Translational investigations of the intersection between Alzheimer’s disease and epilepsy: mechanistic insights and treatment opportunities

Abstract

Alzheimer’s disease (AD) is a neurodegenerative disease characterized by progressive cognitive deficits compromising patients’ execution of daily activities. The incidence of unprovoked seizures has been reported to be higher in AD compared to healthy age matched population and is considered a risk factor for the development of acquired epilepsy. Abnormal electrical activity has also been correlated with increased cognitive decline in these patients and may be associated with a faster progression of AD symptoms. Thus, understanding how seizures develop in AD and how to treat them would be of extreme relevance for developing more effective therapies. The work presented in this thesis used a bidirectional approach to investigate the relationship between epilepsy and AD and potential treatment options.

First, three antiepileptic drugs were assessed in their ability to delay the development of acquired epileptogenesis in the Tg2576 mice. These widely studied AD mice overexpress the human form of the amyloid precursor protein (APP) with a mutation that results in increased deposition of amyloid plaques with age. The amygdala kindling model was used to induce epileptogenesis and previous studies have shown that these mice have increased susceptibility to kindling induced seizures. The antiepileptic drugs brivaracetam (BRV), levetiracetam (LEV) and lacosamide (LAC) were administered via subcutaneously implanted osmotic pumps. BRV and LEV act through ligand binding to the SV2A synaptic vesicle protein and LAC through Na+ channels. After 28 days of continuous treatment with either one of the drugs or vehicle (VEH) animals were submitted to daily electrical stimulations, also under treatment, and the only drug able to slow down the epileptogenic process was BRV.

Therefore, in the second study, the effect of BRV was further investigated. Mice were then pre-treated for 28 days, but pumps were removed, and drug allowed to wash-out for one week before kindling. BRV demonstrated a long-lasting effect and treated Tg2576 mice required significantly more stimulations to experience the first class 5 seizure. Investigations of the potential mechanisms underlying this effect revealed that BRV is not acting through changes in the expression levels of APP protein or SV2 mRNA.

Lastly, with an approach in the opposite direction, a histopathological evaluation of resected tissue from patients with temporal lobe epilepsy (TLE) aimed at correlating the cognitive deficits often found in these patients with the presence of AD lesion hallmarks (tau and amyloid pathology). However, no significant tau deposition was found and although amyloid plaques were present in a higher proportion of patients compared to other similar studies, there was no strong evidence to suggest it is the mechanism responsible for cognitive impairments.

In conclusion, network treatment strategies, such as antiepileptic drugs, could be useful in the treatment of seizures in AD, and in this study a potential antiepileptogenic effect was demonstrated for BRV in an AD-acquired epilepsy model. We have also shown that amyloid and tau pathology may not be the main underlying cause of cognitive deficits in TLE, but understanding these mechanisms might also advance our understanding of cognitive decline in AD.