PLENARY PRESENTATION

MONDAY OCTOBER 18

2.15pm – 2.45pm

Contemporary approaches to extract gene function from the mouse genome

Bradley A

Wellcome Trust Sanger Institute, Cambridge,United Kingdom

The human and mouse genome projects have provided a product of extraordinary scientific value.  Of the estimated 25,000 genes, most have not been examined experimentally.

In parallel to the sequencing effort, enormous progress has been made in our ability to examine gene function in mice by generating and analyzing mutations.  To accelerate genetic analysis further we have been using chromosome engineering to develop genetic resources based on strategies originally developed in Drosophila, balancer chromosomes.  Our mouse balancers are tagged with coat colour genes enabling recessive mutations to be rapidly identified and maintained

The availability of the mouse genome sequence has enabled us to develop indexed reagents, allowing systematic approaches for generating knockouts and chromosomal rearrangements.  These reagents facilitate the rapid construction of knockouts.  100,000 vectors are displayed in the Ensembl genome browser (www.ensembl.org) under the DAS source MICER.  

Finally, I will describe a strategy which enables recessive genetic screens to be conducted in cultured ES cells, by-passing the time consuming step of producing and inter-crossing mutations in mice.  We have exploited the high rate of mitotic recombination in Blm-deficient ES cells to generate a genome wide library of homozygous mutant cells from heterozygous mutations induced with a revertible gene trap retrovirus.  We have screened this library for cells with defects in mismatch repair and demonstrate the recovery of cells with homozygous mutations in known and novel components of this system. The combination of insertional mutagenesis in Blm-deficient ES cells opens a new approach for phenotype based recessive genetic screens in ES cells.

PLENARY PRESENTATION

MONDAY OCTOBER 18

4.15pm – 4.45pm

RNAi DIRECTED THERAPIES FOR HUMAN VIRAL HEPATITIS INFECTION

Grimm D, McCaffrey A, Nakai H, Ehrhardt A, Marion P Kay MA

Depts of Pediatrics and Genetics, Stanford University

RNA interference has great potential for the treatment of viral infections.  Our laboratory has focused on liver directed transfer of siRNAs and the expression of short-hairpin RNAs from the livers of living animals to inhibit transgene expression.  To do this, we injected either the short-inhibitory RNAs or plasmids expressing the short-hairpins along with a plasmid expressing the luciferase transgene, luciferase-HCV fusion gene, or HBV genome into the livers of mice by a hydrodynamic infusion procedure.  We established that:  (1)  RNAi  was functional in whole mammals, (2) Expressed shRNAs were efficient at down-regulating gene expression, suggesting that RNAi-mediated therapies can be adapted to a gene therapy approach, (3) RNAi against hepatitis C virus sequences were inhibitory, and (4) shRNAs directed against hepatitis B virus inhibited viral HBV DNA replication in the liver. To further pursue a gene therapy approach against hepatitis virus infection in animal models, a transgenic mouse that contains and replicates the HBV genome was used for further study. shRNAs expressed from  pol III  promoters were cloned into first generation or helper dependent (adenoviral gene deleted)  adenovirus  and  AAV pseudotype 8  recombinant viral vectors,  and then infused into the HBV transgenic mice.  These three vector systems were selected because they are capable of transducing nearly 100% of the hepatocytes in vivo. Using recombinant  AAV-8 vectors and carefully selecting the appropriate shRNA expression cassette, safe and long-term elimination of the HBsAg from HBV transgenic mice was observed.  HBV DNA titers, liver immunohistochemistry for HBcAg, northern and Southern blots for quantification of HBV nucleic acids are being determined.  Moreover, these studies have revealed important safety considerations. The potential therapeutic value of each of these vectors approaches will be discussed.

VERNE CHAPMAN MEMORIAL LECTURE

MONDAY OCTOBER 18

7.30pm – 8.30pm

25 YEARS OF TRANSGENIC MICE: THE METALLOTHIONEIN PERSPECTIVE

Palmiter RD

Howard Hughes Medical Institute and University of Washington, Seattle, United States

Our venture into mouse genetics started with the cloning of the mouse metallothionein I (Mt1) gene in 1979.  We showed that metals, glucocorticoids, and cytokines induced transcription of the Mt1 gene.  The Mt locus is amplified in cells resistant to cadmium toxicity, but inactivated by DNA methylation in other cell lines.  In pursuit of understanding the transcriptional regulation by metals, we fused the 5’ flanking region of Mt1 gene to the coding region of HSV-thymidine kinase to produce MT-TK and sent it to Ralph Brinster for injection into mouse eggs.  These experiments along with mutational analysis were used to identify the metal response elements that allow induction by metals.  More importantly, they launched a 15-year collaboration with Ralph Brinster making and studying genetically modified mice.  The MT-TK transgenic mice were the first that demonstrated expression, regulation, methylation and germline transmission of a functional transgene.  Input from Vern Chapman and colleagues at Roswell Park in 1981 contributed to the “big mouse” experiments that demonstrated how transgenic technology could alter the phenotype of mice.  While endeavouring to understand how enhancers work, we discover that SV40 T-antigen is a potent oncogene.  Meanwhile, the odd transmission of MT-TK transgene in one line of mice, in which males were fertile but never transmitted the transgene, let to the discovery of HSV-TK toxicity and transgene deletion by homologous recombination.  Later, while studying knockout mice lacking Mt1 & Mt2, we learned that these genes are not essential but do protect against cadmium toxicity and oxidative damage.  Vern Chapman’s research allowed us to understand the female lethality of mice lacking Mt1&2 and the Menkes copper transporter, Atp7a.  We maintain an interest in Mt gene regulation and function, but the lab currently uses an array of contemporary genetic techniques to explore the neurobiology of motivation, appetite, and learning.  But, we would not have gotten there without our foray into Mt genetics. 

PLENARY PRESENTATION

TUESDAY OCTOBER 19

8.30am – 9.00am

RECOMBINEERING: A POWERFUL NEW TOOL FOR MOUSE FUNCTIONAL GENOMICS

Copeland NG 1, Warming S1, Liu P2, Jenkins NA1

1 National Cancer Institute, Frederick, United States, 2 The Wellcome Trust Sanger Institute, Cambridge, United Kingdom

Highly efficient phage-based Escherichia coli homologous recombination systems have recently been developed that enable genomic DNA cloned into plasmids, PACs or BACs to be modified or subcloned directly in E. coli without the need for restriction enzymes or DNA ligases.  This new form of chromosome engineering, termed recombineering, is highly efficient and greatly decreases the time it takes to create transgenic or knockout mouse models by conventional means.  Recombineering also facilitates many kinds of genomic experiments that have been difficult, if not impossible, to carry out before, and should enhance functional genomics studies by providing better mouse models and a more refined genetic analysis of the mouse genome.  In my talk, I will describe some of the many uses of recombineering for mouse functional genomic studies.  I will also describe a new recombineering strain we have developed that makes it possible to efficiently introduce double-strand DNA targeting cassettes (i.e., DNA containing epitope tags or loxP sites) or mutations (i.e. point mutations or small deletions) into cloned DNA by positive-negative selection, thus obviating the need for a linked selectable marker.

PLENARY PRESENTATION

TUESDAY OCTOBER 19

2.00pm – 2.30pm

SEQUENCE OF THE MOUSE Y CHROMOSOME

Alfoldi JE1, Skaletsky H1, Graves T2, Minx P2, Wilson RK2, Rozen S1, Page DC1

1 Howard Hughes Medical Institute, Whitehead Institute, and Department of Biology, Massachusetts Institute of Technology, Cambridge, United States, 2 Genome Sequencing Center, Washington University School of Medicine, St. Louis, United States

Y chromosomes have no partner with which to cross-over during meiosis, and so were once thought to be doomed to degeneration and gene loss. However, the recent discovery of palindromes on the human Y chromosome has given us hope of the continued survival of Y genes, since palindromes could be protecting their genes by gene conversion.

The sequencing of the mouse Y chromosome was begun with the hope that it too has palindromes, and that these palindromes could be studied more easily in a more tractable organism. Now that some mouse Y sequence has been produced, we can see that this chromosome contains not only palindromes, but a remarkable feature never seen before in any genome – the Huge Repeat array.

95% of the sequence of the mouse Y consists of a single large repeat – approximately 200 copies of a 500 kb repeat unit, all found on the Y’s long arm. The repeat units are 90-99.9% identical, euchromatic, internally repetitive and contain two gene families, Ssty (Y-linked spermiogenesis specific transcript) and Sly (Sycp3-like, Y-linked). This Huge Repeat array is unlike anything found on the palindrome-containing human Y chromosome or the heterochromatin-filled Drosophila Y chromosome. It is also extremely unusual within the mouse genome as this genome has very low segmental duplication, especially when compared to the human genome. The short arm of the mouse Y chromosome does not contain Huge Repeat sequence. However, it does contain palindromes, similar to the human Y, as well as several spermatogenesis-related genes.

PLENARY PRESENTATION

TUESDAY OCTOBER 19

4.00pm – 4.30pm

Functional Genome Analysis

O’Brien TP

The Jackson Laboratory, Bar Harbor, ME, United States

Understanding the functional content of the mammalian genome is a major goal of the mouse genetics community. Numerous computational and experimental approaches are being used to understand gene activity and network level interactions. One approach, random mutagenesis, has long served the efforts of the geneticist to build the link between genomics and biology. Mutagenized genomes harboring chemically induced base-pair changes represent a rich source of novel genetic variation and can now be produced using either whole-animals or ES cells. This variation can be tapped using direct sequencing or phenotype-based screens. Phenotypic screens can be designed to survey the entire genome, a defined chromosomal region or sensitized to identify environmental or genetic interactions. One desirable feature of such screens is to isolate mutations as efficiently as possible. Toward this end, genetic schemes often use chromosomal aberrations and markers to facilitate the recovery of mutations. A growing collection of reagents, such as chromosomal deletions and inversions along with coat color and GFP-based chromosomal tags, permits the design of a variety of increasingly efficient genetic screens. These chromosomal reagents as well as ES cell-based and direct sequencing methods open new opportunities for using random mutagenesis to generate allelic series, construct genetic interaction networks, and build regional views to understand the functional genomic landscape.

PLENARY PRESENTATION

TUESDAY OCTOBER 19

5.45pm – 6.15pm

IDENTIFYING COMPONENTS OF GROWTH FACTOR SIGNALLING PATHWAYS THROUGH GENE TRAP MUTAGENESIS

Soriano P

Fred Hutchinson Cancer Research Center, Seattle, United States

We have developed a versatile, high throughput genetic screening strategy by coupling gene mutagenesis and expression profiling technologies.  Using a retroviral gene trap vector optimized for efficient mutagenesis and cloning, genes were randomly disrupted in mouse embryonic stem cells and amplified to construct a cDNA microarray.  With this gene trap array, we demonstrate that PDGF transcriptional target genes can be efficiently and reliably identified in physiologically relevant cells, and are immediately accessible to genetic studies to reveal their in vivo roles and relative contribution to PDGF regulated developmental processes.  The same platform can be used to search for genes of specific biological significance in a broad array of experimental settings, providing a fast track from gene identification to functional validation.

PLENARY PRESENTATION

WEDNESDAY OCTOBER 20

8.30am – 9.00am

A SENSITIZED SCREEN FOR MODIFIERS OF EPIGENETIC PHENOMENA

Champ M1, Vickaryous N1, Hemley S1, Preis JI 1, Arkell R2, Whitelaw E1

1School of Molecular Microbial Biosciences, University of Sydney, NSW, Australia, 2Mammalian Genetics Unit, MRC Harwell, United Kingdom

Phenotype-driven mutagenesis is proving to be a useful way of discovering the function of genes in higher organisms. In the mouse, ENU causes point mutations in progenitor spermatocytes. We are using ENU mutagenesis to identify genes that modify epigenetic state. Transcriptional silencing associated with epigenetic modifications is known to be responsible for parental imprinting and X-inactivation in mammals and there is increasing support for the idea that it plays a critical role in differentiation. Epigenetic silencing can result in mosaic expression of genes within a tissue, termed variegation, and between isogenic individuals, termed variable expressivity. While some of the proteins involved in these processes are known, many remain to be identified.

We have treated male mice carrying an erythroid-specific GFP transgene that displays variegation. We have screened 610 F1 offspring for dominant mutations and have identified six; four are suppressors and two are enhancers of variegation. In all of the cases where we have determined the dominance of the mutant alleles, we have found the they are semi-dominant and homozygous lethal, indicating the obligate requirement for the genes that have been hit. The semi-dominance is consistent with the idea that transcription is determined by a dynamic equilibrium between complexes that silence and those that activate. 120 male F1 mice (without any dominant mutations) have been used in a recessive screen. So far, we have identified five recessive mutations. We have mapped five of the mutations to between 3 and 20 cM intervals. A number of the dominant mutations affect the variable expressivity seen at the endogenous agouti viable yellow allele. This locus is known to display variable expressivity in isogenic mice. Interestingly, in most cases, we see parent-of-origin and sex-specific effects. We are currently investigating whether any of the mutated genes play a role in X inactivation or parental imprinting.

This project has the potential to find novel genes involved in epigenetic phenomena, and to produce hypomorphs and hypermorphs of known genes that alter epigenetic state.

PLENARY PRESENTATION

WEDNESDAY OCTOBER 20

10.45am – 11.15am

NON-CODING RNAS PLAY MULTIPLE ROLES IN INACTIVATION INITIATION

Avner P

Institut Pasteur, Paris, France

X chromosome inactivation, a chromosome wide mechanism of transcriptional silencing, is under the control of a complex master locus, the X-inactivation centre (Xic) that contains the Xist gene, the source of a large 17 kb non-coding RNA critical to the X-inactivation process. The onset of X-inactivation which occurs during early embryogenesis sees an up-regulation of Xist transcripts and the spreading of the Xist RNA along the X chromosome, followed by chromatin modifications involving both extensive histone modifications and the recruitment of polycomb group proteins to the inactivating X and the transcriptional repression of X-linked genes. Female ES cells are extensively used as models for the inactivation process as undifferentiated female ES cells have 2 X’s active and X-inactivation occurs with the onset of differentiation. More recently both TS cells (Trophoblast) and XEN cells (extra-embryonic endoderm) have been used as models for imprinted X-inactivation. Integral to the initiation of X-inactivation is the counting process in which the X chromosome/autosome ratio in the cell is sensed and a choice step in which one of the two X-chromosomes in the female cell will be selected for inactivation. Both of these processes involve a region of the Xic 3’ to Xist which also contains the non-coding Tsix antisense RNA which partially overlaps Xist. The Tsix RNA which acts to repress steady state levels of Xist expression and influences the choice process also plays a possibly critical role in the dynamics of X-inactivation by restricting the Xist RNA to its transcription site. Recent results now also suggest an association of Tsix antisense transcription with H3 Lys-4 methylation and chromatin remodelling around the Xist promoter prior to the onset of X-inactivation.

  Several other novel non-coding RNAs are present in the Xic, in particular in the region 5’ and upstream to Xist which has been suggested to function as a nucleation centre prior to Xist coating of the X chromosome possibly by acting as anchorage sites for Xist. Large scale chromatin modifications occur in this region during the onsed of X-inactivation and these changes may well be correlated with transcription activity within the genomic loci encoding these non-coding RNAs. An overview of recent progress will be given.

PLENARY PRESENTATION

WEDNESDAY OCTOBER 20

7.30pm – 8.00pm

GENOMIC INSTABILITY AND AGING

Martin GM

University of Washington, Seattle, United States

The contributions of genomic instability to processes of biological aging will be reviewed in the context of a classification of gene actions that escape the force of natural selection. Murine models are increasingly utilized to address these issues. Examples have implicated important contributions of mitochondrial and nuclear genomic instability and of epigenetic and genetic instability. Senescent phenotypes will be shown to be associated with early gene actions that are either adaptive or non-adaptive. Major deficiencies of the current state of the field, however, can be attributed to the almost exclusive reliance upon life span assays, an overemphasis upon experimental interventions that shorten, rather than lengthen life span,  and too much reliance upon the pathologies associated with old age. What is missing  are longitudinal measures of the rates of post-maturational declines in suites of specific physiological functions.

PLENARY PRESENTATION

THURSDAY OCTOBER 21

8.30am – 9.00am

Genetics of host resistance to cytomegalovirus infection: Role of h2 and nk receptors

Vidal S

McGill University, Montreal, Canada

Human cytomegalovirus (CMV) infection can cause life-threatening disease in immunodeficient hosts. Experimental infection in mice has revealed that natural resistance to mouse CMV is genetically determined and mediated by binding of the natural killer (NK) cell activating receptor Ly49H to the pathogen-encoded glycoprotein m157. Novel data indicate multiple NK cell receptor gene mechanisms, including genetic and functional interactions with MHC receptor genes, that are crucial to the recognition of the infected cells and the eradication of the infection.

PLENARY PRESENTATION

THURSDAY OCTOBER 21

1.30pm – 2.00pm

GENETICS OF ATHEROSCLEROSIS:  INTEGRATING GENETIC AND GENE EXPRESSION DATA

Schadt E1, Lamb J1, Lum P1, Leonardson A1, Wang S2, Doss S2, Yang X2, Ghazalpour A2, Davis R2, Zhang B2, Horvath S2, Drake T2, Lusis A2

1 Rosetta Inpharmatics, LLC, Seattle WA, United States, 2UCLA School of Medicine, Los Angeles, CA, United States

We are utilizing genetic variation between strains of mice to identify genes and pathways contributing to atherosclerosis and diabetes.  Using QTL mapping, numerous relevant loci have been identified and some positional candidates have been confirmed.  We are now developing approaches that integrate traditional gene mapping and expression array data to help identify genes and elucidate pathways involved in these complex diseases.  Thus far, we have applied expression-based mapping strategies to genetic crosses between strains DBA and C57BL/6 (liver) and strains C3H and C57BL/6 (liver, fat, muscle, brain).  Each cross has resulted in the identification of thousands of expression QTL (eQTL).  About a third of these appear to result from cis-acting variations (that is, the eQTL maps over the gene encoding the transcript in question).  These have been validated using a classical cis-trans test with F1 mice, in which the levels of transcript from each allele were determined using SNPs present in the mRNA.  Trans-acting eQTL are being validated using congenic strains.  For this, we are expression profiling a set of congenic strains between DBA and C57BL/6 containing introgressed segments (about 30-40 cM in size) covering all chromosomes.  A major goal of our studies is to use the combination of genetics and gene expression to construct causal expression networks involved in atherosclerosis and diabetes.

ORAL PRESENTATION

MONDAY OCTOBER 18

2.45pm – 3.00pm

THE FUNCTIONAL IMPLICATIONS OF LINKAGE DISEQUILIBRIUM BLOCKS ON MOUSE CHROMOSOMES

Paigen K, Petkov PM, Graber JH, Churchill G

The Jackson Laboratory, Bar Harbor, ME, United States

Typing 1508 SNPs chosen for high information content on 63 common and wild-derived mouse strains chosen for maximum genetic diversity, we find that over half the mouse genome presents as large blocks of linkage disequilibrium (LD blocks), averaging 5 Mb in length and 100 times the length of LD blocks typically observed in human studies.  The largest block, 40+ Mb, is on the X chromosome.  LD blocks can be segmented, containing short, internal regions in linkage equilibrium, and show internal haplotype switching.

  Strong LD does not result from a deficiency of recombining ability in these regions as their average cM/Mb is similar to the genome wide average; nor does it appear to derive from the common genetic origins of laboratory inbred mice as only 12-13% of unlinked genes show LD.  Rather, a compression of genetic distances in LD regions compared to non-LD regions among RI lines indicates that LD arises from convergent selection of co-adapted sets of alleles at functionally related linked genes during the intense reproductive selection applied during inbred strain formation, allied with the pressures of survival as homozygotes. 

  Importantly, some LD blocks form inter-connnected networks seen as LD between genetically unlinked blocks, presumably reflecting functional relationships of distant chromosome regions.

  To the extent ancient evolutionary forces have driven the arrangement of genes along mammalian chromosomes, putting functionally related genes in proximity to promote their common inheritance, observations of LD in inbred mice provide a means of investigating this functional organization.

ORAL PRESENTATION

MONDAY OCTOBER 18

3.00pm – 3.15pm

COUPLED COMPUTATIONAL AND EXPERIMENTAL APPROACHES TO DISCOVERY OF CODING AND NON-CODING GENES IN THE MOUSE GENOME

Hughes TR, Morris Q, Zhang W, Babak T, Mohammad N, Shai O, Fehlings M, Aubin J, van der Kooy D, Rossant J, Bruneau B, Blencowe B, Frey B

University of Toronto, Toronto, Canada

We are applying three different approaches to identify and characterize new transcripts in the mouse genome.  All three involve hybridizing RNA from diverse mouse tissues to custom Agilent oligonucleotide microarrays designed to detect computationally predicted genes. 

In the first approach, expression of 42,000 “XM” genes (NCBI predictions) was analyzed in 55 different tissues.  Using a stringent threshold, 21,575 genes were detected as “expressed”; among these, more than 5,000 are not present in current cDNA databases, and more than 3,000 are not present in EST databases.  Among the ~2,000 that were represented by an EST but not a cDNA, a random sampling demonstrated that more than half could be amplified by RT-PCR.  In virtually all cases we confirmed the tissue specificity observed on microarrays, which can be used to predict function.

In the second approach, we are using the program QRNA (Rivas and Eddy, 2001), which seeks regions of phylogenetically-conserved secondary structure, to predicting noncoding RNAs.  Using a pilot microarray to querying over 3,000 QRNA predictions across ten tissues, we detected fifty candidate noncoding RNAs as expressed.  Thus far we have confirmed three of these by Northern analysis that are between 50-100 nucleotides long and expressed at levels similar to the spliceosomal RNAs, but do not overtly appear to be members of any known RNA class (tRNA, snoRNA, etc.)

In the third approach, we designed microarrays containing over one million known and predicted mouse exons identified by one or more of five different gene-finding programs, and hybridized these arrays to cDNA from twelve mRNA pools encompassing forty different mouse tissues.  Visual analysis of the resulting data suggests that many sets of co-regulated exons are observed that appear likely to be novel, tissue-specific, multi-exon transcripts (i.e., they are not represented in RefSeq, Ensembl, Fantom, or Unigene).  Computational analysis of these data is in progress.

We anticipate that these analyses will contribute to the assembly of a complete collection of mammalian genes, and to their functional characterization.

ORAL PRESENTATION

MONDAY OCTOBER 18

3.15pm – 3.30pm

IDENTIFICATION OF CO-EXPRESSED GENE CLUSTERS IN A COMPARATIVE ANALYSIS OF TRANSCRIPTOME AND PROTEOME IN MOUSE TISSUES

Mijalski T1, Harder A2, Halder T2, Kersten M2, Horsch M1, Drobyshev A1, Lottspeich F3, Hrabe de Angelis M1, Beckers J1

1 Institute of Experimental Genetics, GSF – National Research Centre for Environment and Health, Neuherberg, Germany, 2 TopLab GmbH, Proteomics-Division, Martinsried, Germany, 3 Max-Planck-Institute for Biochemistry, Martinsried, Germany

A major advantage of the mouse model system lies in the increasingly complete information on its genome, transcriptome and proteome, as well as in the availability of a fast growing number of mutant and genetically engineered alleles. However, little is known about the relationship between the transcriptional and post-transcriptional regulation of gene expression. Data from comparative transcriptome and proteome analyses in the mouse as mammalian model organism is very limited. We use DNA-chip based RNA expression profiling and 2D-gelelectrophoresis combined with mass spectrometry of mouse liver and kidney extracts to explore the general feasibility of such comprehensive gene expression analyses. Whereas protein analyses mostly identify known metabolic enzymes and structural proteins in these tissues, transcriptome analyses reveal the differential expression of functionally diverse genes and, in addition, a significant number of genes that are not functionally described. The comparative analysis of proteins and their corresponding transcripts suggests a positive correlation between transcriptional and translational expression for the majority of genes and a few significant exceptions. Based on RNA expression data from the two hundred most differentially expressed liver or kidney specific genes, we identify chromosomal co-localization of known as well as not yet described gene clusters. The spatial organization and evolutionary conservation of such gene clusters may suggest common gene regulatory functions.

ORAL PRESENTATION

MONDAY OCTOBER 18

3.30pm – 3.45pm

IN-SILICO MAPPING - USE OF INFERRED HAPLOTYPES TO DEFINE QTLS

Pletcher MT1, McClurg P1, Batalov S1, Su A1, Bogue M3, Mural R2, Paigen B3, Wiltshire T1

1 GNF, San Diego, United States, 2 Celera Genomics, Rockville, United States, 3 The Jackson Laboratory, Bar Harbor, United States

Single nucleotide polymorphisms (SNPs) that occur between mouse strains may produce a specific functional change in a gene leading to phenotypic variation but are more often simply markers for an ancestral haplotype.  In silico mapping provides an expedient way to associate the natural diversity of phenotypic traits with ancestrally inherited polymorphisms for the purposes of dissecting genetic traits.  The goal of in-silico mapping is to identify which haplotype patterns track with a specific phenotype with the idea that the tracking haplotype contains a causative mutation.  For in silico mapping to be successful, a dense SNP map is required and multi-strain phenotypes data well defined. In mouse, the current SNP data has lacked the density across the genome and coverage of enough strains to properly achieve this goal.  To remedy this, 467,015 allele calls were produced for 10,917 evenly spaced SNP loci across 48 inbred mouse strains.  Phenotype data for multiple strains is now available through the Phenome project.  Use of the SNP set with statistical models that considered unique patterns within blocks of 3 SNPs as fixed haplotypes could successfully map known single gene traits and a cloned quantitative trait gene.  Application of these methods to high density lipoprotein and gallstone phenotypes reproduced previously characterized quantitative trait loci (QTL).  Also, examining the SNP data in this manner allows for QTL regions to be refined by reviewing the haplotype structures such that candidate genes are then more readily identified.  We have examined a candidate gene Adcy7for an HDL phenotype in this manner and characterized polymorphisms within the gene.

ORAL PRESENTATION

MONDAY OCTOBER 18

4.45pm – 5.00pm

REGION-SPECIFIC SATURATED MUTAGENESIS USING THE SLEEPING BEAUTY TRANSPOSON SYSTEM

Takeda J1, Keng VW1, Yae K1, Hayakawa T1, Mizuno S1, Uno Y2, Kokubu C1, Horie K1

1 Dept. of Social and Environmental Medicine, Osaka University Graduate School of Medicine, Osaka, Japan, 2 The Institute of Experimental Animal Science, Osaka University, Osaka, Japan

Generation of mutant mice is important in determining the function of genes in the post-genome era. Sleeping Beauty (SB) is a synthetic transposon derived from a fish genome and we have previously demonstrated that the SB transposon sytem worked efficiently in mouse germ line, suggesting a potential usage of a large-scale mutagenesis in mice. We also observed that about 80% of transposition events in germ cell preferentially occurred near the donor site and within the same chromosome.  After screening about 600 mice, the number of transposon-integrated gene hits was shown to be clustered within approximately 3 Mb distance of the donor site.  From a double-positive (containing SB transposase and transposon, acting as a mutagen) transgenic mouse with donor site located on chromosome 12, we obtained 16 clustered transposon integrations, corresponding to about 50% of known genes located within the 3 Mb region.  This indicates the usefulness of this system for saturated mutagenesis, which is currently impossible with other large-scale mutagenic systems. We are currently analyzing about 50 mutant lines with transposon integration in both known and novel genes by generating homozygotes and analyzing the resulting phenotype. Taken together, SB transposon system is a rapid and efficient method for region-specific mutagenesis.

ORAL PRESENTATION

MONDAY OCTOBER 18

5.00pm – 5.15pm

INGENOTYPING – AN HIGHLY EFFICIENT APPROACH TO GENERATE GENE TARGETED MOUSE AND RAT MODELS

Laufs J, Sedlmeier R, Peters T, Huffstadt U, Wattler S, Nehls M

Ingenium Pharmaceuticals, Martinsried, Germany

Mouse and rats are the model of choice for in vivo studies of gene function but the development of knockouts and allelic variants, i.e. hypo – and hypermorphs, is time consuming and costly. The generation of mouse models is dominated by gene targeting through homologous recombination, which requires sophisticated ES cell manipulation and chimera production. No such standard procedure for the development of targeted rat models is available despite some first success in nuclear cloning of rats from unmodified somatic cells.  In order to make rat and mouse models with genetic alterations readily available to the scientific community, we developed a patented technology (named INGENOtypingTM) based on gene-saturating chemical mutagenesis. We first applied INGENOtypingTM to generate a pre-set library of mutagenized murine sperm that represents multiple alterations in every mouse gene. Currently somatic DNA and corresponding sperm samples of more than 15,000 mice derived from a well-controlled chemical mutagenesis process using ENU have been archived and can be screened for mutations in any gene of interest within days. Together with the subsequent extremely efficient in vitro fertilization, the adult heterozygous carriers are available within 3-4 months. This approach has now been adapted to rats for the industrialized production of gene targeted rat models.  ENU mutagenesis in our hands has resulted in more than 220 identified mouse mutations which are to a large extent null but also hypomorphic and hypermorphic alleles. The process and examples of ENU derived mutant mouse lines will be presented.

ORAL PRESENTATION

MONDAY OCTOBER 18

5.15pm – 5.30pm

MUTATIONS THAT CAUSE HETEROZYGOUS EYE DEFECTS ARE LETHAL WHEN HOMOZYGOUS AND MODEL HUMAN DISEASE

Cross SH1, Hart AW1, Morgan JE1, McKie L1, West K1, Schneider JE2, Bhattacharya S2, Jackson IJ1

1 MRC Human Genetics Unit, Edinburgh, United Kingdom, 2 Wellcome Trust Centre for Human Genetics, Oxford, United Kingdom

ENU mutagenesis is a powerful tool for revealing gene function. We have produced a collection of 25 ENU-induced mutants with a variety of eye defects and most of these are homozygous lethal. Three– Rwhs, Icst and Dilp2 (an X-linked male lethal) – map to novel loci. We used high-throughput magnetic resonance imaging (MRI) of embryos to determine the mechanism of lethality in these mutants. This enables us to screen 32 embryos per overnight run, retains 3D information and allows embryos to be examined in any section plane. Using MRI, we found that Rwhs homozygotes have pulmonary hypoplasia and Bochdalek congenital diaphragmatic hernia. This malformation is present in 1:2500 live births, and even with corrective surgery has a mortality of ~30%. Icst homozygotes have a ventricular septal defect and aberrant right-sided aortic arches. Dilp2 males have a common arterial trunk, a condition that accounts for 1% of all congenital heart defects. In addition they have palate defects and the absence of a sternum.

The critical genetic intervals for each of these do not contain any previously described candidate genes but are now sufficiently small to start scanning the genes present for mutations. It is noteworthy that numerous of our mutations that give a dominant eye phenotype affect organogenesis and are lethal when homozygous. Importantly, all three model conditions that are relevant for human disease.

ORAL PRESENTATION

MONDAY OCTOBER 18

5.30pm – 5.45pm

HAPLOTYPE ANALYSIS IN MULTIPLE CROSSES TO IDENTIFY A QTL GENE

Wang X, Korstanje R, Higgins D, Paigen B

The Jackson Laboratory, Bar Harbor, United States

Identifying quantitative trait locus (QTL) genes is a challenging task. Herein, we report using a two-step process to identify Apoa2 as the gene underlying Hdlq5, a major QTL for plasma high-density lipoprotein cholesterol (HDL) levels on mouse chromosome 1. First, we performed a sequence analysis of the Apoa2 coding region in 46 genetically diverse mouse strains and found five different APOA2 protein variants, which we named APOA2a to APOA2e. Second, we conducted a haplotype analysis of strains in 21 crosses that have so far detected HDL QTLs and found that Hdlq5 was detected only in the nine crosses where one parent had the APOA2b protein variant characterized by an Ala61 to Val61 substitution. We then found that strains with the APOA2b variant had significantly higher (P < 0.002) plasma HDL levels than those with either the APOA2a or the APOA2c variant. Thus, in addition to identifying Apoa2 as the gene underlying Hdlq5, our haplotype analysis identified the Apoa2 polymophism responsible for the Hdlq5 phenotype. Our haplotype analysis in multiple crosses proved to be a quick and cost-effective approach in testing a candidate QTL gene.

ORAL PRESENTATION

TUESDAY OCTOBER 19

9.00am – 9.15am

SOMATIC SLEEPING BEAUTY TRANSPOSITION PROVIDES A POTENTIALLY POTENT CANCER GENE DISCOVERY TOOL

Jenkins NA, Dupuy AJ, Copeland NG

National Cancer Institute, Frederick, United States

Transposon-induced insertional mutagenesis has been a valuable tool for functional genomics in invertebrates and holds great promise for functional genomic studies in mouse.  Sleeping Beauty (SB), a synthetic “cut-and-paste” transposon of the Tc1/mariner class, transposes at low frequencies in mouse ES cells (3.5 X 10-6 transposition events/cell per generation) and higher frequencies in the mouse germ line (1-2 transposition events per animal born), but these frequencies are too low to be useful for most genetic screens.  Sleeping Beauty is a two-part system consisting of the transposon and the transposase enzyme.  When both elements are present in the same cell, the transposase binds to the inverted repeats present at the ends of the transposon facilitating its excision and reintegration at a new site in the host cell genome.  We have generated an enhanced SB mutagenesis system that holds great promise for cancer gene discovery.  Transgenic mice were created that harbor 100-300 tandem copies of a genetically enhanced SB transposon in addition to a line of mice that carry a genetically enhanced transposase at the Rosa26 locus.  Approximately 75% of the mice produced that carry both elements died during development, presumably as a result of somatic transposition-mediated developmental defects.  Mice that were born died between 6 and 16 weeks of age from aggressive hematopoietic malignancies.  Tumors from these mice contained multiple (10-20 or more) SB transpositions and some of these transpositions were located at sites of known cancer genes. Our experiments demonstrate the power of somatic SB transposition for cancer gene discovery.  SB elements can transpose in both dividing as well as in non-dividing cells unlike retroviruses, which integrate into the genome only during cell division.  In principle, therefore, somatic SB transposition could induce tumors in a wide variety of cell types and provide a nearly universal cancer gene discovery tool.

ORAL PRESENTATION

TUESDAY OCTOBER 19

9.15am – 9.30am

GENOMIC CHARACTERIZATION OF NEURONAL SYNAPSE

Yang S2, Murphy TK2, Hadley D4, Farias M2, Kapfhamer D2, Ungar L1, Kim J3, Bucan M1

1 Penn Center for Bioinformatics, University of Pennsylvania, Philadelphia, United States, 2 Department of Genetics/SOM, University of Pennsylvania, Philadelphia, United States, 3 Department of Computer and Information Sciences/SEAS, University of Pennsylvania, Philadelphia, United States, 4 Department of Biology/SAS, University of Pennsylvania, Philadelphia, United States

The goal of our studies is to use the neuronal synapse, the key functional unit in cell-cell communication in the brain, as a model for a comparative and integrative approach to pathway building and to address the relevance of this pathway to human disease.  We have developed SynapseDB - a database of components (genes/proteins) that are the functional building blocks of a synapse (genomic sequence, multi-species conserved sequences, expression and mutant phenotypes).  Phylogenomic analyses of gene and protein sequences for several large gene families (synaptotagmins, syntaxins, rab proteins) revealed variable rates of evolutionary changes in different gene families, "species-specific" paralogs and evolutionary dynamics of non-coding regions.  Specifically, while coding regions of closely related paralogs showed slow rates of sequence changes as measured by per-gene ratios of Ka/Ks, non-coding regions showed high levels of sequence changes. 

We have also investigated Rab3a, a small GTPase involved in synaptic vesicle trafficking, in more detail.  To address the molecular mechanisms underlying behavioral phenotypes in two alleles (loss-of-function and dominant-negative allele), we carried out transcriptome analysis by the microarray and Q-PCR analysis of RNA isolated from the cortex and hippocampus of wild type and mutant mice (C57BL/6J - N4 generation). These experiments identified a set of differentially expressed genes (wild type vs, mutant) and only one differentially expressed gene (Rabphilin – a Rab3A effector) in common between the two alleles.  The most striking result was the detection of over 20 differentially expressed genes (129/Sv vs. C57BL/6J) located in the 129/Sv-derived chromosomal region surrounding the Rab3A-/- gene.  Our results illustrate the challenges of molecular and behavioral comparison of mutants that arise on different genetic backgrounds.

ORAL PRESENTATION

TUESDAY OCTOBER 19

9.30am – 9.45am

IDENTIFICATION OF REGION-SPECIFIC TRANSCRIPTION FACTOR GENES IN THE ADULT MOUSE BRAIN BY MEDIUM-SCALE REAL-TIME RT-PCR

Suzuki H, Okunishi R, Hashizume W, Katayama S, Ninomiya N, Osato N, Sato K, Nakamura M, Iida J, Kanamori M, Hayashizaki Y

RIKEN GSC, Yokohama, Japan

Many transcription factors play important roles in the development and functions of the brain, although their low levels of expression in the brain’s complex tissues have hindered the exploration of their expression properties by DNA microarray technology. We established a medium-scale real-time RT-PCR system focusing on mouse transcription factors and applied it to their expression profiles in 11 regions of the adult mouse brain. Of a total of 756 transcription factor genes examined, 677 (89.6%) showed significant expression in at least one brain region, reflecting the high level of complexity of the mouse brain. We successfully extracted 179 (23.7%) region-specific genes by clustering analysis; they were observed in abundance in the medulla oblongata, cerebellum, pituitary gland and midbrain, but less so in the posterior cerebral cortex and hippocampus. Interestingly, most of the transcription factors involved in the development of the anterior pituitary were still expressed specifically in the adult pituitary, suggesting that these transcription factors play important roles in both development and maintenance of the pituitary. In addition to providing reliable expression profiles of transcription factors and a unique set of molecular markers for 11 brain regions, these results can help us to identify transcription regulation pathways that may account for the molecular basis of the unique functions of specific regions of the brain.

ORAL PRESENTATION

TUESDAY OCTOBER 19

9.45am – 10.00am

SIGNIFICANCE OF GENETIC VARIABILITY IN THE HUMAN PON1 GENE EXPLORED IN A MOUSE MODEL

Furlong CE1, Cole TB1, Richter RJ1, Li WF1, Carlson C1, Nickerson D1, Jarvik GP1, Lusis AJ2, Tward A2, Shih D2, Costa LG1

1 University of Washington, Seattle, United States, 2 UCLA, Los Angeles, United States

Paraoxonase (PON1), an HDL-associated enzyme, exhibits broad substrate specificity.  While its main physiological role appears to be the metabolism of oxidized lipids and prevention of vascular disease, it also metabolizes organophosphates and several drugs.  More than 160 PON1 SNPs have been identified.  The Q192R protein polymorphism affects catalytic efficiency of hydrolysis of some compounds. The 5’ regulatory region SNP at C-108T is in a SP1 binding site and affects plasma levels of PON1.  The term “PON1 status” reflects both an individual’s PON1 plasma level and Q192R genotype. A “humanized” mouse PON1 model has provided valuable insights into the significance of variability in the human PON1 gene. PON1-/- mice show dramatically increased sensitivity to diazoxon (DZO) and chlorpyrifos oxon (CPO), but not to paraoxon.  Thus, PON1 activity in vitro does not necessarily indicate physiological significance. Differences in sensitivity are explained by the catalytic efficiency of each PON1192 alloform. Injection of purified human PON1Q192 or PON1R192 restored resistance to DZO, but only PON1R192 provided significant resistance to CPO.  These observations are consistent with the sensitivities of the “humanized” PON1Q192R transgenic mice.  Mice expressing hPON1Q192 were nearly as sensitive as the null mice to CPO exposure, whereas the hPON1R192 mice exhibited significant resistance.  These observations are significant in that PON1Q192 homozygotes comprise about 50% of the individuals in many populations.  Supported by ES09601/EPA-R826886, ES09883, ES04696, ES07033, ES11387, T32 AG00057.

ORAL PRESENTATION

TUESDAY OCTOBER 19

10.00am – 10.15am

IDENTIFICATION OF THE IL-17 RECEPTOR-RELATED MOLECULE, IL-17RC AS A RECEPTOR FOR IL-17A AND IL-17F

Kuestner R, Brandt C, Gao Z, Ostrander C, Bort S, Taft D, Bilsborough J, Lewis K, Jaspers S, Dillon S, Lewis P, Topouzis S, Rixon M, Chan C, Moore M, Reardon B, Bukowski T, Moore B, West J, Parrish-Novak J, Levin SD

ZymoGenetics, Inc., Seattle, United States

The pro-inflammatory cytokines IL-17A and IL-17F have a high degree of sequence similarity, share many biological properties, and are both produced by activated T cells.  They have both been implicated as factors that contribute to the progression of various autoimmune and inflammatory diseases including rheumatoid arthritis and asthma.  In fact, reagents that negate IL-17A function significantly ameliorate disease incidence and severity in several mouse models of human disease.  IL-17A mediates its effects through interaction with its cognate receptor, the IL-17 receptor (IL-17R), but the receptor for IL-17F has not yet been identified.  We now report that we have identified the IL-17R-related molecule, IL-17RC as the receptor for IL-17F.  However, we have also noted that this receptor binds to both IL-17A and IL-17F with a similar high affinity.  IL-17R on the other hand, binds IL-17A with high affinity, but binds IL-17F with very low affinity.  Consistent with this, we have shown that a soluble form of IL-17R blocks IL-17A binding and signaling in cells expressing either receptor, but does not interfere with binding or function of IL-17F to IL-17RC.  In contrast, a soluble form of IL-17RC inhibits the interaction and effects of both IL-17A and IL-17F in cells expressing either receptor.  Since IL-17A intervention has been proposed as an effective therapy for several auto-immune diseases, we propose that using soluble IL-17RC to inhibit effects of both IL-17A and IL-17F will have advantages over therapies that target only one of these two cytokines.

ORAL PRESENTATION

TUESDAY OCTOBER 19

2.30pm – 2.45pm

ESTABLISHING THE FUNCTIONAL POTENTIAL OF CONSERVED NON-CODING SEQUENCES

Grice EA1, Emison E1, Portnoy ME2, Rochelle E1, Comparative Sequencing Program NISC2, Green ED2, Chakravarti A1, McCallion AS1

1 McKusick-Nathans Institute of Genetics Medicine, Johns Hopkins University, Baltimore, United States, 2 National Human Genome Research Institute, National Institutes of Health, Bethesda, United States

Mutations in regulatory sequences influence tissue specificity, timing and/or level of the corresponding gene product.  Despite the predicted role of regulatory mutations in human genetic disease, their nature and identity remain largely unknown. Sequence conservation among distantly related species is proposed to be an indicator of sequence functionality.  Sequence analysis of putative regulatory elements at known disease genes may reveal variants that are significantly associated with disease.  However, it provides little direct evidence of causation. Consequently, functional non-coding sequences and the variants that compromise them must be interrogated in model organisms like the mouse.  We have combined the use of computational tools, in vitro pre-screens and human genetic data to preliminarily interrogate conserved sequences.  We are employing the mouse to further examine the functional potential of critical subsets of non-coding conserved sequences.  The RET gene is ideal for the development of such strategies - as coding sequence mutations are known and non-coding mutations are predicted among a known disease population (Hirschsprung disease: HSCR). We recently completed comparative analysis of ~350 Kb of genomic sequence from 13 vertebrates, identifying multi-species conserved sequences (MCSs) at RET.   We will report the analysis of >30 such non-coding MCSs, demonstrating that >85% can function as enhancers/suppressors in vitro. We also report the identification of a HSCR-associated variant in an MCS in RET intron 1 and will discuss its functional impact. Our long-term objective is to develop rapid screens to uncover the nature, identity and phenotypic impact of regulatory mutations in critical developmental / disease genes.

ORAL PRESENTATION

TUESDAY OCTOBER 19

2.45pm – 3.00pm

EXACTPLUS : A PROGRAM FOR DETECTING SMALL CONSERVED GENOMIC REGIONS BY MULTI-SPECIES SEQUENCE COMPARISONS

Antonellis A1, Prasad AB1, Wolfsberg TG1, Program NCS2, Green ED1, Pavan WJ3

1 Genome Technology Branch, NHGRI, NIH, Bethesda, United States, 2 NIH Intramural Sequencing Center, NHGRI, NIH, Gaithersburg, United States, 3 Genetic Disease Research Branch, NHGRI, NIH, Bethesda, United States

Multi-species comparative sequence analysis is emerging as a powerful tool for identifying transcriptional regulatory elements. One obstacle in these analyses is determining the appropriate conservation thresholds that accurately identify sequences most likely to be functional. With this in mind, we developed a software tool (ExactPlus) that identifies short, identical stretches of DNA sequence in multi-species alignments. Relevant factors related to this approach include: (1) transcriptional regulatory elements are likely defined by transcription factor binding sites; (2) transcription factor binding sites may represent very short stretches of DNA; (3) sequence conservation within regulatory elements is likely highest at protein binding sites; and (4) algorithms that identify large, broadly conserved regions may not detect smaller, highly conserved functional elements. ExactPlus input includes a MultiPipMaker alignment file, the basepair length of matches to report, and the number of species that define a match. For example, fragments 6 basepairs or greater identical in at least 5 out of 7 species may be identified. ExactPlus reports a consensus sequence at each match, and a UCSC Genome Browser custom track for positioning results on available genomes.

We used ExactPlus to analyze multi-species sequences of five loci involved in melanocyte development. Importantly, experimentally validated regulatory elements have been identified for a subset of these. We established appropriate thresholds by comparing ExactPlus results obtained from human/mouse/rat alignments with: (1) the known set of regulatory elements, and (2) the ExactPlus dataset generated using additional mammalian species. Our results are relevant for identifying regulatory elements using already available genome sequences.

ORAL PRESENTATION

TUESDAY OCTOBER 19

3.00pm – 3.15pm

POLYMORPHISMS PREDATING THE DIVERGENCE OF THE MUS MUSCULUS SUBSPECIES ARE VERY COMMON IN INBRED STRAINS.

Pardo-Manuel de Villena F1, Ideraabdullah FY1, Doherty HE1, Bell TA1, de la Casa-Esperon E2, Detwiler DA1, Sapienza C2

1 Department of Genetics, UNC-Chapel Hill, Chapel Hill, United States, 2 Fels Institute fro Cancer Research, temple University School of Medicine, Philadelphia, United States

We have estimated the total genetic diversity present in a panel of wild-derived and classical inbred strains. Substitutions and microinsertions/deletions generate a variant at 3.5% of the positions in introns and intergenic regions. In coding sequences the frequency of variants is 1.4%, while UTRs have intermediate levels of diversity. Overall microinsertions/deletions contribute 10% to the total number of variant events and 50% of the total number of variant positions. This represents the highest level of diversity described in a mammalian species yet, despite the limited size of this panel. We have classified all variants into two categories depending on whether the variant arose before or after the divergence of the Mus musculus subspecies. Interestingly, 33% of the variants predate the divergence of the subspecies, and a third of those variants predate the divergence of the musculus, spretus and spicilegus lineages. These ancient variants are distributed uniformly across the genome and interspersed within more recent variation. Therefore, a substantial fraction of the diversity is ancient and has been segregating in the ancestral mouse populations for millions of generations.  its an important factor in explaining the levels of diversity found in mouse. Given their frequency ancient variants are expected to contribute significantly to the total phenotypic variation in mouse. These variants, if unrecognized, may affect the topology and length of branches of phylogenetic trees. Furthermore, ancient variants differ significantly from the more recent variants with respect to the patterns and length of linkage disequilibrium and may pose challenges in genetic studies.

ORAL PRESENTATION

TUESDAY OCTOBER 19

3.15pm – 3.30pm

REGULATORY AND CODING SEQUENCES OF THE PITUITARY-SPECIFIC TRANSCRIPTION FACTOR, PROP1 ARE EVOLUTIONARILY AND FUNCTIONALLY CONSERVED.

Camper SA1, Ward RD1, Cho M-C1, Raetzman LT1, Esposito C1, Rubin EM2, Smith TPL3, Rhodes SJ4, Lyons RH1

1 University of Michigan Medical School,, Ann Arbor, MI, United States, 2 DOE Joint Genome Institute,, Walnut Creek, California, United States, 3 USDA/ARS, Meat Animal Research Center,, Clay Center, NE, United States, 4 Indiana University-Purdue University Indianapolis, Indianapolis, United States

All vertebrates have pituitary glands composed of specialized hormone-producing cells.  The individual hormones are evolutionarily conserved, although their function varies across the classes of Animalia. For example, prolactin influences libido, maternal behavior and lactation in mammals, while it regulates the fish response to salinity, and it facilitates learning territorial bird songs.  Even prochordates such as Ciona savignyi have pituitary-like hormones that function in regulation of growth and gonadotropin production, suggesting that the genes regulating pituitary organogenesis may be conserved throughout the phylum Chordata.  The first known pituitary-specific gene in the genetic hierarchy is Prop1, a paired-like homeodomain transcription factor.  PROP1 mutations are the most common known cause of multiple pituitary hormone deficiency in humans, and mice homozygous for either the spontaneous Ames dwarf mutation or the targeted null allele are good models for the human disease.  We determined the sequence of the PROP1 gene in five primates, including human, gorilla, baboon, howler monkey and lemur, and compared them with the sequences of PROP1 in other orders of mammals including species representing Rodentia, Carnivora, and Artiodactyla, as well as distantly related vertebrates such as Fugu, zebrafish and chicken.  A phylogenetic analysis using parsimony supported the orthologous nature of the sequences, and suggested that PROP1 is an orphan class of paired homeodomain transcription factors.  Alignment of large regions of genomic sequence from four species and VISTA analysis revealed the presence of four highly conserved noncoding sequences (CNS) in and around the PROP1 gene.  Studies in transgenic mice confirm that CNS1 is functionally important; it confers a dorsalized expression pattern in the developing pituitary gland.  Phylogenetic footprinting of CNS1 across several species implicates specific DNA binding sites in regulation of PROP1 expression.  In conclusion, evolutionary comparison has proven to be a valuable approach to understanding regulation of PROP1, the key gene for pituitary organ development in vertebrates. 

ORAL PRESENTATION

TUESDAY OCTOBER 19

4.30pm - 4.45pm

INFLAMMATORY DISEASE AND ABORTIVE PLATELET SHEDDING CAUSED BY A MUTATION IN A PIVOTAL MODULATOR OF ACTIN DYNAMICS IN THE REDEARS MOUSE

Kile BT2, Woodward LS1, Justice MJ1

1 Baylor College of Medicine, Houston, United States, 2 The Walter and Eliza Hall Institute of Medical Research, Parkville, Australia

Recently identified in a balancer chromosome mutagenesis screen, the redears mouse is an intriguing model of inflammatory disease and thrombocytopenia. Animals homozygous for the redears (rd) mutation develop spontaneous inflammatory lesions of the ears and tail characterised by necrosis, edema, epithelial thickening and leukocyte infiltration. Neutrophils are the predominant infiltrating cell, and neutrophilia is observed in the peripheral blood and spleen. Chemotaxis and apoptosis are perturbed in rd/rd neutrophils suggesting these cells are playing a key role in driving the inflammatory process. Unexpectedly, blood platelet numbers are dramatically reduced in rd/rd animals, and a thorough analysis of platelet biogenesis confirms this is the result of defects in the platelet precursor cell, the megakaryocyte. rd/rd megakaryocytes undergo an abnormal maturation, characterised by gross morphological abnormalities, increased ploidy and abortive platelet shedding. We have identified a mutation in a gene related to the yeast actin-interacting protein Aip1, in rd/rd mice.  While its exact role remains controversial, Aip1 interacts with, and increases the activity of cofilin, a key regulator of actin polymerisation. Our data suggests the rd mutation is hypomorphic, and confirm that actin dynamics are dysregulated in rd/rd neutrophils. This provides an explanation for chemotactic deficiencies, and in conjunction with recent evidence implicating cofilin and other actin regulators in the initiation of apoptosis, suggests that this protein might play an essential role in neutrophil cell death. The massive cytoplasmic reorganisation that is required for megakaryocyte maturation and platelet shedding has long been assumed to depend on the actin cytoskeleton. Intriguingly, recent studies suggest the process is caspase-dependent, and represents a form of ‘para-apoptosis’. Thus, the redears mouse not only provides the first in vivo demonstration of the critical role of the actin cytoskeleton in megakaryocyte development and platelet production, but also represents a unique reagent to examine the relationship between actin dynamics, cellular maturation, inflammation and apoptosis.

ORAL PRESENTATION

TUESDAY OCTOBER 19

4.45pm – 5.00pm

NEW INSIGHTS INTO MOUSE HAPLOTYPE STRUCTURE AND SNP HAPLOTYPE MAPPING

Wade CM1, Frazer KA3, Kulbokas EJ2, Cox DR3, Linblad-Toh K2, Daly MJ1

1 Whitehead Institute, Cambridge, MA, United States, 2 Broad Institute of MIT and Harvard, Cambridge, MA, United States, 3 Perlegen Sciences, Mountain View, CA, United States

SNP haplotype mapping is promising to be an important method for mapping genes contributing to complex phenotypes.  To facilitate this we are embarking upon a large scale SNP haplotype map of the common inbred laboratory mouse strains. This exciting undertaking promises to enhance the results of prior QTL mapping experiments, and can be particularly powerful for narrowing causative regions when mapping data are available from three or more inbred laboratory mouse strains.

A recent collaboration using Perlegen Sciences arrays has revealed, for the first time, the fine-structure of SNP haplotypes in the mouse genome over several megabases at 5 different genomic locations. The data strongly support previous assertions that the ancestral segment size in inbred laboratory mice is in the order of one to two megabases.  While inbred mouse phylogeny may at first glance appear to be complicated, at a local level the individual strains can be readily ascribed to one of only a small number of ancestral groupings (most commonly 1 to 3).

In this presentation we aim to share the latest data from the haplotype map, describe public resources available to the mouse community from this work, and outline the pattern of local haplotype structure in a number of common mouse strains.

ORAL PRESENTATION

TUESDAY OCTOBER 19

5.00pm – 5.15pm

SEQUENCE AND PHENOTYPE ANNOTATION OF THE TYRP1 DELETION COMPLEX ON CHROMOSOME 4

Smyth IM1, Taylor M1, Wilming L2, Du X3, Gautier P1, Edgar R1, White S1, Cross SH1, Botcherby M4, Rogers J2, Campbell RD4, Brown SDM5, Beutler B3, Justice MJ6, Jackson IJ1

1 MRC Human Genetics Unit, Edinburgh, United Kingdom, 2 Wellcome Trust Sanger Institute, Hinxton, United Kingdom, 3 Scripps Research Institute, La Jolla, United States, 4 MRC Rosalind Franklin Centre for Genomics Research, Hinxton, United Kingdom, 5 MRC Mouse Genome Centre, Harwell, United Kingdom, 6 Baylor College of Medicine, Houston, United States

The region of chromosome 4 around the Tyrp1 locus has been studied for many years, initially as a target for radiation mutagenesis in the specific locus test, which generated numerous deletions and other rearrangements. Historically these deletions served to define essential genes mapping in the region by examination of the phenotypes of homozygous and compound heterozygous deletion carriers. Subsequently the deletions have allowed screening for chemically induced mutations. Over 25 recessive ENU mutations have been isolated that map to this chromosomal interval.

We produced high quality, finished and annotated sequence of a 173 BAC contig across this 22 Mb region, and have anchored the sequence to the deletion map. The whole region is relatively gene poor, and includes a ~5Mb region that is devoid of coding sequence, but which is nevertheless conserved in other vertebrate genomes.

We have identified candidate genes for the deletion defined phenotypes l(4)1Rn, l(4)2Rn, l(4)3Rn and baf.  We have also identified a novel basonuclein homologue that is misexpressed through a chromosomal inversion in the radiation-induced white based brown mutation. We have characterised a candidate for the classical hair loss mutation, depilated.

We have identified two new ENU alleles of the blebbing mutation heb, and have shown that the gene underlying the phenotype is a novel gene Frem1. The classic heb mutation contains a LINE insertion, an ENU induced allele, bat, has a splice defect. We are using a systematic exon resequencing strategy to identify the base lesions in the other ENU mutations in the region.

ORAL PRESENTATION

TUESDAY OCTOBER 19

5.15pm – 5.30pm

A MOUSE INSERTIONAL MUTATION ON CHROMOSOME 9 CAUSES JUVENILE HYDROCEPHALUS

Schmidt JV, Kalinina EA, Steshina E

University of Illinois at Chicago, Chicago, IL, United States

A series of transgenic mouse lines were generated in our laboratory that carried a minimal promoter driving the lacZ reporter gene.  These transgenes functioned as enhancer traps, expressing lacZ under the control of enhancers located at their individual integration sites.  One transgenic line demonstrated expression in the epiphysis of the diencephalon at midgestation, a tissue that represents the rudiment of the developing pineal gland.  The mammalian pineal is an enigmatic tissue, potentially involved in such functions as circadian rhythms, reproduction and aging.  Hoping to identify the “trapped” gene expressed in the developing pineal, we used anchored PCR to localize the transgene integration to mouse chromosome 9.

Upon generation of homozygous transgenic animals, this line displayed a lethal juvenile hydrocephalus phenotype.  Visible hydrocephalus appears by 3-4 weeks of age, and progresses to death by 6-8 weeks.  This phenotype indicates that, in addition to trapping a gene involved in pineal development, the transgene also disrupted expression of a gene required for normal brain development.  It is not yet known if these two genes are the same, and candidate genes near the transgene insertion are being tested for a possible role.  Histological analysis is being used to further analyze expression of the transgene, as well as to visualize the alterations in brain morphology.  Identification of the trapped gene will allow us to study its role in pineal development and may identify a new mechanism for juvenile hydrocephalus.

ORAL PRESENTATION

WEDNESDAY OCTOBER 20

9.00am – 9.15am

USING ENU MUTAGENESIS TO IDENTIFY NEW GENES INVOLVED IN EPIGENETIC GENE REGULATION

Lossie AC, Justice MJ

Baylor College of Medicine, Houston, United States

DNA methylation is an important biological process that plays a central role in epigenetic gene regulation. Imprinted genes often demonstrate differential methylation between the maternal and paternal chromosomes. In the past few years, studies have shown that DNA Methylation, in concert with chromatin structure, can alter gene expression. Defects in the processes that establish and maintain DNA methylation imprints often cause severe birth defects, including Angelman, Prader-Willi and immunodeficiency-centromeric instability facial-anomalies syndromes. In addition, some spontaneous abortions could be caused by defects in epigenetic gene regulation.

Although many imprinted genes have been characterized, we are just beginning to learn how these genes are regulated. In order to determine the mechanisms underlying mammalian epigenetic gene regulation, we are using N-ethyl-N-nitrosourea mutagenesis to select for homozygous mutant mice that have defects in establishing and/or maintaining well-established DNA methylation profiles. We identified 22 recessive mutant lines that are excellent candidates. These mutants die early during embryogenesis, and some demonstrate a broad phenotypic spectrum. These characteristics are consistent with the disruption of one or more global DNA methylation processes.

We used sodium bisulfite to identify mutant lines that have defective DNA methylation imprints. Since sodium bisulfite converts unmethylated (U) cytosines to uracils while leaving methylated (M) cytosines intact, we designed sets of primers to amplify either the M or U allele of several well-characterized differentially methylated regions (DMR) including: p57Kip2, Kvlqt1, Snrpn, Gnas, H19, Grb10, Ube3a and U2af1-rs1. So far, three lines show abnormal methylation at one or more DMR, and complementation studies indicate that two of the lines are allelic. This complementation group has two distinct phenotypes. The most severe allele dies by E6.5 with a pronounced lack of methylation at U2af1-rs1, while the other survives beyond E8.5, and has a less obvious methylation defect. The third mutant dies at E9.5 with defects in cardiac development and a disruption in methylation at Kvlqt1. Current studies will determine if other DMRs are abnormal in these three mutants. The identification of these mutations will give us a better understanding of the mechanisms underlying DNA methylation during embryogenesis, and provide us with a model to study epigenetic gene regulation in the mammal.

ORAL PRESENTATION

WEDNESDAY OCTOBER 20

9.15am – 9.30am

GENETIC STUDIES OF CANDIDATE METASTASIS MODIFIER GENES ON MOUSE CHROMOSOME 19

Park Y-G, Clifford R, Buetow KH, Hunter KW

National Cancer Institute, Bethesda, United States

The ability of a tumour to metastasize is the major determinant of cancer-patient mortality. Therefore elucidation of the metastasis pathway is an important priority for cancer biology. Studies have demonstrated that tumour progression is a complex process with many different barriers for a tumor to successfully colonize a distant site. To further explore the molecular and genetic events required for tumor dissemination, we have used the highly metastatic MMTV-PyMT transgenic mouse mammary tumor model, and combined medium-resolution haplotype mapping with multiple experimental crosses to reduce the number of potential candidate genes in a complex-trait candidate interval on mouse chromosome 19. Coincident mapping of a modifier gene in multiple experimental crosses is consistent with the common inheritance of a modifier allele. A haplotype map was developed in four inbred strains of mice used in our complex-trait mapping crosses across the proximal 10 cM of proximal chromosome 19 to identify haplotype blocks that segregate appropriately. Only ~23 out of >400 genes met this criteria. This strategy coupled with tissue and expression arrays, as well as our recently described common pathway analysis was used to further reduce the number of high-priority candidates to two. Subsequent analysis using both publicly available databases and direct sequencing of a set of the inbred mouse strains, has revealed multiple silent polymorphisms in both genes, Sipa1 and Map3k11, as well as a single missense mutation in a protein-binding domain of Sipa1 as well as the presence of a polymorphic VNTR in the promoter region of Map3k11. Current experiments have been investigating the role of these polymorphisms on gene function and metastatic progression by a variety of molecular strategies, including quantitative PCR, western blot analysis, mammalian gene expression, and siRNA technologies. This combination of methodologies, which we refer to as trans-omic analysis, may therefore provide a rapid, efficient method to identify and prioritize complex-trait candidate genes without requiring construction of congenic mouse strains.

ORAL PRESENTATION

WEDNESDAY OCTOBER 20

9.30am – 9.45am

POLYGENY, IMPRINTING, AND MATERNAL EFFECT MODIFY SEX DETERMINATION IN ODD SEX (ODS) MICE

Poirier C1, Qin Y1, Adams CP1, Anaya Y1, Moran JL2, Beier DR2, Nadeau JH3, Bishop CE1

1 Baylor College of Medicine, Houston, United States, 2 Harvard Medical School, Boston, United States,

3 Case Western Reserve University, Cleveland, United States

The transgenic insertional mouse mutation Odd sex (Ods) represents a model for the long-range regulation of Sox9. The mutation causes complete female to male sex reversal by inducing a male specific expression pattern of Sox9 in XX Ods/+ embryonic gonads. We previously described an A/J strain specific suppressor of Ods termed Odsm1A. Here we show that phenotypic sex depends on a complex interaction between the suppressor and the transgene. Suppression can only be achieved if the transgene is transmitted paternally. In addition, the suppressor itself exhibits a maternal effect suggesting that it may act on chromatin in the early embryo. Finally, the genetic background, itself, modulates sex determination.

ORAL PRESENTATION

WEDNESDAY OCTOBER 20

9.45am – 10.00am

EPIGENETIC ANTAGONISM AT THE H19/IGF2 IMPRINTING CONTROL REGION

Engel NI1, Bartolomei MS2

1 University of Pennsylvania, Philadelphia, United States, 2 Howard Hughes Medical Institute, Philadelphia, United States

Genomic imprinting is an epigenetic mechanism resulting in parental-specific monoallelic expression. H19 and Igf2 are two closely linked oppositely imprinted genes, with H19 expressed maternally and Igf2 expressed paternally. The imprinting regulation at these loci depends on a differentially methylated domain (DMD) upstream of the H19 gene that contains four highly conserved CG-rich repeats. These repeats bind CTCF, a protein involved in establishing chromatin boundaries, and this has led to the proposal that the DMD is a maternal-specific boundary regulating the usage of 3’ enhancers common to H19 and Igf2. The insulator function of the DMD is epigenetically regulated, since the boundary is not established on the paternally hypermethylated allele. We used gene targeting technology to replace the wild-type DMD region with a mutated DMD in which 9 CpGs have been replaced within the repeats. The mice generated from mutant ES cell lines were analyzed. We found that the mutations allow normal establishment of the maternal insulator but disrupt paternal identity at the endogenous locus. Mice with the paternally inherited mutation show hypomethylation of the DMD and biallelic expression of H19. They also have a growth retardation phenotype, consistent with a dramatic reduction of Igf2 expression. CTCF binds ectopically to the paternally inherited mutant repeats in vivo, suggesting that a functional insulator is established on that chromosome and blocks Igf2 expression.  Thus, nine CpG mutations within the CTCF binding sites reveal that the two parental-specific roles of the H19 DMD, maintenance of methylation and insulator assembly, are competitive and antagonistic.

ORAL PRESENTATION

WEDNESDAY OCTOBER 20

10.00am – 10.15am

CHARACTERIZATION OF THE MOM2 LOCUS: A SPONTANEOUS MUTATION RESULTING IN SUPPRESSION OF INTESTINAL POLYP FORMATION IN APCMIN MICE. 

Buchberg AM, Silverman K, Koratkar R, Zeskind J, McEntee B, Baran A, Grant S, Siracusa LD

Kimmel Cancer Center, Thomas Jefferson University, Philadelphia, United States

We identified a new modifier locus of ApcMin-induced intestinal tumorigenesis called Modifier of Min 2 (Mom2), from a cross between DBA/2J (DBA) and C57BL/6J (B6) ApcMin/+ mice.  The spontaneous Mom2 mutation most likely occurred in a B6 ApcMin/+ male.  A single mutant Mom2R allele results in an ~90% reduction in small intestinal polyp number and colon polyp incidence in ApcMin/+ mice.  We previously localized Mom2 to a 10 cM region on mouse chromosome 18, distal to the Apc gene.  This region is syntenic with human chromosome 18q, which frequently undergoes loss of heterozygosity (LOH) in several human cancers, including colorectal cancer.  An intercross between congenic DBA.B6 Mom2R/+ animals failed to generate mice homozygous for the Mom2R locus, indicating that Mom2R is a recessive embryonic lethal mutation.  We describe an exclusion strategy for mapping the Mom2 locus, which relies on the lethal phenotype as a method of selection.  Since the Mom2 region is highly conserved between the DBA and B6 strains, we established intersubspecific crosses between congenic DBA.B6 Mom2R/+ and Mus musculus castaneus mice to increase genetic diversity and facilitate the mapping of polymorphic markers to define the boundaries of the Mom2 region.  The results further refine the chromosomal location of Mom2 to a 1 Mb gene-rich region.  In addition, they suggest a probable mechanism of action for Mom2 in suppressing ApcMin-induced tumorigenesis.  Efforts are underway to identify the gene encoding Mom2 and determine its involvement in human colorectal cancer susceptibility.  Research supported in part by NCI PO1 CA72027. 

ORAL PRESENTATION

WEDNESDAY OCTOBER 20

11.15am – 11.30am

REQUIREMENT OF DELTA1 DURING EARLY NEUROGENESIS IN MICE

Przemeck G1, Wünsch K1, Gossler A2, Hrabé de Angelis M1

1 GSF- National Research Center, Institute of Experimental Genetics, Neuherberg/Munich, Germany, 2 Institute of Molecular Biology OE5250, Medizinische Hochschule Hannover, Hannover, Germany

During vertebrate neurogenesis, several Notch ligands and receptors are expressed in distinct as well as overlapping domains suggesting specific functions for and combinatorial interactions between ligands and receptors of the Delta/Notch signal transduction pathway. Lateral inhibition mediated by Notch signaling is required in the central nervous system (CNS) for maintaining neuroepithelial progenitors and their coordinate differentiation into neuronal precursors. Here, we present data on the neuronal phenotype of mouse embryos carrying a null mutation of the Notch ligand Delta1 (Dll1). Whereas various populations of neuronal progenitor cells formed prematurely several specific differentiated cell types are reduced or lost. Together, our findings identify Delta1 as a critical regulator of early neurogenesis in mice and demonstrate different requirements for Delta1 signals in different parts of the nervous system.

ORAL PRESENTATION

WEDNESDAY OCTOBER 20

11.30am – 11.45am

THE ADRENOCORTICAL DYSPLASIA (ACD) MOUSE: MUTATION IN A NOVEL GENE CAUSES A PLEIOTROPIC PHENOTYPE OF UROGENITAL DEFECTS AND CAUDAL DYSGENESIS

Hutz JE1, Keegan CE1, Else T1, Adamska M1, Shah SP1, Kent AE1, Looyenga BD1, Howes JM1, Beamer WG2, Hammer GD1

1 University of Michigan, Ann Arbor, MI, United States, 2 The Jackson Laboratory, Bar Harbor, ME, United States

The adrenocortical dysplasia (acd) mouse is a spontaneous autosomal recessive mutant with developmental defects in organs derived from the urogenital ridge: the kidneys, gonads, and adrenals.  Adrenocortical dysplasia and hypofunction are predominant features in surviving adult mice.  Adult female mutant mice exhibit an overall decrease in ovarian folliculogenesis and males have no mature sperm, leading to infertility.  Fifty percent of adult mutants develop hydronephrosis due to ureteral hyperplasia.  Here we report a splice donor site mutation in a novel gene (Ucd1 – Urogenital and Caudal Dysgenesis-1) in acd mice.  Characterization of Ucd1 transcripts produced in acd mutant tissues shows two abnormal transcripts and lack of detectable normal transcript, consistent with a splicing defect.  Expression of a beta-actin-driven Ucd1 cDNA transgene in acd mutants rescues the observed phenotype.  On the original genetic strain (DW/J), mutants are not born in Mendelian ratios.  Analysis of e14.5 mutant embryos reveals a striking caudal regression phenotype and limb defects.  Whole mount in situ hybridization studies at e11.5 reveal normal expression of Shh and a unique Fgf8 expression domain that is expanded in the dorsal-ventral axis but shortened in the anterior-posterior axis of the apical ectodermal ridge.  Expression of Wnt3a and Dll1 in the tail bud is reduced in some mutants and correlates with the observed variability in phenotypic expression.  The combination of caudal regression, limb malformations, and urogenital defects suggests involvement of UCD1 in early embryo patterning regulated by Wnt and retinoic acid signaling.

ORAL PRESENTATION

WEDNESDAY OCTOBER 20

11.45am – 12.00pm

A GLOBAL PROFILE OF GENE EXPRESSION IN MOUSE PRIMORDIAL GERM CELLS

Abe K1, Sugimoto M1, Kobayakawa S1, Noce T2, Qian Y3, Sharov A3, Ko M3, Mise N1

1 RIKEN BRC, Tsukuba, Japan, 2 Mitsubishi Inst Life Sci, Tokyo, Japan, 3 National Inst on Aging, Baltimore, United States

In developing mammalian early embryos, there exist pluripotent stem cells, giving rise to all somatic cells as well as germ line cells. Primordial germ cell (PGC) is the cell-type appeared first in the germ cell lineage, sharing many features with the embryonic stem cells. Unlike differentiated somatic cells, the PGCs possess ability to erase epigenetic modifications on the genome accumulated during development. Thus PGCs can be regarded as the cells programmed to rejuvenate the genomic status. Despite of this biological importance, molecular nature of the PGCs remains largely unknown. We have established systematic methodologies to analyze PGCs and related embryonic cells: PGCs were purified from transgenic mouse embryos, in which the PGCs were marked by GFP-reporter expression, and cDNA libraries were made with the purified PGCs; transcriptome of the PGCs were explored by EST analyses and microarray. ESTs were classified according to the NIA Mouse Gene Index: the analyzed 18,293 ESTs were clustered into 6,159 transcripts and 4,833 genes, of which 435 were unknown genes. Analysis of EST frequency suggested expression level of each gene, and identified 'signature' genes for PGCs. Principal component analysis of EST frequency showed that PGCs are more similar to blastocyst or Trophoblast stem (TS) cells rather than undifferentiated ES cells or embryonic germ (EG) cells, which were derived from PGCs.

We compared gene expression profiles of ES, EG, PGC, and PGC-like cells derived from ES cells in vitro (Toyooka et al., 2003). EG cells have an expression profile quite similar to that of ES cells (only 1-2% of about 20,000 genes showed significant differences ), although EG cells are different from ES cells in terms of genome-reprogramming activity (Tada et al., 1997). In contrast, PGCs have an expression program distinct from ES cells: for example, about 17% of genes showed differences between ES and E13.5 female PGCs, suggesting that dynamic changes in gene expression occur during establishment of germ cell lineage from undifferentiated stem cells. Comparisons of PGCs with the in vitro-formed PGCs identified a set of genes that characterize PGC development. 

Knowledge and resources obtained in this study should facilitate a wide range of research in germ cell and stem cell biology.

ORAL PRESENTATION

WEDNESDAY OCTOBER 20

12.00pm – 12.15pm

A SENSITIZED MOUSE MUTAGENESIS SCREEN FOR NOVEL GENES REGULATING NEURAL CREST CELL CEVELOPMENT

Watkins-Chow DE, Silver DL, Matera I, Baxter LL, Elliott G, Rivas C, Incao A, Pavan WJ

National Human Genome Institute, NIH, Bethesda, MD, United States

We have established an enhancer screen in mouse to identify novel mutations disrupting development of the neural crest derived melanocyte lineage. Our focused phenotype screen is based on the prediction that ENU induced mutations acting synergistically with the neural crest transcription factor, Sox10, will cause white spotting larger than the typical spot (1.3 +/- 1.0 % of total ventral body area) that we observe in Sox10LacZ/+ heterozygous mutant control animals. In screening 100 first generation pedigrees, we have identified 2 heritable phenotypes that display a significant increase in white spotting on the Sox10LacZ/+ heterozygous background. The BALB/cJ and C57BL/6J mixed genetic background of our cross has facilitated subsequent mapping of these ENU induced mutations to two independent regions of the genome distinct from loci previously implicated in known coat color mutants. Positional cloning is currently underway to identify the mutations responsible for these ENU induced alleles. In addition to the second generation phenotype screen, third generation embryos from 20 lines have been screened for both dominant and recessive embryonic phenotypes that alter expression of the Sox10LacZ reporter gene. Work is underway to confirm heritability of a phenotype selected in this embryonic screen that displays a disorganized pattern of gene expression in the region of the developing dorsal root ganglia. DNA samples from cryopreserved first generation mice provide an invaluable resource for recovering live animals from any identified phenotype and for additional sequence-based, non-phenotype driven approaches to mutation detection.

ORAL PRESENTATION

WEDNESDAY OCTOBER 20

8.00pm – 8.15pm

CHROMOSOMAL INSTABILITY LEADS TO BIRTH DEFECTS AND CANCER PREDISPOSITION IN RECQL4-DEFICIENT MICE: A MODEL OF ROTHMUND-THOMSON SYNDROME

Mann MB, Luo G

Department of Genetics, Case Western Reserve University and Ireland Cancer Center, University Hospitals of Cleveland, Cleveland, United States

Rothmund-Thomson Syndrome (RTS) is a genetically heterogeneous autosomal recessive disorder.  Type-II RTS patients have RECQL4 mutations, abnormalities of the skin and skeleton and increased risk for developing osteosarcoma.  Karyotypic analyses of type II patient-derived cells demonstrate unusually high frequencies of chromosomal aberrations including aneuploidy.  While chromosomal instability likely contributes to the patients increased susceptibility to osteosarcoma, the nature of genomic instability induced by RECQL4 deficiency and the disease etiology of RTS remains elusive.  We generated Recql4-deficient mice to study the role of Recql4 in maintaining genomic stability.  Recql4-deficient mice recapitulate the hallmark features of RTS with skeletal abnormalities of the palate and limbs, ultraviolet light induced hyperpigmentation associated with premature photoaging and increased susceptibility to osteosarcoma and lymphoma.  Cells from Recql4-deficient mice display aneuploidy and chromosomal instability with an increased incidence of anaphase lagging chromosomes and high rates of spontaneous micronuclei formation.  Remarkably, in Recql4-deficient cells, aneuploidy manifests in the absence of defects in spindle checkpoints or aberrant centrosomes hyperamplification.  Further analysis revealed that premature centromere separation induced unequal chromosome partitioning is the underlying mechanism leading to aneuploidy in Recql4-deficient cells. Thus, we conclude that Recql4 has a role in mitotic chromosome segregation and maintenance of genome stability. This is the first documentation of this mechanism for the production of aneuploidy in mammalian cells. Our findings also provide the first evidence that in mammals, as in unicellular organisms, the RecQ family of DNA helicases have an important role in chromosome segregation.

ORAL PRESENTATION

WEDNESDAY OCTOBER 20

8.15pm – 8.30pm

PINKIE, THE FIRST VIABLE GERMLINE HYPOMORPH ALLELE OF RETINOID X RECEPTOR ALPHA, REVEALS AN IMPORTANT ROLE FOR RXRa IN TH2 DEVELOPMENT

Du X1, Bigby T2, Tabeta K1, Janssen E3, Crozat K1, Mann N1, Beutler B1

1 The Scripps Research Institute, La Jolla, United States, 2 VA San Diego Healthcare System and the University of California, San Diego, United States, 3 La Jolla Institute for Allergy and Immunology, La Jolla, United States

Pinkie is a recessive ENU-induced mutation, homozygotes develop premature graying and subsequently, progressive alopecia.  Histologically, the skin is characterized by the destruction of hair follicle architecture, formation of dermal cysts and appearance of black papules throughout the ventral skin late in life.  The Pinkie mutation was mapped to the retinoid X receptor alpha (Rxra ) locus (chromosome 2) on 700 meioses and positionally cloned.  As a component of the nuclear receptors for retinoic acid and other bioactive molecular ligands, RXRa plays an irreplaceable role during embryogenesis and a null allele of RXRa inevitably causes embryonic lethality. Pinkie, which results from the amino acid substitution N273I in the ligand-binding domain, is the first viable germline hypomorphic mutation of RXRa. Transfection studies suggest that the abnormal protein retains only 10 to 20% of the activity of the wild type receptor. Pinkie homozygotes develop a progressive phenotype, part of which entails a severe, age-dependent imbalance of Th1/Th2 differentiation upon antigenic challenge. When immunized with ovalbumin (OVA), pinkie homozygotes displayed markedly decreased IgG1 levels in serum as compared with normal C57BL/6 mice; in vitro OVA re-stimulation of pinkie splenocytes causes greatly enhanced IFN-g production and decreased IL-4 production. Upon MCMV infection, pinkie homozygotes showed significantly higher type I interferon levels in serum than normal C57BL/6 mice. These data collectively imply that RXRa is required for Th2 development. This requirement might reflect an influence of RXRa on the expression of pertinent transcription factors (e.g., GATA-3, T-bet), cytokines (IL-4, INF-gamma) and/or cytokine receptors that are involved in Th1/Th2 differentiation. RXRa represents a likely interface between the nutritional environment and the immune status of the host. Consumption of retinoids and vitamin D may therefore be expected to influence the development of the allergic/atopic (Th2) phenotype, and provide a plausible explanation for the increasing frequency of atopic disease observed in developed countries.

ORAL PRESENTATION

WEDNESDAY OCTOBER 20

8.30pm – 8.45pm

MAINTAINING TELOMERE AND CHROMOSOME STABILITY REQUIRE THE RAD51D HOMOLOGOUS RECOMBINATION GENE

Smiraldo PG1, Tarsounas M3, Gruver AM1, Osborn JC1, Munoz P2, Class A3, Blasco MA2, West SC3, Pittman DL1

1 Medical College of Ohio, Toledo, United States, 2 Spanish National Cancer Center, Madrid, Spain, 3 Cancer Research UK, London Research Institute, Hertfordshire, United Kingdom

Homologous recombination (HR) is a double-strand break repair pathway required for resistance to DNA damage and maintaining genomic integrity. In mitotically dividing cells, the primary proteins involved in HR repair are RAD51 and the five RAD51 paralogs, RAD51B, RAD51C, RAD51D, XRCC2, and XRCC3. By investigating Rad51d knockout mice and mammalian cell lines, we demonstrate that RAD51D is critical for maintaining chromosome stability and telomere protection. Homozygous mice defective for Rad51d die prior to birth, and cells derived from mutant embryos fail to proliferate in culture, likely as a result of genomic instability and p53 activation. A p53 deletion is sufficient to extend the lifespan of Rad51d-deficient embryos by up to 6 days and rescue the cell lethal phenotype. Rad51d-/- Trp53-/- mouse embryo derived fibroblasts were sensitive to DNA-damaging agents, particularly interstrand crosslinks, and exhibited extensive chromosome instability including aneuploidy, chromosome fragments, deletions, and complex rearrangements. Additionally, loss of Rad51d resulted in increased centrosome fragmentation and reduced levels of radiation-induced RAD51-focus formation. Spontaneous frequencies of sister chromatid exchange (SCE) were not affected by the absence of Rad51d, but SCE frequencies did fail to be induced upon challenge with the DNA cross-linking agent mitomycin C. The RAD51D protein was recently found to localize at telomeres.  Decreased telomere lengths and increased levels of anaphase bridging and telomere fusions were observed in primary Rad51d-deficient cells. Similarly, inhibition of RAD51D synthesis by RNA interference in human cells resulted in increased telomere fusions.  These findings support a crucial role for RAD51D in maintaining the integrity of the mammalian genome.

ORAL PRESENTATION

WEDNESDAY OCTOBER 20

8.45pm – 9.00pm

STUDYING THE ROLE OF THE KINESIN-LIKE MOTOR PROTEIN, EG5, IN TUMORIGENESIS, GENOME INSTABILITY, AND AGING.

Castillo A, Justice MJ

Baylor College of Medicine, Houston, United States

Formation and stabilization of the mitotic spindle is dependent on the kinesin-like motor protein Eg5.  Chemical and antibody inhibition have shown that disruption of Eg5 function leads to mitotic arrest, monopolar spindles, and aneuploidy.  In cultured post-mitotic axons, inhibition can disturb axonal outgrowth. Additionally, Eg5 has been shown to be regulated by the centrosomal kinase, Aurora-A/STK15, which is often over expressed in numerous cancers. We showed that Eg5 is up regulated in B-cell leukemias from AKXD recombinant inbred mice with proviral insertions at lymphoid viral insertion site (Lvis1).  Transgenic mice overexpressing Eg5 in lymphoid tissues were generated and aged to determine if misexpression of Eg5 can lead to tumor development.  Aged mice develop multiple tumor types, but also display distended bladders, dermatitis, and chronic piloerection.  Based on Eg5’s involvement in cell cycle mechanics and post-mitotic neurons, our observations suggest a premature aging syndrome with potential skin and neurogenic bladder problems.  In parallel, Eg5 loss of function mice were generated from genetrapped ES cells and display early embryonic lethality. Conditional targeting of Eg5 will circumvent this lethality and be used to study neuronal and tissue specific functions.

ORAL PRESENTATION

THURSDAY OCTOBER 21

9.00am – 9.15am

A TLR2-RESPONSIVE LIPID EFFECTOR PATHWAY PROTECTS MAMMALS AGAINST GRAM-POSITIVE BACTERIAL SKIN INFECTIONS

Georgel P, Crozat K, Lauth X, Sovath S, Hoebe K, Du X, Rutschmann S, Jiang Z, Bigby T, Nizet V, Beutler B

1 TSRI, La Jolla, United States, 2 UCSD, San Diego, United States

Surface epithelia constitute the first line of defense against pathogens.  This defense may depend both upon barrier function and upon specific microbicidal effector molecules.  In the course of our germline mutagenesis effort, we isolated  flake (flk), an ENU-induced recessive mutation of C57BL/6 mice, which causes progressive alopecia and chronic exfoliative dermatitis.  These features appeared at weaning age and were more pronounced in older animals. Visible disruption of epidermal integrity and spontaneous skin infections requiring antibiotic therapy prompted us to examine the integrity of innate immune function in these mice. We utilized luminescence-tagged bacterial strains to monitor the progression of dermal infections in vivo and we showed that the flk mutation impairs the clearance of subcutaneous infections by Streptococcus pyogenes and Stapylococcus aureus:  Gram-positive pathogens that elicit innate immune responses by activating Toll-like receptor 2 (TLR2).  Positional cloning, sequencing and genetic complementation analysis revealed that flk is a novel allele of the steroyl CoA desaturase 1 gene (Scd1).  Flake homozygotes are unable to synthesize the monounsaturated fatty acids (MUFA) palmitoleate (C16:1) and oleate (C18:1), both of which are markedly cytotoxic to Gram-positive (but not Gram-negative) bacteria. Intradermal MUFA administration in S. aureus-infected mice improves bacterial clearance, and in normal mice, transcription of Scd 1- a gene with numerous NF-kB elements in its promoter - is strongly and specifically induced by TLR2 signaling.  These observations reveal the existence of a regulated, lipid-based antimicrobial effector pathway in mammals, and suggest new approaches to the treatment or prevention of Gram-positive bacterial infections.

ORAL PRESENTATION

THURSDAY OCTOBER 21

9.15am – 9.30am

SEX DEPENDENT SUSCEPTIBILITY PATTERN TO LISTERIA MONOCYTOGENES INFECTION IS MEDIATED BY DIFFERENTIAL IL-10 PRODUCTION

Kalaydjiev S1, Pasche B2, Franz TJ1, Kremmer E3, Gailus-Durner V4, Fuchs H4, Hrabe de Angelis M1, Busch DH1, Lengeling A2

1 Institute for Medical Microbiology, Immunology, and HygieneTechnical University Munich, Munich, Germany, 2 Junior Research Group Infection Genetics, German Research Centre for Biotechnology (GBF), Braunschweig, Germany, 3 Institute of Molecular Immunology, GSF - National Research Center for Environment and Health, Munich, Germany, 4 German Mouse Clinic, Institute of Experimental Genetics, GSF - National Research Center for Environment and Health, Neuherberg, Germany

Infectious diseases are a major cause of morbidity and mortality worldwide. It is well established that the sex of a host can significantly affect susceptibility to infection. A number of reports have shown that patients of one sex are more likely to get an infectious disease, and gender is often referred to as risk factor for the severity and outcome of an illness. The underlying molecular mechanisms of this predisposition are largely unknown.

Listeria monocytogenes (L.m.) is an intracellular Gram-positive bacterium that causes diseases in immunocompromised individuals and pregnant women, often with deleterious consequences for the fetus. It is also one of the most widely used pathogens in experimental mouse studies that provided the basis for establishing major paradigms in contemporary immunology.

Although most mouse infection models demonstrate higher resistance in females, we made the unexpected observation that female mice are significantly more susceptible to L.m. infection compared to their male littermates. Surprisingly, increased severity of infection in females is correlated with elevated IL-10 plasma levels. Experiments using IL-10-knockout mice, in which no differences between the susceptibility of males and females to L.m. infection could be detected, confirmed the crucial role of this immunosuppressive cytokine for the outcome of disease.

Our findings might be of substantial clinical importance, since similar sex differences in infection with L.m. and other intracellular pathogens have been reported in humans.

This work is supported by the National Genome Research Network NGFN (01GR0102-KB-P5T0513) and by the EU project EUMORPHIA (QLG2-CT-2002-00930).

ORAL PRESENTATION

THURSDAY OCTOBER 21

9.30am – 9.45am

MECHANISM OF RESISTANCE TO PLASMODIUM CHABAUDI IN MICE IS MEDIATED THROUGH THE RED CELL AND A TOTALLY SYNERGISTIC NON-ERYTHROCYTIC PATHWAY

Lin E, Marshall V, Burt RA, Foote SJ

The Walter & Eliza Hall Institute, Parkville, Australia

P. chabaudi infection in mice is a model for the severe anaemia seen in falciparum infections. Inbred mice strains are differentially susceptible to malaria. We have mapped three loci involved in outcome to this infection and have generated mice reciprocally congenic for these loci on several strains of mouse. These congenic animals have phenotypes different from their wildtype parents. Parasitaemias are more informative than clinical outcome in differentiating between the congenic animals. In one case, a congenic animal carrying a locus predicted to encode susceptibility was much more resistant than even the “resistant” animal. Analysis of the ability of red cells from congenic mice to sustain the growth of the parasite has been compared to wildtype using a novel comparative in vivo parasite survival assay. Mice congenic for the lmr2 locus demonstrate a curious effect where the origin of the animal receiving the red cells modulates the effect of the red cells. This means that a red cell factor interacts with another, non-red cell factor in a completely synergistic fashion. Other haematological parameters also implicate the origin of the red cell in susceptibility to disease. Another congenic, for the char3 region on chromosome 17 displays a more resistant phenotype, despite the donor region coming from a susceptible animal. This line produces more lymphocytes at around the time of peak parasitaemia. These are interesting cells expressing cell surface markers from B, T and macrophage cell lineages.

ORAL PRESENTATION

THURSDAY OCTOBER 21

9.45am – 10.00am

RISK ASSESSMENT OF MOUSE HEPATITIS INFECTION (MHV) VIA IN VITRO FERTILIZATION AND EMBRYO TRANSFER WITH INTACT AND LASER MICRODISSECTED OOCYTES.

Peters D1, Marschall S1, Mahabir E2, Schmidt J2, Hrabé de Angelis M1

1 Institute of Experimental Genetics, GSF-National Research Center for Environmental and Health, Neuherberg, Germany, 2 Department of Comparative Medicine, GSF-National Research Center for Environmental and Health, Neuherberg, Germany

Cryobanking of mouse mutants and allocation of specific pathogen free (SPF) animals is of particular importance for the scientific community. Routinely used rederivation methods have to be enhanced by new assisted reproductive technologies (e.g. Zona dissection, ICSI) but these methods need to be investigated with regard to possible pathogen transmission.

The most common viral pathogen in mouse facilities, the Mouse Hepatitis Virus (MHV), was chosen to ensure a well-defined washing procedure used for in-vitro fertilization (IVF) and to estimate the risk increase by zona microdissection with PALM microbeam laser. Zona pellucida (ZP) intact and laser dissected oocytes used for the IVF were scheduled in three experiment groups:

oocytes in IVF were exposed to virus during fertilization and culture without washing procedure

oocytes in IVF were exposed to virus during fertilization and culture with washing procedure

laser microdissected oocytes in IVF were exposed to virus during fertilization and culture with washing procedure

Resulting embryos were transferred to SPF recipients, sera from recipients and offspring were tested for MHV-antibodies using ELISA. All used media were tested for presence and infectivity of MHV.

In experiment 1 nearly all recipients showed immune response whereas in experiment 2 and 3 neither in recipients nor in offspring antibodies against MHV could be detected.

Appropriate washing procedures applied in IVF protect against transmission of MHV even if the zona pellucida – the shield of the embryo – is disrupted.

ORAL PRESENTATION

THURSDAY OCTOBER 21

2.00pm – 2.15pm

JENNA : A NOVEL ENU HYPERACTIVE MOUSE MUTANT WITH AN ENHANCED ACOUSTIC STARTLE RESPONSE

Keays DA 1, Nolan P 2 , Oliver P 3, Tian G 4, Fullerton J 1, Rees M 5, Harvey RJ 6, Cowan NJ 4, Davies KE 3, Flint J 1

1 Wellcome Trust Centre for Human Genetics, Oxford University, Oxford, United Kingdom,

2 MRC Mammalian Genetics Unit, Harwell, Didcot, United Kingdom, 3 Department of Human Anatomy and Genetics, Oxford University, Oxford, United Kingdom, 4 Department of Biochemistry, New York University Medical Center, New York, United States, 5 Swansea Clinical School, University of Wales Swansea, Singleton Park, Swansea, United Kingdom, 6 Department of Pharmacology, The School of Pharmacy, London, United Kingdom

The identification of the genes and molecular pathways that are associated with human neurobehavioral disorders has been difficult because they are heterogeneous, complex and likely involve multiple genetic loci. Mouse models that are based on intermediate signatures of psychiatric disease hold promise as an approach to dissecting the molecular basis of these disorders. N-ethyl-N-nitrosourea (ENU) mutagenesis in the mouse, when coupled to a battery of sensitive behavioural screens, is an effective way of creating and identifying novel mouse behavioral mutants. We screened 9,000 F1 mice in a dominant ENU screen for locomotor activity. We report on the identification of Jenna a hyperactive mouse mutant with an enhanced acoustic startle response. Histological analysis revealed that the Jenna mouse line also has an enlarged superior colliculus, a region of the brain implicated in regulating the startle response. We undertook positional cloning and identified linkage to a 1.2 mb region on distal chromosome 15. We have sequenced all 39 predicted and known genes in this region and identified a mutation in alpha 1 tubulin. This mutation results in a serine to glycine substitution in a highly conserved GTP binding domain of this cytoskeletal protein. We show that this results in a five-fold reduction in GTP binding affinity. We report on the screening of this gene in a cohort of humans with startle disease (hyperekplexia) and attention deficit disorder. This is the first report of alpha tubulin being associated with abnormal behavioral traits in vertebrates.

ORAL PRESENTATION

THURSDAY OCTOBER 21

2.15pm – 2.30pm

ESTIMATING THE CONTRIBUTION OF GENETIC, DIET AND GENDER EFFECTS TO THE PHENOTYPIC VARIATION OBSERVED IN METABOLIC SYNDROME-RELATED TRAITS BETWEEN A/J AND C57BL/6J MICE

Sinasac DS, Nadeau JH

Case Western Reserve University, Cleveland, United States

Metabolic syndrome (MetS) refers to a constellation of cardiovascular risk factors that include obesity, hypertension, dyslipidemia and altered glucose homeostasis leading to diabetes.  Male C57BL/6J mice fed a high fat/high sucrose (HF/HS) diet are a model of MetS.  To evaluate the contribution of genetic background, diet and gender on the MetS phenotype in this model, 21 MetS-related traits were examined in male and female A/J and C57BL/6J.  At 5 weeks of age, 30 mice per group were fed either the HF/HS or maintenance diet for 16 weeks.  After 15 weeks, a fasted IPGTT was performed.  At 16 weeks, body weight/length measures, blood and liver were collected following a fast.  Plasma ALT, AST, cholesterol, glucose, insulin, NEFA and triglycerides as well as liver weight and triglyceride content were measured.  Data were analyzed using a factorial ANOVA design with planned comparison tests per trait and an experiment-wide, modified Bonferroni correction was applied.  Fourteen traits showed significant background effects, 17 significant diet effects, and 16 significant gender effects.  Interactions between factors were also noted for 13 traits.  Comparing A/J and C57BL/6J on the HF/HS diet, males showed significant differences in 12 traits, whereas females showed difference in only six.  These results indicate that the genetic differences between A/J and C57BL/6J, when exacerbated by the HF/HS diet, explains a substantial portion of the variation observed in a subset of MetS-related traits in males.  Genetic mapping studies are currently underway to examine these traits in the newly generated B6-ChrA chromosome substitution strain panel of mice.

ORAL PRESENTATION

THURSDAY OCTOBER 21

2.30pm – 2.45pm

MUTANT ANALYSIS REVEALS WHIRLIN AS A DYNAMIC ORGANISER IN THE GROWING HAIR CELL STEREOCILIUM

Kikkawa Y1, Mburu P1, Morse S1, Kominami R2, Townsend S1, Brown S1

1 MRC, Harwell, United Kingdom, 2 Tokyo Metropolitan Institute of Medical Science (Rinshoken), Tokyo, Japan

Stereocilia on the surface of hair cells are vital for the process of auditory transduction. Stereocilia develop in bundles with a regular staircase pattern, whose actin core is organised such that the barbed ends of actin filaments are located at the stereocilia tips where there is a continuous cycle of renewal of actin filaments by addition of actin monomers. What little we know of the molecular processes that control actin polymerisation and stereocilia growth come from studies of mouse mutants with defects in stereocilia development, including the shaker2 (myosin XVa) and whirler mutants. The PDZ protein whirlin is known by virtue of the whirler mutation to be involved in the process of stereocilia organisation (Mburu et al, Nature Genetics 2003). We have investigated the expression of whirlin and myosin XVa in developing stereocilia in wild-type and whirler and shaker2 mutants. Whirlin localises to the actin-free zone at the tips of stereocilia. Expression of whirlin is dynamic during stereocilia growth demonstrating an ordered appearance and fade-out across the stereocilia rows and revealing a novel molecular gradation of process traversing the developing stereocilia bundle. Fade-out of whirlin in inner hair cells precedes that of outer hair cells consistent with the earlier maturation of inner hair cell stereocilia. In myosin XVa mutants in which stereocilia are shortened, whirlin is still expressed in the stereocilia tips but fade-out is accelerated. The data indicate that whirlin is a critical and dynamic organiser for proteins involved in stereocilia elongation and actin polymerisation.

ORAL PRESENTATION

THURSDAY OCTOBER 21

2.45pm – 3.00pm

DISCOVERY OF A GENE CAUSING HUMAN CONGENITAL DIAPHRAGMATIC DEFECT AND PULMONARY HYPOPLASIA BY CHARACTERIZATION OF AN ENU-INDUCED MOUSE MODEL

Ackerman KG1, Vargas SO2, Beier DR1

1 Brigham and Women's Hospital/Harvard Med Sch, Boston, United States, 2 Children's Hospital/Harvard Med Sch, Boston, United States

We are screening embryonic mice for recessive ENU-induced phenotypes similar to human congenital defects. The spectrum of abnormalities found to date is remarkably varied, and many are similar to human malformation syndromes, including, for example, models of asphyxiating thoracic dystrophy, Robin sequence, cleft palate, polycystic kidney disease, and congenital heart defects.

One mutant phenotype includes pulmonary hypoplasia with lobar agenesis and a diaphragmatic muscularization defect, which is similar to the abnormalities found in the often devastating but poorly understood human disorder congenital diaphragmatic hernia (CDH).  Positional cloning revealed that this line carries a hypomorphic mutation of the transcriptional co-factor Fog2. This result itself validates the utility of phenotype-driven analysis, as a requirement for Fog2 in diaphragm and lung development has not been previously recognized.

A role for this gene in CDH was suggested by the association of the disorder with chromosomal translocations at 8q22.3, where human FOG2 is located. To test this, sequencing was done on 30 autopsy samples, and a heterozygous base change resulting in a premature stop codon was identified in one case.  Analysis of the parents revealed that this was a de novo mutation, which strongly implicates FOG2 as causal for CDH in the affected child. No changes at this site were found in 400 normal controls.  

Our identification of FOG2 as the first causal gene for developmental diaphragmatic defects and congenital pulmonary hypoplasia very clearly demonstrates the utility of ENU mutagenesis as a tool for finding candidate genes for human developmental defects. 

ORAL PRESENTATION

THURSDAY OCTOBER 21

3.00pm – 3.15pm

ENU-INDUCED MOUSE ENAMELIN (ENAM) MUTANTS AS MODELS FOR DIFFERENT CLINICAL SUBTYPES OF HUMAN AMELOGENESIS IMPERFECTA (AI).

Masuya H1, Shimizu K2, Sezutsu H1, Sakuraba Y1, Nagano J1, Shimizu A1, Fijimoto N1, Ishijima J1, Kaneda H1, Kobayashi K1, Maeda T2, Gondo Y1, Noda T1, Wakana S1, Shiroishi T1

1 RIKEN GSC, Tsukuba, Japan, 2 Nihon Univ, Matsudo, Japan

Amelogenesis imperfecta (AI) is a common group of inherited defects of dental enamel formation, which exhibit marked genetic and clinical heterogeneity. The genetic basis of this heterogeneity is still poorly understood. Enamelin is a extracellular matrix molecule that is one of components of enamel, and is affected in human AI (AIH2). We isolated three ENU-induced dominant mouse mutations, M100395, M100514 and M100521, which showed AI-like phenotypes in their incisors and molars. They were mapped closely to the genes encoding enamelin (Enam) and ameloblastin (Ambn) on chromosome 5. Sequence analyses revealed that all three mutations have base substitutions in the Enam. M100395 and M100514 have S to I and E to G (putative) missense mutations located near the N-terminal of enamelin protein. M100521 has a mutation at splicing donor site of intoron 4, which results in a flame shift and gives rise to the stop codon. We demonstrated that degradation of transcript of the mutant allele occurred in M100521. Thus, it appeared that M100521 is a loss-of-function type mutation of the Enam. Heterozygotes of M100521 showed hypomaturation-type AI-like phenotype in the incisors. It must be occurred by the haploinsufficiency of Enam. Homozygotes of M100521 showed complete loss of enamel in the incisor and the molar. By contrast, heterozygotes of M100395 and M100514 showed severe braking of the enamel surface, which is similar to local hypoplastic AI. Thus, we report here that the Enam gene is essential for amelogenesis, and that the mice with different point mutations at Enam represent different clinical subtypes of AI. These mouse mutants provide good animal models for study of variation of human AI.

POSTER 1 - MOUSE GENOME INFORMATICS: AN INTEGRATED RESOURCE FOR MOUSE BIOLOGICAL DATA

Reddy TBK, Blake JA, Bult CJ, Ringwald M, Richardson JE, Kadin JA, Eppig JT

The Jackson Laboratory, Bar Harbor, United States

The Mouse Genome Informatics (MGI) system provides an integrated and curated resource for information about the laboratory mouse. MGI represents extensive primary experimental data sets and also provides a consensus view of the biology of the laboratory mouse. All data associations are supported with evidence and citations. From genotype to phenotype, this resource integrates information about sequences, maps, gene function, expression analyses, alleles, strains and mutant phenotypes. Comparative mammalian data are also presented in the form of comparative maps and mammalian orthology relations.

MGI acquires data by direct data loads from other databases, from labs that generate data and from published literature. The indexing and nomenclature process ensures proper association of a publication with genes and mutant phenotypes described and ensure the use of standard nomenclature. We co-curate data with NCBI, SWISSPROT and other resources to ensure proper Sequence to Gene Marker associations. MGI makes use of several controlled vocabularies as tools for grouping and querying the various datasets comprising the database. The GO annotation system is used to describe a gene product in terms of the function it performs, the process that the function is part of, and the cellular compartment or complex in which it is found. The mammalian Phenotype Ontology is being developed and applied to describe phenotypes of mutant and genetically engineered mice. The mouse anatomical dictionaries are used to consistently annotate expression data, clone libraries and phenotypes.

The combined efforts at MGI produce a community resource for researchers, providing information on official nomenclature for genes/strains, extensive gene characterization, comparative mapping of orthologus genes in other mammals, mouse gene expression data, mouse tumor biology data, sequence to gene associations, strain polymorphisms, mutant allele and phenotype descriptions. Various reports are provided through the ftp server for computational users. These concerted efforts to integrate mouse genome data with mouse biology make the MGI resource unique and support the use of mouse as a model for human disease and biology.

MGI is supported by grants HG00330 and HG02273 from NHGRI, HD33745 from NICHD and CA89713 from NCI.

POSTER 2 - A FUNCTIONAL GENOMICS APPROACH TO THE INVESTIGATION OF HUMAN 6P GENE FUNCTION

Arkell R 1, Bogani D 1, Kuar K 2, Davies J 1, Mirza G 2, Willoughby C 1, McKeone R 1, Denny P 1, Ragoussis J 21 Mammalian Genetics Unit, Oxfordshire, United Kingdom, 2 Wellcome Trust Centre for Human Genetics, Oxford, United Kingdom

The telomeric region of human 6p contains genes required for many aspects of embryonic development.  Monozygosity for this region is associated with a variety of congenital abnormalities that include orofacial clefting and other craniofacial defects, CNS defects, eye abnormalities, ear abnormalities and deafness, heart, kidney and limb defects.  We have used functional genomics to generate mouse models that identify the 6p genes involved in the development of these structures.  To achieve this, we have taken advantage of a large region of synteny between human 6p and the proximal portion of mouse chromosome 13.  Mice monozygous for a region of chromosome 13 (corresponding to 6p25 and 6p22) are viable and fertile and replicate features of the human 6p deletion syndromes.  In addition, homozygous loss of function of the Foxq1 gene that resides within the deleted material produces mice with a glossy coat.  This has enabled a two-generation region specific screen for recessive, ENU induced mutations to be carried out.  Screening of 1729 pedigrees has identified 11 heritable mutations that are linked to Foxq1.  All of the mutations are embryonic lethal and phenotypes include clefting and other craniofacial defects, holoprosencephaly, heart, kidney and lung defects.  Recombination mapping has been used to further refine the location of the mutations within the deleted area and mutation detection in candidate genes is underway for the majority of the mutant lines.  The location and or identity of the mutated genes and associated phenotypes will be discussed with reference to the human 6p deletion syndromes.

POSTER 3 - DRIVING ALLELES AT THE DISTORTER LOCUS IN THE OM MEIOTIC DRIVE SYSTEM PREDATE THE DIVERGENCE OF THE MUSCULUS, SPICILEGUS, AND SPRETUS LINEAGES

Doherty HE 1, Kim K 2, Bell TA 1, De la Casa-Esperon E  3, Ideraabdullah F 1, Pardo-Manuel de Villena F 1

1 Department of Genetics, University of North Carolina Chapel Hill, Chapel Hill, NC, United States, 2 Curriculum in Genetics and Molecular Biology University of North Carolina-Chapel Hill, Chapel Hill, NC, United States, 3 Fels Institute for Cancer Research and Molecular Biology, Temple University School of Medicine, Philadelphia, PA, United States

In mammals, meiotic drive is achieved by preferential segregation of one chromosome over another to the functional product of female meiosis.  Two components are required in all meiotic drive systems, a Responder and a Distorter.  The Distorter is the effector in the system, and “drives” alleles at the Responder.  The goal of this project is molecular identification of the Distorter and the Responder in the Om (Ovum mutant) meiotic drive system.  The Responder maps to the Om locus (Chr 11) while the Distorter is located 2.5 Mbp distal to Om.  We have defined a 315kb candidate interval for the Distorter locus and generated a haplotype map built by sequencing 50 fragments of 450bps.  We compiled the meiotic drive phenotypes for 18 wild-derived and classical inbred strains (representing three species of the Mus genus) with each strain classified as driving or non-driving.  Based on the presence of drive in Mus spicilegus, we conclude that driving alleles have been segregating in the Mus genus for over 1.5 million years.  Within Mus musculus, driving alleles have been observed only in Mus musculus domesticus strains.  Therefore, allelic variation at the Distorter is an example of ancient variants in the Mus musculus lineage that has been maintained in the population over millions of generations.  Unsurprisingly given the age of the mutation, the occurrence of extensive recombination creates special challenges to haplotype mapping.

POSTER 4 - IDENTIFICATION OF MURINE HOMOLOGUES OF THE HUMAN REPEAT D4Z4, DELETIONS OF WHICH ARE CAUSALLY ASSOCIATED WITH FACIOSCAPULOHUMERAL MUSCULAR DYSTROPHY

Clapp J 1, Fantes J 2, Hewitt JE 1

1 Institute of Genetics, University of Nottingham, Nottingham, United Kingdom, 2 MRC Human Genetics Unit, Western General Hospital, Edinburgh, United Kingdom

Facioscapulohumeral muscular dystrophy (FSHD) is a dominant human neuromuscular disorder caused by a deletion within the tandem repeat array D4Z4 on human chromosome 4q35: affected individuals carry <12 repeat units on one D4Z4 allele. The mechanism whereby this mutation causes FSHD is still debated, but current models favour epigenetic effects such as de-repression of 4q35 genes and altered methylation. D4Z4 contains an open reading frame (ORF) potentially encoding a homeodomain protein. However, by DNA hybridization techniques, homologues have only been identified in higher primates supporting the hypothesis of a non-coding function of D4Z4. Here we show that bioinformatics analysis of the draft mouse genome sequence enabled us to identify a potential mouse homologue (mD4Z4). The sequence and organisation of mD4Z4 has been confirmed by a combination of PCR, pulsed field gel electrophoresis and FISH. The repeat unit size in mouse is 4.9kb, compared to 3.3kb in human. The repeat contains an ORF of 2025bp, potentially encoding two homeodomains with 55% aa similarity to those encoded by human D4Z4. There is no significant sequence conservation outside the coding region. The mD4Z4 repeat units are arranged in large tandem arrays, with FISH analysis suggesting a single chromosome locus. A rat homologue of D4Z4 has also been identified, potentially encoding a homeodomain protein with 66% aa similarity to the mouse. The conservation of the ORF between human, mouse and rat, and the identification of several mouse ESTs suggest a coding function for D4Z4. Our findings also provide potential for the development of a mouse model of FSHD.

POSTER 5 - HUMOT (HUMAN AND MOUSE ORTHOLOGOUS ANNOTATION): NAMING OF ORTHOLOGOUS GENES FROM THE HUMAN AND MOUSE TRANSCRIPTOME PROJECTS

Maltais LJ 1, Blake JA 1, Bult CJ 1, Wright MW 2, Wain H 2, Povey S 2, Eppig JT 1

1 The Jackson Laboratory, Bar Harbor, ME, United States, 2 University College London, London, United Kingdom

Identifying all full length mRNA transcripts in both mouse and human through international collaborative efforts is generating an abundance of data. The RIKEN mouse cDNA sets of 60,000 clones, representing 21,000 unique mouse genes and the 42,000 annotated human cDNAs from the Odaiba H-Invitational, represent approximately 23,000 unique human genes. These large cDNA sets provide an opportunity to assign nomenclature to those genes which do not have, as yet,  meaningful symbols and names.  Both the RIKEN set and the Odaiba set are already annotated with data suggesting function, or sequence similarity with other species.  The Mouse Genomic Nomenclature Committee (MGNC) and the HUGO Gene Nomenclature Committee (HGNC), with a long history of working together to provide standardized nomenclature for orthologous genes, will utilize these mouse and human annotated cDNA sets to rapidly identify and name orthologous genes. To date there are approximately 19,000 known genes in both species, which have a meaningful and standardized nomenclature assigned to them. These have been named primarily based on data from the scientific literature, author submissions, and databases, such as, EntrezGene, SwissProt, and other organism orthologs. There is a community need, in both mouse and human, to have the remaining unnamed genes identified and their relationships to genes in other species established. HUMOT will fulfill this need.

Supported by NIH HG00330 and Welcome Trust 072955

POSTER 6 - EXPERIMENTAL CONFIRMATION OF ENHANCER ACTIVITY AT LVIS1, A CONSERVED GENE-DISTANT ANCIENT REGULATORY SITE.

Weiser K, Justice MJ

Baylor College of Medicine, Houston, United States

The recent availability of genome sequence from many species has opened myriad possibilities for comparative genomics.  This allows us not only to chronicle genomic evolution with a focus on genes, but also to more effectively identify sequences distant from genes that are important for the proper development and function of the organism.  Amid an ever-changing sea of “junk” DNA sequences, there are segments that are conserved through millions of years of evolution, yet are distant from genes. These sequences are likely to be enhancers, repressors, locus control regions, matrix attachment regions, insulators, and some may have functions yet to be recognized.  We have identified one of these conserved gene-distant sites through a viral insertional mutagenesis screen for leukemia in AKXD mice.  This site was termed Lymphoid Viral Insertion Site 1 (Lvis1).  Lvis1 is the most frequent site of insertion leading to leukemia in the AKXD lymphoid tumors.  Analysis of mRNA of flanking genes by Northern blot and quantitative real-time PCR revealed two genes that are up-regulated in Lvis1 tumors, Hex (an early expressed divergent homeobox gene) and Eg5 (a kinesin-related motor protein).  These genes are 50-100kb upstream of Lvis1 and transcription from genes equivalent distances downstream is unaffected.  Sequence comparisons of Lvis1 reveal a striking evolutionary conservation with 92.5% sequence identity between Human and chicken over  226 basepairs. DNaseI hypersensitivity assays and mobility shift assays reveal factor binding within Lvis1. Luciferase assays in cell culture demonstrate enhancer activity of Lvis1.  These data imply a mechanism of gene regulation with strict sequence requirements and a mechanism for targeting specific genes for regulation at a distance.  Recent work has identified 156 intergenic sites longer than 199bp with 100% evolutionary conservation between human, rat, and mouse with enrichment of these sites near early developmental genes (Haussler, D. Science 304:1321. 2004).  The striking conservation of numerous sites distant from target genes and the discovery of enhancer activity at Lvis may indicate a common mode of gene regulation involving distant enhancer sites.

POSTER 7 - BUILDING A GENOMIC REGULATORY NETWORK FOR LUNG DEVELOPMENT

Peterson KA 2, Richardson JE 1, King BL 1, Dolan ME 1, Welsh IC 1, Bult CJ 1, O'Brien TP 1

1 The Jackson Laboratory, Bar Harbor, United States, 2 The University of Maine, Orono, United States

A major goal in functional genomics is to identify and understand the collection of cis-regulatory modules necessary for coordinating differential gene expression during development. Comparative studies of the complete genomic sequences for mouse and human indicate that the largest conserved fraction is non-coding, and within this non-coding fraction are sequences that perform regulatory function. One step towards deciphering the function of these conserved non-coding sequences (CNCS) is by interpreting the combinatorial clustering of transcription factor binding sites (TFBS) as an informational pattern that forms a regulatory code. Based upon a signaling network model for lung branching morphogenesis, we developed a novel sequence analysis pipeline for integrating comparative genomic approaches to identify shared regulatory motifs that are derived from constraints imposed by a biological network.  The algorithm was trained on a set of genes known to control lung branching morphogenesis (LungNet). This algorithm is designed to identify novel motifs (e.g. binding sites for factors involved in lung branching) as well as map clusters of TFBS to search for regulatory modules on a genome-wide scale. We used the LungNet training set to derive a set of novel motifs and to identify regulatory modules proposed to be important for lung development. This computational pipeline was implemented using a new data analysis platform and visualization tool, BioDX.  We tested our approach by identifying cis-regulatory modules along mouse chromosome 2 (Mmu2) using comparative genomics, motif clustering and position relative to known genes. Genes predicted to be expressed in the branching lung epithelium were confirmed by RT-PCR and in situ hybridization. We are working to validate and extend the LungNet model as a proof-of-concept so that this tool will provide a powerful new approach for investigating the connection between genomic regulatory architecture and dynamic gene expression throughout development.

POSTER 8 - 3-D VISUALIZATION OF MOUSE EMBRYOS WITH OPTICAL PROJECTION TOMOGRAPHY

Walls JR 1, Sled JG 1, Bruneau BG 2, Sharpe J 3, Henkelman RM 1

1 Mouse Imaging Centre, Hospital for Sick Children, Toronto, Canada, 2 Hospital for Sick Children, Toronto, Canada, 3 Human Genetics Unit, Medical Research Council, Edinburgh, United Kingdom

The ability to visualize the 3D organization of biological tissue is essential to unravelling its complexities.  A new imaging technique called Optical Projection Tomography (OPT), essentially an optical version of X-Ray computed tomography (CT), fills a gap among current imaging modalities by creating molecularly specific, cellular resolution images of specimens up to 1 cc in size.

The visualization of a specific gene’s pattern of expression is essential to gain an understanding of its role.  Mutant comparisons performed with OPT are more sensitive to spatial complexity than comparisons performed with serial sectioning techniques that might alter subtle morphology.  Common optical markers can be used to highlight a region of interest or a particular genetic expression for enhanced visualization.  We have acquired cellular resolution OPT data sets of whole E9.5 to E12.5 mouse embryos with a fluorescently labelled cardiac specific pattern of gene expression to examine the full extent of 3D morphological differences between control and mutant specimens.

Vascular development is difficult to study without 3D visualization.  Resolution of vessels less than 10 microns in size is needed to obtain a complete 3D map of the vasculature.  Vessel imaging is a promising application of OPT as it meets the resolution criteria for smaller vessels while retaining information from the entire specimen.  We have acquired a complete map of the vasculature of an E16.5 mouse head with vessel resolution of 15 microns.  It is expected that the development we are undertaking will allow the modality to resolve the capillary vessels.

POSTER 9 - LINKING MAMMALIAN GENOME EVOLUTIONARY INSTABILITY TO HUMAN CANCER DEVELOPMENT AND PROGRESSION

Zhao S, Delcher D, Shetty J

The Institute For Genomic Research, 9712 Medical Center Drive,, Rockville, MD, United States

Using paired-end sequences from bacterial artificial chromosomes (BACs), we have constructed high-resolution synteny and rearrangement breakpoint maps among human, mouse and rat genomes.  This not only has allowed us to achieve several novel discoveries about mammalian genome evolution, but also provides a unique resource to study human cancer development and progression. 

As closely related species, mouse and rat share the majority of the breakpoints and often have the same types of rearrangements, when compared to the human genome.  However, the breakpoints not shared between them indicate that mouse rearrangements are more often inter-chromosomal, whereas intra-chromosomal rearrangements are more prominent in rat.  Centromeres may have played a significant role in reorganizing a number of chromosomes in all three species. 

Among the >300 syntenic blocks identified, are segments of over 40Mb without any detected interspecies rearrangements, as well as regions with frequently broken synteny and extensive rearrangements.  Computer simulation of the syntenic block length distribution supports the fragile breakage model but not the random breakage model for mammalian genome evolution, indicating that mammalian genomes consist of evolutionary stable and unstable regions.

Statistical analyses indicate that many human cancer-associated genome lesions are directly linked to evolutionary genome instability.  By examining the genomic locations of ~50 genome wide recurrent chromosomal aberrations with their associated genes identified from the Mitelman Database at http://cgap.nci.nih.gov/Chromosomes/RecurrentAberrations, we have found that over 95% of these lesions fall in the evolutionarily unstable sites.  This is consistent with a previous study that examined the relationship between evolutionary breakpoints among primates and chromosomal defects in human cancer.  These analyses indicate that the evolutionarily unstable regions should become logical targets for research on the origins of human malignancy.  We are currently studying the linkage between evolutionary genome instability and colon tumorigenesis.

POSTER 10 - BIOINFORMATICS AND EXPRESSION ANALYSIS OF MOUSE CHROMOSOME 5 (0-75MB REGION)

Shao H 1, Gan T 2, Valladares O 1, Manduchi E 2, Pinney D 2, Stoeckert C 2, Bucan M 1

1Department of Genetics, University of Pennsylvania, Philadelphia, United States, 2 Penn Center for Bioinformatics, University of Pennsylvania, Philadelphia, United States

Although sequences for large eukaryotic genomes (e.g. human, mouse, rat, chicken etc) are being completed, it remains a challenge to identify all encoded genes.  Approaches used to date include ab initio gene prediction, similarity-based annotation, and direct alignment of transcribed sequences.  We are using the proximal portion of mouse Chromosome 5 and several public databases to assemble a gene index for the 75 Mb region – target for region-specific ENU mutagenesis screen.  A Database of Transcribed Sequences (DoTS, http://www.allgenes.org/) has identified over 150 novel transcripts, in addition to 550 genes annotated by the ENSEMBL and Celera databases.  Over 81 % of these transcripts were verified using a custom-built 70-mer oligo microarray hybridized with embryos cDNA (E9.5, E12.5, E14.5 and E17.5) and cDNA from several adult tissues.  The expression map of the chromosome is being assembled by integration of our array data, with expression data from large scale efforts (Hamatani et al. 2004, Su et al. 2004).  This gene index and the expression map will be used to suggest and rank candidate genes for identified embryo-lethal and behavioral mutations, as well as in disease-gene identification in the corresponding portions of the human genome.

POSTER 11 - SEQUENCE AND HAPLOTYPE ANALYSIS OF H-RAS AND DISTAL CHROMOSOME 7 IN 12 MOUSE INBRED STRAINS REVEALS A STRONG ANCESTRAL BIAS

Drew JC, Kastenmeier AS, Drinkwater NR

McArdle Laboratory for Cancer Research, University of Wisconsin Medical School, Madison, United States

Inbred mouse strains vary in the frequency of H-ras initiating mutations in chemically induced liver tumors.  To determine if allelic variation in the H-ras gene accounts for differences in mutant frequencies, we sequenced the 4.5kb H-ras gene from 14 inbred strains:  12 common laboratory strains (129P3/J, 129X1/SvJ, A/J, AKR/J, BALB/cByJ, C3H/HeJ, C57BL/6J, C57BR/cdJ, CBA/J, DBA/J, SM/J, and SWR/J) and 2 distant strains (CAST/Ei and SPRET/Ei).  Surprisingly, eleven inbred strains had identical sequences over the 4.5kb region.  The AKR/J strain exhibited 2 SNPs in non-coding regions signifying a remarkably low rate of polymorphism.  The genomes of common laboratory inbred strains are mosaics of M.m.domesticus and M.m.musculus sequence.  All 12 common inbred strains, including AKR/J, inherited H-ras alleles from M.m.domesticus.

To determine if the mono-allelism of H-ras is unique or represents a larger, selected region of distal Chr. 7, we sequenced 500-600 bp for each of 15 genes spanning the distal 10Mb for the 16 aforementioned strains.  In the 12 common inbred strains, an extremely low SNP rate (5.2x10-5) was observed for a 3-4 Mb interval including H-ras and Igf2, and the genes sampled in this sub-region share the M.m.domesticus allele.  This large block of mono-allelic genes is flanked by genomic regions with an over 60-fold higher SNP rate (3.3x10-3) and multiple archetypal haplotypes represented among 12 inbred strains. It is not known how common such stringently conserved haplotype blocks are in the genomes of the laboratory inbred strains. 

POSTER 12 - HIGH THROUGHPUT SEQUENCE ANALYSIS OF CO-REGULATED GENE SETS

Kirov S 1, Zhang B 1, Locascio P 2, Schmoyer D 1, Crawford O 1, Snoddy J 1

1 GST at Univ. of Tennessee/ORNL, Oak Ridge, United States, 2 Oak Ridge National Lab, Oak Ridge, United States

To understand the how RNA levels are regulated in cells, we have created a high-throughput, automated pipeline which can search for putative cis-regulatory elements (CREs) in the genome or other control regions within the upstream regions 5’ or 3’ UTRs of RNAs. 

We will present the structure of the pipeline and some results, including an example from a set of genes that form the cornified envelope.

This pipeline can input mouse gene sets that might be co-regulated (e.g. very similar gene expression changes under multiple perturbations, participation in a common biological process, etc.). The pipeline automatically retrieves appropriate sequences from the input gene sets. This pipeline also retrieves the relevant orthologs in other sequenced chordates. We use the orthologs to define evolutionarily conserved regions that are more likely to contain the motifs of interest. We search within these regions for short motifs that are overrepresented in the different ortholog sets.  These overrepresented motifs might be functional sites of interest. Typical results often include multiple sequence motifs that are present in a subset of the input gene set (e.g. several putative CREs in promoters shared among half the input genes). The second part of the pipeline is a search for these sequence motifs in the appropriate locations for all the genes in the genome. Frequently, our results from this search include the original genes and a modest number of genes with similar properties.  This process is partly confirmed, as it will sometimes uncover functional motifs that have been seen experimentally.

POSTER 13 - ASSESSING GENOMIC VARIATION IN COMMON STRAINS OF  INBRED MICE

Cervino ACL, Edwards S, Schadt E

rosetta/merck, seattle, United States

We characterized DNA variation in 60 different strains of mice, including 56 common strains ordered from the Jackson Laboratories. Using phylogenetic analysis we investigated how various common strains relate to each other. Insight into heterozygosity between strains and between markers is essential in the selection of both an optimal set of strains  and an optimal set of markers for the experimental crosses aimed at elucidating complex phenotypes. In addition to the survey of genomic variation we also investigated a number of clinically relevant phenotypes and gene expression profiles in a subset of strains.

POSTER 14 - EVOLUTIONARY BREAKPOINTS IN MOUSE, RAT AND HUMAN CHROMOSOMES

Elliott RW

Roswell Park Cancer Institute, Buffalo, United States

Positions of loci on sequence-based maps of several mouse chromosomes have been compared to the positions of orthologs in rat and human chromosomes.  There are about 10 to 15 evolutionary-based breaks for human chromosomes on each mouse chromosome, although mouse chr 15 has only 3 such breaks.  The number of breaks for rat chromosomes is usually much fewer, ranging from 0 to 11.  However, most of the break points between rat and mouse occur at the same position as mouse to human breakpoints.  This suggests that very few breaks have occurred between mouse and rat after the divergence of the rodent lineage, and that most pre-existed in earlier the mammalian genome.  Exceptions appear to be chromosome inversions, which are illustrated by the analysis of mouse Chr 17 and its rat and human orthologs.

POSTER 15 - THE EMAGE 3D GENE EXPRESSION DATABASE

Christiansen JH, Richardson L, Venkataraman S, Burton N, Yang Y, Baldock R, Davidson D

MRC Human Genetics Unit, Edinburgh, U nited Kingdom

The Edinburgh Mouse Atlas of Gene Expression (EMAGE) is a database of spatially mapped gene expression data in the developing mouse embryo that we are developing as part of the Edinburgh Mouse Atlas Project (EMAP).  EMAGE is part of the Mouse Gene Expression Information Resource (MGEIR) which is being built in a collaborative effort with the GXD/MGI group at the Jackson Laboratory.

All EMAGE data is housed in a standard framework: the EMAP Digital Atlas of Mouse Development which consists of at least one representative 3D digital embryo model for most Theiler stages (TS) as well as a standardised nomenclature for anatomical structures that are present at every Theiler stage of development.  As the digital models are 3D, it is possible to cut virtual sections in any plane through them to reveal anatomical detail within that plane.

Incoming EMAGE data images are mapped spatially into the corresponding  regions within the embryo models.  At TS07-14 the 3D standard models have anatomical regions defined within them and as such, data spatially mapped into these models is automatically annotated to the corresponding text terms for these structures.  This is accompanied by further manual text annotation.

Searching the data in EMAGE can be done spatially by a simple point and click method, or by using text terms. 

New developments aimed at identifying groups of co-expressed genes (by heirarchical clustering and domain intersection analysis) will be discussed.

EMAGE is publicly accessible at http://genex.hgu.mrc.ac.uk.

POSTER 16 - CHARACTERIZATION AND MAPPING OF THE YODA MUTATION

Barbaric I 1, Blake S 2, Hunter AJ 2, Brown SDM 1

1 MRC Mammalian Genetics Unit, Harwell, United Kingdom, 2 GlaxoSmithKline Pharmaceuticals, Harlow, United Kingdom

We have characterised the Yoda mouse mutant carrying an ENU-induced dominant mutation that causes an abnormal craniofacial phenotype, disorganised bone growth plate and late-onset osteopaenia. Mice homozygous for the Yoda mutation die during embryogenesis, around dE9.5. At this stage they exhibit severe growth retardation, as well as anterior and heart patterning defects. Yoda homozygous embryos also fail to turn and their allantois fails to fuse to the chorion. To identify the mutated gene that is underlying the Yoda phenotype, mutant backcrossed animals were genotyped with a panel of markers spanning the genome. The mutation has been localized to a 4.5 Mb region on Chromosome 8. Fine mapping of the mutation has been complicated by the apparent shared haplotype between the used strains. Ongoing work involves high resolution mapping to identify the mutation and extended phenotypic characterization of the mutant phenotype.

POSTER 17 - ESSENTIAL FUNCTIONS OF THE PHOSPHATIDYLSERINE RECEPTOR (PTDSR) DURING EMBRYOGENESIS, DISTINCT FROM APOPTOTIC CELL REMOVAL

Böse J 1, Gruber AD 2, Helming L. 1, Schiebe S 1, Wegener I 1, Hafner M 3, Horsch M 4, Clarke K 5, Neubauer S 5, Schneider JE 5, Beckers J 4, Bhattacharya S 5, Lengeling A 1

1 Junior Research Group Infection Genetics,  German Research Centre for Biotechnology (GBF), Braunschweig, Germany, 2 Department of Pathology, School of Veterinary Medicine Hannover, Hannover, Germany, 3 Department of Experimental Immunology, German Research Centre for Biotechnology (GBF), Braunschweig, Germany, 4 Institute of Experimental Genetics, GSF - National Research Centre for Environment and Health, Neuherberg, Germany, 5 Department of Cardiovascular Medicine, University of Oxford, Oxford, United Kingdom

Apoptosis and phagocytosis of apoptotic cells are fundamental processes for normal animal development, homeostasis and immune function. Among the described receptors on phagocytes for the efficient recognition and engulfment of apoptotic cells, the phosphatidylserine receptor (Ptdsr) has been particularly implicated in apoptotic cell removal. To determine the function of Ptdsr in vivo, we generated a null allele in the mouse and established a Ptdsr-b-geo gene-trap line. Targeted inactivation of Ptdsr causes perinatal lethality that is associated with growth retardation and a delay in the terminal differentiation of kidney, intestine, liver and lung. Interestingly, about 14 % of Ptdsr -/- mice are anophthalmic and display fundamentally novel lesions, with induction of ectopic retinal pigmented epithelium in paranasal sinuses. By using the gene-trap mouse line we found that Ptdsr has specific and prominent expression patterns in organs that are affected by Ptdsr-deficiency. Moreover, by using high-resolution magnetic resonance imaging (MRI) we identified ventricular septal defects, double-outlet right ventricle and hypoplasia of the pulmonary artery in mutant embryos, suggesting that Ptdsr -/- mice die due to heart malformations. These results indicate that Ptdsr plays a hitherto unexpected role in cellular differentiation during organ morphogenesis and especially in cardiovascular development. To our surprise we found normal engulfment of apoptotic cells in Ptdsr knockout mice, demonstrating that Ptdsr has essential, novel functions during embryogenesis, distinct from apoptotic cell removal. Recent molecular analysis and expression profiling studies are supporting the function of Ptdsr as a key factor for the control of multiple molecular pathways required for organogenesis.

This work is supported by the EU project EUMORPHIA (QLG2-CT-2002-00930, grant to A.L.) and by the German National Genome Research Network (NGFN-Xpress, grant to J.B.)

POSTER 18 - ANALYSIS OF OVERLAPPING FUNCTIONS OF SMAD2 AND SMAD3 USING CHEMICALLY-INDUCED GERMLINE MOUSE SMAD2 MUTATIONS

Vivian J 1, Magnuson T 2

1 Institute of Maternal-Fetal Biology, University of Kansas Medical Center, Kansas City, KS, United States, 2 Department of Genetics, University of North Carolina, Chapel Hill, NC, United States

Smad2 and Smad3 are closely related intracellular transcriptional regulators and mediate TGF-beta, nodal, and activin signaling. Mice lacking Smad2 die at perigastrulation stages, whereas mice lacking Smad3 develop colorectal carcinoma.  Combination of mutations in these genes will provide further understanding of the overlapping functions of these factors in vivo.  We are exploring this overlap in function using two germline Smad2 mutations identified in a gene-directed screen of chemically mutagenized mouse embryonic stem cells.  One Smad2 mutation used in this study has previously been shown to be hypomorphic in function.  A variety of phenotypes were observed when this hypomorphic allele of Smad2 is combined with a null allele of Smad3.  The embryos lack a well-defined forebrain, and have left-right patterning defects.  More severely affected embryos lack a portion of the anterior notochord.  These phenotypes were not observed in either Smad2 or Smad3 mutant background alone, suggesting these factors share function in forebrain development, left-right patterning, and formation of the notochord.  The molecular deficits underlying these phenotypes are currently being explored.  We are also examining a new mutation in Smad2, which is viable in the homozygous state.  This allele harbors a point mutation within a region of Smad2 altered in some human hepatocellular carcinomas. We are exploring whether this mutation will provide insight into Smad2 function in tumor progression by combining this mutation with the null allele of Smad3.

POSTER 19 - COMPREHENSIVE SAGE ATLAS OF MURINE GENE EXPRESSION THROUGHOUT DEVELOPMENT

Bohacec S 1, Xie Y 1, Kuo B 1, Khattra J 2, Sidiqqui A 2, Helgason CD 4, Hoodless PA 3, Jones S 2, Marra M 2, Simpson EM 1

1 Centre for Molecular Medicine and Therapeutics, Vancouver, Canada, 2 Genome Sciences Centre, Vancouver, Canada, 3 Terry Fox Laboratory, B.C. Cancer Agency, Vancouver, Canada, 4 Cancer Endocrinology, B.C. Cancer Agency, Vancouver, Canada

This project is designed to generate a comprehensive collection of gene expression profiles encompassing various tissues throughout development, from fertilized egg to adult, for the C57BL/6J mouse. The use of mouse as a model of human development enables the systematic collection of embryonic tissues not possible in humans. Detection and quantitation of transcripts is carried out via the Long Serial Analysis of Gene Expression (LongSAGE) technique. This platform provides three major advantages over conventional gene expression analysis approaches: it presents a quantitative profile of the mouse transcriptome, it is not restricted to only detecting known genes, and it can be subsequently expanded as further data becomes available. To date, data sets from 44 libraries, of a planned 200, are publicly available. These libraries total over 5 million sequenced LongSAGE tags and represent manual or laser capture microdissected (LCM) tissues at significant stages in embryogenesis and organogenesis. A third of the projected libraries will profile gene expression in developing neural tissue with the aim of better understanding brain formation and a range of neural disorders. Two software packages (DiscoverySpace and CMOST) have also been developed for use in library analysis, inter-library comparisons and tag-to-gene mapping through numerous databases.

The main objective of the Mouse Atlas Project is to create a publicly accessible on-line expression database that will offer the research community insights into the progression and regulation of mammalian development for more rapid identification of disease genes and targets for new therapeutics. Library descriptions and full data sets are available on our website: http://www.mouseatlas.org/.

This project is supported by Genome Canada, the BC Cancer Agency and NCI-NIH.

POSTER 20 - MIND BOMB1 – CHARACTERIZATION OF THE MIB1 KNOCKOUT IN MOUSE AND ITS ROLE IN THE NOTCH  SIGNALING PATHWAY

Rajendra R 1, Barsi JC 1, Wu JI 2, Artzt K 1

1 Department of Molecular Genetics & Microbiology, Institute of Cellular and Molecular Biology, University of Texas at Austin, Austin, United States, 2 Depatrments of Pathology and Developmental Biology, Howard Hughes Institute, Stanford Medical School, Palo Alto, United States

Early embryonic development is regulated by various signaling pathways, one of the most widely studied and conserved being the effect of Notch-Delta signaling. Experiments with Delta like1 (Dll1) have shown that, like Notch, it is cleaved twice resulting in an intracellular fragment, that is transported to the nucleus, to possibly affect transcriptional factors (Six et al.,2003). Dll1 protein is also ubiquitinated by two proteins Neuralized (Lai etal., 2001) and Mind bomb1 (Itoh et al., 2003). Neuralized is non-essential for embryogenesis, as the knockout is not lethal. On the other hand knockout of Mind bomb1 (mib1) results in embryonic lethality in both zebrafish and mice. Mib1 was also ascertained biochemically, to target the protein Death Associated Protein Kinase (DAPK) for ubiquitination (Jin etal., 2000).

We have generated a mib1 knockout in the mouse that results in embryonic lethality by day E10.5, which is preceeded by abnormal somitogenesis, improper vasculogenesis and premature neural differentiation.  We report here that the entire Notch1-Dll1 signaling pathway is disrupted and have focused mainly on the in vivo effects of the mib1 knock out pertaining to this pathway.The hypothesis based on our observations is that, failure to modulate Dll1 protein disrupts the signaling down-stream of Notch1, resulting in loss of lateral inhibition and premature differentiation into neuroblasts by E9.5. Further, the neuroblasts are unable to regulate DAPK levels in the absence of mib1, resulting in apoptosis. The mib1 knockout system is a useful tool in understanding neuronal differentiation with possible long-term implications for treating neural tumors.

POSTER 21 - SEAL, A MUTATION THAT CONFERS TRANSIENT, INDUCIBLE HIND LIMB PARALYSIS

Tabeta K, Du X, Sovath S, Mudd S, Beutler B

The Scripps Research Institute, La Jolla, United States

Using ENU, we have produced a novel recessive mutation on the C57BL/6 background, associated with an inducible defect of hind limb movement. The mutation was named seal because homozygotes appeared to “waddle” like seals on land.  For a time, transmissibility of seal appeared sporadic.  However, it was eventually realized that homozygous animals show normal locomotion and cage activities if undisturbed, but demonstrate hind limb paralysis after being picked up for examination by grasping the loose skin over the nape of the neck.  Once triggered in this manner, paralysis persists for about 8 days before the hind legs regain function.  However, most of the mice still show a partial residual “seal-like” deficit. About 50% of seal homozygotes also have swollen heels and foot pads.  This lesion is normally unilateral.  Moreover, seal homozygotes seem to be very sensitive to environmental changes, and appear anxious, as exemplified by vigorous tail trembling and self-confinement to corners when a cage is opened. The seal phenotype is fully penetrant on the C57BL/6 and C3H/HeN x C57BL/6 hybrid backgrounds.  The mutation has been mapped to mouse chromosome 11 using a panel of microsatellite markers and at present, has been confined to a 0.7 Mbp critical region on 1300 meioses. Further characterization of phenotype and responsible gene will be presented.

POSTER 22 - FERTILE XY(SRY-) MALES

Qin Y 1, Bishop CE 2

1 Baylor College of Medicine, department of OB/GYN, Houston, United States, 2 Baylor College of Medicine, dept. Molecular & Human Genetics, Houston, United States

We have previously reported the dominant mouse mutant Odd Sex (Ods) in which XX Ods/+ mice, develop as phenotypic males due to male-pattern expression of Sox9 in developing XX Ods/+ embryonic gonads. The testis of XX Ods/+ males are small, devoid of germ cells, and the adults are sterile. This is to be expected as such males  lack all Y chromosome encoded fertility genes and the presence of two copies of the X chromosome in the germ line has been shown to be incompatible with the early post natal mitotic divisions of spermatogenesis. In order to test whether, the specific activation of  Sox9, in embryonic XX Ods/+ gonads was sufficient to initiate the development of a fully functional adult testis and a fertile male, we constructed an XY (Sry-), Ods/+ male which carries one copy of the X, and all Y genes except Sry. Such males were normaly fertile up to 4-5 months of age indicating that Sox9 then. Unexpectedly XY Ods/+ males became sterile by 6-8 months in contrast to XY TgSry+ or XY TgSry+, Ods/+ littermates who remained fully fertile.  When compared to normal XY FVB males, we noticed that the vasculature pattern in adult XY Ods/+ testis was disrupted, whereas that of XY TgSry+ and XY TgSry+, Ods/+ showed the normal pattern. These finding correlate with the failure of XX Ods/+ males to downregulate Wnt4 at E11.5. To test whether this was due to Sry itelf or the level of Sox9, we constructed an XY Ods/Ods homozygote male. In this mouse the testicular vascular pattern was restored to normal. These data indicate that activation of the male sex-determining pathway by Sox9 is sufficient to produce a fully functional testis and a fertile male. They indicate that all genes in this pathway, upstream of Sox9, muts be regulatory in nature. They indicate that subtle changes in gene dosage of Sox9 and Wnt4 can result in anomalies of vascularisation that have a progressive impact on male fertility over time.

POSTER 23 - Homozygosity For Two N-ethyl-N-nitrosourea (ENU)-induced And A Gene-trap Mutations Of The Cytoplasmic FMRP-interacting Protein 1 (Cyfip1) Results In Embryonic Lethal Phenotype In Mice

Li L 1, Mentzer S 2, Nicholls RD 3, Carpenter D 2, Rinchik E 2, Johnson D 2, You Y 2

1Graduate School of Genome Science and Technology, University of Tennessee-Oak Ridge National Laboratory. 2 Mammalian Genetics and Genomics Group, Life Sciences Division, Oak Ridge National Laboratory. 3Department of Psychiatry, University of Pennsylvania.

The functional unit l71Rl has been defined as a peri-implantation lethal locus by complementation analysis of deficiencies in the p-region of mouse Chromosome 7.  Previous characterization of the genomic region indicated that the l71Rl interval contains 4 genes including the Cytoplasmic FMRP (Fragile-X Mental Retardation Protein)-interacting protein 1 (Cyfip1), also known as Shyc or Sra-1Cyfip1 was considered to be the most likely candidate gene among the four to cause the peri-implantation lethal phenotype.  Subsequently, a Cyfip1 gene-trap ES cell line was identified in the BayGenomics gene-trap ES cell clone library, and heterozygous Cyfip1GT/+ mice were generated from the chimeric mice derived from a Cyfip1 gene-trap ES cell clone.  However, Cyfip1GT/GT homozygotes do not show peri-implantation lethality.  Instead, the homozygous Cyfip1GT/GT embryos implant and the development appears normal until 7.5-8.5 dpc, after which the embryos manifest abnormalities and remain in lordotic position.  The mutant embryos are subsequently disintegrating after 10.5 dpc.  Previously, two N-ethyl-N-nitrosourea (ENU)-induced embryonic lethal mutations, stocks 785DSJ and 828DSJ, were mapped within the l71Rl critical interval.  We performed intra- and inter-crosses for these two ENU-induced mutations as well as allelism tests with Cyfip1GT/+ mice.  The results unambiguously indicated that these three mutations are allelic and the homozygous or compound heterozygous mutants display the same embryonic abnormalities after 8.5 dpc. Therefore, we concluded that the two ENU mutations are also Cyfip1 gene mutations and have identified the mutations at the molecular level.  Our preliminary results suggest that Cyfip1 plays an important role in mouse embryonic development.  The function of Cyfip1 in the mouse development is being analyzed by molecular and cellular approaches. 

POSTER 24 - MOLECULAR BASIS OF PITUITARY DYSFUNCTION: MECHANISM OF PROP1 ACTION.

Camper SA, Ward RD, Brinkmeier ML, Nasonkin IO, Vesper AH, Potok MA, Stone BM, Beuschlein F, Suh H, Hammer GD, Raetzman LT

University of Michigan Medical School, Ann Arbor, MI, United States

The Prophet of PIT1 (PROP1) gene was discovered 8 years ago.  Recently animal studies have revealed the mechanism of Prop1 action and provide an explanation for the unusual clinical features in humans with PROP1 mutations. Lesions in PROP1 are the most common known cause of autosomal recessive hypopituitarism in humans.  GH, TSH, PRL, LH and FSH are usually deficient, and hypocortisolism occasionally develops.  Surprisingly, some patients initially have normal pituitary size or hyperplasia that resolves to hypoplasia later in childhood.  We investigated the mechanism of PROP1 action using mice with altered PROP1 levels, including the spontaneous mutant known as Ames dwarf (Ser83Pro), a genetically engineered null mutant that deletes the homeodomain of this transcription factor, and a transgenic over-expressor of PROP1.  Many features of the human patients are mimicked in PROP1 deficient mice. Adult mutant mice are consistently deficient in GH, TSH, PRL, LH and FSH, although there is no evidence of acquired ACTH deficiency through the first year of life. The genetic background has a profound influence on the mutant phenotype, which can include lethal respiratory distress syndrome, juvenile wasting associated with hypoglycemia, or viable adult dwarfism.  PROP1 expression peaks in mice during early embryonic development and decreases to barely detectable levels by birth.  Despite the transient, fetal-specific expression, the total mass of the pituitary prirmordium is nearly indistinguishable in normal and PROP1 deficient mice until a week after birth.  The mutant mice develop pituitary hypoplasia eventually because proliferating progenitor cells are unable to migrate ventrally into the anterior lobe of the organ.  This initially results in dorsal overgrowth of undifferentiated cells, which resolves through enhanced apoptosis in 1-2 week old mice.  This appears analogous to the hyperplasia and degeneration of the pituitaries of some human patients.  We expect that progenitor cells fail to migrate and differentiate because mutants are unable to restrict expression of the HESX1, TLE3, AES repressor complex and unable to activate expression of NOTCH2 and PIT1.  Persistent expression of the repressor complex is sufficient to block differentiation, misexpression of NOTCH2 is sufficient for disruption of gonadotrope differentiation, and PIT1 is required for specification of somatotropes, thyrotropes and lactotropes.  Interestingly, both reduced and elevated expression of Prop1 disrupt gonadotrope function, suggesting that a specific level of PROP1 may be critical for normal pituitary function.  In summary, the Prop1 mutant mice are valuable tools for understanding the mechanism of PROP1 action. 

POSTER 25 - CHARACTERIZATION OF HOLOPROSENCEPHALY IN THE ZIC2KU/KU MUTANT

Warr N, Elms P, Willoughby C, Bogani D, Arkell R

MGU MRC Harwell, Didcot, United Kingdom

Holoprosencephaly (HPE) is a relatively common disorder involving the incomplete development and septation of midline structures in the central nervous system. It has an incidence as high as 1:250 in embryogenesis but only 1:16,000 of live births due to intrauterine lethality. Clinical features vary widely, in the more extreme cases, these may include a single brain ventricle and severe facial anomalies such as cyclopia with a proboscis, a single-nostril nose and cleft lip and in less severe cases mild facial dysmorphia may be observed such as a single front incisor and/or narrowly spaced eyes. A number of genes have been linked to HPE in humans and several are members of the Sonic hedgehog (SHH) and BMP signaling pathways including Sonic hedgehog, Hedgehog-interacting protein, Patched, Smoothened, Gli, Wnt, BMP, Nodal, Smads, One-eyed pinhead, and TG-interacting factor. In addition, mutations in ZIC2, a mammalian gene homologous to the Drosophila pair-rule gene odd-paired, can also give rise to HPE in humans. We have previously isolated and partially characterized a murine Zic2 mutant called kumba (Zic2ku) which has a point mutation in the fourth C2H2 zinc finger domain. Zic2ku/ku homozygotes exhibit holoprosencephaly as well as other developmental defects (heart looping, neural tube closure, hindbrain segmentation and neural crest migration). We have initiated a study into the embryological cause of holoprosencephaly in the Zic2ku/ku mutant using in situ hybridization with probes for midline markers. The preliminary results are presented and the potential interactions between Zic2 and the SHH and/or BMP pathways during forebrain development are discussed.

POSTER 26 - CHARACTERISATION OF SHORTY, AN ENU DERIVED MUTANT MOUSE WITH DEFECTS IN RIB FORMATION

Harboe TL 1, Herron B 2, Beier DR 1

1 Genetics Division, Harvard Medical School, Brigham & Women’s Hospital, Boston, United States, 2 Wadsworth Center NYS Department of Health, Albany, United States, 3 Genetics Division, Harvard Medical School, Brigham & Women’s Hospital, Boston, United States

Here we report the characterization of an ENU-induced mutant mouse we call shorty (srt) that wasidentified in a screen for late embryonic developmental phenotypes.The mutant has a severe malformation of the thoracic vertebrae and is lacking several ribs. The development of the lumbar vertebrae appears to be normal as do the long bones. The defect is lethal and the affected pups are usually stillborn and smaller in size than wild type littermates. The phenotype of srt is similar to a group of human disorders called spondylocostal dysostosis, characterised by abnormal vertebral segmentation. The phenotype is similar to that of pudgy (pu), which is caused by a mutation in Dll3, a Notch ligand. Our mapping analysis has excluded the possibility that srt is an allele of pu, since the mutation have been localised to mouse chromosome 17 in a 1.47 MB region in the interval between D17Mit34 and D17Mit11 by genetic mapping. The region is within the MHC region and hence very gene rich. We are currently sequencing 10 candidate genes. Also under investigation is the expression of various genes involved in somitogenesis such Pax1 and -9, Shh, Dll1, Lfng and -3 and Msx1 and -2. This novel mutant results in abnormal rib development and could provide insight into the patterning of the paraxial mesoderm and somitogenesis.

POSTER 27 - GENETIC VARIABILITY IN PARAOXONASE (PON1): RELEVANCE FOR ACUTE PESTICIDE EXPOSURE DURING DEVELOPMENT

Walter BJ 1, Cole TB 1, Pettan-Brewer C 1, Fisher JC 1, Shih DM 1, Tward A 2, Lusis AJ 2, Costa LG 1, Furlong CE 2

1 University of Washington, Seattle, United States, 2 UCLA, Los Angeles, United States,

Paraoxonase (PON1) is an HDL-associated enzyme that metabolizes organophosphorous (OP) pesticides, drugs, and oxidized lipids. PON1 activity is determined in part by a coding region polymorphism (Q192R) that affects its catalytic efficiency and in part by differences in PON1 levels, which vary widely (>13-fold). To assess the importance of the Q192R polymorphism for OP detoxication during development, transgenic mice were used that expressed human transgenes encoding hPON1Q192 or hPON1R192 at equal levels in place of mouse PON1. Dose-response curves were generated for exposure of developing mice (PND 4, 13, or 21) to chlorpyrifos, diazinon, and their oxon metabolites. Mice expressing hPON1R192 were more resistant than hPON1Q192 mice to CPO toxicity during development, despite a similar onset of expression between the two transgenic lines. These studies, together with other studies in our laboratory, indicate that children less than 2 yr old, especially those homozygous for PON1Q192, are particularly susceptible to OP toxicity. Supported by T32 AG00057, ES09601/EPA-R826886, ES09883, ES04696, ES07033, ES11387.

POSTER 28 - TARGETED GENE DISRUPTION OF HIP/RPL29 REDUCES PLURIPOTENCY OF MURINE ES CELLS

Kirn-Safran CB, Focht RJ, Mari ER, Carson DD

University of Delaware Department of Biological Sciences, Newark, United States

HIP is a heparin/heparan sulfate interacting protein identical to ribosomal protein L29 that is not believed to be essential for translation, but rather plays a role in translation efficiency.  There is increasing evidence that targeted gene disruption of specific ribosomal protein genes is lethal in embryos prior to implantation.  Because high levels of HIP/RPL29 are found in embryonic stem (ES) cells and in all types of proliferating and developing tissues, we hypothesize that the presence of HIP/RPL29 is essential for normal early embryonic growth.  In particular, we believe that a strict regulation of HIP/RPL29 expression is required to maintain proliferation and avoid commitment of stem cell progenitors to specific cellular differentiation programs.

To investigate the role of HIP/RPL29 expression during early development, we targeted one of the two alleles of Hip/RPl29 gene in a pluripotent mouse ES cell line of the 129S7 background.  Mutant ES cell contribution was assessed in vivo by breeding chimeras obtained after microinjection of Hip/Rpl29 +/- ES cells into C57BL/6N host blastocysts.  Our results show that Hip/Rpl29 +/- ES cells display a dramatically reduced germ cell potential. Thus, a significant number of chimeric males highly derived from the ES cell line, as determined by the strong contribution of the dominant Agouti gene to their coat color, primarily transmitted C57Bl/6N sperm to their progeny.  This result suggests that a mutation in the Hip/Rpl29 gene is detrimental to the maintenance of the ES cell undifferentiated phenotype, and is consistent with our previous studies that demonstrated that a knock-down of HIP/RPL29 mRNA is associated with an accelerated differentiation potential of a multipotent stem cell line into cartilage-like cells.  Additional studies are currently focusing on determining whether Hip/Rpl29 -/- null embryos and ES lines are viable.  In conclusion, our data indicate that HIP/RPL29 constitutes a candidate regulator of stem cell differentiation. (Supported by Lalor grant to C.B.K.S and NIH grant HD25235 to D.D.C.)

POSTER 29 - A MOUSE WHOLE-GENOME OLIGO MICROARRAY PLATFORM FOR TRANSCRIPT COPY NUMBER ESTIMATION.

Carter MG 1, Sharov AA 1, VanBuren V 1, Dudekula DB 1, Carmack CE 2, Ko MSH 1

1 National Institute on Aging, Baltimore, MD, United States, 2 Agilent Technologies, Palo Alto, CA, United States

In an effort to enable more comprehensive expression profiling studies of mouse embryos and stem cells, we have created a 60-mer oligo microarray platform with coverage of all genes and transcripts identified in the NIA Mouse Gene Index 2.0.  The NIA Gene Index combines genome assembly sequence, public EST libraries, and human-curated genetic databases to form structure models of over 40,000 genetic loci.  Every transcript model in the index is represented on the microarray by at least one in situ-synthesized 60-mer oligonucleotide probe, for a total of 40,503 genes and 43,830 transcripts.  Dose-response linearity was evaluated, with  extremely high correlation (correlationmedian r = 0.98, with 50% of values between 0.99 and 0.93) between predicted and observed expression log ratios, and a compact distribution of regressions (median slope = 0.94, with 50% of values between 1.0 and 0.97).

While we have used similar microarrays to generate very reliable relative expression data in the past, we have not been able to estimate the absolute copy numbers of mouse transcripts, information which is critical to interpreting expression data, as well as to making decisions about follow-up studies of signature genes.  A set of synthetic “spike-in” RNA control transcripts and oligo probes, based on yeast intronic and intergenic sequences, allows us to estimate copy numbers of endogenous transcripts, as well as control for variations in target labeling and hybridization.

This platform was used to compare the expression profiles of embryonic stem (ES) cells and trophoblast stem (TS) cells.  This dData was compared to previous experiments, and providesproviding insight into what is thought to be the first differentiation step in mammalian development, - the transition from totipotent embryonic cells to multipotent trophoblast and inner cell mass.

POSTER 30 - AN ENU-INDUCED MUTATION IN A NOVEL TRANSMEMBRANE PROTEIN IS ASSOCIATED WITH HOMEOTIC TRANSFORMATIONS IN MICE

Howell GR, Wilson L, Munroe RJ, Schimenti JC

The Jackson Laboratory, Bar Harbor, United States

A region-specific ENU mutagenesis screen has identified 38 embryonic lethal phenotypes (termed L5Jcs1 to L5Jcs38) based on “loss of class” at wean.  Due to the nature of the screen each of the lethal mutations could be localised to a 40 Mb region of mouse chromosome 5, delineated by the Rump white (Rw) inversion.

The recessive phenotype, apt1 (anterior-posterior transformation 1, formally known as L5Jcs1) was originally identified as an embryonic lethal (based on the “loss of class” at wean). Studies now show that mice homozygous for the mutation develop to around e18.5 days, but are either born dead, or survive until around one day post-partum. A more detailed analysis of the skeleton of apt1 mice revealed a posterior to anterior transformation of the thoracic and lumbar vertebrae, similar to abnormalities seen in a number of Hox gene knockouts. Specifically, the eighth rib (thoracic segment 8 – T8) is attached to the sixth sternabra, and the first lumbar vertebra (L1) shows the development of a fourteenth rib. Judging by the rib length it also appears that the ninth to the thirteenth thoracic segments (T9-13) are potentially transformed into more anterior structures. Recombination mapping (carried out in collaboration with Dr Maja Bucan’s lab, Univ Penn, Philadelphia) localised apt1 to a 6.3 Mb region on proximal mouse chromosome 5, between D5Mit353 and D5Mit107. In combination with recombinational mapping, mapping by complementation was also carried out to narrow the region of interest using a series of overlapping deletions generated around the Qdpr gene. Skeletal preparations of deletion/apt1 mice showed the same homeotic-like transformation as apt1/apt1 mice, and this analysis narrowed the critical region to 3 Mb containing only 14 genes.

RNA was recovered from mutant and wild type mice and all genes in the region were screened for mutations by PCR and sequencing. A splice site mutation, causing exon skipping, has been identified in a novel transmembrane protein and confirmation by BAC rescue is underway. This novel gene provides the basis for furthering our understanding of the mechanisms involved in the development of anterior-posterior structures in mice.

POSTER 31 - THE SEARCH FOR HUNTINGTON'S DISEASE MODIFIERS IN THE MOUSE

Acevedo A 1, Chrobot N 1, Rubinsztein DC 2, Brown SD 1

1 MRC Mammalian Genetics Unit, Harwell, United Kingdom, 2 Cambridge Institute for Medical Research, Cambridge, United Kingdom

Huntington’s disease (HD) is an autosomal dominant, progressive, fatal, neurodegenerative disorder caused by an expanded polyglutamine tract (PolyQ) in exon one of the HD gene. Up to 70% of the variance in age of onset can be accounted for by an inverse relationship with polyQ repeat number. The residual variance is likely to be mediated by environmental and genetic factors (modifiers). In order to identify dominant modifier genes for HD, we have initiated an N-ethyl-N-nitrosourea (ENU) mutagenesis screen with a mouse model of HD. First, we carried out extensive behavioural characterisation of two HD transgenic mouse models (R6/1 and N171-82Q) in order to select the one most suitable strain and to determine baseline phenotypes. Both lines show early signs of disease, including tremors, weight loss and motor deficits. However, N171-82Q strain has more rapid disease progression and better fertility, so it was chosen for the modifier screen. We injected BALB/c males (2 x 100mg/kg ENU), crossed them to N171-82Q carrier females and assessed disease progression in F1s using a variety of simple behavioural analyses (modified from SHIRPA) and weight measurement. Preliminary analyses of the progeny to date has identified several mice carrying potential dominant modifiers that are currently undergoing inheritance testing.

POSTER 32 - COORDINATION OF GROWTH AND METABOLISM BY THE IMPRINTED GRB10 GENE

Ward A, Koumanov F, Charalambous M, Garfield A, Allsop JE, Moorwood K, Smith FM

University of Bath, Bath, United Kingdom

The growth factor receptor-bound protein 10 (Grb10) gene acts as a potent inhibitor of both fetal and placental growth during embryogenesis. Grb10 encodes a signalling adaptor protein that can bind to a number of receptor tyrosine kinases, including the insulin-like growth factor type 1 (Igf1r) and insulin (Insr) receptors. Here, we address the physiological relevance of these biochemical interactions. Genetic tests indicate that the effects of Grb10 on growth are essentially independent of IGF signalling. On the other hand, Grb10 interacts with Insr to control growth during embryogenesis in a highly organ-specific manner and, post-natally, to modulate glucose homeostasis.  Thus, Grb10 appears to fulfil a role in coordinating growth with energy supply.

Grb10 is in most tissues predominantly, but not exclusively, expressed from the maternal allele and therefore behaves in a manner consistent with the parent-offspring conflict hypothesis (briefly, that paternally expressed genes will promote growth and exploit maternal resources opportunistically, while maternally expressed genes will counter this by being growth inhibitory).  However, there is in some tissues a striking reversal of this pattern such that expression occurs exclusively from the paternal allele. The reciprocal imprinting of a single gene with a strong growth-regulatory function will be discussed in relation to the parent-offspring conflict hypothesis.

POSTER 33 - SENSITIZED MUTAGENESIS SCREEN FOR MODIFIERS OF THE DELTA/NOTCH PATHWAY IN THE MOUSE

Rubio-Aliaga I, Soewarto D, Wagner S, Przemeck G, Fuchs H, Hrabé de Angelis M

GSF Research Center - Institute of Experimental Genetics, Neuherberg/Munich, Germany

Delta1 encodes a transmembrane protein, which interacts directly with transmembrane proteins of the Notch-family, in a ligand-receptor-like manner. The Delta/Notch signal transduction pathway is well conserved and plays a central role in mediating cell-to-cell communication regulating the determination of various cell-fates during development. Disruption of the Delta1 gene implies alteration of early embryonic development followed by embrionic lethality (Hrabé de Angelis et al., 1997). Moreover, recent studies indicate that the Delta/Notch pathway belongs to a sophisticated system of interactions, which is believed to be involved in a wide range of common diseases including cancer, immunological diseases, infections as well as cardiovascular and neurological diseases. As proven recently, genetic modifier screens, which have been demonstrated to be very powerful in lower organisms, are also feasible in the mouse (Carpinelli et al., 2004). Combining knockout technology and ENU mutagenesis, we are carrying out a sensitized screen with the Delta1 knockout in order to identify enhancers and suppressors of the Delta/Notch pathway. Here we present recent results of this sensitized screen.

POSTER 34 - COORDINATION OF GROWTH AND METABOLISM BY THE IMPRINTED GRB10 GENE

Smith FM1, Charalambous M.2, Koumanov F.1, Garfield A.1, Ward A.1

1 Centre for Regenerative Medicine, University of Bath, Bath, United Kingdom, 2 Centre for Diabetes and Endocrinology, Division of Medicine, London, United Kingdom

Growth Factor Receptor-bound protein 10 (Grb10) is a member of the Grb7 family of SH2 domain-containing adaptor proteins. Murine Grb10 is maternally expressed, located on chromosome 11, and uniparental disomy (UPD) of this region results in reciprocal growth phenotypes. Grb10 has been shown to associate in vitro with a number of tyrosine kinase receptors including Igf1r and Insr, although its function and signalling remain unclear. Mice with disruption of the Grb10 locus were generated (termed Grb10D 2-4) to characterise the physiological role of this signalling protein. Maternal transmission of Grb10D 2-4 resulted in intrauterine and post-natal growth enhancement with disproportionate effects on organ size and increased liver glycogen. Despite Grb10 expression being reminiscent of the Igf2 expression pattern during development, our findings indicate that Grb10 acts via an Igf-independent growth axis. Adult mice displayed reduced adipose tissue content, mild hypoinsulinemia and improved glucose tolerance. Enhanced Insr signalling was observed in cultured adipocytes and muscle lysates of Grb10D 2-4 mice following insulin stimulation. Genetic tests confirm a tissue-specific interaction of Grb10 with the Insr to affect growth and glucose homeostasis. Our findings demonstrate that Grb10 is important in growth and provide the first in vivo evidence for a role as a modulator of insulin action.

POSTER 35 - FINE MAPPING PRION DISEASE INCUBATION TIME QTL USING HETEROGENEOUS STOCK MICE

Lloyd SE 1, Maytham ELG 1, Thompson S 1, Mott R 2, Fisher EMC 1, Collinge J 1

1 MRC Prion Unit, London, United Kingdom, 2 The Wellcome Trust Centre for Human Genetics, Oxford, United Kingdom

Prion diseases are fatal neurodegenerative disorders of both humans and animals which include scrapie, bovine spongiform encephalopathy and Creutzfeldt-Jakob disease. They are characterised by prolonged incubation periods which in mice can vary from 100-500 days.  The main genetic determinant of incubation time is the prion gene, Prnp, where allele a (Leu 108, Thr 189)  and allele b (Phe 108, Val 189) are associated with short and long incubation times respectively.  Large scale linkage studies have now successfully identified additional modifier loci on at least eight different chromosomes. Because the regions of linkage identified in these crosses were broad (10-30cM), we are now employing a number of different approaches for fine mapping and candidate gene identification. One strategy is to use a heterogeneous stock of mice. These are derived from the semi-random mating of eight inbred lines of mice (A/J, AKR/J, BALB/cJ, C3H/HeJ, C57BL/6J, CBA/J, DBA/2J and LP/J) over multiple generations. Approximately 1000 mice at generation 37 were inoculated intracerebrally with RML/scrapie and incubation times were recorded. 400 animals representing both extremes of the incubation time distribution were genotyped at 1cM intervals in regions previously reported to contain quantitative trait loci.  HAPPY linkage analysis of these data successfully identified regions on chromosomes 11 and 15 representing loci of 1-2cM.  These intervals are small enough for individual candidate gene analysis which we are investigating through the use of quantitative RT-PCR and sequencing to identify the molecular basis of the variation.

POSTER 36 - IDENTIFICATION OF NOVEL IMPRINTED GENES ON MOUSE CHROMOSOMES 7 AND 18

Woodfine K.1, Choi JD 1, Wood AJ1, Collins JN 2, Oakey RJ 1

1 Division of Medical Genetics, GKT School of Medicine, London, United Kingdom, 2 Division of Human Genetics, The Children’s Hospital of Philadelphia, Philadelphia, United States

Imprinted genes are expressed in a monoallelic fashion. Their pattern of expression is dependant on the parent of origin. While one allele is expressed the other allele is epigenetically silenced. Mice with uniparental disomies (UpDi) or uniparental duplications (UpDp) contain regions of the genome that have originated from just one parent of origin. RNA’s from these mice can be used to assay genes that are expressed in a parent of origin dependant manner.

RNA from mice with UpDi of Chromosome 7 has been screened using the Affymetrix GeneChip expression system. Gene expression patterns were compared in tissues from embryos with a maternal duplication verses those with a paternal duplication of Chromosome 7. Additional bioinformatic analysis ensured the best candidates for genomic imprinting were selected from the screen. Allele specific RT-PCR analysis has been used to confirm the imprinting status of seven novel imprinted genes thus far.

We present here, results from screening RNA from mice with UpDi 18 to characterise an imprinted region near the imprinted Impact gene This region is homologous to a region on human chromosome 18 known to be associated with Bipolar associated disorder (BPAD). BPAD has been associated with parent of origin affects in this region of Chromosome 18. Therefore study of the homologous region in the mouse with UpDi will be invaluable in trying to understand the parent of origin affects in BPAD and in providing candidate susceptibility genes for this disorder.

POSTER 37 - SUPPRESSORS AND ENHANCERS OF TESTICULAR CANCER SUSCEPTIBILITY IN DOUBLE-MUTANT MICE

Lam MYJ, Youngren KK, Nadeau JH

Case Western Reserve University, Cleveland, United States

Linkage studies with the 129/Sv strain, which has a low rate (1-5%) of spontaneous testicular germ tumors (TGCTs), reveal exceptional complexity in the genetic control of susceptibility. Currently, susceptibility genes have not been identified and pathways that lead from primordial germ cells (PGCs) to stem cells to TGCTs are unknown. Various Mendelian traits in mice affect susceptibility only on the 129/Sv genetic background, suggesting that these mutations act as modifiers of tumorigenesis. We tested pairwise interactions among the five known modifier genes, by comparing TGCT frequency in single and double-mutant mice, to identify combinations that show enhancer or suppressor effects. Trp53+/- MgfSlJ+/- double-mutant males had a tumor frequency that was four-fold lower than the rate expected if each modifier acted independently. Normally, TRP53 arrests cells to allow repair of DNA damage or programmed cell death, whereas MGF is essential for survival of PGCs and TGCT stem cells. Reduced TGCT susceptibility suggests that interactions between DNA repair, apoptosis and MGF-KIT signal transduction modulate tumorigenesis.  Studies are underway to characterize the nature of the interaction between TRP53 and MGF.  Preliminary results indicate that Trp53 mRNA expression is elevated in the testis of double-heterozygous mutants, suggesting a dosage effect resulting from a novel mechanism to up-regulate TRP53 and suppress tumorigenesis.  In  parallel, we discovered other examples of suppressors (Ay and 129/Sv-Chr19M) and enhancers (MgfSlJ and 129/Sv-Chr19M). Together, these results provide insights to apoptosis, cell cycle control, cell lineage determination and signal transduction that control PGC biology and TGCT susceptibility.

POSTER 38 - MODIFIERS OF THE DDK SYNDROME RESULT IN THE DOMINANT RESCUE OF EMBRYONIC LETHALITY

Ideraabdullah FY, Kim K, Pardo-Manuel de Villena F

Dept of Genetics, UNC-Chapel Hill, Chapel Hill, United States

The DDK syndrome is an early embryonic lethal phenotype that is observed in crosses involving females of the DDK inbred strain and males of other strains. This phenotype is the result of an incompatibility between a maternal DDK factor and a non-DDK paternal gene, both of which have been mapped to the Ovum mutant (Om) locus on mouse chromosome 11. Previous studies have shown that F1 hybrid females resulting from crosses involving a non-DDK dam and a DDK sire, express a semi-lethal phenotype when crossed to B6 males. It has been suggested that the semi-lethal nature of the phenotype in these females is the result of allelic exclusion acting on the Om  locus. In this study, we tested the fertility of several F1 hybrid  females derived from crossing females of classical or wild-derived strains with DDK males.  Although most of the crosses are semi-lethal, two crosses involving F1 females from wild-derived Mus musculus domesticus strains, PERA/Ei or PERC/Ei, are viable. Our observations indicate the presence of genetic modifiers of the DDK syndrome that result in dominant rescue of lethality. Further investigation revealed that the PERA/Ei and PERC/Ei modifiers segregate independently of the Om locus.  We hypothesize that these dominant modifiers skew expression of the maternal factor in favor of the non-DDK allele at Om, by means of allelic exclusion. Currently, efforts are being made to identify these modifiers by genome scan and characterize their role in allelic exclusion.

POSTER 39 - THE REQUIREMENT FOR EED IN AUTOSOMAL GENOMIC IMPRINTING MAY BE TISSUE SPECIFIC

Chamberlain SJ, Montgomery ND, Kalantry S, Magnuson T

Department of Genetics, University of North Carolina-Chapel Hill, Chapel Hill, NC, United States

EED and EZH2 –containing Polycomb group (PcG) complexes are dynamic and maintain transcriptionally repressive chromatin states through replication and mitosis. In addition to their role in the maintenance of expression patterns of homeotic genes, these complexes also act upon other genes to maintain cellular differentiation in such processes as development, hematopoiesis, and tumorigenesis.

Mice with mutations in the Eed gene display defects in imprinted X-inactivation and autosomal imprinting.  The major defect is a failure to maintain transcriptional repression in the silenced chromosome or alleles. Data from this lab suggests that the failure to maintain silencing on the inactive X-chromosome occurs only in tissues that undergo imprinted X-inactivation, and that it is most pronounced in differentiating cells of the trophoblast, while the primitive endoderm and the undifferentiated trophoblast cells appear to correctly maintain the inactive X in the absence of EED.  Here we utilize a gene-trap reporter construct, and a conventional targeted mutation to determine whether the defects in autosomal imprinting in Eed mutants are similarly restricted to a particular tissue type or suggestive of a particular differentiation state in mouse embryos.

POSTER 40 - PREDISPOSITION TO MOUSE THYMIC LYMPHOMAS DEPENDS ON COMMON VARIANT ALLELES OF MTF-1 (METAL RESPONSIVE TRANSCRIPTION FACTOR-1)

Kominami R1, Tamura Y1, Maruyama M1, Mishima Y1, Schaffner W2

1 Niigata University, Graduate School of Medical and Dental Sciences, Niigata, Japan, 2 Institute for Molecular Biology, University of Zurich, Zurich, Switzerland

Genetic predisposition to cancers is significant to public health because a high proportion of cancers probably arise in a susceptible human subpopulation. One major model for polygenic predisposition is the common variant-common disease model, in which common variants that have arisen only once, early in the history of the population, underlie disease predisposition in humans and perhaps in mice. The mapping and isolation of such variants in humans are complicated by the multiplicity of unlinked loci, whereas for mouse models of cancer susceptibility experimental strategies exist for fine mapping of complex traits. Using a mouse model of g -ray induced thymic lymphomas, we performed linkage analysis and haplotype mapping that identified Mtf-1, metal responsive transcription factor-1, as a candidate lymphoma susceptibility gene. The Mtf-1 allele inherited from resistant strains exhibited higher radiation inducibility of target genes than that of susceptible strains, and products of the targets such as metallothionein are able to suppress cellular stresses generated by irradiation. Therefore, this suggests that highly inducible strains are refractory to radiation effects and hence are resistant to lymphoma development. Interestingly, the Mtf-1 polymorphism was common to different mouse subspecies, consistent with the common variant-common disease model.

Genetic predisposition to cancers is significant to public health because a high proportion of cancers probably arise in a susceptible human subpopulation. Using a mouse model of g -ray induced thymic lymphomas, we performed linkage analysis and haplotype mapping that identified Mtf-1, metal responsive transcription factor-1, as a candidate lymphoma susceptibility gene. The Mtf-1 allele inherited from resistant strains exhibited higher radiation inducibility of target genes than that of susceptible strains, and products of the targets such as metallothionein are able to suppress cellular stresses generated by irradiation. Therefore, this suggests that highly inducible strains are refractory to radiation effects and hence are resistant to lymphoma development. Interestingly, the Mtf-1 polymorphism was common to different mouse subspecies, consistent with the common variant-common disease model.

Genetic variation plays a key role in determining the range of individual susceptibility to cancers in human. A high proportion of cancers probably arise in a susceptible subpopulation that carry low-penetrance variant alleles. One major model for polygenic predisposition is the common variant-common disease model, in which common variants that have arisen only once, early in the history of the population.

POSTER 41 - EVOLUTIONARILY CONSERVED SEQUENCE ELEMENTS THAT POSITIVELY REGULATE INTERFERON GAMMA EXPRESSION IN T CELLS

Shnyreva M.1, Weaver WM1, Blanchette M.3, Taylor SL2, Fitzpatrick DR2, Tompa M.1, Wilson CB1

1 University of Washington, Seattle, United States, 2 Amgen Corporation, Seattle, United States, 3 McGill University, Montreal, Canada

Our understanding of mechanisms by which the expression of interferon gamma is regulated is limited. Herein, we identify two evolutionarily conserved non-coding sequence elements (IFNgCNS1 and IFNgCNS2) located ~5kb upstream and ~18kb downstream of the initiation codon of the murine Ifng gene. When linked to the murine Ifng gene (-3.4 to +5.6 kb) and transiently transfected into EL-4 cells, IFNgCNS1 enhanced IFN gamma production, and this was amplified further by inclusion of IFNgCNS2. A DNase I hypersensitive site and extragenic transcripts at IFNgCNS2 correlated positively with the capacity of primary T cell subsets to produce IFN gamma. Transcriptionally favorable histone modifications in IFNgCNS2, the Ifng promoter and intron regions, and, though less pronounced, in IFNgCNS1 increased in parallel as naïve T cells differentiated into IFN gamma producing effector CD8 and TH1 T cells, and this was not seen in TH2 T cells. Like IFN gamma expression, these histone modifications were T-bet-dependent in CD4 but not CD8 T cells. These findings define two distal regulatory elements associated with T cell subset-specific IFN gamma expression.

POSTER 42 - MODIFIER LOCI THAT INFLUENCE TPA SKIN TUMOR PROMOTION SUSCEPTIBILITY IN GENETIC CROSSES OF DBA/2 WITH C57BL/6 MICE

Angel JM, Abel EL, Riggs PK, Elizondo L, Caballero M, DiGiovanni J

Univ Texas, MD Anderson Cancer Center, Science Park-Research Division, Smithville, United States

Cancer susceptibility in the general population is a function of multiple, poorly penetrant modifier genes, each of which contributes to, but is not solely responsible for determining the likelihood that a particular type of cancer will develop after exposure to certain environmental carcinogenic agents.  Genetic differences in susceptibility to two-stage skin carcinogenesis in the mouse have been known for many years and the major contribution to susceptibility appears to be at the level of tumor promotion.  Loci that modify the susceptibility to 12-O-tetradecanoylphorbol-13-acetate (TPA) skin tumor promotion have been mapped to several chromosomal regions in genetic crosses of sensitive DBA/2 with relatively resistant C57BL/6 mice. One locus, Psl1, was mapped to an ~40 cM region of distal chromosome 9. Recent tumor studies using interval specific congenic mouse strains suggest that this region consists of at least three loci that modify the sensitivity to TPA skin tumor promotion.  A large number of genes mapping to these loci have been associated with skin phenotypes or cancer development. Additional studies using interval specific subcongenic strains and haplotype mapping are underway to further delimit the map locations of these loci. Furthermore, global gene expression analysis using cDNA microarrays have identified genes mapping to distal chromosome 9 that are differentially expressed in the epidermis of TPA-treated C57BL/6 vs DBA/2 mice, suggesting that these genes may be good candidates for TPA promotion susceptibility loci.

POSTER 43 - MOUSE HAIRY EARS (EH) INVERSION MUTATION DISRUPTS NO GENE TRANSCRIPTS, BUT EXPRESSION OF HOXC GENES IN SKIN IS DISTURBED

Mentzer S 1, Sundberg J 2, Cacheiro N 1, Chao H 3, Carpenter D 1, Johnson D 1, Rinchik E 3, You Y  1

1 Oak Ridge National Laboratory, Oak Ridge, United States, 2 The Jackson Laboratory, Bar Harbor, United States, 3 University of Tennessee, Knoxville, United States

The mouse Hairy Ears (Eh) mutation is a paracentric inversion, In(15)Eh2Rl, in the distal half of Chr 15.  Heterozygous Eh/+ mice have small and hairy ears, whereas Eh/Eh homozygotes in C3H/Rl background typically die before weaning.  We mapped both the proximal and distal inversion breakpoints to BAC clones using fluorescent in situ hybridization (FISH) on metaphase Eh/+ cell spreads, cloned the distal breakpoint, and determined inversion breakpoint sequences.  Consequently, genes surrounding the inversion breakpoints were identified.  No protein-coding transcripts are disrupted by either inversion breakpoints.  The proximal breakpoint occurs between Sntb1 (syntrophin  1, centromerically) and Has2 (hyaluronan synthase 2, telomerically), while the distal breakpoint takes place between the Hoxc4 (Homeobox C4, centromerically) and an uncharacterized transcript (GenBank accession numbers: BF606930 and AK019974, telomerically).  Therefore the inversion spans ~47 Mb. Quantitative RNA analysis by TaqMan real-time PCR assay for the Hoxc genes indicated that expression of many members of Hoxc has been changed in skin during skin/hair morphogenesis and in adult mice.  Hair fiber analysis using scanning electron microscopy revealed that ear-specific hair fibers grow longer in the ears of the Eh mutation.  Histology analysis of ear-skin did not reveal any obvious abnormalities.  Taken together, our results suggest that Hoxc genes play important roles in hair growth.

POSTER 44 - DEVELOPMENTAL PROFILE OF DNA METHYLATION AND GENE EXPRESSION AT THE H19 LOCUS IN MICE LACKING SEQUENCE REQUIRED FOR GENOMIC IMPRINTING

Thorvaldsen J., Fedoriw A., Yang G., Bartolomei M.

University of Pennsylvania, Philadelphia, United States

The 2 kb differentially methylated domain (DMD) 5’ of H19 is exclusively methylated on the paternal allele throughout development and is required for H19 and Igf2 imprinting. In mice harboring a 1.6 kb deletion of the DMD (D DMD) or a 3.8 kb deletion spanning the DMD (D 3.8kb-5’H19), loss of H19 and Igf2  imprinting is detected in all neonatal tissues (G&D 12:3693, MCB 22:2450). In addition, the DMD is required for full expression of H19 and Igf2. Nevertheless, DMD sequence that remains in the D DMD mice, acquires slightly more methylation on the paternal allele than on the maternal allele, indicating that some parental-specific identity is maintained that does not result in differences in parental-specific expression. Here, we investigate allele-specific DNA methylation and gene expression patterns at the H19 locus in the D DMD and the D 3.8kb-5’H19 mice in pre- and post-implantation embryos to determine when loss of H19 and Igf2 imprinting occurs. We detect loss of imprinting at the H19 locus from the deletion alleles after embryo implantation. We have also generated a conditional (floxed) targeted allele of the 1.6 kb DMD sequence. Removal of this sequence in neonatal liver results in loss of Igf2 but not H19 imprinting. Thus, the DMD is required for maintenance of Igf2 but not of H19 imprinting in non-dividing tissues. In summary, analysis of the DMD deletions throughout development is important to determine the sequence requirements and mechanism of imprinting at the H19 locus.

POSTER 45 - CURRENT PROGRESS IN SCREENING FOR MUTANTS AFFECTING GENOMIC IMPRINTING IN THE RIKEN-GSC PROJECT

Suzuki T 1, Furuumi H 2, Hashimoto M 3, Kyouno S 4, Nagashima A 1, Kumaki K 2, Kaneda H 1, Gondo Y 5, Noda T 1, Wakana S 1, Ishino F 4, Sasaki H 2, Shiroishi T 1

1 Mouse Functional Genomics Research Group, Genomics Science Center (GSC), RIKEN Yokohama Institute, Yokohama, Japan, 2 Department of Integrated Genetics, National Institute of Genetics, Mishima, Japan, 3 Gene Research Center, Tokyo Institute of Technology, Yokohama, Japan, 4 Department of Epigenetics, Medical Research Center, Tokyo Medical and Dental University, Tokyo, Japan, 5 Population and Quantitative Genomics Team, GSC, RIKEN, Yokohama, Japan

Genomic imprinting is epigenetic regulation of gene expression depending on parental origin of the genes in mammals. In order to detect and identify novel genes involved in the initiation of the genomic imprinting, we have established a screening system for mutants that show impaired genomic imprinting in the RIKEN mouse ENU-mutagenesis program. In our screening, we measure methylation levels of imprinted genes in the mature gametes using methylation-sensitive enzymes or the COBRA (combined bisulfite restriction analysis). Both of non-methylated and re-methylated imprinted genes are analyzed for one individual to detect the mutants that show aberrant imprinting in the two stages: erasure of parental imprints and re-establishment of sex-different imprints. In this system, we can constantly assay the methylation levels of the imprinted genes using approximately 20 gametes originated from one individual.

Since September 2002, we have tested the phenotypes about 1,100 males for dominant screening, and 20 males and 300 females for recessive screening. Several individuals showed abnormal methylation patterns of the imprinted genes in the gametes. These results indicate efficacy of our screening system. We report the current results of the screening and the several phenodeviants that show abnormal methylation patterns.

POSTER 46 - ANALYSIS OF THE CANDIDATE REGION AND CANDIDATE GENES RESPONSIBLE FOR THE DDK SYNDROME OF EMBRYONIC LETHALITY

de la Casa-Esperon E 1, Gimelbrant A 3, Adey B 1, Briscoe T 1, Hao L 1, Wu G 1, Chess A 3, Pardo-Manuel de Villena F 2, Sapienza C 1

1 Fels Institute, Temple University School of Medicine, Philadelphia, PA, United States, 2 Department of Genetics, UNC-Chapel Hill, Chapel Hill, NC, United States, 3 Whitehead Institute for Biomedical Research, MIT, Cambridge, MA, United States

The DDK syndrome of embryonic lethality is observed in crosses between females of the DDK inbred strain with males of other strains (e.g., C57BL/6), in which the resulting embryos die around the time of implantation, while the reciprocal cross (non-DDK females x DDK males) is viable. The lethality is the result of an interaction between a maternal gene of DDK origin, which is expressed in oocytes, and a non-DDK paternal gene. Both genes (Om) map in the same region of chromosome 11. Interestingly, the Om candidate region in DDK chromosome 11 lacks ≈50 Kb present in the C57BL/6 chromosome. In addition, only half of the offspring of the backcross between (C57BL/6xDDK)F1 females and C57BL/6 males survive. This observation suggests that the maternal Om gene is expressed from only one allele in each single oocyte. This hypothesis is supported by our observation in the Om candidate region of asynchronous replication, which has been associated with monoallelic expression of genes. Consequently, in order to identify the maternal Om gene we are performing a study of expression patterns within the Om candidate region.

POSTER 47 - GENOME-WIDE ASSOCIATION ANALYSIS IDENTIFIES NOVEL MODIFIER LOCI OF HIRSCHSPRUNG DISEASE IN SOX10DOM MICE

Owens SE 1, Broman KW 2, Smith JR 1, Southard-Smith EM 1

1 Vanderbilt University, Nashville, United States, 2 Johns Hopkins University, Baltimore, United States

Hirschsprung disease (HSCR) is caused by abnormal development of the neural crest cells that comprise the enteric nervous system, resulting in an absence or reduction of enteric ganglia in a variable portion of the distal gastrointestinal tract.

Dominant megacolon (Sox10Dom) mice model the aganglionic megacolon seen in HSCR, and SOX10 mutations have been identified in a subset of HSCR patients. Sox10Dom heterozygote mice on a mixed genetic background (B6C3F1) mimic the variable severity of aganglionosis in HSCR patients, and this variation is influenced by genetic background.

To establish linkage to modifier genes that modulate the severity of aganglionosis in Sox10Dom mice we performed a genome-wide scan. Penetrance and extent of intestinal aganglionosis was quantified by whole-mount acetylcholinesterase (AChE) histochemistry on gastrointestinal tracts from postnatal day 7-10 backcross (n=204) and intercross (n=1092) progeny. Mice from the extremes of the phenotype distribution, along with mice in which aganglionosis was not penetrant, were genotyped by fluorescence polarization methods to distinguish 104 SNP markers at an average spacing of 19 cM throughout the genome.  Using R/qtl, nonparametric interval mapping identified five loci with significant linkage results.  All five loci are modifiers of penetrance, and two also contribute to the difference in severity of aganglionosis.  Epistatic interactions between modifiers and effects of gender and imprinting will be presented.

POSTER 48 - A SENSITIZED ENU MUTAGENESIS SCREEN FOR DOMINANT GENETIC MODIFIERS OF THROMBOSIS IN THE FACTOR V LEIDEN MOUSE

Westrick RJ 1, Manning SL 2, Dobies SL 1, Peterson AL 2, Siemieniak DR 2, Korepta LM 1, Ginsburg D 2

1 University of Michigan, Ann Arbor, United States, 2 Howard Hughes Medical Institute, Ann Arbor, United States

Venous thrombosis affects ~300,000 individuals per year in the USA.  A gain-of-function mutation in the factor V gene, Factor V Leiden, (FVL) is the most common known inherited risk factor for venous thrombosis.  Penetrance is incomplete, with only ~10% of FVL individuals experiencing clinically significant thrombosis.  We are performing a whole genome mouse mutagenesis screen to identify modifier gene candidates contributing to the penetrance of FVL in humans.  Previously, we demonstrated synthetic lethality between FVL and genetic deficiency of a key coagulation component, tissue factor pathway inhibitor (TFPI).  Complete TFPI deficiency in mice is embryonic lethal, whereas heterozygosity is compatible with normal survival.  However, homozygosity for FVL (FvQ/Q) in the context of heterozygosity for TFPI (Tfpi+/-) is uniformly lethal due to disseminated perinatal thrombosis.  This synthetic lethal interaction was utilized as a phenotyping tool for a sensitized ENU mutagenesis screen.  We aim to uncover novel dominant mutations that improve hemostatic balance leading to survival of FvQ/Q Tfpi+/- mice.  We have proven our approach by rescuing FvQ/Q Tfpi+/- with tissue factor (Tf+/-) heterozygosity.  Male FvQ/Q mice were ENU mutagenized and bred to FvQ/+ Tfpi+/- double heterozygous females.  Surviving G1 offspring were analyzed to identify rescued mice with the FvQ/Q Tfpi+/- genotype.  Analysis of 2199 G1 offspring thus far has identified 15 mice that survived to weaning.  Of the 8 mutants progeny tested to date, 3 appear to be heritable.  Our preliminary findings demonstrate the feasibility of our sensitized approach in the identification of dominant suppressors of the FvQ/Q Tfpi+/- lethal phenotype.

POSTER 49 - MODIFIERS OF AGANGLIONIC MEGACOLON IDENTIFIED BY EVALUATION OF CANDIDATE GENES IN THE SOX10DOM HIRSCHSPRUNG DISEASE MODEL

Cantrell VA, Owens SE, Chandler RL, Bradley KM, Smith JR, Southard-Smith EM

Vanderbilt University, Nashville, United States

Abnormalities in proliferation, migration, or survival of enteric neural crest (NC) can lead to aganglionosis in a variable portion of the distal intestine, Hirschsprung disease (HSCR).  Cumulative evidence suggests HSCR is the consequence of multiple gene interactions that modulate the ability of enteric NC cells to populate the developing gut. 

One of the essential genes for enteric ganglia development is the NC transcription factor Sox10Sox10Dom mice on a mixed genetic background exhibit variable penetrance and expressivity of aganglionic megacolon reminiscent of human HSCR families.  We have established congenic lines of Sox10Dom mice that fix this allele on distinct genetic backgrounds, C57BL/6J and C3HeB/FeJ.  Differences in survival and degree of aganglionosis between Sox10Dom animals in these lines demonstrate that aganglionosis is influenced by heritable factors.

To identify genes responsible for variation in aganglionic megacolon we have collected mice from the phenotypic extremes of an extended B6C3F F1.Sox10Dom pedigree.  Animals were stratified based either on survival or on extent of gastrointestinal tract affected as determined by acetylcholinesterase histochemistry.  Genotyping was performed with genetic markers closely flanking known HSCR susceptibility loci and genes that affect NC development to test for associations with severity of aganglionosis.  Our single locus association analysis indicates several candidate loci interact with Sox10Dom to influence the severity of aganglionosis in this HSCR model.  Crosses between mutants in the genes that exhibit significant association and Sox10Dom mice support the significance of these gene interactions and illustrate their impact on aganglionosis in double heterozygous progeny. 

POSTER 50 - CHROMATIN DOMAINS THAT ESCAPE X INACTIVATION CHARACTERIZED BY CTCF BINDING AND HISTONE MODIFICATIONS

Cheng MK 1, Filippova GN 2, Disteche CM 1

1 University of Washington, Seattle, United States, 2 Fred Hutchinson Cancer Research Center, Seattle, United States

Genes that escape X inactivation are expressed in the context of inactive chromatin and must therefore employ some mechanism to overcome the repressive structure of the inactive X chromosome. One possible mechanism is binding of the chromatin insulator CTCF, which we hypothesize, acts as a boundary element between inactivated and escape regions on the X. We have found CTCF binding sites at the 5’ CpG islands of Jarid1c and Eif2s3x, two mouse genes that escape X inactivation and are adjacent to an inactivated gene. We have also shown that DNA methylation is very low at the CpG island of Jarid1c. To characterize the chromatin structure of transition regions, we are using chromatin immunoprecipitation and real-time PCR to map the status of histone H3 acetylation, histone H3 lysine 9 di-methylation, histone H3 lysine 27 tri-methylation, and histone H4 lysine 20 mono-methylation in the regions around and within the Jarid1c and Eif2s3x escape domains. We propose that CTCF binding at the boundaries between inactivated and escape domains inhibits the spread of epigenetic modifications characteristic of inactive chromatin such as DNA methylation and histone modifications.

POSTER 51 - SNP AND SSLP HAPLOTYPES, TUMOR MICROARRAYS, AND CONGENIC RECOMBINANTS IN THE IDENTIFICATION OF HCS7, A POTENT LIVER CANCER MODIFIER.

Bilger A, Schneider A, Leutkehoelter K, Sundlov T, Drinkwater N

University of Wisconsin Medical School, Madison, United States

The Hcs7 liver cancer modifier, which accounts for the majority of the high susceptibility of C3H mice relative to B6, lies in a 6.3 Mb region of distal Chromosome 1 (based on recombinant mapping).  This region is orthologous to regions of human chromosome 1q that are amplified in about half of liver and breast cancers.

  We have analyzed SSLP and SNP haplotypes, gene expression in normal and neoplastic tissue, and fine-structure recombinants to identify candidates for Hcs7.  For the haplotype analysis we analyzed 41 of the 114 known and predicted genes in the region for B6, C3H, and two additional strains, CBA and BR, whose sensitivity relative to B6 also maps to distal Chromosome 1.  Haplotypes were derived from ~102 kb of sequence from 260 exons and flanking introns, yielding 656 SNPs.  SSLP markers were chosen at 120 kb intervals.  The SNP and SSLP maps are remarkably similar, revealing the near-identity of B6 and BR throughout the region, as well as large blocks where the closely related CBA and C3H strains resemble or differ from each other.  This haplotype analysis, together with cDNA microarray analysis of expression in the tumors of sensitive and resistant congenic lines, has led us to focus on a small number of related genes that are candidates for Hcs7.

POSTER 52 - IDENTIFICATION OF GENES INFLUENCING PLASMA VON WILLEBRAND FACTOR LEVELS IN MICE

Lemmerhirt HL 1, Ginsburg D 2

1 Department of Human Genetics, The University of Michigan, Ann Arbor, United States, 2 Howard Hughes Medical Institutue, Ann Arbor, United States

Circulating levels of von Willebrand factor (VWF), a blood coagulation protein, vary greatly among human and inbred mouse populations.  Both the Vwf gene and secondary modifier genes are critical in VWF level regulation. One murine modifier gene we previously reported, Mvwf1, lowers VWF levels in a subset of inbred mouse lines.  To identify modifier genes affecting Mvwf1-independent strains, we examined 200 F2 mice from an A/JxCASA/Rk cross. A pooled genome scan found VWF levels strongly correlated with parental genotype at markers D6Mit12 and D9Mit67.  D6Mit12 is closely linked (1.2 cM) to murine Vwf, suggesting strain-specific differences in Vwf expression or protein structure may be partly responsible for the variation observed.  Sequence analysis of Vwf mRNA identified 20 SNPs between these strains, four that encode amino acid changes.  F1 Vwf mRNA analysis does not suggest strain-specific differences in allele expression.  Allelic differences in protein biosynthesis and secretion are being examined through transfection analysis of A/J and CASA/Rk Vwf cDNAs.  Specific candidate genes at the D9Mit67 locus have not yet been characterized.  Collectively, D6Mit12 and D9Mit67 account for ~23% of the VWF level difference observed between these strains.  To identify additional modifier loci contributing to the substantial remaining variation, we are now conducting an extended non-pooled genome scan using all 200 F2 mice.  These data, when combined with our earlier results, should provide novel insight into the genetic complexity of VWF level regulation in mice and may help identify candidates for modifiers of bleeding and thrombotic risk in humans.

POSTER 53 - IDENTIFICATION OF THE MOLECULAR MECHANISMS INVOLVED IN THE DEVELOPMENT OF HAIRLESS PHENOTYPE IN NEAR NAKED HAIRLESS (HRN) MICE

Liu YT 1, Das S 2, Olszewski RE 2, Lu XC 3, Voy BH 2

1 University of Tennessee, Knoxville, United States, 2 Oak Ridge National Lab, Oak Ridge, United States,

3 Lawrence Livermore National Laboratory, Livermore, United States

Recessive mutations at the mouse hairless (hr) locus cause hair loss and increased susceptibility to chemical and UV carcinogenesis.  The near naked (Hrn) allele of hr arose spontaneously in the ORNL mouse colony and is unique in both its mode of inheritance (semi-dominant) and in the fact that, unlike recessive mutations that lead to hair loss after the first hair cycle, affected animals never grow a normal coat.  The hr gene encodes a transcriptional repressor, but little is known about the cellular pathways in which it acts.  We are therefore characterizing the Hrn mutation and using cDNA microarrays and histology to identify early changes in gene expression and skin structure that coincide with phenotype development and precede epidermal lesions that occur in adult animals.  Northern blot analysis and PCR of the hr cDNA indicated normal transcript size and exon composition. We sequenced the open reading frame and the 5’- and 3’- UTRs in Hrn/Hrn mice and in both potential founder strains (C3H and 101), but we did not identify any base pair changes in mutant animals. Quantitative RT-PCR analysis indicated differential hr expression in mutant mice.  In adults, hr expression was down-regulated by approximately two- and seven-fold in heterozygous and homozygous animals, respectively.  However at seven days of age, the age at which mutant animals are phenotypically distinguishable from wild type littermates, hr expression was elevated (2-fold in Hrn/+ and 3-fold in Hrn/ Hrn, respectively, vs. +/+).  These differences were confirmed with Northern blots.   Combined, these data suggest a potential regulatory rather than coding region mutation in Hrn mice.  Microarray analysis of skin from adult animals identified a set of genes that were differentially expressed in Hrn/+ compared to +/+.  Approximately 80% of the genes indicated as disregulated by microarray were confirmed by real time RT-PCR and were also shown to be altered in 7-day old animals.  Many of these genes encode keratin-associated proteins, which provide structural support to the hair fiber matrix.  Given histological analysis indicating an early defect in hair structure, the early nature of changes in gene expression relative to phenotype development, and the fact that hr encodes a transcription factor, we predict that genes critical for proper hair formation are primary targets of hr, and that their normal expression levels are disrupted by abundance of hr protein. 

POSTER 54 - IDENTIFICATION OF MULTIPLE GENETIC LOCI LINKED TO LUNG FUNCTION IN MICE

Reinhard C 1, Meyer B 3, Fuchs H 2, Stöger T 1, Eder G 1, Rüschendorf F 3, Heyder J 1, Nürnberg P 3, Hrabé de Angelis M 2, Schulz H 1

1 Institute for Inhalation Biology, GSF-Research Center for Environment and Health, Neuherberg/Munich, Germany, 2 Institute of Experimental Genetics, GSF-Research Center for Environment and Health, Neuherberg/Munich, Germany, 3 Gene Mapping Center, Max-Delbrück-Centrum, Berlin, Germany

Quantitative trait locus (QTL) linkage analysis was applied to identify chromosomal regions contributing to lung function in mice. The strains C3H/HeJ (C3) and JF1/Msf (JF1) as well as their offspring were phenotyped for the parameters total lung capacity (TLC) and dead space volume (VD) describing lung and conducting airway size, respectively, lung compliance (CL) characterizing lung elasticity as well as diffusing capacity (DCO) measuring the gas exchange. A whole genome scan in 150 animals of both the backcross (N2) and intercross (F2) progeny by use of 210 microsatellite markers revealed significant linkage results on chromosomes 5, 15, 17, and 19. All analyzed traits were significantly linked to a broad region on chromosome 15 suggesting the presence of more than one gene on chromosome 15 influencing lung function. The region of interest on chromosome 17 contains a syntenic region to human chromosome 6q27, which was recently identified to be linked to lung function in humans. Interesting candidate genes are located within the peak regions, e.g. the platelet derived growth factor beta (Pdgfb) gene and the relaxin1 (Rlx1) gene. Both genes are related to lung relevant phenotypes as was demonstrated in transgenic and knock out mouse models. Sequencing of the Rlx1 gene revealed functional polymorphisms between the two parental strains. The results of our QTL linkage analysis demonstrate that lung function in mice is influenced by multiple genetic loci located on at least 4 different chromosomes.

POSTER 55 - 'ANONYMUS' - A LABORATORY INFORMATION MANAGEMENT SYSTEM FOR THE CAPTURE AND DISSEMINATION OF ANIMAL HUSBANDRY AND EXPERIMENTAL DATA

Jones JD 1, Brundrett AC 1, Chandrasekar K 2, Jones PJR 1, Olver GH 2, Oxspring RJ 1, Seymour M 1, Selley RL 1, Johnson R 1

1 Mary Lyon Centre, Medical Research Council, Harwell, United Kingdom, 2 Prion Unit, Medical Research Council, London, United Kingdom

AnonyMus is a Laboratory Information Management System (LIMS) that has been developed for deployment in the Mary Lyon Centre at MRC Harwell, a new facility for functional genomics.

The system has been developed to ensure real-time collection, collation and dissemination of data generated in a paperless barrier facility.

Animal husbandry, phenotyping, facility management and UK Home Office legislative data are unified to form a comprehensive package for genetics laboratory requirements.

AnonyMus went live alongside the Mary Lyon Centre in Spring 2004 and at the MRC Prion Unit based in London. Both units are successfully operating the new software to capture and report critical data.

The software consists of a database on a commercial quality relational database management system and a web-enabled user front-end developed using JavaTM technologies.

Regular updates of newly developed and improved functionality are applied to the live version, created based on feedback from users and any new requirements that may arise. This dynamic development is made possible by the modular architecture of AnonyMus allowing new functionality to be added to the live system with minimal disruption and enabling 24-7 usage.

Since deployment, the AnonyMus team has embarked on an assessment of the genetics community for LIMS requirements.  The conclusion being that there is a niche in the market place for a software solution such as AnonyMus that has been designed to ensure stability, efficiency, accuracy and ease of use in an online environment.

POSTER 56 - IMPAIRED THALAMIC SENSORY GATING AND AFFECTIVE DISORDER IN T-TYPE CA2+ CHANNEL MUTANTS

Shin H-S, Choi S, Lee J, Lee S, Kim C, Kim D

Korea Institute of Science & Technology, Seoul, Korea, South

Low threshold Ca2+ currents mediated by α1G T-type channels are responsible for burst spike activities of relay neurons in the thalamus, and have been implicated in diverse physiological and pathological processes. We have previously reported that the knock-out mice for the α1G T-type channels show an enhanced nociceptive response to visceral pain, accompanied by an increase in pain-encoding tonic spikes in the absence of burst firing in thalamic relay neurons (Science, 2003). These results suggested that α1G T-type channels are involved in the thalamic sensory gating function, blocking the relay of the pain signals to the cortex. We have further examined this point by using several different sensory modalities: startle responses to auditory or tactile stimuli, and responses to chronic inflammatory pain. First, the mutant mice showed an enhanced startle response to auditory stimulations. The enhanced response appeared to be due to a change in the thalamic function instead of a peripheral problem, because auditory brainstem recordings (ABR), which measures primary neural signals induced by auditory stimuli, revealed no difference in the response pattern between the two genotypes. Second, the mutant also displayed an enhanced startle response to tactile stimuli compared to wildtype. Third, the mutant showed a selective increase in the late phase pain response to an intradermal injection of formalin into the hindpaw, a response under the control of a supraspinal mechanism. In summary, therefore, the mutant showed enhanced responses to sensory inputs of four different modalities. These results strongly support the idea that α1G T-type channels play an important role in the sensory gating in the thalamus. We suggest that the burst spikes induced by the low threshold Ca2+ currents are the key element in this thalamic function. Behavioral studies revealed that the mutant mice showed a mania-like affective disorder, increased alcohol preference, and a resistance to alcohol-induced hypnotic effect. These behavioral consequences will be discussed with regard to the sensory gating failure of the mutant.

POSTER 57 - A GENETIC ANALYSIS OF BEHAVIORAL DIVERSITY BETWEEN WILD-DERIVED MOUSE STRAIN, MSM/MS AND C57BL/6

Nishi A 1, Nishi A 3, Yonekawa H 2, Shiroishi T 1, Shiroishi T 4, Koide T 1, Koide T 3

1 The Graduate University of Advanced Studies, Hayama, Japan, 2 The Tokyo Metropolitan Institute of Medical Science, Tokyo, Japan, 3 Mouse Genomics Resource Laboratory, National Institute of Genetics, Mishima, Japan, 4 Mammalian Genetics Laboratory, National Institute of Genetics, Mishima, Japan

The wild-derived mouse strain, MSM/Ms, belongs to a musculs subspecies group, has much different behavioral traits in comparison with C57BL/6, which is a domesiticus subspecies group. MSM/Ms is more active than C57BL/6 in home cage at night phase. And MSM/Ms indicates lower activity than C57BL/6 at open-field test and Light-dark box test. It seemed that MSM/Ms is hyperactive and more anxious than C57BL/6. For identifying genes involved in the difference of behavioral traits between MSM/Ms and C57BL/6, we used consomic strains generated by replacing one of the chromosomes of C57BL/6 by that of MSM. As a first step of this project, we have characterized the behavioral traits of each consomic strains with the use of home-cage activity test and light- dark box test. As a result, C57BL/6-Chr6CMSM, which carries MSM locus from D6Mit166 (1.1cM) to D6Mit12 (49.2cM) showed lower spontaneous activity and higher anxiety than C57BL/6. And there were no apparent sex difference on the test scores of spontaneous activity in this consomic strain, while females of C57BL/6, MSM/Ms and other consomic strains exhibit higher activity than males.

It seemed that the major gene or genes that causes this phenotype are located in the centromere side from D6Mit268 (7.7cM), because C57BL/6-Chr6TMSM, which carries MSM locus from D6Mit268 to D6Mit15 (66.7cM) did not showed similar phenotype as C57BL/6-Chr6CMSM.

Currently, we have been mapping this gene and analyzing the in vivo mechanism of this behavioral trait.

POSTER 58 - DESCRIPTIVE AND TEMPORAL ANALYSIS OF THE OPEN-FIELD BEHAVIOR IN WILD-DERIVED MOUSE STRAINS

Takahashi A 1, Takahashi A 4, Yonekawa H 2, Shiroishi T 1, Shiroishi T 3, Koide T 1, Koide T 4

1 The Graduate University of Advanced Studies, Mishima, Japan, 2 The Tokyo Metropolitan Institute of Medical Science, Tokyo, Japan, 3 Mammalian Genetics Laboratory, National Institute of Genetics, Mishima, Japan, 4 Mouse Genomics Resource Laboratory, National Institute of Genetics, Mishima, Japan

Open-field is one of the widely used methods for measuring emotionality. Recently, genetic mapping studies have reported some QTLs for open-field behavior. Most of them used two major variables, ambulation and defecation, that totaled per test. However, when a mouse is placed into the open-field, they show a variety of behavior such as scout and rear to look around, or groom their body. Also, frequency of those behaviors changes with time as the mouse accustomed to the environment. In this point of view, detailed analysis of the open-field behavior is important to understand psychological status of the mice. The focus of the present work was to investigate temporal analysis of the detail open-field behavior.

The wild mouse strains were established as inbred strains from wild mice. These strains are known to have a large inter-strain genetic diversity, because they were captured from many countries. In the present study, we demonstrated the descriptive analysis of the open-field behavior by scoring 12 component of the behavior on 10 wild mouse strains, a laboratory mouse strain C57BL/6, and a Japanese fancy mouse strain JF1. They showed a great diversity in the frequencies and the temporal changes of those open-field behaviors. The pattern of the behaviors was much more complicated than which were shown in ambulation or defecation. We next performed this detailed analysis of the open-field behavior on consomic mouse strains that were composed of wild MSM and laboratory C57BL/6. We found some cosomic strains showing a significant difference in the detailed open-field behaviors.

POSTER 59 - GENETIC AND BIOCHEMICAL CHARACTERIZATION OF AN ALLELIC SERIES OF SMAD4 MUTATIONS IDENTIFIED IN AN ENU-MUTAGENIZED MOUSE EMBRYONIC STEM CELL LIBRARY

Chen Y, Yee D, Magnuson T

University of North Carolina-Chapel Hill, Chapel Hill, United States

Genotyped-based screens in mutagenized mouse embryonic stem cells represent a rapid and cost-effective means of generating germline mutations valuable for the functional annotation of the mammalian genome.  Smad4 is central to the TGF--related signaling and is implicated in numerous developmental and disease processes.  Targeted mutations in Smad4 cause early lethality, making the study of its later functions difficult.  An allelic series of mutations in Smad4 was obtained by screening a cryopreserved ES cell library mutagenized with ENU using the DHPLC mutation detection technology.  The identified mutations were distributed throughout the coding region of the Smad4 gene.  The functional consequence of the mutations is assessed through germline transmission.  The phenotypes range from embryonic lethality to adult viability.  The genetic and biochemical characterization of this allelic series of mutations in Smad4 provides insights into the in vivo relationship of structure and function of this family of molecules. 

POSTER 60 - IDENTIFICATION OF POSITIONAL CANDIDATE GENES FOR GROWTH AND OBESITY

Aksu S, Neuschl C, Renne U, Koczan D, Thiesen HJ, Brockmann GA

Humboldt-University of Berlin, Berlin, Germany

One of the major challenges in modern biology is to understand the genetic basis of complex traits, such as body weight and fat accumulation. Obesity is often associated with many other severe diseases, for example, hypertension, diabetes, cardiovascular diseases, osteoarthritis, and certain cancers. In our study, we have used the high body weight selected mouse line DU6 and its inbred derivate DU6i, which contain the biggest mice known world-wide, and the mouse lines DBA/2 and DUKs as unselected controls. High body weight selected mice are twice as heavy and three times as fat as unselected control mice. Furthermore, the selection strain DU6 has highly elevated serum concentrations of insulin, leptin, IGF-I and IGF binding proteins. So far, many QTLs have been mapped for body weight, body composition and subcomponents of the endocrine control of growth in the cross between the lines DU6i and DBA/2. Microarray expression analysis showed, that 77 genes are differentially expressed in fat tissues between the selected and non-selected mouse lines. 10 out of 77 genes are located in QTL regions which affect body weight and fat accumulation. Semi-quantitative real-time PCR analysis of these 10 positional candidate genes confirmed the data obtained from microarray GeneChip analyses. Therefore we used these genes for further analyses. We comparatively sequenced promotor regions of 10 positional candidate genes in selected and control mouse lines to find DNA variants which could contribute to differences in gene-expression levels. We found polymorphisms which are identical for DU6 and DU6i. Some of these polymorphisms are located within potential transcription factor binding sites. There are candidate polymorphisms which could underlie the detected QTL effects.

POSTER 61 - ENU MUTAGENESIS IDENTIFIES A RECESSIVE MUTATION IN A DNAJ PROTEIN THAT RESULTS IN RETARDED GROWTH AND POLYDACTYLY

Webb T 1, McKie L 1, West K 1, Peters J 2, Cross SH 1, Jackson I 1

1 MRC Human Genetics Unit, Edinburgh, United Kingdom and 2 MRC Mammalian Genetics Unit, Harwell, United Kingdom

Chromosomal deletions can be used in mutagenesis stategies that screen for recessive phenotypes. Several mouse chromosomal deletions are available which remove the Pax6 gene and result in the small-eye phenotype. The Pax6Sey-1H deletion spans 3Mb and includes about 15 genes. We have used this deletion in an ENU mutagenesis screen for phenotypes revealed when hemizygous with the deletion. We have found one mutation from 233 pedigrees tested. The mutation is lethal against the Pax6Sey-1H deletion. Mice heterozygous for the mutation are phenotypically normal. When homozygous the mutation is generally recessive lethal by E14, and embryos have retarded growth as early as E9. However, some mice are born which are small and have extra digits on one or both hind limbs. Sequencing of the genes in the deletion interval identified a splice site mutation in a novel member of the Dnaj family of molecular chaperones. The mutation causes the in frame skipping of exon 4, leading to an mRNA that encodes a protein with an internal 23 amino acid deletion. We have expressed wild type and mutant proteins in E. coli for further studies. Expression in mammalian cells is underway.

POSTER 62 - THE MOUSE JUVENILE SPERMATOGONIAL DEPLETION (JSD) PHENOTYPE IS DUE TO A MUTATION IN MUTP14B, A MEMBER OF THE SNORNP COMPLEX

Rohozinski J 1, Bishop CE  2

1 Baylor College of Medicine, Dept. Obstetrics & Gynecology, Houston, United States, 2 Baylor College of Medicine, Dept. Molecular & Human Genetics, Houston, United States

The recessive juvenile spermatogonial depletion (jsd) mutation results in a single wave of spermatogenesis, followed by failure of type A spermatogonia to differentiate, resulting in adult male sterility. We have identified a jsd specific, rearrangement in the mouse homologue of the S. cerevisiae gene Utp14, provisionally termed mUtp14b. Confirmation that mUtp14b underlies the  jsd phenotype was obtained by transgenic BAC rescue. We also identified a homologous gene on the MMUX (mUtp14a) which is the strict homologue of the yeast gene, from which the intronless mUtp14b has been derived by retrotransposition. Expression analysis showed that mUtp14b is predominantly  expressed in the germ line of the testis from zygotene through round spermatids whereas mUtp14a, although well expressed in all  somatic tissues, could only be detected in the germ line in round spermatids. In yeast, depletion of the UTP proteins impedes production of 18S rRNA, leading to cell death. We propose that the retroposed autosomal copy, mUtp14b, having  acquired a testis specific expression pattern, could have provided a mechanism for increasing the efficiency and/or numbers of germ cells produced, by meeting the need for more 18S rRNA and protein. Such a mechanism would be of obvious reproductive advantage and be strongly selected for in evolution. Consistent with this hypothesis is the finding of a similar X-autosome retroposition of UTP14 in human which appears to have arisen independently of that in rodents. In jsd homozygotes, which lack a functional copy of Utp14b, insufficient production of rRNA quickly leads to a cessation of spermatogenesis.

POSTER 63 - MUTATIONS IN THE MOUSE AXONEMAL DYNEIN HEAVY CHAIN, DNAHC2, RESULT IN MALE INFERTILITY

Meehan TP, Justice MJ

Baylor College of Medicine, Houston, United States

Infertility is a problem affecting about 15% of couples with nearly equal numbers of male factor and female factor infertility.  The underlying defects in most of these cases remain unknown.  Over 200 mouse models exist for genes involved in infertility, however, since it is estimated that 5000 genes are involved in fertility, more research is needed to determine novel causes of infertility.  Towards this goal, we have utilized ENU mutagenesis coupled with a balancer chromosome.  Previously, we described the isolation and characterization of a novel male infertility locus on mouse chromosome 11 (Clark et al., 2004, BOR 70:1317-1324).  Three non-complementing mutant lines generated by ENU mutagenesis displayed reduced sperm counts and complete loss of sperm motility.  Additionally, defects were found in the ultrastructure of the sperm tail axoneme.  A candidate gene approach was utilized to determine the gene responsible for the defects.  Twenty-two infertile candidate genes are closely linked to the region of chromosome 11 containing the infertile locus.  Based on obtaining three alleles in one gene, we predicted that the mutated gene is large in size, therefore, we focused our investigation on Dnahc2, an uncharacterized member of the axonemal dynein heavy chain family which spans approximately 84 kb of genomic sequence.  Northern blot analysis and real-time RT-PCR revealed decreased Dnahc2 transcript levels in two of three lines.  Additionally, the infertility in the third mutant line shows tight linkage to Dnahc2 by fine structure mapping.  The predicted 79 exons of Dnahc2 result in an 11.7 kb transcript encoding a 3907 amino acid protein.  Sequencing of the Dnahc2 cDNA from mutant lines has revealed mutations in two lines.  One mutation results in a premature stop codon prior to the motor domain of the protein.  The other mutation is a Q to R transition in the fifth of six AAA (ATPases associated diverse cellular activities) domains.  This residue is conserved among other members of the dynein heavy chain family.  Together, these mutations demonstrate the critical role of Dnahc2 in sperm function, and will be useful for dissecting the action of the dynein motor in sperm flagella.

POSTER 64 - MOUSE DEAFNESS MUTANT LINES FROM THE RIKEN ENU MUTAGENESIS PROJECT

Minowa O, Inoue M, Sakuraba Y, Motegi H,  Toki H, Tada M, Kaneda H, Ishijima J, Masuya H, Kobayashi K, Suzuki T, Wakana S, Gondo Y, Shiroishi T, Noda T

RIKEN-GSC, Yokohama, Japan

  Development of syndromic and nonsyndromic deafness mouse models would greatly facilitate understanding the mechanisms of hearing impairment; the models would be useful tools to advance current methods of diagnosis and treatment of hearing impairment. In order to develop a novel human hereditary deafness mouse model, we used a ENU-mutagenesis screening platform established to identify various kinds of mouse models for human diseases. We extensively evaluated various visible, clinical, biochemical, hematological, and cardiovascular phenotypes as well as behavioral anomalies and tumorigenesis.

  Specifically, we performed primary screening for hearing impairment by observing the startle responses evoked by click box stimulation during the modified-SHIRPA procedure. Of the 5,000 G1 mice screened, 14 G1 mutant candidates were subjected to the auditory brainstem response (ABR) measurements to analyze the early phases of auditory function impairment. Of these 14 mutant candidates, 11 G1 mice showed ABR abnormalities and they had been clearly inherited by G2 generations. The heritable hearing phenotype had not been accompanied by any behavioral or other traits, which is compatible with human diagnostic-type nonsyndromic deafness. The threshold ABR values in these G1 founders varied in mice with slight hearing impairment to those with complete hearing loss, suggesting that multiple causative genes and/or a variety of alleles of certain genes exist among the mutants.

POSTER 65 - MICROARRAY ANALYSIS OF GENE EXPRESSION AT THE DEL(13)SVEA36H REGION OF MOUSE CHROMOSOME 13 DURING POSTIMPLANTATION MOUSE DEVELOPMENT

Willoughby C 1, Davies J 1, Underhill P 1, Holmes C 1, Greenfield A 1, Denny P 1, Ragoussis J 2, Arkell R 1

1 Mammalian Genetics Unit, MRC Harwell,, Oxfordshire, United Kingdom, 2 Wellcome Trust Centre for Human Genetics, Oxford, United Kingdom

Radiation induced chromosomal deletions in mouse have been used in the functional analysis of defined regions in a number of studies. Within the Mammalian Genetics Unit at MRC Harwell attention has focused on the annotation of the Del(13)Svea36H deletion on chromosome 13. This annotation has involved the sequencing and comparative sequence analysis of the region, in parallel with a screen for recessive ENU-induced mutations at the region. The screen has identified 11 embryonic lethal mutations and as part of the ongoing analysis we aim to clone the mutated genes. In this type of cloning experiment a key stage is the identification of the expression pattern of genes within the critical region. We have decided to utilise microarray technology to assist in this positional candidate approach to identification of these mutations. We have used whole genome spotted oligo arrays to create an expression profile of the genes in the Del(13)Svea36H deletion between 6.0 days post coitum (dpc) and birth. Since the experiment utilises whole genome mouse arrays it also constitutes a profile of postimplantation mouse development which has not previously been undertaken.

POSTER 66 - MAPPING AND POSITIONAL CLONING OF ENU-INDUCED MOUSE DEVELOPMENTAL MUTANTS WITH A WHOLE GENOME SNP PANEL AND EXON-BASED SEQUENCING

Moran JL 1, Yun Y 1, Brown A 1, Bjork BC 1, Tran P 1, Li C 2, Salinger AP 3, Poirier C 3, Zhang B 1, Montgomery K 1, Bishop C 3, Justice MJ 3, Wiltshire T 4, Kwiatkowski DJ 1, Kucherlapati R 1, Beier DR 1

1 Brigham and Women's Hospital, Harvard Medical School, Boston, MA, United States, 2 Dana Farber Cancer Institute, Harvard School of Public Health, Boston, MA, United States, 3 Baylor College of Medicine, Houston, TX, United States, 4 Genomics Institute of the Novartis Research Foundation, San Diego, CA, United States

ENU mutagenesis has proven to be an efficient means for generating novel developmental mutations in the mouse. However, the task of localizing and characterizing these mutations remains a significant barrier to their general utilization. We have established a consortium for the genetic mapping, phenotypic characterization and positional cloning of ENU-induced mutations affecting developmental processes.  In order to expeditiously map mutations at moderate resolution, we have assembled a genome wide panel of over 400 single nucleotide polymorphisms (SNPs).  Thus far, confirmed linkage of 5 recessive mutants has been obtained by genotyping 10-11 affected mice.  Genetic intervals of 21-38 Mb identified with the SNP panel were subsequently reduced to 2-13 Mb by genotyping additional affecteds with microsatellite markers.  The SNP panel offers several advantages over conventional microsatellite marker panels: higher marker density, which enables the detection of whole genome linkage information, and the ability to map mutations with small numbers of mice independent of genetic background. Preliminary chromosomal localization of a genetic modifier suggests that the panel potentially has use beyond mapping ENU-induced mutations.

In order to identify causative mutations in ENU mutants, a high-throughput pipeline for sequencing candidate genes in recombinant intervals has been developed.  The pipeline is exon-based and includes semi-automated primer design, primer ordering, amplification, sequencing, and data analysis.  Both the SNP genotyping and sequencing platforms are available to interested investigators.  Using these genetic mapping and sequencing methods, we have identified a causative mutation in the mummy mutant, a recessive perinatal lethal mutant with an epidermal permeability barrier defect.

POSTER 67 - PHENOTYPE OR GENOTYPE? THE ORGANISM FROM DIFFERENT BIOLOGICAL PERSPECTIVES

Eppig JT, vandenBorre P, Lu I, Anagnostopoulos A, Burkart DL, Cassell MA, Dene H, Bello SM, Washburn LL, Lutz CM, Goldsmith CW, Smith CL

The Jackson Laboratory, Bar Harbor, United States

The major goal of model organism databases is to provide an integrated resource of biological and genomic information that supports knowledge discovery. The Mouse Genome Database (http://www.informatics.jax.org/) provides this entrée into mouse as a model system. The genome sequence of mouse that underlies the species’ biology is fundamental and recently obtained, and will be a source of revision, reassembly, analyses and discovery for years to come. At the opposite end of the biological system spectrum, the phenotype, representing the manifestation of the “whole organism” genotype is by far more complex. To represent phenotypic information in a useful way, biologists must be able to approach the data from both a phenotype-centric or genotype-centric view. For example, ‘”what phenotypes are associated with mutations in this gene or in this genomic region?” or “In this genomic region what sequence changes affect heart development?” (genotype-centric) versus “What are the mouse models for Hermansky-Pudlak syndrome” or “ Which mice show insulin resistance and normal serum glucose?” (phenotype-centric). Representing these data robustly in MGD in ways that support variable knowledge resolution and integrate data with other biological information is challenging, but good initial steps in this direction are being developed. MGD is collaborating with RGD and Mouse Mutagenesis Centers on developing and annotating phenotypes using a shared structured vocabulary, the Mammalian Phenotype (MP) Ontology. MGD is expanding its phenotype data model and developing a new suite of user interfaces. These user interfaces take advantage of the MP ontology and allow users to query and view data from the genotype or phenotype perspectives. Progress on this work will be described as well as some of the remaining challenges.

Supported by NIH grants HG00330

POSTER 68 - SOPDB: A RESOURCE OF STANDARD OPERATING PROCEDURES FOR MOUSE PHENOTYPING

Green ECJ, Gkoutos GV, Weekes J, Mallon AM, Hancock JM

Mammalian Genetics Unit, Oxfordshire, United Kingdom

EUMORPHIA1 is an integrated research programme involving the development of new approaches in phenotyping, mutagenesis and informatics leading to improved characterisation of mouse models for the understanding of human physiology and disease.

The SOPdb resource is an XML database2 of Standard Operating Procedures (SOPs) that constitutes the European Comprehensive First-Line Phenotyping Protocol for mice (ECFLP)3 with a web interface.  The ECFLP is a precise, evolving, working manual of SOPs for primary pathophysiological analysis of mouse mutants, which will be used in all reference laboratories contributing to the EUMORPHIA project.  SOPdb allows storage, indexing, searching, retrieval and transformation of SOPs taking advantage of the XML technology.

A major challenge in modern biology is to link genome sequence information to gene function. In many organisms this is being done by characterising phenotypes that result from individual mutations. Efficiently expressing phenotypic information requires combinatorial use of ontologies.4 SOPs are vital for describing the phenotype they attempt to measure, the units and the manner of measurement as well as possible interpretations of observations.4 SOPdb will provide links to Assay ontologies to facilitate phenotypic descriptions generated according to our recently proposed schemal4.

References

1.EUMORPHIA: Understanding Human Disease Through Mouse Genetics. Available on line at: http://www.eumorphia.org/

2.eXist: Open Source XML Database. Available on line http://exist.sourceforge.net/

3.Coghill E.L., Lad H. and the EUMORPHIA Consortium, EUMORPHIA and the European Comprehensive First Line Phenotyping protocol, IMGC, 2004.

4.Gkoutos G.V., Green E.C.J., Mallon A.M., Hancock J.M., Davidson D. (2004). Building mouse phenotype ontologies. Pac Symp Biocomput., 2004, 178-89.

POSTER 69 - STUDY OF THE GENES INVOLVED IN THE HYPERTROPIC LANGERHANS ISLAND OF THE KK-AY MOUSE AND THEIR HISTOLOGICAL LOCALIZATION

Kusakabe M 1, Matsuba K 2, Inoue J 2, Aotsuka S 2, Hokao R 1, Hashimoto H 3

1 Institute for Animal Reproduction, Ibaraki, Japan, 2 ANB Tsukuba Institute, Aloka Co., Ltd., Ibaraki, Japan, 3 Jikei Univ. Sch of Med., Dept. Anat., Tokyo, Japan

A KK-Ay mouse is the experimental animal model for type II diabetes. Interestingly, it was found that the Langerhans Island of KK-Ay becomes hypertropic with age. Therefore, the purpose of this experiment is to search the genes, which are involved in the hypertropic Island, and to examine the histological localization of these genes.

The two-way subtraction libraries were made between the pancreas of BALB/cA, 10 week-old, and one of KK-Ay, 10 week-old.  After cloning of these libraries, we picked up at lease 1200 clones from each library and arrayed them on the membrane filters. These filters were then hybridized with the probes prepared from cDNA of each pancreas. The characteristic clones were selected and sequenced. Then, homology and cluster analysis were performed. Next, PCR analysis was done to select the clones, which are different in amount between diabetes and control. Finally, the probes for in situ hybridization were made from the selected clones.  We selected 162 clones from KK-BALB library (K-B) and 153 from BALB-KK library (B-K) as specifically expressed clones. After homology and cluster analysis, 28 genes in K-B and 24 in B-K were identified. Two unknown genes in K-B and 6 in B-K were also detected. After pcr, 9 in K-B and 10 in B-K were identified as more specific genes. Reg I, Reg II and Reg III beta, which are involved in regeneration, were specifically enhanced their expression in the KK-Ay mouse. At present, in situ hybridization to know histological localization of these genes is under way.

POSTER 70 - IDENTIFICATION AND LOCALISATION OF MUTATIONS IN THE DEL(13)SVEA36H DELETION

McKeone R, Rowe C, Polley S, Wells S, Arkell R, Davies J, Bogani D, Dear N, Denny P

MRC Harwell, Didcot, United Kingdom

The Del(13)Svea36H (or Del36H) mutation is associated with a deletion of about 12.7 Mb of mouse chromosome 13.  Mice that are heterozygous for Del36H are viable and fertile and have been used to identify ENU-induced recessive-lethal mutations.

In order to facilitate positional cloning of these recessive mutations, sequence data from BAC clones has been used to generate a set of microsatellite markers across the Del36H region. These markers were then tested for polymorphism between each strain used in the screen.

To complement the lethal recessive screen, a gene-driven screen is underway to discover additional mutations of genes in the region. DNA from the F1 progeny of mutagenised individuals from the MRC Harwell ENU mutagenesis programme has been screened using dHPLC.  The first gene to undergo screening, Foxf2, is a forkhead family transcription factor gene.

A Trp to Arg mutation has been discovered in the DNA binding domain of Foxf2, which occurs within a potential -sheet structure. The tryptophan at this position is conserved in all members of the forkhead family and so may represent a critical residue.  Mice with this mutation have been rederived by in vitro fertilisation using cryopreserved spermatozoa from the ENU archive and outcrossed onto C3H.

Heterozygous mutant animals have undergone a battery of phenotype tests in order to detect dominant phenotypes associated with the Foxf2 mutation; so far, none has been uncovered.  Two homozygotes have been born, each dying several days after birth.  This is in contrast to the phenotype reported for mice homozygous for a null allele of Foxf2 produced by gene targeting, which die by 12 hours after birth.  It is possible that the ENU induced mutant allele is hypomorphic, contributing to observed longer survival time.

POSTER 71 - BEHAVIORAL CHARACTERIZATION AND LINKAGE ANALYSIS OF AN ENU-INDUCED MOUSE MUTANT LINE THAT SHOWS LEARNING DEFICIT, REDUCTION OF BODY SIZE, CONVULSIVE SEIZURE, AND TRANSIENT IMMOBILITY

Furuse T, Wada Y, Masuya H, Kaneda H, Kobayashi K, Kawai A, Kushida T, Nishii R, Gondo Y, Noda T, Wakana S, Shiroishi T

RIKEN GSC, Tsukuba, Japan

We are carrying out a large-scale screening for behavioral mutant mice induced by ENU. A mutant line designated as M100200 was identified in the screening, and showed learning deficit in the passive avoidance test. In further analysis, three characteristic phenotypes, reduction of body size, convulsive seizure, and immobility were observed in the backcross progeny. In-depth characterization indicated reduction of body weight, short body length, and small body mass index specifically in the male mice. The convulsive seizure and immobility start suddenly with vocalization, ataxic gait, lowered posture and slow movement. The symptom starts 30-60 minutes after mice were transferred to new home cage. When the mutants are immobilized, they continuously open their eyes, sometimes move their vibrissae, and show prayer reflex. They move a few steps in response to human contact. Thus, affected mutant mice keep conscious during immobility. A single dominant mutation may cause the multiple symptoms in M100200 line, because the several symptoms are co-segregated in the back-cross progeny (N6). To identify the causative gene, we carried out linkage analysis. Backcross progeny was phenotyped for the immobility and genotyped for SNPs markers. As a result, the causative gene was mapped to chromosome 4. Further fine mapping and detailed behavioral characterization are now underway.

POSTER 72 - THE NOX3 AND NOXO1 GENES, ENCODING PRESUMPTIVE MEMBERS OF AN NADPH OXIDASE COMPLEX, ARE REQUIRED FOR NORMAL VESTIBULAR FUNCTION AND DEVELOPMENT

Bergstrom DE 1, Bergstrom RA 1, Munroe RJ 2, Schimenti JC 2, Gagnon LH 1, Johnson KR 1, Heinzmann U 3, Stumm G 4, Paffenholz R 4

1 The Jackson Laboratory, Bar Harbor, United States, 2 Cornell University, Ithaca, United States, 3 GSF-National Research Center for Environmental Health, Neuherberg, Germany, 4 Ingenium Pharmaceuticals, Martinsried, Germany

The mammalian inner ear consists of the cochlea (required for auditory sensation) and the vestibular system (required for balance sensation). The vestibular system is comprised of the three semicircular canals that detect angular acceleration, plus the saccule and utricle and their associated maculae that detect linear acceleration.

The head tilt (het) and head slant (hslt) loci of mice are required for the normal development and function of the vestibular system. Head tilt and head slant mutant mice display a complete absence of saccular and utricular otoconia, calcium-containing crystals that act as inertial masses affixed to the maculae. Mutant mice that lack these otoconia display several behaviors consistent with vestibular impairment including circling, an abnormal tilting posture, impaired righting reflexes, and inability to swim.

Recombination- and deletion-based positional cloning strategies have identified Nox3 and Noxo1 as the causative genes underlying the het and hslt phenotypes, respectively. Sequence analysis shows that Nox3 is paralogous to gp91phox (Nox2), an NADPH oxidase of immune cells that generates reactive oxygen species as a bacteriocidal weapon against invading microorganisms. Similarly, Noxo1 is paralogous to p47phox (Ncf1), a cytosolic component of the same oxidase complex.

Based on these results, we have hypothesized that a novel NADPH oxidase complex, composed of immune complex paralogs, is present in the vestibular system and acts to initiate otoconial development from components in the endolymph including otoconin 90, Ca2+, otopetrin 1, and HCO3-.

POSTER 73 - HOLOPROSENCEPHALY AT DEL(13)SVEA36H

Bogani D 1, Willoughby C 1, Mirza G 2, Davies J 1, Ragoussis I 2, Arkell R 1

1 Mammalian Genetics Unit, Medical Research Council, Harwell, Didcot, United Kingdom, 2 Wellcome Trust Centre for Human Genetics, Department of Genomics, Oxford, United States

Holoprosencephaly (HPE) is a relatively common developmental defect of the forebrain and often the midface in humans that can be due to both genetic and environmental causes and involves incomplete development and septation of midline structures of the central nervous system. HPE has a prevalence of 1:250 during embryogenesis and, due to lethality, of 1:16000 newborn infants. During a screen for recessive mutations at a defined region of mouse chromosome 13 (Del(13)Svea36H) we have isolated several mutants that exhibit HPE as their main embryonic phenotype. The phenotype is frequently visible at 9.5 dpc and represented by the typical reduction of the distance between sagittally symmetrical structures of the forebrain (optic vesicles, etc.). Monozygosity for the telomeric region of human 6p, largely syntenic with Del(13)Svea36H, has been associated with HPE in aborted foetuses amongst other craniofacial and congenital abnormalities. This points to a new HPE locus in humans and to the possibility that the mutated gene in our mutant lines is the mouse homologue. Recombination mapping results indicate that at least 3 mutations map to overlapping regions of Del(13)Svea36H. Complementation testing and further mapping will confirm whether the mutated gene is the same in these lines. Mutation detection in candidate genes is also being carried out together with more detailed phenotypic analysis. Results of these experiments will be presented.

POSTER 74 - CHARACTERIZATION OF THE REGULATORY MUTATION MVWF1, A MAJOR CAUSE OF PROLONGED APTTS IN MICE.

Johnsen J 1, Ginsburg D2

1 University of Michigan, Ann Arbor, United States, 2 Howard Hughes Medical Institute, Ann Arbor, United States

We previously identified a modifier gene, Mvwf1, as the cause of low levels of von Willebrand factor (VWF) in the inbred mouse strain RIIIS/J. Mvwf1 is a regulatory mutation in the gene encoding an N-acetylgalactosaminyltransferase, GALGT2, resulting in a tissue-specific switch in gene expression from intestinal epithelium to vascular endothelium. This ectopic expression results in aberrant post-translational modification of VWF, leading to accelerated clearance. The specific regulatory DNA sequences responsible for this remarkable transcriptional switch have not yet been identified.  To further characterize this regulatory mutation, we generated transgenic mice with overlapping C57BL/6J bacterial artificial chromosomes (BACs).  These experiments narrowed the genomic region sufficient for wild-type (C57BL6/J) intestine-specific Galgt2 expression to 83kb.  Direct sequence analysis of RIIIS/J identified a 35kb region of 2-3% divergence from C57BL6/J flanking Galgt2 exon 1.  To address the possibility of intergression, we studied strains ~2% divergent from Mus musculus.  Neither Mus spretus nor Mus spicilegus were similar to the Mvwf1 region.  PCR analysis of 49 strains identified nine strains, including four wild-derived strains, which share >55kb of the Mvwf1 haplotype block and the tissue-specific switch in Galgt2 expression.  Analysis of The Jackson Laboratory Mouse Phenome Database indicates that this haplotype and the corresponding Mvwf1 mutation account for five of the six highest aPTTs in the database.  This implies that reduced VWF:FVIII due to this unique Mvwf1 Galgt2 allele is the major cause of prolonged aPTTs in the mouse, and the prevalence of this allele suggests that it has been maintained through positive selective pressure.

POSTER 75 - GENETICS OF GENE EXPRESSION OF THE INSULIN SIGNALING PATHWAY IN POLYGENIC OBESITY

Ferrell AD, Allan MF, Pomp D

University of Nebraska, Lincoln, United States

An estimated 65% of U.S. adults are overweight and ~30% are obese.  Polygenic mouse models play a pivotal role in understanding the roles of specific loci in genetic predisposition to obesity.  We are applying transcriptome mapping, whereby gene expression phenotypes are evaluated within a QTL mapping population, to identify candidate genes for obesity predisposition and to better understand the genetic architecture of complex traits in general.  Using a cross between the M16 (selected for rapid growth rate for 27 generations) and ICR (unselected control) lines, we have created a large (n=1200) F2 population segregating for phenotypes relevant to obesity, food consumption, and NIDDM. We are hybridizing mRNA, extracted from liver samples of F2 mice (n=88) with extreme body fat phenotypes, to probes for 96 well-characterized genes in the insulin signaling pathway.  This macroarray (GEArray Q Series, SuperArray) contains genes representing insulin receptor-associated proteins, the PI-3 kinase and MAPK pathways, primary targets for insulin signaling, secondary effector targets for insulin signaling, and targets for PPARg and SREBP-1. Analyses for gene expression endo-phenotypes and primary energy-balance phenotypes will be merged, and evaluated relative to the overall genetic correlation structure of the population, to reveal cis- and trans-acting expression QTL that combine to regulate the insulin signaling pathway and contribute to the overall regulation of polygenic obesity.

POSTER 76 - TWO NOVEL ENU-INDUCED CIRCADIAN RHYTHM MUTANTS IN MICE

Siepka SM 1, McGurk R 1, Chen M 1, Marusawa A 1, Olson D 1, Walkowiak J 1, Pinto L 1, Takahashi JS 2

1 Center for Functional Genomics, Northwestern University, Evanston, United States, 2 Howard Hughes Medical Institute, Evanston, United States

In the last decade, tremendous progress has been made in identifying genes and gene products that appear to be key components of the mammalian circadian clock. In mammals, genetic screens for circadian rhythms mutations have been limited in both scale and scope. We are currently engaged in a large-scale ENU mutagensis screen for recessive mutations affecting the nervous system and behavior (http://genome.northwestern.edu/neuro). One of the five domains of primary interest in this project is circadian behavior.

As a pilot study, we conducted a small three-generation recessive mutation screen using BTBR/J mice. Approximately 3600 mice from 217 different mutant pedigrees were screened for abnormalities in circadian locomotor activity. Two of these mutant lines produced mice with altered free running periods. The first mutant, part-time, is a recessive mutation with a free running period approximately 1.5 hours shorter (21.5 hours) than that of wild type mice (23.17 + 0.22 hrs). part-time maps to chromosome 10 and is a new mutant allele of the Cryptochrome1 gene. The second mutant, Overtime, is a semi-dominant mutation with a free running period more than 2.5 hours longer (25.9 hrs) than that of wild type mice. Overtime maps to mouse chromosome 14 and is particularly interesting because it maps to a chromosome where there are no known circadian rhythm genes. The identification and characterization of these new circadian mutants suggest that additional circadian genes remain to be identified.

POSTER 77 - AN INSERTION OF A RETROTRANSPOSON IN THE EDNRB GENE CAUSES PIEBALD COAT COLOR OF THE S/S MOUSE

Ohtani S 1, Yamada T 2, Sakurai T 2, Tsuji T 1, Yanagisawa M 2, Kunieda T 1

1 Graduate School of Natural Science and Technology, Okayama University, Okayama, Japan, 2 Howard Hughes Medical Institute, Department of Molecular Genetics University of Texas Southwestern Medical Center, Dallas, United States

JF1/Msf is an inbred mouse strain originated from Japanese fancy mice and shows piebald coat color controlled by s, a recessive allele of endothelin-B receptor (Ednrb) gene. The Ednrb gene plays an important role in differentiation and migration of neural crest cells including melanocytes and ganglion cells and the loss of function of the Ednrb gene cause abnormal coat color, megacolon, and sensorineural deafness. The expression level of this gene is significantly decreased in the s/s mice, but the exact mutation in the Ednrb gene of s allele has not yet been identified. We, therefore, analyzed the nucleotide sequences of the Ednrb gene in JF1/Msf mouse to reveal the causative mutation of the s allele. Southern blot analysis showed an approximately 5-kb insertion in the first intron of the gene. Sequencing analysis of this region revealed that the inserted sequence shows similarity with mouse ETn (Early retrotransposon). Comparison of the transcripts of the gene between wild-type and s/s mice revealed that the insertion of the retrotransposon results in abnormal transcripts caused by a cryptic splicing between the first exon and the inserted retrotransposon, and the decreased expression of the normal transcripts. We concluded that the piebald phenotype of the s/s mouse is caused by the insertion of ETn-like retrotransposon in the first intron of the Ednrb gene.

POSTER 78 - DOSAGE COMPENSATION BY UP-REGULATION OF X-LINKED GENES: MICROARRAY REVEALS TISSUE-SPECIFIC DIFFERENCES

Nguyen DK, Disteche C

University of Washington, Seattle, United States

S. Ohno (1967) hypothesized that up-regulation of genes on active Xchromosome would have occurred together with X inactivation to ensure dosage compensation between X chromosome and autosomes and between the sexes.  Using microarrays, we compared the mean expression of X-linked genes against autosomal genes in several tissue types, mammalian species and array types. The dataset that we analyzed was a combination of 587 arrays collected from different laboratories, public databases, arrays platforms and tissues from males and females. Our analysis showed that, overall, the mean expression of X-linked genes is similar to autosomal genes, consistent with up-regulation of the X chromosome in all mammalian species examined, including mouse, rat, human and other primates.  X chromosome up-regulation appeared to be largely independent of tissue type, gender or types of arrays.  However, there was a marked reduction in the expression of X-linked genes in sperm producing testis, consistent with X inactivation spermatogenesis.  Furthermore, the ratio of the mean expression of X-linked genes versus autosomal genes was significantly higher in the brain tissues of mouse, rat, human and primates, which may reflect the large number of brain-specific genes on the X chromosome.  In conclusion, analysis of the overall expression of X-linked genes in comparison to autosomal genes supports Ohno’s hypothesis of X chromosome up-regulation.

POSTER 79 - COMPARATIVE ANALYSIS OF THE NR1I3 LOCUS BETWEEN THE APN STRAIN AND A CHROMOSOME 1 APN.C3H/HEJ CONGENIC STRAIN WITH AN ALTERED DRUG METABOLISM PHENOTYPE

Casley WL, LeBlanc CA, Lavigne L, Nowakowska M

Health Canada, Ottawa, Canada

The APN mouse shows decreased capacity for caffeine metabolism relative to the C3H/HeJ mouse strain. We had previously demonstrated the presence of quantitative trait loci (QTLs) affecting 3-demethylation of caffeine on chromosomes 1, 4 and 9, using an APN x C3H/HeJ intercross. In the present study, we compare caffeine metabolism between the APN strain and a congenic strain in which the region carrying a QTL from chromosome 1 from C3H/HeJ is isolated on an APN background. The congenic interval for this strain is APN.C3H-(D1Mit495-D1Mit292). Phenotyping of the N13 generation demonstrated no significant difference for the QTL trait of caffeine 3-demethylation (p = 0.31) between the congenic and background strains, but a significant difference between the two for the trait of caffeine 7-demethylation (p = 0.003). The Nr1i3 locus, encoding the constitutive androstane receptor (CAR) lies within the congenic interval. CAR is known to regulate several Cytochrome P450 enzymes involved in drug metabolism. Hepatic expression of CAR mRNA was determined using a novel RT-PCR assay which allows the relative quantitation of both the transcript encoding the active, DNA binding form and the inactive splice variant. CAR mRNA was found to be overexpressed in APN.C3H-(D1Mit495-D1Mit292) relative to APN. Comparative sequencing of the Nr1i3 locus identified numerous SNPs as well as a 7 nucleotide insertion which may influence CAR expression.

POSTER 80 - IMPROVING EFFICIENCY OF TRANSGENIC RAT PRODUCTION

Filipiak WE, Saunders TL

University of Michigan Transgenic Core Facility, Ann Arbor, United States

The rat is an important animal model for cardiovascular, cancer, and pharmacological research. Quantitative trait loci (QTL) associated with these diseases have been mapped in the rat. The application of transgenesis with bacterial artificial chromosomes can refine QTL maps and accelerate the identification of disease genes. Academic transgenic facilities may hesitate to provide transgenic rat services because anecdotal evidence suggests that the production of transgenic rats is less efficient than transgenic mouse production. To compare transgenic efficiency between rats and mice we first established procedures for superovulation and pseudopregnant recipient production. To this end, we compared five different superovulation treatments and four methods of pseudopregnancy induction. The most productive superovulation treatment was 30 IU PMSG followed by 20 IU HCG 48 hours later. The best method to prepare pseudopregnant recipients combined estrus synchronization with LHRH agonist treatment and mating with vasectomized males. These procedures reduced the number of egg donors and the number of rats required pseudopregnant recipient production.

We made three Sprague-Dawley (SD) and one Fischer 344 (F344) transgenic rat models by pronuclear microinjection of DNA. The efficiency of SD transgenesis was one transgenic founder per four egg donors or 1.8% of injected eggs developed into transgenic founders. F344 transgenesis was less efficient: one transgenic founder per fifteen donors or 0.7% of injected eggs developed into founders. In our experience, transgenic efficiency with outbred SD rats is comparable to transgenesis in inbred mouse lines such as FVB/N (one founder per seven donors or 1.0% transgenic eggs) and C57BL/6 (one founder per six donors or 3.3% transgenic eggs). Transgenesis in inbred F344 rats required more effort because of lower egg yields and lower birth rates.

The efficiency of SD rat transgenesis is similar to inbred mice. Transgenic facilities should be encouraged to investigate the feasibility of offering transgenic rat production. Identification of rat lines with desirable superovulation characteristics and birth rates may further improve the efficiency of rat transgenesis in the future.

POSTER 81 - RECIPROCATING WNT- AND BMP-DEPENDENT PATHWAYS DRIVE CARDIOGENESIS IN P19CL6 CELLS

Chen M 2, Asakura M 1, Nakamura T 1, Inoue H 1, Schneider M 1

1 Baylor College of Medicine, Houston, United States, 2 Agilent Technologies, Palo Alto, United States

Early steps for cardiac lineage specification are problematic to study in mammalian embryos, which has favored the use of pluripotent cells that recapitulate the onset of cardiac myogenesis.

In mouse P19 cells or their derivatives, it has been shown that bone morphogenetic proteins drive Nkx2.5 induction both via receptor-activated Smad transcription factors and by TGF beta-activated kinase-1 (MAP3K7), and that Wnts (a reported inhibitor of heart formation in avian or amphibian explants) can activate cardiogenesis both via a non-canonical pathway for Wnt11 and via the canonical beta-catenin pathway, acting prior to and upstream of the endogenous BMPs’ induction.

Here, we have undertaken a dissection of the BMP- and Wnt-dependent pathways in P19Cl6 cells, by DNA microarray expression profiling in the absence or presence of extracellular inhibitors.

Blocking Wnt signaling with soluble Frizzled protein (sFz-8/Fc) and BMP signaling with noggin inhibited the induction of sarcomeric myosin heavy chains, suggesting that sFz-8/Fc and noggin inhibited cardiac differentiation in these cells.  Blocking Wnt signaling with sFz-8/Fc revealed that many reported cardiogenic or mesoderm-inducing genes were Wnt-dependent in this system, including the growth factors Cripto and FGF-8 and the transcription factors brachyury/T and eomesodermin.  Interestingly, blocking BMPs with noggin also prevented the early induction of T, eomesodermin, Cripto, and FGF-8, along with Wnts 3 and 5a and the Wnt receptors.

Thus: (1) Reciprocating Wnt- and BMP-dependent pathways promote cardiogenesis in this model system. (2) Cripto and FGF-8 are inducted by dual Wnt- and BMP-dependent circuits.

POSTER 82 - EPIGENETIC EVENTS INFLUENCE THE VARIATION IN INBRED MOUSE STRAIN INNATE IMMUNE RESPONSES

Wells CA 1, Aung H 1, Himes R 1, Forrest A 1, Ravasi T 1, Grimmond S 1, Kasukawa T 2, Carninci P 2, Hayashizaki Y 3, Hume D 1

1 IMB, University of QLD, Brisbane, Australia, 2 RIKEN Genome Sciences, Wako, Japan, 3 RIKEN Genome Sciences, Yokohama, Japan

Pathogenic challenges are one of the primary evolutionary drivers on mammalian genomes. An effective innate immune system must recognise pathogens and respond appropriately; too little and the pathogen will colonise the host, too much may generate local tissue damage or septic shock.  Inbred mice provide a spectrum of innate immune responses to a variety of pathogens, and we have demonstrated by microarray expression profiling that each strain exhibits a unique transcriptional response to a single pathogenic challenge, LPS.  This study demonstrates that macrophages derived from a single individual also demonstrate heterogeneity of responses to the same pathogenic challenge. We surveyed the genomic databases using RIKEN full-length cDNA clones and demonstrate that most participants in the Tlr4 signalling cascade are alternatively spliced, yielding functionally distinct products.  We propose that most receptors and signalling molecules recruited to innate immune pathways are regulated at an allelic level. Monoallelic expression of key signalling components, coupled with a cellular bias toward particular splice variants, provides a combinatorial molecular diversity within a population of macrophages from a single individual. We demonstrate that this heterogeneity is generated at a clonal level, so presenting a wide repertoire of defensive responses against new and evolving pathogens.

POSTER 83 - DEVELOPMENT OF A WHOLE MOUSE GENOME MICROARRAY WITH EMPIRICALLY OPTIMIZED 60-MER OLIGONUCLEOTIDE PROBES USING AGILENT’S SUREPRINT TECHNOLOGY

Collins PJ, Sun H, Gao J, Nguyen K, Lin E, Doan TB, Giles S, Tang S, Fulmer-Smentek SB, Shannon KW, Webb PG

Agilent Technologies, Inc., Palo Alto, United States

The explosion in mouse genomic sequence availability in recent years has enabled the development of whole genome microarrays with the potential to revolutionize fundamental research.  Agilent’s whole mouse genome microarray was developed using a powerful validation process in which optimal probes were empirically selected to represent each gene in the mouse genome.  The first step was to define the set of gene elements by grouping transcript sequences from well-established public sources, such as RefSeq, Riken, NIA, and Ensembl, based on their similarities to one another and their overlap on the mouse genome.  Next, consensus sequences covering all the high quality transcripts in each of these gene bins were defined, and a set of ten candidate 60-mer probe sequences were computationally determined to represent each consensus region.  These candidate probes were hybridized with ten differentially expressing sample pairs.  From the resulting data one optimal probe was selected from each candidate set to specifically represent each consensus region.  Probes were selected to represent additional transcripts to ensure whole genome coverage.  The final set of 60-mer oligo probes have been in situ synthesized to create a 44,000-feature microarray format using Agilent’s SurePrint technology.  In this presentation we describe the novel probe selection process used in developing this whole genome microarray.  In addition, we demonstrate some aspects of the performance of this microarray in measuring differential gene expression.

POSTER 84 - THE MOUSE PANCCHIP: HIGH THROUGHPUT GENE EXPRESSION ANALYSIS OF THE ENDOCRINE PANCREAS

Mazzarelli J 1, White P 1, Gorski R 1, Brestelli J 1, Arsenlis A 1, Manduchi E 1, Katokhin A 2, Belova O 2, Bogdanova V 2, Elisafenko E 2, Gubina M 2, Nizolenko L 2, Perelman P 2, Puzakov M 2, Shilov A 2, Trifonoff V 2, Vorobjeva N 2, Voronin D 2, Zykov I 2, Kolchanov N 2, Kaestner K 1, Stoeckert C 1

1 University of Pennsylvania, Philadelphia, United States, 2 Institute of Cytology and Genetics SB RAS, Novosibirsk, Russia

The mouse pancreas-enriched microarray or mouse PancChip, provides a powerful tool for studying pancreatic development and diabetes research, allowing the investigation of gene expression for over 13,000 transcripts expressed in the pancreas. In designing the PancChip, one clone from mouse pancreas libraries was picked to represent each unique mouse transcript as identified by computational analysis using the Database of Transcribed Sequences (DoTS).

Computational and manual annotation of the transcripts has identified known and novel transcripts and their functions.  Using the PancChip, we show that these transcripts are expressed in purified islets and the pancreas.  The novel transcripts define new mouse genes when BLAT-aligned to the mouse genome.  Genomic alignment of the transcripts also identifies alternative transcripts of known genes that are expressed in the pancreas.  Interestingly, some transcripts that are expressed are non-coding RNAs and overlap microRNA genes aligned to the genome. 

Information regarding the mouse PancChip and the transcript annotation is accessible via the web site of the Endocrine Pancreas Consortium, EPConDB ( http://www.cbil.upenn.edu/EPConDB/index.shtml).

POSTER 85 - Functional Annotation of Human Genes by Gene-Driven ENU Mutagenesis in Mice

Michaud EJ 1,2, Culiat CT 1,2, Barker G 1, Cain KT 1, Carpenter DJ 1, Easter LL 1, Foster CM 1, Gardner AW 1, Geiger J 3, Guo ZY 3, Houser KY 1, Hughes LA 1, Kerley MK 1, Klebig ML1,2, Liu Z 3, Olszewski RE 1, Pinn I 1, Shaw GD 1, Shinpock SG 1, Wymore AM 1, Johnson DK 1,2, Rinchik EM 1,2,4

1Life Sciences Division, Oak Ridge National Laboratory, Oak Ridge, TN, United States 2The University of Tennessee–Oak Ridge National Laboratory Graduate School of Genome Science and Technology, Oak Ridge, TN; United States, 3SpectruMedix, PA, United States 4Department of Biochemistry, Cellular, and Molecular Biology, The University of Tennessee, Knoxville, TN, United States

The availability of the complete DNA sequence of the mouse genome, coupled with the development of high-throughput methods for rapid detection of single-nucleotide polymorphisms (SNPs), have made it practical to implement genome-wide, gene-driven approaches to mouse germline mutagenesis. Such gene-driven strategies enable the performance of whole-genome mutagenesis, and screening for alterations in any pre-selected gene(s). To complement embryonic stem-cell-based gene-driven mutagenesis resources, such as gene-trap libraries and banks of N-ethyl-N-nitrosourea (ENU)-mutagenized ES cells, we generated a cryopreserved bank of DNA, tissues (for RNAs and proteins), and sperm from 4,000 C57BL/6J mice that each carry a unique load of paternally induced ENU mutations. This ORNL Cryopreserved Mutant Mouse Bank (CMMB) is a source of induced, heritable SNPs in virtually every gene in the genome. The ability to produce an allelic series of mutations in any gene, where each mutation may have different degrees of severity of the mutant phenotype, is invaluable in discovering the full scope of biological functions of a gene. High-throughput Temperature Gradient Capillary Electrophoresis (TGCE) is used to identify mutations in any gene by heteroduplex analysis in pre-selected genes in the CMMB DNA/RNA panel, and mutant stocks are recovered by intracytoplasmic sperm injection (ICSI) from the parallel bank of frozen sperm. Thus, the CMMB will provide mouse models of a wide range of altered proteins for phenotypic, gene/protein-network, and structural biology-type analyses. We will present progress on the identification of ENU-induced mutations in the 4,000-member CMMB by TGCE, and the recovery of mutant mice by ICSI.

POSTER 86 - ISOLATION OF GENES IMPLICATED IN INNATE IMMUNE RESPONSE AGAINST LISTERIA MONOCYTOGENES USING ENU MUTAGENESIS SCREENING IN THE MOUSE

Rutschmann S, Du X, Huser N, Jiang Z, Hoebe K, Beutler B

The Scripps Research Institute, La Jolla, CA, United States

Infectious diseases are a leading cause of morbidity and mortality worldwide. It is known that genetic variation foretells susceptibility to many of these diseases in both humans and animals. Thus, the positional identification of genes that are critical for host defense against infections looms as one of the greatest challenges in immunology. The intracellular bacterium Listeria monocytogenes is one of the most studied models for such infections. Resistance to this pathogen relies on both innate and adaptive immunity, but many aspects of our defenses are still not completely understood. To gain deeper insight into the nature of these immune responses, infectious agents need to be considered in the context of their complex mammalian host. For this reason, an in vivo forward genetic approach has been undertaken. We have implemented a screen for germline mutant mice with enhanced susceptibility to Listeria infection. Mice were inoculated intravenously with Listeria and closely monitored for signs of illness. A total of 4319 mice have been screened, allowing us to isolate 9 transmissible mutations in genes that specifically disrupt the mouse resistance to Listeria. All phenodeviants identified were bred, and the progeny tested to ascertain transmissibility of the mutations. 3 of these mutant lines have been established as homozygous stocks, and are currently being outcrossed for positional cloning. Meanwhile, these mutant mice are subjected to detailed phenotypic analysis, including neutrophils counts and activity studies, inflammatory cytokine production following an infection with Listeria, macrophage phagocytic activity and cytokine production in response to various other stimuli.

POSTER 87- AN IN VIVO ENU MUTAGENESIS SCREEN TO IDENTIFY CYTOMEGALOVIRUS-SUSCEPTIBILITY GENES.

Crozat K, Georgel P, Hoebe K, Du X, Tabeta K, Beutler B

TSRI, La Jolla, United States

Natural Killer (NK) cells play a crucial role in the innate mechanisms directed against viral infections. Mice represent a convenient and well established in vivo model to study the antiviral functions of NK cells, as antibody depletion of these cells leads to severe susceptibility to the Mouse Cytomegalovirus (MCMV, a double-stranded DNA Herpes-family virus). Recently, the Ly49H receptor was shown to be of central importance in MCMV susceptibility. The Ly49H gene (which is absent in the MCMV-sensitive BALB/c strain) encodes a C-type lectin-transmembrane receptor which associates to an accessory factor, DAP12, containing an Immunoreceptor Tyrosine-based Activating Motif (ITAM) to promote target cell lysis. The ligand of the Ly49H receptor is the MCMV-encoded protein m157, which supports the idea that this receptor mediates a specific innate immune response against viruses. In addition, dendritic cells (DC) have been shown to produce type I interferons (IFN) in response to viral infections, thereby assisting the NK-dependent response that eliminates the pathogen. In order to provide a better understanding of the innate immune response against viral infections, we designed an in vivo screen of EthylNitrosoUrea (ENU) mutated MCMV-resistant (C57BL/6) mice and searched for animals with abnormal susceptibility to MCMV infection. A total of 17 MCMV-sensitive lines have been derived from this screen. In addition, we have shown that Toll-like Receptors (TLR) signaling is essential for an efficient anti-viral response, as evidenced by low cytokine production, high mortality rate and increased splenic viral titer upon viral challenge of Tlr9, Tlr3, MyD88 and Trif mutants. We have also identified a coat color mutant, “salt and pepper” (spp), with enhanced susceptibility to MCMV infection.

POSTER 88 - A SCREEN FOR NOVEL GENES INVOLVED IN TLR SIGNALING

Jiang ZF, Du X, Shamel L, Beutler B

The Scripps Research Institute, La Jolla, United States

Toll-like receptors (TLRs) play essential roles in inflammation and innate immunity. Individual TLRs respond to specific molecules of microbial origin. TLR4 has been identified as the sensor for lipopolysaccharide (LPS), a molecular component of Gram-negative bacteria. TLR2 responds to molecules derived from mycobacteria, yeast, and Gram-positive bacteria. TLR5 and TLR9 recognize bacterial flagellins, and unmethylated DNA bearing CpG motifs, while TLR3 and TLR7 recognize viral double-stranded RNA (dsRNA) and single-stranded RNA (ssRNA), respectively.  While many of the molecules involved in TLR-mediated signaling have been identified, others remain obscure, and numerous questions surround the sensing mechanism itself.  ENU mutagenesis has been used to identify novel genes in the process by screening for ENU-caused germline mutants that are defective to different ligands. Previous work in our lab has shown that LPS2/TRIF is required for mediating the signaling through TLR3 and TLR4, and that CD36 is involved in TLR2/6 mediated LTA sensing.  We now report additional mutations (Insouciant, Heedless, and Unmindful) that affect TLR sensing.  Insouciant macrophages show decreased responses to peptidoglycan (PGN), suggesting the mutated gene is involved in responses to one or more molecular components of commercial peptidoglycan preparations.  Heedless displays reduced response to both LPS and the tri-acylated bacterial lipopeptide PAM3CSK4, suggesting that an unknown protein might be involved in TLR4 and TLR2/1 (but not TLR2/6) sensing. Unmindful shows no response to Poly I:C and PAM3CSK4, as well as a reduced response to CpG DNA.  Each mutation is being expanded for positional identification.

POSTER 89 - FINE MAPPING OF THE TMEVP3 LOCUS

Bureau JF, Levy Acobas F, Levillayer F, Brahic M, Szatanik M

Institut Pasteur, Paris, France

Characterizing genes of susceptibility to complex disorders is one of the most promising challenges at the beginning of the XXI century. The demyelinating disease induced by the persistent infection of Theiler’s virus in the mouse spinal cord is studied as a model for Multiple Sclerosis. Susceptibility to the infection amongst mouse strains depends mainly to the ability of their immune system to eliminate the virus.

By crossing and analyzing congenic strains between the SJL/J and the B10.S strains, two loci of susceptibility, named Tmevp2 and Tmevp3, were located on chromosome 10 close to the Ifng gene. We decided to characterize the Tmevp3 locus.

A cluster of cytokines containing at least 3 genes: Tmevpg1, Ifng, and Il-Tif/IL22 is an interesting region located at the telomeric part of the Tmevp3 interval since major loci of susceptibility between the SJL/J and the B10.S are expressed in the immune system. We decided to test this hypothesis by haplotyping this cluster in 15 strains: 11 laboratory strains with known susceptibility to the persistence of Theiler’s virus and 4 wild-type derived strains. We searched for polymorphism located in Tmevpg1, Ifng, Il22, and Mdm1/Mdm2, the 2 first genes centromeric to the cluster. The haplotype of laboratory strains are divided in at least two groups. The first group contains the SJL/J and the NZB strains and the second group the 9 other ones. The good correlation between these haplotypes and susceptibility to Theiler’s persistence strongly suggests that the Tmevp3 locus is located in this cluster.

POSTER 90 - 3D, A NOVEL MUTATION THAT CONFERS DEFECTIVE SENSING BY TOLL-LIKE RECEPTORS 3, 7 AND 9

Tabeta K 1, Du X 1, Georgel P 1, Hoebe K 1, Goode J 1, Mann N 1, Mudd S 1, Sovath S 1, Shamel L 1, Crozat K 1, Csencsits L 2, Tarantino L 2, Cooke MP 2, Beutler B 1

1 The Scripps Research Institute, La Jolla, United States, 2 The Genomics Institute of the Novartis Research Foundation, San Diego, United States

The mammalian Toll-like receptors (TLRs) comprise a key interface between cells of the host and all classes of microbial pathogen.  By sensing nucleic acids, the endosomal TLRs 3, 7, and 9 play a particularly important role in the detection of viral infection, and permit the host to mount an immediate and efficacious anti-viral response.  In an effort to identify novel components of the TLR signaling apparatus, we have pursued a program of germline saturation mutagenesis with N-ethyl-N-nitrosourea (ENU) in mice.  The germline mutants are produced on the C57BL/6 background, and peritoneal macrophages from individual animals are screened for their competence to respond to TLR-dependent microbial inducers.  A strong phenodeviant called 3d was identified in the F3 generation as a non-responder to nucleoside-based molecules such as unmethylated CpG oligodeoxynucleotides, Resiquimod (a drug of the imidazoquinoline class) and Poly I:C.  TNFproduction induced by TLRs 3, 7, and 9 was completely prevented in 3d homozygotes.  However, heterozygotes were unaffected, and the ability to produce TNF in response to other bacterial compounds, sensed by TLRs 1, 2, 4, and 6, was intact. 3d homozygotes show extreme susceptibility to infection by mouse cytomegalovirus (MCMV) infection in vivo. After IP inoculation with MCMV (104 PFU), 3d mutants showed impaired production of type I and type II interferons, TNF, and IL-12 in serum (p<0.001).  With larger inocula (5 x 105 PFU) a splenic viral titer approximately ten-thousand fold higher than that in controls (p<0.05) and rapid mortality (p<0.0001)  were observed.  This novel recessive phenotype is fully penetrant on the C57BL/6 and C3H/HeN x C57BL/6 hybrid backgrounds.  The 3d mutation was mapped to mouse chromosome 19 using a panel of microsatellite markers and at present, has been confined to 0.9 Mbp critical region on 1700 meioses.  3d appears to encode a protein essential for TLR-mediated detection of nucleic acids, which is indispensable for effective innate defense against viral infection.

POSTER 91 - THE NONSENSE ALLELE OBLIVIOUS REVEALS A SENSOR OF DI‑ACYLGLYCERIDES ACTING IN CONJUNCTION WITH TLR2 AND TLR6

Hoebe K 1, Georgel P 1, Rutschmann S 1, Du X 1, Tabeta K 1, Mudd S 1, Sovath S 1, Shamel L 1, Hartung T 2, Zahringer U 3, Beutler B 1

1 The Scripps Research Institue, La Jolla, United States, 2 Department of Biochemical Pharmacology, University of Konstanz, Konstanz, Germany, 3 Research Center Borstel, Leipniz-center for Medicine and Bioscience, Borstel, Germany

The mammalian Toll-like receptors (TLRs) activate cells of the innate immune system when stimulated by diverse ligands of microbial origin.  In some instances, these ligands are directly engaged by the TLRs; however, this is not necessarily true in all cases. TLR2 recognizes multiple, structurally disparate microbial ligands, consistent with a requirement for co-receptors in ligand binding.  Using N-ethyl-N-nitrosourea (ENU), we generated the recessive immunodeficiency phenotype oblivious, in which macrophages show diminished awareness of the S-enantiomer of the di-acylated bacterial lipopeptide MALP-2 and lipoteichoic acid (LTA), together with an increased susceptibility to Staphylococcus aureus infection.  Oblivious macrophages readily detect the tri-acylated bacterial lipopeptide PAM3CSK4 as well as zymosan, revealing that some TLR2 ligands are activated via an Oblivious-independent pathway.  The gene responsible for the oblivious phenotype has been positionally cloned and was proven to result from a nonsense mutation in Cd36, which encodes a double-spanning cell surface protein expressed by murine myeloid cells.  A receptor for endogenous molecules including fatty acids, thrombospondin, oxidized LDL, and b -amyloid, CD36 is also a selective and nonredundant sensor of microbial di-acylglycerides, and makes an important contribution to anti-bacterial defense.  We propose that CD36 may initiate TLR2-mediated inflammatory responses through recognition of both endogenous and exogenous ligands.

POSTER 92 - SUSCEPTIBILITY OF INBRED MICE TO FATAL INFECTION BY MOUSE ADENOVIRUS TYPE 1

Spindler KR, Welton AR

University of Michigan, Ann Arbor, United States

Mouse adenovirus type 1 (MAV-1) causes acute and persistent infections in inbred and outbred mice.  Aspects of the host immune response important for control of MAV-1 infection include T and B cell function and the antiviral cytokines IFN-g and IFN-a/b   Infection of immunodeficient mice can result in pneumonia, hepatitis, encephalitis, gastroenteritis, and disseminated disease involving multiple organs.  These resemble the disease phenotypes occurring in infection of immunocompromised people; the incidence of human adenovirus disease is increasing with the increased number of immunosuppressed patients, particularly pediatric bone marrow transplant recipients.  Identifying host factors involved in susceptibility to adenovirus infections is thus of importance in transplantation and also because of projected uses of adenovirus vectors in cancer treatments.  Study of human adenoviruses in animal hosts is not possible due to the species-specific nature of adenoviruses. 

MAV-1 is an excellent model for studying susceptibility, because there are significant differences in susceptibility to MAV-1 in different inbred mouse strains. 

SJL/J mice are susceptible to MAV-1, succumbing to a low dose infection 8-10 days post infection with high virus loads in the brain and with a 50% lethal dose (LD50) more than 4 log units lower than other strains of mice, including C3H/HeJ, BALB/cJ, 129/SvEv, and C57BL/6J (1).  When resistant mice are sublethally irradiated, they become susceptible to MAV-1 infection. Primary embryo fibroblasts and primary peritoneal macrophages from resistant and susceptible mice produce equal yields of MAV-1 when infected in culture.  These data suggest that systemic factors, such as components of the innate or adaptive immune system, play a role in MAV-1 susceptibility.  Congenic strains of mice that differed only in their H-2 haplotype were tested for susceptibility, and the results indicated that H-2 or closely linked genes are not a factor in susceptibility to MAV‑1.  Infection of (BALB/cJ x SJL/J)F1 indicated that susceptibility is semidominant.  A genome scan analysis of a 200 mouse (BALB/cJ x SJL/J)F1 x BALB/cJ backcross panel is in progress to develop a low resolution linkage map for susceptibility to MAV-1.  Identification of the gene(s) involved in MAV-1 susceptibility will provide important insight into the contribution of host factors to viral infection.

(1) Spindler, K. R., Fang, L., Moore, M. L., Brown, C. C., Hirsch, G. N., and Kajon, A. K. (2001). J. Virol. 75, 12039-12046.

POSTER 93 - DEVELOPMENT OF PATHOLOGY IN A/J, C57BL/6 AND BALB/C MICE FOLLOWING TRYPANOSOMA CONGOLENSE INFECTION

Mburu D 1, Naessens J 1, Noyes H 2, Kemp S 2, Gibson J 3, Iraqi F  1

1 International Livestock Research Institute (ILRI), Nairobi, Kenya, 2 University of Liverpool, Liverpool, United Kingdom, 3 University of New England, Armidale, Australia

Different inbred mouse strains show significant differences in innate resistance to Trypanosoma congolense infection. A/J mice are highly susceptible and C57BL/6 mice are relatively resistant, while BALB/C mice are intermediate. We investigated the development of pathology during the acute phase of trypanosome infection in these three strains. A total of 240 mice of A/J, C57BL/6 and BALB/C were challenged with 104 blood stream form of T. congolense IL-1180 parasite. At day 0, 3, 5, 7, 9, 13 and 17 post-infection the body weight, skin temperature and parasitaemia, were recorded. At each collection point 30 mice were sacrificed and the spleen, liver and kidney were examined and weighed. On day 3 and 13 there was a significant drop in the skin temperature in all the strains. This drop corresponds to the first parasitaemia and the peak of infection, respectively. On day 3 the associated gross pathology showed a marginally enlarged spleen (140%) but no changes were observed on the liver and kidney. On day 13 the spleen increased twelve fold while the liver and kidney increased two fold. In spite of the phenotypic differences in their response to T. congolense infection the gross pathology was consistent across the three strains. From day 5 onwards the lymph nodes were enlarged and petechial haemorrhages were observed in some mice. There was no significant change in body weight and the observed parasitaemia re-affirmed that A/J are highly susceptible. Tissues have been collected for histological studies to investigate the architectural changes.

POSTER 94 - THE CAPACITY TO CONTROL OF ANEMIA DEVELOPMENT AND PARASITEMIA GROWTH IS UNDER DIFFERENT GENETIC MECHANISMS IN AFRICAN MURINE TRYPANOSOMOSIS

Naessens J 1, Mburu D 1, Noyes H 2, Kemp S 2, Gibson J 3, Iraqi I  1

1 International Livestock Research Institute (ILRI), Nairobi, Kenya, 2 University of Liverpool, Liverpool, United Kingdom, 3 University of New England, Armidale, Australia

The A/J (A) mouse strain is more susceptible to infection with African trypanosomes and die earlier than the more resistant C57BL/6 (BL6) mice, while BALB/c (BALB) is known to be intermediately susceptible. In an early study, trypanotolerance QTL associated with survival time, following infection with the disease, were mapped to chromosomes 17, 5 and 1, using two F2 resource populations (C57BL/6 x A/J and C57BL/6 x BALB/c), and designated as Tir1, Tir2 and Tir3, respectively. Subsequently, these QTL were fine mapped using F6 advanced intercross line (AIL) populations. After infections with Trypanosoma congolense, A/J mice developed a higher parasitemia than BL6 mice. In contrast, the BL6 mice never develop such high parasitemia, but start dying in late stages of the infection. Searching for more parameters, which might underline the variation in susceptibility of the different mouse strains, we recorded the kinetics of anemia development in the three mouse strains following infection with T. congolense. BL6 mice developed anemia within a week after infection, and this lack of red blood cells got more severe with time. A/J mice also developed anemia early on, but recovered once the first parasitemic wave subsided. It seems likely that A/J mice die from pathology related to high numbers of parasites in the body, while BL6 mice die from pathology correlated with severe anemia. BALB mice also developed a high parasitemia, but were even better than A/J mice to control their anemia. These data suggest that the capacity to control parasitemia and the capacity to limit anemia are two unrelated genetic mechanisms.

POSTER 95 - MALARIA INDUCED PATHOLOGY OF TUMOUR NECROSIS FACTOR-Α DEFICIENT MICE AFTER INFECTION WITH PLASMODIUM CHABAUDI CHABAUDI 408XZ

Hernandez M 1, Naessens J 1, Musoke A 1, Rihet P 3, Ole-MoiYoi O 2, Iraqi F  1

1 International Livestock Research Institute (ILRI), Nairobi, Kenya, 2 Institute of Molecular and Cell Biology of Africa (IMCBA), Nairobi, Kenya, 3 Université de la Méditerranée, Marseille, France

Tumour necrosis factor-α, TNFa, plays a pleiotropic role during the murine malarial infection. It is involved in controlling the blood parasitaemia levels as well as the mortality rates of the host. Membrane-bound TNFa is involved in the most severe form of the disease, the cerebral malaria. Some studies have correlated TNFa with hypothermia, hyperlactataemia, hypoglycaemia and a suppression of the erythropoietic responses, although others have not supported the role of TNFa in the two latter clinical aspects. We have studied several parameters of pathology in TNFa-/-C57BL/6J and TNFa+/+C57BL/6J mice after infection with Plasmodium chabaudi adami 408XZ. Results have shown that blood glucose and lactate levels, and body weight are not significantly different between TNFa-/-C57BL/6J and TNFa+/+C57BL/6J mice during infection. However, TNFa-/-C57BL/6J mice significantly suffered a more severe anaemia and hypothermia than TNFa+/+C57BL/6J mice. In our model, it appears that TNFa is not involved in all the investigated aspects of the pathology of the malarial infection, although it affects negatively haemoglobin concentrations and body temperatures, essential parameters to overcome the disease. These results could explain the high mortality rates observed in TNFa-deficient mice and in mice treated with anti-TNFa antibodies.

POSTER 96 - EXPRESSION PATTERNS OF TLR GENES WITHIN THE CRITICAL TIR2 REGION

Nganga J 1, Nganga J 2, Iraqi F  1

1 International Livestock Research Institute (ILRI), Nairobi, Kenya, 2 Jomo Kenyatta University of Agriculture and Technology, Nairobi, Kenya

Trypanotolerance quantitative trait loci (QTL) in mice have been mapped to chromosome 1, 5 and 17 and named Tir1, 2 and 3. Tir2 QTL has been fine mapped to less than 1cM using F13 advanced intercross lines. The particular Tir2 region spanning 2KB on the mouse genome was examined for possible candidate genes. Plausible candidate genes in the locus included toll like receptor 1 (TLR1) and TLR6 among others. Rapid defense mechanisms are provided through recognition of pathogen associated molecular patterns by TLRs. Their expression and cooperation is essential for the induction of interleukins through their interaction with membrane glycosylphosphatidylinositol (GPI) anchors. Spleen and liver tissues were collected at day 0, 4, 7, and 10 post infection from 4 groups of each of the resistant C57BL/6J and susceptible AJ mice. Tissue specific expression levels of each of the two genes was determined using a semi quantitative RT-PCR. Analysis of variance of the mean ratios of ß-actin as the house keeping gene with each of the two genes revealed that the genes are regulated in a statistically significant fashion. The level of TLR1 increased between day 4 and day 7 in the resistant C57BL/6J while it declined in the susceptible strain over the same time in the spleen tissue. In the liver tissue, the gene was only expressed after day 4 in both strains. While the level of TLR6 increased at the onset of the infection in both strains, they kept raising in the susceptible AJ strain while its level remained constant but at a lower level in the resistant C57BL/6J strain. By comparing TLR1 and 6 levels in the resistant and susceptible livestock, we can better understand their role in trypanotolerance hence develop new trypanosomosis control strategies in livestock.

POSTER 97 - MODULARITY OF PLEIOTROPIC EFFECTS ON SKELETAL MORPHOLOGY

Kenney JP, Wong B, Pletscher LS, Cheverud JM

Department of Anatomy and Neurobiology, Washington University School of Medicine, Saint Louis, MO, United States

In order to understand the role of genetic variation in morphological evolution, it is important to investigate the genetic architecture of morphological traits. Aspects of genetic architecture include number and size of gene effects, dominance, epistasis, and pleiotropy.

Pleiotropy is predicted to evolve so that functionally and developmentally related traits are influenced by the same loci.  When functional systems evolve in concert, adaptive evolution is facilitated. Thus, the evolution of development is predicted to result in a nested hierarchy of gene effects reflecting the nested hierarchy of functional or developmental effects.  To date, there have been no comprehensive surveys of genetic architecture that reflect functional and developmental hierarchies.

We have undertaken a comprehensive survey of pleiotropic patterns affecting mouse skeletal morphology over the entire skeleton, including aspects of several modules and levels of morphological hierarchy, in order to elucidate the evolved genetic architecture of skeletal morphology and the consequences of this architecture for evolutionary processes. We predict that pleiotropic relationships will mirror the developmental and functional relationships of the mouse skeleton, and reflect the hierarchy of modules within the skeleton.

POSTER 98 - SEQUENCES MEET BIOLOGY: INTEGRATION OF SEQUENCES AS

DATABASE OBJECTS IN MOUSE GENOME INFORMATICS

Corbani LE, Baldarelli RM, King BL, Cousins S, Beal JS, Lewis J, Meiers DB, Walker MB, Kadin JA, Richardson JE, Blake JA, Ringwald M, Eppig JT, Bult CJ

The Jackson Laboratory, Bar Harbor, United States

The Mouse Genome Informatics (MGI) Database is a public resource that provides curated and integrated information on the biology and genetics of the laboratory mouse. With the recent 3.0 release, MGI has radically improved support for sequence data. Nucleotide and protein sequences are represented as distinct database objects, rather than accession ID attributes of other objects.  The full power of relational integration now connects sequence information to the existing MGI integrated platform of curated biological domains such as: genes, gene expression, gene function, phenotypes, strains, mammalian orthology, and chromosomal positioning, and soon to come, SNPs. MGI now stores information for all mouse sequences from GenBank, SWISS-PROT, LocusLink, RefSeq, Ensembl and NCBI gene models, TIGR and NIA Mouse Gene Indices, and DoTS. Information for each sequence includes source attributes such as library, strain, tissue, gender, etc. mapped to MGI controlled vocabularies whenever possible and sequence attributes such as type, description, provider, length, genome assembly coordinates (for gene models), etc. Users can query for sequences using any or all of the biological domains mentioned above and get results back in a variety of formats (HTML, tab delimited, FASTA). An additional aspect of this implementation is a robust and uniform system for loading sequence data, clone data, sequence-to-gene associations, and sequence-to-clone associations from a variety of sources, all within the context of MGI's commitment to high quality data.  Storing and integrating sequences as objects in MGI is a first step toward a much more precise representation of sequence features and attributes through the development of sequence feature maps.  The details and challenges of our implementation of sequences as objects and the benefits this brings to our user community will be presented. The MGI database can be accessed at http://www.informatics.jax.org.

POSTER 99 - IN SILICO POSITIONAL CLONING: A WEB SYSTEM FOR EXPLORATION OF RESPONSIBLE GENES IN MONOGENIC AND MULTIGENIC TRAITS

Toyoda TT, Masuya HM, Kawashima TK, Hasegawa YH, Sezutsu HS, Kaminuma EK, Mochizuki YM, Hirosawa KH, Heida NH, Gondo YG, Kawai JK, Wakana SW, Konagaya AK, Hayashizaki YH, Shiroishi TS

RIKEN Genomic Sciences Center, Yokohama, Japan

RIKEN Genomic Sciences Center has been producing a wide range of bio-resources and comprehensive data collections from genomic-level to phenomic-level. Now, various sorts of large-scale datasets are available for the estimation of responsible genes in the forward-genetics approaches. Here we introduce a web-based system named “Genome ó Phenome Superhighway (GPS),” by which users can browse all the genes and the available bio-resources that exist in a genetic-mapped interval, and can select promising candidate genes based on the available omic datasets, such as literature-based network (LIBNET), gene expressions (READ), transcript annotations (FANTOM), genetic maps (TraitMap), and other species’ syntenic information. Thus, GPS integrates the genome, transcriptome, proteome, metabolome and phenome resources (large-scale ENU-mutant mice), and provides bioinformatics-based strategies for the gene hunting. By specifying a chromosomal interval and a keyword, for example “diabetes,” a user can obtain a ranking list of candidate genes from not only the diabetes-related genes that reportedly involve in the disease, but also from the diabetes-associated genes that is reportedly associated with other known diabetes-related genes. Then the user can pick up most promising candidates out of the list by checking their expression levels in the mouse organs. In addition to the monogenic cases, it is necessary for the system to be effective in multigenic cases, because most ordinary traits are controlled by multiple genes. GPS associates every combination of genetic loci to biomolecular networks, and thus helps us to estimate molecular-level candidate networks responsible for a given multigenic trait. GPS is open to public at http://omicspace.riken.jp/gps/.

POSTER 100 - COMPARISON AND COMBINATION OF THREE MOUSE TISSUE-SPECIFIC MICROARRAY EXPRESSION DATA SETS

Morris QD1, Zhang W2, Robinson MD1, Frey BJ1, Hughes TR2

1 Electrical and Computer Engineering, University of Toronto, Toronto, Canada, 2 Banting and Best Department of Medical Research, University of Toronto, Toronto, Canada

Microarray expression data are useful for predicting gene functions, exploring gene regulation mechanisms, and uncovering novel aspects of physiology.  We have generated a microarray expression data set encompassing 55 tissues analyzed on custom ink-jet microarrays that represent over 42,000 known and predicted genes in the mouse draft genome sequence.  Here, we compare our data to the Novartis Gene Atlas (Su et al, 2004) and the Riken Transcriptome Analysis (Bono et al, 2003). 

To allow uniform comparison, we first mapped the array sequences from all three data sets to 34,343 MGI-curated genes.  Genes in any one data set were retained only if a single MGI gene was detected unambiguously.  We then compared the datasets on the basis of their ability to support the prediction of the known functional roles of genes shared by all three datasets.  We used Support Vector Machines (SVMs) to “learn” the relationship between each gene’s expression profile in each one of the datasets and that gene’s MGI Gene Ontology Biological Process (GO-BP) annotation(s).  Typically SVMs are used to predict new functional roles, here we use them to calculate the predictive value of each dataset for known functions on the basis of precision (% of predicted annotations that were confirmed by the annotation database) and recall (% of confirmed annotations that were predicted).  When restricting profiles to the 14 tissues shared by all three datasets, we found profiles drawn from our dataset to be more predictive of GO-BP annotation those of the other two.  Using the entire expression profiles of the shared genes (20 tissues and/or cell lines in Riken, 122 in Novartis, 55 in our data), we found the Novartis profiles generated more precise predictions at 10% recall but at higher recall, predictions based on our data were more precise.

We are currently producing a combined “validated” data set, by retaining only the gene pairs in each data set that contained a significantly positive “correlation of correlations” (Parmigiani, 2003) to the same pair in one of the other two data sets.  We expect this combined data set be superior to all three of the original data sets and we will distribute this validated dataset to the community once it is completed.

POSTER 101 - QTL REAPER: DEFINING QUANTITATIVE TRAIT LOCI WITH HERITABILITY-WEIGHTED MICROARRAY DATA

Manly KF, Wang J, Williams RW

University of Tennessee Health Science Center, Memphis, TN, United States

Heritable differences in transcribed RNA levels can be mapped as quantitative trait loci (QTLs). One source of data is microarray analysis of RNA transcripts in a set of recombinant inbred lines, such as that available at the WebQTL Web site, www.webqtl.org/search.html. QTL Reaper is software for rapid detection of QTLs in large data sets. It estimates empirical p-values by permutation tests, adapting the number of permutations to the significance of the QTL. The resulting p-values lend themselves to modern multiple-test statistical methods, which can define statistically significant sets of QTLs without excessive numbers of either false positives or false negatives.

Recent work with QTL Reaper has focused on improving methods for combining data from individual transcript-specific oligonucleotide probes. The use of recombinant inbred lines allows replicate measurements on genetically identical individuals, and replicates allow an estimate of the heritability of expression measured by individual oligonucleotide probes. This heritability varies greatly among probes, even among those designed to measure the same transcript. Since heritability is necessary (but not sufficient) to define a QTL, we have tested heritability-weighted averages to define expression of a transcript, averages in which each probe value is weighted by the heritability of that probe relative to the average heritability in the probe set. These averages allow detection of more QTLs than several other methods.

Supported by: Dunavant Chair, UTHSC, Center of Genomics and Bioinformatics, and the Human Brain Project (MH62009, NIMH)

POSTER 102 - THE MOUSE GENE EXPRESSION DATABASE (GXD):  INTEGRATED ACCESS TO EXPRESSION INFORMATION FOR THE LABORATORY MOUSE

Begley DA, Eppig JT, Finger JH, Hayamizu TF, Hill DP, Kadin JA, McCright IJ, Richardson JE, Smith CM, Ringwald M.

The Jackson Laboratory, Bar Harbor, United States

The Gene Expression Database (GXD) collects and integrates gene expression information about the developing laboratory mouse.  By combining diverse types of expression data, GXD provides information about the expression profiles of transcripts and proteins in different mouse strains and mutants, thus enabling insights into the molecular networks underlying developmental and disease processes.  Expression patterns are described using an extensive dictionary of standardized anatomical terms, making it possible to record expression results from assays with differing spatial resolution in a consistent manner.  GXD is integrated with the Mouse Genome Database and interconnected with other community resources, to include expression data in a larger biological and analytical context.  Data is acquired by direct curation from the literature, database downloads and electronic laboratory submissions.  To facilitate the direct submission of data to GXD we have developed the Gene Expression Notebook (GEN).  GEN functions as a laboratory notebook to store and organize expression data, assay details and images for in situ hybridization, immunohistochemistry, RT-PCR and Northern and Western blot experiments.  It also allows the researcher to effortlessly submit selected expression data to GXD.  Capturing these data will further enhance GXD’s utility as a community resource.  GEN is available at http://www.informatics.jax.org/mgihome/GXD/GEN/.  GXD is accessible through the Mouse Genome Informatics web site at http://www.informatics.jax.org/. 

GXD is supported by NIH grant HD 33745.

POSTER 103 - THE GENE ONTOLOGY CONSORTIUM: BIO-ONTOLOGIES FOR GENOME ANNOTATIONS.

Blake JA, Gene Ontology Consortium The

The Jackson Laboratory, Bar Harbor, United States

The goal of the Gene Ontology(GO) Consortium (http://www.geneontology.org) is to produce a controlled vocabulary that can be applied to all organisms even as knowledge of gene and protein roles in cells is accumulating and changing. GO provides three structured networks of defined terms to describe gene product attributes. GO is one of the controlled vocabularies of the Open Biological Ontologies effort.

The Gene Ontology development effort focuses on four domains of molecular information:  molecular function, biological process, cellular component, and sequence features.  Over 17,500 terms have been carefully defined and are represented in structured vocabularies that provide hierarchical representations of relationships between terms.  Research scientists are providing their expertise in particular sub-specialties to help with the continued evaluation and update of GO subtrees.

Most Model Organism Databases, including MGI and RGD, are now actively annotating genes and gene products to the GO vocabularies.  Almost 100,000 genes from over 20 model organisms have been hand-annotated to the GO, and tens of thousands more have computational assignments to GO terms.  All GO associations are supported by evidence and citation information.  Current priorities include analysis of annotation consistency between contributing research groups.

In addition to developing the GO resource and supporting genome annotation groups, the GO Consortium hosts Web and database resources to enable researchers to access the GO and to incorporate GO tools into their research pipelines.  Multiple research groups have contribute GO analysis tools that are hosted on the GO web site and that enable statistical analysis of gene product clusters based on shared GO annotations. 

The Gene Ontology Consortium is supported by a P41 grant from the National Human Genome Research Institute (NHGRI) [grant HG002273] and by grants from the European Union RTD Programme "Quality of Life and Management of Living Resources" [QLRI-CT-2001-00981 and QLRI-CT-2001-00015].

POSTER 104 - SNPLAD; A SNP DISCOVERY PIPELINE FOR IDENTIFYING CANDIDATE SNP IN PUBLIC EST TRACE FILES

Noyes HA, Amigo Lechuga J, Broadhead AM, Hughes M, Morton IG, Rennie K, Kemp SJ

University of Liverpool, Liverpool, United Kingdom

Single Nucleotide Polymorphisms (SNP) are assumed to underlie many of the differences between inbred mouse strains. It is important to identify as many SNP as possible to facilitate functional and mapping studies. Large scale SNP discovery projects are being undertaken by the Whitehead Institute and others by resequencing of shotgun genomic clones. The SNP that emerge form these random sequencing programmes are principally in intergenic regions and will be important for mapping and haplotype identification. However it is expected that relatively few of these SNP will be functional. We have screened EST reads from public databases in order to discover SNP with a higher proportion of functional variants. EST reads have been neglected as a source of candidate polymorphisms since there is only a single read associated with each sequence and hence data quality is uncertain. We have developed a SNP discovery pipeline using PolyBayes (Marth et al., Nature Genetics 1999, 452-456) to screen EST trace files for high quality variant base calls. SNP that are assigned high probabilities by PolyBayes are passed to a script which retrieves metadata about the EST libraries in which the SNP was identified. Another script identifies the genomic position of the SNP in Ensembl and retrieves 200bp flanking genomic sequences that can be used as permanent identifiers of the SNP position. The data is loaded into a MySQL database. Heuristic queries have been developed to screen the Polybayes output and the EST metadata for EST predictions that are most likely to be true positives. All publicly available mouse EST trace files will be scanned and the data submitted to public databases. Approximately 100 SNP predictions are currently being validated in vitro to evaluate the accuracy of the pipeline.

POSTER 105 - INTEGRATING MOUSE SNPS INTO MGI: PLANS, PROGRESS, AND PROTOTYPES

Richardson JE, Blake JA, Bult CJ, Eppig JT, Kadin JA, King BL, Ringwald M

The Jackson Laboratory, Bar Harbor, United States

Strain polymorphisms have always been an important resource for biological analysis. Until recently, polymorphisms, e.g. RFLPs and SSLPs, have been used primarily as mapping landmarks. New technology for large-scale detection of single nucleotide, multiple nucleotide, and deletion/insertion polymorphisms (SNPs, MNPs, and DIPs, respectively) offers the potential to correlate phenotypic differences directly to base-changes in the DNA. Large data sets offering dense coverage across the genome are already available, as are public repositories of these data, e.g., dbSNP.

MGI (http://www.informatics.jax.org/ ) adds value to data by integration, both within and across data domains, and by offering the integrated resource for query, display, and download. The database has always included polymorphism data, e.g. RFLPs and SSLPs. We now plan to integrate mouse SNP data and to provide interfaces that exploit this integration. For example, a user will be able to query for SNPs mapping (within a specified bp range) near genes (mapped to the genome assembly sequence) known to be involved in DNA repair (via GO annotations), where those genes occur within any QTL region for radiation induced apoptosis (e.g., Rapop1, et.al.,  based on the MGD genetic map).

We will describe our plans for integrating SNP data into MGI, including planned data and initial data sets, integration strategy, and Web interface mockups, including query forms, result summary pages, SNP data pages, and other related pages, such as gene details. We will also describe an operational prototype tool, PLAD ( http://proto.informatics.jax.org/prototypes/plad/), which demonstrates the kind of positional candidate analysis expressed by the above example query.

MGI is supported by NIH Grants HG00330,  HD33745, and HG002273.

POSTER 106- HIGH THROUGHPUT YEAST TWO-HYBRID STRATEGIES FOR PROTEOME-WIDE PROTEIN-PROTEIN INTERACTION MAPS

Zhong J 1, Finley R 2

1 Current affiliation: Laboratory of Genetics, National Institute on Aging, NIH,, Baltimore, MD, United States,

2 Center for Molecular Medicine and Genetics, Wayne State University, Detroit, MI, United States

Most cellular activities are carried out by stable or transient protein-protein interactions. A proteome-wide protein interaction map could suggest the functions of individual proteins, help map entire pathways or processes, and suggest how individual pathways and processes are integrated into larger cellular events. The yeast two-hybrid system is an efficient and sensitive way to detect binary protein interactions, and can be modified for high throughput assays. However, most high throughput two-hybrid approaches fail to detect many interactions that should be detected by yeast two-hybrid systems and they produce a significant number of false positives. We developed an alternative high throughput two-hybrid approach  in which two yeast arrays, conditionally expressing AD or BD fusions, are mated using a two-phase pooled mating scheme. We show that our approach allows us to detect interactions not detected by other approaches, including interactions involving proteins which can activate the reporters on their own, and proteins that are toxic to yeast. We demonstrated that our approach is efficient and can be applied to complex organisms such as Drosophila. Combined, our results suggest that multiple studies using different approaches and different systems may be necessary to uncover all interactions that can be detected by yeast two hybrid assays.

POSTER 107 - ASSEMBLING THE MOUSE GENOME: WGS AND CLONE BASED METHODS

Church DM 1, Baily JA 3, Eichler EE 2, Agarwala R 1

1 NIH/NCBI, Bethesda, United States, 2 University of Washington, Seattle, United States, 3 Case Western Reserve University Medical School, Cleveland, United States

The mouse genome is in the unique position of being sequenced using two distinct strategies with all of the data for both strategies publicly available. In late 2002, a Whole Genome Assembly (WGA) was published and is referred to as the MGSCv3. Work is ongoing to produce a high quality finished genome using clone based sequence. Currently, over 50% of the mouse genome is covered by high-quality, finished sequence and the rest is covered by draft sequence and WGS contigs. Finished sequence is hand curated and assembled into non-redundant contiguous sequences (contigs).

In an effort to provide access to the most current data we have been performing genome assemblies using all available data (Whole Genome Shotgun Contigs and BAC based sequence). Performing the composite assemblies has provided us with some insight concerning differences between clone based assemblies and WGA. Recently, we have performed several composite assemblies using different parameters on the same set of sequences. In addition to producing a better mouse genome assembly, we have been able to assess errors that are likely to occur when a given data source is allowed to drive the assembly. Assemblies produced where clone based data drives the assembly can have local order and orientation errors when a large amount of draft sequence is included. By contrast, assemblies where the WGS contigs drive the assembly have an increased rate of artificial duplication. This has led us to more systematically characterize the differences between WGA and draft clone based sequence.

In addition, we have been able to assess the presence of segmental duplication within the mouse genome. Regions of segmental duplication cause assembly problems regardless of the assembly strategy adopted. Preliminary analysis of the MGSCv3 and a small amount of finished data suggested that the amount of large (>10 Kb) regions of segmental duplication are under-represented in the WGA. We are extending this analysis now that over half of the genome is in finished form. Preliminary attempts at characterizing the location and contents of these duplications were limited due to our inability to place them in the assembly. Improved mapping and assembly has improved our ability to assess these regions. The results of these assessments will be discussed.

POSTER 108 - MOUSE GENOME RESOURCES AT NCBI

Church DM, Annotation Team NCBI

NIH/NCBI, Bethesda, United States

Genomic resources for the mouse genome have increased greatly. A Whole Genome Shotgun (WGS) assembly generated by the Mouse Genome Sequencing Consortium (MGSC) was released and published. During the past year, sequencing resources for mouse have shifted towards clone-based (HTGS) sequence. As of June 4, 2004, 1.48 Gb of non-redundant finished sequence and 1.8 Gb of redundant draft sequence were available. Greater than 95% of the HTGS sequence is from the reference strain (C57BL/6J). To leverage all available sequence data, NCBI has been performing composite assemblies that integrate HTGS sequence from C57BL/6J into the MGSCv3. NCBI Build 33 (based on data from May 30, 2004) integrated 1.1 Gb of HTGS phase 3 (finished) sequence.. In addition to producing the reference assembly, NCBI has been producing alternate, strain-specific assemblies. In Mouse Build 33, 6 alternate assemblies were produced. In addition, the publicly available portion of the Celera mouse assembly (Mmu16) is annotated and available as part of the standard NCBI resource set:, Map Viewer, the NCBI ftp site, Entrez Gene, etc.

In addition to producing these assemblies, NCBI provides annotation for all assemblies via a suite of software tools available from our website (http://www.ncbi.nlm.nih.gov). Current annotation provides prediction of gene models (based on alignment and ab initio prediction), clone placement (BACs and fosmids based on end sequence alignment), variation, STSs, Gene Trap clones, human and rat transcripts and phenotypes (via STS connections). Annotation information will be provided for the current mouse assemblies. In addition, improvements in the NCBI MapViewer and associated resources with respect to clone identification and comparative mapping will be discussed.

POSTER 109 -  IN SILICO POSITIONAL CANDIDATE GENE ANALYSIS IN THE MOUSE: NEW TOOLS FROM THE MOUSE GENOME INFORMATICS CONSORTIUM

Bult CJ, Ringwald M, Blake JA, Kadin JA, Richardson JE, Eppig JT, Sequences and Maps Group MGI, Phenotypes Group MGI, Expression Group MGI, Software Group MGI

The Jackson Laboratory, Bar Harbor, United States

The mouse genome sequence serves as a framework for the discovery of genes that underlie biological and disease processes. To leverage the genome sequence effectively depends on how well genetically defined mutant phenotypes are integrated with genome annotations and information about sequence variation (e.g. SNPs), allelic variation, gene expression, homology, and gene function annotations. Because of its long-standing focus on the integration of exactly these kinds of diverse biological data about mouse genes, the Mouse Genome Informatics database is a powerful platform to assist in the computational analysis of integrated biological data with the goal of identifying candidate genes associated with complex genetically defined phenotypes. Associating expression phenotypes with mutant phenotypes has proved to be a powerful approach to identifying candidate genes in regions of the mouse genome to which complex traits have been mapped. To facilitate such associations via MGI we have integrated probe sets from both Affymetrix and Agilent with genes in MGI. As a result, researchers can map expression data generated from these platforms to genes in MGI using both genomic and genetic coordinates. They can also combine expression data with other data including SNPs, phenotypic classifications, and functional annotations. We will show how the more than 1,800 Quantitative Trait Loci in MGI can be combined with genetic, expression, and variation data in support of in silico positional candidate gene analysis.

The MGD and GXD components of the Mouse Genome Informatics Consortium are supported by NIH 5 P41 HG000330-P1 and NIH HD33745, respectively.

POSTER 110 -  COMPLEX BIOLOGICAL SYSTEMS AND BIOINFORMATICS: SOME NEW BIOINFORMATICS TOOLS AND APPROACHES.

Snoddy J

GST at Oak Ridge National Lab and University of Tennessee, Oak Ridge, United States

There are novel bioinformatics challenges posed by the study of biological systems. Highly networked systems create phenotypes from genotypes and the environment. We now need efforts to understand how those genes, gene products and cells function in these networks.  This will require large data sets whose analysis will further require new bioinformatics. While databases of experimental information are necessary—indeed critical—they are not sufficient to help obtain insight from the analysis of large, complex networks.  We need a mathematical language to describe the interconnections of these networks, i.e., robust tools that can help us get insights from use of that language, and data mining tools to see patterns among these networks.

A number of both experimental and computational collaborators have been working together to develop these needed improvements.  Our current work involves elucidating aspects of regulatory networks and co-expression networks.  We are integrating data from different DNA microarray data sets from gene product function data (e.g. GO), gene families, genetic variation, evolutionary conservation of coding and non-coding parts of the genome, and other similar information.  We are developing methods to find and visualize tightly connected subcomponents in networks that are of interest to researchers after data integration.

We have developed several sophisticated data mining environments and tools. Some of these will be shown as they are now ready for wider use and evaluation. Some biological results will be presented that come from this analysis.

POSTER 111 -  THE UTILITY OF HIGH-DENSITY GENETIC MAPS IN EXPERIMENTAL MOUSE CROSSES

Leonardson AS, Edwards SW, Cervino ACL, Schadt EE

rosetta/merck, seattle, United States

We have examined the genetics of gene expression in an F2 mouse intercross using both a microsatellite marker map and a high-density SNP marker map. We report on the differences between these two maps and highlight the advantages of using a high density map in the context of gene expression profiling.

POSTER 112 -  MOUSE PHENOME PROJECT

Bogue MA, Grubb SC

The Jackson Laboratory, Bar Harbor, Maine, United States

The Mouse Phenome Project is an ongoing international collaborative effort to enhance the research potential of the laboratory mouse by a) promoting the phenotypic and genotypic characterization of a set of 40 inbred strains and their derivatives, and b) making the data publicly available through a web-accessible database, the Mouse Phenome Database (MPD; www.jax.org/phenome) [Grubb et al., 2004].  We have collected and annotated over 600 measurements that include a broad range of metabolic, developmental, and behavioral parameters relevant to human disease.  Also, large volumes of SNP data are being incorporated into the MPD to maximize community SNP resources.  More than 450,000 SNPs have been identified by large-scale genotyping consortia in a subset of the Project's 40 priority strains and contributed to the MPD.  Strain characteristics and genotypic data have proven to be a valuable resource for the biomedical research community, as indicated by the increasing use of the MPD and significant advances using the data and associated analytical tools to uncover information on the genetic and biological factors involved in normal and disease pathways.  Expansion of the Mouse Phenome Project will allow the biomedical research community to continue to exploit quantitative phenotypic data together with emerging sequence, SNP, and haplotype data.  Project status and analysis tools will be highlighted.

Grubb SC, Churchill GA, Bogue MA. A collaborative database of inbred mouse strain characteristics. Bioinformatics. 2004 May 6 [Epub ahead of print] PMID: 15130929.

POSTER 113 -  GENETIC ANALYSIS OF GENE EXPRESSION IN MOUSE CNS REVEALS MAJOR PLEIOTROPIC AND POLYGENIC CONTROL OF SYNAPTIC MACHINERY.

Chesler EJ 1, Baldwin NE 2, Zhang B 3, Kirov S 3, Wang J 1, Lu L 1, Snoddy JR 3, Langston MA 2, Manly KF 1, Williams RW 1

1 Center for Genomics and Bioinformatics, University of Tennessee Health Science Center, Memphis, TN, United States, 2 Department of Computer Science, University of Tennessee, Knoxville, TN, United States,

3 Oak Ridge National Laboratory, Oak Ridge, TN, United States

In our ongoing microarray-based analysis of genetic regulation of gene transcription in recombinant inbred strains generated by crossing

C57BL/6J and DBA/2J (BXD RI), we have identified a small number of chromosomal loci that modulate mRNA abundance of several hundred transcripts each. These loci were detected using a novel application of clique analysis for dimension reduction of microarray data. Many of these results can be explored using new tools and external links in WebQTL (www.webqtl.org/search.html), which allows users to rapidly examine multiple traits and interpret their relations and patterns of covariance. Using a combination of genetic correlation analysis, clique analysis, Quantitative trait locus (QTL) analysis, and Gene Ontology category representation analysis, we have identified loci that modulate expression of components of the synaptic vesicle cycling system.  Using Batch Sequence Analysis, we have identified putative novel conserved regulatory sequences for one highly co-regulated group of transcripts. Specific candidate genes that reside at some of these regulatory loci have been identified. A single locus has been identified that regulates the transcription of over 1500 transcripts. The synaptic vesicle structure and function is dependent on a large number of cytoskeletal, anchoring, motor and fusion proteins. Vesicle mediated transport, exocytosis and endocytosis are major means of effecting and controlling cell-cell communication in the CNS. Other related cellular processes that rely on the same gene networks are similarly varied. Genetic variation also affects mRNA translocation, and possibly synaptic gene expression mechanisms. The relationship of these cliques of genetically correlated transcript abundances to systems level phenotypes such as ethanol withdrawal severity demonstrates the utility of this mouse reference population for systems biological research.

POSTER 114 -  A MUTATION IN THE SERUM AND GLUCOCORTICOID-INDUCIBLE KINASE-LIKE KINASE (SGKL) GENE IS RESPONSIBLE FOR DEFECTIVE HAIR GROWTH IN MICE

Masujin K 1, Okada T 2, Izawa N 1, Tuji T 1, Ishii Y 2, Takano K 3, Matsuda J 3, Ogura A 4, Kunieda T 1

1 Graduate School of Natural Science and Technology, Okayama University, Okayama, Japan, 2 Graduate School of Agricultural and Life Sciences, The University of Tokyo, Tokyo, Japan, 3 National Institute of Infectious diseases, Tokyo, Japan, 4 RIKEN Bioresource Center, Ibaraki, Japan

YPC is a mutant mouse strain with defective hair growth characterized by thin, short hairs and poorly developed hair bulbs and dermal papillae. To identify the gene responsible for the phenotype, we performed genome-wide linkage analysis using 1,010 backcross progeny and 123 microsatellite markers covering all chromosomes. The mutant locus (ypc) was mapped to a 0.2-cM region in the proximal part of mouse chromosome 1. This 0.2-cM region corresponds to a 450-kb region of genome sequence that contains two genes with known functions and five ESTs or predicted genes with unknown functions. Sequence analysis revealed a single C to A nucleotide substitution at nucleotide 1,382 in the Sgkl gene, causing a nonsense mutation at codon 461. Sgkl encodes serum and glucocorticoid-inducible kinase-like kinase (SGKL), which belongs to a subfamily of serine/threonine protein kinases and has been suggested to have a role downstream of lipid signals produced by activation of phosphoinositide 3-kinase (PI3K). In the mutant SGKL, a serine residue in the C-terminal end of the protein (Ser486), which is indispensable for activation of SGKL upon phosphorylation, is abolished by premature termination. Specific expression of the Sgkl gene in the inner root sheath of growing hair follicles was also identified by in situ hybridization. Therefore, we concluded that the mutation in the Sgkl gene is responsible for defective hair growth in the ypc mutant mouse and that the signaling pathway involving SGKL plays an essential role in mammalian hair development.  We also found a single nucleotide insertion in the Sgkl gene of fz mouse that is another mutant mouse showing hair abnormalities resembling to that of YPC mouse.  The nucleotide insertion causes a frame shift mutation in the middle part of the protein.

POSTER 115-  MOUSE CHROMOSOME 11 ANALYSIS AND ANNOTATION: THE HIGHLIGHTS

Ashurst J, Frankish A, Gibson R, Grocock R, Hart L, Laird G, Loveland J, Mudge J, Sehra H, Steward C, Swarbreck D, Wilming L

Wellcome Trust Sanger Institute, Hinxton, United Kingdom

Mouse chromosome 11, finished and sequenced from clone based contigs (as opposed to whole genome shotgun), has been analysed and manually annotated. We will present the highlights in the form of interesting features, gene clusters and gene arangements.

POSTER 116 - GENETIC INFIDELITY AND INBRED MOUSE STRAINS

Wiles MV, Chu T, Palmer JH

The Jackson Laboratory, Bar Harbor, United States

Abstract: The use of mice and, in particular, genetically engineered mice to study normal and pathological biology pathways has increased exponentially in the last decade. One impetus for this has been the completion of sequence drafts for human, mouse and recently the rat genomes. The complexity, depth and associated high cost of information which can now be acquired from exploring mouse model systems mean that strict, stable genetic standards are paramount.

The genomes of all living organisms, including inbred mice, are however, continually evolving. Any genetic change in a strain can potentially lead to phenotypic change. For example, it was reported that a C57BL/6 substrain from Olac-Harlan-Sprague Dawley, UK was supplied for a few years with an undetected spontaneous deletion of more than 97kb, covering in part the alpha-synuclein locus, a gene implicated in a number of neurodegenerative diseases. These naturally occurring, accumulated genetic variations (genetic drift) can lead to unexpected results and confound deep data acquisition, making reliable comparative analysis on large data sets derived over time difficult. Furthermore, genetic contamination of inbred strains can occur even in the best mouse house leading to unexpected and confounding outcomes. It is therefore critical for any research organization to ensure that mice used are of the correct genotype and have not become a non standard subline.

We present how The Jackson Laboratory controls for genetic contamination and is reducing genetic drift to almost negligible levels by continual Genetic Quality Control monitoring and the creative use of in vitro fertilization and cryopreservation. The upholding of stable genetic standards for inbred lines enables the international community to develop valid massive comparative data sets which can be used for the foreseeable future. Support Contributed By: HMMI, The Ellison Medical Foundation and The Jackson Laboratory

POSTER 117 - FUNCTION OF RECQ DNA HELICASES BLM AND RECQL5 IN THE SUPPRESSION OF MITOTIC CROSSOVERS AND MAINTENANCE OF GENOMIC STABILITY

Hu Y, Barnes E, Luo G

Case Western Reserve University, Cleveland, United States

Homologous recombination (HR) is important for both mitotic and meiotic cells. HR can lead to either crossover or gene conversion without crossover upon completion. In somatic cells, crossovers can cause genome rearrangements,and must be suppressed by specific mechanisms. Previous studies have shown that BLM, a RecQ DNA helicase, has important roles in such suppression. Mutations in the BLM gene in humans cause Bloom syndrome (BS), a disorder with unusually high tumor susceptibility. The elevated frequency of mitotic crossovers in BS cells leads to a significantly increased frequency of sister chromatid exchange (SCE), as well as triradial and quadriradial structures. Interestingly, while BLM is the only known gene to suppress mitotic crossovers in mammals, recent studies in yeast suggest additional mechanisms exist. Therefore, it is important to investigate whether other pathways are also present in mammals. In addition to our previously reported mouse model for BS, we generated knockout mice for another RecQ family member, Recql5. Analysis in Recql5-/- cells revealed that the frequency of spontaneous SCE increased to a comparable level of that observed in Blm-/- cells. Triradial and quadriradial structures were also seen in these cells. Frequency of SCE was further elevated in the double knockout cells of Blm and Recql5. These observations provide the first genetic evidence that Recql5 suppresses mitotic crossovers in a parallel but non-redundant manner with Blm. Studies with other cell types and knockout mice are in progress to investigate the function of Recql5 and its genetic interaction with Blm in HR regulation and maintenance of genomic stability.

POSTER 118 - APOE GENOTYPE AND AMYLOID-BETA METABOLISM IN A MOUSE MODEL OF ALZHEIMER'S DISEASE

Mann KM, Lamb BT

Department of Genetics, Case Western Reserve University and University Hospitals of Cleveland, Cleveland, United States

The APOE e 4 allele is the most significant genetic risk factor associated with Alzheimer’s disease to date.  Epidemiological studies have demonstrated that inheritance of one or more e 4 alleles affects both the age of onset and severity of pathology development.  Dosage of APOE e 2 and e 3 alleles, however, appear to be protective against the effects of e 4.  Although much of the biology of APOE in peripheral cholesterol metabolism is understood, its role in brain cholesterol metabolism and its impact on AD development is less defined.  Several studies have shown that cholesterol may be an independent risk factor for AD and can directly affect the production of Ab, the constituent of senile plaques.  We have characterized APOE knock-in (KI) mice, which express each human allele under endogenous regulatory elements, on a defined C57BL6/J background.  These mice have significantly different serum cholesterol levels and steady-state brain APOE levels, yet have equivalent brain cholesterol levels. However, the presence of human APOE significantly increases brain Ab levels in a genomic-based model of AD, irrespective of allele.  These data indicate an independent role for APOE in cholesterol metabolism in the periphery relative to the CNS. Additionally, altered levels of cholesterol and APOE are insufficient to influence Ab metabolism in a mouse model of Alzheimer’s disease. Future experiments will address the impact of aging and diet with regards to APOE genotype and amyloid deposition.

POSTER 119 - THE HETEROZYGOUS DEAFWADDLER MOUSE IS A MODEL FOR AGE-RELATED HEARING LOSS

McCullough BJ, Tempel BL

University of Wasington, Seattle, United States

Hearing loss is a neurodegenerative disease that affects over 20 million people in this country alone. Recent experiments in our lab indicate that the heterozygous deafwaddler mouse is a model for hearing loss. This phenotype is the result of loss or dysfunction in the plasma membrane calcium ATPase isoform 2 (PMCA2), which is critical for calcium extrusion in the stereocilia of hair cells in the organ of Corti. Mice homozygous for a mutation in the gene encoding this protein are profoundly deaf and ataxic; heterozygotes display partial hearing loss in a haplo-insufficient manner. In young animals, this hearing loss is as great as ~50 dB at high frequencies, while leaving low frequencies largely unaffected. To better characterize the hearing loss seen in these animals, we have genetically manipulated the amount of PMCA2 activity available and examined the progression of hearing loss over time. Preliminary results demonstrate that hearing loss progresses to include lower frequencies with age. Furthermore, the rate of this progression depends on the amount of PMCA2 activity present. This pattern of hearing loss resembles the age-related hearing loss (AHL), or presbycusis, seen in humans and suggests a mechanism involving the cochlear amplifier.

POSTER 120 - IL-15Ra IS A NEGATIVE REGULATOR OF THE ACTIVATION OF CD4+ T CELLS AND AUTOIMMUNITY

Lee JM, Hou MS, Chung CY, Chiang WW, Yang ST, Liou YH, Liao NS

Institute of Molecular Biology, Academia Sinica, Taipei, Taiwan

Although IL-15 is known to be a T cell growth factor, the function in T cells of IL‑15Ra, its high affinity receptor, remains unclear. We found that IL‑15Ra -/- CD4+ T cells hyperproliferated in response to TCR stimulation, in vitro and in vivo, and displayed a lower TCR activation threshold than wild type (wt) CD4+ T cells. TCR-induced activation of Zap70 and of the PLC-g1-Ca ++-NFATp, Ras-ERK-c-Fos, and Rac-JNK-c-Jun pathways were all augmented in IL-15Ra -/- CD4+ T cells.  This in turn led to earlier IL-2Ra induction, higher IL-2 production and hyperproliferation. Exogenous IL-15 reduced levels of TCR-induced transcription factors, IL-2 and IL-2Ra and division in wt CD4+ T cells. In addition, the majority of aged female IL-15Ra -/- mice developed lupus-like symptoms. These results reveal IL-15Ra to be a negative regulator for CD4+ T cell activation and autoimmunity, and demonstrate a novel layer of regulation of TCR signaling by a cytokine system.

POSTER 121 - SCREENING FOR LONGEVITY GENES IN MICE

Johnson DK 1, Miller DR 1, Moustaid-Moussa N 2, Goldowitz D 3

1 Life Sciences Division, Oak Ridge National Laboratory, Oak Ridge, United States, 2 Dept. of Nutrition, University of Tennessee, Knoxville, United States, 3 Dept. of Anatomy and Neurobiology, University of Tennessee Health Sciences Center, Memphis, United States

With the goal of discovering genes that extend the healthy life span in mammals, we are phenotyping mice carrying new mutations induced by N-ethyl-N-nitrosourea (ENU).  The phenotyping plan includes evaluation for markers of growth trajectory and stress response in pedigrees aged to 28 months and beyond.

For the aging core, four males and four females of the “test class” are group-housed at

weaning to await screening at 18 months for behavioral and neuroanatomical abnormalities and at 19-24 months for factors thought to be predictive of longevity. Mice of the genotype that is homozygous for a mutagenized G0 grandparental target chromosome constitute the “test class” that should manifest any obvious or subtle abnormal recessive phenotypes caused by a new ENU-induced mutation.  This test class can be identified by visual or molecular markers.

To date, we have aged 202 pedigrees beyond 18 months and screened for body weight, blood glucose, leptin, insulin, insulin-like growth factor-1, corticosterone, and ability to maintain body temperature in the cold. We have identified phenodeviants for early low and high body weight, high corticosterone, high and low blood glucose, high core temperature, high IGF-1, high IGF-1 + high insulin, low core temperature + low IGF-1, and low leptin.  We are also developing software-analysis tools for determining bone density and body-fat content from microCT images. Phenodeviant pedigrees are being recovered for confirmation of heritability of the ascertained phenotype by breeding old males or by performing intracytoplasmic sperm injection using sperm routinely frozen from young males.

This work is supported by the National Institute on Aging in grant U01 MH61971 to the Tennessee Mouse Genome Consortium.

POSTER 122 - NATIONAL CANCER INSTITUTE – MOUSE MODELS OF HUMAN CANCERS CONSORTIUM (MMHCC)

Marks CL

National Cancer Institute, NIH, Bethesda, MD, United States

In 1999, the National Cancer Institute (NCI/NIH) confronted the critical need for vastly improved model systems to inform basic, clinical, epidemiologic, and translational cancer investigations.  The ability to manipulate the germline of laboratory mice, coupled with an unprecedented store of data about genetic alterations implicated in human cancer and the rapid acquisition of human and mouse genomic sequences prompted the NCI to implement a collaborative project of mouse cancer modeling. 

The resulting research program, the Mouse Models of Human Cancers Consortium (MMHCC), has expertise in many aspects of basic, translational, clinical, and epidemiological research and mouse genetics.  The MMHCC is continually challenged by the NCI to identify the most pressing questions in cancer biology and translational science and to direct cancer modeling skills to those questions.  The initial program of 19 research groups was recently expanded to 25; the increased size requires the NCI to revisit how to manage the MMHCC to preserve the free exchange of ideas and spirit of cooperation that are hallmarks, and great strengths, of the program.

The 250-member MMHCC cooperates with the NCI to evolve an integrative systems approach to human cancer research.  The NCI Center for Bioinformatics integrates descriptive cancer model information with comparable human disease data.  The Center maintains the Cancer Models Database and the Cancer Images Database to house descriptive data about all types of in vitro and in vivo cancer models.  Any researcher may submit data to these databases to ensure that the information store reflects the experience of the community of cancer researchers who explore how well the models inform human cancer therapy, prevention, early detection, imaging, and population science.  The MMHCC eMICE website ( http://emice.nci.nih.gov) is the interface to all the NCI’s preclinical models programs, resources, and databases. 

The MMHCC members collaborate with the NCI to convene community-based meetings to promote state-of-the-art mouse cancer science and advise the NCI on the resources and infrastructure to sustain development and application of mouse models by anyone in the cancer research community.  One key resource is the Mouse Repository (http://cancermodels.nci.nih.gov), which the NCI established in 2000 to deploy fully developed mouse cancer models and the strains that are used to derive models free-of-charge to the worldwide scientific community.

POSTER 123 - DIET, GENETICS, AND BEHAVIOR

Collins K, Kusek G, Stellone M, Beyer B, Bolivar V, Flaherty L

Genomics Institute, Wadsworth Center, Albany NY, United States

Genetic and environmental interactions are common when studying complex traits.  In the literature, there have been several suggestions that diet, especially a high fat diet, may play a role in mouse behaviour.  Therefore, the effects of a high fat diet on behaviour was investigated in two inbred strains of mice, C57BL/6J (B6) and 129S1/SvImJ (129S1).  Fifty mice of each strain were placed on an atherogenic diet (Paigen’s Diet) for 14 weeks and compared in parallel to 50 mice on a standard laboratory chow diet.  As expected, this diet significantly affected average plasma cholesterol values (136 vs 242 mg/dl in 129S1 and 105 vs 276 mg/dl in B6).  The diet also affected several other blood parameters, including levels of alanine aminotransferase, alkaline Phosphatase, glucose, amylase, and calcium.  Sometimes these changes were observed in only in one of the two strains.  For example, calcium levels were higher in the 129S1 strain on a high fat diet, yet the same was not true for B6.  Two measurements of behaviour were also affected by diet, but only in B6 mice.  On the high fat diet, B6 mice showed lower open field activity and a deficit in habituation compared to mice on standard chow.  Other parameters, such as behaviour on an elevated zero maze, were unaffected.  These results imply that high fat diets may influence mouse behaviour, as well as physiology, but only on certain genetic backgrounds.  Further experiments designed to evaluate the effects of these diets on brain gene expression and brain metabolism are in progress. 

POSTER 124 - DYNAMIC INSERTIONAL MUTAGENESIS PROJECT FOR THE MOUSE

Poirier C, Overbeek PA, Adams CP, Harrison WR, Xiao N, Castile CA, Bishop CE

Baylor College of Medicine, Houston, United States

We developed a 2-step insertional mutagenesis program for the mouse using the FVB/N albino inbred strain. All transgenes carried a visible semidominant coat color marker, tyrosinase, flanked by the inverted terminal repeats, specific for the Sleeping Beauty transposase. In the first step, we generated regular transgenic mice with our tyrosinase transgenes and transmission of the transgene was followed with the visible coat color marker. In the second step, the transgene was transposed from its original insertion site through the “cut and paste” system mediated by the Sleeping Beauty transposase. For transpositional purposes, we generated different transgenic lines in the FVB/N background which expressed the transposase from 4 different promotors. Transposition in the paternal germline was identified by new variations of pigmentation in the progeny. In both steps, heterozygous and homozygous mice were easily identified by their coat color and were screened for inherited abnormal phenotypes. From this phenotype driven mutagenesis program, the causative genes were subsequently cloned using the transgene/transposon as a probe. Several tyrosinase transgene constructs have been generated. The lastest versions contain a LoxP site for further engineering, inversion and deletion within the mouse genome and 2 splice acceptors in opposite orientation for in vivo gene trapping. We will discuss and compare the transposition and mutagenesis efficiency of all the transgenes.

POSTER 125 - MOUSE SPLICE MUTANT GENERATION FROM ENU-TREATED EMBRYONIC STEM CELLS

Greber B, Lehrach H, Himmelbauer H

Max Planck Institute for Molecular Genetics, Berlin, Germany

Despite the availability of mammalian genome sequences, the functions of many genes remain unknown. Mutant mice are valuable for studying mammalian gene functions and for modeling human disease phenotypes. With regard to the latter, a range of diseases involves the missplicing of pre-mRNAs. 

Embryonic stem cell technology offers the possibility to manipulate the mouse genome and select for mutations in vitro. Mutant animals can then be generated from selected clones. We are inducing mutations chemically using ENU, followed by the generation of a mutant clone library in a 96-well format. The library we have generated consists of about 40,000 clones distributed into 96 times 96 samples. Such clone libraries can be specifically screened for mutations in genes of interest. We have developed a protocol that allows the targeted detection of splice mutations based on highly pooled cDNA samples. The procedure does not require any high-cost instrumentation and is time-efficient enabling one worker to screen a given gene within about one week. As a proof of principle, we have isolated splice mutant clones for the mouse Kit gene and obtained germline transmission for a mutation causing the deletion of exon 18 at the transcript level. The heterozygous mutant displayed a dominant phenotype which resembled that of an established targeted null allele.

Acknowledgement of support: Our work is carried out through funding by the German National Genome Research Network (NGFN).

POSTER 126 - EUMORPHIA: PHENOTYPING BEHAVIOUR AND COGNITION IN MICE

Tucci V, Auwerx J, Brown S, Chambon P, Coghill E, Gale K, Golini E, Hölter SM, Jacquot S, Krezel W, Lad H, Mandillo S, Marazziti D, Meziane H, Nolan PM, Ouagazzal A, Parker A, Pedersen V, Rosenthal N, Tocchini-Valentini G, Wurst W

1 MRC, Harwell, United Kingdom, 2 ICS, Strasbourg, France, 3 EMBL, Monterotondo, Italy, 4 CNR, Monterotondo, Italy, 5 GSF, Munich, Germany, 6 EMBL, Monterotondo, Italy

EUMORPHIA is a European Consortium focused on the development of new approaches in phenotyping, mutagenesis and informatics leading to improved characterisation of mouse models for the understanding of human physiology and disease.

European mouse geneticists with different and varied expertise are working together in order to provide the scientific community with a platform for the systematic and standardised phenotyping of mouse models. Eumorphia members from different centres are grouped into workpackages (WP), each dealing with a set of specific body systems. Workpackage 10, responsible for behaviour and cognition phenotyping, identified a high-throughput battery of 9 primary tests to be included in the first-line screen. These tests are: open field, modified SHIRPA, grip strength assessment, rotarod assessment, y-maze spontaneous alternation, pre-pulse inhibition of acoustic startle response, tail flick test, tail suspension test, and swim ability assessment. Here we will present and compare data obtained on four inbred strains (C57Bl/6J, C3HeB/FeJ, BALB/cByJ, 129S2/SvPas) and across all participating centres. From our preliminary analysis we can define the robustness and limits for each test.

To deeply characterize those aspects of behaviour that require more complex procedures we will discuss the ongoing work on the development of secondary screening protocols. A description of some important features of the specific analysis of cognition phenotyping and the use of new algorithms will be discussed.

POSTER 127 - REGIONAL MUTAGENESIS OF THE MOUSE GENOME AND THE DETECTION OF NEUROBEHAVIORAL MUTANTS

Goldowitz D 1, Miller DR 4, Swanson D 1, Chesler E 1, Zuo J 3, Ferkin M 2, Mittleman G 2, Hamre K 1, Matthews D 2, Cook M 2, Jablonski M 1, Smeyne R 1, Elberger A 1, Johnson DK 4

1 UT Health Science Center, Memphis, TN, United States, 2 University of Memphis, Memphis, TN, United States, 3 St. Jude Children's Research Hospital, Memphis, TN, United States, 4 Oak Ridge National Laboratory, Oak Ridge, TN, United States

The Tennessee Mouse Genome Consortium (TMGC) consists of researchers from eight research institutions. Members from five of these institutions are part of one of the three NIH-funded groups to perform ENU-mutagenesis and screen for neurological phenotypes. The mutagenesis protocol is designed to detect recessive mutants whose mutated gene is localized to a specific region of the genome. This “regional” mutagenesis, using visible or molecular markers, permits the a priori identification of test class mice that can be tested as isogenetic cohorts. This helps greatly with the aging of pedigrees to 18 and 28 months, identification of pedigrees that have developmental lethal mutations, and providing more power for statistically-sensitive traits.  To date we have identified over 80 mutant lines of which 26 of these have neurological phenotypes that are identified in the screening protocols and 9 are visible balance or ataxic mutants. The 26 neurological mutants have been identified in screens that examine general behavior, histology, social behavior, drug- and alcohol-related behavior, hearing, or eye. Most recently we have added a screen for adult and pre-weanling seizure activity. The TMGC is engaged in the phenotypic analysis of mutant mice provided by members of the research community and has been distributing mutant mice through our website at tnmouse.org. Currently, mice can be ordered through the common website of the three mutagenesis centers at neuromice.org.

POSTER 128 - CHARACTERIZATION OF ACCUMULATED DATA BY BLOOD TEST SCREENING IN RIKEN MOUSE MUTAGENESIS PROJECT

Motegi H 1, Ohtaki M 2, Toki H 1, Inoue M 1, Masuya H 1, Kaneda H 1, Kobayashi K 1, Suzuki T 1, Wada Y 1, Wakana S 1, Minowa O 1, Gondo Y 3, Shiroishi T 1, Noda T 1

1 Mouse Functional Genomics Research Group, RIKEN GSC, Tsukuba, Japan, 2 Radiation Biology and Medicine, Hiroshima University, Hiroshima, Japan, 3 Population and Quantitative Genomics Team, RIKEN, GSC, Yokohama, Japan

To identify mouse models of human diseases, we are running a blood test screening in a large-scale ENU mutagenesis program in RIKEN GSC since 2000. The blood test screening system consists of hematological test and clinical biochemical test. For each test, blood samples are collected at early onset stage (9 weeks for hematology and 11 weeks for clinical biochemistry) and late onset stage (52 weeks for hematology and 54 weeks for clinical biochemistry).

For dominant screening, we have screened so far about 10000 mice for early onset stage and about 2000 for late onset stage in each test. In early hematological test, about 70 of phenodeviants were tested for heritability and half of phenodeviants were inherited. In the case of early clinical biochemical test, about 200 phenodeviants were preceded to the inheritance test and one forth of them showed heritability. In late onset screening, about 100 phenodeviants including hematological and clinical biochemical test were identified. Inheritance tests for late onset screening are in progress.

Furthermore, recessive screening has started and about 300 mice of 20 pedigrees were screened for early onset screening of each test.

In this presentation, progress of blood screening in RIKEN mouse mutagenesis project will be reported and the feature of early onset, late onset and dominant, recessive screenings from the analysis of large amount of accumulated data will be discussed.

POSTER 129 - GENERATION OF GENETIC TOOLS FOR FUNCTIONAL ANALYSIS OF THE MOUSE CHROMOSOME 15

Chick WSH 1, Mentzer SE 2, Carpenter DA 2, Rinchik EM 1, You Y 2

1 The University of Tennessee, Knoxville, United States, 2 Oak Ridge National Laboratory, Oak Ridge, United States

Chromosomal inversions and deletions are valuable genetic tools for functional analysis of the genome.  We have used homologous recombination to modify the recessive-lethal In(15)21Rk inversion to endow it with a dominant-visible phenotype.  Several ES-cell lines were derived from inversion heterozygotes, and a keratin-14 (K14) promoter-driven agouti minigene was introduced into the inverted Chr 15 in the ES cells by gene targeting.  Mice derived from the targeted ES cells carry the inverted Chr 15 and at the same time, exhibit lighter coat color on their ears and tails, making this modified In(15)21Rk useful as a balancer for proximal mouse Chr 15.  In order to map the ENU-induced mutations generated by the Neuromutagenesis project of the Tennessee Mouse Genome Consortium, to date we have generated three chromosomal deletion complexes by X-radiation on the distal half of mouse chromosome 15, centered around the Oc90, Sox10 and Cpt1b genes, respectively.  Large deletions spanning up to 5 Mb in ES cells were recovered from around the Sox10 locus whereas mostly small deletions (less than 1 Mb) were recovered from Oc90 and Cpt1b loci. Selected deletions captured in ES cells were injected into blastocysts to produce deletion-bearing mice.  At least 2 deletions at Sox10 transmitted to the germline, and the heterozgyous deletion mice exhibit white spotting in the belly characteristic of Dom mice. Two deletions at Oc90 were also transmitted to germline and the heterozygous deletion mice are outwardly normal. These chromosomal deletion complexes should be useful for mapping mutations and analyzing functional units along the chromosomal regions.

POSTER 130 - GENOME-WIDE MUTATION DISCOVERY BY TGCE IN GENE-DRIVEN APPROACH OF ENU-MUTAGENESIS

Sakuraba Y, Sezutsu H, Takahasi R, Uchiyama M, Tsuchihashi K, Fujimoto N, Ichikawa R, Kaneko S, Goda N, Ikeda A, Karashima Y, Inoue M, Kaneda H, Minowa O, Wakana S, Noda T, Shiroishi T, Gondo Y

RIKEN GSC, Yokohama, Japan

It has become practical to generate and recover mutant mice having ENU-induced point mutations in a specific target gene by reverse genetics. We have prepared frozen sperms and genomic DNA from over 7,000 G1 male mice for the gene-driven approach of ENU mutagenesis. How to find the mutations is one of the most important issues in the gene-driven approach. We have adopted a Temperature Gradient Capillary Electrophoresis (TGCE) method. Multiplexing of the TGCE enabled the high-throughput search of point mutations. We have found over 100 mutations from 130 Mb analysis using the TGCE method. Forty-four mutations will alter the amino acid sequences; 36 missense, 2 nonsense, 1 “make-sense” (Stop to Tryptophan) and 5 splicing signal mutations. Eight mutations were found in candidate regulatory regions. The differences of mutation spectra between TGCE and direct-sequencing methods as well as between gene-driven and phenotype-driven screens will be discussed.

POSTER 131 - MOSTRACK –INFORMATICS FOR LARGE-SCALE MUTAGENESIS PROJECT

Yang P, Kuo Y, Ho M, Cheng C, Lin M, Lin W, Kung J

Mouse Mutagenesis Program Core Facility, Taipei, Taiwan

In the high-throughput breeding environment, many hundreds of offspring are generated quickly. In order to maintain high-level productivity and efficient operation, establishment of highly specialized tools is highly critical. Here, we describe our efforts at generating informatics for our Mouse Mutagenesis Program Core Facility based on technology-derived tools that are task-oriented, effective, scalable, and configurable, so as to serve the needs of all members of this facility. The goal of our proposed system, known as MosTrack, is to provide the necessary tools to manage the daily affairs within the Mouse Mutagenesis Program Core Facility, support efforts in maintaining timely data transactions and keeping informed of issues relevant to the mice. Overall success of the design and development of the MosTrack system will be attributed to the following key features:

Flexibility: Generic design makes MosTrack more flexible. Different users have different and often unique needs. MosTrack is easily customized to meet the needs of different user groups.

Extensibility: Modular design is another feature for MosTrack. New features can be easily added or modified.

Visualized Design: Visualization helps research assistants and caretakers to efficiently track the history and current status of each cage.

User-Friendly GUI Design: Consistent user interface design makes this system more user-friendly. Users can easily operate this system without extensive training.

Tailored to the workflow of mouse mutagenesis problem: The development of MosTrack is tailored to the workflow of mouse mutagenesis research. It is a user-friendly system with enhanced efficiency, flexibility and expandability.

POSTER 132 - THE GNF ENU MUTAGENESIS SCREEN:  USING ENU MUTAGENESIS TO IDENTIFY GENES INVOLVED IN NEUROBIOLOGY

Tarantino LM 1, Schwander M 2, Dong H 1, Hice R 1, Wiltshire T 1, Mueller U 2

1 Genomics Institute of the Novartis Research Foundation, San Diego, United States, 2 The Scripps Research Institute, La Jolla, United States

The ENU mutagenesis screen at the Genomics Institute of the Novartis Research Foundation (GNF) was originally set up to focus on neurobiology phenotypes.  Over the last three and a half years, the screen has developed into a major effort including screens in immunology, metabolic disorders and cancer.  However, neurobiology remains a major component of the screen.  The neuroscreen was developed to focus on complex behaviors including activity, anxiety, learning and memory and prepulse inhibition.  While these areas are still being actively pursued, some of the focus has shifted to other areas such as vision and hearing.

To date, we have screened over 700 pedigrees and have identified numerous heritable mutants that are in various stages of heritability testing, mapping and cloning.  By far, one of the most successful screens has been our acoustic startle screen that has led to the identification of ten heritable mutants.  Eight of these were subsequently found to be deaf using the auditory brainstem response (ABR) assay.  Two of these mutants have been mapped and one has been cloned.

The mutant, distracted, was mapped to Chromosome 10 and a point mutation has been identified in the Cdh23 gene.  The characterization of this mutant will be discussed as well as progress on mapping and cloning additional mutants.  The use of acoustic startle as a high-throughput screen for deafness will also be addressed. 

POSTER 133 - CROSS-SPECIES STUDY OF THE ODZ GENE FAMILY

Zheng L, Nakamura H, Lossie A, Jafar-Nejad H, Schulze K, Bellen H, Justice M

Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, United States

ENU mutagenesis generated six mutant alleles at the mouse Odz4 locus. Five of the six alleles cause early embryonic lethality, while the sixth one demonstrates defects in skeletal and hematopoietic development. The Odz4 gene, also called ten-m4, encodes a 270kD transmembrane protein and is conserved among human, mouse and Drosophila. The Drosophila homologue was previously identified to be a pair-rule gene and was thought to be involved in a novel signal transduction pathway.

To elucidate the pathway, we set out to identify the interacting partners by both biochemical and genetic methods with the Drosophila gene (odz) as a starting point. 1) By performing yeast two hybrid screening, we have identified filamin that may physically interact with the odz protein. Filamin is an actin binding protein which is critical for ring canal formation during Drosophila oogenesis. The interacting part of odz, defined here as FID domain, has been determined by deletion mapping. The expression patterns of filamin and odz during embryonic development are partially overlapped. The mouse Odz4 and FilaminA also have overlapping expression at several embryonic stages. The biological significance of this possible interaction in embryonic and later developmental stages is currently under investigation in both Drosophila and mouse. 2) By characterizing several new odz alleles, we found that the odz gene was not a pair-rule gene. Instead, it causes defects in motor neuron axon routing. The segmental nerve b (SNb) does not innervate or innervate into unrelated muscles. We are currently generating odz cDNA transgenic flies to rescue the defects. This study on the biochemical and genetic interactions may partially reveal the important biological roles that odz plays during development.

POSTER 134 - BIOINFORMATICS PIPELINE FOR ENU MOUSE MUTAGENESIS

Liu B, Zhong J, Tran N, Fan M, Justice M

Baylor College of Medicine, Houston, United States

The Mouse Mutagenesis for Developmental Defects project utilizes ENU as a powerful mutagen in mouse spermatogonial stem cells. By determining the function of genes on a mouse chromosome, we can extrapolate to predict function on a human chromosome. Due to the high throughput generation of ENU mutants in this project, it is essential to have an efficient data organizing system to carry out real-time data capture and analysis. We have developed a bioinformatics pipeline that helps management of mouse colonies, analysis of mutant phenotype screens and web presentation of information of ENU mutants. The pipeline includes the following web services components: a) A virtual mouse barcode system, which is a comprehensive web-based mouse colony tracking system. This system simulates the real mouse facility design and can be employed for any mouse work. This system provides advanced securities to track colonies for different institutions, PIs, labs, projects, users as well as offer controlled data exchanging among users; b) A sample submitting and analysis system for phenotypic screenings (Complete Blood Counts, Hormone Assays and Tandem Mass Spectrometry Analysis): which is a web based interface to arrange mutant screening and flag potential mutants based on the standard deviation analysis; c) A cryopreservation database that is a web-based management system for recording organs and DNAs from the ENU mutants; d) ENU mutant resource: a publicly accessible database that presents over 283 ENU mutants in ten different phenotype categories, in which the information for each mutant is regularly updated. Within the ENU mutant resource, an image management process, called ImageView, was developed, which enables internal users to edit the information for each image and external users to view real-time updated information of images; e) A mutant alert, which regularly alerts web subscribers for the mutants they are interested in, and f) An online mouse request, so researchers in the mouse community may request the mutant through this system. To access the public domain of this bioinformatics pipeline, please go to the web site: www.mouse-genome.bcm.tmc.edu

POSTER 135 - NINE STRAIN-SPECIFIC MOUSE BAC LIBRARIES

Nefedov M 1, Zhu B 1, Yoshinaga Y 1, Luo M 2, Wing R 2, Tomkins J 3, Misceo D 1, Hallers B 1, Cao Q 1, Osoegawa K 1, De Jong P 1

1 BACPAC Resources, Children’s Hospital Oakland Research Institute, Oakland, CA94609, United States,

2 Arizona Genomics Institute, Tucson, AZ 85721, United States, 3 Clemson University Genomics Institute, Clemson, SC 29634, United States

With the release of the human and mouse draft DNA sequences, there is a growing need for functional evaluation of various genes and their regulatory elements.  We constructed nine 10-fold redundant mouse strain-specific BAC genomic libraries from inbred mouse strains: Mus musculus C57BL/6J, DBA/2J, NOD, A/J, BALB/cByJ, C3H/HeJ, AKR/J, and feral strains derived from M. m. castaneus and Mus spretus. Providing more than 10 genome equivalents, each library contains 160,000-200,000 clones with average 143-185Kb insert sizes, arrayed in 384-well plates. For screening, BAC clones were gridded onto high density nylon filters. The complete filter sets were screened by hybridization with 30 conserved overgo probes. The average number of about nine hybridization-positive clones per probe is in agreement with the overall redundancy of the libraries. Half of NOD mouse library (98,000 clones) will be end-sequenced to create a mapped BAC resource. These libraries are indispensable tools for the characterization of genes that contribute to complex traits and strain-specific phenotypes.

POSTER 136 - USING REAL-TIME PCR ALLELIC DISCRIMINATION ASSAY TO QUANTITATE TRANSGENES IN PHENOTYPE RESCUE EXPERIMENTS

Bosak NP, Li S, Li X, Reed DR, Beauchamp GK, Bachmanov AA

Monell Chemical Senses Center, Philadelphia, United States

To determine the zygosity of transgenic animals and transgene copy number with currently available techniques demands tedious procedures with sometimes ambiguous results. Real-time PCR is a high throughput quantitative method, accurate and sensitive enough to discriminate a two-fold difference in the amount of target sequences. We investigated the validity of allelic discrimination assay using real-time PCR with allele-specific probes for the transgene quantification. We used DBA/2J;129P3/J-TgN(Tas1r3B6Ge)41Mon mice obtained in a transgenic complementation (phenotype rescue) experiment. The transgene-positive mice were either homozygous or hemizygous for the transgene. The assay was optimized to distinguish between a transgenic and an endogenous allele of the Tas1r3 gene. We used two quantitation approaches: 1) the comparative cycle threshold method and 2) endpoint reading of normalized fluorescence.  Zygosity status was determined using the allelic discrimination assay, and was confirmed by Southern analysis and progeny segregation. Real-time quantitative PCR allows precise quantification of transgene copy number and rapid determination of zygosity in transgenic animals. Furthermore it is more efficient than the traditional methods and particularly suitable for phenotype rescue transgenic experiments.

Supported by NIH grants R01AA11028 (AAB) and R01DC00882 (GKB).

POSTER 137 - FRESH AND CRYOPRESERVED SPERM QUALITY, DNA INTEGRITY, AND IN VITRO FERTILITY IN TWO HYBRID STRAINS OF N-ETHYL-N-NITROSOUREA (ENU) MICE

Cengiz Yildiz CY, Craig Fleming FC, Amanda Cao AC, Colin McKerlie CM

1 The Hospital for Sick Children, Toronto, Canada, 2 The Hospital for Sick Children, Toronto, Canada, 3 The Hospital for Sick Children, Toronto, Canada, 4 The Hospital for Sick Children, Toronto, Canada

Background: Efficient collection, freezing, archiving, and re-derivation of sperm are essential components for random mutagenesis in the mouse. The purpose of this study was to determine the effect(s) of a triple dose of ENU (225 or 300 mg/kg total dose) on fresh and frozen-thawed sperm quality, spermatozoa DNA integrity, and unassisted in vitro fertility rates in C3B6G1 and B6129S1G1 mice that are generated, used, and maintained in the mutagenesis program at the Centre for Modeling Human Disease in Toronto.

Methods:  Sperm was collected from C3B6G1 and B6129S1G1 control and ENU mice. Sperm assessment parameters included progressive motility, plasma membrane integrity, concentration, membrane function integrity, acrosome integrity, and DNA integrity. To assess the effect(s) of oocyte donor strain, the in vitro fertilization rate, embryo culture, and live-born offspring rates from fresh and frozen sperm from G1 and age-matched control wild-type hybrid mice were determined when either of the parental or hybrid strain females were used as oocyte donors.

Results: There were no significant differences in fresh or frozen sperm quality parameters between wild-type and C3B6G1 or B6129S1G1 mice (P>0.05) with these doses of ENU. However, there were significant differences in the in vitro fertilization rate, embryo culture rate, and live-born offspring rates when the background strain of the oocyte donor did not match the sperm donor. Representative data will be presented to confirm that background strain of sperm or oocyte donor is an important consideration and can affect the efficiency of derivation of mice from frozen mouse sperm.

POSTER 138 - PHENOTYPIC, GENETIC AND DEVELOPMENTAL CHARACTERIZATION OF RECESSIVE ENU-INDUCED MOUSE MODELS OF CLEFT PALATE.

Bjork BC, Vieira AR, Van Parijs L, Murray JC, Beier DR

1 Div. Genetics, Brigham & Women's Hospital, Boston, MA, United States, 2 Dept. Pediatrics, The University of Iowa, Iowa City, IA, United States, 3 Center for Cancer Research, M.I.T., Cambridge, MA, United States

We observe a large spectrum of recessive phenotypes similar to human congenital disorders, including several craniofacial defects, in a late embryonic ENU mutagenesis screen. Clefting occurs in many known mouse mutants, but those with isolated clefts are ideal models of human non-syndromic cleft lip/palate (NSCL/P) (1/500-1000 live births).  The cleft palate only 1 (cpo1; isolated cleft secondary palate) and curly tail/cleft palate (ctcp; cleft secondary palate and/or a curly tail) mutations are excellent models.

 The cpo1 phenotype is likely hypomorphic due to inefficient splicing of a Zn finger transcription factor on chr 4 (Hs. 1p36).  A specific defect in palatal shelf elevation is evident upon histologic evaluation.  cpo1 gene expression during craniofacial development is consistent with its playing an important role in palatogenesis.  We identified three potential etiologic missense mutations in a screen of human NSCL/P cases from Iowa and the Philippines, and a CPO1 haplotype is in strong linkage disequilibrium with Filipino NSCL only and shows a borderline association with Iowa NSCP only, thereby supporting an important role for the CPO1 gene in the complex etiology of NSCL/P. 

We identified a causative nonsense mutation in a novel protocadherin gene on chr 3 (Hs. 4q28) in ctcp mutants.  ctcp gene expression during development is consistent with its phenotype. Identification of downstream target genes and interacting proteins for these genes is in progress.  We are also utilizing lentiviral-expressed RNAi to rapidly “knock down” these genes in mouse embryos and provide confirmation of the etiology of these ENU-induced mutations. 

POSTER 139 - REFINEMENT OF BART PROJECT, INSERTIONAL MOUSE MUTAGENESIS PROGRAM AT BAYLOR

Adams CPA, Castile CAC, Petit DCP, Poirier CP, Bishop CEB

Baylor College of Medicine, Houston, United States

We previously outlined a phenotype driven mutagenesis project in the FVB inbred albino background.  Our concatameric transgenes contained tyrosinase and was flagged by repeats specific for Sleeping Beauty transposases.  Transmission of the transgene was followed with pigmented progeny.  By use of a transposase, single copies of the tandem array of transgenes were excised and pasted else where in the genome. “Jumps” were identified by the appearance of new color patterns indicating new expression patterns/loci of tyrosinase.  New integrations, after being separated from the original transgene, were then brought to the homozygous level and screened for abnormal phenotypes.  Given the high frequency of jumps, a single transgene can theoretically be inserted, excised, then inserted again in the germ line.  Given this possibility, associated phenotypes need to be linked to either the genetic disruption caused by insertion of the transgene at the new locus or from the molecular signature left from the previous integration site.  In the case where the two sites are closely linked, we have developed a method to distinguish which loci is responsible for the phenotype. 

Additionally, we have begun using a linearized version of the transgene which improves upon the original design of the project in two important ways.  Firstly, new colors can be unambiguously identified.  Linearized transgenic founders are albino due to vector regions which inhibit expression of tyrosinase.  New colors seen in progeny are therefore unmistakably induced by the integration of tyrosinase at a new locus.  Secondly, a linearized transgene enhances both the rate and ease of cloning; a significant advantage when original transgene site and new integration site are closely linked.

POSTER 140 - Establishing  novel mutant mouse models that mimic human diseases through ENU-based genome-wide mouse mutagenesis

Kung JT

Institute of Molecular Biology, Academia Sinica, Taipei, Taiwan 11529

With the human genome sequenced, the race is on to understand the functions of the nearly 40,000 or so genes. Under the support of the National Science Council and Academia Sinica, a national program of large-scale genome-wide mutagenesis of the mouse has been set up.  Ethyl-nitrosourea (enu) was used to treat founder C57BL/6 (B6) male mice, followed by a 3-generation breeding scheme to generate G3 mice that were screened for recessive phenotypes.  Two screening programs have been set up.  As a service to the research community, the G3 mice are provided as a service to investigators who has received funding to perform their unique phenotypic screens.  The core facility also performs phenotypic screens as part of its internal program.  So far, the core has been screening for blood cell composition, some blood chemistry assays and easily observed traits.  Mutants that have been identified include elevated platelet count, depressed white blood cell count, abnormal limb development, coat color mutant, etc.  Screening results and subsequent heritability testing results of these mutants will be presented and discussed.  Individuals with interests in performing phenotypic screens in a collaborative arrangement are welcome to visit our website at http://mmp.sinica.edu.tw or contact us at mouse@gate.sinica.edu.tw

POSTER 141 - HIGH RESOLUTION MAPPING OF THE TEETERING MOUSE EPILEPSY GENE AND INVESTIGATION OF CANDIDATE GENES

Moffatt S, White AH, Gardiner RM, Rees M

Department of Paediatrics & Child Health, Royal Free & University College Medical School, University College London, London, United Kingdom

The autosomal recessive teetering mutation arose spontaneously in the 1950’s on a C3H/HeJ background.  Homozygous teetering (tn/tn) mice were shown to exhibit progressive ataxia, growth retardation and muscle wasting following weaning with eventual death at 5-6 weeks.  More recently teetering mice were shown to display bilaterally synchronous spike-wave discharges on cortical electroencephalogram recordings (J. Noebels, pers. comm.) linking teetering with a group of mouse models with absence epilepsy.  The teetering gene was originally mapped to the distal end of mouse chromosome 11 (Lane, 1967).  We typed offspring from a (B6C3Fe-a/a-tn x CAST/Ei) F1 interspecific backcross for microsatellite markers on distal chromosome 11 and localised tn to a 1cM region between D11Mit104 and D11Mit69. This region was subsequently narrowed to a smaller region of approximately 0.5 cM (~1Mb) between the genes Nptx1 and Pde6G using novel microsatellite and RFLP markers.  Here we will present data further refining this region to a 420kb interval between two further novel markers  In addition we report data evaluating candidate genes within the region and further characterise teetering brain pathology using immunohistochemical staining of cerebellar purkinje and granule cells. 

Lane LM (1967) [tn]. Mouse News Letter 37:34

POSTER 142 - MODELING AUTISM IN THE MOUSE

Nadler JJ, Moy SS, Zou F, Wright FA, Crawley JN, Threadgill DW, Magnuson TR

1 University of North Carolina, Chapel Hill, United States, 2 National Institute of Mental Health, Bethesda, United States

Autism is a uniquely human disorder comprised of three major behavioral components:  social deficits, repetitive/ritualistic behavior and communication abnormalities.  There is a wide range of severity observed in autistic patients as well as the milder, non-clinical manifestation of the Broader Autism Phenotype (BAP) in family members.  Just as humans can span the range of autistic to BAP to reserved to outgoing, various inbred lines of mice exhibit a range of different behaviors. 

We are correlating molecular profiling with social behaviors and cognitive flexibility of a variety of inbred strains in order to begin to dissect the complex trait of autism.  Data on social behavior are being collected using a three-chambered apparatus containing an unfamiliar mouse in a wire cage on one side, an empty wire cage on the other and a clear middle chamber.  Several measurements are taken over the ten-minute task, including time spent in each chamber, number of entries into each chamber and sniffing directed at the wire cages.  A second ten-minute task involved placing another unfamiliar mouse in the previously empty wire cage to assess preference for the novel mouse over the now-familiar mouse.  Cognitive flexibility is being measured through training and reversal on a t-maze.  Animals are trained to find a food reward in one arm of a t-shaped maze.  Upon reaching criterion, the food reward is switched to the opposite arm, measuring the animal’s ability to relearn the task in a different way.

We are observing that these behavioral characteristics vary in some strains, indicating a strong genetic component.  In order to profile the expression of genes associated with autism-like behaviors, microarray analyses are being conducted on seven brain regions implicated in autism or social behavior.  Data on strain-dependent gene expression will be presented. In this way we hope to begin to dissect the complex genetics of social behaviors and autism.

POSTER 143 - LOSS OF P53 SENSITIZES MICE WITH A MUTATION IN CCM1 (KRIT1) TO DEVELOPMENT OF VASCULAR MALFORMATIONS

Plummer NW 1, Gallione CJ 2, Srinivasan S 2, Louis DN 3, Marchuk DA 2

1 National Institute of Environmental Health Sciences, Research Triangle Park, NC, United States, 2 Department of Molecular Genetics and Microbiology, Duke University Medical Center, Durham, NC, United States, 3 Department of Pathology, Neurosurgical Service and Cancer Center, Massachussets General Hospital and Harvard University, Boston, MA, United States

Cerebral cavernous malformations (CCM) consist of clusters of abnormally dilated blood vessels within the central nervous system.  Hemorrhaging of these lesions can result in headaches, seizures, and lethal stroke.  Three loci are associated with autosomal dominant CCM, and the causative genes have been identified for CCM1 and CCM2.  The allelic series for both genes suggests that disease results from loss of protein function, and it has been proposed that growth of cavernous malformations depends on somatic mutation of the wild-type allele in heterozygotes.  To test this hypothesis, we bred mice that were heterozygous for a targeted mutation of Ccm1 and homozygous for loss of the tumor suppressor Trp53 (p53) which has been reported to increase the rate of somatic mutation due to mitotic recombination, interstitial deletion, and whole chromosome loss and reduplication.  We observed vascular lesions in the brains of 5/9 of the Ccm1+/- Trp53-/- animals but none in littermates of other genotypes.  Histologic and immunohistochemical analyses revealed that the mouse lesions shared distinctive characteristics with human cavernous malformations.  Although the genetic evidence suggested somatic mutation of Ccm1, laser capture microdissection of a mouse lesion failed to reveal loss of heterozygosity.  An alternative explanation for these results is that p53 plays a direct role in formation of the vascular lesions.  The striking similarity of the human and mouse lesions indicates that the Ccm1+/- Trp53-/- mice are an appropriate animal model of cerebral cavernous malformations.

POSTER 144- FITFUL – A NEW MOUSE MODEL OF IDIOPATHIC EPILEPSY AND ATAXIA

Roberts MC, Letts VA, Beyer B, Frankel WN

The Jackson Laboratory, Bar Harbor, United States

The fitful mouse mutation arose spontaneously on the C57BL/6J mouse strain.  Affected mice begin to have recurrent, non-lethal handling induced seizures (with limbic and generalized components) at two to three months of age, and this phenotype is dominant with 50% penetrance.  In order to understand the inheritance and to map the disease, the mice were crossed with FVB/NJ and CAST/EiJ strains. Although on the inbred C57BL/6J strain, the phenotype is homozygous lethal, either in the embryonic or perinatal period, in the mapping crosses homozygotes survive for about three weeks, with ataxic symptoms starting at over two weeks.  These symptoms include ataxic walk, wild runs and seizures.  The homozygous pups are also smaller than the unaffected littermates.  The epilepsy and ataxia were mapped to chromosome 2 in both crosses.  The fitful gene is located in an interval on chromosome 2 flanked by markers D2Mit120 (27.98 Mb) and D2Mit83 (29.27 Mb) that contains 19 genes and transcripts.  None of the genes in the interval appear to be ion channels or other genes that are already associated with seizures and ataxia, suggesting that fitful represents a novel form of idiopathic epilepsy and ataxia in the mouse.  

POSTER 145 - CHOLINE KINASE BETA AND A MURINE MODEL OF MUSCULAR DYSTROPHY

Sher RB, Cox GA

The Jackson Laboratory, Bar Harbor, United States

In August 2001, mice in a colony at The Jackson Laboratory developed a muscular dystrophy (MD) resulting from a spontaneous recessive mutation on Chromosome 15.  The disease is observable around birth with a disordered forelimb bone development, and mice rapidly develop a progressive muscular dystrophy.  The causative gene has recently been identified as Choline Kinase eta (Chk-eta), which is involved in the first step of the production of phosphatidylcholine.  A mutation in Chk-eta does not correspond to any currently identified human muscular dystrophy.

Of the two major forms of MD membrane defects, the first results in inherent instability of the membrane through faulty structural proteins.  The second is due to compromised membrane repair after normal membrane tearing.  Recently, the loss of dysferlin has been shown to be the cause of Limb-Girdle MD due to its role in resealing muscle membrane tears.

Our MD model involves a gene involved in the production of one of the major constituents of muscle cell membranes, and may be involved in membrane repair, and therefore related to the dysferlin dystrophies.  Potential therapies for muscular dystrophies depend upon detailed knowledge of both the genetic and functional causes of the diseases. New animal models such as this may assist in the discovery of novel mechanisms of disease progression, and therefore may aid in the development of therapeutic treatments. Understanding the biology of this mouse line may provide important insight into the biological pathways leading to muscular dystrophy and muscle repair.

POSTER 146 - GENETIC PATHWAY ANALYSIS OF MOM1 CANDIDATE GENES

Cormier R, McAlpine C, Bogan C, Bluemn E

University of Minnesota School of Medicine, Duluth, MN, United States

The Multiple Intestinal Neoplasia (Min) mouse has been extensively studied as a model for the human form of colorectal cancer called familial adenomatous polyposis (FAP).  Min mice carry a germline heterozygous mutation in the Apc gene that predisposes to the development  of numerous adenomas throughout the intestinal tract. Notably, the Min phenotype is mouse strain dependent.  For example, the C57BL/6 strain is highly sensitive to Min-induced intestinal tumorigenesis, while several strains, including AKR, are highly resistant to tumor development.  Linkage analysis of backcross progeny arising from crosses between tumor resistant and sensitive strains has mapped a major resistance modifier of the Min phenotype to distal chromosome four. This tumor resistance locus, called Mom1 (Modifier of Min 1), has been shown to consist of a complex of two or more closely linked candidate genes.  Our laboratory has a major research focus in the identification and functional characterization of the Mom1 genes.  We continue work conducted over a number of years by groups in Madison, WI and Cambridge, MA. These collaborative studies indicate that the proximal Mom1 gene is likely to be Pla2g2a, which encodes for a phospholipase that is active at the plasma membrane. Pla2g2a provides resistance to Min tumorigenesis throughout the intestine but especially in the colon.  In contrast the distal Mom1 component exerts a resistance phenotype that is restricted to the small intestine.  Ongoing experiments include efforts to map and clone the distal Mom1 gene.  Progress on this mapping project will be presented.  In addition, we intend to present results of a functional test of a distal Mom1 candidate gene that carries a targeted knockout mutation that has been introgressed into the C57BL/6-Min strain.  To understand the action of the proximal Mom1 gene, Pla2g2a, we have experiments underway to identify genetic pathways utilized by Pla2g2a.  These include potential interactions with PPARg and Mucin 2.

POSTER 147- A SERIES OF MATURITY ONSET DIABETES OF THE YOUNG, TYPE 2 (MODY2) MOUSE MODELS GENERATED BY A LARGE SCALE MUTAGENESIS PROJECT IN RIKEN GSC

Inoue MI 1, Sakuraba YS 1, Motegi HM 1, Kubota NK 2, Terauchi YT 2, Toki HT 1, Matsui JM 1, Toyoda YT 3, Miwa IM 3, Shigeyama YS 4, Kaneda HK 1, Ishijima JI 1, Masuya HM 1, Suzuki TS 1, Wakana SW 1, Gondo YG 1, Minowa OM 1, Shiroishi TS 1, Noda TN 1

1 RIKEN GSC, Yokohama, Japan, 2 Univ. Tokyo, Tokyo, Japan, 3 Meijo Univ., Nagoya, Japan, 4 Kobe Univ., Kobe, Japan

In RIKEN mutagenesis project, to generate mouse models for common human diseases, we screened for various phenotypes in the mutated animals. We report here a series of new mouse models for human diabetes. We screened 9,375 animals for dominant traits using a clinical biochemical test and thereby identified 11 mutations in the glucokinase (Gk) gene that were associated with hyperglycemia. GK is a key regulator of insulin secretion in the pancreatic beta cell. Heterozygous mutations in human GK cause maturity onset diabetes of the young, type 2 (MODY2) heterozygously, and homozygous mutations result in cause permanent neonatal diabetes mellitus (PNDM). The mutations in our 11 mutant lines were, scattered in the mouse G k k gene and, included six missense mutations, two nonsense mutations, and three mutations in splicing signals. Four lines bears   identical mutations to those found in human MODY2 patients, and another bears the its mutation at the same location that is mutated in a PNDM patient. All of our mutant lines displayed mildly but consistently hyperglycemia, a similar phenotype to that of human MODY2 patients, while  and the homozygotes analyzed here suffered severe hyperglycemia soon after birth, as has also been similarly reported for in human PNDM patients. Some of our Gk mutant lines displayed impaired glucose-responsive insulin secretion and the mutations had different effects on Gk mRNA levels and/or the stability of the GK protein. This collection of Gk mutants will be highly useful for understanding GK gene function, for dissecting the function of the enzyme, and as models of human MODY2 and PNDM.

POSTER 148 - COMPREHENSIVE, STANDARDIZED PHENOTYPING OF MOUSE MUTANTS  IN THE GERMAN MOUSE CLINIC (GMC)

Fuchs H 1, Gailus-Durner V 1, Lengger Ch 1, Reinhard C 2, Schluz H 2, Calcada-Wack J 3, Elvert R 4, Dalke C 5, Franz TJ 6, Grundner-Culemann E 1, Hoelter SM 5, Javaheri A 7, Kalaydjiev S 6, Klempt M 8, Kunder S 3, Mijalski T 1, Horsch M 1, Pedersen V 5, Prehn C 1, Radc I1, Schneider I 1, Ehrhard N 4, Brielmeier M 2, Lengeling A 3, Müller W 4, Reitmeir P 9, Schmidt J 2, Adamski J 1, Beckers J 1, Behrendt H 7, Busch DH 6, Favor J 5, Graw J 5, Heldmaier G 4, Heyder J 2, Höfler H 3, Klingenspor M 4, Klopstock T 1, Meitinger T 5, Ollert M 7, Quintanilla Martinez L 3, Wolf E 8, Wurst W 5, Zimmer A 1, Hrabe de Angelis M 1

1 GSF Research Center, Institute of Experimental Genetics, Munich, Germany, 2 GSF Research Center, Institute of Inhalation Biology, Munich, Germany, 3 GSF Research Center, Institute of Pathology, Munich, Germany, 4 Phillips Univsersity, Department of Biology, Marburg, Germany, 5 GSF Research Center, Institute of Developmental Genetics, Munich, Germany, 6 TUM, Institute of Medical Microbiology, Immunology and Hygieny, Munich, Germany, 7 TUM Division of Environmental Dermatology and Allergy, Munich, Germany, 8 LMU, Lehrstuhl für  Molekulare Tierzucht und Biotechnologie, Munich, Germany, 9 GSF Research Center, IGM, Munich, Germany,

During the last years, large-scale projects like ENU mutagenesis, gene trap or knock-out generation, brought about many new mouse mutants. In many cases those lines were generated by a highly-specialized lab, and had not been comprehensively phenotyped under standardized conditions. Since mutations might have pleiotropic effects causing phenotypes in different organ systems and pathways, it is very likely that additional phenotypes were not uncovered, and the enormous scientific value of those new mutants remained unexploited. To overcome this “phenotyping gap”, the German Mouse Clinic (GMC, www.mouseclinic.de) was established. The GMC offers a phenotyping platform for comprehensive and standardized phenotyping of mouse mutants in the fields of behaviour, dysmorphology, bone and cartilage, neurology, eye function and development, clinical chemistry, immunology, allergy, steroid metabolism, nociception, expression profiling, lung function, cardiovascular diseases, energy metabolism and pathology. Additional screens for host-pathogen interaction are performed at collaborating institutes. In a primary screen more than 200 key parameters covering fundamental biological pathways are determined. When new phenotypes of interest are detected, more detailed analysis can be performed in secondary and tertiary screens.

Twenty mutant lines have already finished the primary phenotyping in the German Mouse Clinic. In nearly 90% of the lines, new, previously unknown phenotypes could be assigned to the mutant. More than 100 significant differences between mutants and their littermate controls were identified. All screens of the German Mouse Clinic contributed new phenotypes in various mutant lines. A more detailed analysis of these mutant lines in secondary screens is currently in progress.

This work is supported by the National Genome Research Network NGFN (GR-010313).

POSTER 149 - A SYSTEMATIC, PHENOTYPE-DRIVEN MUTAGENESIS SCREEN FOR GENE FUNCTION STUDIES: RECENT RESULTS OF THE MUNICH ENU-MOUSE-MUTAGENESIS SCREEN

Soewarto D 1, Wagner S 1, Rathkolb B 3, Fuchs H 1, Moor M 3, Klempt M 3, Howaldt M 3, Kalaydjiev S 2, Franz T 2, Schneider I 1, Marschall S 1, Boersma A 1, Schäble K 1, Tiedemann H 1, Schneltzler E 1, Steinkamp R 1, Alessandrini F 5, Jakob T 5, Binder E 9, Kremmer E 6, Behrendt H 5, Ring J 5, Zimmer A 7, Peters C 1, Flaswinkel H 3, Busch D 2, Pfeffer K 2, Klopstock T 1, Gekeler F 1, Ohl F 9, Balling R 8, Wolf E 3, Hrabe de Angelis M 1

1 GSF Research Center, Institute of Experimental Genetics, Munich, Germany, 2 Institute of Medical Microbiology, Immunology and Hygiene, Technical University of Munich, Munich, Germany, 3 Institute of Molecular Animal Breeding, Gene Center, University of Munich, Munich, Germany, 4 Max-Delbrueck-Centre, Molekulare Genetik und Mikrosatellitenzentrum, Berlin, Germany, 5 Division Environmental Dermatology and Allergology, GSF/TUM, Munich, Germany, 6 Institute of Immunology, GSF Research Center for Environment and Health, Munich, Germany, 7 Division Molecular Neurobiology, Polyclinic for Psychiatry, University of Bonn, Bonn, Germany, 8 German Resaerch Center for Biotechnology, Braunschweig, Germany, 9 Max Planck Institute of Psychiatry, Munich, Germany, 10 Department of Neurology, Clinic Großhadern, University of Munich, Munich, Germany, 11 Institute of Internal Medicine I, Medical Microbiology, University Clinic Freiburg, Freiburg, Germany

Mouse models have proven to play an important role in gene function studies for inherited diseases in humans. We give an update of one of the largest ENU mutagenesis programs in Europe, the Munich ENU Mouse Mutagenesis Project. After having focused on dominant traits in the first years, we put our main efforts during the last years on recessive phenotypes. In addition, we continue to produce F1 animals to further isolate novel dominant alleles of known and new genes.

Currently, more than 34.000 mice have been investigated for dysmorphology and blood based parameters. To date, more than 550 mutant lines have been isolated. Novel dominant or recessive phenotypes have been identified with specific abnormalities comprising congenital malformations, biochemical alterations, immunological defects and complex traits such as behaviour or predispositions to allergies.

Mutants of clinical relevance for inherited diseases in human have been further analysed by backcross mapping and genome-wide microsatellite typing. Many mutant lines deriving from this ENU Screen are under detailed phenotypic characterisation and have been proceeded for fine mapping and positional cloning (Kiernan et al. 2001, Graw et al. 2001). With Beethoven (Bth) (Vreugde et al. 2002) a semidominant mouse model for progressive hearing loss was found. Some new mutant lines with clinical relevance from various screens associated to the ENU project will be shown in this presentation.

Reference

Graw J. et al. 2001. Characterization of a mutation in the lens-specific MP70 encoding gene of the mouse leading to a dominant cataract. Exp Eye Res 73, 867-876

Kiernan A. et al. 2001. The Notch ligand Jagged1 is required for inner ear sensory development. PNAS 98, 3873-3878

Vreugde S. et. al. 2002. Beethoven, a mouse model for dominant, progressive hearing loss DFNA36. Nature Genetics 30, 257-258

POSTER 150 - EXPRESSION AND FUNCTION OF BASP1: A COSUPPRESSOR OF THE WILMS’ TUMOUR SUPPRESSOR PROTEIN, WT1

Allsop JE 1, Moorwood K 1, Charalambous M 1, Lahiri D 1, Dutton JR 1, Roberts GE 2, Ward A 1

1 University of Bath, Bath, United Kingdom, 2 University of Manchester, Manchester, United Kingdom

The Wilms’ tumour suppressor protein, Wt1, is a transcriptional regulator that plays a key role in the development of several organ systems. Wt1 homozygous knock-out mice showed agenesis of the kidneys, gonad and spleen, and also abnormal development of the heart diaphragm and retinal ganglia. The transcriptional activation domain of Wt1 is subject to regulation by an N-terminal suppression domain. This transcriptional suppression domain has been proposed to function by recruiting a cosuppressor protein to Wt1 that can block the function of the transcriptional activation domain. Transfection assays have identified Brain acid-soluble protein (Basp1) as a component of the Wt1 cosuppressor that can regulate Wt1 transcriptional activity. Basp1 is a nuclear factor that associates with Wt1 in cells that naturally express both proteins. Analysis of embryonic and adult kidney sections reveals that Basp1 is present in the developing nephron structures of the embryonic kidney and that, coincident with Wt1, its expression is restricted to the highly specialized podocyte cells of the adult kidney. Basp1 however is also more widely expressed than Wt1, including within the CNS, heart, liver and tongue. We are developing a mouse knockout model to test whether Basp1 and Wt1 interact in vivo, and to ascertain the role Basp1 plays in development.

POSTER 151 - A ROLE FOR SONIC HEDGEHOG IN THE CEREBELLAR DEFICITS OF THE TS65DN MOUSE MODEL OF DOWN SYNDROME

Saran NG, Klinedinst DK, Roper RJ, Baxter LL, Beachy PA, Reeves RH

Johns Hopkins University School of Medicine, Baltimore, United States

Down syndrome (DS), the most common viable autosomal aneuploidy in humans, is caused by trisomy 21. The Ts65Dn mouse is a model with segmental trisomy for approximately 16 Mb at the distal end of mouse chromosome 16 in a region that shows near-perfect conserved synteny with human chromosome 21, and includes orthologs of about half the chromosome 21 genes. Like humans with DS, the Ts65Dn mouse has a significantly reduced cerebellum and reduced density of Purkinje and granule cells that comprise the internal granule layer of the cerebellum. We present a postnatal profile of the maldevelopment in the granule cells that identifies the developmental stage at which the trisomy-related morphological differences first appear. We also show that the granule cell precursors (gcp) are the affected cell population. Further, we investigate the effects of sonic hedgehog (Shh), a potent mitogen that activates transcription of several genes in gcp. We show reduced activation of genes whose transcription is regulated by Shh and a reduced yet dose-dependent response of the trisomic gcp to the Shh-induced mitogenic pathway. This deficit underlies the insufficient generation of granule cells seen in Ts65Dn mice. Using gene expression data from trisomic and euploid gcp stimulated by Shh, we are analyzing the basis for this cell-autonomous response deficit.

POSTER 152 - ATP2B2 EXHIBITS DIFFERENTIAL FIRST EXON USAGE IN MOUSE BRAIN AND COCHLEA

Silverstein RS 1, Tempel BL 2

1 Graduate Program in Neurobiology & Behavior, University of Washington, Seattle, United States, 2 Department of Otolaryngology, University of Washington, Seattle, United States

Deafwaddler (dfw) mice are deaf and exhibit wobbly gait. The dfw mutation was localized to Atp2b2, encoding a plasma membrane Ca2+-ATPase, PMCA2. Heterozygous mutants of Atp2b2, while not deaf, have significant high-frequency hearing loss, implying that tight regulation of its expression is required. PMCA2 is highly expressed in stereocilia of cochlear outer hair cells, cerebellar Purkinje cells, and epithelial cells of lactating mammary gland, suggesting that it is important for hair cell Ca2+ homeostasis. To understand the pathways that control PMCA2 expression, we have examined transcriptional regulation of the Atp2b2 gene. Using antisense primers in the open reading frame, 5’-RACE was performed on RNA from multiple tissues in CBA mice. Sequencing of clones identified three primary alternative exons of transcriptional initiation, implying regulation through distinct promoters. Two of these (1c and 1f) were observed in cochlea and cerebellum. A third (1m) was restricted to lactating mammary gland. Quantitative real-time RT-PCR (qPCR) confirmed the findings from 5’-RACE. To determine the promoter used in hair cells, cochleae were microdissected by separating organ of corti (OC) from modiolus (MOD). qPCR was performed on the OC and MOD samples, using marker genes to confirm quality of dissection. This analysis revealed that exon 1f dominates in the OC (hair cells), while exon 1c is most abundant in MOD (spiral ganglion neurons). In silico analysis of putative regulatory sequences revealed high levels of conservation between mouse, rat, and human. Further analysis by in situ hybridization will confirm cell-specificity of promoter use in the different tissues.

POSTER 153 - EUMORPHIA AND THE EUROPEAN COMPREHENSIVE FIRST LINE PHENOTYPING PROTOCOL

Lad HV, Coghill and The Eumorphia Consortium E L

MRC Mammalian Genetics Unit, Harwell, United Kingdom

EUMORPHIA is an EU supported integrated research programme incorporating the expertise and resources of many European mouse genetics centres.  The focus is on the development of new approaches in phenotyping, mutagenesis and informatics leading to improved characterisation of mouse models for the understanding of human physiology and disease.

One of the key aims is to improve the reproducibility and comparability of phenotype tests across different laboratories and to provide access to new and improved phenotyping platforms. A major achievement has been the establishment of a new robust primary screening strategy – the European Comprehensive First Line Phenotyping protocol (ECFLP) for many of the body systems in the mouse. Validation and refinement of the tests selected for the ECFLP is underway using specific inbred and known mutant lines. Crucial to the development of the ECFLP as a resource has been its integration with a “standard operating procedure” (SOP) XML database as well as appropriate ontologies for improved data analysis, data dissemination and continuous SOP refinement. This resource will be publicly available through the EUMORPHIA website (www.eumorphia.org) from summer 2004.

Two levels of first-line phenotyping have been defined; “primary tests” represent the routine core tests of the ECFLP protocol; “primary extended tests” are those valuable in a first-line screen, but dependent upon availability of specialised equipment, skills and resources. Progress within the “Sensory Systems” workpackage, which covers audition, vestibular function and vision, is described as an example of the type of work undertaken within this programme.

POSTER 154 - RAPID EFFICIENT PLACEMENT OF NEW MUTANT PHENOTYPES ON THE MOUSE GENETIC MAP

Rowe LB, Barter ME

The Jackson Laboratory, Bar Harbor, ME, United States

Major mutagenesis programs are underway at several institutions, providing mice for screening for diverse phenotypes to develop new models for human disease and to help dissect biological pathways. At The Jackson Laboratory, ENU-mutagenized C57BL/6J mice are being screened by three major project groups, Neuromuscular Function (NMF), Heart, Lung, Blood and Sleep Disorders (HLBS), and Reproductive Genomics (ReproGenomics), as well as many smaller project interests (eg. eye defects, hearing defects, obesity, and others). Rapid localization of these new mutations on the mouse genetic map is essential to define remutations at previously known loci or identify newly affected genes. We have developed a high throughput genome scanning system that regularly allows defining chromosomal localization within two days of the availability of animals segregating the phenotype. From as few as 6 affected and a few unaffected N2 or F2 animals (or, less optimally, G3 or G4 sibships), genome scanning by haplotype analysis (Neuhaus and Beier, 1998 Mammalian Genome 9: 150-154) rapidly identifies the candidate region. If more than ten affected and a similar number of unaffected N2, F2 or G3 are available for the mapping, genome scans by pooling (Taylor et al. 1994 Genomics 21: 626-632) are the most efficient method. Using these two scanning methodologies, we have been able to successfully map an average of two new mutations a week. Our lab also provides rapid DNA preparation for genotyping, and genotyping services for other in-house mapping projects including congenic line breeder selection ("speed congenics") and major modifier locus mapping.

POSTER 155 - THE EUROPEAN MOUSE MUTANT ARCHIVE (EMMA)

Zeretzke SZ 1, Hrabé de Angelis MHdA 1, Tocchini-Valentini GTV 2, Herault YH 3, Brown SB 4, Ährlund-Richter LAR 5, Mallo MM 6, Thornton JT 7

1 GSF, Research Center for Environment and Health, Institute of Experimental Genetics, Munich, Germany, 2 CNR, Istituto di Biologia Cellulare, Monterotondo, Italy, 3 CNRS, Centre de Distribution, de Typage et d´Archivage animale, Orleans, France, 4 MRC, Mammalian Genetic Unit, Harwell, United Kingdom, 5 Karolinska Institutet, Unit for Embryology and Genetics, Stockholm, Sweden, 6 FCG, Instituto Gulbenkian de Ciencia, Oeiras, Portugal, 7 EMBL/EBI European Bioinformatics Institute, Hinxton, United Kingdom

The EMMA (European Mouse Mutant Archive) was established as part of a worldwide network of repositories for retaining all generated mutant strains essential for biomedical and medical research. Among cryopreservation of mouse mutant lines in form of embryos or sperm, the distribution of these lines to qualified investigators is the main focus. EMMA maintained lines are supplied as a service to the research community at large and solely for research purposes.

EMMA strains are available to academic institutions from all around the world. Applications for depositing and requesting mutant strains can be submitted through the EMMA website www.emmanet.org.

In order to make the transfer of biological information more effective, appropriate databases will be further developed in the framework of the EMMA and in collaboration with existing databases. An essential role of the EMMA project is to foster the virtual coupling of the EMMA stock centers and related information systems through the establishment and maintenance of a dedicated resource database (EMMA-RDB). This is implemented as a fully integrated web-based database and serves as an interface to the scientific community.

EMMA actively co-operates with "The Jackson Laboratory", USA and other primary repository resources worldwide and is supported by primary public research Institutions of the participating Countries and by the European Commission.

POSTER 156 - AUTOMATED DNA PURIFICATION FROM MOUSE TAILS

Mulrooney C, Howe S, Robinson J, Oultram J, Sayle J

Tepnel Life Sciences, Manchester, United Kingdom

The extraction of high quality genomic DNA is the first and often limiting step in the process of mouse tail genotyping.  Tail snips are difficult samples to work with and most of the chemistries and systems available today fail to consistently produce sufficient DNA of an appropriate quality to meet the requirements of PCR and Southern Blotting.

Tepnel’s Nucleopure system not only produces excellent yields of highly intact and pure DNA from 192 tail snips in less than 2 ½ hours, but can also set up downstream PCR reactions as well.

POSTER 157 - CHARACTERIZATION OF TWO NOVEL ENU-INDUCED CIRCADIAN MUTANT LINES

Godinho SIH, Shaw-Andrews L, Clay J, Bacon Y, Nolan PM

MRC, Harwell, United Kingdom

Large scale mutagenesis screens have provided a powerful tool for the characterisation of behavioural mutants used in the study of gene function. Using the chemical mutagen N-ethyl-N-nitrosourea (ENU) we have screened approximately 2000 progeny of chemically mutagenised mice. Approximately 5% of all animals screened presented an abnormal phenotype with 1% showing robustly inherited mutant phenotypes.

Two mutant lines with dominant rhythm phenotypes will be presented here. One, Play 8 (short circuit, Sci) presents a reduced circadian amplitude, a short circadian period in constant darkness (21.4 to 23hrs) and a reduced phase shift in response to light. This phenotype is reminiscent of some features seen in familial advanced phase sleep syndrome. To date, intercrosses have been unsuccessful in identifying a homozygous phenotype. The mutant locus is a novel circadian locus between 106.9 and 110.5Mb on chromosome 8. A second mutant line, Play 68, presents a subtle but significant lengthening of the free-running period (24.3hrs compared to 23.6hrs in wt). No additional anomalies were observed in heterozygotes. Intercrosses revealed an additional class of phenotype in presumptive homozygotes. In light:dark conditions, activity onsets were delayed by approximately 5 hours and free-running period in constant darkness was approximately 27 hrs. Homozygotes display features of familial delayed sleep phase syndrome in humans. This mutant also maps to a novel circadian locus on chromosome 14, between 91 and 99Mb.

We are presently sequencing several potential candidate genes and undergoing fine mapping using additional polymorphic markers for both mutant lines. In addition we are studying the molecular expression of genes known to be expressed in the suprachiasmatic nucleus (SCN) which are closely involved in the regulation of circadian rhythms. Such mutants may ultimately provide an insight into human behaviour and neurological disorders.

POSTER 158 - ENU MUTAGENESIS REVEALS A NEW ROLE FOR THE AF4 PROTEIN FAMILY IN NEURODEGENERATION

Oliver PL, Bitoun E, Clark J, Davies KE

MRC Functional Genetics Unit, Oxford, United Kingdom

We have used ENU mutagenesis to identify new models of neurological disease and have characterised a novel cerebellar mutant named robotic. It displays a jerky, ataxic gait that is caused by an adult-onset loss of Purkinje cells that occurs in a striking region-specific pattern. The causative mutation was identified in a highly conserved region of the putative transcription factor, Af4, and we have shown that this gene is expressed specifically in Purkinje cells. Af4 has previously been implicated in leukaemia, but its function in the brain is unknown.

From a yeast-two hybrid screen in the brain, we identified an ubiquitin ligase as a novel binding partner of Af4. We were able confirm this interaction and demonstrate co-localisation of these proteins both in vitro and in vivo. In addition, quantitative binding assays revealed a highly significant reduction in affinity of this interaction with mutant protein, suggesting that normal turnover of Af4 by the proteosome is blocked in the robotic mouse. To identify potential targets of Af4 transcriptional regulation, a microarray study was carried out using wild-type and robotic cerebellar samples across a time-course. In situ hybridisation results from differentially regulated genes have provided clues to the pathways involved in Purkinje cell death.

The discovery that Af4 is involved in movement and behavioural disorders has implications for other members of the highly conserved Af4 protein family, including Fmr2, revealing more about their normal function and contribution to human disorders as diverse as leukaemia, mental retardation and neurodegenerative disease.

POSTER 159 - CRANIOFACIAL GENE DISCOVERY IN MICE

Donahue LR, Curtain M, Giggey J, Guido V, Hurd J, Davisson M

The Jackson Laboratory, Bar Harbor, United States

We identify and characterize craniofacial disorders in mice to discover genetic and physiological models for human craniofacial dysmorphologies, and share these models with the scientific public. We screen for spontaneous and induced mutations, do comprehensive genetic and characterization studies, and identify gene mutations by candidate gene testing and positional cloning. We provide notification of new models via our website: www.jax.org/cranio/, offer information and training via our email: faces@jax.org, and distribute live mice to investigators worldwide.

Craniofacial dysmorphologies represent nearly 75% of all human congenital malformations visible in newborns and are a complication in over 700 known human genetic syndromes. This diverse group of disorders includes cleft lip and/or palate, abnormal morphology of the skull, face, or jaws, misshapen or missing teeth, plus eye and ear abnormalities, with or without impaired sight or hearing. Discovery of the genetic cause of such dysmorphologies in humans is difficult due to population heterogeneity and to the diversity of interacting environmental and nutritional variables. Consequently, animal models with defined genetic backgrounds maintained in controlled environments are critical for craniofacial gene discovery.

We investigated the heritability of more than 100 craniofacial phenodeviants and found at least 65% of the traits transmissible. Presently we are characterizing the genetic and biologic features of these models. Phenotypes include: cleft lip; abnormal ear pinnae shape, size, and placement; shortened snouts; shortened, domed and asymmetrical skulls; abnormal dentition; abnormal eye morphology, and deafness. Many deviants have more than one phenotype, and are potential models of human craniofacial syndromes.

POSTER 160 - IDENITIFICATION OF ENU-GENERATED MUTANTS FOR ANXIETY AND ACTIVITY IN AN OPEN FIELD TEST

Pletcher MT, Grabowski L, Wiltshire T, Tarantino L

The Genomics Institute of the Novartis Research Foundation, San Diego, United States

The determination of the complex genetic interactions that control behavioral phenotypes has been slow to progress.  In an attempt to exploit new technologies to unravel aspects of behavioral genetics, a forward genetics screen was established to identify phenotypic outliers in an open field test from a population of offspring from ENU-mutagenized mice.  Values were collected for percent time spent in the center of the open field, total distance traveled, percent time resting, amounts of continuous motion, average velocity, and total rearing.  Percent time spent in the center of the open field and average velocity show good correlation as a marker of anxiety in the mice while total distance traveled, percent time resting, and amounts of continuous motion track potential hyperactive phenotypes.  456 ENU-mutagenized families have been screened by this method with 30 being identified as outliers for anxiety and 19 having abnormal activity levels.  To date, 4 have been found to be heritable for one of the phenotypes screened.  The two lines that have been characterized most extensively are Frantic and Angry MouseFrantic shows an anxious-like phenotype of hyperactivity in novel environments while Angry Mouse is characterized by consistently increased activity level and aggressive behavior.  Currently, the causative mutations for these mouse lines are actively being mapped.

POSTER 161 - NEURAL CREST DEFICIT IN CRANIOFACIAL SKELETON PRECURSORS OF TS65DN MICE, A MODEL FOR DOWN SYNDROME

Roper RJ, Reeves RH

Johns Hopkins University School of Medicine, Baltimore, United States

Ts65Dn mice display several phenotypic similarities to Down syndrome (DS) including craniofacial dysmorphology, a distinguishing feature found in all DS individuals.  Neural crest cells (NCC) contribute prominently to the craniofacial skeleton affected in DS and the Ts65Dn mouse.  Because many tissues affected in DS have a NCC component, it has been hypothesized that trisomy 21 causes a defect in NCC.  However, no direct experiments have proven or disproven an effect on NCC in DS.  To examine the hypothesis that trisomy affects NCC and therefore contributes to the etiology of DS craniofacial dysmorphology, we bred Ts65Dn mice with mice homozygous for the Wnt1-lacZ transgene which is expressed predominantly in NCC at embryonic day 9.5 (E9.5). Ts65Dn x Wnt1-lacZ offspring were dissected at E9.5 and exhibited lacZ-marked cranial NCC.  Trisomic and euploid embryos were identified by FISH analysis of the yolk sac and age-matched by somite number.  Although the frequency of Ts65Dn at weaning is 25-30%, we observed no deviation from Mendelian ratios at E9.5.  Additionally, the average somite number in E9.5 trisomic and euploid mice was not significantly different and provided no evidence for “developmental delay” of Ts65Dn mice.  Unbiased stereological techniques revealed both NCC number and volume of the 1st pharyngeal arch were reduced in Ts65Dn vs. euploid littermates, demonstrating an effect of trisomy on NCC and craniofacial skeleton precursors.  Further analysis of embryos at additional somite-specific developmental stages will investigate the molecular etiology of the NCC deficit and its contribution to craniofacial skeletal dysmorphology in Ts65Dn mice.

POSTER 162 - ANALYSIS OF A NEW DOMINANT HAIRLOSS MUTANT MAPPING TO MOUSE CHROMOSOME 11.

Weber JS1, Rahajeng J1, Kovar J2

1 University of Nebraska, Lincoln, United States, 2 LI-COR BioSciences, Lincoln, United States

Development and maintenance of hair in mammals requires multiple genes that control position, growth cycle, and type of hair. Since most of the genes regulating hair have not been determined, we took a genetics-based approach to identify genes involved in hair formation. An ENU mutagenesis project was initiated to screen mice for a variety of abnormalities including hair problems. Mutagenized C57BL/6 males were bred to produce dominant and recessive mutants. A new dominant mutant was found which exhibited hair loss starting between 14 and 21 days of age.  Affected animals undergo progressive rounds of hair loss ending with hair only around the snout, legs and base of the tail.  To map the Hairloss gene, affected mice were outcrossed to DBA/2J mice to produce F1s. F2 mice were generated by backcrossing affected F1 mice to DBA/2J mice. DNA from 50 Hairloss and 50 wild-type F2 mice were pooled by phenotype and amplified with 103 microsatellite markers equally spaced across the genome and polymorphic for C57BL/6 and DBA/2J alleles. Hairloss was mapped initially from 213 F2 mice to distal mouse Chromosome 11 between D11Mit39 and D11Mit181. Individual genotyping identified 35 recombinants between the linked markers. A tiered-mapping approach was used to narrow down the non-recombinant region to approximately 1.77Mb, the size of the region between new markers D11NEB2 (98.3Mb) and D11NEB1 (100.07Mb). Three mice have recombination breakpoints between D11NEB1 and Hairloss; one mouse has a recombination breakpoint between D11NEB2 and the mutant gene. No new recombinants have been found between D11NEB1 and D11NEB2 after genotyping an additional 144 F2 mice. We have found candidate genes in the nonrecombinant region, including homologs of Gasdermin.  We will present analysis of these loci along with more detailed studies of the hair phenotype. 

POSTER 163 - CORRECTION OF PATHOLOGICAL AGGRESSION IN THE ‘FIERCE’ MOUSE BY

HUMAN NUCLEAR RECEPTOR 2E1

Abrahams BS, Kwok MCH, Trinh E, Budaghzadeh S, Hossain SM, Simpson EM

University of British Columbia, Vancouver, Canada

Mouse mutagenesis is a powerful means of identifying and understanding gene-behavior interactions, but determining which observations from mouse generalize to human remains challenging. Mice deleted for nuclear receptor 2E1 (Nr2e1; known previously as Mtll, Tailless, Tll, Tlx) show nervous-system specific abnormalities including pathological aggression. It is unknown, however, whether human NR2E1 is similarly involved in the modulation of development and behavior. To test the ability of human NR2E1 to modulate behavior, we generated mice carrying human NR2E1 under the control of its endogenous regulatory elements and bred them to ‘fierce’ mutants that lack mouse Nr2e1. The presence of human NR2E1 eliminated structural brain defects and ameliorated eye abnormalities observed in fierce mutants. Excitingly, the behavior of fierce mice transgenic for human NR2E1 was indistinguishable from wild-type controls. That human NR2E1 can correct fierce brain-behavior abnormalities, supports conserved underlying mechanisms for behavior modulation, and suggests that variation at NR2E1 may contribute to human behavioral disorders. More generally, we have established an experimental paradigm in which to functionally evaluate the role of human genes in behavior. Use of this approach may permit the assessment of candidate psychiatric disease genes, or specific allelic variants, for which evidence is either lacking, or exists but is equivocal.

POSTER 164 - SENSITIVITY OF STRIATAL PROJECTION NEURONS TO MITOCHONDRIAL DYSFUNCTION

IN OMI DEFICIENT MND2 MICE

Jones JM 1, Tallaksen-Greene S 2, Albin RL 2, Meisler MH 1

1 University of Michigan, Department of Human Genetics, Ann Arbor, United States, 2 University of Michigan, Department of Neurology, Ann Arbor, United States

The Ser276Cys mutation in the mitochondrial protease OMI (Prss25) is responsible for neurodegeneration and muscle atrophy in the mnd2 mouse (Jones et al, Nature 2003).  Neuronal degeneration and gliosis are first detectable in the striatum, with other brain regions becoming involved later in disease progression (Rathke-Hartlieb et al, Expt. Neurol. 2002).  To identify the sensitive striatal neurons, brain sections from early stages of the disease were stained with antisera identifying striatal neuron subpopulations.  Normal numbers of  somatostatin-containing and cholinergic interneurons were detected.  A dramatic reduction in striatal projection neurons containing substance P and enkephalin was indicated by reduced staining within their target structures, the substantia nigra and globus pallidus, respectively. This is the same pattern of pathology found in Huntington's Disease (HD), consistent with prior suggestions of mitochondrial dysfunction in HD.  Ultrastructural studies of tissues from mnd2 mice reveal swollen and disorganized mitochondria in brain, cardiac muscle and skeletal muscle, confirming the biochemical evidence for mitochondrial dysfunction.  We are testing OMI as a candidate gene for human mitochondrial encephalomyopathies in which muscle and brain are both affected. 

POSTER 165 - CHARACTERIZATION OF NEW MICE MODELS OF MOTOR NEURON DISEASES

Achilli F 1, Nolan P 2, Fisher EMC 1

1 Institute of Neurology, London, United Kingdom, 2 MRC Mammalian Genetic Unit, Harwell, United Kingdom

The aim of this study is to characterize three mutant mice lines (GENA201, 202 and BHV/7) with motor deficits as assessed in the primary screen of the Harwell ENU Mutagenesis Programme. For the purpose of this motor neuron disorder study, we have adopted four parameters to assess the motor coordination and the balance of the three mouse strains The parameters measured include grip strength, rotarod performance, wire manoeuvre and weight from ages 1-12 months.

Following phenotyping at 5, 9, 15 weeks and 7 months, GENA 201 and 202 mutants showed poor grip strength. The results also showed no correlation between grip strength and rotarod performances and poor correlation between grip strength and wire manoeuvre. No significant weight loss was observed in the mice. General tremor was the only visible phenotype observed concurrent with poor grip strength.

On the other hand, BHV/7 mutants showed a very clear phenotype including retropulsion, hyperactivity and aggression. Due to their aggressive and hyperactive behaviour, it was very difficult to characterize their locomotor behaviour.BHV/7 mutants were significantly lighter than wild type but there was no statistical difference between mutants and wild type in the grip strength performance.

These results allow the scoring of mutant and wild type mice in order to map and subsequently clone the defective genes. As the BHV/7 mutants could be easily scored visibly due to their distinctive phenotype, a genome scan was performed.

Gaining access to new genes involved in the pathology of motor neurons will provide the opportunity to investigate the biology of MND so that it will be possible to identify and characterize novel genetic defects associated with motor MND.

POSTER 166 - FURTHER MAPPING OF THE MOUSE DEAFNESS MUTANT BRONX WALTZER

Taylor A 1, Cheong MA 2, Bussoli TJ 2, Kelly A 2, Steel KP 1

1 Wellcome Trust Sanger Institute, Cambridge, United Kingdom, 2 MRC Institute of Hearing Research, Nottingham, United Kingdom

bronx waltzer is an autosomal recessive mouse mutation causing abnormalities in the inner ear which result in mutant mice having deficiencies in both the auditory and vestibular systems. Homozygous mice exhibit hyperactivity, circling behaviour, head tossing and failure to respond to sound. Hearing loss in these mice is due to degeneration of the inner hair cells in the organ of Corti, while the vestibular phenotype is a result of sensory hair cell degeneration in the maculae and cristae. This phenotype is visible from E17.5, shortly after the hair cells differentiate, making bronx waltzer an interesting model for the understanding of the molecular basis of development and function of the inner ear as well as for hereditary deafness.

The mutation is being mapped using a backcross of 1085 mice to the inbred strain 101/H and has been localised to a 1.9Mb region of chromosome 5 between 113.5Mb and 115.4Mb which contains 49 genes in Ensembl Build32. There remain 17 backcross mice with recombinations within this interval but we have been unable to find new markers polymorphic for the two strains, a task made difficult because bronx waltzer arose spontaneously and is maintained on an unknown background. We have carried out sequence sampling of over 200 amplicons in areas where polymorphisms might be expected such as 3’UTRs, tandem repeats and reported SNPs and have found the sequences of the two strains to be remarkably similar in this region.

As a result of the difficulties encountered with refining the localisation of the mutation we are pursuing alternative approaches to identifying the gene responsible for the bronx waltzer phenotype. These include assessment of the candidacy of genes in the region by in situ hybridisation and screening for replication of the phenotype in zebrafish on injection of morpholinos designed to the fish orthologs of mouse candidates. We are also carrying out large scale exon resequencing of the genes in the region to identify the mutation at the sequence level.

POSTER 167 - NEW CANDIDATE GENES FOR WOBBLER, EXPRESSION PROFILING AND MODIFIED PHENOTYPE

Schmitt-John T, Drepper C, Hahn P, Mussmann A, Ulbrich M, Beckers J, Jockusch H

Univerity of Bielefeld, Bielefeld, Germany

The recessive wobbler (wr, Chr 11; phenotype WR) mutation of the mouse causes a spinal muscular atrophy (SMA), a progressive degeneration of motoneurons which secondarily leads to an atrophy of skeletal muscles. The wobbler mouse serves as an animal model for human SMAs and amyotrophic lateral sclerosis (ALS). In addition to the neurological effect the mutation causes a spermiogenesis defect.

Here we report on a further narrowing down of the wr-critical chromosomal interval and the detailed characterization of novel candidate genes for wobbler. A previously reported inversion [1] between mouse and man was not confirmed. In one of the characterized genes a single base mutation was identified, making it a hot candidate.

An extensive microarray based expression profiling of WR CNS mainly revealed secondary consequences of neurodegeneration, astro- and microgliosis, but also surprising similarities to expression alterations in the CNS of human Tay Sachs patients [2].

In crosses of the standard C57BL/6J stock with M. m. castaneus we observed a wide variation of the severity of the neurological phenotype. QTL mapping was performed and a major modifier locus (wrmod1) for the neurological WR phenotype was mapped to Chr 14 [3]. An M. m. castaneus-specific polymorphism within the middle neurofilament gene Nfm affecting a highly conserved KSP-phosphorylation site was detected. This might be responsible for the aggravated neurological WR* phenotype. An altered expression of the Nfm gene was observed in expression profiling of WR spinal cords. NF-M deposits were found in the pericarya of WR* neurons and axonal calibers were found to be reduced. We conclude that NF-M is involved in the wobbler pathogenesis.

Supported by DFG (SCHM1276/1) and NGFN (Xpress).

[1] Fuchs et al. (2002) BMC Genetics 3:14.

[2] Myerowitz et al. (2002) Human Molecular Genetics 11 : 1343-1350

[3] Ulbrich et al. (2002) NeuroReport 13: 535-539.

POSTER 168 - A NOVEL MOUSE MUTATION HEMORRHAGIC HYDROCEPHALUS (HHY) MAPS TO CHROMOSOME 12

Mori N 1, Kuwamura M 2, Kinoshita A 2, Okumoto M 1, Yamate J 2

1 Research Institute for Advanced Science & Technology, Osaka Prefecture University, Osaka, Japan, 2 Laboratory of Veterinary Pathology, Graduate School of Agriculture & Biological Sciences, Osaka Prefecture University, Osaka, Japan

A novel mouse mutation hemorrhagic hydrocephalus (hhy) in the BALB/c background was found at the Osaka Prefecture University in 2000. The hhy gene was inherited in an autosomal recessive manner and the hhy homozygous animals developed hydrocephalus with dilatation of lateral ventricles during post-natal and adult ages. Pathological examination detected mutants with mildly dilated ventricles among normal appearing siblings. Multiple hemorrhages were frequently seen in the meninges as well as throughout the brain parenchyma, but in other viscera of the animals affected by hydrocephalus. The hhy mutants showed neither macroscopic brain malformations nor causative stenoses of the ventricular system. Subcortical heterotopia, a sign of defective neuronal migration, was observed in all of the mutants. To assign the hhy locus, animals heterozygous for the hhy mutation (hhy/+) were mated with STS mice and the (hhy/+)F1 daughters were backcrossed to the hhy/+ males. Genome-wide screening using microsatellite markers identified the region carrying homozygous BALB/c allele on chromosome 12 in 12/12 (100%) affected animals arising in the crosse, which were pathologically diagnosed as hydrocephalus with ventricular dilatation. Linkage was confirmed in other cross. Our study showed that the hhy mutant might be a useful animal model for the study of neuronal abnormalities.

POSTER 169 - BONE MARROW DERIVED CELLS MEDIATE INFLAMMATORY ARTHRITIS PHENOTYPE IN ALI18 MUTANT MICE

Abe K 1, Wagner S 1, Kalaydjiev S 2, Jakob T 3, Franz TJ 2, Busch DH 2, Soewarto D 1, Fuchs H 1, Hrabé de Angelis M 1

1 Institute of Experimental Genetics, GSF National Research Center, Neuherberg, Germany, 2 Institute for Medical Microbiology, Immunology and Hygiene, Technical University Munich, Munich, Germany, 3 Division of Environmental Dermatology and Allergy, GSF National Research Center/Technical University Munich, Munich, Germany

Rheumatoid arthritis (RA) is a common human systemic autoimmune disease that leads to erosive destruction of diarthrodial joints.  Despite world-wide efforts, its etiology and pathogenesis are still incompletely understood.  Because of complex genetic background and environmental effects in humans, experimental animal models have the potential to accelerate dissection of molecular mechanisms leading to inflammatory arthritis.  From the Munich large scale ENU-mutagenesis screen, we have isolated a dominant mutation, Abnormal limb (Ali)18, that shows redness and swelling on the hind paws at adult stage.  Ali18 homozygous animals display very severe inflammation in the front and hind paws and tail: arthritis, dermatitis, vasculitis, and swelling. First, to elucidate how the immune system is involved in this systemic inflammation, we analyzed various immunological parameters by means of ELISA and flow cytometry. Although increased granulocyte populations and elevated levels of IgE were detected in peripheral blood of Ali18 homozygous mice, autoimmune markers such as rheumatoid factor and anti-DNA antibodies were  moderately upregulated. In lymph nodes and spleen of Ali18, increased populations of activated T lymphocytes (CD4+CD25+) were detected. Next, we used disease transfer systems to address what cell types contribute to the phenotypes. Bone marrow cells (BMCs) were injected intravenously to irradiated wild type animals. In contrast to control experiments using BMCs from wild type mice, BMCs from Ali18 mice could transfer the arthritis phenotypes. However, plasma from Ali18 mice failed to transfer the phenotypes. These results indicate that bone marrow derived cells, not  serum factors, play an important role in the inflammatory process. In addition, the dual energy x-ray absorptiometry radiograph (DEXA)  analysis showed a significant lower bone mineral density at the whole body of the Ali18 homozygous mice (p<0.0001). Bone marrow derived cells contribute to bone remodelling, although it is uncertain whether BMCs of Ali18 mice influence this phenotype directly or indirectly through inflammation. In patients with RA, generalised bone loss frequently occurs. Therefore, detailed analysis of bone loss in Ali18 mice provides an insight into mechanisms how systemic inflammation affects bone metabolism. The possible association of the Ali18 phenotypes in rheumatoid arthritis and/or seronegative

POSTER 170 - MOUSE WT SNCA PROTECTS AGAINST HUMAN A53T SNCA-INDUCED SYNUCLEINOPATHY; HUMAN WT SNCA DOES NOT PROTECT

Cabin DE, Nussbaum RL

NHGRI/NIH, Bethesda, United States

a -synuclein (SNCA) has been linked to Parkinson's disease (PD) by rare disease-causing mutations and a chromosomal triplication.  It is also a major constituent of Lewy bodies, neuronal inclusions found in sporadic PD.  Non-physiological overexpression of human A53T mutant SNCA in mouse spinal cord causes a synucleinopathy leading to limb paralysis and death.  We showed that this phenotype is exacerbated in mice that lack endogenous Snca, although the normal level of the mouse protein in spinal cord is low.  We wished to determine if human WT SNCA confers similar protection by crossing a human WT SNCA PAC transgene onto the line lacking endogenous Snca but carrying the mutant human transgene.  The hWT transgene driven by its own promoter shows an expression pattern in mouse brain and spinal cord quite similar to that of mouse Snca.  Doubly transgenic mice that lack mouse Snca are currently developing the spinal cord neuronopathy at the same rate seen in the singly transgenic line, indicating that human WT SNCA does not provide the protection afforded by the mouse protein.  Identical pathology develops, with Wallerian degeneration in ventral roots.  Intracellular SNCA inclusions similar to pale bodies are found in motor neuron cell bodies in both lines.  Thus at least some of the 6 amino acid differences between the mouse and human proteins (besides that at position 53) must confer a significant difference in their properties.  Microarray analysis is currently being performed to better understand how the mutant human SNCA causes the spinal cord synucleinopathy.

POSTER 171 - REDUCED STARTLE RESPONSE AND TREMOR IN THE NEW MOUSE MUTANT CLOTH-EARS

Mackenzie FE 1, Parkinson NJ 2, Brooker D 1, Polley S 1, Glenister P 1, Fray M 1,Brown S D M 1

1 MRC Mammalian Genetics Unit, Harwell, United Kingdom, 2 MRC Prion Unit, London, United Kingdom

Hearing loss is the most common sensory defect in humans and affects one in two thousand children from birth.  A further one in two thousand will be affected by deafness before adulthood. Approximately fifty percent of all deafness cases are thought to have a genetic cause. In humans over seventy non-syndromic deafness loci have been mapped, but the majority remain to be cloned. The UK ENU (N-ethyl-N-nitrosourea) mouse mutagenesis programme has generated numerous deaf mouse mutants, of which the underlying genes for several have been identified by positional cloning. Hearing loss in these mutants is defined by a reduced or absent Preyer reflex (ear flick) and startle response on exposure to a 90dB SPL, 20kHz single tone burst generated by a clickbox. The Cloth-ears mouse mutant displays autosomal dominant reduced startle response from 1 month and autosomal semi-dominant tremor. High resolution genetic mapping places the Cloth-ears mutation within a 2.1Mb region on distal chromosome 15. We present genotypic and phenotypic data for this novel mutant line, and show that the Cloth-ears mouse is a useful model for dissecting the genetic pathways involved in sensory loss and neurological defects.

POSTER 172 - GASTRIN REGULATION OF SONIC HEDGEHOG PROCESSING CORRELATES WITH MAINTENANCE OF THE GASTRIC MUCOSAL ARCHITECTURE

Zavros Y 1, Eaton K 1, Low MJ 2, Samuelson LC 1, Merchant JL 1

1 University of Michigan, Ann Arbor, United States, 2 Vollum Institute, Oregon Health and Science University, Portland, United States

Gastrin-deficient (G-/-) mice develop severe metaplastic changes in the gastric mucosa in response to hypochlorhydria, bacterial overgrowth and inflammation.  Gastrin is a known trophic hormone for the gastric epithelium.  Sonic hedgehog (Shh), a morphogenic peptide produced by murine parietal and chief cells maintains normal gastric mucosal architecture, but the mechanism is unknown.  The present study investigates the role of gastrin in the regulation of Shh protein expression and processing using the G-/- mouse model.  Compared to 12 month old wild type (WT) and hypergastrinemic somatostatin-deficient (SOM-/-) mice, the hypochlorhydric G-/- mice developed parietal cell atrophy, antral inflammation and metaplasia.  60% of the G-/- mice developed gastric tumors that showed loss of the tumor suppressor gene RUNX3 in isolated tissue.  To confirm that the tumors in the G-/- mice were malignant, cells dissociated from tumors alone grew in soft agar.  Immunofluorescence revealed co-localization of Shh with the parietal cells in the WT animals.  However, Shh expression was decreased in the G-/- mice coincident with loss of parietal cells.  Gastrin-infusion in G-/- mice restored Shh peptide in the parietal cells and increased Shh processing from the 45kDa precursor to the 19kDa active peptide.  Isolated G-/- mouse gastric cells treated with gastrin showed a decrease in the 45kDa precursor in whole cell extract and an increase in the secreted 19kDa Shh peptide.  Therefore, Shh protein expressed in the mouse parietal cells is lost in the absence of gastrin due to reduced expression and peptide processing and this is important for maintenance of normal mucosal architecture.

POSTER 173 - JOINT QTL ANALYSIS FOR MAMMARY TUMOR SUSCEPTIBILITY AND OBESITY

Gordon RR 1, Pomp D 1, Hunter KW 2

1 University of Nebraska, Lincoln, United States, 2 NCI/NIH, Bethesda, United States

Genetic predisposition to complex traits results from interactive combinations of relatively small effects of genetic variations within a large number of primarily unidentified polygenes, known as quantitative trait loci (QTL).  Given the growing evidence that obesity may increase the risk of susceptibility to certain forms of cancer, we have created a population of F2 mice cosegregating obesity QTL and the MMTV-PyMT transgene, with a primary objective of determining how polygenic obesity influences age at mammary tumor onset, tumor number and severity.  The F2 population (n=600) resulted from a cross between the polygenic obesity model M16i and FVB/NJ-TgN (MMTV-PyMT)634Mul, which develop synchronously appearing multifocal tumors involving all of the mammary glands with more than 85% of the animals developing pulmonary metastases by 100 days of age.  At four weeks of age all F2 individuals were randomly assigned to either a high fat (45% calories from fat) or a matched normal fat (10% calories from fat) synthetic diet.  Body weights are being recorded at periodic intervals, while body composition will be analyzed at pre- and post-tumor ages using DEXA. All F2 mice will be palpated for tumors at regular intervals to determine age of onset, and will be sacrificed at 100 days of age to determine tumor number and extent of pulmonary metastasis.  Genotyping of a genome-wide panel of microsatellite markers across the F2 population will facilitate a joint QTL analysis for obesity and cancer polygenic and diet-induced phenotypes.

POSTER 174 -  A LOSS-OF-FUNCTION MUTATION IN NATRIURETIC PEPTIDE RECEPTOR 2 (NPR2) GENE IS RESPONSIBLE FOR MOUSE ACHONDROPLASIA

Tsuji T, Kunieda T

Okayama University, Okayama, Japan

The achondroplastic (cn/cn) mouse is a spontaneous mutant found in the AKR/J colony at the Jackson Laboratory, which is characterized by an achondroplastic dwarfism with short limbs and tail due to disturbed chondrogenesis in the endochondral ossification. The cn locus has been mapped to mouse chromosome 4, but precise localization has not yet been determined. In this study, we preciously mapped the cn locus by linkage analysis and identified the causative gene for the cn/cn mutant. By linkage analysis using 115 affected mice of F2 progeny, the cn locus was mapped on a 0.8-cM region of chromosome 4. We found that Npr2 gene localized in this region was most potent candidate gene. The sequence analysis of the gene revealed a T to G transversion leading to an amino acid substitution of highly conserved Leu to Arg in the guanylyl cyclase catalytic domain of NPR2. Npr2 gene encodes a receptor for C-type natriuretic peptide (CNP), which positively regulates longitudinal bone growth by intracellular cGMP production. No significant increase of intracellular cGMP level in response to CNP stimulus was observed in the chondrocytes of the cn/cn mice, while it was evident in the chondrocyte of the wild-type mice. Similar results were also obtained by transfection of mutant or wild-type Npr2 gene in COS-7 cells. These results indicate that the achondroplasia of cn/cn mouse is caused by a loss-of-function mutation of the Npr2 gene.

POSTER 175 - DIETARY RESCUE AND POSITIONAL CLONING OF PEQUEÑO: A SECOND LOCUS FOR SHWACHMAN DIAMOND SYNDROME?

Loftus SK, Cannons J, Incao A, Chen A, Zerfas P, Bryant MA, Schwartzberg P, Biesecker LG, Pavan WJ

1 National Human Genome Research Institute, NIH, Bethesda, United States, 2 Diagnostic and Research Services Branch, Veterinary Resources Program, NIH, Bethesda, United States

Pequeño is inherited in an autosomal recessive pattern and is a mouse model for exocrine pancreatic insufficiency.  The pequeño phenotype is caused by a transgene insertion and 210kb deletion mutation on chromosome 3.  Pequeño mice exhibit progressive postnatal apoptosis of pancreatic acinar cells with nearly complete exocrine acinar cell loss at 8 weeks, while islets and ductal tissue remain present.  Homozygous animals are about 1/3 smaller than heterozygous or wildtype littermates at weaning, have an average lifespan of 4 months under standard housing conditions and have immunodeficiency.  Dietary supplementation pancreatic enzymes can correct the body size, weight, immunodeficiency, and increase the lifespan of pq/pq mice, suggesting that malnutrition is a major cause of the phenotype.  BAC transgenic rescue crosses have identified a single transcript present in the deletion that is responsible for the pancreatic insufficiency.  The pequeño phenotype is similar to that of Shwachman Diamond syndrome (SDS) in humans. SDS is characterized by short stature, pancreatic lipomatosis, neutropenia, and increased risk of myelodysplastic syndrome with progression to acute myeloid leukemia.  The major genetic locus for SDS has recently been identified (Boocock et al., 2003) and is distinct from that of the pequeño locus.  As only 89% of patients with SDS have mutations at the SDS locus, we are currently testing if the pequeño locus may contribute to the genetic heterogeneity found in humans

POSTER 176 - DARK SKIN MOUSE MUTANTS REVEAL NEW GENES INVOLVED IN PIGMENTATION

McGowan K., Van Raamsdonk C, Aradhya S, Fitch K, Barsh G

Stanford University, Stanford, United States

Melanocytes undergo a series of stereotypic behaviors as they move from the neural crest to their final resting position in the skin, including migration, proliferation, programmed cell death and differentiation.  Recently, large-scale mutagenesis projects in mice have enabled our lab to use forward genetics to study a novel class of pigmentation mutants identified by dark skin (Dsk). One class of Dsk mutants, represented by Dsk3, Dsk4, Dsk6, and Dsk11 is characterized by increased epidermal pigment in non-hairy skin.  Each mutant can be identified at weaning age. While Dsk3, Dsk4 and Dsk6 homozygotes die early in embryogenesis, Dsk11 homozygotes expire within minutes of birth.  In addition, Dsk11 homozygotes are born with an open eye phenotype, and exhibit gross and histologic abnormalities consistent with pulmonary hypoplasia. Using melanocyte-specific markers, we have shown that Dsk3 animals have an increased number of epidermal melanocytes in the tail and footpad skin, whereas Dsk4, Dsk6 and Dsk11 animals have dark skin based on mechanisms independent of pigment cell number.  Each mutant maps to a unique location in the genome, and sequence variants have recently been identified in Dsk6 and Dsk11.  This group of dark skin mutants may shed light on the mechanisms that control normal variation in human skin color, a biologic process that is mediated by changes in the activity of epidermal melanocytes. 

POSTER 177 - BONE DEFECTS IN DYSMORPHOLOGY MUTANT MICE FROM THE MUNICH ENU-MUTAGENESIS SCREEN

Abe K, Grundner-Culemann E, Gailus-Durner V, Fuchs H, Hrabé de Angelis M

Institute of Experimental Genetics, GSF National Research Center, Neuherberg, Germany

Bone remodelling is a dynamic process in which new bone is added and old bone is removed.  In humans, the bone formation dominates the bone resorption until the age of twenties when the bone mass reaches a maximum.  The bone mass remains stable for years, and starts to decrease after a certain age. Balance between formation and resorption is regulated by multiple complex systems which include growth factors, hormones, and vitamins. Although defects in bone remodeling in humans such as osteoporosis cause deterioration of quality of life in old age, its mechanisms remain incompletely understood. In a large scale mouse ENU-mutagenesis project, we use 49 visible morphological parameters for mutant screening. So far, 260 dysmorphological mutant lines have been isolated and confirmed. To identify mouse models for human bone diseases, we have started a secondary screening of existing mutant lines in the dysmorphology module of the German Mouse Clinic. Additional bone-related parameters include x-ray radiography, dual-energy x-ray absorptiometry (DEXA), micro-computer tomography, and physiological analysis. We selected mutant lines showing bone-related phenotypes for this screening. One mutant line, Ali14 (abnormal limb) mutant mice showed decreased bone mass density (p<0.001), weight (p<0.001), fat and lean mass (p<0.001), and body size (p<0.001) were detected. Ali14 is a dominant mutation, and only mutant males exhibit swelling on  hind feet and spotted ears around 4 months of age. Homozygous animals have not been detected. By histological examination, inflammatory infiltrations were detected in joints, skin, and bone marrow of hind feet in Ali14 mice. These results suggest that Ali14 mutant mice could be a mouse model for generalized bone loss in inflammatory arthritis. Further detailed analysis using micro-computed tomography for Ali14 mice is underway. We will present a recent update of bone-related screening, and the possible involvement of detected phenotypes in human bone disease will be discussed.

POSTER 178 - TAKE OUR MICE – PLEASE!  ENU-DERIVED MUTANT STRAINS NOW AVAILABLE FROM THE JAX PGA

Svenson KL, Peters LL

The Jackson Laboratory, Bar Harbor, United States

The Center for New Mouse Models of Heart, Lung, Blood and Sleep Disorders was established in 2000 as a National Heart, Lung, and Blood Institute Program for Genomic Applications.  Our goal is to provide the scientific community with new mouse models in these disease areas. We are conducting high-throughput primary phenotyping of third-generation ENU-mutagenized pedigrees derived from background strain C57BL/6J.  A battery of non-invasive phenotypic screens is performed on mature mice over the course of a 9-week protocol, and includes challenge with a high fat, high cholesterol diet.  Testing of each animal includes the following: plasma coagulation factor profile (fibrinogen, antithrombin III, Factor VII); standard whole blood count; plasma total- and HDL-cholesterol, triglycerides, and glucose; metabolic cage monitoring, including a novel method for assessment of sleep behavior; whole body plethysmography; dual energy x-ray absorptiometry for obesity; and evaluation of cardiac function by echocardiography and unanesthetized electrocardiography.  To date, we have identified over 350 phenotypic deviants, of which more than 60 have proven heritable.  Heritability testing and colony development of more than 100 additional lines is ongoing for all phenotypes.  Breeding pairs of all proven heritable mutants are freely available to the scientific public from our web site http://pga.jax.org.  We currently have 23 mutant lines available.  These are:  14 hypertensive; 3 obese; 4 blood disorders; 2 lipidopathies, one of which has severe atherosclerosis. We urge the scientific public to consider taking these established mutants into their laboratories for further characterization and use as they study these important human diseases.

POSTER 179 - CHRONIC LOW-LEVEL EXPOSURE TO PESTICIDES DURING DEVELOPMENT: ROLE OF PARAOXONASE (PON1)

Pettan-Brewer C, Cole TB, Walter BJ, Fisher JC, Richter RJ, Burbacher TM, Costa LG, Furlong CE

University of Washington, Seattle, United States

Paraoxonase (PON1) is an HDL-associated enzyme involved in the metabolism of organophosphorous pesticides. We demonstrate that infants have very low PON1 levels and a variable onset of expression, reaching a plateau at 6-24 mo. To assess the importance of PON1 for protecting against the developmental toxicity of chlorpyrifos oxon (CPO), PON1 knockout (PON1-/-) and wild-type (WT) mice were exposed chronically (PN4 to PN21) to low levels of CPO. Endpoints included cholinesterase activity, histopathology, gene expression, and behavior. As early as PN4, PON1-/- mice were more sensitive to inhibition of brain cholinesterase by CPO. At PN22, and persisting as long as 4 months, chronic developmental exposure to 0.18 mg/kg/d or 0.25 mg/kg/d CPO resulted in perinuclear vacuolization of cells in a discrete area of the neocortex, with an increase in the number of affected cells at 0.25 mg/kg/d. Surprisingly, none of the behaviors chosen for assessment were affected. However, from ~PN16-19, mice exposed to 0.18 or 0.25 mg/kg/d exhibited striking hyperactivity immediately following CPO administration. These studies revealed toxicity associated with low-level OP exposure and indicate that children less than 2 yr old represent a particularly susceptible population for OP exposure. Supported by T32 AG00057, ES09601/EPA-R826886, ES09883, ES04696, ES07033, ES11387.

POSTER 180 - MOUSE MUTANTS WITH HYPERACTIVITY: RECENT PROGRESS IN RIKEN ENU-MUTAGENESIS PROJECT

Wada Y 1, Masuya H 1, Kushida T 1, Kawai A 1, Nishii R 1, Miura I 1, Furuse T 1, Kobayashi K 1, Kaneda H 1, Suzuki T 1, Minowa O 1, Gondo Y 2, Noda T 1, Wakana S 1, Shiroishi T 1

1 Mouse Functional Genomics Research Group, RIKEN GSC, Tsukuba, Japan, 2 2Population and Quantitative Genomics Team, RIKEN GSC, Yokohama, Japan

Recent studies have demonstrated that psychiatric diseases such as schizophrenia and bipolar disorder have some genetic basis. In human populations, however, identification of genes underlying psychiatric disorders is still difficult because of polygenic inheritance and gene-environment interactions. Genetic studies using mouse models have great advantage for identification of molecular pathways of behavior. In order to establish mouse mutants modeling symptoms of psychiatric disorders, we have initiated dominant behavioral screening in RIKEN ENU-mutagenesis project. One of our focuses was “increased locomotor activity” in a novel and/or familiar environment, because hyperactivity in mice has been associated with behavioral symptoms of several human psychiatric disorders; e.g., hyperactivity in attention-deficit hyperactivity disorder (ADHD), and psychomotor agitation in schizophrenia and mania. We have screened about 1,500 G1 animals (DBA/2J x mutagenized C57BL/6J) for home-cage activity, and 2,000 G1s for open-field activity. Thirty phenodeviants with hyperactivity (home-cage 8; open-field 22) were detected and mated with DBA/2J for inheritance testing. Heritability of the seven lines (home-cage 1; open-field 6) was confirmed. They were hyperactive only in either home-cage or open-field, but two open-field mutant lines, M100073 and M100174, showed hyperactivity both in home-cage and open-field. Percentages of hyperactive offspring in N2 progenies were 20-50%, but in the N3 progenies, the percentages were decreased to 3-8 % except in M100073 and M100174. The responsible genes of M100073 and M100174 were mapped to Chr.6 and Chr.2, and fine mapping is underway. We will report the recent progress of gene mapping and behavioral characterization focusing on ADHD and schizophrenia.

POSTER 181 - THE PHENOTYPE OF AKT1, AKT2 AND AKT3 SINGLE AND DOUBLE KNOCKOUT MICE

Chen W, Hay N

1 University of Illinois at Chicago, Chicago, United Kingdom, 2 University of Illinois at Chicago, Chicago, United States

Akt, also known as protein kinase B (PKB), belongs to a serine/threonine kinase subfamily.  So far, three main isoforms of Akt have been identified in mammalian cells: Akt1, Akt2 and Akt3 encoded by 3 separate genes.  Studies have shown that Akt in cells is activated by diverse stimuli, such as hormones, growth factors and extracellular matrix components, and is a downstream target for phosphoinositide 3-kinase (PI3K).  It appears that Akt lies at the crossroads of multiple cellular signaling pathways, since activated Akt phosphorylates and regulates many cellular proteins involved in processes that include apoptosis, glucose metabolism, proliferation, transcription and angiogenesis.  In order to elucidate the physiological role of Akt1, Akt2 and Akt3, we use knockout mice as the model to define the function of Akt1, Akt2 and Akt3.  We will describe the phenotypes of individual and compound knockouts of the 3 Akt isoforms.

POSTER 182 - MODELS OF HUMAN CANCER AT THE CLICK OF A MOUSE: THE MOUSE TUMOR BIOLOGY (MTB) DATABASE

Bult CJ, Krupke DM, Allio T, Vincent M, Sundberg JP, Mikaelian I, Eppig JT

The Jackson Laboratory, Bar Harbor, United States

The Mouse Tumor Biology Database (http://www.tumor.informatics.jax.org) is a freely accessible on-line informatics resource designed to support the use of the mouse as a model system of hereditary and induced cancers. The database was designed around the principle that genetic background is a key factor influencing the kinds and onset of cancers observed in different strains of genetically defined mice. MTB provides cancer genetics researchers with access to data regarding various mouse models for cancer including such information as tumor names and classifications, tumor incidence and latency data in different strains of mice, tumor pathology reports and images, information on genetic factors associated with tumors and tumor development, and literature references.

MTB provides users with a powerful web-base query interface that supports complex ad hoc queries. For example, users can query the database with questions such as "Which transgenic strains of mice have a high incidence of lung adenocarcinomas?" or "Show me tumor records where a point mutation was detected in Kras2" or "What spontaneous tumors are found in the inbred strain A/J?". The key to enabling these sorts of queries in MTB is the integration of mouse tumor data from many sources, primarily the published literature, and the use of standardized nomenclature and controlled vocabularies.Terms for classifying tumors, organ and tissue names, strain type, reproductive status, and mutation type have been standardized in MTB annotations. This allows consistent searching within MTB and provides the ability to compare and link MTB records with those in other resources such as the Mouse Models of Human Cancer Consoritum (http://emice.nci.nih.gov). Nomenclature for genes, alleles, and strains is integrated with data held in other Mouse Genome Informatics resources (http://www.informatics.jax.org), so that users have access, via hypertext links, to a wide range of additional genetic and phenotypic information.

MTB is supported by NCI grant CA89713.

POSTER 183 - NATIONAL INSTITUTE OF ENVIRONMENTAL HEALTH SCIENCES – COMPARATIVE MOUSE GENOMICS CENTERS CONSORTIUM: TO IMPROVE OUR UNDERSTANDING OF THE BIOLOGICAL SIGNIFICANCE OF HUMAN POLYMORPHISMS

Packenham J

National Institute of Environmental Health Sciences, Research Triangle Park, North Carolina, United States

The National Institute of Environmental Health Sciences (NIEHS) Comparative Mouse Genomic Centers Consortium (CMGCC) is a cross-disciplinary, multi-Institutional, program that falls under the auspices of the NIEHS Environmental Genome Project. This program was initiated in 2001 with the goal of developing transgenic and knockout mouse models based on human DNA sequence variants in the environmentally responsive genes discovered under phase I of the EGP.  Ultimately, these models will be used as tools to improve our understanding of the biological significance of human DNA polymorphisms and the role of such variation in environmentally related diseases.  Initially, this program has focused on DNA polymorphisms that occur in DNA repair and cell cycle environmentally responsive genes. 

The Consortium is accomplishing its goals through, mouse model development, technology development, and resources developed by the Consortium.  To date the Consortium has over 32 mouse models under development and has developed through its bioinformatics team a Mouse Federated database, containing three major components: (1) mouse phenotypic assessment that allows for functional prediction of gene variants to model, (2) collection and analysis of phenotypic data including gene expression data, and (3) mouse model dissemination.  Through this database, SNPs are analyzed for their functional impact using SNP location and Haplotype frequency data, structural analysis and prediction, literature mining and potential combinatorial effects through pathway data.  Models are generated and analyzed for pathology and other phenotypic endpoints under various environmental conditions.  This combined knowledge is then integrated to generate knowledge of the human gene in the perturbed mouse background and map new information onto existing knowledge and finally, the experimental data is used to refine and adapt models in order to validate causality pathways. Validated mouse models and resources developed within the CMGCC are available to the general scientific community without charge.  These resources can be found at: http://www.niehs.nih.go

POSTER 184 - FUNCTIONAL ANALYSIS OF THE HYALOID MICROCIRCULATION IN A MOUSE MODEL OF RETINOBLASTOMA

Duckett AS, Leamen L, Cucevic V, Foster FS

Sunnybrook and Womens College Health Sciences Centre,  University of Toronto, Toronto, Canada

High resolution micro imaging tools have enabled non invasive visualization of tissue remodelling in longitudinal studies of the mouse. In mice, the hyaloid vasculature of the eye degenerates from a well defined structure at birth via progressive loss of branching structure.  Hyaloid regression is accompanied through a complex and poorly understood process by retinal vascular maturation.  Until the present it has not been possible to study ocular vascular patterning noninvasively.  Recently developed high frequency ultrasound microimaging (UBM) is well suited for the study of this process because the vascular networks of eye are comparativley superficial and are in juxtaposition to the echolucent tissues of the vitreous and lens. Flow is easily measured in the hyaloid artery (HA), vasa hyaloidea propria (VHP), and tunica vasculosa lentis (TVL), and retina.  In this study, flow waveforms at all 4 locations were measured in normal CD-1 and LHbTAg (retinoblastoma) mice over the period from birth  to 4.5 months.  Ultrasound imaging and Doppler were performed with a Vevo 660 (VisualSonics,Toronto) mouse scanner. Contrast enhanced MicroCT was used as a gold standard for vessel structure and was performed with an MS8 scanner (EVS (now  General Electric), London Ontario).  All imaging was performed under an approved animal protocol.  Observations of normal mice showed marked remodeling of the hyaloid vasculature over the first 16 days.  Peak flow velocities in the hyaloid artery dropped from 3.0 ± 1.0 cm/s to 0 between P0 and P13 while peak velocities in the VHP and TVL dropped from 1.25 ± .25 cm/s to 0 from P0 to P13.  No flow or evidence of a hyaloid vasculature was observed in the control mice after P13. Flow in the retinal vasculature showed an inversely proportional increase from 0.3 ± .05 cm/s to 4.1 ± 0.6 cm/s over the same time period.  In the retinoblastoma mice intraocular tumours were visible at 2 months and by 4.5 months had filled the vitreous cavity.  The blood supply for these tumours is demonstrated to be derived from a persistent and significantly augmented hyaloid vasculature. Conclusion:  Functional flow patterns of the hyaloid vasculature is result from a sensitive balance of angiogenic and apoptotic factors and appear to be inversely proportional to flow in the retinal circulation.  Flow in the hyaloid circulation does not normally persist beyond P13. Ocular tumours in a spontaneous model of retinoblastoma appear to recruit and expand the hyaloid vasculature to sustain an aggressive growth pattern.

POSTER 185 - 3D LASER IMAGING OF MOUSE CRANIOFACIAL MUTANTS ISOLATED AFTER ENU MUTAGENESIS

Keener VL, Garcia M, Dawson B, Bacino C, Lee B, Justice MJ

Baylor College of Medicine, Houston, United States

The mouse is a powerful animal model for studying diseases and defects analogous to those in humans. By using a three-generation mutagenesis screen after treatment of male mice with N-ethyl-N-nitrosourea (ENU), we isolate many strains of mutant mice,  ~10% of which have craniofacial abnormalities.  Craniofacial phenotypes can be caused by perturbations of many developmental processes, including neural crest cell migration, osteoblast differentiation, tooth and palate formation, and nasal and eye morphogenesis. To rapidly classify and prioritize these mutants, we have developed a novel method for characterizing mouse craniofacial mutations.  Our protocol combines observation with Cyberware 3-dimensional surface laser scanning and imaging, coupled with morphometric analysis and Euclidean Distance Matrix Analysis (EDMA). Further classification of mouse mutants includes radiograph studies, pathology, and skeleton preparation.

Of particular interest for our initial studies are mutant lines that demonstrate:  1) a possible CAP Syndrome model with severe craniosynostosis, as well as frontal and parietal foramina, 2) a patterning defect with vertebral and clavicle abnormalities, 3) asymmetric facial development, and 4) complete transformation to a different facial shape.

We have tested software for morphometric analysis of our mouse mutants, and are working on further refinement of our protocol.  Statistical analysis performed on many other lines has identified facial measurements that are abnormal compared with wild type mice. We are able to prioritize our mutants quickly and easily, and rapidly generate morphological data.  Our studies will help to determine parameters that are important for analyzing craniofacial dimensions over time in comparative mouse-human syndromes.

POSTER 186 - MOUSE MODELS OF HUMAN DISEASE:  THE INDUCED MUTANT RESOURCE AND MOUSE REPOSITORY AT THE JACKSON LABORATORY

Lane DB, Lutz CM, Rockwood SF, Linder CC, Davisson MT

The Jackson Laboratory, Bar Harbor, United States

Targeted mutation and transgenic mutant mouse models serve as invaluable research tools to study human disease.  Recent advances in mouse genome manipulation and the information generated by the human and mouse genome projects have led to increased development and use of induced mutant mouse models.  As the number of mutant mouse strains grows, cooperative accessible resources for these research tools are needed. 

The Induced Mutant Resource (IMR) at The Jackson Laboratory imports, preserves and distributes mouse strains important for biomedical research.  The IMR serves as a resource through which researchers can provide their strains to the rest of the scientific community.  Cryopreservation of each IMR strain ensures protection of genetic integrity and its continued availability in the future.  An online database of curated strain information is maintained as a resource for researchers. Mutant strain records in the IMR database include phenotype descriptions, transgene or targeted mutation information, strain development and maintenance procedures, references, and links to other relevant online resources. Currently the IMR contains over 1000 accepted strains and adds approximately 5-6 strains each month. Over 130,000 mice are distributed each year from the IMR.  Newly acquired strains include Alzheimer's Disease related models, mouse models for Spinal Muscular Atrophy and a conditional mutation strain for multiple endocrine neoplasia type 1.

Researchers can submit candidate strains directly through forms available on The Jackson Laboratory website at: http://www.jax.org/resources/submit_strain.html.

The IMR is part of a centralized repository that also includes spontaneous mutation strains and strains valuable for genetic analysis of complex traits.

POSTER 187 - IMPROVING RESEARCH AND REDUCING ANIMAL POPULATIONS THROUGH TECHNOLOGY

Drummond A, Ferreira A, Ferreira J

Auckland University, Auckland, Netherlands

Improving disease research capability while reducing the number of laboratory animals used in testing and is not only possible; the two aims are synergistic. This paper evaluates how post-genomic Information Technology supports more cost-effective high-quality wet-lab research while reducing use of animals to manageable proportions. The paper demonstrates how this is achieved through tightening the loops between analysis and animal management, reducing “schedule 1” waste through process management, pre-emptive supply/demand analysis and reducing redundancy through secure collaboration. It also investigates the regulatory shifts that would allow such post-genomic technology to achieve maximum effectivity.

POSTER 188- STUDYING THE INDUCTION OF MAMMARY TUMOR FORMATION IN TRANSGENIC MICE EXPRESSING PHOSPHO-MUTANT BETA-CATENIN

Kockeritz LK 1, Wood GA 2, Done SJ 3, Woodgett JR 1

1 Department of Medical Biophysics, University of Toronto, Toronto, Canada, 2 Department of Pathobiology, University of Guelph, Guelph, Canada, 3 Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, Canada

The Wnt signaling pathway results in the activation of beta-catenin, a transactivator that works with members of the Tcf/LEF family of transcription factors to drive gene transcription.  Negative regulators of beta-catenin are frequently mutated in human cancer, resulting in inappropriate beta-catenin activation and subsequent gene expression of various downstream targets including many oncogenes.  Activating mutations of beta-catenin itself has also been found in a variety of human tumors. Here we illustrate that a phosphomutant version, typical of what is seen in human cancers, can induce mammary tumor formation in mice.

The development of mammary tumors in our MMTV-beta-catenin phosphomutant (S33, 37, 41, 45A) occurs at a higher frequency in multiparous females than in nulliparous mice, suggesting hormonal influences are at play. Considering that the background strain of mice, FVB, tend to develop prolactin-secreting pituitary adenomas that induce lobuloalveolar development typically seen in pregnant females and that 92% of tumor-bearing females exhibit moderate to extreme lactational changes in non-tumor bearing mammary glands, we are currently investigating a link between prolactin and beta-catenin signaling.  We hypothesize that beta-catenin signaling requires the increase in cellular proliferation provided by prolactin for tumors to develop.  By using microarray and immunohistochemical analyses of tumors developed in our model we hope to better understand the molecular mechanisms of aberrant beta-catenin signaling in cancer development.

POSTER 189 - THE ROLE OF CALPAIN 3 IN MDM MOUSE MUSCULAR DYSTROPHY

Huebsch KA 1, Spencer MJ 2, Cox GA 1

1 The Jackson Laboratory, Bar Harbor, United States, 2 Department of Pediatrics and UCLA Duchenne Muscular Dystrophy Research Center, University of California, Los Angeles, United States

Muscular dystrophy with myositis (mdm) mice exhibit a severe and progressive muscular dystrophy with kyphosis and a rigid gait accompanied by a reduction in calpain 3 (CAPN3) levels in skeletal muscle.  The mdm mutation is a small in-frame deletion within the CAPN3 binding N2A domain of the titin gene.  We are using a genetic approach to test two alternate hypotheses regarding the role of CAPN3 in the mdm disease mechanism.  First, transgenic overexpression of CAPN3 in mdm mice will test the hypothesis that reduction of CAPN3 protein levels are critical to the disease process. Second, crosses of mdm mice with CAPN3-/- null mice will test the hypothesis that aberrant activation of CAPN3 causes the muscular dystrophy in mdm mutant mice. If CAPN3 is critical to the mdm disease process, we expect that either overexpression of CAPN3 or the complete absence of CAPN3 will alter the onset or progression of the mdm muscular dystrophy. Additionally, adding back the CAPN3 binding site in mdm mice by transgenic overexpression of the N2A domain of TTN will determine whether it can function independently to alleviate the mdm muscular dystrophy. 

POSTER 190 - GENCYCLOPEDIA: A COMPREHENSIVE ANIMAL RESEARCH TOOL FOR ENU-MUTAGENESIS AND LARGE SCALE PHENOTYPING

Laufs J 1, Kurzmann M 1, Schöniger M 1, Zinser C 1, Gnodtke A 1, Kukulies M 2, Sedlmeier R 1, Wattler S 1, Nehls M 1

1 Ingenium Pharmaceuticals, Martinsried, Germany, 2 Pieris Proteolab AG, Freising-Weihenstephan, Germany

Controlled breeding of animals for scientific research is a logistic challenge. For Ingenium Pharmaceuticals, which generates hundreds of genetic rodent lines using ENU-mutagenesis and performs complex phenotyping on thousands of mice both for in-house research and also for collaborations, the efficient management of the animal facility is essential.

With GENcyclopedia, Ingenium has developed an electronic platform that supports all the comprehensive aspects of animal management and scientific research.

By supporting the manifold requirements of the different departments and research groups, including project management, animal handling, phenotyping and genotyping, all data are tightly linked and controlled by an accurate rules engine.

As one example, project management is supported by generation of phenotyping schedules, resulting in daily work-lists for animal facility tasks. The execution is monitored by feedback loops.

All activities are controlled and monitored by the middle tier engine according to genetic, administrative and animal-welfare rules. Reporting tools and pedigree-viewer are further benefits. In addition, GENcyclopedia is central for tracking HTS mutation detection of Ingenium’s successful INGENOtyping platform.

As a cutting edge multi-tier J2EE system, GENcyclopedia allows different kinds of clients to be connected to the core rules system, which guarantees data consistency on the backend Oracle RDBMS. For multi-user access, permission is controlled on user or group level, both on performing actions and visibility of data objects. The framework based on an accurate data model keeps the system flexible for customizations and extensions in response to the fast changing needs of rapid and efficient research for Ingenium and its partners.

POSTER 191 - EMERILLON THERAPEUTICS’ ACB/BCA RECOMBINANT CONGENIC STRAINS: GENE MAPPING AND DISSECTION OF COMPLEX TRAITS.

Fortin A, Skamene E

Emerillon Therapeutics Inc., Montreal, Canada

The AcB/BcA gene discovery platform is a series of 36 Recombinant Congenic Strains, which were produced from the second backcross generation of the two progenitor mouse strains A/J and C57BL/6J. Each individual inbred RCS carries 12.5% of the donor genome in 87.5% of the background genome.  As the two parental inbred strains A/J and C57BL/6J are known to vary in the expression of resistance and susceptibility to a considerable number of mouse models of important human diseases, the AcB/BcA RCS platform represents a valuable and versatile genetic tool to study many different phenotypes.

The AcB/BcA RCS platform was used by our group and others to study a series of complex phenotypes including malaria infection, lipid metabolism, nociception and osteoporosis. In each case, the survey of the RC strain panel led to the identification of informative “pheno-deviant” strains, which exhibit a phenotype significantly different from the background progenitor. These strains were selected for further gene identification studies, including linkage mapping in secondary crosses and gene expression analyses in targeted organs. The development of a central database integrating all phenotypic, genetic and expression information produced from the AcB/BcA RCS platform is currently ongoing. This bioinformatics tool, which is linked to public databases, can be used to examine phenotypic data, look for correlation between phenotypes or sub-phenotypes, and investigate candidate genes underlying QTLs.