부산대학교 도서관

The overall aim of this thesis was to investigate the role of the Tuberous Sclerosis (TSC) complex in selective neuronal vulnerability (SNV) of the CA1 region in Alzheimer's disease (AD). This aim was addressed through a combination of neuropathological and in vitro studies. We first investigated whether a Hamartin (TSC1) loss of function predisposes rat hippocampal neurons to Aβ1-42-induced cell death and tau phosphorylation. We employed a rat hippocampal model exposed to TSC1 shRNA or a scrambled control vector, and Aβ1-42 or vehicle. We subsequently assessed cell death over time by high-content imaging and tau phosphorylation by Western blot at both AT-8 and S262 sites. Next, we applied either rapamycin or a human TSC1 (hTSC1) overexpression vector in an attempt to rescue TSC1-induced cell death. We observed that the largest driver of cell death was TSC1 loss of function, which could be rescued by hTSC1 overexpression and to an extent rapamycin treatment. The latter also improved viability in control cells. Furthermore, tau phosphorylation was not altered in our system. Lastly, we demonstrated through overlapping differentially expressed genes (DEGs) acquired through RNA-seq of human material and rat hippocampal experiments, that a TSC1 loss of function led primarily to a reduction of genes in the metabotropic glutamate pathway that has previously been shown to be fundamental for long-term depression (LTD). We concluded that a TSC1 loss of function is pivotal to hippocampal neuronal cell death but not necessarily in an mTORC1-dependent manner. Next, we proceeded to investigate the origin of the TSC1 loss of function in the pyramidal neurons of the hippocampal CA1 region in the AD brain, and the subcellular localisation of TSC1 aggregates. We hypothesised that TSC1 would be aggregated due to a loss of its binding partner and molecular chaperone, Tuberin (TSC2), and that it would be located either to the lysosomal compartment or granulovacuolar degeneration (GVD) bodies. We first investigated the distribution of lysosome-associated membrane protein (LAMP1)- and GVD-positive pyramidal neurons in the human hippocampus, and subsequently the distribution of TSC2-positive pyramidal cells in the human hippocampus. We observed a discordant pattern of GVD and LAMP1-positive cells in the CA1 region but observed a profound TSC2 loss in the CA1 region in AD cases. We furthermore demonstrated that TSC2-/- embryonic stem cell (ESC)-derived cortical neurons possessed more TSC1 aggregates that localised to the Golgi apparatus compared to TSC2+/+ neurons. We therefore concluded that TSC1 aggregation in the CA1 region of the AD hippocampus is directly related to the loss of TSC2. Lastly, we investigated the impact of TSC complex failure on lysosomal biogenesis and Aβ1-42 uptake. To this end, we used a combination of immunocytochemistry, live imaging, Western blot and flow cytometry on ESC-derived cortical neurons. We demonstrated that, although TSC2-/- ESC-derived cortical neurons retain more cytoplasmic transcription factor EB (TFEB), the endolysosomal and autophagosomal compartments were upregulated in TSC2-/- neurons compared to TSC2+/+ neurons. Similarly, we demonstrated an upregulation of the autophagosomal compartment in the presence of mTORC1 hyperactivation. Next, we showed that TSC2-/- neurons contain a proteostatic defect and accumulate aggresomes that are lined with lysosomes. Finally, we showed that TSC2-/- cells take up more fluorescent Aβ1-42 than TSC2+/+ cells. This effect was mainly driven by dead or dying cells. Based on these results, we concluded that the reduction of lysosomes in the CA1 region of AD cases compared to the CA3 region, was likely not related to a TSC2 loss of function. Nevertheless, TSC2-deficient neurons did display proteostatic defects that may be relevant to AD. Overall, we concluded that a loss of function of the TSC complex in AD was an important factor in neuronal CA1 vulnerability. However, the exact mechanism by which this vulnerability is conveyed remains elusive and requires further exploration. In this thesis, we propose that a loss of TSC2 is 'a hit' in a multi-hit process that leads to AD.