Michael F. Seldin - US grants

This proposal is designed to refine the understanding of gene linkage relationships in the mouse and allow precise comparative mapping analyses with the human genome. DNA obtained from a large series of [(C3H/HeJ-gld/gld x Mus spretus) F1 X C3H/HeJ-gld/gld] interspecific backcross mice will be used to identify genetic markers at interval of between 10 to 20 centi-Morgans throughout the genome. As this segregation map develops it will be utilized to identify and define the extent and border of conserved linkage groups common to mice and humans. Detailed analysis of mouse chromosome 1 genetics is the central focus of the proposal. The goals include: i) creation of a distal mouse chromosome 1 specific library; ii) detailed comparative mapping with syntenic groups on human chromosomes 1q, 2q and possibly 18q; iii) physical mapping studies to generate a molecular map of distal mouse chromosome 1 to complement the detailed genetic map and iv) comparative physical mapping of different Mus species and areas of the human genome. The large number of interspecific backcross mice (over 1200 available) combined with the generation of large numbers of markers will allow the resolution of a 0.5 cm genetic map of distal mouse chromosome 1. The refinement of the mouse genomic map at both a genetic and physical level will allow better analyses of the relationships between mouse genetic models for disease and their possible human counterparts as well as fundamental structure/function correlations concerning genomic organization. The studies should provide important insight into aspects of mammalian meiotic recombination and the evolution of mammalian genomes.

Genetic factors contribute to the expression of autoimmune diseases in both mice and humans. These genes are largely uncharacterized; moreover, their gene products and mechanisms of action are unknown. A recessive mutation, lpr, that occurred in MRL mice results in profound lymphadenopathy with expansion of an unusual subpopulation of T cells and generalized autoimmune disease. A "reverse" genetic approach is proposed to identify the mutant or deficient gene products of the lpr locus and MRL genes that interact with lpr to result in renal disease. The method relies on defining the physical location of these genes. RFLV and microsatellite polymorphisms will be used as markers in two different crosses segregating for the recessive genes that are critical to clinical and serologic disease in these mice. The renal histology will be scored and correlated with the genotype analyses and autoantibody profiles. The localization of lpr to mouse Chr 19 will be confirmed and saturation mapping using clones from a flow sorted mouse Chr 19 library will be performed. Of most interest will be defining the MRL quantitative trait loci (QTL) that predispose to nephritis. Detailed mapping of each relevant region of the genome will be undertaken using available clones localized to the specific chromosome segment and clones likely to be in that region of the genome. If necessary, a subchromosome specific library or a chromosome microdissection library will be generated to allow saturation mapping of one of the identified QTL regions. Long range restriction site analyses of the relevant regions of the mouse genome will be performed. If, as is likely, one or more autoimmune loci are within a chromosomal segment largely conserved between the mouse and human genome, these studies may identify the chromosomal location of human "autoimmune genes". A variety of strategies including the use of chromosome jumping and a mouse yeast artificial chromosome library, in addition to long range restriction mapping will be used to more precisely define the molecular location of the critical gene (s) responsible for genetic predisposition. Comparison with MRL-plus-minus/plus-minus mice will be useful for determining the lpr mutation as will comparison of CBA/KJ with a new mutation at the lpr locus in a strain designated CBA/KJ-lpr(cg) . For other localized "autoimmune genes", candidate genes may be suggested by their chromosomal location. Definition of a single gene that predisposes to autoimmunity should allow fundamental insight into molecular basis for disease in mice and humans.

Genetic factors contribute to the expression of autoimmune diseases in both mice and humans. These genes are largely uncharacterized; moreover, their gene products and mechanisms of action are unknown. A recessive mutation, lpr, that occurred in MRL mice results in profound lymphadenopathy with expansion of an unusual subpopulation of T cells and generalized autoimmune disease. A "reverse" genetic approach is proposed to identify the mutant or deficient gene products of the lpr locus and MRL genes that interact with lpr to result in renal disease. The method relies on defining the physical location of these genes. RFLV and microsatellite polymorphisms will be used as markers in two different crosses segregating for the recessive genes that are critical to clinical and serologic disease in these mice. The renal histology will be scored and correlated with the genotype analyses and autoantibody profiles. The localization of lpr to mouse Chr 19 will be confirmed and saturation mapping using clones from a flow sorted mouse Chr 19 library will be performed. Of most interest will be defining the MRL quantitative trait loci (QTL) that predispose to nephritis. Detailed mapping of each relevant region of the genome will be undertaken using available clones localized to the specific chromosome segment and clones likely to be in that region of the genome. If necessary, a subchromosome specific library or a chromosome microdissection library will be generated to allow saturation mapping of one of the identified QTL regions. Long range restriction site analyses of the relevant regions of the mouse genome will be performed. If, as is likely, one or more autoimmune loci are within a chromosomal segment largely conserved between the mouse and human genome, these studies may identify the chromosomal location of human "autoimmune genes". A variety of strategies including the use of chromosome jumping and a mouse yeast artificial chromosome library, in addition to long range restriction mapping will be used to more precisely define the molecular location of the critical gene (s) responsible for genetic predisposition. Comparison with MRL-plus-minus/plus-minus mice will be useful for determining the lpr mutation as will comparison of CBA/KJ with a new mutation at the lpr locus in a strain designated CBA/KJ-lpr(cg) . For other localized "autoimmune genes", candidate genes may be suggested by their chromosomal location. Definition of a single gene that predisposes to autoimmunity should allow fundamental insight into molecular basis for disease in mice and humans.

DESCRIPTION (Adapted from the Investigator's Abstract): The human genome project is entering a sequencing intensive stage. However, it is the premise of this proposal that high resolution gene and EST based map construction still has a role to play in 1) providing additional infrastructure for these efforts and 2) to facilitate the utilization of the emerging genome resources in a wide variety of investigations into genetic disease. Computational algorithms will be used to identify mouse ESTs/genes that are the orthologues of mapped human ESTs/genes. To further develop a high density gene rich map of the mouse genome the investigators will map 600 of these mouse ESTs/genes using their well-characterized interspecific backcross (containing 1200 markers). The mouse ESTs selected for mapping will be based on the human chromosomal location of the orthologous human unigenes. Although homology relationships have been rapidly identified at low resolution the borders of conserved genomic segments as well as intra-segment rearrangement events remain largely ambiguous. The current proposal will allow a higher resolution definition of these mouse/human homology relationships by this selective mapping of ESTs and genes in a single gene dense mouse map. The integration of microsatellite markers in the high-density genetic map will also be a component of these efforts. Approximately 500 microsatellite markers will be chosen to provide an integration of these maps. These studies will facilitate contig construction in the mouse as well as gene hunting studies in both mouse and human. The proposal will provide an additional framework for future biological studies as well as the tools necessary to verify and extend the definitive characterization of the mouse genome and completion of contig construction. An on-line database that provides access to the human/mouse homologies (http:/www3.ncbi.nlm.nih.gov/Homology/) will be maintained and updated as part of this project. This database will provide a simple interface between human and mouse coding sequences as well as nongenic STS markers. The database will be integrated with the online Gene Map of the Human Genome and will be annotated to include unmapped mouse orthologues of human unigenes and their putative chromosomal location.

Abstract
CTS-9986530
Scott Collins, U of CA Davis

This exploratory effort is focused on the development of a high-speed DNA gene sequencing device. The sequencer is based on electrophoretically pulling single-stranded DNA through a nanopore 1-3 nm in diameter. As the individual nucleotides traverse the nanopore, they block, and therefore, modulate the current through the pore. The magnitude of the current modulation is a function of the chemical composition of the nucleotide base, i.e., base sequences may be determined by their characteristic current signatures. The key aspect of this exploratory effort is the feasibility fabrication of the nanopore by anodic dissolution of silicon in HF, e.e., porous silicon. The nanopores will be active component in an integral microdevice design to electrically address the nanopore with sense electrodes and data acquisition circuitry. DNA transport and detection through these pores will be demonstrated.

DESCRIPTION (provided by applicant): Admixture mapping is a potentially powerful tool for finding disease susceptibility genes in complex genetic diseases. The method depends on recent admixture between different ethnic groups for which differences in the distribution of susceptibility genes exist, and ancestry of the chromosomal regions can be distinguished in the admixed population. We believe that this approach will be a useful complement to genome-wide linkage studies that often lack power and association based methodologies that are difficult to apply genome-wide. For providing proof of the applicability of admixture mapping it is necessary to 1) identify a set of markers that distinguish ancestry; 2) develop and apply multilocus statistical tests for linkage in the presence of ancestral association to simulations of real genotyping data; and 3) provide an actual demonstration of the method in a real disease. These are the major objectives of this proposal. Preliminary studies by our group as well as other investigators have shown that African Americans are a large admixed population for which admixture mapping is likely to be applicable. Epidemiological studies suggest that systemic lupus erythematosus is a disease that disproportionately affects individuals with African heritage. Preliminary studies provide confidence that a Ancestry Informative Markers (AIMs) can identify African versus European ancestry and that the vast majority of these AIMs in contrast to other markers have only small intra-ethnic differences within the European and African founding populations. Furthermore, the current results of several groups including the Perlegen Sciences screen of over 1.6 million SNPs suggest that a sufficiently large genome-wide set of AIMs is now available for a robust statistical analysis. This proposal will establish a genome wide set of >4000 AIMs that are characterized in African, European American, and African American populations. The typing of 950 African American SLE cases and 500 matched African American controls will provide the necessary data for simulations and testing as well as the opportunity to explore the usefulness of this method for a disease of major importance to the health of African Americans.

[unreadable] DESCRIPTION (provided by applicant): Recently, substantial progress has been made in development and application of admixture mapping including derivation of appropriate statistical tools and testing in African American populations. The implementation of this approach in Mexican Americans (MA) has been hampered by the lack of a sufficient density of markers that can provide the ancestry information distinguishing the two major parental populations, Amerindians (Al) and European that contribute to this admixed population. The current proposal will remedy this problem by identification, validation and analysis of a panel of >5000 AI/European American (EA) Ancestry Informative Markers (AIMs) distributed over the genome. The panel of AIMs will be tested using a large collected set of MA DNA samples from type 2 diabetics with nephropathy (750 samples) or without nephropathy (600 samples) and a smaller matched set of nondiabetic MA subjects (300 samples). The utilization of this set will enable both case only and case control designs to be examined using a variety of recently developed algorithms. The first phase of the study (identification of AI/EA AIMs) will utilize an enrichment strategy based on pre-selecting AIMs dependent on putative Chinese/EA AIMs that have been identified in a completed 1.6 x 106 SNP screen. Over 23,000 putative Chinese/EA AIMs with Fsts > 0.45 have been identified and these will be used to examine 48 Pima Amerindian samples. Our previous studies conservatively suggest that >25% of these putative Chinese/EA AIMs will have an AI/EA Fst > 0.4. In the second phase > 5000 AI/EA AIMs will be chosen based on both Fst and chromosomal location to achieve a high density AIMs panel. This will be validated utilizing an additional set of 48 Pima Amerindians in addition to testing two other Al groups (96 samples each) and EA (96 samples). The MA populations will be examined using this panel that is expected to include >4000 validated EA/AI AIMs and analyzed to 1) define chromosomal segments derived from EA and Al, 2) perform modeling studies and 3) identify putative diabetes and diabetic nephropathy associated regions distinguishing the major ancestry. Finally, the three strongest signals will be further investigated utilizing a dense set of SNPs in the putative susceptibility regions. [unreadable] [unreadable]

DESCRIPTION (provided by applicant): We will use a combination of second and third generation deep sequencing of selected candidate regions, whole exomes and mRNAs to identify both common and uncommon variants that predispose to PBC susceptibility. The sequencing of selected chromosome regions in 150 cases and 150 controls will target identification of variants that underlie the association of IL12A, SPIB, and a chromosome 17 locus (IKZF3/ORMDL3) that are identified in our PBC GWAS. The paired sequencing of these regions will provide the opportunity to ascertain coding and non-coding variation including copy number variants. The exome sequencing of 400 cases and 400 controls will screen for uncommon genetic variants that are not amenable to GWAS detection and provide the opportunity to test an alternate paradigm that does not depend on the common variant hypothesis. Importantly, mRNA sequencing of two cell populations implicated in PBC pathogenesis (CD8+ and CD4+ T cells) will complement both the chromosome region and exome results. For this aspect, mRNA will be sequenced in 75 cases and 75 controls. This mRNA sequencing together with the targeted chromosomal region sequencing and exome sequencing will provide the ability to correlate sequence variation with 1) gene expression, 2) eQTN data, 3) alternative exon usage, 4) RNA editing and 5) preferential allelic expression. A variety of informatics approaches using the combined data will establish a prioritization of SNPs for validation and testing in large numbers 1100 PBC cases and 2200 controls (not including discovery subject set) using a Golden Gate 1536 SNPlex. Both the sequencing and replication studies will be performed using a homogeneous Italian population. Together this design should maximize our ability to identify uncommon as well as more common variants that are important in the etiopathogenesis of this autoimmune disease.