Molecular mechanisms linking cerebrovascular disease pathologies to AD/ADRD in older adults - ABSTRACT CVD pathologies are a major contributor to AD/ADRD in older adults. However, the underlying mechanisms are unknown. To elucidate the underlying mechanisms, interrogating molecular genetic data including transcriptomes and proteomes is required. Prior studies identifying mechanisms underlying Alzheimer’s disease (AD) have examined multi-omics data at gray matter tissues where AD and other neurodegenerative pathologies predominate. However, it is mainly white matter tissue that is damaged by CVD pathologies and risk factors. In addition, white matter changes have also been observed in relation to AD pathology accumulation such as amyloid-β deposits. Since some genetic risk factors express differently in different tissue environments, it is necessary to complement prior studies with generation of multi-omics data extracted from white matter to identify mechanisms underlying contribution of CVD and AD pathologies to AD/ADRD. The overall objective of this proposal is to identify molecular mechanisms underlying contribution of CVD pathologies to AD/ADRD clinical phenotypes. To accomplish this objective, we will leverage clinical, pathological, and omics data of two studies, the Religious Orders Study (ROS, P30AG072975) and the Rush Memory and Aging Project (MAP, R01AG017917). We will complement the currently available ROSMAP omics data obtained from dorsolateral prefrontal cortex (DLPFC) with transcriptomic and proteomic data of deep frontal white matter obtained by this proposal. This proposal is based on the following preliminary findings. 1) some DLPFC proteins that were associated with AD/ADRD clinical phenotypes were also associated with CVD pathologies. 2) White matter disruption in deep frontal area was associated with faster cognitive decline in participants with vascular cognitive impairment. We extend these findings by examining the following specific aims. In Aim 1, we will measure RNA transcription levels at deep frontal white matter (DFWM). Then, we will use computational biology approaches including clustering methods and Bayesian networks to identify RNAs underlying contribution of CVD and AD pathologies to AD/ADRD. In Aim 2, we measure DFWM proteins and examine them using clustering methods and Bayesian networks to identify DFWM proteins underlying contribution of CVD pathologies to AD/ADRD. In Aim 3, we will use other computational biology approaches to identify colocalized signals between DFWM RNA transcriptions (or proteins) data and publicly available GWAS summary statistics for ischemic stroke and Alzheimer’s dementia to link genetic risk factors to the molecular mechanisms. All three Aims will be done using RNA and proteins quantified in 2 separate dataset: a discovery dataset of 300 participants and a validation dataset of 200 participants. In Aim 4, we will replace DFWM RNA transcriptions and proteins with those of DLPFC to identify if RNAs and proteins underlying contribution of CVD and AD pathologies to AD/ADRD are different in gray vs. white matter.
