Mutational Cloning in Familial Dementia and Alzheimers Disease

描述（由申请人提供）：该申请涉及广泛挑战领域（08）基因组学和特定挑战主题08- ag -106交叉疾病研究，以确定低发病率、神经遗传疾病与高发病率、基于人群的疾病之间的共同靶向途径或机制。本提案的目标是确定负责阿尔茨海默病（AD）或额颞叶痴呆（FTD）亚型的孟德尔，高渗透家族性痴呆（FD）的新基因。这将通过对10个FD家族的全基因组蛋白质编码部分（“外显子组”）进行测序来完成，这些FD家族在华盛顿大学阿尔茨海默病研究中心（ADRC）进行了30年的确定、表型分析和采样。广泛的ADRC收集是发现负责认知能力下降和神经变性的新基因的宝贵资源。FD基因和相关突变的鉴定将有助于从几个方面发现其他致病基因。首先，对新发现或拓宽的生物学途径的分析将为更有针对性的研究带来新的痴呆候选基因。其次，未知原因的FD的遗传异质性将减少，因为根据新发现的FD基因的突变将家族划分为亚群。第三，这些基因反过来又会涉及额外的途径，并将增加治疗干预的目标数量。最后，为了使科学和临床社区受益，并便于在其他样本中证实我们的发现，我们将把FD基因型和表型信息存入在线资源GeneTests和NCBI变异数据库（dbSNP和dbGaP）。在过去的二十年里，基因研究和家庭定位克隆已经确定了许多孟德尔疾病的基因。ADRC的研究人员一直站在这项工作的最前沿，并在发现包括PSEN1、PSEN2和MAPT在内的多个基因方面发挥了重要作用。这项重要的工作目前允许对单基因痴呆的准确诊断，扩展了我们对神经变性生物学的理解，并为可能适用于复杂病因的常见fd的新治疗奠定了基础。然而，对于许多受影响的家庭，致病基因尚未发现。许多这样的家族太小，无法进行位置克隆。以前，尽管有广泛的疾病家族史，但只有少数受影响的人的DNA可用，因此没有方法鉴定新的致病基因。在本提案中，我们将利用最新的大规模平行测序技术，结合基因组注释和生物信息学的进展，将外显子组的突变克隆作为鉴定FD相关基因的一种强大而创新的方法。突变克隆是基于排斥映射的概念。在一个家庭中，所有受影响的个体必须共享导致疾病的突变，所有非共享的变异都被排除在外。从已经测序的个体的基因组中，我们估计每个个体大约有10,000个蛋白质改变。因为导致FD常染色体显性形式的突变必须是罕见的，它们不太可能以规定的频率出现在dbSNP中。此时，大约90%的蛋白质改变变体存在于dbSNP中。这使得每个测序个体平均有1000个独特的变异。这个数字是高估的，因为随着重测序的进一步增加（例如，1000基因组计划），dbSNP中不太常见的变异的数量将会增加。目前对超过2000个克隆孟德尔基因的研究强烈支持致病突变影响高度保守残基的观点。一些变异将预测保守的氨基酸取代不太可能损害蛋白质。根据这些原则，我们估计80%的独特变异可以被消除。根据这些标准，每个外显子组将保留大约200个候选变体，这与定位克隆方法中最小连锁区域中包含的基因数量非常相似。家庭中第二个受影响的人的外显子组将把这些变体的一部分排除在进一步的考虑之外。因为兄弟姐妹平均共享50%的基因组，一对受影响的兄弟姐妹有可能将候选突变的数量减少一半。即使家庭人数略有增加，这个数字也会急剧减少。表亲平均共享12.5%的基因组，可以将候选突变的数量从200个减少到25个。如果该变异不能与家族中其他受影响的人的疾病共分离，将会消除更多的变异。剩余的候选基因将根据基因功能、组织表达、保存、突变的预测影响、动物模型的存在、与人类疾病的关系以及文献中的其他线索，优先用于其他家庭和痴呆症病例的研究。我们准备最大限度地利用基础设施（华盛顿大学基因组科学系），分子遗传学专业知识(博士。Raskind， Shendure和Brkanac)，现有的表征良好的样本（Bird博士），以及测序技术的进步，现在可以在没有预设候选基因的情况下进行全外显子组筛选。

DESCRIPTION (provided by applicant): This application addresses Broad Challenge Area (08) Genomics and Specific Challenge Topic 08-AG-106 Cross-disease research to identify commonly targeted pathways or mechanisms between low incidence, neurogenetic disorders with high incidence, population-based disease. The goal of this proposal is to identify novel genes that are responsible for mendelian, highly penetrant familial dementia (FD) of either Alzheimer's disease (AD) or frontotemporal dementia (FTD) subtype. This will be accomplished by sequencing the whole genome protein-coding portion ("exome") in a collection of 10 families with FD that were ascertained, phenotyped and sampled over a period of 30 years at the University of Washington Alzheimer's Disease Research Center (ADRC). The extensive ADRC collection is a valuable resource for the discovery of new genes responsible for cognitive decline and neurodegeneration. The identification of FD genes and associated mutations will facilitate discovery of additional causal genes in several ways. First, analysis of newly uncovered or broadened biological pathways will bring to light new dementia candidate genes for more targeted studies. Second, the genetic heterogeneity of FD of unknown cause will be diminished as families are assigned to subgroups on the basis of mutations in newly found FD genes. Third, these genes, in turn, will implicate additional pathways involved and will increase the number of targets for therapeutic interventions. Finally, to benefit the scientific and clinical communities at large and to facilitate corroboration of our findings in other samples, we will deposit FD genotype and phenotype information into the on-line resources GeneTests and the NCBI Variation Databases (dbSNP and dbGaP). In the past two decades genetic studies and positional cloning in families have identified genes for numerous mendelian diseases. Researchers from the ADRC have been in the forefront of this effort and have had a major role in discovery of multiple genes, including PSEN1, PSEN2 and MAPT. This important work currently allows for accurate diagnosis of monogenic dementias, expands our understanding of the biology of neurodegeneration and has laid the foundations for new treatments that might be applicable for common FDs with complex etiology. However for many affected families the causal genes have not yet been found. Many of these families are too small for positional cloning. Previously there were no methods for identification of novel causative genes when DNA from only a few affected persons was available, despite an extensive family history of disease. In this proposal we will capitalize on newly available massively parallel sequencing coupled with advances in genome annotation and bioinformatics to apply mutational cloning of the exome as a powerful and innovative approach for identification of genes involved in FD. Mutational cloning is based on the concept of exclusion mapping. Within a family, all affected individuals must share the mutation responsible for the disease, and all nonshared variants are excluded. From the genomes of already sequenced individuals we estimate that each individual bears approximately 10,000 protein-changing alterations. Because mutations responsible for autosomal dominant forms of FD must be rare they are not likely to be present in dbSNP with a stated frequency. Approximately 90 percent of protein- changing variants are present in dbSNP at this time. This leaves an average of 1000 unique variants for each sequenced individual. This number is an overestimate, as with further increases in resequencing (e.g., the 1000 Genomes Project), the number of less common variants in dbSNP will grow. Current experience with over 2000 cloned mendelian genes strongly supports the contention that disease causing mutations affect highly conserved residues. Some variants will predict conservative amino acid substitutions unlikely to damage the protein. We estimate that 80% of unique variants can be eliminated based on these tenets. With these criteria, approximately 200 candidate variants will remain for each exome, quite similar to the number of genes contained in the minimal linkage region from a positional cloning approach. The exome of the second affected person in the family will exclude a portion of these variants from further consideration. Because sibs share on average 50% of the genome, an affected sib pair has the potential to reduce the number of candidate mutations by half. With even slightly expanded families this number will diminish more dramatically. Cousins share on average 12.5% of the genome and can reduce the number of candidate mutations from 200 to 25. Failure of the variant to cosegregate with disease in the other affected people in the family will eliminate yet more variants. The remaining candidate genes will be prioritized for studies in additional families and dementia cases on the basis of gene function, tissue expression, conservation, predicted effect of the mutation, existence of animal models, involvement in human diseases, and other cues from the literature. We are poised to take maximal advantage of the confluence of infrastructure (University of Washington Department of Genome Sciences), expertise in molecular genetics (Drs. Raskind, Shendure, and Brkanac), existing well- characterized samples (Dr. Bird), and advances in sequencing techniques that now make it feasible to do whole exome screening without preconceived candidate genes. PUBLIC HEALTH RELEVANCE: The goal of this proposal is to identify novel genes that are responsible for mendelian, highly penetrant familial dementia (FD) of either Alzheimer's disease (AD) or frontotemporal dementia (FTD) subtype. We will employ newly available powerful sequencing and bioinformatics techniques to detect causative mutations in families with apparent autosomal dominant dementia where the available affected persons are too few for linkage analyses.