Hypoconnectivity between locus coeruleus and medial temporal lobe during novelty predicts accelerated Aβ‐related cognitive decline: The locus coeruleus: Nexus of behavioral and cognitive problems in dementia. (7th December 2020)

• Record Type: Journal Article
• Authors: Engels, Nina; Prokopiou, Prokopis C.; Uquillas, Federico d'Oleire; Scott, Matthew R.; Schultz, Aaron P.; Papp, Kathryn V.; Farrell, Michelle E.; Rentz, Dorene M.; Sperling, Reisa A.; Johnson, Keith A.; Jacobs, Heidi I.L.
• Is Part Of: Alzheimer's & dementia. Volume 16(2020)Supplement 2
• Page Start: n/a
• Page End: n/a
• Publication Date: 2020-12-07
• Subjects: Alzheimer's disease -- Periodicals; Alzheimer Disease -- Periodicals; Dementia -- Periodicals; Démence; Maladie d'Alzheimer; Périodique électronique (Descripteur de forme); Ressource Internet (Descripteur de forme); 616.83
• DOI: 10.1002/alz.041323
• Languages: English
• ISSNs: 1552-5260

Abstract

Abstract: Background: Tauopathy can be detected in the locus coeruleus (LC) decades before clinical symptoms or Aβ‐pathology. Following involvement of the LC, tau progresses to the medial temporal lobe (MTL), possibly via hyperactivity or connectivity. Given LC's modulating role on brain function and its early involvement in proteinopathies that affect LC's interaction with other brain regions, it is conceivable that individual variability in LC functional connectivity (FC) with the MTL will be related to Alzheimer's disease (AD)‐related cognitive functioning. The LC plays an important role in novelty processing, learning and memory. Therefore, we aimed to examine whether LC‐MTL FC during novelty predicts Aβ‐related cognitive decline. Method: All participants ( N = 153; number of observations = 772; Figures 1, 2) were clinically normal and part of the Harvard Aging Brain Study. Baseline PiB‐PET (cut‐off = 1.324 DVR), 3T‐fMRI data on a face‐name encoding task and longitudinal neuropsychological PACC5 data were obtained. The fMRI‐task instructed participants to remember face‐name pairs and stimuli consisted of novel and repeated faces. A dilated ex‐vivo validated LC‐template was registered to each individual using ANTs. SPM and ICA‐AROMA were used for motion and physiological artefact removal. An FDR‐corrected gPPI‐analysis (with age, sex, education covariates) established five regressors: novelty; repetition; LC BOLD‐activity; novelty x LC BOLD‐activity; repetition x LC BOLD‐activity. A linear contrast was created by subtracting the repetition interaction regressor from the novelty interaction regressor. LME models assessed whether the association between PACC5 and FC was dependent on PiB (linearly and quadratic), while correcting for age, sex, education and their interactions with time. Bonferroni correction was applied. Result: The gPPI‐analysis demonstrated negative LC‐MTL FC during novelty versus repetition for every region of interest ( p ‐FDR<0.05; Figure 3), thereby indicating hypoconnectivity. The LME‐analysis revealed that hypoconnectivity of the LC‐Parahippocampal Gyrus predicted worse PACC5 performance, which accelerated at elevated PiB levels ( p = 0.013; f = 0.11; Bonferroni‐adjusted; Figure 4). Similar results were apparent for LC‐hippocampus and LC‐amygdala FC. Conclusion: This study demonstrates that the LC is less functionally connected with the MTL during novelty, and that this hypoconnectivity predicts Aβ‐related cognitive decline. This suggests that LC‐MTL hypoconnectivity might act as a marker of AD‐related processes. Future work will assess the relationship between LC FC and tau pathology.