.2 93.20 bb,bd,ac NP Mat isoD1S2746 p13.2 115.57 ac,ac,bb b Mat hetD1S250 p13.2 116.22 ab,ab,aa a NID1S2687 p13.1 117.79 bc,bc,aa a Mat hetD1S189 p13 119.39 cc,ac,bb b Mat isoD1S252 p12 122.73 cc,ac,bc b Mat isoD1S2669 p12 124.35 bb,ab,bb b Mat isocen, D1Z5 cen 128.0 NP NP NPD1S518 q25.2 194.78 aa,aa,bc NP MatD1S1656 q42.2 244.98 bc,-,ad NP Mat hetD1S180 q44 260.62 ab,ab,cd NP Mat het

The alleles are given in the order patient, mother, father. Allele designations (a-d) are arbitrary.Genetic mapping according to the sex averaged linkage map by the Genetic Location Database(LDB). Mat=maternal, iso=isodisomy, NP=not performed, NI=not informative.

Letters 885

J Med Genet2001;38:885–888

Institut fürMedizinische Genetik,Universität Zürich,Rämistrasse 74,CH-8001 Zürich,SwitzerlandB RöthlisbergerD KotzotD BalmerA Schinzel

Department ofMedical Genetics,Medical Academy ofPostgraduate Study,Kyiv, UkraineT E ZerovaT I Buzhievskaya

Correspondence to:Professor Schinzel,[email protected]

www.jmedgenet.com


http://jmg.bmj.com/


additional small marker chromosome (SMC)in 14 of 40 metaphases (35%) investigated (fig1A). Chromosome analysis of the mother andthe father showed normal karyotypes. In orderto determine the origin of the SMC, amicroFISH experiment with reverse paintingto normal metaphases was performed. It couldbe shown that the SMC was formed by a regionof chromosome 1, 1p21.1→1q12 (fig 1B).Application of the “all telomeres” probe(Vysis® Inc, Downers Grove, IL, USA)showed no telomeric signal on the extrachromosome indicating that it is a ringchromosome (data not shown).

Microsatellite analysis on genomic DNA ofthe patient and the parents showed that thepatient had inherited two maternal alleles (het-erodisomy and isodisomy respectively at diVer-ent loci) but no paternal allele at diVerent lociof chromosome 1 (table 1, fig 2A, B).Investigations of several microsatellites fromvarious other chromosomes were in agreementwith correct paternity. Based on these results,the karyotype of the patient may be describedas 47,XX,UPD(1)mat,+der(1)(p21.1q12)/46,XX,UPD(1)mat.

Probably owing to the patient’s relatively lowlevel mosaicism of leucocytes with the SMC,the microsatellite analysis with pericentromericmarkers did not result in an unambiguousdetermination of the parental origin of theSMC. Although some markers showed a weakpaternal band, this was not obvious enough for

unequivocal parental determination. There-fore, a new approach combining microdissec-tion of the SMC with subsequent PEP-PCRand conventional microsatellite PCR wasperformed. All investigated markers mapping

Figure 1 (A) GTG banded karyotype of the patient. The arrow points to thesupernumerary marker chromosome. (B) Reverse painting to a normal metaphase usingthe probe generated from DNA obtained by microdissection of the supernumerary markerchromosome. Note the fluorescent signals in the pericentric region of both chromosomes 1.

Figure 2 (A) Results of parental origin determination ofthe normal homologues with microsatellite markers. ThePCR products are represented in the following order: father,patient, mother. Allele designations (a to d) are arbitrary.(B) Results of parental origin determination of thesupernumerary marker chromosome 1 (SMC 1) and thenormal homologues with microsatellite markers. The PCRproducts are represented in the following order: father,patient, SMC, mother. Allele designations (a to d) arearbitrary.

D1S2746

D1S180

a

B

b

c

D1S2687

a

A

b

c

d

a

b

c

886 Letters

www.jmedgenet.com


http://jmg.bmj.com/


to the duplicated region clearly showed onepaternal band and no maternal band and werethus in agreement with paternal origin of theSMC (fig 2B). In addition, by investigating thegenomic DNA with these pericentromericmarkers, three markers mapping within theduplicated region on the short arm of chromo-some 1 within a 10 cM distance from the cen-tromere and also two markers at the distal endof the short arm of chromosome 1 showedmaternal isodisomy (table 1).

DiscussionThe great variability of clinical symptoms inpatients with a SMC of the same chromosomalorigin is probably the result of variation of thegenetic content, the degree of mosaicism(especially, as in our case, in ring chromo-somes), and uniparental disomy of the normalhomologues from which the SMC is de-rived.1 12

To allow any karyotype-phenotype correla-tion, comparison of a suYcient number ofaccurately investigated cases is necessary. Thebest way to characterise a SMC is a multistepapproach: (1) classical cytogenetic procedures,which provide information about the degree ofmosaicism by investigating many cells, if possi-ble, from diVerent tissues; (2) microFISH,which gives detailed information about thechromosomal content of the SMC; (3) micro-satellite analysis of the normal homologues,which gives information about uniparental dis-omy and a clue towards the mechanism of for-mation; and (4) the absence of telomeres indi-cated that the SMC is a ring.

In most cases with an SMC originating fromchromosome 1, neither FISH analysis withlocus specific probes nor molecular investiga-tions with microsatellites was performed.12 Thepatients described show great variability oftheir phenotypes and so far no distinctsyndrome has been associated with the groupof cases with a supernumerary marker chromo-some originating from chromosome 1.

The phenotype of four patients with mater-nal UPD(1) reported so far13–16 is normal andtherefore suggests that maternally imprintedloci on chromosome 1, if existing at all,probably do not influence the phenotype.Thus, the congenital anomalies in our patientare probably not the result of the maternalUPD of chromosome 1, but are more likelyrelated to the mosaic duplication of chromo-some 1 (p21.1→q12).

Several authors assumed that subjects withSMCs might have an increased risk for UPD ofthe structurally normal homologues fromwhich the SMCs derived. However, the inci-dence of UPD associated with a SMC is notknown, and so far only one case each of chro-mosomes 6, 7, 20, and X,3 5–7 as well as ninecases of chromosome 15, seven with Prader-Willi syndrome, and two with Angelmansyndrome, have been reported.4 17–22

Coexistence of UPD with a SMC is mainlyexplained by the following two mechanisms.21

First, duplication of the normal homologue ina zygote which has inherited a SMC in place of

the normal corresponding chromosome “res-cues” an aneuploidy. In this case, UPD arisesby mitotic non-disjunction and therefore com-plete isodisomy should always be observed.Second, the zygote may have originated as atrisomy with the single parental chromosomebeing lost through a breakage event or, alterna-tively, a disomic gamete was fertilised by agamete with a SMC formed during meiosis.

Heterodisomy, as reported in our case, canonly be explained by the second mechanism.The distribution of isodisomic and heterodis-omic segments in our case, pericentromericand more distal, respectively, indicates an errorin maternal meiosis II resulting in the maternalUPD(1). Whether the SMC was formedduring paternal meiosis or in an early somaticcell division cannot be diVerentiated. If it wasformed during paternal meiosis, the coexist-ence of a SMC with UPD may be considered asa coincidence. Even if it was formed postzy-gotically, our case might at best not contradictthe hypothesis that subjects being uniparentallydisomic might have an increased risk of bearinga SMC of the same homologue (rather than thereverse). In other words, the presence of twomaternal (or paternal) chromosomes in thezygote might constitute a risk for the formationof a paternal (or maternal) SMC through abreakage event.

Parental origin of the normal chromosomesand of the SMC is sometimes diYcult to deter-mine. In many instances, only quantitativeresults can be obtained (mitotically formedmosaicism, supernumerary marker chromo-somes). The new technical approach applied inthe case reported here allows the unequivocaltracing back of informative alleles from adissected abnormal chromosome to one of theparental homologues.

In conclusion, we present a case withcoexistence of a SMC 1 with maternal UPD(1)and hence mosaicism for complete maternaluniparental disomy and partial trisomy for asmall pericentromeric segment of chromosome

+ To the best of our knowledge, 10 cases ofsupernumerary marker chromosome 1(SMC 1) mosaicism have been reportedso far. Most of these ring shaped markerswere identified by fluorescence in situhybridisation (FISH) with centromerespecific probes.

+ We report another case with SMC 1mosaicism identified by combining chro-mosome microdissection with reversepainting to normal metaphases (micro-FISH).

¶ Microsatellite analysis showed heterodis-omic maternal uniparental disomy(UPD) of the normal chromosomes 1. Inaddition, a new approach combiningchromosome microdissection, primer-extension-preamplification polymerasechain reaction (PEP-PCR) and standardPCR of highly polymorphic microsatel-lites showed paternal alleles only withinthe region of the SMC. This indicatesthat the SMC is of paternal origin.

Letters 887

www.jmedgenet.com


http://jmg.bmj.com/


1 combined with uniparental disomy for therest of the chromosome. By a new technicalapproach, which allows the parental origin ofthe marker chromosome to be determined,even in cases with low level mosaicism, we wereable to unequivocally show the paternal originof the marker.

The authors are grateful to the family for their cooperation. Thestudy was supported by the Swiss National Foundation, grants32-45604.95, 32-56053.98, and 71P51778.

1 Crolla JA. FISH and molecular studies of autosomal super-numerary marker chromosomes excluding those derivedfrom chromosome 15. II. Review of the literature. Am JMed Genet 1998;75:367-81.

2 Gardner RJM, Sutherland GR. Chromosome abnormalitiesand genetic counselling. Oxford: Oxford University Press,1996:263-4.

3 James RS, Temple IK, Dennis NR, Crolla JA. A search foruniparental disomy in carriers of supernumerary markerchromosomes. Eur J Hum Genet 1995;3:21-6.

4 Bettio D, Rizzi N, Giardino D, Gurrieri F, Silvestri G,Grugni G, Larizza L. FISH characterization of smallsupernumerary marker chromosomes in two Prader-Willipatients. Am J Med Genet 1999;68:99-104.

5 Chudoba I, Franke Y, Senger G, Sauerbrei G, Demuth S,Beensen V, Neumann A, Hansmann I, Claussen U. Mater-nal UPD 20 in a hyperactive child with severe growthretardation. Eur J Hum Genet 1999;7:533-40.

6 Miyoshi O, Kondoh T, Tanedo H, Otsuko K, Matsumoto T,Niikawa N. 47,XX,UPD(7)mat,+r(7)pat/46,XX,UPD(7)mat mosaicism in a girl with Silver-Russell syndrome(SRS): possible exclusion of the putative SRS gene from a7p13-q11 region. J Med Genet 1999;36:326-9.

7 Yorifuji T, Muroi J, Kawai M, Uematsu A, Sasaki H, MomoiT, Kaji M, Yamanaka C, Furusho K. Uniparental andfunctional X disomy in Turner syndrome patients withunexplained mental retardation and X derived markerchromosomes. J Med Genet 1998;35:539-44.

8 Senger G, Lüdecke HJ, Horsthemke B, Claussen U. Micro-dissection of banded human chromosomes. Hum Genet1990;84:507-11.

9 Lichter P, Ried T. Molecular analysis of chromosome aber-rations. In situ hybridization. Methods Mol Biol 1994;29:449-78.

10 Röthlisberger B, Schinzel A, Kotzot D. A new molecularapproach to investigate origin and formation of structuralchromosome aberrations. Chrom Res 2000;8:451-3.

11 Dietmaier W, Hartmann A, Wallinger S, Heinmöller E,Kerner T, Endl E, Jauch KW, Hofstädter F, RüschoV J.

Multiple mutation analyses in single tumor cells withimproved whole genome amplification. Am J Pathol1999;154:83-95.

12 Callen DF, Eyre H, Fang YY, Guan XY, Veleba A, MartinNJ, McGill J, Haan EA. Origins of accessory small ringmarker chromosomes derived from chromosomes 1. J MedGenet 1999;36:847-53.

13 Pulkkinen L, Bullrich F, Czarnecki P, Weiss L, Uitto J.Maternal uniparental disomy of chromosome 1 withreduction to homozygosity of the LAMB3 locus in a patientwith Herlitz junctional epidermolysis bullosa. Am J HumGenet 1997;61:611-19.

14 Leigh Field L, Tobias R, Robinson WP, Paisey R, Bain S.Maternal uniparental disomy of chromosome 1 with noapparent phenotypic eVects. Am J Hum Genet 1998;63:1216-20.

15 Dufourcq-Lagelouse R, Lambert N, Duval M, Viot G,Vilmer E, Fischer A, Prieur M, de Saint Basile G. ChediakHigashi syndrome associated with maternal uniparentalisodisomy of chromosome 1. Eur J Hum Genet 1999;7:633-7.

16 Lebo RV, Shapiro LR, Fenerci EY, Hoover JM, Chuang JL,Chuang DT, Kronn DF. Rare etiology of autosomal reces-sive disease in a child with noncarrier parents. Am J HumGenet 2000;67:750-4.

17 Cheng SD, Spinner, NB, Zackai EH, Knoll JH. Cytogeneticand molecular characterization of inverted duplicatedchromosomes 15 from 11 patients. Am J Hum Genet 1994;55:753-9.

18 Christian SL, Mills P, Das S, Ledbetter DH. High risk ofuniparental disomy 15 associated with amniotic fluid con-taining de novo small supernumerary marker 15 chromo-somes. Am J Hum Genet Suppl 1998;63:A11.

19 Ebrahim SAD, Feldman B, Knaus A, Gyi K, Mills PL,Johnson MP, Evans MI. Prenatal diagnosis of maternaluniparental disomy of chromosome 15 in association withde novo supernumerary marker chromosome 15. Am JHum Genet Suppl 1998;63:A162.

20 Woodage T, Deng ZM, Prasad M, Smart R, Lindeman R,Christian SL, Ledbetter DL, Robson L, Smith A, Trent RJ.A variety of genetic mechanisms are associated with thePrader-Willi syndrome. Am J Med Genet 1994;54:219-26.

21 Robinson WP, WagstaV J, Bernasconi F, Baccichetti C, Arti-foni L, Franzoni E, Suslak L, Shih LY, Aviv H, SchinzelAA. Uniparental disomy explains the occurrence of theAngelman or Prader-Willi syndrome in patients with anadditional small inv dup(15) chromosome. J Med Genet1993;30:756-60.

22 Lebbar A, Dupont JM, Cuisset L, Pinton F, Vasseur C, LeTessier D, Denavit MF, Ponsot G, Delpech M, RabineauD. Clinical features of Prader-Willi syndrome in a girl withmethylation status of Angelman syndrome. Eur J HumGenet Suppl 1998;6:97A.

888 Letters

www.jmedgenet.com


http://jmg.bmj.com/


Sponastrime dysplasia: presentation in infancy

A C OYah, M Lees, R M Winter, C M Hall

EDITOR—The case of a white female with spo-nastrime dysplasia is presented.

Case reportThe patient is the first and only child ofhealthy, non-consanguineous parents. Duringroutine ultrasound scanning, the fetus wasnoted to have extremely short limbs and a ten-tative diagnosis of achrondroplasia was made.The pregnancy was otherwise uncomplicated.The baby was delivered at 40 weeks’ gestationby emergency caesarian section because ofbreech presentation. Birth weight was 2460 gand length 44 cm (both below the 3rd centile).During infancy and early childhood, her heightremained well below the 0.4th centile.

Examination showed a small baby with mid-face hypoplasia and a small, slightly upturnednose. With increasing age, the prominent fore-head, saddle shaped nose, and midface hypo-plasia became more obvious. There wasrhizomelic and mesomelic shortening of herupper and lower limbs (fig 1). She was noted tohave short, broad hands and feet with deeppalmar creases, short toes, and dimples in theelbows and knees. In addition, she had markedgeneralised joint laxity except at the elbowswhere extension was limited. A skeletal surveyat 5 months was not diagnostic, but achondro-plasia was excluded.

Chromosomal analysis was normal (46,XX).At 4 months of age, she was noted to have an

eczematous skin rash, and there followedseveral hospital admissions for recurrent chestinfections; investigations showed hypogamma-globulinaemia. Both the skin rash and thehypogammaglobulinaemia gradually normal-ised, eVectively excluding a diagnosis of a shortlimbed dwarfism syndrome in association withimmunodeficiency.

Despite a dysplastic (but not dislocated) lefthip, crawling and walking were not delayed;however at 2 years of age she developed a wad-dling gait, and at 2 years 5 months a dislocatedleft hip was diagnosed. At 3 years, an openvarus reduction and derotation osteotomy wasperformed. She made good recovery from heroperation, and at 4 years 3 months the internalfixator was removed.

Developmental milestones were reached atappropriate ages, and mental development wasnormal.

RADIOGRAPHIC FINDINGS

The radiographic findings are illustrated in figs2-7.

Spine (figs 2 and 3)These x rays show the typical changes in theshape of the vertebral bodies as described byLanger et al in 1996 and again in 19972 3:

platyspondyly improving with the patient’s age;a distinct junction (apparent in early/mid-childhood) between the anterior and posteriorparts of the vertebral bodies (this is as a resultof the anterior portions having convex endplates compared to the straight end plates ofthe posterior portions); a central anterior beak-ing of the vertebral bodies; and increasing con-cavity of the posterior surfaces of the vertebralbodies (posterior scalloping).There is a pro-gressive kyphoscoliosis and mild osteopenia.Additionally, there is loss of the normalincrease in interpedicular distance from L1 toL5 which can be appreciated in the radiographtaken at birth (fig 2A).

Long bones (figs 4-7)Characteristic changes in the proximal femoraconsist of a pronounced bony projection of thelesser trochanter, short femoral necks, and lossof the normal metaphyseal flare. These give a“spanner-like” appearance to the proximal

Figure 1 Clinical photograph of patient aged 4 years 8months. Note the significant (rhizomelic) shortening,prominent forehead, and depressed nasal bridge.

Letters 889

J Med Genet2001;38:889–893

Department ofRadiology, GreatOrmond StreetChildren’s Hospital,London WC1N 3JH,UKA C OYahC M Hall

Department ofMedical Genetics,Great Ormond StreetChildren’s Hospital,London WC1N 3JH,UKM LeesR M Winter

Correspondence to: Dr Hall,[email protected]

www.jmedgenet.com


http://jmg.bmj.com/


femora (fig 4A), which becomes less markedwith time (fig 4B). There was progressivedevelopment of coxa vara deformity, whichalthough complicated by dislocation on theleft, could be appreciated bilaterally.

Vertical metaphyseal striations are best seenaround the knee and in the distal radius. Thesestriations developed with age, and were notdemonstrable radiologically until 4 years 8months (figs 5 and 6).

There is retardation of bone age (fig 5) whencompared to the standards of Greulich andPyle (2 years 6 months at a chronological age of4 years 8 months, SD 11.65 months). A pseu-doepiphysis of the first metacarpal is seen, afinding which has also been seen in severalother patients.3 4 All epiphyses are small for ageand slightly irregular.

Other features include a predominantlyrhizomelic shortening of the limbs and flaringof the distal humeral metaphyses, giving them arather bulbous appearance which becomesmore pronounced with time. In addition, thereis a curious appearance of the proximalhumeral diaphyses (fig 7), consisting of a linearradiolucency aVecting the medial cortex andrunning obliquely. Beneath this, there is corti-cal thickening (buttressing) and mild angula-tion of the humeral shaft. This appearance,reminiscent of focal fibrocartilagenous dyspla-sia, was bilaterally symmetrical.

DiscussionSponastrime dysplasia is a rare but distinctentity which can be categorised as a spondy-loepimetaphyseal dysplasia. The acronym wasderived by Fanconi et al6 from the spondylarand nasal alterations which occur in addition tothe striations of the metaphyses. It has beendocumented in several sets of sibs,2 3 5 6 but inthe 13 cases reported to date (including ours)the parents have been non-consanguineous. Itwould appear that sponastrime dysplasia isinherited as an autosomal recessive disorderalthough germline mosaicism is a possibility.

Other than ours, 12 cases of true sponas-trime dysplasia have been reported,2–6 of whichonly one, described by Langer et al,2 was aninfant. Comparing the findings of Langer et al2

to those in our patient, there would appear tobe specific findings which may allow thediagnosis of sponastrime dysplasia to be madeat birth and in infancy. Clinically these arenon-specific and include midfacial hypoplasia,a saddle shaped nose, short limbs, and shortstature. Radiological features, however, aremore specific; the proximal femora have acharacteristic radiological “spanner-like” ap-pearance with a bony projection of the lessertrochanter, short curved femoral necks, andloss of the normal metaphyseal flare. Thisappearance of the proximal femora becomesless apparent with age. In the spine there is asignificant platyspondyly with loss of thenormal progressive widening of the inter-pedicular distances from L1 to L5. Whilepresent in all eight of the previous cases inwhich the result of spinal radiography wasavailable, this is the first time that the gradualreduction in interpedicular distance has been

Figure 2 (A) AP spine at birth, showing platyspondyly and loss of the normal increase ininterpedicular distance from L1 to L5. There is no significant curvature of the spine. (B)AP spine at 4 years 8 months. A kyphoscoliosis has developed. Note the relative increase inheight of the vertebral bodies (improved platyspondyly) and the mild osteopenia.

Figure 3 Lateral spine at 2 years 6 months. The platyspondyly is less marked than ininfancy. There is a distinct junction between the anterior and posterior portions of thevertebral bodies and posterior scalloping.

890 Letters

www.jmedgenet.com


http://jmg.bmj.com/


confirmed in the neonate and infant. As inother reported cases,4–6 patients may develop ascoliosis, which, as in our patient, may besignificant.

Lachman et al5 obtained biopsies from theiliac crests of two patients with sponastrimedysplasia. Light and electron microscopicexamination findings suggested a specific mor-phological appearance for sponastrime dyspla-sia. Unfortunately, despite open surgery, a his-tological sample was not obtained in ourpatient to confirm this appearance.

Short stature is a universal finding. Theseverity of short stature seen in our patient (fig1) was made worse by the development of aprogressive (and significant) kyphoscoliosis (fig2B).

Our patient developed bilateral coxa varadeformity, which has been documented in twoother patients.3 There have been two publishedcases of patients who both developed thoraco-lumbar scoliosis warranting surgery5; ourpatient has also developed a significant scolio-sis.

Including our patient, mild osteopenia hasbeen described in all eight of the patients inwhom this information is available.

This is the first case of sponastrime dysplasiain which transient hypogammaglobulinaemiahas been reported.

In our patient, there was a predominantlyrhizomelic limb shortening and prominence ofthe distal humeral metaphyses which had arather bulbous appearance. These changeshave not previously been described in sponas-trime dysplasia.

The appearance of the proximal humeri wasreminiscent of focal fibrocartilaginous dyspla-sia, a known cause of tibia vara. On plain film itappears as a tongue-like cortical defect. ItaVects the medial cortex and leads to varusdeformity centred at the lesion. On MRI, thereis no associated soft tissue mass.13 Although ithas previously been reported in the upperlimb,14 to our knowledge it has never been seenbilaterally. Histologically, there is abnormalgrowth and remodelling of fibrocartilaginoustissue at the growth plate interface between thetendon and its bony attachment.15 16 Compar-ing this with the histological findings of Lach-man et al4 in sponastrime dysplasia, we wonderif a similar (abnormal) process is occurring inthe two conditions. Interestingly, neither thehumeral shaft changes nor the metaphysealstriations around the wrist and knees wereradiologically obvious at birth or in infancy, butbecame apparent in mid-childhood (4 years 8months). The natural history of focal fibrocar-tilaginous dysplasia is that of spontaneousresolution17; as we have neither histology norradiographs beyond 4 years 8 months in ourpatient, we feel that radiographic follow up maybe useful.

Fig 8 shows the patient’s metacarpophalan-geal pattern profile (MCPP) performed at 4years 8 months according to the method ofGarn et al.18 All 19 bones had negative Z values,confirming shortening. Z scores ranged from−1.2 (distal phalanx 3) to −3.6 (metacarpal 3).The normal Z score range for the metacarpalsis +2 to −2. All metacarpals in this patient hada value of less than −2, again confirming

Figure 4 (A) The hips at 13 months. (B) The hips at 3 years 1 month. The spanner-likeappearance of the proximal femur seen in the infant period (and early childhood) willbecome less obvious with age. The left hip is subluxed. The patient subsequently developedbilateral coxa vara.

Figure 5 Left wrist and hand at 4 years 8 months. Boneage is delayed (2 years 6 months according to the standardsof Greulich and Pyle). The first metacarpal has apseudoepiphysis. Note the metaphyseal striations of thedistal radius.

Letters 891

www.jmedgenet.com


http://jmg.bmj.com/


brachymetacarpy as a feature of sponastrimedysplasia. Fig 8 illustrates an up and downvariation in the hand pattern, particularlyaVecting the phalanges. In our patient, this isnot as pronounced as in the cases of Cooper etal4 and Fanconi et al.6 The MCPP obtained byCamera et al7 shows a much flatter pattern, fur-ther supporting the likelihood of a diVerentcondition in their patient. The pattern variabil-ity index (OZ) in this patient, based on themethod described by Garn et al,19 was calcu-lated to be 0.51. A score greater than 0.7 is saidto indicate hand dysmorphogenesis; Cooper etal4 calculated a value of 0.73 for their patient.Their patient was 6 years 7 months old andours 4 years 8 months; age is therefore anunlikely explanation for the diVerences andMCPP analyses are required in more patients

with sponastrime dysplasia, in order to evaluateits use in aiding the diagnosis.

Intelligence in patients with true sponas-trime dysplasia is normal. Several patientsreported as cases of sponastrime dysplasia7–9

are likely to be cases of spondyloepimetaphy-seal dysplasia (SEMD) with large joint disloca-tions (first characterised by Hall et al10). Thetwo sisters described by Camera et al11 and thepatient described by Verloes et al12 do not seemto have either sponastrime dysplasia or SEMDwith large joint dislocations. Table 1 summa-rises and compares the findings in these condi-tions.

As previously reported,2 3 the metaphysealstriations are not a prominent feature in theearly stages, and in our patient were not radio-logically apparent until the child was almost 5

Figure 6 Left knee at (A) 2 years 6 months and (B) 4 years 8 months. Metaphysealstriations around the knee become more obvious with age.

Figure 7 Left upper limbat 4 years 8 months. Thereis rhizomelic shortening.The humeral metaphysesare bulbous and there is alinear radiolucency of themedial cortex of theproximal humeral diaphysisrunning obliquely withcortical thickening(buttressing) below this.This appearance becameapparent in mid-childhood.

Table 1 DiVerential diagnosis of sponastrime dysplasia

Feature SponastrimeSEMD with largejoint dislocations10 Camera et al11 Verloes et al12

Intelligence Normal Normal Severe retardation Severe retardationShort stature + + + +Midface hypoplasia + + + +Saddle nose + + + +Rocker bottom feet − − − +Joint laxity + ++ ? +Hypotonia + + + +++Large joint dislocations − ++ − −Osteopenia + − + +++Microcephaly − − + +Wormian bones − − − +Striated metaphyses* +/++ + + −Irregular metaphyseal margins* + − − −Epiphyseal involvement + ++ + −/+Delayed bone age + + + +Scoliosis +/+++ + − +Characterisic age related vertebral changes* ++ − − −Lumbar lordosis + + + −Reduction in interpedicular distances from L1 to L5 + + − −Gracile metacarpals − + ? −Brachymetacarpy + − ? +Pseudoepiphyses of metacarpals + − − +Long slender phalanges − + ? −MCPP Up - down variation ? ? Relatively flat

*Major radiological diagnostic features of sponastrime dysplasia.+ mild; ++ moderate; +++ severe; − absent; ? not mentioned in paper.

Figure 8 Metacarpophalangeal pattern profile (4 years 8months). An up and down pattern variation (especially ofthe phalanges) and brachymetacarpy are illustrated.

0

–1.0

–0.5

–1.5

–2.5

–2.0

–3.0

–3.5

–4.5

–4.0

Metacarpal Proximal Middle Distal

Z s

core

1 2 3 4 5 1 2 3 4 5 2 3 4 5 1 2 3 4 5

892 Letters

www.jmedgenet.com


http://jmg.bmj.com/


years of age, so sponastrime dysplasia was notconsidered as a possible diagnosis; indeed thispatient has previously been presented bySlaney et al1 as a new syndrome of spondy-loepimetaphyseal dysplasia, eczema, and hy-pogammaglobulinaemia. We therefore tend toagree with Langer et al2 3 who feel that lessemphasis should be placed on these striationsand more on the findings in the spine. Theysuggest that the condition be called “spondy-lometaphyseal dysplasia with midface hypopla-sia and depressed nasal bridge”. However,because of the mild epiphyseal abnormalities,and because it is now present in textbooks anddatabases as sponastrime dysplasia, we feel that“spondyloepimetaphyseal dysplasia (SEMD),sponastrime type” is a more appropriate term.

1 Slaney SF, Hall CM, Atherton DJ, Winter RM. A new syn-drome of spondyloepimetaphyseal dysplasia, eczema andhypogammaglobulinaemia. Clin Dysmorphol 1999;8:79-85.

2 Langer LO Jr, Beals RK, LaFranchi S, Scott CI Jr,Sockalosky JJ. Sponastrime dysplasia: five new cases andreview of nine previously published cases. Am J Med Genet1996;63:20-7.

3 Langer LO Jr, Beals RK, Scott CI Jr. Sponastrime dysplasia:diagnostic criteria based on five new and six previouslypublished cases. Pediatr Radiol 1997;27:409-14.

4 Cooper HA, Crowe J, Butler MG. Sponastrime dysplasia:report of an 11-year-old boy and review of the literature.Am J Med Genet 2000;92:33-9.

5 Lachman RS, Stoss H, Spranger J. Sponastrime dysplasia: aradiologic-pathologic correlation. Pediatr Radiol 1989;19:417-24.

6 Fanconi S, Issler C, Giedon A, Prader A. The SPONAS-TRIME dysplasia: familial short-limb dwarfism with

saddle nose, spinal alterations, and metaphyseal striation.Helv Paediatr Acta 1983;38:267-80.

7 Camera G, Camera A, Pozzolo S, Costa P. Sponastrimedysplasia: report on a male patient. Pediatr Radiol 1994;24:322-4.

8 Masuno M, Nishimura G, Adachi M, Hotsubo T,Tachibana K, Makita Y, Imaizumi K, Kuroki Y. Sponas-trime dysplasia: report on a female patient with severe skel-etal changes. Am J Med Genet 1996;66:429-32.

9 Nishumura G, Mikawa M, Fukushima Y. Another observa-tion of Langer-type sponastrime dysplasia variant. Am JMed Genet 1998;80:288-90.

10 Hall CM, Elçioglu NH, Shaw DG. A distinct form ofspondyloepimetaphyseal dysplasia with multiple disloca-tions. J Med Genet 1998;35:566-72.

11 Camera G, Camera A, Di Rocco M, Gatti R. Sponastrimedysplasia: report on two siblings with mental retardation.Pediatr Radiol 1993;23:611-14.

12 Verloes A, Mission JP, Dubru JM, Jamblin P, Le Merrer M.Heterogeneity of SPONASTRIME dysplasia: delineationof a variant form with severe mental retardation. Clin Dys-morphol 1995;4:208-15.

13 Meyer JS, Davidson RS, Hubbard AM, Conrad KA. MRI offocal fibrocartilaginous dysplasia causing tibia vara. J Pedi-atr Orthop 1995;15:304-6.

14 Lincoln TL, Birch JG. Focal fibrocartilaginous dysplasia inthe upper extremity. J Paediatr Orthop 1997;17:528-32.

15 Bell SN, Campbell PK, Cole WG, Menelaus MB. Tibia varacaused by focal fibrocartilaginous dysplasia. Three casereports. J Bone Joint Surg (Br) 1985;67:780-4.

16 Niyibizi C, Vinsconti CS, Gibson G, Kavalkovich K. Identi-fication and immunolocalization of type X collagen at theligament-bone interface. Biochem Biophys Res Commun1996;222:584-9.

17 Bradish CF, Davies SJ, Malone M. Tibia vara due to focalfibrocartilaginous dysplasia. The natural history. J BoneJoint Surg (Br) 1988;70:106-8.

18 Garn SM, Hertzog KP, Poznanski AK, Nagy JM. Metacar-pophalangeal length in the evaluation of skeletal malforma-tion Radiology 1972;105:375-81.

19 Garn SM, Leonard WR, Poznanski AK. Applications of thepattern variability index (OZ) to the quantification of dys-morphogenesis in the hand. Am J Med Genet 1987;27:143-52.

Letters 893

www.jmedgenet.com


http://jmg.bmj.com/


doi: 10.1136/jmg.38.12.852 2001 38: 852-861J Med Genet

Magdalena Duisterhof, Rutger W Trijsburg, Martinus F Niermeijer, et al. making up the balancePsychological studies in Huntington's disease:

http://jmg.bmj.com/content/38/12/852.full.htmlUpdated information and services can be found at:

These include:

References

http://jmg.bmj.com/content/38/12/852.full.html#related-urlsArticle cited in:

http://jmg.bmj.com/content/38/12/852.full.html#ref-list-1This article cites 62 articles, 15 of which can be accessed free at:

serviceEmail alerting

box at the top right corner of the online article.Receive free email alerts when new articles cite this article. Sign up in the

Notes

http://group.bmj.com/group/rights-licensing/permissionsTo request permissions go to:

http://journals.bmj.com/cgi/reprintformTo order reprints go to:

http://group.bmj.com/subscribe/To subscribe to BMJ go to:


http://jmg.bmj.com/content/38/12/852.full.html

http://jmg.bmj.com/content/38/12/852.full.html#ref-list-1

http://jmg.bmj.com/content/38/12/852.full.html#related-urls

http://group.bmj.com/group/rights-licensing/permissions

http://journals.bmj.com/cgi/reprintform

http://group.bmj.com/subscribe/

http://jmg.bmj.com/


Documents

Psychological studies in Huntington's disease: making up the balance