MicroRNAs as Diagnostic and Prognostic Biomarker of Alzheimer's Disease

点击翻译按钮获取中文摘要

ABSTRACT Preventative and therapeutic strategies for late onset Alzheimer’s Disease (AD) depend on access to reliable biomarkers for early detection, ideally minimally invasive and inexpensive with more invasive and costly ones reserved for subsequent diagnostics. Despite the progress in the development of blood biomarkers to detect AD- associated beta-amyloid or Tau pathology, the challenges remain because AD is characterized by dysregulated neural gene expression triggered by combinations of genetic and environmental risk factors. Genome-environment interactions are orchestrated by epigenetic processes that include the function of microRNAs (miRNAs, miRs) which is to regulate gene-expression and proteostasis. While altered microRNA expression is repeatedly observed in AD biospecimens, small sample size and lack of mechanism maintain the gap of knowledge whethermicroRNAs could serve as diagnostic and prognostic AD biomarker. The unprecedented combination of samples from Alzheimer’s Disease Neuroimaging Initiative (ADNI), Framingham Heart Study (FHS), and Germany’s cognitive impairment and dementia study (DELCODE) in this application provides us with the opportunity to bridge that gap. In Aim 1we test the hypothesis that distinct patterns of circulating plasma microRNA levels differentiate among cognitively normal (CN), individuals with mild cognitive impairment (MCI), and dementia patients (AD) meeting clinical criteria for AD. Cross-sectional plasma samples for microRNA-Seq analysis were obtained from 847 participants in ADNI-1/GO/-2 and from 585 participants in DELCODE. We will use complementary computational approaches to compare: AD vs CN, MCI vs CN, and AD vs MCI samples and identify microRNA signatures of MCI and AD. The results will be integrated with the data from the FHS cohort comprising 48 participants with diagnoses confirmed postmortem and 3 longitudinal plasma samples separated by years; the participants who developed AD provided 2 samples before and 1 after their clinical AD diagnosis. We will use statistical and machine learning approaches to generate miRNA-based classifiers. Aim 2 will determine if differential expression of circulating candidate biomarker miRs that discriminate between AD and CN subjects can be observed in vulnerable cells, laser- microdissected from postmortem cortices of Aim 1 FHS participants. In these cells we will quantify the expression of miR-181a-5p, miR-148-3p and miR-146a-5p, a 3-miR plasma signature that uncovered patients at risk for converting from MCI to AD in our exploratory study in press. We will also examine the expression of at least 3 top microRNA AD biomarker candidates validated in Aim 1 analyses. In Aim 3 we will uncover the functions of candidate biomarker microRNAs in vitro and in vivo by manipulating levels of biomarker candidates to establish their functional readouts in human induced pluripotent stem cell- derived cortical neurons and glia and in next generation AD mouse models. Our comprehensive interdisciplinary analyses will integrate observational and experimental data to determine how MCI- and AD-associated microRNA signatures reflect AD-associated pathology and serve as a biomarker for developing AD.