Authors :
Presenting Author: Anna Harutyunyan, PhD – Monash University
Alison Anderson, PhD – Monash University; Patrick Kwan, PhD – Monash University; Nigel Jones, PhD – Monash University
Rationale:
There is increasing recognition that seizures and epilepsy frequently occur in Alzheimer’s disease (AD) patients, and this co-occurrence is associated with accelerated cognitive decline and different treatment responsivity in AD. However, the mechanisms of this interaction remain unknown. Previously, we developed an animal model of dual pathology (epilepsy and AD) by establishing a recurrent seizure phenotype in 5xFAD mice. Here, we aimed to investigate the molecular mechanisms driving the synergy between recurrent seizures and AD pathology by integrating data from multiple modalities (molecular, histopathological, electrophysiological and behavioural).
Methods:
Female transgenic 5xFAD mice (N=20) and WT littermates (N=22) underwent electrical amygdala kindling or were treated as sham. In this experimental paradigm the kindled 5xFAD mice represent the subpopulation of AD patients who develop recurrent seizures and are at risk of accelerated cognitive deterioration, while WT and sham 5xFAD groups serve as respective controls. We employed immunohistochemistry and behavioral testing (Y-maze) to evaluate the effect of recurrent seizures on amyloid plaque load and cognitive performance, respectively. The hippocampal transcriptome was examined through RNA-sequencing. Correlation network analysis was employed to integrate electrophysiological (EEG), cognitive/behavioural, histopathological, and transcriptomic data into a multi-modal network-based model.
Results:
5xFAD mice showed increased hyperexcitable phenotype (p< 0.001) and impaired spatial memory (p< 0.05) compared to WT group. Compared to shams, the kindled 5xFAD showed enhanced amyloid deposition, evidenced by increased amyloid plaque area (p< 0.01) and vascular Aβ deposition in the leptomeningeal vessels. Differential expression analysis with nested comparisons identified a group of 326 genes that responded synergistically in the dual pathology group compared to all other groups (FDR< 0.05, FC >2). Correlation network analysis identified modules of immediate early genes (IEG) and complement cascade-related proteins showing significant (p< 0.00001) correlation with dual-pathology group. Notably, the regulatory hub gene of IEG module, Pcdh8 (Protocadherin 8) is involved in elimination of dendritic spines and was profoundly overexpressed in the dual pathology group.