David R. Borchelt, PhD, University of Kentucky - US grants

The present proposal seeks to clarify the role of aberrant Cu chemistry (peroxidation/nitration) in the pathogenesis of SOD1. One component will examine the abilities of FALS mutant SOD1, particularly enzyme with mutations involving crucial histidine, to bind Cu and to react with H202 and -OONO. A second components will determine whether experimentally mutated SOD1, lacking the capacity to bind Cu and encoding FALS mutations, can cause ALS-like motor neuron disease in mice. The third component of the study will operate under the assumption that, even if aberrant Cu chemistry is not the primary mechanism by which all FALS mutant SOD1 causes disease, some mutants do have enhanced peroxidase and nitration activities, and the level of these associated activities may modulate the severity of the disease.

disease /disorder model; laboratory mouse; pathologic process

One of the prevalent pathologies in Alzheimer's Disease is the accumulation of ubiquitin-immunoreactive material in cell bodies, dendrites, and neurites. The neuritic pathology is recapitulated in transgenic mouse models of beta-amyloid deposition. The prevalence of this pathology in AD, as well as other disorders of the CNS, has led to the hypothesis that a dysfunction of the ubiquitin/proteasome system may play a role in the pathogenesis of disease. However, probing the function of this system in vivo is problematic and it has been difficult to know whether dysfunction of this system is an early or late event. We propose to produce transgenic mice that express proteasome sensors. We will construct two sensors by fusing the coding sequence of ubiquitin (encoding mutations that preclude endoproteolytic processing) to green fluorescent protein and beta-galactosidase. These proteins will be short- lied in normal settings. We will insert these cDNAs into the mouse PrP vector and generate transgenic mice. We will then screen for lines of mice in which expression of mRNA for the sensor is high, while the accumulated levels of protein are negligible. We will then generate primary neuronal cultures and use chemical inhibitors of the proteasome to screen for mice in which inhibition of the proteasome leads to increased accumulation of our sensor proteins. Ultimately, if successful, these animals would provide function in normal aging, and in our transgenic mouse models of Alzheimer's Disease and other neurodegenerative diseases.

DESCRIPTION (provided by applicant): Alzheimer's disease (AD) pathology is characterized by neurofibrillary tangles, astrocytosis, microgliosis, and extracellular deposits of [unreadable]-amyloid peptides (A[unreadable]). The A[unreadable] peptide (40-42 residues) is generated by the constitutive proteolytic processing of the amyloid precursor protein (APP). Many converging lines of evidence suggest that oligomerization and ultimately the deposition of A[unreadable]42 is an early and critical event in the pathogenesis of AD. In the previous grant period, new strains of mice that express high levels of mutant APP were developed using approaches that utilize tetracycline-regulated vectors. Studies of these mice demonstrated that in an in vivo setting, amyloid deposits are very stable structures that persist long after new production of A[unreadable] peptides is suppressed. Recently, however, focus has turned away from amyloid deposits as important mediators of cognitive impairment, focusing instead on more soluble oligomeric forms of the peptide. We have devised five Aims to investigate whether amyloid deposits may serve as reservoirs for oligomeric structures that have been implicated as the more potent mediators of cognitive impairment. This question is of major importance in considering therapeutic interventions that target A[unreadable] production. Aim 1 will use the tet- regulated transgenic mice to suppress A[unreadable] production in an environment of high amyloid burden and then use established methods of assessing cognition with passive peripheral transfer of antibodies to A[unreadable] as a means to determine whether diffusible A[unreadable] oligomeric inhibitors of memory function persist along with deposited amyloid. Aim 2 will use a different approach to address whether diffusible species of A[unreadable] persist along with amyloid deposits;the approach uses transgenic mice expressing mutant APP as recipients of fetal neural grafts from non-transgenic animals. Previous studies have established that the graft environment is very receptive to amyloid deposition. The approach will be to use implant non-transgenic neural grafts in tet-regulated mice with amyloid deposits (and after suppression of new A[unreadable] production) and then assess whether the grafts take up A[unreadable] to form new deposits. Aim 3 proposes to use viral vectors to transduce the grafted cells to express molecules that could be used to further define the nature of these diffusible forms of the peptide. Aim 4 proposes experiments to ascertain the role of resident microglia in modulating amyloid formation. Using a strain of mice that naturally is deficient in microglia, we will examine the rate, distribution and character of amyloid deposition when crossed to mice that express mutant APP. Aim 5 proposes to generate new lines of mice that express high levels of mouse A[unreadable] peptides. Previous studies establish that mouse A[unreadable] does not promote amyloid deposition;the planned studies will confirm whether this conception is true and examine whether mouse A[unreadable] (whether in amyloid or not) is capable of influencing cognitive performance. Collectively, these studies address fundamental questions regarding the biology of amyloid deposition and the role of various forms of these A[unreadable] oligomeric structures in amyloid deposition and cognitive impairment. PUBLIC HEALTH RELEVANCE: Small protein fragments called amyloid beta peptides are crucial mediators of the pathology and symptoms Alzheimer's disease, which is a major cause of disability in the elderly and is currently largely untreatable. Crucial to the disease is the assembly of amyloid peptides into larger structures, which produce molecules that affect systems critical in memory and produce lesions in brain characteristic of the disease called amyloid plaques. The studies proposed in the present application address fundamentally important questions regarding the biology of amyloid peptide assembly into disease-specific structures and the role of such molecules in producing disease-associated symptoms.

DESCRIPTION (provided by applicant): The function and half-life of most cytoplasmic and nuclear proteins, as well as some membrane proteins, is regulated by the covalent linkage of ubiquitin. Ubiquitination a key regulator of degradation by the proteasome. In neurodegenerative diseases such as Alzheimer's disease, amyotrophic lateral sclerosis, and Huntington's disease, ubiquitinated proteins accumulate, indicating a possible dysregulation of ubiquitination or proteasome system. Mutations in enzymes involved in ubiquitination have been associated with familial forms of neurodegenerative disease, including loss of function mutations in parkin for Parkinson's disease. In this application, we propose a proteomic approach to study ubiquitination under normal and disease-like toxic settings. As a first step, we have generated a mouse L cell line that stably expresses a hexa-histidine and GFP tagged ubiquitin protein and demonstrated that this modified ubiquitin can be conjugated to protein targets and then purified on immobilized metal affinity chromatogrphy (IMAC). Proteins purified in this manner can be directly anlayzed by MALDI-TOF and tandem mass spectroscopy to obtain sequence information. We now propose 3 Aims that are designed to characterize protein substrates of ubiquitination and to study how certain types of insults affect the process. In Aim 1, we will study the mouse L cells to characterize the following; 1) steady-state ubiquitinated protein profiles, 2) cell cycle specific profiles, 3) profiles after oxidative insult, 4) profiles after serum deprivation, 5) profiles after heat shock, and 6) profiles after induction of the unfolded protein response. In Aim 2, we will construct lines of human neuroblastoma SY5Y and astrocytoma CCF-STTG1 cells that stably express hexa-his-GFP-Ubq for a similar characterization. Collectively, these studies will lead to a better characterization of the substrates of ubiquitination and how various disease-like insults alter the process. We anticipate that some of the newly identified proteins may be useful biomarkers for certain types of toxicity in vivo.

[unreadable] DESCRIPTION (provided by applicant): The accumulation of ubiquitin-immuno-reactive material in cell bodies, dendrites, and/or axons of neurons are a prevalent pathology of neurodegenerative disease. It has been suggested that the accumulation of this material is a cellular symptom of reduced ubiquitin/proteasome system (UPS) function. In the present application, we propose four Aims that are designed to probe whether and how the proteasome/ubiquitin system is dysfunctional in various models of neurodegenerative disease. In Aim 1, we propose to use genetic approaches to alter the activities of UPS components. We have been provided mice lacking parkin (a biquitin E3 ligase whose loss triggers Parkinson's disease), and we would like to cross these mice to our APPswe/PS1dE9 mice. We hypothesize that amyloid deposition, in a context of parkin deficiency, may induce novel cytoplasmic pathologies, such as Lewy-body-like inclusions. Aim 2 will develop systems to inhibit proteasome function in vivo in transgenic mice, both chronically and acutely, using genetic approaches. Aim 3 will build on recent characterization of a subset of sporadic ALS cases, where we have identified cystatin C as a protein of interest in the disease. To test the role of this protein in ALS, we propose to create transgenic animals that express elevated levels of the human protein. Aim 4 will focus on identifying the protein backbone constituents of the ubiquitin immunoreactive material that accumulates in our mouse models of Alzheimer's disease and ALS. This Aim will involve the development of transgenic mice expressing recombinant ubiquitin molecules carrying peptide motifs that facilitate detection and purification. Collectively, these studies should allow us to examine the role of proteasome dysfunction in disease pathogenesis and perhaps identify some of the mis-folded proteins that accumulate in disease-associated inclusions. [unreadable] [unreadable]

SUMMARY (Project III) Over the past decade, it has become increasingly clear that Alzheimer's disease (AD) is a pathologically complex disorder that evolves over decades. Although the most common pathology of AD is the co-existence of amyloid plaques and neurofibrillary tangles, about 40% of AD cases also show a-synuclein (aS) pathology. It is also common to observe TDP-43 positive inclusions in AD cases. In dominantly inherited forms of AD (fAD), it has become clear that the deposition of A? occurs well before the onset of cognitive symptoms and the appearance of tau pathology. Although the order of events may be less obvious in sporadic AD, the perception is that aS and tau pathologies occur as secondary events in the evolution of disease for fAD. A growing body of literature suggests that the evolution of intracellular aS and tau pathology may involve a prion-like spreading of a misfolded protein conformation along anatomical pathways or between cells in discrete anatomical structures. In studies preliminary to this application, we have developed models in which we can induce aS and endogenous tau pathology by exogenous seeding with aS. In Aim 1, we propose to use a combination of transgenesis and seeding to develop models that more faithfully recapitulate the various pathologies of AD. An important feature of these models is that, by seeding, we are able to establish point of origin and then track the spread of pathology to adjacent structures or anatomically connected structures. If misfolded proteins are moving between cells, as our data and data from other laboratories suggest, and if such proteins are exposed to the intercellular space for a significant interval of time, then antibodies directed against these proteins may be able to bind and inhibit further spread. The development of model systems that mimic the spread, or transmission, of human pathology offers an opportunity to test novel immune therapies to provide proof of concept for moving such therapies to humans. In Aim 2, we propose to use the unique features of these inducible models in proof of concept studies to determine the potential efficacy of antibody therapies. Given that a substantial proportion of AD cases have mixed pathologies, and that these pathologies are independently associated with neurodegeneration, we propose that to achieve optimal clinical benefit it is likely that therapies need to be developed that can target more than one type of protein inclusion pathology. Thus, in this project, we will develop and test immunotherapeutics targeting A?, aS, and possibly tau. Collectively, these studies work towards two major unmet needs in AD, the generation of models that more faithfully reproduce human disease and development of disease modifying therapies.

DESCRIPTION (provided by applicant): Over the past few years, there has been increasing exploration of the potential for gene silencing or knockdown therapies in the treatment of neurodegenerative disorders. The technology has progressed to a point in which a phase 1 human therapeutic trial for familial amyotrophic lateral sclerosis has been initiated. Current knock-down therapies under investigation include viral vector delivery of shRNAi or microRNA mimics, delivery of naked RNAi and RNAi complexed with various reagents to facilitate uptake, and delivery of modified antisense DNA oligonucleotides. The mechanisms of action for these approaches include modulation of mRNA translation, modulation of pre-mRNA splicing, and degradation of mRNA and pre-mRNA. These various approaches have been tested in pre-clinical animal models to varying extents with varying levels of efficacy. A glaring limitation of these studies that have been conducted thus far is that it has generally been impossible to monitor the efficacy of knock-down in real time. To overcome this limitation, we propose two Aims that are designed to build capability to track the efficacy of knock-down in real time and provide proof of concept studies in mouse models of two neurodegenerative diseases that are potential targets for gene silencing efforts. Taking advantage of the expertise of the investigators involved, we plan to focus on models for Huntington s disease and fronto-temporal dementia. These disorders are prime candidates for gene-silencing therapeutics and previous work in modeling these disorders in mice has produced models that recapitulate aspects of each disorder. In the approach described here, we seek to generate models in which we will produce assayable and observable behavioral phenotypes while simultaneously being able to monitor the efficacy of gene silencing reagents in real-time. In Aim 1, we will generate mice that express mutant forms of human tau fused in-frame to luciferase. In Aim 2, we will similarly generate mice that express mutant Nterminal fragments of huntingtin fused in-frame to luciferase. In both constructs we will employ a technique that facilitates post-translational processing of the poly-protein to liberate the luciferase so it can be assayed independently of pathologic accumulations of mutant tau or huntingtin. We propose to use new in vivo imaging techniques to detect and measure bioluminescence catalyzed by the expressed luciferase. In the tau model we propose to generate, we expect the animals to develop measurable memory deficits with neuropathological abnormalities that include neuronal loss and neurofibrillary tangle pathology. In the Huntington s model, we similarly expect to induce assayable phenotypes, which include motor function deficits, reduction in the transcription of a subset of genes in striatum, hypoactivity, and premature death. Thus, one could ultimately have models with dual readout capability in which reductions in expression could be monitored in real-time by monitoring luciferase activity levels while simultaneously having disease-relevant phenotypes to assay.

DESCRIPTION (provided by applicant): Protein aggregation is a major pathological hallmark of neurodegenerative diseases such as Alzheimer's disease (AD) and frontotemporal lobar degeneration (FTLD); and the role of some form of misfolded protein (oligomer or inclusion of ¿-amyloid peptide or tau) in inducing a cascade of events that produces symptoms is largely undisputed. Although originally a 'one protein aggregate, one cause, one disease' hypothesis was the dominant ideology, over the last decade it has become clear that these disorders are more heterogeneous with multiple protein aggregates present in any one specific disease. In most cases, the mixed pathology involves proteins that have been identified as intrinsically vulnerable to self-seeded misfolding and aggregation (TDP-43 is an example). The mechanisms that underlie the appearance of these mixed pathologies is poorly understood with a leading hypothesis being that the accumulation of one misfolded and aggregating protein negatively impacts the function of the network of activities that mediate protein folding and degradation (the proteostasis network). Loss of proteostasis function is proposed to lead to secondary misfolding of vulnerable proteins. We propose to test the hypothesis that the accumulation of Alzheimer-type amyloid and FTLD-type neurofibrillary tangles can induce the aggregation of a secondary reporter protein that is inherently vulnerable to misfolding and aggregation in transgenic mouse models. Mice that express the reporter, which consists of a variant of superoxide dismutase 1 fused to yellow fluorescent protein (SODG85R:YFP), at levels just below the threshold to initiate self-seeded aggregation, will be crossed with both mice that develop Alzheimer-like amyloidosis and mice that develop tauopathy. Outcomes in these mice will be compared to that of crosses of mice that express SODWT:YFP, which is much less prone to aggregation, with the same amyloid and tauopathy models. For reasons explained within the proposal, we contend that the SOD1-based vectors are well suited for the questions posed by this application. Our goal in these experiments is to establish whether the accumulation of one type of misfolded protein in the mammalian CNS diminishes proteostasis function to such a level that the system fails to prevent the secondary misfolding of other proteins that are intrinsically vulnerable to self-seeded aggregation.

DESCRIPTION (provided by applicant): Project Summary Alzheimer's Disease (AD) is now recognized as a disorder with a long incipient phase in which a myriad of pathologic abnormalities occur long before the first disease symptoms appear. The first pathologic event is the deposition of A¿ peptides in deposits of diffuse amyloid and in structures referred to as senile plaques. These changes may begin to occur some 20 years before the onset of symptoms. At death, individuals that exhibit only amyloid pathology are largely cognitively normal. Individuals that exhibit mild cognitive impairment at the time of death may exhibit a range of pathologic features, but a large subset show abundant amyloid pathology and some level of abnormal tau pathology (ranging from accumulation of phosphorylated tau to neurofibrillary tangles). Individuals that exhibit more severe cognitive impairment, meeting clinical criteria for diagnosis of AD, at autopsy will invariably have significant tau pathology alng with amyloid (more variable in severity). This human data argues persuasively that the deposition of amyloid in some manner induces a secondary misfolding of tau. However, the inability to model the staged transition from primarily amyloid pathology to amyloid and tauopathy has impeded our ability to define the molecular mechanisms that underlie the apparent secondary induction of tau pathology in AD. To date there have been multiple attempts to produce models that replicate this important feature of AD using various transgenic and gene-targeting approaches. A key drawback to the transgenic models has been that often there is a need to express high levels of a transgene in order to raise the levels of aggregating proteins high enough that they will spontaneously seed fibrillar aggregation. With this R21, we seek to determine whether seeding amyloid pathology in mice that express much lower levels of mutant human APP and Tau will produce models in which amyloid precedes the appearance of tau pathology in a manner that is completely dependent upon the induction and severity of the amyloid pathology (as appears to occur in humans). We propose that we can generate hosts that would accomplish this goal by generating mice that co- express low levels of mutant human APP (amyloid deposition first appears at 18 months) and mutant human tau (no pathology). We hypothesize that seeded induction of amyloid deposition in mice that co-express these transgenes will create a model in which amyloid deposition in specifically accelerated, followed by a secondary induction of tau pathology. We propose that if successful, such models could be used to better understand the mechanisms that drive the transition between primarily amyloidosis to amyloid with tau pathology.

? DESCRIPTION (provided by applicant): Cu-Zn superoxide dismutase 1 (SOD1)-linked familial amyotrophic lateral sclerosis (fALS) is an extremely heterogeneous disease phenotypically with diverse clinical symptoms that can originate in upper or lower motor neurons and with a wide range of disease durations, from as short as a year to as long as 20 years. The duration of disease is largely a function of the speed with which symptoms spread along the neuraxis until motor neurons involved in respiration become affected. The question of how the disease seems to spread is one of the major unanswered questions in the study of ALS. Over the past few years, there has been increasing evidence that one mechanism by which the disease spreads may involve a prion-like propagation of a toxic misfolded protein from cell to cell along anatomically connected pathways of the CNS. Proteins that can transmit toxic conformations between cells often can also experimentally transmit disease between individual organisms. To survey the ease with which motor neuron disease (MND) can be transmitted, we injected spinal cord homogenates prepared from paralyzed mice expressing mutant superoxide dismutase 1 (SOD1-G93A and G37R) into the spinal cords of genetically vulnerable SOD1 transgenic mice. From the various models we tested, one emerged as showing high vulnerability. Tissue homogenates from paralyzed G93A mice induced MND in 6 of 10 mice expressing low levels of G85R-SOD1 fused to yellow fluorescent protein (G85R-YFP mice) by 3-11 months, and produced widespread spinal inclusion pathology. Importantly, second passage of homogenates from G93A¿G85R-YFP mice back into newborn G85R-YFP mice, induced disease in 4 of 4 mice by 3 months of age. Homogenates from paralyzed mice expressing the G37R variant were among those that transmitted poorly, regardless of the strain of recipient transgenic animal injected, a finding suggestive of strain-like properties that manifest as differing abilities to transmit MND. Although these preliminary findings are very exciting, we recognize that our studies to date are underpowered and we cannot fully assess the ease with which SOD1-linked motor neuron disease (MND) can be transmitted between animals without a much larger effort. Aims 1 and 2 propose such an effort to better understand genotype/phenotype interactions in transmitting MND in these models. We also have very exciting evidence that we might be able to create a model in which we could initiate disease focally, by injecting the infectious tissue homogenates from paralyzed mice in to the sciatic nerves of vulnerable transgenic models. However, again, our preliminary data is limited and much larger effort is required. Aim 3 proposes such an effort. Collectively, our studies are designed to better establish the biological relevance of SOD1-MND transmissibility and to build model systems that would enable investigations into the mechanisms of disease progression.

DESCRIPTION (provided by applicant): One of the major gaps in our understanding of the evolution of Alzheimer's disease is how the deposition of amyloid triggers tauopathy. Moreover, it is now widely recognized that it is common for the CNS of individuals with a neurodegenerative phenotype to develop multiple pathologic abnormalities. The basis for the preponderance of mixed pathology is poorly understood. We hypothesize that insults that compromised function of the proteostasis network may lay the foundation for the development of mixed proteinopathies. The basic concept here is that high levels of misfolded proteins produce an added burden on the proteostatic network by occupying various activities required to dissociate such aggregates and degrade the misfolded proteins. This concept was first uncovered in C. elegans models, where the expression of proteins that produce intracellular inclusions leads to the secondary misfolding of by-stander proteins that are particularly dependent upon the proteostatic network. Recently, we have extended this concept to mammalian model systems. In proteomic studies of brain from mice with high levels of Alzheimer-amyloidosis, Drs. Xu and Borchelt identified a number of cytosolic proteins that appeared to lose solubility - a finding that is consistent with the hypothesis that amyloid deposition can, by some manner, impinge on the function of the proteostatic network to cause secondary misfolding. The Lewis laboratory also recently found that two independent lines of mice that model tau pathology also develop cytoplasmic TDP-43 immunoreactive inclusion pathology. Thus, in our mouse models, we are beginning to uncover evidence that the accumulation of one misfolded protein, can by some manner, impact on the folding of others. Our central hypothesis is that these secondary pathologies are the consequence, at least in part, of a disturbance in the cellular protein quality control network, or proteostasis network, to cause collateral misfolding. In the present application, we propose 3 Aims that seek to determine the contribution of proteostatic network dysfunction to the evolution of AD-related pathology. Aim 1 will create a novel paradigm in which mutant tau expression is induced in a preexisting environment of amyloid pathology and disturbed proteostatic function. Aim 2 will determine how the mixed pathology of AD may synergize to produce by-stander misfolding and whether the severity of such misfolding produces functional deficits in critical cellular processes (e.g. energy metabolism). Aim 3 will determine whether augmentation of proteostatic networks mitigates by-stander misfolding. Collectively, these studies will provide new insight into how AD-related pathology impacts CNS protein homoeostasis and whether augmentation of the network in later stages of disease may provide significant benefit.

? DESCRIPTION (provided by applicant): Cu-Zn superoxide dismutase 1 (SOD1)-linked familial amyotrophic lateral sclerosis (fALS) is an extremely heterogeneous disease phenotypically with diverse clinical symptoms that can originate in upper or lower motor neurons and with a wide range of disease durations, from as short as a year to as long as 20 years. The duration of disease is largely a function of the speed with which symptoms spread along the neuraxis until motor neurons involved in respiration become affected. The question of how the disease seems to spread is one of the major unanswered questions in the study of ALS. Over the past few years, there has been increasing evidence that one mechanism by which the disease spreads may involve a prion-like propagation of a toxic misfolded protein from cell to cell along anatomically connected pathways of the CNS. To investigate this, we initiated studies and obtained tantalizing evidence that we can transmit ALS and that different ALS mutants may have strain-like attributes that modulate transmission. We have now been able to induce paralytic disease with aggregate SOD1 pathology in a strain of transgenic mice that express mutant SOD1 (G85R) fused to YFP. This strain of mice expresses at levels too low to induce disease on their own. Intraspinal injection of homogenates from a paralyzed G93A SOD1-overexpressing mouse transmits motor neuron disease to G85R-YFP mice in 6-11 months, and serial passage of spinal cord homogenates of these animals to naïve G85R-YFP mice produces disease in less than 3 months. This appearance of adaptation is typical of prion disease. In contrast, administration of homogenate from paralyzed G37R SOD1 mice failed to induce disease in the same recipient mice. The question this grant seeks to resolve is whether different SOD1 mutations produce conformations that produce strain-like properties that manifest as differing abilities to transmit motor neuron disease and if these abilities relate to rates of disese progression in patients. To establish whether mutations associated with slowly progressing ALS represent poorly transmissible strains of mutant SOD1, we will test 5 mutants that are found in slowly progressing SOD1-linked fALS and 5 found in rapidly progressing disease and compare these to WT SOD1 as well as negative controls. We will test both soluble and insoluble fractions of cell culture homogenates from cells transiently transfected with mutant SOD1 and recombinant hSOD1 fibrils created in a cell-free system as our inoculum, and test their potential to cause disease by direct spinal injection in G85R-YFP mice at postnatal day P0. Overall, our proposed studies could have an enormous impact in the field as they could lead to the development of an experimentally facile model of transmitted ALS and facilitate exploration of the causes for the heterogeneity of clinical symptoms observed in SOD1-linked fALS patients.

Summary: New Drug Discovery Paradigms for Synucleinopathies Dysfunction in cellular proteostasis leads to abnormal accumulation of misfolded proteins implicated in the pathology of several neurodegenerative diseases. Alpha-synuclein (aS) is the primary component of intracellular inclusions known as Lewy bodies?the cytopathological hallmark of Lewy body dementias, Parkinson?s disease and a frequent pathology in Alzheimer?s disease (AD). Both aS and tau pathology may overlap in AD because of the potential for inter-nucleation and aggregation of these two proteins. Thus, synucleinopathy is a high-priority target that bridges several CNS protein-misfolding disorders. Alpha-synuclein is a 140-amino-acid intracellular protein. It is an intrinsically disordered monomer but can adopt multiple ?- helical conformations on binding to lipid vesicles. This transition from disordered to ?-helical conformations is thought to lead to the formation of misfolded ?-sheet-rich structures that can form soluble oligomers and aggregates. Our hypothesis is that aS dimerization induced by membrane phospholipids is an early, rate- limiting step in protein-misfolding pathways that ultimately leads to synucleinopathies; and that a drug inhibiting this step will have therapeutic value. We have developed an ex vivo split-luciferase protein complementation assay to detect the initial dimerization of aS as it oligomerizes and aggregates. Our proposal is to use this assay in two innovative, high throughput screening (HTS) paradigms to identify novel aS aggregation inhibitors. One approach will take the standard long road to identify novel small molecules. We will screen 100,000 compounds in assay buffer conditioned with phospholipids that enhance aS dimerization. A panel of secondary assays will be used to characterize confirmed HTS hits for specificity, potency, mode of action and cellular activity. The expected Aim 1 milestone is identification of a novel lead suitable for nomination as a preclinical drug development candidate. Because of the mismatch between the small sizes of drug-like molecules compared to the expansive surface area involved in protein-protein interactions (PPI), HTS for protein-misfolding inhibitors is challenging. To potentially accelerate getting new therapies to the clinic, we propose a more unconventional approach in which we will attempt to identify a combination of two FDA approved drugs acting supra-additively by allosteric synergy to prevent the self-association and aggregation of aS. We will use a multiplexed-HTS method we developed for screening mixtures of FDA-approved drugs to discover combinations that stabilize monomeric aS structure, preventing self-association. The expected Aim 2 milestone is identification of a combination of two previously approved drugs that can potentially advance to proof-of-concept clinical testing for treating synucleinopathies faster than an individual new chemical entity. If successful, the new approaches we develop to discover novel PPI modulators could have broader utility for advancing understanding of other neurodegenerative diseases caused by protein misfolding and aggregation.

Is there a form of benign brain amyloidosis in aging? Although relatively rare, the identification of aged individuals with significant amyloid pathology (lacking tauopathy) that are cognitively normal is one of the lines of evidence that argues against the notion that amyloid-? (A?) deposition is a direct mediator of cognitive decline. Our study seeks a potential explanation for cases of amyloidosis that appear benign by testing the hypothesis that the brains of these pathologic aging (PA) individuals, as compared to Alzheimer disease (AD) brain, contain a distinct conformer of A? assembly that is essentially benign. One potential explanation for amyloid pathology that is benign, in regard to cognition, is that A? may assemble into structures that are conformationally distinct, with some conformations possessing an activity that diminishes cognitive function while others do not. To address this question, we will take advantage of recent studies that have shown that rate and character of amyloid deposition in vulnerable transgenic mice can be greatly influenced by injecting homogenates from human Alzheimer?s disease (AD) brains into their brains. Our preliminary studies, presented below, demonstrate that we have extended this model system to include brain homogenates of PA brains. As expected, we observed robust acceleration of A? pathology in transgenic APPswe/ind mice that were injected with two different AD brain homogenates. Although less consistent, we similarly observed accelerated amyloid pathology in mice injected with brain homogenates from 4 PA cases, with one seeding comparably to AD brain. We propose to seed the brains of two distinct lines of transgenic mice that develop amyloid pathology late in life with homogenates from AD and PA brains, and then assess whether pathology induced in these animals causes cognitive impairment. If PA represents a form of A? amyloidosis that is benign, then we would expect that only the mice seeded with AD homogenates will develop cognitive deficits. Apart from clarifying the distinction between PA and AD, our study may have therapeutic implications. If there are forms of A? amyloidosis that are benign, then it might be possible to chemically modulate the conformer of A? that is associated with cognitive decline to favor the benign forms and thus reduce the burden of A? assemblies that diminish cognitive function.

Title: APOE as a modifier of prion-like spread in dementia Abstract: The overall objective of this multidisciplinary project is to test the hypothesis that relative to APOE2 or APOE3, APOE4 facilitates the seeding and spread of misfolded A? and tau. Inheritance of an APOE4 allele increases the risk of developing Alzheimer's disease (AD) dramatically. Most studies suggest that the primary mechanism by which APOE genotype modulates the risk for AD is by influencing the deposition of A? peptide. However, there are studies that suggest APOE may also directly modulate the phosphorylation and misfolding of tau. Somewhat surprisingly, although the first descriptions of A? seeding in mice were reported more than 10 years ago and various human APOE models have been available for many years, there have been no studies of how APOE genotype may modulate seeding or spread of misfolded A?. Similarly, there has been no study of how APOE genotype may influence the spread of misfolded tau. We now propose three Aims to conduct a thorough and systematic assessment of how different isoforms of human APOE impact the prion-like seeding and spread of misfolded A? and tau. A component of our study will also asess whether different APOE isoforms may interact with these aggregating proteins to produce distinct strains of misfolded A? or tau. Aim 1 will determine the relative ability of APOE2, APOE3 and APOE4 to support the seeding of A? pathology, the spread of A? aggregates within the brain, and whether APOE isoforms modulates the strain characteristics of the seeded A? aggregates. Aim 2 will determine whether APOE genotype influences the spread of CNS tau pathology, and whether any observed influence in spread is potentially due to the emergence of distinct strains of misfolded tau. Aim 3 will examine whether APOE isoforms may differentially modulate A?-induced misfolding of tau. Collectively, these studies will produce a unique repertoire of animal models and provide the first assessment of whether different human APOE isoforms may be influencing that pathogenesis of AD by modulating the prion-like seeding or spread of misfolded A? and tau.

The presence of a-synuclein brain aggregates are a hallmark of a spectrum of neurodegenerative disorders, including Parkinson disease, Lewy body dementia and multiple system atrophy, and are associated with disease severity. Several studies suggest that this ?a-synuclein pathology may spread during disease progression by a self-templating mechanism and that nervous tissue containing aggregated a-synuclein could be a risk for disease transmission similar to prion disease. However, we recently demonstared that a non-??- synuclein factor enriched in central nervous tissue-derived white matter is sufficient to induce the progressive formation of a-synuclein inclusion pathology that can mimic prion-like transmission of a-synuclein pathology. These observations suggest the presence of a yet unidentified factor that can trigger a-synucleinopathy, similar to protein X, a putative auxiliary factor in prionopathies. Furthermore, this factor may have a pathogenic role in human disease. To resolve critical issues at the core of whether synucleinopathies should be re-classified as prion disorders and to elucidate mechanisms involved in disease progression we propose the following aims: 1) characterize the biochemical properties of the non-a-synuclein component that can trigger the progressive formation of a-synuclein, 2) determine whether a-synuclein aggregates can truely exhibit properties expected of classical prions, and 3) determine the relative prion-like transmission properties of central nervous tissue from patients with multiple system atrophy and Lewy body dementia, two phenotypically distinctive a-synucleinopathies. These highly integrated aims will provide important mechanistic and biologically relevant insights into the mechanisms and risk of pathology transmission from tissues containing a-synucleinopathies.  

The MPIs for this proposal independently co-developed and thoroughly characterized some of the most commonly used mouse models for research on Alzheimer?s disease and related disorders (ADRD); however, all existing tau and/or amyloid mouse models still have shortcomings which limit their utility and the questions that they can be used to answer. For example, the rTg4510 model which expresses human P301L tau through a doxycline-repressible system is one of the gold standard models in the field; however, rTg4510 is limited, in part, by its dependence on two unlinked transgenes, the early onset of tauopathy and cognitive dysfunction, and the leakiness of the tau expression. The overall goal of this proposal is to develop new models for the Alzheimer?s Disease field that overcome the shortcomings of existing models, ultimately providing an innovative platform in which the sequential nature of amyloidosis and tauopathy and the molecular pathways underlying their interaction can be examined in a streamlined, cost-effective manner. Under Aim 1, we propose to generate a model in which the CamKII-tetracycline transactivator transgene and the tau responder transgene (either WT or P301L) required for the conditional expression of tau will be co-injected and thus co-integrated into the murine genome which can subsequently transmit as a single allele. We will strive to develop a P301L tau/tTA model that will develop pre-tangle pathology at 12-15 months and tangles at 18 months of age. This new model will subsequently be fully characterized biochemically, pathologically, cognitively and structurally using MRI. Once established, these novel, conditional tau transgenics will provide a less expensive, more accessible model that develops tauopathy in mid to late life; enabling both studies aimed at accelerating and at slowing/abrogating the tauopathy. We also anticipate that this model, like the JNPL3 and rTg4510 tau models, will develop neuroinflammation and secondary TDP-43 proteinopathy. Under Aim 2, we propose to create a new conditional APP transgenic model through co-injection/co-integration using the cumate-repressible system to control APPswe/ind expression. No mouse model utilizes the cumate-repressible system in the brain and simply having the APP transgene under this alternatively conditional system positions this model for use in studies to identify interactions between APP and tau, ?-synuclein or other potential interactors. Finally, under Aim 3, we will crossbreed the new, single allele, tetracycline-repressible tau model with the single allele, cumate- repressible APP to allow the dissection of the interaction between tau (tauopathy) and APP (amyloidosis) in a sequential and systematic fashion. This innovative model, requiring a cost-effective, single breeding, permits independent control of both the tau and APP transgenes and will position the field to fill critical gaps in knowledge that no existing animal model in the AD arena currently allows.