Posters P0826-P1013 - Genetic Epidemiology and Population Genetics

P0826. Hsp70 polymorphism and HLA-DR diversity; Implications for tuberculosis susceptibility in the Cape Coloured population of South Africaility in the Cape Coloured population of South Africa

T. Boshoff¹, A. E. Brune¹, E. G. Hoal van Helden², L. Bornman¹

¹Biochemistry Division, Rand Afrikaans University; Auckland Park, South Africa; ²Department of Medical Biochemistry, University of Stellenbosch; Tygerberg, South Africa

LBO@na.rau.ac.za

Tuberculosis (TB) being one of the leading causes of death from infectious disease in the world, is also a major problem in the Cape Coloured population in the Western Cape of South Africa. Multiple genes influence susceptibility to TB. HLA class II polymorphisms have been associated with susceptibility to inflammatory diseases, but linkage disequilibrium of alleles in the major histocompatibility complex (MHC) complicates identification of disease-associated genes. Genes encoding members of the 70 kDa heat shock protein (HSP)family (Hsp70) are localised within the MHC. In the light of the protective role of Hsp70 in infection and immunity it is hypothesized that Hsp70 polymorhisms may contribute to disease susceptibility. Possible linkage disequilibrium of hsp70 alleles with HLA-DR alleles might lead to extended haplotypes which might act as additive TB susceptibility markers. Hsp70 gene polymorphism (hsp70-1, hsp70-2, hsp70-Hom and the polymorphic PstI site within the coding region of hsp70-2) and HLA class II DR polymorphisms were investigated in the Cape Coloured population inhabiting the Western Cape of South Africa. Polymorphic analysis of hsp70 and HLA-DR genes was performed on genomic DNA from patients suffering from TB (n=60) and matched control subjects (n=61) using PCR-RFLP and PCR-SSP respectively. Preliminary results showed no evidence for an independent role of hsp70 gene polymorphism in susceptibility to TB while DR3 (DRB1*0301-0302) and DR53 (DRB4*0101) were present at a higher frequency (p=0.068 and p=0.034 respectively) in TB patients, and DR12 (DRB1*1201-1203) in control subjects (p=0.028). An additional number of 300-400 cases and controls are currently analysed to meet statistical standards and to investigate linkage disequilibrium of HLA-DR and hsp70 alleles. An improved understanding of the underlying mechanisms contributing to tuberculosis susceptibility may open new avenues in the development of novel therapeutic approaches.

P0827. ACE Polymorphism in National Turkish Athletes

F. Ciloglu¹, I. Peker², N. Ēine³, A. Demirkan³

¹GENLAB Medical Diagnostics and Research Laboratory; Istanbul, Turkey; ²Marmara University; Istanbul, Turkey; ³DETAM; Istanbul, Turkey

fciloglu@hotmail.com

Cardiovascular performance is a key element in athletic success and certain genes are thought to be candiates for regulating cardiac and vascular physiology. One of the candidate genes is is in the renin-angiotensin system. The angiotensin converting enzyme (ACE) insertion allele (ACE I) is a genetic marker that might be associated with improved endurance performance. In this study ACE polymorphism was studied in past and present Turkish National male athletes who have been competing in national and international level for several years.. The athletes were seperated into three groups consisting of 40 long distance runners, 25 sprinters and 30 fotball players who are felt to have characteristics of both long and short distance athletes. The findings were compared to the results of 100 randomly selected males from the sedentary Turkish population. Genomic DNA was obtained from white blood cells and and the I and D variants of the ACE gene were identified by PCR amplification of the polymorphic region. The findings show that compared to the normal population, the long distance runners have an excess of the ACE I allele (p<0.05)and the ACE II genotype (p<0.05). On the other hand the sprinters show results very similar to the normal population. The fotball players show a slightly increased ACE I allele and ACE II genotype but it is not enough to be statistically significant. These findings suggest that atleast in the Turkish population, there does seem to be a positive association between endurance performance and ACE I allele.

P0828. Folic acid prevents more than neural tube defects; a registry-based study in the Northern Netherlands

H. E. K. de Walle, J. Reefhuis, M. C. Cornel

Department of Medical Genetics, University of Groningen; Groningen, The Netherlands

H.E.K.de.Walle@medgen.azg.nl

Folic acid (FA) is not only protective for neural tube defects (NTDs) but could be for other defects as well. Using data from the EUROCAT Northern Netherlands registry, we have investigated the possible preventive effect of FA in five selected groups of malformations; NTDs, heart defects, oral clefts, urinary tract defects, and limb reduction defects. A case-control analysis was performed using the years 1981 through 1998. Information on FA use was collected via the physician, pharmacist and mother. Odds ratios (OR) with 95% confidence intervals (CI) were calculated using chi-square tests. Cases were defined as infants with FA sensitive defects (n=2,751), defects part of chromosomal or monogenic disorders were excluded. Controls were all children and foetuses with anomalies other than the sensitive anomalies (n=3,647). Use of FA during the periconceptional period was known for 69.5% (n=1,914) of the cases and for 70.0% (n=2,553) of the controls. Of the FA sensitive cases 67 (3.5%) were periconceptionally supplemented with FA, while this was true for 144 (5.6%) controls. Significantly less case mothers than control mothers took folic acid periconceptionally. More specifically, this study showed a significant reduction in risk of heart defects (OR=0.52, 95% CI; 0.35-0.78) and a strong indication for a reduction in the prevalence of urinary tract defects (OR=0.54, 95% CI; 0.29-1.02). The OR for NTDs are indicative of a protective effect (OR=0.68, 95% CI; 0.35-1.30). Recall bias is an unlikely explanation, since sick controls were used as controls. The explanation probably lies in the power of this study; there were not enough cases to detect a significant protective effect of FA for NTDs. Birth defect registries will continue to monitor the future effect of FA on birth defects. Therefore these registries will remain important tools for determining effects of primary prevention on other severe congenital anomalies besides NTDs.

P0829. Postal collection of buccal wash samples for DNA; A successful method for expansion of the ALSPAC DNA resource.

S. Ring¹, E. Spanakis², T. Beavan¹, B. Barratt³, R. Jones⁴, I. Day², M. Pembrey⁴, L. Tyfield¹, A. Study Team⁵

¹Department of Molecular Genetics, Southmead Hospital; Bristol, United Kingdom; ²Division of Human Genetics, Southampton University Hospital; Southampton, United Kingdom; ³Department of Medical Genetics, Wellcome Trust Centre for Molecular Mechanisma in Disease.; Cambridge, United Kingdom; ⁴Institute of Child Health, London; London, United Kingdom; ⁵Institute of Child Health, University of Bristol; Bristol, United Kingdom

S.M.Ring@Bris.ac.uk

The Avon Longitudinal Study of Pregnancy and Childhood is a large epidemiological study of approximately 14,000 children designed to investigate the interplay between genes and environment. A DNA resource has been created using blood samples from approximately 10,000 children and 9,000 mothers and is being used for several genetic studies. Cell lines will be created from the study participants in order to ensure long-term provision of DNA. Paternal blood is currently unavailable. Therefore a new postal buccal wash collection and extraction method was piloted to create mother;father;child trios for transmission tests and imprinting studies. This method was developed for trio studies by the Southampton group and independently validated for ALSPAC in Bristol. Mouthwash kits were sent to 1386 mothers who were asked if they and their partners would return a mouthwash sample for genetic studies. 63% of mothers and 55% of partners returned a sample. Samples were normally less than 48 hours old when returned although some mothers took up to 3 months to collect the samples. DNA is extracted using a phenol-free method and concentration measured using picogreen. The mean yield is 62µg of DNA (range 2-300µg) per 10ml sample. In initial genotyping studies a PCR success rate of 94.6% has been obtained. Studies are underway to confirm sample identity. We expect to generate trios from 52% of the targeted families when parent samples are associated with existing child samples. In conclusion this is a successful method of collecting DNA for epidemiological studies by post

P0830. Whole Genome Amplification; A Means of Providing DNA from Limited Biological Sources

P. K. Bender¹, M. Schmidt¹, W. Beggs², W. Beggs², J. C. Beck¹, R. T. Johnson¹

¹Coriell Institute for Medical Research; Camden, NJ United States; ²Coriell Cell Repositories; Camden, NJ United States

benderpk@umdnj.edu

Whole genome amplification offers a means to prepare sufficient DNA for detailed genetic analysis from biological materials of limited availability. Samples such as biopsies that cannot be cultured, forensic samples, hair and cheek cell samples, and animal scat represent valuable sources of genetic material if sufficient DNA can be obtained. The WGA method is based on a technique called primer-enhanced-preamplification (PEP). PEP relies on fully degenerate primers to target the PCR amplification of sequences throughout the genome without bias for sequence content. Biological samples including a sliver of skin tissue from a deceased centenarian, keratinocytes from foreskin, blood, and primate scat were subjected to WGA in order to evaluate the method. The WGA method yielded up to a 2,000-fold amplification of the genomic DNA providing up to 2.0 micrograms of WGA DNA from a 1.0 ng aliquot of genomic DNA. Several molecular assays were undertaken to determine the allelic bias in the samples and to assess the utility of the material for genetic analysis. For the human samples, the allelic representation of the X- and Y-chromosome amelogenin alleles and five microsatellites on different autosomes was compared between the WGA products and the original template genomic DNA. The DNA samples from primates similarly are being assayed using genomic and mitochondrial markers. The results with the human samples indicate that WGA generates DNA that is representative of the original genomic material at many loci, and potentially offers a method for amplification of DNA from animal samples collected in the field.

P0831. SNP-based genotyping using arrayed primer extension on the oligonucleotide array.

E. Lõhmussaar¹, A. Lushnikov¹, N. Tõnisson², T. Nikopensius¹, I. Kask¹, A. Kurg¹, A. Metspalu¹

¹Inst. of Molecular and Cell Biology, University of Tartu; Tartu, Estonia; ²AsperBio Ltd.; Tartu, Estonia

elin@ebc.ee

SNP-s offer a number of advantages with respect to population-based analysis of the human genome. We present Arrayed Primer Extension (APEX) technology for SNP scoring all over the genome. The method is based upon an array of oligonucleotides immobilized via a 5'-end amino linker onto amino-silanized glass slide. Oligonucleotides are selected from the sense and antisense genomic sequence so that 3'-ends are one base upstream of the SNP. The amplified DNA template, containing the SNP, is digested enzymatically and then annealed to the immobilized primers, which promote sites for template-dependent DNA polymerase extension reactions. Four unique fluorescently labeled dye terminators are used to extend each primer by only one base. As a result of this reaction each primer identifies one base in the target sequence. The Genorama™ imaging system and software package is used for SNP scoring. In the present study APEX technology was successfully used for SNP-based genotyping. We have selected 68 SNPs over the whole genome and estimated the allele frequencies and heterozygosities of these SNPs in Estonian population analyzing 240 chromosomes. From the 68 analyzed SNPs 58 were polymorphic according to the allele frequencies data. This current oligonucleotide array with 68 SNPs can be successfully used for paternity testing and forensic analysis. Oligonucleotide design, quality, DNA Polymerase, dye terminators, template DNA quality and special software tools are all critical for the optimal results. APEX method is a reliable tool in SNP studies, which seem to have a great potential for large-scale genotyping in the near future, although method has utility already now in paternity and forensic testing, mutation analysis and SNP scoring.

P0832. Evolution and population genetics of the H-ras minisatellite

J. A. Langdon, J. A. L. Armour

Institute of Genetics, Queen’s Medical Centre, University of Nottingham; Nottingham, United Kingdom

mrxjal@nottingham.ac.uk

Studies of the H-ras minisatellite have shown that 94% of all alleles in native Europeans and Americans of European descent consist of 4 common alleles. The remaining alleles are designated as rare. Rare alleles have been reported to be associated with cancer in several studies.

In this study 11 substitutional polymorphisms within the sequence flanking the minisatellite have been detected in 40 Castillian and 20 Pygmy chromosomes, and used to construct an allele tree for this locus. The ancestral state has been inferred by analysing several non-human primates, allowing the allele tree to be rooted. The internal structures of these minisatellite alleles have been deduced by minisatellite variant repeat (MVR) mapping. This information has been integrated with the tree constructed from flanking haplotypes, providing additional information regarding lineage evolution.

In addition a survey of allele lengths has been carried out in a UK control population, demonstrating a higher rare allele frequency than was reported in previous studies. The rare allele frequency of 17% detected is significantly greater than that of 6% reported by Krontiris et al. (1993). This is likely to be due to the increased power of the methods used in this study to detect rare alleles, and may have implications for the interpretation of the previously published case-control studies.

P0833. Classical association and TDT-type methods; power for detecting candidate genes influencing quantitative traits.

D. A. Tregouet, C. Pallaud, C. Sass, S. Visvikis, L. Tiret

INSERM U525; Paris Cedex 13, France

tregouet@idf.inserm.fr

Association studies based on candidate genes are one of the major strategies for identifying genes involved in complex traits. However, they suffer from the fact that they do not overcome the risk of spurious association due to uncontrolled stratification of the population. Therefore, the use of Transmission Disequilibrium Test (TDT)-based methods has been advocated by several authors for checking that an observed association is actually due to linkage and not to other uncontrolled phenomenon. In this paper, an estimating equation procedure is proposed to assess the power and the cost efficiency of a classical association and of two TDT-type analyses for quantitative phenotypes in sibship data. It is shown that an association analysis is generally more powerful, i.e. requires less informative sibs, than a TDT-type method when the allele frequency of the studied marker is close to 0.5. Conversely, TDT methods become more powerful when the allele frequency is low or high, all the more since the sibship size is large and the residual sib-sib correlation is strong. However, TDT methods always require more sibs to be sampled, regardless whether or not they are informative, than an association analysis to get the appropriate number of informative sibs, differences between methods tending to decrease as the sibship clustering increases.

P0834. Regression-based inference for association of haplotype frequencies with the phenotype.

D. V. Zaykin¹, P. H. Westfall², S. S. Young³

¹Statistical Genetics, GlaxoWellcome Inc.; Research Triangle Park, NC United States; ²Texas Tech University; Lubbock, TX United States; ³GlaxoWellcome Inc.; Research Triangle Park, NC United States

dvz90620@glaxowellcome.com

There are increasing efforts to relate clinically important phenotypes, such as disease predisposition, drug efficacy or glucose level to single nucleotide polymorphisms (SNPs). Recent research gives examples where several SNPs in haplotypes affect the phenotype so there are biological and statistical interactions. It is important to develop statistical methods to identify and deal with these interactions. We give statistical tests for association of haplotype frequencies with discrete and continuous traits in samples of unrelated individuals. There is a need to deal with unrelated individuals as many data sets, e.g. clinical trials, will not have pedigree data. We present conditions for asymptotic equivalence of regression-based methods with methods that "double the sample size" in the case of known haplotypes. Then we extend these models to the case of statistically inferred haplotypes. We confirm that our methods have excellent power while maintaining type I error. We present applications to gene mapping using epidemiologic data with adjacent markers and show that our methods improve the efficiency of genome scans by incorporating information from consecutive markers.

P0835. Does accounting for gene-environment interaction increase the power to detect the effect of the gene ?

E. Genin¹, J. M. Norris², H. Selinger-Leneman¹, M. Khlat³

¹INSERM U535; Le Kremlin Bicetre, France; ²University of Colorado, Health Sciences Center; Denver, CO United States; ³INED; Paris, France

genin@kb.inserm.fr

Despite tremendous efforts, few genetic risk factors involved in the susceptibility for complex disorders have been identified so far. One reason for this may be that ignored gene-environment interaction conceals effects of genetic factors on disease. To investigate whether this could be an explanation, we have studied the power to detect gene effects of family-based tests of association in the presence of a gene-environment (GXE) interaction. Family-based association tests that use case-parent trios can allow for the detection of a gene effect and a GXE interaction by the comparison of the allele transmissions from parents to exposed and unexposed cases. We have compared the power of the test that account for the GXE interaction and of the test that does not under different exposure frequencies and several models of GXE interaction. We show that the gain of power due to accounting for GXE interaction is highly dependent on the population exposure frequency and on the exposure effect; it is higher when the exposure is rare and when the exposure effect is low. When the exposure frequency is high (>30%) and/or the exposure effect is strong, there may even be a decrease in power when accounting for the GXE interaction. For given exposure frequency and effect, we found that the gain in power is usually stronger for dominant as compared to recessive models.

P0836. Quantifying genetic diversity in a putative German isolate by multivariate feature vector analysis

K. Hoffmann¹, H. H. Stassen², C. Planitz³, R. P. Bochmann³, M. Skorna⁴, B. Lucke⁴, L. Haucke⁴, W. Gunia⁴, M. Zschornack⁵, A. Reis⁶, F. C. Luft⁷, C. Scharfetter², T. Wienker⁸

¹Gene Mapping Center at the MDC and Franz Volhard Clinic at the Humboldt University; Berlin, Germany; ²Psychiatric University Hospital; Zuerich, Switzerland; ³University Clinic Carl Gustav Carus Dresden, Physiology, TU Dresden; Dresden, Germany; ⁴Gene Mapping Center at the MDC; Berlin, Germany; ⁵Malteser Hospital; Raeckelwitz, Germany; ⁶Institute for Human Genetics at the University Nuernberg-Erlangen, and Gene Mapping Center at the MDC; Erlangen and Berlin, Germany; ⁷Franz Volhard Clinic at the Humboldt University; Berlin, Germany; ⁸Institute of Medical Biometry, Informatics, and Epidemiology, University of Bonn; Bonn, Germany

khoffma@mdc-berlin.de

Purpose; Population isolates are important for mapping both monogenic and complex traits. The success of genetic studies in isolated or founder populations depends on the effective number of founders, and the population history. Since reliable information on population history are not always available, the genetic diversity should be estimated in the present generation. In an isolated population, we would expect a reduced heterogeneity. Ethnological studies on populations use the genotypes of mitochondrial and Y-chromosomal polymorphisms. However, these methods are not absolutely representative for autosomal loci, where we would expect genes responsible for most diseases. Method; To address this issue, we applied a multivariate feature vector approach to quantify genetic diversity. Genotypes of 18 highly polymorphic microsatellite loci were analysed by a genetic similarity function. The inter-individual genetic distances d(xi,xj) between the 18-dimensional feature vectors xi,xj (made up by the allelic patterns of any two subjects i, j) represent the level of genetic diversity within a given population. Population; In an ongoing hypertension study, the population of interest are the Sorbs, a putative slavic population isolate in Eastern Germany (sample n=326). Normalized data from 257 subjects from Kanton Zuerich, and an earlier study of US-American samples and of Hutterites (as a proved isolated population) served as reference values. Results; The US-American population displayed a significantly higher genetic variation compared to the Sorbs and the Kanton Zuerich sample (increase of 50 %). In the Hutterites, the genetic diversity was even more reduced and displayed only 30 % of the variation observed in the Sorbs. Surprisingly, the variability is similar in the Sorbs and the Zuerich sample. Conclusion; As expected, the heterogeneous US population showed the highest and the Hutterites the lowest genetic variability. A significant reduction was observed both in the Swiss and the Sorbs sample, indicating a reduced genetic diversity.

P0837. POPULATION GENETICS; Proposed Guidelines for the Ethical Conduct of Genetic Research in Populations

M. Deschenes¹, G. Cardinal¹, B. Knoppers¹, C. Laberge²

¹University of Montreal; Montreal, PQ Canada; ²Quebec Network of Applied Genetic Medicine (RMGA); Montreal, PQ Canada

mylene.deschenes@umontreal.ca

Now that the human genome is mapped, genetic research will focus on understanding the function of each gene. The genetic profile of an individual is shaped by geographical, cultural and socio-economic considerations, all of which are underpinned by biological evolution. Hence, the renewed interest in population research. The genetic profile of small isolated populations may ease the discovery of gene function. Indeed, population genetics holds the key to individual genetic medicine. Genetic epidemiology in heterogenous populations is also of interest to health authorities. Consequently, we observe emerging national DNA banks in Iceland, Estonia, England, Sweden, Newfoundland (Canada) and Quebec (Canada). Genetic research has been framed by a wide set of legislation and international guidelines. However, genetic research the scale of a population brings issues which are not addressed by current frameworks. The Quebec Network of Applied Genetic Medicine is working on the creation of a population bank to study SNP variation within the general Quebec population. To do so, it is developing guidelines for population genetic research. Our presentation will first discuss the ten fundamental principles on which the guidelines are founded. Second, we will focus on certain themes elaborated in the guidelines such as population assent, communication of research results, commercialisation and benefit sharing.

P0838. Optimal combination of marker-specific statistics in a two-stage SNP selection method for disease association data

A. Wille^1,2, J. Hoh¹, J. Ott¹

¹Rockefeller University; New York, NY United States; ²IMBIE; Bonn, Germany

wille@linkage.rockefeller.edu

Although single nucleotide polymorphisms (SNPs) provide a good setting for the fine mapping of complex traits, their density and abundance burden data analysis more than ever. In order to decide which SNP loci in a large pool synergistically influence a particular disease, the mere marker-by-marker method is not efficient because it ignores correlations among markers and requires conservative multiple testing adjustment. Large-scale approaches in which information on multiple markers is bundled to a test statistic should overcome these difficulties. Earlier, we proposed a two-stage analysis of association data in which, first, a subset of markers is selected, and secondly, interactions between these markers and the disease are modeled. Here, we address a question that arose in the first stage of the analysis; how to combine marker-specific statistics in order to achieve the optimal detection power. In principle, each SNP is initially assigned an "index" that measures the association strength between the disease status and its genotypes. For example, this index can be the usual Chi-square statistic. Then information is to be combined either by simply summing those indices or by the empirical Bayesian method. In doing so, it leads us to explore the whole class of the James-Stein estimators.

P0839. Logistic Regression analysis for case/control genetic association studies; combining genetics and epidemiology

F. Macciardi¹, R. Adolfsson², D. Blackwood³, G. N. Papadimitriou⁴, R. Kaneva⁵, M. Noethen⁶, L. Oruc⁷, A. Serretti⁸, C. Van Broeckhoven⁹, J. Mendlewicz¹⁰, B. Lerer¹¹

¹Centre for Addiction and Mental Health (CAMH); Toronto, ON Canada; ²University of Umea, Dept. of Psychiatry; Umea, Sweden; ³University Department of Psychiatry and MRC Human Genetic Unit; Edimburgh, United Kingdom; ⁴Athens University Medical School, Department of Psychiatry and University Mental Health Research Institute; Athens, Greece; ⁵Laboratory of Molecular Pathology, MaternityUniversity Hospital of Obstetrics, Medical University; Sofia, Bulgaria; ⁶Institute of Human Genetics, University of Bonn; Bonn, Germany; ⁷Department of Psychiatry, University Hospital "Rebro", University of Zagreb; Zagreb, Croatia; ⁸Instituto Scientifico Ospedale San Raffaele, Department of Neuropsychiatric Sciences; Milan, Italy; ⁹Neurogenetics Laboratory, Flanders Interuniversity Institute for Biotechnology (VIB), Born-Bunge Foundation (BBS), University of Antwerpen (UIA), Department of Biochemistry; Antwerp, Belgium; ¹⁰Dept. of Psychiatry, Erasme Hospital, Free University of Brussels; Brussels, Belgium; ¹¹Biological Psychiatry Laboratory, Hadassah Medical Organization; Jerusalem, Israel

fabio_macciardi@camh.net

The ultimate goal of genetic association studies, which should be viewed within the larger framework of epidemiological studies of risk factor/disease associations, is to identify gene(s) responsible for a given disease. To detect a genetic association the easiest design is the case-control approach, which usually apply a chi-square analysis to look at the significance of an association. The test is simple and does not depend on any population genetic theory or model, other than that which supposes the strength of an association between genes increases when physical and genetic distances between them decrease. The major problem with the case-control design is the identification of an appropriate control sample, to avoid spurious associations due to potential confounding factors related to population admixture or stratification. To deal with these issues, various methods have been proposed, as, for example, using a set of unrelated polymorphisms to evaluate the extent of "random" association (Devlin & Roeder, Biometrics, 1999). We propose an alternative strategy for genetic association in a case-control sample by applying a logistic regression (LR) model, which allows for the simultaneous consideration of other risk and/or confounding factors. Using a data set from a case-control multicenter European Collaborative study, with a response variable (i.e., the diagnosis), a predictor (i.e., a risk allele for the 5HT2C gene variant) and a confounder (i.e., the ethnical origin of cases and controls), we show that the LR outcome is equivalent to the Mantel-Haenszel test (MH) for multiple contingency tables. Both the statistical approaches lead to the same result, showing a positive association despite substantial population variability across the participating centers. The advantages of this alternative model will be discussed.

P0840. Advances in Genetic Epidemiology

N. E. Morton

University of Southampton; Southampton, United Kingdom

nem@soton.ac.uk

After two decades developing statistical methods applicable to DNA markers, genetic epidemiology is emerging into the sunlight of modern biology. Advances can be grouped into four categories; mapping, positional cloning, nonclassical inheritance, and complex genetic systems. Physical mapping is now sequence-based, constraining order on the linkage map and interpolating for loci without linkage information. It will be shown that this gives dramatic reduction of error. Further improvement will come from ways to define recombination hot spots and cold spots in small regions that cannot be reliably mapped by linkage. These approaches include the decline of allelic association between SNPs and identification of recombination - prone repeat sequences. Positional cloning depends on linkage and linkage disequilibrium (LD) with markers localised on integrated maps, LD providing the greater resolution. Examples will be given for both major loci and oligogenes, using the Malecot equation for allelic association, together with evidence that the latter is substantially more efficient than alternative metrics. Nonclassical mechanisms include imprinted disease loci, dynamic mutation, and possibly concatenated mutation. Complex inheritance has been approached by weakly parametric ("nonparametric") methods including variance components and sib pairs that do not separate gene frequency and effect, as well as by parametric methods that combine segregation, linkage, and association with ascertainment correction. Both approaches allow synthesis of evidence over multiple studies, as illustrated by asthma and (at higher resolution) by SNPs within the ACE locus. Finally, brief consideration will be given to current ethical, legal, and social issues raised by population genetics.

P0841. Inflation of type I error when model-free statistics are computed between markers

M. C. Babron, E. Genin, P. Margaritte-Jeannin, M. H. Dizier, F. Clerget-Darpoux

INSERM U535; Le Kremlin Bicetre, France

babron@ccr.jussieu.fr

Several model-free methods of linkage analysis compare the observed allele sharing of pairs of affected individuals with the one expected under no linkage given only the genealogical relationship. In multipoint analyses, the allele sharing is estimated taking into account the joint information provided by all the genetic markers in a region. Different statistics, such as the MLS or the NPL, have been proposed. There is evidence for linkage if the value of the statistics at a given marker exceeds a given threshold, function of the number of markers tested. However, the statistics are often evaluated at different locations on and between the markers. When looking at the curves that give the value of the statistics as a function of the position in the map of markers, one striking feature is the little bumps that can be seen between markers indicating that the value of the statistic is higher between than on the markers. The maximum value of the statistics over a region is indeed often found between two markers rather than on the markers themselves although the information content is smaller between the markers. In order to investigate whether this could lead to an inflation of the type-one error, we have performed simulations under the null hypothesis of no linkage and compare the type-one errors under two analysis strategies ; including or excluding intermarker locations. The results are reported for different marker map densities and between-marker steps. The interest of the two strategies is discussed.

P0842. Genetic variation and linkage disequilibrium in 174 human genes

T. Acharya, J. Schneider, R. Jiang, A. Wood, D. Tanguay, S. Stanley, W. Newell, B. Koshy, R. S. Judson, G. F. Vovis, J. C. Stephens

Genaissance Pharmaceuticals; New Haven, CT United States

t.acharya@genaissance.com

2,549 SNPs were discovered in approximately 465 kb DNA sequence from 174 human genes with known genomic organization. DNA sequencing was performed on one chimpanzee and 82 unrelated individuals including African-Americans, Asians, European-Americans, and Hispanic Latinos. The data were collected from several gene regions including exons, exon-intron boundaries, untranslated regions and 5' flanking sequences. The average nucleotide diversity for 166 autosomal genes was p=0.063% and q=0.103%, while the average nucleotide diversity for eight X-linked genes was p=0.052% and q=0.070%. Of the four ethnic groups considered, the African-American sample had the highest average nucleotide diversity followed by the Hispanic-Latino, European-American and Asian groups. Haplotypes were inferred using a variation of the Expectation-Maximization algorithm, and linkage disequilibrium (D') was calculated between pairs of SNPs within each gene. Because allele frequencies affect linkage disequilibrium calculations, only SNPs that occurred at least five times in each population were considered in these analyses. The ~8,000 remaining pairwise comparisons were stratified by ethnic group and physical distance between SNPs. In general, the European-American sample had the highest levels of linkage disequilibrium and the African-Americans the lowest, although the median D' value for all ethnic groups was 1, suggesting high overall levels of intragenic linkage disequilibrium. The observed relationship between linkage disequilibrium and physical distance was highly unpredictable; however, most of our inter-SNP physical distances were less than 10 kb. The complex relationship between linkage disequilibrium and physical distance within genes highlights the need to construct detailed, population-specific, empirical linkage disequilibrium maps of the human genome.

P0843. SNPs in the apo(a) unique kringles-population spectra and effect on the Lp(a) plasma levels

M. S. Ogorelkova¹, H. Kraft², C. Ehnholm³, G. Utermann⁴

¹Institute of Medical Biology and Human Genetics, Univ. of Innsbruck, 6020 Innsbruck, Austria; Montreal, PQ Canada; ²Instute if Medical Biology and Human Genetics; Innsbruck, Austria; ³Natl. Public Health Institute; Helsinki, Finland; ⁴Institute of Medical Biology and Human Genetics; Innsbruck, Austria

miraog@netscape.com

Lipoprotein(a) [Lp(a)] is a complex of apolipoprotein(a) [apo(a)] and low density lipoprotein which is associated with atherothrombotic disease. Lp(a) plasma levels are controlled to a large extend by the apo(a) gene locus. Known polymorphisms in the apo(a) gene including the kringle IV-2 variable number of tandem repeats explain only part of the large interindividual variability and do not explain the differences in Lp(a) concentrations between major human ethnic groups. We here performed screening for single nucleotide polymorphisms (SNPs) in exons and flanking intron sequences of the apo(a) kringle IV types 6, 8, 9, and 10 which represent 1.3 kb of coding sequence in two African (Khoi San, Black South Africans) and one Caucasian (Tyroleans) populations and investigated whether they affect Lp(a) levels. Together 768 alleles were analysed. We identified fourteen SNPs including eleven non synonymous, eight of which involved conserved residues, one splice site and two synonymous base changes. No sequence variants common to Africans and Caucasians were found. Some of the newly identified SNPs showed significant effects on Lp(a) plasma concentrations, e.g. the substitution G17R in K IV-8 was associated with Lp(a) levels significantly below average in the two African samples. On the contrary the R18W substitution in K IV-9 which occurred with a frequency of ~8% in Khoi San resulted in a significantly increased Lp(a) concentration. Together our data show that several SNPs in the coding sequence of apo(a) affect Lp(a) levels.

P0844. CYP21 gene SNPs and C4B Taq I RFLP disclose 54 haplotypes for the steroid 21-hydroxylase deficiency alleles in Brazil

M. P. De Mello, M. de Araujo, S. H. V. Lemos-Marini, G. Guerra-Jr, M. T. M. Baptista

Universidade Estadual de Campinas; Campinas, Brazil

mmello@unicamp.br

Congenital adrenal hyperplasia (CAH) due to steroid 21-hydroxylase deficiency is a common inherited defect of adrenal steroid hormone biosynthesis. Unusually for a genetic disorder, the majority of mutations causing CAH result from recombinations between the CYP21 gene encoding the 21-hydroxylase enzyme and the closely linked, highly homologous pseudogene CYP21P. The CYP21 and CYP21P genes are located in the major histocompatibility complex class III region on chromosome 6p21.3, a region that undergoes high recombination rates. Studies on the molecular basis of steroid 21-hydroxylase deficiency in Brazil have revealed the presence of fourteen mutations distributed among 117 chromosomes. The percentages of individual mutations does not differ from those found in different populations. We have performed haplotype analysis on a total of 111 chromosomes using Taq I CPY21 and C4 RFLP/Southern blotting, mutation-specific PCR and PCR/RFLP for two intragenic SNPs in order to evaluate the chromosomal background of ten recurrent mutations and four new mutations. Fifty four different haplotypes were identified.

The number of haplotypes varied from five to nine for each the five most frequent mutations. These data reflect the wide heterogeneity of the Brazilian population, and show that most recurrent mutations on the CYP21 gene are from various independent origins and do not present founder effect. Supported by Grants from; CAPES, CNPq and FAPESP

P0845. Origin of parkin gene mutations in Europe

M. Periquet¹, C. B. Lücking¹, V. Bonifati², A. Dürr¹, G. De Michele³, T. Gasser⁴, M. M. B. Breteler⁵, N. W. Wood⁶, A. Brice¹

¹INSERM U289; Paris, France; ²Universitą La Sapienza; Rome, Italy; ³Universitą Federico II; Naples, Italy; ⁴Klinikum Grobhadern; Munnich, Germany; ⁵Harvard School of Public Health; Rotterdam, The Netherlands; ⁶Queen Square; London, United Kingdom

periquet@ccr.jussieu.fr

A wide variety of mutations in the parkin gene, including exon deletions and multiplications, as well as point mutations, result in autosomal recessive early-onset parkinsonism. Interestingly, several of these anomalies were found repeatedly in independent patients and may therefore result from recurrent (de novo) mutational events or founder effects. In the present study, haplotype analysis with 10 microsatellite markers covering a 4.7 cM region containing the parkin gene, was performed in 48 families with early-onset autosomal recessive Parkinson's disease, mostly from European countries. The patients carried 14 different mutations of the parkin gene that were detected more than once. Our results support the hypothesis that exon rearrangements occurred independently whereas some point mutations, found in families of different geographical origins, may have been transmitted by a common founder.

P0846. Analysis of polymorphisms in the human beta-defensin 1 and 2 genes

A. L. E. Vankeerberghen¹, O. Scudiero¹, G. Castaldo², H. Cuppens¹, F. Salvatore², J. J. Cassiman¹

¹University of Leuven; Leuven, Belgium; ²University of Naples, Federico II; Naples, Italy

anne.vankeerberghen@med.kuleuven.ac.be

The consequences of pulmonary disease are the main death cause of cystic fibrosis patients. Nevertheless, no complete association between the severity of lung disease and the CFTR genotype has been found. Therefore, it is hypothesized that other genetic factors, like defensins, components of the innate immune system, might influence the sensitivity of CF patients to pulmonary infections. Mutations in these peptides could increase or decrease their activity and thereby influence the sensitivity to bacterial infections. In order to analyze the possible modulatory role of human beta-defensin (hBD)1 and 2 in pulmonary diseases associated with CF, the hBD1 and 2 intron boundaries and exons of 60 Italian homozygous ∆F508 CF patients, showing different levels of pulmonary disease, and of 90 Belgian and 90 Italian control individuals were amplified by PCR and sequenced on the ABI 377. Most of the nucleotide changes found in the 5’UTR, the 3’UTR and the introns are very frequent (more than 20%). However, no frequency differences were found between the control and the affected populations and between the Italian and the Belgian controls. Also, no correlation with the age of first colonization of CF patients by Pseudomonas aeruginosa was observed. In hBD2 intron1, multiple insertion/deletion polymorphisms were found. The coding regions of both genes are prone to rare nucleotide changes with some of them causing amino acid changes, four in hBD1, G22C and C67S in Belgian controls, and H34R and V38I in Italian CF patients, and only one in hBD2, M16I (Italian CF patient and Italian control). The promoters are very polymorphic but these nucleotide changes are mostly very rare. We can conclude that both genes are very polymorphic. No association with CF has been found up to now. Further analysis could, however, increase our insight into the role that defensins play in the innate immune system, our first defense against microbia.

P0847. Homogeneous Phase Assays Utilizing Novel Fluorescent Primers for Detection of Single Nucleotide Polymorphisms in Allele Specific Amplification Reactions.

J. Solus, M. Darfler, A. Gallego, I. Nazarenko

Life Technologies; Rockville, United States

jsolus@lifetech.com

Homogeneous phase fluorescence-based assays provide powerful tools for detection of specific nucleic acid sequences. We have developed assays based on the use of primers labeled with a single fluorescent dye which do not require a specific quenching moiety. These fluorescent primers exhibit a large increase in intensity when incorporated into double stranded DNA. This property provides a platform for real-time or endpoint detection of nucleic acids in a closed tube and requires no additional steps subsequent to PCR. We demonstrate the use of fluorescein-labeled primers to detect product in allele specific amplification (ASA) reactions on SNP targets. The effect of changes in the design of the ASA primer for improved discrimination is shown. We also demonstrate the use of a novel DNA polymerase which improves discrimination. Homogeneous phase detection of the allele specific products is demonstrated by real-time analysis during the PCR or by endpoint analysis in a fluorescence plate reader. Alternatively, multiplex PCR of several SNP targets of different sizes can be resolved on an automated fluorescence sequencer. This approach, although not utilizing the potential for homogeneous phase detection, offers the advantage of higher throughput for analysis of multiple SNP targets. This system is shown to be readily adaptable to a universal allele format where the ASA primer contains a 5’ tail identical to the 3’ portion of the labeled primer. The identification of alternate labels to fluorescein permits single tube genotyping. The combination of improvements to allele specific amplification with a flexible homogeneous phase fluorescent detection system provide a simple and reliable method for genotyping of SNP targets.

P0848. A demogenetic analysis of lipoprotein lipase D9N mutation carriers in the Saguenay population (Quebec, Canada)

M. Tremblay¹, J. Lambert¹, D. Gaudet²

¹University of Quebec at Chicoutimi; Chicoutimi, PQ Canada; ²Lipid Research Group, Chicoutimi Hospital; Chicoutimi, PQ Canada

marc_tremblay@uqac.uquebec.ca

D9N mutation in the lipoprotein lipase gene (LPL) is associated to a dominant risk of coronary disease, with variable expression and penetrance, and to arterial hypertension susceptibility. Clinical studies performed at the Chicoutimi Hospital (Quebec, Canada) suggest a relatively high prevalence of this mutation in the Saguenay population. The present study aims at explaining the distribution and origins of the D9N mutation in this population with the use of genealogical data. A sample of 39 carriers was drawn from files at the Chicoutimi Hospital Lipid Clinic. Genealogical data was obtained from the BALSAC population register. This register was also used to choose control individuals and reconstruct genealogies. These extended and deep-rooted genealogies go back to the 17th century (up to 16 generations), with an average depth of 10 generations. More than 12000 distinct ancestors were identified in both groups (carriers and controls). Measures of kinship, inbreeding and ancestors' concentration showed significantly higher levels within the carriers' group. Intergroup kinship coefficients were also higher than controls' intragroup kinship coefficients. However, no great difference was observed between the two groups as regards to the ancestors' geographical origins. Most regional founders (19th century) came from the adjacent region of Charlevoix, while early founders (17th century) came mainly from the French regions of Normandie, Ile-de-France, Poitou and Aunis.

P0849. Establishment of a DNA bank from the European population as a resource for studies of genetic variability

Y. Hua, S. Cichon, N. Kluck, P. Propping, M. Nöthen

Institute of Human Genetics, University of Bonn; Bonn, Germany

y.hua@uni-bonn.de

A DNA bank is compiled including 96 samples of 30 European countries representing different population size. Cell lines have been established from every individual to ensure an unlimited DNA source. The DNA bank will be available on request to interested research groups from 2001. The DNA bank forms the basis for the second part of the project that aims at a systematical survey of genetic variability in target proteins of CNS-active drugs. Receptor genes of neurotransmitters as well as genes that are located downstream in the intracellular signaling pathways are considered primary candidate genes. We are using a high-throughput sequencing approach to identify genetic variants.

P0850. TaqMan Genotyping of 64 drug metabolizing enzyme polymorphisms and their allelic frequencies in 4 different populations

S. R. Asquith¹, A. Yeo¹, D. J. Dow¹, T. M. Saunders¹, R. Hussein¹, B. Bahari¹, D. S. Montgomery², S. X. Pillai³, M. Dickins¹, E. Foot², I. J. Purvis¹, M. J. Stubbins¹

¹GlaxoWellcome R&D; Ware, United Kingdom; ²GlaxoWellcome R&D; Greenford, United Kingdom; ³GlaxoWellcome R&D; Research Triangle Park, NC United States

ms85772@glaxowellcome.co.uk

The study of polymorphisms in enzymes associated with the absorption, distribution, metabolism and excretion (ADME) of xenobiotics is a critical factor in our ability to rationalise an individual’s response to different medicines. To date, there has been very little data presented on the frequency of these so-called “metabolic polymorphisms” in specific poplulations, the most detail being available currently for the cytochrome P450 isozymes CYP2D6 and CYP2C19. Here we have attempted to expand our knowledge in this area by developing a panel of TaqMan assays for the routine genotyping of over 60 metabolic polymorphisms. This study will also enable us to begin to answer a question which is perhaps more important for the Pharmaceutical Industry - how representative are clinical trials of the “general” population? Further statistical analyses may enable us to correlate the genotype data with various phenotypic parameters (measured during clinical studies), thus enabling us to move towards the prediction of an individual’s “metabolic capability” with respect to a medicine, by virtue of their genotype. In conclusion, this report describes the design and development of “low volume” TaqMan genotyping assays for 64 potentially functional polymorphisms in various ADME-associated genes. We have subsequently used these assays to determine the allele frequencies of these polymorphisms in four defined populations. The information presented here will be a valuable aid to the further understanding of population differences in this important area of research.

P0851. Association of Human IL-1b (-511) Polymorphism with early Onset Periodontal Disease (EOP)

M. R. Bazrafshani¹, W. E. R. Ollier¹, M. H. Thornhill², A. H. Hajeer¹

¹ARC Epidemiology Research Unit; Manchester, United Kingdom; ²Oral Disease Research Center, The Royal London School of Medicine and Dentistry; London, United Kingdom

Mohammad.R.Bazrafshani@fs1.ser.man.ac.uk

Background; Early onset periodontal diseases (EOP) are a group of inflammatory disorders characterised by a rapid rate of periodontal tissue destruction, in young individuals. There is now substantial evidence to suggest that genetic factors play a role in the pathogenesis of EOP. Periodontal disease is a major public health issue of worldwide significance. It is a chronic inflammatory disease of the supporting tissues of the teeth, starting with gingivae and progressing to gradual destruction of the bony support and peridontal attachment of the teeth. This results in significant morbidity, with loosening and loos of teeth the ultimate outcome. Indeed, improved prevention and treatment of tooth decay, has resulted in peridontal disease becoming the most common cause of tooth loss in Europe and the USA. Polymorphisms in cytokine genes, which may underpin inter individual differences in cytokine synthesis and secretion have been associated with other diseases which have an inflammatory pathogenesis, Genetic variations found within candidate genes in EOP patients may represent a mechanism by which individuals are rendered susceptible to disease. Objective; To investigate whether a biallelic polymorphism (A or G) occurring within the promoter region of the IL-1b gene (position -511) is associated with EOP. Methods; The -511 polymorphism was detected using a PCR-RFLP method. IL-1b polymorphism was examined in 97 patients with EOP and 91 healthy matched UK controls.

Result

Conclusion; Several polymorphisms exist in the IL-1 cluster that influences the IL-1b biological activity. Our results demonstrated a possible role for IL-1b gene in the development of EOP. Key Words; 1- EOP 2- Polymorphism 3- IL-1b 4- RFLP

P0852. The Polymorphism Detection of Human Genomic AA-NAT Gene

W. Ying

National University of Singapore; Singapore, Singapore

medp9087@nus.edu.sg

The role of melatonin in normal sleep-wake regulation has been inferred from the temporal relationships between its rhythmic synthesis and the 24 hours cycle. The biochemical basis of this hormone rhythm is the penultimate rate-limiting enzyme in melatonin synthesis, Arylalkylamine N-acetyltransferase (AANAT). [Axelrod, 1974; Klein, 1985; Namboodiri et al., 1985]. Specifically, our current study focuses on characterizing the AANAT gene in a local population and looking for possible genetic variability that may explain the widely documented variation in sleep pattern among normal human individuals. A self-assessed sleep pattern survey was conducted among 210 healthy, young and drug-free human subjects based on a whole week practice. Data pertaining to normal sleep onset, offset, and total nocturnal sleep duration was obtained. Circulating blood samples were collected from 4 individuals with early onset and long sleep length, and 5 individuals with late onset and short sleep length. Polymerase chain reaction (PCR) amplification of the genomic DNA containing all of the four exons of AANAT gene was done. Analysis of PCR products by direct sequencing demonstrated that all the 36 sequences were identical with the published data. Therefore, there is no evidence of genetic variability in the coding region of AANAT gene that may explain the extreme differences in sleep pattern. More studies are being conducted to examine the promoter regions for AANAT.

P0853. Segregation Analyses of Asthma and Respiratory Allergy; The Humboldt Family Study

Y. Chen¹, A. Schnell², D. Rennie³, R. Elston², L. Lockinger³, J. Dosman³

¹University of Ottawa; Ottawa, ON Canada; ²Case Western Reserve University; Cleveland, United States; ³University of Saskatchewan; Saskatoon, Canada

chen@zeus.med.uottawa.ca

We performed segregation analyses of asthma and respiratory allergy based on data from 309 nuclear families comprising 1,053 individuals living in the town of Humboldt, Saskatchewan in 1993, using the REGD program of the S.A.G.E. program package. For adults, information on asthma and history of respiratory allergy was provided by the subjects themselves, and for children by their parents. When asthma was considered as the trait in segregation analysis, models of no major effect, with or without familial effects, were rejected, but were not rejected after adjusting for history of respiratory allergy. The major gene hypothesis was not rejected before adjusting for history of respiratory allergy. When respiratory allergy was analysed as the trait, both major gene and multifactorial models fitted the data well regardless of whether there was adjustment for asthma or not. Other covariates adjusted for in the segregation analyses were age, sex, number of household smokers, current smoking, number of household members, generation, and house type. The data suggest that a major gene related to respiratory allergy may explain the familial aggregation of asthma.

P0854. GLN223ARG SNP in the leptin receptor gene; studies in anorexia

A. I. F. Blakemore¹, N. D. Quinton¹, D. W. Meechan¹, L. F. Pieri², K. M. O. Brown³, H. Eastwood³

¹Sheffield Hallam University; Sheffield, United Kingdom; ²Yorkshire Centre for Eating Disorders; Leeds, United Kingdom; ³University of Leeds; Leeds, United Kingdom

a.i.blakemore@shu.ac.uk

Leptin, a cytokine expressed and secreted by the adipose tissue, is involved in the regulation of body weight. Leptin levels in subjects with anorexia nervosa are low, although the correlation with body mass index is still evident. We, and others, have shown an association between the leptin receptor SNP, GLN223ARG, and fat mass, body mass index (BMI) and leptin levels. The aim of this study was to determine if this association could be found in young female, Caucasian control subjects and if differences in allele frequency existed in female Caucasian subjects with anorexia nervosa. 175 subjects with anorexia nervosa (divided into bingeing/purging and restricting anorexics) were recruited from the Yorkshire Centre for Eating Disorders at Seacroft Hospital. 145 controls were recruited from the University of Leeds. In both cohorts, allele frequencies for GLN223ARG, did not differ significantly from published frequencies (A allele 0.61 and 0.56, G allele 0.39 and 0.44, in anorexia and control cohorts respectively). There was no significant difference in genotype frequency between the control and the anorexic cohorts or between restricting and bingeing subjects with anorexia. The previously published association between BMI and genotype was not observed in either cohort. This may strengthen the existing evidence that this association is only observed in older middle-aged males and postmenopausal subjects.

P0855. Segregation Analysis (sa) Of Heart Rate (hr)

M. B. Bravo-Luna, C. A. R. Marcon, I. R. Tunkieviez, B. M. Perez

Faculty of Medicine; Rosario, Argentina

mbravoluna@hotmail.com

Genetic variation, important feature regarding any trait is desirable. In hypertension loss of heart rate (HR) variability leads to poor prognosis. HR, semiquantitative discrete trait, regulated by the ANS, is involved in blood pressure (BP) regulation, which is admitted that is genetically determined, although the mechanism is controversial, due to the many regulation components. There are papers supporting the BP Mendelian inheritance (MH) hypothesis, though mixed models admitting the involvement of polygenes and environmental influences are more appropriate. We have published (Med Hypotheses(2000)54(2)307-9 a hypothesis considering the probable MH of HR. Objective; To demonstrate through SA methods HR-MH. Methods; 544 individuals (148 pedigrees) were monitored, in basal and comparable conditions, during 5 minutes in 3 different occasions. Data were entered in an ad hoc software to perform SA, mixed model of Morton-Maclean, taking into account 15 parameters to be analysed through 11 proposed models, estimating [-2Ln (LH)] (log. likelihood ratio), in a most parsimonious model. Heritability (H2) (regression coefficient) and D2 [D2=H2+(1-H2)R2], variance proportion of the trait measuring genetic and non-genetic factors, were estimated. Results [X2;d.f.;p;H2;D2], Mendelian model, transmissibility set free; 3.586;2;n.s.;0.60;0.70, proved the best model for this trait admitting MH without ruling out polygenes and environmental influence.

P0856. Analysis of the modifying effects of TAP 1 and TAP 2 genes on Cystic Fibrosis phenotype

F. Ozbas Gerceker¹, U. Ozcelik², D. Anadol², N. Kiper², A. Gocmen², E. Yilmaz³, H. Erdem Yurter³, M. Ozguc¹

¹TUBITAK DNA/Cell Bank & Gene Research Laboratory; Ankara, Turkey; ²Hacettepe University Medical School Pediatric Chest Disease Unit; Ankara, Turkey; ³Hacettepe University Medical School Department of Medical Biology; Ankara, Turkey

ozbas@hacettepe.edu.tr

Cystic Fibrosis is an autosomal recessively inherited genetic disease which results from the mutations in the gene encoding Cystic Fibrosis Transmembrane Conductance Regulator (CFTR)protein. It is important to analyze the modulating and modifying effects on phenotype since we have genotypic and phenotypic heterogeneity among patients. In our study, we investigated TAP 1 and TAP 2 gene polymorphisms in CF. The TAP genes (transporter associated with antigen processing) encode two membrane-spanning ABC proteins that translocate antigenic peptides from the cytoplasm into the endoplasmic reticulum. TAP 1 and 2 genes are localized within the MHC class II region. Polymorphisms in the TAP genes are associated with the specificity of the binding of antigenic peptides. Comparison of 58 CF patients whose CFTR mutations were determined previously and 100 Turkish controls revealed a significant increase in the Ala/Ala dimorphism at position 665 of TAP 2 gene in CF patients as compared to controls (p<0.025). Furthermore, Gly phenotype at position 637 of TAP 1 gene was seen in CF patients in a significantly higher ratio (p<0.05).

P0857. Apolipoprotein E genotypes in a Turkish Population with renal disease; Preliminary findings

A. Beşer¹, H. Sarı², R. Bircan¹, S. Manay¹, M. Koē², I. Güney¹, Ē. Özener², B. Ēırakoğlu¹

¹Marmara University Faculty of Medicine Dept. of Medical Biol. & Genetics; İstanbul, Turkey; ²Marmara University Faculty of Medicine Internal Medicine Nephrology Department; İstanbul, Turkey

arbeser@hotmail.com

Background; Apolipoprotein E plays an important role in regulation of cholesterol and lipid metabolism. This study is performed to investigate the APO E allele genotypes in Turkish population and to determine the association among the APO E genotype, serum cholesterol levels, blood lipid profile and renal disease in a population of 286 people including 186 end stage renal disease patients (100 on hemodialysis and 86 on chronic ambulatory peritoneal dialysis (CAPD))and 100 control cases. Methods; The APO E genotypes are investigated with polymerase chain reaction. Amplified DNA fragments are digested with Hha I enzyme and analyzed with %4 agarose gel electrophoresis. Trigliseride and total cholesterol levels are determined by enzymatic colorimetric assay and HDL by homogenous enzymatic colorimetric assay. VLDL, LDL and body mass index are calculated. Results; The results of apolipoprotein E genotyping obtained in 55 patients are; 20 analyzed in the CAPD group;4 E2/E2, 12 E3/E3, 1 E4/E4, 1 E2/E3, 1 E2/E4, 1 E3/E4; 15 analyzed in the hemodialysis group;11 E3/E3, 4 E3/E4; 20 analyzed in the control group;15 E3/E3, 3 E2/E3, 2 E3/E4. Conclusion; Recording of the biochemical data is complete but the number of subjects whose APO E genotypes have been determined is not sufficient to make statistical calculations. After completion of genotyping, the APO E genotypes will be correlated with the cholesterol levels, lipid profile and the body mass index. Finally the distribution of APO E alleles between the normal population and the patients with ESRD will be evaluated.

P0858. Glutation S-transferase polymorfism in populations with different ethnisity.

S. N. Popova, P. Slominsky

Institute of Molecular Genetics; Moscow, Russian Federation

popova@img.ras.ru

The glutation S-transferase (GST) are widely expressed in mammalian tissues and involved in phase II detoxification reactions. The GST form a supergene family consisting of four distinct families, named alpha (GSTA), mu (GSTM), theta (GSTT) and pi (GSTP). Several of GST are polymorphic in humans. Among the GST genes we examined GSTT1 and GSTM1 polymorphism in some populations from Russian Federation; Russians from Holmogory and Oshevensk (Arkchangelsk region) and three populations with different level of Mongoloid component; Chant, Kalmyk and Buryat. We observed that frequency of GSTM1 null genotype are significantly higher in Chant, Kalmyk and Buryat than in Russians from Arkchangelsk reg. For GSTT1 gene null genotype frequency was statistically higher for Kalmyk (p=0,03), but in Chant and Buryat it was similar as in Russians from Arkchangelsk region. Frequency GSTM1 (0/0), GSTT1 (0/0) in sum for Kalmyk was 18%, there as in other populations it wasn't bigger than 5%. Comparing Mongoloid populations with Russians using Chi squred-criterium and G-statistic one can see that Kalmyk population mostly varies from Russians populations, and differences between Chant, Buryat and Russians are less expressed. We can note that population differences are more conditioned by difference in GSTT1 genotype frequency. Gene GSTM1 has less marked divergence in populations, and this gene frequencies hardly vary for Oshevensk, Holmogory and Chant populations.

P0859. The TNFRSF6 gene is not implicated in familial early-onset Alzheimer’s disease

J. Theuns¹, L. Feuk², M. Cruts¹, J. Del-Favero¹, B. Dermaut¹, G. Roks³, D. Van den Bossche¹, E. Corsmit¹, M. Van den Broeck¹, C. M. van Duijn³, A. Brookes², C. Van Broeckhoven¹

¹Department of Molecular Genetics, Flanders Interuniversity Institute for Biotechnology (VIB), University of Antwerp (UIA); Antwerpen, Belgium; ²Center for Genomics Research, Karolinska Institute; Stockholm, Sweden; ³Department of Epidemiology & Biostatistics, Erasmus University Medical School; Rotterdam, The Netherlands

theuns@uia.ua.ac.be

The tumor necrosis factor receptor superfamily member 6 (TNFRSF6) gene encoding FAS, a cell-surface receptor involved in apoptosis initiation, was recently reported to constitute a genetic risk factor for early-onset Alzheimer’s disease (EOAD). Dynamic allele-specific hybridization (DASH) analysis of the –670 G/A polymorphism showed enrichment of the homozygous GG-genotype in Scottish sporadic EOAD cases, almost completely attributable to enrichment within the subset of APOE4 carriers (Feuk et al.,2000). In this study, we analyzed the promoter polymorphism in a Dutch population-based EOAD case-control sample, using the same technique, but could not detect a significant disease association. Thus our study does not reinforce the hypothesis of an independent involvement of the TNFRSF6 -670 G/A polymorphism in AD risk. In this study we also compared the reliability of DASH and pyrosequencing analysis and demonstrated the robustness of the two techniques by genomic sequencing.

P0860. The frequencies of chemokine receptor CCR5 polymorphisms affecting susceptibility to human immunodeficiency virus (HIV) and the rate of progression to AIDS in Kuwaitis

A. Voevodin

Faculty of Medicine; Kuwait University, Kuwait

voevodin@hsc.kuniv.edu.kw

Chemokine receptor CCR5 serves as the major co-receptor for HIV-1. Several polymorphisms in the CCR5 gene are known to affect the susceptibility/resistance to HIV-1 and/or the rate of progression to AIDS. The frequencies of these polymorphisms as well as their infection/disease modifying effect vary in different populations. Most of the data available in this field were obtained in the studies of Americans of various ethnic/racial origin. Very little in this respect is known about Arabic populations. We studied the frequencies of two CCR5 gene polymorphisms; m303 and 59029-A/G in Kuwaitis. The genotyping for both polymorphisms was carried out by PCR-RFLP tests. The m303 allele was not found in any of 230 Kuwaitis genotyped for this marker (the allele frequency is less that 0.002). The frequency of "protective" allele 59029-G in Kuwaitis (0.66, no. of chromosomes = 266) was higher than that in other populations for which the frequencies of the allele have been reported. The modifying effect of 59029-A/G genotype as well as two other genetic markers associated with slower progression to AIDS (CCR2-64I and SDF-3'A) is being investigated in HIV-infected Kuwaitis. Acknowledgement. This study was supported Research Grant MI 120 from Kuwait University.

P0861. Association of paraoxonase Gln -Arg 192 polymorphism with late-onset sporadic Alzheimer’s disease and coronary artery disease in ultraoctuagenarians.

R. Corbo¹, G. Gambina², T. Vilardo³, L. De Bernardini⁴, R. Scacchi⁵

¹Dept. of Genetics and Molecular Biology, University "La Sapienza"; Rome, Italy; ²Division of Neurology, Hospital; Verona, Italy; ³Surgery for the Prevention of the Vascular and Metabolic Diseases, S. Giovanni Hospital; Rome, Italy; ⁴Geriatric Hospital "Villa delle Querce"; Rome, Italy; ⁵CNR Center of Evolutionary Genetics; Rome, Italy

Corbo@axcasp.caspur.it

Recent advances suggest a possible relationship between Alzheimer's disease (AD) and atherosclerosis. This makes interesting to examine the coronary artery disease (CAD) candidate gene polymorphisms in AD too. In the present investigation we examined a polymorphism in the paraoxonase gene (PON1) due to an A -> G substitution at amino acid position 192, which has been found associated with cardiovascular diseases, though with conflicting results. A group of late-onset sporadic AD patients and a group of patients with CAD were studied in order to compare the results between them and with normal subjects. Most of the patients were ultraoctuagenarians. Significant differences in the allele frequencies between patient groups and correspondent controls were observed. In particular the AD patients aged ³ 80 (n= 152) showed a frequency of the PON1*Q allele significantly higher than the control group (0.832 vs 0.741, p = 0.003). Also CAD patients aged ³ 80 (n=56) had a PON1*Q allele frequency (0.848) higher than their controls (0.743) (p=0.04). A CAD patient group aged < 80 (n= 144) showed an opposite pattern with an PON1*R allele frequency (0.260) significantly higher than their controls (0.154) (p=0.01), confirming the previously reported association of PON1 *R allele with CAD. In summary our data suggest that the association of PON1*Q with both AD and CAD is peculiar to ultraoctuagenarian patients.

P0862. The dynamics of congenital malformations in Tomsk region according the results of 20-year’s epidemiological research

N. Kricunova, L. Nazarenko

Institute of Medical Genetics; Tomsk, Russian Federation

kricunova@img.tsu.ru

The variation of structure and frequencies of congenital malformations (CM) in Tomsk region during the 20- year’s period of monitoring (1979 – 1998) has been analysed. Epidemiological study has revealed the cyclic character of the CM frequency fluctuation in range from 13,9‰ to 35,2‰, with the mean 23,65‰. Significant increasing in the frequency of overall CM spectrum has been disclosed (from 22,8‰ up 26‰, P<0,01). The summary frequency of the isolated forms of CM has increased significantly, but the frequencies of multiple CM and Down’s syndrome have decreased. The average frequency of 19 forms of CM registered according the European register EUROCAT has been estimated as 14,2‰, not differing from the basic level of CM. In CM structure, the most frequent were the defects of osteomuscular system, multiple CM, nervous system, cardiovascular system, face and neck. Regarding the separate CM forms, the most frequent were congenital heart defects, Down syndrome, hydrocephaly and hypospadia.

P0863. Epidemiological data about birth defects in Ural's region (Russia, Ekaterinburg)

M. Devaikina¹, N. Demikova², G. Pavlov¹, E. Nikolaeva¹, I. Sklyar¹, N. Nikitina¹

¹Medico-Genetic Center; Ekaterinburg, Russian Federation; ²Institute of pediatria and child surgery; Moscow, Russian Federation

gvp@online.ural.ru

Since 1992 birth defects of newborns have been registered in Ural’s region.

Since 1999 International Clearinghouse for Birth Defects Monitoring System was introduced into practice. 21 birth defects are due to be registered.

Multiple sources of information are registration forms from maternal houses, child clinics, pathology departments.

From 1992 to 2000  348075 newborns were borne in the region, and 6161 of them had birth defects. Frequency of birth defects was 17 in 1000 births (1,7%).

The structure of birth defect’s diagnoses; heart’s defects – 21%, neural tube defects –13%, defects of musculoskeletal system – 10%, defects of urinary system – 9%, multiple congenital anomalies – 8%.

In 1999  35877 newborns were borne, and 1161 of them had birth defects. Frequency of birth defects was 32 in 1000 births ( 3,2%). The structure of birth defect’s diagnoses; heart’s defects – 25%, defects of musculoskeletal system - 15%, neural tube defects – 12%, multiple congenital anomalies – 11%, defects of urinary system –8%.

Frequency of some birth defects in 1999; anencephaly – 0,6 in 1000, spina bifida – 1,2 in 1000, hydrocephaly – 1,6 in 1000, cleft lip with or without cleft palate – 1,2 in 1000, polydactyly – 1,6 in 1000, anotia and microtia – 0,8 in 1000, total Uml reduction defects – 0,3 in 1000, Down syndrome – 1,3 in 1000 births.

For drawing well- grounded conclusion about epidemiology of congenital anomalies, it’s necessary to monitor birth defects during next two years.

P0864. Polymorphism of Apo AI-CIII Gene Cluster in Korean Essential Hypertensives

J. Shin¹, B. Kang², K. Kim², C. Lee¹

¹School of Biological Sciences, Seoul National University; Seoul, Republic of Korea; ²Seoulin Bioscience Institute, Seoulin Bioscience, Co., Ltd.; Seoul, Republic of Korea

nariann77@yahoo.com

The apolipoprotein AI and CIII (apo AI and CIII) play an important role in the metabolism of plasma lipoproteins and lipids. The apo AI-CIII gene cluster is located in chromosome 11q23 and more than 20 different RFLPs have been described in this gene cluster. To search for a useful genetic marker for the essential hypertension in Korean population, the distribution of two restriction fragment length polymorphism (RFLPs) of the apo AI-CIII gene cluster and their association on essential hypertension was investigated in total 163 Korean individuals. The distribution of the genotypes of all the RFLPs was in Hardy-Weinberg equilibrium in this population. the G-75→A polymorphism of the apo AI gene was significantly associated with essential hypertension in Korean population (P<0.05). Therefore, this result suggest that the G-75→A polymorphism of the apo AI gene may be useful as a genetic marker for essential hypertension in Korean population.

P0865. Apolipoprotein E Genotypes And Diabetic Retinopathy In Iddm Patients

N. S. Shcherbak¹, A. Stroykova², E. Schwartz¹

¹St.Petersburg State Medical University; St.Petersburg, Russian Federation; ²St.Petersburg Diabetic Center; St.Petersburg, Russian Federation

nshch@yahoo.com

Apolipoprotein E (apoE) was discovered as a plasma protein involved in lipoprotein metabolism. There are three common variants of apoE, resulting from common genetic variation, called E2, E3 and E4. Three-allelic variation in the apolipoprotein-E (APOE) gene have been suggested as risk factors for the development of diabetic micro- and macrovascular complications. We investigated the association between APOE genotypes and development diabetic retinopathy by case-control study in a Russian population. We studied 76 type 1 diabetic patients with diabetic retinopathy (34/42 m/f, age (mean+/-SD) 30+/-10 years, diabetes duration 18+/-8 years) and 96 patients without diabetic retinopathy (50/46 m/f, age 25+/-8 years, diabetes duration 16+/-6 years). APOE polymorphisms were detected by the restriction fragment lenght polymorphism method after a polymerase chain reaction. No did APOE allele frequencies (epsilon2/epsilon3/epsilon4) differ between diabetic patients with and without retinopathy; 0.138/0.776/0.086 vs 0.099/0.802/0.099, respectively. Genotype distributions were also similar, n.s. No associations between diabetic retinopathy and APOE polymorphisms were observed. These results suggest that APOE genotype are not assosiated with the development diabetic retinopathy in patients with IDDM in a Russian population.

P0866. Genetics of Human Female Pelvic Morphology and Menarcheal Status; a Twin Study

K. Sharma

Department of Anthropology, Panjab University; Chandigarh, India

neelamkuki@usa.net

To examine the relative role of genetic and environmental factors on pelvic morphology and menarcheal status, data on 60 pairs of female twins (30 MZ and 30 DZ) were analyzed. Various pelvic mesurements were normally distributed except for two of sixteen instances. Association of twin type with the mean value was found in only one out of eight traits, which might be attributed to type 1 error. Heterogeneity of variance between zygosities was observed for four pelvic traits (50%) suggesting considerable amount of hidden environmental influence. No evidence of stronger environmental covariance for MZ than DZ twins was observed. Significant genetic component of variation was observed for pelvic height, bi-ischeal breadth, age at menarche and pelvic area. In instnces where inequality of variances between zygosities were demonstrated, total among pair and within pair mean squares were larger for dizygotic than for monozygotic twins. This is interpreted as evidence of greater environmental influence between zygosities. Environmental modification shawing phenotypic plasticity was not of same magnitude in various pelvic traits thus indicating differential selection pressures. Similar differences were observed in the magnitude of cultural inheritance.

P0867. CC - chemokine receptor CCR5 gene polymorphism analysis in diabetes type I and type II patients in Estonia.

I. Kalev¹, K. Oselin¹, P. Pärlist¹, T. Podar², A. Mikelsaar¹

¹University of Tartu; Tartu, Estonia; ²Clinicum of Tartu University, Department of Endocrinology; Tartu, Estonia

ingridk@ut.ee

Chemokines are small chemotactic proteines, that mediate innate and adaptive immune responses and inflammatory processes through the G-protein-coupled receptors. Previous studies have shown, that the temporal expression of certain CC-chemokines and the CCR5 chemokine receptor in the pancreas is associated with the development of the insulitis and spontaneous type I diabetes. It is known, that some time before onset of diabetes type I, inflammatory process takes place in the pancreas. In the present study we have determined the frequency of the ccr5delta32 allele in diabetes type I (n=38; age 15-62, average 34,5y) and diabetes type II (n=113; age 42-77, average 63,3y) patients. In control group of healthy, unrelated native Estonians (n=504; age 14-94; average 54y) the ccr5delta32 allele frequency was 0,148. In diabetes type II cohort and in diabetes type I group the deleted allele frequency was 0,106 and 0,141 respectively (p>0,05). Results indicated, that the deleted allele of the CCR5 receptor gene has no significant association with disease frequency in both types of diabetes. In the case of type I diabetes there was a significant (p<0,05) difference in the onset time and duration of the disease being earlier and longer in the wild-type homozygotes compared to ccr5delta32 heterozygotes. In the type I diabetes the analysis of the correlation between clinical signs and ccr5delta32 genotype is in progress.

P0868. Genetic Susceptibility To The Evolution Of Liver Disease; Chronic Hepatitis C Model

C. Gritti¹, L. Sonzogni¹, L. Silvestri¹, L. Rossi¹, E. Civardi¹, E. M. Silini²

¹Associazione Studio Avanzato Epatiti Virali (ASAEV); Bonate Sotto (BG), Italy; ²Department of Pathology, University and IRCCS S. Matteo; Pavia, Italy

asaev@tin.it

The association between gene polymorphisms affecting “drug metabolizing enzymes” (DMEs) activity and chronic hepatitis C severity was studied by univariate and multivariate analysis. Two classes of DMEs genes were considered; phase I enzymes involved in oxidative metabolism (CYP2E1, CYP2D6) and phase II enzymes involved in conjugation of toxic compounds (GSTM1, GSTT1, Epoxide Hydrolase-EPHX). Polymorphisms were screened on cohort of 400 anti-HCV+ Italian patients divided into four groups; asymptomatic carriers, chronic hepatitis, cirrhosis and hepatocellular carcinoma (HCC) patients. A cohort of 100 healthy subjects was considered as reference category. Inherited differences in several DMEs activities were found to be significantly associate with liver disease expression. In particular, high activity phase I enzymes variants and low activity phase II enzyme alleles occurred more frequently among subjects with advanced liver disease. This effect was more pronounced in males and considering extreme patient categories (carriers and cirrhosis/HCC). Two multivariate analysis techniques (logistic regression and correspondence analysis) were used to model the unconfounded contribution of these genetic traits and their interaction, controlling for known variables of liver disease progression (age, sex, HCV genotype and alcohol consumption). An independent and synergistic effect in predicting cirrhosis and/or HCC was observed for two characters, EPHX T113H homozygosity, and CYP2D6 EM phenotype. Genetic epidemiology techniques may provide clinically meaningful information useful for the dissection of complex disease traits such as those involved in chronic hepatitis C progression.

P0869. Risk of osteoporosis in carriers of 9G>C polymorphism in osteoprotegerin (OPG) gene.

R. Slomski¹, R. Kalak², W. Horst-Sikorska³, D. Baszko-Blaszyk³

¹Institute of Human Genetics; Poznan, Poland; ²Department of Biochemistry and Biotechnology; Poznan, Poland; ³Endocrinology Clinics, Medical School; Poznan, Poland

slomski@au.poznan.pl

Scanning of the 5 exons of OPG gene was performed for 85 patients with osteoporosis or osteopenia. PCR-HD and PCR-SSCP analyses of genomic DNA revealed 6 polymorphisms localized in the intronic region and one polymorphism in exon 1. Five of these different polymorphic variants were substitutions; IVS1+15C>T, IVS2+4C>T, IVS3-5C>T, IVS4-24C>A, IVS4+8A>C and one was a 2 bp deletion; IVS3+46delTC. One polymorphism, 9G>C, detected in exon 1 resulted in a change in the encoded amino acid from positively charged lysine to neutral asparagine and was previously observed in cDNA clones. Because this polymorphism affects conserved region of the osteoprotegerin signal peptide and could influence of secreted protein, we studied the polymorphic allele frequencies in the patient and control population and looked for evidence of association between the polymorphism and osteoporosis. The change 9G>C does not create a new restriction site, and therefore PCR-SSCP analysis was used to determine allele status, by comparison with sequenced control samples. First, we observed that genotype distribution was in Hardy-Weinberg equilibrium for control population (CC=23/97, GC=45/97, GG=29/97, c2=0.458, df=2, p=0.79) we then determined OPG genotype distribution for patients with severe osteoporosis; CC=13/56, GC=36/56, GG=7/56. The genotype and allele frequencies in a combined group of 85 patients with osteoporosis or osteopenia revealed CC=25/85, GC=47/85 and GG=13/85. There were no significant differences in lumbar spine BMD for groups of osteoporotic women and men with different genotypes but genotypes with allele C were over-represented in patients with severe osteoporosis as compared with population group (c2 =5.69, df=1, p<0.02). This is equivalent to odds ratio for osteoporosis of 2.99 for individuals who carry allele C (95% confidence interval 1.20-7.42). Significant differences in combined group of patients with severe osteoporosis or osteopenia, as compared with population group (c2 =5.31, df=1, p<0.025), was also observed (odds ratio 2.36, 95% confidence interval 1.13-4.94).

P0870. Familial Aggregation of Quantitative Traits for Insulin Resistance; The IRAS Family Study

S. S. Rich¹, J. Norris², M. Saad³, S. Haffner⁴, L. Henkin¹, D. Zaccaro¹, L. Wagenknecht¹, R. Bergman⁵

¹Wake Forest University School of Medicine; Winston-Salem, NC United States; ²University of Colorado Health Sciences Center; Denver, CO United States; ³UCLA School of Medicine; Los Angeles, CA United States; ⁴University of Texas Health Sciences Center; San Antonio, TX United States; ⁵USC School of Medicine; Los Angeles, CA United States

srich@wfubmc.edu

Type 2 diabetes mellitus is a complex disease with strong familial aggregation but relatively low genetic risk (λs ~ 2). The strongest risk factors for type 2 diabetes include family history, obesity and factors associated with insulin resistance and ß-cell dysfunction, each of which may be mediated in part by genetic factors. The IRAS Family Study characterizes important diabetes-related traits with respect to their familial aggregation and, ultimately, linkage to polymorphic DNA markers, in African-American (AA) and Hispanic-American (HA) families.

A total of 61 families (22 AA and 39 HA) comprising 650 individuals with phenotypic data were used for these analyses. Frequently sampled intravenous glucose tolerance tests (FSIGT) were performed on eligible subjects and analyzed using the minimal model to obtain estimates of insulin sensitivity (S_I), glucose effectiveness (S_G) and insulin secretion (acute insulin response to glucose, AIR). The disposition index (DI), a measure of insulin resistance-corrected ß -cell function, was also derived (D_I = AIR*S_I), as was the HOMA measure (HOMA=fasting insulin*fasting glucose/22.5). Variance components analyses were employed to estimate familial aggregation (heritability, h²) using the SOLAR software package.

After adjustment for age, sex, race and BMI, h² estimates were S_I; 19 + 8% (p < 0.001); S_G; 19 + 11% (p < 0.02); AIR; 5 + 8% (p = 0.26); D_I; 4 + 7% (p = 0.29); and HOMA; 15 + 9% (p < 0.03). These results suggest that there is modest heritability for S_I and S_G as diabetes-related quantitative traits in this population.

P0871. Hsp70 polymorphism and HLA-DR diversity; Implications for tuberculosis susceptibility in the Cape Coloured population of South Africa

T. Boshoff¹, A. E. Brune¹, E. G. Hoal van Helden², L. Bornman¹

¹Biochemistry Division, Rand Afrikaans University; Auckland Park, South Africa; ²Department of Medical Biochemistry, University of Stellenbosch; Tygerberg, South Africa

Tuberculosis (TB) being one of the leading causes of death from infectious disease in the world, is also a major problem in the Cape Coloured population in the Western Cape of South Africa. Multiple genes influence susceptibility to TB. HLA class II polymorphisms have been associated with susceptibility to inflammatory diseases, but linkage disequilibrium of alleles in the major histocompatibility complex (MHC) complicates identification of disease-associated genes. Genes encoding members of the 70 kDa heat shock protein (HSP)family (Hsp70) are localised within the MHC. In the light of the protective role of Hsp70 in infection and immunity it is hypothesized that Hsp70 polymorhisms may contribute to disease susceptibility. Possible linkage disequilibrium of hsp70 alleles with HLA-DR alleles might lead to extended haplotypes which might act as additive TB susceptibility markers. Hsp70 gene polymorphism (hsp70-1, hsp70-2, hsp70-Hom and the polymorphic PstI site within the coding region of hsp70-2) and HLA class II DR polymorphisms were investigated in the Cape Coloured population inhabiting the Western Cape of South Africa. Polymorphic analysis of hsp70 and HLA-DR genes was performed on genomic DNA from patients suffering from TB (n=60) and matched control subjects (n=61) using PCR-RFLP and PCR-SSP respectively. Preliminary results showed no evidence for an independent role of hsp70 gene polymorphism in susceptibility to TB while DR3 (DRB1*0301-0302) and DR53 (DRB4*0101) were present at a higher frequency (p=0.068 and p=0.034 respectively) in TB patients, and DR12 (DRB1*1201-1203) in control subjects (p=0.028). An additional number of 300-400 cases and controls are currently analysed to meet statistical standards and to investigate linkage disequilibrium of HLA-DR and hsp70 alleles. An improved understanding of the underlying mechanisms contributing to tuberculosis susceptibility may open new avenues in the development of novel therapeutic approaches.

P0872. Factor VII promoter decanucleotide insert polymorphism not associated with decreased Factor VII activity in African American

A. Ratti¹, R. Kudaravalli¹, T. Tidd¹, R. Santacroce², M. Margaglione², M. Devoto³, P. Fortina¹, E. S. Pollak¹

¹University of Pennsylvania; Philadelphia, PA United States; ²IRCCS Casa Sollievo della Sofferenza; San Giovanni Rotondo, Italy; ³duPont Hospital for Children; Wilmington, DE United States

fortina@mail.med.upenn.edu

Factor VII is a procoagulant protein that initiates blood clotting with its cofactor Tissue Factor. Polymorphic variations in the Factor VII (FVII) gene correlate with cardiovascular disease risk in both familial and non-familial studies. The FVII A2 allele contains a 10 nt insert (CCTATATCTT at -323 in the FVII promoter where +1 is the start of translation) and occurs in 10-20% of the population. The A2 allele is associated with a 15-30% decreased FVII level per allele. Two additional SNPs at -122(C/T) and -401(G/A) have been reported to occur in complete linkage disequilibrium with the decanucleotide insert in the Italian population. In vitro assays have previously shown a significantly decreased FVII promoter strength of the A2 compared to the A1 allele and have further shown that this requires the concurrent presence of the -122 and -401 polymorphisms. Using both DNA sequence analysis and the Nanogen NanoChipTM, we analyzed 28 African-American individuals for the presence of the -401/ -323/ and -122 polymorphisms. Twenty-eight percent were found to have a new allele containing the -401 and -323 but not the -122 polymorphisms. The average FVII level in individuals with this allele was similar to that with the A1 allele whereas the combined -401/-323/-122 allele showed the expected significant 30% decrease in Factor VII levels. In summary, we have shown in vivo that the decanucleotide insert in the absence of the -122 polymorphism does not alone cause a decrease in Factor VII coagulant levels.

P0873. Mutations in Kringle IV-2 of the human apolipoprotein(a) gene

H. G. Kraft¹, W. Parson², H. Niederstätter², A. Lingenhel¹, S. Köchl², F. Fresser¹, G. Utermann¹

¹Institute of Medical Biology and Human Genetics; Innsbruck, Austria; ²Institute of Legal Medicine; Innsbruck, Austria

hans-georg.kraft@uibk.ac.at

The identification of mutations in one of many large identical repeated sequences requires special strategies because the detection of the mutation is hindered by the background of a high number of unchanged sequences. We have applied a brute force approach to detect mutations in the 5.6 kb Kringle IV type 2 domain of human apolipoprotein(a) which is present in 2 up to > 40 almost identical copies in individual alleles and is responsible for the size polymorphism of the protein. The procedure consisted of isolation of a single apo(a) allele from a heterozygote by PFGE, two independent amplifications by PCR, cloning of the amplicons and finally sequencing of a number (93) of clones high enough to represent by chance most individual repeats. The application of this strategy to an apo(a) allele with 26 K-IV-2 repeats yielded the following results; i) K-IV types 2A and 2B were both detected in 74 % of the analyzed clones, ii) the ratio of 2A;2B was different from the published cDNA sequence, iii) an additional K IV type 2 (designated K IV 2C) was detected in 5 % of the clones iv) in addition different mutations (4 silent, 2 nonsense and 5 missense ) were detected, and v) all mutations were detected in the first exon of the K-IV structure and none in the second. The bona fide nature of one non-sense mutation was supported by the finding that the corresponding isoform was smaller than predicted by K IV-2 repeat number and by an independent PCR procedure based on genomic DNA. These results demonstrate a higher than anticipated sequence heterogeneity in K IV-2 of apo(a).

P0874. Heritability of Coronary Artery Calcium in Families with Type 2 Diabetes

D. W. Bowden, L. E. Wagenknecht, J. J. Carr, C. D. Langefeld, B. I. Freedman, S. S. Rich

Wake Forest University School of Medicine; Winston-Salem, NC United States

dbowden@wfubmc.edu

Type 2 diabetes is a major risk factor for atherosclerotic cardiovascular disease. Subclinical atherosclerosis, measured as coronary artery calcification (CAC), depends on both genetic and environmental factors. To evaluate evidence for genetic factors (familial aggregation) contributing to variation in CAC, we studied 122 type 2 diabetics and 13 nondiabetics (median age 59 years; range 38-80) in 56 families. CAC was measured by fast-gated helical computed tomography. Other measured atherosclerotic risk factors included blood pressure, body size, lipids, HbA1c, and medical history. Generalized estimating equations were used to test for association between CAC and the risk factors while accounting for underlying correlation due to family membership. CAC was detectable in 80% of patients with diabetes (median score = 84, range 0 - 5776). CAC, adjusted for age, was associated with male gender (p=0.0003), reduced HDL (p=0.02), albumin-creatinine ratio (p=0.008), and cigarette pack-years (p=0.03). CAC was associated with history of angina, myocardial infarction, stroke and vascular procedure (all p<0.005). HbA1c (p=0.14) and fasting glucose (p=0.08) were positively, but non-significantly associated with CAC. After adjustment for age, sex, race and diabetes status, CAC was significantly heritable (h2 = 0.50; p=0.009). In multivariate analysis with additional adjustment for HDL, BMI, hypertension, and smoking, the residual heritability remained significant (h2 = 0.40; p=0.038). These results suggest that genetic factors, independent of known risk factors for atherosclerosis, contribute to the variance of CAC in type 2 diabetes and that searching for QTLs for CAC will be an important phenotype for mapping genes for atherosclerosis and type 2 diabetes.

P0875. Estimation of individual admixture in Trinidad; application to a case-control study of systemic lupus erythematosus (SLE)

M. Molokhia¹, A. L. Patrick², M. Shriver³, E. Parra³, J. Ye³, A. J. Silman⁴, P. McKeigue¹

¹London School of Hygiene & Tropical Medicine; London, United Kingdom; ²Kavanagh Street Clinic; Port of Spain, Trinidad and Tobago; ³Dept Anthropology, Penn State University; Philadelphia, PA United States; ⁴ARC Epidemiology Research Unit; Manchester, United Kingdom

mariam.molokhia@lshtm.ac.uk

Objective: To distinguish between genetic and environmental explanations for high risk of SLE in west Africans compared with Europeans by studying the relation of risk to individual admixture.

Methods: Cases of SLE and controls were sampled from an admixed population in Trinidad. The first 63 cases and 90 controls were typed with a set of 26 SNPs chosen to have large frequency differentials between the parental populations (west African, European and Native American). Individual admixture was estimated in a Bayesian analysis using Markov chain simulation.

Results: Mean admixture of the population was estimated as 51% west African, [95%CI 46%-56%]; 31% European [95%CI 27%-36%]; and 17% Native American [95%CI 14%-20%]. The estimated proportions with >75% and <25% African ancestry were 27%[ 95%CI 24-34%] and 31%[95%CI 25%-37%], respectively. Marker-based estimates of individual African admixture correlated [r= 0.86] with estimates based on reported grandparental ancestry, which were available for 70% of the sample. The slope of the relationship of risk ratio for SLE and renal SLE to African admixture were estimated as 2.3 (95% CI 0.7-7.2) and 3.2 (95% CI 0.5-19.1) respectively. The confidence interval for these slopes are wide because the current marker set extracts only 23% of the information that we would have if admixture were measured accurately.

Conclusion: Individual admixture varies widely within Trinidad so that the relationship of disease risk to admixture can be studied, but for accurate estimation of this relationship it will be necessary to type more markers informative for ancestry.

P0876. Apolipoprotein E Polymorphism In Patients With Alzheimer Disease In Iranian Population .

M. R. Keikhaee¹, P. Pasalar², F. Sahebjam¹, H. Najmabadi^1,3

¹University of Welfare and Rehabilitation; Tehran, Islamic Republic of Iran; ²Biochemistry dept., Tehran University of Medical Sciences; Tehran, Islamic Republic of Iran; ³Karimi-Nejad Pathology and Genetics Center; Tehran, Islamic Republic of Iran

maryamneishabury@yahoo.co.uk

Alzheimer's Disease (AD) is a progressive neurodegenerative disease, which is one of the most common forms of dementia in the age of 65 years and older. Diagnosis is based upon evaluation of cognitive skills, elimination of other dementia associated diseases and cranial CT examination. However, definitive diagnosis is proved only by neurohistological findings. The etiology of the disease is unknown and effective treatment is not available. Intensive research in the pathomechanism of the disease revealed a connection between increased frequency of apolipoprotein E4 allele and late onset of AD. The objective of this study was to characterize apolipoprotein E polymorphisms in late onset AD patients in Iranian population. Therefor, in this study 100 AD patients were examined. The polymorphic area was amplified by PCR, subjected to 8% polyacrylamide gel electrophoresis, and visualized by silver staining. The results shows that E4 allele was significantly more frequent in both early and late onset groups compared to controls ( p<0.001 ). This data indicate that the Apo E4 is also a risk factor for AD in Iranian population.

P0877. CCR5 and CCR2 gene polymorphisms in essential hypertension

V. Romano-Spica¹, M. Mettimano², A. Ianni¹, A. Migneco², D. Arzani¹, L. Savi²

¹Inst. of Hygiene and Public Health (Dir. Prof. G.C. Vanini), Catholic University Medical School; Rome, Italy; ²Hypertension Center, Dept. of Internal Medicine, Catholic University Medical School; Rome, Italy

vrs@rm.unicatt.it

Blood pressure is a common trait of multifactorial origin. Evidence indicates that the immune system plays a role in the development of hypertension in both animal models and humans. Genes for CC-chemokine receptor (CCR) 5 and CCR2 map to human chromosome 3p21.3, within a cluster of 350 kb. A 32-bp deletion in the CCR5 gene (CCR5D32) and a G-to-A point mutation in the CCR2 gene (CCR264I) have been first associated to resistance to HIV-1 infection and recently involved in several immuno-related disorders. In the present study we evaluated allelic distribution of CCR5 and CCR2 genes in essential hypertension through a case-control study. Genotype frequency was compared between a group of essential hypertensive patients (stage I-II; n=53) and a group of Caucasian unrelated healthy subjects (n=340). Case selection was made by means of strict clinical evaluation and careful laboratory investigation, according to international guidelines. Genomic DNA was isolated from peripheral blood cells, and gene polymorphisms were analyzed by polymerase chain reaction (CCR5) and restriction enzyme analysis after amplification (CCR2). Chi-square analysis and Fisher’s exact tests were used to compare data between cases and controls and test for conformity to the Hardy-Weinberg equilibrium. A statistically significant difference was observed for CCR5 and CCR2 mutant alleles in essential hypertensive patients in comparison with controls (p<0.01). Both CCR5D32 and CCR264I alleles showed a 0.11 frequency among cases. Genotype distribution was in equilibrium among cases and controls, according to the Hardy-Weinberg equilibrium. Our results suggest an association between polymorphisms at CCR genes and essential hypertension.

P0878. Interaction between allelic variants of Cystathionine beta synthase (CBS) and Metilenetetrahydrofolate reductase (MTHFR) genes as risk factor for Neural Tube Defects (NTD) in Argentina.

R. Cortese¹, N. Buzzalino¹, E. Goldschmidt², G. Mercado¹, L. Dain¹

¹Centro Nacional de Genética Médica; Buenos Aires, Argentina; ²Servicio de Genética. Hospital de Nińos "R.Gutiérrez"; Buenos Aires, Argentina

renecortese@s6.coopenet.com.ar

Allelic variants of CBS and MTHFR have been suggested as risk factors for NTD. We have demonstrated an increased risk factor for the allele T of MTHFR in our population. The aim of this study was to assess the presence of a 68 bp insertion (68bp ins) in the CBS gene with the development of NTD, isolated or in association with the mutation C677T in MTHFR gene. We have studied both mutations in a group of 79 Argentine patients with NTD, 78 mothers and 67 fathers. Control group consisted of 105 healthy adults and 39 healthy children. DNA from blood samples was genotyped by PCR and BsrI and HinfI digestion for CBS or MTHFR respectively. We found an allelic frequency of 0.05 for the 68bp ins in our population. No statistical differences were found when the presence of the 68bp ins in the control group and the affected one were compared (0.11 vs. 0.15, OR= 1.39, 95% CI; 0.59-3.21, p>0.1). However, when comparing the 68bp ins in association with the C677T MTHFR mutation with the absence of both allelic variants, a significant difference was found (0.15 vs. 0.33, p<0.05). A frequency of 0.14 (n=145) for both allelic variants in parents of affected children and 0.15 for parents of non-affected ones (n=40) was observed. Our results suggest that the 68bp ins is not a risk factor for NTD in the studied population unless combined with the thermolabile mutation in MTHFR gene.

P0879. Study of two candidate loci to susceptibily for leprosy (NRAMP1 and MBP)

H. Buchmüller, P. H. Cabello

Fundaēćo Oswaldo Cruz; Rio de Janeiro, Brazil

buchlima@infolink.com.br

Genetic factors can contribute to susceptibility to infectious diseases. The Mannose Binding Protein (MBP) is a seric protein that recognizes patogen microorganisms by carbohydrates present on its cell wall. The MBP acts as an opsonic agent inducing the fagocitosis by macrophages. Low concentrations of this protein are considered to be the most common immunodeficiency in humans. In mice, natural resistance to intracellular infectious disease is associated to the gene Nramp1. The effects of the human gene homologue, NRAMP1, are not well known yet. In this study we genotyped the gene MBP and the locus D543N of the gene NRAMP1 of 204 leprosy patients and 211 blood-donors controlling sample, to evaluate the genes effects on infection or disease development by the Mycobacterium leprae. No significant differences were noticed of the genic and genotypic frequencies of these systems, between the affected group and the normal control, neither the haplotypes MBP/NRAMP1 between the different types of leprosy. Therefore, the genetic variations studied do not seem to be related to susceptibility to leprosy, as suggest by some authors in studies done on other populations.

P0880. Inheritance study of hand osteoarthritis (HOA) in Iceland

I. Manolescu¹, H. Jonsson², S. Stefansson¹, T. Ingvarsson³, H. Jonasson¹, A. Manolescu¹, J. Gulcher¹, K. Stefansson¹

¹DeCode Genetics Inc.; Reykjavik, Iceland; ²University Hospital; Reykjavik, Iceland; ³General Hospital; Akurery, Iceland

ileana@decode.is

Objective; To investigate the genetic contribution in a large scale study of HOA in Iceland using an extensive genealogy database of the Icelandic population. Patients and methods; As part of an ongoing study, patients with HOA have been actively registered through nationwide clinical and surgical registers, radiology clinics and ergotherapy centers. At present 2950 HOA patients, comprising 1% of the Icelandic population, have been registered through the above channels. The patient phenotypes used are mild and severe finger OA and mild and severe thumb base OA based on clinical and radiological findings. Three methods were used to investigate the genetic contribution; (1) the Minimum Founder Test (MFT), to estimate the minimum number of ancestors (founders) needed to account for all patients at various years in the past, (2) the average pairwise kinship coefficient (KC) of the patients, and (3) the relative risk of HOA for relatives. In each case the results were compared with 1000 control sets of similar composition with regard to number, age and sex, generated from the genealogy database. Results; The MFT indicated that the study group had descended from fewer ancestors than controls (p<0.001). The KC was significantly higher than in the control sets (p<0.001). The relative risk of HOA for siblings and parents was higher than in spouses. The KC obtained for patients with severe finger OA was notably high whereas that for thumb base OA without finger involvement was not significantly different from controls. Conclusions; Patients seeking medical services for HOA are more related to each other than matched controls, supporting the role of a genetic component in the disease. Our methods may also contribute to the understanding of the inheritance of different subsets of hand OA.

P0881. Risks for Relatives of Patients with Multiple Sclerosis in Central Sardinia, Italy

I. Prokopenko¹, L. Bernardinelli¹, C. Montomoli¹, B. S. Murgia², R. Ferrai², A. Ticca², A. Caria², L. Musu², M. Piras², A. Mander³

¹Dep. Applied Health Sciences, University of Pavia; Pavia, Italy; ²Divisione di Neurologia, Ospedale San Francesco; Nuoro, Italy; ³MRC Biostatistics Unit; Cambridge, United Kingdom

inga@fisher.unipv.it

Objectives: We calculated age-adjusted relative risks for different categories of relatives of Multiple Sclerosis (MS) patients from the case register of Central Sardinia (Italy) and examined some covariates that may influence the risk in sibs.

Background: MS is a multifactorial disease determined by genetic-environmental interactions. A genetic component to MS is indicated by an increased relative risk in siblings, by an increased concordance rate in monozygotic compared to dizygotic twins, by studies from adoptees, half-siblings and by familial recurrence risk data.

Design: Pedigrees were routinely obtained on prevalent patients of the MS case register. As a consequence most of the cases were independently ascertained.    

We used life tables techniques to calculate age-adjusted recurrence risks. We estimated the effect of the studied covariates by fitting a proportional hazard model.

Results: We included in our analysis 313 probands and a total of 14173 relatives. The mean size of a pedigree was 47 family members. The age-adjusted risk in parents was 1.26 (0.60-2.63), in children 9.44 (9.32 -25.25), in sibs 4.76 (3.57-6.32), in second-degree relatives 0.71 (0.41-1.21), and in third-degree relatives 1.75 (1.24-2.45).

The relationship of the studied covariates to the sibling risk was examined using Cox proportional hazards model. Sex of the proband and proband onset age were significant. More specifically, the hazard ratio for siblings of a female MS proband compared with a male proband was 0.62 (p=0.05). The risk for siblings of the oldest onset patient was lower than for siblings of the early onset patients (hazard ratio = 0.50, p=0.023).

Conclusion: We report the results of the first population-based study in Italy in which recurrence risks are computed within a sufficiently large population to provide figures for patient counselling. Due to the ascertainment scheme we used where each case is ascertained independently on the others, our study is not likely to be affected by ascertainment bias (families with a higher number of MS cases are more likely to be ascertained).

The higher risk observed in sibs of the less susceptible sex (male) can be interpreted as evidence in favour of a liability threshold model for MS.

The recurrence risks for different categories of relatives can be used to counsel patients and their families.

Our results agree with those obtained by similar studies.

P0882. The association alleles of VDR3, COL1A1, CALCR genes and severe osteoporosis in women from Russia.

M. Moskalenko¹, M. Aseev¹, I. Zazerskaya², V. Baranov¹

¹Ott's Institute of Obs.& Gyn.; St-Petersburg, Russian Federation; ²Pavlov's State Medical University.; St-Petersburg, Russian Federation

mvmos@yahoo.com

The pathogenesis of osteoporosis is controlled by genetic and environmental factors. Although genetic predisposition seems to be a factor in the pathogenesis of osteoporosis, the precise cohort of genes that may be involved is not well defined. We investigated the association of VDR3, COL1A1 and CALCR genes polymorphism with osteoporosis in two groups of osteoporosis patients and unrelated controls of north-west Russia. The first group of patients consisted of 64 women with osteoporosis complicated with fractures, while the second group (78 women) contained patients with osteoporosis in postmenopausal (surgical or physiological). The statistical analysis proved preponderance of allele t of Taq I polymorphism in 9 intron of VDR3 in the first group of patients (53,1 %) compared to the control group (31,1 %)( р<0.01).The frequency of allele s of Apa I polymorphism in a recognition site for the transcription factor Sp1 of COL1A1 gene was three times higher in patients (48,4%) compared to the controls (16,7%) and this difference appears to be significant (р<0.01). The frequency of allele T of Alu I polymorphism in CALCR gene was also authentically higher (р<0.01) in this the group of patients (91,7%)(72,5% in control). In the second group of patients we had not elucidated any authentic differences in frequencies of alleles of these three genes, compered to the control group. The frequencies of alleles of these genes were 33,9% (VDR3), 21,8% (COL1A1) and 77,6% (CALCR). Our date indicate that different alleles of VDR3, COL1A1 and CALCR genes correlate with severe osteoporosis in women.

P0883. Association of polymorphisms in the NRAMP1 gene and susceptibility to tuberculosis in Chinese

Y. Lam, S. W. K. Im, W. C. Yam

The University of Hong Kong; Hong Kong, China

krissylamyin@yahoo.com.hk

Evidence for the genetic factor, human natural-resistance-associated-macrophage-protein 1 (NRAMP1) gene to have a role in susceptibility to tuberculosis was found in West Africans, Koreans and Japanese. The objective of the study is to investigate whether the polymorphisms in the four regions of the NRAMP1 gene; 5'microsatellite, Intron4, D543N and 3'UTR are associated with the host susceptibility to tuberculosis among Chinese population in Hong Kong. Polymorphisms in NRAMP1 gene were investigated in a case-control study of tuberculosis in Hong Kong, China. Polymerase Chain Reaction -- Restriction Fragment Length Polymorphism (PCR-RFLP) analysis was used to type the polymorphisms and to determine the allelic frequencies of the different regions of the gene among paitents and controls. Patients suffering from tuberculosis were diagnosed by positive findings in chest X-ray and sputum culture, while the controls were healthy blood donors with no history of tuberculosis. Relationship of the polymorphisms in the 4 regions of NRAMP1 and the host susceptibility to tuberculosis among Chinese population in Hong Kong will be highlighted and discussed.

P0884. BRCA1 and BRCA2 founder mutations account for 5% of 233 unselected Finnish ovarian carcinoma patients

L. M. Sarantaus¹, P. Vahteristo², E. Ostermeyer², A. Tamminen², L. Unkila-Kallio², R. Bützow², H. Nevanlinna²

¹Department of Obstetrics and Gynecology, Helsinki University Central Hospital; Helsinki, Finland; ²Dept. of Obstetrics and Gynecology; Helsinki, Finland

Laura.Sarantaus@hus.fi

Germ-line mutations of BRCA1 and BRCA2 predispose to hereditary breast-ovarian cancer syndrome. In Finland, 20 different BRCA1/2 mutations have been identified and 13 of the mutations are founders and account for the vast majority of Finnish BRCA1/2 families. The purpose of our study was to determine the prevalence of BRCA1/2 mutations in unselected Finnish ovarian carcinoma patients and to evaluate the relationship between mutation status and personal/family history of cancer. Two hundred and thirty-three patients were screened for all BRCA1/2 mutations known in the Finnish population. Additionally, a subgroup of patients with personal history of breast cancer and/or family history of breast and/or ovarian cancer was screened for novel BRCA1/2 mutations. Thirteen patients (5.6%) had mutations; eleven in BRCA1 and two in BRCA2. Seven of the 13 known Finnish BRCA1/2 founder mutations were identified in this study, and they accounted for 12 of the 13 mutations detected. A logistic regression analysis was used to determine the odds of mutation for ovarian carcinoma patients. The most significant predictor of a mutation was the presence of both breast and ovarian cancer in the same woman, but family history of breast cancer was also strongly related to positive mutation carrier status. Although BRCA1/2 mutation testing is not warranted in the general ovarian cancer patient population, patients with personal/family history of breast cancer could benefit from referral to genetic counseling and mutation testing.

P0885. Prostaglandin H synthase 2 (PTGS2/COX-2) variant in African Americans and a case-control study of colorectal adenomas.

H. J. Lin¹, K. M. Lakkides², T. O. Keku³, H. D. Frankl⁴, E. R. Lee⁵, R. M. Garavito², R. S. Sandler³, R. W. Haile⁶, W. L. Smith²

¹Harbor-UCLA Medical Center; Torrance, CA United States; ²Michigan State University; East Lansing, MI United States; ³University of North Carolina; Chapel Hill, NC United States; ⁴Kaiser Permanente; Los Angeles, CA United States; ⁵Kaiser Permanente; Bellflower, CA United States; ⁶University of Southern California; Los Angeles, CA United States

henry_lin@humc.edu

Prostaglandin H synthase 2 (PTGS2; also known as cyclooxygenase-2) is thought to take part in prevention of colorectal cancer by nonsteroidal anti-inflammatory drugs, which are inhibitor of the enzyme. We used DNA heteroduplex analysis to screen the PTGS2 gene for naturally-occurring enzyme variants, in order to obtain further information on a biochemical mechanism of prevention. We found a single-base substitution in 10% of African Americans (g.5939T>C; GTT>GCT; Val511Ala). The amino acid change is predicted to open a large cavity near the PTGS2 active site and may change conformations of key residues. No changes in Vmax, Km, or thermal stability were observed for the mutant enzyme in Cos-1 cell assays with arachidonic acid as substrate. However, the conditions of the assay were expected to differ substantially from the normal intracellular environment. Case-control analyses of 380 African Americans from 2 study populations showed odds ratios (and 95% confidence intervals) for colorectal adenomas of 0.50 (0.18-1.44) and 0.56 (0.13-2.44) among subjects with the mutation. The results are potentially consistent with a protective effect of the mutation, mimicking nonsteroidal anti-inflammatory drugs. Larger sample sizes are needed to confirm.

P0886. A Unique Resource for Breast Cancer Research; The Cooperative Family Registry for Breast Cancer Studies

R. T. Senie¹, I. Andrulis², M. Daly³, J. Hopper⁴, S. Buys⁵, D. West⁶, H. Anton-Culver⁷

¹Columbia University; New York, NY United States; ²Cancer care Ontario; Toronto, ON Canada; ³Fox Chase Cancer Center; Cheltenham, PA United States; ⁴University of Melbourne; Carlton, Australia; ⁵Huntsman Cancer Research Center; Salt Lake City, UT United States; ⁶Northern California Cancer Center; Union City, CA United States; ⁷University of California, Irvine; Irvine, CA United States

rts15@columbia.edu

Although increased susceptibility to breast and ovarian cancer has been associated with genetic mutations of BRCA1 and BRCA2, risk may also be modified by lifestyle factors, health behaviors, environmental exposures,and other genetic components. To provide resources for studies of these complex interactions, the National Cancer Institute funded the Cooperative Family Registry for Breast Cancer Studies [CFRBCS], a consortium of six international research centers with an Informatics Support Center. Family recruitment was initiated in 1997 at the three population-based and three clinic-based sites. As of December 2000 more than 6,000 families including 15,000 individuals are participating. Data and biospecimens are collected from probands and their relatives using common instruments and protocols. Coded personal health information, dietary intake, treatment for breast and/or ovarian cancer, and pedigree data are routinely transmitted to the Informatics Center. Biospecimens including blood and tumor tissue samples are banked at each collaborating site following rigid quality control procedures. Genetic analyses are being conducted; currently 350 carriers of BRCA1 or BRCA2 have been identified. The primary purpose of the CFRBCS has been the development of a unique resource for interdisciplinary studies of genetic and environmental risk of breast and ovarian cancer. Several hypothesis-driven research projects are being conducted by CFRBCS investigators and international collaborators using the data and biospecimens; additional investigators are encouraged to develop research proposals for submission to the external Advisory Committee. Information about the CFRBCS and application procedures are available on the NCI website www-dccps.ims.nci.nih.gov/CFRBCS.

P0887. Glutathione S-transferase GSTM1 and GSTT1 polymorphisms and the risk of oral and laryngeal cancer.

A. Hatagima¹, C. F. S. Marques¹, F. S. Rajoy¹, R. J. Koifman¹, J. S. Capelli¹, M. B. Kneipp¹, P. Boffetta², P. Brennan², N. Munoz², R. Herrero², S. Koifman¹

¹Fundaēćo Oswaldo Cruz; Rio de Janeiro, Brazil; ²IARC; Lion, France

hatagima@gene.dbbm.fiocruz.br

Glutathione S-transferases are important xenobiotic metabolizing enzymes and several studies have showed that their genetic polymorphisms may contribute to individual susceptibility for some cancers. Oral and laryngeal cancers are important due to their incidence in several countries. Genetic and environmental factors are involved in the development of these cancers. Many of GST enzymes are involved in the desintoxication of active metabolites from tobacco smoke, what makes them important on modulating the susceptibility to tobacco-related cancers. As part of a wide case-control study conducted in Latin America by the International Agency for Research on Cancer, we studied GSTM1 and GSTT1 genetic polymorphisms in 200 oral and 73 laryngeal cancer patients from Rio de Janeiro (Brasil). Polymorphisms were analyzed in DNA from peripherical blood extraction by Polymerase Chain Reaction and restriction enzymes digestion. The null homozygotes frequencies were 45% (GSTM1) and 24% (GSTT1) in oral cancer patients and 44% (GSTM1) and 21% (GSTT1) in laryngeal cancer group. The observed differences among these frequencies and the ones from a sample of healthy individuals were not statistically meaningful. No associations were observed concerning laryngeal cancer and polymorphisms studied here. However, for the GSTM1 polymorphism both healthy and oral cancer samples were not under the Hardy-Weinberg equilibrium (p<.001 and p=.003, respectively). Among healthy controls it happens due to the excess of AB heterozygotes, but in the oral cancer group this genotype was absent, suggesting its protective role in oral cancer in our population. Support; CNPq, FAPERJ-FIOCRUZ, European Community DG XII.

P0888. Frequency distribution of ATM gene alterations in German breast cancer patients and in the general population

P. Yamini, K. Klöpper, R. Bendix, M. Bremer, D. Rades, J. H. Karstens, T. Dörk

Medizinische Hochschule; Hannover, Germany

doerk.thilo@mh-hannover.de

The ATM gene that is mutated in the autosomal recessive syndrome ataxia-telangiectasia (A-T) encodes a 350 kDa protein kinase which controls radiation-induced DNA damage responses via regulation of downstream targets such as p53, c-abl or BRCA1. Heterozygosity for ATM germline mutations has been implicated as a genetic predisposition in breast cancer and other age-related disorders. We have sought to determine the relative frequencies of common ATM gene alterations in a hospital-based series of 1000 consecutive breast cancer patients and in 500 random individuals from the general population of Lower Saxony in order to gain more insight into the potential clinical relevance of these genetic variations. We have developed a hexaplex ARMS based method for a rapid and specific screening of the most frequent ataxia-telangiectasia mutations in Germany, and we have screened for an additional six common amino acid substitutions using restriction-enzyme or SSCP-based methods. The most frequent ATM truncating mutation in Germany is the leaky splicing mutation IVS10-6T->G that we have identified in seven (0.7 %) breast cancer patients but also in three (0.6 %) control individuals. The most common amino acid substitution, D1853N, was present at an allele frequency of 0.13 in both the breast cancer and control cohorts. However, the rarer ATM amino acid substitutions were, in a composite test, more frequent in the breast cancer patients than in the general population (7.9% vs. 5.3% of alleles, p<0.01). These results suggest that ATM missense substitutions could play a role in the genetic predisposition towards cancer.

P0889. Evaluating the APEX-based resequencing assay of p53 tumour suppressor gene

N. Tõnisson^1,2, J. Zernant¹, A. Kurg², H. Pavel¹, A. Metspalu^2,1

¹Asper Ltd.; Tartu, Estonia; ²Estonian Biocentre / IMCB, University of Tartu; Tartu, Estonia

neemet@asper.ee

Mutation data of the p53 gene have relevant prognostic and therapeutic value in several types of cancer. We are developing and evaluating an APEX (Arrayed Primer Extension) - based test for the gene with goal of getting the full mutation data at both DNA strands in a single assay. A patient DNA sample is amplified, digested enzymatically, and annealed to arrayed primers, which promote sites for template-dependent DNA polymerase extension reactions using four fluorescently labelled dideoxynucleotides. The Genorama™ imaging system and genotyping software are used for imaging and semiautomatic sequence analysis. The p53 gene chip is scanning exons 2 to 9 plus two introns from both strands (1218 bases). An average of 98 % of the arrayed p53 gene sequence can be analysed from either sense or antisense strand and 85 % from both strands. In best cases the readable sequence is up to 99.8 % and 96 %, respectively. Tumour samples with known mutations were analysed in a blind test. Predominantly the results were concordant with TTGE (Temporal Temperature Gradient Electrophoresis) plus dideoxy sequencing. One case showed the presence of double mutation in codon 290. In addition, one heterozygous SNP and an SNP with minor allele were found. The entire resequencing procedure can be completed in less than six hours and at 5 to 10 times less cost compared to ABI sequencing per sample. Once fully developed, the p53 gene chip should become a medium for accurate and efficient DNA sequence analysis of this or other frequently mutated genes.

P0890. Introduction and diffusion of the BRCA1 mutation R1443X in the French Canadian population

H. Vezina¹, F. Durocher², L. Houde¹, C. Szabo³, S. Delos⁴, M. Dumont⁴, M. Jomphe¹, M. Plante⁴, R. Laframboise⁴, J. Chiquette⁵, D. Stoppa-Lyonnet⁶, D. Goldgar³, D. Easton², P. Bridge⁷, J. Simard⁴

¹Universite du Quebec a Chicoutimi; Chicoutimi, PQ Canada; ²CRC Genetic Epidemiology Unit, University of Cambridge; Cambridge, United Kingdom; ³Unit of Genetic Epidemiology, International Agency for Research on Cancer; Lyon, France; ⁴Centre hospitalier universitaire de Quebec, Laval University; Quebec City, PQ Canada; ⁵Hopital St-Sacrement; Quebec City, PQ Canada; ⁶Oncology Genetics, Institut Curie; Paris, France; ⁷Alberta Children’s Hospital, University of Calgary; Calgary, AB Canada

hvezina@uqac.uquebec.ca

In the 17th century, about 5 000 immigrants, coming mostly from France, settled in Canada. Among them, 3 500 to 4 000 have descendants in the contemporary Quebec population and they account for the major part of the gene pool of the 5 million Quebecers of French descent. The present study was designed to analyse the role of this founder effect and the resulting population demogenetic structure in the introduction and diffusion of the BRCA1 recurrent R1443X mutant allele. The highly conserved haplotype observed in 46 R1443X carriers from 11 French Canadian families and generated using seventeen microsatellite markers spanning a 9.3 cM region surrounding the BRCA1 locus confirms that the R1443X mutation is a founder mutation in the Quebec population. Ascending genealogies of one carrier individual per family (n=10) and of controls (n=30) were reconstructed using the BALSAC population register. These genealogies have an average depth of ten generations but many lineages go as far back as 13 generations. We identified the founder couple with the highest probability of having introduced the mutation in the population. Genetic contributions of this founder couple to the contemporary regional populations of Quebec were measured in order to understand the spread in time and space of the mutation and to detect the presence of spatial stratification in the diffusion pattern of the mutation. Finally, we also performed haplotyping analysis of R1443X carriers from 10 French families and although the results are consistent with a common origin for this mutant allele in both populations, a distinct haplotype was obtained in two French families, thus suggesting multiple origins for the R1443X mutation.

P0891. Interaction between BRCA1 and BRCA2 genes and environmental risk factors in early-onset breast cancer. Results from a French population-based study.

V. Bonodona¹, O. Sinilnikova², A. Bremond¹, H. Mignotte¹, P. Mathevet³, A. Martin⁴, C. Zinzindohoue³, J. Bobin³, P. Romestaing³, D. Raudrant³, R. Rudigoz³, G. Lenoir², C. Lasset¹

¹Centre Léon Berard; Lyon, France; ²International Agency for Research on Cancer; Lyon, France; ³Hospices Civils de Lyon; Lyon, France; ⁴Clinique Jeanne D'Arc; Lyon, France

bonadona@lyon.fnclcc.fr

Studies have suggested the role of gene-environment interactions on breast cancer risk, since the effects of known risk factors differ according to the existence or not of a family history of breast cancer. The recent identification of susceptibility genes for hereditary breast cancer provides further insight into the evidence of such interactions. The purpose of this study was to evaluate the prevalence of BRCA1 and BRCA2 mutations in early-onset breast cancer cases and to analyse gene-environment interactions. From a population-based cohort of women with breast cancer diagnosed before age 46 years, we collected information about environmental and familial risk factors and blood sample for genetic testing. We used a case-only design to study gene-environment interactions. A total of 269 women were included and genetic analyses were performed for 232 women. Twenty-one BRCA1 and BRCA2 mutations carriers were identified. Mutation prevalence was higher in women with breast cancer before age 41 years than the other women (12.8% versus 5.2%). We found a non-significant interaction of some reproductive factors; an increased breast cancer risk was associated with a late age at menarche (>15 years) (odd ratio=2.5; p=0.2) and parity (odd ratio=1.8; p>0.2) in carriers of BRCA1 and BRCA2 mutations. No effect of age at first-full term pregnancy and oral contraception were found. These results indicate a possible interaction between BRCA1 and BRCA2 susceptibility genes consistent with the previous studies. However, they need confirmation with further larger studies. The implications of our findings for the high prevalence of BRCA1 and BRCA2 mutation in women with breast cancer before 41 years will be also discussed.

P0892. Characterization of a subcloned fragment (pBA0.6) of pCMM86 located on 17q21 and its potential use in generating an individual-specific DNA profile and detection of genetic variability in Indian population and sporadic breast cancer tissues

A. Saha, R. Bamezai

Jawaharlal Nehru University; Delhi, India

anjana14@hotmail.com; bamezai@hotmail.com

Sequence analysis was carried out of a human clone pBA0.6 generated after ExoIII/S1nuclease digestion and subcloning of pCMM86 (GDB; 168382, D17S74), which was not available in the database. It revealed the presence of a reiterating core motif of 24mer GTGGGTGTGTTGGAGGGGGTGAGG present 23 times (Accession no. AF079321), which was GC-rich and minisatellitic in nature. Genomic blots of HaeIII-digested human DNA when hybridized to pBA0.6, generated a ladder of polymorphic bands which were individual specific in nature .The probability of identity ranged from 5.07X10-14 to 2.64X10-16 in different population groups. Out of three isomorphic bands of 29.0 kb, 2.4 kb, 2.1 kb, 29.0 kb band was observed to be Homo sapiens specific (Saha et al., 2000, DNA and Cell Biology, vol. 19 no. 4; 219-226). Dendrograms based on UPGMA method with Jaccard's coefficient values suggested high genetic diversity in all the population groups suggesting that the samples taken were random. Maximum likelihood estimates through bootstrap sampling method showed that Punjabis, Bengalis and UPites formed one cluster whereas South Indians formed a separate cluster altogether thus showing the proximity of these three population groups as compared to that from South India. Further with another minisatellite sequence (Accession no. AF157691) dendrogram at the individual level led to the formation of several small clusters which were interleaved; also, the subgroups for each of the populations were intermingled with the subgroups for the other populations (Saha and Bamezai, 2000, J Hum Genet vol. 45; 207-211). Further, pair studies comparing the lymphocyte and tumor DNA of 19 sporadic breast cancer patients indicated a genetic variation in 32% of the cases studied, pointing towards its utility in screening for somatic changes in the breast cancer tissues. A preliminary study of Northern hybridization with pBA0.6 resulted in two transcripts of 0.63 kb and 0.29 kb. This was further corroborated with RT-PCR results where 2 amplicons, matching with the expected size of two reading frames within the minisatellite sequence, were obtained. Interestingly, amplicons were also generated when tumour samples of breast cancer patients were analyzed. The role of the two transcripts from the minisatellite sequence is not clear as yet.

P0893. A common variant of the methylenetetrahydrofolate reductase gene (MTHFR, 1p36) is associated with an increased risk of cancer

B. T. Heijmans^1,2, J. M. A. Boer³, C. J. Cornelisse⁴, D. Kromhout³, R. G. J. Westendorp⁵, E. J. M. Feskens³, E. Slagboom²

¹TNO Prevention and Health; Leiden, The Netherlands; ²Dept. of Medical Statistics, Leiden University Medical Centre; Leiden, The Netherlands; ³Dept. of Chronic diseases, National Institute of Public Health and the Environment; Bilthoven, The Netherlands; ⁴Dept. of Pathology, Leiden University Medical Centre; Leiden, The Netherlands; ⁵Gerontology and Geriatrics, Leiden University Medical Centre; Leiden, The Netherlands

bt.heijmans@pg.tno.nl

Folate metabolism is thought to play an important role in carcinogenesis through its involvement in both DNA methylation and nucleotide synthesis. A common Ala/Val variant in the methylenetetrahydrofolate reductase gene (MTHFR) leads to a mildly disturbed folate metabolism. We previously reported that the MTHFR Val/Val genotype was associated with increased cancer mortality in men aged 85 years and over (Eur J Hum Genet 1999;7;197-204). To further explore the deleterious effects of the MTHFR genotype, we studied the association of the genotype with overall cancer risk and the risk of cancer of specific organs in 860 men aged 65-84 years who were followed over 10 years. During follow-up, 150 new cases of cancer occurred among the 793 men without cancer at baseline. The risk of cancer was 1.81-fold (95% CI, 1.09-3.00) increased among men with the Val/Val genotype as compared to men with the Ala/Ala genotype. The higher incidence of cancer could be attributed to an increased risk of cancer of the prostate (RR, 3.48; 95% CI 1.05-11.6), the colorectum (RR, 3.65; 95% CI, 1.07-12.5) and the kidney and bladder (RR, 5.48; 95% CI, 1.67-18.0), but not to an increased risk of lung cancer. The risk of cancer appeared to be particularly increased among men with lower folate intake, higher alcohol consumption and of an older age. In conclusion, our current and previous studies in two independent populations indicate that a common Ala/Val variant in the MTHFR gene may have a deleterious effect on the risk of cancer in the general population.

P0894. Thiopurine methyltransferase polymorphism; genotype-phenotype correlation analysis in a Portuguese sample

S. Alves¹, A. Amorim¹, F. Ferreira², M. Prata¹

¹IPATIMUP; Porto, Portugal; ²Serviēo de Hematologia Clķnica, Hospital Geral S. Joćo; Porto, Portugal

sandra.alves@ipatimup.pt

Thiopurine methyltransferase is a polymorphic enzyme that catalyses the S-methylation of thiopurine drugs used in immunosupressive and anticancerigenous therapy. Therefore, characterisation of all the variation at TPMT gene which can produce changes in TPMT activity should be conducted. We set out to perform genotype-phenotype correlation analysis in 143 Portuguese individuals for several TPMT SNPs (including silent and intronic substitutions) and for a VNTR located in the promoter region of the gene. The SNPs and the VNTR were studied by PCR based methods and the phenotype quantification of TPMT was performed using a HPLC method. As expected a statistical significant association was found between the presence of TPMT mutant alleles and reduced TPMT activity. For the other silent or intronic substitutions analysed, no signs that they could influence TPMT activity were detected. The VNTR at the promoter region is characterised by a composite internal structure due to the presence of 3 different types of repeats - A, B and C. Therefore each VNTR allele is defined by a particular pattern of repeat arrangement. We found a statistical significant association between VNTR*6 and reduced levels of TPMT activity. Since VNTR*6 is the allele with more B type repeats, we can hyphotesize that the high number of B repeats is the responsible for the association registered. However the question needs further support. Finally we reported linkage disequilibrium between VNTR*6 and TPMT*3A and between VNTR*4 and TPMT*2.

P0895. Evidence For Deviation From Hardy-weinberg Equilibrium In Brca1 And Brca2

C. S. Healey¹, A. M. Dunning¹, F. Durocher¹, D. Chase², J. Burn³, B. A. J. Ponder¹, D. F. Easton¹

¹University of Cambridge; Cambridge, United Kingdom; ²Westlakes Research Institute; Cumbria, United Kingdom; ³University of Newcastle; Newcastle, United Kingdom

katie@srl.cam.ac.uk

We have investigated a number of common polymorphisms in both BRCA1 and 2 for an association with low penetrance breast and ovarian cancer susceptibility. Two of the most common variants in BRCA1 (P871L and Q356R) and six variants in BRCA2 (a-26g, N289H, N372H, T1915M, R2034C and K3326X) have been studied in population-based series of ~2000 breast cancer cases from the ABC study, 500 ovarian cancer cases and 1500 controls from the EPIC cohort. We have found a significant increase in risk of breast cancer associated with the BRCA2 N372H polymorphism (OR=1.3 [95%CIs; 1.07-1.61]) The BRCA1 polymorphisms show no association with breast cancer risk. We have also observed a significant deviation from Hardy-Weinberg equilibrium (HWE) in adult females for both the BRCA2 N372H and BRCA1 P871L polymorphisms, with a deficit of homozygotes and an excess of heterozygotes for both variants (p<0.01). In a set of ~2400 new-born individuals, new-born girls also showed a deficit of homozygotes and an excess of heterozygotes for both genes, consistent with the adult females. The new-born boys, however, showed the opposite effect, with an excess of homozygotes, which was significant in BRCA2 (p=0.001). This suggests that common variants of BRCA1 and 2 are subject to selection, which appear to affect foetal survival in a sex-dependent manner.

P0896. Folate and breast cancer; the role of polymorphisms in methylenetetrahydrofolate reductase (MTHFR)

A. C. Schofield, L. Sharp, E. Pavlidov, Z. Miedzybrodzka, S. D. Heys, J. Little, N. E. Haites

University of Aberdeen; Aberdeen, United Kingdom

a.schofield@abdn.ac.uk

Breast cancer is one of the most common malignancies affecting women in the UK. Recent evidence has highlighted the importance of nutrition in the aetiology of breast cancer. In particular, diets with higher intakes of fruit and vegetables are considered protective. Vegetables are a major source of dietary folate, which is involved in DNA synthesis and methylation. An association between higher folate intake and reduced risk of breast cancer has been observed. Functional polymorphisms in MTHFR, an enzyme involved in folate metabolism, have been reported (C677T and A1298C). We undertook a case-control study to investigate associations between breast cancer, dietary folate and MTHFR polymorphisms. Sixty-five breast cancer cases were recruited at Aberdeen Royal Infirmary. Age-sex matched controls (without breast cancer) were selected from general practitioner registers. Subjects completed a food-frequency questionnaire and provided a mouthwash sample to enable MTHFR genotyping. A trend of decreasing breast cancer risk with increasing dietary folate intake was observed (OR=0.49, 95% CI 0.2-1.2). For the C677T polymorphism, the TT genotype had a protective effect (OR=0.38, 95% CI 0.12-1.24). A similar protective effect was observed for women with the CC genotype for the A1298C polymorphism (OR=0.24, 95% CI 0.06-0.97). A significant protective effect was observed for women with homozygosity for the A1298C variant only with higher folate intake (OR=0.09, 95% CI 0.01-0.82). A1298C homozygote individuals with high folate intake appear to have the strongest protective effect against breast cancer. Both dietary and genetic variation in folate status may be implicated in the pathogenesis of breast cancer.

P0897. Relation between childhood leukemia and ABO and Rh blood groups in Serbian population

R. Papovic^1,2, S. Cvjeticanin², B. Trišic²

¹Medical Faculty University of Belgrade; Belgrade, Yugoslavia; ²Institute for Biology and Human Genetics; Belgrade, Yugoslavia

bebrapap@eunet.yu

In the sample of 214 children with leukemia(ALL-acute lymphoblastic N=163;ANLL-acute nonlymphoblastic N=43 and HGL-chronic granulocite leukema N=8)the frequencies of ABO and Rh blood types were similar to the average value of the Serbian population. Comparation of the frequencies of ABO blood types between samle of affected individuals with ALL and control sample we can see that O blood type is slightly increased and A,B and AB blood types are slightly decreased. But, in the sample of affected children with ANLL the frequencies of A,O and AB blood types are slightly increased and B blood type is statisticaly decreased (X2=3.8, p<0.05)compared with incidence of ABO blood groups in Serbian population.If we compared frequencies of ABO blood groups between affected children of ALL and ANLL we can see that percentages are similar for A,O and AB blood types but frequences of blood type B is statisticaly decreased in the groop of ANLL affected (X2=3.9,p<0.05). Taking all this into account we may conclude that most frequently affected individuals in whole sample and group of ALL are with blood types O;A;B;AB. But,in the sample of ANLL the most affected children are with blood types in this order A;O;B=AB.

P0898. Cytochrome P450 (CYP1A1 and CYP2E1) Genetic Polymorphisms and Susceptibility to Oral and Laryngeal Cancer.

C. F. S. Marques¹, F. S. Rajoy¹, S. Koifman¹, R. J. Koifman¹, J. S. Capelli¹, M. B. Kneipp¹, P. Boffetta², P. Brennan², N. Munoz², R. Herrero², A. Hatagima¹

¹Fundaēćo Oswaldo Cruz; Rio de Janeiro, Brazil; ²IARC; Lion, France

tiane.ntg@terra.com.br

Oral and laryngeal cancers are responsible for about 7% of new cancer cases worldwide. Many studies have provided evidences of their association with tobacco and alcohol intake, while others suggest the role of genetic factors contributing to carcinogenesis. Polycyclic aromatic hydrocarbons (PAH) and other chemicals present on tobacco smoke and environmental pollutants are submitted to a biotransformation process. Its first part is mostly attributed to P450 superfamily enzymes, which function is to transform these chemicals into intermediates that will be metabolized by other enzymes into the excretion way. The disability to eliminate toxic products due to an enzymatic deficiency may contribute to individual risk for chemical-induced carcinogenesis. Enzymes encoded by polymorphic CYP1A1 and CYP2E1 genes were previously associated with oral, esophagus and lung cancer. In this case-control study we genotyped 277 patients with oral and laryngeal cancer and 200 healthy controls from Rio de Janeiro (Brasil) in order to compare genotypic and allelic frequencies between the groups. Using Polymerase Chain Reaction and restriction enzymes techniques we analyzed this sample, which is part of a greater study coordinated by the International Agency for Research on Cancer. The observed allelic frequencies between cases and controls were similar for CYP2E1 polymorphism and we found no association with the cancers studied. However, the CYP1A1 polymorphism seems to be associated with an increased risk for oral cancer, as this group had twice more mutant homozygotes (GG) than the control one, although the allelic frequencies were similar between them. Support; CNPq, FIOCRUZ-FAPERJ, Europpean Community DG XII.

P0899. Molecular-Genetic Study of CFTR Gene Mutation delF508 in Cystic Fibrosis Families from Bashkortostan

G. Koritina¹, T. Viktorova¹, E. Khusnutdinova¹, G. Baykova²

¹Institute of Biochemistry and Genetics; Ufa, Russian Federation; ²Bashkir medical university; Ufa, Russian Federation

ekkh@anrb.ru

Cystic fibrosis (CF) is the most common lethal autosomal recessive disease among Caucasians. It is caused by defects in the CF transmembrane counductance regulator (CFTR) gene. More then 900 molecular defects have been reported to date. The distribution of these mutations is modify among different human populations. Fifty five patients with CF from Bashkortostan were tested for mutation delF508. We have analyzed 110 unrelated CF-chromosomes (53 - Slavians, 39 - Tatars,12 - Bashkirs,4 - Chuvashes, 2 - Udmurts). The common mutation delF508 was found only on 31,8% of tested CF-chromosomes; 30 were homozygous and 5 were compound heterozygous. It is one of the lowest incidence of this deletion reported in Russia. Taking into account, that CF-patients from Bashkortostan characterised by high ethnical heterogeneous, we have analyzed frequency of delF508 in different ethnical groups of CF patients. Among Slavians probands (Russians, Ukrainians, Belorussians) delF508 was detected on 45% CF-chromosomes, among Tatars - on 28% and among Bashkirs, Chuvashes, Udmurts delF508 was not found. We suppose, that the absence of delF508 in Bashkirs CF-families could be explained by the features of formation of this population, resulting from mixing of different ethnical components (Turkic, Mongolian and probably Finno-Ugric and Indo-European). By the data of archeology and anthropology the Turkic and Mongolian components in Bashkirs gene pool is highest among other investigated populations. The delF508 practically does not meet in Turkic and Mongolian populations. So our results showed the little portion of Slavians component in Bashkirs gene pool.

P0900. The major cystic fibrosis mutation in Latvia; frequency, origin and age of the ∆F508 mutation

A. Krumina¹, V. Kroshkina¹, V. Svabe¹, Z. Krumina¹, L. Krumina¹, I. Tamane¹, M. Lazdinsh², V. Baumanis²

¹Medical Academy of Latvia; Riga, Latvia; ²Latvian State University; Riga, Latvia

Astrida.Krumina@rsu.lv

CFTR gene mutation ∆F508 is the most frequent cystic fibrosis (CF) mutation, accounting for 70% of CF chromosomes. Strong linkage disequilibrium found between ∆F508 and different polymorphic DNA markers have substantiated the hypothesis of a single origin of this mutation. The aims of our study were 1) to characterize the prevalence of mutation ∆F508 among normal and CF chromosomes;2)to estimate the age and origin of the ∆F508 mutation in Latvia. 31 CF patients,30 their healthy family members and 136 healthy unrelated Latvians were subjected to analysis of mutation ∆F508, three extragenic and three intragenic polymorphic DNA markers. The ∆F508 allele frequency in the overall CF patient population was 59.7%.An average incidence of ∆F508 mutation carriers among healthy Latvians was 1;42.Absolute linkage disequilibrium was found between ∆F508 and loci IVS6a and XV-2c,highly significant -between ∆F508 and loci TUB18 and KM-19.The results of extended haplotype analysis showed different distribution of haplotypes between normal,∆F508 and non-∆F508 CF chromosomes.These results are consistent with the data published for most European populations and confirms the hypothesis that the ∆F508 mutation has derived from a single mutational event.Taking into consideration the values of recombination rates between ∆F508 and extragenic polymorphic markers,the estimate of the number of generations elapsed since the mutation ∆F508 was first introduced into Latvia, approximates 101 (2020 years).These data are in contrast with those published for other populations.

P0901. Why is cystic fibrosis so common?

M. Super¹, J. Chadwick², T. Roberts¹, N. Andrew³

¹Royal Manchester Children's Department of Clinical Genetics; Manchester, United Kingdom; ²Medical Student, Manchester University; Manchester, United Kingdom; ³Molecular Genetics Laboratory; Royal Manchester Children's Hospital; Manchester, United Kingdom

Maurice.Super@man.ac.uk

It is widely accepted that a heterozygote advantage must be operating to maintain a carrier frequency of 1 in 25 associated with the 1 in 2500 incidence children born with cystic fibrosis. This could operating prenatally, through meiotic drive, after birth or both. We have explored meiotic drive by counting the number of carriers born to couples where only one parent is a carrier proven on DNA testing and the partner negative. In 519 offspring born to such couples 313 have been found to be carriers instead of the expected 259 (chi squared =11.4, p<0.01. This phenomenon is seen most strongly in female offspring born to female carriers. Implications and possible mechanisms will be discussed.

P0902. The FV Leiden and prothrombin G20210A mutations in healthy and thrombophilic Yugoslav population

L. B. Rakicevic¹, D. Mikovic², M. Kovac², V. Djordjevic¹, D. Radojkovic¹, A. Savic¹

¹Institute of Molecular Genetics and Genetic Engineering; Belgrade, Yugoslavia; ²Hemostasis and Hemophilia Center, National Blood Transfusion Institute; Belgrade, Yugoslavia

qwert@eunet.yu

Purpose; The development of thrombotic disorders in humans is one of the most common causes of morbidity and mortality in the Western world. The risk of venous thrombosis is increased in individuals who carry specific genetic abnormalities in blood coagulation proteins.The FV Leiden, and prothrombin G20210A mutations are the most prevalent genetic defects which increase risk of venous thrombosis. Their allelic frequencies in Caucasian populations vary between 1 and 8%, and 0.35 and 2% respectively. For Yugoslav population there are no data for either one. Purpose of this study was to establish the prevalence of the FV Leiden and prothrombin G20210A mutations in normal and thrombophilic subjects inYugoslavia. Methods and matherials; A study was carried in a group of 100 unrelated blood donors and in 87 unrelated thrombophilic patients. Criteria for selection of patients for this study were; the occurrence of the first episode of thrombosis at the age below 50 years and at least one clinical feature of the inherited thrombophilia. FV Leiden and prothrombin G20210A mutations were detected by polymerase chain reaction, followed by digestion with allele-specific restriction enzymes. Normal and mutated genes were distinguished by the sizes of the restriction fragments. Results; In 100 control subjects 4 carriers of FV Leiden and 3 carriers of prothrombin G20210A mutations were detected, wich gives the frequencies of mutated alleles 2% and 1.5% respectively. Among 87 trombophilic patients, the frequencies of mutated alleles were higher. For FV Leiden mutation, 23 carriers (21 heterozygous and 2 homozygous) were detected, giving the allelic frequency of 14%. For prothrombin G20210A mutation, 8 heterozygotic carriers were discovered, i.e. the allelic frequency of 4.5%. Conclusions; The frequencies of FV Leiden and Prothrombin G20210A mutations in our population are within the range for general Cucasian populations. The prevalences of these mutations in thrombophilic patients, show the usual higher frequencies than in control.

P0903. The correlation of the genotype and phenotype in Slovak Huntington`s population.

M. Kvasnicovį¹, L. Kadįši², M. Lukįsovį³, F. Cisįrik⁴, J. Barošovį⁵, E. Nebesįkovį⁶

¹Hospital Roosevelt; Banskį Bystrica, Slovakia; ²Institut of Molekular Biology Scientific Faculty; Bratislava, Slovakia; ³Centrum of Medical Genetics Faculty Hospital; Bratislava, Slovakia; ⁴Department of Medical Genetics; Žilina, Slovakia; ⁵Department of Medical Genetics; Nitra, Slovakia; ⁶Department of Medical Genetics Faculty Hospital; Košice, Slovakia

fdr@isternet.sk

Since Huntington disease (HD) mutation was identified a direct molecular test has become an integral part of diagnosis of the disease. The aim of our study was to find out the relationship between number of CAG repeats in IT-15 gene and some phenotype signs of disease in Slovak Huntington`s patients. Since molecular test has been available in our country diagnostic positive molecular test has been confirmed in 63 individuals. In accordance with the range of mutation these individuals were divided into 3 groups; I. With the number of CAG repeats 40 - 45 II. -"- 46 - 50 III. -"- over 50 The majority of our patients belong to the group I. The mean age at onset HD in this group was 40 years (with the range 20 - 59 years). The less patients belong to the group III with the mean age at onset of disease 22 years (with the range 15 - 29 years). Except the relationship between the size of mutation and the age of onset HD we have found also the relationship between the mutation and the first observed symptoms of the disease. While in the I. and II. group majority of patients showed neurological symptoms at the onset of the disease in the III. group that were the changes in behavior and cognition. Our study was the first study of the Slovak Huntington`s population. We confirmed correlation of the size of mutation in HD gene with the age at onset disease and with some of the clinical symptoms of the disease.

P0904. Congenital Cataract – Autosomal Recessive Form – in An Ethnic Isolate, Intensely Inbred

D. Bembea¹, L. Spineanu¹, M. Bembea², O. Vancsik², C. Jurca², R. Spineanu²

¹Clinical Adults Hospital, Department of Ophtalmology; Oradea, Romania; ²Clinical Children Hospital; Oradea, Romania

diabem@yahoo.com

Introduction; The most common lens defect in the neonate is cataract. It may occur as an isolated defect or it may be associated with other ocular or systemic abnormalities. Cataract may be inherited or sporadic. The most common type of inheritance is autosomal dominant; autosomal recessive inheritance occurs less frequently and it is sometimes found in populations with high rates of consanguinity. Objective; The study of congenital cataract in an isolate, in order to identify the aetiology, the type of inheritance and the genetic and clinical characteristics. Material and method; The study group comprises an isolate population of approx. 200 individuals, with 8 cases of congenital cataract identified so far. The pedigree, segregation study and the probability calculation using the Bayes formula have been accomplished for each family. Results; Endogamy in this isolate is of 80%. The consanguinity coefficient of the population forming the isolate is 0.004373 and the frequency of consanguineous marriages is of 20%. Congenital cataract has been noticed in 8 patients of 4 siblings, which yields a prevalence of the disease of 7.4% inside the isolate (1/14). The calculated risk for the 4 families varies between 1/513 and 1/5. The frequency of heterozygous carriers is 32%. The chances of maintaining the frequency of the disease in the isolate and particularities of genetic counselling in this case are discussed. Conclusions; Congenital cataract, usually rare, can reach an unusually high incidence in isolates. Its study becomes more interesting today, as isolates are disappearing.

P0905. Geographic clustering of left-right axis abnormalities in Finland - evidence of a founder mutation?

E. Kajantie¹, A. Eerola², T. Boldt², K. Virkola³, M. Eronen², K. Aittomäki⁴

¹Department of Clinical Genetics and The Hospital for Children and Adolescents, Helsinki University Central Hospital; Helsinki, Finland; ²The Hospital for Children and Adolescents, Helsinki University Central Hospital; Helsinki, Finland; ³Department of Radiology, Helsinki University Central Hospital; Helsinki, Finland; ⁴Department of Clinical Genetics, Helsinki University Central Hospital; Helsinki, Finland

eero.kajantie@hus.fi

Backgorund; The population history of Finland with regions of old (around 2000 years ago) and new (about 500 years ago by a small number of founders) settlement has created an unique model to study genetic disorders on the basis of their geographical distribution across the country. Left-right axis abnormalities constitute a clinically well-defined but genetically heterogeneous disorder manifesting as congenital defects in the sidedness of the heart, great vessels, bronchi, spleen, liver and intestine. Objective; To determine whether left-right axis abnormalities in Finland occur more frequently within the new founder settlement than the old settlement. Methods; The database of the Helsinki University Central Hospital was searched for suggestive diagnoses which were confirmed by chart review. In Finland, almost all patients having left-right axis abnormalities with cardiovascular malformations are evaluated in this hospital. To exclude the effect of modern migration, the birthplaces of the patients' grandparents, collected from population records, were analysed. Results; 42 patients were found, among them 2 sibpairs (total number of siblings 131). An excess, 94 (59%), of the grandparents was born in provinces corresponding to the region of new settlement, which comprise of 37% of the Finnish population in the 1930s (khi square = 15.4, p < 0.001). Clustering of the grandparents' birthplaces with in a few smaller areas was also observed. Conclusions; Founder mutation(s) predisposing to left-right asymmetry abnormalities are likely to be present in the region of new settlement of Finland. Given the small number of sibpairs found, a complex inheritance pattern is more likely than monogenic.

P0906. CAG/CTG repeat polymorphism of IT-15 and DMPK genes and prevalence of myotonic dystrophy and Huntington disease in different populations

O. Malysheva, T. Ivaschenko, V. Baranov, V. Baranov

Institute of Obstetrics and Gynecology; St.Petersburg, Russian Federation

omal99@mail.ru

It is considered that in many cases the frequency of alleles with tandem trinucleitide repeat number more than 20 correlates with prevalence of corresponding expansion diseases in population. We studied distributions of CAG/CTG repeat alleles of DMPK and IT-15 genes in four populations; North-West Russia (St-Petersburg), Moldavians, Georgians and Uzbeks. The frequencies of more than 20 CTG-repeat alleles were in a good correlation with prevalence of MD in Russians, Uzbeks and Europeans (literature data). There weren’t a statistically significant differences between frequencies of IT-15 gene alleles with more then 20 CAG repeats in Russians and in Uzbeks compared to East England (Rubisztein et al., 1994). HD is the most common in England, 2-5 times more rare in Russia and practically absent in Uzbeks. We also compared the distributions of long normal CAG/CTG repeat alleles of IT-15 and DMPK genes in many populations (ссылка). In some populations there weren’t alleles with more then 16-17 CTG-repeats in DMPK gene. In IT-15 gene alleles with 20-22 CAG repeats were common in all populations, while longer alleles were represented only in part of them. Taken together, these facts indicate that the alleles groups which are a reservoir for de novo mutations could differ in their repeats number in two genes. According to our suggestion, the threshold value between stable alleles and alleles predisposed to expansion is 17-18 repeats for DMPK gene and 22-23 repeats for IT-15.

P0907. Genetic epidemiology of mendelian diseases in Yakutia

L. Nazarenko, A. Nogovicina, O. Saljukova

Institute of Medical Genetics; Tomsk, Russian Federation

lnaz@img.tsu.ru

We have studied the incidence and prevalence of monogenic disorders in 100000 inhabitants of native population of Yakutia, situated at nothern-east part of Siberia. The maximal morbidity rate of autosomal dominant (AD) diseases was defined for eastern Yakutian subpopulations (2.4 x 10-3) and minimal – for western subpopulations (1.4 x 10-3). The values of morbidity rate of autosomal recessive (AR) diseases ranging from 0.4 to 0.9 x 10-3. The mean values of morbidity rate of mendelian diseases for Yakutian population were 1.7 x 10-3 for AD, 0.7 x 10-3 for AR and 0.4 x 10-3 for X-linked diseases. The total morbidity rate of monogenic diseases is 2.8 per 1000 person. In Yakutian population we observed a high frequency of spiniocerebellar ataxia I (SCA1 - MIM 164400), dystrophia myotonica (DM - MIM 160900) and methemoglobinemia (NadH-Cytochrome b5 Reductase deficiency - MIM 250800). The study of hereditary diseases will give us the opportunity to create a prophylactic register of hereditary diseases and come to the prospective medico-genetic consultations.

P0908. Haplotype Analysis of Apolipoprotein B Gene in Korean Essential Hypertensives

B. Kang¹, K. Kim¹, J. Shin², C. Lee²

¹Seoulin Bioscience Institute, Seoulin Bioscience, Co., Ltd.; Seoul, Republic of Korea; ²School of Biological Sciences, Seoul National University; Seoul, Republic of Korea

kby_99@yahoo.com

Essential hypertension is considered to be a multifactorial disease that is influenced not only by environmental factors but also by genetic factors. Alterations of lipid metabolism in plasma have been reported to be related to an increased risk of essential hypertension. The purpose of this study was to estimate haplotype frequencies of apolipoprotein B(apo B) gene in Korean population and investigate the relationship between haplotypes of this gene and essential hypertension. In order to estimate the haplotype frequencies, Pvu II, Xba I and Eco RI RFLPs of apo B gene were used as genetic marker. There were no significant differences in allele, genotype or haplotype frequencies between normotensives and essential hypertensives. However, Xba I and Pvu II RFLPs of apo B gene were significantly associated with plasma total cholesterol and triglyceride levels in essential hypertensive groups, respectively(P<0.05). Therefore, our result suggest that these two RFLPs of apo B gene may be genetic components of cardiovascular risk factors in Korean essential hypertensives.

P0909. High Incidence of 550delA Mutation in Limb-Girdle Muscular Dystrophy Type 2A (LGMD2A) in Croatia

A. Milic¹, B. Kovac², F. Leturcq³, N. Zurak¹, N. Canki-Klain¹

¹Croatian Institute for Brain Research and Department of Neurology, Zagreb University Medical School; Zagreb, Croatia; ²Department of Neurology, University of Osijek; OSIJEK, Croatia; ³Laboratoire de Biochimie et GenetiqueMoleculaire, Hopital Cochin; Paris, France

stella24hrv@yahoo.com

We report preliminary data concerning 28 LGMD2A patients from Croatia's population of 4,8 million people. Sequence analysis of patients from six apparently non related families from a small rural community revealed homozygocity for a 550delA. Analysis of additional 4 families from different parts of the country revealed homozygous 550delA patients in two families, compound heterozygotes 550delA / Y537X in one family, and compound heterozygotes for a 550delA /one unknown mutation. These results motivated us to develop rapid screening method for 550delA. The fact that 550delA creates restriction site for Bsa AI enzyme was used to distinguish wild type PCR product (210bp) unchanged after the incubation with the enzyme, from mutated type (cut in two bands of 120 and 90 bp). Application of this method on 28 patients from 18 families permitted us to identify; homozygous 550delA patients in 8/18 families; compound heterozygotes 550delA in 7/18 families. Second mutated allele is not identified in 4 families. In five typical LGMD2A patients from 2 families both alleles are unknown. In conclusion our preliminary study shows that 550delA mutation accounts for 63,8 %( 23/36) of CANP3 chromosomes.

P0910. Distribution the most common cystic fibrosis mutations in Balkans, Central & Eastern European populations.

M. Macek Jr.¹, M. Macek¹, T. Doerk², E. Dequeker³, J. Cassiman³

¹CF Center-Institute of Biology and Medical Genetics; Prague 5, Czech Republic; ²IHG, MNH; Hannover, Germany; ³Cnt. Med. Genet. Univ.; Leuven, Belgium

pavel.roubic@lfmotol.cuni.cz, milan.macek.jr@lfmotol.cuni.cz

We have analyzed 151 CFTR mutations in representative groups of CF patients from 20 Balkans, Central and Eastern European countries (BCEC; 73 AL, 290 A, 94 BY, 216 BG, 138 HR, 247 CZ, 30 EE, 359 EL, 488 HU, 18 LV, 47 LT, 113 RM, 863 PL; 28 RO, 1,281 RU, 117 SK, 66 SI, 335 TR, 260 UA, 199 YU; a total of 5262 cases). All centers screened for the 12 most common CF mutations, while the entire CFTR coding region was scanned in A, BG, CZ, EE, EL, PL, RM, RU, SI and TR. The 25 most common CF alleles comprise; deltaF508 (n=5636/53.55% of all CF chromosomes), G542X (227/2.16), N1303K (164/1.56), CFTRdele2,3/21kb/ (120/1.14), 3849+10kbC->T (67/0.64), 621+1 G->T (63/0.59), R553X (62/0.59), W1282X (56/0.53), 1677delTA (47/0.45), G551D (44/0.42), 1717-1G->A (34/0.32), 2184insA (32/0.30); 2143delT, R334W (28/0.27 each); R347P (27/0.26); 2183AA->G (24/0.23), 2789+5 G->A (21/0.20), R1162X (19/0.18), 394delTT (17/0.16), 1898+1G->A (16/0.15), R1070Q (15/0.14), G85E (13/0.12); 457TAT->G, E822X (13/0.12 each) and R1158X (11/0.1). The spectrum and distribution of mutations markedly differs from that observed in Western Europe (Hum Mutat 10;135,1997 extrapolated data) with statistically significant differences being observed in deltaF508, G542X, CFTRdele2,3 /21kb/ and G551D mutations (p<0.001). Inclusion of common BCEC mutations into screening panels will considerably improve the genetic diagnosis of CF. Supported by grants;#6250-3, 6411-3, 000000064203; #111300003, ME258, OK192; LN00A079; ERB IC20 CT96 0058. (*)We acknowledge contribution of members of the INCO-BIOMED CF Mutation Analysis Consortium that could not be listed by their names due to abstract space constraint.

P0911. The mutation spectrum of Hyperphenylalaninaemia in Southern Ireland; the population genetics of the Irish revisited

D. T. Croke

Royal College of Surgeons in Ireland; Dublin, Ireland

dtcroke@rcsi.ie

Phenylketonuria (PKU), the most severe form of hyperphenylalaninaemia (HPA), has an incidence in the Republic of Ireland of approximately 1 in 4,500 live births. The purpose of this study was to investigate the HPA mutation spectrum in the Republic, to determine mutation-haplotype associations and to investigate genotype-phenotype correlations. Mutational analysis was carried out on a cohort of 279 unrelated hyperphenylalaninaemic patients (558 independent alleles) by standard methods. Mutations were detected in 92% of alleles revealing a total of 29 mutations. A large proportion of alleles (63.6%) were accounted for by three mutations; R408W (41.0%), F39L (12.1%) and I65T (10.1%). The predominant haplotype associations for these common mutations were; R408W-1.8-242, F39L-1.8-238 and I65T-1.8-246. The R408W-1.8 mutation exhibits an East to West gradient of increasing relative frequency across Europe peaking in Ireland. Our data demonstrates that this gradient continues across Ireland and peaks in Connacht, the most Westerly province. Combining our data with that from Northern Ireland (Zschocke et al., Am J Hum Genet 1995; 57; 1311-1317) we have produced a merged data-set for the island of Ireland. Our analysis of this data concurs with previous suggestions that the English contribution to the Irish genepool was small and that gene flow between Ireland and Scandinavia was unidirectional. It further demonstrates that the province of Ulster has been a zone of population admixture between Ireland and Scotland.

P0912. Study of the mutations causing familial hypercholesterolemia in St.-Petersburg

J. Tatisheva, F. Zakharova

Institute for Experimental Medicine; St.-Petersburg, Russian Federation

July@postbox.spb.ru

Familial hypercholesterolemia (FH) is an inherited metabolic disease with a dominant mode of inheritance. It is quite common in most human populations (1;500) and results in drastical reduction of low-density lipoprotein (LDL) catabolism followed by premature coronary heart disease (CAD). The disease is caused by mutations in the LDL receptor gene. In order to develop presymptomatic diagnostic tools for management of the FH we aimed to study the LDL receptor gene mutation spectrum in St.-Petersburg (Russia). We have created a DNA bank from 100 unrelated patients with clinical picture of FH. Separate exons of the receptor gene were amplified by polymerase chain reaction (PCR) and screened for presence of the mutations via combined single-strand conformation polymorphism-heteroduplex analysis (SSCP-HA). The fragments showing shifted mobility in polyacrylamide gel electrophoresis were sequenced by method of Sanger. Up-to-date, we have identified six mutations - C74X, 347delGCC, А130Р, С146R, C188Y, G578E, all of them besides C74X were novel. Rapid methods for mutation detection were developed. Cosegregation of mutations and high cholesterol levels proves the role of mutations in disease development. St.Petersburg population was found to be polymorphic in many postions of the LDL receptor gene - 447 Т/С, 750C/T, 1170A/G, 1413 G/A, 1545 С/Т, 1773 T/C, 2177 C/T, 2231 G/A (exons 4, 5, 8, 10, 10, 12, 15, 15 correspondingly). Using DNA methods we have confirmed or set diagnosis of FH in 34 patients and confirmed absence of FH in 14 relatives of patients.

P0913. International variation in the prevalence of the C677T variant of the 5,10 methylenetetrahydrofolate reductase (MTHFR) gene among well-defined populations.

R. Tenconi¹, P. Mastroiacovo¹, -. MTHFR study group², -. MTHFR study group²

¹International Center for Birth Defects; Rome, Italy; ²International Clearinghouse Birth Defects Monitoring Systems; Italy

icbd@icbd.org

MTHFR encodes a critical enzyme in folate and homocysteine metabolism. The C677T allele of the MTHFR gene is reportedly common in some populations and has been associated with an increased risk for spina bifida and for adult cardiovascular diseases. Although many reports the C677T prevalence are available, few were based on well-defined populations. We report data from 12 countries in Europe (Finland, France, Hungary, Italy, Netherlands, Spain), Israel, Russia, China, and the Americas (Canada-Alberta, United States-Georgia, Mexico). We attempted to sample populations that were well defined; our approaches included using newborn blood spots from specific geographic areas or from consecutive liveborn babies from hospitals. Typically these populations were also under birth defect monitoring. We found a high prevalence of C677T homozygotes in Mexico (32%), Northern China (20%), and Southern Italy (20% in Catania, 26% in Campania). Such prevalence was intermediate in Northern Italy, France, and Spain (15, 13, and 12%, respectively), and it was low in Finland, the Netherlands, and Russia (4,6, and 7%, respectively). In the sample from the United States (Georgia), we noted racial variations; prevalence was high among Hispanics (18 percent), low among African Americans (3 percent) and intermediate among Whites (11 percent). These findings underscore the geographical and racial variability of the C677T variant. Such data should be useful to better design studies of gene-environment interactions involving the MTHFR gene, as well as for future applications in population-based prevention of conditions related to such gene.

P0914. Two novel mutations in the HEXA gene found in TSD carriers of Iraqi Jewish origin and haplotype analysis suggesting a founder effect.

L. Peleg¹, M. Karpati¹, B. Goldman¹, E. Gazit², A. Frisch³

¹Danek Gertner Institute of Human Genetics, Sheba Medical Center; Tel Hashomer, Israel; ²Department of Pediatrics, Sheba Medical Center; Tel Hashomer, Israel; ³Felsenstein Medical Research Center; Rabin Medical Center, Petah-Tikva, Israel

lea_peleg@hotmail.com

Among Jews, Tay-Sachs disease (TSD) was thought to be restricted to individuals of Ashkenazi (carrier frequency of 1;29) and Moroccan (1;110) descent. However, an elevated carrier frequency of 1;140 was found also among Iraqi Jews (IJ). A G749 to T change resulting in the substitution of Glycine250 to valine was recently identified by us in 41% of the TSD carriers (24/58). We now report an additional novel mutation; a C1351 to G transversion resulting in a change of Leucine451 to valine in 29% (17/58) of IJ carriers. Neither of the mutations was found in 100 non-carriers of the same ethnic group. Haplotype analysis was conducted using 6 markers closely linked to the HEXA gene. Mutation G749 to T was associated with allele #1 of D15S131 in 70% of heterozygotes compared with 10% in ethnically matched controls (chi square=46.9, p=10-5) and with allele #2 of D15S1025 in 68% of heterozygotes compared with 43% in controls (chi square=4.4, p=0.03). The calculated linkage disequilibrium was 0.66 and 0.44 respectively. Mutation C1351 to G was also associated with allele #1 of D15S131 (chi square=7.0, p=0.008) yielding a linkage disequilibrium of 0.3 yet, no significant linkage was found with D15S1025. These results suggest a possible common ancestor for each of the HEXA mutations in Iraqi heterozygotes. The C1351 to G mutation had probably occurred earlier than G749 to T, as it shows less association with the adjacent markers. Since these mutations have not been documented in Ashkenazi Jewish carriers, they were probably introduced after the 70 AD Exile.

P0915. Notification of Cystic Fibrosis as primary cause of death in South and Southeast Brazil, from 1981 to 1995

C. L. A. Paiva, L. Janotti, C. Naurath, A. Essinger, R. Gonēalves, E. Baltazar, S. R. Middleton, S. R. dos Santos

Universidade do Rio de Janeiro; Rio de Janeiro, RJ, Brazil

clapaiva@prolink.com.br

Cystic Fibrosis (CF) is an autossomal recessive chronic disease mapped to 7q31-q32. Lung disease accounts for ~95% of its morbidity and mortality. The lungs of CF patients are compromised by persistent bacterial infection from early age. Of the congenital disorders CF is one of the most common with ~1 in 2000 Caucasians affected, with carrier frequency as high as one in 25 in Caucasians of northern European ancestry. There was ample motivation to investigate the notification of CF as primary cause of death in South and Southeast Brazil, since these regions have a multiracial mixed population including Caucasians from European origin. Our hypothesis was that there was under estimation of deaths from CF, since the disease has not been currently recognised mainly in the inland parts of the country. CF symptoms, such as pulmonary infection, and diarrhoea,as well as malnutrition (caused by lack of pancreatic enzymes secretion) are also common causes of death of non-CF infants in Brazil. Our data were extracted from the Brazilian CD-rom entitled "Sistema de informacao sobre mortalidade/1979-1997" which contained data of all death declarations of the country.The frequencies of CF death notifications per 100,000 inhabitants per year for Brazil and for each south or southeast state were calculated for comparison. Chi-square test was performed by comparing the values of each state against that of Brazil and of Sao Paulo(SP),the more developed state. Our results may indicate that SP has a more precise notification of CF as primary cause of death than other states of Brazil

P0916. Intragenic recombination events and de novo mutations in the DMD gene; rapid detection by analysis of microsatellite markers.

M. Kaczmarek¹, D. Napierala¹, M. Szalata¹, R. Slomski^1,2

¹Institute of Human Genetics; Poznan, Poland; ²Department of Biochemistry and Biotechnology; Poznan, Poland

slomski@au.poznan.pl

Studies of allele frequencies of three microsatellite markers revealed few new alleles of each marker in Polish population. All new alleles are very rare. The number and frequency of alleles are similar for Polish and Caucasian population. Only one discrepancy occurred between our studies and previous data and it concerned size of the most common allele of the DMD(CA)5'II marker. The three markers were run in a multiplex reaction to determine carrier status for sisters of affected child from 36 DMD/BMD families. In two families carrier analysis concerned an aunt of an affected child and in one case a cousin. At least one marker was informative in all studied families. All three markers were informative in seven families. In seven cases deletion of the DMD(CA)45int marker in affected children occurred. It gave possibility of direct identification of mutations in female relatives. Additionally direct carrier detection was possible in one family with large deletion including the DMD(CA)5'II marker. However, in order to distinguish homozygous and hemizygous females, DNA samples of both parents are necessary. Three cases of homo-/hemizygous mother occurred in our investigation. Unfortunately, grandparents' DNAs were not available and in these cases carrier status of mothers was not determined. Direct approach failed also in two families, in which both mother and father had the same allele and it was impossible to estimate whether a sister is homo- or hemizygous. New mutations were observed in seven families. In four of them deletion of marker occurred in an affected child while mother was heterozygous. In three other cases the same haplotype occurred in an affected child and a healthy brother. Additionally, seven cases of intragenic recombination events were detected. Five of them occurred in DNA obtained from sisters. Only two markers were informative in all these cases, so it was impossible to determine which marker is linked with a mutation.

P0917. (AC) repeats of 5'beta globin gene and its relation with sickle cell anemia in Mexican population.

R. I. E. Peńaloza¹, P. Delgado¹, T. Sedas², C. Hernandez³, L. Buentello⁴, D. Arenas¹, F. Salamanca¹

¹Unit of Medical Investigation in Human Genetics; Mexico City, Mexico; ²Regional Hospital; Veracruz, Mexico; ³Regional Hospital; Acapulco, Mexico; ⁴Institute of Antropological Investigations, UNAM; Mexico City, Mexico

rosenda@servidor.unam.mx

The (AC) repeats are been found in the human genomic very frequently. Some of them are related with pathologies. The (AC)n of 5' beta globin gene end was studied in normal Mexican population. We founded three alleles. The first A was the most common, B was intermidiat frequency and C only in Purépecha and Mestizo population was found . Now we reported its relation with sicle cell anemia of Mexican patients. The allele B it was found more frequente in beta-S mutation. Moreover, two news alleles there are in the Mestizo Mexican population studied, that have high African component. This finding are useful for better knowling of this human group.

P0918. The Distribution of Bardet-Biedl Syndrome Loci in the Newfoundland Population

W. S. Davidson¹, M. O. Woods², J. S. Green², P. S. Parfrey²

¹Simon FRaser University; Burnaby, BC Canada; ²Memorial University; St. John's, NF Canada

wdavidso@sfu.ca

Bardet-Biedl Syndrome (BBS) is a rare, genetically heterogeneous (>6 loci), recessive disorder characterized by dysmorphic extremities, retinal dystrophy, obesity, male hypogenitalism, and renal disease. We have identified 22 families with BBS in Newfoundland, 17 of whom participated in a genetic linkage study. A founder effect was associated with BBS1 in 5 families, 2 families have BBS2, and 1 is linked to BBS3. Another Newfoundland family positioned BBS5 on chromosome 2q. No BBS4 families have been identified. A genome wide scan of family B13 (excluded from BBS loci 1-5) suggested that the locus for BBS6 was at 20p12. Fine mapping reduced the critical interval to 1.9cM encompassing a chaperonin-like gene associated with McKusick Kaufman Syndrome (MKKS). Mutation screening of the MKKS gene identified combinations of 3 mutant alleles in five BBS families and confirmed that MKKS and BBS6 result from mutations in the same gene. Of the 3 unassigned families, one has been excluded at BBS1-6, implying that another disease locus exists. The remaining 2 families are uninformative due to pedigree structure. We conclude that, unlike the rest of North America and Europe, BBS6 accounts for approximately one third of BBS in Newfoundland. We have no explanation at this time for the high incidence of BBS in this relatively small, isolated population or for the presence of such a diversity of loci (at least 6) or the number of different mutations (at least 8).

P0919. Analysis of autosomal dominant polycystic kidney disease (ADPKD) in the Newfoundland population.

D. S. Compton¹, W. S. Davidson¹, J. S. Green², E. L. Dicks², P. S. Parfrey²

¹Simon Fraser University; Burnaby, Canada; ²Memorial University of Newfoundland; St. John's, Canada

dcompton@sfu.ca

The Newfoundland population is ideal for examining founder effects and for studying the incidence of hereditary disease in what is essentially a closed community. Autosomal dominant polycystic kidney disease (ADPKD) is one of the most common genetic diseases in humans. Mutations in any one of at least two different genes are thought to be responsible for this disease and two genes, PKD1 and PKD2, have been mapped to 16p13.3 and 4q21-23, respectively. Fifteen Newfoundland families with ADPKD have been examined using microsatellite markers flanking and within the PKD genes on chromosome 16 and chromosome 4. Eight of the fifteen families showed linkage to PKD1 whereas four were linked to PKD2. The other three families were uninformative due to the structure of the family. Comparisons of the haplotypes of the families linked to PKD1 revealed that four families share a partial haplotype suggestive of an ancestral founder disease chromosome. The other four PKD1 haplotypes suggest that these all resulted from independent mutation events. Comparisons of the PKD2 haplotypes revealed that two families shared a haplotype and sequencing of exon 6 showed that both families shared the same R464X mutation. The other two PKD2 haplotypes suggest that they resulted from independent mutation events. Mutational analysis is currently underway to determine the exact number of founder mutations for ADPKD in the Newfoundland population. (Supported by the Kidney Foundation of Canada and the Medical Research Council of Canada)

P0920. Frequency of the C282Y mutation of the HFE gene in Ukraine.

V. Pampuha¹, A. Cherchenko², V. Rozumenko³, I. Lubyanova⁴, L. Livshits⁵

¹Institute of Molecular Biology & Genetics National Academy of Science of Ukraine; Kyiv, Ukraine; ²Institute of Neurosurgery Academy of Medical Science of Ukraine; Kyiv., Ukraine; ³Institute of Neurosurgery Academy of Medical Science of Ukraine; Kyiv, Ukraine; ⁴Institute for Occupational Health Academy of Medical Science of Ukraine; Kyiv, Ukraine; ⁵Institute of Molecular Biology and Genetics National Academy of Science of Ukraine; Kyiv, Ukraine

livshits@imbg.org.ua

Hereditary hemochromatosis (HH) is a common autosomal recessive genetic disorder in Caucasians. Hemochromatosis is characterized by increased gastrointestinal iron absorbtion in excessive uptake and persistant accumulation of iron in the body. Classical symptoms of hemochromatosis include liver disease, diabetes mellitus, myocardiopathy, arthralgia, skin siderosis. The gene of hemochromatosis HFE is located on the short arm of the chromosome 6 telomeric to major histicompatibility complex. A single missence mutation C282Y is responsible for hemochromatosis in approximately 80% of all cases. C282Y frequencies of 1-10 % are observed in Europeans. We analysed the frequency of the C282Y in 260 individuals with Slavic origin from three Ukranian regions (West, Central , East) and in 100 individuals from isolated Crimean tartars. Frequency of C282Y mutation in different regions of Ukraine ranged from 2% to 3%; frequency of this mutation in pooled Ukrainians was 2,5% and in Crimean tartars was 1%. The high prevalence of the C282Y mutation in Ukraine suggests that population screening for the C282Y mutation could be highly advantageous in terms of preventive health care.

P0921. Detection of the CFTR gene mutations in cystic fibrosis patients and in men with azoosperma from Estonia.

M. Teder¹, T. Kahre², K. Kaasik³, M. Punab⁴, A. Sõritsa⁵, A. Metspalu⁶

¹Estonian Biocentre; Tartu, Estonia; ²Centre of Molecular Diagnostics, Tartu University Clinics; Tartu, Estonia; ³Institute of Cell and Molecular Biology, University of Tartu; Tartu, Estonia; ⁴Surgery Clinic, Tartu University Clinics; Tartu, Estonia; ⁵Tähe Private Clinic; Tartu, Estonia; ⁶Institute of Molecular and Cell Biology, Centre of Molecular Diagnostics, University of Tartu; Tartu, Estonia

mteder@ebc.ee

Since the identification of the cystic fibrosis transmembrane conductance regulator (CFTR) gene more than 900 mutations, responsible for CF have been described. Involvement of CFTR gene mutations has also been confirmed in otherwise healthy but infertile men with obstructive azoospermia from congenital bilateral or unilateral absence of vas deference, and epididymal or ejaculatory duct obstruction. The aim of the study was to identify the spectrum of CFTR gene mutations in CF patients and in infertile men with azoospermia and oligozoospermia in Estonia. Thirty families with CF patients were studied. All 27 exons and their flanking sequences were scanned by SSCP and DGGE analyses. Two CFTR gene mutations were found to be common in Estonia; ∆F508 in 31 (51.7%) alleles and 394delTT in 8 (13.3%) alleles. Eight rare mutations were detected, all described earlier in other populations. We did not find any of the mutations more common in the European populations like G542X(2.6%), N1303K(1.6%), G551D(1.5%) or W1282X(1.0%). In conclusion, about 80% of mutations were identified in CF patients. The study group of men with fertility problems included 20 individuals with obstructive and 29 individuals with nonobstructive azoospermia and 107 men with severe oligozoospermia. Patients were tested for three CF mutations ∆F508, 394delTT, R117H and a IVS8-T variant. The patients with obstructive azoospermia were additionally tested for more mutations by INNO/LipaTMCFTR12 and INNO/LipaTMCFTR17+Tn kits. In the result of these tests only ∆F508 mutation and 5T allele were revealed. We identified two heterozygotes ∆F508/5T and two heterozygotes with 5T allele in one chromosome in obstructive azoospermia group. One compound heterozygote ∆F508/P508C, one 5T homozygote and one patient with only one 5T allele were detected in nonobstructive azoospermia group. Among the oligozoospermic men 14 patients revealed 5T in one allele and one carried ∆F508 muation on one chromosome. In comparison with the control group of normal men, the frequency of ∆F508 mutation and 5T allele was considerably higher only in patients with obstructive azoospermia; 5% versus 0,6% for ∆F508 and 10% versus 4,2% for 5T.Therefore CFTR gene mutation screening and genetic counselling should be recommended to males with obstructive azoospermia and there partners to reduce there risk for CF child by assisted reproduction procedures.

P0922. Factor V Leiden and Factor II Mutation frequencies in Turkish pregnant women ; preliminary findings

R. Bircan¹, C. Erzik¹, F. Akbas², I. Güney¹, E. Oral², B. Cirakoglu¹

¹Marmara University Faculty of Medicine Dept of Medical Biol&Genetics; Istanbul, Turkey; ²Istanbul University Cerrahpasa Medical Faculty Dept of Obstetrics and Gynecology; Istanbul, Turkey

rifatbircan@yahoo.com

Objective; The study's objective was to evaluate association between hereditary coagulation abnormalities including factor V Leiden and prothrombin G20210A and obstetrical complicalitions such as preeclampsia, abruptio placentae, intrauterine fetal growth retardation (IUFGR)and stillbirth in Turkish population. Material and Methods ; DNA was extracted from whole blood of 84 women -42 complicated pregnancies and 42 control cases. The polymerase chain reaction was used to amplify exon 10 of factor V gene and untranslated end of prothrombin gene followed by enzymatic digestion with Mnl I and Hind III for mutation detection respectively. Results ; The mutation at nucleotide 1691 in factor V gene was detected in 8 of the women with obstetrical complication and in 3 of the women with normal pregnancies (19% and 7.1% respectively; p<0.005). This mutation was detected 5 of the 28 women who had preeclampsia (17.9% p<0.005), 2 of the 5 women who had stillbirth (40%) and 1 of the 7 pregnant women who had IUFGR (14%). Prothrombin G20210A polimorphism detection results under evaluation. Conclusions ; The Leiden mutation is relatively common in pregnant women with preeclampsia. Although the mutation was detected in women who had stillbirth and IUFGR, the number of subject is not meaningful for statistical evaluation. It is suggested that factor V Leiden and prothrombin G20210A mutation detection screening can be used for pregnant women under risk.

P0923. Variable expression of presenilin 1 is not a major determinant of risk for late-onset Alzheimer's Disease

R. Rademakers¹, G. Roks², M. Cruts¹, J. Theuns¹, J. Houwing-Duistermaat², S. Serneels¹, A. Hofman², M. Breteler², C. Van Broeckhoven¹, C. van Duijn², B. Dermaut¹

¹Flanders Interuniversity Institute for Biotechnology (VIB); Antwerpen, Belgium; ²Erasmus Medical Center; Rotterdam, The Netherlands

rrademak@uia.ua.ac.be

We have demonstrated a significant association between early-onset Alzheimer's disease (EOAD) and an allele in the promotor of presenilin 1 (PSEN1), significantly decreasing PSEN1 expression in vitro. For late-onset Alzheimer's disease (LOAD), numerous studies have reported inconsistent associations with a polymorphism in intron 8 of PSEN1. We hypothesized that these conflicting reports in LOAD might be due to linkage disequilibrium between this intronic PSEN1 polymorphism and the functional promotor polymorphism. To determine the genetic contribution of the PSEN1 intronic and promotor variations to LOAD we analyzed both variations in 356 LOAD patients and 230 controls in a population-based case-control study. In addition, we reanalyzed all published literature on the PSEN1 intronic polymorphism in a meta-analysis. No significant association was found with the PSEN1 intronic or promotor polymorphism in our case-control sample. In the meta-analysis no major differences between patients and controls were found for the PSEN1 intronic variation. Together our results do not support a major role for variable expression of PSEN1 in LOAD.

P0924. Genotype and gene frequency analysis of beta thalassemia in Chaharmahal Bakhtiary Province (Iran)

M. Hemmat

Shahrekord University; Shahrekord, Islamic Republic of Iran

MO_hemmat@yahoo.com

The beta thalassemia is an autosomal ressesive genetic disease, caused by a loss in activity in one or both alleles of the beta globin gene. The beta thalassemia is well known as one of the major genetic disease in ten provinces located in north and south of Iran. It has been estimated that more than 3000000 gene carriers exist in these regions. The present study has been focused on beta thalassemia in Chaharmahal Bakhtiary province located in south of Iran. Hardy-Weinberg principle was utilized in order to determine the genotype and gene frequencies and estimation of the percentage of homozygote and heterozygote carriers in this population. The frequency of all the affected genes (q)and normal gene (p)were estimated as 0.016 and 0.984 respectively. The frequency of the number of people expected to be homozygous normal, heterozygous carriers (minor thalassemic) and homozygous affected (major thalassemic) were determined as p2=0.9682, 2pq=0.0315 and q2=0.0003 respectively. From these data it was concluded that, 96.82 percent of human population in this province were homozygous normal, 3.15 percent were heterozygous carriers (minor thalassemic)and 0.03 percent were homozygous affected and demonstrated major thalassemia.

P0925. Incidence of XmnI RFLP among Iranian Phenylketonuria Patients as a Marker Contributing to Carrier Detection

M. seifati¹, M. Hosseini Mazinani¹, N. Hosein Pur¹, M. Koochmeshgi², B. s. Behboudi³

¹national research center for genetic engineerig and biotechnology; tehran, Islamic Republic of Iran; ²Iran medical university; Tehran, Islamic Republic of Iran; ³Tehran university; Tehran, Islamic Republic of Iran

mseifati@nrcgeb.ac.ir

Phenylketonuria (PKU) is the most common error of amino acid metabolism. It is an autosomal recessive disorder caused by a variety of mutations in the Phenylalanine Hydroxylase (PAH) gene. These mutations exhibit a high degree of association with specific haplotypes in different populations. These haplotypes are composed of different kinds of genetic markers. An Extended- haplotype is a combination of different RFLPs (EcoRV, XmnI, MspI, EcoRI, PvuIIb, PvuIIa, BglII,…), while a mini-haplotype consists of one specific STR, one specific VNTR and XmnI RFLP of the PAH gene. Since STR and VNTR are polymorphic markers, they have higher polymorphism information contents (PICs) in comparison with RFLPs (each having only two forms), and thus they represent to be more suitable markers in carrier detection procedure done by segregation analysis. However, some RFLPs are needed beside these polymorphic markers to yield highly efficient carrier detection results. One of the several markers we have studied in order to identify their degree of heterozygosity and PIC is XmnI, which is the only common marker in extended- and mini-haplotype. For this purpose we extracted DNA from PKU patients’ blood samples and amplified exon 8 of PAH gene in them. Among 124 alleles, only 8 alleles (6.5%) were digested with related enzyme (ASP 700) . Thus the degree of heterozygosity and PIC of XmnI RFLP in Iranian PKU patients are 0.12155 and 0.11416 respectively. Combining this RFLP with STR and VNTR polymorphic markers in segregation analyses will increase the efficiency of carrier detection procedure.

P0926. Identification of the Major Mutation in PAH Gene among Iranian Phenylketonuria Patients

N. Hosein Pur, M. Hosseini Mazinani, J. Koochmeshgi, K. Moshiri, Z. Khazaee, K. Anisi, M. Seifati

National Research Center for Genetic Engineering and Biotechnology; Tehran, Islamic Republic of Iran

nasim@nrcgeb.ac.ir

Phenylketonuria (PKU), the most common inborn error of amino acid metabolism, is an autosomal recessive disorder caused by more than 400 mutations in Phenylalanine Hydroxylase gene (PAHdb). This gene, 90 kilo bases long, includes 13 exons and is located in 12q22-q24 region. Distribution patterns of mutations in the PAH gene are specific to each population. In this study,the incidences of IVS10nt546, R261X, R261Q and R408Q mutations among a group of Iranian PKU patients from all over the country were determined. IVS10nt546, occurring in intron 10 of PAH gene, is the most common mutation among Mediterranean populations. The next two ones have a good chance to occur since they belong to a hotspot in exon 7 of this gene,and R408Q mutation,existing in exon 12, is one of the only two reported mutations among Persian ethnicity in PAHdb to date. In order to study these mutations, we amplified exons 7, 11 and 12 of PAH gene, extracted from blood samples of PKU patients. PCR products were then digested with DdeI, HinfI, DdeI and HaeIII restriction enzymes for detecting R261X, R261Q, IVS10nt546 and R408Q mutations respectively. Among 130 alleles studied, 40 alleles (33.3%) were positive for IVS10nt546 mutation and among 80 alleles, 5 alleles(6.25%) were positive for R261Q and 2alleles(2.5%) for R261X mutations. No allele was positive for R408Q mutation. According to this data, IVS10nt546 mutation seems to be the major mutation in Iranian PKU patients. This result is in accordance with the prevalent mutation in Mediterranean populations.

P0927. Molecular diagnosis of Rett syndrome; Detection of the prevalent mutations in MeCP2 gene in Czech and Slovak patients.

R. Rosipal¹, J. Zeman¹, J. Hadač², N. Mišovičovį³, S. Nevšķmalovį¹, P. Martįsek¹

¹Charles University, Faculty of Medicine I; Praha, Czech Republic; ²Thomayer Hospital; Praha 4, Czech Republic; ³Faculty Hospital; Martin, Slovakia

rrosipal@hotmail.com

Rett syndrome is an X-linked dominant neurodevelopmental disorder affecting 1 in 10,000 to 15,000 females worldwide (1) and is caused by mutations in X-linked MECP2 gene, encoding for methyl-CpG-binding protein 2 (2,3). It plays an important role in the regulation of gene expression. There are 5 prevalent mutations in methyl-CpG binding protein 2, causing Rett syndrome. Four of them are detectable by restriction analysis. In this study we present the results of the search for four prevalent mutations in the gene for methyl-CpG binding protein 2 in Czech and Slovak patients with Rett syndrome. Twenty two females with clinical diagnosis of Rett syndrome from Czech and Slovak Republics were investigated. Restriction analysis and direct sequencing of PCR products of methyl-CpG binding protein 2 gene revealed 3 different mutations [473 C to T (T158M) in 3 patients, 502 C to T (R168X) in 2 patients, and 808 C to T (R270X) in 1 patient] in six unrelated patients with Rett syndrome. The results of restriction analysis were confirmed by bi-directional direct sequencing. The first study of Rett syndrome in Slavic population shows that three from hot spot mutations in exon 3 of MECP 2 gene are present in 27% of the patients and dictates therefore the strategy in molecular diagnosis of Rett syndrome. Mutation R306C, frequent in Swedish patients (4), was not found in 22 Czech and Slovak patients. 1. Rett, A.; Wien Med. Wochenschr., 116, 1966, 723-726. 2. Amir, R.E., et al. ; Nature Genet., 23, 1999, 185-188. 3. Sirianni, N., et al.; Am. J. Hum. Genet., 63, 1998, 1552-1558. 4. Xiang, F., et al.; J. Med. Genet., 37, 2000, 250-255. Supported by Czech Granting Agency (GACR 301/01/P068 and 302/99/0648).

P0928. Haplotype Analysis of the Most Common Beta-Thalassemia Mutation Intervening Sequence II-1 (IVSII-1) in Iran

T. Parsian¹, N. Ghobadian¹, A. Javadian¹, S. Teimourian², F. Pourfarzad¹, S. Amjadi¹, H. Najmabadi^1,2

¹University of Welfare and Rehabilitation; Tehran, Islamic Republic of Iran; ²Karimi-Nejad Pathology and Genetics Center; Tehran, Islamic Republic of Iran

maryamneishabury@yahoo.co.uk

IVSII-1 is considered the most common beta-thalassemia mutation in Iran. In the North of Iran by the Caspian Sea, over 50% of affected individuals carry this mutation and the overall frequency in Iran is 30%. Restriction enzyme mapping for beta-globin gene cluster is important in systematic studies of single copy genes causing beta-thalassemia. The most frequent mutation in the Iranian population, which leads to beta-thalassemia is the substitution of G to A in the intervening sequence II, nucleotide number 1. We have studied ten affected homozygous for IVSII-1. After restriction digestion for six polymorphic loci (XmnI, HindIII/Ggamma, HindII3'/wb, AvaII/b,RsaI/b, HinfI/b) . We constructed chromosomal haplotype that are related to this mutation. Our result shows that the IVSII-1 mutation is essentially in company with two types of haplotypes about 40% and 30% respectively. These haplotypes are quite different from Mediterranean haplotype for IVSII-1.

P0929. MEFV gene mutations and high carrier rates for familial Mediterranean fever in Turkish FMF patients and healthy population

B. Balci, E. Yilmaz, S. Ozen, R. Topaloglu, N. Besbas, A. Bakkaloglu, M. Ozguc

Hacettepe University; Ankara, Turkey

banubalci@hotmail.com

Familial Mediterranean fever (FMF) is a recessive disorder characterised by episodes of fever and neutrophil-mediated serozal inflammation. The most serious complication of FMF is the development of amyloidosis. Colchicine has been shown to be effective in preventing the attacks of FMF as well as the development of amyloidosis1. The disease affects primarily ethnic groups living around the Mediterranean basin; Ashkenazi Jews, Armenians, Turks, and Arabs. The gene causing FMF has been cloned in 1997, and several mutations characterised in MEFV gene. The aim of this study was to determine the carrier rate in the Turkish population and the mutation frequency in the clinically diagnosed FMF patients. We found a high frequency of carriers in the healthy Turkish population (20%). One healthy asymptomatic individual was found to carry mutations in both alleles (V726A/E148Q). The distribution of the five most common MEFV mutations among healthy individuals (M694V 3%, M680I 5%, V726A 2%, M694I 0% and E148Q 12%) was significantly different (p <0.005) from that found in patients (M694V 51.55%, M680I 9.22%, V726A 2.88%, M694I 0.44%, and E148Q 3.55%). The high incidence of MEFV gene mutations in the Turkish population (gene frequency of 0.011; 95% CI = 0.06-0.15, and carrier rate of 0.20; 95% CI = 0.12-0.28) indicate that new-born screening may be discussed in the future.

P0930. Carrier frequencies of DHCR7 mutations in West-Austria indicate that Smith-Lemli-Opitz syndrome is among the most common autosomal recessive disorders

J. Loeffler, M. Witsch-Baumgartner, H. J. Menzel, H. G. Kraft, G. Utermann

Institute of Medical Biology and Human Genetics; Innsbruck, Austria

witsch-baumgartner@uibk.ac.at

Smith-Lemli-Opitz syndrome (SLOS, MIM 270400) is an autosomal recessive multiple congenital anomaly/mental retardation syndrome of variable severity. Its incidence has been estimated to be approximately one in 20 000 to 60 000 based on case frequency surveys in populations of Caucasian origin. Identification of the underlying gene defect has paved the way for molecular methods to identify carriers. Using allelespecific PCRs we screened 640 anonymous and random blood donor samples from Tirol (part of West-Austria) for the presence of the three most common caucasian SLOS mutations (IVS8-1G>C, W151X, V326L). For the mutation IVS8-1G>C a carrier frequency of 1 in 50 (1;26-1;75, 95% CI) was identified. W151X was much less frequent (1;141-0, 95% CI). V326L was not found in the sample analysed. Based on this data an incidence of SLOS of about 1;10.000 (1;1666 – 1;22 201, 95% CI) is expected. Remarkably, we know only of two unrelated Tirolean SLOS patients. The discrepancy between the expected and observed incidence could be due to misdiagnosed severe cases, death prior to diagnosis, or fetal loss. The high frequency of the functional null mutation IVS8-1G>C suggests a founder effect and/or heterozygote advantage.

P0931. The Genealogy of Fanconi Anaemia Patients Homozygotic for the Type I and Type II Afrikaner Mutations

T. Pearson¹, N. V. Morgan², A. J. Tipping², L. P. Kuyt³, C. G. Mathew², S. Jansen¹

¹Division of Human Genetics, University of the Orange Free State; Bloemfontein, South Africa; ²Division of Medical and Molecular Genetics, GKT School of Medicine; London, United Kingdom; ³Department of Human Genetics, Free University of Amsterdam; Amsterdam, The Netherlands

gnmgtp@med.uovs.ac.za

Fanconi anaemia (FA), a rare autosomal recessive genetic disorder, is known to occur at a higher than normal incidence among the white Afrikaans speaking population of South Africa. The disease is characterised by progressive bone marrow failure, starting predominantly in childhood, various phenotypic abnormalities as well as an increased sensitivity to the clastogenic agents diepoxybutane (DEB) or mytomycin C (MMC), resulting in chromosomal aberrations. Initial genealogical investigations on 12 families affected by FA but with unknown mutational status, identified GN and/or his spouse JDLB as common ancestors. A founder effect, the cause of the high incidence of FA among this population group, was further substantiated by results obtained through complementation studies and mutation screening, showing the predominance of complementation group A and mutation types I, II and III. The existence of three mutation types among the descendants of GN and JDLB raised the possibility of more than one founder pair or family being involved. For this reason an additional genealogical investigation was carried out, limited to seven and six FA parents carrying respectively the Afrikaner types I and II mutations. Apart from GN /JDLB, HDP and wife CD also featured prominently as possible founders for the FANCA Afrikaner type I mutation, whereas either PV/MDP are candidate founders of the FANCA Afrikaner type II mutation.

P0932. Analysis of CTG polymorphism in untranslated 3' region of SCA8 gene in Polish control group

A. Sulek, D. Hoffman-Zacharska, M. Bednarska-Makaruk, J. Zaremba

Institute of Psychiatry and Neurology; Warsaw, Poland

sulek@ipin.edu.pl

Autosomal dominant spinocerebellar ataxias (SCA) are a group of neurodegenerative disorders characterized by progressive ataxia, dysartria and other neurological signs. Genes responsible for different types of SCA (SCA1, 2, 3, 6, 7) have been cloned, showing in coding regions an expanded CAG repeats in mutant alleles. In 1999 a novel CTG trinucleotide expansion in 3' untranslated region of SCA8 gene on chromosome 13q21 was identified. Studies established on controls and patients with linkage to chromosome 13q21 showed tetramodal distribution of CTG repeat number. The prevalence of three non-pathogenic ranges with different frequency was shown. The pathogenic range was estimated as approximately 100 and more CTG repeats and was observed in patients with clinical signs of SCA (we observed such alleles in 3 families). However, alleles with more than 100 CTG repeats were also observed rarely in healthy controls and in unaffected patients' relatives. This phenomenon may be the result of reduced penetrance of the gene. To establish a non-pathogenic range of CTG repeats for Polish population we performed an analysis of CTG polymorphism in a group of 250 Polish healthy controls. In one subgroup consist of 100 persons aged 18-69 years two alleles were found with number of CTG repeats 82 and 100, respectively. In the second subgroup 150 Polish controls age of 65-82 we observed one allele with 95 CTG repeats. The results of our study suggest that only CTG repeat number above 100 may relate to neurological signs.

P0933. Sensoryneural Hearing Loss (snhl); A Complex Trait

A. Z. Spadoni¹, M. T. Trincavelli¹, G. E. Dotto², M. C. Fumagalli¹, M. V. Dalmaso¹, M. V. Chinellato¹, A. Jannelli², M. B. Bravo-Luna¹

¹Faculty of Medicine-UNR; Rosario, Argentina; ²Faculty of Medicine; Rosario, Argentina

mbravoluna@hotmial.com

SNHL is a mono or multifactorial hearing deficit of great significance in our culture, that can be prevented in some cases. Objective; to assess, in a first stage, the prevalence of SNHL etiology in a qualified population aged 0-15 yr. in order to analyse the involvement of the genetic component. Methods; Retrospective study through the analysis of professional records of children and adolescents (N= 250) with hearing disorders, attending public and private specialised institutions of the city of Rosario, that were diagnosed of hearing impairment by audiometry and evoked potentials. Results (%) are presented according the following summarised etiological classification; I) Congenital [32%]; 1) Genetic [35]; 2) Non genetic [65]; a) infections; [96]; b) toxics; [4]. II) Acquired [19]); 1) infections [66]; 2) toxics [17; 3) others [17]. III) Indeterminate [40]; and IV) multifactorial [9]. Conclusions; It is striking the high prevalence of the indeterminate and multifactorial groups, within these groups it is probable the occurrence of genetic disorders, either syndromic or no syndromic, that have not been adequately diagnosed. Presently, we are studying this population, considering SNHL as a complex trait, which is our hypothesis, through segregation analysis, and improving diagnosis regarding its etiology.

P0934. Mutation Spectrum of Connexin 26 in Iranian Patients with Autosomal Recessive Non-syndromic Sensorineural Hearing Loss

S. Sahebjam¹, N. Kouchakian¹, S. Arzhangi¹, M. Daneshmandan¹, M. Javan¹, B. Cucci², M. Menezes², R. Smith², H. Najmabadi¹

¹Genetics Research Center,Welfare Sciences and Rehabilitation University; Tehran, Islamic Republic of Iran; ²Molecular Otolaryngology Research Laboratories,Department of Otolaryngology Head and Neck Surgery,University of Iowa; Iowa, IA United States

sahebjam.s@angelfire.com

Hereditary deafness is one of the most common inherited sensory disorders, which affects 1;2000 newborns. About 70% of hereditary deafness is non-syndromic. Autosomal recessive forms (DFNB) make up about 85% of these cases. Mutations in the connexin 26 (Cx26) gene are associated with a type of autosomal recessive non-syndromic sensorineural hearing loss (ARNSHL) known as DFNB1. DFNB1 accounts for about 50% of congenital severe to profound ARNSHL. The most common mutation 35 del G is found in over two-thirds of individuals with DFNB1. The present study investigates the mutations of connexin 26 found in patients with ARNSHL in the Iranian population. Fifty-five unrelated patients with congenital ARNSHL were studied. DFNB1 mutations were found in 9 patients. Four patients were homozygotes for 35 del G mutation. Compound heterozygousy of 35 del G/ W24X and R184/ splice site mutation G-to-A 5’ss was the cause of ARNSHL in 2 patients. Three patients carried only a single 35 del G mutation. A second mutation is under sequencing. Mutations of other patients are under study. The frequency of 35 del G allele was much lower than reported studies from other countries. These results suggest that the different types of mutations affect ARNSHL according to ethnic background. The study is ongoing.

P0935. Why are Mutations at the Connexin 26 Locus Such a Frequent Cause of Genetic Deafness?

W. E. Nance, A. Pandya, X. Liu

Virginia Commonwealth University; Richmond, VA United States

nance@hsc.vcu.edu

Heterosis, founder effects and mutation hot spots, have been suggested to explain the high frequency of Cx26 deafness. Alternatively, we have proposed that Cx26 deafness may have been specifically amplified during the past 200-300 years by relaxed selection and assortative mating. In the past, the genetic fitness of the deaf may have approached zero. However, the social, economic and educational circumstances of the deaf have improved dramatically, and have been accompanied by an increase in fertility, and the onset of assortative mating in some but not all countries. Conventional wisdom holds that curing a rare recessive disease will not affect the phenotype frequency appreciably for many generations. However, this conclusion assumes random mating, and with phenotypic or genotypic assortative mating, the approach to a new equilibrium can be much more rapid. Since Cx26 marriages account for a disproportionate share of non-complementary matings and can only produce deaf offspring, intermarriage among the deaf has preferentially increased this phenotype. With data from the 19th century, we have shown that the frequency of Cx26 deafness has doubled during the past 100 years. Furthermore, the current prevalence is lower in countries without a long tradition of intermarriage among the deaf. If our hypothesis is correct, the availability of premarital genetic testing to either avoid or ensure the birth of deaf children could have even greater effects on the phenotypic frequency, but only if genotypic mate selection became a universal and uniform practice among the deaf.

P0936. Spectrum of Connexin 26 gene (GJB2) mutations in Turkish families with inherited non-syndromic deafness and determination of the GJB2 35delG mutation carrier frequency in Turkish population

O. Uyguner¹, M. Emiroglu², G. Hafiz², M. Yuksel-Apak¹, A. Ghanbari¹, N. Basaran², B. Wollnik¹

¹Division of Medical Genetics, Child Health Institute, Istanbul University; Istanbul, Turkey; ²Ear, Nose, and Throat Department, Istanbul Medical Faculty, Istanbul University; Istanbul, Turkey

muyguner@superonline.com

Congenital deafness occurs in about 1 in 1000 infants and approximately 80% of the cases are non-syndromic and inherited in an autosomal recessive pattern. Mutations in the gene encoding connexin 26 (Cx26), GJB2, are the most common cause of recessive non-syndromic hearing loss (NSHL). Cx26 mutations account for 30-60% of recessive NSHL in European and American populations. A specific mutation in the GJB2 gene, known as 35delG, represents a „hot spot“ mutation and is found in approximately 2/3 of the detectable Cx26 mutations. Nothing is known yet about the spectrum of Cx26 mutations in Turkish population. In the present study, we have studied 37 index patients from consanguineous Turkish families with autosomal recessive NSHL. The entire coding region of the Cx26 gene was sequenced and a homozygous mutation was found in 27% of the cases. The 35delG mutation was present in 78% of all Cx26 mutations identified. One new mutation and two polymorphisms were identified. To determine the prevalence of the 35delG mutation in Turkish population we established a recently described screening method using PCR-mediated site-directed mutagenesis, followed by a BsiYI digestion. We tested 359 unrelated individuals and found 3 heterozygous carriers giving a carrier frequency of 0.84% (1 in 120). Our data show that both, frequency of Cx26 mutations in recessive NSHL and the carrier rate of Cx26 35delG mutation in normal Turkish population is much lower than described for other Mediterranean countries. This finding reflects the mixed origin of Turkish population.

P0937. Determination of the carrier frequencies of two common Cx26 mutations (35delG and 167delT) in the Hungarian population

A. Bors¹, H. Andrikovics¹, L. Kalmar¹, Z. Liska¹, A. Borbely¹, S. Galambos², A. Losonczi², S. Furedi³, I. Balogh⁴, C. Szalai⁵, A. Tordai¹

¹National Institute of Haematology and Immunology; Budapest, Hungary; ²Jewish Community Hospital; Budapest, Hungary; ³Institute for Forensic Sciences; Budapest, Hungary; ⁴University of Debrecen; Debrecen, Hungary; ⁵Heim Pal Paediatric Hospital; Budapest, Hungary

bors@biomembrane.hu

The most common form of non-syndromic neurosensory autosomal recessive deafness (NSRD) is caused by mutations in the gene GJB2 located on chromosome 13, encoding the gap-junction protein connexin 26. One mutation, which represents a deletion of a guanine within a strech of six Gs between nucleotide positions 30 and 35 of the GJB2 cDNA (35delG) is found in 40-70% of NSRD cases, and was shown to have a high allele frequency (2-5%) in several randomly selected populations. Another deletion, the loss of a timine in position 167 was found in the Ashkenazi Jewish population in 4% frequency. A simple polymerase chain reaction (PCR)-based test was used to detect both mutations in order to establish the allele frequencies in the Hungarian control, Gypsy, and Ashkenazi populations. We screened 173 randomly selected Hungarian first time blood donors and 351 unrelated, unaffected Gypsy individuals for the Cx26 35delG mutation. We found 2 heterozygous in the control population, which gives a carrier frequency of 1,17% (0%-2.81%) which is lower, but does not differ significantly from the published data for Caucasian populations. The heterozygous carrier rate in the Gypsy population was 1.4% (0.1%-2.7%) (5 heterzygous from 351). We identified 2 heterozygous individuals for mutation 167delT among 48 Ashkenazi samples analysed so far. Our data support the assumption that the G track containing the deletion may represent a region with higher mutation rate, since we detected similar carrier frequencies in two populations with distant origin.

P0938. A carrier frequency of the 35delG deafness mutation in several populations of Russia

A. Anichkina¹, T. Kulenich¹, I. Shagina¹, T. Viktorova², A. Polyakov¹, E. Khusnutdinova², E. Ginter¹, O. Evgrafov¹

¹Research Center for Medical Genetics; Moscow, Russian Federation; ²Ufa Research Center; Ufa, Russian Federation

dnalab@online.ru

The connexin-26 (Cx26) gene has been shown to be a major cause of nonsyndromic recessive deafness. Although many mutations have been found, a single mutation is responsible for the majority of alleles in different populations (35delG). A high frequency of this mutation was revealed in populations of Northern, Central and Southern Europe with higher prevalence in Southern Europe, and the highest percentage in Estonia. We decided to expand this study further to the East, and investigate several populations of Russia. First, we consider this investigation to be important for defining the strategy of diagnosis and genetic counseling of congenital deafness in Russia. Second, this study could clarify the origin and the history of 35del mutation. Taking into account high frequency of this mutation in Estonia, we paid a special attention to related Finno-Ugric populations, namely Mari and Komi. Three other populations, Bashkirs, Chuvashs and Yakuts, belong to Turkish speaking populations. In ethnogenesis of Bashkirs and Chuvashs participated not only Turkish but also some other populations including Finno-Ugric populations. Altogether 560 persons from 5 ethnic groups of Russia were analyzed for 35delG mutation. Twelve mutations were found, resulting in average carrier frequency 1/46,7. It corresponds to the frequency of more than 1% chromosomes with mutation.

P0939. Hereditary Deafness - Genetic And Epidemiological Aspects -

S. Bembea¹, M. Bembea², O. Vancsik², C. Skrypnyk², C. Jurca², S. Ciuchina²

¹Clinical Adult Hospital, ORL Department; Oradea, Romania; ²Clinical Children Hospital; Oradea, Romania

diabem@yahoo.com

Introduction; The relative incidence of different types of neuro-sensorial deafness has clearly changed for the last 25 years. Maternal rubella, post-natal ototoxicity and meningitis are much better controlled today. Meanwhile, there's only little achievement in what prevention or reduction of hereditary deafness (HD) incidence is concerned. Therefore, the tendency in the present and probably for the coming years, is an increase in the HD relative incidence, with a lowering of acquired deafness of any kind. Objective; Determining the incidence and the genetic characteristics of the HD in Bihor county (with a population of 630000). Material and method; A number of 59 cases have been studied. The pedigree, following the segregation and establishing the type of heredity have been studied for each case. Results; The incidence of HD is approx. 1/10000. The model of inheritance is autosomal recessive in 55% of cases, autosomal dominant in 40% of cases and undetermined in 5% of the cases. Most of the times, HD is neuro-sensorial. Syndromic and non-syndromic HD are described, with examples from the personal experience of rare cases of deafness, like Deafness + Sinphalangism S., Waardenburg S., Deafness + Oculo-cutaneous albinism S., Alport S., Treacher-Collins S. Conclusions; Any case of deafness should be suspected as having a genetic cause. Therefore, the family history, a meticulous clinical examination, audiogram and a try to situate the case in one of the genetic mechanisms are all recommended.

P0940. Connexin26 mutation 35delG in non-syndromic hearing loss in Northern Finland

T. Löppönen¹, M. Väisänen¹, M. Allinen¹, P. Lindholm², M. Luotonen², H. Löppönen², E. Mäki-Torkko², M. Sorri², M. Väyrynen², J. Uusimaa³, J. Leisti¹

¹Department of Clinical Genetics, Oulu University Hospital; Oulu, Finland; ²Department of Otorhinolaryngology, Oulu University Hospital; Oulu, Finland; ³Department of Pediatrics, Oulu University Hospital; Oulu, Finland

tuija.lopponen@oulu.fi

OBJECTIVE; The aims of the study were to evaluate the role of the most common mutation reported, 35delG, of the connexin26 gene (Cx26/GJB2) in children with non-syndromic sensorineural hearing loss and to evaluate the carrier frequency of this major mutation in Northern Finland. STUDY DESIGN; The study population comprised children, who were referred to the Oulu University Hospital and who had non-syndromal congenital sensorineural hearing loss. A total of 70 patients were analyzed. The estimation of carrier frequency was carried out in a random sample of 313 control persons. METHODS; Detection of the 35delG mutation was performed by PCR amplification of the Cx26 sequence followed by direct sequencing of purified PCR products. Simultaneous sequencing to both directions was carried out by automated sequencer (Li-Cor 4200) using forward and reverse primers labeled with IRD800 and IRD700, respectively. Conformation sensitive gel electrophoresis (CSGE) of PCR products was applied to detect the heterozygous carriers in control population and positive samples in this screening were confirmed by direct sequencing. RESULTS; Of the 70 patients with the hearing loss 11 were found to be homozygous for the 35delG mutation. These belonged to 7 families, three from one, two from two and a single patient from four different families. Out of these 11 patients three had severe and 8 profound hearing loss. In the control population four individuals heterozygous for the 35delG mutation were found, which donotes a carrier frequency of 1/78 in Northern Finland. The screening of the Cx26 coding sequence for other mutations is under study. CONCLUSIONS; The present results indicate that the Cx26 mutation 35delG is an important contributor to recessive inherited non-syndromic hearing loss also in Finland. The carrier frequency of this mutation was found to be of the same magnitude as reported previously in Central and Northern Europe.

P0941. Statistical analysis of non-syndromic genetic deafness in Iranian population

N. Kouchekian¹, N. Daneshmandan¹, H. Sheikh-Al-Eslamy¹, H. Najmabadi^1,2

¹University of Welfare and Rehabilitation; Tehran, Islamic Republic of Iran; ²Karimi-Nejad Pathology and Genetics Center; Tehran, Islamic Republic of Iran

maryamneishabury@yahoo.co.uk

Genetic deafness affects 50% of patients. A great portion of this kind of disorder is Non-syndromic which respect to the mode of inheritance is categorized to autosomal recessive (AR),autosomal dominant (AD) ,X-linked and mitochondrial. In order to illuminate these ratios in Iranian population, we started a retrospective study. A total of 180 patients were selected based on genetic background. The genetic counseling and pedigree analysis were performed for each individual. Our results demonstrate 155 families (86.11%) AR, 21 families (11.67%) AD and for 4 families (2.2%) we could not determine the mode of inheritance exactly. The statistical analysis of data showed that distribution of mendelian inheritance of non - Syndromic genetic deafness in Iranian population is similar to other studies. In our study the results showed that 38% of AR had moderate to severe and 61.9% had profound hearing impairment. More over among AD cases 61.9% had moderate to severe while 38% had profound hearing loss. The results also revealed that 57% of AD and only 17% of AR cases had progressive hearing loss. The investigation of consanguine marriage of the parents was revealed as 79.4% of AR and 7.4% of AD cases. From all the patients 68.6% had consanguineous parents who 99.2% of them were AR and 0.8% were AD. In the future most of these families can be used for linkage analysis.

P0942. The prevalence of Usher syndrome in Sweden; A nation - wide epidemiological and clinical study.

C. G. Möller¹, M. Sadeghi², W. Kimberling³, L. Tranebjœrg⁴

¹dep. audiology Gothenburg Univ sweden; 41345 Gothenburg, Sweden; ²Dept. of Audiology, Sahlgrenska University Hosptial; Gothenburg, Sweden; ³Dept. of genetic; Omaha, NE United States; ⁴Dept. Of Medical Genetics; Tromsoe, Norway

claes.moller@audiology.gu.se

BACKGROUND. Usher syndrome (USH) is characterized by hearing loss, vestibular dysfunction and Retinitis Pigmentosa (RP). Three clinical types of USH are recognized. USH I; profound congenital sensorineural deafness, vestibular areflexia and RP. USH II; congenital moderate to severe hearing loss, normal vestibular function and RP. USH III; progressive hearing loss, progressive vestibular deficiency and RP. MATERIALS AND METHODS. The study was an epidemiological, population-based survey, based on examination of data from medical researches, schools and organizations. A large number of the patients underwent clinical and genetic assessment by the authors. RESULTS. A total of 366 deaf-blind subjects were assessed. 21% (77 subjects) previously diagnosed as having USH were excluded because of incorrect diagnosis. In total, 140 patients with USH I, 122 patients with USH II and 27 patients with USH III were diagnosed. The prevalence of USH I was1.6/100 000, USH II 1.4/100 000 and USH III 0.3/100 000. The prevalence rate varied in different counties. USH I was most common in the three northernmost counties, (7.4 /100 000). CONCLUSION. The total prevalence rate of USH is lower then previously reported. A large variation was found in USH I . While the three major types of USH are clearly different, the subtypes within type I, II are so far believed to be clinically indistinguishable.Thus, there is a critical need to further study phenotype-genotype correlation in USH. This study of USH is the most complete ascertainment that so far has been undertaken.

P0943. Haplotypic determinants of instability in the FRAX region; concatenated mutation or founder effect?

S. M. Ennis¹, A. Murray², N. E. Morton¹

¹Wessex Human Genetics Institute, Southampton General Hospital; Southampton, United Kingdom; ²Wessex Regional Genetics Laboratory, Salisbury District Hospital; Salisbury, United Kingdom

se@soton.ac.uk

The fragile X triplet repeat expansion at Xq27.3 has been shown to be associated with mutation or instability 600 kb distal at the FMR2 repeat locus. Concatenated mutation, whereby a mutation at one locus somehow interacts with mutation, recombination, deletion or transposition at another locus is a possible explanation. In this study we examine evidence from a sample of over 7000 independent haplotypes from the FRAX region. We adopt the use of cladistic groups to more thoroughly define the properties of these haplotypes, and in doing so isolate one group of haplotypes which may be predisposed to the phenomenon of concatenated mutation. Distinguishing concatenated mutation from founder effects is difficult within a single population. We present our evidence for and against concatenated mutation, and in the process describe a previously undefined mutation at FRAXE.

P0944. Different distribution of DXS548 and FRAXAC1 haplotypes between normal and fragile X population in Croatia

R. Bago¹, S. Hecimovic¹, I. Barisic², V. Culic³, K. Pavelic¹

¹Rudjer Boskovic Institute; Zagreb, Croatia; ²Children's Hospital Zagreb; Zagreb, Croatia; ³Clinical Hospital Firule; Split, Croatia

bago_r@yahoo.com

The fragile X syndrome is caused by expansion of the (CGG)n repeat in 5' end of the FMR1 gene. In order to look for linkage disequilibrium between the fragile X locus and its flanking markers, we analyzed the DXS548 and FRAXAC1 microsatellite markers in normal and unrelated fragile X males of Croatian origin. Different distribution of alleles and haplotypes was found between these two samples. A significant increase in frequency of DXS548 allele 2 was found among fragile X patients when compared to normal individuals (31,3% vs. 2.86%). We also noticed a different distribution of FRAXAC1 allele A (18.8% in fragile X group vs. 10.0% in normal population). Haplotype 7-C was the most represented in normal population (57.14%), while haplotypes 2-C, 8-C and 2-A were more frequent in fragile X group (accounted for 43.75% of all fragile X chromosomes and less than 4% of normal population). This difference may suggest the existence of linkage disequilibrium between the two loci and/or selective advantage of this haplotypes among fragile X affected individuals in Croatia.

P0945. Epidemiological Study Of Some Sentinel Anomalies and Down's Syndrome

E. Nikula, O. Tymchenko, O. Turos, T. Pokanevich, L. Piotrovich, O. Makaruk

Institute of hygiene and medical ecology; Kyiv, Ukraine

elena@ushc.kiev.ua, otimch@usch.kiev.ua

It is known that the sentinel anomalies (SA) and Down's syndrome (DS) make a considerable contribution into infant morbidity and mortality. These defects are diagnosed well and can provide important clues in the detection of teratogenic agents. An anencephalia, spina bifida, limb reduction, oesophageal atresia and traheo-oesophageal fistula, ano-rectal atresia, cleft of a labium with / or without cleft of a palate, multiple congenitals anomalies and Down syndrome were taken into account. Objective. To estimate the relative risk (RR) of the live-borns with SA and DS in different Ukrainian regions. Methods. Data on live-borns only were received from the annual reports of seven medical genetic centres of Ukraine (Southern, Western, North, Northern-Eastern, Central, South-Eastern, South-Western). Period of the observation - from 1993 to 1999. The precise data about still-borns with birth defects are unknown in Ukraine. RR was calculated according to Epi Info program for each selected defects. Results. It was established that RR evenly disseminated among the new-borns of the country. RR was lower in the Northern and the South-Eastern regions (0,86 at the 0.81-0.90 confidence interval; 0.88; CI 0.84-0.93). RR was the highest in the Southern (1.10; CI 1.02-1.19) and the Northern-Eastern regions (1.19; CI 1.14-1.25). Conclusion. The distinctions in the rates of observed defects can reflect not only the difference of incidence in population, but mainly the difference in the levels and the quality of medical care and other socio-economic factors.

P0946. Epidemiology and Relative Incidence of Rare Metabolic and Genetic Disorders in IRAN and the results of 50 Prenatal Testings

Y. Shafeghati^1,2, R. Karimi-Nejad¹, F. Azimi¹, M. Karimi-Nejad¹, J. Huijmann³, W. J. Kleijer³, O. VanDiglen³

¹Karimi-Nejad Pathology and Genetics Center; Tehran, Islamic Republic of Iran; ²University of Welfare and Rehabilitation; Tehran, Islamic Republic of Iran; ³Genetics and Metabolic Department of Erasmus University; Rotterdam, The Netherlands

maryamneishabury@yahoo.co.uk

Since 10 years ago, our Genetics Center has been evaluating 128 families including 196 cases with the collaboration of the Genetic and Metabolic Department of Erasmus University based in Rotterdam, the Netherlands. The result has been the diagnosis and discovery of these rare but very important disorders in the families under scrutiny. Out of the above number of families mentioned, the final diagnosis in 41 was MPS, the most common type being MPS III. Also, 48 families with 77 patients were suffering from Lipid Storage Diseases. The most common disorder was Metachromatic Leukodystrophy followed by Niemann-Pick, Gauchers and Tay-Sachs diseases respectively. 39 families with 67 cases were diagnosed to have micromolecular Metabolic Diseases. Organic Acidemias was the most common type. Totally, in 50 families prenatal testing was carried out showed that 11 (22%) of the fetuses were affected. Details of different clinical subtypes would be presented in this article. In dealing with this patients especially while planning for prenatal testings, genetic counselling was carried out to provide the families with detailed information about their suffering children. In doing so, Ethical principles also fully recommended in Islam for different contexts such as "Autonomy, Beneficence, Non-maleficence, Justice and confidentiality were carefully observed.

P0947. Screening for fragile X syndrome among mentally retarded children in Greece

C. Sofocleous¹, H. Fryssira¹, L. Thomaidis², A. Kolialexi¹, A. Mavrou¹, C. Metaxotou¹

¹Medical Genetics, Athens University, School of Medicine; Athens, Greece; ²1st Dept. of Pediatrics, Athens University; Athens, Greece

csofokl@cc.uoa.gr

Fragile X syndrome (FXS), second most common cause of mental retardation (MR) affects 1; 4000 males and 1;7000 females. The CGG repeat of FMR-1 gene varies from 6 to 54 triplets in normal individuals, whereas in FXS families prermutations (PM) of 55 to 200 repeats and full mutations (FM) of more than 230 repeats are observed. Hypermethylation of CpG island, upstream the CGG repeat, causes the transcriptional shutdown of FMR-1 gene resulting in the absence of FMRP protein in FXS patients. 1687 individuals, 590 MR boys and 88 MR girls, their parents and other family members (759) were tested for FXS and 250 normal individuals were used as controls. DNA extraction from peripheral blood lymphocytes followed by modified non radioactive Southern blot and PCR (CGG sizing) analysis were performed. Immunocytochemical detection of FMRP on blood smears was also performed for 300 individuals. The results confirmed the clinical diagnosis of FXS in 28 males and 2 females with FM and 1 male with PM. In the remaining individuals an allele distribution of 7 to 58 CGG repeats with a mean of 28-30 triplets was observed. The FMRP antibody test confirmed the diagnosis in 15 patients and revealed 3 more FXS families with no CGG expansion. The combination of non radioactive Southern and PCR analysis, enhanced by the FMRP test was highly efficient for diagnosis and thus supportive of large scale population screening especially since a rough estimation, revealed an FXS frequency at about 1;4250 boys in Greece.

P0948. Social selection against X-linked intelligence?

N. V. Kovaleva

Institute of Obstetrics and Gynecology; St. Petersburg, Russian Federation

kovaleva@robotek.ru

Social selection against X-linked intelligence? Natalia V. Kovaleva Institute of Obstetrics and Gynaecology RAMS, Mendeleev line, 3, 199034, St.Petersburg, Russia, e-mail; kovaleva@robotek.ru During the past decade, several different genes involved in learning capacity were identified on human X-chromosome. They were discovered because of mutations causing nonspecific mental retardation in males. On the other hand, these genes are reasonably considered to be the genes for higher intelligence and even have been discussed as genes "by which Homo may have become sapiens" [Turner, 1996]. As the pattern of X-chromosome inheritance is very specific (men transfer X-chromosome to their daughters only), the process of transmission and spreading of such genes in population may be influenced by social factors. In times when there was no effective birth control, and even women with high intellectual potency had no or little chance of self-realization, outstanding and successful men and their daughters had many offspring, which ensured maintenance of X-linked genes for intelligence in population. At present, intelligent women who had inherited excellent genetic material from their father tend to have few children because childbearing hampers their social development. This leads to a decrease in frequencies of the "clever" alleles in population. Thus, social selection against genes for higher intelligence has been putting into effect through successful women. The genetic drift of the "clever" alleles away from initial values is expected to be stronger in societies with traditional attitude to gender roles in household.

P0949. Trinucleotide repeat polymorphism of some X-linked loci in a normal West Siberian Slavic population

E. N. Tolmacheva, V. N. Evdokimova, S. A. Nazarenko

Institute of Medical Genetics; Tomsk, Russian Federation

kate@img.tsu.ru

AR, FMR1 and FMR2 genes have been mapped to the X-chromosome. The expansion of CAG (AR), CGG (FRAXA/FMR1) and GCC-repeats (FRAXE/FMR2) is the mechanism associated with Kennedy disease and fragile X-syndromes correspondingly. The distribution of these loci is varied and characterized by the positive asymmetry among the world populations. This means they tend to accumulate the repeats over time. A hypothesis does exist that the transition from normal-sized to expanded alleles is a result of a mutational bias towards larger repeat sizes of the repeat arrays in the normal population. The distributions for AR, FRAXA and FRAXE loci were analyzed in a West Siberian Slavic population. Automated genotyping of these loci was perfomed by PCR amplification with fluorescently labeled primers on the ABI PRISM 310 Genetic Analyzer and Genescan softwere. We detected among 113 unrelated healthy individuals; 12 alleles for the AR locus, ranging in size from 21 to 32 repeat units; 31 alleles for the FRAXA locus, ranging in size from 8 to 56 repeats; and 18 alleles for the FRAXE locus, ranging in size from 9 to 27 repeats. The allele's frequency distribution for AR and FRAXE loci was unimodal with the size of modal allele in 27 repeats (18%) and 17 repeats (19.9%) correspondingly. The FRAXA locus showed a bimodal distribution with the modal alleles in 18-20 repeats (11.5%) and 28-29 repeats (24.6%). Allele's distribution for AR locus in Slavic population had a narrower spectrum and a higher size of modal allele compared with another studied populations. The FRAXA locus showed very high frequency of "gray zone" alleles. Alleles containing more than 40 repeats share 12.4% and more than 45 - 8.0% of all alleles in population. In contrast, the allele's distribution for FRAXE locus was characterized by the predominance of the short-size alleles (<15 repeats). Intriguingly, each investigated locus has the special features in allele's distribution compared with another world populations.

P0950. Etiology of moderate to profound mental retardation ; a retrospective study of 1242 patients

A. Destree¹, A. Vokaer², Y. Gillerot¹, L. Van Maldergem¹

¹Centre de Genetique Humaine, Institut de Pathologie et de Genetique; Loverval, Belgium; ²CHU Brugmann, Centre de Périnatalogie; Brussels, Belgium

secret.genet@ipg.be

Objective ; To carry out a retrospective epidemiological survey of the etiology of moderate to profound (IQ £ 50) mental retardation in a population of 1242 patients and to compare this data with those from previous studies. Method ; We analysed the files of 729 moderately to profoundly retarded patients evaluated in our centre for Human Genetics. Most of these patients underwent karyotyping, brain imaging and a search for inborn errors of metabolism. The files of 513 additional patients from institutions for the mentally retarded were also studied. In these cases, the work-up was often minimal. The study of this heterogeneous population gave the following results ; Constitutional disorders ; Down Syndrome ; 193 cases, partial chromosomal duplication and/or monosomy ; 32 cases, apparently balanced reciprocal translocations ; 5 cases, sex chromosomes aneuploidy ; 6 cases, mosaic triploidy ; 1case, partial tetrasomy ; 2 cases, microdeletion syndromes ; 22 cases, autosomal dominant disorders ; 31 cases, autosomal recessive disorders ; 79 cases, X-linked MR syndromes ; 39 cases, Known MCA/MR syndromes ; 18 cases, undiagnosed MCA/MR ; 281 cases, CNS-malformations ; 5 cases. An acquired etiology was likely in 171 cases. No etiology was found for 349 patients. Conclusion ; The study of a heterogeneous population of 1242 moderately to profoundly retarded patients indicates a proportion of 28.1% for which no etiological clue is found. This population includes 513 patients with a minimal work-up.

P0951. Consanguinity and Mental Retardation in North India

M. Afzal¹, M. S. Ahmad², B. Shadab³, G. G. H. A. Shadab¹

¹Section of Genetics, Department of Zoology, AMU; Aligarh, India; ²Section of Genetics, Department of Zoology, AMU; India; ³Department of Human Genetics, Guru Nanak Dev University; Amritsar, India

hep_amu@usa.net

We have studied mental retardation from different States of North India viz. Bihar, Uttar Pradesh, Delhi and Punjab. Consanguinity is prevalent widely among muslim groups, the first cousin unions being most common. From our field survey (Ansari and Qureshi groups) and hospital samples (Jawahar Lal Nehru Medical College and Hospital, Aligarh & All India Institute of Medical Sciences, New Delhi), we tried to find out the genetic load due to mental retardation classified as chromosomal disorders, fragile sites, single gene disorders and low IQ in polygenic variation. Among chromosomal disorders, trisomy 21 was most conspicuous, again free trisomy 21 being the most prevalent one, followed by G/G and D/G translocation. Turner’s, Klinefelter’s and intersex cases showed lower IQ levels. Fragile site cases comprised Xq.27 cases being most common. Amino acidopathies, and enzymatic disorders were more common among consanguineous offsprings than non-consanguineous ones. Surprisingly age of the mother show highest number of defectives at around 40 years, but the downship invariably was toward the last/later parity. Verbal, performance and total IQ, scores on WISC R-74 (Hindi/Urdu rendition by the author) shows lowering of IQ, the effect being statistically significant. Loss of mental age due to consanguinity, percentage inbreeding depression and genetic load have been calculated. We are looking ahead for molecular markers, and pedigree analysis for non-specific aetiology of mental retardation, and work out a model for genetic epidemiology for North India.

P0952. Autosomal allele frequencies and the search of ancestors in Hungarian populations

C. R. Guglielmino¹, A. De Silvestri², J. Beres³

¹Dipartimento di Genetica e Microbiologia, Universitą di Pavia, & Istituto di Genetica Biochimica ed Evoluzionistica- CNR; Pavia, Italy; ²Dipartimento di Genetica e Microbiologia, Universitą di Pavia; Pavia, Italy; ³National Center for Epidemiology, Dep. of Human Genetics and Teratology; Budapest, Hungary

guglielmino@ipvgbe.igbe.pv.cnr.it

Autosomal gene frequencies allow to describe the genetic structure of populations, in particular to reveal the effects of drift or of admixture. Applying specific models to population samples, it is also possible to infer a rate of admixture of a population with respect to hypothetical parent populations. Some Hungarian ethnic groups claim to be descendant either of Turks , Iranian or of the ancient Huns. Hungarian nation was founded in the 9th century from people caming from eastern regions close to the Ural mountains and since then people speak an Uralic language belonging to the Finno-Ugric language group. We compared the gene frequencies of eight ethnic groups and seven hypothetical ancestral populations, including Uralics, applying a model of admixture. The results, most of which confirm historical hypotheses or the oral tradition, show that only one ethnic group highly resembles the Uralic population.

P0953. Occurance of ABO and Rh blood groups in three samples of Serbian population

S. Cvjeticanin^1,2, B. Trišic³, R. Papovic³

¹Medical Faculty University of Belgrade; Belgrade, Yugoslavia; ²Institute of Biology and Human Genetics; Belgrade, Yugoslavia; ³Institute for Biology and Human Genetics; Belgrade, Yugoslavia

bebrapap@eunet.yu

In the group of patients with congenital hip dislocation (N=93) the frequencies of ABO blood types were similare to the average value of Serbian population, while the percentage of blood group A is slightly increased.Comparing frequencies of Rh blood groups we can see that there is no differences between tested samples. In the sample of children affected with acute lymphoblastic leukemia (N=163)the frequencies of ABO blood types was similar to the average value of Serbian population, while the percentage of blood group O is slightely increased.The percentages of Rh blood types are not showing any differences between tested samples. It is interesting that in the group of top sportsment, including 61 winners of international champions, the frequencies of A and B blood groups were decreased,AB was absent and O group was about twice more frequent (71%)than an average value of Serbian population.The percentage of Rh negative individuals is significantly lees (10%)in the group of top sportsmen comparing with percentage of those individuals in Serbian population(17%). Taking all this into account we may see that variation in physical abilities can bee correlated with their immunogenetics properties.

P0954. Haptoglobin (Hp) and Transferrin (Tf) polymorphism - a population genetic study of Romanian population

G. G. Meianu¹, V. Stoian², A. Rodewald³, C. Glavce⁴

¹Biology Faculty Bucharest Universyti; Bucharest, Romania; ²Laboratory of Genetics - Faculty of Biology, Bucharest University; Bucharest, Romania; ³Institute of Human Biology, University of Hamburg; Hamburg, Germany; ⁴Center of Anthropological Research - Academy of Science; Bucharest, Romania

pisica_ii@hotmail.com

The aim of this study is to present the distribution of the Hp subtypes (by means of the polyacrylamide gel electrophoresis - method by Linke, 1984) and of the Tf subtypes (by polyacrylamide gel isoelectric focusing at pH 4-6, method described by Weidinger et al., 1984) in the Romanian population. EDTA - plasma samples were obtained from 120 unrelated donors from a premountain Carpatian region of Romania (Prahova - Valley), and only this individuals were considered as autochthonous. X2 - tests and genetic distance analysis were used to compare the data with those of other studies of Mediterranean and European populations. The number of the phenotypes observed in both systems showed no significant differances to the data expected according to the Hardy-Weinberg equilibrium. The comparison with the other European and Mediterranean populations shows that the genes and the phenotype frequencies of the tested samples lines nearer to those of Southeuropean -Mediterranean population than to Balkan or Slavic groups. Our results are interpretable in the background of historical and ethnical events, which have played an important role in changing the genepool of the Romanian population. Linke R.P.; Analytical Biochemistry, 141, 55-61 (1984) Weidinger et al.; Human Genet., 66, 356-360 (1984)

P0955. Mesolithic-Neolithic population relationships in Portugal; the evide