Structural and Functional Brain Connectivity in Middle-Aged Carriers of Risk Alleles for Alzheimer's Disease

Abstract

Single nucleotide polymorphisms (SNPs) in APOE, COMT, BDNF, and KIBRA have been associated with age-related memory performance and executive functioning as well as risk for Alzheimer’s disease (AD). The purpose of the present investigation was to characterize differences in brain functional and structural integrity associated with these SNPs as potential endophenotypes of age-related cognitive decline. I focused my investigation on healthy, cognitively normal middle-aged adults, as disentangling the early effects of healthy versus pathological aging in this group may aid early detection and prevention of AD. The aims of the study were 1) to characterize SNP-related differences in functional connectivity within two resting state networks (RSNs; default mode network [DMN] and executive control network [ECN]) associated with memory and executive functioning, respectively; 2) to identify differences in the white matter (WM) microstructural integrity of tracts underlying these RSNs; and 3) to characterize genotype differences in the graph properties of an integrated functional-structural network. Participants (age 40-60, N = 150) underwent resting state functional magnetic resonance imaging (rs-fMRI), diffusion tensor imaging (DTI), and genotyping. Independent components analysis (ICA) was used to derive RSNs, while probabilistic tractography was performed to characterize tracts connecting RSN subregions. A technique known as functional-by-structural hierarchical (FSH) mapping was used to create the integrated, whole brain functional-structural network, or resting state structural connectome (rsSC). I found that BDNF risk allele carriers had lower functional connectivity within the DMN, while KIBRA risk allele carriers had poorer WM microstructural integrity in tracts underlying the DMN and ECN. In addition to these differences in the connectivity of specific RSNs, I found significant impairments in the global and local topology of the rsSC across all evaluated SNPs. Collectively, these findings suggest that integrating multiple neuroimaging modalities and using graph theoretical analysis may reveal network-level vulnerabilities that may serve as biomarkers of age-related cognitive decline in middle age, decades before the onset of overt cognitive impairment.