Revealing Epistasis in Alzheimer Disease

阿尔茨海默病(AD)是导致老年人痴呆的主要原因。识别阿尔茨海默病的易感基因将有助于风险评估、诊断和了解疾病的病因。几个染色体区域已被确定为含有晚发性AD的基因，但只有一个基因APOE一直被证明与疾病风险直接相关。虽然载脂蛋白E可能解释了晚发性阿尔茨海默病遗传效应的一半，但其余的易感基因一直难以识别。大量的候选基因已经通过对无关个体的病例对照测试或基于家庭的关联测试进行了关联测试，但不同研究的结果差异很大。我们的假设是，在决定AD风险方面，多个基因之间的上位性效应比任何单个基因的独立效应发挥着更重要的作用。许多候选基因关联研究的多个阳性和阴性报告实际上可能是由于研究设计只评估每个候选基因的主效应，而这些主效应与所有其他基因无关。这项建议的目的是确定上位性或非加性基因-基因相互作用在晚发性AD风险中的作用。为了实现这一目标，我们建议对十个AD候选基因中的几个单核苷酸多态进行鉴定和分型，这些基因的关联研究一直存在冲突。我们将使用以家庭为基础的设计，使用受影响和未受影响的兄弟姐妹，并建议研究800个兄弟姐妹，大约1600个人。我们将把最近开发的系谱不平衡检验(PDT)扩展到在基因水平上测试关联性，并允许将其纳入新的多因素降维(MDR)方法。为了测试这些数据中复杂的遗传交互作用，我们将使用这种改进的MDR-PDT方法。我们预计这些结果将至少解释阿尔茨海默病候选基因关联研究中出现的一些不一致之处。此外，从拟议的研究中获得的知识将对预防和治疗阿尔茨海默病的发生、发展和严重程度的公共卫生努力具有非常宝贵的价值。

Alzheimer disease (AD) is a leading cause of dementia in the elderly. Identifying susceptibility genes for AD will aid in risk assessment, diagnosis and understanding the etiology of the disease. Several chromosomal regions have been identified as harboring genes for late-onset AD, but only one gene, APOE, has been demonstrated consistently to be directly involved in disease risk. While APOE may explain up to half of the genetic effect in late-onset AD, the remaining susceptibility genes have been difficult to identify. A large number of candidate genes have been tested for association using either case-control tests in unrelated individuals or family-based association tests, but the results have varied greatly between studies. It is our hypothesis that epistatic effects among multiple genes play a more important role in determining risk of AD than the independent effects of any single gene. The multiple positive and negative reports for numerous candidate gene association studies may in fact be due to study designs that only evaluate each candidate gene for main effects that are independent of all other genes. The goal of this proposal is to determine the role of epistasis or nonadditive gene-gene interactions on the risk of late-onset AD. To accomplish this goal, we propose to identify and genotype several single nucleotide polymorphisms in ten AD candidate genes for which there have been conflicting association studies. We will use a family-based design using affected and unaffected siblings and propose to study 800 sibships containing approximately 1600 individuals. We will extend the recently developed pedigree disequilibrium test (PDT) to test for associations at the genotypic level and allow its incorporation into the new multifactor dimensionality reduction (MDR) method. To test for complex genetic interactions in these data, we will use this modified MDR-PDT method. We anticipate that these results will explain at least some of the inconsistency arising from Alzheimer disease candidate gene association studies. Further, knowledge gained from the proposed research will be invaluable for public health efforts to prevent and treat the initiation, progression, and severity of Alzheimer disease.