The aggregation of misfolded proteins into amyloid fibrils is implicated in the pathogenesis of several human degenerative diseases, including Alzheimer’s, Parkinson’s and Type II diabetes. Links between the deposition of amyloid fibrils and the progression of these diseases are poorly understood, with much of the current research focused on monomer misfolding and subsequent assembly of oligomers and mature fibrils. This project examines the formation of human apolipoprotein (apo) C-II amyloid fibrils, with a focus on the stability and reversibility of amyloid fibril assembly. / The initial stages of the project were to develop a model for apoC-II amyloid fibril formation. This was achieved by analysis of the concentration dependent kinetics of apoC-II amyloid fibril formation, and correlation of these data with the final size distribution of the fibrils, determined by sedimentation velocity experiments. On the basis of these studies, a new reversible model for apoC-II amyloid fibril formation is proposed that includes fibril breaking and re-joining as integral parts of the assembly mechanism. The model was tested by rigorous experimentation, with antibody-labelling transmission electron microscopy providing direct evidence for spontaneous fibril breaking and re-joining. / The development of this model for apoC-II fibril assembly provided the foundation for experiments to investigate factors that promote, inhibit or reverse amyloid fibril formation. Factors that were considered include a molecular chaperone protein, αB-crystallin, and a chemical modification, methionine oxidation. Investigations on the effect of αB-crystallin revealed that the inhibition of apoC-II fibril formation occurs by two distinct mechanisms: transient interaction with monomer preventing oligomerisation, and binding to mature fibrils, which inhibits fibril elongation. Studies on the effect of methionine oxidation on apoC-II fibril formation showed that both the assembly and stability of the fibrils was affected by this modification. ApoC-II contains two methionine residues (Met-9 and Met-60), and upon oxidation of these residues fibril formation was inhibited. In addition, the treatment of pre-formed fibrils with hydrogen peroxide caused dissociation of the fibrils via the oxidation of Met-60, located with the fibril core structural region. The final chapter details the development of antibodies that specifically recognise the conformation of apoC-II amyloid fibrils, which provide the foundation for future studies to examine the role that apoC-II amyloid fibrils play in disease. / Overall, this thesis reveals the dynamic and reversible nature of amyloid fibril formation. New insight is also obtained of the general stability of amyloid fibrils and the processes that may regulate their formation, persistence and disease pathogenesis in vivo.

Neurodegenerative disorders, including Alzheimer’s disease (AD) and Parkinson’s disease (PD), are increasingly common in our ageing society, are remain incurable. A major obstacle encountered by researchers in their attempts to find effective therapies is represented by the current lack of understanding of the molecular origins of these disorders. It is becoming clear that, although the aggregation of specific proteins, including amyloid β (Aβ) and tau in AD and α-synuclein in PD, hallmark these disorders, such behaviour is a consequence of a wider, system-level disruption of protein homeostasis. In order to identify the genetic factors contributing to such a disruption, the transcriptional changes that occur during neurodegenerative disease progression have received considerable scientific attention in recent years. In our approach, we considered another hallmark of these diseases - their characteristic patterns of spreading across the brain - to identify the nature of the transcriptional signature which underlies tissue vulnerability to protein aggregation. By understanding why tissues succumb in their characteristic sequential pattern in neurodegenerative diseases, and why some tissues remain almost completely resistant throughout, we hoped to obtain insight into the molecular origins of these disorders. Our results show that the AD progression can be predicted from a transcriptional signature in healthy brains related to the protein aggregation homeostasis of Aβ, tau, and the wider proteome. We highlight a relationship between a specific subproteome at high risk of aggregation (formed by supersaturated proteins), and the vulnerability to neurodegenerative diseases. We thus identify an AD-specific supersaturated set of proteins - termed the metastable subproteome, whose expression in normal brains recapitulates the staging of AD, with more vulnerable tissues having higher metastable subproteome expression. We find evidence of these vulnerability signatures transcending the tissue level of interrogation, with cellular and subcellular analysis also showing elevated levels of proteins known and predicted to predispose the aberrant aggregation of Aβ and tau. These results characterise the key protein homeostasis pathways in the inception and progression of AD, and establish an approach with the potential to be applied to other protein misfolding diseases, in the brain and beyond.

La formation d'agrégats protéiques insolubles et fibreux, appelés fibrilles amyloïdes, est impliquée dans une large variété de maladies humaines. Parmi elles, figurent entre autres, le diabète de type II, l'arthrite rhumatoïde et, notamment, les atteintes neurodégénératives débilitantes, telles que les maladies d'Alzheimer, de Parkinson ou encore de Huntington. Actuellement, il n'existe ni traitement, ni diagnostic précoce pour aucune de ces maladies.De nombreuses études ont montré que la capacité à former des fibrilles amyloïdes est une propriété inhérente à la chaîne polypeptidique. Ce constat a conduit au développement d'un certain nombre d'approches computationnelles permettant de prédire les propriétés amyloïdogéniques à partir de séquences d'amino-acides. Si ces méthodes s'avèrent très performantes vis à vis de courts peptides (~ 6 résidus), leur application à des séquences plus longues correspondant aux peptides et protéines en lien avec les maladies, engendre un nombre trop élevé de faux positifs. Le principal objectif de cette thèse consiste à développer une meilleure approche bioinformatique, capable de prédire les régions amyloïdogéniques à partir d'une séquence protéique. Récemment, l'utilisation de nouvelles techniques expérimentales a permis de mieux appréhender la structure des amyloïdes. Il est ainsi apparu que l'élément caractéristique de la majorité des fibrilles amyloïdes impliquées dans les maladies, était constitué d'une structure étagée (β-arcade), résultant de l'empilement de motifs « feuillet β – coude – feuillet b » appelés « β-arches ». Nous avons mis à profit cette particularité structurale pour créer une approche bioinformatique permettant de prédire les régions amyloïdogéniques d'une protéine à partir de l'information contenue dans sa séquence. Les résultats provenant de l'analyse des structures de type β-arcade, connues et modélisées, ont été compilés et traités à l'aide d'un algorithme écrit en langage Java, afin de créer le programme ArchCandy.L'application de ce programme à une sélection de séquences protéiques et peptidiques, connues pour leur lien avec les maladies, a permis de démontrer qu'il était en mesure de prédire correctement la majorité de ces séquences, de même que les séquences mutées impliquées dans les maladies familiales. Outre la prédiction de régions à haut potentiel amyloïde, ce programme suggère la conformation structurale adoptée par les fibrilles amyloïdes. Le séquençage de génomes entiers devenant toujours plus abordable, notre méthode offre une perspective de détermination individuelle des profils à risque, vis à vis de maladies neurodégénératives, liées à l'âge ou autres. Elle s'inscrit ainsi pleinement dans l'ère de la médecine personnalisée. / A broad range of human diseases are linked to the formation of insoluble, fibrous, protein aggregates called amyloid fibrils. They include, but are not limited to, type II diabetes, rheumatoid arthritis, and perhaps most importantly, debilitating neurodegenerative diseases such as Alzheimer's disease, Parkinson's disease, and Huntington's disease. There currently exists no cure, and no means of early diagnosis for any of these diseases. Numerous studies have shown that the ability to form amyloid fibrils is an inherent property of the polypeptide chain. This has lead to the development of a number of computational approaches to predict amyloidogenicity by amino acid sequences. Although these methods perform well against short peptides (about 6 residues), they generate an unsatisfactory high number of false positives when tested against longer sequences of the disease-related peptides and proteins. The main objective of this thesis was to develop an improved bioinformatics based approach to predict amyloidogenic regions from protein sequence.Recently new experimental techniques have shed light on the structure of amyloids showing that the core element of a majority of disease-related amyloid fibrils is a columnar structure (β—arcade) produced by stacking of β-strand-loop-β-strand motifs called “β-arches”. Using this structural insight, we have created a bioinformatics based approach to predict amyloidogenic regions from protein sequence information. Data from the analysis of the known and modeled β-arcade structures was incorporated into a rule based algorithm implemented in the Java programming language to create the ArchCandy program. Testing it against a set of protein and peptide sequences known to be related to diseases has shown that it correctly predicts most of these sequences and a number of mutated sequences related to the familial diseases. In addition to the prediction of regions with high amyloidogenic potential, a structural arrangement of the amyloid fibril is also suggested for each prediction. As whole genome sequencing becomes cheaper, our method provides opportunity to create individual risk profiles for the neurodegenerative, age-related and other diseases ushering in an era of personalized medicine.

The aim of my thesis was to develop and characterise PrP transgenic Drosophila melanogaster of various genotypes to study the process of prion-induced neurodegeneration in this model. Prion diseases are caused by the occurrence of an abnormally-folded form of PrP (PrPSc) protein that arises either from the environment as an acquired disease, from mutation in the PrP-coding gene as a genetic disease or sporadically from causes unknown. The PrPSc then recruits PrPC, the normal form of PrP, that is ubiquitously present in the mammalian CNS and triggers neurotoxicity and neurodegeneration that is transmissible between individuals of the same or even different species. All prion diseases are currently incurable, fatal and the mechanism of prion-induced neurodegeneration remains to be discovered. In this thesis, Drosophila transgenic for ovine (chromosome 3 and dual PrP transgenic flies), hamster, humanised murine, human and cervid PrP were characterised for expression and biochemical properties. The ultimate goal of my thesis was investigation of cell-to-cell spread of misfolded PrP in Drosophila CNS. To achieve this, a mutant form of PrP that is thought to misfold was co-expressed with the normal form PrPC that served as a substrate in the same dual PrP-transgenic fly. The process was modelled using hamster, humanised murine or ovine PrP transgenes that carry human mutations associated with the spontaneous onset of transmissible neurodegeneration in the natural host. Various approaches towards independent spatial expression of PrP in Drosophila were exploited here in both single and dual PrP expressing flies. Moreover, the ability to initiate misfolding and the impact of this on the fly phenotype was investigated. Both apparent misfolding and phenotypic changes were seen in different fly models suggesting the models were successful. To this extent, PrP transgenic Drosophila were developed to allow for relatively rapid modelling of mammalian prion disease in this invertebrate organism.

Amyotrophic lateral sclerosis (ALS) is characterized by motor neuron degeneration resulting in progressive paralysis and death. The only known cause of typical ALS is mutations in SOD1; these predominantly missense mutations produce a toxic gain-of-function in the enzyme Cu/Zn superoxide dismutase (SOD1). The prevailing hypotheses regarding the mechanism of toxicity were a) oxidative damage from aberrant SOD1 redox chemistry, and b) misfolding of the mutant protein. The goal of this thesis was to investigate the molecular mechanisms of the mutant SOD1 (mSOD1) misfolding and toxicity. We proposed that oxidative damage to SOD1 itself could cause its misfolding and aggregation. To investigate this hypothesis, we subjected purified SOD1 in vitro to metal catalyzed oxidation. Oxidation of SOD1 produced aggregates reminiscent of those observed in ALS pathology. Aggregation propensity of zinc-deficient SOD1 and several mSOD1s known to have lower zinc-binding affinity was proportional to partial unfolding. Oxidation of SOD1 caused conversion of several His residues to 2-oxo-histidine. Because oxidation of SOD1 primarily affected the metal-binding His residues, we hypothesized that oxidation of wild-type, holo-SOD1 should lead to aggregation. Increasing the concentration of wild-type SOD1 in oxidation reactions produced aggregates similar to those observed earlier. Both wild-type and mSOD1 aggregation kinetics revealed an initial decrease in particle size rather than a monotonic increase using dynamic light scattering. This was consistent with the conversion of SOD1, normally an obligate homodimer, into monomers prior to aggregation. This observation was confirmed using analytical ultracentrifugation. The common aggregation pathway for wild-type and mSOD1 suggested a mechanism for sporadic ALS caused by SOD1 misfolding. To interrogate the in vivo misfolding pathway of SOD1, we used its high-resolution structure to create an antibody that reacts with monomer/misfolded SOD1 but not the native dimer. Upon verifying the reactivity of this antibody, we showed that monomer/misfolded SOD1 is found in a human case of familial ALS and in transgenic animal models of ALS. Misfolded SOD1 is found primarily in affected cells, motor neurons. Misfolded SOD1 is also initially absent, but appears prior to symptom onset. These observations together suggest a causal role for SOD1 misfolding through a monomeric intermediate in ALS pathogenesis.

Amyotrophic lateral sclerosis (ALS) is characterized by motor neuron degeneration resulting in progressive paralysis and death. The only known cause of typical ALS is mutations in SOD1; these predominantly missense mutations produce a toxic gain-of-function in the enzyme Cu/Zn superoxide dismutase (SOD1). The prevailing hypotheses regarding the mechanism of toxicity were a) oxidative damage from aberrant SOD1 redox chemistry, and b) misfolding of the mutant protein. The goal of this thesis was to investigate the molecular mechanisms of the mutant SOD1 (mSOD1) misfolding and toxicity. We proposed that oxidative damage to SOD1 itself could cause its misfolding and aggregation. To investigate this hypothesis, we subjected purified SOD1 in vitro to metal catalyzed oxidation. Oxidation of SOD1 produced aggregates reminiscent of those observed in ALS pathology. Aggregation propensity of zinc-deficient SOD1 and several mSOD1s known to have lower zinc-binding affinity was proportional to partial unfolding. Oxidation of SOD1 caused conversion of several His residues to 2-oxo-histidine. Because oxidation of SOD1 primarily affected the metal-binding His residues, we hypothesized that oxidation of wild-type, holo-SOD1 should lead to aggregation. Increasing the concentration of wild-type SOD1 in oxidation reactions produced aggregates similar to those observed earlier. Both wild-type and mSOD1 aggregation kinetics revealed an initial decrease in particle size rather than a monotonic increase using dynamic light scattering. This was consistent with the conversion of SOD1, normally an obligate homodimer, into monomers prior to aggregation. This observation was confirmed using analytical ultracentrifugation. The common aggregation pathway for wild-type and mSOD1 suggested a mechanism for sporadic ALS caused by SOD1 misfolding. To interrogate the in vivo misfolding pathway of SOD1, we used its high-resolution structure to create an antibody that reacts with monomer/misfolded SOD1 but not the native dimer. Upon verifying the reactivity of this antibody, we showed that monomer/misfolded SOD1 is found in a human case of familial ALS and in transgenic animal models of ALS. Misfolded SOD1 is found primarily in affected cells, motor neurons. Misfolded SOD1 is also initially absent, but appears prior to symptom onset. These observations together suggest a causal role for SOD1 misfolding through a monomeric intermediate in ALS pathogenesis.

Protein misfolding diseases (PMDs) are a broad group of disorders including Alzheimer’s, Parkinson’s and prion diseases. They are characterised by the presence of aggregated, misfolded host proteins which are thought to cause cell death. Prion diseases are associated with misfolded prion protein (PrPSc), which has a tendency to form fibrillar aggregates. By contrast, Alzheimer’s disease (AD) is associated with misfolded amyloid beta (Aβ), which aggregates to form characteristic Aβ plaques. A feature which is common across PMDs is that small assemblies (oligomers) of the misfolded proteins are thought to be the important neurotoxic species, and it has been proposed that there may be a shared mechanism leading to cell death across PMDs caused by oligomers. In this study, the toxicity of different misfolded forms of recombinant PrP (recPrP) and recombinant Aβ (recAβ) and the mechanisms leading to cell death were investigated using a primary cell culture model. In addition, the importance of the disulphide bond in recPrP in relation to oligomer formation was explored using size exclusion chromatography and mass spectrometry, the toxicity of the different resulting oligomer populations were also investigated. Both recPrP oligomers and fibrils were shown to cause toxicity to mouse primary cortical neurons. Interestingly, oligomers were shown to cause apoptotic cell death, while the fibrils did not, suggesting the activation of different pathways. By contrast, recAβ fibrils were shown to be non-toxic to cortical neurons, Aβ oligomers, however, were shown to cause toxicity. Similar to recPrP, my data showed that it is likely that recAβ 1-42 oligomers also cause apoptosis. However, by contrast this seemed to be caused by excitotoxicity, which was not found to be the case for recPrP. Additionally, I have shown that the presence or absence of the disulphide bond in PrP has a profound effect on the size of oligomers which form. RecPrP lacking a disulphide bond leads to the formation of larger oligomers which are highly toxic to primary neurons. Findings from this study suggest that structural properties such as the disulphide bond in PrP can affect the size and toxicity of oligomers, furthermore, whilst oligomers have been shown to be important in both AD and prion diseases, they may not trigger the same pathways leading to cell death.

Alzheimer’s disease (AD) is the most common cause of dementia, affecting almost 40 million people worldwide, and it is predicted that this number will rise to nearly 150 million by 2050. It results not only in enormous distress for affected individuals and carers but also a substantial economic burden on society. Although more than 100 years have passed since its discovery, no cure for AD exists, despite enormous efforts in basic and clinical research over the past few decades, due to limited understanding of its underlying mechanisms. Neurodegenerative disorders, of which AD is an example, are highly complex disorders characterized by extensive neuronal dysfunction associated with the misfolding and aggregation of a specific set of proteins, including amyloid plaques and neurofibrillary tangles in AD. One promising avenue for progress in the field is to improve our understanding of the mechanisms by which cellular dysfunction arises from the initial protein aggregation events. The studies described in the thesis are based on the recent finding that a large number of proteins are inherently supersaturated, being expressed at concentrations higher than their solubilities, and constituting a metastable subproteome potentially susceptible to aggregation. These studies illustrate the dependence of aggregation prone metastable proteins on the cellular degradation machineries. They also study the role of metastable proteins and their homeostasis complement in the vulnerability of various body and brain tissues to protein aggregation diseases. Using extensive sequencing data and network based systems biology approaches, they elucidate how fundamental physicochemical properties of an individual or group of proteins relate to their biological function or dysfunction.

RATIONALE: Transgenic mice with cardiac specific overexpression of mutated αB-crystallin (CryABR120G) display Desmin-related myopathy (DRM) with dilated cardiomyopathy and heart failure. Our previous studies showed the presence of progressive mitochondrial abnormalities and activation of apoptotic cell death in CryABR120G transgenic hearts. However, the role of mitochondrial dysfunction and apoptosis in the overall course of the disease was unclear. OBJECTIVE: We tested the hypothesis that prevention of apoptosis would ameliorate CryABR120G pathology and decrease morbidity. METHODS AND RESULTS: We crossed CryABR120G mice to transgenic mice with cardiac specific overexpression of Bcl-2. Sustained Bcl-2 overexpression in CryABR120G hearts prolonged CryABR120G transgenic mice survival by 20%. This was associated with decreased mitochondrial abnormalities, restoration of cardiac function, prevention of cardiac hypertrophy, and attenuation of apoptosis. CryABR120G misfolded protein aggregation was significantly reduced in the double transgenic. However, inhibition of apoptotic signaling resulted in the upregulation of autophagy and alternative death pathways, the net result being increased necrosis. CONCLUSION: Although Bcl-2 overexpression prolonged life in this DRM model, in the absence of apoptosis, another death pathway was activated.

The misfolding and aggregation of proteins is closely associated with more than fifty human disorders, including Alzheimer's and Parkinson's diseases, all of which are currently incurable and many represent a major threat to human life. The mechanism of protein aggregation is subject to extensive studies. The damaging effects associated with protein aggregation have been attributed to amyloidogenic species that are present during the misfolding process. In particular, oligomeric species are, however, intrinsically difficult to study as a consequence of their low abundance and highly heterogeneous nature. The first chapter of my thesis gives an introduction into the field of protein folding and misfolding with a focus on the study of protein aggregation, and toxic effects relevant to human disorders. The second chapter of my thesis describes the development of a methodology that enables the study of aggregate induced lipid bilayer permeability, possibly the most general mechanism of protein aggregate toxicity. Surface-tethered lipid vesicles functioning as optochemical probes sensitive to membrane integrity are imaged using total internal reflection microscopy. It is shown that oligomeric species of the 42-residue form of the Aβ peptide (Aβ42) are responsible for the membrane disruption. The methodology can be applied to the study of other proteins such as α-synuclein and tau, and the ability of antibodies and chaperones to counteract the aggregate induced lipid bilayer permeability can be assessed. Furthermore, lipid bilayer permeability induced by aggregates formed in human induced pluripotent stem cells can be studied. The third chapter presents a new approach for the measurement of protein aggregation kinetics by following the development of the lipid bilayer permeability over the course of the aggregation process of Aβ42. The aggregation kinetics can be modulated with molecular chaperones and pre-formed seed fibrils, which allows secondary nucleation to be identified as the process that drives the formation of species responsible for the lipid bilayer permeability. The fourth chapter describes the development of a three-pronged strategy to study the mechanism of α-synuclein amyloid formation. The aggregation is studied in the presence of lipid vesicles or pre-formed fibrils at neutral or acidic pH of the solution. The influence of single-point mutations on the aggregation of α-synuclein is described. Furthermore, the strategy is applied to the characterisation of the ability of antibodies and small molecules to inhibit the aggregation, and thus has the potential for the development of therapeutical agents. The work presented in the fifth chapter characterises the amyloid fibril populations formed by α-synuclein and mutational variants associated with familial Parkinson's disease. X-ray crystallography, circular dichroism spectroscopy, Fourier transform infrared spectroscopy, transmission electron microscopy and atomic force microscopy have all been applied to the analysis of these amyloid fibrils. Finally, the sixth chapter summarises the results described in this thesis and points out future opportunities in the context of fundamental and translational studies related to the research area of protein misfolding disorders.