The search for common genetic determinants of pre-clinical Alzheimer's disease-related traits
Etchill, Eric W.
MetadataShow full item record
Background: As Alzheimer's disease (AD) becomes an increasingly urgent public health problem, much effort has been made by medical researchers and public health professionals to further our understanding of AD etiology and causal factors that contribute toward its onset, progression, and severity. AD research has recently focused on identifying and examining preclinical traits associated with the onset and progression of AD with the goals of early detection discovery, mechanism elucidation, and ultimately treatment and prevention. By understanding the genetic components that make one susceptible to pre-clinical traits that are correlated with future development of AD, we can increase our understanding of what and how genes are involved in the molecular pathways as well as identify risk factors that contribute to the onset of a presently elusive, costly, and devastating disease. Specific Aims: This study sought to complete the following; identify MRI and cognitive endophenotypes that are co-heritable with the future onset of Alzheimer's disease; determine cross-trait co-heritability of endophenotypes heritable with AD; select a co-heritable endophenotype combination that has predictive value for AD development; assess which potential genes may be responsible for the co-heritability of this endophenotype combination and prepare for further candidate gene analysis. Methods: Data from the Framingham Heart Study was analyzed to determine heritability of MRI and cognitive endophenotypes with AD development, and co-heritability of the traits with each other. A bivariate, Genome-Wide Association Study (GWAS) was then conducted on the selected MRI-cognitive endophenotype combination to determine which single nucleotide polymorphisms (SNPs) are most associated with co-heritability of the selected trait combination. The results of the GWAS directed further study, including gene-based association studies and assessment of protein interaction pathways, that resulted in selection of potential candidate genes. Results: Twenty-four MRI and cognitive traits were found to be heritable with future AD onset. The most co-heritable MRI and cognitive trait combination among the AD-heritable traits was temporal horn brain volume (THBV) and trail-making test B (TrB) scores. This combination was selected to drive our hypothesis of a significant association between pre-clinical endophenotypes and future AD. The GWAS results indicate that several SNP markers are associated with the co-heritability of THBV and TrB scores. Subsequent regional analyses implicate several genes likely to be causally involved in expression of this endophenotype expression, including SPON1, which directly interacts with APP. Based on a combination of evidence, the genes selected for candidate gene analysis are SPON1, DICER1, GPR177, ADCY9, VAMP3, and FOXF2. Conclusion: Our study validated several previous findings of MRI and cognitive heritability with AD. We then used estimates of co-heritability to successfully run a bivariate analysis of two pre-clinical traits to identify genes explaining their impact on future AD development. The genes selected for candidate gene analysis appear to be involved in amyloid precursor protein processing, which is the primary pathway that leads to plaque deposition and AD pathology. Further study of these genes, especially SPON1, will allow for determination of replicable findings, assessment of whether the candidate genes are causative, and determination of whether these genes are also related to progression and severity of AD symptoms. Importantly, this method of gene discovery, through bivariate analyses of pre-clinical traits, allows for insights into disease susceptibility, molecular pathways, early detection, and more focused treatments.
Thesis (M.A.)--Boston University PLEASE NOTE: Boston University Libraries did not receive an Authorization To Manage form for this thesis or dissertation. It is therefore not openly accessible, though it may be available by request. If you are the author or principal advisor of this work and would like to request open access for it, please contact us at firstname.lastname@example.org. Thank you.