Alzheimer's disease (AD) is the most common form of dementia, and a devastating neurodegenerative disease in which there is no cure or effective treatments to delay disease onset. One of the major culprits of the disease is the protein amyloid-beta (A&beta), which accumulates as both soluble and insoluble aggregates in the brain of AD patients, leading to cellular toxicity and neuronal degeneration. To gain a better understanding of the toxic effects of A&beta, we used the yeast Saccharomyces cerevisiae as a model organism. The advantage of using yeast was that we could conduct an unbiased, high-throughput screen to identify loss-of-function genetic modifiers of A&beta toxicity. In addition, yeast is a simple, eukaryotic system to study the intracellular pathways that might be affected by A&beta accumulation. Genetic manipulation of the yeast genome is technically easy: genes of interest can be deleted, fusion proteins can be tagged to endogenous genes to track their localization within the cell, and expression of heterotrophic genes can be introduced. Furthermore, many tools are available to study physiological processes and pathways in yeast, such as intracellular trafficking, induction of stress responses, and growth rates measurements.

Our studies show that the expression of A&beta in yeast causes cellular toxicity, reduced growth rate, and induction of the unfolded protein response. However, these effects are not seen in A&beta mutant peptides that have a reduced aggregation propensity. This suggests that the accumulation of aggregated A&beta is a key component of toxicity. In addition, we observed that degradation of A&beta is impaired, resulting in slow turnover of the protein in the cell. Fluorescence microscopy studies showed that A&beta localizes to the endoplasmic reticulum (ER), where misfolded or aggregated A&beta may be retained. A high-throughput screen identified FAB1 as a loss-of-function genetic enhancer of A&beta toxicity. Loss of FAB1 further retards A&beta degradation, which may enhance A&beta accumulation and toxicity; however, the precise mechanisms for A&beta toxicity are still unknown. Our model provides a potential tool to study the quality control mechanisms that maintain intracellular A&beta in a properly folded and non-toxic state.

Huntington's disease (HD) is a fatal, incurable neurodegenerative disorder caused by an expanded polyglutamine (polyQ) tract in the protein huntingtin (htt). Recent studies have suggested the potential importance of immune cells in HD pathogenesis. Of particular interest, microglia have early activation and dystrophy in the brains of HD patients and mouse models. HD patients also have increased levels of plasma pro-inflammatory cytokines and chemokines before the onset of motor symptoms that correlate with disease progression. HD patients' monocytes also have increased cytokine levels in response to stimulation. Because htt is ubiquitously expressed, it may directly contribute to the dysfunction of microglia and peripheral immune cells as they have similar origins and functional similarities.

To investigate the effects of mutant htt in immune cells, we first established microglial cell lines that stably express this protein and obtained primary microglia from HD mouse models. We evaluated canonical activation responses and found that microglia expressing mutant htt selectively have increases in pro-inflammatory cytokines and an impairment in stimulus-dependent migration. This migration defect is also present in peripheral immune cells isolated from HD patients, such as monocytes and macrophages. We also observed defective migration of peripheral macrophages towards the site of inflammation upon induction of peritonitis in HD transgenic mice. In the brains of these mice there was also abnormal microglial process extension and retraction, and a delayed response toward laser ablation injury.

Based on our results that mutant htt impairs some immune cell functions and previous evidence of abnormal immune activation in HD, we determined the effects of bone marrow transplantation (BMT) on pathogenesis in vivo. HD donated mice with wild-type (WT) bone marrow had modest but significant improvements of behavioral phenotypes, increased levels of synaptophysin, and normalized levels of peripheral inflammatory cytokines. Our results collectively indicate that mutant htt indeed impairs the function of peripheral and central immune cells in a manner that could be a contributing in a disease-modifying manner to pathogenesis in HD. The cell-autonomous effect of mutant htt in immune cells may also explain the early immune abnormalities in HD. This work has potential implications for HD patients, as there are drugs currently used to treat inflammation that could be tested in HD patients.

Many studies of immune responses in neurodegeneration assert that neuroinflammation is a double edged sword - i.e. some aspects of an immune response are beneficial while others are detrimental. In this thesis we investigated two enzymatic cascades that are activated in response to inflammatory stimuli and that we hypothesized were relevant for Huntington's Disease (HD).

In Chapter 2 we explored the role of the complement cascade in the R6/2 mouse model of HD. The complement cascade is a well characterized immune response mechanism and it is activated in brains of HD patients. Surprisingly, we found that this upregulation of the complement cascade is not mirrored in the R6/2 mouse model of HD, and that genetic deficiency in complement component C3 does not alter disease progression in these mice. This illustrates that normal complement cascade activity is not necessary for disease progression in R6/2 mice, with the caveat that complement activation in HD patients is not faithfully reproduced in this mouse model of HD.

In Chapter 3 we turned our attention to a pathway known to be relevant in HD, the Kynurenine Pathway. Here, instead of seeking a link between the pathway and HD, we sought to understand how two types of Kynurenine Pathway activity, basal and immune stimulated, are controlled. We employed mice genetically deficient in either tryptophan 2,3-dioxygenase (TDO) or indoleamine 2,3-dioxygenase (IDO) because these two enzymes can catalyze the rate-limiting step of the Kynurenine Pathway. We evaluated the contribution of each enzyme to basal and immune stimulated production of several Kynurenine Pathway metabolites that play a role in a variety of highly prevalent diseases, including HD. TDO deficient mice show evidence of several compensatory changes in basal Kynurenine Pathway metabolism. These changes suggest the existence of important unknown mechanisms that regulate basal Kynurenine Pathway metabolism. In contrast, IDO deficient mice appear to be a useful tool for cleanly blocking inflammation induced increases in brain Kynurenine Pathway metabolites without altering basal Kynurenine Pathway metabolism. Both of these results have important implications for development of therapeutics targeting the Kynurenine Pathway.

a-synuclein, a major player in idiopathic and familial forms of Parkinson's disease pathogenesis, is thought to impair vesicle trafficking as a toxic gain of function through elevated levels of the protein. Genetic screens for modifiers of a-synuclein toxicity in model organisms have overwhelmingly identified a significant subset of genes involved in vesicle trafficking. While recent studies in yeast have shown that at toxic levels a-synuclein disrupts Rab homeostasis, causing an initial ER-to-Golgi block that precedes a

generalized trafficking collapse, further studies in mammalian systems suggested that several Rab proteins downstream of ER-Golgi trafficking can also rescue a-synuclein toxicity. Here, we show that constitutive expression of a-synuclein in yeast causes an accumulation of endocytic vesicles and impairs late-exocytic, early-endocytic and/or recycling trafficking and, further, that this vesicle trafficking impairment can be attenuated by yeast casein kinase 1 and the small rab protein Ypt1p. Our work in mammalian cell culture demonstrates that human orthologs of vacuolar protein sorting

genes involved in endosomal transport modulate the toxicity of a-synuclein. Furthermore, our findings suggest that a-synuclein-induced toxicity is due to impairment in the endosomal trafficking pathway of neuronal cells. Taken together, our results provide evidence of the potential contribution of endosome anomalies to the pathogenesis of Parkinson's disease.