Research Article: Interactive effects of dopamine transporter genotype and aging on resting-state functional networks

Date Published: May 8, 2019

Publisher: Public Library of Science

Author(s): Christian Baeuchl, Hsiang-Yu Chen, Yu-Shiang Su, Dorothea Hämmerer, Manousos A. Klados, Shu-Chen Li, Fabio Sambataro.


Aging and dopamine modulation have both been independently shown to influence the functional connectivity of brain networks during rest. Dopamine modulation is known to decline during the course of aging. Previous evidence also shows that the dopamine transporter gene (DAT1) influences the re-uptake of dopamine and the anyA9 genotype of this gene is associated with higher striatal dopamine signaling. Expanding these two lines of prior research, we investigated potential interactive effects between aging and individual variations in the DAT1 gene on the modular organization of brain acvitiy during rest. The graph-theoretic metrics of modularity, betweenness centrality and participation coefficient were assessed in 41 younger (age 20–30 years) and 37 older (age 60–75 years) adults. Age differences were only observed in the participation coefficient in carriers of the anyA9 genotype of the DAT1 gene and this effect was most prominently observed in the default mode network. Furthermore, we found that individual differences in the values of the participation coefficient correlated with individual differences in fluid intelligence and a measure of executive control in the anyA9 carriers. The correlation between participation coefficient and fluid intelligence was mainly shared with age-related differences, whereas the correlation with executive control was independent of age. These findings suggest that DAT1 genotype moderates age differences in the functional integration of brain networks as well as the relation between network characteristics and cognitive abilities.

Partial Text

During alert but task-free states, the brain’s sensorimotor and higher cognitive systems display organized temporal correlations between spontaneous fluctuations of brain activity in different brain regions. Over the past decade, functional magnetic resonance imaging (fMRI) research on the organized patterns of spontaneous brain activity suggests that resting-state functional connectivity (rsFC) is a promising avenue for investigating the dynamics of coherent brain activity patterns that are organized into distinct systems or networks (for a recent review see [1]). The rsFC can be calculated using different methodologies, including: a seed-based approach by selecting voxels in the regions of interests (ROI) as seeds for cross correlational analyses (e.g., [2], [3]), independent component analysis (ICA; [4]), as well as the more recent clustering methods (e.g., [5]). A multitude of factors have been shown to be associated with interindividual differences in the patterns of rsFC, notably: brain maturation (for a review see [6]), brain aging (for review see [7]), neuromodulation (e.g., [8]), and clinical conditions (for reviews see [9], [10]).

As an overview, results for the metric PC showed significant age x genotype interactive effects after FDR correction and are presented in details with follow-up analyses and corresponding figures below. Results for the other two metrics (Q & BC) did not show any significant effects after FDR correction, and are thus only briefly described below.

Before discussing the results, the key findings are summarized here to provide an overview. At the brain-wise level, neither age nor genotype main effects were observed in the measures modularity (Q) and betweenness centrality (BC). In contrast, however, the measure PC, which quantifies the extent to which the network connectivity is evenly distributed across modules, yielded significant effects of age, genotype and their interactions. Specifically, the rsFC is less evenly distributed across modules (lower values of PC) in older than in younger adults in carriers of the DAT1 anyA9 genotype, but not in A10 homozygotes. Furthermore, follow-up analyses revealed that the nodes contributing to the age and genotype effects with respect to PC mainly resided in the DMN network. None of the remaining six cortical networks using the Yeo et al. [50] parcellation–i.e., the visual (VIS), somatomotor (SOM), limbic (LM), dorsal attention (DOR), ventral attention/saliency (VEN), and frontoparietal (FPC) networks–exhibited age or genotype effects. Moreover, the values of PC correlated with individual differences in general cognitive ability and executive control function in DAT1 anyA9 carriers, but not in A10 homozygotes. The functional relevance of PC in general cognitive ability is mainly shared with the age effect, whereas PC predicted individual differences in executive control above and beyond the effects of age.




Leave a Reply

Your email address will not be published.