VERNE CHAPMAN LECTURE                SUNDAY 21 OCTOBER              18.00HRS – 19.00HRS





Jean-Louis Guénet

Institut Pasteur

25 rue du Docteur Roux

75724 PARIS Cedex 15












The founder colony of modern inbred laboratory strains was, most probably, a very small sized hybrid population derived from progenitors of the Mus musculus complex of species. This fact is substantiated by historical records as well as by the observation that most of the classical strains share the same mitochondrial DNA of Mus musculus domesticus origin and that most strains carry a Y chromosome from the Mus musculus musculus species. It follows that the genetic polymorphism present in these strains is not abundant and originates from three species: Mus m. domesticus, Mus m. musculus or Mus m. castaneus. Such a restricted genetic pool together with the relatively short period that has elapsed since the construction of present-day laboratory strains explains why such mice exhibit relatively little allelic variation compared, for example, with man. This disadvantage was overcome when mouse geneticists decided to take advantage of the diversity existing among wild specimens of the Mus genus, to develop new inbred strains what allowed the establishment of highly polymorphic inter-specific or inter-subspecific crosses. Except in a few cases, in particular when t-haplotypes were segregating in the wild progenitors, most attempts at establishing such inbred strains have been successful with wild mice of the Mus musculus complex of species. With progenitors of the other species of the Mus genus, such as Mus spretus or Mus spicilegus, the derivation of new inbred strains has proven more difficult but scientists have succeeded on several occasions.


The use of inbred strains derived from wild specimens is particularly appropriate in the design of experimental crosses for positional cloning of mouse genes because this allows the establishment of high resolution, high density maps of the flanking regions. In addition to their value as a source of genetic polymorphism, wild mice also represent an invaluable source of morphological variations at the level of karyotype. In laboratory strains a very large variety of reciprocal translocations or inversions exist that are by-products of the many experiments made to evaluate radiation hazards. On the contrary, Robertsonian translocations, or centric fusions, are uncommon. However, this type of chromosomal variations is fairly common in wild populations of the Mus m. domesticus species. These naturally occurring rearrangements have been extensively used for the generation of a wide variety of monosomies and trisomies. They have also been used for the demonstration of chromosomal imprinting. They are also very useful tools for gene localisation by in situ hybridisation because they assist in the discrimination between paired chromosomes, something that is often difficult using the normal mouse chromosomal complement. It is no risky prediction to say that, in the future, the use of wild mice will expand to other fields than gene mapping or the study of genomic imprinting. They will certainly become an essential tool for the study of epistatic interactions and for studies about cancer predisposition or susceptibility to infectious diseases.

PLENARY SESSION                              MONDAY 22 OCTOBER             09.15HRS – 09.45HRS




John D McPherson

Genome Sequencing Center

Washington University School of Medicine

St. Louis

MO 63108











The mapping and sequencing capacity of the International Sequencing Consortium was greatly increased during the production of the working draft of the human genome.  Until recently, this capacity has been utilized primarily for completing this genome.  The finishing phase of the human genome will only occupy approximately 10% of this available capacity leaving large resources available to be applied to other genomes.  A major focus of the Sanger, Washington University and Whitehead Institute genome sequencing centers is the generation of a working draft of the mouse genome as rapidly as possible and the completed sequence by 2005.  To date, a BAC physical map has been generated by restriction digest fingerprinting and consists of 300,000 BAC clones ordered into less than 700 contigs.  These contigs have been aligned to the mouse RH map and the human sequence map.  Clones have been selected from these contigs for shotgun sequencing.  To provide a working draft of the mouse genome in advance of the BAC-based sequence, a three-fold coverage of the genome has been generated in largely paired-ends reads.  Plans are underway to supplement these sequences with two-fold more genome coverage using larger insert clones by the end of the year.  The physical map and the assembled whole genome sequence reads will be a valuable tool for mouse research.  Availability of these resources, progress to date and future plans will be presented.


ORAL PRESENTATION             MONDAY 22 OCTOBER             09.45HRS – 10.00HRS              





Kerstin Lindblad Toh

Whitehead Institute

320, Charles Street






Co-Authors: 1,6)Brown DG, 2)Ainscough R, 1)Batzoglou S, 1)Birren B, 3)Brent M, 2)Clee C, 1)Jaffe D, 4)Kent J, 1)Lander ES, 1)Linton LM, 5)Marra M, 3)McPherson JD, 2)Mortimore B, 2)Mullikin J, 1)Nusbaum C, 2)Plumb B, 2)Rodgers J, 3)Sekhon M, 1)Stange-Thomann N, 3)Waterston RH,3) Wylie K, 2)Willey D, 3)Wilson RK, 1)Zody MC

Institutions:1)Whitehead Institute/MIT, 2) Sanger Centre, Cambridge, 3)Washington University, 4)University of California Santa Cruz, 5)British Columbia Cancer Agency, 6)University of Waterloo




Given the significance of the mouse as a model organism and the value of a second complete mammalian genome for comparative studies, there is great scientific interest in producing a complete sequence of the mouse genome. The Sanger Centre, Washington University, and the Whitehead Institute, have now turned significant sequencing capacity towards large scale mouse genome sequencing. This is being done in two main phases: an initial random shotgun survey of the whole genome followed by a BAC-based hierarchical sequencing approach with the intent of producing a finished genome (keynote: J. McPherson).


The first phase has been the generation of 16,5 million reads constituting a ~2.5-3.0-fold whole genome shotgun (WGS) coverage (traces available at: Annotation of the human genome by aligning the mouse reads to the human genome and a preliminary shotgun assembly of them is underway. This will shed light on conservation between the mouse and human genome in general and for different features such as exons, introns and intragenic regions. Although further study will be required, the conserved regions will help confirm and improve existing gene predictions, identify possible novel genes, and help highlight likely regulatory regions in both genomes.  One distinctive feature is the presence of several long, highly conserved, non-exonic features between mouse and human.  Preliminary analysis to understand characteristics of these features is underway.

ORAL PRESENTATION             MONDAY 22 OCTOBER             10.00HRS – 10.15HRS





Shaying Zhao

The Institute for Genomic Research (TIGR)

9712 Medical Center Drive


MD 20850





Co-Authors:  Shetty J, Shatsman S, Ayodeji B, Geer K, Tsegaye G, Krol M, Gebregeorgis E, Shvartsbeyn A, Russell D, Overton L, Jiang L, Dimitrov G, Tran K, Malek J, Feldblyum T, Nierman W, Fraser CM

Institutions: The Institute for Genomic Research




A large scale BAC end sequencing project at The Institute for Genomic Research (TIGR) has generated one of the most extensive set of sequence markers for the mouse genome to date.  With a sequencing success rate of over 80%, an average read length of 485 bp and an average Q20 bases of 406 bp, we have generated 449,234 mouse BAC end sequences (mBESs) from 257,318 clones from libraries RPCI-23 and RPCI-24, representing 15X clone coverage, 7% sequence coverage, and a marker every 7 kb across the genome.  A total of 191,916 BACs have sequences from both ends providing 12X genome coverage.  ABI3700 sequencers and the sample tracking system ensure that over 95% of mBESs are associated with the right clone identifiers.  The annotation results of mBESs for the contents of repeats, human and mouse genomic sequences, ESTs and STSs indicate that this resource will be valuable to many research fields.


Besides the mouse, a large scale BAC end sequencing project has started for the rat at TIGR, where the goal of the project is to generate paired-ends from 200,000 rat BAC clones in one year.  We have generated more than 60,000 sequences from library CHORI-230.


These mouse, rat BAC end sequences along with the BAC end sequences from primates including human, chimpanzee, baboon rhesus macaque and lemur have been mapped to the human genome to compare three versions of the human genome assemblies (GoldenPath, NCBI and Celera) and to study the genome evolution.


ORAL PRESENTATION             MONDAY 22 OCTOBER             10.15HRS – 10.30HRS





Muriel Davisson

The Jackson Laboratory

600 Main Street

Bar Harbor







Co-Authors:  2)Gardiner K, 1)Akeson EC, 1)Bechtel LJ, 2)Fortna A, 2)Slavov D, 1)Schmidt C

Institutions:  1)The Jackson Laboratory, 2)The Eleanor Roosevelt Institute



Human Chr 21q is conserved, centromere to telomere, in mouse, Chrs 16, 17 and 10. Ts65Dn mice are trisomic for much of the Chr 16 segment. To facilitate the construction of appropriate mouse models of Down syndrome we have mapped additional genes in mouse Chrs 16 and 17 and identified additional genes in the Ts65Dn segment. The annotated complete genomic sequence of human Chr 21 was used to identify relevant genes, followed by database searches to identify the homologous mouse genes. The mouse T-31 radiation hybrid (RH) panel, FISH, and genomic sequence analysis were used to map these genes to the mouse chromosomes. These data show that the mouse Chr 17 component is about 1.5 Mb and contains possibly as few as 21 genes. The Chr 16 breakpoint of the Ts65Dn chromosome lies between Ncam2 and App/Gabpa and the Ts65Dn chromosome contains ~60% of the human Chr 21 homologs. The mouse Chr 16 - Chr 17 evolutionary breakpoint is between human genes ZNF295 and UMODL1. The Chr 17 - Chr 10 breakpoint is between KIAA0179 and PDXK, and appears to have involved a duplication of the PDXK gene, which is present as a pseudogene, Pdxk-ps, in mouse Chr 17.

ORAL PRESENTATION             MONDAY 22 OCTOBER             10.30HRS – 10.45HRS





AM Mallon

MRC UK Mouse Genome Centre and Mammalian Genetics Unit



OX11 0RD




Co-Authors: 1)P Denny, 2)MRM Botcherby, 4)P Gautier, 5)H Hummerich, 4)S Cross, 4)V van Heyningen, 2)N Leaves, 2)J Greystrong, 2)L Greenham, 2)S Jones, 2)K Maggott, 2)S Manjunath, 2)G Strachan, 3)M Strivens, 2)P North, 2)M Campell, 2)G Hunter, 2)G Kimberley, 2)L Cave-Berry, 3)L Mathews, 3)S Simms, 3)S Gregory, 3)R Evans, 3)T Hubbard, 3)R Durbin, 1)M Cadman, 1)R Mc Keone, 1)C Sellick, 5)M Iravani, 4)S White, 6)P Little, 4)I Jackson, 3)J Rogers, 2)RD Campbell,  1)SDM Brown

Institutions: 1) MRC UK Mouse Genome Centre and Mammalian Genetics Unit, 2) MRC UK-HGMP Resource Centre, 3) Sanger Centre, 4) MRC Human Genetics Unit, Western General Hospital, 5) MRC Prion Unit, St. Mary’s Hospital Medical School, 6) School of Biochemistry & Molecular Genetics, University of New South Wales



The UK mouse sequencing programme commenced in October 1999 with the aim of sequencing 50 Mb of the mouse genome targeted to four core regions:


·         The WAGR-homologous region on mouse chromosome 2

·         The brown deletion complex on mouse chromosome 4

·         The Del(13)Svea36H chromosome 13 deletion

·         Dmd-Ar region on chromosome X


For each of these four regions, there are already significant research efforts ongoing within the UK focused on the systematic determination of gene function through mutagenesis and other studies. The work of the consortium complements the MRC ENU mutagenesis programme, as three of its major target regions will be sequenced. In addition, 10% of the sequencing capacity is being devoted to additional requests by UK genomics researchers.


The programme will focus on the delivery of finished, contiguous sequence for these regions by 2002, improving the efficiency of mutation scanning and the identification of genes underlying mutations of interest.  The discovery of candidate genes and features in these targeted regions is being undertaken by annotation generated from the Ensembl project ( and also by cross-species comparison. 


The mapping stage is nearing completion and has been carried out by the three mapping groups at HGU, Harwell and Imperial College. Significant draft and finished sequence is now available from all regions. A progress report on the sequencing will be presented along with an update on the analysis of sequence generated in the four core regions.


ORAL PRESENTATION             MONDAY 22 OCTOBER             1045HRS – 11.00HRS





Jim Thomas


49 Convent Dr MSC4431, Rm 2C28


MD 20892





Co-Authors: 1)Summers TJ, 1)Lee-Lin SQ, 1)Maduro VVB, 1)Idol JR, 1)Prasad AB, 1)Ryan JF, 1)Touchman JW, 2)Bouffard GG, 2)Thomas PJ, 2)Beckstrom-Sternberg SM, 2)Blakesley RW, 2)McDowell JC, 2)Dietrich NL, 2)Green ED

Institutions: 1)National Human Genome Research Institute, 2)NIH Intramural Sequencing Center



The sequence of the human genome provides a reference point by which we can compare ourselves to other organisms.  To date, most comparative sequencing has focused on pair-wise comparisons between human and a limited number of other vertebrates, such as mouse.  Thus, little is presently known regarding the value of comparative sequencing across multiple species.  As an extension of our human-mouse comparative mapping and sequencing that focused on the ~150 Mb of the mouse genome orthologous to human chromosome 7, we are now establishing sequence-ready BAC maps in 10 additional vertebrates for genomic segments corresponding to 15 Mb of human chromosome 7.  Mapping and sequencing progress in chimpanzee, baboon, cow, pig, dog, cat, rat, chicken, fugu and zebrafish has already led to the selection of ~900 clones (~135 Mb) for sequencing, with ~44 Mb of sequence currently deposited in GenBank.  As part of the mouse genome sequencing effort, the NIH Intramural Sequencing Center (NISC) is committed to finishing the portion of the mouse genome orthologous to these 15 Mb.  An additional 30 Mb of draft mouse sequence orthologous to human chromosome 7 has already been deposited in GenBank by NISC.  The establishment of this comparative sequence resource should facilitate the development of new computational tools for multi-species sequence comparisons and address the benefits of sequencing species from a range of evolutionary distances.


PLENARY SESSION                              MONDAY 22 OCTOBER             11.30HRS – 12.00HRS





Eugene W. Myers

Celera Genomics

45 West Gude Drive









Co-Authors: Sutton G, Mural R, Li P, Yandell M, Halpern A, Smith H, Venter JC

Institution: Celera Genomics



The assembly of a shotgun data set of end-sequence reads was controversial in 1998 with critics claiming that it would require an impossibly large computation and result in a very fragmented and error-ladden assembly.


In 1999 the informatics research team at Celera produced an assembly of the 130Mbp Drosophila genome from a 13X whole genome shotgun data set, followed by an assembly of the Human genome in 2000 with a 5.1X data set and synthetic reads from public data.  This year we produced an assembly of the Mouse genome from a 5.3X data set of 3 different mouse strains in equal proportions.


Our results from the mouse project prove unequivocally that whole genome shotgun sequencing is effective at delivering a highly reliable reconstruction with long-range order and orientation and dense polymorphism information.  The fact that assembly is achieved at only 5.3X implies that a relatively complete picture of a large vertebrate genome can be obtained in six to nine months at very competitive cost with current technology.  We demonstrate that the 5.3X assembly is a solid substrate for annotation and that it has great syntenic power when compared against the human genome.

ORAL PRESENTATION             MONDAY 22 OCTOBER             12.00HRS – 12.15HRS





Tim Wiltshire


3115 Merryfield Row

San Diego

CA 92121






Co-Authors: 1)Walker J, 1)Hakak Y, 1)Su A, 1)Ching K, 1)Hogenesch J, 1)Cooke P, 1)Schultz P, 2)Patapoutian A, 2)Moqrich A,

Institutions: 1)GNF, 2)The Scripps Research Institute



The recent description of the human genome is the first in a series of milestones to be joined in the future by comprehensive descriptions of the transcriptome and the proteome.  Here we report results from an analysis of a significant portion of the transcriptome and its integration with genomic sequence data.   We used high-density oligonucleotide arrays to interrogate the mRNA expression of 8882 human and 6139 mouse genes in a diverse set of tissues, cell lines, organs, and pathological samples.  A total of 46 human and 45 mouse tissues were analyzed initially, resulting in more than 2,500,000 independent measurements of gene expression.  Three strategies were chosen to validate data; RT-PCR, Northern blot analyses and in-situ hybridization.  These data have been curated and will be made publicly available with annotation gathered from the public domain. 


This extensive set of transcript profiles, representing a significant portion of the mammalian transcriptome makes it possible to address a number of fundamental questions in transcription and adds to our understanding of global transcriptional control, physiological and molecular gene function, transcriptional regulation of functional pathways and co-regulated genes, and disease etiology.  We also performed comparative analyses between mouse and human expression patterns of ortholog gene pairs.  These data indicate good concordance between mouse and human gene expression but also highlights differences between tissues.


ORAL PRESENTATION             MONDAY 22 OCTOBER             12.15HRS – 12.30HRS





Lee Smith

Medical Research Council

MRC Mammalian Genetics Unit



OX11 0RD




Co-Authors: 1)Siggers P, 1)Van Hateren N, 1)Larder R, 1) Walsh J, 2) Stephens R, 2) Freeman T, 1)Greenfield A

Instututions: 1) MRC Mammalian Genetics Unit, 2) MRC HGMP Resource Centre




The mouse gonad and mesonephros originate from primordia that initially exhibit no morphological differences between males and females. The gonadal primordium (genital ridge) develops into a testis or ovary depending on the presence or absence of Sry in the gonadal supporting cell lineage between 10.5 and 13.0 days post coitum (dpc). The mesonephros also has a sexually dimorphic fate, depending on the presence or absence of hormonal signals from the testis. Several genes regulating these processes have been identified, but many gaps exist in these pathways.


We have used DNA microarrays to screen for genes exhibiting sexually dimorphic patterns of expression at different stages of mouse gonad and mesonephros development. These microarrays have been constructed using cDNA clones from a variety of embryonic sources. We present an overview of this work focussing on new genes whose patterns of expression suggest important roles in sexual development.     



ORAL PRESENTATION             MONDAY 22 OCTOBER             12.30HRS – 12.45HRS





Harukazu Suzuki

RIKEN Genomic Sciences Center

1-7-22, Suehiro-cho, Tsurumi-ku






Co-Authors: Saito R, Kanamori M, Miki R, Kagawa I, Bono H, Okazaki Y, Hayashizaki Y

Institutions: Laboratory for Genome Exploration Research Group, RIKEN Genomic Sciences Center (GSC), Japan and Genome Science Laboratory, RIKEN Tsukuba Institute and Dept of Medicine, Tsukuba University



The RIKEN mouse encyclopedia project has already collected large number of mouse full-length enriched cDNAs, where many of them correspond to novel genes of unknown function. In order to uncover their function as a systematic genome-wide approach, we have developed a high-throughput protein-protein interaction (PPI) assay system based on a modified mammalian two-hybrid method (reported in the last IMGC meeting). We have now obtained >500 PPIs using 6,000 RIKEN cDNAs. For the detailed analysis of these PPIs, we have developed software, the PPI network viewer; we exhibited the strength of each interaction as a thickness of network line since the luciferase reporter activity roughly parallels the strength of interaction. This seems to be useful evaluating biological significance of the PPIs since two-hybrid data involve the pseudo-positive PPIs. Several intriguing PPIs that we found by the viewer will be reported. Further, we have made links from the viewer to our gene expression profiles and gene mapping data for the integrative analysis. As one of analyses, we have studied the correlation coefficient of gene expression profiles between the PPI pairs. The results show that the correlation between the pairs seems to depend on both the protein functions and the set of selected expression data. We will report the results and discuss how these analyses contribute to uncover the function of the novel genes.



ORAL PRESENTATION             MONDAY 22 OCTOBER             12.45HRS – 13.00HRS





William L Stanford

Samuel Lunenfeld Research Institute

600 University Ave, Rm. 989A

Toronto, Ontario

M5G 1X5







Co-Authors: Bernstein A, Cohn JB

Institutions: Samuel Lunenfeld Research Institute



We are performing gene trap-based expression and genotypic screens to generate new mouse mutations that will help delineate the molecular controls of specific developmental programs. Using a polyA trap vector with recombination sites for post-insertional manipulations, gene trap insertions are screened using multiplexed in vitro differentiation and induction assays. An expression profile is being generated at a rate of greater than 5000 per year. Sequence tags for all insertions demonstrating restricted expression patterns (about 20%) are now being generated. A database is being developed that will be searchable by expression pattern, sequence, and phenotype. The clones will be available as a resource to researchers worldwide. One of the clones identified in our ongoing screen, SNAG-1, is expressed by hematopoietic stem cells, neural endothelial cells, cardiomyocytes and sensory nerves. SNAG-1 mutant embryos die mid-gestation with neural vascular hemorrhaging and cardiac edema. SNAG-1 encodes a protein containing an N-terminal SH3 and a C-terminal PX domain. Insertion of the vector into the SNAG-1 locus resulted in a fusion transcript of the upstream SH3 domain and the lacZ reporter, deleting the PX domain. SNAG-1 is differentially regulated upon cellular activation and is localized to the Golgi Apparatus. The vector insertion leads to mis-localization of SNAG-1, suggesting that the PX domain is required for localization and function within the Golgi.


PLENARY SESSION                              MONDAY 22 OCTOBER             15.00HRS – 15.30HRS





Jean Weissenbach

Genoscope and CNRS UMR-8030

2 rue Gaston Cremieux

CP 5706

Evry Cedex 91057






Co-Authors: Olivier Jaillon, Michael Levy, Hugues Roest Crollius, Alain Bernot, Laurence Bouneau, Cécile Fischer, 1) Catherine Ozouf-Costaz,  William Saurin

Institutions:  Genoscope and CNRS UMR-8030, 1) Museum d'Histoire Naturelle



Despite the availability of most of the human genome sequence, the accurate identification of genes on the DNA sequence is not yet definitive and has to be continuously improved and updated. This relies on the combined use of a variety of tools including similarity searches and exon/gene prediction programs.


Similarity searches rely on a set of sequence data that has been increasing very substantially and rapidly during the past months. These include human cDNAs, mouse genomic DNA and cDNAs and puffer fish genomic sequences. With these resources, the definition of a vertebrate set of genes is progressively

taking shape.


To contribute to the generation of these new datasets, we have undertaken a whole genome sequencing of the fresh water pufferfish Tetraodon nigroviridis. This programme relies on the establishment of a physical map based on several approaches including BAC fingerprinting, in situ hybridization, STS mapping etc. Shotgun DNA sequence reads representing nearly 3 X coverage of the genome of Tetraodon nigroviridis were produced.


In addition to their assembly, currently in progress, these sequence data were used for direct similarity searches with sequences from other genomes.Results of these large scale sequence comparisons will be presented.

PLENARY SESSION                              MONDAY 22 OCTOBER             15.30HRS – 16.00HRS





Dr Tim Hubbard

Sanger Centre

Wellcome Trust Genome Campus



CB10 1SA.










I will discuss the current state of human and mouse sequence analysis as provided by the Ensembl project and by gene structure curation efforts at the Sanger Centre.  The Ensembl project ( is an established system for automatic annotation of genomic sequence data.  It provides both a portable, open-source software system and a full analysis of current human and mouse sequence data.  Automatic annotation is essential in order to be able to handle the current flood of data rapidly.  However, in order to generate "gold standard" annotation additional curation and experimental investigation is required.   Efforts in Sanger have already annotated 10% of the human genome in this way and the results are both submitted to the EMBL sequence database and made available via Ensembl.  I will also discuss work to extend analysis beyond gene structures to promotors (




ORAL PRESENTATION             MONDAY 22 OCTOBER             16.00HRS – 16.15HRS





Dr Deanna Church


Building 38A, rm6S614K

8600 Rockville Pike








Co-Authors:  Maglott D, Shkeda A, Chen H-C, Schuler GD

Institution:  National Institute of Biotechnology Information



Currently, there are greater than 17 million Whole Genome Shotgun (WGS) reads from C57BL6/J, representing roughly 3-fold coverage of the mouse genome. In addition, more than 400 Mb of sequence from greater than 2300 accessioned clones is also available. The majority of this sequence is contained within HTGS phase 1 sequence (322 Mb in 1696 accessions). The average phase 1 accession contains 24 fragments, with an average fragment length of 7.7 Kb. HTGS phase 3 sequences are being assembled into non-redundant, annotated sequence contigs. Our effort to assemble and annotate mouse genomic sequence will be discussed. The WGS reads are currently being assessed for quality, repeat-content, pairing rates.  In addition, these WGS are being compared to the human genome. When using well-annotated regions of the genome, there are currently as many sequence alignments in non-coding regions as there are in coding regions.


In addition to analyzing genomic sequence, we have been utilizing mRNA sequences found in LocusLink, EST sequences found in UniGene, and curated information from the Mouse Genome Database (MGD) to establish orthologs relationships between human and mouse. Using the human draft sequence as a guide, we have built human-mouse homology maps using two different human genome assemblies and two different mouse maps. The utility of the conserved synteny map as a tool to assess genome assemblies, as well as the virtual mapping of >3000 mouse genes will be discussed.

ORAL PRESENTATION             MONDAY 22 OCTOBER             16.15HRS – 16.30HRS





Carol Bult

The Jackson Laboratory

600 Main Street

Bar Harbor, Maine






Co-Authors:  King BL, Zhu YS, Baldarelli R, Ramachandran S, Bradt D, Cousins S, Beal J, Mani P, Kadin J and the Mouse Genome Informatics Staff

Institution:  Mouse Genome Informatics Group, The Jackson Laboratory



Computational sequence annotation pipelines have been developed to support large-scale genome sequencing efforts.  While computational annotations are useful as entry points into genome biology, they are usually not deeply integrated with existing biological knowledgebases. In contrast to these automated, computational annotation methods, community model organism database groups, such as the Mouse Genome Informatics (MGI) group ( rely heavily on the expertise of domain specialists to analyze and interpret biological data from diverse sources as part of an overall strategy to create and maintain a highly integrated and well-curated genome biology database for the laboratory mouse.  A major challenge facing MGI, and all model organism databases, is how best to incorporate information from large and constantly changing genomic sequence data streams with curation processes that rely heavily on human reasoning and interpretation. The MGI group is developing an annotation infrastructure that combines both automated and human-centric curation processes. We will present an overview of this infrastructure and discuss issues relative to our current work on integrating and updating the annotations of the emerging mouse genome sequence with the wealth of genetic and phenotypic data about the laboratory mouse that is already available from MGI.

PLENARY SESSION                              TUESDAY 23 OCTOBER                        09.00HRS – 09.30HRS





Adrian Bird

Wellcome Trust Centre for Cell Biology

Institute of Cell and Molecular Biology

University of Edinburgh

The King’s Buildings

Mayfield Road






Co-Authors: Brian Hendrich, Jacky Guy, Egor Prokhortchouk, Anna Prokhortchouk, Helle Jørgensen

Institutions: Wellcome Trust Centre for Cell Biology, Inst for Cell Biology, University of Edinburgh



Methylation affects cytosines at the DNA sequence 5’CpG in mammalian DNA and causes local silencing of gene expression. Proteins that bind to methylated DNA are likely mediators of silencing. MeCP2 is a relatively abundant chromosomal protein whose localisation in the nucleus is dependent on CpG methylation. MeCP2 can recruit histone deacetylases and therefore links DNA methylation with changes in chromatin structure. Mutations in the MECP2 gene are the primary cause of Rett syndrome, a progressive neurological disorder that affects girls. We used conditional gene disruption in mice to approach the function of MeCP2. Homozygous Mecp2-null mice appear normal until several weeks after birth when neurological symptoms appear. Mecp2+/- heterozygotes also initially appear normal and can raise normal litters, but eventually develop neurological symptoms that recall Rett Syndrome, including gait ataxia and breathing arrhythmia. MBDs 1 - 4 constitute a family of proteins containing a motif related to the methyl-CpG binding domain of MeCP2. One of these proteins, MBD2, also acts as a deacetylase-dependent transcriptional repressor in vivo. In the absence of MBD2, a component of the “MeCP1” complex in mouse cells is lost and repression of methylated genes is compromised in vivo. We recently identified yet another complex containing a novel methyl-CpG binding protein. This protein, Kaiso, binds tightly to a specific methylated sequence, but is unrelated to the MBD protein family. In order to address the functional significance of these proteins, we are disrupting the genes in mice and analysing the phenotypic consequences.

ORAL PRESENTATION             TUESDAY 23 OCTOBER                        09.30HRS – 09.45HRS





David Beier

Brigham & Women's Hospital/Harvard Medical School

20 Shattuck Street

Boston, MA






Co-Authors:  1)Liu S, 1)Lu W, 1)Kuida S, 2)Obara-Ishihara T, 2)Drummond I

Institutions:  1)Genetics Division, Brigham and Women's Hospital, Harvard Medical School, 2)Renal Division, Massachussetts General Hospital 



A defect in a novel nek-family kinase causes cystic disease in the mouse and in zebrafish


The juvenile cystic kidney (jck) mutation results in progressive polycystic kidney disease. By positional cloning we identified a mutation in a novel Nek kinase protein, which we call Nek8. The mutation results in a gly-to-val substitution in a highly conserved amino acid motif. To prove that this defect causes cystic disease, we performed a cross-species analysis taking advantage of the evidence that morpholino anti-sense oligos can be used to abrogate gene function in developing zebrafish. Treatment of embryos with an oligo corresponding to the 5' end of the zebrafish ortholog of Nek8 resulted in the formation of pronephric cysts. In vitro studies reveal that expression of either jck mutant Nek8 gene or a construct mutated in the kinase domain results in markedly enlarged and multinucleated cells, which also demonstrate a striking loss of actin stress fibers. Mutant jck kidneys do not show multinucleation in vivo and immunohistochemical analysis does not show mislocalization of membrane proteins. However, EM analysis reveals that, prior to cyst formation, the collecting duct epithelia appears to be lifting off the basement membrane.We suggest that a disruption in actin cytoskeletal assembly by the jck mutation affects cellular adhesion functions, which results in the development of cysts. Further, we suggest that a comparative analysis of gene function in different model systems represents a powerful means to annotate gene function.

ORAL PRESENTATION             TUESDAY 23 OCTOBER                        09.45HRS – 10.00HRS





Michael Coleman

University of Cologne

Institute for Genetics & Center for Molecular Medicine (ZMMK),

Zuelpicher Strasse 47





Co-Authors:  1)Mack TGA, 1)Reiner M, 1)Beirowski B, 1)Mi W, 2)Emanuelli M, 1)Wagner D, 3)Thomson D, 3)Gillingwater T, 4)Conforti L, 5)Fernando FS, 5)Tarlton A, 1)Addicks K, 2)Magni G, 6)Perry VH, 3)Ribchester RR

Institutions:  1)University of Cologne, 2)University of Ancona, 3)University of Edinburgh, 4)Mario Negri Institute, Milan, 5)University of Oxford, 6)University of Southampton 



The Wld gene: a unique neuroprotective factor for axons


Axon degeneration in both injury and disease (Wallerian degeneration) is greatly delayed by a dominant mutation in the C57BL/WldS mouse.  Here we identify the protective gene by reproducing the WldS phenotype in transgenic mice.  We confer a dosage-dependent block of Wallerian degeneration by expressing an in-frame fusion protein of Ubiquitination factor E4B (Ube4b) and Nicotinamide mononucleotide adenylyltransferase (Nmnat).  Five days after transection, up to 96% of distal axons and 93% of neuromuscular junctions were structurally and functionally preserved in transgenic mice, compared to none in wild-type mice.  70% of axons survive for at least 14 days.  Thus, degeneration of axotomised nerves can be prevented.  Nmnat enzyme activity is present in the protective protein and increases four-fold in WldS tissues without altering NAD+ content.  These data indicate that ubiquitination or pyridine nucleotide metabolism underlies Wallerian degeneration and axon protection. 


Degeneration of the axon, the largest compartment in many neurons, has been inexplicably neglected in studies of neurodegenerative disease.  Studies showing that the WldS mutation also protects from peripheral neuropathy (Samsam, M. and Martini R., in preparation) and motoneuron disease (Ferri, A. and Kato, A., personal communication) indicate that the Wld gene fills an important gap in potential therapeutic strategies.  With the identification of the gene, it is now possible to protect axons as well as neuronal cell bodies.

ORAL PRESENTATION             TUESDAY 23 OCTOBER                        10.00HRS – 10.15HRS





Grant MacGregor

Emory University School of Medicine

Center for Molecular Medicine,

1462 Clifton Rd NE, 403-E

Atlanta, GA

30322, USA


Co-Authors: 1,2)Campbell P, 2)Waymire K, 1,2)Heier R, 3,4)Sharer C, 4,5)Day D, 6)Friedrich G, 7)Burmeister M, 4,5)Bartness T, 8)Russell L, 3,4)Young L 9)Zimmer M, 10)Jenne D

Institutions: 1)Graduate Program in Genetics and Molecular Biology, Emory University, 2) Center for Molecular Medicine, University School of Medicine, 3) Dept of Psychiatry, Emory University, 4) NSF Center for Behavioral Neurosciences, Emory University, 5) Dept of Biology, Georgia State University, 6) Dept of Molecular Genetics, Baylor College of Medicine, 7) Mental Health Research Institute, University of Michigan, 8) Dept of Physiology, Southern Illinois University School of Medicine, 9) Institute of Clinical Biochemistry & Pathobiochemistry, University of Würzburg, 10) Dept of Neuroimmunology, Max Planck Inst of Neurobiology



The ROSA22 gene trap line of mice was identified during a screen for recessive mutations that cause male sterility. The basis for the sterility involves defective formation of the spermatid flagellum. The mutation is pleiotropic. Homozygous mutant ROSA22 animals display reduced body fat mass as adults. In addition, although they exhibit normal mating behavior, mutant males fail to display inter-male aggression. The gene mutated by the single retroviral insertion is located on chromosome 10 and is closely flanked by two previously identified genes. Expression of the flanking genes appears to be unaffected by the retroviral insertion. The gene is predicted to encode a novel transmembrane protein, and contains motifs involved in intracellular protein sorting. The trapped allele is expressed in polarized epithelial cells in a wide range of tissues, as well as in the CNS and PNS and developing male germ cells. When expressed in COS cells, the gene product localizes in a punctate manner to the ER with occasional co-localization with microtubules. Orthologs for the mutated gene appear to be present in humans, frogs and fish, but not in fly or worm. Models will be presented for the possible mechanism in which this novel gene product functions.


ORAL PRESENTATION             TUESDAY 23 OCTOBER                        10.15HRS – 10.30HRS





Philomena Mburu

MRC Mammmalian Genetics Unit & Mouse Genome Centre



OX11 0RD




Co-Authors: Varela-Carver A.,Holme RH, Hardisty RE, White D, Paige A, Fleming J, Rogers M, Kiernan BW, Steel KP, Brown SDM

Institutions: MRC Mammalian Genetics Unit and Mouse Genome Centre, MRC Institute of Hearing Research



Genetic deafness is highly prevalent in the human population, affecting 1 in 2000 births. Many of these show primary abnormalities of the sensory epithelia of the inner ear, as do several mouse mutants. In the whirler (wi) mutant the stereocilia of the inner hair cells of the cochlear duct are considerably shorter than wild-type while outer hair cell stereocilia take on a more rounded U shape compared to the normal V or W shape. Cloning of the defective gene underlying wi will provide insight into the molecular processes involved in normal development of stereocilia as well as providing valuable insights into the causes of neuroepithelial deafness.


The wi non-recombinant region is contained within a minimal tiling path consisting of 2 BACs and a PAC. One of the BACs has been used in transgenic rescue experiments and been shown to rescue the inner hair cell phenotype of wi  mutant mice, while there is partial rescue of the outer hair cell phenotype. Whirler mice carrying the BAC also demonstrate a normal Preyer reflex up to P95 and an absence of the usual circling and head tossing behaviour normally associated with the wi mutant. One of the genes in the BAC carries a mutation in wi mice and is expressed from the BAC in the rescued mutant mice. The gene encodes a PDZ protein and represents a provocative novel candidatemolecule involved in steriocilia development.

ORAL PRESENTATION             TUESDAY 23 OCTOBER                        11.00HRS – 11.10HRS





Alexander Bachmanov

Monell Chemical Senses Center

3500 Market Street

Philadelphia, PA






Co-Authors:  1)Li X, 1)Reed DR, 2)Ohmen JD, 1)Li S, 1)Chen Z, 1)Tordoff MG, 2&3)de Jong PJ, 2)Wu C, 2)West DB, 2)Chatterjee A, 2)Ross DA, 1&4)Beauchamp GK

Institutions:  1)Monell Chemical Senses Center, 2)Pfizer Global Research and Development, 3)Children's Hospital Oakland Research Institute, 4)University of Pennsylvania



Differences in sweetener intake among inbred strains of mice are partially determined by allelic variation of the saccharin preference (Sac) locus.  Genetic and physical mapping limited a critical genomic interval containing Sac to a 194-kb DNA fragment.  Sequencing and annotation of this region identified a gene (Tas1r3) encoding the third member of the T1R family of putative taste receptors, T1R3.  Introgression by serial backcrossing of the 194-kb chromosomal fragment containing the Tas1r3 allele from the high-sweetener preferring C57BL/6ByJ strain onto the genetic background of the low-sweetener preferring 129P3/J strain rescued its low sweetener preference phenotype.  Polymorphisms of Tas1r3 that are likely to have functional significance were identified using analysis of genomic sequences and sweetener preference phenotypes of genealogically distant mouse strains.  Tas1r3 has two common haplotypes, consisting of six single nucleotide polymorphisms: one haplotype was found in mouse strains with elevated sweetener preference and the other in strains relatively indifferent to sweeteners.  This study provides compelling evidence that Tas1r3 is equivalent to the Sac locus and that the T1R3 receptor responds to sweeteners.

ORAL PRESENTATION             TUESDAY 23 OCTOBER                        11.10HRS – 11.20HRS





Murray Brilliant

University of Arizona College of Medicine

Department of Pediatrics

1501 N. Campbell Ave

Tucson, AZ





Co-Authors: 1)Newton JM, 2)King RA, 1)Gardner JM, 1)Hagiwara N, 1)Cohen-Barak O, 1)Tran WN, 3)Davisson MT

Institutions:  1)University of Arizona, 2)University of Minnesota, 3)The Jackson Laboratory


Identification of the mouse underwhite (uw) gene and its human ortholog: oculocutaneous albinism type 4 (OCA4)


Mutations in the mouse underwhite (uw) locus lead to hypopigmentation.  The fur of mice homozygous for the most severe allele (uw) is a light beige color with white underfur and the eyes are pink at birth but darken somewhat with age.  A series of alleles at this locus has been previously described (Sweet et al., J. Heredity, 89:546-51, 1998; Lehman et al., J. Invest. Dermatol., 115:601-606, 2000), with the uw mutant phenotype and residual melanin composition closely resembling that of p mutant mice.  Two alleles, uw and uw-dense, are recessive, with uw homozygotes lighter than uw-dense homozygotes.  A third allele, Uw-dominant brown,  is semidominant. The dominance heirarchy in the allelic series is Uw-dominant brown, wild-type, uw, followed by uw-dense. We have cloned the mouse uw gene and found a dramatic reduction in expression of the encoded transcript in melanocytes homozygous for the uw allele.  Point mutations have been found in the uw-dense and Uw-dominant brown alleles.  The human ortholog has also been cloned.  Humans with mutations in this gene have a new type of albinism, OCA4 that phenotypically resembles OCA2.  Thus, the uw protein is a major determinant of normal pigmentation.   Studies are underway to determine the function of the encoded protein in the normal production of melanin.


ORAL PRESENTATION             TUESDAY 23 OCTOBER                        11.20HRS – 11.30HRS





Lydia Matesic


Mouse Cancer Genetics Program

Bldg 539, Room 234

National Cancer Institute            


21702, USA


Co-Authors: 1)Wu X, 2)Yip R, 2)Reuss AE, 2)Swing DA, 2)O'Sullivan TN, 3)Fletcher CF, 1)Hammer III JA, 2)Copeland NG, 2)Jenkins NA

Institutions: 1)Laboratory of Cell Biology, National Heart, Lung & Blood Institute, 2) Mouse Cancer Genetics Program, National Cancer Institute, 3)Genomics Institute of the Novartis Research Foundation



The d, ash, and ln coat color mutations provide a unique model system for the study of mammalian vesicular transport.  All three mutant loci encode genes that are required for the polarized transport of melanosomes, the specialized, pigment-containing organelles of melanocytes, to the neighboring keratinocytes and eventually into coat hairs.  Genetic evidence predicts that these genes function in the same or in overlapping pathways and is supported by biochemical studies showing that d encodes an actin-based melanosome transport motor, MyoVa, while ash encodes Rab27a, a protein that localizes to the melanosome and is thought to serve as the MyoVa receptor.  Here we demonstrate that ln encodes melanophilin (Mlph), a protein that localizes to the melanosome and has sequence similarity to Rab effectors such as granuphilin, Slp3-a, and rabphilin-3A, all of which are thought to be involved in vesicular trafficking.  Like all of these effectors, Mlph possesses two Zn2+-binding motifs and a short region rich in aromatic amino acids that is critical for Rab binding.  However, Mlph does not contain the two Ca2+-binding C2 domains found in these and other proteins involved in vesicle transport, suggesting that it is a member of a novel class of Rab effectors.  Collectively, our data suggest that Mlph functions as part of a transport complex with Rab27a and with MyoVa.


ORAL PRESENTATION             TUESDAY 23 OCTOBER                        11.30HRS – 11.40HRS





Kazuhiro Kitada

Kyoto University



Kyoto 606-8501





Co-Authors: 1)Muraguchi T, 2)Ueno M, 2)Kuwamura M, 3)Guénet JL, 1)Serikawa T

Institutions: 1) Institute of Laboratory Animals, Kyoto University, 2)College of Agriculture, Osaka Prefecture University, 3)Unité de Génétique des Mammiferes, Institut Pasteur



The qc (queue courte) rat mutation was found in the ACI/Pas stock owing to its short tail. Homozygotes also exhibit a circling behavior. The main pathological changes in the CNS are hypoplasia of the cerebellum and hippocampus as well as maldeveloped corpus callosum and choroid plexus. These phenotypes are inherited as an autosomal recessive trait, and the causative gene qc was mapped to rat Chr 13. Interestingly, the segment where the qc locus was assigned shows homology with mouse Chr 1, the region of which includes the dreher (dr) locus. The mutation dr causes a phenotype similar to the one in qc rat. In the mouse the dr gene was recently identified as Lmx1a, a LIM homeobox gene that is expressed in the roof plate along the neuraxis. Isolating the orthologous gene in qc rats, we found that the LIM encoding gene was truncated. By this truncation, its homeobox domain was deleted, resulting in a loss-of-function allele. Moreover analysis for the normal expression pattern showed that Lmx1a was expressed in the CNS not only in embryonic stages, but also in postnatal stages, corresponding to architectural development of the cerebral cortex and cerebellum. Our study supported the hypothesis that Lmx1a plays a crucial role for normal development of the CNS throughout all stages, perhaps via regulation of its downstream molecules such as diffusible signaling molecules or trophic factors.


ORAL PRESENTATION             TUESDAY 23 OCTOBER                        11.40HRS – 11.50HRS




Tomoji Mashimo
Unité de Génétique des Mammifères

Département d'Immunologie

Institut Pasteur
25, rue du Docteur Roux

75724 Paris, France


Co-Author: 1)Simon-Chazottes D, 2)Lucas M, 2)Frenkiel MP, 1)Montagutelli X, 3)Deubel V, 2)Desprès P, 1)Guénet JL

Institution: 1)Unité de Génétique des Mammifères, Institut Pasteur, Paris, 2)Unité des Arbovirus et Virus des Fièvres Hémorragiques, Institut Pasteur, 3)Unité de Biologie des Infections Virales Emergentes, Centre de Recherche Mérieux-Pasteur


The susceptibility of mice to the encephalitogenic effects of flaviviruses is controlled by a major locus (Flv - Chr 5) with at least two alleles. Most inbred strains are homozygous for the recessive allele Flvs and develop fatal encephalitis 7 to 11 days after infection. Most strains derived from recently trapped wild mice are resistant (Flvr). With the aim of identifying the responsible gene by positional cloning, we refined the mapping of the Flv locus within 0.46-cM interval on Chr 5. Combining the data from mouse T31 Radiation Hybrid map within the same interval and merging information from the homologous region of human chromosome 12q, we established a high resolution composite/consensus map of the region. Among several candidate genes in this region, the Oas cluster, encoding the multimember family of a/b interferon induced 2’-5’ oligoadenylate synthetase enzymes, appeared the most likely. Comparing the cDNA sequences of the 3 isoforms (L1, L2 and L3) of the 2’-5’OAS genes, we identified a STOP codon in exon 4 of Oas1 gene in all susceptible strains (8 strains), which is not present in resistant strains (6 strains derived from unrelated wild specimens of the genus Mus). The perfect concordance between the occurrence of a STOP codon with the Flv phenotype of resistance/susceptibility suggests that the truncated form of Oas1 is the cause of the innate susceptibility to flavivirus infection.

ORAL PRESENTATION             TUESDAY 23 OCTOBER                        11.50HRS – 12.00HRS





Pam Grewal

Institute of Genetics

Queen’s Medical Centre

University of Nottingham






Co-Authors: 1)Bittner R, 2)Hewitt J

Institutions: 1)Institute of Neuromuscular Research Department, University of Vienna, 2)Institute of Genetics, Queen’s Medical Centre, University of Nottingham



The autosomal recessive mouse mutation myodystrophy (myd) produces a severe, progressive muscular dystrophy phenotype. High-resolution comparative mapping showed myd  maps to the region of MMU8 displaying homology to HSA19p13 and 22q12. The availability of ordered draft and finished sequence data for these two chromosomes and homologous mouse genomic regions allowed us to use a database-facilitated approach to develop a gene map around the myd locus.


Using this comparative, positional cloning approach, we identified the gene mutated in myd as a glycosyltransferase (Large). In myd, an intragenic deletion removes three exons, producing a frameshift and premature termination of the protein. Given the specific muscular phenotype of myd mice, we hypothesised that the target of this glycosyltransferase might be a component of the dystrophin associated glycoprotein complex (DGC). Mutations in various components of the DGC are associated with muscular dystrophy in humans and rodents. Western blotting data suggests glycosylation of a-DG is markedly reduced in myd mice. We are currently investigating our finding that the abnormal dystroglycan in myd has altered affinity for laminin in skeletal and cardiac muscle. The underglycosylation of this key component of the DGC resulting in a reduced affinity for laminin could possibly be the pivotal event that effects their dystrophic phenotype.


PLENARY SESSION                              TUESDAY 23 OCTOBER                        14.00HRS – 14.30HRS





Monica Justice

Department of Molecular and Human Genetics

Baylor College of Medicine

One Baylor Plaza


TX 77030, USA



Co-Authors: 1)Salinger A, 1)Clark A, 1)Hentges K, 1)Box N, 1)Holdsworth A, 1)Maffucci J, 1)Ross M, 1)Liu B, 2)Behringer B,  3)Bradley, A 1)Justice M

Institutions: 1) Baylor College of Medicine, 2)The University of Texas M.D. Anderson Cancer Center, Houston, TX 2) The Sanger Centre, Hinxton, UK



The post-sequencing challenge is to define the function of genes, and large-scale high-throughput Mouse mutagenesis is one of the best avenues for determining mammalian gene function.  A powerful approach for mutagenesis combines chromosome engineering in embryonic stem cells with phenotype-driven ethylnitrosourea (ENU) mutagenesis. Chromosome engineering allows molecularly defined inversions to be marked with a K14-agouti transgene that confers a dominant yellow coat color, providing a tool called a balancer chromosome for simple mapping, stock maintenance, and genetic screens. 


A Cre-loxP engineered balancer chromosome is being used in a three generation pedigree mating scheme to isolate recessive mutations on mouse Chromosome 11, which shows extensive linkage conservation with human Chromosome 17.  The new ENU-induced mutations include recessive mutations mapping to mouse Chr 11, as well as those segregating genome-wide, demonstrating the dual benefits of this genetic strategy for large-scale mutation isolation.  The phenotypes are relevant to human disease, causing early embryonic and postpartum death, infertility, as well as skeletal, hematopoietic, neurological, urogenital, skin/coat and metabolic defects.  Surprisingly, nearly 2/3 of the new mutations mapping to mouse Chr 11 cause early death or infertility, phenotypes that would be difficult to manage without the use of a balancer.  In future studies, we will generate balancer chromosomes that will cover 25% of the mouse genome.  New mutants, physical maps, and engineered chromosome rearrangements are described at  As new mouse mutations are generated, the comparative sequence information will allow predictions of gene function in the human.   Many of these new mutations will be powerful models of human diseases because mouse and human biological systems are similar.  The accumulation of new mutations will alter views of mammalian gene function and developmental pathways.

ORAL PRESENTATION             TUESDAY 23 OCTOBER                        14.30HRS – 14.45HRS





Karen P Steel

MRC Institute of Hearing Research

University Park






Co-Authors: 1)Steel K, 1)Kiernan A, 1)Erven A, 1)Rhodes C, 2)Tsai H, 2)Hardisty R, 2)Nolan P, 2)Peters J, 2)Brown SDM, 3)Hunter AJ, 4)Ahituv N, 4)Hertzano R, 4)Vreugde S, 4)Avraham K, 5)Fuchs H, 6)Balling R, 5) Hrabé de Angelis M, 7)JL Guénet

Institutions: 1)MRC Institute of Hearing Research, 2)MRC Mammalian Genetics Unit and Mouse Genome Centre, 3)GlaxoSmithKline, 4) Dept of Human Genetics and Molecular Medicine, Sackler School of Medicine, 5)Institute of Experimental Genetics, GSF Research Centre for Environment and Health, 6)Institute of Mammalian Genetics, GSF Research Centre for Environment and Health, 7)Institut Pasteur



Mouse mutants with ear defects are valuable as models for human hereditary deafness, as well as giving us tools for understanding normal cochlear development and function.  53,000 F1 offspring of males treated with N-ethyl-N-nitrosourea (ENU) were screened for balance defects or deafness.  We found 50 new mutations with dominant inheritance, and describe here the first 17 studied.


Eight new mutants map to proximal chromosome 4 and show lateral semicircular canal truncations;  these may be new alleles of the Wheels locus.  Two mutants (Headturner and Slalom) have been identified as Jag1 mutations, and show truncation of the posterior and anterior semicircular canals as well as pattern defects in the organ of Corti, a unique phenotype.  Three mutants have middle ear defects:  Doarad on chromosome 13 has misshapen ossicles, Pardon on chromosome 19 has malformed ossicles combined with supernumary hair cells in the organ of Corti, and Jeff shows a predisposition to middle ear inflammation, the mutation mapping to chromosome 17.  Two mutants have abnormal development of stereocilia bundles, Tailchaser on chromosome 2 and Headbanger on chromosome 7.  Beethoven on chromosome 19 shows progressive postnatal degeneration of hair cells.  Dearisch mutants have progressive hearing loss but the mutation is not yet mapped. 


Eight of these mutants show novel phenotypes, suggesting that the mutagenesis programmes will continue to be a rich resource for investigating auditory function and development. 

(Supported by the MRC, EC contract CT97-2715, Defeating Deafness and GlaxoSmithKline)

ORAL PRESENTATION             TUESDAY 23 OCTOBER                        14.45HRS – 15.00HRS





Heinrich Flaswinkel


Institut für Med. Mikrobiologie

Troger Str. 4a


81675, Germany


Co-Authors: 2)Rathkolb, B, 3) Faerber, C, 1) Servatius, A, 4) Soewarto, D, 5) Kremmer, E, 1)Sandholzer, N, 4) Hrabe de Angelis, M, 6)Balling, R, 2) Wolf, E, 1)Pfeffer, K

Institutions: 1) Institute of Medical Microbiology, Immunology & Hygeine, Universität München, 2)Institute of Molecular Animal Breeding, Gene Center, 3) Ingenium Pharmaceuticals, 4) Institute of Experimental Genetics, GSF Research Center for Environment and Health, 5) Institute of Molecular Immunology GSF Research Center for Environment and Health, 6) Gesellschaft für Biotechnologische Forschung



Analysis of gene function in vivo is currently mostly performed by transgenic insertion, inactivation of the respective gene by homologous recombination in embryonic stem cells and gene trapping. We have complemented this approach by isolation of mouse lines with phenotypic immunological alterations which were produced within the Munich ENU-mouse mutagenesis project.


In this phenotypic approach C3HeB/FeJ are randomly mutagenized using the potent mutagen ethyl-nitrosourea (ENU). ENU is most potent during spermatogenesis and therefore mutagenesis is performed on male mice which are subsequently crossed to female WT mice. Progeny thereof are analyzed for inherited immunological alterations using a panel of immunological parameters.


Out of more than 50 mutant lines isolated with this approach we have analysed two in more detail and mapped the corresponding mutation. Data from this analysis will be presented. <<>> in Microsoft Word 6.0 format with margins of 2.5cm together with all of the details requested in this form.


ORAL PRESENTATION             TUESDAY 23 OCTOBER                        15.00HRS – 15.15HRS





John Schimenti

The Jackson Laboratory

600 Main St.

Bar Harbor, ME








Co-Authors: Libby B, Ward J, Munroe R, Shima N, Bergstrom R

Institutions: The Jackson Laboratory



Mutations affecting gametogenesis and DNA repair were isolated by mutagenesis of mice and ES cells. The germ cell mutations affect stem cell development, meiosis and spermiogenesis.  We have also employed a novel, high throughput screen to detect potentially cancer susceptible mice.  This involves a flow cytometric assay to detect genome instability.  We will present the paradigm and concentrate on one particular mutant that appears to be defective in double strand break repair.



ORAL PRESENTATION             TUESDAY 23 OCTOBER                        15.15HRS – 15.30HRS





Jay Vivian

Department of Genetics

University of North Carolina

Campus Box 7264

Chapel Hill

NC 27599





Co-Authors: Chen Y, Magnuson T

Institutions: Department of Genetics, University of North Carolina




Developing a series of mutations of a gene of interest has largely been limited to invertebrate genetic model systems such as Drosophila and C. elegans. An efficient means of creating an allelic series in the mouse would greatly aid the analysis of complex gene function and in the development of mouse models for disease states. Our laboratory has recently described a technique to mutagenize mouse embryonic stem cells with ENU, while maintaining the capacity for germline contribution of the ES cells. We have extended this mutagenesis strategy to identify mutations in non-selectible genes, using a DHPLC-based mutation detection scheme, focusing on the mouse Smad2 locus. The SMAD proteins are a family of intracellular proteins important in TGF-beta signaling. Smad2 was chosen for this screen due to its diverse functions throughout embryonic and adult development. Several ES cell clones carrying mutations in the coding regions of Smad2 have been identified. Germline analysis of one of the Smad2 mutations shows it is a hypomorphic allele. Mice homozygous for the ENU-induced allele have a variety of embryonic defects, including defects in anterior neurectoderm development. Our work demonstrates that a large number of mutations in nonselectible genes can be readily identified by combining ENU mutagenesis in ES cells with genotype-based mutation detection schemes, and can be extended to any gene of interest.


ORAL PRESENTATION             TUESDAY 23 OCTOBER                        15.30HRS – 15.45HRS





Emma Coghill

MRC Harwell

MRC Mammalian Genetics Unit & UK Mouse Genome Centre,








Co-Authors:  1)Hugill A, 2)Hunter AJ, 1)Cox R, 1)Brown SDM

Institutions: 1)MRC Mammalian Genetics Unit and UK Mouse Genome Centre, 2)GlaxoSmithKline



A gene-driven approach to the identification of enu mutants in the mouse connexin 26 gene


The establishment of parallel DNA and sperm archives from ENU mutagenised mice represents a powerful approach to gene-driven mutagenesis, allowing for rapid DNA screening and swift mutant recovery. A spectrum of single base changes could be recovered, potentially uncovering the full array of functional changes including hypomorphs, antimorphs and neomorphs. We provide proof of principle of this gene-driven approach, reporting the identification of a nonsense mutation in the connexin 26 (Gjb2) deafness gene. GJB2 gene mutations are the most common cause of human non-syndromic deafness. A mouse Gjb2 gene knock-out is embryonic lethal, therefore, no mouse model for this form of deafness is currently available. The gene-driven approach was used in an attempt to recover new alleles at the mouse Gjb2 locus and investigate their phenotype. Screening of 500 archive DNAs, identified one mouse carrying a stop mutation at codon 119, which was recovered by IVF and the Gjb2 stop mutation confirmed. Heterozygous mice showed normal hearing and intercrosses produced no homozygous progeny, demonstrating embryonic lethality. We have begun to apply a high-throughput gene-driven approach for the recovery of mouse mutants in a variety of cochlear-expressed genes, including a search for further alleles of the Gjb2 gene.

PLENARY SESSION                  WEDNESDAY 24 OCTOBER                              09.00HRS – 09.30HRS





Ian Wilmut

Roslin Institute



EH25 9PS











The production of cloned offspring by nuclear transfer is still a very inefficient process with only 1-4% of reconstructed embryo developing to become viable offspring. This outcome is similar for all 5 species in which somatic cell nuclear transfer has been achieved, regardless of species, donor cell type or method of nuclear transfer. However, the precise stage when loss occurs may vary between different cell types. The low overall efficiency is the cumulative result of failure at all stages of development through from cleavage to the post-natal period. The unusually high loss, spread throughout development is assumed to reflect the inappropriate expression of a number of genes whose lethal effects are exerted at different stages of development. Both direct and indirect evidence is accumulating of errors in expression of imprinted genes, but this does not indicate that these genes are the only ones to be expressed inappropriately. At present very little is known of the molecular mechanisms involved in either the "reprogramming" of gene expression or the causes of death of cloned embryos. Significant improvements in the efficiency of nuclear transfer may depend upon modifications to the procedure to enhance the ability of the oocyte cytoplasm to remodel the chromatin of the transferred nucleus and achieve appropriate developmental regulation of gene expression.



PLENARY SESSION                              WEDNESDAY 24 OCTOBER                  09.30HRS – 10.00HRS





Kevin Eggan

Whitehead Institute for Biomedical Research

Department of Biology

Massachusetts Institute of Technology

9 Cambridge Center


MA 02142, USA



Co-Authors: 1,2)Eggan K, 1,2)Humpherys D, 3)Akutsu H, 1)Hochedlinger K, 1)Rideout III W, 3)Yanagimachi R, 1,2)Jaenisch R

Institution: 1)Whitehead Institute for Biomedical Research, 2)Massachusetts Institute of Technology, 3)John A. Burns School of Medicine, University of Hawaii



Cloning by nuclear transfer is an inefficient process with most clones dying prior to birth and survivors often displaying growth abnormalities. In an effort to correlate gene expression with survival and fetal overgrowth, we have examined imprinted gene expression and X-chromosome inactivation in cloned mice.  Although normal X-inactivation was recapitulated in cloned mouse embryos, variation in imprinted gene expression was observed in most mice cloned from embryonic stem (ES) cells, even in those derived from ES cells of the same sub-clone. Many of the cloned animals survived to adulthood despite improper expression of several imprinted genes, indicating that mammalian development may be rather tolerant to epigenetic abnormalities. Our results suggest that epigenetic processes ordinarily transpiring after fertilization may occur normally in cloned mice but that epigenetic information customarily imposed on the genome during gametogenesis may not be recovered if lost during differentiation or aging of the donor cell nucleus.  Therefore, even apparently normal cloned animals may have subtle abnormalities in gene expression.



ORAL PRESENTATION             WEDNESDAY 24 OCTOBER                  10.00HRS – 10.15HRS






Jesse Mager

University of North Carolina - Chapel Hill

Magnuson Lab

102 Mason Farm Rd. rm 11-116

Chapel Hill






Co-Authors: Jainbo Wang, Terry Magnuson

Institutions: University of North Carolina at Chapel HIll



In mammals, dosage compensation of X-linked genes is achieved by transcriptional silencing of one X chromosome in the female.  This process, called X-inactivation, is usually random in the embryo proper.  However, in marsupials and the extraembryonic region of the mouse, X-inactivation is imprinted: the paternal X chromosome is preferentially inactivated while the maternal X is always active.  Having more than one active X is deleterious to extraembryonic development in mouse.  Here we show that eed (embryonic ectoderm development), a member of the Polycomb-group genes, is requred for primary and secondary giant cell development in female embryos.  Results from mice carrying a paternally inherited X-linked GFP transgene implicate eed in the stable maintenance of imprinted X-inactivation in extraembryonic tissues.  Based on the recent finding that Eed interacts with histone deacetylases, we suggest that  one component of this maintenance activity involves hypoacetylation of the inactivated paternal X in extraembryonic tissues.


ORAL PRESENTATION             WEDNESDAY 24 OCTOBER                  10.15HRS – 10.30HRS






Bernadette Holdener

State University of New York at Stony Brook

Dept of Biochemistry & Cell Biology

Stony Brook, NY







Co-Authors: Lee L, Zhang L, Brown K, DeRossi C

Institutions: Dept of Biochemistry and Cell Biology, State University of New York at Stony Brook



Mouse embryos lacking the mesoderm development (mesd) deletion interval fail to form mesoderm or a primitive streak and do not survive to term. Characterization of gene expression in mesd mutant embryos demonstrates that failure to form a primitive streak results from a defect in the proximal epiblast. Further, anterior visceral endoderm and epiblast gene expression is expanded, demonstrating that the establishment of the A/P axis is dependent upon antagonism between head and trunk signals. Transgenic rescue of the mesd phenotype demonstrates that the mesd gene(s) and regulatory elements are entirely contained within a 75 kb BAC clone. Comparative sequence analysis identifies two closely linked candidate genes, Mesdc1 and Mesdc2. Human homologues map to a conserved region on Chromosome 15q and cosegregate with a locus for autosomal dominant nocturnal frontal lobe epilepsy (ENFL2). Both candidate genes are ubiquitously expressed.  Protein prediction and immunofluorescence data suggest that Mesdc1 is a nuclear and cytoskeleton-associated protein and that Mesdc2 is a novel ER resident protein. Significantly, the Mesdc2 transgene rescues the early polarity defects in deletion homozygotes.  We are currently investigating the role of Mesdc1 later in development.  These results suggest that Mesdc2 plays an important role in gastrulation, potentially acting as a molecular chaperone to regulate folding or secretion of critical signaling pathway molecules.


ORAL PRESENTATION             WEDNESDAY 24 OCTOBER                  11.00HRS – 11.15HRS





Karen Fitch

Stanford University

279 Campus Drive

Beckman Center, B281

Stanford, CA







Co-Authors:  1)McGowan KA, 1)Van Raamsdonk CD, 2)Fuchs H, 2)Hrabé de Angelis M, 1)Barsh GS  

Institutions:  1)Stanford University, 2)Institute of Experimental Genetics, GSF



Among nearly 100 mouse coat color mutations collected over the past century, those that cause dark hair by affecting Agouti signaling have identified key components of an important paracrine signaling pathway. Large-scale mutagenesis projects offer the potential to further the application of forward genetics, to determine if the collection of coat color mutations has been saturated, and potentially, to discover new genes that control melnaocyte homing to the intra vs. inter-follicular microenvironment.


We have characterized ten new ENU-induced dominant mutations that cause dark coat color (Dcc) or dark skin (Dsk), and therefore are likely to interfere with pigment switching or follicular homing of melanocytes, respectively. Each maps to a different position, with one of the Dsk and two of the Dcc mutants likely to represent new alleles of Egfr, Agouti, and Sox18. Double mutant studies between Sox18 and Agouti or Mc1r reveals that Sox18 is genetically upstream of the other two mutants, suggesting that Sox18 directly regulates Agouti transcription.


Using a melanocyte specific lacZ reporter transgene reveals two distinct classes of Dsk mutants, distinguished by whether melanocytes accumulate beneath the dermal-epidermal junction or within the interfollicular epidermis. In both cases, intrafollicular pigmentation is normal or increased, suggesting that instructive cues for follicular homing are acquired by developing pigment cells prior to visible pigment production.

ORAL PRESENTATION             WEDNESDAY 24 OCTOBER                  11.15HRS – 11.30HRS





Teresa Gunn

Department of Biomedical Sciences

T4 018 VRT

Cornell University

Ithaca, NY







Co-Authors: He L, Barsh GS

Institutions: Stanford University



Mutations of the mouse Attractin (Atrn; formerly mahogany) gene, which encodes a widely expressed type I transmembrane protein containing multiple plexin and EGF repeats, were originally recognized because they suppress Agouti pigment type-switching. Surprisingly, homologs of Atrn are found in fruit flies and nematodes, even though Agouti and/or Agouti-related protein are found only in vertebrates. Insight into this apparent paradox now comes from studies of different Atrn alleles, in which we find hyperactivity, abnormal myelination, and widespread CNS vacuolation whose severity is inversely proportional to the amount of normal Atrn mRNA. Neurodegeneration is also caused by the classical coat color mutation mahoganoid, whose effects on pigment type-switching and epistatic interactions are nearly identical to Atrn. We have now positionally cloned mahoganoid, and find that it predicts a 54 kD intracellular protein whose pattern of evolutionary conservation is similar to Atrn. Taken together with studies of invertebrate Atrn homologs, these results suggest that Atrn and mahoganoid function as closely linked components of a metazoan pathway required to maintain normal neuronal architecture, that mahoganoid lies downstream of Atrn, and that this pathway was recruited during vertebrate evolution by the Agouti-melanocortin system to control coat color.


ORAL PRESENTATION             WEDNESDAY 24 OCTOBER                  11.30HRS – 11.45HRS





Bruce Herron

Harvard Medical School

Division of Genetics

Brigham & Womens Hospital

75 Francis Street

Boston MA

02115-6110, USA



Co-Authors: 1) Rao C, 1) Pacella L, 1) Semsarian C, 1) Seidman C, 3) Stubbs L, 2) Millar S, 1) Beier DR    

Institutions: 1) Brigham and Women's Hospital, Harvard Medical School, 2) Dept of Dermatology, University of Pennsylvania, 3) Human Genome Center, Lawrence Livermore National Laboratory



We have previously described a spontaneous mutation that affects the development of skin and heart in mice called waved 3(wa3). While wa3 is similar to wa1 and wa2 mice, it also includes a perinatal onset cardiomyopathy with regions of cardiomyocyte necrosis. This disorder progresses with age, resulting in transmural ventricular fibrosis and dilated cardiomyopathy.


We have identified a mutation in a partially characterized gene previously called RAI. Comparative sequencing has revealed a 15 base pair deletion in wa3 mice that removes a splice donor site. This mutation results in a truncation of the translated gene product.


Characterization of this gene uncovered additional 5‘ sequences. Database analysis indicates it is most similar to P53 binding protein 2, a gene previously shown to interact with P53, Bcl-2, and NFkB. In-situ analysis has indicated that this gene is expressed in epidermal cell types. Cardiovascular expression was also detected in the atria, ventricles, and in the arterial endothelial cells. 


We propose that a loss of function mutation in 53bprp results in abnormal epidermal cell growth. This results in open eyes at birth and may contribute to the hair phenotype. Furthermore, the focal regions of cardiac necrosis seen in these mice are consistent with an ischemic injury. Abnormalities in endothelial cell growth in coronary arteries may result in ischemia, leading to cardiomyocyte death.



ORAL PRESENTATION             WEDNESDAY 24 OCTOBER                  11.45HRS – 12.00HRS





Kenro Kusumi

U.Penn/Children's Hospital of Philadelphia


3615 Civic Center Blvd

Philadelphia PA

19104, USA




Co-Authors: 1)Dunwoodie, S., 2)Krumlauf, R.

Institutions: 1)The Victor Chang Cardiac Research Institute, 2)The Stowers Institute for Biomedical Research



Genes in the Notch signaling pathway play key roles in embryonic patterning.  We have previously shown that the pudgy (Dll3pu) mouse and human spondylocostal dysostosis (SD, MIM277300) are mutations in the Notch ligand, Delta-like 3.  Both Dll3pu and SD display extensive vertebral and rib malformations, resulting from defective segmentation of the paraxial mesoderm. The Notch modulator, lunatic fringe (Lfng), is dynamically expressed in the presomitic mesoderm during each somite cycle.  In Dll3pu mutants, altered Lfng cycling is detected at 8.5 dpc, and completely disrupted at 9.5 dpc.,  This suggests that Dll3 is required for maintenance of Lfng cycling, and that defects in the somite clock underlie spondylocostal dysostosis.


Dll3 is also expressed throughout neural patterning. In particular, we have identified dynamic expression of Dll3 between 9.0 and 11.5 dpc in hindbrain development.  Interestingly, SD patients exhibit facial torticollis, a defect potentially related to neural development, but do not show overt signs of mental retardation or gross neurological impairment.  Dll3 and Notch pathway genes are expressed in laminar domains during neural development, and in Dll3pu mutants, we have characterized ependymal laminar defects.  Examination of SD patients and analysis of Dll3pu mice are ongoing, to define Dll3 function in neural patterning.


ORAL PRESENTATION             WEDNESDAY 24 OCTOBER                  12.00HRS – 12.15HRS





Galina N. Filippova

Fred Hutchinson Cancer Research Center

1100 Fairview Ave. N, C2-023

PO box 19024

Seattle, WA




Co-Authors: 1) Galina N.Filippova, 1) Sara Fagerlie, 1) James M. Moore,1) Kevin P. Foley, 2) Joanne Whitehead 1) Chris Kemp, 2) Rolf Ohlsson, 3) Victor Lobanenkov

Institutions: 1)Human Biology Division, Fred Hutchinson Cancer Research Center, 2)Department of Development and Genetics, Uppsala University, 3) National Institute of Allergy and Infectious Diseases, National Institutes of Health




The CTCF gene encodes an evolutionary conserved, 11 zinc finger (ZF) transcription factor that is involved in different aspects of gene regulation including promoter activation or repression, hormone-responsive gene silencing, methylation-dependent chromatin insulation and genomic imprinting. Somatic missense mutations within the CTCF 11ZF DNA-binding domain have been observed in breast, prostate and Wilms’ tumors. These mutations  selectively change rather than completely disrupt CTCF function. To study the role of CTCF in cancer and normal development we isolated the mouse CTCF gene and generated a null allele via homologous recombination in embryonic stem cells by deleting all coding exons of CTCF. Heterozygous mice were vital and showed 2-3 fold increase in tumor development in comparison to their wild type littermates. However, when intercrossing heterozygotes, homozygous CTCF mice were selectively absent from the offspring. Furthermore, we could not detect CTCF -/- embryos even at day 3.5 of the development. In vitro growth experiments with early preimplantation stage embryos obtained from heterozygous intercrosses did not reveal any CTCF -/- embryos either.


These results establish the essential role of the CTCF protein in mammals and in cellular viability and are consistent with the notion than only subtle CTCF mutations are found in tumors.

ORAL PRESENTATION             WEDNESDAY 24 OCTOBER                  12.15HRS – 12.30HRS





Susan Ackerman

The Jackson Laboratory

600 Main Street

Bar Harbor, ME







Co-Authors:  Park C, Edgar JH, Longo-Guess CM

Institution:  The Jackson Laboratory



Disruption of the aN-catenin gene in cerebellar deficient folia mutant mice results in cerebellar and hippocampal lamination defects


Cerebellar deficient folia (cdf) is a recently identified spontaneous mouse mutation causing ataxia and cerebellar abnormalities. Our results demonstrate that the cdf mutation disrupts the positioning of a specific subset of Purkinje cells during embryonic development, a process known to depend on the extracellular protein reelin, which also regulates migrations in the hippocampus and neocortex.  Analysis of cdf«ROSA26 chimeric mice demonstrated that the cdf mutation is intrinsic to Purkinje cells.  Consistent with a role for the cdf gene product in reelin signaling, we observed lamination defects in the hippocampus of cdf mutant mice, although neocortical defects were not seen. In addition, ectopic Purkinje cells in cdf/cdf mice express an increased level of Dab1 protein, as previously observed in mice with mutations in genes in the reelin signaling pathway.


We have fine mapped the cdf mutation to a 0.28 cM interval on Chromosome 6. A deletion in cdf mutant mice that removes the last exon of the aN-catenin (Catna2) gene and continues proximally into adjacent DNA was identified. Cerebellar and hippocampal structure was rescued in cdf mutant mice expressing the Catna2 cDNA. Our findings suggest that Catna2 and its associated cadherins function in positioning neurons in the developing cerebellum and hippocampus.


PLENARY SESSION                              WEDNESDAY 24 OCTOBER                  14.30HRS – 15.00HRS





Alain Balmain

UCSF Comprehensive Cancer Center

234D Sutter Street

Box 0875, Room S227

San Francisco

CA 94143










The risk of developing cancer is due to a combination of environmental factors and inheritance of high or low penetrance cancer modifier genes.  Mouse models using interspecific crosses between Mus spretus, which is relatively cancer resistant, and inbred strains of Mus musculus, which are relatively cancer susceptible have been used to map cancer modifier genes.  We have mapped 13 mouse skin cancer modifier loci using this approach.  One locus, Skts13, on distal chromosome 2, shows strong linkage for skin tumor susceptibility with a p-value of 5.5 x 10-7 (Wilcoxon rank sum test).  In addition, in another mouse cancer model, approximately 63% of lymphomas from F1 mice show tumor modifier gene maps to this locus in mouse.  To narrow down the locus, we used a combination of linkage analysis together with haplotype analysis in a backcross between outbred Mus Spretus and NIH/Ola.  Linkage analysis localized the QTL to an interval of approximately 20cM, but a common haplotype spanning approximately 1.5cM was present only in the haplotypes associated with papilloma resistance.  The orthologous region of the human genome, 20q13.2, is amplified in a variety of tumor types including breast cancer, colon cancer and ovarian cancer.  A series of candidate SNP’s in four genes in this interval are being tested in association studies of human breast cancer cases and controls.  The results suggest that this combination of linkage and haplotype analysis can circumvent many of the problems associated with the fine mapping of tumour modifier loci.



ORAL PRESENTATION             WEDNESDAY 24 OCTOBER                  15.00HRS – 15.15HRS






Ayo Toye

Diabetes QTL and Modifier Loci Group

Medical Research Council

Harwell Nr Didcot


OX11 0RD





Co-Authors: Goldsworthy M, Bentley L, Moir L, Ritson D, Haynes A.E, Mijat V, Cox R.

Institutions: Diabetes QTL and Modifier Loci Group, Medical Research Council



Our aim is to identify novel genes that contribute to glucose regulation and its pathologies including impaired glucose tolerance and type 2 diabetes in humans and other mammals.


Our approach uses a whole genome mapping approach to identify mouse loci that cosegregate with and by implication determine two biochemical phenotypes (plasma glucose and plasma insulin levels) that are perturbed in diabetes. We surveyed 4 inbred mouse strains  (C57BL/6J, C3H, DBA/2 and BALB/C) for variation in glucose tolerance during an intraperitoneal glucose tolerance test. We observed larger strain differences in male than female mice. Male C57BL/6J mice were significantly less glucose tolerant than other strains (N=20, P<0.000001). In order to map the underlying genes, we produced an F2 intercross cohort of 700 male and female mice from reciprocal crosses between C57BL/6J and C3H parents then their F1 progeny. F1 males (N=75) were significantly (P<0.0001) different from C57BL/6J (N=20) but not C3H parents (N=20). Glucose tolerance was positively skewed in F2 males (N=350), and about 25% appeared like their C57BL/6J grandparents. To date, we have genotyped mice in the top 17.5% of the F2 glucose tolerance phenotype distribution, and have identified 4 distinct loci that contribute significantly (P<0.00001) to glucose tolerance. We are currently genotyping the remaining 82.5% of the phenotype distribution in order to exploit their full information content.


ORAL PRESENTATION             WEDNESDAY 24 OCTOBER                  15.15HRS – 15.30HRS





Sarah Lloyd

MRC Prion Unit

Department of Neurogenetics

Imperial College School of Medicine at St Mary's

Norfolk Place


W2 1PG


Co-Authors: 2)Onwuazor O, 2)Uphill J, 2)Beck J, 3)Farrall M, 4)Targonski P, 1)Collinge J, 1)Fisher E

Institutions: 1)MRC Prion Unit and Dept of Neurogenetics, Imperial College of Medicine at St Mary’s, 2)MRC Prion Unit, 3)Dept of Cardivascular Medicine, University of Oxford, The Wellcome Trust Centre for Human Genetics, 4)Dept of Internal Medicine, Mayo Clinic



Prion diseases are fatal neurodegenerative disorders which include sheep scrapie, bovine spongiform encephalopathy (BSE) and its human form, variant Creutzfeldt-Jakob disease (vCJD).  Prion diseases have prolonged incubation times which are known to be affected by amino acid polymorphisms in the prion protein in humans, mice  and sheep.  In mice, Prnpa (Leu-108, Thr-189) and Prnpb (Phe-108, Val-189) are associated with short and long incubation times, respectively. However, studies with inbred lines of mice, all carrying Prnpa , show significant differences in incubation time suggesting that other genes contribute to this variation.  To identify these loci we generated an F2 intercross between two Prnpa strains of mice with significantly different incubation times when inoculated intracerebrally with Chandler/RML scrapie prions.  A whole genome screen was carried out in two stages using a total of 1009 F2 animals.  Interval mapping identified three highly significantly linked regions on chromosomes 2, 11 and 12 and composite interval mapping suggests that each of these regions includes multiple linked quantitative trait loci.  Suggestive evidence for linkage was obtained on chromosomes 6 and 7.  We will also present data from another smaller F2 intercross using the same mouse strains but challenged with a different strain of prion: mouse adapted BSE.



ORAL PRESENTATION             WEDNESDAY 24 OCTOBER                  15.30HRS – 15.45HRS





Leonid Bystrykh

State University of Groningen

Ant.Deusinglaan 1


9713 AV

The Netherlands



Co-Authors:  1) Bystrykh L, 1)Weersing E, 2)Geiger H, 3)Ivanova N, 2)Van Zant G, 3)Lemischka I, 1)Vellenga E, 1)de Haan G

Institutions: 1)State University of Groningen, 2)University of Kentucky, 3)Princeton University



We have previously shown that hematopoietic stem cells from DBA/2 and C57BL/6 mice differ in their cell cycle activity We have mapped a major quantitative trait locus (QTL) associated with variation in cell proliferation to chromosome 11. In order to detect candidate genes causing these differences we performed subtractive hybridizations and gene array assays using cDNA from highly purified stem cells from both strains. The results revealed an unexpectedly high frequency of differently expressed genes (ESTs) mapping to chromosome 11. Intriguingly, detailed mapping of those transcripts showed that they occured in three distinct clusters. The largest cluster colocalizes exactly with the stem cell cycling QTL. Using newly derived backcross panels and the BXD recombinant inbred database, we found the three clusters of differentially expressed transcripts to coincide with regions of increased frequency of meiotic recombination. In addition, sequence analysis of selected genes in those clusters demonstrated profound sequence variations. Taking into account the high functional heterogeneity of the clustered genes we postulate that the QTL underlying variation in stem cell proliferation may not be due to a single gene but rather may involve a genomic disparity affecting a higher order of chromosomal organization. These differences lead to increased rates of meiotic recombination and aberrant expression of a large cohort of clustered genes. It is tempting to speculate that many QTLs may result from anomalous expression of a collection of highly linked genes. In addition, a similar model may apply for tumorigenic chromosomal translocations.


ORAL PRESENTATION             WEDNESDAY 24 OCTOBER                  15.45HRS – 16.00HRS





Roger Reeves

Johns Hopkins University Schl. Med.

Dept. of Physiology


MD 21205




Co-Authors: 1&2)Richtsmeier J, 3)Zumwalt A, 4)Carlson E, 4)Epstein C, 5)Reeves R

Institutions: 1)Dept of Anthropology, The Pennsylvania State University, 2)Centre for Craniofacial Development and Disorders, The Johns Hopkins Hospital, 3)Department of Cell Biology and Anatomy, Johns Hopkins University School of Medicine, 4)Dept of Pediatrics, University of California, 5)Department of Physiology, Johns Hopkins University School of Medicine



Trisomy for chromosome 21 (Chr 21) is the most complex genetic disorder compatible with human survival.  It has profound effects on development that result in a constellation of phenotypes known as Down syndrome (DS).  Distinctive craniofacial dysmorphology is among the few features common to all individuals with DS.  The characteristic DS facies result primarily from maldevelopment of the underlying craniofacial skeleton and mandible.  The Ts65Dn mouse, which has segmental trisomy 16 producing dosage imbalance for about half the genes found on human Chr 21, exhibits specific skeletal deformations corresponding directly to the craniofacial dysmorphogenesis in DS (Richtsmeier et al., 2000).  Here we demonstrate that Ts1Cje mice, which are at dosage imbalance for about 70% of the genes triplicated in Ts65Dn, demonstrate a very similar pattern of anomalies in the craniofacial skeleton. However, one character of Ts65Dn mice, a broadening of the cranial vault contributing to brachycephaly, is not seen in Ts1Cje mice.  These observations independently confirm that dosage imbalance for mouse genes orthologous to those on human Chr 21 has corresponding effects in both species. The subtle differences in the craniofacial phenotypes of Ts1Cje and Ts65Dn mice have implications for elucidation of the mechanisms by which this aneuploidy disrupts development.    


ORAL PRESENTATION             WEDNESDAY 24 OCTOBER                  16.00HRS – 16.45HRS





David Threadgill

University of North Carolina

Dept of Genetics, CB#7264

Lineberger Rm 11-109

Chapel Hill

NC 27599



Co-Authors: 1)Threadgill D 2) Airey DC, 2) Lu L, 3) Manly KF, 2) Williams RW

Institutions: 1)Dept of Genetics, University of North Carolina at Chapel Hill, 2)Center of Genomics and Bioinformatics, University of Tennessee, 3)Dept of Molecular and Cell Biology, Roswell Park Cancer Institute



RIX mapping is a new complex trait mapping method that extends the power of existing RI strains. RIX mapping involves generating F1 hybrids by intercrossing all, or a subset of, parental RI strains to generate a genetically complex, but reproducible population of n(n-1)/2 unique individuals, where n is the number of parental RI strains. We have performed a theoretical analysis by simulating the mapping of QTLs using actual CXB and BXD genotypes. Furthermore, we have empirically validated the RIX approach using all 13 CXB parental strains, 78 non-reciprocal F1 RIX hybrids, and 4 reciprocal RIX hybrids to detect and map genes controlling body weight. RIX mapping has many significant advantages over existing mapping populations including: 1) greatly increased power to detect QTLs, particularly those with complex epistatic interactions; 2) improved mapping resolution; 3) mapping of additive, dominance, and epistatic effects; 4) analysis of multiple identical heterogeneous genomes minimizes non-genetic variance; 5) no genotyping is required; 6) data are cumulative enabling multivariant approaches; 7) parental effects can be mapped through analysis of reciprocal RIX crosses; and 8) phenotypic variances are smaller than inbred genomes. The major disadvantages of RIX mapping include: 1) a paucity of parental RI strains; and 2) existence of strong statistical association between non-syntenic chromosomal intervals in RI strains.