Search

Aged individuals often accumulate multiple brain pathologies that contribute individually and synergistically to cognitive decline and dementia. Alzheimer’s disease (AD) is the most commonly diagnosed form of dementia, constituting over 50% of dementia diagnoses, and poses an enormous burden on human health and well-being. In the United States alone, over six million individuals are living with AD, and the disease imposes financial costs of treatment and care of over $300 billion annually. The most common form of AD, late-onset AD (LOAD), presents in individuals aged 65 years and older and is highly heritable, with twin and family studies estimating its heritability at ~60%. Genome-wide association studies (GWAS), in which millions of genetic variants are individually assessed for association with a trait through generalized linear regression models, have been an enormously successful approach for studying the genetic risk of LOAD. Recent GWAS of AD have included upwards of 700,000-1,200,000 participants and have identified more than 70 individual genetic risk loci for LOAD. However, the simple clinical or proxy definitions of LOAD used in these studies do not reflect the complexity of the underlying disease. In addition to the pathognomonic LOAD neuropathologies of amyloid-beta plaques and tau neurofibrillary tangles, the majority of dementia patients also have TDP-43, alpha-synuclein, cerebrovascular pathology, or some combination thereof. Individual neuropathologies likely have both independent and shared genetic risk factors, and studying precisely defined neuropathologic phenotypes through GWAS can act as a complementary approach to large case-control based studies; however, neuropathologic endophenotypes have been relatively neglected in GWAS relative to clinical phenotypes. Brain volumes are another class of endophenotypes associated with a variety of important life and health outcomes, including LOAD. GWAS have identified genetic loci associated with brain MRI volumes measured via magnetic resonance imaging (MRI), although these studies have either exclusively studied genetic variants with minor allele frequencies (MAF) ≥1% or have primarily focused on participants of European ancestry. Investigating the relationship between rare genetic variants (MAF In the first study, we performed a genome-wide association study of brain arteriolosclerosis. We then performed a gene-based analysis to prioritize genes and Bayesian colocalization analyses to identify functional effects of risk loci. In the second study, we expanded our investigation, performing GWAS and functional analyses on 11 neuropathologic endophenotypes. We then performed targeted analyses to confirm association between DNA methylation and RNA expression with neuropathologic endophenotypes. In the third study, we investigated the genetic factors of four brain volume endophenotypes (intracranial volume, total brain volume, hippocampal volume, and lateral ventricular volume) using whole-genome sequence data in a cohort of participants with diverse ancestry.

The Mediterranean diet (MedDiet) has been linked with preservation of brain structures. Brain structural changes associated with cognitive impairment may occur during midlife, and timing of exposure may be critical. This study aims to examine the association of cumulative average MedDiet score from early through middle adulthood, and MedDiet scores at individual timepoints in relation to the midlife MRI brain measures. This study will also assess statistical mediation of brain measures on the association between cumulative average MedDiet score adherence and four cognitive domains: executive function, global cognition, processing speed, and verbal memory. 515 participants of the Coronary Artery Risk Development in Young Adults (CARDIA) study were included. Cumulative average MedDiet scores were calculated by averaging scores from baseline, year 7, and year 20 (mean age 25, 32, and 45, respectively). MRI brain scans were obtained at year 25 (mean age 50). General linear models adjusted for demographic variables, lifestyle factors, and vascular and metabolic comorbidities were used to examine all exposure-outcome relationships. Cumulative average MedDiet scores measured in tertiles were not associated with brain volumes. MedDiet scores examined separately at years 0, 7, and 20 were not associated with midlife brain measures. Higher cumulative average MedDiet scores were associated with better executive function (β= -0.38, ptrend