Nikolaos K. Robakis - US grants

proteins; gene expression;

Alzheimer disease (AD) is marked by a progressive degenerative neuropathology in the cerebral cortex and hippocampus. The primary morphological signs of the disease are the presence of intracellular neurofibrillary tangles as well as neuritic plaques and cerebrovascular amyloidosis. The neuritic plaque is a complex structure composed of a central amyloid core surrounded by a cluster of dystrophic neurites and glia. The major component of the neurite plaque is a 4.2 kDa peptide termed the amyloid beta-protein (AbetaP) or A4 peptide. AbetaP exists as a component of at least three distinct precursor proteins, referred to as APP695, APP751, APP770. The pathological process leading to the production of the amyloid protein is not known, although it seems that aberrant degradation of APP is involved. Research in our laboratory has shown that the full length Amyloid precursor (APP) protein is associated with the cytoskeletal component of the cell in both glial and neuronal cell cultures and in brain tissue. This association is modulated by at least two factors, e.g. cell density and protein kinase C, and it requires intact microtubules. In addition, the human and rat brains contain truncated APP forms associated with the cytoskeleton, which derive from the C-terminal portion of the APP protein and encompass all of the AbetaP sequence. The presence of these forms in brain tissue suggests that in addition to the non-amyloidogenic cleavage pathway which produces the secreted APP forms, other potentially amyloidogenic cleavage pathways exist which produce APP fragments containing the intact AbetaP sequence. These observations have implications for both the physiological function of APP and the production of the AbetaP from its precursor. Here we propose to further characterize the factors mediating the cytoskeletal association of all APP forms, and to examine any differences in the cytoskeletal association of APP between normal and AD brain.

The major component of the neuritic plaque cores and cerebrovascular amyloid of the Alzheimer's disease (AD) exists as a component of at least three distinct precursor proteins, referred to as APP695, APP751, APP770. APP is normally metabolized by at least two pathways: one pathway involves cleavage within the AbetaP sequence, thus preventing formation of the amyloid peptide. The second "amyloidogenic" pathway results in the production of the amyloid peptide. Recent genetic studies showed that certain mutations in the APP gene that result in amino acid substitutions may cause Familial AD. However, it is still not clear whether these mutations cause AD by increasing production of the amyloid peptide or by altering the biological function of the APP protein. Since the structural elements of a biomolecule define both its biological function and its metabolic pathway, determination of the structural characteristics of the APP is important for the elucidation of its role in the development of AD. Research in our laboratory showed that APP exists as the core protein of a chondroitin sulfate proteoglycan (CSPG), ranging in apparent molecular size from 140 to 250 kDa, secreted by a glial cell line. Most of the secreted nexin II form oa APP occurs in the proteoglycan form. We also obtained evidence that the APP proteoglycan is present in human brain and neuroblastoma cells. Proteoglycans are molecules with important biological functions including cell adhesion and migration, cell-cell communication, modulation of growth factor activities and neural patterning. Dysfunction of the CSPG form of APP may contribute to the neuronal degeneration observed in AD. We find two potential chondroitin sulfate glycosaminoglycan (CSG) attachment sites in close proximity to both the N- terminus of the AbetaP sequence of APP and the secretase cleavage site, suggesting that the CSG chains may affect the proteolysis of APP and production of AbetaP. Here we propose to determine the attachment sites and length of the CSG chains, the structure of the carbohydrate residues attached to APP and the role of the APP CSPG in cell adhesion. In addition we will examine the expression of this novel APP form in normal and AD brains and its developmental regulation in rat brain.

Alzheimer disease (AD) is caused by heterogeneous genetic and probably environmental factors. Although the etiology of the disease is still not clear, several lines of evidence indicate that the integrity and number of cellular organelles that sustain neuronal vesicular axoplasmic transport, including endoplasmic reticulum (ER), Golgi, endosomes, and large dense core vesicles (LCDCVs) are compromised in AD. These observation suggest those factors that compromise that comprise neuronal vesicular transport may also be causally involved in the development of AD. This hypothesis, while does not disregard the pathological significance and consequence of neurofibrillary tangles (NFTs) and neuritic plaques (NPs), it emphasizes the need to search for abnormalities in neuronal protein transport upstream of the formation of NFTs and NPs. The association of the Alzheimer's amyloid precursor protein (APP) with the cytoskeleton and its axoplasmic transport raise the possibility that familial AD (FAD)-linked APP mutations alter the function and vesicular transport of APP. Presenilin l (PSl) is an integral membrane protein of unknown function. It is cleaved post-translationally to yield an N-terminal fragment and a C-terminal fragment. Many PSl mutants have been linked to the development of FAD. We obtained preliminary data that PSl proteolytic fragments are expressed in LDCVs, chromaffin granules (CGs), and somatodendritic clathrin coated vesicles (SDCCVs) suggesting that this protein play a role in vesicular function. This observation raises the possibility that the FAD PSl mutations may interfere with the function of these vesicles. The purpose of this proposal is to further examine the vesicular localization of PSl and its proteolytic fragments, to test the hypothesis that PSl has a vesicular function and to examine the effects of FAD-linked PSl mutations on vesicular transport. The results of our research should further out understanding of the mechanisms involved in the neuropathology of AD.

Presenilin1 (PS1) is an integral membrane protein involved in the development of familial Alzheimer disease (FAD). PS1 is processed intracellularly into N and C-terminal fragments. Classic cadherins are type I transmembrane glycoproteins that control critical events in cell-cell adhesion and recognition. Cadherin-based cell adhesion is mediated by catenins, which are cytosolic proteins that link surface cadherin to the cortical cytoskeleton. We used confocal microscopy to show that in cell cultures PS1 accumulates at cell-cell contact sites. At these sites, PS1 colocalizes with components of the cadherin-based adherens junctions and is linked to the cortical cytoskeleton. Both PS1 fragments form complexes with components of the cadherin-based adhesion system including E-cadherin, beta- catenin, gamma-catenin, and alpha-catenin. Furthermore, PS1 fragments are found in a single complex with E-cadherin and beta- catenin. PS1 forms complexes with cell surface E-cadherin, and in epithelial tissue it concentrates at cell-cell contact sites. Together, these data show that PS1 incorporates into the cadherin/catenin adhesion system where it probably functions in cell-cell adhesion and recognition. In the brain, PS1 fragments form complexes with cadherins which are known components of the synapse. Thus, PS1 mutations may act in FAD through disruption of cadherin-based cell-cell interactions. Based on these findings, we propose to define the bindings between PS1 and other components of the cadherin-based adhesion system and to explore how PS1 FAD mutants affect these bindings. We will also examine whether PS1 functions in cadherin-based cell-cell adhesion, and define the effects of FAD mutations on PS1 function. Finally, we will examine whether PS1 is part of the cadherin/catenin apparatus at the synapse.

Presenilin1 (PS1), is a polytopic integral membrane protein that plays important roles in the development of familial Alzheimer's disease (FAD). Classic cadherins are type I transmembrane glycoproteins that control critical events in cell adhesion, recognition, and contact-mediated inhibition of cell growth. Recently we reported that PS1 concentrates at cell-cell and synaptic contact sites where it incorporates into the cadherin/catenin cell-cell adhesion system. Here we present preliminary data that PS1 binds directly to E-cadherin and regulates Ca++-dependent cell-cell adhesion. In addition, our results indicate that PS1 and a gamma-secretase activity control production of a cadherin cytoplasmic peptide of about 20kDa (termed pC20) and regulate expression of both cyclin D1 and cyclin-dependent kinase inhibitor p27Kip1, two important cell cycle proteins. pC20 seems to be derived from a larger cadherin precursor termed pC35. pC35 is produced following an extracytoplasmic cleavage of the full length cadherin. Expression of a cadherin cytoplasmic peptide in PS1-deficient cells mimicked the pC20 effects and reduced the expression levels of both cyclin D1 and p27Kip1. Taken together, our preliminary data suggest that PS1, cadherins, and gamma-secretase are involved in a novel signal transduction mechanism that regulates expression of cell cycle proteins. This mechanism may use the cytoplasmic cadherin-derived peptide pC20 as a messenger which translocates to the nucleus where it controls expression of cell cycle proteins. PS1 and gamma-secretase activity may thus control the cadherin signal transduction pathway by regulating expression of the cadherin-derived message. This application proposes to characterize the cadherin-PS1-gamma-secretase signal transduction pathway and to examine its involvement in Alzheimer's disease.

[unreadable] DESCRIPTION (provided by applicant): Increased cell death is a hallmark of neurodegenerative diseases including Alzheimer's disease (AD). The PI3K/Akt signaling pathway plays critical roles in cell survival. Its targets include components of the cell death machinery, like the BCL-2 and FOXO families, transcription factors important for cell survival, and kinases like GSK3_ involved in the generation of neurofibrillary tangles (NFT) of Alzheimer's disease. Presenilin 1 (PS1) is a transmembrane protein involved in familial Alzheimer's disease (FAD). Classic cadherins, including epithelial (E)- and neural (N)-cadherins, are major cell-cell adhesion receptors involved in the development, maintenance and function of almost all solid tissues. PS1 binds cadherins and regulates their function and processing. Using PSI+/+ and PS1 knockout (PS1-/-) fibroblasts we noticed that absence of PS1 correlates with apoptotic cell death and decreased activity of the PI3K/Akt cell survival pathway. Re-introduction of PS1 in PSI-/- cells activates the PI3K/Akt pathway and rescues cells from apoptosis suggesting that PS1 mediates transmission of survival signals. PSI-induced cell survival requires PI3K activity. These data indicate that PS1 activates the PI3K/Akt pathway. Cadherin adhesion stimulates the PI3K/Akt pathway by promoting cadherin association with the p85 subunit of PI3K. Our data show that PS1 stabilizes the cadherin/p85 association suggesting a mechanism for the PS1 cell survival effects. Furthermore, we obtained evidence that PS1 is important for insulin growth factor (IGF)-induced stimulation of the PI3K/Akt pathway, suggesting that PS1 may be involved in tyrosine kinase receptor signaling. Several PS1 FAD mutants showed a decreased ability to activate Akt or to phosphorylate GSK3[3 kinase. Here we propose to investigate the cell survival function of PS1, the mechanisms involved in the PSI-mediated activation of the PI3K/Akt pathway and the effects of PS1 FAD mutations on the activation of the PI3K/Akt pathway and on cell survival. [unreadable] [unreadable]

Glutamate-induced excitotoxicity and oxidative stress are mechanisms implicated in the pathogenesis of Alzheimer's disease (AD). We found that ephrinB ligands (ephrinBLs) protect primary neuronal cultures from glutamate- and oxidative stress-induced toxicity and that the neuroprotective activities of ephrinBLs depend on EphB receptors (EphBRs) and presenilini (PS1), a protein involved in familial AD (FAD). We also observed that PS1 mediates the neuroprotective effects of brain-derived neurotrophic factor (BDNF). Interestingly, the neuroprotective activities of ephrinBLs and BDNF depend on both alleles of PS1 because absence of either one or both alleles results in the inhibition of trophic factor-induced neuroprotection. Furthermore, the neuroptotective activities of ephrinBLs and BDNF are significantly reduced in neurons either heterozygous or homozygous for FAD mutants of PS1. In contrast, the neuroprotective function of basic fibroblast growth factor (bFGF) is affected neither by WT nor by mutant PSl Our data suggest a novel mechanism of neurodegeneration in FAD where brain neurons expressing FAD mutants of PS1 may have reduced ability to respond to neuroprotective factors and therefore are more vulnerable to toxic insults then wild type neurons. This decreased neuroprotection may result in increased rates of neuronal loss and over many years, in severe depletion of neuronal populations. Here we propose tp investigate further the effects of PS1 and its FAD mutants oh the neuroprotective function of trophic factors in vitro using primary neuronal cultures and in vivo using PS1 knockout and FAD knock-in mouse models. We will use the same technology we employed for PS1 to explore the neuroprotective activities of presenilin 2 (PS2) and the amyloid precursor protein (APP). In addition, we will explore mechanisms involved in the neuroprotective function of PS1 including the role of MP/y-secretase receptor processing and endocytosis in the neuroprotecive activities of PS1, BDNF and ephrinBLs and the effects of PS1 and FAD mutants on ligand-induced receptor phosphorylation and downstream signaling. RELEVANCE (See instructions): Excitotoxicity and oxidative stress are implicated in neurodegeneration and Alzheimer disease (AD). We found that PSl, a protein involved in familial AD, mediates the neuroprotective functions of brain factors like ephrinB and BDNF. These factors are unable to protect neurons in the absence of PS1. Interestingly, genetic changes that promote AD interfere with the neuroprotective function of PS1. Our findings have implications for the mechanism of neurodegeneration and mav provide novel targets for therapeutic intervention in AD.

[unreadable] DESCRIPTION (provided by applicant): Increased cell death is a hallmark of neurodegenerative diseases including Alzheimer's disease (AD). The PI3K/Akt signaling pathway plays critical roles in cell survival. Its targets include components of the cell death machinery, like the BCL-2 and FOXO families, transcription factors important for cell survival, and kinases like GSK3_ involved in the generation of neurofibrillary tangles (NFT) of Alzheimer's disease. Presenilin 1 (PS1) is a transmembrane protein involved in familial Alzheimer's disease (FAD). Classic cadherins, including epithelial (E)- and neural (N)-cadherins, are major cell-cell adhesion receptors involved in the development, maintenance and function of almost all solid tissues. PS1 binds cadherins and regulates their function and processing. Using PSI+/+ and PS1 knockout (PS1-/-) fibroblasts we noticed that absence of PS1 correlates with apoptotic cell death and decreased activity of the PI3K/Akt cell survival pathway. Re-introduction of PS1 in PSI-/- cells activates the PI3K/Akt pathway and rescues cells from apoptosis suggesting that PS1 mediates transmission of survival signals. PSI-induced cell survival requires PI3K activity. These data indicate that PS1 activates the PI3K/Akt pathway. Cadherin adhesion stimulates the PI3K/Akt pathway by promoting cadherin association with the p85 subunit of PI3K. Our data show that PS1 stabilizes the cadherin/p85 association suggesting a mechanism for the PS1 cell survival effects. Furthermore, we obtained evidence that PS1 is important for insulin growth factor (IGF)-induced stimulation of the PI3K/Akt pathway, suggesting that PS1 may be involved in tyrosine kinase receptor signaling. Several PS1 FAD mutants showed a decreased ability to activate Akt or to phosphorylate GSK3[3 kinase. Here we propose to investigate the cell survival function of PS1, the mechanisms involved in the PSI-mediated activation of the PI3K/Akt pathway and the effects of PS1 FAD mutations on the activation of the PI3K/Akt pathway and on cell survival. [unreadable] [unreadable]

DESCRIPTION (provided by applicant): Presenilin 1 (PS1) plays important roles in development and in familial Alzheimer's disease (FAD). Cadherins are cell-cell adhesion receptors that control critical events in development and cell survival. The PI3K/Akt cell survival signaling is activated by growth factor receptors and promotes survival in almost all cell types including neurons, by downregulating the activity of pro-apoptotic factors and upregulating survival factors. Recent evidence however, shows that this pathway is also activated in response to cadherin-dependent cell adhesion and we showed that PS1 binds cadherins and promotes survival of confluent fibroblast cells by activating the cadherin/PI3K/Akt signaling. Confluent PS1 null (PS1-/-) fibroblasts display increased rates of apoptotic cell death, a phenotype rescued by exogenous PS1. PS1 promotes activation of PI3K/Akt signaling by stimulating the association of PI3K with cadherins. By activating this pathway, PS1 promotes phosphorylation/activation of Akt kinase while it downregulates the activities of both GSK-3 kinase and apoptotic caspase-3. This function of PS1 is independent of its role in gamma-secretase activity. Here we present data that survival of mature primary neuronal cultures depends on the ability of PS1 to activate the cadherin/PI3K/Akt signaling. Absence of PS1 activity at a developmental stage marked by formation of extensive neuronal contacts in vitro results in the inhibition of Akt and in the activation of both GSK-3 and apoptosis. Here we propose to examine the role of the PS1/cadherin/PI3K/Akt signaling in neuronal development and apoptosis and whether PS1 FAD mutants promote neuronal apoptosis and tau phosphorylation by interfering with this signaling. Since neuronal apoptosis and GSK-3 have been shown to increase Abeta production, we will examine the effects of the cadherin/PI3K/Akt signaling on Abeta. Finally, we will identify adaptor molecules promoting the PS1/cadherin/PI3K complex.

[unreadable] DESCRIPTION (provided by applicant): Increased cell death is a hallmark of neurodegenerative diseases including Alzheimer's disease (AD). The PI3K/Akt signaling pathway plays critical roles in cell survival. Its targets include components of the cell death machinery, like the BCL-2 and FOXO families, transcription factors important for cell survival, and kinases like GSK3_ involved in the generation of neurofibrillary tangles (NFT) of Alzheimer's disease. Presenilin 1 (PS1) is a transmembrane protein involved in familial Alzheimer's disease (FAD). Classic cadherins, including epithelial (E)- and neural (N)-cadherins, are major cell-cell adhesion receptors involved in the development, maintenance and function of almost all solid tissues. PS1 binds cadherins and regulates their function and processing. Using PSI+/+ and PS1 knockout (PS1-/-) fibroblasts we noticed that absence of PS1 correlates with apoptotic cell death and decreased activity of the PI3K/Akt cell survival pathway. Re-introduction of PS1 in PSI-/- cells activates the PI3K/Akt pathway and rescues cells from apoptosis suggesting that PS1 mediates transmission of survival signals. PSI-induced cell survival requires PI3K activity. These data indicate that PS1 activates the PI3K/Akt pathway. Cadherin adhesion stimulates the PI3K/Akt pathway by promoting cadherin association with the p85 subunit of PI3K. Our data show that PS1 stabilizes the cadherin/p85 association suggesting a mechanism for the PS1 cell survival effects. Furthermore, we obtained evidence that PS1 is important for insulin growth factor (IGF)-induced stimulation of the PI3K/Akt pathway, suggesting that PS1 may be involved in tyrosine kinase receptor signaling. Several PS1 FAD mutants showed a decreased ability to activate Akt or to phosphorylate GSK3[3 kinase. Here we propose to investigate the cell survival function of PS1, the mechanisms involved in the PSI-mediated activation of the PI3K/Akt pathway and the effects of PS1 FAD mutations on the activation of the PI3K/Akt pathway and on cell survival. [unreadable] [unreadable]

DESCRIPTION (provided by applicant): Excitotoxicity, a form of neuronal damage due to excessive activation of glutamate receptors and oxidative stress are mechanisms of neuronal injury implicated in the pathogenesis of Alzheimer's disease (AD). We found that the ephrinB family of ligand proteins protects primary neuronal cultures from both glutamate- and oxidative stress- induced death. The neuroprotective activities of ephrinB ligands (ephrinBLs) are mediated by their receptors (EphB receptors, EphBRs) and depend on PS1, a protein involved in familial AD (FAD). Interestingly, the neuroprotective effect of ephrinB depends on both PS1 alleles because absence of one allele (haploinsufficiency) results in severe reduction of the ephrinB neuroprotection. Furthermore, we obtained preliminary data that FAD mutants of PS1 interfere with the ephrinB-dependent neuroprotection and that -secretase activity may be involved in the neuroprotective functions of the ephrinBL/EphBR system. Here we propose to further investigate the effects of PS1 FAD mutants and ?-secretase on the neuroprotective function of ephrinBLs and to elucidate molecular mechanisms by which PS1 mediates this function. We will explore whether PS1 regulates the binding of ephrinBLs to EphBRs and the ephrinBL-induced phosphorylation of both EphB and NMDA receptors. To this end we will use cortical primary neuronal cultures from our PS1 knockout and FAD mutant knock-in transgenic mouse colonies as well as EphBR knock-out colonies available in our laboratory. We will also use our mouse models to examine the neuroprotective function of the ephriBL/EphBR system in vivo and to ask whether PS1 FAD mutations affect this function. Finally, we will ask whether PS2, a homologue of PS1 also involved in FAD, may also be involved in the ephrinB neuroprotection. PUBLIC HEALTH RELEVANCE: Excitotoxicity and oxidative stress are neurotoxic mechanisms implicated in the neurodegeneration of Alzheimer's disease (AD). We found that a protein called ephrinB, has neuroprotective functions and rescues neurons from excitotoxic and oxidative death;these neuroprotective effects of ephrinB depend on PS1, a protein involved in AD. Our findings have implications for the mechanisms by which genetic mutations cause neurodegeneration and may provide novel targets for therapeutic intervention in AD.

Project Summary Recerch in the last decade indicates that reduced cerebral glucose utilization is an early sign of AD that correlates with clinical progression of the disease. In vivo imaging studies indicate severe reduction of glucose metabolism in regions of the AD brain affected at early stages of the disease such as hippocampus, entorhinal formation and temporal cortex. Reduction in brain glucose metabolism is also observed in familial AD (FAD) patients and may occur years before disease onset in carriers of presenilin-1 (PS1) FAD mutations implicating PS1 in the regulation of brain metabolic activity. Glucose deprivation leads to neuronal death, a process associated with expression changes of genes involved in neuronal survival. In recent years a new mechanism of transcriptome regulation via the action of microRNAs (miRs) has emerged. MiRs are short, non-coding RNAs that suppress gene expression by post-transcriptional mechanisms including destabilization and inhibition of translation of mRNAs. By regulating gene expression, miRs regulate physiological processes including neuronal survival, synaptic plasticity and memory and emerging evidence supports a role for miRs in neurodegenerative disorders. We obtained evidence that the PS1/¿-secretase system regulates the expression of miR-212 that targets the mRNA of PED/PEA-15 (PEA15), a protein known to promote cell survival under reduced glucose. We also observed that absence of PS1 sensitizes neurons to glucose deprivation and correlates with increased expression of neuronal miR-212 and decreased levels of both PEA15 mRNA and protein. Furthermore, we obtained evidence that PS/¿-secretase regulate neuronal survival by controlling the expression of miR-212. Based on our data, we hypothesize that the increased expression of miR-212 and the resultant decrease in PEA15 contribute to the increased vulnerability displayed by PS1 null neurons under reduced glucose. Importantly, in this submission, we present new data that downregulation of neuronal PS1 with siRNA technology increases miR-212 and decreases PEA15 indicating PS1 controls a cascade that regulates production of cell survival factor PEA15. Here we examine the relationship between PS1, miR-212 and its target protein PEA15 and test their effects on neuronal survival under stress conditions. In addition, we ask how the ¿-secretase function affects miR-212 and PEA15 and whether PS1/¿-secretase modulate the glucose deprivation-induced survival signaling of PEA15. We also ask whether expression of miR-212 and its target PEA15 protein changes in brains of AD patients and in FAD mutant knockin animal models. Our data may indicate whether PS1 promotes neuronal survival by controlling the PS1/miR- 212/PEA15 cascade and whether antagonists of miR-212 rescue neurons from stress-induced death.

? DESCRIPTION (provided by applicant): Evidence in the last decade implicates cerebral microvasculature abnormalities in the genesis of AD neuropathology. Additional literature shows that the EphB4/ephrinB2 system is an important regulator of the development and function of the vascular system. Binding of the extracellular sequence of EphB4 receptor to its transmembrane ligand ephrinB2 protein on the surface of endothelial cells of blood vessels, stimulates angiogenesis and growth of new vessels from existing vasculature. Furthermore, in vitro assays show that treatment of ephrinB2expressing endothelial cells with the extracellular domain of EphB4 stimulates cell sprouting and tube formation, processes considered crucial initial steps in angiogenesis, while transgenic mouse experiments indicate that the intracellular (cytoplasmic) domain of ephrinB2 protein is necessary for ephrinB2dependent angiogenesis. We found that EphB4 stimulates the metalloproteinase (MP) and PS/?secretase processing of ephrinB2 producing cytosolic peptide ephrinB2/CTF2 and that the EphB4induced sprouting and tube formation of endothelial cells depends on ?secretase activity. These observations raise the possibility that the endothelial EphB4/ephrinB2 system regulates angiogenesis through PS/?secretase. In support of this hypothesis, we obtained data that peptide ephrinB2/CTF2 stimulates sprouting of endothelial cells in vitro. Recent literature shows that a crucial step in angiogenic factorinduced angiogenesis is formation of complexes between Raf1/ Rok? and Vascular Endothelial cadherin (VEcadherin) and we made the novel observation (preliminary data) that treatment of endothelial cell cultures with EphB4 increases these angiogenic complexes. Together, our observations suggest that PS1/?secretase may affect angiogenesis by regulating processing of transmembrane protein ephrinB2, a critical step in EphB4induced angiogenesis. Here we propose to explore the mechanisms via which the EphB4/ephrinB2 and PS1/?secretase systems interact to promote endothelial cell sprouting and angiogenesis and to examine whether any of these mechanisms are altered in Alzheimer disease (AD) brains. Furthermore, we and others reported that PS1 familial AD (FAD) mutations may affect the epsilon (?) cleavage of PS1/?secretase substrates thus decreasing production of CTF2 peptides including ephrinB2/CTF2 (see Significance). Thus, we will ask whether PS1 FAD mutants alter the EphB4/ephrinB2dependent angiogenesis.

Mechanisms of neurodegeneration in Alzheimer disease (AD) are imperfectly understood contributing to the lack of efficient therapeutic methods. Mutants of protein Presenilin1 (PS1) cause autosomal dominant neurodegeneration in Familial AD (FAD). Thus these mutants constitute a system where mechanisms involved in AD-like neurodegeneration can be studied. Since all AD forms share clinical and neuropathological phenotypes, data derived from FAD may also illuminate pathways involved in sporadic AD (SAD). Neurotrophins and ephrins are two classes of receptor ligands that function in neuronal survival and synaptic activity and literature shows that ephrinB (efnB) and BDNF stimulate interactions of their respective receptors with the NMDAR. Recently we reported that the ability of efnB1 and BDNF (called here Factors) to protect neurons from toxicity depends on PS1 and here present data that PS1 also regulates the Factor-stimulated complexes of their cognate receptors with the NR1 subunit of the NMDAR. We also found that these complexes play crucial roles in neuroprotection. In contrast to WT neurons, neurons expressing FAD mutants are not protected by Factors from toxicity and FAD mutants inhibit the Factor-stimulated association of their receptors with NMDAR. Surprisingly, PS1FAD mutant-expressing neurons contain increased amounts of PS1- NMDAR complexes. Furthermore, PS1FAD mutant-containing NR1 complexes (necrocomplexes) do not change in response to Factor treatment. Together, our data suggest that the structure of the PS1-NR1 association of WT neurons differs from that of FAD mutant expressing neurons. Here we will elucidate mechanisms by which PS1 FAD mutants block neuroprotective activities of brain Factors by changing the structure of the PS1-NMDAR interaction. Here we also present data that the excitotoxicity-induced neuronal death in vitro may be necroptotic. We also found that PS1 FAD mutants sensitize mouse brain neurons to MCAO (ischemia)-induced death in vivo. Since MCAO-induced neuronal death involves excitotoxicity and is mediated by necroptosis, we will ask whether ischemia-induced neuronal death in FAD mutant-expressing brains correlates with necroptosis. Necroptosis was recently found activated in postmortem brains of SAD patients. Thus, we will ask whether the necroptotic pathway is activated in human FAD brains and test if survival complexes modulate the necroptotic pathway in vivo. We also obtained data that NMDAR currents decrease in PS1FAD mutant-expressing hippocampal neurons and here we will test whether FAD mutants affect the synaptic distribution of NMDAR. Since both BDNF and efnB1 modulate NMDAR currents, we will also test if FAD mutants affect modulation of NMDAR currents by Factors. Since in WT neurons Factors induce formation of survival complexes, we will test whether these complexes mediate the Factor-dependent modulation of NMDAR currents and synaptic plasticity.

A large amount of evidence links brain vascular disorders to cognitive impairment and dementia including Alzheimer's disease (AD). It is believed these defects impair neuronal health and function by restricting transport of nutrients and oxygen to neurons. Additional studies indicate that cerebrovascular dysfunction plays important roles in the pathogenesis of neurodegenerative disorders, a theory also supported by data that the risk of AD increases three fold in stroke patients. Crucial to vascular tissue function and repair is the process of angiogenesis, the generation of new blood vessels from pre-existing vasculature. Angiogenesis is promoted by complex mechanisms including endothelial cell sprouting and plays critical roles in neovascularization, the generation of new blood vessels in damaged tissue. Insufficient angiogenesis in AD brains may represent an important pathogenic mechanism affecting repair of vasculature and leading to neuronal dysfunction. There is now evidence that type 2 diabetes (T2D), a disease also associated with vascular abnormalities in the brain, is a risk factor for dementia. Thus, brain vasculature abnormalities observed in both AD and T2D may provide etiological links between the two disorders. Literature shows that the EphB4/ephrinB2 bidirectional signaling promotes angiogenesis and that the cytoplasmic domain of ephrinB2 is required for this function. Additional reports indicate that endothelial cell sprouting is promoted by angiogenic complexes of Raf-1, Rok-?, and vascular endothelial cadherin (VE- cadherin) and that presenilin1 (PS1) plays important roles in the development, maintenance, and integrity of brain vasculature. Recently, we obtained data that EphB4 stimulates angiogenic complexes between Raf-1, Rok-?, and VE-cadherin and increases sprouting of endothelial cells in vitro in a ?-secretase-dependent manner. In addition, we found that PS1/?-secretase mediates the EphB4-induced cleavage of ephrinB2 and stimulates production of cytoplasmic peptide ephrinB2/CTF2 that regulates cell sprouting. Together, our observations suggest that the PS1/?-secretase system regulates angiogenesis via proteolytic processing of ephrinB2 and production of ephrinB2/CTF2. Here we propose to use animal models to ask whether AD and T2D impair brain neovascularization in response to ischemic injury and whether formation of angiogenic complexes during neovascularization is affected by AD and T2D. In addition, we will examine whether angiogenic complexes in AD and T2D human brains differ from those in normal controls. We will also ask whether peptide ephrinB2/CTF2 promotes neovascularization in vivo and whether it acts as a protective factor against AD- and T2D-linked vascular impairments.

Mount Sinai ADRC (Sano): Research Education Component (REC) ? Research Summary The Research Education Component (REC) of the Alzheimer's Research Center (ARC) will provide critically needed training for junior faculty, senior postdoctoral fellows, and clinical research track residents and fellows, to conduct research on Alzheimer's disease-related disorders (ADRD). REC will support mentored research experiences for junior investigators during two vulnerable periods of their career development, early in their careers as they approach the end of residency or fellowship training, when they are too junior to obtain career development awards, or later when they have completed a career development award (e.g. K award), have a junior faculty position, but have yet to obtain RO1 or equivalent grant funding. Through the participation of distinguished senior faculty mentors in an intellectually and technologically rich academic environment at the Icahn School of Medicine at Mount Sinai (ISMMS), REC will also provide a mechanism to support gifted and highly motivated junior investigators who are new to AD research. Complementing state-of-the-art research training with teams of multidisciplinary REC mentors, REC scholars will receive an individually tailored didactic curriculum and exposure to multiple career-building activities, including a work- in-progress seminar series, science communication course, translational neuroscience and AD seminar series, grant and publication writing workshops and course, an academic survival and leadership seminar series, optional advanced coursework in neuroscience, genetics/genomics, and quantitative analyses, and a variety of additional resources available at ISMMS. REC scholars, moreover, will take advantage of well- designed, accessible ARC cores which will assist in their training allowing timely completion of their research program milestones and assisting scholars to meet goals set in their individual development plans (IDPs). REC objectives include: (1) To support trainees to conduct research to test questions and mechanisms important to the health of ADRD patients; (2) To provide advanced training in approach and methodologies needed to conduct high quality, ethical, and multidisciplinary research on ADRD disorders; (3) To provide multidisciplinary mentorship, with at least two interdisciplinary mentors, and an individually tailored career development plan; (4) To provide multiple forums that will encourage development of trainee presentation skills. (5) To prepare and assist trainees to submit and obtain external grant funding that is appropriate for their career stage (e.g. K award for postdocs and clinical fellows, or R21/RO1 for junior faculty), to sustain long-term academic careers as independent investigators and future leaders in the basic, translational, and clinical research of ADRDs. REC leadership, mentoring teams, and of course REC scholars themselves, will work together to closely assess short-term trainee progress and outcomes, and long-term trainee career development, providing critical metrics to evaluate overall success and guide future improvement.

Cerebral microvasculature abnormalities, such as degeneration of the capillary endothelium, are implicated in the genesis of Alzheimer's disease (AD) neuropathology. In addition, ischemic lesions that cause neuronal damage are often found in AD brains. The brain responds to ischemia by stimulating tissue neovascularization via sprouting angiogenesis; impairment of this function renders the brain vulnerable to the insult. It has been hypothesized that decreased angiogenesis in AD leads to insufficient blood flow and neuronal dysfunction in affected areas. The EphB4/ephrinB2 (efnB2) ligand-receptor system is known to regulate brain angiogenesis in response to ischemia. We present evidence that Presnilin1 (PS1), an important factor in familial AD (FAD), regulates angiogenic functions of EphB4/efnB2 such as stimulation of VE-cadherin angiogenic complexes in brain endothelial cells (ECs). We also found that this function is impaired by PS1 FAD mutants. In addition, we found that ischemia stimulates formation of the same angiogenic complexes in the brain and that this function is attenuated by PS1 FAD mutants together with ischemia-induced neovascularization and cerebral blood flow while neuronal death is increased. We also found that PS1 FAD mutants decrease the ?-secretase processing of efnB2 and production of the angiogenic peptide efnB2/CTF2 thus impairing the response of brain ECs to angiogenic factors. Together, our data support the hypothesis that PS1 FAD mutants inhibit ischemia-induced angiogenic functions of ECs by decreasing the ?-secretase processing of efnB2 and impairing the EphB4/efnB2 signaling, thus decreasing neovascularization in the adult brain and leading to increased vulnerability to this toxic insult and neuronal death. In this application we examine the roles of PS1 FAD mutants and ?-secretase on ischemia-induced angiogenic complexes, blood flow, angiogenesis and neuronal death. In addition, we propose to test a small peptide, which derives from the proteolytic processing of efnB2 by ?-secretase and which we found to rescue angiogenic functions of PS1 FAD ECs, for its pro-angiogenic and neuroprotective functions in brains expressing FAD mutants. In addition we aim to search for novel ischemia-induced angiogenic pathways that are impaired by PS1 FAD mutants and examine whether VE-cadherin angiogenic complexes and angiogenic protein expression are impaired in brains of human PS1 FAD and AD patients.