JMed Genet 1995;32:135-155 Abstracts of the British Medical Genetics Conference held in York on 12 to 14 September 1994 (sponsored by the Clinical Genetics Society, Clinical Molecular Genetics Society, Genetical Society, and Genetic Nurses and Social Workers Association) Mutations of the RET proto- oncogene in the multiple endocrine neoplasia type 2 syndromes and Hirschsprung disease C ENG, L M MULLIGAN, S LYONNET*, P EDERY*, D P SMITH, J B J KWOK, E GARDNER, C S HEALEY, M A PONDER, A MUNNICH*, B A J PONDER CRC Human Cancer Genetics Research Group, University of Cambridge, Cambridge; *Hospital Necker Enfants Malades, Paris, France. The autosomal dominantly inherited multiple endocrine neoplasia (MEN) type 2 syn- dromes and Hirschsprung disease (HSCR) are disorders involving neural crest and its derivatives. MEN 2A is characterised by me- dullary thyroid carcinoma (MTC), phaeo- chromocytoma (50%), and parathyroid hyperplasia (15-20%); FMTC by MTC only; and MEN 2B by MTC, phaeochromocytoma and developmental abnormalities. In MEN 2A and FMTC, mutations in the RET proto- oncogene affect one of five cysteine residues in the cysteine rich region of the extracellular domain. There is an association between mu- tation at cysteine 634 and the presence of phaeochromocytoma. There is a correlation between a cysteine to arginine change at codon 634 and parathyroid involvement. In contrast, in MEN 2B, a single missense mu- tation within the catalytic core of the tyrosine kinase domain has occurred in over 90% of cases. The methionine to threonine change is postulated to alter the specificity of the catalytic core for its substrate. In addition, approximately 25% of sporadic MTC and less than 10% of sporadic phaeochromocytoma carry somatic MEN 2B type mutations. HSCR is caused by the absence of enteric ganglia, resulting in intestinal obstruction. In some families with HSCR, heterozygous point mutations are scattered throughout RET, which lead to amino acid substitution or stop codons. In these cases, an inactivation of RET is postulated to be responsible for the aetiology of the disease. This contrasts with the MEN 2 syndromes. Various coliagen mutations produce widely varying clinical phenotype F M POPE, A C NICHOLLS, P NARCISI, A J RICHARDS Dermatology Research Group, MRC Clinical Research Centre, Harrow. At least 19 collagen types with in excess of 25 genes have now been identified. These form two broad groups: the interstitial col- lagens with a highly conserved GlyXY central helix and the others in which there are glob- ular interruptions with resultant kinking. The clinical diseases range from bone and arterial fragility (as in COLlAl, 1A2, and 3A1 mut- ations) to more subtle disturbances of epi- dermal adhesion as in COL7A1 mutants. More abundant genes such as COL3A1 can be analysed by combination of protein and molecular mapping techniques. Less abund- ant ones are best analysed by molecular map- ping. Phenotypic severity of COLlAl, 1A2, and 3A1 mutants correlates well with mu- tational position. The role of the androgen receptor in male sexual differentiation M N PATTERSON, C L BEVAN, H R DAVIES, P A CLARKSON, B D BROWN, I A HUGHES Department of Paediatrics, University of Cambridge. Androgens are responsible for many aspects of male sexual differentiation. The effects of androgens are mediated by the androgen receptor, a typical steroid receptor. The re- ceptor binds to androgen inside the cell, and subsequently regulates the transcription of target genes by binding to promoter se- quences termed androgen response elements. Androgen receptor defects cause a range of defects of male sexual development, collectively termed androgen insensitivity syndrome (AIS). Clinically, androgen insensitivity is classified into two forms: in complete androgen insensitivity syndrome (CAIS), 46,XY subjects develop with a nor- mal female external phenotype, although they possess testes internally; the partial androgen insensitivity syndrome (PAIS) is associated with a wide range of undervirilisation. We have used single strand conformation poly- morphism analysis and, more recently, semi- automated DNA sequencing to identify mut- ations in the androgen receptor gene. We have found mutations in 21 patients, which has allowed us to provide information for genetic counselling, and are now studying the functional consequences of the mutations. Mitochondrial disease J POULTON Department of Paediatrics, University of Oxford, Jrohn Radcliffe Hospital, Headington, Oxford OX3 9DU. Heteroplasmy (the presence of more than one distinct population of mitochondrial DNA (mtDNA) within a single person) plays a central role in the unique transmission ge- netics and the phenotypic variability of mtDNA diseases. The level of mutant may vary between tissues, between cells, between regions of a single cell, and across time. The factors affecting the segregation of mtDNA during cell division and proliferation of sub- populations within a cell are poorly un- derstood. The phenotypes of Kearns-Sayre syndrome (KSS) and chronic progressive ex- ternal ophthalmoplegia (CPEO) are closely associated with deletions of mtDNA. We have recently shown that more than one type of closely related rearranged mtDNAs may be present in KSS, namely large duplications, deletions, and deletion dimers. We suggested that some or all of the deletions were derived from duplications. In contrast, deletion monomers are the only recombinant mito- chondrial DNA easily detectable in patients with CPEO. We suggest that duplication of mtDNA is characteristic of the early onset disease KSS, particularly those with diabetes mellitus. The balance of mtDNA re- arrangements appears to change as the phenotype evolves and may be central to the pathogenesis of this unique group of dis- orders. There are no parallels for these com- plex interactions between families of recombinant mitochondrial and nuclear gen- omes in closely related organisms, although similar events do occur in plants. Vertebrate developmental genetics: from one genome many species IAN J JACKSON MRC Human Genetics Unit, Western General Hospital, Crewe Road, Edinburgh EH4 2XU. Despite the superficial differences, all ver- tebrates share a common body plan and mode of development. The genes that instruct de- velopmental and adult processes are re- 135 The next British Medical Genetics Conference will be held at the University of York on 11 to 13 September 1995. Further details are available from Dr Peter Farndon, Clinical Genetics Society, Bir- mingham Maternity Hospital, Edgbaston, Bir- mingham B15 2TG (Tel/fax 021 627 2634.) on May 23, 2021 by guest. Protected by copyright. http://jmg.bmj.com/ J Med Genet: first published as 10.1136/jmg.32.2.135 on 1 February 1995. Downloaded from
Transcript
JMed Genet 1995;32:135-155
Abstracts of the
British Medical Genetics Conferenceheld in York on 12 to 14 September 1994 (sponsored by the ClinicalGenetics Society, Clinical Molecular Genetics Society, Genetical Society, andGenetic Nurses and Social Workers Association)
Mutations of the RET proto-oncogene in the multipleendocrine neoplasia type 2syndromes and Hirschsprungdisease
C ENG, L M MULLIGAN,S LYONNET*, P EDERY*, D P SMITH,J B J KWOK, E GARDNER,C S HEALEY, M A PONDER,A MUNNICH*, B A J PONDER
CRC Human Cancer Genetics ResearchGroup, University of Cambridge, Cambridge;*Hospital Necker Enfants Malades, Paris,France.
The autosomal dominantly inherited multipleendocrine neoplasia (MEN) type 2 syn-dromes and Hirschsprung disease (HSCR)are disorders involving neural crest and itsderivatives. MEN 2A is characterised by me-dullary thyroid carcinoma (MTC), phaeo-chromocytoma (50%), and parathyroidhyperplasia (15-20%); FMTC byMTC only;and MEN 2B by MTC, phaeochromocytomaand developmental abnormalities. In MEN2A and FMTC, mutations in the RET proto-oncogene affect one of five cysteine residuesin the cysteine rich region of the extracellulardomain. There is an association between mu-tation at cysteine 634 and the presence ofphaeochromocytoma. There is a correlationbetween a cysteine to arginine change atcodon 634 and parathyroid involvement. Incontrast, in MEN 2B, a single missense mu-tation within the catalytic core of the tyrosinekinase domain has occurred in over 90% ofcases. The methionine to threonine changeis postulated to alter the specificity of thecatalytic core for its substrate. In addition,approximately 25% ofsporadicMTC and lessthan 10% of sporadic phaeochromocytomacarry somatic MEN 2B type mutations.HSCR is caused by the absence of entericganglia, resulting in intestinal obstruction.In some families with HSCR, heterozygouspoint mutations are scattered throughoutRET, which lead to amino acid substitutionor stop codons. In these cases, an inactivationof RET is postulated to be responsible forthe aetiology of the disease. This contrastswith the MEN 2 syndromes.
Various coliagen mutationsproduce widely varying clinicalphenotype
F M POPE, A C NICHOLLS,P NARCISI, A J RICHARDS
Dermatology Research Group, MRC ClinicalResearch Centre, Harrow.
At least 19 collagen types with in excess of25 genes have now been identified. Theseform two broad groups: the interstitial col-lagens with a highly conserved GlyXY centralhelix and the others in which there are glob-ular interruptions with resultant kinking. Theclinical diseases range from bone and arterialfragility (as in COLlAl, 1A2, and 3A1 mut-ations) to more subtle disturbances of epi-dermal adhesion as in COL7A1 mutants.More abundant genes such as COL3A1 canbe analysed by combination of protein andmolecular mapping techniques. Less abund-ant ones are best analysed by molecular map-ping. Phenotypic severity of COLlAl, 1A2,and 3A1 mutants correlates well with mu-tational position.
The role of the androgenreceptor in male sexualdifferentiation
M N PATTERSON, C L BEVAN,H R DAVIES, P A CLARKSON,B D BROWN, I A HUGHES
Department of Paediatrics, University ofCambridge.
Androgens are responsible for many aspectsof male sexual differentiation. The effectsof androgens are mediated by the androgenreceptor, a typical steroid receptor. The re-ceptor binds to androgen inside the cell, andsubsequently regulates the transcription oftarget genes by binding to promoter se-quences termed androgen response elements.Androgen receptor defects cause a rangeof defects of male sexual development,collectively termed androgen insensitivitysyndrome (AIS). Clinically, androgeninsensitivity is classified into two forms: incomplete androgen insensitivity syndrome(CAIS), 46,XY subjects develop with a nor-mal female external phenotype, although theypossess testes internally; the partial androgeninsensitivity syndrome (PAIS) is associatedwith a wide range of undervirilisation. Wehave used single strand conformation poly-morphism analysis and, more recently, semi-automated DNA sequencing to identify mut-ations in the androgen receptor gene. Wehave found mutations in 21 patients, which
has allowed us to provide information forgenetic counselling, and are now studying thefunctional consequences of the mutations.
Mitochondrial disease
J POULTON
Department of Paediatrics, University ofOxford, Jrohn Radcliffe Hospital, Headington,Oxford OX3 9DU.
Heteroplasmy (the presence ofmore than onedistinct population of mitochondrial DNA(mtDNA) within a single person) plays acentral role in the unique transmission ge-netics and the phenotypic variability ofmtDNA diseases. The level of mutant mayvary between tissues, between cells, betweenregions of a single cell, and across time. Thefactors affecting the segregation of mtDNAduring cell division and proliferation of sub-populations within a cell are poorly un-derstood. The phenotypes of Kearns-Sayresyndrome (KSS) and chronic progressive ex-ternal ophthalmoplegia (CPEO) are closelyassociated with deletions ofmtDNA. We haverecently shown that more than one type ofclosely related rearranged mtDNAs may bepresent in KSS, namely large duplications,deletions, and deletion dimers. We suggestedthat some or all of the deletions were derivedfrom duplications. In contrast, deletionmonomers are the only recombinant mito-chondrial DNA easily detectable in patientswith CPEO. We suggest that duplication ofmtDNA is characteristic of the early onsetdisease KSS, particularly those with diabetesmellitus. The balance of mtDNA re-arrangements appears to change as thephenotype evolves and may be central to thepathogenesis of this unique group of dis-orders. There are no parallels for these com-plex interactions between families ofrecombinant mitochondrial and nuclear gen-omes in closely related organisms, althoughsimilar events do occur in plants.
Vertebrate developmentalgenetics: from one genome manyspecies
IAN J JACKSON
MRC Human Genetics Unit, Western GeneralHospital, Crewe Road, Edinburgh EH4 2XU.
Despite the superficial differences, all ver-tebrates share a common body plan and modeof development. The genes that instruct de-velopmental and adult processes are re-
135
The next British Medical Genetics Conference willbe held at the University of York on 11 to 13September 1995. Further details are available fromDr Peter Farndon, Clinical Genetics Society, Bir-mingham Maternity Hospital, Edgbaston, Bir-mingham B15 2TG (Tel/fax 021 627 2634.)
markably conserved. Human developmental,physiological, and disease processes are vari-ations on the consistent theme that runsthrough all vertebral development. Detailedgenetic mapping in humans and mice in-dicates that not only are their genes con-served, but their respective genomestructures, shown by conservation of synteny,are highly similar. Data from other speciesshow that each mammalian genome derivesfrom a common ancestor by a small numberof rearrangements. It is not yet known howsimilar to mammals are the genomes of lessrelated species such as zebra fish and pufferfish, but they will probably have significantimpact on the understanding of human ge-netics and development. If all of vertebrategenetics can be viewed as the study of a singlegenome controlling a single developmentalprogramme, then many problems in humangenetics are best solved by studying otherspecies. In many cases the mouse is the or-ganism of choice. Mutations identified inhuman patients have commonly been foundin mouse mutant strains, and vice versa. Fur-thermore, mice have proven to be particularlyinformative for the study of polygenic effects.With detailed genetic maps now availablefor mouse and human, and physical mapsbecoming available, there will be increasingcrossover between species in the search forgenes.
The Caenorhabditis elegansapproach to genome analysis
I A HOPE
Department of Pure & Applied Biology andDepartment of Genetics, The University ofLeeds.
The nematode worm, Caenorhabditis elegans,was specifically selected for investigations ofnervous system function and of mechanismsofanimal development because this organismis a simple metazoan which is ideally suitedfor genetic analysis. The developmental ana-tomy of this species is already well described.Details of the development and functioningof the organism are now being shown at themolecular level. To facilitate the cloning ofC elegans genes identified through mutation,John Sulston and Alan Coulson constructeda physical map of the 100 Mb genome. Thisfully catalogued library of overlapping gen-omic DNA clones has now become the baseof the project to sequence the entire C elegansgenome. The C elegans genome sequencingproject is at the forefront of a series of similarprojects for several animal species. The 4 Mbof sequence generated so far is already beingput to use. Discovery of homologous geneswithin the sequence data has led those work-ing with other species to begin using C elegansto investigate the function of their gene ofinterest. Developmental gene expression pat-tems are being examined for genes predictedpurely from the genome sequence data. Thesequence of entire genomes will radically alterthe nature of biological research.
Dissecting human genetic diseasethrough mouse models
KAREN P STEEL
MRC Institute of Hearing Research, UniversityPark, Nottingham NG7 2RD.
Mouse models can be invaluable for in-vestigating the causes of human genetic dis-eases. Three examples illustrate theusefulness of mouse models, all relating tohuman hereditary deafness. Firstly, mice arepowerful tools for positional cloning of genes.This is particularly valuable in cases suchas non-syndromic recessive deafness, whichoccurs in about 1 in 3000 births but is highlyheterogeneous, making direct positional clon-ing in humans very difficult. We have usedthe mouse to generate over 1000 meioses forstudy in deaf mice known to carry the samerecessive mutation, and are progressing to-wards identifying the mouse gene and itshuman homologue. Secondly, mice can beused to study the development of a defect,including prenatal stages. Deafness is as-sociated with pigmentation defects in a num-ber of human syndromes, and we have usedwhite spotted mouse mutants to show that alack ofmelanocytes in the inner ear causes thedeafness, and that the melanocyte precursorcells migrate from the neural crest duringembryogenesis but die before they reach thecochlea. Thirdly, specific mutations can becreated in mice by transgenic techniques tomimic known human diseases and investigatefurther the pathological mechanisms in-volved.
Cell cycle regulation and signaltransduction mechanisms inyeast: the relevance to humandisease
D A HUGHES
Chester Beatty Laboratories, Institute ofCancer Research, Fulham Road, LondonSW3 6JB.
There is increasing evidence that geneticchanges in components of cell cycle controland growth factor signalling pathways play amajor role in the causation of human cancerand the resulting 160 000 deaths per year inthe UK. The remarkable evolutionary andfunctional conservation of genes involved inthe cell cycle and signal transduction hasensured that research using the unicellularyeasts Saccharomyces cerevisiae (buddingyeast) and Schizosaccharomyces cerevisiae(fission yeast) plays a major role in our un-derstanding of these processes in eukaryotes.In the cell cycle field genetic analyses in yeastfirst indicated the crucial role of p34cdc2/cyclincomplexes in controlling cell cycle pro-gression. Complementation ofyeast cell cyclemutants has allowed the cloning of mam-malian homologues of cdc2 and cyclin. Yeastsand mammalian cells use a conserved sig-nalling pathway, the "MAP kinase" pathway,to transduce extracellular signals. We andothers have shown that it is possible to re-constitute mammalian signalling pathways inyeast, and we are currently using such modelsystems to try to identify new mammaliangene ftunctions.
Molecular analysis ofsegmentation in vertebratedevelopment
D G WILKINSON
Laboratory of Developmental Neurobiology,National Institute for Medical Research, TheRidgeway, Mill Hill, London NW7 1AA.
An important theme emerging from studiesof the molecular genetics of embryogenesis isthe remarkable evolutionary conservation ofdevelopmental mechanisms. This is reflectedin the conservation between different ver-tebrates of many aspects of morphologicaldevelopment. Moreover, certain de-velopmental mechanisms have even been con-served between invertebrate and vertebratespecies. As a consequence, results from avariety of model systems have advanced ourunderstanding of mammalian development.Experimental assets of different systems haveilluminated molecular mechanisms that un-derlie segmental patteming in the vertebratehindbrain and the branchial arches. Studiesin the mouse have shown critical functions ofHox genes in anterior-posterior specification,and of the zinc finger gene, Krox-20, in theformation of hindbrain segments. Functionalanalyses in Xenopus and zebra fish have im-plicated the receptor tyrosine kinase Sek incell-cell interactions that restrict gene ex-pression to specific hindbrain segments.These studies provide some initial glimpsesof the regulatory cascade of transcriptionalregulation and cell-cell signalling underlyingpattem formation in the branchial region ofthe head.
Mouse model systems and theiruse in the design of gene therapy
JULIA R DORIN
MRC Human Genetics Unit, Western GeneralHospital, Edinburgh EH4 2XU.
Genetic disease affects over 1% of all births.Random mutagenesis can reveal useful mousemodels ofhuman disease but once a defectivegene has been isolated, gene targeting in EScells provides the opportunity to create micewith the same genetic disruption. The valueof these models is in clarification of diseasepathogenesis, genotype-phenotype cor-relation, identification of other relevant fac-tors, and as the optimal test system for newtherapeutic intervention. Correction of thebasic defect by somatic gene therapy is anattractive approach to disease treatment. Res-toration of gene function can be easily mon-itored in a mutant animal and thus both safetyand efficacy can be determined. An animalmodel which displays a relevant clinicalphenotype also allows the question to beaddressed of what level of gene expression isenough to have a clinical effect? Our recentexperiments using a mouse model for cysticfibrosis (CF) have shown that aerosolisedDNA liposome mediated somatic gene ther-apy to the lung can partially correct the CFelectrophysiological defect and has promptedthe first UK clinical trial of this type.
Marfan syndrome: fibrillinexpression and microfibrillarabnormalities in familial cases
S J DAVIES, C M KIELTY*
Institute of Medical Genetics, UniversityHospital of Wales, Cardiff; *School ofBiological Sciences, University of Manchester,Manchester.
Marfan syndrome is a clinical diagnosis madein the presence of ocular, skeletal, and cardio-vascular abnormality and inherited as an
autosomal dominant disorder. Mutations inthe fibrillin gene (FBN1) on chromosome 15have been identified in both sporadic andfamilial cases of Marfan syndrome. The fib-rillin protein is a complex multi-domain struc-ture and the organisation of these monomersinto microfibrils or the effect of mutationson both the structure and function of thesemicrofibrils is as yet poorly understood. Wepresent the fibrillin and microfibrillar ab-normalities manifest in cell cultures and tis-sues from members of Welsh families withMarfan syndrome and different clinicalphenotypes.
Screening of Marfan syndromepatients for mutations in thefibrillin (FBN1) gene
C HAYWARD, L J LOGIE,M E M PORTEOUS, D J H BROCK
Human Genetics Unit, University ofEdinburgh, Western General Hospital,Edinburgh EH4 2XU.
Linkage analysis and mutation identificationhave confirmed fibrillin (FBN1), on chro-mosome 15q21.1, as the defective gene re-
sponsible for the Marfan syndromephenotype. FBN1 is a large highly fragmentedgene consisting of 65 exons in 110kb ofgenomic DNA. Characterisation of the prim-ary structure and intron/exon boundary se-
quences has enabled exon by exon screeningof the FBN1 gene. We have screened 60unrelated Marfan syndrome patients for mut-ations in FBN1. To date, we have analysed,by non-radioactive PCR and SSCP, 50 of the65 FBN1 exons and identified 30 differentband shift patterns. Of these, 27 are ap-
parently unique and probably due to diseasecausing mutations and three are present inseveral patients as well as in unaffected con-
trols and are polymorphisms. Nine of the 27mutations have been fully characterised andconsist ofseven missense mutations, one non-
sense mutation, and one splicing defect. Twoof the three polymorphisms have been char-acterised and they are both single basechanges producing a conserved amino acid.
Connexin 32 mutations in Xlinked dominant Charcot-Marie-Tooth disease
N HAITES, C BELL, S COCHRANE,N FAIRWEATHER, T MONACO,A JOHNSTON
Department ofMedical Genetics, University ofAberdeen, Foresterhill, Aberdeen, and ICRFLaboratories, Institute ofMolecular Medicine,John Radcliffe Hospital, Oxford.
X linked dominant Charcot-Marie-Tooth dis-ease (CMTX1) is one of a group ofperipheralneuropathies characterised clinically by aslowly progressive weakness and wasting ofthe distal muscles. Males affected withCMTX1 present with features similar to thedominant type 1 form of the disease, withevidence of de- and remyelination and re-duced nerve conduction velocities. Femalesoften present with milder clinical features.CMTX1 has been localised to the peri-centromeric region of the X chromosomeand has recently been shown, in a study of15 families, to be flanked by markers DXS 106and DXS559, a distance of approximately2-3 Mb. Contained within this interval is thegene connexin 32 (locus designated GJp1),which encodes a membrane bound, gap junc-tion protein. We have sequenced the codingregion of exon 2 from affected persons in 13families with the clinical features of CMTX1and have found mutations in 12. These find-ings are consistent with the disease CMTX1being the result of mutations affecting thegene connexin 32 (Cx 32).
The impact of mutations inrepeat 5 of the ligand bindingdomain of the LDL receptor onthe clinical phenotype of familialhypercholesterolaemia (FH)
V GUDNASON, I N M DAY,S HUMPHRIES
Department of Medicine, Division ofCardiovascular Genetics, UCLMS, The RayneInstitute, University Street, London WClE6_JJ.Cholesterol is cleared from the circulation byLDL receptor mediated endocytosis of apoBand apoE containing lipoproteins. The clin-ical characteristics of heterozygous FH are atwo to three fold rise ofserum cholesterol andpremature coronary artery disease. Mutationsin the LDL receptor gene cause FH which isfound at a frequency of 1 in 500 in mostpopulations. Broadly the mutations can bedivided into those causing null alleles andthose producing a defective receptor protein,and more than 150 different mutations havebeen identified in the LDLR gene worldwide.In the UK 8 to 10% of FH patients have amutation in repeat 5 of the ligand bindingdomain. Comparison of those patients withpatients with mutations causing either a nullallele or a protein with a defect affecting otherparts of the protein shows that those with amutation in repeat 5 have significantly(p<0-001) higher serum total and LDL cho-lesterol (9-37 mmol/l) than patients with mu-tations elsewhere producing defective pro-teins (7-78 mmol/l), but similar levels to thosewith a null allele (9-54 mmol/l). Since repeat
5 is the sole binding repeat responsible forbinding both the ligands ofthe LDL receptor,the most likely explanation is that both mis-sense mutations in repeat 5 and mutationscausing null alleles completely destroy re-ceptor function while less serious mutationselsewhere leave some residual receptor ac-tivity. This represents the first demonstrationof a genotype-phenotype effect in hetero-zygous FH in a heterogeneous population.
Description of two families in thenorth east of Scotland withglucocorticoid remediablealdosteronism
L J GATES, A MACCONNACHIE,N BENJAMIN, N HAITES*
Departments ofMedicine and Therapeutics,*Molecular and Cell Biology, University ofAberdeen, Polwarth Building, Foresterhill,Aberdeen AB9 2ZD.
Glucocorticoid remediable aldosteronism isan autosomal dominant cause of primary al-dosteronism. Cases have been described ofsevere hypertension presenting at a young ageand in association with familial occurrence ofhypertension and of cerebral haemorrhage.We describe two pedigrees in which 25 per-sons with glucocorticoid remediable al-dosteronism have been identified by thepresence of a new 6-3 kb fragment detectedby Southern blotting and due to the presenceof a chimaeric gene in which the regulatorysequences of 11 3-hydroxylase are fused to thecoding sequences of aldosterone synthase.This rearranged gene results in the abnormalsynthesis of aldosterone in response to ad-renocorticotrophic hormone leading to thedevelopment of hypertension. Gluco-corticoids inhibit adrenocorticotrophic hor-mone secretion and therefore will suppressaldosterone production and normalise bothblood pressure and the metabolic ab-normalities of primary aldosteronism. Inthose persons identified, there is a spectrum ofdisease ranging from few with normotension,the majority indistinguishable from those withessential hypertension, and few with severehypertension. Those persons with severe hy-pertension have attendant early morbidity andmortality, primarily owing to strokes. Thisform ofhypertension is noteworthy because itmay be successfully managed by appropriatemedical therapy; expected benefit from iden-tification and treatment is therefore high.
Genotype-phenotype correlationsin von Hippel-Lindau (VHL)disease
P A CROSSEY, F M RICHARDS,J S GREEN*, H P H NEUMANNt,K FOSTER, A PROWSE, N A AFFARA,M A FERGUSON-SMITH, E R MAHER
Cambridge University Department ofPathology, Cambridge; *Division ofCommunity Medicine, Memorial University ofNewfoundland, Canada; tDepartment ofNephrology, University of Freiburg, Germany.
VHL disease is a dominantly inherited fa-milial cancer syndrome predisposing to a vari-ety of neoplasms, most frequently retinal and
CNS haemangioblastoma, renal cell car-cinoma, and phaeochromocytoma. Markedinterfamilial differences in predisposition tophaeochromocytoma are recognised. Weidentified germline mutations in 85 unrelatedVHL disease patients (22 large germline de-letions, 44 different intragenic mutations: 21missense, six nonsense, 16 deletions or in-sertions, one splice donor site mutation), andcorrelated this information with phenotype in65 kindreds. Large deletions or intragenicmutations predicted to cause a truncated pro-tein were found in 36 of 53 families withoutphaeochromocytoma but only two of 12families with phaeochromocytoma (X2 = 8-58,p<0-01). Of 12 families with phaeo-chromocytoma, 10 had missense mutationscompared to 13 of 53 kindreds withoutphaeochromocytoma (X2 = 12-33, p<0-001).Five percent of kindreds with large deletionsor intragenic mutations predicted to cause atruncated protein were phaeochromocytomapositive, compared to 43% of kindreds withmissense mutations. In particular, sub-stitution of an arginine at codon 238(Arg-.Trp or Arg-+Gln) was associated witha high risk (62%) of phaeochromocytoma.The identification of germline mutations inVHL disease not only allows presymptomaticdiagnosis, but can also indicate the risk ofphaeochromocytoma.
D BOURN, S MASON, S TEKES,S A CARTER, D G R EVANS*,T STRACHAN
Division ofHuman Genetics, University ofNewcastle upon Tyne; *Department of MedicalGenetics, St Mary's Hospital, Manchester.
Type 2 neurofibromatosis (NF2) is a dom-inantly inherited disorder characterised bya predisposition to multiple tumours of thecentral nervous system. The NF2 tumoursuppressor gene has recently been isolatedusing positional cloning methods. We havescreened for germline mutations in the NF2gene in patients with a range ofclinical pheno-types, using SSCP and heteroduplex analysisof PCR amplified exons or cDNA fragments.DNA sequence analysis of cloned am-plification products showed single nucleotidesubstitutions in 12 unrelated persons, ofwhich six were nonsense mutations, two weresplice site mutations, and four were missensemutations. Four frameshift mutations werealso detected, a lbp deletion in two cases, a2 bp deletion in one case, and a 4 bp insertion.Mutation screening has improved the pro-spects for diagnosis and predictive testing inaffected families, and tentative correlationscan be drawn between the type of mutationdetected and the resultant clinical phenotypefor some persons. Mutations giving rise to atruncated gene product appear to cause asevere phenotype, whereas the effect of mis-sense mutations is more variable.
Parents and offspring with thesame unbalanced chromosomerearrangements
J C K BARBER, M N COLLINSON,C E BROWNE, C M CAMPBELL,P L CAMPBELL, N M GREGSON,C A JOYCE, N M SILLITOE,I K TEMPLE*
We have ascertained 13 families in which thesame unbalanced karyotype has been foundin two or more generations. These includeeuchromatic deletions, duplications, and un-balanced translocations. When our 13 famil-ies are combined with 39 comparable familiesfrom published reports the 52 families can bedivided into three groups. Group I consistsof 21 families with no detectable phenotypicabnormality in parents or offspring. As-certainment was predominantly through pre-natal diagnosis for maternal age. Within thisgroup two of our own duplications suggestthat genetic activity in man can berepressed by chromatin condensation or bythe proximity of heterochromatin. Group IIconsists of another 21 families in which bothparents and children are phenotypically ab-normal. Ascertainment was predominantlythrough the phenotype of the affected child.Group III consists of 10 families in whichphenotypically normal parents have abnormaloffspring. Imprinting, or alternatively the co-incidence of an abnormal phenotype withan unrelated chromosome abnormality, mayexplain at least a proportion of the cases inthis group. In all three groups transmissionof the unbalanced karyotype by a mother (39/52) is far more frequent than transmission bya father (10/52). Transmission by both malesand females was found in the other threefamilies.
Molecular and fluorescence insitu hybridisation (FISH) studiesof supernumerary marker 15chromosomes
patients with mar(1 5)s not containing thePWS/AS regions were also observed and al-ternative explanations for their phenotypes,such as uniparental disomy, were excluded.Molecular results showed that the six cyto-genetically de novo mar(I 5)s which includedthe PWS/AS regions were maternal in originsuggesting that a dosage imbalance betweenmaternal and paternal copies ofthis imprintedregion results in mental retardation.
Possible role of imprinting in theTurner phenotype
C E CHU*, M D C DONALDSONt,C J H KELNAR4, P J SMAIL§,S A GREENEII, W F PATERSONt,J M CONNOR*
*Duncan Guthrie Institute, Glasgow;tDepartment of Child Health, Glasgow;tChild Life and Health, Edinburgh; §RoyalAberdeen Hospital, Aberdeen; JINinewellsHospital, Dundee.
Imprinting is postulated to be one of thefactors influencing the phenotype in Turner'ssyndrome (TS). The aim of this study wasto determine the parental origin ofthe normalX chromosome in 63 TS patients by Southernblotting or PCR. It was maternal (XM) in 43cases and paternal (XP) in 20 cases. In 36patients with height data, there was a stronglypositive correlation between the child's pre-treatment height centile and the mother'sheight centile for XM patients (r= 0-607,p<0-01) but not for Xp (r=-0-259, p<005).Child's pretreatment height centile and fath-er's height centile showed a non-significantcorrelation for XM patients (r= 0-17, p>0-05)and Xp (r= 0-225, p>0 05). Ten of63 patientshad cardiac anomalies, seven of 63 significantneck webbing, and eight of 53 renal an-omalies. If these data are pooled with pub-lished data, there is a highly significantassociation between XM and cardiac mal-formation (0-01>p>0 001) and neck webbing(p<005) but no association with renal mal-formation (p>0Q05). This study suggests thatsome of the phenotypic features in TS arethe result of genomic imprinting of areas ofthe X chromosome.
J A CROLLA, F L SITCH, J F HARVEY,N R DENNIS*
Wessex Regional Genetics Laboratory,Salisbury District Hospital, Salisbury,Wtltshire SP2 8BJ; *Wessex Clinical GeneticsService, The Princess Anne Hospital, CoxfordRoad, Southampton, Hants S09 4HA.
Clinical and molecular studies of 58 patientspresenting with either familial or de novo
autosomal supernumerary marker chro-mosomes showed that 17 were of chro-mosome 15 origin. Using a combinedconventional cytogenetics, FISH, Southernblotting, and multiplex polymerase chain re-
action approach, it was possible to subclassifythe 17 mar(I 5)s into six morphological andmolecular groups. Primers and cosmids fromthe Prader-Willi and Angelman syndromes(PWS/AS) critical regions were used to studyDNA and metaphase spreads from thepatients and their parents and the resultsshowed that all six patients with mar(I 5)scontaining these regions were mentally re-
tarded. However, two mentally retarded
Functional disomy of a locus(loci) in Xp and hypomelanosisof Ito
E HATCHWELL, I W CRAIG
Genetics Laboratory, Department ofBiochemistry, University of Oxford, SouthParks Road, Oxford OXI 3QU.
The description of a number offemales mani-festing the phenotype of incontinentia pig-menti (IP)/hypomelanosis of Ito (HI), inassociation with balanced X;autosome trans-locations, has led to the suggestion that alocus (loci) for this disorder resides at thecommon cytogenetic breakpoint, XpI 1.21.Linkage studies in familial IP, however, havefailed to confirm this localisation and pointto the existence of a locus at Xq28. Detailedmolecular analysis of a number of theXpl 1.21 breakpoints has shown that they areheterogeneous, and are separated by three
YAC contigs that, together, span a distanceof not less than 5 Mb. The order of the threebreakpoints mapped is Xcen-X;17-DXS14contig-X;9-ALAS2 contig-OATL2 contig-X;5. This result makes the likelihood of a singlegene being implicated extremely unlikely. Itis our hypothesis that partial functional dis-omy ofsequence(s) in proximal Xp may resultin the phenotype of HI, as a consequence of"atypical" X inactivation of translocated X;autosome chromosomes. An analogousmechanism may be implicated in HI resultingfrom the presence, in mosaic form, of ring Xchromosomes that fail to inactivate.
A Y;15 translocation with twodifferent unbalanced segregants
We report a family in which a phenotypicallynormal male with a balanced t(Y;l 5) (p1 1;q13) translocation has had three offspring.A dead daughter was obese and mentallyretarded. She had one normal 15, and theother 15 replaced by the derived Y (Yqter-+Yp 1::15ql3-+15qter); she was thereforemonosomic for proximal 1 5q (pter-q13)which is consistent with Prader Willi syn-drome. Her younger brother is phenotypicallynormal and has inherited the translocation inbalanced form. The youngest son is hypo-gonadal and has mild learning difficulties.He has two normal 15s and the derived 15(15pter--+ql3::Ypll -Ypter) which effect-ively substitutes for a Y chromosome in hiskaryotype. FISH with cosmids for theD15SIl and GABRB3 loci shows that hisderived 15 includes the Prader Willi/Angel-man (PW/AS) critical region and that hetherefore has three copies of proximal 15q.Most patients with duplications of the PWS/AS region have additional maternally derivedcopies which are associated with severe men-tal retardation, seizures, and behavioural ab-normalities. The youngest son is very unusualin having an extra paternally derived copy ofthe PWS/AS region associated with a sig-nificantly milder phenotype. This familytherefore suggests that phenotypic con-sequences of duplications as well as deletionsof the PWS/AS region are subject to theeffects of imprinting.
ionising radiation. We have used in situ hy-bridisation with whole chromosome paintingprobes to investigate the chromosome contentof radiation induced micronuclei. Fourlymphoblastoid cell lines established from aperson with ataxia telangiectasia, a personwith aniridia, an infant with mitotic in-stability, and a healthy person were irradiatedwith 2-0 Gy. Slides with micronuclei wereprepared using the cytokinesis block tech-nique or by flow sorting. The slides werepainted with probes for chromosomes 1, 7,and 14 in separate experiments, and an av-erage of 200 micronuclei were scored forhybridisation signals. We did not find anysignificant difference in the numbers of mi-cronuclei with positive hybridisation signalsin these cell lines. Although this may bebecause of the small numbers of micronucleiscored, these preliminary results suggest thatthe chromosome breaks induced by ionisingradiation and the incorporation of damagedDNA into micronuclei is probably randomfor these chromosomes.
The Human Genetic andChromosome AbnormalityRegister
The Human Genetic and Chromosome Ab-normality Register (HGCAR) acts as a cent-ralised repository for records ofchromosomaland genetic abnormalities contributed by theEuropean Human Cell Bank and the ap-proximately 50 laboratories around the UKwhich perform routine diagnostic karyo-typing. This resource enables researchers tolocate cases of interest without having tocontact all of the laboratories individually.The database now stores over 50 000 recordsofchromosome abnormality, for each holdingthe sex, date of birth, sample type, and karyo-type, as well as the diagnosis or reason forreferral and details of the laboratory fromwhich the record was collected. It also holdsinformation on the availability of storedsamples and cell lines where these are avail-able. Any researcher can access the data freeof charge. Please contact Dr Mercer by faxon (0865) 226006.
odology (acceptance policy (if any) and tech-niques) (29 returns). Laboratories thenundertook a prospective survey of a con-secutive series of samples (maximum 40) en-tering the laboratory (26 returns). Analysis ofdata indicated a wide variation in acceptancepolicy and technical procedures. A numberof factors including Transit Time, TransportMedium, Origin of Specimen, Type of Speci-men Received, and Tissue Cultured havebeen examined with respect to successful out-come, abnormality rates, and sex ratios ofcultured tissues. This study has allowed usto draw conclusions as to what constitutes"good practice" which it is hoped will be ofbenefit to all laboratories.
Cancer families: what risks arethey given and do the risks affectmanagement?
E M ROSSER, J A HURST,C CHAPMAN
Department of Clinical Genetics, OxfordRadcliffe Hospital, The Churchill, OxfordOX3 7LJ.
The number ofpersons referred with a familyhistory of malignancy is increasing rapidly.We noted that reference texts gave differentrisk estimates and also disagreed aboutscreening protocols. To determine whetherthis was leading to different clinical man-agement we devised a questionnaire on riskestimation and screening. This was sent toall consultant geneticists earlier this year. Itcontained information on four families andasked what risks would be given to a probandin each family, whether screening was ne-cessary for the proband, and, if appropriate,what screening should be offered. The replieswere compared to those obtained using ref-erence texts. The four pedigrees were all takenfrom referrals we had seen in clinic, andshowed a family history ofeither breast, colon,possible breast-ovarian, or multi-site cancer.There was consistency between answers forthe colon cancer family, but a wide variationbetween replies for other families, both interms of risk given and suggested man-agement. This variation was also present inthe reference texts. This study illustrates thedifficulties in advising patients. However, itis important that different family members,who may be seen in different centres, aregiven consistent advice.
A M SLAVOTINEK, E J THOMSON,J FANTES, G M TAYLOR*,V VAN HEYNINGEN, M NUSSEt
MRC Human Genetics Unit, Western GeneralHospital, Edinburgh EH4 2XU; *DepartmentofMedical Genetics, St Mary's Hospital,Manchester; tGSF-Institut furBiophysikalische Strahlenforschung,Oberschleissheim, Germany.
Ionising radiation is a potent inducer of cyto-genetic damage which is detectable by scoringmicronuclei. Little is known about the natureof the chromosome breakpoints induced by
ACC Solid Tissue WorkingParty: a progress report
J J WATERS, M R CREASY,M FITCHElT, C MALIZEWSKA,N PRATT, C S RODGERS
Solid tissue samples are given a low priorityby most laboratories, but are neverthelesstechnically demanding and time consumingto process. UK NEQAS data show a widevariation in success rates and abnormalityrates. An ACC survey of all UK CytogeneticsLaboratories was undertaken. Laboratorieswere asked to describe their solid tissue meth-
The impact of geneticcounselling on risk perception inwomen with a family history ofbreast cancer
D G R EVANS*, V BLAIRt,P HOPWOODt, A HOWELL§
*Department ofMedical Genetics, St Mary'sHospital; tCRC Paediatric and FamilialCancer Research Group; tCRC PsychologicalMedicine Group; §CRC Department ofMedical Oncology, Christie Hospital,Manchester.
Women with a family history of breast cancergenerally self refer because they have a feeling
that their risk is high. However, they have, ingeneral, only a hazy notion of the populationrisk of breast cancer and their own risk inrelation to this. It is probable that they arehelped by genetic counselling and, if at sub-stantial risk, annual mammography. How-ever, the psychological impact of assigningtrue risk and the value of mammographyneed to be evaluated. We have assessed riskperception in 519 new referrals to a familyhistory clinic and 200 women returning tothe clinic at least one year after counselling.Correct assessment ofpopulation lifetime riskof breast cancer in the pre-counsel group was16% and in the post-counsel group 33%.Similarly personal risk was correct in 11%and 41% respectively. Post-counsel womenwere significantly more likely to retain in-formation if they were sent a post-clinic letteror if they assessed their personal risk as toohigh initially. Compliance with screening inour family history clinic is exceptionally high(98%) in contradistinction to clinics in NorthAmerica.
Factors influencing decisions onwhether to proceed withpredictive testing for breastdovarian cancers
A BINCHY, D G R EVANS, C ENG*,B A J PONDER*, D CRAUFORD
Clinical Genetics Department, St Mary'Hospital, Manchester; *CRC Human CancerGenetics Research Group, University ofCambridge.
Recent data from the International Con-sortium for linkage in familial breast andovarian cancer suggests that a gene on chro-mosome 17q accounts for the majority offamilies in which early onset breast andovarian cancer occur. This breast and ovarianlocus has been labelled BRCA1. We are con-ducting a study assessing the impact of theavailability of this test on five families in eachof which the probability of linkage to 1 7q12-21 is estimated to be greater than 95% andin whom clinical contact is well established.All family members who have an affected firstdegree relative or second degree where theintervening relative has died and are agedbetween 25 and 70 years are being offered apreliminary discussion in their own homesabout the test and its implications. Seventypersons are eligible, 41 (58%) are female,and 19 (47%) have undergone prophylacticoophorectomies and three prophylactic mast-ectomy. At present the options for womenwho have inherited the gene are screeningand surgery. Reasons for choosing to proceedamong both sexes include a desire for cer-tainty and to provide information to youngergenerations. Women who have undergonepreventative surgery appear to find making adecision to proceed with the test less daunt-ing.
Attitudes to predictive testing forBRCA1
S MOHAMMED, C BARNES, S WATTS,S MICHIE*, S HODGSON
SE Thames Regional Genetics Centre;*Psychology and Genetics Research Group,Guy's Hospital, London SEI 9RTJ
We have undertaken a pilot study to assessthe likely uptake and perceived impact ofundergoing predictive testing for breast can-cer susceptibility, when available. Postal ques-tionnaires explaining BRCA1 testing weresent to 50 women attending the Clinical On-cology Unit because of a family history ofbreast cancer in first degree relatives. Eightysix percent of women responded, 81% in-dicating a desire to undertake testing. Com-mon motivations for requesting testing wereto assess personal risk (58%) and to optimisescreening (62%). Sixty percent anticipatedfeeling moderately/extremely anxious on test-ing positive, and 9% on testing negative (p=0-002). Twenty eight percent felt that a "lowrisk" result would not relieve anxiety. Con-sultand's age, extent of family history, or pre-vious perception of risk had no significanteffect on the stated desire to undertake test-ing. Women daunted by the complexities ofpredictive testing were not deterred from re-questing such a test when available. Despitethe limitations of the hypothetical nature ofthis study, the results reflect the experienceof testing for other late onset disorders. Theimplications of predictive testing on women'sperceptions of cancer risk, their un-derstanding of genetic testing, and possiblepsychological sequelae are likely to have im-portant consequences for future counsellingprotocols.
A pragmatic randomised trial was performedto compare stepwise and couple approachesto antenatal carrier screening for cystic fib-rosis (CF). Uptake of screening was the samefor both approaches (90%). After delivery,most women remembered test results andtheir meaning, but 21% ofthose with negativecouple test results had forgotten that repeattesting would be advisable if they had a preg-nancy with a new partner. With stepwisescreening, women identified as carriers hadhigh levels of anxiety when results were re-ceived. This dissipated with a reassuring part-ner's result, although anxiety levels remainedslightly higher than among those given a neg-ative result. Women receiving negative resultsfrom stepwise screening were significantly lessanxious than those receiving negative resultsfrom couple screening. Stepwise screeninggives information for future pregnancies with
different partners, allows carriers' relatives tobe informed, and avoids the ethical difficultyofwithholding information. Couple screeningallows carriers to avoid transient high levelsof anxiety, but it is associated with moreanxiety and false reassurance among thosewho screen negative.
Genetic counselling problems infamilies with multiplehaemostatic disorders
F G H HILL, B THEOPHILUS,M S ENAYAT
Haematology Department, The Children'sHospital, Birmingham.
The existence of more than one disorder willonly be suspected if clinical symptoms haveoccurred and been investigated. Multiple hae-mostatic disorders are very rare in individualpeople and in families. We report the fol-lowing. (1) Coinheritance of haemophilia Ain a boy from his carrier mother and typeIIA von Willebrand's disease (vWD) fromhis asymptomatic father. Haemophilia A andvWD was diagnosed in the boy by phenotypicinvestigations. Recently, using genetic tech-niques, it was found thatvWD had risen froma Leu 894 Ile substitution on vWF gene andhaemophilia from Arg 2163 His substitutionin the FVIII gene. (2) Inheritance of hae-mophilia A and B in a family with establishedmild haemophilia A. A mother already witha son affected by haemophilia A had a secondson with normal VIII:C levels, but a pro-longed PTT' and 0% FIX indicated severehaemophilia B. Gene sequencing showed, inthis boy but not the other, A 31,314 C trans-version resulting in a Tyr 398 Ser substitution(Giannelli, personal communication), andverified that the mother and grandmotherare carriers of haemophilia B. Although rare,these two families show that during geneticcounselling it must be stressed that the coun-selling is about the known disorders and oc-casionally new disorders can also emerge.
General practitioners'expectations of the clinicalgenetics service in two districtsin the West Midiands
P A FARNDON, J AFFLECK,J MOORE, B GIBBONS, A BINKS,J HOWARD
Clinical Genetics Unit, Birmingham MaternityHospital, Birmingham B15 2TG.
Twenty-five percent of referrals in District Awere from General Practitioners (GP), com-pared with 15% in District B. A's local geneticclinic had been in operation for over 10 years,while B's clinic had been established for twoyears. We wished to investigate the views andopinions of GPs regarding referral to theservice. A postal questionnaire showed thatopinions about the successful outcome ofgenetic counselling were the same in eachdistrict: accurate estimation of genetic risks,information and support to families, and areduction in incidence of genetic disease (inthat order). The proportions of GPs whowould deal with genetic diseases rather thanrefer was broadly similar in each district:
DMD 3.5%, CF 20%, FAP 38%, trans-location 6%, HD 6%, NF1 35%. However,for most of the common genetic disorders upto 16% of GPs were uncertain as to whetherto refer. About 40% of GPs would telephonefor advice. There were many requests formore information on basic genetics, and themanagement and necessity for referral. Aninformation and education package is beingdeveloped; a booklet for GPs has already beencirculated.
Gene patenting: a historicalreview
T CLANCY
Department of Clinical Genetics, St Mary'sHospital, Manchester.
Biotechnology has attracted considerable aca-demic and commercial interest because ofits humanitarian and economic importance.Along with this has come the application ofintellectual property law to the field. Thisarea of law exists to protect the rights ofinventors, and to encourage their endeavours.Patents are the mechanism of choice in bio-technology because of the strength of pro-tection provided. Although there is asubstantial history of patenting in bio-technology, gene patenting is a recent phe-nomenon. The significant events of the past15 years include the seminal cases ofDiamondv Chakrabarty, the Harvard oncomouse, andthe first patents for claims pursued for humantissues and cells. Gene patenting came ontothe agenda in October 1992 when Ventnerannounced that the NIH intended to applyfor patent protection for cDNA sequences.This move caused furore within the scientificand clinical communities. The NIH ap-plication has been unsuccessful so far, and isgenerally thought to be likely to fail underUS, UK, and European law. New ways ofoffering intellectual property protection forgenes (which are being developed in an at-tempt to satisfy both commercial and sci-entific/clinical interests) are being developed.
Characterisation of a geneencoding a novel peroxisomalmatrix protein, PXEL
DAVID FITZPATRICK, DAVID VALLE
Howard Hughes Medical Institute, JrohnsHopkins University School of Medicine,725 N Wolfe St, Baltimore, MD 21205, USA.
To identify the genes involved in disorders ofperoxisomal biogenesis and function we haveused a subtractive hybridisation method toisolated rat liver cDNAs upregulated bytreatment with peroxisome proliferators (clo-fibrate and DEHP). In a pilot study of 20/53 (37-7%) upregulated cDNAs isolated andsequenced using this method were knownperoxisomal genes. Two of the remainingclones were fragments of a previously un-known cDNA that showed >20-fold in-duction. The full length cDNA was isolatedand predicts a protein product of 36 kDa witha C-terminal peroxisomal targeting signal(-SKL). This protein was epitope tagged witha C-myc dodecapeptide and found to be ef-
ficiently imported into peroxisomes inHEK293 cells by double label im-munofluorescence. A search of the proteinsequence against the public databases showedhomology to enoyl-CoA hydratases from awide variety of species. We have named thisgene peroxisomal enoyl-CoA hydratase-like(PXEL). We have also isolated human PXEL(>85% identity in both nucleotide and aminoacid sequence when compared to rat PXEL)and mapped it to 19qI3.1 in a contig 3' tothe ryanodine receptor using hybridisation tosomatic cell hybrid DNA and chromosome19 specific cosmid. Northem blot analysisof tissue distribution showed high levels ofexpression of a 1-4 kb message in skeletal andheart muscle with a detectable transcript inevery tissue examined. To investigate thefunction of this gene we are examiningpatients with disorders of peroxisomal b-oxi-dation for mutations in the PXEL gene.
Screening for rare CF mutationsusing a joint SSCP/heteroduplexstrategy
nique is ideal. In our hands the most sensitivetechnique other than direct sequencing isthe chemical cleavage of mismatch (CCM)method of Cotton et al. This method requiresa great deal of hands-on time and the useof toxic chemicals. By analogy with DNAsequencing technology we sought enzymaticalternatives to the CCM method, using themismatch binding protein MutS, which bindsto all mismatches in heteroduplex DNA. Het-eroduplex molecules were produced by PCRamplification of CFTR heterozygotes usingfluorescent primers. MutS was bound to het-eroduplex in a 4:1 molar ratio. DNA wasthen subjected to exonuclease attack from the3' ends. MutS binding protected heteroduplexDNA at the site of binding, so that partiallydegraded molecules could be resolved by gelelectrophoresis. Using this technique we wereable to localise four different point mutationsin a 492 base fragment of CFTR exon 11.Further templates and mutations are underinvestigation. Problems with the assay includethe low stability of E coli MutS and its lowaffinity for some mismatches.
A J WALLACE*, M TASSABEHJI*,A DALTONt
*Regional Molecular Genetics Laboratory,Department ofMedical Genetics, St Mary'sHospital, Manchester; tCentre for HumanGenetics, 117 Manchester Road, Sheffield.
We are using silver staining to detect SSCPsand heteroduplexes in CF patients and knownCF carriers with mutations not detected bythe Standard Cellmark Diagnostics multiplexARMS assay. Multiplexes ofbetween two andfour exons are electrophoresed at +4C on
32cm polyacrylamide gels with a 50:1 ratioof cross linking for at least 16 hours afterwhich they are silver stained and dried be-tween sheets of cellophane. SSCPs and het-eroduplexes are visualised simultaneously onthe same gel and only samples which shownew shifts are sequenced. Dual analysis in-creases the sensitivity of mutation detectionsince some mutations give rise to visible het-eroduplex shifts and not SSCPs and viceversa. Analysis of 22 of the 27 CFTR exons
increased our detection rate to 97% (607/626) of CF mutations. In addition we haveidentified seven mutations and one poly-morphism not previously observed by othermembers of the International CF Con-sortium. In total 43 separate mutations havebeen identified in our sample of CF patients,22 ofwhich are only found in a single person.We have now integrated this technique intothe standard diagnostic CF service.
Detection of "unknown"mutations using the MutEx assay
G R TAYLOR, L A ELLIS,E A SHERIDEN, K J LEACH,R F MUETL R
DNA Laboratory, Clinical Genetics Service, Stjames's University Hospital, Leeds LS9 7TF.
The wide variety oftechniques now describedfor the detection of mutations, particularlypoint mutations, suggests that no one tech-
Identification of three mutationsin the porphobilinogendeaminase gene in four SouthAfrican patients with acuteintermittent porphyria
P M L ONG, W G LANYON,M R MOORE*, J M CONNOR
Duncan Guthrie Institute of Medical Genetics,Yorkhill, Glasgow G3 8S_J; *NationalResearch Centre for Environmental Toxicology,The University of Queensland, 39 KesselsRoad, Coopers Plain, Queensland 4108,Australia.
The RNA of four South African patientswith acute intermittent porphyria (AIP) wereanalysed by direct sequencing of asym-metrically amplified cDNA transcripts fol-lowing a reverse transcriptase polymerasechain reaction (RT-PCR). Three differentmutations of porphobilinogen deaminase(PBG-D), two involving single base sub-stitutions and one a single base insertion, weredetected. RI 16W (found in two patients)and R173Q affect highly conserved residues.Restriction endonuclease analysis was per-formed using ActI for RI 16W and Nci[ forR173Q. Argll6, equivalent to ArglOl in Ecoli, interacts with the interdomain hingeregion by forming a salt bridge with Glu231and a hydrogen bond with the backbone car-bonyl group of residue 198. Argl55, equi-valent to ArgI73 in the human, forms saltbridges with the propionate group of the firstpyrrole ring of the cofactor and the carb-oxylate side group of the substrate molecule.Its replacement inhibits substrate binding andelongation. The third mutation is a T in-sertion at nucleotide position 771, resultingin a stop codon 33 codons downstream fromthe insertion. Approximately 20% of the pro-tein is lost.
Phylogenetic analysis ofvertebrate and invertebratedystrophins and utrophins
R G ROBERTS, L NICHOLSON,Q BONE*, M BOBROW
Paediatric Research Unit, Guy's Hospital,London; *Manne Biological AssociationLaboratory, Plymouth.
Dystrophin is a cytoskeletal protein expressedin muscle and neural tissue, and is defectivein Duchenne and Becker muscular dystrophy(DMD, BMD). Its only close homologue,utrophin, is more ubiquitously expressed butis yet to be associated with a genetic disorder.The function of neither protein is known.Human utrophin differs sufficiently in prim-ary structure from the known dystrophinsto suggest that the dystrophin and utrophinbranches of this gene family diverged beforethe vertebrate radiation. Using degeneratereverse transcript PCR, we have characteriseddystrophin and utrophin transcripts from arange of vertebrate and invertebrate animals.Phylogenetic analysis of the resulting se-quence data suggests that the gene du-plication leading to distinct dystrophin andutrophin sequences occurred near to thedivergence ofurochordates from the cephalo-chordate-vertebrate clade. Characterisa-tion of the role of these primitive dys-trophins may shed light on the function ofdystrophin and utrophin in higher organismssuch as man. In addition, our results representa substantial body of sequence data whichwill aid the interpretation ofDMD and BMDmissense mutations in the dystrophin cysteinerich and C-terminal domains. We believe theutrophin sequences to be the first known non-human utrophins, and find these to be moredivergent than cognate dystrophins.
Analysis of point mutations inthe dystrophin gene
R J GARDNER, M BOBROW,R G ROBERTS
Paediatric Research Unit, 8th Floor, Guy'sTower, Guy' Hospital, London SE1 9RT
Using reverse transcription and nested PCR(RT-PCR) followed by the Protein Trun-cation Test (PTT), we have analysed thedystrophin gene in patients with Duchennemuscular dystrophy (DMD) who have nodeletion as detectable by multiplex analysis.The RNA is extracted from blood lympho-cytes and amplified in a set of 10 overlappingsections. The 5' nested primer is modifiedto contain a T7 promoter and eukaryotictranslation initiation signals. The product isthen transcribed and translated in vitro andanalysed by SDS-PAGE. Fifteen mutationscomprising eight nonsense mutations, a singlebase pair insertion, a single base pair deletion,a deletion of four base pairs, an exon du-plication, two single exon deletions, and adeletion of two adjacent exons were detected.One of the nonsense mutations was detectedin a female patient. We have seven remainingpatients in whom a small mutation or otherstructural abnormality has not been found.These patients are being analysed by chemicalcleavage of mismatch in order to determinewhether a missense mutation may be thecause of the disease.
Mutation detection and carrierdiagnosis in Duchenne musculardystrophy by pulsed field gelelectrophoresis
We have applied the technique ofpulsed fieldgel electrophoresis to mutation detection andcarrier diagnosis in Duchenne and Beckermuscular dystrophies. Over the past threeyears we have analysed 402 persons from 209families with a suspected dystrophinopathy.Although affected persons or obligate carrierswere not available in all families we identifiedaltered sized SfiI restriction fragments in 71families (56 deletions and 15 duplications)and hence there is a qualitative test of carrierstatus. In order to increase confidence in thespecificity of the analysis we have additionallyanalysed 120 unaffected males and in thesepersons no abnormally sized SfiI restrictionfragments were detected. The technique hasproved to be an effective means of detectingduplications, and in the vast majority of de-letion/duplication families the identificationof carrier status is unambiguous. In 21 of thedeletion/duplication families no affected malewas available for analysis by molecular tech-niques and mutations were identified andcharacterised in female carriers.
On the mechanisms of sexchromosomal gene expressionwith particular reference to Xinactivation
M A HULTEN, S J ARMSTRONG
LFS Research Unit, Regional Genetic Services,Birmingham Heartlands Hospital, YardleyGreen Road, Birmingham B9 5PX.
We have used multicolour fluorescence in situhybridisation (FISH) to study the in-tranuclear spatial orientation and chromatincondensation patterns of the sex chro-mosomes (1) in the soma ofmales and femaleswith normal karyotypes as well as con-stitutional sex chromosome aberrations withparticular respect to the Barr body, and (2)in premeiotic spermatogonia and first sper-matocytes in "sex vesicle" formation, in-activating the X and Y In both situationsthere is a similar complex series of chro-mosome movement and differential chro-matin condensation, involving in particular aclosed ring formation by the X bending at asite near the centromere on Xq and pairingof its telomeres. On the basis of these ob-servations we propose that (1) the XIST(human)/Xist (mouse) gene functions to im-pose a chromatin reorganisation, leading tobending at the X inactivation centre (XIC)at both somatic mitosis and meiosis, (2) thedifferential X and Y segments are protectedfrom potentially deleterious meiotic ex-changes by their separate spatial orientation,and (3) the differential decondensation of theYq heterochromatic block in the premeioticspermatogonia may activate genes involvedin initiating and maintaining spermatogenesisby a mechanism similar to heterochromatin
position-effect variegation in Drosophila. Wefurther suggest that our observation may helptowards understanding genomic imprintingin general.
Mutational bias provides a modelfor the evolution of Huntington'sdisease and predicts a generalincrease in disease prevalence
D C RUBINSZTEIN*, W AMOSt,J LEGGO*, S GOODBURN*,R S RAMESAR4t, J OLD§, R BONTROPII,R McMAHON*, D E BARTON*,M A FERGUSON-SMITH*
*East Anglian Regional Genetics ServiceMolecular Genetics Laboratory andDepartment of Clinical Genetics, Box 158,Addenbrooke's NHS Trust, Hills Road,Cambridge, CB2 2QQj tUniversity ofCambridge Department of Genetics;tDepartment ofHuman Genetics, University ofCape Town; §Institute of Molecular Medicine,John Radcliffe Hospital, Oxford; IrMedicalBiological Laboratories TNO, Riiswijk, TheNetherlands.
Huntington's disease (HD) correlates withabnormal expansion in a block of CAG re-peats in the Huntington's disease gene. Wehave investigated HD evolution by typingCAG alleles in several human populationsand in a variety of primates. We find thathuman alleles have expanded from a shorterancestral state and exhibit unusual asym-metrical length distributions with more chro-mosomes lying above rather than below themodal allele length. Computer simulationsare used to show that the human state canbe derived readily from a primate ancestor,without the need to invoke natural selection.The key element is a simple length-dependentmutational bias towards longer alleles. Ourmodel can explain a number of empiricalobservations, and predicts an ever increasingincidence of HD.
A major susceptibility gene forpsoriasis maps to 17q in Britishfamilies and evidence for geneticheterogeneity
R C TREMBATH*, J A ROSBOTHAM*t,N PERIAM*, J A L ARMOUR*,J N W N BARKER
*Departments ofMedicine and Genetics,University of Leicester; tInstitute ofDermatology, UMDS, St Thomas's Hospital,London.
Psoriasis is a common and serious in-flammatory disorder of the skin, the aetiologyof which remains unclear. We have recruitedand clinically documented psoriasis in a totalof 60 multigeneration families (affected n=162) from the UK. We have excluded a majordisease susceptibility gene at HLA-C using aDNA sequence specific assay. In a genomesearch, D17S26, a minisatellite, was foundto be linked in eight families, Zmax = 3.42 at0 = 0-06. In a further two psoriatic kindreds,exclusion of a major disease susceptibilitylocus at 17q was confirmed. No significantdifference in clinical severity or age of disease
onset could be identified between chro-mosome 17q linked and unlinked families. Acandidate gene, interleukin enhancer bindingfactor, which maps within the linkage con-fidence interval is now under investigation.This is the first major disease locus to beidentified in a common cutaneous disorder.
Mapping of the tylosis cancer
gene in a Liverpool pedigree withpalmoplantar keratoderma andoesophageal cancer
J K FIELD*, J M RISK*,J WHITTAKER*, A FRYER*, A ELLIS*,J M SHAW*, E A FIELD*,P S FRIEDMANN*, T BISHOPt,J BODMERt, I M LEIGHt
*The University of Liverpool, PO Box 147,Liverpool L69 3BX; tICRF, London.
Tylosis (palmoplantar keratoderma) is an
autosomal dominant disease that is readilyidentified by a thickening of the skin of thepalms and soles. We have recently updatedthe largest of the Liverpool families (S family)which now contains 345 persons, of whom21 have died of oesophageal cancer. The
anatomical sites affected in this Liverpoolfamily are characterised by the expression ofkeratins 6, 16, and 17. The keratin gene
clusters are located on chromosome 12 (typeII) and on chromosome 17 (type I). We havedetermined genotypes for 58 persons fromgenerations IV to VI of the "S" tylotic familyusing microsatellite markers on chromosomes12 and 17. No linkage was shown for markerson chromosome 12. Thirty three polymorphicmicrosatellite markers from chromosome 17were tested in this pedigree. A number ofthese markers mapped to the keratin genecluster at 17ql2-q21 and included KRT10and KRT9 microsatellites. Close linkage tothe markers D17S250, THRA1, KRT1O,KRT9, D17S800, D17S810, D17S791, andD17S579 in the 17ql2-q21 region was ex-cluded. However, we have found a sig-nificantly positive lod score with more distalmarkers on chromosome 17. It is proposedthat the late onset tylotic cancer gene locusshould be termed "LOTCG".
Mapping a gene for Noonansyndrome to the long arm ofchromosome 12
C R JAMIESON*, I VAN DER BURGt,A F WADE*, M VAN REENt,M ELSAWI*, F HOLSt, S JEFFERY*,M A PATTON*, E MARIMANt
*St George's Hospital Medical School, LondonSW17 ORE; tNijmegen University Hospital,Nijmegen, The Netherlands.
Noonan syndrome (NS) is an autosomaldominant disorder of unknown origin char-acterised by typical facial features, short stat-ure, and congenital heart disease, and is thesecond most common syndrome associatedwith congenital heart disease after Down'ssyndrome. To localise the gene for NS, link-age studies were carried out in 19 mul-tigeneration families. A total of 57 patientsaffected out ofa total of 106 members from 19
families were analysed. All affected patientsfulfilled the minimum diagnostic criteria. Inthe largest family, nine out of 29 memberswere affected. Genotypes using 200 (CA)nmicrosatellite markers were analysed and twomarkers, D12S105 and D12S79, producedsignificantly positive lod scores in this family(Zmax=4-32 at 0=0). Analyses using theseand other markers were carried out in the 19families and the lod scores obtained wereZmax= 373, Zmax=4-98, and Zmax= 558at 0 = 0 1 at the D 1 2S105/D 1 2S354/D1 2S79loci respectively. This is the first time that a
locus for NS has been mapped.
Consanguinity as a determinantof pre-reproductive mortality: a
global assessment
A H BITLES*, JV NEELt
*Department ofHuman Biology, Edith CowanUniversity, Perth, Western Australia;tDepartment ofHuman Genetics, University ofMichigan Medical School, USA.
Marriage between close biological relatives isstrongly preferred by many major pop-ulations. To obtain a representative value forconsanguinity associated mortality, the resultsof surveys conducted on 38 mainly urbanpopulations located in east, south, and westAsia, West Africa, South America, and west-ern Europe were analysed. The total data setcomprised 622 188 pregnancies or live births,each individual study was based on a mini-mum of 750 families, and all information wascollected within the last two generations. Foreach population, mean numbers of deathsfrom late abortions/miscarriages to a medianage of 10 years were compared at first cousinand non-consanguineous levels. By plottingan unweighted linear regression, mean excess
mortality at first cousin level was found to be4-4% (± 4 6). R2 for the regression was 0-80,with no evidence to suggest an increasedprevalence ofconsanguinity associated deathsat higher levels of background mortality. Itis proposed that the excess mortality thusdetermined, that is, 44 deaths/1000 at firstcousin level (F=0-0625), can be adopted as
a reference value to predict the adverse effectsof consanguinity on early survival across allpopulations in which consanguineous unionsare strongly favoured, including migrant com-munities.
Microsatellite instability insquamous cell carcinoma of thehead and neck
H KIARIS*t, P HOWARDt,E D VAUGHAN*, D A SPANDIDOSt,A S JONES§, J K FIELD*
*Department of Clinical Dental Sciences, TheUniversity of Liverpool, PO Box 147,Liverpool L69 3BX; tNational HellenicResearch Foundation, Athens; tRegionalCytogenetic Unit, Royal Liverpool HospitalTrust; §Department of OtorhinolaryngologyThe University of Liverpool.
Genomic instability at simple repeated se-
quences or microsatellite instability (MI) has
been recognised in carcinoma ofthe colon andin a number of other carcinomas. Although anumber of reports exist on MI in varioustumours, its real significance in tumour pro-gression is unknown. We have investigated34 microsatellite markers in 10 chromosomesin squamous cell carcinoma of the head andneck (SCCHN). Fifty-six tumours have beenstudied of which 36 have been investigatedwith microsatellite markers in at least twochromosomes; 22/36 (61%) of the tumourswere found to have MI at one locus and 7/22 had evidence of two or more loci with MI.No correlations were found between MIand previously untreated/previously treatedpatients, site, histological differentiation, pos-itive nodes at pathology, or a history ofalcoholintake. Three of four TNM stage 1 tumourshad one or more MI, indicating that thesechanges occur early in the disease process.An association has been found between MIshifts and clinical outcome; none of the 14patients without MI has died, whereas 6/22of the patients with MI died. Two or moremarkers of MI were found in three of fournon-smokers compared to one of 17 in thesmoking group of patients which suggest anovel mechanism of carcinogenesis in non-smokers.
The population genetics ofphenylketonuria in NorthernIreland
J ZSCHOCKE*, C A GRAHAM*,J P MALLORYt, F J STEWART*,D J CARSONt, N C NEVIN*
By direct sequencing we were able to de-termine all mutations in a random sample of70 Northern Irish PKU alleles. We found21 different mutations, with three mutations(R408W, I65T, and F39L) accounting for67% of PKU chromosomes. The spectrumof mutations reflects the history of the peoplein Northern Ireland and the British Isles as awhole. I65T, the second commonest PKUmutation in Ireland and also frequent inSpain, was probably present in the originalPalaeo-/Mesolithic settlers who belonged toa west European group of people and arrivedin Ireland after 7000 BC. R408W, in contrast,the commonest Irish PKU mutation, wasprobably prevalent in the Neolithic farmerswho settled after 4500 BC. Several less fre-quent mutations can be traced to Scandinaviaand were probably introduced into Irelandby Norwegian Vikings. The rarity in NorthernIreland of IVS 1 2ntl, the commonest muta-tion in Denmark and England, is a vividexample that drastic changes in culture andlanguage of a population may occur withor without major changes of the gene pool.Similarly, the cultural and technologicaltransformation from Bronze Age Ireland to aCeltic society probably occurred without largescale immigration of "Celtic" people, and theterm "Celtic" should not be used to describe"Irish genes".
Brachydactyly and mentalretardation: a newhaploinsufficiency syndrome
L C WILSON*t, M E M OUDELUTTIKHUIS*, J FLINT*,J V LEONARDt, R C TREMBATH*
*Departments of Genetics and Medicine,University of Leicester; tInstitute of ChildHealth, London; 4Jnstitute of MolecularMedicine, Oxford.
Combined brachydactyly and mental re-tardation is a feature of Albright's hereditaryosteodystrophy (AHO) where it is typicallyassociated with a 50% reduction in Gsa bio-activity and heterozygous mutations in theGsot gene on chromosome 20q13. Phenotypicsimilarities occur in acrodysostosis and bra-chydactyly type E. We have recruited 40patients on the basis of their AHO phenotypeand have identified five patients with normalGsoa levels. All had normal prometaphasekaryotypes. Following a report of cyto-genetically visible 2q37 deletions in twopatients with features of AHO, we screenedthis subset for microdeletions of 2q. In onepatient we have identified a maternally de-rived deletion using markers D2S90 andD2S125 which map within 2q37.3 and con-firmed this by fluorescent in situ hybridisation(FISH) with D2S90. A second patient isapparently homozygous for all 11 2q miniand microsatellites tested so far. The mi-crodeletion identified is the smallest 2q ter-minal deletion reported to date and providesevidence for the localisation of genes im-portant in skeletal morphogenesis and neuro-development to 2q37.3. Clearly this is acandidate region for brachydactyly type E.Haploinsufficiency at 2q37 should be ex-cluded in patients with brachydactyly andmental retardation and is potentially amen-able to analysis by FISH.
Only 27% of those with renal involvementhad any symptoms from their renal disease,the commonest symptom being haematuria.There was no correlation with other featuresofTSC. We have drawn up guidelines for themanagement of renal disease in TSC.
"DIDMOAD" syndrome: adegenerative dilemma
T G BARRETT, S BUNDEY
Clinical Genetics Unit, Birmingham MaternityHospital, Edgbaston, Birmingham.
"DIDMOAD" is the acronym given to thesyndrome of Diabetes Insipidus, juvenileonset Diabetes Mellitus, Optic Atrophy, andDeafness. The estimated prevalence is1:100 000 children. Because of its rarity, noclinical study has included more than sevenpatients. As part of a wider biochemical, gen-etic, and mitochondrial case finding studywe present the clinical findings and initialinvestigations on the first 20 patients, outof 53 ascertained, and discuss the possiblemechanisms ofinheritance and aetiology. Theages of our patients ranged from 7 to 45years. They all had our defining criteria ofjuvenile onset diabetes mellitus and optic at-rophy. Fourteen of the 20 had high tonesensorineural deafness. Diabetes insipiduswas also found in 12 patients. Renal tractabnormalities were found in seven out of 10investigated. Magnetic resonance imaging onfour patients showed widespread brain at-rophy, and absence of the posterior pituitary.Eight patients had severe truncal ataxia, andtwo had a brainstem or cortical infarct. Onepatient had biochemical evidence of primarygonadal atrophy. Three index patients hadconsanguineous parents, which would sug-gest an autosomal recessive mode of in-heritance. This should be regarded as aninherited degenerative ataxia which presentswith diabetes mellitus.
Renal involvement in tuberoussclerosis
J COOK*, R F MUETLLER*, K OLIVER*,J SAMPSONt
*Department of Clinical Genetics, St JamesHospital, Leeds; tInstitute of Medical Genetics,Cardiff.
Renal disease is one of the commonest ma-
nifestations of tuberous sclerosis and yet verylittle is known about the incidence or naturalhistory. Between the centres of Leeds andCardiff we have looked at the notes of 106patients with TSC who have had renal in-vestigations. Three types ofrenal involvementhave been reported in TSC: angio-myolipomas, renal cysts, and malignantrenal tumours. The incidence of renal in-volvement was 60% with the majority of le-sions being angiomyolipomas. The incidenceof angiomyolipomas increased pro-
portionately with age. This is compatible witha "second hit" hypothesis being involved inthe aetiology of angiomyolipomas. The in-cidence of renal cysts was 29% and did notalter with age indicating a different aetiologyto angiomyolipomas. Renal cysts were thecommonest lesion in the under 5s. The in-cidence ofrenal carcinoma was 2-2% which ismuch greater than the population incidence.
Ascertainment of myotonicdystrophy through cataract byselective screening
A KIDD, P TURNPENNY, K KELLY,C CLARK, W CHURCH,C HUTCHINSON, J DEAN, N HAITES
Departments of Medical Genetics andOphthalmology, Foresterhill Hospital,Aberdeen.
Myotonic dystrophy is almost always the re-
sult of the expansion of an unstable (CTG)nrepeat. The mutation can be detected directly.Affected patients with cataracts often haveminimal other signs of the disorder, but allare at risk of life threatening complications.We have studied the efficacy of detecting newfamilies with myotonic dystrophy by se-
lectively screening cataract patients. Selectioncriteria were: age under 60 with no obviousprecipitating factor (except NIDDM);patients ofany age with other signs suggestiveof myotonic dystrophy detected by the oph-thalmologist. One hundred patients were
tested prospectively; 18 others under 55 were
screened retrospectively. All patients were
counselled by a geneticist before testing. The
patients' DNA was analysed using GB2.6/EcoRI, KB1.4/Bgl and sometimes PCR. Fivepatients have been found to have a mutationout of the 96 results available so far. A sixthpatient from Nigeria also has an expansionwith KBA.4/Bgl, but is being investigated torule out a polymorphism known to occur inthe southern black African population.
Prenatal diagnosis ofX linkedhydrocephalus
M JOUET*, C GARRETTt, R LAXOVA4,S KENWRICK*
*Department of Medicine, CambridgeUniversity, Cambridge; tClinical ResearchCentre, Northwick Park Hospital, Harrow,Middlesex; tClinical Genetics Center, Madison,Wisconsin, USA.
X linked hydrocephalus accounts for only asmall percentage of congenital hydrocephaluscases, and wide variation in severity and clin-ical signs makes counselling of families withsporadic hydrocephalic males difficult. Wehave shown that X linked hydrocephalus iscaused by mutations in the gene for neuralcell adhesion molecule LI. To facilitate coun-selling we have screened two sporadic casesof severe hydrocephalus for LI mutations.In both, new mutations were detected thatwould result in truncation of the Li proteinand eliminate its expression on neuronal ax-ons. In one case, the mother was found tocarry the mutation and a CVS sample wasused to exclude hydrocephalus in a secondpregnancy. In the second case, the motherdid not carry the mutation in somatic cells,but CVS testing was carried out for a secondpregnancy in order to account for potentialgonadal mosaicism. Mutation detection inthis case was particularly valuable as the sec-ond boy carries the same Stl4 allele as hisaffected brother and may have been ter-minated on genetic grounds. These resultsshow the value of mutation screening in casesofsporadic male hydrocephalus and representthe first cases of prenatal exclusion based onLI mutations.
Fragile X syndrome is lesscommon than previouslyestimated
J E MORTON, P M RINDL,S BULLOCK, S BUNDEY, T WEBB
Clinical Genetics Unit, Birmingham MaternityHospital, Birmingham B15 2TG.
In 1986, a population based study of schoolchildren in Coventry found the prevalence offragile X syndrome in school children agedbetween 11 and 16 to be 1/952. This figurewas revised to 1/1038 after the identificationofthe "common" fragile site in 1990. In 1991,the FMR1 gene was cloned. The Coventryschool children have been reinvestigated withmolecular analysis. Fragile X syndrome wasexcluded in a 13 further study children re-sulting in a drop in the estimated frequencyto 1/2197. The clinical features of20 childrenin the West Midlands still thought to havefragile X syndrome have been re-evaluated.IQ lay between 35 and 70 in 12/14 affected
persons with only 2/14 having IQs below 35and none less than 20. In accordance withprevious findings, the children tended to havelarge heads (15/18 >50th centile), macro-orchidism (13/14 >50th centile), and largeor protruberant ears (12/17). None of thechildren had neurological abnormalities onexamination and only 1/20 had epilepsy. Con-clusions are: (1) fragile X syndrome is lesscommon than previously estimated. (2) Fea-tures useful in suggesting the diagnosis offragile X syndrome in a boy with learningdifficulties are testicular volume >50th cent-ile, head circumference >50th centile, largeor prominent ears, and IQ >35.
Hemizygosity at the elastin locus(7q11) in Williams syndrome: aclinical tool?
N W PERIAM*, R PUGSLEY*,O QUARRELLt, E MAHER4,R C TREMBATH*
*Departments of Genetics and Medicine,University of Leicester, tDepartment of HumanGenetics, Sheffield; tDepartment of ClinicalGenetics, Addenbrooke' Hospital, Cambridge.
Williams syndrome is a developmental dis-order characterised by facial dysmorphism,mental retardation, and congenital heart dis-ease. Implication ofthe elastin gene in familialsupravalvular aortic stenosis led to the iden-tification of patients with Williams syndromewho are hemizygous at this locus. Haplo-insufficiency may be assessed by FISH ana-lysis using cosmids from the deleted region.However, we wished to assess hemizygosityat the elastin locus in relation to clinicalvariability in Williams syndrome and haveinvestigated the information content ofelastingene polymorphisms to determine parentalorigin of 7q11.23 deletions. We studied acohort of 14 sporadic patients with the diag-nosis of Williams syndrome. Of six patientsand their parents analysed to date five areinformative for an intragenic elastin genedinucleotide repeat polymorphism. Hemizy-gosity was identified in four patients, ofwhichonly one was paternal in origin.
succeeding generations in preference to oth-ers; preferential transmission of large CTGalleles may account for their continued ex-istence in the gene pool. There is evidencethat a CTG allele with > 19 repeats maygradually increase in repeat number overmany generations until it is sufficiently largeto cause a DM phenotype. We report a studyof 495 transmissions from parents, het-erozygous for CTG repeat numbers withinthe normal range (5-30), to their children.Alleles were simply classified as large or smallrelative to the other allele in a person; 126/242 paternal transmissions (52-1%, p=0 -52)and 146/253 maternal transmissions (57 7%,p=0-014) were of the larger allele. This isstrong evidence that females preferentiallytransmit the larger DMPK allele, contrary toa previous report ofmeiotic drive in the male.
Use of linked markers in breast/ovarian cancer families
A SCHOFIELD, B MILNER, L ALLAN,N HAITES
Department ofMedical Genetics, University ofAberdeen.
The majority of families with breast/ovariancancer predisposition have been found to belinked to the BRCA1 locus on chromosome17q. In a large family with a lod score ofgreater than 1-5 to tightly linked flankingmarkers, haplotype analysis and allele lossstudies on familial tumours have allowed us todefine the at risk haplotype. As with predictivetesting for Huntington's disease, such testingproduces unexpected results including newmutations at microsatellite loci, non-pa-ternity, and complex recombinant eventswhich preclude reliable predictive results forcertain persons. While prophylactic pelvicclearance seems to be an acceptable form ofpreventative surgery for the majority of highrisk women, only a few are willing to considerprophylactic mastectomies. However, thosewho do take this option appear to benefitpsychologically and to find the cosmetic resultof reconstruction to be very satisfactory. Itwill be important to have established clear cutprotocols for dealing with the psychologicalimpact of predictive testing and managementoptions.
POSTERS
Evidence for meiotic drive at themyotonic dystrophy locus
A M SHAW, R A BARNETSON,M F PHILLIPS, P S HARPER,H G HARLEY
Institute of Medical Genetics, UWCM, HeathPark, Cardiff.
The molecular basis of myotonic dystrophy(DM) is an expanded CTG repeat within the3' untranslated region of a putative serine-threonine protein kinase gene (DMPK). An-ticipation, increasing disease severity, andearlier age of onset in successive generations,and the fact that the origin of the disease hasbeen attributed to a founder chromosomesuggests that, in time, DM should die out.Meiotic drive has been described as a wayin which certain alleles are transmitted to
Cranial desmoid tumour as thefirst presentation of familialadenomatous polyposis (FAP)and the demonstration of the roleof the APC gene in thepathogenesis of this tumour
D DE SILVA*, D STEVENSON*,E GRAYt, J DEAN*, M DUNLOP4t,N HAITES*
*Department ofMedical Genetics, Aberdeen;tDepartment of Pathology, Aberdeen; tMRCHuman Genetics Unit, Edinburgh.
A 2 year old girl presenting with a rapidlyenlarging desmoid tumour of the foreheadhas a family history of familial adenomatouspolyposis (FAP) and has been shown to haveinherited a germline mutation of the APC
gene from her affected father. Cytogeneticanalysis of the tumour has shown partial lossof the long arm of chromosome 5. Thesefindings suggest that the loss of APC genefunction is implicated in the pathogenesis ofdesmoid tumours and further support pre-vious cytogenetic findings in desmoid tu-mours.
Experience of familial ovariancancer in Aberdeen
H GREGORY, J SEMPER, D DE SILVA,N E HAITES
Department ofMedical Genetics, MedicalSchool, Foresterhill, Aberdeen AB9 2ZD.
Since 1990, 968 patients have been referredfor genetic counselling in Aberdeen in con-nection with their family history of cancer.On our Family Cancer Register, the majorityof patients have a family history of breastcancer, bowel cancer, or other family cancersyndromes. Sixty-six patients from 53 ped-igrees have a family history including ovariancancer. In 15 pedigrees, only ovarian cancerwas seen, in 22 pedigrees both breast andovarian cancer was seen, and in the remaining16 pedigrees ovarian and other cancers wereseen. Thirty-one pedigrees are consistent withautosomal dominant inheritance. Of these,only three pedigrees are thought to be suitablefor linkage studies. Therefore for the majorityof persons at high risk of ovarian cancer inour population, prediction of risk by linkagewill not be possible. Should direct mutationtesting prove feasible, persons are likely toseek testing and will require further pretestcounselling, with major resource im-plications.
D R DAVIES*, J G ARMSTRONG*,S P GUY*, I FRAYLING*, T CLANCY*,T W WARNESt, D G R EVANS*
*Department ofMedical Genetics, St. Mary'Hospital, Manchester; tDepartment ofGastroenterology, Manchester Royal Infirmary.
We have used single strand conformationpolymorphism (SSCP) analysis and the pro-tein truncation test (PTT) to screen for mut-ations in the APC gene in a series of 44unrelated persons with FAP on the North-West Regional Polyposis Register. The site ofmutation within the APC gene was correlatedwith the phenotype. Mutations near the 5'end of the gene were shown to be associatedwith a relatively mild phenotype; however,there was marked intrafamily heterogeneity.Conversely, a common mutation at codon1309 was associated with a severe phenotype.Congenital hypertrophy ofthe retinal pigmentepithelium (CHRPE) occurred only when themutation was distal to exon 9. There was noclear correlation between the location of themutation and radio-opaque jaw lesions,dental abnormalities, desmoid disease, or epi-dermoid cysts. Our study shows that thephenotypic heterogeneity seen in FAP is re-flected to some extent at the molecular level.
hMSH2 screening is continuing, and onemissense mutation in a highly conservedcodon 322 in exon 6 has been identified.
G A FEWS, E V DAVISON, M GILES,C FELIX, L J WILLIAMSY WALLIS*, D MORTONt,
C McKEOWNt, J MORTON:,F MACDONALD*
*DNA Laboratory, Birmingham HeartlandsHospital; tDepartment of Surgery, QueenElizabeth Hospital; tDepartment of ClinicalGenetics, Birmingham Maternity Hospital,Birmingham.
We have investigated phenotype-genotypecorrelations for extracolonic manifestations(ECM) of FAP in 125 patients. We haveshown congenital hypertrophy of the retinalpigment epithelium (CHRPE) to be the onlyECM where expression is directly controlledby the position ofAPC gene mutation. Of 54FAP kindreds with a known CHRPE status,29 have a characterised constitutional APCgene mutation. Nine of these patients pos-
sessed mutations before exon 9 and did notexpress CHRPE, whereas the other 20patients with mutations after this exon allexpressed the CHRPE phenotype. The ex-
pression of CHRPE lesions therefore has im-portant diagnostic implications with respectto facilitating the detection of constitutionalmutations.
Molecular genetic analysis ofhereditary non-polyposiscolorectal cancer syndrome
N J FROGGATT*t, D J KOCHt,R DAVIESj, D G R EVANS:,R PARSONS§, B VOGELSTEIN§,S V HODGSONII, B A J PONDER¶,D E BARTONt, E R MAHER*
*Cambridge University Department ofPathology, Cambridge; tMolecular GeneticsLaboratory, Addenbrooke' Hospital,Cambridge; tFMedical Genetics, St Mary'sHospital, Manchester; §Oncology Center, JtohnsHopkins, Baltimore, USA; IlMedical Genetics,Guy's Hospital, London; ¶CRC HumanCancer Genetics Group, Cambridge UniversityDepartment of Pathology.
HNPCC is estimated to account for 5 to10% of colorectal cancer. Recently genes forHNPCC have been mapped to chromosomes2p and 3p and candidate genes (hMSH2and hMLHl) have been identified. We havepreviously reported genetic linkage studies on14 HNPCC families with markers close tohMSH2 and hMLHl. Although significantlinkage was obtained with D2S123 (Zmax=3-77 at 0=0 0) and locus heterogeneity was
confirmed (linkage to hMSH2 and hMLHlwas excluded in six and five families re-
spectively), no individual family gave a lodscore of >3 with any marker, and only a
minority of our HNPCC families have beensuitable for genetic linkage analysis. We there-fore screened affected persons from 37 un-
related kindreds for mutations in hMSH2and exons 3 and 4 of the APC gene usingexon specific primers and SSCP analysis. Noabnormalities were found in the APC exons
suggesting that mutations in these APC 5'exons are not a common cause of HNPCC.
FAP patients with completedeletions of the APC gene
T HAMZEHLOEI, S WEST, A CURTIS,P CHAPMAN, J BURN
Northern Region Genetics Service and theDepartment ofHuman Genetics (University ofNewcastle upon Tyne), 19120 Claremont Place,Newcastle upon Tyne NE2 4AA.
We have identified four persons from a totalof 48 unrelated patients with familial ad-enomatous polyposis who have a germlinedeletion of one complete copy of the APCgene. The deletions were detected because ofthe apparent non-inheritance from one parentof alleles for intragenic polymorphic markers.One deletion was de novo and three were
inherited from an affected parent. Mutationsof this nature are undetectable by the com-
monly used mutation screening proceduresthat rely on conformational mispairing or
instability between "normal" and "mutant"DNA molecules. We have characterised theextents of each deletion using pulse field gelelectrophoresis and the many polymorphicmarkers surrounding the APC gene.
Mutation detection in the APCgene by heteroduplex analysisfollowing RT PCR
T HAMZEHLOEI, S WEST, A CURTIS
Northern Region Genetics Service and theDepartment ofHuman Genetics (University ofNewcastle upon Tyne), 19120 Claremont Place,Newcastle upon Tyne NE2 4AA.
The APC gene consists of 15 exons of whichthe final exon is 6-5 kb. Exon 15 can bedivided into segments of suitable length forPCR amplification and subsequent mutationanalysis. The first 14 exons, however, consistof a total coding sequence of just 2 kb. Ratherthan amplify and analyse these 14 exons in-dividually, we have designed PCR primerswhich divide this part of the APC codingsequence into six overlapping segments ofaverage size 400 bp. mRNA was extractedfrom the lymphocytes or lymphoblastoid celllines of patients from 15 families on theNorthern Region Polyposis register, cDNAwas prepared, and PCR amplification per-formed using the overlapping primer sets toexons 1-14. Each PCR product was analysedfor the presence ofmutations by heteroduplexanalysis on Hydrolink gels. Despite the pos-sible requirement for fresh blood samplesfrom patients, we have found RT PCR fol-lowed by heteroduplex analysis and sub-sequent sequencing to be an efficient methodfor the identification of mutations in exons1-14 of the APC gene.
Regional Cytogenetics Unit, BirminghamWomen's Health Care NHS Trust(Birmingham Maternity Hospital), Edgbaston,Birmingham B15 2TG.
Following an abnormal ultrasound finding ofhydrocephalus and bright gut, a CVS samplewas taken at 20 weeks' gestation. Extensiveanalysis of both direct and long term culturesshowed an abnormal 47,XY, + 16 karyotype.Fetal blood sampling showed a 46,XY/47,XY, +mar karyotype. In situ hybridisationshowed the marker to be derived from chro-mosome 16. Analysis of amniotic fluid takenat the same time as the fetal blood sampleshowed a karyotype of 46,XYl47,XY, +16/47,XY, + mar. Molecular investigations todate have neither confirmed nor excludeddisomy. We propose this is an example of thegenetic mechanism of trisomic zygote rescue
by formation of a marker chromosome.
Parental attitudes to the prenataldiagnosis and treatment ofcongenital adrenal hyperplasia
S A SMITH*, I D YOUNGt,D I JOHNSTON*
*Department of Child Health, UniversityHospital, Nottingham; tDepartment of ClinicalGenetics, City Hospital, Nottingham.
This study was designed to establish parentalattitudes to the prenatal diagnosis and sub-sequent antenatal treatment of congenitaladrenal hyperplasia aimed at preventingvirilisation in affected girls. Thirty-four famil-ies (with 39 affected children) are knownlocally and these families were contacted andinterviewed in person where possible (22families) or by telephone and postal ques-tionnaire. Responses were as follows. Twenty-five questionnaires were completed by one or
both parents jointly. Twenty-two respondentswere still in a relationship with the otherparent of the affected child. The youngestchild was the only one affected in 10 familiesand in a total of 15 families the diagnosis wasfelt to have limited the eventual family size(in a further three families the parents hadopposing views on this). Twenty-three re-
spondents would choose (or have chosen inthree cases) to undergo prenatal diagnosisand treatment aiming to prevent virilisation,one would not, and one set of parents dis-agreed. Twelve respondents would considerterminating an affected fetus. We concludethat parents of a child with congenital adrenalhyperplasia should be alerted to the optionand possible benefits of prenatal treatment.
A prenatal diagnosis involvingtrisomy 16 mosaicism
only, indicating a sporadic form of disease.Five of the remaining 11 samples possessedan altered codon 918.
H C SOLTAN*, R F ARMSTRONGt,A D BOCKINGt
*Regional Medical Genetics Centre,
tDepartment of Pathology, Victoria Hospital;tDepartment of Obstetrics and Gynecology, St
J7oseph's Health Centre, University of WesternOntario, London, Ontario, Canada.
The major anatomical abnormality was ex-
tensive thickening of the scalp (1 cm) withrugose arrangement of overlying skin espe-cially in the occipital region but extendingover the chin. Subcutaneum was composedof loose connective tissue. The neuro-
pathological examination showed a normalbrain and spinal cord on both gross andmicroscopic examination. Other dysmorphicfeatures included depressed nasal bridge, cleftpalate, low set ears, receding jaw, and fingersoverriding the second digit. Fluorescent andgiemsa banded karyotype from amniotic fluidcells was normnal. This was the sixth preg-nancy of a 34 year old woman proceedingwithout complication until suspected non-
immune hydrops and cystic hygroma were
identified at 14 weeks. The previous five preg-nancies included two spontaneous and twoinduced abortions and another which in-volved rupture of membranes at 21 weekswith delivery of a stillborn fetus and pa-thologically confirmed placentitis. There is norelevant family history and no consanguinity.
Molecular analysis of MEN-2Aand sporadic MTC tumours inthe north east of Scotland
D U BATY, A ROSS, D GOUDIE,M BOXER
Human Genetics, Department of Pathology,Ninewells Hospital and Medical School,Dundee, Scodand.
Mutations within the RET oncogene on chro-mosome 1Oql 1.2 are responsible for the mul-tiple endocrine neoplasias MEN-2A andMEN-2B, as well as familial medullary thy-roid carcinoma (FMTC). MEN-2A is char-acterised by medullary thyroid carcinoma(MTC), phaeochromocytomas, and para-thyroid hyperplasia. Approximately 80% ofMEN-2A cases result from point mutationswhich alter codon Cys634 in exon 11. Mu-tations in one offour conserved Cys codons inexon 10 account for another 17%. Similarly,93% of MEN-2B samples and 38% of spor-adic MTC tumours involve a point mutationat codon 918 in exon 16. The exon 11 and16 mutations alter restriction enzyme sitesallowing detection by digestion of PCR amp-lified samples. Identification of exon 10 mu-tations requires direct sequence analysis.Fifteen Scottish families were analysed forMEN-2A mutations. Four carried the com-mon Cys634 mutation, while five of the re-
maining families were sequenced, showing analtered Cys620 codon in only one. Fourteenparaffin embedded sporadic MTC tumourswere also screened. Two possessed Cys634mutations in both tumour and somatic tissueidentifying these as familial cases, while onecarried the same mutation in tumour tissue
One hundred and ninety fourprenatal diagnoses for Duchennemuscular dystrophy
R C MOUNTFORD, A P READ,R G EL ES
Regional Molecular Genetics Laboratory,Department ofMedical Genetics, St MaryIHospital, Manchester.
We present the results of 194 prenatal diag-noses for Duchenne muscular dystrophy car-
ried out between April 1985 and June 1994.Sixty-two percent (120/194) of the cases weremale. This is biased by the fact that somecentres referred only male samples for DNAanalysis. Twelve percent of samples were forfetal sexing only. All these samples were re-
ferred in the first five years ofthe programme.Of the remaining male samples 35% werefound to be at high risk after DNA analysis.The remainder were assigned a low risk. Therisks in this category varied considerably re-
flecting the wide variety of individual cir-cumstances and the changing nature ofmolecular tests available. Currently all pre-natal tests are PCR based, using multiplexdeletion screening or intragenic and flankingmarkers. Results are reported within threeworking days after receipt of the sample.
Increased infections associatedwith immunologicalabnormalities in Noonansyndrome
M SHARLAND, M ELSAWI,E KAMINSKI, M A PATTON,E G DAVIES
Department of Child Health, St George'Hospital, London.
A recent study reported an apparent excess ofinfections in patients with Noonan syndrome(NS). A structured infection questionnairewas administered prospectively to a un-selected cohort of 30 children with NS (14M:16F, mean age 11 9 years), and control chil-dren (9M:15F, mean age 6-3 years). Bloodwas taken for immune function tests. A his-tory of recurrent otitis media (60% v 0%),lower respiratory tract infections (33% v 0%),and serious bacterial infections (3% v 0%),were all commoner in the NS group thancontrols. Only in the NS group were de-ficiencies of IgG (three cases), IgM (onecase), and IgA (nine cases) noted. Ab-normally low levels of IgA combined withIgG2 were seen in 20% of the NS groupcompared with 7% of controls. Levels ofIgGland 3 were not significantly different in thetwo groups. This type of immunological ab-normality is usually associated with frequentcapsulated bacterial infections of the re-
spiratory tract. The deficiencies are similar tothose previously reported in other cardiofacialsyndromes. It is important that geneticistsbring this increased susceptibility to infectionsto the attention ofboth parents and clinicians.
Comparison of genotype andphenotype in Northern IrelandPKU patients homozygous forthe mutations R408W, 165T, andR408Q
F J STEWART*, J ZSCHOCKE*,D CARSONt, C GRAHAM*, A GREER
*Department of Medical Genetics, Belfast CityHospital; tDepartment of Child Health,Queen' University of Belfast.
Northern Ireland has a high prevalence ofPKU at 1:4000 live births compared with1:11 000 live births in the UK as a whole. Weare genotyping all PKU patients and at-tempting to correlate this with their clinicalphenotype. In particular, we have looked atintelligence and phenylalanine intake. Phenyl-alanine intake has been identified as strict,<500 mg/day; relaxed, 500-1000 mg/day;free, >1000 mg/day. The median value ofthe monthly Guthrie test for each year wasmeasured. This has been described as highor satisfactory to take account ofthe changingguidelines over the years. Six patients werehomozygous for R408W. All required verystrict dietary restriction. Despite this, somepatients were unable to maintain Guthrieswithin the recommended guidelines. Twopatients were homozygous for 165T and onewas homozygous for R408Q. These threepatients were able to maintain acceptablephenylalanine levels on a more relaxed diet.Almost all patients had an IQ within thenormal range. Although these are small num-bers, there does appear to be a variation ofphenotype with genotype in PKU. Changesin the recommended phenylalanine levelsover the years and variation in methods ofIQ testing make more detailed comparisonsbetween patients difficult. It is also verydifficult to assess dietary compliance.
Phenotype-genotype correlationsin Huntington's disease patientsin Northern Ireland and theRepublic of Ireland
P J MORRISON, C A GRAHAM,N C NEVIN
Department ofMedical Genetics, Belfast CityHospital, Belfast BT9 7AB.
Triplet repeat analysis was carried out onDNA from affected Huntington's disease(HD) patients obtained after virtually totalascertainment in Northern Ireland and partialascertainment in the Republic of Ireland.Over 120 patients were sampled, includingseven juvenile HD patients (JHD 10-20years), four childhood onset patients(COHD, 0-10 years), and 12 patients withan onset over 60 years. One parent-child pairhad anticipation from a clinically asympto-matic 77 year old with an "intermediate"allele of 39, to a 35 year old clinically affectedoffspring with a repeat of 48. A threegeneration parent-juvenile-childhood onsettransmission is described with exponentialexpansion in repeat size and phenotype se-verity. Seven patients had psychiatric pre-sentations including three with paranoidpsychosis. One patient presented with inertiaand one with isolated tongue dystonia. AllJHD patients had repeat sizes greater than 50.
Three COHD patients with grandmaternal-paternal-child transmission had repeats in themid 40s. Phenotype and age at onset oftenhave no direct relation to trinucleotide repeatsize. The only consistent feature was thatrepeat size >55 was specific for JHD, butrepeat size <55 did not exclude JHD orCOHD, or correlate with later onset of thedisease.
Direct mutation analysis forHuntington's disease (HD): theOxford experience
E M ROSSER*, G NORBURYtR GLEW*, L JONES*, S M HUSON*
*Department of Clinical Genetics andtMedical Genetics Laboratory, OxfordRadcliffe Hospital, The Churchill, OxfordOX3 7LJ.
We report our first year's experience of HDmutation analysis. We have had no majortechnical difficulties and all cases have hadunequivocal results; in one case of juvenileHD the expanded allele was only detectableby Southern blotting. Our HD programmehas seen a five-fold increase in referrals. Thesehave fallen into the following categories: 34presymptomatic tests (50% risk); two pre-symptomatic tests (25% risk, at risk parentdead); four re-tests; four prenatal diagnoses(two exclusion and two mutation tests), and21 diagnostic tests. A further three personsat 25% risk, where their at risk parent wasunkeen to be tested, have also been referred.Counselling is continuing about how to re-solve the issues with regard to the at riskparent being involved. The mutation test ontwo dead affected persons with a postmortemdiagnosis of HD was negative; in one, nec-ropsy review has led to a revised diagnosis andthe outcome of the other review is awaited.Despite the ease of mutation testing from thetechnical viewpoint, the counselling issuesremain complex and have required a sig-nificant increase in clinical genetic and nursecounsellor time.
"Dysmorphic unknowns": doesfollow up lead to diagnosis?
M PENMAN SPLITT, J GOODSHIP
Northern Regional Genetics Service, 19120Claremont Place, Newcasde upon Tyne.
Many patients referred for syndrome diag-nosis remain "unknown" after the initialwork up. The aim of this study was to findout how often long term follow up of thisgroup leads to a diagnosis. All photographstaken in the four year period 1990 to 1993labelled "diagnosis unknown" were iden-tified. Of these, 168 patients were new re-ferrals. Of this total, 69 (41%) have beendischarged, four (2%) have died, and in 14(8%) cases follow up was planned but thefamilies failed to attend. Follow up has ledto five new diagnoses (3%): one Rubenstein-Taybi, one Kabuki syndrome (twins), onechromosome 5q deletion, two Smith-Lemli-Opitz (sibs), and one 22ql deletion. In oneother case the child was discharged at thefirst visit and a diagnosis was subsequentlymade on re-referral three years later. In thesame period at least five diagnoses of dys-
morphic unknowns were made by chancereview of photographs including twoShprintzen syndrome, two a thalassaemiamental retardation, and one Pallister-Killian.
Alagille syndrome: is the facespecific?
F V ELMSLIE*, A J VIVIAN*,A BAKERt, A P MOWATt,R M WINTER*
*Mothercare Department of Clinical Geneticsand Fetal Medicine, Institute of Child Health,London; tThe Children's Liver Unit, King'sCollege Hospital, London.
Alagille syndrome (AGS) is one of the majorcauses of cholestatic liver disease in child-hood. A characteristic face has been describedas part of the syndrome since the originaldescriptions. The features are said to be abroad, prominent forehead, deep set eyes, abroad nasal bridge, and a pointed chin. Aprevious study has cast doubt on the spe-cificity of the face. We aimed to determinewhether the face is specific or a reflection ofcholestasis in infancy. Seventy clinical gen-eticists of all grades were shown 20 slides ofchildren and adults with AGS and other formsof cholestatic liver disease. They were askedto identify which had AGS. Overall, 67% ofresponses were correct with a range of 39%to 90%. There was no significant differencebetween responses obtained from consultantsor junior staff. Certain children were con-sistently correctly identified as having AGS,and one child who had a 1 antitrypsin de-ficiency was consistently incorrectly identifiedas having AGS. We conclude that a char-acteristic face exists but that it is not presentuniversally. Caution must be exercised in pla-cing too much emphasis on the presence ofthe facial features as a diagnostic tool inAlagille syndrome.
Fetal valproate syndrome: followup of infants diagnosed at birth
C M BREWER, J L TOLMIE
Duncan Guthrie Institute of Medical Genetics,Royal Hospital for Sick Children, Glasgow G38S_J.The use of sodium valproate in pregnancyhas been associated with a characteristic dys-morphic syndrome, developmental delay, car-diac anomalies, and neural tube defects. Wesought to ascertain the long term progress offive infants in whom the diagnosis of the fetalvalproate syndrome (FVS) was made in thenewborn period. All the infants were bornto epileptic mothers, three of whom wereprescribed valproate monotherapy in preg-nancy. At the time of reassessment the agesof the children were between 13 months and9 years. Each child had some facial char-acteristics previously described as part of theFVS: a short, upturned nose, shallow phil-trum, and infraorbital creases were the mostconsistent findings. In three cases an ad-ditional feature was nasolacrimal duct ab-normality including bilateral fistulae. Fourwere found to have abnormal developmentranging from mild speech delay to globalretardation in two children. This small study
of infants diagnosed in the newborn periodconfirms that the prognosis for psychomotordevelopment is variable with some childrenhaving severe learning difficulties.
PROMM: proximal myotonicmyopathy, a new myotonicdisorder
Myotonic dystrophy (DM), myotonia conge-nita (MC), and paramyotonia congenita (PC)are well recognised autosomal dominantlyinherited myotonic disorders. The ion chan-nel disorders, MC and PC, are caused bymutations in the CLCN1 and SCN4A gene,respectively. Recently we described threefamilies with an autosomal dominant as yetunclassified myotonic disorder, which is char-acterised by late onset myotonia, mild prox-imal leg weakness, capsular cataract, and slowprogression of symptoms. Linkage to theCLCN1, the SCN4A, and the DM1 genewas excluded. We extended our study to anadditional 10 families, in which affected sub-jects exhibit an identical phenotype. Our clin-ically thoroughly investigated patients led usto believe that PROMM was indeed an un-recognised myotonic disorder until now,which may have been confused in the pastwith the more common disorder DM. In allaffected probands we excluded an am-plification ofthe CTG repeat sequence typicalof the molecular defect in DM. Further in-vestigations from different populations areneeded to establish the natural history of thedisease PROMM and the molecular defect.
Monozygotic twins discordant forhypomelanosis of Ito
S M PRICE, A HOLTON,R C TREMBATH
Department of Clinical Genetics, LeicesterRoyal Infirmary, Leicester LEI 5WW
Hypomelanosis of Ito is characterised by pig-mentary disturbance, sometimes associatedwith skeletal abnormality, developmentaldelay, and seizures. The inheritance of thisdisorder is unclear. Familial cases have beendescribed, but most of the evidence supportsmosaicism of a numerical or structural chro-mosome abnormality, often only dem-onstrable on fibroblast culture, as theunderlying cause. We report monozygotictwins, proven by hypervariable DNA analysis,discordant for hypomelanosis of Ito as furtherevidence that a postzygotic mutational eventis the most likely mechanism leading to thisclinical pattern. The affected female twin wasreferred aged 2 years because ofdelayed mile-stones and unsteadiness. She has marked fa-cial asymmetry and streaky depigmentationinvolving the inner aspect ofboth lower limbs
and left arm following Blaschko's lines. In-vestigation has shown her to have an ab-normal MRI scan. Her twin is normal.
Prenatal exclusion test of a Wolf-Hirschhorn family using PCRamplification with Huntington'sdisease primers
A B DODGE, L GAUNT, J DORE,V GILLANDERS
Medical Genetics, St Mary's Hospital,Manchester M13 0JH.
We have used the Huntington's disease (HD)primers to carry out a prenatal exclusion teston a family with a history ofWolf-Hirschhomsyndrome. The father and grandmother are
carriers of a balanced 4;12 translocation, ka-ryotype t(4;12) (pl6.3;pI3.3). The break-point occurs between exons 40 and 41 of theHD gene (67 exons) and is centromeric to theHD mutation site. The fetus's chromosomesappeared normal on cytogenetic analysis buttrisomy could not be excluded. DNA analysiswith the HD primers and silver stainingshowed two alleles. The fetus did not inheritthe der(12) chromosome allele and thereforeis considered unaffected. The balanced trans-location is not associated with any detectableabnormal HD phenotype either in the fatheror grandmother. Thus heterozygous dis-ruption of the HD gene does not have cata-strophic consequences for development.Furthermore it is unlikely that the expandedCAG in HD acts by simply inactivating theallele containing it.
Marfan's syndrome: aorticdistensibility (measured by MRI)and family phenotype
J C S DEAN*, J N ADAMSt,M BROOKES*, J R GRAY§,T W REDPATHII, F W SMITHII,S WALTONt, R J TRENTt
Departments of Medical Genetics*,Cardiologyt, Radiology3, and MagneticResonance Imagingll, Aberdeen Royal HospitalsNHS Trust; §Department of Pathology,Ninewells Hospital NHS Trust, Dundee.
Premature mortality in Marfan syndrome isusually the result of rupture of the ascendingaorta, and risk of rupture correlates with in-creasing aortic diameter. Reduced aortic dis-tensibility has been shown in Marfansyndrome, reflecting abnormal aortic wallelasticity. We studied aortic root size anddistensibility by gated magnetic resonance
imaging and echocardiography in 12 adultMarfan patients and 12 age matched controls.The patients were drawn from six families, ofwhom halfhad ocular features. The ascendingaortic distensibility was lower in Marfanpatients than in controls, but there was lessdecline in distensibility with age. There wasa trend towards lower age adjusted dis-tensibilities in persons whose affected rel-atives had died or suffered aortic rupture atyounger ages. Gated MRI was well tolerated,and showed better reproducibility than echo-cardiography. The degree of reduction ofaortic distensibility seen in Marfan syndromemay correlate with severity of cardiovascular
disease within families, and may be a usefulparameter in the assessment of risk of aorticrupture.
Ll mutation screening:identification of a non-
conservative missense mutationthat does not segregate withhydrocephalus in one family
J MACFARLANE, M JOUET,ALAN FRYER*, DIAN DONNAIt,ROBIN WINTER4, R CHARLTON§,J HURST§, S KENWRICK
Department of Medicine, CambridgeUniversity, Cambridge; *Regional ClinicalGenetics Service, Royal Liverpool Hospital,Liverpool; tRegional Genetics Service, StMary' Hospital, Manchester; tlnstitute ofChild Health, University of London, London;§Oxford Regional Genetics Service, ChurchillHospital, Oxford.
X linked hydrocephalus and MASA syn-
drome represent a clinical spectrum caused bymutations in the gene for neural cell adhesionmolecule Li and 10 independent mutationshave been reported. We have analysed a fur-ther six affected male cases using SSCP. Fournew mutations were detected including a Gto A transition in a family with three affectedboys in a single generation. Surprisingly, thismutation did not segregate with the diseasealthough it results in a non-conservative acidsubstitution (arginine to glutamine). Themutation was found not only in a hy-drocephalic boy but also in a healthy malesib. The other two affected brothers do notcarry the mutation, thus ruling out gonadalmosiacism. Segregation of flanking markerSt14 agrees with this result and the analysishas been reproduced using fresh bloodsamples. The mutation could not be detectedin 40 unrelated X chromosomes indicatingthat it is not a common polymorphism. The
results indicate that an LI mutation is notresponsible for hydrocephalus in this familywhich therefore represents a rare form ofeither autosomal recessive or X linked hydro-cephalus. Furthermore, certain non-con-
servative amino acid substitutions in the LIprotein are not pathological.
Dominantly inherited dilatedcardiomyopathy in a Somersetfamily
Dilated cardiomyopathy in adults (as distinctfrom the hypertrophic type, HOCM), is notoften thought of as genetic, but dominant,recessive, and X linked inheritance are alldescribed. We present a family conformingto dominant inheritance with seven affectedpersons from three generations, including onemale to male transmission. The diagnosis offamilial dilated cardiomyopathy was madeafter the sudden deaths of two brothers atages 38 and 43 years, and the symptomatic
presentation of a third brother at 40 years.On screening, their sister aged 44 years wasfound to be similarly affected. Their motherhad died aged 66 years from cardiac failurewith a grossly enlarged heart. Cardiologicalscreening offered to those at risk in the thirdgeneration showed a further two affected per-sons; a female aged 31 years with obviouscardiac enlargement and the 6 year old ofoneof the affected males whose echocardiogramshowed reduced cardiac contractility. In noneof these subjects was there any cardiac musclehypertrophy or outflow obstruction, and thedistinction from HOCM is clear.
Comparison of facial features ofDiGeorge syndrome (DGS) dueto deletion 10pl3-lOpter withDGS due to 22qll deletion
S LYNCH*, J BROWN*, I CROSS*,D MILLIGANt, J GOODSHIP*
*Department ofHuman Genetics, University ofNewcasde upon Tyne; tDepartment ofPaediatrics, Royal Victoria Infirmary,Newcastle upon Tyne.
DiGeorge syndrome (DGS) is a congenitalanomaly consisting of cardiac defects, hy-poplasia of the thymus and parathyroidglands, and dysmorphic facial features. Themajority ofDGS cases have a submicroscopicdeletion within chromosome 22ql 1. How-ever, DGS has also been described in childrenwith lOp deletions. We describe a further10p deletion case and suggest that the facialfeatures in children with DGS due to de-letions of lOp are different from those as-sociated with chromosome 22 deletions. Theproband was born at 39 weeks' gestation tounrelated white parents, birth weight 2580 g(1Oth centile). The main dysmorphic facialfeatures were a broad nasal bridge with veryshort palpebral fissures. Echocardiographyshowed a large subaortic VSD and overridingaorta. She had a low ionised calcium and lowparathyroid hormone level. T cell subsetsand PHA response were normal. Abdominalultrasound showed duplex kidneys. Her ka-ryotype was 46,XX,-10,der(10)t(3;10) (p23;p13)mat. The dysmorphic facial features inthis baby are strikingly similar to those ob-served in other cases of DGS resulting from10p deletion and distinct from the face seenin children with DiGeorge syndrome resultingfrom interstitial chromosome 22 deletions.
The Nijmegen breakagesyndrome is not an allelic formof ataxia telangiectasia
A J GREEN*, N BARNESt,C J BILLINGt, A M R TAYLOR§,J RW YATES*
*Department of Clinical Genetics,Addenbrooke' NHS Trust and Department ofPathology, University of Cambridge;tDepartment of Paediatrics and tDepartmentof Cytogenetics, Addenbrooke' NHS Trust;§Department of Cancer Studies, University ofBirmingham Medical School.
Nijmegen breakage syndrome is a rare auto-somal recessive condition characterised bysevere microcephaly, chromosome breakage,
and immunodeficiency. It has been proposedas an allelic form of ataxia telangiectasia,which shows identical chromosome breakage.An 18 month old girl was diagnosed withNijmegen breakage syndrome. She presentedwith severe microcephaly of prenatal onset,but normal developmental milestones. Therewas no history of excessive infection, and notelangiectasia or ataxia. Twenty percent ofherlymphocytes showed translocations involvingchromosomes 7p13, 7q35, 14ql1, and14q32, and her lymphocytes showed in-creased x ray induced chromosome damage.She was immunodeficient, with a low CD3count, but normal immunoglobulins. She hada normal AFP. Her parents' and elder sister'skaryotypes were normal. An affected brotherwas born shortly after her diagnosis. He hadmicrocephaly, similar lymphocyte chro-mosome breakage, with increased x ray sens-itivity, and a low CD3 count. Haplotypeanalysis was carried out in the family usingDNA markers spanning the ataxia te-langiectasia locus on chromosome 11 q. Theaffected male had a different haplotype fromhis affected sister across the region. Thus, inthis family, the Nijmegen breakage syndromeis not linked to the ataxia telangiectasia re-gion, and the two conditions are geneticallydistinct.
Training in clinical genetics inthe UK: experiences of trainees
R A NEWBURY-ECOB*, J A RAEBURNt
*Centre for Human Genetics, Sheffield;tInterdisciplinary Centre for Human Genetics,Nottingham.
A confidential questionnaire was completedby 10 clinical genetic trainees in 1992. Ques-tions covered qualifications and previous ex-perience, clinical teaching and laboratorytraining, opportunities for training in alliedsubjects and specialist clinics, training inmanagement and counselling, method ofsupervision and appraisal, and re-commendations for improving training. Sixtrainees were in full time and four in parttime training. All trainees held MRCP, twohad higher degrees, and six were completingan MD. Previous specialities were paediatrics(7) and adult medicine (3). Six trainees hadless than one hour of teaching per week andfour had none. Eight had laboratory trainingvarying from four days to six months. Onlythree trainees had any management trainingand only four had any teaching in counsellingskills. Half the trainees had no system ofreview/feedback. Most (9) had no knowledgeofhow clinical genetics training was organisedwithin their region. Recommendations fromtrainees for enhancing training included amore structured training with a core cur-riculum, formal training in counselling skillsand dysmorphology, and a system of ap-praisal.
Experience of maternal serumscreening for Down's syndromewith a high "cut off" risk
M BARROW*, D DUCKETT*,P GARRICK*, C STEWARTt
*Department of Clinical Genetics andChemical Pathology, Leicester Royal Infirmary;tDepartment of Obstetrics and Gynaecology,Leicester General Hospital.
Maternal serum screening for Down's syn-drome (using AFP and free beta hCG) wasintroduced in Leicestershire in 1992, using a"cut off" risk of 1 in 100 rather than themore common level of 1 in 200 or lower. Theother component of screening included theprovision of amniocentesis to all women overthe age of 35 years. During the first year,8061 patients (82-9% uptake) underwent bio-chemical screening. Six cases out of a totalof 15 (40%) of trisomy 21 were detected afteramniocentesis was performed on 256 highrisk subjects. Only two false negatives oc-curred and one normal pregnancy loss. Fourfurther cases of trisomy 21 were detected byamniocentesis for maternal age, giving anoverall detection of Down's syndrome pre-natally of 66% (10/15). This alternative ap-proach to screening appears to be effectiveand cost saving, another major benefit beingthat fewer women experience the anxiety ofa high risk result.
An evaluation of sonographicdiagnostic criteria for autosomaldominant polycystic kidneydisease 1 Antenatal video counselling for
maternal serum screening is aneffective method ofcommunication
D RAVINE*§, R N GIBSONt,R G WALKER4, L J SHEFFIELD*,P KINCAID-SMITHt, D M DANKS*
*Murdoch Institute, Royal Children's Hospital,Melbourne, Australia; tDepartment ofRadiology, University of Melbourne, RoyalMelbourne Hospital; tDepartment ofNephrology, Royal Melbourne Hospital.§Current address: Institute of Medical Genetics,University Hospital of Wales, Cardiff
Although ultrasound is commonly used forscreening subjects at risk for PKD 1, there hasbeen no evaluation of sonographic diagnosticcriteria. We used DNA linkage among sub-jects from 128 sibships within 18 PKD1families as the basis for an evaluation ofultrasound sensitivity. Positive and negativepredictive values were then calculated to allowassessment of different diagnostic cut offpoints in previously undiagnosed cases. Cur-rently used criteria (bilateral cysts with at leasttwo in one kidney) provided good sensitivity(88-5% at 15 to 29 years and 100% at 30years and above) but performance could beimproved by less stringent criteria in thoseaged 15 to 29 years and more stringent criteriain older family members, in whom simplerenal cysts are frequent. The presence of atleast two renal cysts (unilateral or bilateral)in persons at risk and younger than 30 yearsmay be regarded as sufficient to establish a
diagnosis; among those aged 30 to 59 years,the presence of at least two cysts in eachkidney may be required, and among thoseaged 60 years and above, at least four cystsin each kidney should be required.
What do general practitionersexpect from the clinical geneticservice?
J AFFLECK, J MOORE, B GIBBONS,A BINKS, P FARNDON
Clinical Genetics Unit, Birmingham MaternityHospital, Birmingham B15 2TG.
A postal questionnaire was sent to generalpractitioners in two Health Districts in theWest Midlands Region. Of 403 ques-tionnaires sent, 183 were returned (overallresponse rate 45%). The questionnaire ex-
plored four main areas: (1) Knowledge ofservices provided by the Clinical GeneticsUnit: 63% of GPs were uncertain whetherwe provided presymptomatic testing usingDNA probes. (2) Successful outcomes ofgen-etic counselling: 94% of GPs indicated thatgiving genetic risks was a successful outcome;43% of GPs mentioned avoidance of all ab-normal pregnancies. (3) Conditions referredto the Genetics Unit/conditions managed byGPs: interestingly 87% of GPs would notrefer a patient with a family history of cancer.
(4) Where GPs obtain information on ge-netics. Results and useful comments madeby the GPs led to the production of a user
friendly handbook. Further work will identifywhether the booklet influenced the pattern ofreferrals.
J BURN*, S FAIRGRIEVE*,A STEPHENSON*, P FRANKSt
*Division ofHuman Genetics, University ofNewcastle upon Tyne; tDepartment ofObstetrics, Ashington General Hospital,Ashington, Northumberland.
In a pilot study of population screening forDown's syndrome, 1884 questionnaires wereposted to participants in Northumberland(62% return) and 745 questionnaires weredistributed on the postnatal ward (81%return). Despite overwhelming support(97-8%) for the availability ofDown's screen-ing, 24% had not received a leaflet, 33-3%would have preferred more explanation, and40 7% more information. Twelve testedwomen denied having had the test. A 15minute counselling video written and de-livered by JB was produced and shown to allwomen attending the Ashington antenatalclinic. Uptake of screening remained at 80%but there was a significant improvement inopinion of the blood test. Among 285 womenwho had seen the video and a redesignedleaflet only 17-9% would have liked moreexplanation and 23-5% more information. Inthe overall study the number of women whohad a normal result on amniocentesis ("falsepositives") was 154 in the first cohort and120 in the second. Asked it they would havethe blood test again, the response wasaffirmative in 79 4% and 91-7% respectively,a significant increase (p = 0 0024). Effective
As a result ofanalysing over 600 cystic fibrosis(CF) patients from the north west ofEngland,we have so far identified 30 different mu-
tations of the CFTR gene. These account forover 93% of CF mutations in the north westof England. This information has been ap-plied in a number of ways: (1) "cascadescreening" of relatives of CF patients; (2)confirmation of CF in neonates where symp-toms of CF have occurred, for example, me-conium ileus; (3) testing of sperm and eggdonors; (4) assistance with diagnostic queries,since only 0 5% of those with definite CF are
negative for all 30 mutations.
Identification of cystic fibrosismutations in Asians and easternEuropeans
In over 600 CF chromosomes from patientsfrom the north west of England, we haveidentified 30 different mutations which to-gether account for over 93% of CF genesin this population. We have analysed CFchromosomes from 11 Pakastani, one Indian,and 20 Romanian CF patients for all of theCF mutations seen in the north west of Eng-land. Ofthese only AF508 has been identified,accounting for 7/22 (Pakistani), 0/2 (Indian),and 13/40 (Romanian) CF chromosomes,illustrating the low detection rate of CF mu-tations in these populations. Further analysisby direct DNA sequencing has shown thefollowing new mutations in Pakistani patients:1161delC (exon 7) and 296+12(T>C)(intron 2). A new mutation V520I (exon 10)has been identified in the Indian patient.Analysis of the Romanian patients by SSCPand direct DNA sequencing has identified a
previously described mutation 2183(AA>G)(exon 13).
Apolipoprotein E4 allele is notassociated with sporadicAlzheimer's disease in BritishAsians
J P BOURKE*, C M CASTLEDEN*,D PLAHAt, R TREMBATHt
* University Department ofMedicine for theElderly, Leicester General Hospital;tLeicestershire Genetics Service, Leicester RoyalInfirmary
The apolipoprotein (apo) E4 allele is over-represented in populations of patients withlate onset (>60 years) familial and sporadicAlzheimer's disease. We sought to replicatethese findings in a study of white and Asiansubjects with diagnosed sporadic Alzheimer'sdisease (AD). All patients met clinical re-search criteria forAD and had head CT scansto exclude vascular disease. Asian controlsconsisted of spouses or age, sex, and racialorigin matched controls. Apo-e genotypeswere determined from genomic DNA usingPCR amplification and HhaI digestion. WhiteAD(25), Asian AD(8), and Asian controls(6) did not differ in respect of age (white AD75 + 5, Asian AD 70 + 7, and Asian control72 + 3) or gender (female 68%, 62%, and60%). ApoE 3/4 was the most frequent geno-type in white AD (43%) whereas 3/3 was inAsian AD (62-5%). ApoE allele frequenciess3, s4 in white AD were 0-56 and 0-36 akinto recent UK data. Asian gene frequencies ofc3 and c4 were in AD 0-81, 0-18 and controls0 75, 0-25, not statistically different (X2 1-02,p<0-2). The apo s4 allele was higher in whitethan Asian subjects with AD (X2= 10.1,p<0-001). Our findings confirm that furthergenetic and environmental factors are sig-nificant in the aetiology of this disease. Thisstudy also showed a low rate of case findingamong the Asian community for AD whichrequires further scrutiny.
Instability of a normal (CTG).allele in the DM kinase gene
D J DOW*, D C RUBINSZTEIN*,M A FERGUSON-SMITHtt,J R W YATEStt, D E BARTON*
*Molecular Genetics Laboratory andtDepartment of Clinical Genetics,Addenbrooke' NHS Trust, Hills Road,Cambridge; tCambridge UniversityDepartment of Pathology, Tennis Court Road,Cambridge.
The generation of an abnormal allele from anormal allele has not yet been described inDM, although in vivo instability of a normalsized allele has recently been reported in aCEPH family. We have reinvestigated thisCEPH family and determined the (CTG)repeat numbers. We report on an additionalDM family where at least two mutationalevents have occurred on the same non-DMchromosome. One event was a (CTG)33(father) - (CTG)34 (daughter) mutation. Thisis only the second described mutation of anallele within the normal range. The secondmutational event could not be fully char-acterised since both grandparents were deadbut was associated with (CTG)33 (father) and(CTG).9 (father's sister). The persons withthese repeat numbers were asymptomatic.
This family represents a counselling problemsince the (CTG)n sequence on this chro-mosome is showing instability and the riskassociated with a 39 repeat allele is unclear.
Are cataract patients at high riskof being mmal DM genecarriers?
M F PHILLIPS, A M SHAW,R A BARNETSON, D J SHAW,D HEATH, L BECK, P S HARPER,H G HARLEY
Institute ofMedical Genetics and Departmentof Ophthalmology, UWCM, Cardiff, Wales.
Myotonic dystrophy (DM) is classified intothree types, minimal, classical, and con-genital, all of which may be characterised bypolychromatic lens opacities which progressto cataracts. In minimal DM cataracts maybe the only sign of the disorder. We thoughtit possible that a high CTG repeat numberin the DM-PK gene may be found in a sub-stantial proportion of cataract patients.Mouthwashes were taken for DNA analysisfrom a series of 106 unrelated patients at-tending for cataract surgery. None had a CTGrepeat of greater than 30, and the distributionofrepeat sizes was similar to that observed in anormal control population. It was concludedthat there is no evidence that the populationof cataract patients is a major reservoir forDM nor that cataracts are associated with aCTG repeat number in the upper end of the"normal" range.
Comparison of methods forstudying X inactivation status infemales
E WATKISS, T WEBB
Department of Clinical Genetics, University ofBirmingham, Birmingham Maternity Hospital.
Various methods are now available for de-termining the X inactivation status in females.This study compared three of these, M27p,HUMARA PCR, and FISH, at different sitesalong the X chromosome. Methylation stud-ies had previously been carried out usingM27J on 22 normal female monozygotic twinand triplet families. Methylation tests wererepeated on 55 of the females using the HU-MARA PCR method. A total of 37 femaleswere informative for both; 21/37 gave thesame methylation result and those whoshowed skewed X inactivation with both pro-cedures always did so in the same direction.Sixteen of 37 females gave different resultsusing two different methods. Some of thesediscrepancies may be because of the difficultyin assigning a cut off point for skewedness.Levels of skewing using the two procedureswere remarkably similar; 17/37 femalesshowed skewed X inactivation with HU-MARA, and 19/37 showed skewed X in-activation with M27[. X inactivation statusof two female controls was compared usingHUMARA PCR and FISH. The same resultswere obtained by both methods, with oneshowing skewed X inactivation and the otherrandom X inactivation.
Linkage studies in a kindredfrom Oklahoma with familialbenign hypocalciurichypercalcaemia (FBH) indicategenetic heterogeneity and a thirdlocus for FBH
D TRUMP*, M P WHYTEt,C WOODING*, J T PANG*,D KOCHERf, R V THAKKER*
*MRC Molecular Medicine Group,Hammersmith Hospital, London W12;tMetabolic Research Unit, Shriners Hospitalfor Crippled Children, St Louis, Missouri,USA; tThe Cooper Clinic, Fort Smith,Arkansas, USA.
Linkage studies of FBH (MIM 145980), anautosomal dominant disorder, have indicatedheterogeneity and mapped two loci to chro-mosomes 3cen-3q24 and l9pl3.3. In ad-dition, mutations in a calcium sensing re-ceptor located on chromosome 3 have beenidentified in three unrelated FBH families.We have investigated a five generation kindred(17 affected, two obligate carriers, and 21unaffected) from Oklahoma USA in whomFBH was associated with a developmentalrise in serum parathyroid hormone (PTH)and osteomalacia. Linkage studies using 17genetic markers from candidate chromosomalregions 3cen-3q24, l9pl3.3, l1q13, andllpl5 to which the genes for FBH, multipleendocrine neoplasia type 1 (MENI), andPTH have been respectively mapped ex-cluded linkage (lod scores <-2-00 at (0) =0-05 to 0-25). In addition, an analysis ofmultipoint crossovers and use of the LINK-MAP program helped to exclude these re-gions further. Thus this variant of FBH,designated the Oklahoma variant FBH(ok),indicates the presence of a third locus forFBH.
Linkage studies in two familieswith blepharophimosis-ptosisconfirms linkage to 3q21-3q23
H HARRAR, S JEFFERY, M A PATTON
Medical Genetics Unit, St George's HospitalMedical School, London SW17 ORE.
Two three-generation families with ble-pharophimosis-ptosis (BPES) syndrome havebeen evaluated clinically and with linkagestudies. The families show full penetrance forthe striking facial phenotype. There appearsto be no decrease in male fertility but affectedfemales have reduced ovarian function withnormal pituitary hormone levels. One affectedperson shows markedly reduced mental func-tion associated with agenesis of the corpuscallosum. Chromosomal rearrangementshave suggested the locus for BPES is between3q21-3q23. Thirteen of the 20 family mem-bers in the two families are affected. Theprobes D3S1238 and D3S1237 gave lodscores of 3-2 at 0=0 confirming the chro-mosomal assignment.
Prenatal diagnosis ofhaemophilia A by analysis offactor VIII intron 22 inversions
M J MITCHELL, J HAYES,M R A LALLOZ, D M LAYTON
South Thames Regional Centre for PrenatalDiagnosis of Blood Disorders, Department ofHaematological Medicine, King's CollegeHospital, London.
Inversions resulting from intrachromosomalrecombination of the F8A gene within intron22 of the FVIII gene and either of twotelomeric homologous regions were recentlyreported to be responsible for about half ofall severe cases of haemophilia A. Analysis ofthis mutation by Southern blots of BclIdigested DNA probed with a F8A gene frag-ment allowed definitive diagnosis in eightfamilies referred for prenatal diagnosis. Thesecomprised five sporadic cases of haemophiliaA, including one in which there was no livingproband, and three familial cases un-informative for conventional polymorphicmarkers. In one non-obligate putative car-rier, absence of the inversion excluded car-riership. In 3/5 pregnancies where chorionicvillus sampling was performed an affectedmale was diagnosed by direct analysis for theintron 22 inversion. Analysis of FVIII intron22 inversions enables definitive carrier andfetal diagnosis in a high proportion offamilieswith severe haemophilia A and should su-persede RFLP/VNTR analysis as the first lineinvestigation in such families.
Extending availability ofhaemophilia A carrier detectionto all families with haemophilia
M S ENAYAT, B THEOPHILUS,F G H HILL
Haematology Department, The ChildrensHospital, Birmingham.
In some families and in those with so called"sporadic haemophilia", carrier detectionusing RFLPs is not appropriate. However,the recently identified Flip tip inversion andgene sequencing provides an alternative andmore certain approach. Eleven haemophiliacshave had their gene defect defined: four bythe inversion using BclI restriction followedby Southern blotting and seven (five pointmutations and two deletions) by direct mu-tation detection and sequencing of the FVIIIgene. Three of the four identified by theinversion had sporadic haemophilia and in-vestigation of their mothers showed the sameabnormality in them, confirming them as car-riers. Gene sequencing proved similarly usefulin carrier identification in one sporadic andone established haemophilia A family.
Identification of mutations ofexon 13 of the CFTR gene bySSCP and direct sequencing
We have so far identified over 93% of themutations responsible for cystic fibrosis in thenorth west of England, but over 100 CFchromosomes still possess undefined mu-tations. The combination of SSCP and het-eroduplex analysis enabled 6/8 (75%) of exon13 mutations to be identified (5/5 in the 5'halfand 1/3 in the 3' half of the exon). Duringthe course of analysing the first 68 samples,we identified an as yet undescribed mutation,L633P, which we have so far seen only once.A further six chromosomes have shown SSCPand heteroduplex band shifts consistent withsequence variations in exon 13.
Screening for Cu/Zn superoxidedismutase gene mutations infamilial and sporadic cases ofamyotrophic lateral sclerosis
A E ELSHAFEY, W G LANYON,J M CONNOR
University Department of Medical Genetics,Duncan Guthrie Institute, Glasgow G3 85Jt.
Amyotrophic lateral sclerosis (ALS) is a dis-ease ofmotor neurones ofunknown aetiology.Clinically, ALS is a progressive muscle at-rophy combined with symptoms of corti-cospinal tract degeneration. Both familial(FALS) and sporadic forms of ALS exist.In most instances, both forms are clinicallysimilar. Mutations in the Cu/Zn superoxidedismutase (SOD-1) gene, mapped to2 1q22. 1, have been reported in both familialand sporadic forms of the disease. The SOD-1 gene in two familial and 70 apparentlysporadic ALS patients has been screened forsequence alteration using SSCP. One mis-sense point mutation and two silent pointmutations were detected in exons 4 and 5 ofthe gene. The missense point mutation wasdetected in a familial form of the disease andit was shown to affect one of the conservativeamino acids which may disrupt critical in-teractions between the subunits of the homo-dimer, altering the structural integrity of theenzyme. No effective mutations could be de-tected in the patients with the sporadic form,raising the possibility that other gene(s) mighthave a role in this form of the disease and inthe familial form which shows no linkage tothe chromosome 21 locus.
Hamartomas from patients withtuberous sclerosis show loss ofheterozygosity for chromosome9q34
A J GREEN, T SEPP, J RW YATES
Department of Clinical Genetics, Addenbrooke'sHospital, Cambridge; Department ofPathology, University of Cambridge.
We have previously shown allele loss in ham-artomas from cases of tuberous sclerosis(TSC) for markers in the region of the re-cently characterised TSC2 gene on chro-mosome 16pl3.3. Germline deletions in theTSC2 gene have been shown in 5% ofpatientswith TSC. These data strongly suggest thatthe TSC2 gene acts as a tumour suppressorgene. We hypothesised that TSC hamartomascan also show allele loss for markers on chro-mosome 9q34, in the region of the TSC1
gene. We studied seven hamartomas (threerenal angiomyolipomas, three giant cell as-trocytomas, and a cardiac rhabdomyoma)from seven cases of TSC, none of whichshowed allele loss for markers on chro-mosome 16p1 3.3. Eight microsatellite mark-ers were analysed, from centromeric totelomeric, ASS-D9S64-D9S149-D9S 150-DBH-D9S66-D9114-D9S67. Two ham-artomas (one renal angiomyolipoma and one
giant cell astrocytoma) showed allele loss forat least two markers. The region of allele lossinvolved the TSC1 locus, but did not includeD9S149 or D9S67. We have shown allele lossin two of seven TSC hamartomas in theregion of the TSC1 gene on 9q34. Based on
this deletion mapping, we suggest that theTSC1 gene lies between D9S149 and D9S67.These data would imply that the TSC1 geneon 9q34, like the TSC2 gene, acts as a tumoursuppressor gene.
X inactivation studies in patientswith Turner's syndrome and ringX chromosomes
C E CHU*, M D C DONALDSON*,C KELNARt, P SMAIL, S GREENE§,E BOYDII, J M CONNORI
*Department of Child Health, Glasgow;tChild Life and Health, Edinburgh; 4AberdeenRoyal Hospitals, Aberdeen; §NinewellsHospital, Dundee; IlDuncan Guthrie Institute,Glasgow.
Most patients with Turner's syndrome (TS)have normal intelligence. However, some TSpatients with small ring X chromosomes are
mentally retarded and may have particulardysmorphic features. The phenotype isthought to occur because the small ring chro-mosome is active in the same cells as thenormal X. The aim of this study was to lookat the X inactivation patterns of TS patientswith ring X chromosomes and normal in-telligence to see if the ring remained active.Eighteen patients with ring X chromosomeswere studied; 15 of the patients attendednormal school but six required extra help.One attended special school and informationwas not available for two. There was nodifference in phenotype between patients withdifferent sized rings or different percentagesofrings in blood. Three techniques were used,Southern blotting with probe M271, reverse
transcription PCR with primers from the Xinactive specific transcript (XIST), and PCRwith primers from the androgen receptor(HAR) gene. Both patients investigated withRT-PCR were found to show normal ex-
pression of XIST. One of these showed totalinactivation of the ring X on PCR but theother showed that the ring was active in somecells. In nine other cases the ring X wasfound to be active in some cells. Educationaldifficulty did not correlate with activity of thering X. This study indicates that activity ofthe ring X per se may not have a profoundeffect on educational ability.
Analysis of rhodopsin splice sitemutations in retinitis pigmentosa
J L WHITEHEAD, C BELL,N E HAITES
Department ofMedical Genetics, University ofAberdeen.
Retinitis pigmentosa (RP) describes a het-erogeneous group of human inherited ret-inopathies. Causative mutations have beenfound in the genes encoding rhodopsin, per-ipherinlRDS, the cGMP phosphodiesterasef-subunit, and the rod cell cGMP-gatedchannel. Using SSCP analysis and direct se-
quencing we have identified two splice sitemutations in the rhodopsin gene; one in theintron 1 donor splice site in a person withisolated RP and the other in the intron 4acceptor splice site in affected members ofan autosomal dominant family. To investigatethe effect ofthese mutations on splicing, rhod-opsin transcripts were amplified from patientperipheral blood lymphocytes by RT-PCRusing primers which flanked the respectivesplice sites. Two bands, one of the expectedcDNA size and the other corresponding to a
cDNA in which the intron had been retained,were observed in affected persons on elec-trophoresis. Control samples showed only theexpected cDNA sized band. Since inclusionof intron 1 or intron 4 results in the presenceof a premature stop codon in the processedtranscript, these mutations are strongly sus-
pected to be pathogenic. The ability to am-
plify candidate gene transcripts therefore pro-
vides opportunities to study the effect of sim-ilar mutations in RP.
Characterisation of a nonsense
mutation and a two baseinsertion in exon 42 of theneurofibromatosis type 1 gene
S M PURANDARE, W G LANYON,J S PATERSON, H R DAVIDSON,J M CONNOR
University Department ofMedical Genetics,Duncan Guthrie Institute, Yorkhill, GlasgowG3 8SJ.
Identification and characterisation of germ-line mutations within the NF-I gene was
carried out using chemical mismatch cleavageanalysis and direct sequencing. Upon analysisofexon 42, we identified a nonsense mutation(R2496X), and a two base insertion (7486ins-GG), in two unrelated and sporadic cases ofNF-1, both the mutations affecting codon2496 of the NF-I gene. Direct sequencing ofexon 42 showed the presence of an insertionof two guanine bases at nucleotide position7486. The mutation causes a shift in thereading frame, and creates an inappropriatestop codon at position 2502. A protein prod-uct of 2501 amino acids instead of the normal2818 amino acids is predicted. The nonsense
mutation, a C to T transition at a hy-permutable CpG dinucleotide at position7486, converts the codon for arginine (CGA)to a premature stop codon (TGA) at position2496, and is predicted to produce a truncatedprotein product consisting of 2495 aminoacids. Genotype-phenotype correlation stud-ies were carried out in both the patients whoshowed the classical features of NF-1, and it
was seen that the patient with the two baseinsertion had specific learning difficulties,while the patient with the nonsense mutationhad no added complications of NF-1.
Screening for the fragile X A andE mutations by automatedfluorescent analysis
Q WANG, E P GREEN, M BOBROW,C G MATHEW
Division of Medical and Molecular Genetics,Guy's Hospital, London SEI 9RT
Direct DNA analysis of the fragile X ex-pansion mutation has begun to replace cyto-genetic analysis for the laboratory diagnosisof this disorder. We have established a fluor-escent PCR assay to screen for both theFRAXA and FRAXE unstable triplet repeatmutations simultaneously based on the pres-ence or absence of triplet repeat alleles ofnormal size in the sample tested. PCR primersflanking the FRAXA and FRAXE triplet re-peats were labelled with fluorescent dyes ofa different colour, co-amplified, and the prod-ucts detected and sized on an automatedDNA sequencer. A blind test of 193 samples(45 normal males, 48 affected males, 54 nor-mal females, 46 female carriers) was carriedout, and all samples were correctly identified.The normal range of the number of GCCtriplet repeats in the FRAXE gene in 435unrelated normal chromosomes was 4 to 48copies. The most common alleles were from15 to 20 copies (79%). The heterozygosityofFRAXE alleles in 268 normal females was73%. Family studies showed that the allele(GCC)28 was stable in six meioses, and allele(GCC)39 in two meioses. We conclude thatthe assay is a safe, accurate, and efficientscreen for FRAX mutations.
Molecular DNA testing of afamily manifesting fragile Xsyndrome in both the fraX-A fullmutation and mosaic forms
S BULLOCK*, V H LINDLEY*,K STEVENSON*, T COLEt
*Regional Cytogenetics Unit, BirminghamWomen' Health Care NHS Trust(Birmingham Maternity Hospital), Edgbaston,Birmingham B15 2TG; tClinical GeneticsUnit, Birmingham Women's Health Care NHSTrust.
A 20 month old boy was referred for fragile Xtesting as a consequence ofslow developmentand the fact that a 6 year old male cousin onhis mother's side had already been diagnosedas fragile X positive by cytogenetic analysis(18/75 fraX positive cells). The proband had50% fraX positive cells on cytogenetic ana-lysis and showed a large full mutation fragileX-A insert in the FMR-1 gene when testedmolecularly with the probe StB 12.3 followingan EcoRI/BstZI double digest of his DNA.Subsequently, various members of his largefamily were tested for fragile X-A showing aprecursor premutation expanding to a fullmutation in one branch of the family butleading to mosaicism in another.
The use of molecular DNAtesting to distinguish betweenmutations at the fragile X-Aand -E sites
S BULLOCK*, C A FELIX*,M A BUTTERWORTH*,K STEVENSON*, D WILLIAMSt
*Regional Cytogenetics Unit, BirminghamWomen's Health Care NHS Trust(Birmingham Maternity Hospital), Edgbaston,Birmingham B15 2TG; tClinical GeneticsUnit, Birmingham Women's Health Care NHSTrust.
A 3 year old boy with language delay wasreferred for cytogenetic investigation to ex-clude fragile X syndrome. A fragile X chro-mosome was observed in 22/55 cellsexamined (40% expression) and the fragilesite at Xq27.3 could not be distinguishedfrom that in classic FraX-A families. Sub-sequent molecular analysis of the boy's DNAusing the probe StB 12.3 after double di-gestion with EcoRI and BstZI did not showan expanded trinucleotide repeat region inthe FMR-1 gene. Fluorescent in situ hy-bridisation using cosmid probes which spanthe fraX-A and fraX-E region of Xq27-28showed that the fragile site seen in this boyis consistent with the folate sensitive fraX-Esite. Cytogenetic analysis of the boy's mothershowed a fragile X chromosome in 7/30 cellsexamined (23% expression). This case em-phasises the importance of molecular con-firmation of suspected positive fraX-Acytogenetic results.
The use of microsatellite markersin monitoring the progression ofbone marrow grafts in patientswith severe combinedimmunodeficiency syndrome
A CURTIS*, V SPENCER*,A JACKSON*, M ABINUMt, A J CANTt
*Northern Region Genetics Service, 19120Claremont Place, Newcastle upon Tyne;tNorthern Supra Regional Bone MarrowTransplantation Unit, Newcastle GeneralHospital, Newcastle upon Tyne.
We have used various highly polymorphicmicrosatellite markers to monitor the successofbone marrow transplantation in 12 patientswith SCIDS. An initial analysis is carried outpre-transplant on DNA extracted from wholeblood from the patient and the donor to findan informative marker. The gain of donoralleles was normally evident in whole blood30 to 40 days post transplant if the graft hadbeen successful. This confirmatory evidenceof engraftment usually coincided with theoccurrence of clinical and immunologicalevidence. We estimate a detection limit ofone donor cell in 50 to 100 recipient cells.Earlier evidence of engraftment was obtainedif bone marrow cells were analysed directly.The progression of the grafts was monitoredat suitable intervals post-transplant. In thecase of a father who was the donor for hisdaughter, the chromosome 5 markers DPIand CA25 were informative and the gainof the "non-inherited" paternal allele anddepletion of the maternal allele in the re-cipient was evident 30 days post transplant.However, after splenectomy several months
later owing to massive splenomegaly, the mo-lecular genetic monitoring of the patient de-tected a significant depletion in the donorallele before any deterioration in clinical con-dition. This meant that the patient was as-sessed more closely and has recently led to a"top-up" transplant.
Comparison of microsatellite andmethylation studies for thediagnosis of Prader-Willi andAngehnan syndrome
K J LEACH, G R TAYLOR, S J LYNCH,V A MURDAY, C B BENNETT,R F MUELLER
DNA Laboratory, Clinical Genetics Unit,Ashley Wing, St J7ames's Hospital, LeedsLS12 3RP?
Prader-Willi syndrome (PWS) and Angelmansyndrome (AS) are associated with chro-mosome 15q deletions. In PWS the loss ofpaternal alleles is seen, whereas in AS thematernal markers are lost suggesting a gen-omic imprinting mechanism. Thus mo-lecular diagnosis of PWS and AS can beattempted by three different techniques: mi-crosatellite analysis to detect deletions or uni-parental disomy, methylation analysis, andfluorescent in situ hybridisation. This studycompares the ease and sensitivity of mi-crosatellite analysis using five microsatellitesfrom 1 ql 1 -q13 (PWS and AS critical region)with an ABI Genescanner with methylationstudies using probe PW71B on a HindIII/HpaII digest of genomic DNA. Of the 25pedigrees analysed for PWS, eight were con-firmed by microsatellite analysis and an ad-ditional one was detected with PW7 1 B. Threeof the nine AS pedigrees were confirmed byboth techniques. We conclude that methyl-ation studies offers the following advantagesover microsatellite analysis: (1) parentalsamples are not required; (2) it is more in-formative than the panel of microsatellites weused. Microsatellite analysis was faster, butspeed is rarely an issue in these cases. The lowdetection rate seen with these cases probablyreflects broad based selection criteria used toidentify such cases.
A null allele phenotype in aseverely affected Marfansyndrome patient with reducedlevels of fibrillin
C M BLACK*, A P WITHERS*,Z AL-GHABAN*, D U BATY*,J R GRAYt, L McLEISH*, M BOXER*
*Molecular Genetics Laboratory, Departmentof Pathology, Ninewells Hospital and MedicalSchool, Dundee; tInstitute of MedicalGenetics, University of Wales College ofMedicine, Cardiff, Wales.
MFS patients were screened for an RsaI poly-morphism in the 3' untranslated region ofFBN1 and RNA, extracted from fibroblastsof patients that were heterozygous for thispolymorphism, was then analysed by RTPCR(reverse transcriptase PCR) amplificationand RsaI digestion of the products. In one ofsix patients only one allele was amplified. Nolow level expression of the other allele wasobserved despite RTPCR amplification in-
corporating 32P-dCTP, thus showing a nullallele phenotype. The patient had survivedan acute aortic dissection and exhibitedarachnodactyly, narrow, high arched palate,had an arm span exceeding her height, jointlaxity, and bilateral lens subluxation. Analysisof the family showed affected members inthree generations and that FBN1 segregatedwith the disease. Immunohistochemical stud-ies of the patient's tissues and fibroblastsshowed lowered levels in staining of mi-crofibrillar structures compared with agematched controls. However, immunoblottingof proteins secreted by the patient's dermalfibroblasts showed only normal sized fibrillin.Therefore, low level of expression of fibrillincan lead to classical severe MFS.
A case of Angelman syndromearising as a result of a de novoRobertsonian 15/15 translocation
J CLAYTON-SMITH*, S C RAMSDEN*,L GAUNT*, A SERES-SANTAMARIAt
*Department of Medical Genetics, St Mary'sHospital, Manchester; tDepartment of MedicalGenetics, Barcelona, Spain.
A 4 year old child presented with typicalclinical features of Angelman syndrome.Cytogenetic analysis showed that he had aRobertsonian 15/15 translocation. This ap-peared to have arisen de novo, both parentshaving normal chromosomes. FISH analysiswith Oncor probes 1 5A and 1 5B, which mapto the Sli and GABRB3 loci respectively,showed that both chromosome 15s appearedintact, with no evidence of a 15qll-13 de-letion. Molecular analysis using microsatelliterepeats within the 15qll-13 region showedthat both chromosomes involved in the Rob-ertsonian translocation were of paternal ori-gin, that is, he had paternal uniparentalisodisomy for chromosome 15. This is the firstreported instance in which both chromosome15s have been involved in a Robertsoniantranslocation. It is likely that the translocationarose during paternal meiosis giving rise toa conceptus that was initially trisomic forchromosome 15. The maternal 15 was sub-sequently lost. This patient thus had an ap-parently balanced translocation but presentedwith problems because the chromosome in-volved contained loci subject to the effects ofgenomic imprinting.
Detection of germline mutationsin the RBl gene usingamplification mismatch detection(AMD) analysis and directsequencing of PCR products
M DUNDAR, W G LANYON,J M CONNOR
Duncan Guthrie Institute of Medical Genetics,Yorkhill Hospital, Glasgow G3 8SJ.
In this study, we used RT-PCR combinedwith amplification mismatch detection(AMD) analysis to detect mutations in 15unrelated patients with bilateral re-tinoblastoma. To characterise all types ofmutations in the coding sequence of the RB 1gene, the entire cDNA was amplified usingfive primer sets covering almost the entiregene. We identified two mutations: a missense
mutation (T to G transversion) at nucleotideposition 1587 causing the substitution of hi-stidine by glycine and a splice acceptor sitemutation at position -2 of intron 16 causingthe skipping of exon 17. The approach usedin this study has proved to be an effectivemethod for the detection of germline mu-tations in the RB1 gene.
Molecular pathology of the RB1gene in retinoblastoma, breastand bladder tumours
M DUNDAR, W G LANYON,J M CONNOR
Duncan Guthnie Institute ofMedical Genetics,Yorkhill Hospital, Glasgow G3 8S_J.It has been shown that mutations in the RB 1tumour suppressor gene commonly occur inother cancers, notably breast and bladdercarcinoma. SSCP analysis and heteroduplexanalysis together with direct sequencing wereused to detect mutations in 18 patients with
retinoblastoma. We also looked for mutationof the RB1 gene in 39 breast carcinomas and40 bladder carcinomas. Analysis showed fourmutations in patients with bilateral Rb andone mutation from a patient with bladdercarcinoma. The four mutations from patientswith Rb were: a G insertion at nucleotideposition 2251, a splice site mutation (A to Ctransversion) at position 1636 in exon 16, anA-+G transversion in intron 19, and a T toC transition at position 1617. A G to Ctransversion at position +1 of the splicedonor site of intron 12 was detected in apatient with bladder carcinoma. No al-terations ofthe RB 1 gene in breast carcinomawere detected.
Detection of mutations in theRYRI gene in malignanthyperthermia susceptible persons
A M ADEOKUN*, F R ELLISt,P M HOPKINSt, P J HALSALLt,J CURRANt, A D STEWARTt,S P WEST*
*Department ofHuman Genetics, University ofNewcastle upon Tyne; tDepartments ofAnaesthesia and Genetics, The University ofLeeds.
The commonly used inhalation anaestheticspresent a serious hazard for persons ge-netically predisposed to malignant hy-perthermia (MH) because these agentsinduce a state of hypermetabolism in skeletalmuscle causing a rapid rise in body tem-perature. If not diagnosed and treated im-mediately, death follows rapidly. Defects inthe gene encoding the skeletal muscle ry-anodine receptor (RYRI) are believed to un-derlie malignant hyperthermia susceptibility(MHS) in about 50% of families. We haveused RT-PCR followed by heteroduplex ana-lysis in an attempt to identify the mutationsin 50 unrelated MHS persons. Given thehigh levels of genetic heterogeneity alreadydetected in malignant hyperthermia, the na-ture of the mutations detected so far, and theexistence ofan alternative (although not ideal)diagnostic test, we consider that it is not yetappropriate to offer genetic presymptomaticdiagnosis for this condition.
