Margit Burmeister - US grants

Neurological diseases are among the most common inherited diseases in humans. The primary cause of most neurological inherited diseases is not known, but many neurological mutations in mouse have been shown to have an equivalent human counterpart. Since mice are easier to analyze and can be bred at will, most murine neurological mutations have already been mapped to chromosomes. A molecular cloning of such murine mutations will be important to understand the basic functioning of the brain, and may help in defining the cause of some human neurological diseases. We propose here to characterize and clone the genes that when mutated cause the murine neurological mutations jittery and mocha. Mocha mice are characterized by inner ear defects, reduced bleeding times and a coat color defect, Notably, they sometimes show spike-wave discharges coupled to behavioral inactivity suggestive of absence seizures. Mocha homozygous mice also show a rhythmical, synchronous, high voltage activity on EEG, which is unique to this mutant and believed to be one of the first brain rhythm traits known to arise from a single gene defect. Jittery mice are characterized by progressively worsening dystonia and ataxia with tonic- clonic seizures, and die within one month after birth. These murine mutants map genetically close to each other and are located in a region of murine chromosome 10 that is homologous to a well- characterized region of human chromosome 21. The gene for Baltic progressive myoclonus epilepsy (Unverricht-Lundborg) maps to this region. This genetic proximity suggests that jittery may be a genetic murine model for Baltic progressive myoclonus epilepsy. We will approach mapping and ultimately cloning the genes involved in these two mutants by classical genetic methods, i.e. an intraspecific backcross with Mus musculus casjaneus, and molecular genetic techniques that take advantage of the large number of tools that have been developed by us and others for the study of human chromosome 21. Isolation of these genes will allow us to study and understand the primary cause of these specific neurological mutations, increase our understanding of brain function in general, and determine whether jittery is a murine model for Baltic progressive myoclonus epilepsy.

APPLICANT'S ABSTRACT: There is a growing consensus that alcoholism (alcohol abuse and dependence) develops over time through a complex interaction between a maturing person and the environment, where expression of risk is dependent upon individual experience as well as diathesis. Environmental settings will trigger or suppress risk over the life span. However, developmental changes are difficult to demonstrate without clear knowledge of initial vulnerability markers and environmental risk enhancers that are specific to alcoholism. The complexity of etiology and the inconsistent results from genetic studies have discouraged interest in research on single-gene effects in alcoholism. However, definitive research with homogeneous subtypes and truly relevant candidate genes has not been conducted. Biological markers have been identified for the Antisocial Alcoholism subtype, specifically lower serotonin (5-HT) function. The consistency of this research provides a strong rationale for testing associations with candidate genes that relate to 5-HT metabolism. Single gene effects may be detectable with homogeneous subtypes. The proposed study uses a family-based design, the haplotype relative risk method (a study design that prevents confounding by ethnic differences between control and affected groups). This proposal requests sufficient funding for a definitive study of autosomal serotonin markers (tryptophan hydroxylase, the serotonin transporter, and three serotonin receptor types [lA, lB, 2A]) in male antisocial alcoholics (n=50 probands and their 100 parents), and funds for conducting two feasibility studies: (l) X-linked markers (monamine oxidase A and B and one serotonin receptor type [2C]) in 10 male probands and their 20 parents; and (2) autosomal and X-linked markers in 10 women and their 20 parents. For this exploratory/developmental study, a sample with homogeneous ethnic background will be sought. The specific aims are to test genetic associations between antisocial alcoholism and polymorphisms in tryptophan hydroxylase, the serotonin transporter, monamine oxidase A and B, and four serotonin receptor types (lA, lB, 2A, and 2C). The research plan will also allow preservation of DNA for future use and assessment of the feasibility of future studies involving sib-pair linkage, recruitment of adequate samples of other races, and longitudinal studies of genetic and environmental effects on chronicity of the disorder.

gene mutation; genetic mapping; genetic disorder; myoclonus epilepsy; human genetic material tag;

gene mutation; genetic disorder; sensorineural hearing loss; ear hair cell; developmental neurobiology; genetic models; developmental genetics; genetic markers; disease /disorder model; animal genetic material tag; laboratory mouse; genetic mapping; molecular cloning;

DESCRIPTION: (Adapted from investigator's abstract) The P.I. has recently identified the mutant gene in mocha mice as the delta subunit of the adapter-related complex AP-3. They have also found that the ZnT-3 transporter is not transported correctly to synaptic vesicles, resulting in a lack of zinc in cortex and hippocampus. In this application, the P.I. proposes to genetically map all subunits of the AP-3 complex to determine if other mutants with a similar phenotype are caused by mutations in these AP-3 subunit genes, and test for interaction between mocha locus and pale-ear, the mouse homologue of HPS, the gene most commonly mutated in human Hermansky-Pudlak syndrome (HPS). They will focus on the neurological phenotype of mocha, and determine if mocha may be a mouse model for epilepsy, ADHD, autism or other neurological disorders. Mocha mutant mice have an HPS-like phenotype as well as neurological deficits (seizures, hyperactivity, spike-wave discharges, a hypersynchronized electrocortigram, increased auditory gating). In contrast, pearl mice have HPS but none of these neurological phenotypes, which they postulate is because pearl mice miss the non-neuronal form of the beta subunit, Ap3b1, but not the neuronal form of AP-3 beta, Ap3b2, whereas the delta subunit mutated in mocha is ubiquitously expressed. Dr. Burmeister postulates that inactivation of the neuronal form of AP-3 beta will result in a mouse with the neurological defects of mocha without the HPS-like phenotypes and higher fertility and viability than mocha mice. They will prepare a LoxP construct to knock out Ap3b2 in such a way that they can not only generate a complete knockout in ES cells, but also, by mating to mice in which Cre is under region-specific promoters, mice in which the AP-3 complex is missing only in specific brain regions. The P.I. will characterize the behavior of mocha, mh-2J, ZnT-3 deficient mice as well as the proposed knockout mice for the nature of hyperactivity (is it generally more active, has increased startle, or stereotypic behavior), seizure propensity, anxiety, learning and memory and electrophysiological parameters. To determine if AP3B2 plays a role in human neurological disorders, Dr. Burmeister will isolate and characterize the human AP3B2 gene and search for mutations or polymorphisms that may be present in the normal population or in patients. Given the mocha phenotype, it is anticipated that this gene may be involved in human neurological disorders characterized by increased seizure frequency and hyperactivity (e.g. autism, OCD, ADHD and epilepsy). Such polymorphisms will be made available for the scientific community to test as a candidate gene for other neurological or psychiatric disorders if justified by the results of the behavior tests.

[unreadable] DESCRIPTION (provided by applicant): The etiology of most recessive cerebellar ataxias is unknown. We have recently identified a novel ataxia gene, ATCAY/Atcay, for Ataxia, Cayman type. Mutations in this gene cause recessive ataxia in mouse (alleles of jittery) and human (Cayman Ataxia). The jittery phenotypes range from a barely detectable hesitant, slightly ataxic walk with normal life span in a mild allele to severe ataxia, dystonia with seizures and early death in other alleles. Cayman Ataxia is a recessive ataxia found so far exclusively on one region of Grand Cayman island where it arose by founder effect. Cayman Ataxia patients show severe truncal ataxia, ocular movement abnormalities, and mental retardation. We will identify all causal mutations in mouse and determine which of two mutations that are present in all Cayman Ataxia patients causes the disorder. [unreadable] [unreadable] The protein, CAYTAXIN, is expressed at high levels exclusively in neuronal tissues, including brain, spinal and enteric ganglion cells. The C-terminal end of the protein has a CRAL-TRIO domain which is typical for proteins that bind or transport small molecules. The C-terminus is also homologous to a BCH (BNIP2-CDC42GAP homology) domain present in CDC42GAP-like proteins. To characterize the functions of this protein, we will study its expression, the intracellular localization of the protein, its interacting proteins and its potential functions in apoptosis and/or signalling. Our studies may reveal a novel pathway to the etiology of ataxia. [unreadable] [unreadable]

DESCRIPTION (provided by applicant): Identifying genetic variants underlying the risk for complex disorders such as depression is extremely difficult. To complement linkage and association studies of depression, we propose a genetic association study of Neuroticism. The genetics of Neuroticism overlaps significantly with that of depression, but it constitutes a quantitative trait that can be measured in a population sample. Neuroticism is measured with the psychometrically sound, validated NEO personality inventory. Our sample consists of about 1500 subjects in about 500 families ascertained through a hypertensive proband but unselected otherwise, and about 200 hypotensive controls. More than half of the subjects have also been administered the NEO-PI, and all have given blood for genetic DMA analysis. With the first 470 subjects, we found two genetic variants associated with Neuroticism, which were replicated by other laboratories and also found associated with affective disorder in other studies. These findings confirm our underlying hypothesis that identification of genetic variants affecting Neuroticism scores will be relevant for affective disorders. In addition, a whole genome scan has been performed on our sample and will be analyzed for linkage to personality traits, including Neuroticism. In this pilot grant, we propose to continue candidate gene analysis. Candidate genes will be informed by our preliminary data, ongoing microarray analyses, the literature, and linkage analyses. We will start with candidate genes involved in the serotonin system due to our preliminary data. SNPs in these candidate genes will be either known functional (coding or promoter) variants, or selected from the HAPMAP. DNA analysis will be performed with a novel, multiplex (15-30 SNPs) genotyping system and will be compared to commercial genotyping methods for optimization and error rate determination. The nuclear family structure allows both population-based and family-based association studies to be carried out. A variance component model-based association test called QTDT, which is able to account for familial resemblance, is an ideal tool for this type of analysis Haplotype analyses are reported to be more informative than single marker analysis in association studies, and we will use generalized linear models to estimate haplotype effects. At the end, we will compare the results obtained from different association tests, and follow up on the most interesting signals. It is anticipated that this R21 will lead to a more thorough, genome-wide genetic study of Neuroticism in this valuable sample.

DESCRIPTION (provided by applicant): Positional cloning has helped identify the molecular causes of many of Mendelian disorders. Genes involved in the more common forms of deafness and ataxia have been identified and are often clinically tested. However, the cause of most genetically rare forms is still unknown because such families are often too small to provide by themselves sufficient statistical power to unambiguously identify linkage. In addition, even if families are large enough to identify a linked locus, the regions are typically so large that hundreds of genes are implicated. In some situations, linkage disequilibrium analysis can further narrow a region. Recently, global analysis of gene expression has become a reality. We hypothesize that a subset of mutated genes directly or indirectly leads to abnormal expression levels, and can be identified by combining genetic analysis with gene expression analysis of mRNA from lymphoblastoid cell lines (LCLs). Using such a combined approach, a novel epilepsy gene was recently identified in only three affected subjects. We have recently identified a novel auditory neuropathy gene, and have encouraging data for identification of a second novel deafness gene and for several novel ataxia genes. In this pilot/feasibility R21 grant, we will explore the generalizability of this approach in several families with unidentified causes of deafness or ataxia. We will perform genetic linkage, linkage disequilibrium and global gene expression analysis. Candidates genes will be identified as the intercept between genes in linkage regions and those that show significant expression changes. Computational approaches to define the intersect between genetic and gene expression data will be developed, implemented and tested. While in some cases combining genetic linkage and linkage disequilibrium with global gene expression analysis may directly identify the mutant gene, in other cases the mutation may result in functional differences in a pathway without affecting the expression of its own gene directly. In that case, more complex analysis of affected gene expression changes downstream of a mutant gene will be necessary to identify the deficient pathway and the mutation. Pathways from expression analysis, animal models and the literature and public bioinformatic databases will be used for this purpose. Candidate genes emerging from these approaches will be sequenced to identify variants, and potential mutations tested in unrelated controls. PUBLIC HEALTH RELEVANCE: The genetic cause of many forms of deafness or ataxia is still unknown. Our research will result in the identification of new genes and metabolic pathways involved in these disorders. These findings can improve accurate diagnosis, presymptomatic testing, family planning and personalizing treatment. It may also ultimately lead to new treatments.

This Integrative Graduate Education and Research Traineeship (IGERT) award supports a new graduate training program in open data sharing and data reuse in e-science at the University of Michigan. The purpose of this program is to generate new policies, practices, and technologies for data sharing that will accelerate the pace of scientific discovery and the transfer of scientific knowledge into useful products and technologies. It will build on computer science and information science research that identifies basic principles for acquiring, managing, sharing, and archiving data and for developing new tools and technologies to put these principles into practice in bioinformatics and materials research. The program will train a cohort of doctoral students in a new way of thinking about open data. These students will be capable of contributing high quality data to repositories and effectively leveraging large volumes of data in scientific research. Students from Michigan's doctoral programs in Bioinformatics, Materials Science, Chemical Engineering, Electrical Engineering and Computer Science, and Information will be selected as IGERT trainees. The trainees will participate in a new graduate course in data curation, a multi-disciplinary public seminar series with expert speakers, and short workshops. Program faculty will also develop a supplemental summer program for undergraduates to attract diverse students and strengthen the program's impact in other scientific fields. IGERT is an NSF-wide program intended to meet the challenges of educating U.S. Ph.D. scientists and engineers with the interdisciplinary background, deep knowledge in a chosen discipline, and the technical, professional, and personal skills needed for the career demands of the future. The program is intended to catalyze a cultural change in graduate education by establishing innovative new models for graduate education and training in a fertile environment for collaborative research that transcends traditional disciplinary boundaries.

DESCRIPTION (provided by applicant): Ataxias are a heterogeneous group of neurological disorders caused by environmental as well as genetic factors. While mutations in more than 40 genes have been implicated in hereditary ataxia, and genetic testing has become an increasingly routine clinical practice, most sporadic and many inherited cases are still of unknown origin. Individual families with unexplained ataxia are often too small to allow gene identification by traditional positional cloning approaches; and the lack of understanding of ataxia etiology in these cases continues to hamper accurate diagnosis and identification of novel targets for effective and personalized therapy. Recently, massively parallel sequencing allows efficient discovery of nearly all variants in a genome, or at least its coding portion. However, as each person carries thousands of novel variants, sequencing alone does not permit immediate identification of causative mutations. Here, we hypothesize that genetic linkage, sequencing-based variant discovery, and gene expression analysis, while not sufficiently informative separately, provide complementary information and allow identification of novel ataxia genes in families with a minimum of two recessive or three dominantly affected members. We already have recruited and started to analyze 12 such pedigrees, and have developed expertise to collect and analyze the relevant genomic data. We propose to study 22 pedigrees using an integrated pipeline that combines linkage analysis, gene expression profiling, next-generation sequencing, and gene network analysis. Since different ataxia families likely carry distinct variants, none of these techniques alone will be sufficient, but their combination will be highly informative to pinpoint candidates. Candidate mutations will be tested for absence in control samples, and candidate genes will be screened for additional mutations in unrelated cases. By using such an integrated approach we have recently identified an auditory neuropathy gene and have found a credible novel candidate gene for dominant central nuclear myopathy. We also have promising preliminary data for several ataxia pedigrees. In addition to global analyses of complementary genomic datasets, we are committed to rapid functional follow-up using appropriate neuronal cells. This is achieved by reprogramming fibroblasts from biopsy cultures from selected families into neurons, and testing the effect of potential mutations on splicing and amount of mRNA, on other genes within the relevant cell type, and on cellular phenotypes. We predict that this pipeline will lead to the identification of new rare ataxia gene mutations that were not previously possible using purely genetic positional cloning strategies. The discoveries will enable better diagnosis and prognosis which will immediately help the affected families. It may lead to personalized treatment & generate new hypotheses to study the more common sporadic forms of ataxia. The identification of validated & functionally characterized molecular lesions is expected to facilitat informed drug development. Our experience with this approach will also establish a useful paradigm for other rare Mendelian disorders. PUBLIC HEALTH RELEVANCE: The cause of many forms of ataxia is still unknown, even when inheritance in families clearly points to a genetic cause. Recent genomic tools and next generation sequencing will allow us to identify several novel ataxia genes in small families. Finding these new genes will not only help the affected families with diagnosis, family planning, and prognosis, but also aid in the development of comprehensive ataxia networks of genes that will lead to future personalized diagnosis, prognosis and treatment.

DESCRIPTION (provided by applicant): Integrating large-scale genomics data has huge potential to accelerate the identification of disease genes in human. Three major challenges lie in the current integrative approach for predicting disease genes. First, previous integrations in general limit genomic data input to one species at a time, while disease datasets are often generated in multiple model organisms. Second, public functional genomic datasets are dominated and biased by certain data types and accessible tissues, which can be addressed by expert curation of input datasets. Third, when multiple tissue-specific networks have been generated, a mathematical formulation is lacking to prioritize among these competing networks for the specific disease under consideration. This collaborative proposal aims at addressing the above challenges by exploring a prototype of bioinformatics tools to integrate multiple relevant global and tissue-specific networks across mammalian species targeting a specific disease, here ataxia. This proposal is based on our preliminary data in developing both global and cerebellum-specific networks to prioritize ataxia associated genes, and on the two PIs' complementary expertise in genomic data integration and experimental ataxia gene confirmation. We will 1) use domain-specific and multiple species data to establish global, brain, cerebellum, related tissue, and ataxia-specific networks, and develop web tools to explore these networks; and 2) develop multiple kernel learning algorithms to weigh and integrate multiple networks to predict ataxia-associated genes. Although the algorithms will be developed targeting ataxia only, we envision that this expert-driven integrative approach will be adaptable to other disease gene identification scenarios.

? DESCRIPTION (provided by applicant): Despite convincing evidence for a heritable component in alcohol use disorders (AUD) and other drug addictions (DA), robust and reproducible evidence for genes involved in AUD or DA is rare. One of the few consistent findings over the past 10 years are genetic associations of addiction and related endophenotypes such as brain oscillations and impulsivity with SNPs in GABRA2, the gene encoding the GABA(A) alpha 2 receptor. The biological basis of this association has, however, remained elusive. Associated SNPs are in strong linkage disequilibrium (LD) with each other. Most Caucasians carry one or two common haplotypes differing at 299 SNP alleles, which do not affect an amino acid or measured overall mRNA levels. Since GABRA2 is only expressed in brain, functional studies have so far been difficult. Here, we hypothesize that SNPs within GABRA2 affect gene regulation and/or splicing. Computational integration of chromatin data across 30+ cell types indicates the presence of brain-specific enhancers within the GABRA2 gene, with several SNPs of the relevant haplotype within predicted regulatory regions. Preliminary RNASeq and Sanger Sequencing data suggest allelic imbalance, with the non-ancestral, risk increasing haplotype associated with increased transcription. In Aim 1, we will test the enhancer capability of 8 DNA fragments with either of the two SNP alleles in- vitro in existing human embryonic stem cells differentiated into neurons using standard reporter assays. In Aim 2, to investigate the effect of GABRA2 haplotypes on gene regulation, we will generate neurons that differ by GABRA2 haplotype by generating induced pluripotent stem cell lines (iPSCs) from lymphoblastoid cell lines (LCLs) from four sets of siblings, hence in a similar genetic background. These cell lines will allow, in a genetically similar and well controlled background, and throughout the differentiation from iPSC to neurons, the study of a) transcription level as allelic imbalance, b) GABRA2 splicing at several differentially spliced site, and c) changes of methylation of CpG sites within GABRA2 during differentiation. Our studies will elucidate the path from genetic variation to addiction. Our studies will generate iPSC lines from related individuals who differ by GABRA2 haplotype, which will allow study of development of GABRA2 gene expression over differentiation to neurons. These cell lines will also be made available to others, as they may be useful for the study of other common neuron-specific variants. Future studies building on this research and the tools generated by this R21 will include genetic fine mapping of the molecular phenotype in iPSCs of recombinant cell lines, electrophysiology, reactivity of neurons to GABA, and response to electrical stimulation, i.e. the cellular and physiological phenomena that go beyond proximal RNA and protein, on the path to brain circuitry.

? DESCRIPTION (provided by applicant): This is a proposal to renew and grow the University of Michigan Bioinformatics Training Program (U-M BITP), now in year 10, to 8 trainee slots/year for 2-3 years of pre-doctoral training, usually between years 1 and 3. BITP continues to be a part of our now well established and widely respected U-M Bioinformatics Graduate Program (BGP). The BGP (and its BITP) is an interdisciplinary graduate training program in bioinformatics and computational biology, drawing faculty from the School of Medicine, College of Engineering, College of Literature, Sciences and the Arts (LS&A; Including the Departments of Mathematics, Statistics, Chemistry, and Physics), the School of Public Health, School of Nursing, the College of Pharmacy, and the School of Information. The BGP and BITP are embedded in the U-M Center for Computational Medicine and Bioinformatics (CCMB), an interdisciplinary research and education center that provide the interdisciplinary research and training context. CCMB currently has 127 affiliate faculty members across the U-M, 48 of whom are participating in this BITP as potential primary mentors. CCMB is hosted within our University of Michigan Medical School Department of Computational Medicine and Bioinformatics (DCM&B), which currently has 14 primary faculty appointments, plus 12 affiliate faculty, and 4 research track core faculty members. All core DCM&B faculty members are eligible mentors of BITP trainees. BITP trainees have a full curriculum of Bioinformatics, Statistics, Data Science, and Biology/Biomedicine graduate courses to choose from, journal clubs, seminars, workshops, and special events. The BITP dissertation training utilizes a dual-mentor approach, which combines quantitative/computational and DBP application elements. The U-M-based tranSMART Foundation, cancer biostatistics, proteome informatics activities, NIDDK Metabolomics Center, and the CTSA Biomedical Informatics Program have all achieved national recognition, and are a natural magnet for BITP trainees. The goal of the BITP is to train students in bioinformatics and applied computational biology by engaging them in a rigorous curriculum and pre-doctoral training experience. BITP trainees engage in cutting-edge collaborative research featuring a strong driving biological problem (DBP) application, often leading BTP trainees to have a strong T1 Translational Research orientation. The BGP has graduated 55 Ph.D. trainees in bioinformatics since its first in 2006. Within this trainee cohort, 10 BTP traines have graduated to date. These graduates have launched exciting careers in industry, academics, and government. The BGP and BITP has established a Data Science Training Track, and is actively engaged in the emerging Michigan Institute for Data Science (MIDAS), which will be offering a Data Science Certificate to enhance bioinformatics training for Data Science Concentrators. The overall objective of the BTP is to provide the finest Bioinformatics Training environment and trainee experience available in the US.