Neuronal network changes in Alzheimer's disease

Alzheimer’s disease (AD) is characterized by neuronal network dysfunction. Functional imaging studies showed that oscillatory activity is altered in AD patients. Interestingly, network dysfunction can be observed at an early stage of the disease: subjects with mild cognitive impairment, or with subjective cognitive decline, already show alterations in network activity. The relative ease with which network dysfunction can be detected, as well as its presence during early stages of the disease, make it a valuable starting point for the development of treatments. To further facilitate the development of network-based therapies we need a better understanding of the mechanisms underlying network dysfunction as well as how disease progression in mice translates to the different stages along the AD continuum in humans. Chapter 1 provides an introduction into the field of neuronal network impairments in AD patients and mouse models of AD. In chapter 2, we explored the mechanisms underlying network dysfunction in AD mice with a focus on early disease stages. As a recent study found that hippocampal parvalbumin (PV) interneurons, which play an important role in the generation of network oscillations, are hyperactive at an early disease stage, and that restoring the activity of these cells can mitigate cognitive symptoms in the APP/PS1 mouse model of AD, we investigated the mechanisms underlying PV hyperexcitability. Sequencing data indicated that altered expression of genes related to potassium ion transport might underlie early PV hyperexcitability. In chapter 3, we took a closer look at the network alterations taking place in APP/PS1 mice. Local field potential (LFP) recordings, performed in 3-12-month-old mice, showed that network alterations occurred already at an early age and persisted over time. To obtain a better understanding of how AD-related changes in network activity translate between mice and humans, we also performed MEG recordings in pre-symptomatic human subjects carrying similar mutations in the APP and PSEN1 genes. The very different and often opposite nature of network changes observed in mice and humans raises questions regarding the underlying mechanisms. Findings in mice should be interpreted with care, and additional experiments, for example in AD mouse models bearing both amyloid and tau pathology, are needed to further establish potential translational links between both species. In chapter 4, we explored network alterations in the context of AD-relevant behavior. AD mouse models show a deficit in nesting behavior, an essential behavior that is shared by many species. Even though the nesting test is widely used in the AD field, relatively little is known about how a deficit in nesting behavior relates to AD. A detailed behavioral characterization of nesting behavior in APP/PS1 and wildtype mice showed that, even though APP/PS1 mice were capable of performing the different behavioral components of nesting behavior, they frayed their nestlet with a delay compared to wildtype mice and showed delayed and reduced arranging (pulling in and fluffing) of nest material. These findings suggested that the deficit in nesting behavior in APP/PS1 mice might be related to reduced motivation to build a nest. In chapter 5, we aimed to design a behavioral task that would allow to measure cognitive function and record network alterations in parallel in a longitudinal manner. To this end, we developed an automated figure-8-maze to test spatial working memory. However, we did not find clear differences in memory performance between APP/PS1 and wildtype mice from 2 to 6 months of age. In chapter 6 we provide a systematic review of neuronal network changes in mouse models of AD. This review will help to provide direction for future studies to improve translatability of AD-related network changes between mice and humans