Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The role of inflammation and amyloid beta in Alzheimer disease pathology Dickstein, Dara L. 2004

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2005-99466x.pdf [ 16.13MB ]
Metadata
JSON: 831-1.0092329.json
JSON-LD: 831-1.0092329-ld.json
RDF/XML (Pretty): 831-1.0092329-rdf.xml
RDF/JSON: 831-1.0092329-rdf.json
Turtle: 831-1.0092329-turtle.txt
N-Triples: 831-1.0092329-rdf-ntriples.txt
Original Record: 831-1.0092329-source.json
Full Text
831-1.0092329-fulltext.txt
Citation
831-1.0092329.ris

Full Text

THE ROLE OF INFLAMMATION AND AMYLOID BETA IN A L Z H E I M E R DISEASE PATHOLOGY by Dara L. Dickstein B . S c , York University, 1997 A THESIS S U B M I T T E D I N P A R T I A L F U L F I L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F D O C T O R OF P H I L O S O P H Y in T H E F A C U L T Y O F G R A D U A T E S T U D I E S (Genetics) We accept this thesis as conforming to the required standard T H E U N I V E R S I T Y O F B R I T I S H C O L U M B I A November 2004 © Dara Lynn Dickstein, 2004 Abstract Alzheimer disease (AD) is the most common form of dementia. Due to longer life-spans the number of affected individuals is expected to triple over the next few decades. A s a consequence, a great deal of research is focused on determining the many processes by which the disease manifests as well as in discovering biomarkers and therapeutics to aid in diagnosis and disease prevention. The neuropathological hallmarks of A D include extracellular deposits of amyloid into senile plaques, accumulation of abnormal Tau filaments into neurofibrillary tangles, extensive neurodegeneration and inflammation. Although significant advances have been made in A D neurodegeneration, there still remain many unanswered and unforeseen aspects to the disease. It has been established that microglia, the immune cells of the brain, become activated in response to amyloid; however, the precise intracellular responses of microglia to amyloid and the relationship between microglia and amyloid deposition or clearance is unresolved. There have been many genes identified whose expression is upregulated in activated microglia and many of them have been proposed to be used as markers for inflammation. It has been demonstrated in humans that serum levels of melanotransferrin (p97), an iron binding molecule, is elevated in individuals affected with A D and that it is the activated, plaque-associated microglia that are responsible for this upregulation. This thesis further investigated the association between microglial activation and p97 gene expression and found that the levels of p97, both m R N A and protein, are increased in activated microglia in culture. The change in gene expression occurred largely in response to amyloid treatment possibly by the regulation of the AP-1 transcription factor downstream of the p38 mitogen-activated protein kinase pathway. n Moreover, p97 expression was altered by the treatment of activated microglia with anti-inflammatory drugs indicating that p97 may be used as a marker specific for amyloid-induced inflammation. The production and degradation of amyloid in the brain appears to be in a strict equilibrium. In A D , it is thought that the production of amyloid occurs at a faster rate than its removal and degradation and it is this shift in equilibrium that leads to plaque development. This study addressed the role of microglia in amyloid plaque formation using an A D transgenic model mouse that exhibits dysfunctional microglia. These mice accumulated amyloid deposits at the same rate as A D model mice however, limited numbers of mice did not allow for definite conclusions. Interestingly, these mice also displayed a shift in amyloid distribution, as indicated by increased vascular deposits, whereas normal A D model mice did not. Microglial activation and subsequent removal of amyloid deposits is one of the mechanisms suggested to explain the success of the amyloid beta vaccination treatment protocols. Immunization with amyloid and anti-human amyloid antibodies has resulted in the decrease in amyloid plaque burden, neurodegeneration, gliosis, early Tau pathology and cognitive and memory deficits. One aspect of A D not previously investigated was the effect of immunization on the integrity of the blood-brain barrier ( B B B ) . The studies performed in this thesis show that there was a decrease in B B B permeability after amyloid immunization. These data further support amyloid immunization as a treatment for A D as well as provide an explanation of the mechanism by which immunization effectively reduces A D pathology. 111 Table of Contents Abstract i i Table of Contents iv List o f Figures v i List of Tables v i i i List of Abbreviations ix Acknowledgements and Dedication x i i Chapter 1: Introduction 1 1.1 Alzheimer disease 1 1.2 Genetics of Alzheimer disease 2 1.2.1 Early onset gene candidates 3 1.2.2 Late onset gene candidates 5 1.3 The amyloid precursor protein and amyloid beta 7 1.3.1 The amyloid precursor protein 7 1.3.2 A P P processing 9 1.3.3 A m y l o i d beta, structure and function 14 1.3.4 Putative amyloid beta receptors 20 1.3.5 Animal models 22 1.4 Microgl ia , inflammation and Alzheimer disease 27 1.4.1 Microgl ia as immune cells of the brain 27 1.4.2 Morphological plasticity of microglia 28 1.4.3 The function of microglia in the central nervous system 30 1.4.4 The role of activated microglia in Alzheimer disease 31 1.4.5 Signal transduction pathways and microglial activation 34 1.5 The blood-brain barrier and Alzheimer disease 36 1.5.1 The blood-brain barrier, structure and function 36 1.5.2 B B B integrity and Alzheimer disease 39 1.6 Therapeutic strategies 41 1.7 Project rationale and general approach 44 Chapter 2: Materials and Methods 45 2.1 M i c e 45 2.1.1 Tg2576 A D model mice 45 2.1.2 Colony stimulating factor-1 deficient mice (op/op) 45 2.1.3 Generation of Tg/+;op/op mice 46 2.2 Preparation of reagents 46 2.3 Cel l culture 48 2.4 Ce l l stimulation 48 2.5 Creation of stable B V - 2 transfectant cell lines 49 2.6 R N A Isolation 53 2.7 Reverse transcriptase and Polymerase Chain Reaction 53 2.8 Real-time Polymerase Chain Reaction 54 2.9 T N F - a E L I S A assay 55 2.10 Western blot analysis 55 2.11 Immunohistochemistry 57 2.12 A(3 and antibody injection 58 2.13 Vaccination protocol 59 iv 2.14 Evans blue assay 60 2.15 Statistical analysis 61 Chapter 3: P97 expression in activated microglia 62 3.1 Rationale 62 3.2 Results 67 3.2.1 Microgl ia l activation 67 3.2.2 P97 expression in B V - 2 cells 69 3.2.3 P97 expression in Tg2576 A D model mice 72 3.2.4 M A P K pathways control the expression of p97 74 3.2.5 P97 expression in B V - 2 cells after treatment with N S A I D s 76 3.3 Discussion 80 Chapter 4: The role of microglia in amyloid deposition 85 4.1 Rationale 85 4.2 Results 88 4.2.1 Characterization of Tg/+;op/op mice 88 4.2.2 Amylo id burden in Tg/+;op/op mice 90 4.2.3 Microgliosis in Tg/+;op/op mice 95 4.3 Discussion 98 Chapter 5: Ap immunization and the blood-brain barrier 104 5.1 Rationale 104 5.2 Results 107 5.2.1 Ap peptide and anti-Ap antibodies and their ability to cross the B B B . . . 107 5.2.2 Anti-Ap antibody titres in immunized animals 113 5.2.3 A m y l o i d plaque burden in immunized animals 115 5.2.4 Microgliosis in immunized animals 119 5.2.5 B B B permeability in immunized animals 122 5.3 Discussion 126 Chapter 6: Concluding remarks and future directions 132 Appendix I: Domain Structure of A P P 138 Appendix II: Regional diagram of the brain 139 References 140 v List of Figures Figure 1.1. Mutations in A P P genetically linked to E O F A D 4 Figure 1.2. Schematic diagram of A P P and its metabolic derivatives 13 Figure 1.3. Morphological and functional plasticity of microglia 29 Figure 1.4. Schematic of the B B B 38 Figure 2.1. pEGFP-1 vector and multiple cloning site 51 Figure 2.2. Gel of digested p E G F P and p97 promoter construct 52 Figure 3.1. T N F - a production increased in B V - 2 cells treated with various known activators 68 Figure 3.2. P97 expression in treated cells 70 Figure 3.3. p97 expression is increased in affected brain regions in Tg2576 mice 73 Figure 3.4. The p97 promoter appears to be regulated by the p38 M A P K pathway 75 Figure 3.5. N S A I D treatment decreased T N F - a production in activated B V - 2 cells 77 Figure 3.6. p97 expression is decreased in B V - 2 cells treated with Ibuprofen 78 Figure 3.7. p97 expression is decreased in B V - 2 cells treated with Nimesulide 79 Figure 4.1. P C R genotyping of Tg2576 A D model mice 89 Figure 4.2. Tg/+;op/op mice are smaller than control littermates 90 Figure 4.3. A m y l o i d plaque burden in 9 month Tg/+;op/op mice compared to controls. 91 Figure 4.4. A m y l o i d accumulation in cerebral blood vessels of Tg/+;op/op mice 93 Figure 4.5. Reduced number and altered morphology of microglia in Tg/+;op/op mice. 96 Figure 5.1. A p peptides can cross the B B B in both transgenic and wild-type mice 109 Figure 5.2. anti-Ap antibodies cannot cross the B B B in both transgenic and wild-type mice I l l v i Figure 5.3. Antibody titre in serum of transgenic and non-transgenic mice immunized with either A p or P B S 114 Figure 5.4. A m y l o i d Pathology in Tg2576 Mice Immunized with A p or P B S 116 Figure 5.5. Cerebral amyloid levels are reduced in Tg2576 mice following A p immunization 118 Figure 5.6. Microgliosis in Immunized Mice 120 Figure 5.7. B B B permeability as determined by Evans Blue in cortical regions of A p and P B S immunized mice 124 v i i List of Tables Table 1. Summary of the primary APP-based transgenic mouse models of A D 26 Table 2. List of primer sequences and product 47 v i i i List of Abbreviations Amylo id beta A |3l-40 40-residue C-terminal variant of amyloid beta AP . .42 42-residue C-terminal variant of amyloid beta A D Alzheimer disease A D A M s A disintegrin and metalloproteinase A p o E Apolipoprotein E A P P Amylo id precursor protein Aph-1 Anterior pharynx-defective-1 B A C E P-site A P P cleavage enzyme B B B Blood-brain barrier bp Base pair B S A Bovine serum albumin C A A Cerebral amyloid angiopathy Caspases Cysteine aspartyl proteases C C Cerebral cortex C-terminal Carboxy-terminal C F A Complete Freund's adjuvant C N S Central nervous system C O X Cyclo-oxygenase C R E B c A M P responsive element binding protein CSF-1 Colony stimulating factor 1 Cu(II) Copper II D M E M Dulbecco's modified Eagle's medium D M S O dimethylsulfoxide c D N A Complementary deoxyribonucleic acid dNTP Deoxyribonucleotide triphosphate D T T dithiothreitol E L I S A Enzyme linked immuno-absorbant assay E O F A D Early onset familial Alzheimer disease E R , Endoplasmic reticulum E R K extracellular signal-regulated kinases F B S Fetal bovine serum F c R Fc receptor Fe(III) Iron III F P R L 1 formyl peptide receptor-like 1 G F P Green fluorescent protein h A P P Human amyloid precursor protein HI hippocampus H R P Horse radish peroxidase Ibu Ibuprofen I C F A Incomplete Freund's adjuvant IL Interleukin IFN-y Interferon gamma i.p. intraperitoneal i.v. Intravenous I N K c-jun N-terminal kinases L O A D Late-onset Alzheimer disease L P S Lipopolysaccharide LRP-1 Low-density lipoprotein receptor related protein M A P K Mitogen-activated protein kinase m R N A Messenger ribonucleic acid N . D . Not determined N-terminal Amino terminal N i m Nimesulide N F K B Nuclear factor K B NFTs Neurofibrillary tangles N S A I D s Non-steroidal anti-inflammatory drugs O D Optical density op/op CSF-1 deficient mouse P B S Phosphate buffered saline P C R Polymerase chain reaction P D Parkinson's disease Pen2 Presenilin enhancer 2 P F A Paraformaldehyde P P A R y Peroxisome proliferators-activated receptor gamma PS Permeability coefficient x surface area PS1 Presenilin 1 PS2 Presenilin 2 R A G E Receptor for advanced glycation end product R O S Reactive oxygen species sAPPa Secreted amyloid precursor protein alpha sAPPp Secreted amyloid precursor protein beta SDS page Sodium dodecyl sulfate-polyacrylamide gel electrophoresis SR Scavenger receptor Smac Second-mitochondria derived activator of caspase TfR Transferrin receptor TGF-p Transforming growth factor beta Tg/+ Tg2576 A D model mouse T N F - a Tumor necrosis factor alpha V E G F Vascular endothelial growth factor V L D L - R Very low-density lipoprotein receptor X I A P X chromosome linked inhibitor of apoptosis protein Zn(II) Zinc II Z O Zonula occudens proteins +/+ Wild-type mouse x i Acknowledgements and Dedication This thesis could not have been completed without the help of many people. First and foremost I would like to thank my supervisor, Dr. Wilfred Jefferies, for bringing me into his lab and giving me the opportunity to work on this exciting project. His encouragement and enthusiasm in this project, through the good times and the bad, have propelled me to excel and have taught me how to face things head on, to not give up and to think beyond the obvious. He has taught me how to be a successful scientist and for that I am eternally thankful. I am grateful to my committee members, Dr. Keith Humphries, Dr. Peter Reiner, Dr. Gerry Krystal and Dr. Elizabeth Simpson for their invaluable guidance and assistance, to the Genetics Graduate Program, the Biotechnology Laboratory and the Biomedical Research Centre. I would like to thank all of my lab members, past and present. In particular, Dr. Maya Kotfuri, Dr. Cheryl Pfeifer, Dr. M a k i Ujiie, and Jason Grant for their technical advice, their help with proof reading my thesis and most importantly for their friendship. A special thanks to Andy Jeffries and Ray Gopaul, for their assistance in experimental procedures, and their exceptional care of my animals, Brian Chung and Janet Lee for their help with many experiments and Dr. Aruna Somasiri, Arthur Legg, and Kenny To of Wax-it Histology for all their hard work and technical assistance. I also acknowledge the Alzheimer Society of Canada who granted me a Doctoral research scholarship for 4 years of my Ph.D. Finally I would like to recognize my family and friends for all their love and support and constant encouragement throughout the years. In particular I would like to express my gratitude to my parents, who have always inspired me to strive for what I believe in and to attain each goal I put before myself. xn This work is dedicated to the memory of my grandmother, Ruth Arbuck, 1914-2003 xi i i Chapter 1: Introduction 1.1 Alzheimer disease Alzheimer disease ( A D ) was first described in 1907 by Alios Alzheimer as a neuropathological syndrome characterized by progressive dementia and deterioration of cognitive function along with neurological lesions identified as dark staining plaques and fiber-like tangles \ A D accounts for approximately 65% of all dementia cases in the elderly 1 . It is estimated that there are approximately 20 million people affected worldwide with either A D or mild cognitive impairment. The prevalence of A D , in the general population, increases with age as the rate is 3% in those between 65 and 74 years compared with 47% among those over 85 years of age 3 . Progressive memory and cognitive decline as well as difficulty in language, praxis and visual perception are the clinical manifestations which characterize the disease. The decline in intellectual function progresses at a slow but inexorable rate and leads to severe debilitation and death within 12 years after onset 4. Pathologically, A D selectively damages brain regions and neural circuits critical for cognition and memory and is distinguished by the presence of proteinaceous deposits in the brain, comprised of extracellular amyloid plaques and accumulated paired helical filaments of hyper-phosphorylated Tau in intracellular neurofibrillary tangles (NFTs), dystrophic neuritis and neuropil threads 5 ' 6 . Inflammation also plays a key role in A D and appears to be facilitated by activated microglia, the immune cells of the central nervous system (CNS). It took 75 more years to determine that the main constituent of the plaques was a 40-42 amino acid peptide referred to as amyloid p (AP).7. AP is a 1 metabolic product resulting from the proteolytic cleavage of the amyloid precursor protein (APP) and its aggregation into fibrils is thought to be the central event of A D . The "amyloid cascade hypothesis" proposes that A P precipitation into fibrils initiates the formation of amyloid plaques which in turn contribute to the formation of neurofibrillary tangles, initiate complement cascades and inflammatory processes and ultimately culminate in cell death 8 . However, this hypothesis is not consistent with recent advances that implicate inflammation, NFTs and oxidative stress as independent processes that may even be upstream of A P aggregation. 1.2 Genetics of Alzheimer disease The inheritance of predisposing genetic factors appears to play an important role in A D . After age, family history is the second greatest risk factor for A D . There are two classes of A D : early-onset familial A D ( E O F A D ) , where the age of onset is less than 60 years of age; and late-onset A D ( L O A D ) , where age of onset is greater than 60 years. The genetics of A D are complex since A D is a heterogeneous genetic disorder in which numerous genetic factors, with both minor and major effects, play independent, simultaneous and interdependent roles. Further complexity arises since mutations and polymorphisms in multiple genes are acting together with many environmental factors. Moreover, A D represents a dichotomous situation where genes which cause E O F A D are rare in prevalence but 100% penetrant, while genes which confer increased risk for L O A D are highly prevalent with low penetrance 9 . Present research is focused on elucidating the genes responsible for the various aspects of A D . This wi l l make it 2 possible to create and estimate a person's genetic susceptibility profile that w i l l aid in both early diagnosis and the development of preventative treatment. 1.2.1 Early onset gene candidates The first candidate gene for A D was discovered in the early 1980's. Linkage analysis and subsequent positional cloning techniques were performed on multigenerational families who all had E O F A D . The inheritance pattern in all these families appeared to be autosomal dominant and highly penetrant. The findings from these families showed linkage to chromosome 21 and focused on a candidate gene encoding the A P P 1 0 . Although association was later found to be a false positive, it did lead investigators to a compelling candidate gene. More persuasive evidence for the involvement of A P P in A D was the fact that Down's Syndrome patients, who have trisomy 21, had strikingly similar brain pathology to those suffering from A D " . In addition, Down's patients had increased A P P messenger ribonucleic acid ( m R N A ) expression along with elevated A P P and A P in the serum and brain 1 2 ' 1 3 . Finally, mice overexpressing mutant forms of A P P developed A D - l i k e pathology, further implicating A P P as a genetic determinant for A D . The first A P P mutation was found in 1990 in a Dutch family where individuals were affected with cerebral hemorrhage with significant amyloid deposits in accordance with an autosomal dominance inheritance pattern 9 . Since then 20 mutations in the A P P gene have been found, all of which are missense mutations located close to or within the coding region of the A p fragment in the P and y-secretase cleavage sites (Figure 1.1) 1 0 . Overall, mutations in the A P P gene result in an increase in the production of Api_42 peptide, the more amyloidogenic and toxic species of A p . Other mutations, such as the 3 London mutation (V7171), cause an increase in the ratio of APi_42 peptide to A P M O peptide l 4 , whereas the Swedish mutation (K670N/M671L) causes an increase in the production of both species of ApMo and AP142 ' 5 . Individuals with A P P mutations have an average year of onset of 49 ± 8 years and disease duration of approximately 12 years . Mutations in the A P P gene account for 5-7% of all E O F A D cases which account for less than 2% of all A D cases. p-secretase a-secretase i G (Arctic) K (Italian) N (Iowa) t ...SEVKM Hi (Swedish) N L DAEFRHDSGYEVHHQKLVFFAEDVGSNKGAIIGLMVGGVVI i i (Flemish) G Q (Dutch) y-secretase (Belgian) A (French) MV (Florida) It gTVIVITLVMLKK... / / | \ |S I I F G P (Austrian) (London) (Australian) Figure 1.1. Mutations in A P P genetically linked to E O F A D The A P coding region within the A P P is expanded and shown by amino acid code. The arrows indicate residues with known missense mutations which have been identified in E O F A D and/or hereditary cerebral hemorrhage with amyloidosis. Mutations of A P P near the positions 670, 693 and 715 have been found to increase the risk of A D . A double mutation at K670 and M671 increases the production of both A P M O and Ap42, while mutations near the y-secretase cleavage site favors the production of Api_42- Three-digit numbers refer to the residue number of A P P (Adapted from Selkoe 1 6 ) . A year after the discovery of mutations in the A P P gene, another F A D locus was found to be linked to chromosome 14 and later identified as the presenilin 1 (PS1) gene 9 . Shortly after, another gene encoding a second member of the presenilin family called presenilin 2 (PS2) located on chromosome 1 was also identified . To date there are 4 approximately 140 mutations in PS1 and 10 mutations in PS2, and together presenilin mutations account for the majority of EOFAD cases 1 0" 1 6. PS1 encodes a 7 transmembrane spanning protein that functions as an aspartyl protease and is required for y-secretase activity 1 7 . At first it was thought that PS1 was the y-secretase. However, many studies have demonstrated that PS1 is part of a protease complex consisting of at 18 19 20 least three other proteins including, nicastrin, APH-1, and Pen-2 ' ' . Mutations in presenilin genes result in the over production of A P 1 - 4 2 , presumably by altering y-secretase activity 2 1 . Individuals with PS1 mutations have an earlier age of disease onset (43.5 ± 0.7 years) and have shorter disease, duration (4.5 ± 0.7 years), while PS2 S 22 mutations have a later age of onset and longer duration . Interestingly, PS2 mutations are not highly penetrant compared to PS1 mutations which are completely penetrant. Overall, mutations in PS genes account for greater than 55% of all EOFAD mutations. 1.2.2 Late onset gene candidates LOAD accounts for approximately 95% of AD cases and is classified as sporadic AD. The genetic contributions to LOAD are more difficult to address due to the lack of complete family histories and fewer living subjects from which to obtain blood samples. There are a few gene candidates for LOAD which have been identified by association or linkage disequilibrium studies. Co-segregation of particular allele(s) and disease phenotype was one of the strategies employed to identify an allele of the gene encoding apolipoprotein E (ApoE) as a risk factor for AD 2 3 . There are three alleles of the ApoE gene; s2, s3, and s4. Individuals who have the e4 allele, either heterozygous or homozygous, have a greater risk of developing AD; however, there are many individuals who are homozygous for the 84 allele and do not develop dementia. Therefore, the e4 5 allele is neither necessary nor sufficient for the development of A D . A p o E is a 299 amino acid glycoprotein that normally functions in cholesterol and lipid metabolism 2 4 . It is thought that the s4 allele acts as a modifying gene by decreasing the age of onset in a dose dependent manner 1 0 . The s4 isoform has also been reported to promote A p aggregation by mechanisms not yet elucidated. It is thought that e4 is less capable of clearing A P from the neuropil and that 84 binds to A p with a lower affinity than the other isoforms 2 5 . Genetic linkage studies have also given rise to many more candidate genes including a 2-macroglobulin, low-density lipoprotein receptor related protein (LRP-1) , insulin degrading enzyme, urokinase plasminogen activator and the very low-density lipoprotein receptor ( V L D L - R ) . a2-Macroglobulin and L R P - 1 , both located on chromosome 12, appear to play a role in A P clearance and degradation. LRP-1 can also serve as a receptor for A p o E , a2-macroglobulin and secreted forms of A P P and polymorphisms in exon 3 and 6 have been linked to A D 9 ' 2 6 . Insulin degrading enzyme and urokinase plasminogen activator, located on chromosome 10, have also been suggested to play a role in A P degradation 2 1 . Finally, V L D L - R , located on the short arm of chromosome 9, may be a receptor for lipoproteins such as A p o E 9 . There are also many inflammatory factors that are deemed to have high risk alleles. Polymorphisms in interleukin ( IL) - la , IL-1 P, IL-6, tumor necrosis factor a (TNF-a) and d\-antichymotrypsin have all been shown to influence A D risk 2 8 . There are approximately 10 polymorphisms that have been found in the general population and individuals carrying one or more of these alleles are hypersensitive to oddities resulting in 28 inflammatory processes that can cause increased degeneration . Since the genes 6 identified to date are believed to only account for approximately 30% of the genetic variance in A D , the search for genetic factors associated with A D is an ongoing research effort. 1.3 The amyloid precursor protein and amyloid beta 1.3.1 The amyloid precursor protein With the identification of A P P being a genetic determinant for E O F A D , research began into clarifying the biological role of A P P and its proteolytic cleavage products. A P P is a type 1 transmembrane glycoprotein that is localized to chromosome 21 (21q21.2-3), contains 18 exons and spans 170 kilobases 2 9 . A P P is expressed ubiquitously throughout the body and can exist in at least three isoforms: APP770, APP751, and APP695, arising from alternative splicing with different isoforms expressed in specific tissues 3 0 . The full length protein contains zinc and copper binding sites, a cysteine rich subdomain and an acidic region at the amino terminus. There is also a kunitz protease inhibitor domain, although the functional significance of this domain is unclear 4 . In the brain the APP695 isoform is the predominant form and is expressed in glial, endothelial and neuronal cells. A P P has been observed to be localized to many membranous structures including the endoplasmic reticulum (ER), Golgi compartments, plasma membrane, postsynaptic densities, axons, and dendrites. In addition, A P P has been found in cholesterol and G M 1 ganglioside rich membrane microdomains . The exact function of A P P is not clear and many studies have focused on the role of A P P and its fragments. It had been demonstrated that A P P and its fragments, such as 7 secreted APPa (sAPPa), secreted APPP (sAPPP) and Ap, are powerful regulators of neuronal functions including synaptic plasticity and transmission, neuritic outgrowth, cell excitability, cholesterol metabolism, cell adhesion, cell death and gene transcription '. One possible function of full length APP is as a cell surface G-protein-coupled receptor. Binding of G-proteins to APP causes the G-protein to dissociate into its dimeric form resulting in the activation of multiple signal transduction molecules such as phospholipase C, voltage dependant calcium channels, adenylyn cyclase and many enzymes involved in apoptotic signaling cascades. However, the role of APP mediated G-protein activation is not fully understood. APP has also been implicated in cell adhesion. Immunohistochemical studies have demonstrated that cell surface APP co-localizes with adhesion proteins such as P-l integrin 3 1 and telencephalin 3 2 . Moreover, APP contains several domains that facilitate the binding of heparin, collagen and laminin Other possible functions include synaptic transmission and plasticity and more recently memory and cognition. Nerve process outgrowth is thought to be regulated, in part, by APP. Studies focusing oh embryonic development and APP expression have shown that APP expression is at its highest between embryonic day 6-9, when neuritic outgrowth is at its maximum 4 . Moreover, in vitro experiments with synthetic APP demonstrate that APP is able to stimulate neuritic outgrowth in PC 12 neuronal cells 4 . Research on APP function is conflicting since most AD animal models are transgenic for the human form of the protein that may interfere in the evaluation of the endogenous mouse protein function. 8 1.3.2 A P P processing The processing of A P P into its metabolites involves three different enzymes: P-, a-, and y-secretases and two cleavage pathways (Figure 1.2). In A P P , there are two main cleavage sites on the extracellular side of the membrane and one site within the transmembrane domain. P- and a-secretase cleavage appears to be mutually exclusive events, each generating soluble carboxyl-terminal (C-terminal) fragments. The remaining amino-terminal (N-terminal) fragment of A P P is then cleaved by y-secretase 4 . In the non-amyloidogenic pathway, A P P is cleaved by a-secretase at a cleavage site located within the A P peptide. This cleavage precludes the production of A p producing a 612 amino acid soluble protein, sAPPa , which has been shown to have neuroprotective and memory enhancing effects 3 3 . Cleavage at this site is dependent on secondary structure and on the location of the cleavage site from the membrane. In addition, a-secretase activity can be regulated by protein kinase C activity as well as by other molecules such as estrogen, testosterone, various neurotransmitters and growth factors 3 4 . Although the exact identity of a-secretase is still unknown, three proteins have been described that appear to have a-secretase-like activities. Studies with protease inhibitors have shown that a-secretase is a zinc metalloproteinase and belongs to the family of a disintegrin and metalloproteinase ( A D A M s ) . The three candidate proteins are A D A M I 7 (also known as T N F - a convertase), A D A M 10 and A D A M 9 (also known as M D C 9 ) 3 3 . A D A M s are type 1 integral membrane proteins with a multi-domain structure and function in cell adhesion, cell fusion, intracellular signaling and proteolysis 3 4 . The expression and localization of each these proteins in A D are altered and gives rise to the potential role of each in A P P processing. A D A M I 7 has been found to localize to 9 neurons as well as to senile plaques and tangles, although there is no change in its expression in A D . In contrast, A D A M 10 protein levels are significantly reduced in A D , as are levels of s A P P a and a-secretase ac t iv i ty 3 3 . The emerging hypothesis, as suggested by studies involving knock out mice and enzymatic inhibition, is that all three A D A M enzymes are involved to a similar extent in A P P processing acting together at the a-secretase cleavage site to varying degrees depending on cell type. It is possible that A D A M 10 is involved in both constitutive and regulative pathways of a-secretase, whereas A D M A 1 7 may be involved solely in the regulatory pathway and A D A M 9 in the actual cleavage event 3 3 . In the amyloidogenic pathway of A P P processing, A P P is first cleaved by P-secretase followed by cleavage with y-secretase. These events generate two metabolites, sAPPp and the AP peptide. Two enzymes capable of cleavage at the P-secretase cleavage site are the P-site A P P cleavage enzymes ( B A C E 1 ) and B A C E 2 6 . B A C E 1 , also referred to as Asp2 or memapsin2, is a transmembrane aspartyl protease and exists in three isoforms. B A C E 1 is expressed at high levels in the pancreas, moderate levels in the brain and at low levels in most peripheral tissue. However, B A C E 1 activity is highest in neuronal tissue and is increased in A D 3 : > . Intracellularly, B A C E 1 can be found primarily in the trans-Golgi network, and the endosomal compartments, although it can also be found in the endoplasmic reticulum and on the cell surface 3 4 . Cleavage by B A C E 1 is the rate limiting step in the generation of A p . B A C E 1 cleaves at two positions depending on intracellular localization, either at the A s p l (P-site) or at G l u l 1 (p'-site) of AP, generating an APi_4o/42 fragment or an APi 1.40/42 fragment, respectively. p-site cleavage predominantly occurs in the E R while P'-site cleavage occurs in the Golgi 3 4 . Further 10 evidence to support the role of B A C E 1 as p-secretase comes from studies on various A D animal models. In A P P transgenic mice expressing the Swedish mutation, there is a 10 fold increase in B A C E cleavage activity. Moreover, mice which overexpress human B A C E have a significant increase in A01.40/42 levels and knock out of this gene perturbs A P P processing and prevents AP generation 3 6 . Finally, in B A C E l ' / A P P mice there is a lack of AP compared to other A P P transgenics 3 6 . B A C E 2 , another aspartyl protease, is 55% identical to B A C E 1 and shows similar substrate specificity to B A C E 1 . B A G E 2 is widely expressed in peripheral tissues but is not highly expressed in the brain. It cleaves AP at position A s p l and at position Phel9 or Phe-20 of the A p peptide 3 4 . The precise role of B A C E 2 in A P P processing is unclear. There is no direct evidence for the role of B A C E 2 in A D however, cleavage at positions 19 and 20 located within A p is affected by the F A D Flemish mutation 3 4 and B A C E 2 maps to chromosome region 21q22.2-22.3 3 7 . The final event in A P P processing is the cleavage of the C83 or C99 A P P fragments by y-secretase. Cleavage occurs at a site located within the A P P transmembrane domain. It was first thought that PS1 and PS2 were y-secretases since mutations in PS1 and PS2 cause an increase in the production of A P i ^ 2 3 S - In addition, PS1 and PS2 contain aspartic acid residues in transmembrane domain 6 and 7, sites necessary for aspartyl protease activity and hence y-secretase activity 1 7 . PS1 is expressed in its full length form as an integral membrane protein with 8 membrane spanning regions. In this form, PS1 is thought to be inactive. To form an active presenilin complex, the presenilin protein is cleaved between membrane domains 6 and 7. This yields C- and N-terminal fragments, which remain coupled and form a 11 heterodimeric active complex \ If one or both of the aspartic residues in presenilin are mutated or knocked out, the production of A(3 and the P3 peptide is greatly reduced along with an associated increase in the amount of C-terminal fragments, s A P P a and sAPP(3 1 7 . This theory was questioned and an alternative hypothesis proposed that states that y-secretase is multifaceted and is made up of several proteins. The role of presenilins in this complex would be in A P P trafficking to sites where cleavage by the complete y-secretase complex can take place. This viewpoint was corroborated by biochemical and genetic studies which have identified four membrane proteins as components of y-secretase: heterodimeric presenilin, (composed of its N - and C-terminal fragments); anterior pharynx-defective-1 (Aph-1); nicastrin, (glycosylated); and presenilin enhancer 2 (Pen-2) 2 0 . 12 .SEVKM DAEFRHDSGYEVHHQKLVFFAEDVGSNKGAI IGLMVGGVVI TVIVITLVML.. ft-secretase sAPPB y-secretase a-secretase s A P P a Y-secretase Figure 1.2. Schematic diagram of APP and its metabolic derivatives A P P is a type 1 membrane protein and exists in three major isoforms, A P 7 7 0 , A P 7 5 1 , and AP695. Enlargement of the A p peptide region shows the amino acid sequence and primary sites of cleavage by the various secretases. Cleavage of A P P by either P- or a-secretase results in soluble N-terminal fragments, sAPPp and s A P P a and two membrane-bound C-terminal fragments, C99 and C83, respectively. Subsequent cleavage by y-secretase gives rise to a non-pathogenic A p peptide and P3 (Adapted from Small, 4 ) . Nicastrin is a type 1 transmembrane glycoprotein that is 709 amino acids long and contains 4 conserved cysteine residues at the N-terminus. It is encoded on chromosome lq23, a region previously identified as a susceptibility locus for L O A D 3 4 . Nicastrin has been shown to interact with PS1 and PS2 as well as with the C-terminal A P P fragments, C83 and C99. Mutations in either PS 1/2 directly effect nicastrin trafficking and post-translational modifications thereby preventing its maturation and function. In this state, the y-secretase complex cannot form and A P production is abolished 3 4 . Aph-1 and Pen-2 were two proteins discovered through a genetic screen of Caenorhabditis elegans. These 1 3 genes both encode multi-pass transmembrane proteins that have the capability of binding to PS1, PS2, and nicastrin with high affinity. The specific role of Aph-1 and Pen-2 and the protein-protein interactions responsible for assembling Aph-1 and Pen-2 into the multimeric protease complex remains unknown. Recently, it was revealed that Aph-1 binds to nicastrin in its immature, non-glycosylated form and remains bound after glycosylation and maturation 3 9 . In regards to how this complex forms, it is thought that Aph-1 and nicastrin initially form a subcomplex. Once nicastrin is modified and matured the Aph-1/nicastrin complex binds to and stabilizes presenilin. Pen-2 then binds, confers the y-secretase activity and facilitates endoproteolysis of presenilin which serves as the catalytically active core of the y-secretase complex 3 9 . There are many details about the active y-secretase complex that remain to be uncovered These include determining which y-secretase co-factors physically interact with one another and identifying the protein-protein binding domains that govern these interactions. 1.3.3 Amylo id beta, structure and function A P is a 4 kiloDalton peptide and is a normal cellular product and exists in two predominant forms with different C-termini, Api^o and Api42- A P contains seven positively and seven negatively charged residues at the N-terminal region and a 2 1 hydrophobic domain at residues 17-21 and 29-40/42 (Figure 1.2) . The A P protein exists as monomers, dimers, and oligomers and can undergo further aggregation to yield full fibrils. The conformation of A P appears to be in a dynamic flux and is dependant upon environmental conditions, metal binding and interactions with various other proteins. In its monomeric state A P exists as a random coil containing two a-helices between residues 15-23 and 31-3 5 4 0 . Transition of these a-helical coils into P-pleated 14 sheets facilitates fertilization where the p-pleated sheets are anti-parallel to one another. It is these full fibrils that are found to make up the bulk of the senile plaques and are thought to have neurotoxic properties 4 1 . However, pools of soluble A p have been found in the brains of A D patients indicating that monomeric, dimeric and oligomeric forms of A p may facilitate the pathology of A D 4 1 . The proportion of soluble A P M 2 in these pools is significantly higher than soluble A P M O levels. Soluble A P i ^ 2 dimers have also been shown to have a higher magnitude of neurotoxicity than Api^o 4 1 • The deposition of amyloid is not found in all regions of the brain. A P plaques exhibit a spatio-temporal distribution pattern where regions such as the cortex and hippocampus contain many plaques and regions such as the cerebellum exhibit little to no A p . One possible explanation for the distribution pattern of A p has been proposed stating that the presence of A p in the neuropil of neurons sets in motion a cascade of reactions including inflammation, neuritic sprouting and the upregulation of A P P expression 4 . This model is based on a positive feedback loop which relies on neuritic sprouting and A P P expression. It is known that as the brain ages, certain areas become myelinated and mechanisms which facilitate neuritic outgrowth are suppressed. However, regions of the brain responsible for memory have ongoing synaptic remodeling, continuation of neuritic outgrowth and are thus the most vulnerable. These include the cortex, hippocampus and olfactory system all of which are affected in A D 4 . In addition, regions of the brain that are the last to undergo myelination, such as the temporal cortex, are the first regions to be affected. Therefore, myelin proteins may prevent neuritic sprouting and subsequent A P P expression thereby protecting a given brain region from degeneration 4 2 . 15 The aggregation of A|3 into fibrils and toxicity can occur in response to a variety of actions, including a shortage in the supply of energy in the cell, oxidative stresses, calcium dysregulation and apoptosis. Studies have shown that a disruption in mitochondrial energy metabolism and dysregulation of calcium homeostasis can up regulate the expression of A P P as well as promote the pro-amyloidigenic processing of A P P to A p and subsequent plaque formation 4 3 ' 4 4 . Oxidative stresses like hydrogen peroxide, ultraviolet light, and superoxide radicals have been shown to increase the production of neuronal Ap . Moreover, high concentrations of extracellular A P had been shown to induce oxidative stress and in turn render cells vulnerable to exocitoxicity and apoptosis through the dysregulation of calcium homeostasis. Metal ions have also been shown to cause A P aggregation and promote the formation of diffuse plaques. In the A D brain, there are significant amounts of transition metals, such as copper II (Cu(II)), zinc II (Zn(II)) and iron III (Fe(III)), in both A p plaques and neuropil 4 5 . Histidine residues located in the A p peptide bind to Cu(II), Zn(Il) and Fe(III) as sequester it in the brain during times of oxidative stress and inflammation, since the interaction of these metals with hydrogen peroxide, which is generated in both processes, would cause the production or reactive oxygen species 4 5 . Treatment of post-mortem A D brains with metal-chelators confirmed this finding since the chelators were able to enhance the solubilization of A P 46. The generation of reactive oxygen species has many harmful consequences including disruption of calcium homeostasis by impairing ATPase activity, increase lipid peroxidation and altering the activity of the anti-oxidant enzyme, superoxide dismutase 4 1 . 16 It was originally demonstrated in non-neuronal cells, that APP metabolism can elevate intracellular calcium levels which enhances A P production. This creates a feed back loop since an increase in A p results in the increase of intracellular calcium 4 7 . Soon after studies in neuronal cells, other studies reported that exposure to fibrillar A P causes a disruption in calcium homeostasis generally leading to an increase in cytosolic calcium 47,48,49 j j i e m f l u x 0 f calcium is thought to be mediated by L-type voltage-gated calcium 48 channels and is regulated by the mitogen-activated protein kinase (MAPK) pathways . Under normal physiological conditions, calcium influx in neuronal cells has been implicated in promoting synaptic plasticity and thus beneficial. However, if too much calcium enters the cell or if the entry of calcium occurs at a dramatic rate, it could cause down-regulation of neuronal plasticity mechanisms due to the self-inhibitory mechanisms of calcium '. Pierrot et al. have recently established that high cytosolic calcium concentrations favour the amyloidogenic pathway of APP by preventing the cleavage of APP by a-secretase possibly through modifications of APP by phosphorylation 4 9 . In addition, high calcium levels resulted in the increase in AP142, with no change in A p ^ o , suggesting that calcium may also regulate y-secretase activity. Nonetheless, alterations in calcium homeostasis are detrimental to neuronal survival due to their high polarized nature. Recent advances in AD research suggest that in addition to extracellular A p , intracellular neuronal A P plays a significant role in neurodegeneration. Increasing evidence indicates that A P is more toxic to cells in its protofibrillar state rather than in its P-pleated conformation 1 6 . Initial studies by Wertkin et al. demonstrated that NT2N neuronal cells produced A P and either stored it in the cell or secreted it into the medium 17 . Examination of brain sections from young and old animal models illustrated an anterior to posterior gradient of intra-neuronal A p immunoreactivity in the hippocampus which increased with age in distal processes and synaptic compartments 5 1 . What effect intraneuronal A p exerts on neurons is unclear. Current studies in h A P P and PS transgenic rats propose that physiological levels of A p play a role in synaptic plasticity via activation of c A M P Responsive Element Binding Protein ( C R E B ) directed gene expression 5 2 . A pathological increase in intraneuronal A p would cause abnormal phosphorylation patterns that end up dysregulating these pathways. This is noteworthy since CREB-dr iven gene expression has been found to be important for learning and memory. It was also found that A p caused an upregulation of extracellular signal-regulated kinases ( E R K ) , and E R K then phosphorylates a number of proteins, including Tau, inferring a further link between A p and Tau pathways 5 2 . In addition, both aged animal models and humans exhibit high levels of AP immunoreactivity in the cholinergic neuron of the basal forebrain. The accumulation of intracellular A p in neuronal processes and synapses is suggested to be associated with the manifestation of cognitive decline seen in A D , since accumulation of A p is concordant with abnormal cytoskeletal architecture and synaptic dysfunction 5 3 . Synaptic dysfunction is one of the proposed causes of cognitive decline. Moreover, oligomer A p neuronal accumulation and cognitive decline occur before plaque deposition. These findings proposed the "Synaptic A p Hypothesis" that proposes A P accumulates in neuronal synapses in the brain. Once the A p load in the neuron becomes too cumbersome, the cell dies and releases the amyloid into the extracellular space. The released amyloid can then act as a seed and attract soluble A P MO/42 thereby forming a plaque 5 3 . 18 Apoptotic processes are viewed to play a major role in A D pathology. Evidence suggests that selective neuronal loss seen in A D involves the activation of cysteine aspartyl proteases (caspases), which initiate and execute apoptosis 5 4 . Both extracellular A p oligomers and intracellular A p may activate caspases. Extracellular A P initiates caspase activity via activation of cell surface death receptors. This pathway involves signaling through cell surface death receptors, such as the T N F receptor, which are regulated by decoy receptors and Fas-associated death domain proteins, however, direct binding of A p or A p oligomers to death receptors remains to be shown 5 5 . It is speculated that the binding of A p to these receptors causes downstream activation of various caspases such as caspases 2 and 8, leading to the activation of caspases 3, 6 and 7, all of which are important initiators of the apoptotic signaling cascade 5 4 . Alternatively, intracellular A P may activate caspases through a process that involves E R stress or mitochondrial stress. This leads to the downstream activation of caspase 9 and caspases 3, 6 and 7 5 6 . Caspase activation is thought to facilitate the cleavage of Tau thereby favouring conformational changes of the protein into paired helical filaments. The accumulation of the altered Tau proteins causes cytoskeletal disruption and the consequent failure of axoplasmic and dendritic transport that culminates in neuronal death 5 4 . A P has also been shown to elicit its toxic effects by the initiation of host immune responses. The action of A P on glial cells results in an inflammatory reaction and progressive amyloid deposition promotes the chemotaxis and subsequent activation of microglia 2 6 . Studies by McDonald et al. have established that exposure of fibrillar A P to microglia or monocytes activates M A P K pathways involving p38 and the E R K , as well 19 as tyrosine kinase-dependent signaling pathways, involving Lyn , Syk and F A K ' . The activation of these pathways leads to changes in the expression of various cytokine and pro-inflammatory genes and generates reactive oxygen intermediates, respectively, leading to further neurotoxicity and degeneration. The extensive effect of A p on the inflammatory response elicited by glial cells w i l l be discussed in greater detail later. 1.3.4 Putative amyloid beta receptors The binding of amyloid to plasma membranes has been implicated in many of the pathological features of A D . Binding of A P to various receptors elicits neurotoxicity in neurons and cerebral vascular endothelia and activation of inflammation in microglia. Due to its structure, A p can bind to a variety of molecules including proteins, proteoglycans, and lipids 5 9 . On microglia, scavenger receptor A (SR-A) and BI (SR-BI), CD36, heparin sulfate proteoglycan, formyl peptide receptor-like 1 (FPRL1) and a complex of CD36, a 6 p 1-integrin, and CD47 have been shown to bind to A p . Neuronal receptors include the N-methyl-D-aspartate receptor, the a7-nicotinic acetylcholine receptor, the p75 neurotrophin receptor and the CLAC-P/col lagen type X X V receptor 5 9 . Receptors common to glial cells, neurons and cerebral endothelia include the receptor for advanced glycation end product ( R A G E ) , L R P , the insulin receptor, integrins and the serpine-enzyme complex receptor 5 9 . These receptors vary in their ability to bind monomeric or fibrillar A p . A few of these receptors wi l l be discussed below. S R - A and SR-BI receptors are expressed on microglia, bind fibrillar A p and mediate the clearance of amyloid aggregates 6 0 . These receptors have also been implicated in inflammatory responses and the production of reactive oxygen species; 20 however they are not needed for microglial activation. Expression of S R - A and SR-BI are developmental^ regulated and are not expressed normally in adult mouse brain 6 1 ; however, expression of S R - A and SR-BI m R N A and protein are both upregulated in-vitro and in-vivo upon exposure to lipopolysaccharide (LPS), interferon gamma (IFN-y) and IL-1 a 6 2 . Another microglial receptor for A p is CD36. CD36 is a class B scavenger receptor that has been shown to bind to fibrillar A p and contribute to microglial activation. A p binding to CD36 elicits many signaling cascades involving Src family kinases Lyn and Fyn, and the mitogen-activated protein kinases ( M A P K ) as well as the production of reactive oxygen species. CD36 is also involved in a multi-receptor complex with a6pl-integrin and CD47 , and this complex mediates the binding of microglia to fibrillar A p and the subsequent activation of pro-inflammatory pathways, respiratory bursts, adhesion and cell migration . Both S R - A and CD36 are elevated in microglia in brains of A D patients compared to controls 6 4 . F P R L 1 is a seven transmembrane, G-protein coupled protein and is thought to initiate pro-inflammatory effects in response to both soluble and fibrillar A P 65. Activation of F P R L 1 can lead to the production of reactive oxygen species and pro-inflammatory cytokines. In addition, F P R L 1 is involved in mediating the chemotactic activity of A p on microglia. Other studies have shown that FPRL1 can form a complex with A p and can be internalized into cytoplasmic compartments resulting in the accumulation of intracellular amyloid aggregates 6 6 . Under normal condition FPRL1 is expressed at low levels. In A D , F P R L 1 expression is increased and is seen at high levels in plaque infiltrating mic rog l i a 6 5 . The R A G E receptor is a multi-ligand receptor in the immunoglobulin superfamily found in neurons, microglia and cerebral endothelial cells. It binds many ligands such as 21 AP, the S1 OO/calgranulin family of pro-inflammatory cytokine-like mediators and the high mobility group 1 D N A binding protein amphoterin 6 7 ' 6 7 . The interaction of Ap with R A G E initiates signaling cascades that result in oxidative stress via the production of reactive oxygen species and lipid peroxides as well as the enhanced expression of macrophage colony stimulating factor, which stimulates microglial proliferation and receptor expression. It has also been implicated in Ap internalization on microglia 6 8 . Its expression is ligand dependent and is up-regulated in A D , particularly in the vasculature. R A G E expression is also increased in neurons and microglia, but not as much as in cerebral endothelia. Many in vitro and in vivo studies.have demonstrated that R A G E mediates AP transport, therefore, R A G E is thought to be the major influx receptor for peripheral Ap at the B B B 6 9 ' 7 0 ' 6 7 . LRP is a member of the L D L receptor family and is a multifunctional scavenger receptor and signaling receptor. Its ligands include biomolecules such as ApoE and 0C2-macroglobulin, tissue plasminogen activator, APP and lactoferrin 7 1 . LRP has been genetically linked to L O A D , but the exact mechanism by which LRP affects disease onset in not known 9 . Expression of LRP in A D is decreased and many animal studies involving LRP deficient mice and A D mice results in increased cerebral amyloid load and increase in parenchymal amyloid plaques 7 2 ' 7 3 . As a result, LRP is thought to 70 regulate AP clearance by controlling its efflux from brain to blood . 1.3.5 Animal models With the multiple genetic mutations associated with the APP gene, many transgenic mouse models have been created to mimic the diverse pathological features of AD. To date, there are approximately 20 APP transgenic mice made, some of which are 22 described in detail below (Table 1). The first generation of A D model mice involved expression of the wild-type human A P P (hAPP) complementary deoxyribonucleic acid ( cDNA) 7 4 . These mice, with the exception of the N S E A P P mouse, did not develop amyloid deposits. In the N S E A P P mouse the h A P P 7 5 i isoform is expressed under control of the neuron-specific enolase promoter. These mice develop amyloid pathology; however, the deposits in the brain were diffuse and did not resemble the compact plaques seen in the brains of A D patients 7 5 . The next generation of mice focused on overexpressing the mutated h A P P protein at levels well above the endogenous gene level. This required vectors that provide transgene expression specific to the C N S . Examples of these promoters include the Thy-1 promoter, which exhibits neuron specific gene expression 7 6 , and the hamster prion promoter, which is largely C N S specific with 77 expression in peripheral organs such as the heart . More recently, transgenic mice have 7 X been generated expressing the complete h A P P by using a yeast artificial chromosome . Second generation A D mice include P D A P P , Tg2576 and APP23 . The P D A P P transgenic mouse expresses all isoforms of the h A P P gene and contains the Indiana mutation, V717F, under control of the platelet-derived growth factor P-chain mini-promoter 7 9 . These mice exhibit many of the pathological features of A D including neuritic plaques in the hippocampus and neocortical regions, dystrophic neurites, astro-and microgliosis and synaptic degeneration. However, N F T s and clear 79 80 neurodegeneration have not been identified in these animals * . Radial arm maze testing and Morris water maze testing showed that these mice also exhibit memory impairment as early as 3 months of age, in the pre-plaque stage, and that these 23 81 82 impairments were correlated with increased age and plaque load ' . The pathological features observed in these animals are evident as early as 8 months of age 8 3 . A second transgenic model, which is one of the most widely used A D model, is the Tg2576 mouse. The Tg2576 mouse expresses the hAPP695 transgene containing the Swedish double mutation K 6 7 0 N / M 6 7 1 L under control of the prion promoter at levels approximately 5 fold higher than the endogenous gene 8 4 ' 8 5 . These mice develop amyloid plaques at approximately 9 months of age, with the first deposition in the entorhinal and piriform cortices 8 6 ; exhibit increased gliosis, as well as a decrease in synaptic activity in the hippocampal C A 1 region, dentate gyrus; and exhibit a reduction in long term potentiation 8 7 (see Appendix II for diagram of brain regions) . Cognitive deficits, as determined by Y-maze alternation and water maze learning, occur at 9-10 months of age 8 5 . The one shortfall of these animals is that there is a lack of neurodegeneration 8 3 . This may be due to genetic background since genetics influences susceptibility to excitotoxic cell death 8 8 . Another mouse model which expresses the Swedish mutation A P P 7 5 1 gene is the APP23 mouse. This mouse differs from the Tg2576 by expression of the gene under control of the Thy-1 promoter, resulting in a 7 fold increase in gene expression. A s early as 6 months of age, these mice develop plaques and have astro- and microgliosis, memory deficits as well as hyperphosphorylated Tau 8 9> 9 0 ' 9 1 . in contrast to other mouse models, APP23 mice exhibit a 14% decrease in the number of C A 1 neurons compared to control mice and have deposits of cerebral A p , preferentially in capillaries and arterioles, and develop cerebral amyloid angiopathy ( C A A ) 9 2 ' 9 3 . 24 In addition to transgenic mice expressing mutated h A P P , mice expressing PS1 mutations or PS-1 knock out mice have been generated. Expression of the presenilin gene did not result in the formation of plaques, but did result in an increase in the production of A P 1 4 2 3 8 • These mice were then crossed with the aforementioned mouse models to develop mouse strains that would better represent human pathology. In PS 1/hAPP mice there is an elevation in the levels of A p and most interestingly, acceleration in the rate of amyloid plaque depositions with plaque formation occurring at 6 months of age 9 4 . In addition, other transgenic mice, for example A p o E knock out, ApoE4, B A C E overexpressing, B A C E knockout mice and Tau mice have been crossed to A P P mice in the hope to provide models which accurately mimic A D and help aid in future diagnosis, treatment and prevention. To date no single mouse model faithfully demonstrates the classic triad of human A D pathology; A p containing plaques, NFTs and wide spread neuronal loss in the hippocampus and cortical areas, and as such none can be considered a complete model of the disease. 25 Tg line Transgene Mutation Age at onset Plaque Micro- and Neuronal Cognitive NFTs Ref (background) promoter of AP location Astrogliosis loss deficits deposits Tg2576 Hamster (HUAPP69S) 9-12 months HI, CC, yes Not yes AT8 84, 85 , (C57BL/6 x PrP A P P n l 1 K 6 7 0 M / N 6 7 1 L (Congo red amygdala detectable positive 9 5 , 8 3 SJL) Positive) 79 , 80, PDAPP PDGF-p APPV 7 1 7 1; 6 months HI, corp yes Not Yes AT8 (C57BL/6 x (diffuse and call, CC detectable positive 82 DBA/2 x compact) Swiss Webster) 8 9 , 9 2 TgAPP23 Murine A P P r Y 1 L K 6 7 0 M / N 6 7 1 L 6 months HI, yes Yes, N.D. AT8 (C57BL/6 x Thy-1 (congophilic neocortex -25% in positive DBA/2) and diffuse) CA1 at M I S 1 H-1 o months TgAPP22 Human A P P K 6 7 0 M / N 6 7 1 L 18 months HI yes N.D. N.D. AT8 89 (C57BL/6) Thy-1 andv7ni positive around Congo red plaques (no NFTs) 38,94 Tg2576/PSl Hamster A P P K 6 7 0 M / N 6 7 1 L 6 months HI, CC, yes N.D. Yes AT8 (PSAPP) PrP and PS1 (Congo red amygdala positive (C57BL/6 x p Positive) SJL) M I 4 6 L 96 ,97 TgAPP/Ld/2 Murine A P P V717I 13-18 HI, CC yes No overt yes AT8 (FVB/N) Thy-1 months loss positive (diffuse, mostly APi. 4 2 ) TgAPP/Sw/1 Murine A P P ^ " K . 6 7 0 M / N 6 7 1 L 18 months HI, CC yes No over N.D. AT8 98 (FVB/N) Thy-1 (diffuse, loss positive mostly Ap,_ 4 0 ) Table 1. Summary of the primary APP-based transgenic mouse models of AD Genetic parameters, onset of Ap plaque formation and summary of inflammation and cognitive deficits are described above for some of the APP transgenic mice generated. Abbreviations: HI - hippocampus, CC - cerebral cortex, NFTs - Neurofibrillary tangles, N .D. - not determined (Adapted from Janus et al. 1 0°, Gotz et al. m). 26 1.4 Microglia, inflammation and Alzheimer disease The involvement of inflammatory processes in A D pathology has been established by multiple lines of evidence. The upregulation of many inflammatory markers co-localize to regions of the brain affected by A D , in close proximity to A p deposits and 26 NFTs . There is also direct evidence of inflammatory mediated neurotoxicity as illustrated by the extensive amount of neuronal death in areas with dense microglia. Finally, many epidemiological studies have suggested that the use of non-steroidal anti-inflammatory drugs (NSAIDs) reduces the risk and delay the onset of A D 1 0 2 ' 1 0 3 . The inflammatory processes seen in A D appear to be mediated by microglia 1 0 4 . 1.4.1 Microglia as immune cells of the brain Microglia were first functionally described by del Rio-Hortega in 1932 as mesodermal pial cells that invaded the brain during embryonic development. They have migratory and phagocytic properties and exist in different resting and activated states 1 0 5 . It is estimated that microglia make up 5-15% of the brain and are considered to be permanent residents that do not move like macrophages 1 0 6 . There is still much debate as to the ontogenic origin of microglia. Initially, microglia were thought to be of mesodermal origin. Further studies confirmed this and found that microglial precursor cells are found in the yolk sac and in the mesenchymal tissue associated with the neuroepithelium. These cells enter the brain, become amoeboid microglia and spread throughout the brain 1 0 7 . Alternatively, carbon particle tracing and immunohistochemical techniques showed that ramified microglia originated from circulating blood monocytes that entered the developing brain during development and assumed the form of amoeboid microglia that ultimately evolve to become the ramified microglia 1 0 7 . Lastly, others state 27 that microglia are not derived from monocytes, rather they are derived during fetal development from neuroectodermal cell precursors; glioblasts or cells of the germinal matrix 1 0 5 . Regardless of their origin, it is agreed that microglial cells constitute the immune system of the brain. They carry out the innate immune response observed in the central nervous system during injury and infection. 1.4.2 Morphological plasticity of microglia It has been established that microglia can assume various morphological appearances that can correlate to a distinct functional state (Figure 1.3). Studies performed in the facial nerve system first demonstrated this phenomenon by showing that motor neuron regeneration occurred after injury as a result of a microglial response. Microglia appeared at the site of injury within minutes in response to neuronal signals, became activated and assumed a highly ramified morphology, proliferated and surrounded the injured neuron. During the degenerative phase of the response, microglia became rounded and phagocytosed the dying neuron 1 0 8 . In the normal adult brain, microglia exist in a resting state and also in a highly ramified state. Amoeboid microglia are also seen but these are present predominantly in the developing brain and can transform into ramified cells with age 1 0 7 . Resting microglia are small with flattened or angular nuclei and have been shown to be non-dividing. Once activated, microglia take on a different shape that is dependent on environment and brain architecture 1 0 5 . 28 Hyper-ramified microglia (itiicrmediate stage) Figure 1.3. Morphological and functional plasticity of microglia Microglia can assume different types of morphological shapes in response to environmental stimuli. Signals from damaged or dying neurons cause resting microglia to become active. The amount of signal dictates what conformational state the microglia adopt. In a resting state, microglia have small cell bodies and thin ramified processes with few branches. Intermediate signals from mildly injured neurons cause a hyper-ramified state but in'most cases microglia become reactive in response to stimuli. This activation causes the cell to adopt a larger cell body and thicker more highly branched processes. In some instances microglia can become amoeboid in shape and take on a phagocytic role, removing dying cells from the brain (Adapted from Streit et al. 104). When activated, microglia adopt a reactive conformation with ramified thick processes and abundant cytoplasm. These microglia are usually found in the gray matter of the brain. The hyper-ramified state represents an intermediate stage between the resting and reactive form and signifies the beginning of microglial activation 1 0 4 ' 1 0 5 . Microglia have also been seen as rod-shaped with elongated bodies and thin ramified processes. These can be found in the brain white matter as well as in the aged brains of normal individual. In this instance, aged microglia lose contact inhibition and fuse with one another to create rod cells 1 0 4 1 0 5 . Finally, upon neuronal cell death, microglia morph into brain macrophages, with even more cytoplasm than the reactive state and short thick 29 processes. It is this conformation that microglia are able to phagocytose and degrade cell debris 1 0 5 . 1.4.3 The function of microglia in the central nervous system Microglia are a highly responsive population of cells with the potential to engage in various functions. These include response to invading pathogens and injury, elimination of dying cells and debris, initiation in inflammatory processes as well as regulation of homeostatic mechanisms and neurodevelopment. When microglia respond to neuronal stress signals ( A T P , neuropeptides and neurotransmitters) they become activated, release several secretory proteins such as complement (classical and alternative pathway proteins), proteinases, cytokines, chemokines, excitotoxins, reactive nitrogen and oxygen intermediates, and have altered gene expression 2 6 - 1 0 9 . Microglia can also function as antigen presenting cells, albeit not as well as peripheral macrophage. Microglia residing in the C N S rarely encounter T cells and as a result, may not need to present antigen efficiently 1 0 9 . Microglial activation in response to acute C N S injury is short lived and is a beneficial process. Within 24 hours of insult, microglia arrive to the area and upregulate surface protein expression. Proliferation occurs within 2-3 days and immunophenotypic changes are evident by day 7 m . Microglia maintains a very close physical association with injured neurons in order to facilitate targeted delivery of growth factors and to displace afferent synapses thereby aiding in neuronal regeneration. In non-reversible injury, such as ischemic damage, reactive microglia secrete neurotoxic factors that aid in kill ing the damaged neuron preparing it for removal and degradation by phagocytic microglia 1 0 4 . 30 In circumstances that engage chronic activation, microglia may exert detrimental effects by transforming into autoaggressive effector cells that attack healthy cells, also known as bystander damage. Chronic C N S activation, as in A D , is primarily caused by bystander damage and is characterized by slow, progressive neurodegeneration that takes decades to develop. Exposure of cells to the various proteases and toxic molecules, produced by activated microglia for a long period of time causes extensive neuronal damage and death. Moreover, many of the pro-inflammatory cytokines and chemokines that are secreted contribute to a positive feedback mechanism, causing further attraction and activation of more microglia. These same factors have also been implicated in initiating signal transduction cascades on both microglia and neuronal cells that results in the transcription of I L - l p , transforming growth factor-P ( T G F - P ) and more cytotoxic agents via the nuclear factor K B ( N F K B ) transcription factor 1 0 5 . Finally, many of the inflammatory products secreted by microglia can alter the processing of A P P in neurons to favour the production of AP1 -42 , thereby resulting in more plaque formation, microglial activation and neuronal death . 1.4.4 The role of activated microglia in Alzheimer disease Many reports imply that microglia may exacerbate the pathological states in neurodegenerative diseases. Persistent activation of microglia is an accepted hallmark of A D and a substantial amount of the neuronal damage observed in A D may be due to the inflammatory response mediated by microglia. A p , N F T s and degenerating neurons are most likely the stimulants for inflammation in A D . A s in the periphery, there is not just one system of inflammation that is involved in a response. In the A D brain there is substantial evidence for the multiple interactions of complement pathways, cytokine and 31 chemokine pathways, cyclo-oxygenase ( C O X ) and acute phase proteins. In vitro experiments with A p aggregates, in vivo experiments in A D mice and A p infused animals, and in situ studies on post-mortem A D brains have all demonstrated that A p can induce microglia to upregulate the expression of cytokines ( IL - ip , TNF-a , IL-6), I chemokines (IL-8, macrophage inflammatory protein-la and macrophage chemoattractant peptide-1) chemokine receptors (CCR3 and C C R 5 ) , complement proteins ( C l q and C 3 , C4) and major histocompatibility complex II and other genes such as melanotransferrin (p97) 1 1 0 1 1 1 heme oxygenase 1 1 1 2 and urokinase plasminogen-activating receptor " 3 . Activated microglia also release the excitotoxins quinolinic acid 1 1 4 and glutamate 1 1 5 . Since endotoxins only cause limited damage to synapses and dendrites, it is thought that they may contribute to the neuronal dysfunction rather than complete neuronal death 2 6 . Together, all of these factors contribute to the inflammatory mechanisms observed in the disease. It has been shown in-situ that microglia cluster around the sites of amyloid deposition both in humans and in A P P transgenic mice 1 1 6- l i 7> 9 0> 1 1 8 The clustering of microglia around the senile plaques can be explained due to the chemotactic signaling of a number of molecules including A p , signals from damaged/dying neurons and also the many pro-inflammatory mediators which are found in the area of the plaque. Still under debate is the relationship between microglia and the development of A P plaques. One possibility is that microglia may directly contribute to amyloidosis by participating in the synthesis and processing of A P P to A p fibrils. It has been demonstrated that cultured microglia are able to secrete A p ; however, in vivo, microglia do not appear to express A P P m R N A 2 6 . Therefore the likelihood that microglia secrete A P and contribute to A P plaque formation 32 is debatable and requires more research. Another possible mechanism relates to the potential of microglia to convert soluble AP into a more fibrillar and aggregative form. Many studies have stated that most activated microglia appear to associate around neuritic plaques rather than diffuse plaques in both mice and humans m ' 1 1 8 ' 9 0 ; however, there are some microglia associated with diffuse plaques 1 2 0 - 1 2 1 . These data suggest that microglia play a role in transforming nonfibrillar amyloid into the more insoluble fibrillar form and aid in the transformation from diffuse plaques to neuritic plaques. Finally, microglia may contribute to plaque formation by their ability to phagocytose and degrade Ap . It is proposed that there is a dynamic balance between amyloid deposition and removal in the brain. It has been established in culture, that microglia migrate to and internalize microaggregates of A p ^ , via a type A scavenger receptor, and deliver it to late endosomes and lysosomes, where it is degraded 1 2 2 ' 6 0 . This uptake 123 appears to occur rapidly; however, the degradation process occurs at a slow rate Furthermore, it has been reported that microglia found in the cortex of AD patients contain fragments of A P , most notably C-terminal fragments, inferring that the A p was phagocytosed 1 2 4 , 1 2 5 . This rapid internalization of A p microaggregates coupled with the slow degradation can lead to the accumulation of A p inside the cell. Therefore, in the case of AD, the overproduction and thus the persistence of A P may become too overwhelming for the microglia and thereby disrupting the dynamic balance between A p deposition and removal. This was demonstrated in a study by Rogers et al. where microglia cultured with A p were able to migrate to and phagocytose the A P ; however, degradation of the fibrils took between 2-4 weeks 1 2 6 . Another study found that although some degradation of A p by microglia was observed over 3 days, no further degradation 33 was observed over the next 9 days. Instead, there was a slow release of intact A P Understanding the mechanism of A p clearance by microglia is important in determining the steps of A p plaque formation. It is possible that there is a fine balance between A p processing and deposition and degradation. A n y disruption of the A p equilibrium can thereby lead to the accumulation of A p and the formation of amyloid plaques. A n alternative theory to the role of microglia in A D was offered by Streit 1 0 9 and states that microglia become senescent and/or dysfunctional with normal aging, therefore their ability to support neurons would be diminished resulting in neuronal degradation. The rate of deterioration can be affected by genetic and environmental risk factors thereby causing some individuals to develop A D . Some indication of dysfunction may be in the structural abnormalities seen in microglia as they age and in A D . Microglia which cluster around plaques have bulbous swellings in their cytoplasmic processes, called spheroids, as well as the formation of long stringy processes 1 0 9 . However, there is still little experimental evidence of microglial dysfunction in A D . 1.4.5 Signal transduction pathways and microglial activation Many in vitro studies have demonstrated that the exposure of microglia to fibrillar A p results in the activation of complex kinase and phosphatase signal transduction pathways that lead to the activation of various transcription factors involved in inflammation. These include: the S T A T family members, peroxisome proliferator-activated receptor y (PPAR-y) , c-fos, and c-jun; N F k B ; and members of the C / E B P family of transcription factors 5 8> 1 2 8- 1 2 9> 1 3 0 Subsequent immunohistochemistry studies in A D brains of both human and animal models have revealed the upregulation of many 34 signal transduction proteins in microglia and neurons that are in close proximity to the plaque 1 3 1 . One of the most important kinase families involved in the inflammatory response and consequent neuronal damage/death is the M A P K family. The complex molecular interactions between M A P K s and proteins associated with the evolution of A D form a cornerstone in the knowledge of a still burgeoning field. The M A P K . family comprises four subfamilies: 1) E R K 1 / 2 (p44/42 M A P K ) ; 2) c-jun N-terminal kinases ( INK); 3) p38 M A P K ; and 4) E R K 5 / b i g M A P kinase 1 1 3 0 . Activation and subsequent phosphorylation of E R K 1 / 2 , I N K and p38 have been detected in A D . In the brain, the ERK1/2 kinase pathway has been implicated in eliciting stress responses, including oxidative stress, and in the regulation of intracellular calcium levels l 3 ° . J N K pathways have also been shown to respond to cell stresses and in addition, evidence has confirmed that the phosphorylation of J N K leads to the activation of death domain receptors by T N F - a 1 3 2 . The p38 M A P K family is activated by inflammatory molecules and is involved in cell cycle, cell growth and cell differentiation I 3 0 . In A D mice, phosphorylated p38 levels are increased approximately 3 fold in plaque-associated microglia. Many cell culture experiments have demonstrated that exposure of microglia and co-cultures of microglia and neurons to both oligomer and fibrillar A p leads to the activation of p38, E R K and J N K in microglia 5 7 1 3 3 ' 1 2 8 . Moreover, M A P K s have been found to be integral to the induction of I L - i p , T N F - a and reactive nitrogen intermediates from L P S and A p treated cells 5 7 , 1 3 3 . The molecular mechanisms by which M A P K signaling contribute to glial mediated responses in A D are not fully elucidated. It is thought that the different M A P K pathways are activated in response to different mediators, leading to a differential 35 response by the microglia. Therefore, delineation of the downstream targets of these signaling cascades is also critical in order to further unravel their role in A D as well as in creating potential therapeutics. 1.5 The blood-brain barrier and Alzheimer disease 1.5.1 The blood-brain barrier, structure and function The blood-brain barrier ( B B B ) , found in all vertebrates, prevents the free diffusion of circulating molecules and cells into the brain interstitial space. The barrier is formed by the presence of epithelial-like, high resistance tight junctions that fuse brain capillary endothelia together into a continuous cellular layer separating blood and brain interstitial space. There are fine structural differences from the endothelium in the brain compared to that of other organs. These include tight junctions between adjacent endothelial cells, paucity of pinocytotic vesicles and lack of fenestrations 1 3 4 . The endothelial cells of the B B B have high mitochondrial content. It is thought that the energy generated in the mitochondria provide energy to the cell in order to create and sustain high-resistant tight junctions 1 3 4 . The junctions that comprise the B B B are made up of adherens junctions and tight junctions. Adherens junctions are made up of cadherins which form adhesive contacts between cells by binding to cadherins on the surface of neighboring cells. Inside the cell, cadherins bind to the actin cytoskeleton via proteins such as catenins 1 3 4 . Tight junctions consist of three integral membrane proteins, claudin, occludin and junction adhesion molecules. There are also many cytoplasmic accessory proteins that link tight junctions to the cytoskeleton of the cell. Claudins are 36 the major proteins found in tight junctions and bind to other claudins on adjacent cells to create the primary junction. The carboxy terminus of claudins also associates with cytoplasmic proteins, such as zonula occudens proteins (ZO) which, in turn, bind to actin. Occludins are also found with claudins in the cell membrane. These proteins also bind to Z O proteins and are implicated in the regulation of cell permeability 1 3 5 . The B B B is also comprised of a basement membrane, astrocyte foot processes that surround the vessel, and pericytes embedded within the basement membrane (Figure 1.4) l 3 4 . The basement membrane is a heterogeneous mixture of proteins that are secreted from the endothelial cells and glial cells and include laminin, fibronectin, collagen type IV and proteoglycans. The presence of the basement membrane in the B B B is to provide elasticity to the vessels as well as to aid in the selective transport of highly charged molecules l 3 6 . The contribution of astrocytes to the B B B appears to be that of influencing the morphogenesis and organization of the endothelial cells that make up the vessel wall . Studies involving the transplantation of cultured astrocytes into areas of the brain with leaky vessels have demonstrated that astrocytes induced the tightening of the junctions between the endothelial cells 1 3 7 . This is thought to be mediated by astrocyte derived soluble factors, such as TGF-P and IL-6, and may require direct contact between the astrocytes and the endothelium 1 3 8> 1 3 4 . in contrast, it has been shown that upon glial cell activation, many cytokines and chemokines, such as T N F - a and IL-8, are produced resulting in an increase in vascular permeability and lymphocyte entry into the brain 1 3 6 . Pericytes wrap around endothelial cells and appear to play a role in structural integrity, formation of tight junctions and the differentiation of the intact vessel as well as in angiogenesis. Some studies have shown that pericyte loss can result in microaneurysm formation 1 3 9 1 4 0 37 However, little is known how these processes are mediated since few studies on pericyte function have been conducted 1 3 4 . Figure 1.4. Schematic of the BBB. The B B B is created by the tight apposition of endothelial cells lining blood vessels in the brain, forming a barrier between the circulation and the brain parenchyma. The B B B is permeable to small and lipophilic molecules. Larger and hydrophilic molecules can only be transported across via an active transport system. Blood-borne immune cells (lymphocytes, monocytes and neutrophils) cannot penetrate this barrier. A thin basement membrane, comprised of laminin, fibronectin and other proteins, surrounds the endothelial cells, associated astrocytes and pericytes, providing both mechanical support and a barrier function The fluid of the central nervous system differs in composition from the non-neural extracellular fluid due to the selective permeability of the B B B 1 4 1 . The functional events that define the B B B occur at the capillary level. The B B B , under normal physiological conditions, regulates the passage of peptides, proteins and other molecules between the 38 periphery and the brain. The physiochemical properties of the penetrating substance or solute, determine whether it can be transported across the B B B . Hydrophobic solutes can penetrate across the B B B while hydrophilic substances are preferentially transported across the B B B by specific membrane carrier proteins for which they have a high affinity 1 4 2 . In contrast, most foreign substances cannot penetrate the B B B since they are not recognized by the carrier systems. The barrier works in both directions. In addition to keeping unwanted substances out, the B B B retains brain synthesized compounds, such as neurotransmitters, in the brain. The cerebral endothelia forms tighter junctions than other endothelia and contains specific proteins, which are specifically expressed in the brain. These include alkaline phosphatase, y-glutamyl transpeptidase, the glucose transporter (Glut-1) 1 4 3 , and the molecules involved in metal transport such as the transferrin receptor (TfR) and p97 1 4 4 . 1.5.2 BBB integrity and Alzheimer disease The integrity of the B B B is an area of great contention in A D research. Studies comparing the vasculature of A D patients to controls have conflicting results. Upon analysis of vasculature from post-mortem A D brains there was a decreased mitochondrial content and increased pinocytotic vesicles as compared to values obtained previously in endothelium from multiple sclerosis patients and in A D animal models. Other findings such as accumulation of collagen in vascular basement membranes and focal necrotic changes in endothelial cells were present in A D patients indicating B B B disruption 1 4 5 . In contrast, Caserta et al. found that there was no difference in B B B integrity between A D cases and controls as indicated by blood-to-brain transport (KI) and tissue to blood 39 efflux (k2) of meglumine iothalamate, a radiopaque medium that binds to plasma proteins and is commonly used for angiography and venography 1 4 6 . In animals models the role of the B B B integrity in A D is also debated. Increased B B B permeability for some proteins, such as insulin, is evident and suggests some structural alterations at the B B B , however, these alterations do not support the concept of extensive B B B damage l 4 7 . What's more, it appears that leukocytes are able to cross an intact B B B without causing concomitant B B B breakdown . Alternatively, it has recently been established that the B B B integrity is compromised in h A P P overexpressing A D mice compared to controls and that the breach in B B B integrity precedes plaque formation 1 4 9 . Many studies have indicated that A p peptides promote pro-apoptotic and pro-angiogenic signals in endothelial cells 1 5 0> l 5 1> 1 5 2 ? a n ( i that systemic-derived inflammation, either triggered by A P or neuronal signals, causes B B B tight junction disruption and increased paracellular permeability 1 5 3>1 5 4>1 5 5 In addition to the neurodegeneration observed in A D , there are also considerable cerebrovascular abnormalities. It is becoming evident that the relationship between A D and C A A is growing more complex and it is harder to distinguish one disease being distinct from the other. It is estimated that the prevalence of C A A in A D patients varies from 70 to 100% 1 5 6> , 5 7. 1 5 8> 1 5 9. C A A is characterized by the deposition of amyloid, primarily A p j 3 9 and A P M 0 , in the cerebral vessel wall 1 6 0 . Since A D and vascular disease share common risk factors and since a history of strokes may be a risk factor for A D 1 6 1 , it is important to understand the relationship between dementia, neurodegeneration and cerebrovascular abnormalities. It is possible that this relationship is attributed to alterations in the B B B . Furthermore, as evidenced by the first phase of 40 human clinical trials of the amyloid vaccine where cerebral complications ensued 1 6 2 ' 1 6 3 ' there is still much to learn about the relationship between neurodegeneration, dementia and cerebral abnormalities induced by A p . 1.6 Therapeutic strategies The treatment of A D up until now has focused on the cholinergic system and symptoms thereof. Cholinergic therapeutics, such as Aricept, an acetylcholinesterase inhibitor, and psychotrophic drugs have had relative success in stabilizing the cognitive decline and behavioral deficits attributed to A D . Nevertheless, these therapies only treat the symptoms and do not delay or prevent disease progression. Based on the widely accepted "amyloid cascade hypothesis", research is now centered on the generation of anti-amyloid therapies. Therapeutic strategies directed at lowering A p levels and decreasing levels of toxic A p aggregates through: (1) inhibition of A P P processing to A P , (2) inhibition, reversal or clearance of A P aggregation, (3) cholesterol reduction and (4) A p immunization are under development A s a target to alter A P metabolism, inhibition of both P- and y-secretases are being pursued. Treatment of mice with y-secretase inhibitors has resulted in a significant reduction in brain A p levels and an attenuation of A p deposition 1 6 4 ' . Inhibition of p-secretase is also under study and is thought to be a better target than y-secretase inhibitors since y-secretase cleavage is associated with the cleavage of other proteins, such as N O T C H , which is important for development. Studies on B A C E 1 ~'~ mice demonstrate an absence in brain A p and no other obvious non-AD related pathologies 1 6 5 . Moreover, in B A C E 1 7 7 A P P mice there is an absence of amyloid plaques, microgliosis and dystrophic 41 neuritis in the brain. A t present, inhibitors for y-secretase are in phase 1 clinical trials; however, more research is needed before such inhibitors are readily available for treatment. In vivo, in vitro and epidemiological studies suggest that cholesterol may play a role in the generation of amyloid and its subsequent accumulation by its role in lipid metabolism. Retrospective analysis of people taking (3-hydroxy-P-methylglutaryl-coenzyme A reductase inhibitors, or statins, show that there was up to a 70% reduced risk for developing A D , whereas individuals with high cholesterol levels had a higher risk for A D 1 6 6 , 1 6 7 . In vitro studies on primary neuronal cultures have shown that membrane cholesterol affects the cleavage sites for both P- and y-secretase, favouring the production of APi_42, and that the removal of membrane cholesterol caused a decrease in A p production 1 6 8 1 6 9 . Other studies, in which A P P mice were fed high cholesterol diets, found that there was a decrease in s A P P a and an increased the A P P cytosolic fragment A I C D and plaque burden in high cholesterol diet mice compared to mice on a control diet 1 7 0 ' 1 7 1 . Furthermore, feeding mice with a statin called atorvostation, commercially known as Lipitor, resulted in a 50% decrease in plaque load and an increase in the fonnation of s A P P a in A P P transgenic mice 1 7 2 . Similar results are seen in humans, where statin use resulted in an approximate 40% decrease in A p serum levels 1 7 3 . Taken together, these data suggest that cholesterol modulating drugs may have a significant clinical benefit to the treatment and prevention of A D . Over the past few years the role of inflammatory mechanisms in A D pathogenesis is becoming more evident. N S A I D s , such as ibuprofen (Ibu), indomethacin and nimesulide (Nim), have been used as therapeutics for the treatment of A D . These drugs 42 have been shown to have anti-inflammatory properties by reducing the enzymatic activity Of of C O X enzymes and thus the production of prostaglandins . These drugs have also been demonstrated to have the capability of altering A ( 3 M 2 production, presumably by acting on the y-secretase complex and shifting cleavage towards the shorter less toxic forms of A P 1 7 4 . Treatment of neuroblastoma cells with the above N S A I D s resulted in the stimulation of a-secretase and secretion of s A P P a 3 3 . Recent clinical studies with N S A I D s are inconclusive due to small sample sizes and a large drop out rates. Thus, while the exact targets of anti-inflammatories remain to be clarified, new clinical studies, with larger cohorts and more extensive research into new drugs are ongoing and w i l l hopefully illuminate the practice of anti-inflammatory therapeutics for A D . The most recently developed an t i -AP therapy is aimed at the reversal and/or clearance of A P aggregates and employs immunization with either A P peptides or anti-A P antibodies. A P immunization appears to be effective in reducing amyloid deposition and plaque formation, neuritic dystrophy, astro- and microgliosis, memory and cognitive 175 176 177 178 179 180 181 deficits and early Tau pathology " J >" D >'">"°- H q w i m m u n i z a t i o n reverses or attenuates disease pathology remains to be elucidated. It is thought that A p or anti-Ap antibodies act by eliciting one or more of the following: 1) preventing fibril formation, 2) blocking the toxic effects of soluble and aggregated Ap , 3) disrupting A p fibrils, and 4) enhancing the clearance of A P by microglia 1 6 4 . Cl inical trials of active A p peptide immunization were undertaken but were suddenly halted in phase II when 5% of the patients exhibited meningio-encephalitis. New administration techniques, such as type of adjuvant and dose of A p peptide/anti-Ap antibody, have been explored and currently new clinical trials, involving both active and passive immunization, are being pursued. 43 1.7 Project rationale and general approach The focus of this thesis was to add to the basic understanding of the relationship between microglia and amyloid beta and to the various aspects of A D pathology. Both in vitro and in vivo experiments were performed in order to examine the role of microglial gene expression in response to Ap stimulation as well as the role of microglia in plaque formation. It was hypothesized that in response to general activation, microglia upregulate the expression of specific genes, in particular, p97, and that these changes in gene/protein expression may be used as a means to determine anti-inflammatory drug efficacy. Moreover, it was hypothesized that activated microglia exacerbate the pathology of A D and contribute to plaque formation. Finally, it was hypothesized that the damage that activated microglia and Ap incur on the B B B can be reversed by Ap immunotherapy. Taken together, the studies presented here provide new ways by which A D progression and treatment can be monitored, as well as providing a new therapeutic target for A D treatments and disease prevention. 44 Chapter 2: Materials and Methods 2.1 Mice 2.1.1 Tg2576 AD model mice The transgenic mice used in this study are the Tg2576 A D model mouse (Tg/+), which expresses the Swedish mutant of the amyloid precursor protein (K670N/M671L) 8 4 under control of the hamster prion protein promoter. Mice were maintained by mating Tg2576 males to C57B6/SJL FI females. This strategy was undertaken since Tg2576 mice are heterozygous for the h A P P gene. Tg2576 mice were distinguished from control littermates by P C R at 2 weeks of age as previously described 8 4 , 8 5 . Briefly, primers specific to the Hamster prion promoter and human A P P were used along with an internal control (1501, 1502, 1503; Table 2) to determine i f mice contains the human A P P gene. Animals were fed lab chow and water ad libitum and kept under a 12 hr light/dark cycle. Mice were group housed where possible, although the occasional male mouse had to be housed alone due to aggression. Wi ld type (+/+) littermates were used as controls. For the vaccination studies described below, 6 week and 11 month old mice were used. 2.1.2 Colony stimulating factor-1 deficient mice (op/op) Colony stimulating factor -1 (CSF-1) deficient mice, referred to as op/op mice, are osteopetrotic model mice. These mice have a spontaneous mutation resulting in a base pair insertion approximately 280 base pairs into the coding region of the CSF-1 gene generating a stop codon and a nonfunctional CSF-1 I82>183>184_ CSF-1 is a major growth 1 85 factor for macrophages in vivo controlling survival, proliferation and differentiation 45 The osteopetrosis phenotype is characterized by the lack of osteoclasts, thus impairing bone remodeling accompanied by retarded skeletal growth, excessive accumulation of bone, and the absence of incisors. In addition there is an absence of monocyte derived macrophage. Mature macrophages are produced from other precursor cells by the influence of granulocytes and macrophage colony stimulating factor. There is also a reduction in the number and an alteration in the morphology and function of microglia in the brain 1 8 6 ' 1 8 7 . These animals also have a reduced viability and poor breeding performance, op/op mice were fed with powdered chow in infant milk formula and water ad libitum and kept under a 12 hr light/dark cycle. Homozygous op/op mice were distinguished phenotypically from normal littermates by the absence of incisors and by a domed skull at day 10 1 8 8 . 2.1.3 Generation of Tg/+;op/op mice • . ' Tg/+;op/op mice were first generated by crossing female Tg2576 A D model mouse, described above, to male op/op mice. The Tg/+;op/+ offspring were interbred to produce double mutant Tg/+;op/op mice. Littermates Tg/+;+/+, +/+;op/op, +/+;+/+ served as controls. A l l procedures were conducted with approval by the University o f British Columbia Animal ethics committee. 2.2 Preparation of reagents The amyloid beta peptides ( A p M 0 and A p 4 0 }) used in the present study were synthetic peptides of human A p (Bachem, Torrance, C A ) . The peptides were dissolved in double distilled water at 100 p M concentration and incubated at 37°C with 5% C 0 2 46 for 5 days t o allow for fibrils-to form. The final concentration of Ap for cell stimulation was 10 uM. This concentration was chosen as a result of a dose response experiment which showed optimal activation at 10 uM Ap. Flourescent-labelled A p M Q (Calbiochem, La Jolla, CA), used for BBB studies, was resuspended in distilled water and used immediately after reconstitution. LPS from Salmonella typhimurium and mouse IFN-y (Sigma-Aldrich Canada Ltd, Oakville, ON) was resuspended in sterile phosphate buffered saline (PBS) as 1 pg/ml stocks. For NSAID studies, Ibu and Nim (Sigma) were dissolved in dimethylsulfoxide (DMSO) at a ImM concentration. Inhibitors of MAPK pathways, SB203580 (SB; p38 MAPK inhibitor) and PD98059 (PD; MEK1/2 inhibitor) (Calbiochem, La Jolla, CA) were made as 20 mM stocks in DMSO. All chemicals were of analytical grade. All primers used in this study were synthesized by Sigma Genosis and are listed in Table 2. Primer name Primer sequence Product size (bp) 1501 (mouse APP) 5'-AAGCGGCCAAAGCCTGGAGGGTGGAACA-3' -760 1502 (PrP promoter) 5'-GTGGATAACCCCTCCCCCAGCCTAGACCA-3' 1503 (hAPP) 5 '-CTG ACC ACTCG ACC AGGTTCTGGGT-3' -420 P97+1 5 '-G AGGGTG ACTTTTTGGCTACT-3' -450 P97-1 5'-AACGGA AGGCTCC ACTGAGC-3' SI5 sense 5'-TTCCGCAAGTTCACCTACC-3' -360 SI5 antisense 5'-CGGGCCGGCCATGCTTTACG-3' P1800 FWD 5 '-GGCACGGGTAGTAGTAGG GAA-3' - 1800 P1800 REV 5'-GGCAACGTTGGGTTGGCT-3' Table 2. List of primer sequences and product Primers were custom synthesized by Sigma genosis, made up in distilled water at a concentration of 20 pM, aliquoted and kept at -20°C. 47 2.3 Cell culture The murine microglial cell line, B V - 2 , provided by Dr. Michael McKinney (Mayo Clinic , Jacksonville, F L ) , was derived from vraf/v-myc-transfected primary murine microglia. These cells exhibit morphological and functional features similar to primary microglia, such as cytokine secretion and phagocytosis 1 8 9 . It has been reported that phenotype changes can occur in B V - 2 cells after multiple passages 1 9 0 . To avoid this, all cells used were passaged no more than 3 times. B V - 2 cells were cultured in Dulbecco's modified Eagle's medium ( D M E M ) (Invitrogen Life Technologies, Burlington, O N ) supplemented with 10% (v/v) fetal bovine serum (FBS) (Medicrop, Montreal, PQ.), 2 m M L-glutamine (Sigma) and 20 m M Hepes (Sigma). J B / M S , a murine melanoma cell line which expresses high levels of p97, were provided by Dr. Vincent J. Hearing (National Institute of Health, Bathesda, M D ) and bEnd.3, a murine brain endothelial cell line, obtained from American Tissue Type Culture Collection ( A T C C ) , were maintained in D M E M supplemented with 10% (v/v) F B S , 2 m M glutamine and 20 m M Hepes. 3T3, a murine fibroblast cell line was obtained from A T C C and maintained in D M E M supplemented with 10 % (v/v) bovine calf serum, 4 m M L-glutamine and 20 m M Hepes. A l l cells were cultured at 37°C in a 5% CO2 humidified incubator. 2.4 Cell stimulation A l l cell lines described above were maintained and prepared the same way. Twenty-four hours before treatment, cells were washed with P B S , plated at a density of 5x10 5 cells/ml in serum free D M E M and incubated overnight at 37°C. For determining 48 the effect of microglial activation on p97 expression, the cells were then washed twice with serum free D M E M and treated with either 10 u M A ( 3 M 0 , 50 ng/ml L P S , 5 ng/ml IFN-y, 10 u M A | 3 4 0 } or P B S for 24 hours. The supernatant was harvested and stored at -80°C for T N F - a E L I S A . The cells were then washed with ice cold PBS and R N A was extracted for R T - P C R as described below. For N S A I D studies, cells were treated with 10 u M A p M Q or 50 ng/ml LPS in the presence or absence of 10 u M drug or left untreated. The N S A I D s used in this study were Ibu, a non-selective inhibitor of the C O X enzyme family, and Nim, a C O X - 2 specific inhibitor. After 24 hours of treatment, the supernatant was harvested and stored at -80°C for T N F - a E L I S A and the cells were washed with ice cold PBS . m R N A was then extracted for R T - P C R or cell lysates were harvested for Western blot analysis. 2.5 Creation of stable BV-2 transfectant cell lines The Balb/c mouse brain genomic D N A was used to clone the promoter region of mouse p97. The p97 promoter, spanning approximately 1800bp from nucleotides -1641 to +59 (nucleotides numbered according to GenBank accession N M 013900 from the National Centre for Biotechnology Information database), was cloned by P C R using the sense p i800 F W D primer and antisense p i800 R E V primer (see Table 2 for primer sequences). The P C R product was subcloned into pCR2.1 -TOPO vector (Invitrogen Life Technologies Burlington, ON) and sequenced using standard m l 3 R (5'-C A G G A A A C A G C T A T G A C C - 3 ' ) and T7 primers ( 5 ' - T A A T A C G A C T C A C T A T A G G G -3') at the N A P S Facility (Nucleic acid-Protein Service Unit, University of British 49 Columbia, Vancouver, B C ) . After digestion with A p a l and HindiII (New England BioLabs), the p97 promoter was subcloned in-frame into the pEGFP -1 expression vector (Figure 2.1) (Clonetech, Palo Alto , C A ) . pEGFP -1 was chosen as a reporter system, since it lacks its own promoter and produces green fluorescence upon E G F P gene expression initiated by a ligated promoter. The resulting p l800-GFP construct was sequenced with p i 800 F W D and E G F P - N sequencing primer (5'-C G T C G C C G T C C A G C T C G A C C A G - 3 ' ) and digested with Apa I and Hind III to ensure correct D N A sequence and orientation (Figure 2.2). 50 M C S fco0109 I Sful TAGCGCTACCG G AC TCAGATCTC GAGCTC AAG CTT CG A ATT CTC CAG TCG ACG GTA CCG CGG GcC CBS GAT CCA CCG GTC GCC ACC ATS GTS teMU Be/tl JfA-Ti V U l fool I M l fell i f I "1 *\"~§»™H I V I Stc\ Aec\ 4io7t* I \ Si»139 I Xmi I Figure 2 . 1 . p E G F P - 1 vector and multiple cloning site This vector encodes a red-shifted variant of wild-type green fluorescent protein G F P which has been optimized for brighter fluorescence and higher expression in mammalian cells. pEGFP-1 encodes the GFPmut l variant which contains the double-amino-acid substitution of Phe-64 to Leu and Ser-65 to Thr. pEGFP-1 is a promoterless E G F P vector which can be used to monitor transcription from different promoters and promoter/enhancer combinations inserted into the M C S located upstream of the E G F P coding sequence. The coding sequence of the E G F P gene contains more than 190 silent base changes which correspond to human codon-usage preferences. Sequences flanking E G F P have been converted to a Kozak consensus translation initiation site to further increase the translation efficiency in eukaryotic cells. SV40 polyadenylation signals downstream of the E G F P gene direct proper processing of the 3' end of the E G F P m R N A . A neomycin-resistance cassette (Neo 1) allows stably transfected cells to be selected using G418. A bacterial promoter upstream of this cassette confers kanamycin resistance in E. coli. (Adapted from Clontec). 51 Figure 2.2. Gel of digested p E G F P and p97 promoter construct. Each construct, p E G F P - p l 8 0 0 and p E G P F , were digested with Apa l and H i n d l l l for 3 hours at 37°C. The digested products were analyzed on a 1% agarose gel and visualized. The 4200 bp fragment indicates the pEGFP-1 vector while the smaller 1800 bp product indicates the promoter region of p97. B V - 2 cells, at approximately 70% confluency, were transfected with the 5 pg of the p l800-GFP construct or the promoter-less pEGFP-1 vector by using Lipofectamine™ 2000 reagent (Life Technologies, Inc., Gaithersburg, M D ) under serum free conditions. Cells containing either the p l800-GFP construct or the promoter-less pEGFP-1 vector were selected by adding 500 pg/ml G418 (Invitrogen). The transfectants continued to grow for approximately 3 weeks with 500 pg/ml G418, and then were sorted using the F A C S D i v a ( B D Pharmingen, San Jose, C A ) into single cell clones. Clonal cell lines continued to grow until confluency under selective conditions and were then analyzed for G F P expression. Since G F P expression is under control o f the p97 promoter, cells were treated with 10 p M A P M O for 24 hours and then sorted for high G F P expressing cells. Cells transfected with pEGFP-1 vector alone were sorted for low G F P expression. Three days later, cells were sorted again for low expressing G F P to isolate cells in an inactive, 52 resting state. These cell lines continued to grow for until confluency under selective conditions and were then used for experiments described. 2.6 RNA Isolation m R N A from B V - 2 , J B / M S , 3T3 and b.End3 cells, as well as from mouse brain regions was extracted using the RNeasy mini K i t (Qiagen Inc., Mississauga, ON) according to the manufacturer's instructions. Cells were washed once in cold P B S , lysed with lysis buffer containing P-mercaptoethanol and spun down in a spin column. The cells were then washed with an appropriate buffer and R N A was isolated. R N A concentrations were determined using spectrophotometry at U V wavelengths of 260 and 280. Final R N A concentration was obtained form the following calculation O D 2 6 o X 40 = [RNA] pg/ml 2.7 Reverse transcriptase and Polymerase Chain Reaction Reverse Transcriptase and Polymerase Chain Reaction (RT-PCR) was performed with oligonucleotide primers custom synthesized by Sigma Genosis listed in Table 2. The expression of p97 and SI5 (loading control) was examined. To make the c D N A , 1 pg of R N A was used along with 1 p l of deoxyribonucleotide triphosphate (dNTPs) mix ( lOpl each of d A T P , dCTP, dGTP and dTTP) (Invitrogen), l u l of Oligo dT . (Invitrogen) and Rnase free water to a total volume of 12 p l . The mixture was incubated at 65°C for 5 minutes followed by a quick chil l on ice. Next, 4 p l of 5x first strand buffer, 2 p l o f 0 .1M dithiothreitol (DTT) and 1 p l of RNase O U T (Invitrogen) was added and incubated 53 at 42°C for 2 minutes followed by the addition of 1 p l of Superscript II (Invitrogen). The solution was mixed by gentle pipetting and incubated at 42°C for 50 minutes. After incubating, the reaction was then inactivated by heating the solution at 70°C for 15 minutes, followed by the addition of 1 p l of Rnase H (Invitrogen) and incubation at 37°C for 20 minutes. P C R amplification was performed through 35 cycles at 94°C for 30 sec, 56°C for 30 sec, and 72°C for 45 sec. The P C R was stopped with a final extension for 10 min at 72°C. Samples were then loaded into a 1% agarose gel and images were digitally captured and analyzed. 2.8 Real-time Polymerase Chain Reaction R T - P C R results were confirmed by semi-quantitative real-time P C R for p97 and for S15 m R N A (used as internal control). R N A ( lpg) from each sample was reverse transcribed using Superscript'" II Reverse Transcriptase (Invitrogen). Real-time P C R was performed using the Roche Light Cycler. In brief, 1 p l of c D N A and gene specific primers were added to SYBR® Green Taq ReadyMix™ ( S Y B R Green dye, Taq D N A polymerase, JumpStart Taq antibody, dNTP mix and optimal buffer components; S I G M A ) and subjected to P C R amplification. The following protocol was used: denaturation (95°C for 5 minutes), amplification and quantification (40 cycles at 95°C for 5 seconds, 56°C for 5 seconds and 72°C for 30 seconds), melting curve program (65-95°C with a heating rate of 0.2°C per second and a continuous fluorescent measurement) and a final cooling step to 40°C. The primers used for p97 and SI5 were synthesized by Sigma Genosis and are shown in Table 2. The amplified transcripts were quantified using 54 the comparative C T method. Briefly, fold increase was calculated by determining the cycle at which the fluorescence passes the fixed threshold (CT) for each sample. Below are formulas used to calculate the fold induction for each sample (as presented by Applied Biosystems). A C T = C T (gene of interest) - C T (endogenous control) A A C T - A C T (sample X ) - A C T (control sample) Fold Induction = 2 " A A C | 2.9 TNF-a ELISA assay T N F - a produced by B V - 2 microglial cells was determined in culture supernatants using a specific murine T N F - a E L I S A kit capable of detecting levels as low as 5.1 pg/ml ( R & D Systems, Inc., Minneapolis, M N ) according to the manufacturer's protocol. T N F -a was chosen as a measurement of microglial activation since it has been established that microglia upregulate the production and secretion of T N F - a upon activation 2 6 . After 24 hours of treatment the cell culture supernatant was removed and stored at -80°C until assayed for T N F - a . E L I S A measurements were performed using the Spectra Max 190; (Molecular Devices) using the standard and instructions supplied by the manufacturer. 2.10 Western blot analysis Antibodies The phosphorylated forms of E R K and p38 M A P K were visualized using rabbit polyclonal primary antibodies raised against phosphor-(Thr202/Tyr204) E R K (1:1000) and phosphor-(Thrl80/Thr182) p38 M A P K (1:1000), (Cell Signaling, 55 Beverly, M A ) . G F P was visualized using a mouse monoclonal raised against G F P (1:1000) (Santa Cruz Biotechnology, Santa Cruz, C A ) . Loading controls for E R K , p38 M A P K and G F P were determined by using extracellular regulated kinase 1/2 (Erkl /2) (1:1000) polyclonal antibody, extracellular regulated p38 antibody (1:1000) and G A P D H (1:10000) (Santa Cruz Biotechnology) respectively. Secondary goat anti-rabbit (1:10000) and goat anti-mouse (1:10000) antibodies conjugated to horse radish peroxidase (Jackson ImmunoResearch Laboratories, West Grove, PA) were used. B V - 2 cells transfected with the p i800 promoter region of p97 fused to G F P were washed, resuspended at 5 x 10 5 cells/ml in D M E M containing G418 and incubated overnight at 37 °C. Cells were then washes 3 times in serum free media and incubated with 10 ( i M Api-40 or 50 ng/ml LPS for 24 hrs. For signal transduction studies, cells were treated with 20 u M S B or 50 U.M P D for 30 min and then with either 10 u M Api^o or 50 ng/ml LPS for 6 hrs. Following treatment, cells were lysed in 50 p i of lysis buffer containing 50 m M Tr i s -HCl (pH 7.5), 150 m M N a C l , 10% glycerol, 1% Nonidet P-40, 5 m M E D T A , 1 m M sodium vandadate, 5 m M sodium fluoride, 1 m M sodium molybdate, 5 m M P-glycerol phosphate, in the presence of 10 pg/ml soybean trypsin inhibitor, pepstatin, and 40 pg/ml phenylmethylsulfonyl fluoride. Protein concentrations were determined using the B C A protein assay (Pierce, Rockford, IL). Equal amounts of proteins were denatured by boiling in SDS sample buffer (0.16% (w/v) SDS, 0.002% (w/v) bromophenol blue and 1% (w/v) DTT.) run on 12% S D S - P A G E gel along prestained broad range standard (Bio-rad, Hercules, C A ) and transferred to nitrocellulose membrane. Membranes were blocked for non-specific binding in 5% non-fat dry skim milk in P B S and 0.1% (v/v) Tween 20 detergent (Bio-rad). Primary antibodies to 56 phospho-p38 and phosphor-p44/42 were incubated in PBS/0 .1% Tween 20 and the G F P primary antibody was incubated in blocking buffer. Primary antibodies were incubated with membranes overnight at 4°C. After incubation with HRP-l inked secondary antibodies, proteins were detected using E C L detection reagent (Amersham Biosciences, Piscataway, NJ). After development, the blots were stripped in 62.5 m M Tr i s -HCl (pH 7.5), 0.2% SDS, and 100 m M 2-mercaptoethanol for 30 min at 50 °C and then reprobed with either a monoclonal antibody to G A P D H , extracellular regulated kinasel/2 (Erkl /2) (K-23) polyclonal antibody and extracellular regulated p38 (C-20) (Santa Cruz Biotechnology) as a protein-loading control. 2.11 Immunohistochemistry Adjacent tissue sections were immunostained for the detection of amyloid deposits (4G8) and activated microglia (F4/80). Brains were fixed in 4% paraformaldehyde, paraffin embedded and cut into 8 p M sections. Sections were then deparaffinized, hydrated and incubated in D A K O Target retrieval Solution (DakoCytomatiom) in a steamer for 20 minutes to ensure proper unmasking of the antigen. The samples were then cooled at room temperature for 30 minutes and washed in water 3 times for 5 minutes each. Slides were then incubated in 3% H 2 0 2 for 30 minutes, rinsed 3 times in water and blocked in the appropriate blocking solution for 40 minutes. Slides were then incubated in either 4G8 (1:500, Signet Labs Inc., Dedham, M A ) or F4/80 (1:10, Serotec, Oxford, U K ) overnight at 4°C, washed and incubated with a secondary biotinylated anti-mouse antibody for 4G8 or a biotinylated anti-rat antibody for F4/80 ( D A K O LSAB+system, DakoCytomation) for 25 minutes. Sections were developed by D A B 57 (Vector Laboratories Inc., Burlington, ON) , counterstained with Mayer's hematoxylin, dehydrated and mounted. Slides were examined under a Zeiss microscope and images were captured using OpenLab software. The number of plaques per cortical brain section per mouse was counted and analyzed. Data were collected from 4 equally spaced sections. The values for all sections from one mouse were averaged to obtain a single sample for statistical analysis. 2.12 Ap and antibody injection Fluorescent-labelled A | 3MO , PBS , bovine serum albumin (BSA) conjugated to Texas red (Sigma), anti-human A(3 antibody (4G8), Biotin labeled anti-human A p antibody (4G8) (Signet Labs Inc.) and biotin (Sigma) were injected intravenously (i.v.) into the tail veins of mice. For injections, a total volume of 200 ul was injected into each mouse containing 200 ug of protein. One hour after injection, the.mice were anesthetized with ketamine (25 mg/kg i.p.) and xylazine (5 mg/kg i.p.) and perfused with 1 X P B S followed by 4% P F A . Following perfusion, the brains were dissected, embedded into paraffin and sectioned into 8 u M sections. Brain sections from mice injected with fluorescent A P , P B S and Texas-red conjugated B S A were examined directly by confocal microscopy without any processing. Brain sections for mice injected with anti-Ap antibodies (both unlabelled and biotin labeled) were stained with secondary antibodies or processed with D A B respectively. Briefly, slides were deparaffinized and hydrated as described above. For mice injected with anti-Ap antibodies, slides were blocked in 10% skin milk in P B S for 30 minutes, washed 2 times for 5 minutes in water and then incubated in Alexa 488 nm conjugated goat-anti mouse IgG secondary antibody (1:500) 58 (Molecular probes, Eugene, OR) diluted in 2% B S A / P B S for one hour. The slides were then washed 3 times for 5 minutes. The coverslip was mounted with 5 pl Antifade (Molecular probes) before sealing with nail polish. For mice injected with biotin labeled antibody or biotin alone, slides were developed with D A B (Vector laboratory) washed 3 times for 5 minutes in water, dehydrated in 3 washes of xylene and mounted with a coverslip using permount. The slides were then examined using a confocal microscope or a light microscope. A l l mice used for these studies were 6 weeks old and of similar weight. 2.13 Vaccination protocol Prior to immunization each mouse was bled and serum collected. Two groups of mice were vaccinated beginning at 6 weeks and 11 months of age and sacrificed at 12 months and 15 months respectively. A(3 peptide was freshly prepared from lyophilized powder for each set of injections. For immunizations, 2 mg of A(3 (human A f $ M 0 ; Bachem) was added to 0.9 ml of deionized water and mixed until a solution of uniform suspension was obtained. Then 100 p l of 10X P B S was added to obtain a final I X P B S concentration. The solution was vortexed and placed at 37°C overnight until use the next day. A P (100 pg antigen per injection) or PBS (control) was mixed 1:1 (v/v) with complete Freund's adjuvant ( C F A ) for the first immunization. This was followed by a boost with A P M Q (100 pg) or P B S mixed 1:1 (v/v) with incomplete Freund's adjuvant ( ICFA) at two weeks and monthly thereafter. The 6 week old mice were vaccinated for a total of 11 months, and the 11 month old mice were vaccinated for a total of 4 months. 59 A p or PBS alone were injected from the fifth immunization onward. Antibody titres for anti-Ap antibodies was assayed after the second immunization. Anti-AP antibody titres were determined by serial dilutions of sera against aggregated A p , which had been coated in microtitre wells. Detection was done by using goat anti-mouse immunoglobulin conjugated to horseradish peroxidase and ABTS (2'2-AZINO-bis (3-ethylbenzthiazoline-6-sulfonic acid; Sigma) as substrate. A fluorescence plate reader (Spectra Max 190; Molecular Devices) then measured fluorescence at 405 nm. 2.14 Evans blue assay Quantitative Evans blue analysis was performed as previously described by Ujiie et al. 1 4 9 In brief, Tg2576 mice vaccinated with either A P or PBS alone and their age-matched controls were weighed and injected intra-peritoneal (i.p.) with 50 u.g/g Evans blue dye (Sigma) in 1XPBS. Three hours after injection, the mice were anesthetized with ketamine (25 mg/kg i.p.) and xylazine (5 mg/kg i.p.) and perfused with 1XPBS for 5 minutes. Following perfusion, the brains were dissected, olfactory and cerebellum removed, weighed and dounce homogenized in 0.5 ml of 50% trichloroacetic acid. The sample was then centrifuged at 10000 rpm for 10 minutes. The supernatant was collected and diluted 1:4 in 100% ethanol. An EL1SA plate reader (Spectra Max 190; Molecular Devices) then measured Evans blue fluorescence at 620 nm. Values are graphed as OD62o/brain weight, and the data were statistically analyzed with the student t-test. As a control for Evan blue distribution throughout the mouse, OD620 values for the liver, a 60 tissue known to be highly permeable, were also evaluated. A l l mice used in this study exhibited high levels of Evans blue dye in the liver. 2.15 Statistical analysis A l l analyses were performed using the GraphPad Prism software. T N F - a E L I S A data and Real-time P C R data were analyzed by performing A N O V A tests followed by Bonferroni analysis. A l l experiments were performed at least 3 times and in triplicate within each experiment. Statistical significance was established at a level of P < 0.05. A P vaccination studies were analyzed by comparing A p vaccinated transgenic mice and P B S vaccinated transgenic mice within each group by the student t-test. Transgenic and non-transgenic mice injected with A p and transgenic and non-transgenic mice injected with P B S were also compared to one another respectively. Statistical significance was established at a level of P < 0.05. 61 Chapter 3 : P 9 7 expression in activated microglia 3.1 Rationale It is now becoming widely accepted that inflammatory processes are involved in the pathogenesis of AD. There have been extensive studies over the past 10 years which have characterized the increased expression of many pro-inflammatory cytokines, cell surface markers and various neurotoxins corresponding to the accumulation of AP and disease pathology. The findings that amyloid plaques are surrounded by activated microglia also provide evidence for the contribution of inflammation in A D . These observations provide the rationale for developing therapeutics for A D targeted against inflammation and thus activated microglia. Further support of such therapies came from epidemiological evidence from arthritic patients who were on long term N S A I D s in which NSA1D use correlated with a significant decreased risk for A D by approximately 60-80% 1 0 2 . Furthermore, N S A I D treatment appears to slow down disease progression, delay onset and diminish symptom severity 1 0 3 . While it has been established that the target of N S A I D s are microglia, it is unclear what intracellular process in microglia mediate the response elicited by N S A I D s . It has been suggested that either the C O X - 1 / 2 enzymes or the transcription factor, P P A R - y , are involved ' . Finding a specific marker for inflammation resulting from Ap accumulation would aid in the diagnosis of A D as well as in evaluating the efficacy of new anti-inflammatory therapeutics. There have been many biomarkers suggested to aid in the diagnosis of A D . These are mostly limited to the ratio between Ap42 and Ap4o or total Tau or phosphorylated Tau levels in the CSF . There have also been some markers of 62 inflammation that have been investigated in AD, such as levels of 1L-6 1 9 1 and TNF 1 9 2 however; these markers are general for all forms of CNS inflammation and not specific for AD related inflammation. An alternative candidate that is believed to be specific for AD is a protein called melanotransferrin. Melanotransferrin, also known as the human melanoma associated antigen p97, was originally identified as a cell surface marker associated with human skin cancer 1 9 3 . Woodbury et al. analyzed the expression of p97 in normal adult, fetal, and neoplastic tissues and found that the protein was highly expressed in fetal colon, lung and umbilical cord as compared to adult tissues 1 9 4 . Further studies revealed that p97 is also highly expressed in fetal liver, placenta and sweat glands 1 9 3 . Due to high p97 expression in fetal tissues, it has been suggested that p97 may play a role in fetal development. In addition to its expression in human tissues, p97 has been found to be expressed in many cultures of normal cells including liver, intestinal epithelial cells and fetal intestinal cells 195,196,197,198 e x i s t s i n two forms: one is a glycosylphosphatidylinositol (GPI)-linked cell surface form and the other is a secreted soluble form generated by alternative splicing 199,200,201 j s a m e r n D e r of a group of iron binding proteins that include serum transferrin, lactoferrin and ovotransferrin and maps to the same region of chromosome 3q in humans as the genes of other proteins involved in iron transport 2 0 2 . Human transferrin and p97 share 40% sequence identity and p97 is able to bind one iron molecule 2 0 3 . Thus, p97 may be one of the many uncharacterized iron transport molecules that operate in a transferrin independent fashion. The relationship between p97 and AD remains to be elucidated in molecular detail. The distribution of p97 in the brains of AD and non-AD subjects was studied to identify 63 locations where p97 is expressed 2 0 4 , 1 1 0 . In previous studies comparing A D brains with brains of patients who died from neurodegenerative diseases other than A D , such as Huntington's disease or Parkinson's disease (PD), the distribution of p97 in the non-AD brains appeared to be limited to the endothelial cells, which make up the B B B , with occasional positive staining for astrocytes and oligodendrocytes. Interestingly, in the A D brain, p97 was also detected in reactive microglia associated with senile plaques. Reactive microglia that stained positive with H L A - D R antibody but were not closely positioned near senile plaques did not express p97 n 0 . Further immunohistochemical and in situ hybridization studies 1 1 1 revealed a high expression of p97 m R N A in the reactive microglia associated with senile plaques and lower levels of p97 around endothelial cells. In the non-AD brain, endothelial cells stained weakly for p97 m R N A and there was no p97 m R N A or protein detected in microglia. It is possible that there is an increased requirement for the utilization and/or scavenging of iron by reactive microglia associated with senile plaques since, in addition to p97, increased concentrations of iron, transferrin and ferritin ' have been noted in the region. With the finding that p97 is detected on the B B B and in the reactive microglia associated with the senile plaques of A D patients 1 1 0 it was postulated that p97 may be used as a biological marker for A D . Initial studies by Kennard et al. investigated this putative relationship and found evidence that the soluble form of p97 is elevated, up to 6 fold, in the serum taken from A D patients as compared to controls (n=17) m . Further regression analysis of these data revealed that there was no correlation between p97 levels and age. To eliminate the possibility that environmental factors may result in the increased levels observed, serum samples from A D patients and their cognitively normal 64 spouses (n=10) were obtained . Again it was found that the levels of p97 in AD patients compared to their spouses were significantly elevated thereby suggesting that environmental factors have no influence on serum p97 levels. A subsequent study by Feldman et al, involving a larger cohort of subjects, showed that there is a 2 fold significant increase in serum p97 levels from patients with AD compared to non-AD controls (41 ng/ml versus 20 ng/ml, p < 0 . 0 0 1 ) 209. These results were corroborated by an independent group of investigators 2 1 0 who reported that there was a 3 to 5 fold significant increase in the serum p97 levels in AD cases compared to non-AD cases of dementia and controls. In addition, there was no correlation between p97 serum levels and age or severity of disease. Collectively these are compelling findings which independently support the potential of serum p97 levels as a biomarker of AD. The role of p97 as an iron transporter in AD pathology is unclear. It has been established that in many neurodegenerative diseases, such as AD and PD, there is a 211 dysfunction in iron metabolism and hence homeostasis (reviewed in Qian, 1998) How exactly iron gets into the brain remains unresolved. It is thought that most iron transport in the body is mediated by the classical transferrin-transferrin receptor (TfR) pathway. However, in regards to iron transport in the brain, it is possible that other proteins may be involved. This is supported by looking at TfR, transferrin and iron distribution in the brain. There appears to very little overlap between TfR and iron distribution and between TfR and transferrin distribution 2 1 2 . In addition, there is a difference in the rate of transfer of iron and transferrin across the BBB indicating that iron transport across the BBB is likely to be facilitated by other means. It has been 65 previously established that p97 is able to bind and internalize iron into cells independent of transferrin and TfR 1 9 9 ' 2 0 0 . Furthermore, the rate of iron uptake by p97 was equivalent to the rate of iron uptake seen with transferrin/TfR. These data suggest that p97 may transport iron across the B B B in a manner similar to the TfR. Dysregulation of p97 may be one of the causes of iron deposition in A D . Studies have confirmed the presence of an iron-responsive element in the 5 'UTR of the A P P and that at the biochemical level, copper, zinc and iron are shown to accelerate the aggregation of the A p peptide and amyloid plaque formation 4 5 . Microgl ia may express p97 in response to A P as a means to help clear A p plaques by binding to iron deposited in the plaques and destabilizing it. Alternatively, p97 expression in microglia may be a harmful response. If there is an initial dysregulation of iron in the brain causing a decrease in the concentration of free iron, (for example sequestration in plaques) p97 may cross the B B B into the periphery, bind iron and cross back into the brain. Recently, it has been demonstrated that p97 may participate in the vascularization of solid tumors and promote endothelial cell migration 2 I 3 . This angiogenic activity may depend on activation of endogenous vascular endothelial growth factor ( V E G F ) expression 2 ' 3 . We have established that the B B B is compromised in A D model mice 1 4 9 and it is known that there is an increase in V E G F expression in A D 2 1 4> 2 1 5> 2 1 6 Thus p97 may contribute to the increased permeability of the B B B in A D . More studies are needed in order to determine i f an increase in brain iron levels is associated with an increase in p97 expression. Also, the distribution of p97 in different brain regions and different cells types may help elucidate the role of p97 in brain iron homeostasis and possibly A D pathology. 66 The aims of this study were to determine if microglia upregulate the expression of p97 upon general activation or whether this occurs in response to AD specific stresses, and to investigate which signal transduction pathway is responsible for this up-regulation. Moreover, if p97 could be used as a marker for inflammation, it may be possible to use it as an indicator of anti-inflammatory drug efficacy. To further investigate this possibility, p97 expression was examined in cells treated with NSAIDs. 3.2 Results 3.2.1 Microglial activation As a prerequisite for analysis of levels of p97 in activated microglia upregulate the expression of p97, it is important to confirm the state of microglia after treatment with known stimulants was confirmed. It has been well established that once microglia become activated they secrete many pro-inflammatory cytokines including IL-1 P and TNF-a, as well as reactive oxygen and nitrogen radicals 2 6 . Therefore to determine if microglia were indeed in an activated state, TNF-a levels were measured by a sandwich ELISA with antibodies specific for murine TNF-a. Cultured BV-2 cells were incubated for 24 hours with known activators of microglia including 50 ng/ml LPS, 5 ng/ml IFN- y or 10 uM fibrillar Api_4o, which is a stress specifically associated with AD. As negative controls, cells were treated with 10 U.M of the reverse Ap peptide (Ap 4 0 }) or PBS. Stimulation of BV-2 cells with A P M Q , LPS and IFN-y lead to an increase in the 67 production of T N F - a (Figure 3.1). Treatment with the A p 4 Q } had some effect on T N F - a production likely due to non-specific effects of addition of peptides to the media.. NT LPS IFN-y ApY40 A p 4 ( M PBS Treatment * P < 0.001; ** P < 0.01 Figure 3.1. T N F - a production increased in BV-2 cells treated with various known activators. Cells were plated at a density of 5 x l 0 5 cells/ml and treated for 24 hours with 50 ng/ml L P S , 5 ng/ml IFN-y, l O u M A p M 0 , 10 p M A p 4 0 - i and PBS . Cel l culture supernatant was collected and T N F - a levels were measured using an E L I S A specific for murine TNF-a . There was a significant increase in T N F - a production in cells treated with L P S , IFN-y and A p M 0 compared to no treatment (NT) ( A N O V A , * P < 0.001, ** P < 0.01). Treatment with the reverse A p peptide had some effect, likely due to non-specific effects of added peptides while PBS had no significant effect on T N F - a production. A l l experiments were performed at least 3 times in triplicate. Error bars represent ± SD. The results above are representative of all replicates. 68 3.2.2 P97 expression in BV-2 cells The expression of p97 was determined by R T - P C R in B V - 2 cells under non-stimulating conditions and following treatment with fibrillar A ( 3 M 0 , LPS and IFN-y. The results show that expression of p97 was largely increased in B V - 2 cells in response to treatment with A[3 (Figure 3.2 a). Treatment with other known activators of microglia, such as IFN-y had no significant effect on the expression of p97. There was an increase in treatment with L P S ; however, not as large a magnitude as with treatment with A p . There was no effect on the expression of SI5 , a ribosomal subunit m R N A used as a loading control. In addition, there seemed to be no change in the expression of p97 in murine melanoma cells ( JB/MS) , murine brain endothelial cells (bEnd.3), nor in murine fibroblasts (3T3) in response to these stimuli (Figure 3.2 a). Real-time P C R analysis of p97 expression in B V - 2 cells was performed in order to quantitate the change in m R N A expression. Treatment with A p resulted in an approximate 6 fold increase in p97 expression. The 6 fold increase corresponds to the increased protein serum levels 208 209 210 previously reported in humans ' * . Treatment with LPS exhibited an approximate 4 fold increase in p97 expression (Figure 3.2 b). Since there is no monoclonal antibody to murine p97, B V - 2 cells were transfected with the pEGFP -1 promoterless vector where the expression of G F P is under control of the p97 promoter (pl800-GFP). To investigate i f protein levels of murine p97 corresponded to the increase in m R N A expression, Western blot analysis was performed on p l800-GFP transfected B V - 2 cells and blotted for G F P expression. Treatment with A P and to a lesser extent LPS resulted in an increase in G F P expression compared no treatment (Figure 3.2 c). 69 Figure 3.2. P97 expression in treated cells. p97 expression in B V - 2 cells was upregulated ~6 fold in response to l O p M A p compared to no treatment and treatment with other known activators (IFN-y) of microglia. Treatment with 50 ng/ml L P S resulted in an -3.5 fold increase in p97 expression, (a-b). In addition, treatment of endothelial cells (bEnd.3), melanoma cells ( JB/MS) , and fibroblast cells (3T3) with 10 u M A p , 50 ng/ml L P S and 5 ng/ml IFN-y had no effect on p97 expression. This increase in p97 expression was also observed at the protein level in response to 10 p M A P and 50 ng/ml LPS (c). SI5 m R N A was used as a loading control for R T - P C R and G A P D H for western blot, (a) R T - P C R of p97 expression in B V - 2 , b.End3, J B / M S and 3T3 cells (b) P97 expression as determined by real-time PCR. Relative gene expression was determined by calculating the cycle threshold (CT) for p97 as well as for SI5 . Treated sample thresholds were then compared to non-treated and fold induction of gene expression determined. Error bar represent ± S D for C T values, (c) Western blot analysis of G F P expression in p l800-GFP transfected B V - 2 cells after treatment with A P and L P S . A l l experiments were performed at least three times and data indicated here are representative of all trials. 70 G F P G A P D H 3.2.3 P97 expression in Tg2576 AD model mice Finally, the m R N A expression levels of p97 in different brain regions of Tg2576 A D model mice was examined. Sixteen month old Tg2576 A D model mice, where there is a substantial presence of amyloid plaques and activated microglia, and their control littermates, were perfused with P B S and the brains were dissected into 6 different regions; frontal, hippocampal, occipital, olfactory, parietal, and temporal. The brains were then processed for total R N A extraction and subjected to R T - P C R to assess for p97 expression. Results illustrate that p97 expression is increased in the Tg2576 A D mouse in the brain regions associated with heavy plaque burden and activated microglia such as the temporal cortex and hippocampus (Figure 3.3). 72 F R HI O C O L P A T E Tg2576 SJL /B I6 Figure 3.3. p97 expression is increased in affected brain regions in Tg2576 mice. Brains were dissected into 6 regions, R N A extracted and R T - P C R performed for p97 gene expression. Areas of the brain known to have significant amyloid deposition and activated microglia, FR, HI, P A and T E , also have an increase in p97 gene expression. Non-transgenic mice exhibit no change in the expression of p97. There was also no change in S15 m R N A levels in all brain regions in both transgenic and non-transgenic mice. FR-frontal; Hl-hippocampal; OC-occipital; OL-olfactory; PA-parietal; T E -temporal. A l l experiments were performed at least three times and data indicated here are representative of all trials. 7 3 3.2.4 MAPK pathways control the expression of p97 To seek further insight in what signal transduction pathway is involved in p97 expression, the promoter region of p97 was studied to identify possible transcription factor binding sites. Within this region there are 3 AP-1 transcription factor binding sites, indicating a possible regulation by MAPK-dependent pathways. Moreover, Roze-Heusse et al. found that increased m R N A levels of Jun/Fos, the transcription factors which 217 constitute A P - 1 , correlated with m R N A levels of p97 in human melanoma cells . To directly test which M A P K pathway is involved in the expression of p97, p l800-GFP transfected B V - 2 cells were stimulated with 10 u M A(3MO or 50 ng/ml LPS with and without the addition of selective inhibitors of the p38 M A P K and E R K 1 / 2 pathways. SB203580 (SB) is a selective inhibitor of p38 M A P K and has no effect on E R K 1 / 2 or J N K while PD98059 (PD) is a selective inhibitor of upstream kinases, M E K 1/2, in the E R K pathway. Cells were incubated with each of the M A P K inhibitors for 30 minutes and then stimulated with either 10 p:M Ap^o or 50 ng/ml LPS for 6 hours: N o treatment and treatment with inhibitor alone were used as controls. G F P protein levels were measured by Western blot. Treatment of p l800-GFP B V - 2 cells with the inhibitor of p 3 8 M A P K (SB, 20 uM) resulted in a decrease in G F P expression in both A P and LPS treated samples (Figure 3.4 a). Treatment with inhibitors to E R K 1 / 2 (PD; 50 uM) had no effect on G F P expression in stimulated cells (Figure 3.4 b). Cells transfected with vector alone, pEGFP-1 , showed no change in G F P expression in all cases, but did show a difference in phospho-p38 and phospho-ERK expression upon treatment with A p or LPS with and 74 without respective inhibitors (Figure 3.4 b,d). This suggests that the p97 promoter region is regulated by the p38 M A P K pathway. NT A p L P S 20 pM S B G F P p-38 anti-p38 NT Ap L P S 50 pM PD G F P p-ERK E R K NT A P LPS 20 p M S B G F P p-38 anti-p38 NT Ap L P S 50 p M P D G F P p E R K ant i -ERK Figure 3.4. The p97 promoter appears to be regulated by the p38 M A P K pathway. B V - 2 cells transfected with p l800-GFP were stimulated with 10 p M A P i ^ 0 and 50 ng/ml L P S as well as with inhibitors for p38 (20 pM) and p - E R K (50 p M ) for 6 hours. G F P expression was measured by Western Blots of cell lysates with antibodies against G F P and the activated (phosphorylated) forms of p38 and p - E R K . Levels of extracellular regulated M A P K s were used as loading controls. G F P expression was decreased when treated with A p or L P S in conjunction with SB implying that the p38 M A P K pathway controls the regulation of the p97 promoter region. There was no change in G F P levels in cells treated with P D or in cells transfected with an empty vector, (a) G F P expression in p l800-GFP cells treated with p38 M A P K inhibitor, S B , (b) G F P expression in pEGFP-1 cells treated with p38 M A P K inhibitor, SB, (c) G F P expression in p l800-GFP cells treated with E R K 1/2 inhibitor, P D and (d) G F P expression in pEGFP-1 cells treated with E R K 1/2 inhibitor, P D . Each gel is a representative gel from three separate experiments. 75 3.2.5 P97 expression in BV-2 cells after treatment with NSAIDs Next, the expression of p97 in microglia in the presence of N S A I D s , was tested. To assess i f N S A I D treatment reduced the activation of microglia, T N F - a production was again measured in cells either treated with stimulant alone or stimulant and N S A I D . Treatment with two N S A I D s , 10 p M Ibu and 10 p M N i m , appeared to reduce the production of T N F - a from cells treated with both stimulant and drug (Figure 3.5 a-b). When examining the gene expression of p97, it appears that the expression was decreased in cells treated with both Ibu and N i m , where R T - P C R and real-time P C R show that the expression of p97 in N S A I D treated cells was reduced close to the levels of non-treated cells (Figure 3.6 a-b and 3.7 a-b). Since previous studies have demonstrated that serum levels of p97 are elevated in A D , protein levels were also examined. Western blot analysis with p l800-GFP transfected B V - 2 cells also shows a decrease, similar to the levels of m R N A . G F P levels were also decreased in cells treated with drug compared to treatment with A P and L P S alone (Figure 3.6 c and 3.7 c). 76 A . Figure 3.5. NSAID treatment decreased TNF-a production in activated BV-2 cells. Cells were plated at a density of 5 x 105 cells/ml and treated for 24 hrs with 10 U.M A ^ M O or 50 ng/ml LPS with or without the addition of 10 p.M Ibu or 10 uJVl Nim. There was a significant decrease in TNF-a production in cells treated with drug compared to treatment with A p M 0 or LPS alone (ANOVA, *, ** P < 0.001). (a) TNF-a production in BV -2 cells after treatment with 10 (iM Ibu and (b) TNF-a production in B V - 2 cells after treatment with 10 | i M Nim. A l l experiments were performed at least 3 times in triplicate. Error bars represent ± SD. The results illustrated above are representative of all replicates. 77 NT Aft 1-40 L P S O CM X W f . VHBP :'**i*»Wv :.%HM»" ri — *** PT? 10 u M IBU p97 S 1 5 O w/out IBU • w/IBU AB1-40 Treatment NT ApV40 L P S + 1 0 u M I B U »«» G F P G A P D H Figure 3.6. p97 expression is decreased in BV-2 cells treated with Ibuprofen. Cells were plated at a density of 5 x 105 cells/ml and treated for 24 hrs with 10 | i M A|3MO or 50 ng/ml LPS with and without the addition of 10 urn Ibu. There was a significant decrease in p97 expression in cells treated with A|3 or L P S in conjunction with Ibu compared to A p M 0 and LPS treatment alone, (a) R T - P C R of B V - 2 cells treated with A p Y 40, LPS and Ibu. (b) p97 expression as determined by real-time P C R . Relative gene expression was determined by calculating the cycle threshold (CT) for p97 as well as for SI5 . Treated sample thresholds were then compared to non-treated and fold induction of gene expression determined. Error bar represent ± S D for C T values, (c) Western blot of G F P protein levels in p l800-GFP transfected B V - 2 cells after treatment with 10 | i M A p Y 40 and 50 ng/ml L P S with and without the presence of 10 u M Ibu. The levels of the G A P D H loading control are also shown. A l l experiments were performed at least three times and data indicated here are representative of all trials. 7 8 NT A(3i-40 L P S o CN + x 10 pM NIM p97 S15 AB1-40 T r e a t m e n t C. NT Api-40 L P S + 10 uM NIM m* G F P « • G A P D H Figure 3.7. p97 expression is decreased in B V - 2 cells treated with Nimesulide. Cells were plated at a density of 5 x 105 cells/ml and treated for 24 hrs with 10 p M A p ^ o or 50 ng/ml LPS with and without the addition of 10 pm N i m . There was a significant decrease in p97 expression in cells treated with A p or L P S in conjunction with N i m compared to A p M 0 and LPS treatment alone, (a) R T - P C R of B V - 2 cells treated with Ap,_ 40, LPS and Nim. (b) p97 expression as determined by real-time PCR. Relative gene expression was determined by calculating the cycle threshold (CT) for p97 as well as for SI5 . Treated sample thresholds were then compared to non-treated and fold induction of gene expression determined. Error bar represent ± S D for C T values, (c) Western blot of G F P protein levels in p l800-GFP transfected B V - 2 cells after treatment with 10 p M Ap,_ 40 and 50 ng/ml LPS with and without the presence of 10 p M Nim. The levels of the G A P D H loading control are also shown. A l l experiments were performed at least three times and data indicated here are representative of all trials. 79 3.3 Discussion In this study, it was demonstrated that fibrillar A(3 significantly promoted the up-regulation of p97 at both the m R N A and protein level in microglia. Furthermore, the signal transduction pathway involving p38 M A P K appears to be important in p97 production by stimulated microglia whereas the pathway involving E R K is not. It was also shown that the expression of p97 can be down-regulated with the use of NSAIDs . There have been a few studies that have proposed that p97 may be a putative biomarker for A D , with an ~4 fold increase in serum levels of p97 in A D patients 2 0 8> 2 0 9> 2 1 0 \ n addition, p97 appears to be expressed in the active microglia closely associated with amyloid plaques in the A D brain ' " . However, there was no direct evidence that microglia upregulate p97 expression as a result of activation. In this regard, the results from this study are very intriguing. Not only is it demonstrated that p97 is upregulated in activated microglia, but the upregulation is largely specific to activation via A p \ a stress specific to A D . It has been previously established by many groups, that in response to stimulation from molecules such as L P S and IFN-y, microglia upregluate the expression and secretion of many cytokines and other proteins such as I L - i p , T N F - a and urokinase plasminogen-activator receptor 2 , 8 , 1 1 3 . Therefore, it is possible that p97 is one of few proteins whose expression may be regulated by A(3 in A D . It has been shown here by real-time P C R that the expression of p97 increased approximately 6 fold after treatment of microglia with 10 p M fibrillized A p . A previous gene expression study 2 1 9 using gene array chips supports these results and demonstrated that the expression of p97 increased 2 fold after 24 hrs of treatment with 2.5 p M A p . 80 A number of different parameters of microglial activation in A D have been defined both in vivo and in vitro. In particular, the phosphorylation of various proteins involved in the M A P K . signal transduction pathway, such as p38 and E R K , are increased in activated microglia. The pathways downstream of these proteins share a common transcription factor, A P - 1 , which has 3 potential binding sites in the promoter region of 220 p97, suggesting a regulatory role for this transcription factor Further evidence in human melanoma cells support this and show that the R N A levels of Jun and Fos, which make up A P - 1 , is correlated to the expression of p97 2 X 1 . It is known that in microglia, L P S and A P stimulate the M A P K pathways and we have shown that A p and LPS treatment cause an increase in p97 expression. The present study also examined the expression of p97 in stimulated cells with and without the addition of p38 and E R K specific inhibitors. Inhibition of the p38 pathway resulted in a decrease of G F P expression in p1800-GFP transfected B V - 2 cells. Inhibition of E R K had no effect on G F P expression. Thus, it would appear that the expression of p97 in A p and L P S stimulated microglia may be under control of the p38 M A P K signal transduction pathway. Since p97 expression was not completely abolished by S B , it is possible that p97 expression may also be controlled by other mediators or pathways. The role of inflammation is becoming more evident in A D pathogenesis and it has been recently hypothesized that the neurodegeneration observed in A D is the consequence of an inflammatory response to A p and NFTs rather than as a result of these hallmarks 2 6 . Early epidemiological studies offer support for the role of inflammation in A D and have demonstrated that the use of N S A I D s confers protection against A D and appears to slow down disease progression ' '. In vitro experiments have shown that 81 N S A I D s , such as Ibu and indomethacin inhibit the gene expression of nitric oxide 221 222 synthase in macrophage and decrease cytokine production in neuronal cells ' . In addition, N S A I D s including Ibu, indomethacin and flurbiprophen, have been shown to reduce the amounts of APi_4 2 in human glioma cells by shifting the cleavage of AP to its shorter derivatives 2 2 3 . In animal models, it has been established that oral administration of Ibu approximately around the time of initial disease pathology, attenuated plaque pathology resulting in the decrease in both the size and number of plaques as well as a decrease in the number of dystrophic neuritis and activated microglia 2 2 4 . Moreover, A p i . 42 brain levels and the presence of activated microglia in N S A I D treated mice were 22 S 223 significantly lowered "' ' . Conversely, recent clinical trials examining a variety of N S A I D s failed to report any beneficial effects of N S A I D treatment. These trials are confounded by their small size and large withdrawal rates 2 2 6 . The findings from the many in vitro, in vivo and clinical studies suggest that the molecular targets of N S A I D s play a key role in the development of brain amyloidosis, however, what the exact targets are remain unclear. It is evident that more clinical trials are needed that focus on determining i f anti-inflammatory drugs can delay and/or prevent disease onset. To date, there are no end-point criteria that can be used as a means to assess drug efficacy. Plasma levels of A p are not correlated with AP brain levels indicating that measuring A p plasma levels as an indicator of drug efficacy is inconclusive and erroneous 2 2 3 . Here it is shown that the expression of p97, at both the m R N A and protein level, is affected by the presence of Ibu and N i m , potent anti-inflammatory drugs, after 24 hours of treatment. Thus it is possible that p97 may be a potential marker to determine drug efficacy. It is still unclear as to what aspect of inflammation in A D N S A I D s target or i f they target 82 inflammatory cells at all. Mechanistically, it has been shown that the beneficial effects of N S A I D s are a result of the drugs' ability to reduce or attenuate the activity of C O X enzymes thereby precluding the synthesis of prostaglandins and thromboxanes from arachidonic acid 2 6 . There is still much debate as to the pathological role of each of the C O X enzymes, C O X - 1 and C O X - 2 . Alternatively, N S A I D s have been shown to work in a C O X independent manner reducing the amounts of Api.49 by targeting the y-secretase complex 2 2 7 > 2 2 3 . in this study, the effects of both a non-selective C O X inhibitor, Ibu, and a C O X - 2 specific inhibitor, N i m , were examined. In both instances, p97 expression was decreased. This suggests that p97 expression may be dependent or related to the upregulation of the both the C O X - 1 and C O X - 2 enzymes. Both E R K and p38 M A P K pathways have been demonstrated to increase the transcription and to regulate the 228 stabilization C O X - 2 . In contrast, Ibu has also been shown to reduce A P 1 . 4 2 levels with little inhibition of C O X enzymes 1 7 4> 2 2 3 . i n order to resolve this, treatment with N S A I D s that are C O X - 1 specific inhibitors and have no effect on the activity of C O X enzymes, such as Thalidomide which inhibits T N F - a , needs to be examined. Nonetheless, these data indicate that expression of p97 is upregulated in response to microglia activation and that treatment with N S A I D s causes a down regulation of m R N A expression. Therefore, it is possible that serum levels of p97 may be used in future N S A I D clinical trials as a criterion to resolve whether a certain drug is beneficial. In summary, these data establish that p97 m R N A is upregulated in microglia in response to treatment with A p and that the expression of p97 is regulated by the AP-1 transcription factor downstream of the p38 M A P K pathway, which has also been shown to be highly active in microglia. Moreover, the expression of p97 can be modified by 83 anti-inflammatory drug treatment. It has been previously, proposed that p97 may be a plausible biomarker for A D . These studies show that in addition to being a diagnostic biomarker for A D , p97 may be used as an indicator of A P specific mediated inflammation/microglial activation. P97 may also be used as an aid in determining the efficacy of new and already established anti-inflammatory therapies for. A D . 84 Chapter 4: The role of microglia in amyloid deposition 4.1 Rationale Activation of microglia is among the first cellular changes in the injured CNS. In an acute inflammatory response activated microglia produce many beneficial neuro-trophic and neuro-regenerative factors, in addition to pro-inflammatory mediators and potential neurotoxins. This response is short-lived and only the injured neurons are regenerated or removed. Alternatively, in a chronic inflammatory response, neurons are exposed to large amounts of neurotoxins for a prolonged period of time and this exposure ultimately leads to their demise. It has been proposed that microglia bring about the neurodegenerative changes in A D by eliciting an inflammatory response and by contributing to amyloid deposition. However, the decisive role of microglia in the accumulation of A p and the subsequent propagation of amyloid plaques remain to be clearly elucidated. It has been demonstrated in A D , that activated microglia co-localize with and infiltrate amyloid plaques in both humans and APP transgenic mice l l 6- l l 7>n 8>9 0 ultra-structural studies by Wisniewski et al. have found that plaque associated microglia displayed intracellular channels containing amyloid fibrils 2 2 9 . A subsequent study also found A P to be present in the secondary lysosomes of macrophage-like cells suggesting 230 that microglia are involved in both the production and phagocytosis of A p Furthermore, microglia may be able to drive the fibrillization of A P monomers/oligomers into fibrils and plaques 1 1 9 , 2 3 1 . It is also been established that microglia are able to rapidly internalize and degrade microaggregates of A p by scavenger receptors or by activation of 85 complement pathways ' ' ' "". Regardless of the mechanism of internalization, the degradation of amyloid appears to be the main issue in A p accumulation. The ultimate fate of phagocytosed AP is still unclear. It has been shown that A p is degraded by microglia; however, this degradation is slow and can lead to the accumulation of AP inside the cell. Therefore, in the case of AD, the overproduction and thus the persistence of A p may become too overwhelming for the microglia thereby disrupting the dynamic balance between AP deposition and removal. Regardless of the mechanism it would appear that activated microglia contribute to the formation of amyloid plaques. Many studies have focused on the role of complement in amyloid deposition and have shown that the activation of complement, both the classical and alternative pathways, results in the production of many complement opsonin proteins that bind to A p 116,234 j/hese opsonins promote microglial phagocytosis of A p , thereby implying that 23 ^  microglial activation promotes amyloid removal, precluding plaque formation " . In a study by Wyss-Coray et al. it was found that in mice which overexpress astroglial T G F -p i and human APP695,751,770, there was a 3 fold reduction in the number of A p parenchymal plaques and an overall 50% reduction in amyloid burden 2 3 6 . These mice also exhibited an increase in complement protein C3 as well as activated microglia suggesting the T G F - p i promote the degradation of A p by microglia via complement activation 2 3 6 . Further studies involving the inhibition of C3 activation in A D model mice by expression of the soluble complement receptor-related protein, found that in C3 inhibited mice there were significant increases in amyloid plaque burden, the levels of APi ^ 2 , and the number of degenerating neurons 2 3 7 . Overall, these data suggest that certain inflammatory defense mechanisms are indeed neuroprotective. In contrast, other 86 studies involving complement C l q deficient hAPP mice showed that the absence of C l q resulted in a decrease in the degree of degenerating neurons compared to C l q sufficient mice, with no difference in amyloid burden 2 3 8 . This implies that complement plays a detrimental role in disease pathology. A better understanding of the mechanisms that regulate amyloid accumulation and degradation may help facilitate the generation of therapeutics that w i l l prevent the accumulation or enhance the degradation of A p and ameliorate A D treatments. It is hypothesized that microglia are activated in response to A p and phagocytose it removing it from the parenchyma. In time, the activated microglia may become overburdened and unable to remove the amyloid thereby facilitating its aggregation and subsequent plaque formation. This study addresses the role of microglia in plaque deposition using the Osteopetrotic (CSF-1 deficient; op/op) mouse. A s a model for osteopetrosis, this mouse has a spontaneous frameshift mutation that results in a complete deficiency of the CSF-1 factor, an important mitogen for brain microglia promoting survival, proliferation and 182 183 differentiation ' . As a result, these animals possess a reduced number of mature, functioning microglia in their brains 1 8 7 ' 1 8 6 . In culture, microglia isolated from op/op mice can be restored to full functionality with the addition of CSF-1 . Moreover, daily CSF-1 administration, before B B B formation, can largely restore microglial function 2 3 9 . Animal models with a defective microglial response, such as osteopetrosis, provide an approach to explore the many relationships fostered by microglia in the C N S . To explore the course of A P plaque development, the op/op mouse was crossed with the Tg2576 A D model mouse to generate an A D mouse with reduced microglial capabilities. This new mouse model w i l l hopefully help resolve the contribution of microglia to A D 87 pathogenesis. In this study, the role of microglia in amyloid deposition was investigated. In accordance with previous data on complement and amyloid pathology, it is hypothesized that in A D mice with reduced microglial function, there w i l l be an increase in amyloid burden in the brain. 4.2 Results 4.2.1 Characterization of Tg/+;op/op mice Tg2576;CSF-l deficient (Tg/+;op/op) mice were generated by initially crossing Tg2576 (Tg/+;+/+) mice by CSF-1 deficient (+/+;op/op) mice. The F i generation of these mice (Tg/+;op/+) were crossed to get Tg/+;op/op mice. To distinguish between Tg/+ mice and wild-type (+/+;+/+), all mice were genotyped for the h A P P transgene (Figure 4.1). +/+;op/op mice were distinguished from +/+;+/+ mice by the absence of incisors and a domed skull 10 days after birth. Tg/+;op/op and +/+;op/op mice were separated from Tg/+;+/+ and +/+;+/+ mice and fed on wet chow. Tg/+;op/op mice were significantly smaller than their control littermates (Figure 4.2). Mouse growth and body weight was followed for a 3 month period. At 3 months, op/op mice were approximately 33% smaller from +/+ littermates. Body weight was also evaluated at 6 and 9 months of age at time of sacrifice. A s depicted in Figure 4.2, Tg/+;op/op mice had a significantly lower body weight that control mice. In accordance with the phenotype of the op/op mice, Tg/+;op/op mice appeared to be infertile. In addition, the viability of the Tg/+;op/op mice was reduced, as many mice died before the 9 month time point. The average life span of the op/op mouse is reported 88 to be 7 months of age (Jackson Laboratories); however, we have maintained this mouse for periods exceeding 12 months. Figure 4.1. P C R genotyping of Tg2576 A D model mice. Tg2576 h A P P transgenic mice and littermate controls were genotyped from ear punch D N A by P C R as previously described. The P C R products were analyzed on a 1% agarose gel and visualized. Tg2576 mice amplify a single 420 base pair (bp) product while control mice do not amplify any product. Each P C R was performed twice. Lane 1, Tg2576 founder control; Lane 2, C57/SJL non-transgenic control; Lane 3-6 genotyped mice. The gel shown is representative of all PCR reactions. 89 35-i 30-| ra 3 20\ £ 15-I 10-j o4 0.0 1 Tg/+;op/op ' Tg/+;+/+ +/+;op/op +/+;+/+ —f-2.5 5.0 Time (months) - i — 7.5 10.0 Figure 4.2. Tg/+;op/op mice are smaller than control littermates. Mice were weighed beginning at one month of age until 9 months of age. Tg/+;op/op were significantly smaller than control littermates ;Tg/+;+/+ and +/+;+/+ at every age (P < 0.01, two-way A N O V A followed by Bonferroni analysis). Tg/+;op/op mice were smaller than +/+;op/op mice at all time points, however the difference was not significant (P > 0.05; two-way A N O V A followed by Bonferroni analysis). 4.2.2 Amylo id burden in Tg/+;op/op mice Amyloid plaques are first seen in Tg2576 mice at approximately 9 months of age, therefore this time point was used to evaluate plaque burden. In addition, Tg/+;op/op mice have a high mortality and generally die at 10 months of age. A t 9 months of age, Tg/+;op/op mice exhibited similar amounts of A p immunostaining as Tg/+;+/+ mice (Figure 4.3 a-c. P = 0.1; t-test). There was no presence of A P in +/+;op/op and +/+;+/+ mice (Figure 4.3 d and e). A t 6 months and 3 months there was no A P immunostaining present in all animal brains. Interestingly, there appeared to be Ap deposition in cerebral blood vessels of Tg/+;op/op mice compared to Tg/+;+/+ mice (Figure 4.4). Therefore, it is possible that the absence of microglia increases vascular angiopathy. 90 Figure 4.3. Amyloid plaque burden in 9 month Tg/+;op/op mice compared to controls. Amylo id plaques in cortical sections from mice were visualized with an anti-human AP antibody. There was no significant difference in the total number of Ap plaques in Tg/+;op/op mice compared to Tg/+;+/+ controls. There was no detection of AP plaques in +/+;op/op and +/+;+/+ mice (a) Quantitative assessment of plaque burden in Tg/+;op/op mice and Tg/+;+/+ mice, T-test; P = 0.1. Qualitative assessment of A P in (b) Tg/+;op/op mice, (c) Tg/+;+/+ mice, (d) +/+;op/op mice, and (e) +/+;+/+ mice. Brain sections shown are representative of their respective groups (n=3 for Tg/+;op/op, n=4 for Tg/+;+/+, +/+';op/op and +/+;+/+). 91 92 Figure 4.4. Amyloid accumulation in cerebral blood vessels of Tg/+;op/op mice. Amylo id accumulation in cerebral blood vessels (thin arrow) in the brains of Tg/+;op/op mice compared to Tg/+;+/+ controls as visualized with an anti-human A B antibody, (a) Tg/+;op/op mice (b) Tg/+;+/+ mice. Brain sections shown are representative of their respective groups (n=3 for Tg/+;op/op, n=4 for Tg/+;+/+). 93 94 4.2.3 Microgliosis in Tg/+;op/op mice Many studies examining the number and morphology of microglia in op/op mice are controversial 1 8 3 , 1 8 6 , 1 8 7 > 2 4 0 . This study demonstrates that there is a decrease in the number of activated microglia in the Tg/+;op/op mice compared to the Tg/+;+/+ mice. In addition, the majority of the microglia in the Tg/+;op/op mice exhibited a resting morphology with small cell bodies and thin ramified processes whereas the microglia in the Tg/+ mice had a more activated morphology with more condensed cell bodies and thicker processes (Figure 4.5). These results are supported by previous studies indicating a reduced number and altered morphology of microglia in the op/op mouse I 8 7 . 95 Figure 4.5. Reduced number and altered morphology of microglia in Tg/+;op/op mice. Brain sections were stained with F4/80 to reveal the presence of microglia. Tg/+;op/op mice exhibited an overall decrease in the number of microglia than Tg/+;+/+ mice and fewer activated microglia than Tg/+;+/+ mice. +/+;op/op mice and +/+;+/+ mice exhibited resting microglia, with +/+;op/op mice having a reduced number i f cells. Activated microglia can be distinguished from resting microglia by the presence of ramified processes and condensed cell bodies (black arrow) compared to small cell bodies and thin ramified processes (white arrow). Control littermates exhibited no microgliosis (a) Tg/+;op/op mice, (b) Tg/+;+/+ mice, (c) +/+;op/op mice, and (d) +/+;+/+ mice. Brain sections shown are representative of their respective groups (n=3 for Tg/+;op/op, n=4 for Tg/+;+/+, +/+;op/op and +/+;+/+). 96 • '• v V - • V Y *• - . * .»- . • V I -. - ** • \ ' * * - 1 >v t Y **Y* > .100|iM 97 4.3 Discussion It has been hypothesized that microglia play an integral role in the formation of amyloid plaques and results from many in vitro and in vivo studies support this premise. These previous studies established that activated microglia are attracted to and surround amyloid plaques 1 1 6- l l 7> 1 1 8> 9 0 However, whether microglia promote plaque formation by converting monomeric/oligomeric A P to fibrillized A p or whether microglia prevent A p deposition by complement mediated phagocytosis in vivo remains unresolved. The present study uses a CSF-1 deficient h A P P transgenic model to assess the implication of activated microglia in A D pathogenesis. Tg2576 were bred with mice deficient for C S F -1 and hence lacked mature, differentiated microglia. The results obtained here indicated that the Tg/+;op/op mouse was smaller in weight than control littermates and had a higher mortality rate. This was in accordance with previous studies on op/op mice, which found that op/op mice were smaller than +/op and +/+ littermates, had a decreased life-span and poor reproductive performance 1 8 8> 1 8 2 . in this study, the oldest age group was 9 months of age, the age of initial plaque formation in the Tg2576 A D mouse. Older age groups were not examined due to the fact that no Tg/+;op/op mice survived past 10 months of age. The amount of plaque pathology has been investigated in many mouse models of A D . In the mouse model created in this study, levels of fibrillar A P in Tg/+;op/op mice appeared to accumulate at rates comparable to Tg/+;+/+ mice since there was no significant difference in plaque burden in the cortices of 9 month old mice. No plaques were observed in the brains of 6 month old animals. It was also observed that there was more A p accumulated in the cerebral blood vessels of the Tg/+;op/op mice than in the 98 Tg/+;+/+ mice. With the high incidence of C A A in A D such deposits are likely to be found. Since there were few Tg/+;op/op mice in the study, it was difficult to clearly determine a difference in plaque deposition as well as establish an explanation in the shift of A p deposition patterns. It is possible that with a decrease in the amount of activated microglia in the brains of these animals, other cells, such as astrocytes and pericytes, compensated and upregulated the production of certain cytokines and other important factors. For instance, TGF-P, a known immunosuppressive cytokine, is upregulated in A D and has been implicated in A D in astrocytosis, microglial activation, and 26 accumulation and regional distribution of A p . Recently, TGF-pl was implicated as an inducer of vascular amyloid deposition 236>241. i n addition, Lesne et al. demonstrated that TGF-P 1 potentiated A p production in human astrocytes and, as a result, may enhance the 242 formation of amyloid deposits . However, the role of TGF-P, beneficial or detrimental, in the CNS is still unresolved. It is also feasible that with the functional deficiency of microglia in the op/op mouse the number or proportions of factors produced by activated microglia is skewed. This, in turn, may be amplifying certain pathological changes such as amyloidogenesis and amyloid accumulation in the blood vessels. A more detailed analysis on brain A p levels and vascular A p levels is required in order to determine if there is indeed a shift towards vascular amyloid deposition and what cellular and molecular processes are involved. There is still great controversy as to the state of microglia in the op/op mouse. Initial studies by Blevins and Fedoroff found that microglia were not affected by the mutation in CSF-1, in that they have normal morphology and are present at normal 186 frequency . They did find that the microglia isolated from op/op mice needed CSF-1 in 9 9 order to differentiate and proliferate in culture, implying a possible deficiency in function 1 8 6 . Other studies established that there was indeed a reduction in the number of microglia in the brains of op/op mice; 47% in the corpus callosum, 37% in the parietal cortex and 34% in the parietal cortex compared to +/+ and +/op controls 2 4 3> 2 4 4> 1 8 7 Morphological differences, such as, smaller size and shorter cytoplasmic processes were 187 also found and limited to the frontal cortex. . There have also been some studies focused on the response of microglia to acute C N S injury in the op/op mouse. In the facial motor nucleus paradigm, there was a lack of cell proliferation, mobilization and change in morphology in op/op mice compared to controls 2 4 5 . Microgl ia in the op/op mouse responded to neuronal signals but there was a decrease in neuronal regeneration due to the lack of proliferation and mobilization. Further studies examining the role of microglial mediated neurodegeneration found similar results. Here, endotoxins were injected directly into the brains of animals and microglial activation and neuronal degeneration were assessed. It was found that microglia in the op/op mouse have normal activated morphological phenotype, however, only 30-40% of activated microglia were observed in the brains of op/op mice compared to controls 2 4 0 . Moreover, levels of T N F -a were increased in response to injury 2 4 6 and were at. par with the amount of activated cells with op/op mice having 48% of levels to that of controls 2 4 0 . In regards to neurodegeneration, it appears that microglia in the op/op mice were able to promote neurotoxicity potentially by T N F - a mediated expression of I L - i p , IL-6 and other pro-inflammatory cytokines 2 4 6 ' 2 4 0 . Therefore, it is now evident that the microglia in op/op mice are deficient in their proliferative ability but are able to elicit neurodegeneration and removal of debris. It should be noted that all the aforementioned experiments focused on 100 acute C N S injury. The response of microglia in op/op mice in chronic inflammatory reactions is not well documented. This is where the experiment outlined in this study is unique. From the data presented in the literature one can speculate that, with respect to the CSF-1 deficient hAPP , there may be more AP deposition at a later age due to the inability of microglia to proliferate. The microglia present in the brain may be able to accommodate and degrade accumulated A p at the initial stages of the disease, but as the amount of A p increases there is not enough activated microglia to remove it. This also is in agreement with the complement experiments by Wyss-Coray et al. where inhibition of complement (which results from microglial activation) results in an increase in plaque development 2 3 1 . With fewer cells able to express complement proteins and activate complement pathways there may be a reduction in the effective removal of A p . According to the "microglial dysfunction" hypothesis 1 0 9 , one can also anticipate an accelerated accumulation of A P since with age it is thought that microglia become senescent and/or dysfunctional therefore their ability to support neurons and to act as phagocytic macrophage would be diminished. With fewer microglia present in the brain, the degree of senescence would be greater in Tg/+;op/op mice than Tg/+;+/+ mice perhaps hastening A D pathology. Moreover, the extent of neurodegeneration, like A p deposition, may start at the same rate in Tg/+;op/op and Tg/+;+/+ mice. The A P deposition may then accelerate in Tg/+;op/op mice due to the increase in amyloid and its toxic effects on neurons, as well as, the presence of microglial neurotoxins. Also, it is possible that microglia in op/op mice have a diminished ability to undergo phagocytosis and support neuronal survival due to the lack of CSF-1 which may result in an alteration 101 in the expression and response of many cytokines and other neurotrophic factors that microglia provide to the C N S . The number and ages of mice used in this study were limited due to the high mortality of the Tg/+;op/op mouse. It is possible that with a larger subject cohort and/or at later age, there could be differences in plaque burden between the different mice. Moreover, differences in other pathological hallmarks of A D , such as neurodegeneration, may be evident at later ages. In a study by Fonseca et al, C l q deficient h A P P transgenic mice had similar plaque burden to h A P P mice at 9 and 16 months of age where older C l q deficient h A P P mice exhibited less neuropathology than controls. 2 3 8 . Studies by Wyss-Coray et al. on T G F - p i overexpression hAPP mice and on C3 inhibited h A P P mice, demonstrated brain pathology at 10-12 months of age after significant plaque deposition 236 217 occurs ' " . In addition, the h A P P mice used in these studies express mutated hAPP695,751,770 under control of the platelet derived growth factor p chain promoter and develop plaques at approximately 6 months of age 2 4 1 . Therefore, since the op/op mouse has a shortened life span it would be beneficial to use an A D mouse that develops significant A p deposition prior to 9 months of age to evaluate the relationship between microglial activation and plaque formation. Many studies have demonstrated that genetic background regulates the extent of A P P processing, A p deposition and mouse mortality. Initial attempts by Hsiao et al. to 84 create transgenic mice failed in part due to host strain effect on transgene expression . These mice were generated in the F V B / N inbred strain. The second attempt utilized a C57BL/6 x C57B6/SJL hybrid proved more successful and is the basis for the Tg2576 A D transgenics used in this study. These mice also had some initial complicating factors. 102 The life span of the Tg2576 mouse on the inbred C57BL/6 background is approximately 6 months whereas on a mixed background of C 5 7 B L / 6 ; S J L mice can survive for over 2 years (Younkin, personal communication). Gene dosage also has an effect on mouse mortality. For example, the Tg2576 A D mouse exists only as a hemizygote and is lethal in a homozygous state. It is possible that the Tg/+;op/op generated in this study contained more of the C57BL/6 background which impacted its life span, since very few of these mice survived past 9 months of age. Recently studies on A P P proteolysis, A p metabolism and A P deposition in transgenic mice have demonstrated that these processes are regulated by genetic background 2 4 8 . The C57BL/6 background exhibits greater plaque burden, earlier age of A p deposition, and increased levels of brain and plasma levels of A P M O and A P i ^ , compared to D B A / 2 J and 129Sl/SvImJ strains 2 4 S . These results suggest that certain gene alleles in some mouse strains, such as C57BL/6 , are dominant over alleles present in other strains, such as D B A / 2 J , and alter AP-related pathologies in A D . As a consequence of host effects, most transgenic mice used in A D research have hybrid backgrounds. The diversity of the response seen from this study and many others examining microglia and their induction of inflammatory and complement pathways, indicates the complexity of microglial responses and the multiplicity of microglial activation states in A D . A considerable amount of investigation is required and necessary to elucidate these interactions. 103 Chapter 5: Ap immunization and the blood-brain barrier 5.1 Rationale Conflicting with the concept that A B M O and AB1-42 are cytotoxic are noteworthy advances in recent A D research which advocated the use of peripherally administered A p peptides as a vaccine in an effort to reduce senile plaque loads of A D model mice While the exact relationship between A p production and deposition and its role in A D neurodegeneration remains unclear, one cannot dispute that treatments aimed at reducing the amyloid burden have resulted in reduction in senile plaques and behavioral benefits. Recent vaccination studies aimed at reducing the burden of all A p types in the brain have proven successful in transgenic mice. The first of such studies by Schenk et al. injected fibrillar Api.42 into six week old P D A P P mice, prior to A D symptom onset, and in 11 month old P D A P P mice, after noticeable A P deposition has occurred in the brain 1 7 5> 2 4 9 . After eleven months, mice which had been vaccinated since they were six weeks old showed an almost total absence of amyloid plaques, and a reduced amount of dystrophic neuritis, astrogliosis and microgliosis in the brain compared to P B S injected controls. After five months, mice which had been vaccinated since they were 11 months old showed a significant decrease in plaque burden and in the amount of diffuse A p , as well as a reduction in dystrophic neuritis and astro- and microghosis 1 7 5 . Many other immunization studies, utilizing both active and passive immunization protocols, have been performed in the same and different A D model mice and have found similar results 177 250 251 in regard to the pathological features of A D ' ~ ' . On the other hand, some studies in different A D mouse models have found that immunization had little effect on plaque 104 burden i f dense plaques were pre-existing " . Studies focused on memory impairment have found that immunization with A(3 or passive transfer of antibodies against A p protected mice from learning and age-related memory deficits behavioral impairment 178,179 a g w e j j a g | n a s i g n i f i c a n t i y delay the onset of memory deficits 2 5 3 - 2 5 4 . Recently, it was determined that A P immunotherapy can clear early Tau pathology but not late hyperphosphorylated Tau aggregates 1 8 1 . With the success o f amyloid immunization in mice, clinical trails in humans were undertaken. These trials were halted when approximately 5% of the test subjects showed signs of inflammation of the C N S , clinically described as aseptic meningoencephalitis 2 5 5 , 2 5 6 . Subsequent studies in APP23 mice found the same results as in the human trials. In these mice A p immunization did reduce plaque burden, as in other studies; however, there was also an increase in cerebral microhemorrhages associated with amyloid-laden vessels 2 5 1 . Regardless of the early success in animals, it remains unclear how peripheral immunization can affect plaque burden in 'immunoprivileged' tissues such as the brain that possess vasculature that limit immune surveillance. This vasculature, also known as the B B B , is formed by a continuous layer of capillary endothelium joined by tight junctions that are generally impermeable, except by active transport, to most large molecules including antibodies and other proteins. With this in mind, three possible mechanisms have been proposed to explain the effects of vaccination in A D . The first suggests that the antibodies against A p , normally excluded from the C N S , enter the brain by an obscure mechanism involving receptor mediated transport and then proceed to bind to A p . Entry into the brain may be facilitated by binding of the antibody to A P in the periphery creating an Ap-antibody complex which is then transported in the brain via 105 receptor mediated transcytosis by receptors that recognize A p " . The Ap - a n t i b o d y complex is then degraded by the microglia via Fc receptor (FcR) mediated phagocytosis 259 176 260 261 " ' ' ' . However, when fragments of antibodies were administered there was reduction in A P load in the brain ' . This indicates that non-FcR mediated clearance of A P is also occurring. A second possible mechanism suggests that antibodies against A p are actively transported into the brain, again by an unknown mechanism, and bind directly to A p fibrils resulting in plaque disaggregation. In this instance, the antibodies sequester the A p and prevent further fibril/plaque formation 2 6 1 . Finally, it is also postulated that there is an equilibrium formed by active transport of A P into and out of the brain by receptor mediated transcytosis and by antibodies against A p , binding A p in the periphery removing it from circulation. This then causes a shift in the equilibrium between A p in the plasma and C N S , resulting in a suppression of A P deposition in the 1 77 OK/l 9 / ^ 9 / ^ / i brain ' ' ' . Studies involving intravenous administration of A p specific antibodies demonstrated an efflux of A P from the brain to the plasma 2 6 4 . In addition, administration of a l 2 5 I-AP - a n t i b o d y complex into the periphery inhibited the movement of l 2 5 l - A p into the brain 2 6 6 . The permeability of the B B B to A P peptides and antibodies to A P is addressed in this study. In addition, the effect of A P immunization on the integrity of the B B B is also examined. It is hypothesized that, as with previous immunization studies where there was a reversal of pathology, amyloid immunization w i l l restore B B B integrity in Tg2576 mice. These results w i l l aid in further substantiating this form of therapy in A D research and wi l l also serve as a means to explain the mechanism by which immunization reverses plaque pathology. 106 5.2 Results 5.2.1 Ap peptide and anti-Ap antibodies and their ability to cross the BBB To investigate the possible mechanisms involved in Ap vaccination, the ability of both the APi^o peptide and antibodies against Ap to cross from the periphery into the brain was examined. First, fluorescent labeled Api_4o or PBS were injected i.v. into 6 week Tg2576 mice and control littermates. The fluorescent Ap was detected in the brain parenchyma of both groups of animals (Figure 5.1 a-d). Since it has been previously shown that AP can gain entry into the brain via receptor mediated transcytosis, this data also supports this hypothesis. Other studies focused on the movement of Ap into the brain support these results. Yan et al. found that the endothelial cells which make up the B B B express the R A G E receptor 6 8 . This receptor has been shown to bind to Ap with high affinity and transport AP into the brain. As a control for B B B permeability, BSA, a protein known not to be able to cross the B B B , was conjugated to Texas Red and injected into mice. As seen in Figure 5.1 (e and f), the B S A was primarily localized in the vessels indicating an intact B B B ; however, there is some BSA found in the brain parenchyma of transgenic mice. The presence of B S A in the brain may be an indicator of the beginning of stages of increased B B B permeability in the Tg2576 mouse. Next an anti-human Ap antibody, clone 4G8, was injected into 6 week old transgenic and non-transgenic mice. In contrast to the Ap peptide, there was no presence of anti-Ap antibodies in the brain parenchyma (Figure 5.2 a-d). This experiment was repeated using a biotin-labeled anti-AP antibody. In this instance either the labeled antibody or biotin alone was injected. As with the above experiment, biotin conjugated anti-Ap antibodies were only present in the 107 vessels and not in brain parenchyma (Figure 5.2 e-h). This would indicate that anti-AP antibodies cannot gain access to the brain from the periphery. However, recent studies using iodinated antibodies against A P indicated that antibodies can cross the B B B but with very low efficiency 2 5 8 . In the study by Poduslo et al. the mice used were significantly older than the ones used here 2 5 8 and it has been established that the B B B in Tg2576 A D model mice is compromised by 4 months of age 1 4 9 . Thus at 6 weeks of age, there is little damage to the B B B and it is possible that antibodies cannot gain access to the brain. 108 Figure 5.1. A p peptides can cross the BBB in both transgenic and wild-type mice. Fluorescent labeled Api^o was injected i.v. into 6 week old transgenic and control littermates. A p was able to cross the B B B and enter the brain parenchyma in both transgenic and non-transgenic mice thereby suggesting a receptor mediated transport across the B B B . (a) cortical section of a Tg2576 mouse injected with A P , (b) cortical section of a control littermate injected with A p , (c) cortical section of a Tg2576 mouse injected with P B S control and (d) cortical section of a control littermate injected with PBS . As a control for B B B integrity, Texas red conjugated B S A was injected into both mice. A s seen in (e) Tg2576 mouse and (f) non transgenic littermate, the B B B is relatively intact. There appears to be a small amount of B S A in the brain parenchyma of the transgenic mouse indicating the beginning of the B B B breakdown. *, brain parenchyma with A p peptide, arrow, brain vasculature. 109 Figure 5.2. anti-Ap antibodies cannot cross the B B B in both transgenic and wild-type mice. Anti-AP antibodies, clone 4G8 against human APi_4 2 , were injected i.v. into 6 week old transgenic and control littermates. Both unlabeled antibodies (visualized by immunofluorescence) and biotin labeled antibodies (visualized by light microscopy) were used. In both cases anti-Ap antibodies were localized to the vasculature (white and black arrows) and were unable to cross the B B B and access the brain in both Tg2576 and control littermates. (a) cortical section of a Tg2576 mouse injected with anti-AP antibodies, (b) cortical section of a control littermate injected with anti-AP antibodies, (c) cortical section of a Tg2576 mouse injected with P B S control and (d) cortical section of a control littermate injected with P B S . (e) cortical section of a Tg2576 mouse injected with biotin labeled anti-Ap antibodies, (f) cortical section of a control littermate injected with biotin labeled anti-Ap antibodies , (g) cortical section of a Tg2576 mouse injected with biotin (control) and (h) cortical section of a control littermate injected with biotin (control). I l l 4G8 P B S biotin-4G8 5.2.2 Ant i - A p antibody titres in immunized animals To assess the immune response of the mice to immunization with A P , antibody titres for antibodies against A p were measured. Serological analysis of serum samples collected from transgenic and non-transgenic mice vaccinated with A p and P B S were analyzed for the titres of anti-Ap antibodies by E L I S A using synthetic AP1 .40 peptide. Serum samples from mice in all groups were collected after the second injection and results indicated that transgenic and non-transgenic mice vaccinated with A P produced a high IgG response to A P i ^ o - No detectable antibodies were detected in transgenic and non-transgenic mice vaccinated with P B S (Figure 5.3). Mice vaccinated with A p , which did not exhibit high an t i -AP antibody titres, were not used in the study. 113 1000001 .1 10000' o o o cu 1000H 100' 10H Tg/+ +/+ Tg/+ +/+ PBS B. 10000n 1000' « o o o 0£ 100H 10H Tg/+ +/+ Tg/+ T * 1 +/+ 100001 •c- • o § 1000' CO o o o a: 100H ioi •'A MA. PBS Tg/+ +/+ Tg/+ +/+ PBS Figure 5.3. Antibody titre in serum of transgenic and non-transgenic mice immunized with either Ap or PBS. Serum antibody titres were measured after the second vaccination by ELISA. In all cases, mice (transgenic and non-transgenic) immunized with fibrillar APi_4o exhibited a high anti-Ap antibody titre against Ap peptide. No anti-AP antibodies were detected in mice immunized with PBS. A. 15 month age group, B. 12 month age group, C. 6 month age group. 114 5.2.3 Amyloid plaque burden in immunized animals Next, amyloid plaque burden was assessed in both A p and P B S immunized transgenic and non-transgenic animals. As noted in other immunization s tud i e s 1 7 5 ' 1 7 6 , 2 6 7 ' 1 7 8 ' 1 7 7 ' 1 8 0 ' 1 8 1 there was a significant reduction in the plaque burden in the cortex and hippocampus of transgenic mice immunized with A p compared to those immunized with PBS . In 15 month old mice, which were immunized with A p after disease onset, there was a decrease in plaque burden with a reduction in both the size and number of plaques, whereas brain sections from PBS-treated transgenic mice contained numerous amyloid deposits (Figure 5.4 a-b; Figure 5.5 a). These data agree with previous studies where there was not a total elimination or prevention of plaques in these animals. In mice immunized with A p prior to disease onset (12 month) there was an almost complete prevention of A p deposition. In A P immunized transgenic mice, 4 out of 6 had no detectable amyloid deposits. Two mice from this treatment group had a single isolated plaque in the 4 brain sections examined. A s with the 15 month group of mice, 12 month old, PBS-treated transgenic mice exhibited numerous amyloid deposits in their cortical and hippocampal regions (Figure 5.4 e-f; Figure 5.5 b). There were no detectable plaques in non-transgenic controls vaccinated with either A p or PBS (Figure 5.4 c-d, g-h). No amyloid plaques were found in 6 month old mice since amyloid plaques do not manifest in Tg2576 mice until 9 months of age. Non-transgenic mice injected with either A P and P B S exhibited no plaque burden (Figure 5.5 c-d, g-h). This confirms that immunization with A P may prevent plaque formation and thereby protect against disease. 115 Figure 5.4. A m y l o i d Pathology in Tg2576 M i c e Immunized with Ap or P B S . Amylo id plaques in cortical sections from mice were visualized with 4G8, an antibody against human Ap . There was a significant reduction in the total number of A P plaques in Tg2576 mice vaccinated with A p compared to those vaccinated with P B S at 11 months of age and at 6 weeks of age. Very few 4G8 positive plaques were found in brains of mice vaccinated at 6 weeks of age with A p . (a) 15 month Tg2576 mice vaccinated with P B S , (b) 15 month Tg2576 mice vaccinated with A p , (c) 15 month wild-type controls vaccinated with PBS and (d) 15 month wild-type controls vaccinated with Ap , (e) 12 month Tg2576 mice vaccinated with P B S , (f) 12 month Tg2576 mice vaccinated with A p , (g) 12 month wild-type controls vaccinated with P B S and (h) 12 month wildtype controls vaccinated with Ap . N o plaques were present in all 6 month old mice. Brain sections shown are representative of their respective treatment groups. 116 117 A. 16 14 12 I 1 0 o CD (/> C CD go Q) cr _CU CL * , J , A B P B S Treatment Figure 5.5. Cerebral amyloid levels are reduced in Tg2576 mice following A p immunization. Plaques were detected using an anti-human A p antibody, 4G8, on sequential brain sections. The presence of discrete plaques made it feasible to count the number of plaques in the entire section. Plaques were counted by visual inspection under the microscope for each of 4 sections at equal plane for each mouse. Total averaged number of plaques is presented. There was a significant reduction in the total number of A p plaques in Tg2576 mice vaccinated with A p compared to those vaccinated with P B S for both the 15 moth and 12 moth age groups. No plaques were seen in all 6 month old mice, (t-test * P < 0.05). 118 5.2.4 Microgl iosis in immunized animals The presence of activated microglia in immunized and non-immunized animals was also analyzed. Activated microglia are usually found associated with the amyloid plaques and can be distinguished from resting microglia by the expression of specific proteins, such as IL-1 p, CD1 lb and major histocompatibility class II, and by distinct morphology 2 6 . Activated microglia exhibit an altered morphology from resting microglia first by the presence of thickened ramified processes and larger cell bodies which progresses to a final amoeboid/phagocytic state 1 0 4 , 1 0 5 . Sections of mouse brains were reacted with an antibody specific for the F4/80 antigen. The F4/80 antigen is 268 expressed by a majority of mature macrophages, including microglia . In agreement with previous studies 1 7 5 , there appeared to be a reduction in the presence of activated, microglia in the brains of A p immunized mice compared to P B S controls suggesting a dampening of the inflammatory response. In 15 month transgenic mice immunized with P B S there are more plaque infiltrating microglia present compared to those immunized with A p (Figure 5.6 a,b). There were more densely stained F4/80 positive microglia with swelled cell bodies and thickened processes in PBS-treated transgenic mice compared to Ap-treated mice. Similarly, in 12 month old transgenic mice immunized with P B S there were more amoeboid shaped and thick ramified microglia, whereas there microglia in the A P immunized mice have smaller cell bodies and more extensively ramified processes (Figure 5.6 e,h). The microglia present in non-transgenic mice displayed largely a non-activated morphology with small cell bodies and thin, highly branched processes (Figure 5.6 c-d, g-h). 119 Figure 5.6. Microgliosis in Immunized Mice . Brain sections were stained with F4/80 to reveal the presence of microglia. Activated microglia can be distinguished from resting microglia by the presence of ramified processes and condensed cell bodies. Cortical sections from Tg2576 mice immunized with P B S or with A p , both at 11 months and at 6 weeks, show that A p immunized mice have reduction of activated, plaque associated microglia (black arrow). Control littermates exhibited no microgliosis (a) 15 month Tg2576 mice vaccinated with PBS , (b) 15 month Tg2576 mice vaccinated with A p , (c) 15 month non-transgenic controls vaccinated with P B S and (d) 15 month non-transgenic controls vaccinated with A P , (e) 12 month Tg2576 mice vaccinated with P B S , (f) 12 month Tg2576 mice vaccinated with A P , (g) 12 month non-transgenic controls vaccinated with P B S and (h) 12 month non-transgenic controls vaccinated with Ap . Brain sections shown are representative of their respective treatment groups. 120 5.2.5 BBB permeability in immunized animals To test the effect of vaccination on the B B B , B B B permeability was assessed in mice at various time points. The first group was immunized with either A p peptide or P B S at 11 months of age, after disease onset, and immunized for a 4 month period (15 month old mice). The second group was immunized beginning at 6 weeks of age, well before any pathological disease symptoms, for an 11 month period (12 month old mice), similar to previous studies by Schenk et al. 115. The final group was immunized beginning at 6 weeks of age for a 4 month period (6 month old mice). This last group was only immunized for a short duration since it has been demonstrated that B B B permeability can be compromised as early as 4 months of age. B B B permeability was assessed using the quantitative Evans blue assay ' . Organs with a selective barrier, such as the brain, only take up a small amount of the dye. If there is a breach in the B B B then Evans blue w i l l enter the parenchyma, whereas it is excluded from the brain with an intact B B B . In all groups of mice (15 month, 12 month and 6 month mice), the integrity of the B B B of transgenic mice was compromised compared with that of non-transgenic controls (** p < 0.05). These data concur with the study by Ujiie et al. where B B B integrity was compromised in 4 month and 10 month old Tg2576 mice 1 4 9 . Upon immunization with A(3, the A D transgenic mice displayed a significant decrease in B B B permeability, indicating a possible restoration of the B B B . Transgenic mice injected with A(3 had a significantly lower amount of Evans blue dye in their brain parenchyma compared to A D transgenic mice injected with P B S alone (Figure 5.7 a,b; * P < 0.05). This was true for both 15 month old mice and 12 month old mice immunized with A p . In the 12 month old mice, it is possible that vaccination with A P maintained an intact B B B , with little to no 122 compromise in B B B function. There was no change in B B B permeability in non-transgenic mice injected with A p or PBS . This latter observation is interesting as A P has been shown to be cytotoxic to endothelial cells and neuron in-vitro by eliciting apoptosis 27115154 272 , " , and angiogenesis , yet the low amounts used in this study do not appear to affect B B B integrity in normal mice. Finally, results from the 6 month old group are intriguing. In this instance, transgenic mice vaccinated with A p appeared to exhibit a breach in B B B integrity, similar to that of transgenic mice immunized with PBS (Figure 5.7 c). There was no effect elicited by A P immunization on the B B B as seen with the other groups. As demonstrated above, transgenic mice vaccinated with P B S had a compromised B B B in comparison to non-transgenic mice and no changes in B B B integrity were evident in non-transgenic mice injected with either A p or PBS . There was in an increase in the permeability of the B B B between transgenic and non-transgenic mice (Figure 5.7 c; ** P < 0.05). These data agree with previous data demonstrating that B B B integrity can become compromised as early as 4 month of age 1 4 9 ' . 123 Figure 5.7. BBB permeability as determined by Evans Blue in cortical regions of and PBS immunized mice. In all groups the permeability of the B B B in Tg2576 is greater than non-transgenic controls (t-test, **P < 0.05) (A) 15 months Tg2576 mice immunized with A(3 (n=5) show a decrease in the permeability of the B B B compared to PBS Controls (n=4) (t-test, *P < 0.05). The level of B B B permeability in A p immunized mice is similar to the permeability of non-transgenic littermates. (B) 12 months Tg2576 mice immunized with A P (n=6) show a decrease in the permeability of the B B B compared to P B S controls (n=6) (t-test, *P < 0.05). The level of B B B permeability in A p immunized mice is similar to the permeability of non-transgenic littermates. (C) 6 month Tg2576 mice immunized with A p (n=9) show no difference in B B B permeability compared to P B S controls (N=9). Tg2676 mice (both A p and P B S vaccinated) do exhibit an increase in B B B permeability compared to control littermates. 124 5.3 Discussion Research focused on systemic AP-lowering strategies is becoming more important in the development of therapeutics for A D . Recent advances have focused on immunotherapy using amyloid. These studies have proven successful in mice and have reached clinical trials in humans. What remains unclear is the mechanism by which A p immunotherapy alleviates the pathology seen in A D . This issue was first addressed by examining the ability of A p peptides to pass through the B B B and demonstrated that A p was able to access the brain in both transgenic and non-transgenic mice. This suggests that A P can gain entry into the brain via receptor mediated transcytosis 2 7 3 , 2 7 4 , 7 0 . A p transport across the B B B has been investigated by many groups and all results indicate an active transport of A P across the B B B . In a recent study it was shown that soluble A P can be transported from the C N S to the plasma since direct injection of radiolabeled A p into the brain was recovered in the plasma. Moreover, when radiolabeled A P was injected into the periphery, radioactive counts were found in the brain 7 0 . Banks et al. examined the influx of both A P M 2 and A P M O (mouse and human) in the different mouse models, C D - I and S A M P 8 2 7 5 . They found that all forms of A p were transported into the brain, with mouse 1-42 and human 1-40 being the fastest. In addition, all forms of A p were also transported out of the C N S . When examining the permeability coefficient x surface area product (PS) of proteins known to cross the B B B via receptor-mediated transcytosis (i.e. insulin) and of proteins known to have limited entry into the brain (i.e. albumin) it was shown that the PS value of A p was similar to that of insulin ' \ Further studies have suggested that the R A G E receptor is a receptor on the B B B that is responsible for the influx of A P from the blood to the brain 6 8> 2 7 3 . R A G E , is a multiligand 126 receptor in the immunoglobulin superfamily, binding ligands such as Ap, the SlOO/calgranulin family of pro-inflammatory cytokine-like mediators, the high mobility group 1DNA binding protein amphoterin and products of nonenzymatic glycoxidation 6 1 . The expression of RAGE is ligand dependant. In transgenic mice, RAGE expression correlates with AP deposition and is localized to affected cerebral vessels, neurons and activated microglia. The increase in expression may facilitate an increase in Ap transport 67 into the brain thereby exacerbating cellular dysfunction and disease progression . Other studies in animal models have shown that RAGE mediates the transport of Ap from the blood to the brain and deletion of RAGE results in the inhibition of Ap transport thus protecting the CNS from the accumulation of peripheral AP pools 6 9 . When an antibody against human Ap, was injected into mice there was no apparent presence of antibody in the brain parenchyma. Whether antibodies can cross into the brain from the periphery is under debate. It is possible that in this study, the amount of antibody used was insufficient to cross the BBB. Antibody injected into the periphery could have bound to circulating peripheral amyloid thereby preventing its movement into the brain. In this study the mice used were 6 weeks old. In other studies, where there was a presence of antibody in the brain, the mice were significantly older. It has been previously shown that the BBB is compromised in AD mice as early as 4 months of age. Thus, it is possible that at 6 weeks of age antibodies cannot gain access into the brain whereas in older animals the BBB is damaged thereby facilitating the entrance of antibodies to the brain. When evaluating the PS values of antibodies across the BBB, Podsulo et al. found that the PS value of non-specific IgG and AP specific antibodies were similar, 0.5 to 1.1 X 10"6 ml/g/s and 0.6 to 1.4 X10"6 ml/g/s respectively, and less 127 than that for albumin 1 4 7 . This indicates that the passage of antibodies across the BBB occurs at a very low efficiency 2 5 8 . Furthermore, Demattos et al. were unable to identify any plaque associated anti-Ap antibodies after immunization 1 7 7 . Conversely, there is abundant evidence which illustrates that anti-AP antibodies can cross the BBB. In a study using active immunization with A p peptides there was the appearance of anti-Ap antibodies associated with microglia localized with plaques 1 7 5 . In addition, antibodies to AP may be able to enter the brain once in complex with A p 2 5 8 . It has also been demonstrated that anti-Ap antibodies can enter the brain at a rate similar to that of albumin; however, due to the long half-life of the antibody in the blood there was a longer clearance time from the brain 2 7 6 . Taken together, the disruption of the BBB likely facilitates an increase of antibodies across the BBB efficiently enough to explain the action of the vaccine. The most significant finding in this study demonstrates that immunization with A p tends to repair the damage found in BBB in Tg2576 AD mice. In 15 month old transgenic mice, which have been inoculated for 4 months, there is a decrease in Evans blue uptake compared to PBS-treated transgenic mice. The significant decrease in Evans blue uptake observed in A p immunized transgenic mice compared to the PBS mice suggest that AP immunization results in the restoration of the BBB. In 12 month old mice inoculated for 11 months starting at 6 weeks of age, A p immunization may be able to actually prevent further disease progression. In these mice, Evans blue uptake is comparable to non-transgenic mice, where there is no measurable decreased integrity of the BBB. One possible explanation of the restoration of the BBB in older mice is that the vaccination leads to the decrease in the amount of circulating A p that could directly or 128 indirectly affect the function of endothelium in the B B B . A s demonstrated in this study 17^17X250 180 251 181 a and in many other studies ' ' " ' ' ' , there is a significant decrease in A p deposits and microgliosis following immunization. With the removal of A p from the brain and the subsequent deactivation of microglia, there wi l l be a decrease in the brain in the amount of reactive oxygen and nitrogen compounds and inflammatory cytokines produced by these cells which have been shown to activate vascular endothelial cells 277 153 ' . Studies focused on vasculature activation found that in response to systemic inflammation from either the periphery or brain parenchyma, endothelial cells up-regulate the expression of molecules that are generally associated with inflammatory processes such as prostaglandin E2, nitric oxide, C D 4 0 and C O X - 2 1 5 3 - 1 5 4 > 1 5 5 These then can act agonistically on microglia and neuronal cells to further intensify inflammation and disease pathology. A(3 has also been demonstrated to elicit many pro-apoptotic and pro-angiogenic responses in the endothelial cells that make up the B B B 1 5 2 > 1 5 1 ; 1 5 0 These effects are dependent on the amount of A P present. It has been exhibited in vitro that exposure of 272 271 151 endothelial cells to u M concentrations (5-25 uM) elicits pro-apoptotic signals ' ' whereas treatment with n M concentrations (50-250 nM) of A p elicits pro-inflammatory signals and increased monocyte migration with minimal disruptions to the endothelial monolayer ' Treatment of primary cerebral mouse endothelial cells with AP25-35 resulted in the activation of AP-1 and the subsequent expression of B i m , a member of the B H 3 only family of proapoptotic proteins 1 5 1 . Moreover, A p treatment also resulted in the translocation of second-mitochondria derived activator of caspase (Smac), a regulator of apoptosis, from the mitochondria to the cytosol where it can bind to the X 129 chromosome linked inhibitor of apoptosis protein ( X I A P ) , resulting in cell death 1 5 1 . Cytochrome C release from the mitochondria and the subsequent activation of several caspases, such as caspase 8 and caspase 3, all important events on the apoptotic cascade, have also been reported 1 5 ° . Several studies have also shown that exposure of endothelial cells to A p results in cytotoxic damage elicited from the generation of reactive oxygen species and increased calcium levels causing alterations in endothelial cell structure and function ' . Angiogenesis is also observed in endothelial cells in A D . Inflammatory mediators such as T N F - a , I L - l p , and IL-6 stimulate angiogenesis 2 8 1 . These inflammatory mediators, as well as C O X - 2 and amyloid, have been found to cause an increase in the expression of may angiogenic factors including V E G F , T G F -P and T N F - a 215 282 ' . Moreover, in A D patients there is an increase in V E G F expression in the bram as well as an increase in the serum and C S F 2 1 4 ' 2 1 5 . Recently, it was discovered that V E G F binds to A P with high affinity and specificity and is thus co-localized to the plaques. Therefore, it is possible that in response to pro-angiogenic and pro-apoptotic signals from A p and activated microglia, tight junctions disappear, cells round up creating a leaky barrier 2 1 6 . The endothelial cells in turn release neurotoxins, free oxygen radicals, A P P and more pro-angiogenic factors such as V E G F and T G F -p . With the removal of signal provided by A p and inflammation, it is possible that the endothelial cells quiesce, terminate the release of toxic substances, reform tight junctions reforming an intact, tight barrier. In regards to the mechanism by with A P immunization effectively reduces A D pathology; it is possible that the debate as to whether antibodies can cross the B B B is resolved. With a leaky B B B , antibodies are able to gain access into the brain, bind to A p 130 and are either degraded by microglia via FcR mediated phagocytosis or the AP-antibody complex can leave the brain. Alternatively, with a leaky B B B the efflux of A P from the brain can occur at a faster rate than by active transport by specific receptors, such as L R P 7 0 , thereby re-establishing equilibrium between plasma and C N S A p pools 2 5 8 . Once the excess amyloid is removed, by microglia or by efflux into the periphery, amyloid levels return to tolerable a threshold, microglia become deactivated, the B B B reforms and the brain becomes an 'immunoprivileged' site once again. In summary, these new observations provide an intellectual framework for understanding the efficiency of vaccination to modify disease progression. In addition, resealing of the B B B provides a previously undescribed intervention point for modifying disease outcome in amyliodopathies such as A D . 131 Chapter 6: Concluding remarks and future directions Since the eponym "Alzheimer's disease" was coined, there has been much debate as to the exact etiology and pathology of the disease. Many view that A D is not a single, uniform disease. Rather, A D is more likely a heterogeneous and multifactorial disease influenced by the interactions of various susceptibility genes and environmental factors. Intriguingly, as with other medical conditions, A D represents a disease with heterogeneity in its origins ( E O F A D compared to L O A D ) and homogeneity in its clinical appearance and pathology. Therefore, it is hypothesized that A D results from a complex sequence of steps involving multiple factors that extend well beyond the accumulation of Ap . This thesis concentrates on many facets of one of the main foundation of A D , the contribution of microglia in disease progression. In particular the relationship of A p to the activation of microglia, subsequent signal transduction pathways and gene expression and the impact of microglia on A P deposition and clearance and its effect on another pathological hallmark, the B B B , were addressed. According to the "amyloid cascade hypothesis" it is thought that AP accumulation is the principle event in A D . A possible discrepancy with this theory is that there is no direct evidence linking A p accumulation to neuronal death. Neuroinflammatory processes may be this missing link and have been suggested to exacerbate neuronal damage/death seen in A D . With the current understanding of the role of activated microglia in disease progression many therapeutic strategies directed against inflammatory processes have been pursued. Epidemiological studies on N S A I D use 102 103 demonstrated a reduced risk and decrease in the rate of cognitive decline ' and treatment of h A P P transgenic mice with anti-inflammatories decreased plaque size and 132 slowed down cognitive deficits . However, the use of N S A I D s appears to be effective only when administered prior to disease onset and not once symptoms are manifest, as indicated from current clinical trials where no beneficial effect of N S A I D use is reported 2 2 6 . The development of new anti-inflammatory therapies that interfere with one or more specific steps in the inflammatory cascade is underway. Nevertheless, it is essential to establish a specific, reliable and non-invasive biomarker to assess the efficacy of new drug treatments. Results from the studies performed in this thesis support the use of p97 as a marker of inflammation. The levels of p97, m R N A and protein, were upregulated in activated microglia. Interestingly, the increase in p97 expression occurred largely in response to A P stimulation and not in response to other known A D stresses such as IFN-y. Further investigation into the expression of p97 found that p97 production appears to be controlled by the p38 M A P K signal transduction pathway. This is intriguing since many studies have demonstrated that A P is able to induce the activation of p38 M A P K in vitro and that p38 M A P K phosphorylation is increased in affected brain regions in the 57 128 129 A D brain ' ' . In addition, this thesis has demonstrated that p97 levels can be altered with treatment of activated microglia with NSAIDs , Ibu and N i m , promoting its potential role as an A D inflammatory biomarker. Understanding the mechanism of amyloid aggregation and clearance w i l l ultimately lead to new advances in therapeutic development. Clearance of A p through microglial activation, chemotaxis, proliferation and phagocytosis has received a lot o f interest over the years, in particular from the recent immunization studies. Activated microglia displaying a variety of cell surface markers that distinguish them from resting microglia are often found within and immediately surrounding maturing amyloid plaques 133 ' ' . Whether activated microglia contribute to or aid in the clearance of plaques is controversial. Using a h A P P A D mouse model which lacked functional microglia it was found that, at the early stages of the disease, there was no significant difference in plaque burden in Tg/+;op/op and Tg/+;+/+ mice. Interestingly there were deposits of cerebral amyloid in the Tg/+;op/op mice implying a possible shift in amyloid deposition patterns. These results are intriguing but more mice need to be examined in order to resolve the complex relationship between activated microglia and A p aggregation and clearance. One of the latest therapies to combat A D employs active or passive immunization with A p peptide or antibodies against A p . This method was successful in transgenic mice, but failed during clinical trials due to the incidence of meningio-encephalitis 2 5 6 > 2 5 5 . It has previously been established that there is increased permeability in the B B B of Tg2576 A D mice compared to age match controls at 10 months of age, as the signs of the disease become manifest, and as early as 4 months of age, prior to disease onset and plaque deposition 1 4 9 . Therefore, it was hypothesized that disruption of the B B B is another pathological hallmark of A D that may explain the mechanism by which immunization therapies have proven successful. Specifically', peripheral antibodies are able to pass into the brain via a breached barrier and bind to and sequester brain amyloid. The antibody-amyloid complex is then degraded by F c R mediated phagocytosis by activated microglia. The present study addressed the effect of A p immunization on the integrity of the B B B and revealed that immunization with A P , prior to and after disease onset, resulted in a decrease in permeability and a restoration of B B B integrity. These observations provide an intellectual framework for understanding the efficiency of 134 vaccination to modify disease progression as well as providing new possible therapeutic interventions. It is clear that there are several directions this project could take in the future. For instance: studies focused on further establishing p97 as a biomarker for A D and as an indicator for drug efficacy in an in vitro and in vivo system could be addressed. Measuring p97 levels in primary mouse and/or human microglia with or without exposure to A p and in A D model mice treated with various putative drug treatments would clarify i f indeed p97 can be used as a monitoring system. If successful, this assay could serve as a means to test new drug development initially in an in vitro system and, i f efficacious, move into an in vivo system before clinical trials. In order to do these experiments monoclonal antibodies to mouse p97 need to be generated, ln addition, it is possible that, since p97 is an iron transport molecule, it may play a role in disease progression. Therefore, determining i f p97 contributes to disease pathology could be investigated. One way to address this hypothesis would be by creating transgenic mice overexpressing p97 and breeding these mice to h A P P mice and examining the brains for signs of A D pathology. In regards to the role of activated microglia in A p accumulation and plaque formation, it was found that the amount of A P plaques in h A P P mice with dysfunctional microglia was similar to the amount in hAPP mice. However, the number of animals used in this study was small and since there is variability between mice, a larger cohort is needed in order to draw a more definite conclusion. Moreover, since CSF-1 deficient mice have a short life span it would be beneficial to use an A D mouse model that develops pathological symptoms at an earlier age than the Tg2576 mouse model. A s 135 well , it would be interesting to assess the other hallmarks of A D in these mice, namely neurodegeneration, vascular amyloid burden and B B B permeability. Furthermore, measuring the levels of certain complement proteins and other indicators of microglial activation would be of interest to see i f AP had a direct or indirect effect on microglial maturation and function. It was demonstrated in this thesis that A p immunization restores B B B integrity in mice immunized after disease onset and possibly prevents B B B deterioration when immunized prior to disease onset. In this thesis, the permeability of the B B B was investigated on a global level. It would be interesting to see i f there is a difference in permeability in specific brain regions, in particular, regions of the brain more severely affected in A D and i f the decrease in B B B permeability coincides with the decrease in A p burden. Some studies have shown the occurrence of micro-hemorrhages in the brain after immunization. However, these micro-hemorrhages may have been due to the type of anti-Ap antibody used as an immunogen. It is possible that immunization could result in small hemorrhages in specific areas of the brain but an overall restoration of B B B integrity. This needs to be investigated more thoroughly. Dissecting the brains of immunized mice into various regions and performing the Evans blue assay is a quantitative way of assessing regional permeability. Alternatively, qualitative analysis using sections from Evans blue perfused animals can be done since Evans blue fluoresces under specific light wavelengths. Recently, Ujiie, et al. developed a new technique for assessing B B B permeability using succinimidyl ester of carboxyfluorescein diacetate 1 4 9 . Perfusing immunized mice with succinimidyl ester of carboxyfluorescein diacetate is another way to assess regional permeability. Moreover, it w i l l allow one to determine i f 136 the B B B is more permeable in plaque laden regions. Finally, it would be interesting to examine which cells in the brain are responsible for the damage and re-establishment on the B B B . Overall, there is an important link between microglial activation, plaque formation and degradation and the progression of other pathological hallmarks, which needs to be clearly elucidated. Determining the mechanism of A p clearance in the normal brain and upon A p immunization is required in order to facilitate the design of specific treatment regimens, allowing exclusive targeting of plaques without inducing detrimental side effects. Such experiments are difficult since transgenic mice have less genetic variability than humans, and their plaques have a different chemical composition, making them far more soluble and easier to remove. Furthermore, there is no transgenic mouse to date that can clearly mimic all the hallmarks of A D . In regards to immunization regimes, the consequences are different between human and mouse. Vaccination of transgenic mice removes human A p while leaving endogenous mouse A p intact, whereas in humans the immune response is directed against an endogenous target that occurs naturally and plays an essential role in maintaining healthy brain tissue. In conclusion the results from this thesis shed significant new light on activated microglial gene expression in A D and also on the role of microglia in plaque development and clearance possibly due to alterations in B B B integrity. 137 Appendix I: D o m a i n S t ruc tu re of A P P signal acidic KPl peptide Cys-och domain domain W | 1 | l i f e • • CuBDZnBD Thr-rich OX-2 -secretase Schematic representation of the domain structure of APP. Full length A P P contains a signal peptide at the N-terminal end; a cysteine residue-rich region (Cys-rich) with copper binding (CuBD) and zinc binding sites (ZnBD) ; an acidic domain; a threonine residue-rich domain; Kinitz protease inhibitor domain (KPI); and an O X - 2 homology domain ( O X 2 ) . The A p fragment is flanked by the P- and y-secretase cleavage sites and contains the a-sectretase cleavage site. 138 Appendix II: R e g i o n a l d i a g r a m of the b ra in Parietal Lobe Frontal Lobe Occipital Lobe Hippodampus CA1 Region Dentate G y r u s Region References 1. Turner, P. R., O'Connor, K . , Tate, W. P. & Abraham, W. C. Roles of amyloid precursor protein and its fragments in regulating neural activity, plasticity and memory. Prog Neurobiol 70, 1 -32 (2003). 2. Verbeek, M . M . , Ruiter, D . J. & de Waal, R. M . The role of amyloid in the pathogenesis of Alzheimer's disease. Biol Chem 378, 937-50 (1997). 3. Selkoe, D . J. & Schenk, D . Alzheimer's disease: molecular understanding predicts amyloid-based therapeutics. Annu Rev Pharmacol Toxicol 43, 545-84 (2003). 4. Small, D . H . The role of the amyloid protein precursor (APP) in Alzheimer's disease: does the normal function of A P P explain the topography of neurodegeneration? Neurochem Res 23, 795-806 (1998). 5. Price, D . L . , Tanzi, R. E . , Borchelt, D . R. & Sisodia, S. S. Alzheimer's disease: genetic studies and transgenic models. Annu Rev Genet 32, 461-93 (1998). 6. Nunan, J. & Small, D . H . Regulation of A P P cleavage by alpha-, beta- and . gamma-secretases. FEBS Lett 483, 6-10 (2000). 7. Allsop, D . , Landon, M . & Kidd , M . The isolation and amino acid composition of senile plaque core protein. Brain Res 259, 348-52 (1983). 8. Hardy, J. A . & Higgins, G . A . Alzheimer's disease: the amyloid cascade hypothesis. Science 256, 184-5 (1992). 9. Tanzi, R. E . A genetic dichotomy model for the inheritance of Alzheimer's disease and common age-related disorders. J Clin Invest 104, 1175-9 (1999). 10. Tanzi, R. E . & Bertram, L. New frontiers in Alzheimer's disease genetics. Neuron 32, 181-4 (2001). 11. Wisniewski, K . E . , Wisniewski, H . M . & Wen, G . Y . Occurrence of neuropathological changes and dementia of Alzheimer's disease in Down's syndrome. Ann Neurol 17, 278-82 (1985). 12. Rumble, B . et al. Amylo id A 4 protein and its precursor in Down's syndrome and Alzheimer's disease. NEnglJMed320, 1446-52 (1989). 13. Tanzi, R. E . Neuropathology in the Down's syndrome brain. Nat Med 2, 31 -2 (1996). 14. Suzuki, N . et al. A n increased percentage of long amyloid beta protein secreted by familial amyloid beta protein precursor (beta APP717) mutants. Science 264, 1336-40(1994). 15. Cai , X . D . , Golde, T. E . & Younkin, S. G. Release of excess amyloid beta protein from a mutant amyloid beta protein precursor. Science 259, 514-6 (1993). 16. Selkoe, D . J. Alzheimer's disease: genes, proteins, and therapy. Physiol Rev 81, 741-66(2001). 17. Wolfe, M . S. et al. Two transmembrane aspartates in presenilin-1 required for presenilin endoproteolysis and gamma-secretase activity. Nature 398, 513-7 (1999). 18. Luo, W. J. et al. PEN-2 and APH-1 coordinately regulate proteolytic processing of presenilin \ .JBiol Chem 278, 7850-4 (2003). 140 19. Gu, Y. et al. APH-1 interacts with mature and immature forms of presenilins and nicastrin and may play a role in maturation of presenilin.nicastrin complexes. J Biol Chem 278, 7374-80 (2003). 20. Kimberly, W. T. et al. Gamma-secretase is a membrane protein complex comprised of presenilin, nicastrin, Aph-1, and Pen-2. Proc Natl Acad Sci U S A 100, 6382-7 (2003). 21. Selkoe, D. J. The cell biology of beta-amyloid precursor protein and presenilin in Alzheimer's disease. Trends Cell Biol 8, 447-53 (1998). 22. Russo, C. et al. Presenilin-1 mutations in Alzheimer's disease. Nature 405, 531-2 (2000). 23. Strittmatter, W. J. et al. Apolipoprotein E: high-avidity binding to beta-amyloid and increased frequency of type 4 allele in late-onset familial Alzheimer disease. Proc Natl Acad Sci USA 90, 1977-81 (1993). 24. Mahley, R- W. Apolipoprotein E: cholesterol transport protein with expanding role in cell biology. Science 240, 622-30 (1988). 25. LaDu, M. J. et al. Isoform-specific binding of apolipoprotein E to beta-amyloid. J Biol Chem 269, 23403-6 (1994). 26. Akiyama, H. et al. Inflammation and Alzheimer's disease. Neurobiol Aging 21, 383-421 (2000). 27. Selkoe, D. J. Clearing the brain's amyloid cobwebs. Neuron 32, 177-80 (2001). 28. McGeer, P. L. & McGeer, E. G. Inflammation, autotoxicity and Alzheimer disease. Neurobiol Aging 22, 799-809 (2001). 29. Yoshikai, S., Sasaki, H., Doh-ura, K., Furuya, H. & Sakaki, Y. Genomic organization of the human amyloid beta-protein precursor gene. Gene 87, 257-63 (1990). 30. Tanaka, S. et al. Tissue-specific expression of three types of beta-protein precursor mRNA: enhancement of protease inhibitor-harboring types in Alzheimer's disease brain. Biochem Biophys Res Commun 165, 1406-14 (1989). 31. Storey, E., Beyreuther, K. & Masters, C. L. Alzheimer's disease amyloid precursor protein on the surface of cortical neurons in primary culture co-localizes with adhesion patch components. Brain Res 735, 217-31 (1996). 32. Annaert, W. G. et al. Interaction with telencephalin and the amyloid precursor protein predicts a ring structure for presenilins. Neuron 32, 579-89 (2001). 33. Allinson, T. M., Parkin, E. T., Turner, A. J. & Hooper, N . M. ADAMs family members as amyloid precursor protein alpha-secretases. J Neurosci Res 74, 342-52 (2003). 34. Ling, Y., Morgan, K. & Kalsheker, N . Amyloid precursor protein (APP) and the biology of proteolytic processing: relevance to Alzheimer's disease. Int J Biochem Cell Biol 35, 1505-35 (2003). 35. Vassar, R. et al. Beta-secretase cleavage of Alzheimer's amyloid precursor protein by the transmembrane aspartic protease BACE. Science 286, 735-41 (1999). 36. Cai, H. et al. BACE1 is the major beta-secretase for generation of Abeta peptides by neurons. Nat Neurosci 4, 233-4 (2001). 37. Acquati, F. et al. The gene encoding DRAP (BACE2), a glycosylated transmembrane protein of the aspartic protease family, maps to the down critical region. FEBS Lett 468, 59-64 (2000). 141 38. Duff, K. et al. Increased amyloid-beta42(43) in brains of mice expressing mutant presenilin 1. Nature 383, 710-3 (1996). 39. LaVoie, M. J. et al. Assembly of the gamma-secretase complex involves early formation of an intermediate subcomplex of Aph-1 and nicastrin. J Biol Chem 278, 37213-22 (2003). 40. Roher, A. E. et al. Oligomerizaiton and fibril assembly of the amyloid-beta protein. Biochim Biophys Acta 1502, 31-43 (2000). 41. Small, D. H. & McLean, C. A. Alzheimer's disease and the amyloid beta protein: What is the role of amyloid? J Neurochem 73, 443-9 (1999). 42. Braak, H. & Braak, E. Development of Alzheimer-related neurofibrillary changes in the neocortex inversely recapitulates cortical myelogenesis. Acta Neuropathol (Berl) 92, 197-201 (1996). 43. Gabuzda, D., Busciglio, J., Chen, L. B., Matsudaira, P. & Yankner, B. A. Inhibition of energy metabolism alters the processing of amyloid precursor protein and induces a potentially amyloidogenic derivative. J Biol Chem 269, 13623-8 (1994). 44. Mattson, M. P. & Pedersen, W. A. Effects of amyloid precursor protein derivatives and oxidative stress on basal forebrain cholinergic systems in Alzheimer's disease. Int J Dev Neurosci 16, 737-53 (1998). 45. Atwood, C. S. et al. Amyloid-beta: a chameleon walking in two worlds: a review of the trophic and toxic properties of amyloid-beta. Brain Res Brain Res Rev 43, 1-16 (2003). 46. Cherny, R. A. et al. Aqueous dissolution of Alzheimer's disease Abeta amyloid deposits by biometal depletion. J Biol Chem 214, 23223-8 (1999). 47. Querfurth, H. W. & Selkoe, D. J. Calcium ionophore increases amyloid beta peptide production by cultured cells. Biochemistry 33, 4550-61 (1994). 48. Ueda, K., Shinohara, S., Yagami, T., Asakura, K. & Kawasaki, K. Amyloid beta protein potentiates Ca2+ influx through L-type voltage-sensitive Ca2+ channels: a possible involvement of free radicals. J Neurochem 68, 265-71 (1997). 49. Pierrot, N., Ghisdal, P., Caumont, A. S. & Octave, J. N. Intraneuronal amyloid-beta 1-42 production triggered by sustained increase of cytosolic calcium concentration induces neuronal death. J Neurochem 88,1140-50 (2004). 50. Wertkin, A. M. et al. Human neurons derived from a teratocarcinoma cell line express solely the 695-amino acid amyloid precursor protein and produce intracellular beta-amyloid or A4 peptides. Proc Natl Acad Sci USA 90, 9513-7 (1993). 51. Takahashi, R. H. et al. Intraneuronal Alzheimer abeta42 accumulates in multivesicular bodies and is associated with synaptic pathology. Am J Pathol 161, 1869-79 (2002). 52. Echeverria, V. et al. Rat transgenic models with a phenotype of intracellular Abeta accumulation in hippocampus and cortex. JAlzheimers Dis 6, 209-19 (2004). 53. Takahashi, R. H. et al. Oligomerization of Alzheimer's beta-amyloid within processes and synapses of cultured neurons and brain. J Neurosci 24, 3592-9 (2004). 142 54. Dickson, D . W. Apoptotic mechanisms in Alzheimer neurofibrillary degeneration: cause or effect? J Clin Invest 114, 23-7 (2004). 55. Schultz, D . R. & Harrington, W. J . , Jr. Apoptosis: programmed cell death at a molecular level. Semin Arthritis Rheum 32, 345-69 (2003). 56. Lustbader, J. W . et al. A B A D directly links Abeta to mitochondrial toxicity in Alzheimer's disease. Science 304, 448-52 (2004). 57. McDonald, D . R., Bamberger, M . E . , Combs, C. K . & Landreth, G. E . beta-Amylo id fibrils activate parallel mitogen-activated protein kinase pathways in microglia and THP1 monocytes. J Neurosci 18, 4451-60 (1998). 58. McDonald, D . R., Brunden, K. R. & Landreth, G . E . Amylo id fibrils activate tyrosine kinase-dependent signaling and superoxide production in microglia. J Neurosci 17, 2284-94 (1997). 59. Verdier, Y . , Zarandi, M . & Penke, B . Amyloid beta-peptide interactions with neuronal and glial cell plasma membrane: binding sites and implications for Alzheimer's disease. J'Pept Sci 10, 229-48 (2004). 60. Paresce, D . M . , Ghosh, R. N . & Maxfield, F. R. Microgl ial cells internalize aggregates of the Alzheimer's disease amyloid beta-protein via a scavenger receptor. Neuron 17, 553-65 (1996). 61. Husemann, j . , Loike, J. D . , Anankov, R., Febbraio, M . & Silverstein, S. C. Scavenger receptors in neurobiology and neuropathology: their role on microglia and other cells of the nervous system. Glia 40, 195-205 (2002). 62. Grewal, R. P., Yoshida, T., Finch, C. E. & Morgan, T. E . Scavenger receptor m R N A s in rat brain microglia are induced by kainic acid lesioning and by cytokines. Neuroreport 8, 1077-81 (1997). 63. Bamberger, M . E., Harris, M . E . , McDonald,"D. R., Husemann, J. & Landreth, G. E . A cell surface receptor complex for fibrillar beta-amyloid mediates microglial activation. J Neurosci 23, 2665-74 (2003). 64. Christie, R. H . , Freeman, M . & Hyman, B . T. Expression of the macrophage scavenger receptor, a multifunctional lipoprotein receptor, in microglia associated with senile plaques in Alzheimer's disease. Am J Pathol 148, 399-403 (1996). 65. Cu i , Y . , Le, Y . , Yazawa, H . , Gong, W. & Wang, J. M . Potential role of the formyl peptide receptor-like 1 (FPRL1) in inflammatory aspects of Alzheimer's disease. J Leukoc Biol 72, 628-35 (2002). 66. Yazawa, H . et al. Beta amyloid peptide (Abeta42) is internalized via the G -.protein-coupled receptor FPRL1 and forms fibrillar aggregates in macrophages. Faseb J15, 2454-62 (2001). 67. Yan, S. D . et al. Receptor-dependent cell stress and amyloid accumulation in systemic amyloidosis. Nat Med 6, 643-51 (2000). 68. Yan , S. D . et al. R A G E and amyloid-beta peptide neurotoxicity in Alzheimer's disease. Nature 382, 685-91 (1996). 69. Deane, R. et al. R A G E mediates amyloid-beta peptide transport across the blood-brain barrier and accumulation in brain. Nat Med 9, 907-13 (2003). 70. Shibata, M . et al. Clearance of Alzheimer's amyloid-ss(l-40) peptide from brain by L D L receptor-related protein-1 at the blood-brain barrier. J Clin Invest 106, 1489-99(2000). 143 71. Zlokovic, B . V . Clearing amyloid through the blood-brain barrier. J Neurochem 89,807-11 (2004). 72. Kang, D . E . et al. Modulation of amyloid beta-protein clearance and Alzheimer's disease susceptibility by the L D L receptor-related protein pathway. J Clin Invest 106, 1159-66 (2000). 73. Van Uden, E . et al. Increased extracellular amyloid deposition and neurodegeneration in human amyloid precursor protein transgenic mice deficient in receptor-associated protein. J Neurosci 22, 9298-304 (2002). 74. Marx, J. Major setback for Alzheimer's models. Science 255, 1200-2 (1992). 75. Quon, D . et al. Formation of beta-amyloid protein deposits in brains of transgenic mice. Nature 352, 239-41 (1991). 76. Andra, K. et al. Expression of A P P in transgenic mice: a comparison of neuron-specific promoters. Neurobiol Aging 17, 183-90 (1996). 77. Borchelt, D . R. et al. A vector for expressing foreign genes in the brains and hearts of transgenic mice. Genet Anal 13, 159-63 (1996). 78. Kulnane, L. S. & Lamb, B . T. Neuropathological characterization of mutant amyloid precursor protein yeast artificial chromosome transgenic mice. Neurobiol Dis 8, 982-92 (2001). 79. Games, D . et al. Alzheimer-type neuropathology in transgenic mice overexpressing V717F beta-amyloid precursor protein. Nature 373, 523-7 (1995). 80. Irizarry, M . C. et al. Abeta deposition is associated with neuropil changes, but not with overt neuronal loss in the human amyloid precursor protein V717F (PDAPP) transgenic mouse. J Neurosci 17, 7053-9 (1997). 81. Dodart, J. C. et al. Behavioral disturbances in transgenic mice overexpressing the V717F beta-amyloid precursor protein. Behav Neurosci 113, 982-90 (1999). 82. Chen, G . et al. A learning deficit related to age and beta-amyloid plaques in a mouse model of Alzheimer's disease. Nature 408, 975-9 (2000). 83. Irizarry, M . C , McNamara, M . , Fedorchak, K . , Hsiao, K . & Hyman, B . T. A P P S w transgenic mice develop age-related A beta deposits and neuropil abnormalities, but no neuronal loss in C A 1 . JNeuropathol Exp Neurol 56, 965-73 (1997). 84. Hsiao, K . K. et al. Age-related C N S disorder and early death in transgenic F V B / N mice overexpressing Alzheimer amyloid precursor proteins. Neuron 15, 1203-18 (1995). 85. Hsiao, K. et al. Correlative memory deficits, Abeta elevation, and amyloid plaques in transgenic mice. Science 274, 99-102 (1996). 86. McGowan, E . et al. Amylo id phenotype characterization of transgenic mice overexpressing both mutant amyloid precursor protein and mutant presenilin 1 transgenes. Neurobiol Dis 6, 231-44 (1999). 87. Chapman, P. F. et al. Impaired synaptic plasticity and learning in aged amyloid precursor protein transgenic mice. Nat Neurosci 2, 271-6 (1999). 88. Schauwecker, P. E . & Steward, O. Genetic determinants of susceptibility to excitotoxic cell death: implications for gene targeting approaches. Proc Natl Acad Sci USA 94, 4103-8 (1997). 144 89. Sturchler-Pierrat, C. et al. Two amyloid precursor protein transgenic mouse models with Alzheimer disease-like pathology. Proc Natl Acad Sci USA 94, 13287-92(1997). 90. Stalder, M . et al. Association of microglia with amyloid plaques in brains of APP23 transgenic mice. Am J Pathol 154,1673-84 (1999). 91. Kel ly, P. H. et al. Progressive age-related impairment of cognitive behavior in APP23 transgenic mice. Neurobiol Aging 24, 365-78 (2003). 92. Calhoun, M . E . et al. Neuron loss in A P P transgenic mice. Nature 395, 755-6 (1998) . 93. Calhoun, M . E . et al. Neuronal overexpression of mutant amyloid precursor protein results in prominent deposition of cerebrovascular amyloid. Proc Natl Acad Sci USA 96, 14088-93 (1999). 94. Holcomb, L . et al. Accelerated Alzheimer-type phenotype in transgenic mice carrying both mutant amyloid precursor protein and presenilin 1 transgenes. Nat Med 4, 97-100(1998). 95. Carlson, G . A . et al. Genetic modification of the phenotypes produced by amyloid precursor protein overexpression in transgenic mice. Hum Mol Genet 6, 1951-9 (1997). 96. Moechars, D . , Lorent, K , De Strooper, B . , Dewachter, I. & Van Leuven, F. Expression in brain of amyloid precursor protein mutated in the alpha-secretase site causes disturbed behavior, neuronal degeneration and premature death in transgenic mice. Embo J15, 1265-74.(1996). 97. Moechars, D . , Gil is , M . , Kuiperi, C , Laenen, I. & Van Leuven, F. Aggressive behaviour in transgenic mice expressing A P P is alleviated by serotonergic drugs. Neuroreport 9,3561-4 (1998). 98. Moechars, D . et al. Early phenotypic changes in transgenic mice that overexpress different mutants of amyloid precursor protein in brain. J Biol Chem 274, 6483-92 (1999) . 99. Janus, C , Chishti, M . A . & Westaway, D . Transgenic mouse models of Alzheimer's disease. Biochim Biophys Acta 1502, 63-75 (2000). 100. Janus, C. & Westaway, D . Transgenic mouse models of Alzheimer's disease. Physiol Behav 73, 873-86 (2001). 101. Gotz, J. et al. Transgenic animal models of Alzheimer's disease and related disorders: histopathology, behavior and therapy. Mol Psychiatry (2004). 102. McGeer, P. L . , Schulzer, M . & McGeer, E . G . Arthritis and anti-inflammatory agents as possible protective factors for Alzheimer's disease: a review of 17 epidemiologic studies. Neurology 47, 425-32 (1996). 103. Stewart, W. F., Kawas, C , Corrada, M . & Metter, E . J. Risk of Alzheimer's disease and duration of N S A I D use. Neurology 48, 626-32 (1997). 104. Streit, W. J., Walter, S. A . & Pennell, N . A . Reactive microgliosis. Prog Neurobiol 57, 563-81 (1999). 105. Nelson, P. T., Soma, L. A . & Lavi , E . Microglia in diseases of the central nervous system. Ann Med 34, 491-500 (2002). 106. Perry, V . H. & Gordon, S. Macrophages and the nervous system. Int Rev Cytol 125, 203-44(1991). 145 107. Kaur, C , Hao, A . J., Wu , C. H . & Ling, E . A . Origin of microglia. Microsc Res Tech 54, 2-9 (2001). 108. Streit, W. J. & Graeber, M . B . Heterogeneity of microglial and perivascular cell populations: insights gained from the facial nucleus paradigm. Glia 7, 68-74 (1993). 109. Streit, W. J. Microglia as neuroprotective, immunocompetent cells of the C N S . Glia 40, 133-9 (2002). 110. Jefferies, W. A . et al. Reactive microglia specifically associated with amyloid plaques in Alzheimer's disease brain tissue express melanotransferrin. Brain Res 712,122-6(1996). 111. Yamada, T. et al. Melanotransferrin is produced by senile plaque-associated reactive microglia in Alzheimer's disease. Brain Res 845, 1-5 (1999). 112. Schipper, H . M . et al. Evaluation of heme oxygenase-1 as a systemic biological marker of sporadic A D . Neurology 54, 1297-304 (2000). 113. Walker, D . G . , Lue, L . F. & Beach, T. G . Increased expression of the urokinase plasminogen-activator receptor in amyloid beta peptide-treated human brain microglia and in A D brains. Brain Res 926, 69-79 (2002). 114. Espey, M . G . , Chernyshev, O. N . , Reinhard, J. F., Jr., Namboodiri, M . A . & Colton, C. A . Activated human microglia produce the excitotoxin quinolinic acid. Neuroreport 8, 431-4 (1997). 115. Piani, D . , Spranger, M . , Frei, K . , Schaffner, A . & Fontana, A . Macrophage-induced cytotoxicity of N-methyl-D-aspartate receptor positive neurons involves excitatory amino acids rather than reactive oxygen intermediates and cytokines. Eur J Immunol 22, 2429-36 (1992). 116. McGeer, P. L . , Akiyama, H . , Itagaki, S. & McGeer, E . G . Immune system response in Alzheimer's disease. Can J Neurol Sci 16, 516-27 (1989). 117. Styren, S. D . , Civ in , W. H . & Rogers, J. Molecular, cellular, and pathologic characterization of H L A - D R immunoreactivity in normal elderly and Alzheimer's disease brain. Exp Neurol 110, 93-104 (1990). 118. Frautschy, S. A . et al. Microglial response to amyloid plaques in APPsw transgenic mice. Am J Pathol 152, 307-17 (1998). 119. Itagaki, S., McGeer, P. L , Akiyama, H . , Zhu, S. & Selkoe, D . Relationship of microglia and astrocytes to amyloid deposits of Alzheimer disease. J Neuroimmunol 24, 173-82(1989). 120. Mackenzie, I. R., Hao, C. & Munoz, D. G . Role of microglia in senile plaque formation. Neurobiol Aging 16, 797-804 (1995). 121. Sasaki, A . , Yamaguchi, H . , Ogawa, A . , Sugihara, S. & Nakazato, Y . Microglial activation in early stages of amyloid beta protein deposition. Acta Neuropathol (Berl) 94, 316-22 (1997). 122. Ard , M . D . , Cole, G . M . , Wei , J., Mehrle, A . P. & Fratkin, J. D . Scavenging of Alzheimer's amyloid beta-protein by microglia in culture. J Neurosci Res 43, 190-202(1996). 123. Paresce, D . M . , Chung, H . & Maxfield, F. R. Slow degradation of aggregates of the Alzheimer's disease amyloid beta-protein by microglial cells. J Biol Chem 272, 29390-7 (1997). 146 124. Frackowiak, J. et al. Ultrastructure of the microglia that phagocytose amyloid and the microglia that produce beta-amyloid fibrils. Acta Neuropathol (Bed) 84, 225-33 (1992). 125. Akiyama, F f . et al. Granules in glial cells of patients with Alzheimer's disease are immunopositive for C-terminal sequences of beta-amyloid protein. Neurosci Lett 206, 169-72(1996). 126. Rogers, J. et al. Elucidating molecular mechanisms of Alzheimer's disease in microglial cultures. Ernst Schering Res Found Workshop, 25-44 (2002). 127. Chung, H . , Brazi l , M . I., Soe, T. T. & Maxfield, F. R. Uptake, degradation, and release of fibrillar and soluble forms of Alzheimer's amyloid beta-peptide by microglial cells. J Biol Chem 274, 32301-8 (1999). 128. Combs, C. K , Johnson, D . E . , Cannady, S. B . , Lehman, T. M . & Landreth, G . E . Identification of microglial signal transduction pathways mediating a neurotoxic response to amyloidogenic fragments of beta-amyloid and prion proteins. J Neurosci 19, 928-39 (1999). 129. Savage, M . J., L i n , Y . G . , Ciallella, J. R., Flood, D . G . & Scott, R. W. Activation of c-Jun N-terminal kinase and p38 in an Alzheimer's disease model is associated with amyloid deposition. J Neurosci 22, 3376-85 (2002). 130. Koistinaho, M . & Koistinaho, J. Role of p38 and p44/42 mitogen-activated protein kinases in microglia. Glia 40, 175-83 (2002). 131. Hensley, K. et al. p38 kinase is activated in the Alzheimer's disease brain. J Neurochem 72, 2053-8 (1999). 132. X i e , Z. , Smith, C. J. & Van Eldik, L . J. Activated glia induce neuron death via M A P kinase signaling pathways involving J1MK and p38. Glia 45, 170-9 (2004). 133. Pyo, H . , Jou, I., Jung, S., Hong, S. & Joe, E . H . Mitogen-activated protein kinases activated by lipopolysaccharide and beta-amyloid in cultured rat microglia. Neuroreport 9, 871-4 (1998). 134. Ballabh, P., Braun, A . & Nedergaard, M . The blood-brain barrier: an overview: structure, regulation, and clinical implications. Neurobiol Dis 16, 1-13 (2004). 135. Hirase, T. et al. Occludin as a possible determinant of tight junction permeability in endothelial cells. J Cell Sci 110 ( Pt 14), 1603-13 (1997). 136. Prat, A . , Biernacki, K . , Wosik, K . & Antel, J. P. Gl ia l cell influence on the human blood-brain barrier. Glia 36, 145-55 (2001). 137. Janzer, R. C. & Raff, M . C. Astrocytes induce blood-brain barrier properties in endothelial cells. Nature 325, 253-7 (1987). 138. Ramsauer, M . , Krause, D . & Dermietzel, R. Angiogenesis of the blood-brain barrier in vitro and the function of cerebral pericytes. Faseb J16, 1274-6 (2002). 139. Kern, T. S. & Engerman, R. L . Capillary lesions develop in retina rather than cerebral cortex in diabetes and experimental galactosemia. Arch Ophthalmol 114, 306-10(1996). 140. Lindahl, P., Johansson, B . R., Leveen, P. & Betsholtz, C. Pericyte loss and microaneurysm formation in PDGF-B-deficient mice. Science 211, 242-5 (1997). 141. Saunders, N . R., Habgood, M . D . & Dziegielewska, K . M . Barrier mechanisms in the brain, I. Adult brain. Clin Exp Pharmacol Physiol 26, 11-9 (1999). 142. Kastin, A . J., Pan, W. , Maness, L . M . & Banks, W. A . Peptides crossing the blood-brain barrier: some unusual observations. Brain Res 848, 96-100 (1999). 147 143. Jolliet-Riant, P. & Tillement, J. P. Drug transfer across the blood-brain barrier and improvement of brain delivery. Fundam Clin Pharmacol 13, 16-26 (1999). 144. Rothenberger, S. et al. Coincident expression and distribution of melanotransferrin and transferrin receptor in human brain capillary endothelium. Brain Res 712, 117-21 (1996). 145. Claudio, L. Ultrastructural features of the blood-brain barrier in biopsy tissue from Alzheimer's disease patients. Acta Neuropathol (Bed) 91, 6-14 (1996). 146. Caserta, M. T., Caccioppo, D., Lapin, G. D., Ragin, A. & Groothuis, D. R. Blood-brain barrier integrity in Alzheimer's disease patients and elderly control subjects. J Neuropsychiatry Clin Neurosci 10, 78-84 (1998). 147. Poduslo, J. F., Curran, G. L., Wengenack, T. M., Malester, B. & Duff, K. Permeability of proteins at the blood-brain barrier in the normal adult mouse and double transgenic mouse model of Alzheimer's disease. Neurobiol Dis 8, 555-67 (2001). 148. Perry, V. H., Anthony, D. C , Bolton, S. J . & Brown, H. C. The blood-brain barrier and the inflammatory response. Mol Med Today 3, 335-41 (1997). 149. Ujiie, M. , Dickstein, D. L., Carlow, D. A. & Jefferies, W. A. Blood-brain barrier permeability precedes senile plaque formation in an Alzheimer disease model. Microcirculation 10, 463-70 (2003). 150. Xu, J. et al. Amyloid beta peptide-induced cerebral endothelial cell death involves mitochondrial dysfunction and caspase activation. J Cereb Blood Flow Metab 21, 702-10(2001). 151. Yin, K. J., Lee, J . M. , Chen, S. D., Xu, J. & Hsu, C. Y. Amyloid-beta induces Smac release via AP-l/Bim activation in cerebral endothelial cells. J Neurosci 22, 9764-70 (2002). 152. Vagnucci, A. H., Jr. & Li, W. W. Alzheimer's disease and angiogenesis. Lancet 361, 605-8 (2003). 153. Wong, M. L. et al. Inducible nitric oxide synthase gene expression in the brain during systemic inflammation. Nat Med 2, 581-4 (1996). 154. Ek, M. et al. Inflammatory response: pathway across the blood-brain barrier. Nature 4W, 430-1 (2001). 155. Uchikado, H. et al. Activation of vascular endothelial cells and perivascular cells by systemic inflammation-an immunohistochemical study of postmortem human brain tissues. Acta Neuropathol (Berl) 107, 341-51 (2004). 156. Attems, J., Lintner, F. & Jellinger, K. A. Amyloid beta peptide 1-42 highly correlates with capillary cerebral amyloid angiopathy and Alzheimer disease pathology. Acta Neuropathol (Berl) 107, 283-91 (2004). 157. Weller, R. O., Massey, A., Kuo, Y. M. & Roher, A. E. Cerebral amyloid angiopathy: accumulation of A beta in interstitial fluid drainage pathways in Alzheimer's disease. Ann N YAcad Sci 903, 110-7 (2000). 158. Vinters, H. V. Cerebral amyloid angiopathy. A critical review. Stroke 18, 311-24 (1987). 159. Bergeron, C , Ranalli, P. J. & Miceli, P. N. Amyloid angiopathy in Alzheimer's disease. Can J Neurol Sci 14, 564-9 (1987). 148 160. Prelli, F., Castano, E . , Glenner, G . G . & Frangione, B . Differences between vascular and plaque core amyloid in Alzheimer's disease. J Neurochem 51, 648-51 (1988). 161. Plassman, B . L. & Breitner, J. C. Recent advances in the genetics of Alzheimer's disease and vascular dementia with an emphasis on gene-environment interactions. J Am Geriatr Soc 44, 1242-50 (1996). 162. Munch, G . & Robinson, S. R. Potential neurotoxic inflammatory responses to Abeta vaccination in humans. J Neural Transm 109, 1081-7 (2002). 163. Nico l l , J. A . et al. Neuropathology of human Alzheimer disease after immunization with amyloid-beta peptide: a case report. Nat Med 9, 448-52 (2003). 164. Golde, T. E . Alzheimer disease therapy: can the amyloid cascade be halted? J Clin Invest 111, 11-8 (2003). 165. Luo, Y . et al. Mice deficient in B A C E 1 , the Alzheimer's beta-secretase, have normal phenotype and abolished beta-amyloid generation. Nat Neurosci 4, 231-2 (2001). 166. Wolozin, B . , Kellman, W. , Ruosseau, P., Celesia, G . G . & Siegel, G . Decreased prevalence of Alzheimer disease associated with 3-hydroxy-3-methyglutaryl coenzyme A reductase inhibitors. Arch Neurol 57, 1439-43 (2000). 167. Notkola, 1. L . et al. Serum total cholesterol, apolipoprotein E epsilon 4 allele, and Alzheimer's disease. Neuroepidemiology 17, 14-20 (1998). 168. Simons, M . et al. Cholesterol depletion inhibits the generation of beta-amyloid in hippocampal neurons. Proc Natl Acad Sci U S A 95, 6460-4 (1998). 169. Fassbender, K . et al. Simvastatin strongly reduces levels of Alzheimer's disease beta -amyloid peptides Abeta 42 and Abeta 40 in vitro and in vivo. Proc Natl Acad Sci US A 98, 5856-61 (2001). 170. Shie, F. S., Jin, L . W. , Cook, D . G . , Leverenz, J. B . & LeBoeuf, R. C. Diet-induced hypercholesterolemia enhances brain A beta accumulation in transgenic mice. Neuroreport 13, 455-9 (2002). 171. George, A . J. et al. A P P intracellular domain is increased and soluble Abeta is reduced with diet-induced hypercholesterolemia in a transgenic mouse model of Alzheimer disease. Neurobiol Dis 16, 124-32 (2004). 172. Refolo, L . M . et al. A cholesterol-lowering drug reduces beta-amyloid pathology in a transgenic mouse model of Alzheimer's disease. Neurobiol Dis 8, 890-9 (2001). 173. Friedhoff, L . T., Cullen, E . I., Geoghagen, N . S. & Buxbaum, J. D . Treatment with controlled-release lovastatin decreases serum concentrations of human beta-amyloid (A beta) peptide. Int J Neuropsychopharmacol 4, 127-30 (2001). 174. Weggen, S. et al. A subset of N S A I D s lower amyloidogenic Abeta42 independently of cyclooxygenase activity. Nature 414, 212-6 (2001). 175. Schenk, D . et al. Immunization with amyloid-beta attenuates Alzheimer-disease-like pathology in the P D A P P mouse. Nature 400, 173-7 (1999). 176. Bard, F. et al. Peripherally administered antibodies against amyloid beta-peptide enter the central nervous system and reduce pathology in a mouse model of Alzheimer disease. Nat Med 6, 916-9 (2000). 149 177. DeMattos, R. B . et al. Peripheral anti-A beta antibody alters C N S and plasma A beta clearance and decreases brain A beta burden in a mouse model of Alzheimer's disease. Proc Natl Acad Sci USA 98, 8850-5 (2001). 178. Morgan, D . et al. A beta peptide vaccination prevents memory loss in an animal model of Alzheimer's disease. Nature 408, 982-5 (2000). 179. Janus, C. et al. A beta peptide immunization reduces behavioural impairment and plaques in a model of Alzheimer's disease. Nature 408, 979-82 (2000). 180. Lemere, C. A . et al. Evidence for peripheral clearance of cerebral Abeta protein following chronic, active Abeta immunization in P S A P P mice. Neurobiol Dis 14, 10-8 (2003). 181. Oddo, S., Bill ings, L . , Kesslak, J. P., Cribbs, D . H . & LaFerla, F. M . Abeta Immunotherapy Leads to Clearance of Early, but Not Late, Hyperphosphorylated Tau Aggregates via the Proteasome. Neuron 43, 321-32 (2004). 182. Yoshida, H . et al. The murine mutation osteopetrosis is in the coding region of the macrophage colony stimulating factor gene. Nature 345, 442-4 (1990). 183. Wiktor-Jedrzejczak, W. et al. Total absence of colony-stimulating factor 1 in the macrophage-deficient osteopetrotic (op/op) mouse. Proc Natl Acad Sci USA 87, 4828-32 (1990). 184. Pollard, J. W. , Morgan, C. J., Dello Sbarba, P., Cheers, C. & Stanley, E . R. Independently arising macrophage mutants dissociate growth factor-regulated survival and proliferation. Proc Natl Acad Sci U S A 88, 1474-8 (1991). 185. Stanley, E . R. et al. Biology and action of colony—stimulating factor-1. Mol ReprodDev 46, 4-10 (1997). 186. Blevins, G . & Fedoroff, S. Microglia in colony-stimulating factor 1-deficient op/op mice. J Neurosci Res 40, 535-44 (1995). 187. Wegiel, J. et al. Reduced number and altered morphology of microglial cells in colony stimulating factor-1-deficient osteopetrotic op/op mice. Brain Res 804, 135-9(1998). 188. Marks, S. C , Jr. & Lane, P. W. Osteopetrosis, a new recessive skeletal mutation on chromosome 12 of the mouse. J Hered 67, 11-18 (1976). 189. Blasi , E . , Barluzzi, R., Bocchini, V . , Mazzolla, R. & Bistoni, F. Immortalization of murine microglial cells by a v-raf/v-myc carrying retrovirus. J Neuroimmunol 27, 229-37 (1990). 190. L i , M . , Pisalyaput, K . , Galvan, M . & Tenner, A . J. Macrophage colony stimulatory factor and interferon-gamma trigger distinct mechanisms for augmentation of beta-amyloid-induced microglia-mediated neurotoxicity. J Neurochem 91, 623-33 (2004). 191. Hampel, H . et al. Decreased soluble interleukin-6 receptor in cerebrospinal fluid of patients with Alzheimer's disease. Brain Res 780, 356-9 (1998). 192. Fil l i t , H . et al. Elevated circulating tumor necrosis factor levels in Alzheimer's disease. Neurosci Lett 129, 318-20 (1991). 193. Brown, J. P., Woodbury, R. G . , Hart, C. E . , Hellstrom, I. & Hellstrom, K. E . Quantitative analysis of melanoma-associated antigen p97 in normal and neoplastic tissues. Proc Natl Acad Sci USA 78, 539-43 (1981). 150 194. Woodbury, R. G . , Brown, J. P., Loop, S. M . , Hellstrom, K . E . & Hellstrom, I. Analysis of normal neoplastic human tissues for the tumor-associated protein p97. Int J Cancer 27, 145-9(1981). 195. Brown, J. P. et al. Human melanoma-associated antigen p97 is structurally and functionally related to transferrin. Nature 296, 171-3 (1982). 196. Real, F. X . et al. Class 1 (unique) tumor antigens of human melanoma: identification of unique and common epitopes on a 90-kDa glycoprotein. Proc Natl Acad Sci USAS5, 3965-9 (1988). 197. Sciot, R. et al. In situ localization of melanotransferrin (melanoma-associated antigen P97) in human liver. A light- and electronmicroscopic immunohistochemical study. Liver 9, 110-9 (1989). 198. Alemany, R. et al. Glycosyl phosphatidylinositol membrane anchoring of melanotransferrin (p97): apical compartmentalization in intestinal epithelial cells. J Cell Sci 104 ( Pt 4), 1155-62 (1993). 199. Food, M . R. et al. Transport and expression in human melanomas of a transferrin-like glycosylphosphatidylinositol-anchored protein. J Biol Chem 269, 3034-40 (1994). 200. Kennard, M . L . , Richardson, D . R., Gabathuler, R., Ponka, P. & Jefferies, W. A . A novel iron uptake mechanism mediated by GPI-anchored human p97. Embo J 14,4178-86(1995). 201. McNagny, K . M . , Rossi, F., Smith, G . & Graf, T. The eosinophil-specific cell surface antigen, EOS47, is a chicken homologue of the oncofetal antigen melanotransferrin. Blood 87, 1343-52 (1996). 202. Plowman, G . D . et al. Assignment of the gene for human melanoma-associated antigen p97 to chromosome 3. Nature 303, 70-2 (1983). 203. Baker, E . N . et al: Human melanotransferrin (p97) has only one functional iron-binding site. FEBS Lett 298, 215-8 (1992). 204. Jefferies, W. A . et al. Transferrin receptor on endothelium of brain capillaries. Nature 312, 162-3 (1984). 205. Loeffler, D . A . et al. Transferrin and iron in normal, Alzheimer's disease, and Parkinson's disease brain regions. J Neurochem 65, 710-24 (1995). 206. Grundke-Iqbal, I. et al. Ferritin is a component of the neuritic (senile) plaque in Alzheimer dementia. Acta Neuropathol (Berl) 81, 105-10 (1990). 207. Kaneko, Y . , Kitamoto, T., Tateishi, J. & Yamaguchi, K . Ferritin immunohistochemistry as a marker for microglia. Acta Neuropathol (Berl) 79, 129-36(1989). 208. Kennard, M . L . , Feldman, H . , Yamada, T. & Jefferies, W. A . Serum levels of the iron binding protein p97 are elevated in Alzheimer's disease. Nat Med 2, 1230-5 (1996). 209. Feldman, H . et al. Serum p97 levels as an aid to identifying Alzheimer's disease. J Alzheimers Dis 3, 507-516 (2001). 210. K i m , D . K . et al. Serum melanotransferrin, p97 as a biochemical marker of Alzheimer's disease. Neuropsychopharmacology 25, 84-90 (2001). 211. Qian, Z. M . & Wang, Q. Expression of iron transport proteins and excessive iron accumulation in the brain in neurodegenerative disorders. Brain Res Brain Res Rev 27, 257-67(1998). 151 212. Dwork, A . J., Schon, E . A . & Herbert, j . Nonidentical distribution of transferrin and ferric iron in human brain. Neuroscience 27, 333-45 (1988). 213. Sala, R. et al. The human melanoma associated protein melanotransferrin promotes endothelial cell migration and angiogenesis in vivo. Eur J Cell Biol 81, 599-607 (2002). 214. Kalaria, R. N . et al. Vascular endothelial growth factor in Alzheimer's disease and experimental cerebral ischemia. Brain Res Mol Brain Res 62, 101-5 (1998). 215. Tarkowski, E . et al. Increased intrathecal levels of the angiogenic factors V E G F and TGF-beta in Alzheimer's disease and vascular dementia. Neurobiol Aging 23, 237-43 (2002). 216. Yang, S. P. et al. Co-accumulation of vascular endothelial growth factor with beta-amyloid in the brain of patients with Alzheimer's disease. Neurobiol Aging 25, 283-90 (2004). 217. Roze-Heusse, A . , Houbiguian, M . L . , Debacker, C , Zakin, M . M . & Duchange, N . Melanotransferrin gene expression in melanoma cells is correlated with high levels of Jun/Fos family transcripts and with the presence of a specific A P 1 -dependent ternary complex. Biochem 7318 ( Pt 3), 883-8 (1996). 218. McGeer, E . G . & McGeer, P. L . Inflammatory processes in Alzheimer's disease. Prog Neuropsychopharmacol Biol Psychiatry 27, 741-9 (2003). 219. Walker, D . G . , Lue, L. F. & Beach, T. G . Gene expression profiling of amyloid beta peptide-stimulated human post-mortem brain microglia. Neurobiol Aging 22, 957-66 (2001). 220. Nakamasu, K . et al. Structure and promoter analysis of the mouse membrane-bound transferrin-like protein (MTf) gene. Eur J Biochem 268, 1468-76 (2001). 221. Ogawa, O. et al. Inhibition of inducible nitric oxide synthase gene expression by indomethacin or ibuprofen in beta-amyloid protein-stimulated J774 cells. Eur J Pharmacol 408, 137-41 (2000). 222. Blasko, I. et al. Ibuprofen decreases cytokine-induced amyloid beta production in neuronal cells. Neurobiol Dis 8, 1094-101 (2001). 223. Eriksen, J. L . et al. N S A I D s and enantiomers of flurbiprofen target gamma-secretase and lower Abeta 42 in vivo. J Clin Invest 112, 440-9 (2003). 224. L i m , G. P. et al. Ibuprofen suppresses plaque pathology and inflammation in a mouse model for Alzheimer's disease. J Neurosci 20, 5709-14 (2000). 225. Netland, E . E . , Newton, J. L . , Majocha, R. E . & Tate, B . A . Indomethacin reverses the microglial response to amyloid beta-protein. Neurobiol Aging 19, 201-4(1998). 226. Etminan, M . , G i l l , S. & Samii, A . Effect of non-steroidal anti-inflammatory drugs on risk of Alzheimer's disease: systematic review and meta-analysis of observational studies. Bmj 327, 128 (2003). 227. Weggen, S. et al. Evidence that nonsteroidal anti-inflammatory drugs decrease amyloid beta 42 production by direct modulation of gamma-secretase activity. J Biol Chem 278, 31831-7 (2003). 228. Matsuura, H . et al. Regulation of cyclooxygenase-2 by interferon gamma and transforming growth factor alpha in normal human epidermal keratinocytes and squamous carcinoma cells. Role of mitogen-activated protein kinases. J Biol Chem 274, 29138-48 (1999). 152 229: Wisniewski, H. M., Wegiel, J., Wang, K. C , Kujawa, M. & Lach, B. Ultrastructural studies of the cells forming amyloid fibers in classical plaques. Can J Neurol Sci 16, 535-42 (1989). 230. Wisniewski, H. M., Barcikowska, M. & Kida, E. Phagocytosis of beta/A4 amyloid fibrils of the neuritic neocortical plaques. Acta Neuropathol (Berl) 81, 588-90(1991). 231. Wegiel, J., Wang, K. C , Tarnawski, M. & Lach, B. Microglia cells are the driving force in fibrillar plaque formation, whereas astrocytes are a leading factor in plague degradation. Acta Neuropathol (Berl) 100, 356-64 (2000). 232. Brazil, M. I., Chung, H. & Maxfield, F. R. Effects of incorporation of immunoglobulin G and complement component Clq on uptake and degradation of Alzheimer's disease amyloid fibrils by microglia. J Biol Chem 275, 16941-7 (2000). 233. Webster, S. D. et al. Antibody-mediated phagocytosis of the amyloid beta-peptide in microglia is differentially modulated by Clq. J Immunol 166, 7496-503 (2001). 234. Rogers, J. et al. Complement activation by beta-amyloid in Alzheimer disease. Proc Natl Acad Sci USA 89, 10016-20 (1992). 235. Bradt, B. M., Kolb, W. P. & Cooper, N. R. Complement-dependent proinflammatory properties of the Alzheimer's disease beta-peptide. J Exp Med 188,431-8 (1998). 236. Wyss-Coray, T. et al. Alzheimer's disease-like cerebrovascular pathology in transforming growth factor-beta 1 transgenic mice and functional metabolic correlates. Ann N Y Acad Sci 903, 317-23 (2000). 237. Wyss-Coray, T. et al. Prominent neurodegeneration and increased plaque formation in complement-inhibited Alzheimer's mice. Proc Natl Acad Sci USA 99, 10837-42 (2002). 238. Fonseca, M. I., Zhou, J., Botto, M. & Tenner, A. J. Absence of Clq leads to less neuropathology in transgenic mouse models of Alzheimer's disease. J Neurosci 24, 6457-65 (2004). 239. Wiktor-Jedrzejczak, W. et al. Correction by CSF-1 of defects in the osteopetrotic op/op mouse suggests local, developmental, and humoral requirements for this growth factor. Exp Hematol 19, 1049-54 (1991). 240. Rogove, A. D., Lu, W. & Tsirka, S. E. Microglial activation and recruitment, but not proliferation, suffice to mediate neurodegeneration. Cell Death Differ 9, 801-6 (2002) . 241. Wyss-Coray, T. et al. Amyloidogenic role of cytokine TGF-betal in transgenic mice and in Alzheimer's disease. Nature 389, 603-6 (1997). 242. Lesne, S. et al. Transforming growth factor-beta 1 potentiates amyloid-beta generation in astrocytes and in transgenic mice. J Biol Chem 278, 18408-18 (2003) . 243. Naito, M. et al. Abnormal differentiation of tissue macrophage populations in 'osteopetrosis' (op) mice defective in the production of macrophage colony-stimulating factor. Am J Pathol 139, 657-67 (1991). 244. Witmer-Pack, M. D. et al. Identification of macrophages and dendritic cells in the osteopetrotic (op/op) mouse. J Cell Sci 104 ( Pt 4), 1021-9 (1993). 153 245. Raivich, G . , Moreno-Flores, M . T., Moller, J. C . & Kreutzberg, G . W. Inhibition of posttraumatic microglial proliferation in a genetic model of macrophage colony-stimulating factor deficiency in the mouse. Eur J Neurosci 6, 1615-8 (1994). 246. Bruccoleri, A . & Harry, G . J. Chemical-induced hippocampal neurodegeneration and elevations in TNFalpha, TNFbeta, IL-1 alpha, IP-10, and MCP-1 m R N A in osteopetrotic (op/op) mice. J Neurosci Res 62, 146-55 (2000). 247. Mucke, L . et al. High-level neuronal expression of abeta 1-42 in wild-type human amyloid protein precursor transgenic mice: synaptotoxicity without plaque formation. J Neurosci 20, 4050-8 (2000). 248. Lehman, E . J. et al. Genetic background regulates beta-amyloid precursor protein processing and beta-amyloid deposition in the mouse. Hum Mol Genet 12, 2949-56 (2003). 249. Masliah, E . et al. Comparison of neurodegenerative pathology in transgenic mice overexpressing V717F beta-amyloid precursor protein and Alzheimer's disease. J Neurosci 16, 5795-811 (1996). 250. Mohajeri, M . H . et al. Passive immunization against beta-amyloid peptide protects central nervous system (CNS) neurons from increased vulnerability associated with an Alzheimer's disease-causing mutation. J Biol Chem 211, 33012-7 (2002). 251. Lemere, C. A . et al. Amyloid-beta immunization in Alzheimer's disease transgenic mouse models and wildtype mice. Neurochem Res 28, 1017-27 (2003). 252. Das, P., Murphy, M . P., Younkin, L . H . , Younkin, S. G. & Golde, T. E . Reduced effectiveness of Abetal-42 immunization in A P P transgenic mice with significant amyloid deposition. Neurobiol Aging 22, 721-7 (2001). 253. Dodart, J. C. et al. Immunization reverses memory deficits without reducing brain Abeta burden in Alzheimer's disease model. Nat Neurosci 5, 452-7 (2002). 254. Hock, C. et al. Antibodies against beta-amyloid slow cognitive decline in Alzheimer's disease. Neuron 38, 547-54 (2003). 255. Orgogozo, J. M . et al. Subacute meningoencephalitis in a subset of patients with A D after Abeta42 immunization. Neurology 61, 46-54 (2003). 256. Check, E . Nerve inflammation halts trial for Alzheimer's drug. Nature 415, 462 (2002). 257. Pfeifer, M . et al. Cerebral hemorrhage after passive anti-Abeta immunotherapy. Science 298, 1379 (2002). 258. Poduslo, J. F. & Curran, G. L. Amylo id beta peptide as a vaccine for Alzheimer's disease involves receptor-mediated transport at the blood-brain barrier. Neuroreport 12, 3197-200 (2001). 259. Bacskai, B . J. et al. Imaging of amyloid-beta deposits in brains of living mice permits direct observation of clearance of plaques with immunotherapy. Nat Med 1, 369-72 (2001). 260. Solomon, B . , Koppel, R., Hanan, E . & Katzav, T. Monoclonal antibodies inhibit in vitro fibrillar aggregation of the Alzheimer beta-amyloid peptide. Proc Natl Acad Sci US An, 452-5 (1996). 261. Solomon, B . , Koppel, R., Frankel, D . & Hanan-Aharon, E . Disaggregation of Alzheimer beta-amyloid by site-directed mAb. Proc Natl Acad Sci USA 94, 4109-12 (1997). 154 262. Bacskai, B . J. et al. Non-Fc-mediated mechanisms are involved in clearance of amyloid-beta in vivo by immunotherapy. J Neurosci 22, 7873-8 (2002). 263. Wilcock, D . M . et al. Intracranially administered anti-Abeta antibodies reduce beta-amyloid deposition by mechanisms both independent of and associated with microglial activation. J Neurosci 23, 3745-51 (2003). 264. DeMattos, R. B . , Bales, K . R., Cummins, D . J., Paul, S. M . & Holtzman, D . M . Brain to plasma amyloid-beta efflux: a measure of brain amyloid burden in a mouse model of Alzheimer's disease. Science 295, 2264-7 (2002). 265. DeMattos, R. B . et al. Plaque-associated disruption of C S F and plasma amyloid-beta (Abeta) equilibrium in a mouse model of Alzheimer's disease. J Neurochem 81, 229-36 (2002). 266. Pan, W., Solomon, B . , Maness, L . M . & Kastin, A . J. Antibodies to beta-amyloid decrease the blood-to-brain transfer of beta-amyloid peptide. Exp Biol Med (Maywood) 221, 609-15 (2002). 267. Games, D . et al. Prevention and reduction of AD-type pathology in P D A P P mice immunized with A beta 1-42. Ann N YAcad Sci 920, 274-84 (2000). 268. Austyn, J. M . & Gordon, S. F4/80, a monoclonal antibody directed specifically against the mouse macrophage. Eur J Immunol 11, 805-15 (1981). 269. Uyama, O. et al. Quantitative evaluation of vascular permeability in the gerbil brain after transient ischemia using Evans blue fluorescence. J Cereb Blood Flow Metab 8, 282-4 (1988). 270. Methia, N . et al. A p o E deficiency compromises the blood brain barrier especially after injury. Mol Med 1, 810-5 (2001). 271. Blanc, E . M . , Toborek, M . , Mark, R. J., Hennig, B . & Mattson, M . P. Amylo id beta-peptide induces cell monolayer albumin permeability, impairs glucose transport, and induces apoptosis in vascular endothelial cells. J Neurochem 68, 1870-81 (1997). 272. Thomas, T., Thomas, G . , McLendon, C , Sutton, T. & Mullan, M . beta-Amyloid-mediated vasoactivity and vascular endothelial damage. Nature 380, 168-71 (1996). 273. Mackie, J. B . et al. Human blood-brain barrier receptors for Alzheimer's amyloid-beta 1- 40. Asymmetrical binding, endocytosis, and transcytosis at the apical side of brain microvascular endothelial cell monolayer. J Clin Invest 102, 734-43 (1998). 274. Poduslo, J. F., Curran, G . L . , Sanyal, B . & Selkoe, D . J. Receptor-mediated transport of human amyloid beta-protein 1-40 and 1-42 at the blood-brain barrier. Neurobiol Dis 6, 190-9 (1999). 275. Banks, W . A . , Robinson, S. M . , Verma, S. & Morley, J. E . Efflux of human and mouse amyloid beta proteins 1-40 and 1-42 from brain: impairment in a mouse model of Alzheimer's disease. Neuroscience 121, 487-92 (2003). 276. Banks, W. A . et al. Passage of amyloid beta protein antibody across the blood-brain barrier in a mouse model of Alzheimer's disease. Peptides 23, 2223-6 (2002). 277. Guo, J. T., Y u , J., Grass, D . , de Beer, F. C. & Kindy, M . S. Inflammation-dependent cerebral deposition of serum amyloid a protein in a mouse model of amyloidosis. J Neurosci 22, 5900-9 (2002). 155 278. Paris, D . et al. Soluble beta-amyloid peptides mediate vasoactivity via activation of a pro-inflammatory pathway. Neurobiol Aging 21, 183-97 (2000). 279. Gi r i , R. et al. beta-amyloid-induced migration of monocytes across human brain endothelial cells involves R A G E and P E C A M - 1 . Am J Physiol Cell Physiol 279, C l 772-81 (2000). 280. Suo, Z. , Fang, C , Crawford, F. & Mullan, M . Superoxide free radical and intracellular calcium mediate A beta(l-42) induced endothelial toxicity. Brain Res 762, 144-52 (1997). 281. Grammas, P. & Ovase, R. Inflammatory factors are elevated in brain microvessels in Alzheimer's disease. Neurobiol Aging 22, 837-42 (2001). 282. Pogue, A . I. & Lukiw, W . J. Angiogenic signaling in Alzheimer's disease. Neuroreport 15, 1507-1510(2004). 4 283. Bamberger, M . E . & Landreth, G. E . Microglial interaction with beta-amyloid: implications for the pathogenesis of Alzheimer's disease. Microsc Res Tech 54, 59-70(2001). 156 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0092329/manifest

Comment

Related Items