The retromer is a protein complex whose function is to mediate the recycling of proteins from the endosome to either the plasma membrane or the trans-Golgi network. A deficit in retromer function has been associated with multiple neurodegenerative disorders, including Alzheimer’s disease (AD) and Parkinson’s disease (PD). In both AD and PD, deficiencies have been found in retromer expression both in patient tissues and in animal models of disease. Furthermore, mutations in the retromer and in retromer-associated genes have been strongly linked with both diseases. Despite ample evidence of the link between the retromer and neurodegeneration, little is known about the retromer in the context of amyotrophic lateral sclerosis (ALS), another common neurodegenerative disorder. ALS is an adult-onset neurodegenerative disorder of the upper and lower motor neurons (MNs) characterized by muscle wasting and weakness leading to death within 3-5 years after diagnosis. To date, the most commonly used model of ALS is a transgenic (Tg) mouse that overexpresses an ALS-causing G93A mutation in the human superoxide dismutase 1 (SOD1) gene. In this study, I first establish a link between the retromer and ALS by showing that cells from ALS patients as well as tissues and cells from SOD1G93A-Tg mice express lower protein levels of the retromer core components—vacuolar protein sorting 35 (Vps35), Vps26a, and Vps29. I then establish that deficiencies in retromer core proteins have functional consequences in an in vitro model of ALS. Having found significant deficiencies in the retromer in SOD1G93A-Tg mice, I then followed the model of studies performed in mouse models of other neurodegenerative disorders by investigating whether repletion of retromer levels, either virally or pharmacologically, in SOD1G93A-Tg mice confers a therapeutic benefit. Surprisingly, I find that rather than ameliorating disease, repletion of retromer levels in SOD1G93A-Tg mice exacerbates it, resulting in a faster decline in motor performance, earlier mortality, and a decrease in MNs in the spinal cord. Finally, since retromer repletion causes deleterious effects on SOD1G93A-Tg mouse disease progression, I study the effect of a single allele deletion of Vps35 in SOD1G93A-Tg mice and find that this depletion of the retromer results in amelioration of disease, including delayed onset of symptomatology, slower decline of motor deficits, delayed mortality, and an increase in MNs in the spinal cord. Altogether, the findings reported herein, support the notion that a mild defect in retromer develops over the course of the disease, which, rather than being deleterious may be therapeutic in mutant SOD1-induced MN degeneration. Perhaps this unexpected outcome may be explained by the fact that the observed mild nature of the defect is not sufficient to kill MNs but enough to alter the trafficking of specific cargos such as AMPA receptors, allowing MNs to better withstand the neurodegenerative process.

Amyotrophic lateral sclerosis (ALS) is a progressive neurodegenerative disorder characterized by the degeneration of cortical and spinal motor neurons. Animal models of ALS based on known ALS-causing mutations are instrumental in advancing our understanding of the pathophysiology of motor neuron degeneration. Recent identification of mutations in the genes encoding RNA-binding proteins TDP-43 and FUS has suggested that aberrant RNA processing may underlie common mechanisms of neurodegeneration in ALS and focused attention on the normal activities of TDP-43 and FUS. However, the role of the normal functions of RNA-binding proteins in ALS pathogenesis has not yet been established. In this thesis I present my work on novel FUS-based mouse lines aimed at clarifying the relationships between ALS-causing FUS mutations, normal FUS function and motor neuron degeneration. Experiments in mutant FUS knock-in mice show evidence of both loss- and gain-of-function effects as well as misfolding of mutant FUS protein. Characterization of mice expressing ALS-mutant human FUS cDNA in the nervous system reveals selective, early onset and slowly progressive motor neuron degeneration that is mutation dependent, involves both cell autonomous and non-cell autonomous mechanisms and models key aspects of ALS-FUS. Using a novel conditional FUS knockout mutant mouse, I also demonstrate that postnatal elimination of FUS selectively in motor neurons or more broadly in the nervous system has no effect on long-term motor neuron survival. Collectively, our findings suggest that a novel toxic function of mutant FUS, and not the loss of normal FUS function, is the primary mechanism of motor neuron degeneration in ALS-FUS.

published_or_final_version / Anatomy / Master / Master of Philosophy

Amyotrophic Lateral Sclerosis (ALS) is a fatal neurodegenerative disease caused by the death of motor neurons in the spinal cord and brain. ALS is a genetically complex disease; diverse mutations cause motor neuron death by disrupting various interrelated pathways. To date, no therapy targeting a single factor can rescue motor neuron loss, nor is it known how or why sub-populations of motor neurons are particularly vulnerable in disease. Many studies have pointed to the Transforming Growth Factor Beta (TGF-𝝱) signaling superfamily as a modifier of disease in human patients and in animal models. Here, we have used the SOD1G93A model of ALS to investigate if and how TGF-𝝱 signaling in motor neurons changes pathology in these animals. In the first part of this study we characterize canonical TGF-𝝱 activation in motor neurons in SOD1G93A animals compared to controls. We have found that a vulnerable motor neuron subpopulation upregulates TGF-𝝱RII, a receptor necessary and unique to the classical arm of the TGF-𝝱 signaling family, in a disease dependent manner. Despite the upregulation of TGF-𝝱RII in these cells, there is not a corresponding activation of downstream canonical TGF-𝝱 effectors in diseased motor neurons. Through in vivo genetic manipulation we found that TGF-𝝱RII is dispensable in motor neurons, but that ablation of TGF-𝝱RI, a key receptor in multiple arms of the TGF-𝝱 superfamily, decreases motor neuron survival in SOD1G93A animals. To further understand how this manipulation changes TGF-𝝱 activation in motor neurons, we performed iterative indirect immunoflourescence imaging. We have identified that knocking out TGF-𝝱RI from motor neurons disrupts downstream canonical TGF-𝝱 activation in these cells. To identify how TGF-𝝱 signaling changes gene expression in these cells we have used Visium, a spatial RNAseq method, on lumbar spinal cords from these animals We have identified and are currently investigating potential downstream targets of TGF-𝝱 signaling in motor neurons in SOD1G93A animals. These data suggest that motor neurons rely on TGF-𝝱 signaling for survival in disease and that TGF-𝝱 signaling is important to the biology of a known vulnerable population of motor neurons.