Oligodendrocyte dysfunction in dementia and Alzheimer's disease

Alzheimer's disease (AD) is the most common cause of dementia, occurs predominantly in those over the age of 65 and impacts many aspects of daily life (Mastroeni et al., 2011). A key pathological hallmark of AD pathogenesis is the accumulation of brain amyloid beta peptides which aggregate in plaque formations, and are accompanied by inflammation and focal demyelination (Galasko & Montine, 2010; Mitew et al., 2010). Alzheimer's disease progression manifests as clinical symptoms including memory loss, social dysfunction, and alterations in executive function, following a loss of neurons and synapses in respective brain regions (Arn‚àö¬8iz & Almkvist, 2003; Bondi, Edmonds, & Salmon, 2017; Klimova, Maresova, Valis, Hort, & Kuca, 2015). While neuron loss and inflammation in Alzheimer's disease has been extensively researched, there has been limited study on myelin changes. In the central nervous system, the oligodendrocytes are myelin-forming glial cells which mature from oligodendrocyte precursor cells (OPCs) by membrane and cytoskeletal alterations. They provide metabolic support for neurons and mediate continual remodelling of myelin sheaths, which occurs throughout life in response to learning (Funfschilling et al., 2012; Kaplan et al., 2001; Young et al., 2013). Previous studies of Alzheimer's disease, including animal models, show loss of myelin in the vicinity of amyloid plaques (Bartzokis, 2011; Dean 3rd et al., 2017; Mitew et al., 2010). Plaque forming amyloid beta peptides are generated from cleavage of the amyloid precursor protein (APP) by ˜í‚â§-secretase and ˜í‚â•-secretase enzymes resulting in a number of peptides; however A˜í‚â§40 and A˜í‚â§42 are suggested to be key in AD progression, with A˜í‚â§42 being more hydrophobic and fibrillogenic and most commonly present within the amyloid plaques (Selkoe, 2001). Several mutations in the APP and ˜í‚â•-secretase genes have been described in familial AD. These mutations result in increased amyloid levels in the brain and have been utilized in human studies and animal models to further our understanding of AD pathogenesis. However, there have been a limited number of studies investigating how increased brain amyloid affects myelination in the brain, including generation of new oligodendrocyte lineage cells and maturation of OPCs. This thesis seeks to understand the role that increased levels of amyloid beta peptides have on the maturation of oligodendrocytes, and the capacity for myelination induced by learning a new task. Previous work in our laboratory has demonstrated demyelination in cortical layer V in mouse models and human cases of AD (Mitew et al., 2010). However, the cause and effect of amyloid on oligodendrocyte lineage cells cannot be directly studied in humans. Rodent models represent useful paradigms for investigating the effects of increased brain amyloid on specific cell types. Therefore, in this study, I first investigated myelin and oligodendrocyte lineage cells in a rat model of induced amyloidosis. For this study, I used the TgF344-AD rat model, which overexpresses human APP with the Swedish mutation (KM670/671NL; APP\\(_{swe}\\)) as well as the human Presenilin-1 ˜ívÆ exon 9 mutation (PS1˜ívÆE9), both driven by the mouse prion promoter. Previous studies have demonstrated amyloid plaque formation in the hippocampus and cortex between 6 and 26 months of age and tau pathology at 16 months in this model (Cohen et al., 2013). I first characterized alterations in the TgF344-AD rat model of amyloidosis, at 18-25 months of age when extensive amyloid plaques were present. I hypothesised that the presence of elevated A˜í‚⧠would alter the myelination of axons in these animals. I quantitated the total oligodendrocyte numbers, in the cortex, hippocampus and corpus callosum of transgenic and wildtype (WT) animals using immunohistochemistry and examined the axons in the corpus callosum using electron microscopy. Analysis of changes to total oligodendrocyte numbers by two-tailed student t-test demonstrated significant (p<0.05) decreases in the cortex, alongside a significant increase inproteolipid protein (PLP) expression in the corpus callosum of TgF344-AD rats. No differences in myelination of axons were present in the corpus callosum when comparing WT and TgF344-AD rats in this study. To further understand the relationship between amyloid beta, oligodendrocytes, and myelination, I investigated the effect of amyloid beta on oligodendrocyte maturation. I hypothesised that extracellular amyloid beta peptides would alter health and maturation of OPCs. For this aim, I developed an in vitro cell culture model of OPC maturation. OPCs were derived from post-natal Sprague Dawley rat mixed glial cultures and oligodendrocyte were cultured in maturation media across a 5-day period. During this time, OPCs undergo extensive alterations to their morphology and expression of maturation stage specific markers. Previous studies have demonstrated that A˜í‚â§40 and A˜í‚â§42 are toxic to oligodendrocytes in vitro at high concentrations (10 to 20 ˜í¬¿M) and generate an inflammatory response (Horiuchi et al., 2012; J. Xu et al., 2001). Therefore, I examined the effects of lower concentrations of A˜í‚â§40 and A˜í‚â§42 on cell health, and oligodendrocyte development through examining alterations in morphology and the expression of markers specific to OPC development. My study demonstrated that A˜í‚â§40 reduced the length and number of oligodendrocyte branches following a 24-hour exposure, however, upon removal of the amyloid, the cells rapidly recovered, suggesting an acute response to amyloid in maturing oligodendrocytes. Conversely, 24-hour exposure to amyloid peptides early in their development, increased the number of branches present by 5 days in vitro when the oligodendrocytes were more mature. These data suggest amyloid beta interactions may potentially modulate proliferation of oligodendrocyte processes, which result in alterations in the oligodendrocyte maturation process. Lastly, I examined the effect of brain amyloid beta peptides on the ability of rodents to learn a new task, which involves the proliferation of OPCs and the laying down of new myelin. I hypothesised that the presence of increased levels of amyloid beta peptides in the brain would affect maturation of oligodendrocyte lineage cells, impairing myelination and thus the ability to learn a new task. To test this, I used the APP/PS1\\(_{swe}\\) mouse model. This model overexpresses APPswe and PS1˜ívÆE9 and has been demonstrated to generate cortical amyloid plaque deposits from 4 months of age (Garcia-Alloza et al., 2006) and cognitive alterations from 7 months (Serneels et al., 2009). For this study, I used the model of amyloidosis at 3 months of age, which allowed me to examine the effect of high levels of amyloid in the brain in the absence of other pathological changes, such as amyloid plaques and dystrophic neurites. To examine how high brain amyloid affects the myelination process during learning, a skilled reaching task was used as a model (Bacmeister et al., 2020; Sampaio-Baptista et al., 2013). This involves a mouse learning to repeatedly reach for a chocolate pellet through a narrow slit in a barrier over a 10-day period. This task involves considerable motor skills while reinforcing associated motor learning pathways for the dominant paw, and has been shown to generate structural changes in white matter and a learning-related increase in myelination (Sampaio-Baptista et al., 2013). For this study, mice were treated with 5-ethynyl-2¬¨¬•-deoxyuridine (EdU) during the reaching task period, to distinguish newly mature oligodendrocytes from existing populations. No significant differences in total or newly dividing oligodendrocyte or OPCs were present in the cortex, hippocampus, or corpus callosum in 3-month-old APP/PS1 following learning a skilled reaching task. In addition, no differences were present in myelin thickness in the corpus callosum of these animals compared to WT controls. In this study however, food deprivation, a method used to utilize food rewards to encourage behavioural testing participation, lead to a significant decrease in newly dividing oligodendrocytes in the corpus callosum and hippocampus in both WT and TgF344-AD rats, along with a decrease in cortical OPC numbers. In addition to examining alterations to myelination, I determined whether the learning task affected amyloid beta peptide production through examining levels in cerebrospinal fluid (CSF). CSF samples were analysed using the highly sensitive single molecule array (Simoa, Quanterix) system and demonstrated a significantly elevated level of CSF and serum A˜í‚⧠levels in mice that did not perform the learning task and remained on a normal diet, compared to mice that were food deprived. In summary, my thesis has added to the limited data examining how amyloid affects myelination and oligodendrocyte. These results indicate that amyloid beta peptides affect the development of oligodendrocytes and the formation of myelin in rodent models of AD, which may have implications for myelin associated plasticity and suggests a mechanism for deficits in cognition and learning in AD.