Grant: Canada Postdoctoral Research Award (CPRA) program

Alzheimer's disease (AD) is a major threat to the ageing population due to its high prevalence (10 - 35%) among people aged 65 and above. AD impairs multiple cognitive aspects, including memory, reasoning, language ability and emotional well-being. The cause of AD has been attributed to neuronal death that is accompanied by aberrant protein aggregations, such as beta-amyloid plaques. However, astrocytes, a type of non-neuronal brain cell as abundant as neurons, are equally affected in AD. Astrocytes are known to support and regulate normal neuronal functions. Hence, they are also essential to memory and cognition. Emerging evidence suggests that astrocyte pathological phenotypes can be found at early stages of AD, and astrocytes would excessively engulf neuronal components. This raises the need to investigate the changes in the astrocyte-neuron interface in AD. The challenge of doing so is that astrocytes are one of the most complex cell types, and they possess important processes with sizes in the sub-micron scale. In this project, I will use Focused Ion Beam-Scanning Electron Microscopy (FIB-SEM), a powerful microscopy technology, to visualize astrocytes at nanometer resolution. An artificial intelligence will be trained to extract all the relevant astrocyte areas from the large datasets automatically. I will compare the structural features of the astrocyte-neuron interface in mice and non-human primates, marmosets, in healthy and AD situations. I will reconstruct the extracted EM images into digital 3D models for analyses of (1) cellular morphologies, (2) cellular content or organelle distributions and (3) astrocyte contacts with neurons. This project will help reveal new details about the structure-function relationship of astrocytes and the ways they interact with neurons in healthy and AD situations. The use of marmosets in the project also helps to bridge the gap between mice and humans and increase the translatability of findings in mice to humans.