ESHG Posters 14

J. M. Flanagan¹, O. Tighe¹, C. O'Neill², E. Naughten², P. D. Mayne², D. T. Croke¹;
¹Royal College of Surgeons, Dublin, IRELAND, ²The Childrens Hospital Temple St., Dublin, IRELAND.

Classical, or transferase deficient, galactosaemia is an inherited metabolic disorder, caused by mutation of the Galactose-1-Phosphate Uridyl Transferase (GALT) gene. Previous studies have shown that classical galactosaemia has an overall incidence of 1 in 21,000 live births in Ireland, higher than elsewhere in Europe. Although there is a spectrum of causative mutations in the GALT gene, one mutation (Q188R) predominates in Caucasians. The frequency of this Q188R mutation exhibits an East to West gradient of increasing frequency across Europe, peaking in Ireland (89.1%). This raises questions as to the origins and introduction of the Q188R mutation to Ireland.
Through a combination of bioinformatics, PCR, dHPLC and Restriction Fragment Length analysis (RFLP), four novel Single Nucleotide Polymorphism (SNPs) and three flanking Short Tandem Repeat (STR) markers linked to the GALT gene have been identified. A haplotype system for Galactosaemia has been established through the combination of these markers with two previously known SNPs. Sixty Irish control DNA samples have been analysed demonstrating that 8 distinct haplotypes exist in the Irish population. The analysis of the haplotype backgrounds of the common GALT gene mutations present in the European population is now underway. This should help resolve the origins and migrations of the GALT gene mutations.

CA repeats in the first intron of the CFTR gene in cystic fibrosis patients and healthy Latvians

A. Krumina¹, L. Krumina¹, M. Lazdins², V. Svabe¹, Z. Krumina³, U. Teibe¹, V. Baumanis²;
¹Medical Academy of Latvia, Riga, LATVIA, ²Latvian State University, Riga, LATVIA, ³State Medical Genetics Centre, Riga, LATVIA.

A dinucleotide CA repeat within intron 1 of the CFTR gene, highly informative in tracing unknown mutations causing cystic fibrosis (CF) in families as well as in population genetic studies, has been identified by D.S. Moulin in 1997. The aims of our study were to analyse CA repeat polymorphism in CF patients and healthy Latvians, and to compare our results with the data on other populations. 22 CF patients, 24 their family members and 42 healthy Latvians were subjected to DNA analysis. CA repeats were studied by denaturing gel electrophoresis of amplified PCR products and three control samples were chosen for sequencing analysis. Absolute linkage disequilibrium was found between the CF mutation dF508 and 21 - CA repeat allele. When considering allele distribution, Latvian population presented a unimodal distribution, showing peak at 22 repeats. Expected heterozygosity, estimated as 1 - S p_i², where p_i is the frequency of the ith allele in the locus, for Latvian population was 0.696 that is lower than in other described European populations. The data obtained in our study support conclusions of E.Mateu et al. (1999) about high informativeness of this CFTR gene marker for family and population studies.

H. R. Kelly¹, A. Yoneda², S. Kelly², D. Shields³, C. Molony³, A. Green¹, P. Puri², D. Barton¹;
¹National Centre for Medical Genetics, Dublin, IRELAND, ²Children's Research Centre, Dublin, IRELAND, ³Royal college of Surgeons, Dublin, IRELAND.

Vesicoureteral reflux (VUR), the retrograde flow of urine from the bladder into the ureter and kidneys, is a common disorder, found in 1-2% of children. VUR can cause kidney damage and is the most common cause of end-stage renal failure and severe hypertension in children. It is caused by a shortening of the segment of the ureter which runs through the submucosal layer of the bladder wall. There is a strong genetic component to VUR but the mode of inheritance is still unknown.
We have collected 480 DNAs from 97 families with more than one child affected with primary VUR. We are using this resource to search for VUR susceptibility genes.
Candidate genes or regions are selected based on literature reviews. The uroplakins are 4 integral membrane proteins that physically strengthen the bladder wall. Based on uroplakin knockout mice studies, Hu et al., 2000 suggested the existence of a VUR subtype distinct from that caused by the deletion of the AT II receptor,. This report led to the selection of the first group of genes investigated. Two candidate regions were also investigated: chromosome 10q, based on a report of a patient with a deletion in this region associated with VUR and chromosome 1q, where markers over a 20cM region are reported to show strong evidence of linkage.
Results so far have shown evidence for lack of linkage to VUR in these genes or regions.

The mutation rates and polymorphisms of ten autosomal tetranucleotide loci in Western Siberian population

Microsatellite loci are powerful and convenient markers for genetic studies, but it is necessary to evaluate the applicability of specific STR markers for practical aims and for description of population history. Tetranucleotide STRs have several technical advantages because of lesser prone to artificial slippage. However the mutation rates of this class of microsatellites differ significantly between loci. The purpose of this study was to evaluate the mutation rates and allelic frequencies of ten autosomal tetranucleotide loci in Western Siberian population. Allelic polymorphisms were studied in 200 – 290 unrelated individuals from Tomsk region, and mutation rates were estimated as frequency of parent-child mismatches in the segregation analysis of STR alleles in 134 families with proven paternity. We used polyacrilamide gel electrophoresis following PCR amplification for tetranucleotide loci D2S1242, D11S1983, D16S2624, D17S1185, D19S601, D20S161, D20S168, D21S1413, D21S11and D21S1435. All loci studied have an average heterozigosity > 0.65, PIC > 0,550 and amplicon sizes < 300 b.p. It is revealed five new allelic variants in four different loci in Tomsk population. We analyzed 1198 events of allelic transmission from the parents to children and detected nine de novo mutations events, that corresponds to average mutation rate of the complex of the loci studied 7.5x10-3 per locus per meiosis. No mutations were found in four loci (D11S1983, D16S2624, D20S161, D20S168), and two mutations were observed in each of three markers D17S1185, D19S601, D21S1435. Most of events were single step mutations, and mutations of paternal origin happened three times often than maternal ones.

B. Neve¹, P. Froguel¹^,2, L. Corset¹, E. Vaillant¹, V. Vatin¹, P. Boutin¹;
¹Institut Biologie de Lille/Institut Pasteur de Lille, Lille, FRANCE, ²Queen Mary and Westfield college, University of London, London, UNITED KINGDOM.

Positional cloning of complex disease-susceptibility genes by linkage disequilibrium(LD)-mapping involves genotyping a vast amount of single nucleotide polymorphisms(SNPs). A general aim is to to achieve a density of at least 1SNP/10kb in previously identified chromosomal regions and to screen all coding and regulatory regions of candidate genes. A combined SNP map for a manageable region(±5Mb), may require genotyping 300-500 SNPs. Especially with such a high throughput, individual genotyping of SNPs remains expensive. Although current methods may be accurate, robust and adequate for large scale application, a cheaper and less time-consuming alternative to individual genotyping is using allele frequencies determined in DNA pools. We have developped an accurate and reproducible protocol for direct allele frequency determination using pyrosequencingTM technology in large genomic DNA pools(374 individuals). A correlation of 0.980 was measured between the allele frequency detected by this method and by individual genotyping. In the context of disease-associated SNPs studies, we compared the allele frequencies between disease and control groups. The measured difference varied with 1.5±0.9% between the two detection methods. It may be concluded that the protocol with large DNA pools could reliably detect a 4% difference between populations. Furthermore, it is economic with regard to amounts of DNA and reagents required. The method is currently used to screen a region of chromosome 2(6Mb) and 10(3.5 Mb) for SNPs with an allele frequency differences between type-2-diabetes or obesity and control groups. This allows a rapid identification of interesting SNPs for further association studies and susceptibility-gene discovery in these complex diseases.

A. Begonja, E. Topic, A. Simundic, M. Stefanovic;
Clinical Hospital Sestre Milosrdnice, Zagreb, CROATIA.

In the promoter region of the plasminogen activator inhibitor-1 (PAI-1) gene a common 4G5G polymorphism was described. A 4G allele is associated with increased transcription of PAI-1 protein due to different binding of transcription regulating proteins than in 5G site. This may change the fibrinolitic capacity, decreasing the ability to lyse clots. Many studies showed that 4G5G polymorphism was associated with increased risk for cardiovascular disease. Still, conflicting data are reported for cerebrovascular disease, even suggesting protective role of 4G allele.
In order to investigate relation between 4G5G polymorphism and stroke incidence, 52 stroke patients and 126 healthy subjects were genotyped by PCR-SSCP analysis. Genotype distribution among controls was: 4G4G 24% (30), 4G5G 48% (60) and 5G5G 29% (36). The allelic frequencies in control group were: 4G 48% and 5G 52%. In patients group the genotype distribution was: 4G4G 23% (12), 4G5G 40% (12) and 5G5G 37% (19). The frequencies for 4G and 5G alleles were 43% and 57%, respectively. There was no significant difference for genotype and allele frequencies between control and patient group (c²test). A trend towards a lower prevalence of the 4G allele in the patients group was observed (43% vs. 48% in control group, OR 0.8, 95% CI 0.53-1.32). These preliminary findings suggest that 4G allele is not a risk factor for stroke and may even be related to its reduced incidence. Further studies are needed on a larger number of subjects to evaluate the role of 4G allele in cerebrovascular disease.

Autosomal dominant Juvenile Amyotrophic Lateral Sclerosis (ALS) and distal Hereditary Motor Neuropathy (distal HMN) with pyramidal tract signs

J. I. Devolder-Irobi¹, E. De Vriendt¹, V. Van Gerwen¹, M. Auer-Grumbach², B. Plecko³, K. Wagner⁴, M. Kennerson⁵, G. Nicholson⁵, D. Zhu⁵, H. P. Hartung⁶, V. Timmerman¹, P. De-Jonghe¹^,7;
¹Molecular Genetics Department, Flanders Interuniversity Institute for Biotechnology (VIB), Born-Bunge Foundation (BBS), University of Antwerp (UIA), Antwerpen, Belgium, Antwerpen, BELGIUM, ²Department of Neurology, Karl-Franzens University Graz, Austria., Graz, AUSTRIA, ³Department of Paediatrics, Karl-Franzens University Graz, Austria, Graz, AUSTRIA, ⁴Institute of Medical Biology and Human Genetics, Karl-Franzens University Graz, Austria, Graz, AUSTRIA, ⁵Molecular Medicine and Neurobiology Laboratories, ANZAC Medical Research Institute, University of Sydney, Concord Hospital NSW, Australia, Sydney, AUSTRALIA, ⁶Department of Neurology, Karl-Franzens University Graz, Austria, Graz, AUSTRIA, ⁷Division of Neurology, University Hospital Antwerpen (UZA), Antwerpen, Belgium, Antwerpen, BELGIUM.

Autosomal dominant juvenile amyotrophic lateral sclerosis (ALS) is a rare disorder and so far only one family has been reported and mapped to chromosome 9q34 (ALS4). The diagnosis of ALS in this family is based on almost exclusive lower motor neuron pathology in combination with less prominent pyramidal tract signs. Atypical features include normal life expectancy, the absence of bulbar involvement and the symmetrical distal distribution of atrophy and weakness. We performed a molecular genetic study in 3 families that we had diagnosed as distal hereditary motor neuronopathy (distal HMN), i.e. distal spinal muscular atrophy or spinal Charcot-Marie-Tooth syndrome, and found linkage to the ALS4 locus. The clinical phenotype in these 3 families of different geographic origin (Australian, Austrian and Belgian) is strikingly similar to the original ALS4 family except for a younger onset age in the distal HMN families. These data suggest that ALS4 and distal HMN with pyramidal tract signs may be one and the same disorder. In all 3 families, a disease-associated haplotype was present in all affected individuals. Genotyping data demonstrated that the families do not share a common disease haplotype. Cumulative significant LOD-scores (z > 3) are reached with seven STR markers. These results indicate that the distal HMN locus with pyramidal tract signs is located within the 5 cM ALS4 region, i.e. between the flanking markers D9S64 and D9S164. This region represents a 3.5 Mb region according to the Golden Path at UCSC.

Atb0/slc1a5 Gene. Fine Localisation And Exclusion Of Association With The Intestinal Phenotype Of Cystic Fibrosis

S. Larriba, L. Sumoy, M. D. Ramos, J. Giménez, X. Estivill, T. Casals, V. Nunes;
Institut de Recerca Oncologica, L'Hospitalet de Llobregat, Barcelona, SPAIN.

Genetic heterogeneity in the 19q13.2-13.4 region called Cystic Fibrosis Modulator Locus 1 (CFM1) seemed to be associated to the intestinal phenotypic variation of cystic fibrosis (CF). The Na+-dependent amino acid transporter named ATB0 was previously found to be located in 19q13.3. In the present study, we performed fine chromosomal mapping of ATB0 on radiation hybrid (RH) panels G3 and TNG. Based on the most accurate location results from TNG-RH panel, mapping analysis evidenced that ATB0 is localised between STS SHGC-13875 (D19S995) and STS SHGC-6138 in 19q13.3, that corresponds with the immediately telomeric/distal segment of the strongest linkage region within the human CFM1 (hCFM1) syntenic region. The position in relation to the hCFM1 syntenic region, besides the functional characteristics of the encoded protein and its apparent relevance to meconium ileus (MI) led us to evaluate the possible implication of ATB0 in the intestinal phenotype of CF.
Regarding to the genomic structure, Blast-N program allowed us to determine that ATB0 gene is organised into eight exons. An exhaustive mutational study of the gene was performed by SSCP analysis in CF patients with and without MI. Several sequence variations in the ATB0 gene were identified, although none of them seemed to be related to the intestinal phenotype of CF. Even though no particular allele or haplotype in ATB0 appears to be associated to CF-MI disease, new SNPs identified should be useful in segregation and linkage disequilibrium analyses in families affected by other disorders caused by the impairment of neutral amino acid transport.

Homozygosity Mapping of a Weill-Marchesani Syndrome Locus to Chromosome 19p13.3-p13.2.

L. Faivre¹, A. Mégarbané², A. Alswaid³, L. Zylberberg⁴, N. Aldohayan⁵, A. Campos-Xavier¹, D. Bacq⁶, L. Legeai-Mallet¹, J. Bonaventure¹, A. Munnich¹, V. Cormier-Daire¹;
¹Département de Génétique et INSERM U393, Hôpital Necker-Enfants Malades, Paris, FRANCE, ²Unité de Génétique Médicale, Université Saint Joseph, Beirut, LEBANON, ³Clinical Genetics, Riyadh Armed Forces Hospital, Riyadh, SAUDI ARABIA, ⁴CNRS UMR 8570, Université Denis Diderot, Paris, FRANCE, ⁵Department of Ophthalmology, Riyadh Armed Forces Hospital, Riyadh, SAUDI ARABIA, ⁶Centre National de Génotypage, Evry, FRANCE.

Weill-Marchesani syndrome (WMS) is a rare disease characterized by short stature, brachydactyly, joint stiffness, and characteristic eye abnormalities including microspherophakia, ectopia lentis, severe myopia and glaucoma. Despite clinical homogeneity, both autosomal dominant and autosomal recessive inheritance with occasional brachymorphism in heterozygotes have been reported. Here we report on homozygosity mapping of the WMS gene in two large Lebanese and Saudian families. All affected individuals (n=5) fulfilled the criteria for WMS. A genome-wide search was performed using microsatellites markers at an average distance of 10 cM and revealed linkage of the disease-causing gene to chromosome 19p13.3-p13.2 (Zmax = 5.99 at q=0 at locus D19S906). A recombination event between loci D19S905 and D19S901 defined the distal boundary and a second recombination event between loci D19S221 and D19S840 defined the proximal boundary of the genetic interval encompassing the WMS gene (12.4 cM). Interestingly, the collagen V alpha 3 gene has been assigned to this region, and appeared to be a good candidate gene by its function, but RT-PCR analysis of skin fibroblast mRNAs failed to detect any pathogenic mutations in one family. These results were confirmed by transmission electron microscopy of skin biopsies showing normal diameter and striation of collagen fibrils. Ongoingstudies will hopefully lead tothe identification of thedisease-causing gene.

K. Verhoeven¹, E. De Vriendt², V. Van Gerwen², P. De Jonghe¹^,3, V. Timmerman²;
¹Molecular Genetics Department, Flanders Interuniversity Institute for biotechnology (VIB), Born-Bunge Foundation (BBS), University of Antwerp (UIA), Antwerpen, BELGIUM, ²Molecular Genetics Department, Flanders Interuniversity Institute for Biotechnology (VIB), Born-Bunge Foundation (BBS), University of Antwerp (UIA), Antwerpen, Belgium, Antwerpen, BELGIUM, ³Division of Neurology, University Hospital Antwerp (UZA), Antwerpen, BELGIUM.

Charcot-Marie-Tooth (CMT) disease is a clinically and genetically heterogeneous disorder of the peripheral nervous system. It is characterized by progressive weakness and atrophy, initially of the peroneal muscles and later of the distal muscles of the arm. Most CMT patienst can be classified as having either CMT1 or CMT2 by use of a cut-off value of 38 m/s for the motor median nerve conduction velocity (NCV). However, in some CMT families, patients have motor median NCVs ranging from 25 to 45 m/s. This CMT type has been designated as “intermediate CMT”. Recently, two loci for autosomal domiant intermediate CMT have been identified; the first locus maps to chromosome 10q24.1-q25.1 in an Italian family, the second locus maps to 19p12-p13.2 in an Australian family. Here, we performed a genome wide scan in a Belgian family with autosomal dominant CMT with 386 short tandem repeat (STR) markers of the Applied Biosystems Prism Linkage mapping set MD-10 (ABI). Fragment analysis was performed on an ABI 3700 automated sequencer. Two-point linkage analysis was performed for all STRs. Additional and flanking STRs were genotyped when LOD scores for ABI markers reached the LOD value of 1. A LOD score of 3.3 was found with markers D19S586 and D19S226, which locates the gene responsible for the CMT phenotype in the Belgian family in the region of intermediate CMT on 19p12-p13.2. However, we were not able to the delineate the 19p12-p13.2 linkage region of 16 cM.

M. Pettersson, M. Bylund, A. Alderborn;
Pyrosequencing AB Vallongatan 1, Uppsala, SWEDEN.

The identification of haplotypes from a combination of single nucleotide polymorphisms (SNPs) on one chromosome is a powerful tool for genetic research. Haplotyping is usually performed statistically by computational analysis or by time consuming cloning techniques. Here we present a simple molecular approach for reliable haplotype determination on an individual basis. The procedure is based on allele-specific PCR in combination with PyrosequencingTM technology.
Allele-specific PCR primers were designed with the 3´end at the SNP position and primers specific for each allelic variant of the SNP were tested in two separate reactions. A PCR product in both reactions implied a heterozygote SNP, whereas a PCR product from just one reaction denoted the homozygote for which the allele-specific primer was complementary. A mismatch introduced in the second base from the 3´end was shown to dramatically improve the discriminatory ability between alleles. Amplified allele-specific fragments too long for analysis by Pyrosequencing technology were subdivided by nested PCR. Analysis of the SNPs using the PSQTM 96 System (Pyrosequencing AB) allowed identification of the haplotype. The genotyping after allele-specific PCR showed a typical “homozygous pattern” of either allelic variant depending on the specificity of the allele-specific PCR primer. This procedure was used for haplotyping of fragments up to 10 kb.
Haplotype determination by the described procedure proved to be highly reliable. The results gained from Pyrosequencing technology have the benefit of being quantitative. Any amplification of a non-specific allele would therefore be detected by the system and illustrated as a peak in the pyrogramTM.

I. Eerola¹, L. Boon², S. Watanabe³, H. Grynberg⁴, J. Mulliken⁵, M. Vikkula²;
¹Laboratory of Human Molecular genetics, Christian de Duve Institute of Cellular Pathology, Université catholique de Louvain, Brussels, BELGIUM, ²Laboratory of Human Molecular Genetics, Christian de Duve Institute of Cellular Pathology, Université catholique de Louvain, Brussels, BELGIUM, ³Division of Plastic Surgery, Showa University School of Medicine, Tokyo, JAPAN, ⁴Medisud Medical Center, Brussels, BELGIUM, ⁵Vascular Anomalies Center, Children's Hospital, Harvard Medical School, Boston, MA.

Vascular anomalies comprise a heterogenous group of disorders the severity of which varies from life-threatening lesions to cosmetic harm. They are defects of vasculogenesis/angiogenesis and thereby provide a tool to study the mechanisms involved in these processes. Capillary malformation (CM, or “port-wine stain”) is the most common vascular malformation occurring in 0,3% of newborns. CMs are small flat cutaneous lesions that consist of an increased number of ectatic capillary-like channels within papillary dermis. Vascular birthmarks, such as salmon patch, are milder variants of CM that occur up to 40% of newborns. Unlike common macular stains, the reddish coloration of CMs does not disappear, but becomes darker with advancing age.
Increased incidence of lesions in first-degree relatives of CM patients and several reported familial cases suggest that genetic factors may play a role in the pathogenesis of CM. We performed a genomewide linkage analysis on 13 families with inherited CM. In non-parametric linkage analysis, statistically significant evidence of linkage (peak NPL score 6.72, p-value 0.000136) was obtained in an interval of 69 cM on 5q11-5q23. Parametric linkage analysis gave a maximum combined HLOD score of 4.84 (a-value 0.67) from the same region and the analysis using only the linked families, defined a smaller, statistically significant locus of 23 cM (LOD score 7.22). This locus contains several genes implicated in angiogenesis, such as RASA1and MEF2C. (vikkula@bchm.ucl.ac.be).

Molecular and Phenotypic Analysis of Mutations in COL8A2 in the Corneal Dystrophies

J. E. Yardley¹, S. Biswas¹^,2, F. Munier³^,4, R. Perveen¹, N. Hart-Holden¹, P. Cousin⁴, J. E. Sutphin⁵, B. Noble⁶, M. Batterbury⁷, C. Kielty⁸, A. Hackett⁹, R. Bonshek², A. Ridgeway², D. McLeod², V. C. Sheffield⁵, E. M. Stone⁵, D. F. Schorderet⁴, G. C. M. Black¹^,2;
¹University Department of Medical Genetics, St Mary's Hospital, Manchester, UNITED KINGDOM, ²Academic Department of Ophthalmology, Manchester Royal Eye Hospital, Manchester, UNITED KINGDOM, ³Hôpital Jules Gonin, Lausanne, SWITZERLAND, ⁴Division Autonome de Génétique Médicale, Lausanne, SWITZERLAND, ⁵Department of Ophthalmology and Visual Sciences, The University of Iowa College of Medicine, Iowa City, IA, ⁶Department of Ophthalmology, The General Infirmary at Leeds, Leeds, UNITED KINGDOM, ⁷Department of Ophthalmology, Royal Liverpool University Hospital, Liverpool, UNITED KINGDOM, ⁸School of Medicine, University of Manchester, Manchester, UNITED KINGDOM, ⁹Hunter Genetics, Newcastle, AUSTRALIA.

The corneal dystrophies are a group of conditions which include Fuchs’ Endothelial Dystrophy of the Cornea (FECD) and Posterior Polymorphous Dystrophy (PPCD). These conditions are characterized by a loss of corneal clarity due to an abnormally functioning endothelium (Waring et al. 1982). It is thought that this is a result of defects in neural crest differentiation (Bahn et al. 1984)
A genome-wide search of a three generation family (FECDPed1) with FECD demonstrated significant linkage with D1S2830 (Zmax = 3.72, q = 0.0). The critical region was refined to 1p34.2 - p32. The COL8A2 gene lies within this region and encodes the a2 chain of type VIII collagen. Developmental studies have suggested a role for type VIII collagen in cell differentiation (Shuttleworth 1997) and the protein is a component of endothelial basement membrane, making COL8A2 an attractive candidate gene.
Analysis of the coding sequence of COL8A2 within FEDPed1 revealed a missense mutation (Gln455Lys) within the triple helical domain. This missense mutation was demonstrated in 2 further families with FECD and one family with PPCD. Haplotype analysis of these families suggested the presence of a common founder mutation within two out of three FECD families and the PPCD family, all of whom originate from Northern England. Haplotype analysis also suggested the presence of an identical but independent mutation a third Australian FECD family.
This is the first description of a molecular basis for the corneal endothelial dystrophies and the first association of defects in COL8A2 with human disease.

Tau negative frontal lobe dementia at 17q21: Significant finemapping of the candidate region to a 4.8cM interval

R. Rademakers¹, M. Cruts¹, B. Dermaut¹, K. Sleegers², S. M. Rosso³, M. Van den Broeck¹, H. Backhovens¹, J. van Swieten³, C. M. van Duijn², C. Van Broeckhoven¹;
¹University of Antwerp, Antwerpen, BELGIUM, ²Erasmus University Rotterdam, Rotterdam, NETHERLANDS, ³University Hospital Rotterdam Dijkzigt, Rotterdam, NETHERLANDS.

We report the results of a genome-wide search in a 4-generation pedigree with autosomal dominant early-onset dementia (mean onset age: 64.9 years, range 53 - 79 years). In this family we previously excluded the known Alzheimer's disease genes based on linkage analysis and mutation screening of the amyloid precursor protein gene (exons 16 and 17) and the presenilin 1 and 2 genes. In addition we excluded mutations in the prion protein gene and exons 9 to 13 of the microtubule associated protein tau (MAPT) gene. We obtained conclusive linkage with chromosome 17q21 markers with a maximum multi-point LOD score of 5.51 at D17S951 and identified a candidate region of 4.8 cM between D17S1787 and D17S958 containing MAPT. Recent clinical and neuropathological follow-up of the family showed that the phenotype most closely resembled frontotemporal dementia (FTD) characterized by dense ubiquitin-positive neuronal inclusions that were tau negative. Extensive mutation analysis of MAPT identified 38 sequence variations in exons, introns, untranslated regions and the 5’ regulatory sequence, however none were comprised within the disease haplotype. Although our findings do not entirely exclude a mutation in a yet unanalyzed region of MAPT, the apparent absence of MAPT mutations combined with the lack of tau pathology is highly suggestive for another defective gene at 17q21 responsible for FTD in this family.

An infrequent haplotype of the PCTA-1 gene, located in the susceptibility region 1q42.2-43 (PCaP), indicates association to prostate cancer

C. Maier¹, K. Rösch¹, T. Paiss², S. Bochum¹, K. Herkommer², J. Häußler¹, O. Cussenot³, W. Vogel¹, G. Assum¹;
¹Abteilung Humangenetik, Ulm, GERMANY, ²Abteilung Urologie, Ulm, GERMANY, ³CeRePP, Evry, FRANCE.

Prostate cancer (PCa) is a complex disease with diverse genetically predisposing factors.
The underlying "defects" may range from severe rare mutations in yet unidentified genes to common variants which, in compromise to their frequencies, contribute with reduced penetrance. While high risk alleles well explain familial aggregation of PCa, observed in about 10% of all affecteds, low risk alleles may account for a higher portion of the disease, especially "sporadic" cases. Concerning a putative predisposition to PCa, we investigated the gene encoding the Prostate Carcinoma Tumor Antigen-1 (PCTA-1), mapping to 1q42.2-43, a major susceptibility region identified in a French and German genome wide search. The open reading frame of PCTA-1, which was found to be free of deleterious mutations previously, harbours several SNPs altering the amino acid sequence in four residues. In turns of a population based association study six SNPs were genotyped in 265 controls, 216 sporadic and 57 familial patients. None of the examined SNPs per se revealed association, but one out of five resulting haplotypes, which we call "C2 allele". Elevated numbers of heterozygous carriers of the C2 allele were found in sporadic (10,3% versus 3,8% in controls, p=0,0039), and familial cases (12,3%, p=0,0173). Logistic regression analyses produced relative risks of 2,9 (CI: 1,39-6,13) and 3,6 (CI: 1,30-9,83) in the affected subgroups, respectively. This association between prostate cancer and the C2 allele may be due to the conspicuous haplotype itself, because it codes for a unique PCTA-1 protein, or could reflect linkage disequilibrium to a neighbouring susceptibility gene.

N. J. Prescott, R. M. Winter, S. Malcolm;
Institute of Child Health, London, UNITED KINGDOM.

Nonsyndromic cleft lip and palate is a common congenital anomaly with a complex genetic component. The etiology is likely to be influenced by environmental factors. Folic acid is essential to early embryonic development and recent studies have suggested a role for “folate genes” in cleft lip. We genotyped over 200 parent-case triads for the 5,10-methylenetetrahydrofolate reductase (MTHFR) C677T polymorphism to determine whether this functional variant is responsible for nonsyndromic cleft lip and palate in affected individuals. We could find no distortion in the transmission frequency of MTHFR parental alleles tested by TDT. Examination of Hardy-Weinberg equilibrium detected an over-representation of variant MTHFR homozygotes amongst mothers of affected children when the mothers were themselves affected (odds ratio 4.61 95% CI 1.35-15.77). We postulate that these results are direct evidence of a multifactorial interaction in these families involving maternal folate status, MTHFR genotype and another locus.

Further heterogeneity in human malignant infantile osteopetrosis: evidence for a novel locus on chromosome 6q21

A. Ramirez¹, J. Faupel¹, C. Stöckel¹, C. Hasan², U. Bode², P. Propping¹, C. Kubisch¹;
¹Institute for Human Genetics, Bonn, GERMANY, ²Department of Pediatrics, Bonn, GERMANY.

Malignant infantile osteopetrosis (OMIM #259700) is a rare autosomal recessive disease. The characteristic clinical findings as osteosclerosis, hepatosplenomegaly and pancytopenia become apparent during the first months of life. Mutations have been found in the a3 subunit of the H+-ATPase and the voltage-gated chloride channel CLCN7, demonstrating that osteoclast dysfunction with inadequate bone resorption is the underlying cause of this severe disorder. In a small inbred family with malignant infantile osteopetrosis no mutation could be found in the two known osteopetrosis genes. We therefore pursued a candidate locus approach to unravel the causative gene defect in this family. We excluded homozygosity for several candidate gene loci which in mice cause severe osteopetrosis, among them c-FOS on 14q24, c-SRC on 20q11, OPGL on 13q14, and CSF-1 on 1p13. We also investigated a locus on chromosome 6q21, which is syntenic to a region on mouse chromosome 10 where the causative gene for the osteopetrotic mouse "grey-lethal" has been mapped. We detected homozygosity in a region spanning approximetly 15cM between markers D6S1717 and D6S287. In this region, there are several candidate genes, like e. g. Fyn-related-kinase (FRK), BET3, and MARCKS. We are currently conducting a systematic mutation analysis of several candidate genes in this region. In summary, we provide evidence for a novel locus for human infantile malignant osteopetrosis on chromosome 6q21 which probably represents the human ortholog of the murine "grey-lethal" locus.

Molecular Diagnosis of Haemophilia A in Bulgaria by DNA Analysis Using Polymorphisms Linked to Factor VIII Gene Locus

R. D. Petkova¹, R. Kabakchieva², I. Kalev³, I. Kremensky⁴;
¹University Hospital of Obstetrics and Gynaecology "Maichin dom", Sofia, BULGARIA, ²First Clinic of Paediatrisc, Sofia, BULGARIA, ³First Clinic of Paediatrics, Sofia, BULGARIA, ⁴Univeristy Hospital of Obstetrics and Gynaecology "Maichin dom", Sofia, BULGARIA.

Twenty - seven families affected by haemophilia A, consisting of 138 members (among these, 30 haemophilia A - affected males) were typed for polymorphisms within and outside the Factor VIII gene in order to determine carrier status of female relatives and to estimate possibilities for prenatal diagnosis when needed. Twenty of these families (74 %) did not have previous history of haemophilia A.
A panel of 8 informative DNA polymorphisms, linked to the Factor VIII gene locus was used, as follows: dinucleotide repeats in introns 13, 22 and 25; polymorphisms in presence/abcense of restriction sites in introns 18, 19 and 22; a SSCA polymorphism in intron 7 and a tandem repeat in the DXS52 locus, located outside of the Factor VIII gene locus.
Twenty five (82 %) families showed informativity at at least one marker locus. Only 2 families (8 %) were uninformative at all the markers used.
Pedigree data identified 29 women at reproductive age (mothers of affected boys not included) which were of high risk of having a haemophilia A - affected boy. DNA analysis allowed determination of carrier status in 25 women (83 %), of those 16 women were identified to be haemophilia A carriers and 9 noncarriers.
Ten prenatal diagnoses were performed by analysis of DNA polymorphisms. Seven of these pregnancies were carried to term, producing 5 girls and 2 healthy boys. One male fetus was diagnosed to be at high risk to be affected by haemophilia A, nevertheless, parents did not opt for an abortion.

J. C. Stephens, J. A. Schneider;
Genaissance Pharmaceuticals, New Haven, CT.

We have investigated the level of DNA-based variation (both SNPS and haplotypes) for over 5,400 human genes. In addition, we have characterized how this variation is distributed in a number of biologically and clinically important ways. First, we have determined how SNPs are distributed in human genes: where they occur relative to various functional regions; levels of variability of human SNPs; pattern of the molecular sequence of SNPs; and how these compare to the corresponding sequence of a chimpanzee. Second, we have determined how these aspects of SNP distribution vary among four human population samples. All genes were sequenced on DNA obtained from 82 unrelated individuals: 20 African-Americans, 20 East Asians, 21 European-Americans, 18 Hispanic-Latinos and 3 Native Americans. In particular, we looked at patterns of SNP and haplotype sharing among the four larger population samples. Third, we have determined the patterns of linkage disequilibrium among SNPs, which of course determines the haplotype variability of each gene. This pattern also varies substantially among populations. In order to connect important clinical variability (e.g., genetic disease or susceptibility, variable drug response) to the DNA variability of human genes, an understanding of these patterns of variability within and among human genes is a fundamental prerequisite.

Evidence for linkage of aggressive prostate cancer to chromosome 7q32 in German prostate cancer families

S. Jainta¹, S. Woerner¹, B. Patino-Garcia¹, K. Herkommer², J. Haeussler¹, W. Vogel¹, T. Paiss²^,1;
¹Department of Human Genetics, University of Ulm, Ulm, GERMANY, ²Department of Urology, University of Ulm, Ulm, GERMANY.

Chromosome 7q32 has been suggested to contain genes that influence the progression of prostate cancer from latent to invasive disease by Witte et al. [AJHG 67:92-99,2000]. This locus did not show up in previous genome wide scans and emerged by QTL analysis in sib pairs. We looked for linkage of prostate cancer aggressiveness to chromosome 7q by stratification in 108 German prostate cancer families according to lymph node affection and grading of the tumor. A panel of 8 polymorphic markers ( PE Applied Biosystems) on 7q was used. We found no evidence of linkage between a prostate cancer susceptibility locus and markers on chromosome 7q in the complete set of 108 families. However, the subset of families with aggressive prostate cancer (positive lymph nodes in one family member or two cases of GIII tumors) had an NPL of 1.50 (p = 0.06). Most evidence for linkage came from a few families with aggressive and late onset disease, mean age of onset >65 years and NPL of 2.56 (p=0.01). D7S640, the marker with the higest NPL in our analysis, is located exactly between the markers that gave the strongest signals in the study of Witte et al. [AJHG 67:92-99,2000]. In the German population, chromosme 7q32 is linked to a type of prostate cancer that is characterized by a late onset and an aggressive course of the disease.

Syndactyly type 1 (SD1) is the most common type of syndactyly, inherited in an autosomal dominant fashion, and characterized by complete or partial webbings between the third and fourth fingers and/or between the second and third toes. We recently encountered an Iranian family in which 33 members in six generations were affected with SD1. As a locus of SD1 in a German family has recently been assigned to chromosome 2q34-q36, we performed a linkage analysis of the Iranian SD1 in order to know whether the disorder is genetically homogeneous. With the analysis on 15 affected and 16 unaffected persons using dinucleotide repeat polymorphisms as markers, we mapped the SD1 locus to 2q34-q36 with a maximum LOD score of 6.92 at a recombination fraction q = 0.00 (penetrance = 1.00) for the D2S2179 locus. The result not only confirmed the gene assignment but also suggests genetic homogeneity of the disease.

Identification of athe locus for ichthyosis-prematurity syndrome on chromosome 9.

J. Klar¹, M. Pigg, MD¹, T. Gedde-Dahl, Jr², A. Vahlquist³, N. Dahl¹;
¹Department of Genetics and Pathology, Uppsala, SWEDEN, ²Institute of Forensic Medicine, Oslo, NORWAY, ³Department of Medical Sciences, Uppsala, SWEDEN.

Autosomal recessive congenital ichthyosis (ARCI) is a heterogeneous group of skin disorders. Ichthyosis-Prematurity Syndrome (IPS), is a rare form of ARCI with a relatively high prevalence in the Norwegian population.
Key features are thick caseous desquamating epidermis and complicated pregnancies probably due to polyhydramnion and an opaque amnion fluid caused by the shedding of large amount of epidermally derived cells. This results in premature birth of the affected child.
Thirteen families with at least one affected member and one healthy or affected sibling were identified. Altogether, the 13 families included 17 affected members and 13 healthy siblings. All families are related to a defined region in middle-Norway and Sweden.
A genome-wide linkage analysis gave indication for linkage to chromosome 9q, and further analysis resulted in a maximum cumulative lod-score of 3.73 in the 9q34 region.
Haplotype analysis of meiotic recombination events refined the genetic interval to a 9 cM region between markers D9S250 and D9S63. This restricts the IPS locus to a physical distance of 8 Mb in the chromosome 9q33.3-q34.13 region.

L. Villéger¹, M. Abifadel¹^,2, D. Allard¹, J. Rabès¹^,3, J. Weissenbach⁴, M. Varret¹, C. Junien¹^,3, C. Boileau¹^,3;
¹INSERM UR383, Paris, FRANCE, ²Faculté de Pharmacie, Université Saint-Joseph, Beirut, LEBANON, ³Service de Biochimie, CHU Ambroise Paré, Boulogne, FRANCE, ⁴Centre National de Sequençage, Evry, FRANCE.

Autosomal Dominant Hypercholesterolemia (ADH) is one of the most common hereditary diseases, characterized by a selective increase of LDL particles, giving rise to premature mortality from cardiovascular complications. ADH results from molecular defects in the LDLR gene (Familial Hypercholesterolemia), the APOB gene (Familial ligand-Defective apolipoprotein B-100) and the FH3 gene (1p32-p34.1). We identified a large French ADH family (HC6) in which the involvement of these 3 genes had been excluded suggesting the existence of a fourth locus (FH4). We undertook the identification of the FH4 locus using linkage analysis in family HC6 by a candidate region approach. After the exclusion of 30 candidates, we undertook a whole genome approach with 220 new microsatellite marker. Linkage was obtained with a LOD score of 3.86 (q=0) and confirmed by a multipoint LOD score analysis. Suggestive linkage was also obtained for 4 other nonLDLR/nonAPOB/nonFH3 families conforting the localization of the FH4 locus in this genomic region. Finally, a critical region of 1.9 cM was defined which contains the FH4 gene. We constructed a 3.8 Mb physical map covering our genetic region and identified 82 genes, that we are testing by direct sequencing in collaboration with the Genoscope.

Chromosome 20p is linked to prostate cancer susceptibility in the German population

T. Paiss¹^,2, F. Kurtz², S. Wörner², K. Herkommer¹, J. Häussler², W. Vogel²;
¹Department of Urology, University of Ulm, GERMANY, ²Department of Human Genetics, University of Ulm, GERMANY.

Chromosome 20q13 has been suggested to harbour a prostate cancer (PC) susceptibility locus in families that were characterized by a low number of affected familiy members, late onset disease and no male-to-male transmission. As this epidemiologic profile is characteristic for many German PC families we performed linkage analysis using 7 markers on chromosme 20. There was no evidence for linkage in the whole sample of 108 PC families and in the subsets with mean age of onset < or > 66 years or with the number of affecteds/family < or > 3 as criteria for stratification. In a previous study we identified a subset of families that were linked to chromosome Xq27-28 (NPL > 1) but at the same time had an affected individual in the paternal line of their pedigree. In this subset of 26 families with the conflicting characteristics of male-to-male transmission and X chromosomal allele sharing, multipoint parametric linkage analysis (Genhunter 1.3) showed a maximum LOD score of 4.10 (p=0.001) at D20S112. This may indicate interaction of two genetic determinants of PC susceptibility. Marker D20S112 is localized on 20p11.2-12 and is distinct from the susceptibility locus (D20S887) suggested in the initial study on chromosome 20q.

Linkage disequilibrium mapping of the HLA-linked MYAS1 locus for autoimmune myasthenia gravis.

C. Vandiedonck, G. Beaurain, B. Eymard, C. Tranchant, P. Gajdos, H. J. Garchon;
INSERM U25, Paris, FRANCE.

Myasthenia gravis (MG) is an autoimmune disorder of the neuromuscular junction characterized by production of auto-antibodies against muscle acetylcholine receptors (AChR). Recently, we demonstrated linkage and association of the extended HLA-DR3 haplotype with the form of MG associated with thymus hyperplasia and high auto-antibodies titers, defining the MYAS1 locus. In the present study, given the strong linkage disequilibrium across the DR3 haplotype, we sought to refine the localization of the MYAS1 locus by combining TDT and ANOVA of AChR antibody titers. A panel of fourteen microsatellites and six SNPs, evenly spaced across a 2.5 Mb region between the DRB1 gene and the D6S265 microsatellite, ~100 kb centromeric to HLA-A, was genotyped in 717 MG patients and, for 228 of them, in their relatives. Using single-locus TDT, the strongest association was observed with the TNFd*1 allele, in the class III region (P=1x10^-5). No association was observed with D6S265, which therefore defines the telomeric boundary of the MYAS1 interval. Two-locus TDT bearing on ancestrally-recombinant haplotypes reduced the MYAS1 interval to 700 kb excluding both DRB1 (P=0.008) and C2.4.5 (P=0.007) which is located between HLA-B and -A. ANOVA of anti-AChR antibody titers further narrowed the MYAS1 interval. Variance was best explained by MH*169 (P=7x10^-4) and TNFd*1 (P=2.6x10^-3) and by their combination (P<1x10^-12). A primary region where to look for the MYAS1 gene may be therefore a 40 kb DNA segment limited by these two markers. It currently includes four genes which are under active investigation.

Refinement of a locus for a distinct syndrome of autosomal dominant cleft lip and palate to 2q35

R. M. Winter¹, N. J. Prescott¹, M. M. Lees², S. Malcolm¹;
¹Institute of Child Health, London, UNITED KINGDOM, ²St George's Hospital Medical School, London, UNITED KINGDOM.

Orofacial clefting is a common craniofacial anomaly that occurs in many multiple congenital anomaly (MCA) syndromes. Kumar et al (1996) reported a dominant MCA with cleft lip and palate and characteristic facies in two families of Caucasian origin. The phenotype in these families overlaps with both Hypertelorism, Microtia, clefting syndrome (HMC) and Fronto-nasal dysplasia but we believe this dominant cleft lip syndrome may be recognisable as a distinct entity. We have carried out a whole genome scan in these two families and identified co-segregation to a locus on 2q35. For the larger family we have used additional family members to refine this locus to a region of 10 Mb across a haplotype of four markers which co-segregate in 6 affected individual and one apparently unaffected individual. We have excluded the candidate genes DLX1 and DLX2 by sequencing and fine mapping. All other regions of the genome have been excluded except for the short arm of the X chromosome and work is currently underway to confirm exclusion at this locus. These results demonstrate a major dominant cleft lip locus with high but possible incomplete penetrance.

Genetic analysis of two unrelated Italian Families with non specific X-mental retardation

M. Miano¹, I. Annunziata¹, F. Di Leva¹, G. Fimiani¹, S. Russo², F. Cogliati³, G. Casari⁴, A. Ciccodicola¹, M. Ursini¹, M. D'Urso¹;
¹IIGB, Naples, ITALY, ²Istituto Auxologico Italiano, Milano, ITALY, ³Istituto Auxologico Italiano, Milan, ITALY, ⁴Istituto San Raffaele, Milan, ITALY.

X-linked non specific mental retardation (MRX) accounts for ~ 25% of mental retardation in males. Despite this high frequency, little is known about the molecular defects underlying this disorder, mainly because of the clinical and genetic heterogeneity which is evident from linkage studies. A wide variety of MRX loci have been mapped on X chromosome. At least 8 MRX genes have been identified , but each accounts for only 0.5-1.0% of MRX cases. Here we report two MRX families. The first family has nine males in two generations with classic X-linked inheritance of variable degree of non specific mental retardation.We have performed on this family a two point linkage analysis that shows tight linkage for marker GATA72E05 with Lod Score of 3.14 at q=0.00. Two point linkage interval corresponds to roughly 23 cM in the pericentromeric region of X chromosome. According to linkage data and their functional characteristic, we are performing the mutational screening of some genes in this region to prove their involvment in this patology.
The second family is composed by eight males in three generation characterized by a mild to severe X-linked mental retardation. Previous analysis linked this family in Xq28 between marker DXS1073 and F8c (Lod Score=2.71 at q=0.00). We have performed mutational analysis for 10 candidate genes present in this region by sequencing and RT-PCR analysis and we have found some known and unknown polymorphisms. We are now looking for other candidate genes and we are performing further analysis to esclude genomic rearrangements.

C. R. Cantor, A. Braun, M. Shi, C. Rodi, R. Macdonald;
Sequenom, Inc., San Diego, CA.

SEQUENOM has worked with Incyte Pharmaceuticals and Glaxo Smith Klein to develop a set of validated SNP assay portfolios that altogether number about 200,000 polymorphic SNPs. These assays can be run on pooled DNA samples to generate accurate estimates of allele frequencies in populations of interest. Allele frequency differences in phenotypically-stratified populations can reveal genes with strong associations to phenotypes. Alternatively SNPs can be genotyped on individual DNAs and allele and genotype-phenotype correlations done subsequently in silico. The relative advantages and costs of the two procedures will be compared. For individual genotyping multiplexed sets of SNP assays can cut costs considerably. Progress in developing a multiplexed genome scanning set will be described. Using these methods individually, and in combination, a number of interesting gene associations to complex disease phenotypes have been discovered. The potential importance and utility of some of these associations will be demonstrated.