Bradley T. Hyman, MD PhD - US grants

The major pathological alterations of Alzheimer's disease are neurofibrillary tangles and neurotic plaques. In addition, there is a recently-described accumulation of both amyloid protein and abnormal neurites in the neuropil. The goal of this application is to examine the anatomical relationships of these pathological changes to one another in the perspective of known neural system anatomy. The central hypothesis is that these pathological changes disrupt behaviorally important neural systems by involving their projection neurons, axons and dendrites, and terminal zones. The applicant's previous histopathological studies have suggested that particular projection neurons in the hippocampal formation characteristically and specifically undergo neurofibrillary degeneration. The studies outlined will build on these results by examination of the anatomical correlates of Alzheimer-related pathology in the hippocampal formation, amygdala, temporal neocortex and portions of the diencephalon using both immunohistochemical and histopathological methods. These areas are selected because they are established sites of Alzheimer-related pathology, are areas where the anatomical connection of neural systems have been documented in homologous regions in nonhuman primates, and are areas likely relevant to understanding Alzheimer-related cognitive changes. The exact cytoarchitectural and laminar location of neurofibrillary tangles and neuritic plaques will be determined by histochemical techniques. The localization of amyloid core protein (A4 immunoreactivity), neuritic fibrils in the neuropil (tau and Alz-50 immunoreactivity) and synaptosomal protein markers (synaptophysin and S7B8 immunoreactivity) will be determined by immunohistochemistry. These data will be used to test the specific hypotheses that (1) neurofibrillary tangles occur predominantly in cortical projection neurons; (2) neuritic plaques and amyloid protein occur in the terminal zone of degenerating neurons; (3) the diffuse neuropil alteration represents dystrophic dendrites and axons of affected neurons; and (4) there is a loss of synaptosomal markers in terminal zones of neurons undergoing neurofibrillary degeneration. A potential further benefit of these studies is the identification of specific neuronal populations that consistently undergo neurofibrillary degeneration. This can be used to establish a hierarchical table of relative vulnerability, and provide a construct to assess putative markers of neuronal degeneration by identifying neurons likely to be at risk for neurofibrillary changes. These studies hold the promise of defining, on a neural systems level, the anatomical relationships between the major histopathological markers of Alzheimer's disease. Moreover, these studies provide an opportunity to understand the contribution of these pathological changes to the demential of Alzheimer's disease.

A central mandate of the ADRC is to understand the pathophysiologic processes that underlie dementia. Nonetheless, the physical basis of cognitive and psychiatric symptoms in Alzheimer's disease (AD) remains uncertain, and there is a gap in our knowledge about age related changes in nondemented elderly. The Massachusetts ADRC has been collecting brain tissue from individuals who have been longitudinally clinically studied for 10 years. We have accumulated enough material to finally begin to construct a detailed model of clinical-pathological relationships in AD. We now propose to expand studies we recently initiated aimed at uncovering the temporal sequence of neuropathological changes in AD, and at answering whether or not neurofibrillary tangles (NFT), senile plaques (SP), neuronal loss an/or synaptic loss correlate with the clinical symptoms of AD. We will study AD patients, Down syndrome patients, and nondemented elderly individuals. Our results so far have led to several unexpected findings: SP do not correlate at all with duration or severity of dementia. We hypothesize that SP are in a dynamic equilibrium, both forming and resolving during the course of the disease. We suggest that A beta may be mobilized from the neuropil by ApoE or other potential A beta chaperone proteins. We will explore the recently reported association between sporadic AD and the ApoE-E4 allele by studying the influence of ApoE genotype on neuropathological phenotype. We also postulate that interactions of SP with inflammatory or proteolytic processes could lead to turnover of SP, and potentially to loss of neurons and synapses as well. We will examine SP turnover using a novel strategy based on the nonenzymatic accumulation of advanced glycosylation end products on long lived proteins. We and others find that NFT number in high order association cortices correlate with global measures of dementia. Synaptic loss has also been suggested as a physical substrate of dementia. We find that estimates of neuronal loss in high order association cortex correlate with clinical condition remarkably well. Surprisingly, however, our preliminary data suggest that neuronal loss far outstrips NFT formation, and that NFT formation may account for only a small minority of neuronal loss in the neocortex. Finally, neuropathological variables will be combined with the clinical data base. This will allow us to examine clinical and neuropathological heterogeneity, and to test hypotheses about clinical-pathological correlations. Looking to the future, we expect that additional candidate genes that are risk factors for or causative of AD will emerge, and the tissue samples and data developed in this project will be an ideal resource for determining clinical-pathological and genotype- phenotype correlations.

DESCRIPTION: (Applicant's Abstract) Apolipoprotein E and Alzheimer Disease The apo-e4 allele of apolipoprotein E has recently been found to be in striking genetic disequilibrium with the development of Alzheimer's disease (AD) in patients with late onset familial AD (Strittmatter et al., 1993). This observation has been extended by our own data and at least 4 independent groups to show that apo-e4 is found at a much higher than expected frequency in sporadic AD as well, and apo-E4 has accordingly been suggested to be an heritable risk factor for AD. Neuroanatomical, biochemical and molecular biological tools will be used to gain insight into the biological factors that mediate this increased risk. The results suggest one of two possibilities: either the apo-E4 protein interacts with pathophysiological processes to affect the course or likelihood of AD, or the apo-e4 allele is a genetic marker suggesting the presence of a causative or susceptibility factor nearby. Both possibilities will be explored; emerging evidence of biological interactions between apo-E and AB, a molecule firmly established as playing a role in AD, favors the first possibility. The working model is that apo-E binds AB in the neuropil, and transports it to apo-E receptors that are linked to clathrin coated pits for subsequent metabolism. The investigator hypothesizes that apo-E4 protein performs this "clearing" function less well than apo-E3. Already preliminary data support key features of this model: The investigator has confirmed genetic disequilibrium of apo-E4 and AD in an autopsy confirmed population of sporadic AD. The investigator also found that the amount of AB deposition is related to apo-E genotype, increasing from 3/3 to 3/4 to 4/4. Apo-E immunostains senile plaques (SP) in AD brains. Low density lipoprotein receptor related protein (LRP) immunostaining colocalizes with apo-E on SP, placing this apo-E receptor in close proximity to AB and apo-E. The investigator has developed in vitro models to study these interactions. Using ligand blots they have found specific and high affinity binding of AB to apo-E with Kd in the 10-20 Nm range. Furthermore, they have preliminary data using fibroblasts that express a high amount of LRP support the idea of cellular uptake of AB complexes in vitro. The investigator will characterize this uptake and study the cellular consequences and metabolism of AB. The application has 3 major aims: (1) to investigate the anatomical relationship among apo-E, its receptors, and AB deposition in AD brain tissue; (2) to investigate the interaction of AB with apo-E3 and apo-E4 in vitro. Determine if apo-E facilitates AB uptake or binding to cells in culture via LRP or other members of the LDL receptor family; and (3) to test the hypothesis that amount of apo-E4 expression is related to degree of Alzheimer neuropathological change or risk for dementia. These experiments will study the role apo-E and its receptors play in the pathophysiology of AD, and explore biological mechanisms that may underlie the increased risk for AD mediated by apo-e4.

DESCRIPTION (Investigator Abstract): This competing continuation application proposes to build upon our studies of the anatomic specificity of neurofibrillary tangles (NFT) and senile plaques (SP) to ask the next generation of questions regarding the disruption of the complex topography of human brain architecture by Alzheimer's disease (AD). We have found that the neuropathological changes of AD seemingly obey anatomical principles by selectively and specifically destroying projection neurons within limbic and association areas, leading to loss of feed forward and feedback projections and of neural systems that underlie aspects of normal memory and cognition. We now propose to move beyond anatomical descriptions to quantitative analyses in order to generate and test new hypotheses about AD pathophysiology. We have merged newly developed stereological anatomical techniques with computerized image analysis systems to create quantitative maps of cytoarchitecture and of AD related lesions.A major new thrust is a collaboration with Dr. Gene Stanley, an international leader in applying the mathematical techniques of complex systems analysis to biological systems. Together we will ask how the pathological alterations of AD deform the neural landscape. Quantitation will allow the comparison of individuals with different clinical duration or severity of illness, and different risk factors such as ApoE genotype or Down syndrome. Four specific aims will examine SP deposition, NFT formation, neuronal loss and neuronal cellular integrity. Already our preliminary results suggest that the approach of combining state of the art quantitative anatomical techniques with powerful mathematical analyses will yield new clues about the pathophysiology of AD. For example, we have discovered that the distribution of SP sizes fits a log-normal plot quite well, which argues in favor of some, and against other, hypotheses about SP formation. The SP size distribution plot has led us to generate a new hypothesis about AD deposition in AD (aim 1). Quantitative assessment shows that neuronal loss and NFT, but not SP, correlate strongly with duration of dementia. Moreover, neuronal loss parallels but outstrips NFT formation by an order of magnitude. We will map the locations of NFT and neuronal loss to determine whether they occur in register, to test.specific hypotheses about the relationship of NFTs and neural death. Surprisingly we find that there are long range correlations of the locations of NFT, and we have generated testable hypotheses to explain this observation (aim 2). We have devised methods to quantitate the features that classical anatomists use to distinguish cytoarchitectural fields - lamination, packing density, and topographic patterns - and found that these quantitative analyses reveal specific alterations in AD (aim 3). We will study the structural and metabolic integrity of individual neurons in relationship to NFT and SP (aim 4). In sum, we will develop and use quantitative techniques and analyses in order to test new hypotheses about the structure organization of the normal human cerebral cortex and the specific alterations that occur in AD.

Specific cytoarchitectural areas, subfields, and even lamina develop Alzheimer pathological changes, while other are spared. A pattern of hierarchical vulnerability defines areas that are at risk. This proposal will test hypotheses about molecular characteristics that could underlie this pattern of vulnerability. Our previous studies show neurons that develop neurofibrillary tangles (NFT) occur mainly in specific lamina in limbic and association cortices. Senile plaques (SP) are found in a different but consistent hierarchical pattern, with some brain areas substantially more affected than others. We now propose to combine quantitative anatomic analyses of pathological changes, immunohistochemical and in situ hybridization studies, and RNA analyses from microdissected brain regions to examine how neurons that develop NFT differ from those that do not, and how regions that accumulate SP differ from those that are spared. In the first aim, we will pursue preliminary results which that a change occurs in AD in the pattern of tau isoforms expressed. The second and third aims focus on possible tau kinases. Recent studies show that the MAP Kinase (MAPK) family of Pro directed Ser/Thr kinases (ERKs) phosphorylate tau in vitro in a pattern resembling the changes that occur in NFT. We will study ERKs 1 and 2, ERK3 and the p54 MAPKs. We will test the hypothesis that p54 MAPK phosphorylates tau in a PHF tau fashion. Glycogen synthase kinase 3 has also been implicated in tau phosphorylation. These kinases are all highly expressed in brain, but the neuroanatomical regional and cellular distribution of their expression is unknown. We will study the normal patterns of their expression a swell as determine whether or not changes occur in neurons that develop NFT and in the microenvironment surrounding SP. Pilot experiments will be aimed at determining the activation status of these kinases in the AD brain. In AIM 4, we continue our studies of molecules potentially related to SP formation. Our previous studies, in collaboration with Dr. Tanzi (project 4), analyzed expression of APP major splice forms and found no difference between AD and controls. We now turn to the anatomic distribution and expression of a novel isoform of APP, L-APP (lacking exon 15), as well as new members of the APP family, APLP1 and APLP2. We have observed that nitric oxide synthase neurons are spared in AD. In Aim 5, in collaboration with Dr. Young, (project 1), we will determine the phenotype of these spared neurons in terms of tau, kinase, and glutamate receptor expression. In sum, the overall goal of this project will be to advance the profile of vulnerability from anatomical description towards molecular phenotype by generating profiles of tau, the newly described ERK family of protein kinases, and APP family members in vulnerable and spared areas and cells.

disease /disorder etiology; laboratory mouse; nerve /myelin protein; pathologic process

PS1 mutations in humans invariably lead to an autosomal dominant form of early onset Alzheimer disease (AD). Transgenic mice expressing mutant PS1 increase the ratio of Abeta42/40, a hallmark of AD producing genes. We will study the PS1 and mutant PS1 transgenic mice developed in Core C, crosses with APP over-expressers (from Dr. Games) and PS1 null mice (from DR. Tonegawa). The 3 specific aims are to test the hypothesis that either a gain of function of PS1 or insufficiency of PS1 predisposes towards AD- like neuropathological changes, and to further explore the normal functions of PS1 or insufficiency of PS1 predisposes towards AD-like neuropathological changes, and to further explore the normal functions of PS1 and Notch/PS1 interactions. One of the strongest clues to PS1 function comes from the observation that a PS1 homologue in C elegans (sel-12) facilitates the function of lin-12, the C elegans homologue of the neurologic protein Notch. Sel-12 mutations in C elegans lead to an egg- laying defect that can be rescued with human PS1 or, less efficiently, with mutant PS1. Quite recently, two reports of PS1 knockout mice demonstrated severe developmental abnormalities, fetal or perinatal death, and alteration in Notch expression. One of the key answered questions about PS1 is, how do these observations of developmental changes relate to neurodegenerative changes in terminally differentiated neurons? We have attacked the problem of learning more about the functional roles of Notch and PS1 in post-mitotic neurons by developing a system that allows transfection of constitutively active Notch into primary neurons in culture. We have discovered that Notch robustly impairs or delays neurite outgrowth. This discovery shows a unique role for Notch activity even in terminally differentiated neurons, suggesting perhaps a role in neuroplasticity. Importantly, it also provides an assay to examine the interaction of PS1. We used this assay to test the hypothesis that over- expression of PS1 facilitates Notch function and preliminary data suggest that this is the case. We will now examine how this relates to mutant PS1, as well as physiologic interactors of endogenous Notch such as the inhibitor Numb. Taken together, this project aims at examining the neuropathological phenotypes of PS1 transgenic, and heterozygote PS1 null transgenic mice with age, to study the effect of these genetic manipulations on neural development; and to use primary neuronal cultures to examine the functional relationships between Notch and PS1 in vitro. These studies are central to the overall themes of the program project, and it is our belief that together these experiments will strengthen our understanding of the normal functions of PS1, and lead to insight into how PS1 dysfunction predisposes to AD.

Recent data have linked two different rare mutations in the a synuclein gene with early onset familial Parkinson's disease. We and other have found a synuclein immunoreactivity to be present in Lewy bodies and other pathological lesions in the Parkinson disease brain, even in sporadic cases. These data suggest that a synuclein plays a critical role in the disease process. a Synuclein is predominantly expressed in neurons. Its function is unknown. This proposal will study a synuclein using three complementary systems: human brain samples, tranfected primary hippocampal or brain stem neurons, and over-expressing transgenic mice. We have developed novel applications of confocal scanning laser microscopy and fluorescence resonance energy transfer (FRET) to study the subcellular distribution of a synuclein and a synuclein-ubiquitin conjugates in brain. Transfection of primary neurons with Green Fluorescent Protein-alpha synuclein fusion constructs reveals a membrane association at presynaptic specializations in living neurons- application of FRET techniques to these neurons will be used to study a synuclein intramolecular interactions with membranes, and gain insight into a synuclein (and mutant a synuclein) conformation. Moreover, a synuclein GFP changes its location and merges with the membrane upon depolarization. We will study the impact of the mutations and of other stimuli such as MPP+ on this phenomenon using 2 photon confocal microscopy. We will also propose to generate transgenic mice with wild type and both mutant forms of a synuclein using the hamster PrP promoter which has previously been successful in other neurodegenerative disease models. We will build on the results of aims #1 and #2 to anatomically phenotype the mice, as well as relying on the resources of the Center to examine the mice for behavioral or electrophysiological (Dr. Graybiel) or neurochemical (Dr. Penny) abnormalities. Together, we propose an integrated, comprehensive approach to understanding the role of a synuclein in Lewy body diseases.

laboratory mouse; transfection /expression vector

This project will use anatomical and neuropathological methods to assess the kinetics and life history of Abeta deposition in the Tg2576 human APP (hAPP-Sw) transgenic mouse, which develops age related cerebral amyloidosis, behavior deficits, and electrophysiologic deficits. In the first aim, we will use quantitative anatomical, immunohistochemical and in situ techniques to characterize the time course of cerebral Abeta deposition, and evaluate the consequences of hAPP and Abeta deposition. Stereological techniques are employed to assay neuronal loss, synaptic loss, alterations in mRNA expression, and gliosis. We have recently observed phosphotau positive neurites surrounding Abeta deposits; we will characterize them immunohistochemically at the light and EM level. Dr. Chapman (project 4) has found diminished long term potentiation in the hippocampus of aged mice; we will examine glutamate receptors using quantitative immunohistochemical, in situ, and ligand binding assays. The second aim takes advantage of our observation that Abeta deposits occur specifically in the outer molecular layer of the dentate gyrus, in a region overlapping with the perforant pathway terminal zone. We will lesion the perforant pathway in young mice to test the hypothesis that Abeta in the deposits come from axonal terminals of hAPP expressing neurons. We will also lesion the perforant pathway that plaques have already formed to test the hypothesis that diminishing the among of Abeta synthesized (in the terminal zone) may uncover clearance mechanisms. Lesions of the perforant pathway are known to cause retrograde degeneration of layer II of entorhinal cortex, the cells of origin of the projection. We will examine the influence of the hAPP695Sw and hAPP695 transgenes and of Abeta deposits on the degree of degeneration. The third aim uses genetic manipulations to test hypotheses about Abeta production, deposition and clearance. We will use crosses of Tg2576 with knockout or over-expressing lines to evaluate the roles of apolipoprotein E, presenilins, a putative Abeta receptor (macrophage scavenger receptor) on the deposition of Abeta. We will also study, in collaboration with projects 1 and 2, the influence of strain background, and the impact of inducible or repressible hAPP transgenes on Abeta deposition and neuronal and synaptic loss. In sum, this project will test specific hypotheses about Abeta deposition and its consequences, and provide histopathological correlates for each of the other projects in the PPG.

DESCRIPTION (Reproduced verbatim from the applicant's abstract): This competing renewal application will extend the applicant's studies on the relationship of the a2M receptor/ LRP to AD. LRP, a large multifunctional receptor with multiple ligand binding domains, is the major neuronal receptor for apolipoprotein E (apoE), a2M, and the KPI-containing forms of the APP, all implicated in the pathophysiology of AD by both biochemical and genetic evidence. Our overarching hypothesis is that Abeta deposition is central to AD, and that LRP is in a critical position to influence the balance of Abeta generation and clearance. Both apoE and a2M form stable complexes with Abeta. Aim 1 tests the hypothesis that apoE/Abeta and a2M/Abeta complexes are cleared and that this clearance is via LRP. We will determine the specific domains on LRP that mediate these processes. Aim 2 explores our recent observation that LRP binds and internalizes full length APP, and that blocking this endocytic process decreases Abeta generation and secretion. The discovery that the LRP intracellular domain binds the adapter protein Fe65, a protein which is known to form a complex with the intracellular domain of APP, also implicates APP-LRP complex formation as being important in APP trafficking and potentially Abeta generation. We have also found that LRP undergoes PKC-mediated serine phosphorylation, and will test the hypotheses that this affects LRP trafficking, or interactions with Fe65 or APP. Aim 3 tests the hypothesis, based on new preliminary data, that LRP may serve a role both as an endocytic receptor and as a novel type of signaling receptor in neurons. We have observed that activated a2M induces a RAP-blockable, calcium response in neurons (but not non-neuronal cells), and will characterize this calcium response. Ligand-competent a2M also induces an increase in LRP amount and cell surface LRP; our working hypothesis is that the amount of cell surface LRP is regulated in part by this calcium response. The proposed studies will continue a highly productive and long standing collaboration among investigators at the Massachusetts General Hospital and the American Red Cross in order to pursue new exciting leads which impact our understanding of pathophysiologic mechanisms in AD, and which: will lead to additional insight in the neurobiology of LRP.

There are two overarching goals for the Neuropathology Core. The first involves neuropathological evaluation of brain tissue obtained at autopsy on subjects in the Program Project. These analyses will be performed: (1) to provide a neuropathological diagnosis on all subjects who come to autopsy, (2) to perform quantitative neuropathological data on all subject who come to autopsy, and (3) to perform biochemical analysis for Abeta and synaptophysin on the brain samples of all subjects who come to autopsy. In addition, Core D will evaluate Abeta 40 and Abeta42 in plasma samples that are collected annually, to determine if these measures can serve as biomarkers of 'conversion' to AD.

anatomy; tau proteins; biological models; genetically modified animals; laboratory mouse;

The scientific core consists of two major imaging laboratories to support program project members with state of the art equipment and methodologies. Aim 1. Provide support to investigators to design and carry out anatomically based analyses utilizing quantitative neuroanatomical techniques including stereology based statistically unbiased approaches. Aim 2. Assist investigators in imaging blood brain barrier breakdown and MMP uregulation, both in vivo and ex vivo, using extravasation of fluorescent indicators. Aim 3. Assist investigators in utilizing laser speckle contrast imaging in studies of cerebral blood flow and in the application of multi-spectral in-vivo imaging of intrinsic (e.g hemoglobin concentrations) and extrinsic (e.g. plasma leakiness indicators and MMP activatable fluorescence markers) contrast agents. Advance the methodology to provide more robust measures of blood flow and facilitate investigator driven advanced analyses. Aim 4. Develop robust high-field MRI measurements of CBF in rodents, and assist investigators in the acquisition and analysis of anatomical and functional MRI data. Taken together, these aims both support the scientific aims of the program project as a whole, while the needs of the projects serve to advance the methodology to facilitate investigator driven analyses.

Misfolded, hyperphosphorylated tau is the major constituent of neurofibrillary tangles in Alzheimer disease, as well as the inclusions in Pick disease and progressive supranuclear palsy. Twenty different mutations in the tau gene cause neurodegeneration in fronto-temporal dementia Parkinsonism-17, some influencing splice ratios and others changing tau's sequence. We postulate that alterations in tau conformation are a common underlyiing theme in these disorders. We propose to develop new technologies to tackle the question of tau conformation and tau expression at the cellular level, using neuropathological material from the ADRC brain bank and from colleagues across the country. In the first aim, we will develop novel fluorescence resonance energy transfer (FRET) techniques using fluorescence lifetime imaging to monitor the proximity of different domains of the tau protein in neurofibrillary tangles. Already we have found a unique folded structure in which the N terminus is folded back upon the microtubule binding domain region, while the C terminus is in close proximity to the proline rich domain. The second aim utilzes laser capture microdissection methods to identify and select individual neurons with (or without) neurofibrillary tangles for protein and mRNA analyses. The latter will be carried out using quantitaitve PCR approaches as well as a new method (called polony exon typing) that determines all 6 tau isoforms from microscale amounts of mRNA. In aim 3, the results obtained from the study of neurofibrillary tangles in Alzheimer disease will be compared to parallel studies of non-Alzheimer tauopathies: sporadic Pick disease, progressive supranuclear palsy, and cases of FTDP-17 with known tau mutations. Development of these novel methods will allow for examination of protein structure, expression, and mRNA patterns with a cellular level of resolution. These novel technologies will not only address important questons about tau's role in neurodegenration, but also provide a platform for further investigations of protein and mRNA in Alzheiemr and related dementias. Furthermore, the proposed program integrates closely with the core resuorces of the ADRC and with the other scientific projects within the ADRC, as well as other ADRCs and scientific centers of excellence across the country and in Europe.

Alpha-synuclein-containing Lewy bodies are a central pathological feature of Parkinson's disease. Fly, rat and mouse models of alpha-synuclein overexpression have been developed that lead to inclusion formation and toxicity of dopaminergic neural systems. By analogy to studies performed in models of trinucleotide repeat diseases, Bonini and colleagues used the transgenic fly model of alpha-synuclein overproduction and showed dramatic protection against dopaminergic cell death by overexpressing the molecular chaperone, heat shock protein 70 (Hsp70). We now show that Hsp70 and related molecular chaperones stain Lewy bodies in human neuropathological tissue, and overexpression of Hsp70 and cochaperone proteins prevent alpha-synuclein inclusions in an in vitro mammalian cell system. We hypothesize that molecular chaperones play a role in preventing alpha-synuclein misfolding, and perhaps in directing misfolded alpha-synuclein towards degradation. We will study the extent of HSP involvement in Lewy body disease by examining Lewy body tissue using confocal microscopy and the new fluorescent lifetime imaging technique, FLIM, to evaluate the conformation of alpha-synuclein in pathological specimens and relate that to HSP expression. We have found that alpha-synuciein in cells, transgenic mice and human DLB appears in a triton X-100 insoluble fraction; our preliminary data suggest that overexpression of Hsp70 blocks this in cells and transgenic mice. We will extend our in vitro studies of alpha-synuclein aggregation to determine which set of HSP cochaperones are most effective at preventing alpha-synuciein inclusions and toxicity, and whether down-regulation of HSPs leads to increased inclusions and toxicity. Finally, we will use transgenic animals and new gene transfer techniques to determine the effects of overexpression of HSP on alpha-synuclein inclusions and dopaminergic terminal toxicity observed in a transgenic mouse overexpressing alpha-synuclein, Line D. Taken together we propose an integrated set of aims to study the role of HSPs in alpha-synuclein aggregation and toxicity.These projects also integrate tightly with the aims described in the Lindquist, Graybiel and Standaert projects.

[unreadable] DESCRIPTION (provided by applicant): Neuroanatomical alterations in Alzheimer disease. Alzheimer disease (AD) dementia affects 1 in 10 American families, and costs more than 90 billion dollars a year. Genetic and neuropathological evidence highlight a pathogenic role for Abeta, but the exact mechanism of how Abeta alters neural system function is uncertain. Nonetheless, Abeta is a key therapeutic target, and clinical trials using anti-Abeta approaches are underway. In the last 5 years of this grant we developed in vivo multiphoton microscopy techniques to image Abeta deposits in living transgenic mouse models, and discovered that anti-Abeta immunotherapy can clear existing plaques. Our goal is to test the hypothesis that Abeta induces synaptic failure, and to examine the consequences in both mice and humans of anti-Abeta immunotherapy. Aim 1 examines hypotheses linking plaques with alterations in axons. Whether axonal dystrophies precede or follow plaque formation will be examined by imaging neuronal processes and plaques in young APP X YFP mice, then re-imaging them at later times. Transport of molecules and organelles near and distant from plaques will be examined in vivo using photoactivatable GFP constructs, introduced by AAV and Lentiviral vectors. Aim 2 builds on the observation that postsynaptic elements, including dendritic spines, are lost near plaques. We will examine whether dendritic spines are less stable near plaques by imaging spine turnover. We will also determine if dendritic segments near plaques are functionally less active using a new genetically encoded reporter of synaptic activation, a 3'-5' UTR CAMKIIalpha/EGFP chimeric molecule. Passive immunization against Abeta, and application of oligomeric Abeta, will be used to test the hypothesis that Abeta is directly responsible for these axonal and dendritic pathologies. The third aim proposes parallel studies of pre and postsynaptic structures in human AD autopsy material, including studies of autopsy tissue from individuals who had participated in the first immunotherapy trial, AN1792. Case reports show that plaque clearance occurs after vaccination. We have now organized a collaborative group of neuropathologists from multiple sites to examine systematically human tissue in which antibody mediated clearance of plaques has occurred. We believe that the questions posed in this application have high clinical relevance, and the severity and prevalence of the disease add urgency to these studies. [unreadable] [unreadable] [unreadable]

OVERVIEW: Massachusetts Alzheimer?s Disease Research Center (MADRC) The MADRC is an interdisciplinary Center at Massachusetts General Hospital, Harvard Medical School, with two major strategic goals: 1) to understand the underpinnings of heterogeneity of Alzheimer?s disease and Alzheimer?s Disease Related Disorders, and 2) to develop techniques, strategies, and technologies to accelerate a cure. To approach these goals, we propose 7 Cores (Administrative, Clinical, Biomarker, Imaging, Neuropathology, Data, and Outreach). We will use advanced Imaging, Biomarker, and analytic approaches to study our Research Cohort, which we view as a state of the art clinical phenotyping laboratory. Autopsy follow up for validation of biomarkers and imaging, and therapeutic targets, is essential. We plan to study a diverse population, both in the sense of various pathophysiologies, and also with varying racial and ethnic backgrounds, and rely on a vigorous ORE core both to fulfill our own recruitment needs and to support clinical trials and other affiliated programs. We have refocused our Center?s activities by remodeling the Research Cohort to include subjects who are cognitively normal, have mild cognitive impairment, or dementia, targeting diverse causes of dementia including Alzheimer?s disease and related disorders (Frontotemporal dementia, Lewy Body Disease, and vascular lesions). We postulate that heterogeneity of progression, even among individuals with the same ?disease?, reflects underlying differences that can be uncovered, and that doing so will enhance clinical trial design by parsing variability from the very large, long cohort trials now in common use. Each Core is challenged to develop tools to help accelerate a cure ? from intermediate imaging/biomarker targets that may help with diagnosis (a patient?s state) or prognosis (a patient?s fate), to performing autopsies to validate those targets. We will develop markers of inflammation and neurodegeneration, believing that innate immunity and resilience play a role in addition to tangles and plaques. Development of statistical and data methods are critical for anlaysis. These tools will allow us to test hypotheses about faster or slower rates of progression in a deeply phenotyped cohort. A final goal, central to our mission, is to build the future by training the next generation of scientists, which we propose to do under the framework of the Research Education Component. Together, we hope to make a difference in these devastating diseases.

Gamma-Secretase is the final enzymatic step generating Abeta via intramembranous cleavage of APP. The same enzymatic activity appears to be responsible for cleaving other substrates, including Notch. Presenilin, initially identified as a gene in which mutations account for the vast majority of early onset autosomal dominant AD, is a major component of gamma-secretase. Enzymatic activity also depends on nicastrin, Aph- 1, and pen-2. We propose a model in which gamma-secretase components assemble, interact with substrates at a docking site, then cleave and release substrates. To test this model, we have developed a novel morphological technique based on advanced fluorescent microscopy methods, Fluorescence Lifetime Imaging (FLIM), as well as molecular and biochemical assays of APP and Notch cleavage. FLIM allows us to examine protein-protein proximity in living cells. Already we have identified APP-PS1 and Notch-PS1 interactions that are present even when gamma-secretase inhibitors or dominant negative PS-1 mutations are used to block gamma-secretase activity, supporting a hypothesis that there is a non-catalytic docking site closely associated with gamma-secretase. Aim 1 focuses on understanding where in the cell interactions between gamma-secretase and APP and Notch occur, and examining in depth the molecular characterization of the putative docking site. Because FLIM provides pixel level resolution, we will test the hypothesis that Notch interacts with PS 1 after activation by ligand at or near the cell surface. In aim 2, we will test the hypothesis that substrates compete with one another, and that PS-1 mutations alter substrate interactions with gamma-secretase. Finally, in collaboration with Core C and Project 1, aim 3 outlines experiments targeted at additional gamma-secretase components (Nicastrin, Aph 1a and b, and pen-2) using RNAi, dominant negative, and over-expression approaches, to determine if genetic or pharmacologic manipulations alter gamma-secretase -substrate interactions. Taken together, our studies address 4 fundamental questions: the spatial paradox of where in the cell gamma-secretase and its substrates come together, whether there are distinct and separable docking and catalytic sites, where they are in the gamma-secretase complex, and whether substrates that give rise to novel signal transduction cascades compete with one another. Taken together, we will perform a detailed, integrated study of gamma-secretase-substrate interactions that may lead to new avenues for therapeutic interventions.

Description (provided by applicant): This "Challenge grant" application addresses the Broad challenge area: translational science: Specific challenge topic: 15-NS-103. Demonstration of "proof of concept" for a new therapeutic approach in a neurological disease. Our goal is to utilize a new concept in gene therapy -viral transduction of brain endothelial cells - to provide a reservoir of therapeutic molecules for the treatment of Alzheimer disease (AD). New techniques form Dr Davidson's laboratory have provided ways to engineer the capsid of adenoassociated viruses (AAVs) so that they are targeted specifically to brain capillary endothelial cells. Since the brain capillary network is extraordinarily dense, this provides a mechanism to bathe the CNS in proteins generated by the transduced cells. Dr Hyman has studied amyloid deposition in transgenic models of AD using in vivo multiphoton imaging, which allows determination of the rate of amyloid deposition by longitudinal imaging. The current proposal aims to combine these two state of the art laboratory's efforts. It is already established that inheritance of the apolipoprotein E4 allele (APOE4) increases risk for AD by about 3 fold compared to the common APOE3, whereas inheritance of the rare APOE2 allele is protective, and decreases risk by about half. Age of onset is similarly impacted, with the APOE2 gene associated with a 2 decade delay in onset of dementia compared to APOE4. APOE4 is also associated with much more amyloid deposition in the AD brain than APOE3 or APOE2. We propose to develop, in transgenic mouse models, a way to introduce the APOE2 gene product into the central nervous system by peripheral injection of an AAV targeted specifically to brain endothelia, and test the hypothesis that its intracerebral expression will delay progression of amyloid deposition. Similarly, we have developed a single-chain anti-Abeta antibody which we will test as an alternative means of diminishing amyloid deposition. We postulate that the effective delivery of APOE2 or antibodies to the CNS may lead to a breakthrough in Alzheimer therapy.

The amyloid p-protein (Afi) is closely linked to the pathophysiological processes that lead to Alzheimer disease (AD), and likely are ultimately responsible for the neural system collapse that causes the clinical syndrome of dementia. However the mechanistic role of Afi and the form in which it is toxic remain controversial. One hypothesis is that the insoluble amyloid plaque itself is deleterious to the brain. Contrasting this argument is the hypothesis that the damage is caused by soluble Aft oligomers. The overall goal of his Project is to examine these two competing hypotheses critically and determine whether one or both are more likely. In this research plan, we propose three specific aims built around advances in biochemical and morphological techniques. To learn whether and how soluble A(3oligomers, monomers and/or plaque cores correlate with the histopathological changes (gliosis, Ap deposits, neurofibrillary tangles), we will quantify the levels of soluble oligomers and monomers by new Size Exclusion Chromatography and sensitive IP/Western techniques and correlate these findings with each subject's detailed neuropathological phenotype and with clinical information. A second approach to these questions will utilize a newly developed histological preparation that allows visualization of individual synaptic elements, fibrillar Ap, and oligomeric Ap simultaneously. We have observed that oligomeric Ap-directed antibodies reveal a halo around plaques that corresponds to the region around plaques that have diminished dendritic spine density. Furthermore, oligomeric Ap-directed antibodies demonstrate puncta that co-localize with PSD95 positive dendritic spines. These observations motivate an analysis of oligomeric Apand synaptic change that will take aim at fundamental mechanisms of Ap-associated synaptic loss. These studies will address a central unresolved question: how do soluble oligomeric forms relate to the fibrillar, histologically detected forms of Ap. Moreover, our experiments will also directly address whether Ap in cognitively normal controls who have Ap deposits differ from those in subjects with AD. By taking advantage of well characterized material from the MADRC Neuropathology Core's Brain Bank, Clinical Core evaluations, and Statistical Core expertise, we plan to test hypotheses directed at the relationship of Ap to cognitive impairment and neuronal toxicity. RELEVANCE (See instructions): Project 3 of the Massachusetts ADRC will test the competing hypotheses of the role of Ap protein and the form in which its toxicity is associated with cognitive dysfunction and Alzheimer's disease neuropathology.

The Administrative Core's goals are to set the overall direction for the Center, ensure optimal utilization of Center resources, and provide a sense of "Centerness". The Massachusetts ADRC's goals are to provide the infrastructure and environment to enhance and facilitate cutting-edge basic, translational, and clinical research in Alzheimer disease and related disorders, as well as provide a unique and exciting training environment for the next generation of clinicians and scientists working in these arenas. Our aims are outlined in the RFA, and include creating an administrative structure and infrastructure to facilitate the work of the Center. This includes administrative support for the Mass. ADRC's Clinical, Data Management and Statistics, Neuropathology, Education and Information Transfer Cores and research projects. We provide formal mechanisms to augment collaborative activities with the scientific and lay communities, and to enhance collaborations across departments, units, local institutions, and national efforts to create an energetic multi- and cross-disciplinary approach to understanding AD. The Administrative Core will also provide fiscal accountability and business management expertise for the Center, and ensure compliance with institutional, NIH and HIPAA policies and requirements related to human subjects and animal research. A second major effort of the Administrative Core is to organize 4 major Committee programs: An Executive Committee consisting of Core and Project leaders and additional Key Personnel, to advise the Director on scientific and administrative matters;an External Advisory Board to provide guidance to the Center on an annual basis;a Pilot Project Review Committee to recommend 3 pilot projects on an annual basis;and an Internal Advisory Board consisting of senior scientists, clinicians, and administrators from our constituent institutions to coordinate ADRC efforts with local resources. Additional committees and infrastructure are described to recruit pilot projects, to co-ordinate IRB approvals for projects, to coordinate recruitment of subjects for studies, and to monitor the scientific and administrative functions of the Center. RELEVANCE (See instructions): The Administrative Core of the Massachusetts ADRC provides the infrastructure and management of resources to enhance the training environment for clinical research in Alzheimer's disease and related disorders.

DESCRIPTION (provided by applicant): This application, responding to RFA-OD-09-005 Recovery Act Limited Competition: Core Centers for Enhancing Research Capacity in U.S. Academic Institutions, proposes to recruit and support two new Assistant Professor Faculty to the Massachusetts Alzheimer Disease Research Center/ (ADRC) Mass General Institute for Neurodegenerative Disease (MIND) at Massachusetts General Hospital/Harvard Medical School. The recruited faculty will either be a clinician scientist or a PhD scientist, and focus on translational aspects of the basic science programs. They will participate in activities and take advantage of resources of the ADRC, MIND, the Massachusetts General Hospital and the Harvard Medical School-wide Harvard NeuroDiscovery Center. This request is to provide resources to fill a position focused on neuroplasticity, recognizing the marked advances that have taken place in the context of developmental biology and the technical advances that have made possible monitoring and manipulating axonal and dendritic structure and function in the mature nervous system. The second position is aimed at a better understanding of Brain aging, in the absence of neurodegenerative disease, motivated in part by the recognition that aging is the single greatest risk factor for neurodegeneration, and in part by the still mysterious underlying biology that leads to subtle impairments in cognitive function with age, even in the absence of neurodegenerative changes. Both programs of research complement existing basic science and clinical programs ongoing at the Massachusetts General Hospital and MIND, including substantial depth in related studies at Harvard Medical School and across the Harvard campuses. Senior members of the Harvard faculty will serve as a Selection committee and as a Mentoring committee, and a detailed 4 year program to support the new faculty recruits has been outlined including formal mentoring, review of project designs, and continued financial support. The candidates will be encouraged to design independent research programs that take advantage of, and complement, existing infrastructure in clinical and research projects, and thus enhance the research capacity of the ADRC/MIND Center. PUBLIC HEALTH RELEVANCE: We plan to use Recovery Act funds to jump start the recruitment and hiring of two new young faculty, whose missions will be to complement the extensive ongoing research at Massachusetts General Hospital and Harvard Medical School in the areas of brain response to neurodegenerative disease induced injury (neuroplasticity) and "normal" brain aging.

DESCRIPTION (provided by applicant): Neurofibrillary tangles are one of the primary hallmarks of Alzheimer disease, and closely correlate with clinical severity. They consist of hyperphosphorylated tau that aggregates in the neuronal cell body. Cross sectional histological studies suggest that tangles kill neurons. Tau disaggregation is thus a major target of the pharmaceutical industry, with at least one drug already in human clinical trials. However, recent data suggest that tangles may not be toxic. We propose to test an alternative hypothesis, in which soluble tau induced toxicity leads to activation of apoptosis related cascades, followed by tangle formation. In order to test this model, we have developed in vivo multiphoton based longitudinal imaging techniques for tangles, caspase activation, and propidium iodide (a marker of loss of membrane integrity). Aim 1 applies this approach to Tg4510 P301L tau mice that develop NFT and neuronal loss. Surprisingly, we see caspase activation that appears to precede and predict tangle formation. We will also use array tomography, an advanced microscopic method using ultrathin sections of tissue, to determine the characteristics of the caspase positive, tangle negative cells to test the hypothesis that kinase activation leads to phosphorylation of tau, and that caspase activation leads to the truncation of tau -the combination of which causes tau aggregation into a tangle. We will use gene transfer and pharmacological studies to further dissect this pathway, to ask whether caspase activation, and caspase truncation of tau, are necessary or sufficient to cause tangle formation, neuronal distress or death. Aim 2 asks if NFT remain in viable cells and are long lived, or if they are toxic to the neurons in which they are found either in terms of structure or function. These experiments take advantage of the power of longitudinal in vivo multiphoton imaging to follow the fate of individual tangle bearing neurons for weeks to months. We will also test the hypothesis that tangle bearing neurons are excluded from participation in normal neural system activation from physiological stimuli by exposing animals to an enriched environment, then examining individual neurons in hippocampal subfields for immediate early gene (Arc) expression and the presence or absence of tangles. Aim 3 will rigorously test the hypothesis that wild type, nonmutant tau, undergoes similar phenomenon and that the toxic effects of tau over expression are due, in large part, to soluble tau. We take advantage of 3 models: examination of an alternative transgenic model, the hTau mice which express a minigene of wild type human tau (on a tau null background), introduction of wild type tau (or truncated forms of tau) into wild type mice using gene transfer approaches with AAV2 gene vectors, and the Tg4510 mice, which harbor a tet-response element driving the tau gene. These models allow comparison of wild type and mutant tau, and will also allow us to distinguish te effects of soluble tau from those of misfolded or aggregated tau. Taken together, our proposed studies will test the hypotheses that a non-tangle related mechanism of tau toxicity initiates apoptotic cascades, and that tangles may be a relatively nontoxic, long lived species. The results will have direct impact on design of therapeutic agents destined for clinical trials. PUBLIC HEALTH RELEVANCE: Neurofibrillary tangles are widely believed to be the cause of neuronal death in Alzheimer disease, and so the cause of the dementia symptoms that mark this common and devastating disease. The current application will examine the molecular and temporal relationships between neurofibrillary tangles, neuronal death and neuronal dysfunction using advanced microscopy methods and model systems. Based on these results, therapies aimed at reducing the damage done by neurofibrillary tangles will be explored, in an attempt to better understand what causes the damage and how to prevent it.

DESCRIPTION (provided by applicant): This R21 application proposes to develop and characterize a new model for Alzheimer disease (AD) research. In patients with AD, the earliest neuronal lesions occur primarily in the entorhinal cortex. Thereafter a hierarchical march of lesions involves limbic and association areas. Destruction of the perforant pathway, the entorhinal cortex projection to the dentate gyrus, is known to deafferent the hippocampus in AD. As the disease progresses, disruption of these pathways and closely related structures in the medial temporal lobe are believed to explain why memory impairments so dominate the clinical symptoms of AD. We have established a transgenic model in which P301L tau overexpression is limited to the same set of stellate neurons most vulnerable for tangle formation in AD, layer II of the entorhinal cortex, utilizing an established tet response promoter known to have specificity for these neurons and a tauP301L responder. This line, referred to as rTauP301L EC, expresses tau robustly in the EC. This leads to early (by 3 months of age) localization of misfolded tau in the perforant pathway and its terminals in the dentate gyrus terminal zone. Over the next 18 months, synaptic loss occurs, a robust neuroplasticity phenomenon with sprouting of AChE positive fibers into the deafferented zone occurs, tangles form, and (months later) neuronal loss ensues. Intriguingly, in animals ~18 months of age, neurons anatomically connected to the EC, but which do not express the transgene, develop tau positive inclusions, presumably from misfolded endogenous tau and/or from translocation across synapses of misfolded tau protein. A second line is proposed as well, utilizing a newly generated wild type tau responder line (rTg21221) that is entirely comparable to the rTauP301L EC line but lacks the mutation. We propose to characterize these models in terms of plasticity phenomenon, determine whether suppressing the transgene at various points can prevent ongoing degeneration, and determine whether the animals develop a behavioral phenotype of memory impairments in order to establish a model of the early changes of AD and of progressive pathology. We believe that it is critical to establish models of early disease processes to both understand the pathophysiology of progression and to generate a system in which to test neuroprotective therapies - a first step in designing interventions at a presymptomatic time point where they might be most effective from a clinical perspective. PUBLIC HEALTH RELEVANCE: Alzheimer disease starts with neurofibrillary lesions in a special brain area, the entorhinal cortex, which is responsible for memory related brain functions. We propose to make a model of this stage of the disease by genetically engineering a mouse to develop these same lesions in only this brain area; doing so will allow us to study the earliest phase of the disease, and to learn about whether early lesions lead to disease progression.

This project focuses on the hypothesis that one of the critical downstream consequences of Ab induced calcium elevation is activation ofthe only calcium sensitive phosphatase in neurons, calcineurin. Calcineurin is known to have multiple effects, ranging from alterations in cell surface trafficking of neurotransmitter receptors to activation ofthe transcription factor NFAT and initiation of transcriptional cascades. We have observed that neurons cultured from Tg2576 (APPSw) mice develop the same sort of neurodegenerative phenotype that occur near senile plaques in the adult Tg mouse or human Alzheimer brain - loss of dendritic spines, simplification of dendritic arborizations, and neuritic dystrophies. These neurons also show evidence of elevated Calcium and activation of calcineurin. Blockade of calcineurin prevents these neurodegenerative changes. Moreover, conditioned media from these neurons leads to the same neurodegenerative phenotype in wild type neurons. This can be blocked by either immunodepletion of Ab or blockade of calcineurin. Introduction of calcineurin inhibitors in vivo improves plaque-associated neuritic abnormalities in adult transgenic mice. Introduction of a constitutively active form of calcineurin, without Ab present, is sufficient to also lead to this same phenotype. Preliminary data suggest that activation of NFAT is critical for these phenomena, since the NFAT specific inhibitor VIVIT can also block neurodegenerative changes. Our project will extend these observations to determine the specific type of Ab that induces these changes, to examine the mechanism whereby calcineurin activation leads to neurodegeneration, and to come full circle - to critically test the hypothesis that the mechanism of Ab induced neurodegeneration is via activation of calcineurin. Together we will be in an outstanding position to fill in a critical gap in our understanding ofthe mechanisms of Ab induced neurodegeneration.

MADRC Administrative Core - Summary The MADRC Administrative core provides infrastructural support and integrative activities for the many activities of the MADRC Cores, projects, affiliated programs, and local and national initiatives in which we participate. We provide the overall framework for Center activities, direct its programs, and provide a sense of Centerness by facilitating smooth interactions among the Center's Cores and Projects. The Core acts as the interface of the MADRC both internally within our institutions and externally for collaborative projects. The Administrative Core helps organize resources and harmonizes Alzheimer research related activities within our multi-institutional environment, as well as with the Alzheimer Association and other lay and community groups interested in Alzheimer disease and dementias. Five aims encompass our mission: 1) First, leadership in the Core sets the tone and course of research within the Center. We ensure the integrity of our funded projects, and encourage and support new dementia related research initiatives by leveraging MADRC resources and infrastructure, attract new investigators to the field and support their efforts. 2) We recruit, identify and fund Pilot Projects targeted primarily at young investigators just establishing their research careers. We developed a successful collaboration with the Harvard Neurodiscovery Center's pilot program that allows us to increase the number of grantees; there is a common review system and a successful Pilot Awardee mentoring program. 3) The MADRC is an active participant in many national programs including NACC, ADCS, ADNI, DIAN, and ADGC studies, and the Administrative Core coordinates our involvement in these programs as well as providing leadership for many national and international Alzheimer related projects. Similarly, the Administrative Core is responsible for orchestrating interactions within the local environment, including the clinical operations in the MGH's Memory Disorder Unit and research programs in the Martinos Neuroimaging Center. 4) The Administrative Core is charged with establishing a formal External Scientific Advisory Board; we also have established an Internal Scientific Advisory Board enlisting some of the key institutional stakeholders and collaborators in Alzheimer research. 5) Finally, we provide administrative support with grants administration, financial responsibility, IRB interactions, organization of committees (including the Executive Committee of Core and Project leadership), assess needs, and balance distribution of resources aimed at achieving the Center's scientific goals: understanding the early phase of dementing illness as well as providing support for ongoing and collaborative studies in established dementia. Overall, the Administrative Core takes responsibility for ensuring Center responsiveness to new initiatives, and supports new research opportunities directed towards uncovering the etiology, pathogenic mechanisms, and ultimate cure or prevention of Alzheimer's and related disorders from the earliest phase.

PROJECT THREE SUMMARY/ABSTRACT The central theme of the MADRC is to examine the earliest features of the Alzheimer disease process. In keeping with this theme, the Center aims to understand dysfunction in neural systems prior to overt clinical symptoms - using novel clinical assays, advanced neuroimaging, and neuropathological studies focused on amyloid positive, cognitively intact individuals. Animal models would be an additional important approach to these early phases of disease where there is a gap in our knowledge; unfortunately, although animal models develop amyloid plaques and/or neurofibrillary tangles and reliably reproduce the molecular pathology of AD, they do not reproduce the unique patterns of anatomical changes that occur in early AD. Thus, no current animal models of AD provide a platform to study the anatomically restricted pathological changes that are known to occur in human patients. To address this problem, therefore, we have generated a transgenic mouse line (rTauEC) that over-expresses human mutant P301L tau primarily in the medial entorhinal cortex and develops tangles in those neurons in a pattern that is reminiscent of the early Braak II stage of human AD. We will examine the natural history of this model, examining the temporal relationship of tangles, synapse loss, and neuronal loss, to get at chicken-and-egg issues not possible to disambiguate in human autopsy tissue. We will use behavioral paradigms and molecular markers of neural system activation to test hypotheses about functional deafferentation of neural systems at early time points, before onset of behavioral abnormalities. Entorhinal neurons in rTauEC mice develop aberrant tau-filled axons and altered axonal projections, ultimately losing synaptic terminals in the dentate gyrus. This model also has the attribute of developing tau inclusions in the neurons that are the target of the entorhinal projection, in the dentate gyrus, despite not expressing human tau mRNA in those neurons. This has been interpreted as supporting the idea that there is a trans-synaptic propagation of pathological tau. We have crossed the rTauEC mice with APP/PS1 overexpressors to develop a model of tangles in entorhinal cortex and plaques throughout the cortex, analogous to the human pathology of many early cases of AD changes. Surprisingly, the addition of plaques seems to robustly accelerate the tangle propagation phenotype and also exacerbate the axonal dystrophies, developing more severe neuritic lesions in the hippocampus. Tau overexpression can be regulated with doxycycline in the rTau EC mice, mimicking some forms of anti-tau therapies. This model will therefore allow us to dissect a detailed time course of neural system degeneration, test hypotheses about tau-amyloid interactions in a defined neural system, and examine the consequences of reducing tau at various points in the disease process. Together these experiments will help provide insight into the pathobiology of the earliest phases of AD as well as highlight potential opportunities for therapeutic intervention early in the disease.

? DESCRIPTION (provided by applicant): Alzheimer's disease (AD) causes a progressive, devastating dementia that will affect over 6 million Americans this year. There is an urgent need to develop new therapies. A key element of the NIH AD Research Summit 2012 recommendations was to reexamine our approach to therapies to take into account emerging omics data and exploit network and systems based approaches (e.g. quantitative and systems pharmacology (QSP)) as an adjunct to traditional one-gene, one-receptor, one-mechanism based methods. We propose a planning group to design and begin to implement an approach to therapeutics that focuses on understanding patient heterogeneity (right patient), differences in disease pathophysiology at each stage of the disease (right time) and novel validated drug targets with cognate pharmacological probes by collecting and modeling systems levels perturbations in model systems (right drug). The planning group is designed to leverage emerging -omics data from various time points in the disease process in order to discriminate systems wide changes that occur as a consequence of genetic risk, of early degenerative changes, associated with classical neuropathological and neuroinflammatory changes, and ultimately due to neural system collapse. Both Alzheimer and non-Alzheimer domain expertise has been and will be recruited to design a pipeline based on profiling platforms first in model systems and then in humans to create actionable biomarkers for different endophenotypes, target engagement, and surrogate endpoints for clinical trials. In addition, the planning effort wil examine issues highlighted by RFA 14-017 as potential barriers to the rapid and effective development of therapeutics: we will broadly consider clinical trials input at each stage of the planning and strategic funneling of proposed targets; we will critically examine the institutional infrastructure required to break down non- scientific barriers to successful sharing of information and program expertise; we will organize and carry out workshops, seminars, and a series of face to face meetings to enhance collaborations and knowledge base of Alzheimer and non-Alzheimer domain experts in relevant overlapping topics of interest, with special emphasis on the iterative processes enabled by QSP approaches. This effort will include opportunities for mini- sabbatical exchanges of AD research scientists in QSP focused laboratories. An informatics superstructure will be designed to both enhance collating of data streams form a variety of -omics programs and also enhance communication and collaboration among the work groups envisioned. In additional to a fully-developed plan for a translational Center for Alzheimer's Therapeutic Science, concrete deliverables from the planning process include: (i) evaluation of informatics platforms for storage and distribution of AD data (ii) an opinion or review paper on the application of root cause (failure) analysis to clinical trials for AD and (iii a dozen or so webinars on the application of QSP approaches to AD from leading investigators across multiple disciplines.

Abstract The search for effective treatments for Alzheimer's disease (AD), the leading cause of late-onset dementia, has proven challenging. While recent successes in identifying more than a dozen new genes contributing to late-onset or sporadic AD (sAD) have generated considerable excitement in AD research, it is clear from large population studies including GWAS and whole-exome sequencing projects that many single nucleotide polymorphisms (SNPs) contributing to elevated sAD risk reside in non-coding intragenic or regulatory regions. The biological significance of these noncoding SNPs with respect to sAD pathogenesis is not clear. In the current application, we propose a scalable discovery platform for discerning which AD risk SNPs are associated with functional enhancers in specific neural cell types derived from human induced stem cells (hIPSCs). These hIPSCs, created from fibroblasts of sAD patients with a wealth of phenotypes that clearly lead to AD heterogeneity, will enable us to obtain a high-resolution map of AD risk SNPs associated with enhancers and their putative target genes in varied cell types. We will utilize CRISPR/Cas9/dCas9 technologies to directly determine the cell biological consequences of these AD risk genomic variants via 2D and 3D cytosystems. Our comprehensive strategy will identify novel genetic elements and unexpected regulatory pathways contributing to AD pathogenesis and progression that will lead to new therapeutic avenues.

The background of this proposal rests on three observations: First, that neurofibrillary tangles occur an anatomically related set of regions, leading to the possibility that tangle formation is linked to anatomical connectivity. Second is the discovery that a HMW form of hyperphosphorylated, soluble tau can be taken up by neurons, lead to tau aggregation, and propagate trans-synaptically. This form of tau can be isolated from Alzheimer brain and transgenic mice overexpressing mutant tau. Third is the observation from human neuropathology that tangles rarely spread from medial temporal lobe to the cortex without the presence of Abeta in the cortex, consistent with parallel observations in transgenic mice. We have recently found that the presence of Abeta also leads to several potentially related phenotypes: higher levels of tau propagation, higher levels of the unique, seed competent HMW tau, and increased neuroinflammation. These observations support the idea that there is synergy between Abeta and tau in the cortex, perhaps mediated by glia, and the core of this application is to understand this interaction. Aim 1 uses biochemical measures and a biosensor cells to ask how Abeta changes tau, in mice and in humans, and if microglia play a role. Aim 2 builds a nontransgenic model of tau uptake and aggregation, and evaluates changes in tau neurotoxicity due to age and Abeta. Aim 3 uses pharmacological agents to block soluble Abeta species (BACE inhibitor), clear fibrillar Abeta (anti-Abeta immunotherapy) and doxycycline transgene suppression to reduce tau. Already we find evidence of interesting interactions: Prolonged tau transgene suppression leads to clearance of the HMW species in Tg4510 mice, but the same HMW tau species remains remarkably stable in the APP/PS1xTg4510 cross. These data pave the way for combination therapy, because both Abeta and tau therapeutics may be necessary to target these long lived HMW tau species. Our model systems also allow us to test hypotheses that have direct clinical impact: we will test the concept that Abeta initiates tau misfolding, which can propagate without Abeta participation ? i.e. the hypothesis that there is an early ?amyloid dependent? and a later ?amyloid independent? phase of the disease. Indeed, at 3 and 6 months, APP/PS1 x Tg4510 have markedly increased HMW tau activity compared to Tg4510s, but 12-month old mice no longer show a pattern of enhanced HMW tau. If, as we propose, the spread of tangles starts out dependent on Abeta, but becomes increasingly less so as more misfolded tau accumulates, it follows that anti-Abeta therapeutics would be effective only during the ?amyloid dependent? phase, and help explain the multiple failures of clinical trials of anti-amyloid agents given later in the disease. Mapping the timing of these phases would have crucial import in informing the next generation of trials.

Neurofibrillary tangles are aggregates of the protein tau a normal neuronal cytoskeletal protein that becomes highly phosphorylated and ultimately forms fibrils within neuronal soma. As Alzheimer?s disease (AD) progresses the number of tangles increases and their location spreads throughout the brain. One idea for the spread is the propagation hypothesis which suggests that spread of pathology is due to the presence in the extracellular space of misfolded tau its uptake into other neurons and aggregation of naïve endogenous tau into the misfolded conformation which in a prion-like way leads to neurotoxic consequences in the downstream neurons. This hypothesis underlies current human clinical trials of anti-tau immunotherapy which are designed to block tau propagation. The current proposal explores and tests critical aspects of this model. The form of tau that might be taken up is uncertain ? our own previous work highlights a high molecular weight phosphorylated soluble species that is only ~1% of the soluble tau in an Alzheimer brain but which can be taken up by neurons in culture or in vivo and lead to aggregation. Other reports highlight other species including a sarkosyl preparation of sonicated fibrils that can also be taken up. We will study which if either of these types of post translationally modified tau are bioactive using cellular animal model and human neuropathological materials. The proposed studies will evaluate the exact biochemical characteristics of the tau species that is competent to be taken up by neurons (aim 1) the kinetics of aggregation once taken up (aim 2) and determine whether there are any marks of toxicity of propagated aggregated tau in recipient neurons both in animal models and in human neuropathological specimens (aim 3). Advanced microscopy methods including in vitro time lapse photography and image analysis to determine rates of uptake/aggregation in neurons and IPS cells in culture and longitudinal in vivo intravital multiphoton microscopy of intact living mice exposed to various tau preparations will determine the kinetics and efficiency of uptake and aggregation of the different preparations of tau derived from human brain; a set of morphological readouts evaluating subcellular aspects of tau toxicity will be employed in these experimental systems to assess whether uptake and aggregation of tau ?matters? ? ie what are the consequences of propagation? Those results will be compared to identically prepared samples from human Alzheimer cases testing the idea that uptake and aggregation of exogenous tau leads to the type of neuronal changes that actually occur in the human disease. The results of the proposed study will thus test the underlying assumptions of the ?propagation hypothesis? define the molecular characteristics of the propagating specie(s) and thereby inform interpretation of ongoing clinical trials and the design of future trials for AD

Complexity and Heterogeneity of Alzheimer Disease Patients with Alzheimer disease progress in a stereotypical fashion, starting with neurofibrillary tangles (NFTs) in the medial temporal lobes and mild memory deficits, and progressing to lesions throughout the cortex, and marked dementia. While the pattern of progression is frequently similar, the pace at which this occurs varies dramatically among patients, for unknown reasons. The spread of tau containing tangles through the cortex seems to be due, at least in part, to propagation of misfolded tau molecules across neural circuits, in a manner analogous to prion induced templated misfolding disorders. We hypothesized that patient specific differences in tau seeded templated misfolding properties might underlie differences in propagation and rate of tangle formation as well as rate of clinical progression. Similar to prion ?strains? this would imply a different biology and biochemistry to tau proteins in different individuals. In this patient based study, we find that tau seeding activity, assessed with a TauRD FRET biosensor, does indeed vary dramatically among patients with AD. More aggressive tau seeding correlates with both an increase in the number of NFTs and with the rate of clinical deterioration. The seed competent tau from different cases have very different epitope maps, suggesting different conformations. In accord with the idea that tau varies among individuals, tau from different AD cases shows differential susceptibility to protease digestion, and different properties on size exclusion chromatography. One implication of these observations is that, at least for some anti-tau reagents being developed for therapeutics, one size might not fit all. These preliminary data support the idea that different tau strains exist even within what would conventionally be viewed as sporadic Alzheimer disease patients. This application will directly test a series of hypotheses that follow from these observations. Aim 1 tests the hypothesis that different tau strains occur in the brains of different individuals with Alzheimer disease. Aim 2 will use newly developed super-resolution microscopy methods (STORM) to test the hypothesis that more aggressive seeding properties are linked to more aggressive neurodegenerative phenotype, especially in terms of synaptotoxicity. Aim 3 examines the possibility that tau seeding can be detected in CSF of living patients, opening the opportunity to examine tau strain properties in life, potentially impacting patient care, the design of clinical trials, and providing a ?personalized medicine? approach. Together this application tests a fundamental question about Alzheimer disease ? is the heterogeneity that distinguishes one Alzheimer patient from another simply clinical chance, or is it due to an underlying difference in tau pathobiology?

Massachusetts Alzheimer?s Disease Research Center: Administrative Core The Administrative Core of the Massachusetts Alzheimer?s Disease Research Center (MADRC) leads the Center?s efforts to contribute to our understanding of Alzheimer?s disease and Alzheimer?s disease related disorders (AD/ADRD). The Administrative Core sets the strategic priorities and outlines the tactics: We created the strategic plan for this application, with a focus on two closely related topics: heterogeneity of dementing illness and accelerating a path towards a cure. We coordinate efforts of the various components of the Center with each other, and coordinate planning for optimal utilization of resources within the Center and by leveraging institutional capabilities. We also act as an interface for the Center to optimize connections with the lay public, with institutional stake holders, and with national efforts. We take an active role in future planning, faculty recruitment and retention, and training of future leaders. The Administrative core coordinates interactions to develop trans-ADRC and other outside research projects. We oversee the research and grant administrative processes, including human subjects, animal welfare (if needed by a Developmental Project), scientific integrity, data and sample sharing, and financial policy requirements. We ensure timely and complete communication with NACC, timely and complete submission of samples to NCRAD, and continued participation in NIA summits and similar national forums. The Core is assisted and advised by an Executive Committee, composed of Core leaders and institutional leaders, and by a series of special purpose committees, including Developmental Projects, Clinical Studies Committee (which also oversees clinical trials interactions), Tissue and Biofluid distribution committee, Data Committee, ORE and the Community Advisory Board committee, and a REC committee. We are also advised by and assisted by an External Advisory Committee that will provide input on all aspects of Center activities. Importantly, we recruit, select, and monitor, and mentor the recipients of the exciting new Developmental Projects program, to which we are devoting maximal allowed resources. The Chair of the External Advisory Committee also chairs the selection committee; we hope to use this program to further enhance the growth of new science and new investigators, as another tool to build the future of our program. In sum, the Administrative Core provides leadership and support for both the broad efforts of the Center and support for the day-to-day activities that allow the Center have a maximal impact in our fight against AD/ADRD.

APOE genotype is the strongest genetic risk factor for late-onset Alzheimer?s disease (AD) with apoE4 increasing risk and apoE2 decreasing risk. In addition to apoE?s strong effects on A? clearance and aggregation, it has other effects that are important in how it may influence AD pathogenesis in an isoform- dependent fashion. We found that either lowering apoE or removing apoE/A? complexes in APP/PS1- 21;human apoE knockin (KI) mice produces beneficial effects in mouse models of A? deposition or tauopathy. Lowering apoE levels with anti-sense oligonucleotides (ASOs) early in life decreased A? deposition. When lowered after A? seeding had occurred, it did not lower A? deposition but it significantly decreased the amount of A?-induced neuritic dystrophy. In a mouse model of primary tauopathy (P301S Tau Tg mice), apoE4 markedly increased tau-mediated neurodegeneration, and both brain atrophy and the innate immune response were blocked in P301S mice lacking apoE. The data suggest that apoE is exacerbating the brain?s innate immune response in the setting of neural damage to worsen injury. Recently, we found that Trem2, an apoE receptor on microglia, strongly inhibits A?-induced neuritic tau seeding and spreading. To further understand whether targeting apoE should be in some way move forward therapeutically as well as further define whether its effects are Trem2-dependent, we propose to utilize models where 1) we have evidence that apoE is driving tau-mediated neurodegeneration and 2) a model in which A? pathology and tau seeding/spreading occur in a fashion that mimics types of A? and tau AD pathology seen in human AD. We hypothesize that apoE, in an isoform-dependent fashion, influences A?-mediated tau seeding/spreading and tau-mediated neurodegeneration in a TREM2-dependent fashion and that therapeutically reducing apoE levels in the adult CNS will attenuate amyloid-induced damage to neurons, amyloid-induced tau seeding/spreading, and tau-induced neurodegeneration. We propose these aims. Aim 1. To determine whether lowering apoE levels either before or after the onset of tau pathology with ASOs or via over-expression of the low density lipoprotein receptor (LDLR) decreases neurodegeneration, synaptic loss, the innate immune response, and improves function in P301S;apoE KI mice. We will also decrease human TREM2 with ASOs or activate it with agonizing TREM2 antibodies in P301S;apoE KI mice also expressing human TREM2. Aim 2: To determine the effects of targeting apoE by lowering apoE levels either before or after the onset of human AD-tau seeding with ASOs targeting apoE, by administering an antibody to non-lipidated apoE, or via over-expression of LDLR in APPNL-F ;E2, APPNL-F ;E3, and APPNL-F ;E4 mice. Aim 3: We will utilize a mouse neuronal/mixed glial culture system to assess the effects of apoE isoforms on tau-mediated neurodegeneration and determine 1) whether secreted apoE is responsible for the effects on tau-mediated neurodegeneration and 2) whether TREM2 is required for the effects of apoE and is either upstream or downstream of apoE.

Liquid liquid phase separation and tau biology Tau is a component of neurofibrillary tangles in Alzheimer disease and of analogous aggregates in closely related neurodegenerative diseases called tauopathies. What initiates a change from a very soluble, microtubule associated protein to an aggregate is unknown. The current application is based on a series of very recent experiments that may put a new twist on decades of observations of tau. In the first aim we further explore the new observation that tau (if phosphorylated, mutant, or exposed to polyanions like RNA) can undergo liquid liquid phase separation forming a coherent highly concentrated and reversible droplet. Intriguingly, liquid-liquid phase separation has currently been implicated in the pathobiology of Fus, TDP43, hnRNP, and C9orf72 proteins in ALS; with our new data on tau and Alzheimer's and frontotemporal dementia, the analogy that liquid liquid phase transitions underlie some aspects of neurodegeneration is compelling to explore. The structure of these droplets is fascinating, and over time they undergo a gelation and finally remodel to have some beta pleated sheet structure, potentially on the pathway towards the highly ordered structure of neurofibrillary tangles. The second aim examines the function of these droplets which, when ?fresh?, have the ability to interact with tubulin and to nucleate microtubule structures, ?wetting? the surface of the microtubule bundle and then flowing along it, stabilizing the extending structure. Given the long history of studies showing conclusively that tau stabilizes microtubules, it seems inescapable that droplet formation is linked to this fundamental physiological role of tau. The third aim is to bring these observations full circle, from in vitro observations with recombinant proteins to the tau that is present in the brain ? both in animal models of tau induced neurodegeneration and in human Alzheimer, Frontotemporal dementia (tau positive or TDP43 positive), and control brain. Our preliminary data show that soluble tau isolated from Alzheimer brain ? using biochemical separation techniques to enrich for high molecular weight hyperphosphorylated oligomers and an immunoaffinity column, retains the ability to form LLPS. Our proposed studies will further examine this fraction of tau, compare its properties to other fractions, and (together with aims 1 and 2) allow us to explore how reduced systems studying tau relate to the tau present in human disease. We suggest that the newly described liquid liquid phase separation properties of tau are an important clue that connects the structure and function of tau ? and phosphorylated and mutant forms - with the disease relevant initiation of aggregation.

PROJECT SUMMARY/ABSTRACT Alzheimer's disease (AD) is the most common form of dementia in the elderly, affecting more than 5 million Americans, as well as their families and caregivers. Unfortunately, aging of the global population is only worsening the AD ?epidemic?, as incidence is projected to triple by 2050. Despite intense research, there is currently no cure for this devastating neurodegenerative disorder. Thus, understanding the intrinsic molecular mechanisms that drive AD pathology and progression is critical to devising effective treatments. Postmortem examination of human brains has revealed that AD-associated neuropathologies, such as neurofibrillary tangles (NFTs) and neurodegeneration, generally arise in a conserved spatio-temporal pattern, affecting transentorhinal regions first, and later extending to limbic and isocortical areas. The molecular and neurochemical bases for such selective neuronal vulnerability (SNV) have long been pursued, as they underlie disease progression and may hold the key to understanding the molecular underpinnings of neurodegeneration, but to date these mechanisms remain elusive. Here, cutting-edge, single-cell technologies will be used to generate a comprehensive, multi-omic atlas of cell types within AD-vulnerable brain regions across different stages of disease. The hypothesis is that specific cell types most dramatically affected by AD pathology within susceptible brain regions are characterized by distinct molecular pathways (transcription factors, signaling cascades, gene networks) that drive SNV. Moreover, that these pathways are executed in a sequential spatio-temporal pattern by changes in chromatin architecture and gene regulatory elements. Tracking the molecular changes exhibited by these neuronal cell populations in the continuum of AD pathology will better define AD onset and progression, and potentially indicate new therapeutic targets. Postmortem brain samples will be obtained from healthy controls, or patients who at death exhibited different stages of AD pathology, namely early (Braak III/IV), or late (Braak V/VI). Changes occurring at different stages of AD progression will be analyzed to identify the cell types and molecular pathways most critical for the initiation and spread of AD-related pathology. Regions analyzed will be hippocampus (CA1/sub), inferior temporal cortex (BA20), frontal cortex (BA9), and visual cortex (an AD- resistant region). In Aim 1, samples from control subject will be subjected to single cell analyses to characterize the methylome and chromatin architecture jointly (sn-m3C-seq), as well as chromatin accessibility together with transcriptome (Paired-seq), of cell types within AD-vulnerable brain regions. Integration of these datasets will create a multi-omic atlas of relevant cell types that will serve as the foundation for understanding AD onset and progression. In Aim 2, these analyses will be extended to AD patients. Comparing data sets across brain regions and disease stages will reveal the specific cell types most affected in AD, as well as the molecular pathways (based on changes in methylation patterns, chromatin architecture, etc.) that drive SNV in AD (Aim3).

Neurofibrillary tangles (NFTs) are one of the characteristic features of Alzheimer disease (AD) neuropathology. They are made primarily of the microtubule associated protein tau (MAPT) that is highly phosphorylated, mislocalized to the cytoplasm from the axon, and aggregated in a complex, dense ??pleated sheet that are paired helical filaments (PHFs) as determined by cryo-electron microscopy. The distribution of NFTs in the brain is overwhelmingly consistent across cases of AD: NFT occur initially in the entorhinal cortex, then ?spread? to other limbic and association areas over more than a decade; this spread corresponds to the clinical symptoms of the disease, and correlates with neuronal loss. It was recognized early on that the pattern of spread largely followed neuroanatomical connections, and it was demonstrated that at least part of the reason for this could be explained by propagation of misfolded tau across synaptic elements. Recently, it has been discovered that the LDL receptor-related protein binds tau and participates in tau propagation. The Hyman and Strickland laboratories have worked together on LRP1 related projects since 1993 and have collaborated to confirm these observations. Using fractions isolated from AD patient brains, we confirm that LRP1-expressing cells, but not LRP1-deficient cells, promote tau seeding, demonstrating that LRP1 mediated uptake can lead to escape of tau proteopathic seeds into the cytoplasm. The mechanism(s) of how this occurs are currently not known and will be investigated in Aims 1 and 3 of this grant. We also identified some residual uptake that we now show to be due, in part, to SORL1, another apoE receptor that is implicated in trafficking, and ? importantly- is also clearly implicated in the genetics of AD. The role of SORL1 in tau uptake and processing will be examined in Aims 2 and 3). These data and new questions lead us to propose a multi-PI application to explore the following aims: (1) Identify mechanisms by which LRP1 promotes proteopathic seeding of tau; (2) Define the contribution of SORL1 and SORL1 mutants to tau proteopathic seeding; (3) Identify mechanisms responsible for the endolysosomal escape and tau seeding