Genetic risk factors of Alzheimer's Disease disrupt resting-state functional connectivity in cognitively intact young individuals

Abstract

Background: Past evidence shows that changes in functional brain connectivity in multiple resting-state networks occur in cognitively healthy individuals who have non-modifiable risk factors for Alzheimer's Disease. Here, we aimed to investigate how those changes differ in early adulthood and how they might relate to cognition.

Methods: We investigated the effects of genetic risk factors of AD, namely APOEe4 and MAPTA alleles, on resting-state functional connectivity in a cohort of 129 cognitively intact young adults (aged 17-22 years). We used Independent Component Analysis to identify networks of interest, and Gaussian Random Field Theory to compare connectivity between groups. Seed-based analysis was used to quantify inter-regional connectivity strength from the clusters that exhibited significant between-group differences. To investigate the relationship with cognition, we correlated the connectivity and the performance on the Stroop task.

Results: The analysis revealed a decrease in functional connectivity in the Default Mode Network (DMN) in both APOEe4 carriers and MAPTA carriers in comparison with non-carriers. APOEe4 carriers showed decreased connectivity in the right angular gyrus (size = 246, p-FDR = 0.0079), which was correlated with poorer performance on the Stroop task. MAPTA carriers showed decreased connectivity in the left middle temporal gyrus (size = 546, p-FDR = 0.0001). In addition, we found that only MAPTA carriers had a decreased connectivity between the DMN and multiple other brain regions.

Conclusions: Our findings indicate that APOEe4 and MAPTA alleles modulate brain functional connectivity in the brain regions within the DMN in cognitively intact young adults. APOEe4 carriers also showed a link between connectivity and cognition.

Keywords: Alzheimer’s disease; Cognition; Functional connectivity; Functional neuroimaging; Large-scale networks; Risk factors.

Fig. 1

Identified resting-state networks. A DMN. B Dorsoattentional network. C Sensorimotor network. D Salience network. E Frontoparietal network. F Language network. G Visual network

Fig. 2

Violin plots of the functional connectivity differences in the DMN calculated based on the peak connectivity values in each cluster. A APOEe4 carriers vs non-carriers in Cluster 1. B MAPTA carriers vs non-carriers in Cluster 2

Fig. 3

The spatial maps of clusters that exhibited significant differences in connectivity between carriers and non-carriers. A Cluster 1 comprising of the right angular gyrus and superior occipital cortex that showed significant differences between APOEe4 carriers and non-carriers. B Cluster 2 comprising of the left middle temporal gyrus that showed significant differences between MAPTA carriers and non-carriers

Fig. 4

The spatial map of clusters that exhibited an additional reduction of functional connectivity from Cluster 2

Fig. 5

Spearman’s correlations between functional connectivity in respective clusters and Stroop performance (i.e., the accuracy in the colour-incongruent condition) between groups with fitted linear regression lines. APOEe4 carriers showed a positive relationship between connectivity in Cluster 1 and Stroop performance (R = 0.42, p = 0.037), while no other relationship was observed in APOEe4 non-carriers (R = − 0.16, p > 0.05), MAPTA carriers (R = 0.08, p > 0.05), and MAPTA non-carriers (R = 0.2, p > 0.05)