ESHG Concurrent Symposia

A. Swillen, J. P. Fryns;
Center for Human Genetics, University Hospital Gasthuisberg, Leuven, BELGIUM.

A major challenge in both clinical practice and research in the field of mental retardation and of learning disorders is to identify the underlying causes : the genetic, chromosomal and environmental factors that have an important influence on a person’s development and behaviour.
Advances in clinical genetics have lead to an increased recognition of specific syndromes. In recent years, cytogenetic and molecular genetic tools have resulted in the identification of the underlying genetic defects in a large number of these disorders.
For many years, interest has focused on the delineation of the somatic aspects of the phenotypes and their underlying pathogenetic mechanisms. However, in the last decade, researchers in this field paid more attention to the cognitive and behavioural features of various genetic conditions, the so-called “behavioural phenotypes” (B.P.).
A behavioural phenotype is “a behavioural pattern, including cognitive processes and social interaction style, that is consistently associated with, and specific to, a syndrome which has a chromosomal or a genetic etiology”. This approach has proven to be of practical importance for the patients both regarding an earlier syndrome diagnosis, as well as in the multidisciplinary management and follow-up. In addition, the link between a specific behavioural phenotype and a genetic defect represents an unique opportunity to gain insight in the complex neurobiological processes underlying human behaviour.
In this lecture, focus lies on the clinical goals of research in B.P. , and this will be illustrated using the Velo-Cardio-Facial syndrome (VCFS), the Smith-Magenis syndrome and the Prader-Willi syndrome (PWS).

The tortuous path from genotype to phenotype: genes and cognition in mutant mice

H. P. Lipp;
Institute of Anatomy, University of Zurich, Zurich, SWITZERLAND.

Genetically modified mice are increasingly used to analyze the relations between genes, brain and behavior. One approach concentrates on deleting or controlling genes of theoretical importance for memory and learning, e.g., genes encoding proteins for intracellular signalling pathways. The other focuses on deleting genes in the mouse homologous to human mutations causing mental retardation and cognitive impairment. Both approaches face (i) problems of elucidating the causal relations between genes, brain and behavior, and (ii) the difficulties of matching cognitive processes in mice and men.
(i) Phenotypic expression of mutation effects is often masked or altered by homeostatic regulation at the intracellular level, and by brain plasticity and environmental effects. This will be illustrated by examples taken from the analysis of genetically modified mice carrying different targeted mutations.
(ii) A cross-species comparison of cognition must take into account that the associative cortex in mice is small and largely confined to the hippocampal formation and proximally connected structures. On the other hand, hallmarks of cognition in men (e.g., general problem solving abilities and memory), and of other mammals (e.g., spatial memory and learning of rats) might be of lesser importance for mice. This will be illustrated by a comparison of cognitive impairments in the laboratory and under naturalistic conditions.
In conclusion, the availability of mutant mouse model represents a fundamental methodological progress in modeling the genotype-to-phenotype pathway characteristic for human mental retardation, but we must learn to understand the constraints imposed by species differences. Supp by SNF and NCCR "Neural Plasticity and Repair".

B. Carmichael;
Genetic Nurse Specialist, Southend Hospital, Westcliff on Sea, UNITED KINGDOM.

Fragile X Syndrome is the commonest inherited cause of learning difficulty. Although X-linked, it affects both girls and boys. While learning difficulties are a consistent feature, fragile X is also associated with characteristic behavioural problems in many affected children.
Having a child with a learning difficulty has an impact on any family, as has the presence of an inherited disorder in a family. Many of the challenges faced by Fragile X families are no different from those faced by other families with genetic disorders, or with learning-disabled children. However, there are features of fragile X which make it unique.
Some mothers of fragile X children are themselves affected, making parenting and behavioural management difficult. Women with fragile X do not always have good social skills, and this can lead to estrangement from their families and a lack of peer support. Social anxiety and lack of confidence can make it hard for some affected females seek help for themselves or for their children.
Finding the underlying genetic mechanism which causes the condition has made possible reliable diagnosis and carrier detection. However, more than 30 years after the first observation of the fragile site on the X chromosome, prenatal diagnosis can still be problematic.
These issues will be discussed from the perspective of my work as a genetic counsellor, as an active member of the UK Fragile X Society, and from my experience of living in a Fragile X family.

M. Stratton;
The Sanger Centre, Wellcome Trust Genome Campus, Cambridge, UNITED KINGDOM.

A major application for the human genome sequence in elucidating oncogenesis will be as a template subserving genome-wide searches for mutations in cancer cell genomes. A full description of changes at the DNA level in cancer cells will require information on all types of abnormality; copy number changes, rearrangements, point mutations and methylation. Currently there is no single technology that practically can address all these simultaneously. We have embarked upon genome-wide searches for homozygous deletions and for small intragenic mutations (base substitutions and small insertions / deletions) in cancer cell lines. These searches are beginning to yield fruits in terms of newly identified somatically mutated cancer genes. They are also beginning to reveal insights into global patterns of mutation that differ between individual cancers and cancer types.

T. Visakorpi;
Institute of Medical Technology, University of Tampere, Tampere, FINLAND.

The molecular mechanisms underlying the development and progression of prostate cancer are inadequately understood. Over the past 10 years, genetic alterations in prostate cancer have been identified using techniques, such as linkage analysis, loss of heterozygosity analysis (LOH), fluorescence in situ hybridization (FISH) and comparative genomic hybridization (CGH). These analyses have implicated several chromosomal regions in the development of prostate cancer. Linkage analyses and subsequent positional cloning have now revealed two prostate cancer susceptibility genes, HPC1/RNASEL, and HPC2/ELAC2. However, they seem to explain only a small fraction of the hereditary prostate carcinomas. The most common epigenetic event in the sporadic prostate tumors seems to be hypermethylation of GSTP1 gene found already in premalignant prostate lesions. The chromosomal arms that most frequently contain losses in prostate tumors are 6q, 8p, 10q, 13q, 16q and 18q. Except the PTEN at 10q23, the target genes for these deletions are not known. Although gains and amplifications of chromosomal regions are rare in early prostate cancer, they are found in late stage, especially at 7p, 7q, 8q, 18q, and Xq. Of these, the best characterized is amplification of Xq, found in 30% of hormone-refractory prostate carcinomas, and which affects androgen receptor (AR) gene. Amplification of AR gene leading to its overexpression, as well as, in lesser extent, mutations in the AR gene, seem to play critical role in emergence of hormone-refractory prostate cancer. The most commonly gained region in prostate cancer is 8q. However, the true target genes for this aberration have remained unclear.

P. Workman;
CRC Centre for Cancer Therapeutics, The Institute for Cancer Research, Sutton, Surrey, UNITED KINGDOM.

Cancer drug discovery has now firmly entered the postgenome era. This is characterised by two features: 1) A focus on new therapeutic targets defined by the molecular pathology of cancer; and 2) Use of modern technologies, particularly genomics, high throughput screening, combinatorial chemistry, structural biology, cassette dosing pharmacology and molecular biomarkers, to increase success and accelerate the pace of drug development (see Workman P, Curr Opin Pharmacol 2 342-352 2001; Workman P, Curr Opin Invest Drugs 2 1180-1135 2001).
The expectation is that such molecularly targeted, genome-based drugs will be more effective and less toxic than cytotoxic agents and will be used chronically in long term disease management. A wide range of novel genome-based agents are in clinical and preclinical development and some have now received regulatory approval (see Workman P and Kaye SB, A Trends Guide to Cancer, Trends in Molecular Medicine, 8 (4) Suppl 2002). Herceptin (trastuzumab) is approved for treatment of ErbB2 positive breast cancer and Glivec (imatinib) for use in chronic myelogenous leukaemia and gastrointestinal stromal tumours (GIST) as a result of its action on the Bcr-Abl and mutant c-Kit tyrosine kinases that drive these diseases. Iressa (ZD1839) shows activity in various cancers through inhibition of the epidermal growth factor receptor tyrosine kinases. Hsp90 inhibitors, such as 17AAG, are of special interest because of their ability to cause depletion of multiple oncogenic proteins. Progress and challenges with the gene to drug approach in cancer will be reviewed, with examples from recent work.

M. Koenig, H. Puccio, P. Bomont, M. C. Moreira, D. Simon, S. Klur, M. Gribaa, C. Lagier-Tourenne, M. Schmitt;
IGBMC (CNRS-INSERM-ULP), Illkirch, Strasbourg, FRANCE.

With the exception of metabolic disorders, only two forms of recessive ataxia were well characterised some twenty five years ago : Friedreich ataxia (FRDA) and ataxia-telangiectasia (AT). DNA linkage studies allowed to identify the molecular cause of these frequent forms of inherited ataxia and to start to unravel the complex heterogeneity of the remaining cases. In a collaborative endeavour, we identified the gene defective in FRDA and its product, frataxin, a mitochondrial protein whose deficiency results in disturbances of iron homeostasis and of iron-sulfur proteins. The molecular pathogenesis of FRDA is defined by severely reduced levels, but not absence, of normal frataxin as a result of the intronic trinucleotide expansion mutation present in homozygous, and occasionally in compound heterozygous, patients. We have used conditional and inducible knock-out strategies to artificially recreate this partial frataxin deficiency in mouse models that present with several clinical, histological and biochemical features of the human disease, in different combinations. Idebenone, an antioxidant proposed by P. Rustin to protect against iron excess toxicity, showed a moderate but significant effect on survival on the first model that we tested. We pioneered the use homozygosity mapping for primary localisation of autosomal recessive conditions and applied it successfully to ataxia with isolated vitamin E deficiency (AVED), giant axonal neuropathy (GAN), ataxias with oculomotor apraxia (AOA1 and AOA2), Refsum disease (RD) and ataxia with visual impairment and deafness (van Bogaert ataxia). We subsequently identified the defective gene for the first three conditions and used the genetic knowledge to clinically delineate the AOA forms. The identification of many more genes involved in recessive ataxias should shed light on general pathological mechanisms and suggest therapeutic interventions.

P. Rustin, V. Geromel, N. Darin, A. Munnich, A. Rötig;
INSERM U-393 - Hopital Necker Enfants Malades, Paris, FRANCE.

In 1996, mutations in the gene encoding frataxin, a mitochondrial protein of yet unknown function, were shown to cause Friedreich's ataxia, the most common hereditary ataxia with cardiomyopathy. In more than 97% of the patients, a GAA triplet expansion in the first intron of the gene results in a lack of function of frataxin, due to hampered transcription of the gene. Decreased frataxin was subsequently found to cause a generalized mitochondrial iron-sulfur protein deficiency in the heart from Friedreich's ataxia patients. Post-mortem sample analysis confirmed the functional impairment of these proteins in both heart and brain of the patients. The exact function of frataxin remains a matter of debate, the protein being possibly involved in either mitochondrial iron export or storage, or in iron-sulfur protein synthesis. Whatever the exact function of the frataxin could be, the disabled recruitment of early antioxidant defenses recently reported in patient cell lines should participate to increased sensitivity to the oxidative insults noticed in Friedreich ataxia patients. Given the strong evidences for an increased oxidative stress, antioxidant therapy appeared a reasonable therapeutic option. Patients were therefore treated with idebenone, a potent antioxidant ubiquinone analogue. A preliminary study reported a spectacular improvement of the cardiomyopathy in three patients after 6 months of idebenone oral supplementation (5 mg/kg/d). These preliminary results were afterwards confirmed on a larger cohort of patients with a significant decrease of heart hypertrophy measurable in about half of the patients after 6 months of treatment. The variable response to the treatment, and its inability to counteract the ataxia so far, should prompt identification and development of new molecules possibly targeting the induction of cells early antioxidant defenses.

A. Dürr;
INSERM U289 et Dép.de Génétique, Cytogénétique et Embryologie, Hôpital de la Salpêtrière, Paris, FRANCE.

Functions and Biogenesis of Peroxisomes and the Metabolic and Molecular Basis of Peroximal Disorders

J. A. Wanders, H. R. Waterham;
Academic Medical Center, University of Amsterdam, Laboratory Genetic Metabolic Diseases, Departments of Pediatrics/Emma Children’s Hospital and Clinical Chemistry, Amsterdam, NETHERLANDS.

Peroxisomal disorders (PDs) are relative newcomers in the area of genetic diseases which now comprise over 20 different disorders with Zellweger syndrome as the prototype. The PDs can be subdivided into 2 groups including 1. the peroxisome biogenesis disorders (PBDs) and 2. the single peroxisomal enzyme deficiencies. In PBD patients peroxisomes are strongly deficient resulting in a generalized loss of peroxisomal functions. Within the PBD group there is not only profound clinical heterogeneity ranging from Zellweger syndrome with early death to infantile Refsum disease with survival into adult life but also marked genetic heterogeneity with the involvement of at least 12 different genes. Most of these genes, called PEX-genes, have been identified allowing molecular diagnosis in virtually all patients. Peroxisomes catalyze a range of important metabolic functions including fatty acid beta-oxidation, etherphospholipid biosynthesis, fatty acid alpha-oxidation and glyoxylate detoxification, a.o. In recent years many peroxisomal disorders have been identified resulting from single peroxisomal enzyme deficiencies in each of these pathways with X-linked adrenoleukodystrophy as the most frequent single peroxisomal defect. This advanced knowledge has led to the development of highly reliable methods for the post- and prenatal laboratory diagnosis of patients at the metabolite, enzyme and DNA-level. Despite these many achievements much remains to be learned about the pathogenesis of peroxisomal disorders and about treatment strategies.

A. Monaco¹, .. IMGSAC², .. SLIC³;
¹Wellcome Trust Centre for Human Genetics, University of Oxford, Oxford, UNITED KINGDOM, ²The International Molecular Genetic Study of Autism Consortium (http://www.well.ox.ac.uk/~maestrin/iat.html), ., UNITED KINGDOM, ³The Specific Language Impairment Consortium (http://www.well.ox.ac.uk/monaco/dianne/index.shtml), ., UNITED KINGDOM.

Autism is characterised by impaired social interaction and communication, and is accompanied by repetitive and stereotyped behaviours and interests. Autism has an onset in the first three years, persists throughout life, and is associated with mental handicap and epilepsy although it can include milder, but related impairments in individuals of normal intelligence. From family and twin studies there is substantial evidence that autism has a strong yet complex genetic component. In order to identify autism susceptibility genes the IMGSAC has collected approximately 250 families with more than one child or relative affected. 83 sibling-pairs families with autism were screened for linkage using a whole genome scan and areas of increased allele sharing were genotyped with 119 markers in a further 69 sibling pair families. Four regions of linkage were identified including chromosome 2q (MLS 3.74), 7q (MLS 3.20), 16p (MLS 2.93) and 17q (MLS 2.34), three of which have been replicated by other groups. Candidate gene and association studies are now in progress in order to isolate the susceptibility genes at these loci. On chromosome 7q31, the FOXP2 gene is mutated in a severe monogenic form of speech and language impairment and encodes a transcription factor containing a polyglutamine tract and a forkhead domain. FOXP2 was tested for association and mutation in both complex language impairments and autism. No association or coding-region variants in FOXP2 were found and the gene is therefore unlikely to play a major role in autism or more common forms of language impairment.

R. Baumeister;
ABI/Biochemistry, Laboratory of Molecular Neurogenetics, Ludwig-Maximilians-University, Munich, GERMANY.

The different genome projects have resulted in an exponential increase in sequence information available in the databases. At the same time, the number of functionally characterized genes is only increasing linearly. How can we increase the speed of functional genomics to make full use of the data mining? Model organisms have helped significantly to understand the roles of particular genes in an organism. However, the time and effort to perform even single targeted gene manipulations in mouse or Drosophila melanogaster is significant, and the complexity of these organisms prevents in many cases the detailed analyses of the KO consequences. The nematode C. elegans offers several advantages: About 60 % of the human disease genes are represented by homologues in C. elegans. In addition, the animals are small enough to be kept in large numbers in a format that allows mass manipulations (microtiter plates) and knock-outs of candidate genes can be obtained in a matter of 4-6 weeks. In addition, C. elegans is the only multicellular organism for which the development of each single cell and the entire connectivity of its nervous system are known. At the same time, the cellular diversity of the C. elegans nervous system is remarkable, including the same neurotransmitters, although the total number of neurons is only 302. We will focus on C. elegans models of genes involved in human neurodegenerative diseases, in particular on methods and technologies that can be automated in order to accelerate functional genomics.

C. Haass;
Ludwig-Maximilians-University, Adolf-Butenandt-Institute, Munich, GERMANY.

P. Lichter;
Deutsches Krebsforschungszentrum, Abt. Organisation komplexer Genome, Heidelberg, GERMANY.

M. R. Speicher^1,2, J. Kraus^1,2, R. Gangnus¹, C. Maierhofer¹, I. Jentsch¹, S. Langer¹, G. Lederer¹, C. Keri¹, C. Fauth^1,2;
¹Institut für Humangenetik, Technische Universität München, Munich, GERMANY, ²Institut für Humangenetik, GSF Forschungszentrum für Umwelt und Gesundheit, Neuherberg, GERMANY.

For our multicolor-FISH applications we are currently using seven different fluorochromes for probe labeling on a routine basis. DAPI is used in addition for DNA-counterstaining. This multitude of different fluorochromes allows a wide range of different multicolor approaches for diagnostic applications and basic research. In 24-color karyotyping (multiplex-FISH/M-FISH) this increase in fluorochromes has several advantages including a significant reduction of probe complexity, facilitation of image analysis, and most importantly improvement of resolution. We estimate that interchromosomal rearrangements can be diagnosed if the translocated or inserted segment has a size in the range of 230 kb to 2.6 Mb. Furthermore, application of the latest software developments has allowed to simultaneously visualize and analyze multiple small region specific probes (BACs, PACs, YACs) in an automated fashion. Thus, sophisticated probe sets can now be tailored for specific purposes, e.g. a high resolution screen of all subtelomeric regions in one hybridization.In addition, multicolor FISH is now amenable to applications in interphase-FISH. Our technology of 3D-deconvolution microscopy employs an epifluorescence microscope equipped with a motorized table to collect a stack of images at defined levels in z-direction. Images are processed by deconvolution to remove out-of-focus information. Subsequently, 3D-reconstruction algorithms are applied. This approach allows the simultaneous analysis of up to 13 different region-specific-probes in lymphocyte-preparations and up to 7 different probes on thick (30 mm) tissue sections. Applications to breast and ovarian tissue samples will be presented which revealed an unprecedented insight into heterogeneity and organization of these tumors on a single cell level.

J. A. L. Armour, E. J. Hollox;
Institute of Genetics, University of Nottingham, Queen's Medical Centre, Nottingham, UNITED KINGDOM.

Measuring the copy number of single-copy DNA segments can be used to screen chromosomal regions for deletion or duplication. Multiplex Amplifiable Probe Hybridization (MAPH) involves a combination of hybridization and PCR to measure copy number at up to 60 loci simultaneously, using standard preparations of genomic DNA. Short segments of DNA (100-500bp) are used as probes, so that high-resolution measurement of copy number can be made at (for example) individual exons of genes such as BRCA1. We have assembled and characterised a set of probes in which each of the 41 unique human subtelomeric regions is represented at least once. Using this probe set, copy number at all chromosome ends in one individual can be screened in a single gel lane. A series of positive controls has been used to demonstrate the sensitivity of the probes, and 83 normal control individuals were used to assess the frequency of polymorphic copy number with no apparent phenotypic effect. At some chromosome ends, notably XpYp, we have detected copy number polymorphisms on a scale (1-2kb) too small to detect by FISH. In screening for pathological rearrangements, the quantitative data produced by MAPH can be analysed statistically to test a null hypothesis of normal copy number, and diagnostic thresholds can be adjusted to vary the rates of false negatives and false positives. The ease with which large numbers of samples can be screened suggests the use of MAPH in primary screening of subtelomeric copy number, prior to definitive diagnosis by FISH.

Multiplex PCR of Short Fluorescent Fragments: a simple, fast and reliable method for the detection of heterozygous genomic rearrangements

T. Frebourg^1,2, F. Charbonnier², F. Casilli¹, G. Raux¹, P. Saugier-Veber^1,2, N. Drouot², M. Tosi¹;
¹INSERM EMI 9906, Faculty of Medicine, Rouen, FRANCE, ²Department of Genetics, CHU, Rouen, FRANCE.

The detection of heterozygous genomic deletions and duplications is technically difficult and represents a serious limitation to the complete diagnosis of many genetic diseases. We have developed Multiplex PCR of Short Fluorescent Fragments, a simple method which is based on: (i) the simultaneous amplification of several short genomic sequences using fluorescently labeled primers, (ii) the use of a limited number of cycles, (iii) the superposition of the fluorescent electropherograms and, (iv) comparison, between patients and controls, of the peaks representing the fluorescence of each amplicon. This method has already been adapted to the following genes: MSH2, MLH1, MSH6, C1NH, SMN, BRCA1 et RB1. It has already been included into our diagnostic routine of the HNPCC syndrome, of the hereditary forms of breast and ovarian cancer and of retinoblastoma and has improved the genetic counseling of spinal muscular atrophy. Moreover, this method has allowed us to characterize precisely the boundaries of a large number of heterozygous deletions or duplications. This method is much more sensitive and rapid than the Southern blot technique commonly used, is better suited than quantitative real time PCR to the analysis of genes containing large numbers of exons, and appears to be more flexible than the MAPH (Multiplex Amplifiable Probe Hybridization) method, particularly because it can be rapidly adapted to large numbers of genes.

A. Hovnanian;
UPR2163 Unité de physiopathologie cellulaire et moléculaire, CHU Purpan, Toulouse, FRANCE.

M. De Luca¹, E. Dellambra¹, G. Pellegrini¹, L. Guerra¹, S. Bondanza¹, F. Mavilio²;
¹Laboratory of Tissue Engineering, Istituto Dermopatico dell'Immacolata, Rome, ITALY, ²Institute of Biochemistry, University of Modena, Modena, ITALY.

Cell therapy is an emerging therapeutic strategy aimed at replacing or repairing severely damaged tissue with cultured cells. Since surface epithelia experience a continuous self-renewal process during life, the success of keratinocyte-mediated cell therapy requires cultivation and transplantation of epithelial stem cells. Under the appropriate culture conditions, epithelial stem cells can be cultivated and generate autologous sheets suitable for transplantation. Cultured keratinocytes are currently used to restore severe epithelial defects.
JEB is a group of severe inherited skin diseases caused by mutations in the genes encoding laminin 5 or other components of the hemidesmosome. We show here full phenotypic correction of the adhesion properties of stratified epithelium obtained from epidermal stem cells isolated from patients suffering from laminin-5-deficient JEB, and transduced ex vivo with a retroviral vector expressing the ß3 chain of laminin-5. We also propose a non invasive surgical procedure that allows transplantation of cultured epidermal sheets in local anesthesia.
Thus:
· the possibility of cultivating large areas of epidermis
· the availability of surgical protocols for grafting large skin areas
· the demonstration of sustained transgene expression and stable gene correction in epidermal stem cells from JEB-patients
prompt us to propose the implementation of a phase I/II clinical trial aimed at ex vivo gene therapy of selected JEB patients.

Anhidrotic ectodermal dysplasia with immunodeficiency is associated with genetic defects in the NF-kB pathway

J. L. Casanova;
Laboratoire de Génétique Humaine des Maladies Infectieuses, Faculté de Médecine Necker-Enfants Malades, Paris, FRANCE.

The molecular basis of X-linked recessive anhidrotic ectodermal dysplasia with immunodeficiency (EDA-ID) has remained elusive until hypomorphic NEMO mutations were found in male EDA-ID patients from several kindreds and two patients with a related and hitherto unrecognised syndrome of EDA-ID with osteopetrosis and lymphedema (OL-EDA-ID). Mutations in the coding region are associated with EDA-ID, and stop codon mutations with OL-EDA-ID. NEMO encodes the regulatory subunit of the IKK complex, which is essential for NF-kB signalling. Germline loss-of-function mutations in NEMO have been shown to be lethal in male foetuses. In contrast, NEMO mutations causing OL-EDA-ID and EDA-ID are milder, as they impair but do not abolish NF-kB signalling. EDA results from impaired NF-kB signalling through the Eda receptor. Abnormal immunity in OL-EDA-ID patients results from impaired cell responses to at least Lps, IL-1b, IL-18, TNFa and CD154. In conclusion, impaired but not abolished NF-kB signaling in humans results in two related X-linked syndromes which associate specific developmental and immunological defects. Other patients with EDA-ID still lack a genetic cause, implying that the etiological investigation of the EDA-ID syndrome should provide a molecular dissection of the NF-kB pathway.

E. E. Eichler;
Department of Genetics and Center for Human Genetics, Case Western Reserve School of Medicine and University Hospitals of Cleveland, Cleveland, OH.

It has been estimated that 5% of the human genome consists of interspersed duplicated material that has arisen over the last 30 million years of evolution. Two categories of recent duplicated segments can be distinguished: segmental duplications between non-homologous chromosomes (transchromosomal duplications) and duplications largely restricted to a particular chromosome (chromosome-specific duplications). A large proportion of these duplications exhibits an extraordinarily high degree of sequence identity at the nucleotide level (>95%) spanning large (1-100 kb) genomic distances. Through processes of paralogous recombination, these same regions are targets for rapid evolutionary turnover among the genomes of closely related primates. The dynamic nature of these regions in terms of recurrent chromosomal structural rearrangement and their ability to generate fusion genes from juxtaposed cassettes suggests that duplicative transposition has been an important force in the evolution of our genome. Cycles of segmental duplication over periods of evolutionary time may provide the underlying mechanism for domain accretion and the increased modular complexity of the vertebrate proteome. Further, our data suggest that a small fraction of important human genes may have emerged recently through duplication processes and will not possess definitive orthologues in the genomes of model organisms. I will discuss the organization of recent segmental duplications within the human genome and their impact in terms of disease, gene innovation and rapid restructuring of the primate genome.

The relationship between genome organisation and gene expression at a human telomeric region

D. R. Higgs;
Weatherall Institute of Molecular Medicine, Oxford, UNITED KINGDOM.

A major challenge in the post-genomic era is to understand the relationship between genome structure and function (transcription, replication, repair and recombination). Over the past ten years it has become clear that the interaction of DNA with chromatin and the associated epigenetic modifications (DNA methylation, changes in replication timing, histone tail modification, nuclear sublocalisation) play a major role in elaborating the information encoded in DNA. However, the hierarchy of these epigenetic phenomena and the order of events in regulating nuclear processes appear complex and are largely unknown at present.
To learn more about the relationship between chromosome structure and function we have been characterising the most telomeric 300 kb region of the short arm of human chromosome 16. This is a GC-rich, Alu-dense region containing a variety of widely expressed and tissue-specific genes, including the embryonic and adult alpha-like globin genes. We have extensively characterised the structure and epigenetic modifications of this region enabling us to define a segment of the chromosome which has been maintained as a well-defined conserved syntenic region throughout evolution. This region appears to contain most, if not all, of the information required to fully regulate alpha globin gene expression in experimental assays. The critical cis-acting sequences which mediate these aspects of chromosome function are being characterised and have been shown to be dispersed throughout a region of at least 100 kb.

C. Maurange, M. Prestel, G. Rank, L. Ringrose, H. Ehret, A. Kuhrs, R. Paro;
ZMBH, Universität Heidelberg, Heidelberg, GERMANY.

In Drosophila the proteins of the Polycomb (PcG) and trithorax group (trxG) are necessary to maintain throughout development the differential expression patterns of developmental regulators such as the homeotic genes. PcG and trxG proteins read the activity state of their target genes, as set during the stages of determination, and “lock” the surrounding chromatin either in an active or repressed state. The control of the chromatin state is occurring through the binding of PcG and trxG proteins to common cis-elements termed Cellular Memory Module (CMM). In order to study the mode of CMM action in more detail we have established a transgene system that allows us to switch CMMs from a repressed to an activated state that is mitotically heritable. Results from this system show that transcription through a CMM is necessary for switching the activity state, which might explain the function of several non-coding RNAs transcribed close to genes potentially controlled by CMMs. In a complementing approach, we study the role of CMMs in determination and transdetermination processes in Drosophila imaginal discs.The molecular nature of the epigenetic mark maintaining either silencing or activation of a CMM throughout development is not known. We are investigating histone modifications as candidates for such a mark and indeed find a correlation between H3K9 methylation and PC protein localization. To determine whether this methylation plays a structural role in anchoring the PcG at CMMs, we challenged PC binding on polytene chromosomes in permeabilised salivary glands with various competitor peptides. Strikingly, only a subset of PC bound loci were competable by the H3K9 methylated peptide, suggesting that other additional histone modifications may mark the CMM.

F. Cucca;
Dipartimento di Scienze Biomediche e Biotecnologie, University of Cagliari, Cagliari, ITALY.

The genetic analysis of a complex trait like type 1diabetes (T1D) is complicated by many factors. Aside from IDDM1, the major disease superlocus located in the HLA region on chromosome 6p21, low penetrance, compounded by small individual genetic effects of the other loci create severe difficulties. Interlocus and allelic heterogeneity might further complicate the analysis. Some of the aforementioned factors could be alleviated by concentrating on an isolated and relatively homogenous population such as that from Sardinia. This island has, together with Finland, the highest incidence of T1D in the world. Children with Sardinian parents, who live in the Italian mainland, where the indigenous incidence of T1D is much lower, have the same incidence of T1D as the Sardinian children living on the island, thus supporting the role of genetic factors in the high incidence of the disease. The Sardinians represent a genetic isolate in which the substantial lack of population sub-structure reduces the risk of artifacts due to population admixture. Finally, the present time Sardinian population seems to be the result of a fixation of alleles and haplotypes, rare or absent in other European derived populations, that are particularly useful for trans-ethnic analysis to fine map the etiological variants.
Our results illustrate the advantages deriving from the genetic analysis of T1D in the Sardinian population. The relative importance in the dissection of this complex trait of population specific variables as well as of the disease related factors and of the chromosome-region specific effects will be discussed.

J. M. Soria;
Unitat d'Hemostasi i Trombosi, Hospital de la Santa Creu i Sant Pau, Barcelona, SPAIN.

Thrombosis is a complex disease. Multiple interactions between genetic and environmental factors contribute to the development of the disease. Presently, we know of six or seven genetic risk factors for venous thrombosis, that can explain about 60% of families with thrombophilia. To identify new genetic risk factors for thrombosis we design the GAIT project (Genetic Analysis of Idiopathic Thrombosphilia).
This project included 397 individuals from 21 Spanish families. In all of them we measured 43 quantitative phenotypes, and we genotyped a total of 500 highly informative genetic makers. The statistical genetic analysis has been performed using a variance component model included in the software package SOLAR.
Our results demonstrate the importance of genetic factors in determining variation in hemostasis-related phenotypes. Most importantly, over 60% of the variation in susceptibility to thrombosis is attributable to genetic factors.
From the genome-wide scan, the first undertaken to identify regions containing genes influencing variation in susceptibility to thrombotic disease and its intermediate phenotypes, we demonstrated that the G20210A mutation is functional in relation to prothrombin plasma levels and the risk of thrombosis, and the polymorphism responsible for the ABO blood group is functional in relation to plasma levels of factor VIII and factor von Willebrand .
Moreover, our analyses revealed a strong linkage between a QTL influencing FXII levels and the FXII gene (specifically the 46C/T FXII DNA variant; LOD = 10.21). In addition, a region on chromosome 1 showed strong evidence of linkage with free PS levels (LOD = 4.07). Another interesting result suggests that multiple loci are influencing the normal variation in APCR, and FV DNA variants play a relatively minor role in this normal variation in APCR. These examples as part of the current results confirm the valuable potential of this approach as a basic tool for mapping the genes of complex diseases.

Repairing and protecting neurones, a dual goal for cell based therapy to the brain

For more than a dozen years, a major combined biological and clinical research endeavour has been dedicated to the set up of new therapeutics based upon cell and gene therapy for neurodegenerative diseases. This research essentially takes into account two determinant characteristics of all these diseases, that can be briefly summarised as follows: 1. a neurodegenerative disease is due to the loss of one or a small number of specific populations of neurones, allowing in some cases for focal intra-cerebral intervention; 2. this neuronal loss is always progressive, according however to a schedule which is quite different from one disease to another, therefore providing a potential "therapeutic time window" for protective intervention. Without a discrete knowledge of the physiopathology of the diseases, and of their specific molecular and cellular mechanisms, one can envision basically two different, and complementary therapeutic modalities for these diseases. First, one may consider substituting to the degenerated neurones, homologous cells that are capable of replacing them anatomically and functionally. This so-called "substitutive" therapeutics is essentially based, at this moment, upon the use of neural cells obtained from human foetuses following elective abortions. Second, proteins have been identified that are able to protect neurones against various experimental aggressions in animals and are, therefore, good candidates to rescue neurones affected, though not yet degenerated, during disease progression. This so-called "conservative" therapeutics is essentially attempted, at this moment, by cell or viral based transfer of a gene of interest into the brain. I will present data on Huntington's disease.

E. Cattaneo;
Department of Pharmacological Sciences and Center of Excellence on Neurodegenerative Diseases, University of Milano, Milan, ITALY.

Huntingtin is a cytoskeletal protein which is important for neuronal survival and activity. Attention onto this protein stems from the knowledge that an expansion in the variable CAG tract in the encoding gene causes Huntington’s Disease (HD), an inherited, fatal,autosomal dominant neurodegenerative disorder characterized by selective loss of the striatal neurons (HDCRG, Cell, 1993).
Evidence indicate that HD occurs through a gained toxicity of mutant huntingtin. More recently, the possibility that loss of normal huntingtin function may contribute to HD has gained considerable attention (Rigamonti, J. Neuroscience 2000; Cattaneo, Trends in Neuroscience, 2001). We indeed report that normal (but not mutant) huntingtin is able to increase the transcription of Brain Derived Neurotrophic Factor in cortex, which is then delivered to striatum via the cortico-striatal afferents and thereby acting, within striatum, as a survival factor (Zuccato, Science, 2001). Loss of this activity occurs in HD due to huntingtin’s mutation leading to striatal vulnerability. Strategies aimed at restoring normal huntingtin activities in HD may therefore be beneficial.

R. M. Cassidy;
Department of Cell and Molecular Biology, Medical Nobel Institute, Karolinska Institute, Stockholm, SWEDEN.

Over the past decade it has become clear that stem cells in the adult mammalian brain continuously generate new neurons, predominantly in the hippocampus and olfactory bulb. Data generated in our laboratory demonstrate that ependymal cells lining the ventricular system of the brain and spinal cord function as neural stem cells in the adult CNS. Ependymal cells divide rarely to give rise to subventricular zone progenitor cells, which generate neuroblasts that migrate to the olfactory bulb. In response to a spinal cord injury, ependymal cells lining the central canal are induced to proliferate and generate migratory progeny that differentiate into astrocytes and contribute scar formation.
Recent data from our laboratory further demonstrate that these stem cells also generate neurons in unexpected regions of the brain, suggesting that adult neurogenesis is even more widespread than previously thought.
Given the expanding implications of adult neurogenesis, the central issue of whether neurons generated in the adult mammalian brain actually participate in functional synaptic circuitry has yet to be resolved. We have used virus-based transsynaptic neuronal tracing to demonstrate that neurons generated in different regions of the adult brain integrate correctly into the existing synaptic circuitry. Furthermore, we demonstrate that neurons generated in the adult brain respond to a physiological stimulus and are thus functional. Taken together, these findings may have implications for our understanding of the pathogenesis of neurodegenerative disorders and further provide a promising foundation for the development of therapeutic strategies to stimulate neurogenesis in the adult brain.