USF Libraries
USF Digital Collections

Promoting and preventing alzheimer's disease in a transgenic mouse model

MISSING IMAGE

Material Information

Title:
Promoting and preventing alzheimer's disease in a transgenic mouse model apolipoprotein e and environmental enrichment
Physical Description:
Book
Language:
English
Creator:
Costa, David Antonio
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla.
Publication Date:

Subjects

Subjects / Keywords:
Alzheimer's disease
Transgenic mouse
Amyloid
Apolipoprotein e
Parabiosis
Environmental enrichment
Dissertations, Academic -- Biochemestry and Molecular Biology -- Doctoral -- USF   ( lcsh )
Genre:
government publication (state, provincial, terriorial, dependent)   ( marcgt )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Summary:
ABSTRACT: Besides age, inheritance of the apoE-E4 allele is the main risk factor for late-onset AD. To determine the role of apoE in amyloid deposition, we studied mice expressing both mutant human amyloid beta-protein precursor (APP) and presenilin 1 (PS1) that were either normal or knocked-out for apoE. By 7 months, amorphous Abeta deposition developed equally in both lines, indicating that Abeta alone is sufficient for deposition to occur. In contrast, filamentous amyloid deposition was catalyzed at least 3000 fold by apoE. Electron micrographs further illustrate the filamentous nature of these plaques. These results and other, behavioral, data indicate that the primary function of apoE in AD is to promote the polymerization of Abeta into mature, neurotoxic, amyloid. ApoE is also synthesized in the liver and is crucial in cholesterol metabolism, for mice lacking apoE exhibit hypercholesterolemia.We investigated neuropathology in mice using an uncommon technique, parabiosis, to determine whether apoE in the peripheral circulation influences brain amyloid formation. This surgical procedure allows exchange of proteins via peripheral circulation. We show that plasma apoE is found in parabiosed PS/APP/apoE-KO mice, rescuing their hypercholesterolemia. Unexpectedly, amyloid deposition is reduced in parabiosed PS/APP/apoE-KO mice compared to PS/APP controls. ApoE in the periphery seems to slightly reduce amyloid burden, by likely promoting efflux of Abeta;from the brain. These findings reinforce that the mechanisms whereby apoE affects Abeta metabolism are complex, and the modulation of peripheral apoE metabolism is not likely to impact AD neuropathology. Since cognitive stimulation is associated with lower risk of AD, we sought to investigate the preventative potential of environmental enrichment (EE) using our mouse model.
Thesis:
Thesis (Ph.D.)--University of South Florida, 2005.
Bibliography:
Includes bibliographical references.
System Details:
System requirements: World Wide Web browser and PDF reader.
System Details:
Mode of access: World Wide Web.
Statement of Responsibility:
by David Antonio Costa.
General Note:
Includes vita.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 269 pages.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001670320
oclc - 62253415
usfldc doi - E14-SFE0001179
usfldc handle - e14.1179
System ID:
SFS0025500:00001


This item is only available as the following downloads:


Full Text

PAGE 1

Promoting and Preventing Alzheimer’s Dis ease in a Transgenic Mouse Model: Apolipoprotein E and Envir onmental Enrichment by David Antonio Costa A dissertation submitted in partial fulfillment of the requirement s for the degree of Doctor of Philosophy Department of Biochemist ry and Molecular Biology College of Medicine University of South Florida Major Professor: Hunt ington Potter, Ph.D. Larry P. Solomonson, Ph.D. W. Lee Adair,Jr., Ph.D. R. Kennedy Keller, Ph.D. David G. Morgan, Ph.D. Date of Approval: July 19, 2005 Keywords: Alzheimer's Disease, Transgeni c Mouse, Amyloid, Apolipoprotein E, Parabiosis, Environmental Enrichment Copyright 2005, David Antonio Costa

PAGE 2

To my father, for all his love and countless sacrifices. and To my wife, Heather, for her endless moral support and a reluctant understanding that I will most lik ely be a student forever.

PAGE 3

i TABLE OF CONTENTS LIST OF TABLES.................................................................................................iii LIST OF FI GURES...............................................................................................iv ABSTRACT..........................................................................................................v INTRODUCTION..................................................................................................2 ALZHEIMER’S DISEASE...............................................................................2 APOLIPOPROT EIN E.....................................................................................7 ENVIRONMENTAL ENRICHMEN T..............................................................16 Paper I: Apolipoprotein is required for the formation of filamentous amyloid, but not for amorphous A depos ition, in an APP/PS double transgenic mouse model of Alzheimer’s disease................................23 Paper II: Parabiosis as means to invest igate the role of peripherally derived proteins in Alzheimers disease pa thology .........................................42 Paper III: Environmental Enrichment Protects Alzheimer’s Disease Mice from Cognitive Impairment throu gh Reductions in A Deposition and Beneficial Changes in Gene Expression and can be Mimicked by Inhibition of P hosphodiesterase 4 (PDE 4).....................75 DISCUSSION...................................................................................................123 FUTURE DIRE CTIONS....................................................................................136 REFERENC ES.................................................................................................138 APPENDIX A Source Code for High Output Morphometry Examination Routine (HOM ER)............................................................................157

PAGE 4

ii APPENDIX B Previous Publicat ions...............................................................191 ABOUT THE AUTH OR.........................................................................END PAGE

PAGE 5

iii LIST OF TABLES Paper II: Table 1. Apolipoprotein E (A poE) derived from the blood through parabiosis restores hypercholesterolemia in A-producing ApoE Knockout Mice....................................................................................59 Paper III: Table 1. Comparative microarray analysis of hippocampal gene expression between environmentally enriched and standard housed PS1/PDAPP mice.............................................................................110 Table 2. qRT-PCR verification of select transcripts from hippocampal microarray anal ysis...........................................................................111

PAGE 6

iv LIST OF FIGURES Paper I: Fig. 1. Little or no effect of ApoE on total A deposition Total A deposition, diffuse and compact, in 7-month-old PS1M146L+/-, APPV717F+/+, A poE+/mi ce................................................................34 Fig. 2. Strong catalytic effect of ApoE on mature amyloid deposition Thioflavine S plaque staining is increased over 3200 fold in 7month-old PS1M146L+/-, APPV7 17F+/+, ApoE +/mice.......................36 Paper II: Fig. 1. Apolipoprotein E is detectable in parabi osed apoE-knockout mice.......60 Fig. 2. A-immunoreactive (6E10) st aining in parabiosed or control PS1+/APP+/+,apoE-/and PS1+/-, APP+/+,apoE+/ mice...............................62 Fig. 3. Apolipoprotei n E immunostaining in parabiosed and control PS1+/APP+/+,apoE-/and PS1+/-, APP+/+,apoE+/ mice...............................64 Fig. 4. Thioflavine S tissue staining for the detection of filamentous A...........66 Paper III: Fig. 1. Environmental enrichment (EE) protects PS1/PDAPP mice against cognitive impairment across multip le behavioral tasks administered between 4-6 mont hs of age..............................................................102 Fig. 2. A and compact amyloid pl aque levels are severely reduced in Tg+ mice both environmentally enriched and behaviorally tested despite dendritic defi cits..................................................................................104 Fig. 3. Environmental Enrichment re sults in a set of genetic changes that culminate in anti-apoptotic/neuroprot ective BAD phosphorylation......106 Fig. 4. Administration of Rolip ram to Tg+ mice results in memory improvement which mimics envir onmental enrich ment.......................108

PAGE 7

v Promoting and Preventi ng Alzheimer’s Disease in a Transgenic Mouse Model: Apolipoprotein E and Environmental Enrichment David Antonio Costa ABSTRACT Besides age, inherit ance of the apoE4 allele is the main risk factor for late-onset AD. To determine the role of apoE in amyloid deposition, we studied mice expressing both mutant human amyl oid -protein precursor (APP) and presenilin 1 (PS1) that we re either normal or knocked-out for apoE. By 7 months, amorphous A deposit ion developed equally in bot h lines, indicating that A alone is sufficient for deposition to occur. In contrast, filamentous amyloid deposition was catalyzed at least 3000 fo ld by apoE. Electron micrographs further illustrate the fila mentous nature of these plaques. These results and other, behavioral, data indicate that the pr imary function of apoE in AD is to promote the polymerization of A into mature, neurot oxic, amyloid. ApoE is also synthesized in the liver and is crucial in cholesterol metabolism, for mice lacking apoE ex hibit hypercholesterolemia. We investigated neuropathology in mice using an uncomm on technique, parabiosis,

PAGE 8

vi to determine whether apoE in the peripheral circulation influences brain amyloid formation. This surgical procedure allo ws exchange of protei ns via peripheral circulation. We show that plasma apoE is found in parabiosed PS/APP/apoE-KO mice, rescuing their hypercholesterolemia. Unexpectedly, amyloid deposition is reduced in parabiosed PS/APP/apoE-KO mice compared to PS/APP controls. ApoE in the periphery seems to slightly reduce amyloid burden, by likely promoting efflux of A from the brain. These fi ndings reinforce that the mechanisms whereby apoE affects A metabolism are complex, and the modulation of peripheral apoE met abolism is not likely to impact AD neuropathology. Since cognitive stimulati on is associated with lower risk of AD, we sought to investigate the prevent ative potential of environmen tal enrichment (EE) using our mouse model. At weaning, mice were placed into ei ther enriched or standard housing (SH). Behavioral testing at 4-6 months showed that EE-PS1/APP mice outperformed mice in SH, and we re behaviorally indistinguishable from nontransgenic mice. PS1/APP mice given both EE and behaviora l testing had 50% less brain -amyloid (A), but di d not exhibit changes in dendritic morphology. Microarray analysis of hippocmapal RNA re vealed large EEinduced changes in the expression of genes/proteins related to memory, neuroprotection, and A seques tration. Inhibition of one such protein, PDE4, a cAMP-phosphodiesterase, by Rolipram, mi micked the cognitive benefits of EE.

PAGE 9

2 INTRODUCTION ALZHEIMER’S DISEASE Alzheimer’s disease (AD) is the most prevalent form of dementia in people over the age of 65, and it is estimated that for ev ery five years a person lives over the age of 65, the chance of develop ing AD doubles. In fact, approximately 58% of those living at the age of 95 develop AD (Ebly et al., 1994). Using data from the year 2000 United St ates census, it was estimated that there were roughly 4.5 million Americans liv ing with the disease, and that, due to the rapidly aging population, this figure should triple by the year 2050 (Hebert et al., 2003). At a current estimated cost of $100 billion per year to the US economy, AD is a burden to more than patients, families, and health care providers. The main overt symptoms of AD, init ially described by Alois Alzheimer in 1907, are progressive memory loss and dementia. From a pathological perspective, AD results in two distinct brain lesions, neur itic plaques and neurofibrillary tangles, with concomitant neuronal loss, occurring primarily in the hippocampus and cerebral cortex, two brai n regions intimately associated with memory and higher cognitive functions. Neurofibrillary tangles are highly in soluble cytosolic aggregates made of paired helical filaments (PHF) of an abnormally phosphorylated (hyperphosphorylated) microtubule associ ated protein, tau. Tau normally

PAGE 10

3 stabilizes and promotes the assembly of microtubules, but w hen phosphorylated at either serine 262 or 214, it detaches from t he microtubule and is made available for polymerization (Mandelko w and Mandelkow, 1998). Tau is necessary for the outgrowth of neurites, so it is likely that alterations in microtubule binding could cont ribute to a breakdown in intracellular transport and the resultant dying back of neurons indicati ve of AD. It has also been shown that cyclin dependant kinase 5 (cdk5) can be cons titutively active in patients with AD. Cdk5 can hyperphosphorylate tau, therefore inducing NFT formation, and has been shown to accumulate in tangle-be aring neurons, adding yet another potential mechanism for the development of AD pathology (Patrick et al., 1999). Since the mouse models employed in our re search do not develop neurofibrillary tangles, our efforts focus specifically on the other main pathol ogical hallmark of AD, neuritic plaques. Neuritic, or amyloid, pl aques are extracellular prot ein aggregates typically surrounded by both dystrophic neurites and reactive as trocytes. The primary constituent of these lesions is a short peptide called amyloid-be ta (A). It was the 1984 isolation and characterization of the A peptide by Glenner and Wong that ushered in the modern era of AD research (Glenner and Wong, 1984). Later, it was discovered that A is really the proteol ytic cleavage product of a larger, membrane bound, pr otein termed the amyloid precursor protein (APP) (Goldgaber et al., 1987; Kang et al., 1987; Robakis et al., 1987; Tanzi et al., 1987). APP, a protein whose function is still not known, is ubiquitously expressed, and its cleavage, resulting in A producti on, seems to be a normal

PAGE 11

4 cellular event. Discerning the mechanism wh ereby cellular homeostasis is altered to confer a disease state has been challe nging for the field. One known change that does occur in many AD patients, especially those with FAD, is an altered ratio between the two major species of A, the shorter form ending in amino acid (AA) 40, and the longer 42 AA form. Though still far less abundant than A1-40, a higher percentage of A1-42 is found in diseased brains This is of particular interest since A1-42 seems to have a greater propensity for aggregation (Jarrett et al., 1993). Furthermore, it has been sugges ted that pre-amyloid, or diffuse, plaques which are immunoreacti ve, but not classified as amyloid due to a lack of birefringengence, are m ade predominantly of A1-42, and serve as the foundation for later A1-40 accumulation and eventual amyl oid formation (Lemere et al., 1996). Greater than 97% of all AD cases ar e sporadic, or of unknown etiology. The remaining cases are attributable to genet ic defects. Most of what is known about the cellular events of sporadic AD comes from re search on the relatively rare inherited forms of the disease. The majority of familial AD cases are early onset and due to autosomal dominant mutati ons in the APP, presenilin 1 (PS1), or presenilin 2 (PS2) genes. Another inher ited variety of AD takes the form of trisomy 21, or Down’s syndrome (DS). Phenotypically, autosomal dominantly inherited AD is indistinguishable from s poradic AD with the exception of an earlier onset age. Mutations in the APP gene, which is on chromosome 21, are thought to alter its proteolytic processing. The mo st common mutations are found as point

PAGE 12

5 mutations adjacent to the two proteolytic cleavage sites utilized to excise the A fragment from the APP hol oprotein. Both N-terminal -site (AA 670/671) and Cterminal -site (AA717) mutations increase the amounts of extracellular A1-42, thereby promoting the initial deposition of this amyl oidogenic species. Similarly, mutated presenilin proteins, which are as lo inherited in an autosomal dominant manner, shift the ratio of A1-40/A1-42 towards A1-42. Presenilin is an intramembrane pr otease which is thought to be the secretase, or at least an inseparabl e cofactor in the APP cleavage pathway (Iwatsubo, 2004). Since a genetic ablation of this gene is lethal by birth (Qian et al., 1998), inherited mutations most likel y invoke pathogenesis through a toxic gain of function, which is further support ed by the fact that, with two exceptions, all of these mutations cause amino acid substituitions (i.e. no frameshifts, stops, etc…). Just increasing the bulk amount of APP also confer s a disease state. By having one extra copy of chromosome 21, such as in DS, these individuals develop large amounts of A deposits at an early age, and eventually the memory deficits associated with AD. Insight into the role of APP and A in AD may be provided by genetic and biochemical studies that have explored the roles of other proteins thought to be involved in the disease process. For exam ple, A amyloid d eposits also contain 1-antichymotrypsin (ACT) and apolipopr otein E (ApoE), which are overexpressed in affected regions of the AD brain as part of an inflammatory process (Abraham et al., 1988; Xu et al., 1999). ACT levels are also increased in AD serum and CSF (see Licastro et al., 1995 for data and discussion). It was

PAGE 13

6 proposed at the time of their discoveries that ApoE and ACT might function as amyloid promoters or “Pathol ogical Chaperones”, and both in vitro and in vivo studies support this model. As with APP, the importance of in flammation and specifically ApoE and ACT in AD is supported by genetic studies. Inheritance of an ApoE allele is the strongest risk factor for AD besides age, and an ACT/A signal peptide variant that increases mature glycosylated ACT available for secretion also increases AD susceptibility and pathology (Kamboh, 1 995; Kamboh et al., 1995). Similarly, polymorphisms in the IL-1 promoter gr eatly increase the risk of AD. These genetic variations increase IL-1 expres sion during inflammation and therefore promote amyloid formation by increas ing the production of both APP and ACT. Genetic and biochemical studies have identified key proteins in the AD pathogenic pathway and much research has so ught to ascertain the role that A, ApoE, and ACT play in the formation of amyloid and the re sultant cognitive dysfunction of AD. To begin to answer these questions, we performed cognitive studies with AD mouse models in which the inflammatory proteins and the A deposition they influence could be regul ated without changing the level of monomeric A peptide. To this end, we generat ed and analyzed a number of ACT and/or apoE-expressing/ non-expressing mouse m odels of AD and found that: 1) ApoE and ACT independently and synergi stically promote A immunoreactive and mature amyloid deposition without initially affecting A levels and 2) cognitive impairment in aged AD mice depends on the amyloid promoting effect of ApoE and/or AC T (Nilsson et al., 2004, APPENDIX B).

PAGE 14

7 The compilation of manuscripts presented in this volume seeks, using transgenic mouse models of AD, to more clearly define the role of apoE in amyloidogenesis, and to asce rtain whether the systemic circulating pool of apoE is able to exert an affect on pathogenic A lesions localized in the brain. Furthermore, we explore the beneficial pathological and genetic effects of cognitive stimulation on the Alzheimer br ain. Both avenues of study, apoE and environmental enrichment, are discussed in detal below. APOLIPOPROTEIN E Besides age, the apo lipoprotein E gene is the greatest known genetic susceptibility factor for AD. Three apoE alleles exist, 2, 3, and 4, each conferring a different relative risk of developing the disease ( 4 > 3 > 2). There exists an allelic dose response for the apoE 4 allele in particular which is exemplified by the 20 year accelerati on of the mean age of onset for those people with two copies of apoE 4 (average onset age = 70) vs. those with a 2/ 3 combination (average onset age = 90) (Strittmatter and Roses, 1996). Another particularly interesting example of the affect of apoE 4 comes from studies of the Japanese population as a whole, where the allelic frequency of individuals homozygous for apoE 4 is lower than that of the United states, which results in an overall prevalence of the disease that is therefore lower, and an observed population age of onset of sporadic AD that is shown to be higher, furthing implicating apoE 4 as a risk factor in AD (Uek i et al., 1993). Though not

PAGE 15

8 known with any great deal of certainty, there are a number of putative roles that apoE plays in relation to AD: ApoE, 1) forms stable complexes with the A peptide; 2) In an isoform-specific manner alters the aggregation properties of A; 3) inhibits A-induced neurotoxicity; and, 4) promotes the clearance of A in a receptor mediated fashion. ApoE is a 34 kDa (299 amino acid s in length) lipoprotein produced predominately in the liver w hose primary function is to transport lipids via either low density lipoproteins (LDL ), very low density lipop roteins (VLDL), or high density lipoproteins (HDL). The brain has the second highest amount of ApoE in the body (after the liver) where it is predominantly produced by astrocytes (Driscoll and Getz, 1984). In the course of AD, both the levels of apoE mRNA and protein are increased in astrocyt es, and though neurons do not produce apoE, for no apoE mRNA is detectable, apoE protein is present in neurons of patients with AD. Besides genetic linkage studies implic ating apoE’s role in AD, there are a number of reasons that apoE is thought to be directly in volved in the process of amyloidogenesis. Several apoE promoter mutations exist whereby the gross levels of apoE are increased which is correlated with an increased risk of AD (Strittmatter and Roses, 1996). Furthermo re, apoE has been shown to bind to A, and lastly, apoE accu mulates in extracellula r amyloid plaques. In vitro both apoE 3 and 4 bind, in an irreversible manner, to the A peptide (Strittmatter et al., 1993). Both isoforms catalyze the formation of fibrils, but the 4 isoform promotes A polymerization r oughly 25-times faster than apeE 2 or 3 (Ma et

PAGE 16

9 al., 1994; Wisniewski et al., 1994). Furt hermore, the character of the fibrils formed in the presence of apoE adopt a simple 10nm fiber shape, as compared to a non-apoE catalyzed aggregate which can best be described as a twisted ribbon-like conformation. Interestingly, it has been shown t hat apoE confers a greater degree of protease resistance to A upon binding, which could be responsible for the irreversible nature of this interaction (Zhou et al., 1996), and may help explain the propensity for apoE -mediated fibril accumulation. The vast majority of existing liter ature would suggest that ApoE does not affect APP production. Transcription or translation of APP does not change in the presence of apoE, and this has been de monstrated with both mouse and the and human isoforms (Bales et al., 1999; Koistinaho et al., 2004). Even though there is a reported gene dose effect of mouse apoE on A immunoreactivity and Thioflavine S stai ning (apoE+/+ > apoE+/> apoE -/-), at two months of age there is no differenc e in A levels, as shown by enzyme linked immunosorbent assay (ELISA) (Bales et al., 1997; Nilsson et al., 2004, APPENDIX B). The immunohi stochemical staining pa tterns of transgenic AD mouse models suggests that A1-42 deposition precedes ApoE deposition, while Ab1-40 follows (Terai et al., 2001), further indicating that it is likely that apoE plays an important role in the maturation of plaques, but not necessarily initial APP processing. Other authors have conversely reported that hippocampal levels of A are suppressed in APP mice expressing human apoE 3 or 4 (Fagan et al., 2002). These results, combined with t he findings that apoE can complex with both soluble and fibrillar A (Permanne et al., 1997; Russo et al., 1998), has led

PAGE 17

10 to an isoform-dependant dual-role hypothesis for apoE. First, as previously mentioned, the isoform of apoE promotes the fibrillarization of deposited preamyloid A1-42 aggregates into mature amyloid. Secondly, since one manuscript has reported that more soluble A is found in mice with human apoE4 than those expressing apoE3 (Bales et al., 1997), it has been proposed that apoE 3 inhibits A oligomer deposition (short oligomers of A are thought to be neurotoxic and the precursors to longer fibrils) and that pre-fibrillar A/apoE3 interactions could foster apoE-mediated clearance of A. This precedent has been set by other A-associated molecules such as transthyretin, which have been shown to sequester A (Goldgaber et al., 1993). The classical amyloid cascade hypot hesis describes a process whereby APP is processed into A, which in turn leads to plaque deposition and ultimately neuronal cell dysfunction and death. Unfo rtunately, the temporal and spatial distribution of brain amyloi d deposition does not correlate well with the clinical progression of the disease (Ma et al., 1996; Naslund et al., 2000). There is a large body of research im plicating the soluble A o ligomer, rather than mature, deposited, amyloid as t he causative agent in AD. It has been shown in vitro that neuronal viability is adversely affected by oligomeric A 10fold more than the fibrillar form, and roughly 40-fold more th an the monomeric peptide (Dahlgren et al., 2002). Furthermore, A oligomers cause an increase in inflammatory markers such as interleukin-1 (IL-1) or inducible nitric oxide synthase (iNOS), while fibrillar A does not (Manelli et al., 2004). It therefore seems that oligomeric A is far more neurotox ic than unaggregated or fibrillar A.

PAGE 18

11 Interestingly, ApoE3 has been shown to pr event A-induced neur otoxicity, while ApoE4 has not (Jordan et al., 1998). Spec ifically, apoE3 attenuates oligomer toxicity (Manelli et al., 2004). This is not a general anti-inflammatory mechanism of apoE, for other inflammatory agents su ch as LPS are not attenuated in the same manner as A (Hu et al., 1998). Furthermore, Sinc e APP/ApoE-KO mice perform poorly in behavioral tasks designed to measure memory deficits, even with a lack of ApoE-catalyzed plaque forma tion, a detrimental effect on memory independent of fibrilla rzation events exists (Dodart et al., 2000). This supports the beneficial role of apoE in the inhibition of o ligomer-induced neurotoxicity and inflammation while not discounting apoE ’s catalytic role in promoting A fibrillarization. ApoE’s role in the brain inflamma tory response and neuritic pathology is well characterized, but not widely agreed upon. It is known that apoE levels rise due to brain injury (Poirier et al., 1993), and that that there is a marked reduction in gliosis (as measured by GFAP immunostain ing) in apoE knockout mice (Bales et al., 1999). Since A is known to increase apoE produc tion, it has been hypothesized that apoE negat ively feeds back to limit further inflammation caused by this molecule. It has been not ed by more than one author that apoEKO mice do not develop dendritic or long term potentiation deficits (Dodart et al., 2000; Holtzman et al., 2000) The neuritic pathology of APP mice (i.e. – swollen and distorted neurites) is absent when apoE is selectively ablated, and dystrophic neurites re-appear when either mApoE, hApoE3, or hApoE4 is expressed, indicating that apoE-induced fibrillar plaque formation is most likely

PAGE 19

12 responsible for neuritic pat hology (Holtzman et al., 2000). Opposing research has reported that apoE-KO mice pres ent a 15 to 40% loss in synaptophysin staining and a decrease in tubulin imm unoreactivity. This lends credence to the idea that apoE might play an important ro le in maintaining dendritic and synaptic stability and plasticity, and any alterations could be responsible for typical ADrelated synaptic and cytoskeletal deficits (Ma sliah et al., 1995). Since apoE3, but not apoE4, delays age-dependant decline of synaptophysin levels in the presynaptic terminals of APP mice, the fa ct that the cognitive decline in AD correlates better with synaptophysin reacti vity than plaque load (DeKosky and Scheff, 1990) could be relevant to apoE-conferred risk of AD and an apoE-driven early onset. Furthermore, apoE mi ght even bind to tau and prevent its phosphorylation (Strittmatter et al., 1994) in an isoform specific manner, for apoE forms an irreversible complex with tau, but ApoE does not. The same holds true for another microtubule associated pr otein, MAP2c. Therefore, apoE may play a role in A fibrillarization, in flammation, synaptic health, and neurofibrillary tangle formation. The apoE molecule contains two f unctionally important domains: an LDL receptor (LDLr) binding domain, and a lipoprotein binding domain. Neurons express several apoE receptors, includ ing the LDLr, the LDLr-related protein (LRP), and the VLDL receptor (Terai et al., 2001). Neur ons are thought to uptake apoE and their associated lipid par ticles in a receptor mediated fashion, for In vitro apoE is taken up from the growth media by neurites and neuronal growth cones (Strittmatter and Roses, 1996). After neuronal injury, cholesterol is

PAGE 20

13 endocytosed via apoE and the LDLr in order to support the growth of neurons undergoing reinnervation (Poirier et al., 1993), indicating yet another role for apoE in brain maintenance – lipid transport due to CNS injury. It is likely that apoE’s role in the brain is, like in the periphery, to r egulate plasma lipid transport and clearance as the ligand for lipoprotein receptors. ApoE isoform differences emerge yet again, for ApoE and bind, with high affinity to the LDL receptor, while apoE does not (Weisgraber et al., 1982) This readily explains the atherosclerotic deposit ion found in apoE 2/ 2 and apoE-KO mice, due to respectively decreased or abolished rece ptor mediated apoE endocytosis. A reason that apoE may be detrimental in AD, aside from its fibrillogenic properties, is that it is t hought to be a less efficient chol esterol and lipid mobilizer and transporter during CNS injury (Mahley et al., 1989). Furthermore, apoEenriched VLDL particles hav e been shown to modulate neurite outgrowth in an LRP mediated and isoform specific manner, with apoE stimulating outgrowth to a larger extent than (Holtzman et al., 1995). A has been found to be preferentially internalized by astrocytes but ApoE-KO astrocytes do not degrade A, and blocking LRP also blocks t he degradation. It has therefore been concluded that apoE is essent ial for astrocytes to associ ate with, internalize, and degrade soluble, but not fibrillar A (Koi stinaho et al., 2004), once again pointing to the dual role that apoE plays in AD.

PAGE 21

14 Given the importance of apoE, we hav e decided to further examine the role of endogenous mouse apoE in a tr ansgenic mouse model of AD. To determine which part or product of the depos ition process is controlled by apoE, we examined a mouse model in which the production of A is not rate limiting, thus allowing the unique effect of apoE to be determined. We used animals expressing mutant forms of both the human APP and presenilin-1 transgenes which express high ratios of A1-42 /A1-40 and develop large amounts of both amorphous and filamentous A (amyloid) deposits. Specifically, we crossed PS1M146L+/-, APPV717F+/doubly transgenic animals with an apoE knockout mouse (Bales et al., 1999) in order to explore the effect(s) of apoE in a mouse model that produces large amounts of A1-42 and develops early and severe AD-like pathology. We predicted two likely alternat ive results: 1) If apoE affects A availability or clearance, it should have little positive or negative effect in our model due to the already large amounts of PS1-enhanced A production in these doubly transgenic mice, or 2) If apoE prom otes A fibrillarization, then its expression should elevate the amount, and/ or change the character of the brain A deposition. Mouse brains were analyzed for AD pathology at seven months of age, and we observed that large amounts of amorphous, non-fibrillar, A had been generated in both sets of animals in statis tically similar amounts, reflecting the presence of large levels of A1-42. The physical appearance and distribution of the immunoreactive deposits was quite di fferent. We found that mature plaque deposition was strikingly dep endent on the presence of apoE In particular, apoE

PAGE 22

15 expression induced a more than 3000-fold increase in the formation of mature, fibrillar, amyloid (to our knowledge the lar gest reported effect of any molecule on amyloid formation other than the mutant APP itself). In the absence of apoE, the area occupied by the few thioflavine S-pos itive A deposits was not sufficiently large to be statistically significantly di fferent from the background thioflavine S staining in a completely non-transgenic mouse. Once apoE’s effect on A maturation was determined, we wished to determine the source of apoE most responsible for this occurrence. It is a curious fact that most (if not a ll) of the proteins in volved in Alzheimer amyloid formation are also present in the ci rculation. It is therefore important to know whether the amyloid deposits of the brain are derived in part from the circulating proteins or only from loca l proteins produced in the brain. The evidence that brain expre ssion of A, apoE, and ACT c an lead to local amyloid deposition is strongly suppor ted in the transgenic mouse models of amyloid formation. However, in human, these proteins are made in equal or larger amounts in the rest of the body and, as mentioned abov e, are present in large amounts in the blood. Certain facts point to the possibility that blood-derived proteins may contribute significantly to the brain deposits. For example, A levels, particularly A1-42 are elevated in the bl ood of FAD and many SAD patients (Scheuner et al., 1996). Serum AC T and apoE levels are also generally increased in AD patients (Licastro et al ., 1995; Lieberman et al., 1995; Blain et al., 1997) and apoE4 has been shown to shuttle A across the blood brain barrier (Martel et al., 1997). It is therefore reasonabl e to hypothesize that apoE

PAGE 23

16 from the circulation may c ontribute importantly to th e amyloid deposits in the brain. We attempted to ascertain whether circulating apoE could contribute to Alzheimer amyloid formation using a no t often used surgical technique called parabiosis whereby mice are su tured together early in life as to allow the sharing of their blood circulation (Martinez et al., 1959). In practice, any circulating protein produced in one mous e will be passed to the other and vice versa. By parabiosing an endogenously apoE-expressing mouse to an apoE-KO mouse, we sought to determine whether the pathological chaperone, apoE, must be synthesized in the brain or can be derived from the blood. Th e significance of this experiment for drug development is derived from the fact that circulating proteins are far easier to administer versus those given intra-cortically. ENVIRONMENTAL ENRICHMENT The development of drugs to effectiv ely treat and/or prevent AD has proven a most difficult task for the resear ch community. Therefore, significant effort has been placed in studying non-ph armacological means of delaying or preventing AD. One such av enue of study is that of preventative an d therapeutic cognitive stimulation regimens. The co rrelation between basic cognitive abilities in later life and early performance measures such as IQ and educational status have been studied by a number of research groups. For example, Plassman et al. (1995) established that performance on the Army General Classification Test directly correlated with late-life (i.e. 50 y ears later) cognitive fitness. Similarly,

PAGE 24

17 low levels of education have also been a ssociated with later cognitive deficits and an increased risk of AD (Letenneur et al., 1999). One such study even proposed that each year of education redu ces the risk of generating AD by 17% (Evans et al., 1997). Though these result s are maybe optimistic, a similar study found that high occupational a ttainment is also protecti ve against AD (Stern et al., 1994). No doubt, these reports led to one of the most famous psychological inquiries in the AD comm unity; the “nun study.” Here, Snowdon et al. (1996) examined early life linguistic ability, in a population of nuns, as a marker for later cognitive decline. The c ontent of autobi ographies for admission into their convent written by 93 nuns before the y ear of 1917 (Average age = 22) were studied. It was found that the scores fr om cognitive testing (Mini Mental State Exam) administered roughly 50-60 years late r were highly correlated with both idea density and grammatical complexity as well as years of education. Furthermore, of those nuns who died duri ng the course of the study (n=14), 100% of those diagnosed wit h AD had low idea density in their essays. Idea density was also correlated with the severi ty of AD pathology (Snowdon et al., 1996). Besides education and job attainment, overall low activity levels (physical and especially mental) in mid life have also been shown to increase the risk of AD (Friedland et al., 2001). Recently, several cognitive stimulat ion/rehabilitation programs have been developed in an attempt to slow or revers e the cognitive decline in AD (Davis et al., 2001; Wenisch et al., 2005). These programs have generally involved a relatively short 4–12 week period of cognitive training classes and/or daily

PAGE 25

18 caregiver-directed mental activities, resu lting in modest degrees of improvement. One such study showed that specific c ognitive rehabilitation techniques given to patients with mild AD over a 12-week period improved a va riety of cognitive skills with benefits seen even 3 months followi ng CS (Loewenstein et al., 2004). Nonetheless, the therapeutic potential of in tensive, longterm CS has not yet been evaluated in AD patients. Moreover, the biochemical and potential pathological changes due to CS cannot be easily stud ied in humans. Furthermore, for the aforementioned risk factors, early-life education, activity, and cognition, the cause and affect relationship between AD and poor early-life performance markers has not been determined. The fa ct that AD exerts its affects on the brain at a young age is ex emplified by the finding that patients with Down’s syndrome can have A1-42 accumulation in the brain as early as 12 years old. Do education and mental activity buffer agai nst AD, or does AD predispose one, from an early age, to have a low prop ensity for participation in mentally stimulating activities? Since retros pective studies like those above cannot sufficiently answer this question, and lo ng-term prospective studies are not practical, the use of rodents, specifical ly mouse models of AD, must be employed for this direction of study. Rodents undergoing cognitive stimulation (CS) by being placed in an enriched environment (e.g. socially hous ed in large cages containing toys, tunnels, running wheels, etc.) are s hown to exhibit improved cognitive performance in behavioral testing regime ns (e.g. – Morris wa ter maze, Radial arm water maze, etc…) (Kobayashi et al., 2002). EE protects rats against kainic

PAGE 26

19 acid-induced seizures (Young et al., 1999), and can even reverse the cognitive deficits in rats subjected to long-term lead exposure (Guilarte et al., 2003). In addition, a number of structural and bioc hemical changes have been noted in rodents due to an enriched environment. For example, the overall thickness of the cerebral cortex has been shown to be increased (Diamond et al., 1972). There are increases in hippocampal dentate gyrus neurogenesis (Kempermann et al., 1997; Brown et al., 2003), incr eases in the density of synapses (Greenough and Volkmar, 1973; Ramirez-Amaya et al., 1999), and the extent of dendritic branching (Globus et al., 1973). EE is also prot ective, in that Young et al (1999) observed a 45% decrease in spontaneous brain cell death as measured by TUNEL assay. Besides physical and structural changes in brain architecture, a large number of genetic and neurochemical measures have been described as being altered as a result of exposure to a cognitively stimulating environment. For example, mRNA and protein levels of gl ial-derived neurotrophi c factor and brainderived neurotrophic factor both of which have been associated with neuronal plasticity and survival, are increased in the rat hippocampus after exposure to an enriched environment (Fischer et al., 1992). In addition, the levels of another growth-related molecule, nerve growth fa ctor (NGF) are increased due to EE. When injected into the rat forebrain, NGF improves spatial learning and memory (Fischer et al., 1991). NGF is thought to be important for cholinergic neuron maintenance, and has been shown to increase levels of choline acetyltransferase, an enzyme whose levels decrease in the AD brain (Gnahn et

PAGE 27

20 al., 1983). Various receptor and recepto r-related proteins also undergo changes in expression due to EE. For example, NMDA receptors, which are known to play a pivotal role in hippocampal synaptic plasticity (Singer, 1990; Gu, 1995), are essential for triggering long-term potent iation and increasing the expression of memory related molecules such as phospholipase A2 (PLA2) and the immediate early gene (IEG) nerve growth factor induced-A (NGFI-A) (Diamond et al., 1972). PLA2 expression is further linked to synaptic plasticity via Ca2+dependant and AMPA receptordependant mechanisms (Fi scher and Bjorklund, 1991), for GluR1-4 subunits (i.e. AMPA rec eptor subunits) are increased due to EE. Additionally, the IEG, NGFI-A is thought to be involved in the early stages of memory formation, for its ex pression is rapidly induced after neuronal activity in brain regions with the most pronounced EE -induced structural changes (Wallace et al., 1995). ARC, or acti vity regulated cytoskeletal protein, is another IEG upregulated due to EE and known to be import ant for learning, for inhibiting it blocks both LTP and long-term memory. T hese are just a few of the described neurochemical changes induced by cognitiv e stimulation. Our current study of environmental enrichment hopes to el ucidate some of the genetic and biochemical changes most important for preserving memory and reducing pathology in a mouse model of AD. A previous study from our laboratory sought to elucidate the cognitionenhancing potential of long-term environmenta l enrichment (EE) as a therapeutic intervention in aged transgenic mouse model of AD, which overexpresses the Swedish doubly mutant (K670N,M671L) am yloid precursor protein and bears

PAGE 28

21 moderate cortical/hippocampal A deposition within ma ture AD-like neuritic plaques (Arendash et al., 2001a, APPENDIX B). We found that environmental enrichment begun relatively late during aging in APPsw transgenic mice provided cognitive benefit across multiple cogniti ve domains (reference learning/memory, working memory, recognition, strategy s witching – For detail ed descriptions of behavioral tasks, see (King and Arendash, 2002)) through mechanisms that do not seem to involve a reduction in brai n A deposition. Thos e results suggested that long-term, intensive cognitive stimul ation could be therapeutic in stabilizing or improving cognitive functi on in Alzheimer’s disease. One aim of this dissertation sought to answer the key question regarding whether pre-emptive EE, as a means of prophylaxis, can protect against the pathology of AD, or its asso ciated mental decline, and to further characterize the resultant molecular changes. At wean ing, doubly transgeni c mouse models of AD (hAPPV717F / hPS1M146L) were placed into eith er standard housing or an enriched environment. These mice were then behaviorally tested between the ages of 4-6 months. We found that AD mice raised in EE outperformed those raised in SH across a variety of behaviora l/memory tasks, in which they were statistically indistinguishable from nontransgenic (NT) mice. We found that, brain A deposition in Tg+ mice was not affected by EE alone, but only when in combination with the behavioral testing paradigm which is in and of itself a form of cognitive stimulation. When combined with behaviora l testing, EE resulted in large reductions in brain A deposition. Golg i staining revealed that the extent of dendritic branching and dendrit ic spine number in the brain were unchanged by

PAGE 29

22 enrichment, indicating that one effect of EE is to overcome the dendritic deficits present in transgenic AD mice. Micr oarray analysis of hippocampal tissue revealed that transgenic mice exposed to EE exhibited increase d expression of multiple memory-related genes, a number of neuroprotective genes involved in anti-apoptotic BAD phosphoryl ation such as IGF2, IGF BP2, and PRLR, as well as increased expression of a known A sequestering molecule, transthyretin.

PAGE 30

23 Paper I: Apolipoprotein is required for the forma tion of filamentous amyloid, but not for amorphous A deposition, in an APP/PS double transgenic mouse model of Alzheimer’s disease David A. Costa1,2, Lars N.G.Nilsson3, Kelly R. Bales4, Steven M. Paul4 and Huntington Potter1, 2, 5 1Department of Biochemistry and Molecula r Biology and Suncoast Gerontology Center, University of Sout h Florida, Tampa, FL 33620 2Johnnie B. Byrd Sr. Alzheimer’s Center and Research Inst itute, 15310 Amberly Dr., Tampa FL, 33647 3Department of Public Health and Cari ng Sciences, Uppsala University, Dag Hammarskjlds Vg 20, S-75185 Uppsala, Sweden 4Neuroscience Discovery Research, Lilly Research Laboratories, Indianapolis, Indiana, 46285 5H. Lee Moffit Cancer Center and Res earch Institute, Tampa FL, 33612 Corresponding author: Dr. Huntington Potter, Johnnie B. Byrd Sr. Alzheimer’s Center and Research Instit ute, Tampa FL 33647 tel. (813) 866-1600, fax (813) 866-1601, email. hpotter@hsc.usf.edu

PAGE 31

24 ABSTRACT To determine the role of apolipopr otein E (apoE) in the deposition of different forms of Alzheimer amyloid deposit, we studied mice expressing both mutant human amyloid -protein precursor (A PP) and presenilin 1 (PS1) that, in addition, were either normal or knockedout for apoE. By 7 months of age, extensive deposits of am orphous amyloid (A had developed equally in both lines, indicating that, when present in high amounts, A alone is sufficient for such deposition to occur. In contra st, filamentous, thio flavine S-positive amyloid deposition in APP/PS mice was catalyzed at least 3000 fold by apoE. Electron micrographs further illustrated the filam entous nature of A deposits in mice expressing apoE. These and other behavior data indi cate that the primary function of apoE in Alzheimer’s disease is to promote the po lymerization of A into mature, beta pleated sheet filament s, a process that is necessary for inducing cognitive decline. Thus, prevent ing apoE from binding to A may prove to be an effective means of therapeutic intervention. Key Words: Alzheimer’s Dis ease, Amyloid , Amyloid Protein Precursor Apolipoprotein E, Transgenic Mouse Model

PAGE 32

25 INTRODUCTION Several lines of investigation indicate that the deposition of A peptide into amyloid underlies the neurodege neration and cognitive dec line that characterize Alzheimer's disease (AD) (Hardy and Se lkoe, 2002). Using experimental animal models of AD, two proteins have been shown to be essential for both the formation of such deposits and for the cogniti ve decline in these animals (Games et al., 1995; Bales et al., 1999; Holtzman et al., 2000; DeMattos et al., 2004; Nilsson et al., 2004). The first is the A-peptide itself, particularly the A1-42 isoform. The second prot ein is apolipoprotein E (apoE) which, when produced from the 4 allele, is the strongest known risk factor for developing sporadic AD besides age itself (Wisniewski and Fr angione, 1992; Strittmatter and Roses, 1995). Increases in both A and apoE ex pression promote A deposition in a dose dependent manner (Games et al., 1995; Bales et al., 1999; Holtzman et al., 2000; Fryer et al., 2003). Because cognitive deficits in seve ral models of memory/learning observed in mutant APP-expressing mouse models of AD are dependent upon pathological chaperones such as apoE (and the A deposition that they promote) (Nilsson et al., 2004), it is important to know which part or product of the deposition process is controlled by apoE We therefore examined a mouse

PAGE 33

26 model of AD in which the production of A (and in particular, C-terminally extended and highly fibrillogenic A species such as A1-42) is not rate limiting, thus allowing the unique effect of apoE to be determined. Spec ifically, we took advantage of the fact that animals expressing mutant forms of both human APP and presenilin-1 transgenes from families ca rrying inherited Alzheimer's disease have a high ratio of A1-42 /A1-40 and develop large amount s of both amorphous and filamentous A (amyloid) deposits, resembling that of end-stage human AD (Duff et al., 1996; Scheuner et al., 1996; Citron et al., 1997; Holcomb et al., 1998). We then crossed PS1M146L+/-, APPV717F+/+ doubly transgenic animals with an apoE knockout mouse(Bales et al., 1999) to explore the effect(s) of apoE in a mouse model that develops early and severe AD-like pathology due to the relatively high levels of brain A1-42. Current hypotheses predict two alternative outcomes: 1) If apoE affects A availabi lity or clearance, it should have little positive or negative effect in our model due to the alr eady large amounts of PS1enhanced A production in PS1M146L+/-, APPV717F+/+ mice, or 2) If apoE promotes A polymerization, then its expressi on should increase the amount, and/or change the character of the A deposits in the brains of these mice. At seven months of age, both the apoE+/and apoE-/animals were euthanized, and the brains analyzed fo r AD-related pathology. Very large amounts of amorphous, non-fibr illar, A were generated equally in both sets of animals, reflecting the presence of large amounts of A1-42 in the brain. However, the quality of the deposits was strikingly dependent on the presence of apoE. Specifically, apoE expression induc ed a more than 3000-fold increase in

PAGE 34

27 the formation of thioflavine S-positive f ilamentous amyloid—to our knowledge the largest reported effect of any molecule on amyloid formation other than the mutant APP transgene itself. Indeed, in the absence of apoE, the area occupied by the few thioflavine S-positiv e A deposits was not sufficiently high to be statistically significantly different from the background thioflavine S staining in a completely non-transgenic mouse. MATERIAL AND METHODS Construction of transgenic mice All animals were housed in shoe box cages with static microisolator tops under climate-c ontrolled conditions on a 12 hour light/12 hour dark cycle, fed Harlan Teklad Globa l Diet #2018 and provided with tap water ad libitum. Heterozygous PDGF -hAPP(V717F) mice [Swiss-Webster x C57BL/6 x DBA/2] were crossed wit h PDGF-hPS1(M146L) heterozygotes to generate mice with an APP+/-,PS1+/genotype. These mice were subsequently crossed with ApoE-/mice [Swiss-Webster x C57BL/ 6 xDBA/2] to generate mice with genotype APP+/-,PS1+/-ApoE+/-. These mice were bred to generate either APP+/-,PS1+/-,ApoE+/, or APP+/-,PS1+/-,ApoE+/mice. All offspring were screened by PCR for identity the PDGF -hAPP gene(Games et al., 1995), the PDGF-hPS1 gene(Duff et al., 1996), as we ll as the mouse ApoE gene and neo gene (Bales et al., 1997). Immunohistochemical Procedures Mice that have been fasted over night were anesthetized with Nembutal (0.1mg/g). T he animals were then be intracardially

PAGE 35

28 perfused with 0.9% NaCl (25ml) followed by 50ml 4% paraformaldehyde in 1x Sorenson’s phosphate buffer. The mounted sections were processed through antigen retrieval in prewarmed 25mM citrate buffer (pH 7.3) at +82 C for 5 min and further processed as previously de scribed(Nilsson et al., 2001). The sections were incubated with primary antibodies against A (6E10, dil 1:5000) overnight at +4 C. Secondary antibody was anti-m ouse IgG developed with NovaRED substrate kit (Vector). Thioflavine S staining was performe d as previously described(Nilsson et al., 2001). Image Analysis Data were collected from three equally spaced coronal tissue sections for both dorsal hi ppocampus and overlying parietal cortex (Bregma 1.30 to -2.30 mm) for each mouse. The sections were examined with a Nikon Eclipse E1000 microscope using either 4X or 10X Plan Fluor objective lenses. A Retiga 1300 CCD (Qimaging) with a Qima ging RGB LCD-slider was used to capture images. For thioflavin S, a Nik on BV-2B fluorescence filter cube was used. Customized software written in Visu al Basic 6.0 (Micros oft) utilizing AutoPro function calls (Image Pro Plus, Medi a Cybernetics) was used to segment and quantify images. A deposition was calculated as percent area of interest (=Area Stainedtot/Area Measuredtot). Results were analyzed using a two-tailed, unpaired student’s t test with Welch’s correction. Electron Microscopy Tissue was fixed in 2.5% phosphate buffered glutaraldehyde (Electron Microscopy Sci ences) overnight at 4 degrees C. The cells were post-fixed in 1% osmium tetr oxide (Electron Microscopy Sciences) in

PAGE 36

29 the above buffer for 1 hour at 4 degrees C. Following buffer rinsing, the cells were dehydrated in a graded series of acetone, then in filtrated and embedded in LX112 plastic (Ladd Research Industries, Inc). Blocks containing tissue were polymerized overnight at 70 degrees C. Ce ll blocks were sectioned on a Reichert Ultracut E ultramicrotome (Leica, Inc), and sections were stained with 6 % uranyl acetate and Reynold's lead citrate. Sections were observed and photographed with a Philips CM10 electron micr oscope (FEI, Inc) at 60kV. RESULTS Total A deposition was measured in both hippocampus and parietal cortex as the percentage of 6E10 antibod y immunoreactive staining. Figures 1ab show that overall A deposition was not much affected by the presence or absence of apoE in these PS/APP mice. No significant difference was found in either the hippocampus (apoE +, 5.5%1.3 vs. apoE KO, 8.3%3.2) or parietal cortex (apoE+, 3.0%1.3 vs. apoE KO, 1.5%0.3) (Figure 1c). Although total A deposition was st atistically similar in the apoE+/and apoE-/mice, there was a clear differenc e in the regional distribution and morphology of the A deposits. The presence of apoE promoted the formation of more compact-type plaque structures th roughout the hippocam pus and parietal cortex (Figure 1a), while mi ce without apoE contained few, if any, compact-like plaques anywhere in the brai n, instead manifesting a rather diffuse pattern of deposition. Furthermore, mice lacking apoE developed almost no A in the

PAGE 37

30 granular and molecular layers of the dent ate gyrus of the hippocampus (Figure 1b). To confirm the distinction between am orphous A deposition and dense core plaques, sections were stained with thioflavine S, which detects -pleated sheet structures characteristic of mature amyloid plaques. Thioflavine S staining was elevated in PS1M146L+/-, APPV717F+/mice containing one copy of endogenous apoE by 3200 fold in both hippocampal and cortical regions as compared to mice with the apoE-/genotype (Figure 2). The increase was statistically significant in both the hippocampus (apoE+/-, 0.64%0.18 vs. apoE-/-, 0.0002%0.0001, p=0.0004) and t he parietal cortex (apoE+/-, 0.35%0.04 vs. apoE-/-, 0.0001%0.00005, p=0.0001). Becaus e there was a difference in staining between apoE+/and apoE-/mice of three orders of magnitude, a logarithmic scale was necessary to resolve the low levels of thioflavine S staining in the apoE-/animals (Figure 2c-d). Electron microscopy confirmed the presence of extensive filamentous amyl oid deposits in the PS1/APP/apoE+/animals and only amorphous deposits in the PS1/APP/apoE-/animals (Figure 2). DISCUSSION In sum, we report that doubly transgenic mice with both human PS1M146L and APPV717F mutations do not develop significant thioflavine S-positive amyloid plaque deposits by 7 months in the absence of apoE expression, despite the accumulation of large amounts of immunor eactive A deposition. Only the presence of apoE allows mature filam entous amyloid deposits to form. These

PAGE 38

31 data contrast with previous reports ut ilizing mice expressing only a human APP mutation (PDAPP mice) in which apoE has been shown to promote both total A deposition and amyloid formation. Eviden tly, because the APP/PS1 double transgenic mice express very high levels of A1-42 in the brain, perhaps overriding any effect of apoE on A production or clear ance, they provide a clear proof that apolipoprotei n E directly catalyzes the polymerization of A into sheet fibrils and its consequent aggregation to form mature amyloid plaques. In vitro (Ma et al., 1994) and in vivo (Nilsson et al., 2001) studies show that both of the major pathogenic chaper ones found in amyl oid plaques, apoE and ACT, promote A filament formation, with apoE show ing the higher catalytic activity. Since ACT and apoE act synergi stically with regard to A deposition, future studies with ACT expressing mice co uld show even more drastic amyloid deposition. Furthermore, since endogenous m ouse-apoE is most related to the human 4 allele of apoE, the hu man isotype-specific e ffects of the human apoE alleles, specifically the pr otective properties of the 2 allele should also be further explored in this mouse m odel of AD. Inasmuch as the presence of am yloid plaques is pathonomic for Alzheimer's disease, preventing apoE from binding to A and stimulating its polymerization into -sheet filaments ma y prove to be a pr omising therapeutic approach for AD drug development. An indi cation that such an approach may be effective has been provided by in vitro experiments in which small fragments of the A peptide were shown to comp letely inhibit the apoE-catalyzed

PAGE 39

32 polymerization of A1-42 into neurotoxic amyloid filaments (Ma et al., 1994; Ma et al., 1996).

PAGE 40

33 ACKNOWLEDGEMENTS The research was supported by a grant AG09665 from the National Institute on Aging. H.P. occupies the Eric Pfeiffer Chair for Research in Alzheimer’s disease at the Suncoast Ger ontology Center at the University of South Florida and is the CEO of the Johnni e B. Byrd Sr. Alzheimer’s Center and Research Institute. We thank Ed Haller and the support, in part, of the Pathology Core Facility, Electron Microscopy Laboratory, at the University of South Florida, College of Medicine and at H. Lee Moffitt Cancer C enter& Research Institute for the electron micrographs.

PAGE 41

34 Fig. 1. Little or no effect of ApoE on total A deposition Total A deposition, diffuse and compact, in 7-month-old PS1M146L+/-, APPV717F+/+, ApoE+/mice (a) and 7-month-old PS1M146L+/-, APPV717F+/+, ApoE-/mice (b), as measured by 6E10 immunostaining. (c) Quantitative image analysis of total A deposition from all investigated animals. Hippocampal amyloid load (left) or cortical amyloid load (right). Solid bar, PS1M146L+/-, APPV717F+/+, ApoE+/(n=3); open bar, PS1M146L+/-, APPV717F+/+, ApoE-/(n=3)

PAGE 42

35

PAGE 43

36 Fig. 2. Strong catalytic effect of ApoE on mature amyloid deposition Thioflavine S plaque staining is incr eased over 3200 fold in 7-month-old PS1M146L+/-, APPV717F+/+, ApoE+/mice (a) compared to 7-month-old PS1M146L+/-, APPV717F+/+, ApoE-/mice (b). Quantitative im age analysis of compact amyloid deposition from all investigat ed animals is represented on a linear y-axis (c) or logarithmic y-axis (d) m easuring hippocampal amyloid load (left) or cortical amyloid load (right). Solid bar, PS1M146L+/-, APPV717F+/+, ApoE+/(n=3); open bar, PS1M146L+/-, APPV717F+/+, ApoE-/(n=3); ***, p<0.001. Transmission electron microscopy reveals the filamentous nat ure of plaques in the apoE+/mice (E, 6300X and G 44,000X) as compared to the more amorphous deposits found in the apoE-/mice (F and H).

PAGE 44

37 ApoE ApoE KO ApoE ApoE KO 0.00 0.25 0.50 0.75 1.00 Hippocampus Cortexc ***%Area Thioflavin S*** ApoE ApoE KO ApoE ApoE KO 0.0001 0.0010 0.0100 0.1000 1.0000 Hippocampus Cortexd% Area Thioflavin S *** ***a b g h e f

PAGE 45

38 REFERENCES Bales KR, Verina T, Cummins DJ, Du Y, Dodel RC, Saura J, Fishman CE, DeLong CA, Piccardo P, Petegnief V, Ghe tti B, Paul SM (1999) Apolipoprotein E is essential for amyloid deposition in the APP(V717F) transgenic mouse model of Alzheimer's disease. Proc Natl Acad Sci U S A 96:15233-15238. Bales KR, Verina T, Dodel RC, Du Y, Altsti el L, Bender M, Hyslop P, Johnstone EM, Little SP, Cummins DJ, Piccardo P, Ghetti B, Paul SM (1997) Lack of apolipoprotein E dramatically reduce s amyloid beta-peptide deposition. Nat Genet 17:263-264. Citron M, Westaway D, Xia W, Carlson G, Diehl T, Levesque G, Johnson-Wood K, Lee M, Seubert P, Davis A, Kholodenko D, Motter R, Sherrington R, Perry B, Yao H, Strome R, Lieberburg I, Rommens J, Kim S, Schenk D, Fraser P, St George Hyslop P, Selkoe DJ (1997) Mutant presenilins of Alzheimer's disease increase production of 42-residue amyloid be ta-protein in both transfected cells and transgenic mice. Nat Med 3:67-72. DeMattos RB, Cirrito JR, Parsadanian M, May PC, O'Dell MA, Taylor JW, Harmony JA, Aronow BJ, Bales KR, Paul SM, Holtzman DM (2004) ApoE and Clusterin Cooperatively Suppr ess Abeta Levels and Deposition. Evidence that ApoE Regulates Extrace llular Abeta Metabolism In Vivo. Neuron 41:193-202.

PAGE 46

39 Duff K, Eckman C, Zehr C, Yu X, Pr ada CM, Perez-tur J, Hutton M, Buee L, Harigaya Y, Yager D, Mor gan D, Gordon MN, Holcomb L, Refolo L, Zenk B, Hardy J, Younkin S (1996) Increased am yloid-beta42(43) in brains of mice expressing mutant preseni lin 1. Nature 383:710-713. Fryer JD, Taylor JW, DeMattos RB, Bales KR, Paul SM, Parsadanian M, Holtzman DM (2003) Apol ipoprotein E markedly facilitates age-dependent cerebral amyloid angiopathy and spont aneous hemorrhage in amyloid precursor protein transgenic mice. J Neurosci 23:7889-7896. Games D, Adams D, Alessandrini R, Bar bour R, Berthelette P, Blackwell C, Carr T, Clemens J, Donaldson T, Gillespie F, et al. (1995) Alzheimer-type neuropathology in transgenic mice ov erexpressing V717F beta-amyloid precursor protein. Nature 373:523-527. Hardy J, Selkoe DJ (2002) The amyloid hypothesis of Alzheimer's disease: progress and problems on the road to therapeutics. Science 297:353-356. Holcomb L, Gordon MN, McGowan E, Yu X, Benkovic S, Jantzen P, Wright K, Saad I, Mueller R, Morgan D, Sanders S, Zehr C, O'Campo K, Hardy J, Prada CM, Eckman C, Younkin S, Hsiao K, Du ff K (1998) Accelerated Alzheimer-type phenotype in transgenic mice carrying both mutant amyloid precursor protein and presenilin 1 transgenes. Nat Med 4:97-100.

PAGE 47

40 Holtzman DM, Bales KR, Tenkova T, Fagan AM, Parsadanian M, Sartorius LJ, Mackey B, Olney J, McKeel D, Wozniak D, Paul SM (2000) Apolipoprotein E isoform-dependent amyloid deposition and neuritic degeneration in a mouse model of Alzheimer's disease. Proc Natl Acad Sci U S A 97:2892-2897. Ma J, Brewer HB, Jr., Potter H (1996) Al zheimer A beta neurotox icity: promotion by antichymotrypsin, ApoE4; inhibiti on by A beta-related peptides. Neurobiol Aging 17:773-780. Ma J, Yee A, Brewer HB, Jr., Das S, Potter H (1994) Amyloidassociated proteins alpha 1-antichymotrypsin and apolipoprotei n E promote assembly of Alzheimer beta-protein into filam ents. Nature 372:92-94. Nilsson L, Arendash GW, Leighty RE, Costa DA, Low MA, Garcia MF, Cracciolo JR, Rojiani A, Wu X, Bales KR, Paul SM Potter H (2004) Cognitive impairment in PDAPP mice depends on ApoE and ACT-cata lyzed amyloid formation. Neurobiol Aging. Nilsson LN, Bales KR, DiCarlo G, Gor don MN, Morgan D, Paul SM, Potter H (2001) Alpha-1-antichymotr ypsin promotes beta-sheet amyloid plaque deposition in a transgenic mouse model of Alzheime r's disease. J Neurosci 21:1444-1451.

PAGE 48

41 Scheuner D, Eckman C, Jensen M, Song X, Citron M, Suzuki N, Bird TD, Hardy J, Hutton M, Kukull W, Larson E, Le vy-Lahad E, Viitanen M, Peskind E, Poorkaj P, Schellenberg G, Tanzi R, Wasco W, Lannfelt L, Selkoe D, Younkin S (1996) Secreted amyloid beta-protein similar to that in the senile plaques of Alzheimer's disease is increased in vivo by the presenilin 1 and 2 and APP mutations linked to familial Alzheimer's disease. Nat Med 2:864-870. Strittmatter WJ, Roses AD (1995) Apoli poprotein E and Alzheimer disease. Proc Natl Acad Sci U S A 92:4725-4727. Wisniewski T, Frangione B (1992) Apolip oprotein E: a pathological chaperone protein in patients with cerebral and syst emic amyloid. Neurosci Lett 135:235238.

PAGE 49

42 Paper II: Parabiosis as means to investigate the ro le of peripherally derived proteins in Alzheimers disease pathology D.A. Costa1,7, L.N. Nilsson3,7, S. Gografe2,7, T. Hughes1, D. Dressler4, K.R. Bales5, S.M. Paul5and H. Potter1,6 1Department of Biochemistry and Molecula r Biology and Suncoast Gerontology Center, 2Division of Comparative Medicine, University of South Florida, Tampa, FL 33612 3Department of Public Health and Cari ng Sciences, Uppsala University, Dag Hammarskjlds Vg 20, S-75185 Uppsala, Sweden 4Balliol College and Oxford Univer sity, Oxford England, OX13QU 5Neuroscience Discovery Research, Lilly Research Laboratories, Eli Lilly and Company, Indianapolis, IN 46285 6 Johnnie B. Byrd, Sr. Alzheimer’s Center and Research Institute and Department of Biochemistry and Molecu lar Biology, College of Medicine, 7These authors contributed equally Financial Support: Funding provided by National Institute on Aging grant #AG09665 & Alzheimer's Asso ciation grant #IIRG-013111 Corresponding author: Dr. Huntington Potter, Departm ent of Biochemistry and Molecular Biology, College of Medicine, MDC07, Univer sity of South Florida, Tampa, FL 33612, tel. (813) 974-53 69, fax (813) 974-7357, email. hpotter@hsc.usf.edu

PAGE 50

43 ABSTRACT Apolipoprotein E (apoE) is expre ssed in the brain and present in Alzheimer’s disease (AD) amyloid plaques. The apoE4 allele, the main genetic risk factor for late-onset AD, increases amyloid-beta peptide fibrillization and amyloid plaques density. ApoE is also synthesized in the liver and plays a crucial role in cholesterol metabolism, for transgenic mice devoid of apoE show hypercholesterolemia and cardiovascular li pid deposition. Here we investigate the neuropathology of transgenic PS1+/-,APP+/Alzheimer’s disease mouse models, either with or wit hout the apoE gene, using par abiosis to determine if apoE in the peripheral circulation influ ences amyloid formation. This surgical procedure allows free exchan ge of proteins via periphera l circulation. We show that apoE is found in the plasma of parabiosed PS1+/-,APP+/-,apoE-KO mice, resulting in a reversion of hyperchol esterolemia. Unexpectedly, amyloid deposition is reduced in parabiosed PS1+/-,APP+/-,apoE-KO mice as compared to unoperated controls. Our findings demonst rate that apoE in the peripheral circulation slightly reduces amyloid bur den, most clearly in transgenic mice devoid of apoE, by likely promoting effl ux of amyloid-beta peptides from the brain. These findings further reinforc e that the mechanisms whereby apoE affects amyloid-beta metabolism are comp lex, and that the modulation of peripheral apoE metabolism is not likel y to profoundly impact Alzheimer’s disease neuropathology.

PAGE 51

44 INTRODUCTION Amyloid deposits, which are primarily composed of the amyloid beta peptide (A), are a key feature of Alzheimer’ s disease (AD) neuropathology, and their formation is likely to contribute to t he neuronal degeneration that underlies clinical dementia. Certainly, genetic data point to A as the central component of the Alzheimer’s pathogenic pathway. Howeve r, several other proteins are found in amyloid deposits, and some of these have been proven to alter the rate of amyloid filament formation. The best c haracterized of thes e amyloid promoters are apolipoprotein E (apoE) and 1-antichymotrypsin (ACT), which have been shown to facilitate A deposition in vitro and in APP transgenic mouse models (Ma et al., 1994; Bales et al., 1997; Holt zman et al., 1999; Holtzman et al., 2000; Nilsson et al., 2001). Indeed, without apoE or ACT, ev en at 2 years of age, roughly 50% of homozygous PDAPP mice still do not develop filamentous amyloid (Nilsson et al., 2004). These findin gs agree nicely with the demonstration that postmortem AD brain of ApoE4 carriers have increas ed plaque density (Corder et al., 1993; Rebeck et al., 1993). A long-standing question in AD research is whether A and other proteins that are present in the peripheral circul ation are involved in amyloid formation and the associated brain pathology. Plasma A levels, particularly A1-42, are elevated in some familial and sporadic AD patients (Scheuner et al., 1996) and have been suggested as a potent ial AD biomarker. In addition, serum ACT and apoE levels have been shown to be generally increased in AD patients (Licastro

PAGE 52

45 et al., 1995; Lieberman et al., 1995; Bl ain et al., 1997), and apoE has been shown to shuttle A across the blood br ain barrier (Martel et al., 1997). It is therefore important to analyze whether A peptides apoE, and ACT in the peripheral circulation may have a pathogenic role in AD with respect to amyloid deposition in the brain. Parabiosis is a technique that enables shared perip heral circulation between surgically joined animals and thus give s us the ability to address such questions (Martinez et al., 1959). It allows the infl uence of circulating factors or hormones to be studied, since any circulating pr otein produced in one mouse will be passed to its parabiosed part ner and vice versa (Finerty, 1952). In the present study, we used parabiosis to investigate the effect of apoE present in the peripheral circulation on brain amyloi d deposition in a transgeni c model of Alzheimer’s disease. Transgenic mice expressing both the human mutant presenillin-1 (PS1M146L) and the human mutant amyloi d precursor protein (APPV717F) genes were parabiosed. However, one of t he parabiosed mice was heterozygous for endogenous apoE, while its parabiosed par tner lacked apoE expression altogether. PS1+/-,APP+/transgenic mice rapidly develop robust A-immunoreactive deposition by 3 to 4 months of age. The pathology main ly consists of mature cored plaques, which is totally prevent ed in the absence of apoE in this transgenic model (Costa et al ., 2004). We considered PS1+/-,APP+/mice, with or without apoE, to be an excellent choice for parabiosis experiments due their rapid onset of amyloid pathology. Furt hermore, the strong impact of apoE on

PAGE 53

46 amyloid deposition in this model is ideal, since it is likely that even minute amounts of apoE in the cent ral nervous system will strongly affect the amyloid phenotype of a parabiosed PS1+/-,APP+/-,apoE-KO mouse. Thus, parabiosis of such transgenic mice would allow us to determine whether the pathological chaperone, apoE, needs to be synthesized within the brain or whether its presence in the peripheral circulation is sufficient to promote amyloid formation. This knowledge will allow strategies for therapeutic interventions against Alzheimer disease to be more appropriately devised. MATERIAL AND METHODS All described procedures were approv ed by the Institutional Animal Care and Use Committee and conducted in comp liance with the “Gui de for the Care and Use of Laboratory Animals”. All anim als were housed in shoe box cages with static microisolator tops under climate-c ontrolled conditions on a 12 hour light/12 hour dark cycle, fed Harlan Teklad Globa l Diet #2018 and provided with tap water ad libitum The animal facility maintains a specific pathogen free status based on a sentinel system including colony represent atives and quarantine of animals received from non-approved vendors or other institutions, respectively. Parabiosis Six week-old siblings of the same sex were selected for parabiosis. Both animals were transgenic for APP (PDGF-hAPPV717F, human mutant amyloid precursor protein) and PS1 (PDGF-hPS1M146L, human mutant presenilin-1).

PAGE 54

47 Subsequent crossings of these mice with an apoE-/mouse generated the APP+/-,PS1+/-,apoE+/genotype. This schedule allowed t he generation of siblings with genotypes PS1+/-,APP+/-,apoE+/and PS1+/-,APP+/-,apoE-/as partners for the parabiosis. Animals were an esthetized with 100 mg/kg, Xylazine 20 mg/kg, and Acepromazine 3 mg/kg. For the parabiosis the mice were placed in a parallel orientation, and a left lateral incision was made on one mouse while a right one was made on the partner mous e, extending from the bas e of the ear toward the middle of the femur of the extended pelvic extremity. The incision included skin and muscle along thorax and ab domen. Starting at the la st rib, the opening was extended into the abdominal cavities to accomplish convergence. The peritonea and muscle layers of the two animals were joined by simple interrupted suture with 4-0 PDS (polydioxanone) wher eas the skin was closed via stainless steel clips. The animals were allowed to recover in a wa rm, clean environment before being transferred into the husbandry area. Due to the expected immunosuppression, prophylactic antibioti c treatment (Enrofloxacin, 5mg/kg), was started one day prior to surgery, and was continued for three days. All animals received analgesic/anti-inflammato ry treatment (Acetyl salicylic acid 5mg/kg), for 14 days. Six months afte r surgery, the parabiotic pairs were euthanized for brain analysis at seven months of age.

PAGE 55

48 Immunohistochemi cal Procedures Parabiosed mice were fasted overni ght and subsequently anesthetized with Nembutal (0.1mg/g body weight). Blood was collected by cardiac puncture and immediately supplemented with 0.1% (w/v ) EDTA followed by centrifugation (2000xG, 15min). Plasma was aspirat ed and total cholesterol was measured with a colorimetric assay (Infinity C holesterol Reagent procedure 401, Sigma). The animals were then intrac ardially perfused with 0.9% NaCl (25ml) followed by 50ml 4% paraformaldehyde in 1x Sore nson’s phosphate buffer. Brains were cryoprotected through sequential imme rsion in 10%, 20% and 30% sucrose and sectioned (25 m) using a sledge microtome. The mounted brain sections were processed through antigen retrieval in prew armed 25mM citrate buffer (pH 7.3) at +82 C for 5 min and further processed as previously described(Nilsson et al., 2004). The sections were incubated with primary antibodies against A (6E10, dil 1:5000, Signet and rA40, dil 1:3000, QCB) and apoE (AB947, dil 1:5000, Chemicon) overnight at +4 C. Secondary antibodies were anti-rabbit IgG or antimouse IgG (1:300, Vector). The imm unostaining was visualized with a NovaRED substrate kit (Vector). Thioflavine Sstaining was performed as previously described (Nilsson et al., 2004). For each m ouse, data were coll ected from three equally spaced coronal tissue secti ons for both dorsal hippocampus and overlying parietal cortex (Bregma -1.30 to -2.30 mm). The sections were examined with a Nikon Eclipse E1000 microscope using ei ther 4X or 10X Plan Fluor objective lenses. A Retiga 1300 CCD (Qimaging) wit h a Qimaging RGB LCD-slider was used to capture images. For thioflavin S stained sections, a

PAGE 56

49 Nikon BV-2B fluorescence filter cube was used. Customized software, written in Visual Basic 6.0 (Microsoft) utilizing Au to-Pro function calls (Image Pro Plus, Media Cybernetics) was used to s egment and quantify images A deposition was calculated as percent ar ea of interest (=Area Stainedtot/Area Measuredtot) from no less than seven microscope fields Results were analyzed using a twotailed, unpaired student’s t test with Welch’s correction. RESULTS Apolipoprotein E is transferred to the parabiosed apoE-KO mice via blood circulation. Endogenous apoE in the plasma of nontransgenic mice can be detected at up to 100-fold dilutions. Plasma apoE expression was examined with western blot analysis to verify that apoE from the apoE-positive parabiont was able to gain access to the apoE-kno ckout mouse. ApoE, in PS1+/-,APP+/+,apoEKO mice that had been parabiosed to PS1+/-,APP+/+,apoE+/mice, was detected at a level corresponding to about 5% of that found in nontransgenic mice, as determined by densitometric analysis. In contrast no apoE wa s found in control non-parabiosed apoE-KO mice (Figure 1A). This finding could be interpreted as restricted exchange of plas ma proteins, such as ap oE, between the partners. Alternatively the low plasma apoE le vels in the parabiosed apoE-knockout partner could be due to t he need to quickly sequester lipoprotein particles from the blood stream in these mice. This would tend to increase turnover and lower steady state levels of apoE in the plasma We decided to determine whether the parabiosis allowed free ex change of blood cells betw een parabiosed partners by

PAGE 57

50 analyzing serum DNA. PCR analysis revealed that parabiosed apoE-KO mice contain almost equal amount s of apoE DNA in their blood as their apoE heterozygous parabiosis partners. This DNA analysis demonstrates a successful and efficient white blood cell transfer betw een the two parabiont s (Figure 1B). Apolipoprotein E transferred th rough parabiosis prevents hypercholesterolemia in the apoE-KO mice. Mice with targeted disruption of the mouse-apoE gene are known to devel op severe hypercholesterolemia and atherosclerosis (Plump et al., 1992; Z hang et al., 1992). Accordingly, the PS1+/-,APP+/+,apoE-KO mice used in these experiment s show elevated levels of serum cholesterol as compared to nontransgenic, wild type, mice and PS1+/-,APP+/+ mice with the endogenous murine apoE gene (Table 1). Nontransgenic mice had, on average, 1056 mg/dl (n=19) of to tal cholesterol in their plasma, while PS1+/-,APP+/+ transgenic mice with only one copy of apoE (PS1+/-,APP+/+,apoE+/-) had levels averaging near 796 mg/dl (n=8). In contrast, th e levels of total cholesterol were roughly four times higher in PS1+/-,APP+/+,apoE-KO mice(392121 mg/dl, n=3) and five time s higher in apoE-KO mice lacking APP expression (50139 mg/dl, n=13). In PS1+/-,APP+/+,apoE-KO mice that had been parabiosed with a partner harboring on e copy of the murine apoE gene, cholesterol levels in the apoE knock-out mice were rescued (125 mg/dl for a 5month PS1+/-,APP+/+,apoE-KO mouse and 87 mg/dl for a 7 month PS1+/-,APP+/+,apoE-KO mouse, Table 1). The la tter result was reproduced in all

PAGE 58

51 parabiosed pairs whose A-immunoreacti ve and amyloid deposition was quantified. Apolipoprotein E in the peripheral ci rculation alone does not reach the brain parenchyma and is unable to promote brain amyloid deposition Immunohistochemical analysis of A-immu noreactive deposition using the 6E10 monoclonal anti-A antibody on coronal br ain sections from parabiosed partners and their respective non-parabiosed controls at seven months of age revealed only minor differences (Figure 2). In the ce rebral cortex no statistically significant difference in A burden between the parabiosed (1.55%0.5; n=6) and the nonparabiosed (1.50%0.1; n=3) PS1+/-,APP+/+,apoE-KO mice was found. This suggests that apoE present in the peripheral circulat ion is not sufficient to promote brain A-immunoreactive depositi on. There also was no statistically significant difference in A burden in cerebral cortex between the parabiosed (2.2%0.5; n=6) and the nonpar abiosed (3.0%0.7; n=3) PS1+/-,APP+/+,apoE+/mice. However, a slightly greater amount (p=0.043) of A deposition was found in the hippocampus, of the nonparabiosed PS1+/-,APP+/+,apoE+/mice (5.5%0.7; n=3) as compared to the parabiosed PS1+/-,APP+/+,apoE+/mice (3.3%0.7; n=6; Figure 2F). Again, there was no statistica lly significant difference in A burden between parabiosed and control PS1+/-,APP+/+,apoE-KO mice in the hippocampus (Figure 2F). ApoE-immunoreacti vity was detected in virtually all amyloid plaques of parabiosed PS1+/-,APP+/+,apoE+/in the brain (Figure 3a,b) and in astrocytes of nontrans genic mice (Figure 3c,d). In contrast, parabiosed

PAGE 59

52 PS1+/-,APP+/+,apoE-KO with an A burden showed some apoE-immunoreactive staining in the choroid plexus, but none in the brain parenchyma (Figure 3e), suggesting that apoE in the peripheral circulation is essentially unable to cross the blood brain barrier to reach the brai n parenchyma thereby limiting its impact on AD neuropathology. Control exper iments verified that no apoEimmunostaining was present in c horoid plexus of unoperated PS1+/-,APP+/+,apoE-KO mouse (Figure 3f). Amyloid deposition is reduced in parabiosed mice lacking ApoE. Previous studies have shown a striking diffe rence in the structure of amyloid and A immunoreactive deposition in PS1+/-,APP+/+ mice expressing endogenous apoE as compared to PS1+/-,APP+/+,apoE-KO mice (Costa et al., 2004). This structural difference is not associated with similar quanti tative differences in the A-immunoreactive burden. In fact, im munohistochemical detection of A immunoreactive deposition sh owed equal amounts in PS1+/-,APP+/+ mice with and without apoE, albeit with a different anat omic distribution. In the present study, Thioflavin S staining was performe d to compare the levels of mature, compact, amyloid between the parabiosed mi ce pairs and their respective nonparabiosed controls. No statistically si gnificant difference in the %-area of Thioflavin S-staining wa s found between parabiosed PS1+/-,APP+/+,apoE+/mice and non-parabiosed controls of the same genotype (Figure 4a). Because the deposition of compact amyloid is a specific effect of apo E, in general, only minute amounts of compact amyloid were found in apoE-/mice. The few, detectable,

PAGE 60

53 plaques were counted at 400x magnification. Rema rkably, even though apoE knock-out mice develop only a very low amount of amyloid deposition, this number was further reduced in PS1+/-,APP+/+,apoE-KO mice that had been parabiosed. Both, the hippocampal and cortical regions showed a significant difference between parabiosed PS1+/-,APP+/+,apoE-KO mice (hippocampus, 17.25.4; n=6 and cortex, 13.0 3.0; n=6) and non-parabiosed PS1+/-,APP+/+,apoE-/mice (hippocampus, 28.33.0 n=3; and cortex, 20.75.2; n=3) in the number of plaques per brain secti on (Fig. 4b). Thus, the amyloid plaque loads were 53% and 112% higher for the hippocampus and the cerebral cortex respectively in non-parabiosed mice versus parabiosed apoE-KO mice. Similarly, the optical density of thio flavine S plaque de position was also significantly higher in the apoE-KO contro l mice as compared to those that were not parabiosed (data not shown). DISCUSSION It has long been debated to what extent proteins derived from the blood stream contribute to Alzheimers diseas e pathology. A hemat ogenous origin of amyloidogenic proteins in AD is suppor ted by the fact that systemic amyloid disorders are driven by high concentrations of these proteins in the peripheral circulation. Furthermore, the frequent pr esence of cerebral amyloid angiopathy (CAA) in AD postmortem brain might affect BBB permeability and further increase influx and efflux of amyloidogenic proteins su ch as apoE and A. It is known that A in the blood stream can reach the brain and adhere to amyloid

PAGE 61

54 deposits, as demonstrated through intra-arteri al injection of la beled [125I]-A in squirrel monkeys (Ghilardi et al., 1996). Serum amyloid P component, which is derived from the peripheral ci rculation and believed to render amyloid plaques in AD brain less prone to clearance, is an example of a peripheral protein that is suspected to impact AD n europathology (Pepys et al ., 2002). Developing an AD therapy against a peripheral target ha s clear pharmacological and clinical advantages. Here we show that parabiosis is a feasible approach to determine whether a given protein in the blood str eam is involved in AD amyloidosis. This procedure allows free flux of apoE between the parabiosed partners, as assessed by the level of blood cell DNA. Despite this fa ct, we observed rather low steady state levels of apoE (5% of nontransgenic mice ) in parabiosed apoE knockout mice. We speculate that the low steady state levels of plasma apoE are caused by the rapid turnover of apoE and the high demand for apoE to clear lipoprotein particles in the apoE knockout mice. The very rapid clearance of plasma apoE (t1/2= 20 min), through uptake by the liver (Vogel et al., 1985; Mahley et al., 1989), does support this hypothesis. The low amounts of apoE transferred through parabiosis are however sufficient to normalize the hyperchol esterolemia of apoE-knockout mice. Previous studies have also shown that a rather small amount of apoE is sufficient to restore essentially normal cholestero l metabolism in mice. Normalization of hypercholesterolemia of apoE knockout mice has been shown by supplementation with apoE through bone marro w transplantation (Linton et al.,

PAGE 62

55 1995) or from hypomorphic apoE alleles (Raffai and Weisgraber, 2002). In fact only 3% of wild-type apoE levels in plasma appear to be required to maintain normal plasma cholesterol levels, and even less plasma apoE is needed to reduce atherosclerosis and lipid deposition in the aorta (Thorngate et al., 2000). ApoE gene dosage also str ongly determines the onset and extent of amyloid deposition in amyloid precur sor protein transgenic mice The absence of apoE anatomically redistributes A immunoreactivi ty in the brain and essentially blocks amyloid deposition (Bales et al., 1999; Holtzman et al., 2000; Nilsson et al., 2004). Experiments in bigenic mice containing both presenilin-1 and APP demonstrated that the main effect exerted by apoE is on the fo rmation of senile (Thioflavine S positive) plaques (Costa et al., 2004). Thus, we considered the crossed PS1xAPP animal model to be an i deal animal model to evaluate the role of peripheral apoE on A and amyloid pathol ogy. It is reasonable to anticipate that even a very small amount of apoE entering the CNS compartment would profoundly facilitate amyloi d pathology in such a transgenic model. To our surprise we did not see any enhanced amyloid deposition in parabiosed PS1+/-,APP+/-,apoE-/mice. In fact, on the contrary, we found a reduced number of Thioflavine S-positive deposits in the parabiosed mi ce that were de void of apoE. These results show that very little, if any, apoE synthesized in the periphery reaches the brain and that apoE derived from the per iphery does not accelerate the amyloid pathology. We were only able to detect apoE-immunoreactivity in the choroid plexus of parabiosed PS1+/-,APP+/-,apoE-/mice, which is in

PAGE 63

56 agreement with studies where 125I-labeled apoE has been infused via the carotid artery in guinea pigs (Martel et al., 1997). Amyloid deposition, at least in APP transgenic mice, is a rather dynamic process whereby the size of an individual plaque can grow or shrink dramatically within months (Christie et al., 2001). Clear ly different pools of A in the body exist in a dynamic equilibrium prior to onset of amyloid deposition. There is, for example, a strong correlation between A levels in the CS F, plasma, and brain of young PDAPP mice (DeMattos et al., 2002b) and this equilibrium can be quickly altered by injecting a high-affinity A ant ibody that is able to sequester soluble A in the periphery (DeMattos et al ., 2001; DeMattos et al., 2002a). It is interesting to note that peripherally inject ed A is rapidly cleared by the liver in nontransgenic mice, but not in apoE knoc kout mice (Hone et al., 2003). Thus, it could be that our finding of reduced amyloid deposition in parabiosed PS1+/-,APP+/+,apoE-KO mice is due to apoE in the periphery clearing A. The peripheral clearance of A w ould then shift the equilibrium towards efflux of A out of the brain and thereby reduce/delay amyloid depos ition in the brains of parabiosed PS1+/-,APP+/+,apoE-KO mice. ApoE could thereby act as a sink to drain A out of the brai n and into the peripheral circulation for subsequent clearance by the liver. The sink hypothesis of A me tabolism in APP transgenic mice was introduced as a model to explai n the effects of passive vaccination against A. Anti-A, but also apoE, wa s found to sequester A in a dialysis system (DeMattos et al., 2002a). A sim ilar mechanism might also explain the tendency of reduced A and am yloid burden in the brains of parabiosed PS1+/-

PAGE 64

57 ,APP+/+,apoE+/mice as compared to their unoperat ed siblings. Thus, drainage of apoE in the blood of the parabiosed PS1+/-,APP+/+,apoE+/mice, through the shared circulation with the apoE KO mice could be compensated for by an efflux of apoE from the brain to the periphery in these mice Such transfer of apoE would tend to transport A out of t he CNS compartment in the parabiosed PS1+/-,APP+/+,apoE+/mice, and thereby slow amyloid deposition. Previous transgenic experiments do suggest that human apoE can shift the equilibrium of A between different compartment s and favor transport of A out of the brain. The mechanisms of A efflux from the brain mi ght be both by bulk flow of ISF into the CSF, as well as across the blood brain barrier (Shibata et al., 2000; Silverberg et al., 2003). The low-density lipoprotein recept or-related protein, which is an apoE receptor, is a suspected receptor for A efflux (Zlokovic, 2004). No apoE-A complexes relevant to such transpor t mechanism could be shown in these experiments (Shibata et al., 2000), how ever other findings are strongly suggestive of apoE-A complexes existing in the ISF and also regulating A clearance. Microdialysis experiments hav e shown that the presence of apoE extends the half-life of A in the ISF, which could be due to the existence of a reservoir of A as apoE-A complexes in the ISF (DeMattos et al., 2004). Furthermore, the human glial fibrillary acidic protein promoter GFAP-apoE3 transgene, when crossed to an APP/apoE-K O genotype, lowered both soluble A levels in the brain and the ratio of CSF/plasma A42 in a dose dependant manner prior to onset of amyloid depositi on (DeMattos, 2002). The latter findings strongly suggest that apoE ex erts dual functions with respect to A metabolism

PAGE 65

58 by both facilitating amyloid deposition, but also by clearing A from the CNS to plasma through efflux mechanisms. In su m, our findings suggest that apoE in the peripheral circulation does not promote A deposition in the brain, but clears A by favoring its efflux from the brai n into the periphery. These findings reinforce the idea that agents that ar e able to strongly sequester and clear soluble A monomers in the periphery, such as A antibodies (DeMattos et al., 2001) or other high-affinity A bi nders (Matsuoka et al., 2003), could be promising novel ther apeutics in AD.

PAGE 66

59 Table 1 – Apolipoprotein E (ApoE ) derived from the blood through parabiosis restores hypercholesterole mia in A-producing ApoE Knockout Mice Genotype Age Total Cholesterol (mg/dl) n Control Nontransgenic 7 months 105 6 15 PS1-/-APP+/+apoE+/7 months 73 8 7 PS1+/-APP+/+apoE+/7 months 79 6 8 PS1+/-APP+/+apoE-/7 months 392 121 3 ApoE-/7 months 501 39 13 Parabiosed pairs 1a. PS1+/-APP+/+apoE+/5 months 118 1 1b. PS1+/-APP+/+apoE-/5 months 125 1 2a. PS1+/-APP+/+apoE+/7 months 87 1 2b. PS1+/-APP+/+apoE-/7 months 86 1 3a. PS1+/-APP+/+apoE+/7 months 106 1 3b. PS1+/-APP+/+apoE-/7 months 90 1 4a. PS1+/-APP+/+apoE+/5 months 89 1 4b. PS1+/-APP+/+apoE-/5 months 90 1 Table 1 Total plasma cholesterol measurements in randomly selected parabiosed pairs compar ed to control mice of various genotypes. Apolipoprotein E derived from peripheral circulation can restore the hypercholesterolemia typi cal of Apolipoprot ein E knockout (PS1+/-,APP+/+,apoE-/-) mice.

PAGE 67

60 Figure 1 ( A ) Apolipoprotein E can be detected in a parabiosed apoE-knockout mouse (lane 7) but not in a apoE knock out mouse that has not been parabiosed (lane 5). Lane 6 shows parabiosed partner that is heterozygous for the apoE gene. Lanes 1-4 are ten-fold dilutions of plasma from a nontransgenic mouse. ( B ) Parabiosed PS1+/-,APP+/+,apoE-/and PS1+/-, APP+/+,apoE+/mice show an equal extent of murine apoE and neo DNA in blood cells indicating successful and efficient white blood cell transfe r between parabiosed partners.

PAGE 68

61

PAGE 69

62 Figure 2. A-immunoreactive (6E10) staini ng in parabiosed or control PS1+/-,APP+/+apoE-/and PS1+/-,APP+/+,apoE+/mice. The staining in PS1+/-,APP+/+,apoE-/mice, which is essentially only diffuse A deposition, is extensive in the hippocampus ( B and D ).In contrast, PS1+/-,APP+/+,apoE+/mice show an abundance of amyloid plaques ( A and C ). The A immunostaining is essentially unaffected by parabiosis (n=6) ( A and B ) in both the cerebral cortex ( E ) and the hippocampus ( F ), as compared to unoperated mice (n=3) ( C and D ). No Aimmunoreactive staining was detected in nontransgenic mice (data not shown).

PAGE 70

63

PAGE 71

64 Fig. 3 – Apolipoprotein E immunostainin g in parabiosed and control PS1+/,APP+/+,apoE-/and PS1+/-,APP+/+,apoE+/mice. ApoE is detected in amyloid plaques in parabiosed PS1+/-,APP+/+,apoE+/mice ( A and B ) and PS1+/-,APP+/+,apoE+/controls (not shown) as well in astrocytes of nontransgenic control animals ( C and D ). In PS1+/-,APP+/+,apoE-/control mice, no apoE immunoreactivity is detrected ( F ), but parabiosed mice of the same genotype present with a small amount of staining in choroid plex us, while none is present in the brain parenchyma ( E ).

PAGE 72

65

PAGE 73

66 Fig. 4 – Thioflavine S tissue staining for t he detection of filamentous A. ( A ) There is no statistical significant difference in the area of thioflavine S staining, as determined by densitometric analysis, in PS1+/-,APP+/+,apoE+/between parabiosed (n=6) and control (n=3) mice in either the hippocampus or cerebral cortex. ( B ) The number of plaques counted per brain section in parabiosed PS1+/-,APP+/+apoE-/(n=6) mice is significantly smaller than that of nonparabiosed control (n=3) mice of t he same genotype in both hippocampus and cerebral cortex.

PAGE 74

67

PAGE 75

68 REFERENCES Bales KR, Verina T, Cummins DJ, Du Y, Dodel RC, Saura J, Fishman CE, DeLong CA, Piccardo P, Petegnief V, Ghe tti B, Paul SM (1999) Apolipoprotein E is essential for amyloid deposition in the APP(V717F) transgenic mouse model of Alzheimer's disease. Proc Natl Acad Sci U S A 96:15233-15238. Bales KR, Verina T, Dodel RC, Du Y, Alts tiel L, Bender M, Hyslop P, Johnstone EM, Little SP, Cummins DJ, Piccardo P, Ghetti B, Paul SM (1997) Lack of apolipoprotein E dramatically reduce s amyloid beta-peptide deposition. Nat Genet 17:263-264. Blain H, Jeandel C, Merched A, Visvikis S, Siest G (1997) Apolipoprotein E level in cerebrospinal fluid increases wit h aging. J Am Geriatr Soc 45:1536. Christie RH, Bacskai BJ, Zipfel WR, Williams RM, Kajdasz ST, Webb WW, Hyman BT (2001) Growth arrest of indi vidual senile plaques in a model of Alzheimer's disease observed by in vivo multiphoton microscopy. J Neurosci 21:858-864. Corder EH, Saunders AM, Strittmatter WJ, Schmechel DE, Gaskell PC, Small GW, Roses AD, Haines JL, PericakVance MA (1993) Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer's disease in late onset families. Science 261:921-923. Costa DA, Nilsson LN, Bales KR, Paul SM Potter H (2004) Apolipoprotein is required for the formation of filamentous amyloid, but not for amorphous Abeta

PAGE 76

69 deposition, in an AbetaPP/ PS double transgenic mouse model of Alzheimer's disease. J Alzheimers Dis 6:509-514. DeMattos RB, Bales KR, Cummins DJ, P aul SM, Holtzman DM (2002a) Brain to plasma amyloid-beta efflux: a measure of brain amyloid burden in a mouse model of Alzheimer's dis ease. Science 295:2264-2267. DeMattos RB, Bales KR, Cummins DJ, Dodart JC, Paul SM, Holtzman DM (2001) Peripheral anti-A beta antibody al ters CNS and plasma A beta clearance and decreases brain A beta burden in a mouse model of Alzheimer's disease. Proc Natl Acad Sci U S A 98:8850-8855. DeMattos RB, Bales KR, Parsadanian M, O'Dell MA, Foss EM, Paul SM, Holtzman DM (2002b) Plaque-associated di sruption of CSF and plasma amyloidbeta (Abeta) equilibrium in a mouse model of Alzheimer's disease. J Neurochem 81:229-236. DeMattos RB, Cirrito JR, Parsadanian M, May PC, O'Dell MA, Taylor JW, Harmony JA, Aronow BJ, Bales KR, Paul SM, Holtzman DM (2004) ApoE and Clusterin Cooperatively Suppr ess Abeta Levels and Deposition. Evidence that ApoE Regulates Extrace llular Abeta Metabolism In Vivo. Neuron 41:193-202. DeMattos RBP, M.; ODell MA, Taylor JW, Bales KR, Paul SM, and Holtzman DM (2002) Apolipoprotein E3 dose dependent modulation of Amyloid-beta deposition in a transgenic mouse m odel of Alzheimers disease. SocNeurosciAbstr, 32.

PAGE 77

70 Finerty JC (1952) Parabiosis in physi ological studies. Physiol Rev 32:277-302. Ghilardi JR, Catton M, Stimson ER, Roger s S, Walker LC, Maggio JE, Mantyh PW (1996) Intra-arterial infusion of [ 125I]A beta 1-40 labels amyloid deposits in the aged primate brain in vivo. Neuroreport 7:2607-2611. Holtzman DM, Bales KR, Wu S, Bhat P, Parsadanian M, Fagan AM, Chang LK, Sun Y, Paul SM (1999) Expression of human apolipoprotein E reduces amyloidbeta deposition in a mouse model of Alz heimer's disease. J Clin Invest 103:R15R21. Holtzman DM, Fagan AM, Mackey B, Tenkova T, Sartorius L, Paul SM, Bales K, Ashe KH, Irizarry MC, Hyman BT (2000) Ap olipoprotein E facilitates neuritic and cerebrovascular plaque formation in an Al zheimer's disease model. Ann Neurol 47:739-747. Hone E, Martins IJ, Fonte J, Martins RN (2003) Apolipoprotein E influences amyloid-beta clearance from the murine periphery. J Alzheimers Dis 5:1-8. Huang Y, Liu XQ, Wyss-Coray T, Brec ht WJ, Sanan DA, Mahley RW (2001) Apolipoprotein E fragments present in Alzheimer's disease brains induce neurofibrillary tangle-like intracellular inclus ions in neurons. Proc Natl Acad Sci U S A 98:8838-8843. Licastro F, Parnetti L, Mo rini MC, Davis LJ, Cucinotta D, Gaiti A, Senin U (1995) Acute phase reactant alpha 1-ant ichymotrypsin is increased in cerebrospinal fluid

PAGE 78

71 and serum of patients with probable Alzheimer disease. Alzheimer Dis Assoc Disord 9:112-118. Lieberman J, Schleissner L, Tachik i KH, Kling AS (1995) Serum alpha 1antichymotrypsin level as a marker for Alzheimer-type dementia. Neurobiol Aging 16:747-753. Linton MF, Atkinson JB, Fazio S (1995) Prevention of atherosclerosis in apolipoprotein E-deficient mice by bone marrow transplantation. Science 267:1034-1037. Ma J, Yee A, Brewer HB, Jr., Das S, Potter H (1994) Amyloidassociated proteins alpha 1-antichymotrypsin and apolipoprotei n E promote assembly of Alzheimer beta-protein into filam ents. Nature 372:92-94. Mahley RW, Weisgraber KH, Hussain MM, Greenman B, Fisher M, Vogel T, Gorecki M (1989) Intravenous infusion of apolipoprotein E accelerates clearance of plasma lipoproteins in r abbits. J Clin Invest 83:2125-2130. Martel CL, Mackic JB, Mats ubara E, Governale S, Miguel C, Miao W, McComb JG, Frangione B, Ghiso J, Zlokovic BV ( 1997) Isoform-specific effects of apolipoproteins E2, E3, and on cerebral capillar y sequestration and bloodbrain barrier transport of circulating Alzheimer's amyloid beta. J Neurochem 69:1995-2004.

PAGE 79

72 Martinez C, Smith JM, Shapiro F, Good RA (1959) Transfer of acquired immunological tolerance of skin homografts in mice joined in parabiosis. Proc Soc Exp Biol Med 102:413-417. Matsuoka Y, Saito M, LaFrancois J, Gaynor K, Olm V, Wang L, Casey E, Lu Y, Shiratori C, Lemere C, Duff K (2003) Novel t herapeutic approach for the treatment of Alzheimer's disease by per ipheral administration of agents with an affinity to beta-amylo id. J Neurosci 23:29-33. Nilsson LN, Bales KR, DiCarlo G, Gor don MN, Morgan D, Paul SM, Potter H (2001) Alpha-1-antichymotr ypsin promotes beta-sheet amyloid plaque deposition in a transgenic mouse model of Alzheime r's disease. J Neurosci 21:1444-1451. Nilsson LN, Arendash GW, Leighty RE, Costa DA, Low MA, Garcia MF, Cracciolo JR, Rojiani A, Wu X, Bales KR, Paul SM, Potter H (2004) Cognitive impairment in PDAPP mice depends on ApoE and ACT-catalyzed amyloid formation. Neurobiol Aging 25:1153-1167. Pepys MB, Herbert J, Hutchinson WL, Tennent GA, Lachmann HJ, Gallimore JR, Lovat LB, Bartfai T, Alanine A, Hertel C, Hoffmann T, Jakob-Roetne R, Norcross RD, Kemp JA, Yamamura K, Suzuki M, Taylor GW, Murray S, Thompson D, Purvis A, Kolstoe S, Wood SP, Hawk ins PN (2002) Targeted pharmacological depletion of serum amyloid P component for treatment of human amyloidosis. Nature 417:254-259.

PAGE 80

73 Plump AS, Smith JD, Hayek T, Aalto-Setala K, Walsh A, Verstuyft JG, Rubin EM, Breslow JL (1992) Severe hyperchol esterolemia and atherosclerosis in apolipoprotein E-deficient mice creat ed by homologous recombination in ES cells. Cell 71:343-353. Raffai RL, Weisgraber KH (2002) Hypomorphic apolipoprotein E mice: a new model of conditional gene repair to examine apolipoprotein E-mediated metabolism. J Biol Chem 277:11064-11068. Rebeck GW, Reiter JS, Strickland DK, Hy man BT (1993) Apolipoprotein E in sporadic Alzheimer's disease: allelic va riation and receptor interactions. Neuron 11:575-580. Scheuner D, Eckman C, Jensen M, Song X, Citron M, Suzuki N, Bird TD, Hardy J, Hutton M, Kukull W, Larson E, Le vy-Lahad E, Viitanen M, Peskind E, Poorkaj P, Schellenberg G, Tanzi R, Wasco W, Lannfelt L, Selkoe D, Younkin S (1996) Secreted amyloid beta-protein similar to that in the senile plaques of Alzheimer's disease is increased in vivo by the presenilin 1 and 2 and APP mutations linked to familial Alzheimer's disease. Nat Med 2:864-870. Shibata M, Yamada S, Kumar SR, Calero M, Bading J, Frangione B, Holtzman DM, Miller CA, Strickland DK, Ghiso J, Zlokovic BV (2000) Clearance of Alzheimer's amyloid-ss(1-40) peptide from brain by LDL receptor-related protein1 at the blood-brain barrier. J Clin Invest 106:1489-1499.

PAGE 81

74 Silverberg GD, Mayo M, Saul T, Rubenstein E, Mc Guire D (2003) Alzheimer's disease, normal-pressure hydroc ephalus, and senescent changes in CSF circulatory physiology: a hypothes is. Lancet Neurol 2:506-511. Thorngate FE, Rudel LL, Walzem RL, Williams DL (2000) Low levels of extrahepatic nonmacrophage ApoE inhibit atherosclerosis without correcting hypercholesterolemia in ApoE-deficient mice. Arterioscler Thromb Vasc Biol 20:1939-1945. Vogel T, Weisgraber KH, Zeevi MI, B en-Artzi H, Levanon AZ, Rall SC, Jr., Innerarity TL, Hui DY, Tayl or JM, Kanner D, et al. (1985) Human apolipoprotein E expression in Escherichia coli: structural and functional identit y of the bacterially produced protein with plasma apolipoprotein E. Proc Natl Acad Sci U S A 82:8696-8700. Zhang SH, Reddick RL, Piedrahita JA, Maeda N (1992) Spontaneous hypercholesterolemia and arterial lesions in mice lacking apolipoprotein E. Science 258:468-471. Zlokovic BV (2004) Clearing amyloid through the blood-brain barrier. J Neurochem 89:807-811.

PAGE 82

75 Paper III: Environmental Enrichment Prot ects Alzheimer’s Disease Mice from Cognitive Impairment through Reductions in A Deposition and Beneficial Changes in Gene Expression and can be Mimicked by Inhibition of Phosphodiesterase 4 (PDE4) David A. Costa1,2, Jennifer R. Cracchiolo1,3, Adam D. Bachstetter4, Tiffany F. Hughes2, Kelly R. Bales6, Steven M. Paul6, Ronald F. Mervis4,5, Gary W. Arendash1,3, and Huntington Potter1,2,* 1Johnnie B. Byrd Sr. Alzheimer’s Center and Research Inst itute, 15310 Amberly Dr.,Tampa, FL 33647, USA. 2Department of Biochemistry and Molecu lar Biology and Suncoast Gerontology Center, University of South Florida, Tampa, FL 33612, USA 3Memory and Aging Research Laboratory, SCA 110, University of South Florida, Tampa, FL 33620, USA 4NeuroStructural Research Laboratories, University Center Dr. Suite 180, Tampa, FL 33612, USA 5Center for Aging and Brain Repair, University of S outh Florida, Tampa, FL 33612, USA 6Neuroscience Discovery Research, Lilly Research Laboratorie s, Indianapolis, IN 46285, USA *Correspondence: hpotter@hsc.usf.edu

PAGE 83

76 ABSTRACT Although lifelong cognitive stimulation is associated wit h lower risk of Alzheimer’s disease (AD), prospective long-term human studies needed to prove that environmental enrichment (EE) protects against AD are impractical. We therefore sought to investigate the pr eventative potential of EE using mice expressing both human mutant presenili n-1 and amyloid precursor protein (PS1/PDAPP). At weaning, mice were pl aced into either an enriched or standard housing (SH) environment. Behavioral te sting at 4-6 mont hs showed that EEPS1/PDAPP mice outperformed mice in SH, and were behaviorally indistinguishable from NT mi ce. PS1/PDAPP mice given both EE and behavioral testing showed greater than 50% less brai n -amyloid (A), but did not exhibit changes in dendritic morphol ogy. Microarray analysis using hippocampal tissue revealed large EE-induced changes in t he expression of genes/proteins related to memory, neuroprotection, and A seques tration. Inhibition of one such protein, PDE4, a cAMP-phosphodiesterase, by Rolipram treatment for two weeks, mimicked the cognitive benefits of EE.

PAGE 84

77 INTRODUCTION Alzheimer’s disease (AD) is a common neurodegenerative disorder characterized by parenchymal -amyloi d deposition, neurof ibrillary tangle formation, neuronal loss, and cognitive decli ne. A lifelong pattern of high mental activity (Wilson et al., 2002) and educatio nal attainment (Stern et al., 1994) correlates with lower risk of AD and may be protective. Furthermore, high levels of linguistic ability early in life are a ssociated with a reduced risk of the disease (Snowdon et al., 1996; Riley et al., 2005). These studies suggest that extra “cognitive reserve” developed throughout life may help buffer against the consequences of later dementia. Howeve r, despite some encouraging results (Loewenstein et al., 2004) the extent to which cognitive stimulation (i.e. environmental enrichment, EE) protects agains t AD remains difficult to assess in humans because: 1) retrospective st udies cannot unequivocally isolate environmental enrichment from other factors affecting cognition over a lifetime, and 2) long-term intervention in hum ans is impractical. Furthermore, epidemiological human studies give no insights about the potential mechanisms by which EE may protect against AD. The key question about whether EE intervention can protect against AD pathology, or its associated mental decli ne, was addressed in the present study using a transgenic mouse model of the dis ease. The mice chosen express both the human mutant amyloid precursor protein (hAPPV717F) and mutant Presenilin 1 (hPS1M146L) genes, which result in moderate brain -amyloid plaque deposition and significant behavioral impairment by 56 months of age. At weaning, mice

PAGE 85

78 were placed into either standard housin g or an enriched environment, and were behaviorally tested between 4-6 months of age. It is known from previous studies that non-transgenic r odents subjected to cogniti ve stimulation (i.e. EE) perform better in water mazes, ex hibit increased dendritic branching and dendritic spine formation (Globus et al., 1973; Comery et al., 1995; Turner et al., 2003), increased synaptogenesis (Ramirez-A maya et al., 1999), and increased neuronal plasticity-related gene expression (P inaud et al., 2001), while exhibiting decreased levels of apoptotic cell death (Young et al., 1999). Our previous experiments determi ned that an EE paradigm used therapeutically in aged AD transgenic mice with severe A plaque deposition provided cognitive benefit s without affecting th e amyloid plaque burden (Arendash et al., 2004a). Here we sought to determine the extent to which preemptive EE protects memory, impacts A deposition, and affects dendritic morphology. To characterize these chang es at the molecular level, hippocampal gene expression was analyzed using a whole mouse genome microarray. We found that PS1/PDAPP mice rais ed in EE outperform ed those raised in SH across a variety of behavioral/ memory tasks, in which they were statistically indistinguishabl e from non-transgenic (NT) mi ce. Interestingly, brain A deposition in Tg+ mice was not affect ed by EE alone, but in combination with behavioral testing (an additional enriching experience), EE resulted in large reductions in brain A deposition. Alt hough the extent of dendr itic branching and dendritic spine numbers in cortex/hippoc ampus were unchanged by enrichment, Tg+ mice given EE did exhibit increased expression of multiple memory-related

PAGE 86

79 genes, a number of neuroprotec tive genes involved in anti-apoptotic BAD phosphorylation, as well as increased expression of transthyretin (a known A sequestering molecule). RESULTS Behavioral Testing Cognitive performance fo r both standard-housed (SH) and environmentally-enriched (EE) mice was determined between 4-6 months of age through five cognitive-based ta sks (Radial Arm Wate r Maze, Platform Recognition, Morris maze, Y-maze, and Circular Platform) – each measuring discreet cognitive domains. Although we found significant effects of EE on multiple tasks and measures t herein, two of the tasks were particularly important. Mice were tested in the radial arm water maze (RAWM), a task designed to assay working memory. Over days 7-9 of testing, transgenic mice raised in environmental enrichment (Tg+/EE) had signi ficantly lower escape latencies than their SH counterparts for the last acqui sition trial (T4; p<0.001) and also the memory retention trial (T5; P<0.00001) (F igure 1A). Intere stingly, EE also improved the overall RAWM performance of non-transgenic (NT) mice, as evidenced by their significantly lower esca pe latencies on T4 compared to their SH counterparts (p<0.02). The only ot her task in which EE improved upon the already fine performance of NT/SH mice wa s the circular platform task of spatial learning, wherein NT/EE mice had signific antly shorter escape latencies overall compared to NT/SH mice (data not shown) Since mice raised in an enriched environment had larger cages and access to mazes and running wheels, their

PAGE 87

80 potential athletic ability could have been s uperior to mice raised in standard housing. This was not the case, howev er, since analysis of the number of seconds taken per arm choice revealed no statistical differences in swim speed between SH and EE groups for either NT or Tg+ mice Furthermore, Tg+/EE mice made significantly fewer RAWM e rrors than Tg+/SH mice, and again were not statistically different from NT mice (SH or EE) (Data not shown). In the platform recognition task, which requires mice to use a search/identification strategy rather than the spatial strategy of Morris maze and RAWM, similar behavioral benefit was observed for Tg+ mice. Tg+/EE mice had significantly lower escape latencies than Tg+/SH mice over all days of testing (p<0.005), as well as on all but the initial day of te sting (P<0.01) (Figure 1B). Furthermore, Tg+/EE mice were not statistically signifi cantly different from both NT groups in this task. To determine the effects of EE on overall cognitive performance, we performed discriminant function analysis (DFA) across 8 cognitive measures taken from our test battery that repres ent multiple cognitive domains (working memory, reference learning/memory, ident ification/recognition). DFA determines whether groups can be distinguished fr om one another based on their overall behavioral performance across multiple cognit ive measures. As shown in Figure 1C, DFA was easily able to distingu ish the impaired overall cognitive performance of Tg+/SH mice from the mu ch better performance of both Tg+/EE mice and NT/SH controls (p<0.001 for both comparisons). The latter two groups

PAGE 88

81 could not be distinguished from one another by DFA. Thus the effect of EE in Tg+ mice was global in spannin g multiple cognitive domains. Alzheimer’s Disease Pathology To study the effect of environmental enrichment on AD pathology, overall A im munoreactivity and mature -amyloid (A) deposition were measured. In order to eliminate behavioral testing as an interfering form of cognitive stimulation t hat could alter brain pathology, we raised two cohorts of mice in both SH and EE conditions through 6 months of age. A immunohistochemistry was measured in both cohorts, but only one cohort underwent behavioral testing between 4-6 months of age. Total A immunoreactivity, using the 6E10 monoclona l antibody, revealed that there was no difference in total A load between Tg+/EE and Tg+/SH mice in either cerebral cortex or hip pocampus for the “non-behavio rally tested” group (Figure 2A,B). A 1-42 enzyme linked immunosor bent assay (ELISA) further verified that the A levels (either soluble or inso luble) were not statis tically significantly different between non-behaviorally tested SH vs. EE transgenic groups (Data not shown). Since some pathological chaperon es, such as apolipoprotein E, exert their contributory effect on AD pathology by promoting compact plaque formation rather than overall A levels (Costa et al., 2004), it is not unreasonable to expect EE to affect a change in only mature (com pact) A deposition. We find however, that mature A plaque loads (as measur ed by thioflavine S staining) were also not significantly different between non-behaviorally te sted EE and SH transgenic groups in either cerebral cortex or hippocampus (Figure 2C,D).

PAGE 89

82 Importantly, we did observe that behaviorally-tested EE mice had significantly lower A deposition than those mice that underwent only EE. This was true for both total and compact A deposition and for bo th cortex and hippocampus (Figure 2A-D). For behavio rally-tested Tg+/EE mice, their 57.6% and 69.0% decreases in compact A deposition in cortex and hippocampus, respectively, compared to non-behaviorall y tested Tg+EE mice are particularly noteworthy. Our A deposition results from both cohorts thus indicate that EE or behavioral testing alone is not sufficient to alter A pathology, but when they are combined there is a large statistically sign ificant effect provided by both of these forms of cognitive stimulation in concert with one another. Although A deposition was reduced by a co mbination of EE and cognitive testing, there were no statistically significant correlations in the Tg+/EE mice between such AD pathology and any behavioral m easure (data not shown), i ndicating that the mice had experienced EE-induced cognitive improv ement that was at least partially independent of any reduc tion in A pathology. The extent and distributi on of dendritic arborizatio n in Layer V cortical neurons was measured in behaviorally test ed mice through Golgi staining. Using the Sholl method of quantification, there was no obser ved difference in dendritic branching between transgenic EE and SH groups (Figure 2E ). However, both Tg+/SH and Tg+/EE groups had significantly redu ced dendritic arborization compared to non-transgenic single-housed (NT/SH) mice (p<0.001), indicating that EE could not increase de ndritic arborization in Tg+ mice to the level of NT controls. Additional evidence that EE did not affect dendritic morphology is

PAGE 90

83 provided by analysis of dendritic spine co unts in both cortex and hippocampus of Tg+ mice (Figure 2F). No EE-induced incr ease in dendritic spin es was evident in Tg+ mice for either cortical Layer V pyramidal neurons or for hippocampal CA1 neurons. Thus, Tg+/EE mice performed i dentically to NT/SH controls despite having reduced dendritic branching and synaptic spine counts that were identical to Tg+/SH mice. Evidentially, EE must induce changes in the brains of PS1/PDAPP Tg+ mice that allow them to compensate for any cognitive damage that may be imparted by the dendritic branching defic its and the A pathology typical of these transgenic mouse lines. Microarray To begin to investigate the mechanism(s) by which EE protects against AD-induced cognitive decline, we used microarray analysis to examine gene expression chang es in the hippocampus t hat might contribute to the aforementioned memory enhancement. Non-behaviorally tested EE (n=4) and SH (n=4) transgenic mice were anal yzed using an Affymetrix GeneChip designed for the mouse genome. In all, over 120 genes showed statistically significant expression changes in res ponse to EE. Of these genes, roughly 70 were changed by a factor of 2.0 or mo re. Many of the more robust changes occurred in genes already implicated in one or more aspects of memory or AD (TABLE 1). For example, insulin-like growth factor ( IGF-2 ), which has also been shown to play a neuroprotective role against A (Dore et al., 1997) was upregulated in EE mice by 2.5 fold. This gene, along with Insulinlike growth factor binding protein-2 ( IGFBP-2 ) (up-regulated 3.0 fold in EE), and the prolactin receptor ( PRLR ) (up-regulated 21.1 fold in EE) have all been implicated as

PAGE 91

84 neuroprotective molecules acting in c oncert upstream of the Akt and Erk1/2 pathways that ultimately le ad to BAD phosphorylation, an event shown to be both neuroprotective and anti-apoptotic (Bonni et al., 1999; Thompson and Thompson, 2004). A 10-fold in crease in transthyretin ( TTR ), an A binding protein shown to sequester A and inhibi t amyloid formation, was also observed in the EE mice. These microarray results were verified by quantitative real-time PCR (Table 2). Two genes involved in memory improvement, phospholipase A2 and cytochrome C oxidase were up-regulated in EE mice by 6.7 and 3.1 fold respectively. Furthermore, phosphodieste rase 4B was down-regulated 2.2 fold and the cholecystokinin B receptor was dow n-regulated 3.2 fold. Both of these genes, when experimentally inhi bited, are known to improve memory (Lemaire et al., 1994; Zhang et al., 2000; Bourtchouladze et al., 2003; Zhang et al., 2004). Calcineurin knockout mice exhibit increased tau phos phorylation, abnormal neuronal cytoskeleton, and cognitive deficit s, making the increased expression of this gene by EE also interesting (Ka yyali et al., 1997; Zeng et al., 2001). An examination of the functions of the proteins whose expression is changed by EE reveals that many partici pate in a signaling pathway that can confer neuroprotection in several exper imental systems (See Figure 3). The benefit of deriving such a neuroprotective pathway from the microarray results is that potential targets of ther apeutic intervention can be ident ified. For example, if the down-regulated genes id entified by microarray anal ysis are responsible for the improvement in cognitive function, we would predict that pharmacological

PAGE 92

85 inhibition of the products of one or mo re of these genes should have similar beneficial effects. To test this hypothesis, we attempt ed to pharmacologically intervene in the pathway shown in Figure 3 in hopes of emulating the beneficial memory effects of EE. In our init ial experiments, we chose to inhibit PDE4 since it is down-regulated by EE, and a there exists a highly specific PDE4 inhibitor, Rolipram, which has already been found to alleviate memory deficits in rodent models. To test this possibility, we initia ted experiments in which Rolipram, a specific inhibitor of cAMP phosphodiestera se 4 (PDE4) (Figure 3, Purple), has been injected over a two-week period in to. Cognitively deficient PS1/PDAPP mice were injected with Rolipram over a two-week period, and then subjected to 6 days of RAWM testing. The inject ed mice had significantly lower escape latencies than those mice receiving vehicl e alone for the last acquisition trial (T4; p<0.05) and also the memory retention trial (T5; P<0.05), indicating that Rolipram treatment effectively mimics the cogniti ve benefits of envir onmental enrichment (Figure 4). Indeed, a compar ison of the results shown in Figures 1 and 4 reveals that only two weeks of Roli pram treatment is able to restore normal function to severely impaired AD mice.

PAGE 93

86 DISCUSSION Our previous studies, and studies from others, have shown that mice that overexpress a mutant human APP gene linked to AD are si gnificantly impaired in a number behavioral tasks (Arendash et al., 2001b; Arendash et al., 2004b; Nilsson et al., 2004; Jensen et al., 2005; O gnibene et al., 2005). Here we show that doubly transgenic PS1/ PDAPP Alzheimer’s mice raised in a mentallystimulating environment were “protect ed” from otherwise certain cognitive impairment, as evidenced by their superior performance in a variety of behavioral tasks compared to transgenic mice rais ed in standard housing conditions. In fact, long-term EE produced transgenic mice whose “overall” performance across multiple cognitive measur es was indistinguishable from non-transgenic mice. Thus, the cognitive benefits of EE were global in nature, af fecting multiple cognitive domains. Only by adding s ubsequent behavioral-testing to EE were profound reductions in brain A depos ition achieved, suggesting an A limiting/sequestering action of EE. Mo reover, EE produced benef icial changes in expression of multiple genes involved with A sequestration, memory, and neuroprotection. Our results not only prov ide the first unequivocal evidence that EE can protect against Alzheimer’s-like co gnitive impairment, but also provide insight into the multi-faceted mechani sms involved in that protection. Our previous work has shown that t he levels of amyloid deposition in AD (i.e. APP) transgenic mice correlate with extent of cognitive impairment (Arendash et al., 2004b; Leighty et al., 2004 ; Nilsson et al., 2004). For example, when different A promoting genes, namely Apolipoprotein E (apoE) and -1

PAGE 94

87 antichymotrypsin (ACT) are introduced into a mouse with an APP, apoE-KO, ACT-/background, the severity of amyl oid deposition depends on the presence of these promoters, and, mo re importantly, directly correlates with the levels of cognitive decline (Nilsson et al., 2004). In our current study, we determined that neither EE nor the 6-week behavioral test period alone was sufficient to affect brain A deposition. However, comb ining both these forms of cognitive stimulation (i.e. long-term EE and intense behavioral testing during the final 6 weeks) resulted in dramatic (up to 69% ) reductions in both total and mature (compact) A deposition. Si nce it is unlikely that A deposition during the final 6 weeks of EE would have increased by 23 fold in non-tested mice, it is reasonable to conclude that the combinat ion EE experience actu ally resulted in removal/sequestration of brain A from both diffuse and compact deposits. We hypothesize that long-term EE primed mo lecular/genetic pathways in the brain (see below) for complementary and/or syner gistic actions provided by behavioral testing, resulting in profound reductions in brain A deposition. These results are in direct conflict with anot her recent study (Jankowsky et al., 2003) that focused on A deposition changes in another transgenic mouse model of AD (APPsw/PS1dE9), and reported that EE promotes A burden and deposition. Since Jankowsky et al. (2003) did not behavio rally test their animals and used an atypical EE methodology, a direct com parison between the two studies is not practical. In contrast to our EE me thodology of maintaining the same stable group of animals in any gi ven EE cage, Jankowsky et al. (2003) continually introduced young animals and removed older ani mals from their EE cages. This

PAGE 95

88 likely added a continual source of stre ss, which could have contributed to their observed increase in A burden. EE was not able to protect against the dendritic branching defects observed in PS1/PDAPP mice despite cogni tive benefits observed in the same animals. Analysis of Golgi-stained Layer V pyramidal cells in cortex revealed Tg+/SH and Tg+/EE were nearly identical in their extent of dendritic branching, with both groups having sign ificantly less dendritic branching than NT/SH controls. Dendritic spine counts in both cortex and hippocampus further underscored the lack of EE effects on dendrit ic morphology in Tg+ mice. By contrast, prior studies involving normal rodents have provided evidence for EEinduced changes in dendritic /synaptic morphology. In those studies, EE induced: 1) greater synaptophysin levels (an i ndex of synaptic surface area, but not number of synapses) in hippocampus and neocortex (Frick and Fernandez, 2003), and 2) an increase in Golgi-stai ned dendritic branches and dendritic spines in neocortex (Comery et al., 1995; Tu rner et al., 2003). The inability of EE in the present study to increase dendritic branching or dendritic spines in Alzheimer’s Tg+ mice suggests that EE’s pr otective effect in those same mice does not primarily involve changes in dendritic/synaptic morphology. Nonetheless, subtle morphologic chang es may have occurred or neuronal populations, other than those presently evaluated, may have been effected by EE. To study possible biochemical/genetic changes that might be independent of, or additive to, A reducing/sequesteri ng effects of EE in transgenic mice, we

PAGE 96

89 performed microarray analysis using an Affymetrix 430 2.0 mouse GeneChip which probes over 39,000 kn own unique transcripts of the mouse genome. Hippocampal RNA from eight non-behaviora lly tested transgenic mice (EE n=4, SH n=4) was pooled in pairs, and used to probe four microarrays. The resultant expression should reflect the effects of environmental enrichment and identify genes that are either upor down-r egulated by EE and that may underlie the cognitive protection evi dent in behaviorally-tested transgenic mice. Of the approximately 70 known genes that were eit her up or down regula ted by at least 2-fold due to EE, we focused our attent ion on those that were most likely to impact AD and/or cognitive function. A number of genes known to enhance or inhibit memory were upregulated or down-regulated, respectively In particular, phospholipase A2 (PLA2), a protein involved in the infla mmatory response, membrane remodeling, and phospholipid metabolism (Sun et al ., 2004) and a proposed target for the treatment of inflammatory brain relat ed disorders (Strokin et al., 2004) was upregulated approximately 6.7-fo ld in AD mice raised in EE conditions. Expression of a mutated PLA2 homolog in drosophila introduces memory defects (Chiang et al., 2004), as does treatment of mice with bromoenol lacton e, a potent PLA2 inhibitor (Fujita et al., 2000). Furthermo re, reduced PLA2 expression has been reported in human AD brains (Gattaz et al., 2004). We also observed a 3.1 fold increase in cytochrome C oxidase (CO) message. Often used as a metabolic marker for neuronal activity, CO, an impor tant enzyme involved in mitochondrial electron transport, is reportedly lowered in AD patients (Mutisya et al., 1994). In

PAGE 97

90 addition, mice trained in spatial wo rking memory tasks show increased CO activity in the mammillary bodies, a regi on of the brain closely related to the hippocampus and memory formation. It is also interesting that transthyretin was up-regulated between 6and 10-fold in tr ansgenic mice raised in EE. TTR is a plasma and CSF protein that is known to bind to (Tsuzuki et al., 1997; Tsuzuki et al., 2000) and sequester (Schwarzman et al., 1994) A, thus inhibiting plaque formation. It is hypothesized that high le vels of kidney TTR prevent A plaque formation despite the fact that kidneys hav e the highest levels of A 1-40 and 142 in the body, after the brain (Tanzi et al., 1987; Tsuzuki et al., 2000). In vitro work has shown that TTR co-loc alizes with amyloid plaques, and in vivo inhibition of TTR increases A deposition (Stein et al., 2004). Previous microarray experiments have shown an increase in TTR in an APPsw mouse model of AD as compared to NT mice, and this is hypothesized to be partly responsible for the lack of neurodegeneration typically characteri stic of AD mouse models (Stein and Johnson, 2002). Similarly, we show an increase in TTR expression due to EE. Perhaps the reduced levels of TTR in human AD patients (Serot et al., 1997) contribute to their neurodegeneration an d may therefore be alleviated by cognitive stimulation. Despite TTR’s known A sequestering activity, the 6-10x expression increase in non-behaviora lly tested transgenic s was unable to limit/decrease their brai n amyloid deposition. Converse to EE-induced gene up-regul ation, two genes associated with memory repression phosphodiesterase 4B and cholecystokinin B were downregulated by 2.2and 3.2-fold, respective ly, in transgenic mice raised in EE.

PAGE 98

91 Inhibiting PDE4 in a m ouse model of Rubinstein-Taybi syndrome, a disease characterized by mental retardation, ameliorates the long-term memory deficits typical of this disorder. Similarly, a recent study has shown that specifically inhibiting PDE4 in an APPsw/PS mouse model of AD ameliorates both deficits in long-term potentiation and memory (G ong et al., 2004). Likewise, CCK-B receptor antagonists are shown to enhance memory in rats (Lemaire et al., 1994). In sum, the data show that environmental enrich ment increases relative levels of memory enhancing transcripts while lowering the le vels of multiple known inhibitors of memory. Together these genetic changes may help explain the cognitive enhancement induced by environmental enrichment. The change in PDE4 suggests that the ERK/MAPK signaling cascade, known to be essential for memory consolidation (Impey et al., 1999; Schafe et al., 2000) may contribute to th e observed EE effect. MEK inhibitioninduced memory impairment in a mouse model can be reversed by treatment with a PDE4 inhibitor (Zhang et al., 2004), indicating the potential involvement of PDE4 in MEK/ERK signal ing. In addition, inhibition of PDE4 has been shown to increase cAMP concentrations (Barad et al., 1998), and the cAMP/PKA pathway is also known to modulate memory fo rmation (Frey et al., 1993). Environmental enrichment significantly lowers PDE4 le vels, and therefore likely promotes both the ERK/MAPK and cAMP/PKA pathw ays. We noticed a pattern emerge, in that many of the genes changed on the micr oarray (i.e. PDE4, PRLR, IGF-2, and IGFBP2) fit into a few related metabo lic pathways with a common end-point (Figure 3). Interestingly, these ki nase cascade pathways culminate in the

PAGE 99

92 phosphorylation of BAD, an event known to suppress apoptosis and promote cell survival. The neuroprotective nature of BAD is evidenced by mouse experiments in which high dietary levels of docosahe xanoic acid (DHA; an essential omega-3fatty acid whose consumption is associ ated with a reduced risk of AD) increased PI3-K, BAD phosphorylation, and provided some protection against behavioral deficits found in mice fed a DHA-deficient diet (Calon et al., 2004). Activation of PI3-K is an IRS-1 mediated event. We have found that the PRLR, IGF-2, and IGFBP-2 are all up-regulated in the EE mice. IGF-2 and IGFBP-2 bind to the IGF-1 receptor and, together with the action of the PRLR, are able to activate IRS-1, leading to AKT-mediated BAD phosphorylation. Our findings that PS1/PDAPP mice raised in EE have alte red expression of PDE4, PRLR, IGF-2, and IGFBP-2 suggests that these proteins and the pathway linking them together (See Figure 3) may constitute a neuroprotec tive mechanism that is inducible by environmental enrichment. Finally, we determined that the comb ination of two forms of mental stimulation (long-term EE and intense behav ioral testing) resulted in dramatic reductions in both total and mature (compact) A deposition occurring over the 6 week period of behavioral te sting. Since neither EE nor behavioral testing alone was sufficient to decrease A depositi on, the EE-induced changes in gene expression depicted in Figure 3 may have primed the brain for “behavioral testing-induced” synergist ic and/or complementary effects that resulted in reduced brain A and increases cognitive function. We propose that both Alowering and favorable gene expression path ways are involved in the cognitive

PAGE 100

93 protection afforded by EE. Thus, therapeutics designed to impact these pathways may provide cognit ive protection against, or viable treatment for, Alzheimer’s disease. This prediction has been strikingly supported by the finding that pharmacological inhibition of PDE4 by Rolipram can mimic the cognitive benefits of EE. EXPERIMENTAL PROCEDURES Construction of transgenic mice All procedures involving experimentation on animal subjects were done in accord with the guidelines set forth by the University of South Flor ida’s Institutional Animal Care and Use Committee. Heterozygous PDGF-hAPP(V717F) mice [Swiss-Webster x C57BL/6] were crossed with PDGF-hPS1(M146L) heterozy gotes [Swiss-Webster x C57BL/6] to generate mice with an APP+/-,PS1+/genotype. All offspr ing were screened by PCR to identity the PDGF-hAPP (Games et al., 1995) and the PDGF-hPS1 gene (Duff et al., 1996). Environmental Enrichment. At weaning mice were place into two groups that were exposed to standard single housing (PS1/PDAPP n=32; NT=23) or environmentally enriched housing (PS1/P DAPP n=27; NT n=19) for 4-5 months. All SH animals were housed in shoe box cages with static microisolator tops under climate-controlled conditions on a 12 hour li ght/12 hour dark cycle, fed Harlan Teklad Global Diet #2018 and provided with tap water ad libitum. Although some enrichment studies in normal rodent s have involved socially-housed animals as the standard housing co ntrol, we have found that both singleand socially-

PAGE 101

94 housed double transgenic mice are equally impaired acro ss all of our cognitivebased tasks vs. NT mice, and have elected to utilize standard single housing as our housing control for this study. Enrich ed mice had the same diet and light cycle as SH mice, but were socially housed by sex (6-8 to a cage) in cages containing toys, tunnels, running wheels, etc. (Arendash et al., 2004a) In addition to their enriched housing, EE mice were placed in novel complex environments 3 times weekly for several hours ov er the enrichment period. Behavioral Testing. Beginning at 4.5 to 6 months of age, while continuing in either EE or SH housing, approximately half of the mice (n=57) were tested in five cognitive-based tasks as previously describ ed in detail (Arendash et al., 2001a): Ymaze, Morris water maze, circular platfo rm, platform recognition, and radial arm water maze (RAWM) in that order. All testing was conducted during the light phase. Because this repor t only presents results from platform recognition and RAWM in detail, the methodology of th ose tasks is described. The platform recognition task measures the ability to search for and identify/recognize a variably-placed elevated platform. It requi res animals to ignore the spatial cues present around a 100 cm circular pool, wh ich was the same pool used in earlier Morris water maze testing. Mice were gi ven four successive trials/day over a 4day period. Latencies to find an elevated pl atform (9 cm dia.), bearing a prominent cone-shaped Styrofoam ensign on a wire pole, were determined. For each trial (60 sec. max.), animals were placed into the pool at the same location and the platform was moved to a different one of four possible locations.

PAGE 102

95 For the RAWM task of working memory an aluminum insert was placed into the above pool to create 6 radially-distribut ed swim arms emanating from a central circular swim area. The latency to loca te which one of the 6 swim arms contained a submerged escape platform (9 cm dia.) was determined for 5 trials/day over 9 days of testing. There was a 30 min. time delay between the 4th trial (T4; final acquisition trial) and 5th trial (T5; memory retention tr ial). The platform location was changed daily to a different arm, with different start arms for each of the 5 trials semi-randomly selected from the remaining 5 swim arms. During each trial (60 sec. max.), the mouse was returned to that trial’s start arm upon swimming into an incorrect arm. If the mouse did not find the platform within a 60s trial, it was guided to the platform for the 30s stay. Escape latencies during Trials 4 and 5 are both considered indices of working memory. For both of the above tasks, statistical analysis involved ei ther one-way or two-way repeated measure ANOVAs, followed by post-hoc planned comparisons between groups using the Fisher LSD test. Discriminant Function Analysis (DFA) was performed across all 8 cognitive measur es, from multiple tasks, that loaded together as the primary cogni tive factor in Factor A nalysis. DFA was performed using the DISCRIM subroutine of the Systat software package. Following completion of the behavioral testing at 6 to 7.5 months of age, mice were then anesthetized with Nembutal (0.1mg/g). T he animals were intracardially perfused with 0.9% NaCl (25ml), and their brains were removed. Immunohistochemical Procedures Brains were immersion fixed for 24 hours in 4% paraformaldehyde in 1x So renson’s phosphate buffer, followed by

PAGE 103

96 cryoprotection via three sequential overni ght sucrose immersions, ending in 30% sucrose. 25 m sections were cut using a slid ing microtome (Spencer Lens Co.) and a freezing stage (Physitemp). The sections were mounted and processed through antigen retrieval in prewarmed 25mM citrate buffer (pH 7.3) at +82 C for 5 min. and further processed as previ ously described (Nilsson et al., 2004). The sections were incubated with primary anti bodies against A (6E10, dil. 1:5000) overnight at +4 C. Secondary antibody was anti-mouse IgG developed with NovaRED substrate kit (Vector). Thioflav ine S staining was accomplished using a 5 min. incubation in 1% Thioflavine S fo llowed by 5 min. diffe rentiation in 70% ethanol. Image Analysis Data were collected from three equally spaced coronal tissue sections for both dorsal hippocam pus and overlying parietal cortex (Bregma -1.30 to -2.30 mm) for each mouse. The sections were examined with a Nikon Eclipse E1000 microscope using eit her 4X or 10X Plan Fluor objective lenses. A Retiga 1300 CCD (QImaging) wi th a QImaging RGB LCD-slider was used to capture images. For thioflavi ne S, a Nikon BV-2B fluorescence filter cube was utilized. Customized software wri tten in Visual Basic 6.0 (Microsoft) utilizing Auto-Pro function calls (Image Pr o Plus, Media Cybernetics) was used to segment and quantify images. A deposition was calcul ated as percent area of interest (=Area Stainedtot/Area Measuredtot). Results were analyzed using a twotailed, unpaired student’s t te st with Welch’s correction. Golgi Staining Coronal brain slices 2-3 mm thick were stained en bloc using the Rapid Golgi modification of (Valverde, 1993). Tissue blocks were

PAGE 104

97 initially immersed in a mixture of osmium tetroxide and potassi um dichromate for 5-6 days, then rinsed and blotted, and t hen subsequently immersed in a silver nitrate solution for 36-48 hours. T he blocks were then dehydrated and embedded in nitrocellulose. The stained blocks were then cut at 120 m on a sliding microtome, cleared in alpha-terpineol, rinsed in xylene, and coverslipped under Permount. All slides were coded fo r subsequent quantitative analyses. For dendritic branching analysis, randomly se lected layer V pyramids of the parietal cortex (6 neurons per brai n) had camera lucida drawings prepared of their basilar dendritic arbors. These were subs equently quantified for the amount and distribution of their dendritic domain s using the Sholl analysis (Method of Concentric Circles; (Sholl, 1953)). Pairwise statistical comparisons of the Sholl profiles utilized the repeat ed measures ANOVA with t he post-hoc tuckey test. For dendritic spine analysis, from the cod ed slides, spines were counted directly on the Zeiss research microscope us ing 100x long-working distance oilimmersion objectives. Spines were counted from two neur onal populations: along 30 micron terminal tip segments of the basilar tree of the layer V pyramids, and along 30 micron long terminal tip segments of the basilar tree of CA1 pyramids of the hippocampus. Spines were counted on 3-5 terminal tip segments from each of 6 neurons per brain in each brain region. Only flanking dendritic spines were counted, e.g., sp ines which were not obscured by the dendritic branch itself. Statistical analys is of the spine counts was carried out using ANOVA with a post-hoc tukey test.

PAGE 105

98 ELISA Analysis of A levels Protein extracts from the above eight hippocampi were used for ELISA and western blot analysis. ELISA was performed as previously described (Dodar t et al., 2002), with the modification that PBS rather that TBS was used to extract soluble A. Microarray. Both experimental groups, EE and SH, were represented by two microarrays each, and each microarray had RNA pooled from two mice (n=8 total). Immediately follo wing saline perfusion, brains were micro-dissected and the hippocampus isolated and fr ozen in liquid nitrogen. Total RNA was isolated from each hippocampus using an RNeasy kit (Qiagen). The microarray used, GeneChip 430 2.0 Mouse Expression Se t, contains over 45,000 probe sets designed from GenBank, dbEST, and RefS eq sequences clustered based on build 107 of the UniGene database. The clusters were further refined by comparison to the publicly available dr aft assembly of the mouse genome. An estimated 39,000 distinct transcripts ar e detected including over 34,000 well substantiated mouse genes. Five micr ograms of total RN A pooled from each brain sample served as the mRNA source for microarray analysis. The poly(A) RNA was specifically converted to cDNA and then amplified and labeled with biotin following a previously descri bed procedure (Van Gelder et al., 1990). Hybridization with the biot in labeled RNA, staining, and scanning of the chips followed the proscribed proc edure outlined in the Affy metrix technical manual and has been previously described (Wa rrington et al., 2000). Scanned output files were visually inspected for hybr idization artifacts and then analyzed using Affymetrix Microarray 5.1 software. Si gnal intensity was sca led to an average

PAGE 106

99 intensity of 500 prior to comparison anal ysis. Using the default settings, MAS 5.1 software identifies the in creased and decreased genes between any two samples with a statistical algorithm that asse sses the behavior of 16 different oligonucleotide probes designed to detect the same gene (Liu et al., 2002). Probe sets that yielded a change pvalue less than 0.0045 were identified as changed (increased or decreased) and those that yielded a pvalue between 0.0045 and 0.006 were identified as margi nally changed. A gen e was identified as consistently changed if it was i dentified as changed in all replicate experiments. Rolipram Administration. Twelve-month-old PS1/PDAPP/APOE+/were subcutaneously injected daily with either 0.03 mg/kg Ro lipram (n=5) or vehicle only (n=4) for two weeks. After one day without treatment all mice were behaviorally tested 6 for days in t he radial arm water maze. Real-Time quantitative PCR. RNA samples were treated with DNase (DNAfree, Ambion), to reduce the chance of genomic DNA contamination. cDNA synthesis was achieved using t he SuperScript III First-Strand Synthesis SuperMix kit (Invitrogen) according to t he manufacturer’s instructions. Primer pairs for real-time PCR were designed us ing the web-based app lications Primer3 (http://fokker.wi.mit.edu/cgi-bin/primer3 /primer3_www.cgi) and the Oligo Toolkit (Operon Technologies). The following pr imer pairs were used: Prolactin Receptor primer 1-ATCATTG TGGCCGTTCTCTC, Primer 2TGGAAAGATGCAGGTCATCA; Transthyretin, primer 1ATGGCTTCCCTTCGACTCTT, primer 2-GC ATCCAGGACTTTGACCAT; Insulin

PAGE 107

100 like growth factor 2, primer 1-CCCTCAGCAAGTGCCTAAAG, primer 2TTAGGGTGCCTCGAGATGTT; Insulin-like gr owth factor binding protein 2, primer 1-GCCGGTACAACC TTAAGCAG, primer 2GTGTTGGGGTTCACACACC; 18s RNA, primer 1GTAACCCGTTGAACCCCATT, primer 2CCATCCAATCGGTAGTAGCG. 25 l PCR reactions were prepar ed using iQ SYBR Green S upermix (Bio-Rad), and repeated in triplicate. A tw o-step PCR reaction was run us ing an iCycler iQ RealTime PCR detection system (Bio-Rad) as follows: 1 cycle of 95C for 3 min. followed by 40 cycles of 95C for 30 s. and 60C for 30 s. All primer sets yielded a single fluorescence peak via melt curv e analysis, indicating a lack of mispriming or primer-dimer ar tifacts. Fold-change values were calculated for all experimental wells using the comparative CT method (2Ct) (Livak, 1997).

PAGE 108

101 ACKNOWLEDGEMENTS The research was supported by a grant to H.P from the Nati onal Institute on Aging, a grant to G.A. from the Alzheimer’s associati on, the Eric Pfeiffer Chair for Research in Alzheimer’s disease at t he Suncoast Gerontology Center at the University of South Flori da and the Johnnie B. Byrd Sr Alzheimer’s Center and Research Institute.

PAGE 109

102 Figure 1. Environmental enrichme nt (EE) protect s PS1/PDAPP mice against cognitive impairment acro ss multiple behavioral tasks administered between 4-6 months of age. (A) In the radial arm wate r maze task of working me mory, SH Tg+ mice given showed impaired performance across days 79 of testing on working memory Trials 4 and 5 (T4, T5), while EE Tg+ mi ce exhibited working memory on trials 4 and 5 that was indisti nguishable from non-transgeni c standard-housed (Tg-/SH) mice. On working memory T4, EE also significantly improved performance of non-transgenic mice raised in EE compared to NT/SH mice. All groups were similar in escape latencies during the semi -randomized initial trial (T1) (B) In the platform recognition task of search/ident ification, the performance of Tg+/EE mice over all 4 days of testing was si gnificantly better than impaired Tg+/SH controls and identical to Tg-/SH mice. Th is effect was evident on individual Days 2-4 of testing. (C) En vironmental enrichment comple tely protected mice from overall cognitive impairment across 8 behavioral measures evaluated by discriminant function analysis (DFA). This Canonical Scores Pl ot depicts a linear representation of the two f unctions resulting from DFA (Wilks’ lambda for overall discrimination was p<0.0001) representing the over all poorer cognitive performance of control Tg+/SH mice relative to both enriched Tg+/EE mice and control NT/SH mice, which could not be discriminated from one another. In (A) and (B), means SEM’s are plotted. = p<0.002 or higher level of significance for Tg+/SH vs. both Tg+/EE and NT/SH; # = p<0.02 for NT/SH vs. NT/EE; ** = p<0.05 or higher level of significa nce for Tg+/SH vs. all other gr oups for that day.

PAGE 110

103 0 30 50Tg+/EE Tg+/SH NT/EE NT/SH Ret.Trial* *T1T4T5Latency (sec) 1 2 3 4 0 10 20 30** ** **DayLatency (sec) A B.# 40 20 10 60 C.Tg+/EE Tg+/SH NT/SH

PAGE 111

104 Figure 2. A and compact amyloid plaque levels are severely reduced in Tg+ mice both environmentally enriched and behaviorally tested despite dendritic deficits. (A) Immunohistochemical analysis, usi ng the monoclonal antibody 6E10, reveals a 51.5% reduction in area immunostained ce rebral cortex (P=0.025) in those mice behaviorally tested and raised in EE (n=8) vs. mice only raised in EE (n=11). (B) Similarly, in hippocampus, mice behaviorally tested and raised in EE had a 44.1% (P=0.022) reduction in immunost aining vs. those only raised in EE. There was no statistically significant differ ence in immunostaining in either cortex or hippocampus between EE and SH mice in either tested or non-tested cohorts. (C) Compact plaque deposition, as measured by thioflavi ne S staining, shows, in cerebral cortex, a 57.6% reduction in mature plaque deposition (p=0.008) and (D) a 69.0% reduction (P<0.0001) in hippocampal deposition in mi ce behaviorally tested and raised in EE vs. those onl y raised in EE. As with 6E10 immunostaining, there was no statistically significant difference in thioflavine S staining in either cortex or hippocampus between EE and SH mice in either tested or non-tested cohorts. (E) EE d oes not rescue the transgene-induced dendritic arborization deficits, as show n by Golgi staining (p<0.0001) (Sholl method). (F) Dendritic spine counts furt her verify that EE has no affect on dendritic morphology.

PAGE 112

105

PAGE 113

106 Figure 3. Environmental Enrichment results in a set of genetic changes that culminate in anti-apoptotic/neu roprotective BAD phosph orylation. A number of proteins (repr esented as red in this figure) that affect BAD phosphorylation were changed in their expression due to EE. Insulin-like growth factor ( IGF-2 ) was up-regulated 2.5 fold, and Insulin-like growth factor binding protein-2 ( IGFBP-2 ) was up-regulated 3.0 fold in Tg+/EE mice, which acts through the Akt and Erk1/2 pathways to phosphorylate BAD. Working only through the Akt pathway, the prolactin receptor ( PRLR ) was up-regulated 21.1 fold in Tg+/EE mice. Phosphodiestera se 4B was down-regulated 2.2 fold, resulting in a diminishing of cAMP inhibition and a resultant increase in Mek/Erk mediated BAD phosphorylation. Also shown is Rolipram, a specific PDE4 inhibitor, whose ability to mimic EE was te sted (Figure 4). A 10-fold increase in transthyretin ( TTR ) was also observed in the Tg+/EE mice, an A binding protein known to sequester A and inhibit amyloid formation. (Bole-Feysot et al., 1998; Bergmann, 2002; Stein and Johnson, 2002). The figure shows how the protective affects of BAD phosphoryl ation and TTR may inhibit neuronal degeneration and dysfunction caused by Alzh eimer’s-related changes in A peptide and -phosphorylation.

PAGE 114

107

PAGE 115

108 Figure 4. Administration of Rolipra m to Tg+ mice results in memory improvement which mimics envi ronmental enrichment. ( A) In the radial arm water maze task of working memo ry, Tg+ mice injected with Rolipram showed significantly improved performance across days 4-6 of testing on working memory Trials 4 and 5 (T4, T5 ), while Tg+ injected with vehicle only exhibited working memory deficits typica l of PS/PDAPP mice. Both groups were similar in escape latencies during t he semi-randomized initial trial (T1).

PAGE 116

109 Vehicle Rolipram Latency(sec) Ret.Trial

PAGE 117

110 Table 1. Comparative mi croarray analysis of hippocampal gene expression between environmentally enriched and standard housed PS1/PDAPP mice Classification Gene Accession Number Affymetrix ID SLR FC Amyloid Sequestration Transthyretin NM_013697 1451580_a_at 3.425 10.7 1455913_x_at 3.075 8.4 1454608_x_at 2.575 6.0 1459737_s_at 2.5 5.7 Neuroprotection Prolactin Receptor NM_008932 1448556_at 4.4 21.1 1425853_s_at 2.2 4.6 1421382_at 1.875 3.7 Insulin-like growth factor binding protein 2 NM_008342 1454159_a_at 1.575 3.0 Insulin-like growth factor 2 NM_010514 1448152_at 1.325 2.5 Memory Related Phospholipase A2, group V NM_011110 1417814_at 2.75 6.7 Cytochrome c oxidase, subunit VIIIb NM_007751 1449218_at 1.625 3.1 Cholecystokinin B receptor NM_007627 1454770_at -1.675 -3.2 Phosphodiesterase 4B, cAMP specific NM_019840 1442700_at -1.125 -2.2 Plasticity Related Calcium/calmodulin-dependent protein kinase II NM_009792 1452453_a_at 0.475 1.4 Aging Related Klotho NM_013823 1423400_at 2.175 4.5 Extracellular Matrix Procollagen, type VIII, 1 NM_007739 1418441_at 1.975 3.9 1418440_at 1 2.0 Procollagen, type IX, 3 NM_009936 1460693_a_at 1.875 3.7 1420280_x_at 1.425 2.7 Procollagen C-proteinase enhancer protein NM_008788 1437165_a_at 1.4 2.6 Procollagen C-endopeptidase enhancer 2 NM_029620 1451527_at 1 2.0 Immediate Early Genes Arc NM_018790 1418687_at -0.825 -1.8 Egr1 NM_007913 1417065_at -0.825 -1.8 Presynaptic Proteins Doc2a BB543070 1436862_at -2.075 -4.2 Synaptotagmin-like 2 NM_031394 1421594_a_at -0.55 -1.5 Syntaxin 1A (brain) NM_016801 1437390_x_at -0.35 -1.3 1448366_at -0.325 -1.3 Calcium/calmodulin -dependent protein kinase II NM_178597 1423941_at -0.3 -1.2 1423942_a_at -0.25 -1.2 AD Related -site APP cleaving enzyme (BACE) NM_011792 1455826_a_at -0.225 -1.2 amyloid (A4) precursor-like protein 2 NM_009691 1432344_a_at 0.325 1.3 Calcineurin NM_024459 1450368_a_at 0.35 1.3 Other Steap (six transmembrane epithelial antigen of the prostate) NM_027399 1424938_at 5.425 43.0 growth hormone NM_008117 1460613_x_at -4.225 18.7 1437522_x_at -3.05 8.3 1460310_a_at -1.725 3.3 SLR, Signal log ratio; FC, Fold change (approximate) = 2(SLR)

PAGE 118

111 Table 2. qRT-PCR verification of select transcripts from hippocampal microarray analysis. Gene CT Fold Change (2CT) Prolactin Receptor -3.59 1.02 12.06 Transthyretin -5.39 1.28 41.98 Insulin-like growth factor 2 -1.1 0.86 2.144 Insulin like growth factor binding protein 2 -0.99 0.75 1.99 Comparative threshold cycle: CT = CT(EE) CT(SH) CT = CT (target) CT(reference)

PAGE 119

112 REFERENCES Arendash GW, Garcia MF, Costa DA, Cra cchiolo JR, Wefes IM, Potter H (2004a) Environmental enrichment improves cognition in aged Alzheimer's transgenic mice despite stable beta-amyloid deposition. Neuroreport 15:1751-1754. Arendash GW, King DL, Gordon MN, Mor gan D, Hatcher JM, Hope CE, Diamond DM (2001a) Progressive, age-related behavio ral impairments in transgenic mice carrying both mutant amyloid precur sor protein and presenilin-1 transgenes. Brain Res 891:42-53. Arendash GW, Lewis J, Leighty RE, McGo wan E, Cracchiolo JR, Hutton M, Garcia MF (2004b) Multi-metric behav ioral comparison of APPsw and P301L models for Alzheimer's disease: linkage of poorer cognitive performance to tau pathology in forebrain. Brain Res 1012:29-41. Arendash GW, Gordon MN, Diamond DM, Au stin LA, Hatcher JM, Jantzen P, DiCarlo G, Wilcock D, Morgan D (2001b) Behavioral assessment of Alzheimer's transgenic mice following long-term Abeta vaccination: task specificity and correlations between Abeta deposition and spatial memory. DNA Cell Biol 20:737-744. Barad M, Bourtchouladze R, Winder DG, Golan H, Kandel E (1998) Rolipram, a type IV-specific phosphodiesterase inhibitor, facilitates the establishment of long-

PAGE 120

113 lasting long-term potentiati on and improves memory. Proc Natl Acad Sci U S A 95:15020-15025. Bergmann A (2002) Survival signa ling goes BAD. Dev Cell 3:607-608. Bole-Feysot C, Goffin V, E dery M, Binart N, Kelly PA (1998) Prolactin (PRL) and its receptor: actions, signal transduct ion pathways and phenotypes observed in PRL receptor knockout mice. Endocr Rev 19:225-268. Bonni A, Brunet A, West AE, Datta SR Takasu MA, Greenberg ME (1999) Cell survival promoted by the Ras-MAPK signaling pathway by transcriptiondependent and -independent mechanism s. Science 286:1358-1362. Bourtchouladze R, Lidge R, Catapano R, Stanley J, Gossw eiler S, Romashko D, Scott R, Tully T (2003) A mouse model of Rubinstein-Taybi syndrome: defective long-term memory is ameliora ted by inhibitors of phosphodiesterase 4. Proc Natl Acad Sci U S A 100:10518-10522. Calon F, Lim GP, Yang F, Morihara T, Te ter B, Ubeda O, Rostaing P, Triller A, Salem N, Jr., Ashe KH, Frautschy SA, Cole GM (2004) Docosahexaenoic acid protects from dendritic pathology in an Alzheimer's disease mouse model. Neuron 43:633-645. Chiang AS, Blum A, Barditch J, Chen YH Chiu SL, Regulski M, Armstrong JD, Tully T, Dubnau J (2004) radish encodes a phospholipase-A2 and defines a neural circuit involved in anesthesia-re sistant memory. Cu rr Biol 14:263-272.

PAGE 121

114 Comery TA, Shah R, Greenough WT (1995) Differential rearing alters spine density on medium-sized spiny neurons in the rat corpus striatum: evidence for association of morphological plasticity with early response gene expression. Neurobiol Lear n Mem 63:217-219. Costa DA, Nilsson LN, Bales KR, Paul SM Potter H (2004) Apolipoprotein is required for the formation of filamentous amyloid, but not for amorphous Abeta deposition, in an AbetaPP/ PS double transgenic mouse model of Alzheimer's disease. J Alzheimers Dis 6:509-514. Dodart JC, Bales KR, Johnstone EM, Little SP, Paul SM (2002) Apolipoprotein E alters the processing of the beta-am yloid precursor protein in APP(V717F) transgenic mice. Brain Res 955:191-199. Dore S, Kar S, Quirion R ( 1997) Insulin-like growth fact or I protects and rescues hippocampal neurons against beta-amyloidand human amylin-induced toxicity. Proc Natl Acad Sci U S A 94:4772-4777. Duff K, Eckman C, Zehr C, Yu X, Pr ada CM, Perez-tur J, Hutton M, Buee L, Harigaya Y, Yager D, Mor gan D, Gordon MN, Holcomb L, Refolo L, Zenk B, Hardy J, Younkin S (1996) Increased am yloid-beta42(43) in brains of mice expressing mutant preseni lin 1. Nature 383:710-713. Frey U, Huang YY, Kandel ER (1993) Effect s of cAMP simulate a late stage of LTP in hippocampal CA1 neurons. Science 260:1661-1664.

PAGE 122

115 Frick KM, Fernandez SM (2003) Enrichment enhances spatial memory and increases synaptophysin levels in aged female mice. Neurobiol Aging 24:615626. Fujita S, Ikegaya Y, Nishiyama N, Matsuki N (2000) Ca2+-independent phospholipase A2 inhibitor impairs spatia l memory of mice. Jpn J Pharmacol 83:277-278. Games D, Adams D, Alessandrini R, Bar bour R, Berthelette P, Blackwell C, Carr T, Clemens J, Donaldson T, Gillespie F, et al. (1995) Alzheimer-type neuropathology in transgenic mice ov erexpressing V717F beta-amyloid precursor protein. Nature 373:523-527. Gattaz WF, Forlenza OV, Talib LL, Bar bosa NR, Bottino CM (2004) Platelet phospholipase A(2) activity in Alzheimer' s disease and mild cognitive impairment. J Neural Transm 111:591-601. Globus A, Rosenzweig MR, Bennett EL, Diamond MC (1973) Effects of differential experience on dendritic spine co unts in rat cerebral cortex. J Comp Physiol Psychol 82:175-181. Gong B, Vitolo OV, Trinchese F, Liu S, S helanski M, Arancio O (2004) Persistent improvement in synaptic and cognitive f unctions in an Alzheimer mouse model after rolipram treatment. J Clin Invest 114:1624-1634.

PAGE 123

116 Impey S, Obrietan K, Storm DR (1999) Making new connections: role of ERK/MAP kinase signaling in neur onal plasticity. Neuron 23:11-14. Jankowsky JL, Xu G, Fromholt D, Gonzales V, Borchelt DR (2003) Environmental enrichment exacerbates am yloid plaque formation in a transgenic mouse model of Alzheimer disease. J Neuropathol Exp Neurol 62:1220-1227. Jensen MT, Mottin MD, Cracchiolo JR, Leighty RE, Arendash GW (2005) Lifelong immunization with human beta-am yloid (1-42) protects Alzheimer's transgenic mice against cognitive impai rment throughout aging. Neuroscience 130:667-684. Kayyali US, Zhang W, Yee AG, Seidm an JG, Potter H (1997) Cytoskeletal changes in the brains of mice lacking calcineurin A alpha. J Neurochem 68:16681678. Leighty RE, Nilsson LN, Potter H, Cost a DA, Low MA, Bales KR, Paul SM, Arendash GW (2004) Use of mu ltimetric statistical analysis to characterize and discriminate between the performance of f our Alzheimer's transgenic mouse lines differing in Abeta deposition. Behav Brain Res 153:107-121. Lemaire M, Barneoud P, Bohme GA, Piot O, Haun F, Roques BP, Blanchard JC (1994) CCK-A and CCK-B receptors enhance olfactory recognition via distinct neuronal pathways. Learn Mem 1:153-164.

PAGE 124

117 Liu WM, Mei R, Di X, Ryder TB, Hubbell E, Dee S, Webster TA, Harrington CA, Ho MH, Baid J, Smeekens SP (2002) A nalysis of high density expression microarrays with signed-rank call al gorithms. Bioinformatics 18:1593-1599. Livak K (1997) Comparative CT Me thod (Separate Tubes), ABI PRISM 7700 Sequence Detection System. The Pe rkin-Elmer Corporation:11-15. Loewenstein DA, Acevedo A, Czaja SJ, D uara R (2004) Cognit ive rehabilitation of mildly impaired Alzheimer disease pati ents on cholinesterase inhibitors. Am J Geriatr Psychiatry 12:395-402. Mutisya EM, Bowling AC, Beal MF (1994) Co rtical cytochrome oxidase activity is reduced in Alzheimer's dis ease. J Neurochem 63:2179-2184. Nilsson LN, Arendash GW, Leighty RE, Costa DA, Low MA, Garcia MF, Cracciolo JR, Rojiani A, Wu X, Bales KR, Paul SM, Potter H (2004) Cognitive impairment in PDAPP mice depends on ApoE and ACT-catalyzed amyloid formation. Neurobiol Aging 25:1153-1167. Ognibene E, Middei S, Daniel e S, Adriani W, Ghirardi O, Caprioli A, Laviola G (2005) Aspects of spatial memory and behavioral disinhibition in Tg2576 transgenic mice as a model of Alzheime r's disease. Behav Brain Res 156:225232. Pinaud R, Penner MR, Robertson HA, Cu rrie RW (2001) Upregulation of the immediate early gene arc in the brains of rats exposed to environmental

PAGE 125

118 enrichment: implications for molecular pl asticity. Brain Res Mol Brain Res 91:5056. Ramirez-Amaya V, Escobar ML, C hao V, Bermudez-Rattoni F (1999) Synaptogenesis of mossy fibers induced by spatial water maze overtraining. Hippocampus 9:631-636. Riley KP, Snowdon DA, Desrosiers MF, Markesbery WR (2005) Early life linguistic ability, late life cognitive func tion, and neuropathology: findings from the Nun Study. Neurobiol Aging 26:341-347. Schafe GE, Atkins CM, Swank MW, B auer EP, Sweatt JD, LeDoux JE (2000) Activation of ERK/MAP kinase in the amygdala is required for memory consolidation of pavlovian fear c onditioning. J Neurosci 20:8177-8187. Schwarzman AL, Gregori L, Vitek MP, Ly ubski S, Strittmatter WJ, Enghilde JJ, Bhasin R, Silverman J, Weisgraber KH, Coyle PK, et al. (1994) Transthyretin sequesters amyloid beta protein and prevent s amyloid formation. Proc Natl Acad Sci U S A 91:8368-8372. Serot JM, Christmann D, Dubost T, Couturier M ( 1997) Cerebrospinal fluid transthyretin: aging and late onset Alz heimer's disease. J Neurol Neurosurg Psychiatry 63:506-508. Sholl DA (1953) Dendritic organization in the neurons of the visual and motor cortices of the cat. J Anat 87:387-406.

PAGE 126

119 Snowdon DA, Kemper SJ, Mortimer JA, Greiner LH, Wekstein DR, Markesbery WR (1996) Linguistic ability in early life and cognitive function and Alzheimer's disease in late life. Findings fr om the Nun Study. Jama 275:528-532. Stein TD, Johnson JA (2002) Lack of neurodegeneration in transgenic mice overexpressing mutant amyloid precurso r protein is associated with increased levels of transthyretin and the activation of cell survival pathways. J Neurosci 22:7380-7388. Stein TD, Anders NJ, DeCarli C, C han SL, Mattson MP, Johnson JA (2004) Neutralization of transthyretin reverses the neuroprotective e ffects of secreted amyloid precursor protein (APP) in APPSW mice resulting in tau phosphorylation and loss of hippocampal neurons: support for the amyloid hypothesis. J Neurosci 24:7707-7717. Stern Y, Gurland B, Tatemichi TK, T ang MX, Wilder D, Mayeux R (1994) Influence of education and occupation on the incidence of Alzheimer's disease. Jama 271:1004-1010. Strokin M, Sergeeva M, Reiser G (2004) Role of Ca2+-independent phospholipase A2 and n-3 pol yunsaturated fatty acid docosahexaenoic acid in prostanoid production in brain: perspectives for protection in neuroinflammation. Int J Dev Neurosci 22:551-557.

PAGE 127

120 Sun GY, Xu J, Jensen MD, Simonyi A ( 2004) Phospholipase A2 in the central nervous system: implications for neur odegenerative diseases. J Lipid Res 45:205-213. Tanzi RE, Gusella JF, Watkins PC, Bruns GA, St George-Hyslop P, Van Keuren ML, Patterson D, Pagan S, Kurnit DM, Neve RL (1987) Amyloid beta protein gene: cDNA, mRNA distribution, and genetic linkage near the Alzheimer locus. Science 235:880-884. Thompson JE, Thompson CB (2004) Pu tting the rap on Akt. J Clin Oncol 22:4217-4226. Tsuzuki K, Fukatsu R, Yamaguchi H, Ta teno M, Imai K, Fu jii N, Yamauchi T (2000) Transthyretin binds amyloid bet a peptides, Abeta1-42 and Abeta1-40 to form complex in the autopsied human kidney possible role of transthyretin for abeta sequestration. N eurosci Lett 281:171-174. Tsuzuki K, Fukatsu R, Hayashi Y, Yoshid a T, Sasaki N, Takamaru Y, Yamaguchi H, Tateno M, Fujii N, Tak ahata N (1997) Amyloid beta pr otein and transthyretin, sequestrating protein colocalize in no rmal human kidney. Neurosci Lett 222:163166. Turner CA, Lewis MH, King MA (2003) Environmental enrichment: effects on stereotyped behavior and dendritic mor phology. Dev Psychobiol 43:20-27.

PAGE 128

121 Valverde F (1993) The rapid Golgi technique for staining CNS neurons: Light microscopy. Neuroscience Protocols 1:1-9. Van Gelder RN, von Zastrow ME, Yool A, Dement WC, Ba rchas JD, Eberwine JH (1990) Amplified RNA synthesized fr om limited quantities of heterogeneous cDNA. Proc Natl Acad Sci U S A 87:1663-1667. Warrington JA, Nair A, Mahadevappa M, Ts yganskaya M (2000) Comparison of human adult and fetal expressi on and identification of 535 housekeeping/maintenance genes. Physiol Genomics 2:143-147. Wilson RS, Mendes De Leon CF, Barnes LL, Schneider JA, Bienias JL, Evans DA, Bennett DA (2002) Participation in c ognitively stimulating activities and risk of incident Alzheimer disease. Jama 287:742-748. Young D, Lawlor PA, Leone P, Dragunow M, During MJ (1999) Environmental enrichment inhibits spontaneous apoptos is, prevents seizures and is neuroprotective. Nat Med 5:448-453. Zeng H, Chattarji S, Barbarosie M, Rondi -Reig L, Philpot BD, Miyakawa T, Bear MF, Tonegawa S (2001) Forebrain-specific calcineurin knockout selectively impairs bidirectional synaptic plastici ty and working/episodic-like memory. Cell 107:617-629. Zhang HT, Crissman AM, Dorairaj NR, Chandler LJ, O'Donnell JM (2000) Inhibition of cyclic AMP phosphodiestera se (PDE4) reverses memory deficits

PAGE 129

122 associated with NMDA receptor ant agonism. Neuropsychopharmacology 23:198204. Zhang HT, Zhao Y, Huang Y, Dorairaj NR, Chandler LJ, O'Donnell JM (2004) Inhibition of the phosphodiesterase 4 (PDE 4) enzyme reverses memory deficits produced by infusion of the MEK inhibito r U0126 into the CA1 subregion of the rat hippocampus. Neuropsychopharmacology 29:1432-1439.

PAGE 130

123 DISCUSSION In vitro (Ma et al., 1994) and in vivo (Nilsson et al., 2001) studies have found that the chief pathogenic chaperone found in amyloid plaques, apoE, is able to catalyze the polymerization of A into neurotoxic amyloid filaments and change the anatomical distribution of immunoreactivity in the brain. Furthermore, the deposition of brain AD pathology is strongly influenced, both in intensity and age of onset, by the gene dosage of apoE. Specific binding experiments, site directed mutagenesis, and structural anal ysis have identif ied the primary molecular interactions between t he pathological chaperone, apoE, and A (Potter, 1991; Strittmatter et al., 1993; Ma et al., 1996; Janciauskiene et al., 1998). A previous study from our laboratory provides the first in vivo demonstration that the increased A immunoreactive and mature plaque deposition caused by expression of the am yloid promoter, apoE, is associated with impaired spatial learning (N ilsson et al., 2004, APPENDIX B) We found that apoE promotes t he development of both diffuse, immunoreactive, A deposits and mature amyloid plaques in A-overexpressing mice without affecting the levels of monomeric A itself and that impaired cognitive performance in two different spatial tasks requires both A peptide and either ApoE or ACT as an amyloid pr omoter, with apoE showing a higher catalytic activity. Furthermore, both diffuse A deposits and mature amyloid

PAGE 131

124 plaques were correlated with impaired spatial l earning ability. Indeed, 18 month old APP mice that didn’t express ApoE deposited very little A and displayed no cognitive deficits in the RAWM or Morr is Water Maze, similar to nontransgenic mice. These findings imply that the pr ocess and/or product of ApoE catalyzed amyloid formation is more critical for t he cognitive decline in AD than is the amount of monomeric A-pepti de. In sum, our previous findings of a strong association between cognitive decline and both diffuse A deposits and mature amyloid plaques in transgenic mice indicate that a major pathological role of ApoE in Alzheimer’s disease is to pr omote A polymerization and deposition, possibly with slightly different effects in terms of time course and anatomical distribution of the pathology. Given that the doubly transgenic APP/PS1 mouse model used in the currently reported study expresses se verely elevated levels of brain A1-42, most likely superseding any possible effect of apolipoprotein E on A peptide production or clearance (if t here are indeed any), we o ffer clear proof that apoE directly catalyzes the polymerization of A into -sheet fibrils and its consequent aggregation to form mature, Thioflavine S-positive, amyloid plaques. This directly contrasts with previous publicat ions only utilizing singly transgenic, PDAPP, animals, whereby apoE had been shown to promote both total A deposition and amyloid formation ( Nilsson et al., 2004, APPENDIX B). Transgenic mice expressing both human PS1M146L and APPV717F mutations do not develop significant fibrillar amyloid deposits by 7 months of age, in the absence of apoE, despite copious amounts of immunoreactive A

PAGE 132

125 deposition. We find that mature, Thiofl avine-S positive, am yloid deposits are allowed to form only the presence of apoE, and are the first authors, to our knowledge, to describe the gross mor phological chages of amyloid plaques, using electron microscopy, due to apoE. Although we find that ApoE clearly pr omotes A polymerization directly, this protein likely has additional roles to play in AD and other brain diseases. For example, other authors have sh own that ApoE affects the level or location of APP processing so as to generate fewer A peptides and more potentially toxic C-terminal fragments (in animal s slightly older than ours’ and at an age known to show the initial stages of am yloid deposition) (Dodart et al., 2002). Particularly interesting is the finding t hat ApoE4 is less neuroprot ective than ApoE3 both in vitro and in vivo (Buttini et al., 2002). ApoE is also known to affect chol esterol metabolism (Poirier, 2000), for ApoE knockout mice develop hyperchol esterolemia and atherosclerosis that could have an effect on amyloid formation. It is therefore wise to consider the possible direct and compensatory consequenc es of expressing or not expressing ApoE during development. However, het erozygous PDAPP/ApoE+/mice have normal serum cholesterol levels and develop amyloid deposits at a rate intermediate between full PDAPP/ApoE-K O and PDAPP/ApoE+/+ animals (Bales et al., 1999), suggesting that the lower am yloid deposition induced by knocking out ApoE is not caused by high serum cholesterol due to ApoE deficiency. Furthermore, except for hypercholestero lemia in ApoE-KO mice, we have not

PAGE 133

126 observed gross differences in physiology or brain morphology due to knocking out ApoE. Preventing apoE from binding to A and stimulating its polymerization into -sheet filaments may prove to be a prom ising therapeutic approach for AD drug development. It is possible that the process of amyloid formation, or an intermediate formed therein, rather than the end product (mature plaques), causes the typical neuronal dysfuncti on and cognitive decline of AD. These results showing the essential role of ApoE in amyloid fibril formation, together with the previous findings that apoE promotes cognitive decline and does not alter the steady state level of A, sugg est that pharmacologic inhibition of A/ApoE interactions should be a prime tar get for therapeutic intervention in AD. Previous experiments indicate that such an approach may be effective, for small fragments of the A peptide were shown to effectively prevent ApoE-catalyzed polymerization in vitro (Ma et al., 1994; Ma et al., 1996). A pharmacological target in the per iphery as a therapy against AD has clear pharmacological and clinical advantages There is much debate in the field of Alzheimer’s disease regarding the c ontribution of peripheral circulating proteins to disease pathology. The fact t hat disease-related proteins such as A and apoE are elevated in t he serum of AD patients, and that there exists a notable prevalence of vascular cerebral amyloid angiopathy indicates that circulating proteins are at least linked causally to the disease process. Since we observed such a drastic catalytic affect of apoE on the polymer ization of A, and since apoE is known to shuttle A ac ross the blood brain barrier, we surmised

PAGE 134

127 that if circulating apoE has an affect on brain pathology it would be easily detectable in our APP/PS1 transgenic mouse model. We demonstrate here that the unorthodox surgical technique of parabiosis is a feasible approach for studying the c ontribution of circulating proteins and hormones to Alzheimer’s disease pathology. We found that apoE is efficiently transferred from one parabiont to another using PCR analysis of white blood cells. Though transferred efficiently, the levels of circulating apoE protein determined via western blot were onl y 5% of that found in a nontransgenic mouse. These low steady state levels of apoE in the apoE-KO parabiont are most likely due to massive sequestration of apoE for use in lipoprotein particles and cholesterol transport. This is suggested by the fact that the hypercholesterolemic phenotype of apoE-knockout mice is rescued in these same mice when parabiosed to a m ouse expressing apoE. It should be mentioned that work by other authors has shown that it takes less than 3% of normal levels of apoE to prevent ather osclerotic plaque deposition and to maintain typical cholesterol leve ls (Thorngate et al., 2002). Based upon our findings regarding apoE-mediated th ioflavine-S catalysis, we anticipated that even minute amounts of apoE entering the central nervous system of PS1/APP/apoE-KO mice could promote large am ounts of fibrillar amyloid deposition. Surpri singly, there was no increase in A deposition, diffuse or fibrillar, whatsoever. Conversely, we found fewer fibrillar, Thioflavine Spositive, deposits in the par abiosed mice that expres sed apoE as compared to control mice. Immunohistochemical analysi s reveals that peripheral apoE is only

PAGE 135

128 detectable in the choroids plexus whic h is in agreement with another previous study (Martel et al., 1997). Since A injected into the circulat ion of mice is cl eared rapidly, the observed lowering of A deposition in parabiosed PS1+/-,APP+/+,apoE+/mice may be due to apoE-mediated peripheral A clearance. It has been shown that there exists a dynamic equilibrium between brain A levels and the levels in the CSF, which was demonstrated by usin g a peripheral antibody injection to sequester A (DeMattos et al., 2002). Besi des apoE’s known role as a promoter of fibrillization, ther e is increasing evidence for a dual role of apoE, in that it may also promote the clearance of A. For example, in vitro astrocytic internalization and degradation of A is found to be dependent on the presence of apoE (Koistinaho et al., 2004). Furthermore, intra-venous injections of A, which would normally be quickly cleared by the li ver, are not detectably cleared in mice devoid of apoE (Hone et al., 2003). Ther efore our study, and those of others, suggests that apoE may be involved in th e peripheral clearance/sequestration of A, acting as a sink which may c hange the equilibrium between brain and peripheral A. This could therefore alter the circulating:CSF/brain A equilibrium, shifting it towards the blood, re sulting in an efflux of A from the brain, consequently resulti ng in the observed reduction in amyloid deposition. This shift in A equilibr ium between discreet compartm entalized pools has also been suggested by the work of others (Silverberg et al., 2003). We conclude that peripheral apolipop rotein E does not contribute to the brain pathology found in Alzheimer’s dise ase, and may actually help promote the

PAGE 136

129 clearing/efflux of brain A by a mechani sm that remains open to speculation. Based on the amelioration of hypercholesterolemia due to parabiosis, we are confident that this uncomm on technique has merit for studying the contribution of circulating factors to Alzheimer’ s and other neurodegenerative diseases. Another, seemingly unrelated, primary topic of study in our laboratory is that of environmental enrichm ent as an intervention strategy for the treatment of AD. Initial environmental enrichment studies from our laboratory indicate that a long-term period of EE in aged AD transgenic mice results in superior overall cognitive performance as determined by a number of behavioral measures (Arendash et al., 2004, APPENDIX B). Although a reduction in brain A deposition has been associated with behavioral benefit of A immunotherapy in similar AD transgenic mice (C order et al., 1993), our initial results indicated that mechanisms independent of A deposition are sufficient for behavioral benefit, for we observed no changes in brain pathol ogy in EE mice. Al ong this line, EEinduced enhancements in neurogenesis (A nderson et al., 1998), synaptogenesis (Anderson and Higgins, 1997), growth factor levels (Arendash et al., 2001b), and gene expression (Arendash et al., 2001a) have all been s een in normal mice and are thus potentially involv ed in the behavioral benefits of EE we reported in aged APPsw mice (Arendash et al., 2004, APPENDIX B). In this prior study, designed to evaluate the potential of EE as a therapeutic means in aged AD transgenic mice, our result s suggested that longterm intensive enrichment/cognitive stimulat ion could be useful in stabilizing or slowing the cognitive decline of AD and it’s predecessor, mild cognitive

PAGE 137

130 impairment (MCI) without a need to r educe A burden. Largely because of practical considerations, prospective longitudinal studies have not been done in humans to confirm controversial retrospec tive studies reporting AD risk reduction with “lifelong” educatio n/occupation-related intellectual activity (Bales et al., 1997; Bales et al., 1999). A recent 5-year longitudinal study in volving non-demented 75+ year olds did find participation in cogn itively-stimulating leisure activities to be associated with a lower risk of dement ia (Dodart et al., 2002). However, a cause and effect relationship can not be established because the leisure activities were self-chosen. The current study sought to both elimin ate any cause/effect ambiguity by randomly assigning transgenic mice destined to develop AD and their NT counterparts to either EE or SH and to ascertain the pr eventative benefits of long term enrichment. We find that mice raised in a cognit ively stimulating environment were protected from certai n cognitive decline as measured by multiple behavioral assays. Our result s also indicate that, alone, neither environmental enrichment nor behavioral testing could alte r brain A levels. The cognitive benefits of prevent ative EE are theref ore at least partl y independent of AD brain pathology, for we observed w hat may be a cognitive stimulation threshold for the clearance of A that does not affect cognitive measures performance, for only the combination of both long-term environmental enrichment and intense behavioral testing resu lted in a lowered level of brain A deposition. In fact, we observed up to 69% decreases in both diffuse and compact A levels, and surmise that a decrease of this magnitude could not

PAGE 138

131 possibly be from only a halting in A deposition, but from an as yet unknown removal or sequestration mechanism. EE has been shown to prime the brain for increased maze training-induced choline acet yltransferase activity (Levi et al., 2003), and we observe what is most likely a similar priming by EE for behavioral testing-induced pathology changes. A dditionally, microarray analysis revealed a number of beneficial gene expressi on changes related to memory, neuroprotection, and A sequestration. Previous studies have revealed that EE is capable of inducing beneficial structural changes such as increases in the number of de ndritic spines and branching complexity (Comery et al., 1995; Turner et al., 2003). In our experiment, we observed no such changes due to EE. In fact, all the transgenic mice in our study displayed dendritic def ects, as determined by Golgi staining, and these were unaffected by EE, indicating that 1) these physical defects could not be rescued by EE, and 2) that the obs erved memory improvement due to EE in these mice was independent of dendritic structural changes. Microarray analysis was performed to study the possible genetic changes that could be responsible for the improved cognitive performance and the observed lowering of AD brain pathology due to cognitive stimulation. We focused on the roughly 70 known genes that were si gnificantly changed, either up or down, by at least 2-fold due to EE, and provided an in-detail inquiry for those transcripts most likely to affect Al zheimer’s disease or cognitive function. We found that a number of genes known to be involved in pathways which culminate in the phosphorylation of BAD an event known to be anti-apoptotic

PAGE 139

132 and neuroprotective, had altered expressi on levels. Previous experiments wherein mice where fed a diet rich in a compound (docosahexa noic acid) known to promote BAD phosphorylation helped prot ect mice against cognitive deficits (Calon et al., 2004). In particular, we found four transcripts upstream of bad phosphorylation that were c hanged due to EE: phosphodies terase 4B, insulin-like growth factor-2, insulin-like growth fact or binding protein-2, and the prolactin receptor. Docosahexanoic acid-induced BAD phosphorylation is an IRS-1/PI3-K mediated event, and three transcripts for the proteins, IGF-2, IGFBP2, and PRLR which are also known to promote BAD phosphorylation via the IRS-1/PI3-K pathway were significantly upregulated. IGF-2 and IGFBP2 are also able induce BAD phosphorylation through an alternativ e MEK/ERK mediated pathway which has been show to be imperative in the fo rmation of long term memory (Impey et al., 1999; Schafe et al., 2000). Interesti ngly, PDE4, which was downregulated by EE, acts as an inhibitor of this same pathway (Zhang et al., 2004), and its inhibition has recently been shown to promote both memory and LTP in a mouse model of AD (Gong et al., 2004). By inject ing Rolipram, an inhibi tor of PDE4, into 8-month-old transgenic mice we have rescued the transgene-induced cognitive deficits, effectively mimicking the effect of environmental enrichment. Taken as a whole, the changes in these relat ed transcripts suggest the memory improvement in mice that were exposed to EE may be due an inducible mechanism that confers neuroprotec tion. A molecule know to bind to (Tsuzuki et al., 1997; Tsuzuki et al., 2000) and sequester (Schwarzman et al., 1994) A, transthyretin, was up-regulated roughly

PAGE 140

133 10-fold in transgenic mice raised in EE. TTR binds to A and is thought to inhibit A aggregation, for inhibiti on of TTR has been shown to increase A deposition (Stein et al., 2004). In fa ct, TTR is thought responsible for the prevention of peripheral A deposition, for the kidney’s le vel of this protein is quite high. Furthermore, TTR levels are reduced in human AD patients, and are hypothesized to be at least partly responsib le for the accumulation of A and the resultant neurodegeneration, so there exists a discrepancy, for in those mice only undergoing EE there was no decrease in pathology despite cognitive improvements. Besides these changes, some transcrip ts were altered due to EE that are known to improve memory. For exampl e, CCK-B was down -regulated, and past experiments have shown that inhibiting t he CCK-B receptor enhances memory in rats (Lemaire et al., 1994). Another protein, which is known to be reduced in human AD patients, phospholipase A2, wa s upregulated due to EE, and memory defects are known to be induced by its pharmacological inhibition. A number of other genes or gene groups experienced significant changes, but their direct relevance to our model and AD is unclear. For example, STEAP (six transmembrane epithelia l antigen of the prostate ), a seemingly ADand memory-unrelated transcript was up-regul ated by approximately 43-fold, the largest observed change on the microarray. Also a number of procollagen and procollagen-related transcripts were al so up-regulated. These proteins are potentially important for Sc hwann-cell extracellular matrix (Greenberg et al., 1980; Roytta et al., 1988), but their role in memory formation or improvement is

PAGE 141

134 unknown. Klotho, a gene encoding a me mbrane protein known to cause premature aging in mice when mutat ed (Kuro-o et al., 1997), was also upregulated by 4.5-fold. This could be an EE-induced compensatory mechanism against an aspect of aging that has yet to be determined. A recent report of a similar experim ent run in parallel by another group (Lazarov et al., 2005) using a different mouse model (APPsw/ PS1dE9), further confirms our findings that EE reduces amyloid deposition. It should be noted that we do not see any changes in the levels of neprilysin, an A degrading enzyme that these authors attribute to the EE-indu ced lowering of A, therefore indicating that other sequestration/degr adation mechanisms are most likely responsible for the changes in pathology that we observed. Furthermore, a differential sensitivity to A reduction due to EE likely depends on genotype, possibly explaining the lowered A deposition only in animals that underwent EE and intense behavioral testing. Jankowsky et al., in direct c ontrast to both of the above findings, reported that EE promotes A deposition in an APPs w/PS1dE9 mouse model of AD. Unfortunately, these authors c hose an unusual enric hment methodology whereby new mice were continually introduced to the cages, adding a confounding source of stress which ma y explain the observed increase in pathology. This therefore makes comparisons between this study and ours’ of little value. In conclusion, we found that the combination of both long-term EE and intense behavioral testing, both forms of cognitive stimulation, can reduce the levels of A deposition. Furthermo re, the gene expression changes due to EE

PAGE 142

135 may prime the brain for further changes in AD pathology indu ced by additional intense cognitive stimulati on. Therefore, pharmacological alteration of these genetic pathways, as exemplified by our inhibition of PDE4 by Rolipram, may prove to be valuable targets in the fight against the cognitive decline of Alzheimer’s disease.

PAGE 143

136 FUTURE DIRECTIONS The paths for a number of future av enues of study have been cleared by the recent research of our laboratory. For example, we show that parabiosis is a useful method for determining the contributi on of a circulating pr otein (in the case of our current research, apoE) to AD pat hology. Further experiments should be repeated with different par abiont combinations. For example, a mouse expressing circulating A, such as our PS1/APP mouse, should be parabiosed to a non-transgenic mouse, for any brain pathology present in the transgenic partner would be a result of circulating A, therefore answering the question of whether peripheral A contributes to brain pathology. If so, antibody injections or the administering of A-sequestering agents may prove useful as a stand-alone, or more likely as an adjunct AD therapy. If on the other hand there is no affect on brain pathology, efforts to curb brain pathology via the peripery should possibly be reevaluated. Similarly, ex periments using mice which overexpress peripheral 1-antichympotrypsin, another pat hogenic chaperone, should also be initiated for the same r easons. Since our environmental enrichment par adigm has a number of differing aspects which may be responsible for the observed memory improvement and genetic changes, it will be usef ul to parse out the aspect or aspects of EE which

PAGE 144

137 are most responsible for these benefic ial changes. Our “whole” EE setting entails social, physical, and cognitive el ements, and by systematically isolating and differentiating which element exerts the greatest cognitive and genetic benefits, using behavioral testing and further microarray analysis, both recommendations for using enrichment ther apy in humans and more appropriate pharmacological targeting can be dev ised with greater accuracy. Since pharmacological inhibition of PDE4 using Rolipram rescued the cognitive deficits in our transgenic mouse m odel of AD, further animal trials using other available phosphodiesterase inhibito rs (e.g. Aminophylline, cilomilast, etc…) should be initiated, and if successful should lead to eventual clinical trials for the treatment or prevent ion of Alzheimer’s disease. This strategy should furthermore be applied to the pathways and proteins whose expression changed due to EE that converge in BAD phosphorylation such as IGF-2, IGFBP2, and PRLR. Lastly, it has been shown that envir onmental enrichment-induced memory and learning improvements in transgenic mi ce are apoE allele-dependant (Levi et al., 2003), so future studies should ther efore concentrate on combining the two main foci of my work, apolipoprotein E and environmental enrichment, in order to elucidate the differing genetic changes and behavioral changes of the various apoE alleles among varying environmental enrichment conditions so as to develop a clearer understanding of the relevant pathways and define the most efficacious and personalized treatment regi mens possible in relation to ones’ given apoE allele.

PAGE 145

138 REFERENCES Abraham CR, Selkoe DJ, Potter H (1988) Immunochemical identification of the serine protease inhibitor alpha 1-antichymo trypsin in the brain amyloid deposits of Alzheimer's disease. Cell 52:487-501. Anderson R, Higgins GA (1997) Absence of central cholinergic deficits in ApoE knockout mice. Psychopharma cology (Berl) 132:135-144. Anderson R, Barnes JC, Bliss TV, Cain DP, Cambon K, Davies HA, Errington ML, Fellows LA, Gray RA, Hoh T, St ewart M, Large CH, Higgins GA (1998) Behavioural, physiological and morphological analysis of a li ne of apolipoprotein E knockout mouse. Neuroscience 85:93-110. Arendash GW, Garcia MF, Costa DA, Cra cchiolo JR, Wefes IM, Potter H (2004) Environmental enrichment improves c ognition in aged Alzheimer's transgenic mice despite stable beta-amyloid deposition. Neuroreport 15:1751-1754. Arendash GW, King DL, Gordon MN, Mor gan D, Hatcher JM, Hope CE, Diamond DM (2001a) Progressive, age-related behavio ral impairments in transgenic mice carrying both mutant amyloid precur sor protein and presenilin-1 transgenes. Brain Res 891:42-53. Arendash GW, Gordon MN, Diamond DM, Au stin LA, Hatcher JM, Jantzen P, DiCarlo G, Wilcock D, Morgan D (2001b) Behavioral assessment of Alzheimer's transgenic mice following long-term Abeta vaccination: task specificity and

PAGE 146

139 correlations between Abeta deposition and spatial memory. DNA Cell Biol 20:737-744. Bales KR, Verina T, Cummins DJ, Du Y, Dodel RC, Saura J, Fishman CE, DeLong CA, Piccardo P, Petegnief V, Ghe tti B, Paul SM (1999) Apolipoprotein E is essential for amyloid deposition in the APP(V717F) transgenic mouse model of Alzheimer's disease. Proc Natl Acad Sci U S A 96:15233-15238. Bales KR, Verina T, Dodel RC, Du Y, Al tstiel L, Bender M, Hyslop P, Johnstone EM, Little SP, Cummins DJ, Piccardo P, Ghetti B, Paul SM (1997) Lack of apolipoprotein E dramatically reduce s amyloid beta-peptide deposition. Nat Genet 17:263-264. Blain H, Jeandel C, Merched A, Visvikis S, Siest G (1997) Apolipoprotein E level in cerebrospinal fluid increases wit h aging. J Am Geriatr Soc 45:1536. Brown J, Cooper-Kuhn CM, Kempermann G, Van Praag H, Winkler J, Gage FH, Kuhn HG (2003) Enriched environment and physical activity stimulate hippocampal but not olfactory bulb neurogen esis. Eur J Neurosci 17:2042-2046. Bullido MJ, Guallar-Castillon P, Artiga MJ Ramos MC, Sastre I, Aldudo J, Frank A, Coria F, Rodriguez-Artale jo F, Valdivieso F (2000) Alzheimer's risk associated with human apolipoprotein E, alpha-2 macroglobulin and lipoprotein receptor related protein polymorphisms: absence of genetic interactions, and modulation by gender. Neurosci Lett 289:213-216.

PAGE 147

140 Buttini M, Yu GQ, Shockley K, Huang Y, Jones B, Masliah E, Mallory M, Yeo T, Longo FM, Mucke L (2002) Modulation of Al zheimer-like synaptic and cholinergic deficits in transgenic mice by human apol ipoprotein E depends on isoform, aging, and overexpression of amyloid beta pept ides but not on plaque formation. J Neurosci 22:10539-10548. Calon F, Lim GP, Yang F, Morihara T, Te ter B, Ubeda O, Rostaing P, Triller A, Salem N, Jr., Ashe KH, Frautschy SA, Cole GM (2004) Docosahexaenoic acid protects from dendritic pathology in an Alzheimer's disease mouse model. Neuron 43:633-645. Comery TA, Shah R, Greenough WT (1995) Differential rearing alters spine density on medium-sized spiny neurons in the rat corpus striatum: evidence for association of morphological plastici ty with early response gene expression. Neurobiol Lear n Mem 63:217-219. Corder EH, Saunders AM, Strittmatter WJ, Schmechel DE, Gaskell PC, Small GW, Roses AD, Haines JL, PericakVance MA (1993) Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer's disease in late onset families. Science 261:921-923. Dahlgren KN, Manelli AM, Stine WB, Jr., Ba ker LK, Krafft GA, LaDu MJ (2002) Oligomeric and fibrillar species of amyl oid-beta peptides differentially affect neuronal viability. J Biol Chem 277:32046-32053.

PAGE 148

141 Davis RN, Massman PJ, Doody RS (2001) Cognitive intervention in Alzheimer disease: a randomized placebo-controlled study. Alzheimer Dis Assoc Disord 15:1-9. DeKosky ST, Scheff SW (1990) Synapse loss in frontal cortex biopsies in Alzheimer's disease: correlation with cogni tive severity. Ann Neurol 27:457-464. DeMattos RB, O'Dell M A, Pa rsadanian M, Taylor JW Harmony JA, Bales KR, Paul SM, Aronow BJ, Holtzman DM (2002) Clusterin promotes amyloid plaque formation and is critical for neuritic toxi city in a mouse model of Alzheimer's disease. Proc Natl Acad Sci U S A 99:10843-10848. Diamond MC, Rosenzweig MR, Bennett EL, Lindner B, Lyon L (1972) Effects of environmental enrichment and impoveris hment on rat cerebral cortex. J Neurobiol 3:47-64. Dodart JC, Mathis C, Bales KR, Paul SM, Ungerer A (2000) Behavioral deficits in APP(V717F) transgenic mice deficient fo r the apolipoprotein E gene. Neuroreport 11:603-607. Dodart JC, Bales KR, Johnstone EM, Little SP, Paul SM (2002) Apolipoprotein E alters the processing of the beta-am yloid precursor protein in APP(V717F) transgenic mice. Brain Res 955:191-199. Driscoll DM, Getz GS ( 1984) Extrahepatic synthesis of apolipoprotein E. J Lipid Res 25:1368-1379.

PAGE 149

142 Ebly EM, Parhad IM, Hogan DB, Fung TS (1994) Prevalence and types of dementia in the very old: results from the Canadian Study of Health and Aging. Neurology 44:1593-1600. Evans DA, Hebert LE, Beckett LA, Scherr PA, Albert MS, Chown MJ, Pilgrim DM, Taylor JO (1997) Education and other m easures of socioeconomic status and risk of incident Alzheimer disease in a defined population of ol der persons. Arch Neurol 54:1399-1405. Fagan AM, Watson M, Parsadanian M, Bales KR, Paul SM, Holtzman DM (2002) Human and murine ApoE markedly alte rs A beta metabolism before and after plaque formation in a mouse model of Al zheimer's disease. Neurobiol Dis 9:305318. Fischer W, Bjorklund A (1991) Loss of AChEand NGFr-labeling precedes neuronal death of axotomized septal-d iagonal band neurons: reversal by intraventricular NGF infusion. Exp Neurol 113:93-108. Fischer W, Bjorklund A, C hen K, Gage FH (1991) NGF im proves spatial memory in aged rodents as a function of age. J Neurosci 11:1889-1906. Fischer W, Chen KS, Gage FH Bjorklund A (1992) Progressive decline in spatial learning and integrity of forebrain c holinergic neurons in rats during aging. Neurobiol Aging 13:9-23.

PAGE 150

143 Friedland RP, Fritsch T, Smyth KA, Koss E, Lerner AJ, Chen CH, Petot GJ, Debanne SM (2001) Patients with Alzheimer's disease have reduced activities in midlife compared with healthy control-group members. Proc Natl Acad Sci U S A 98:3440-3445. Glenner GG, Wong CW (1984) Alzheimer's disease: initial report of the purification and characterization of a nov el cerebrovascular amyloid protein. Biochem Biophys Res Commun 120:885-890. Globus A, Rosenzweig MR, Bennett EL, Diamond MC (1973) Effects of differential experience on dendritic spine co unts in rat cerebral cortex. J Comp Physiol Psychol 82:175-181. Goldgaber D, Lerman MI, McBride OW, Saffiotti U, Gajdusek DC (1987) Characterization and chromosomal loca lization of a cDNA encoding brain amyloid of Alzheimer's disease. Science 235:877-880. Goldgaber D, Schwarzman AI, Bhasin R, Gr egori L, Schmechel D, Saunders AM, Roses AD, Strittmatter WJ (1993) Sequestr ation of amyloid bet a-peptide. Ann N Y Acad Sci 695:139-143. Gnahn H, Hefti F, Heumann R, Schwab ME, Thoenen H (1983) NGF-mediated increase of choline acetyltransferase (ChAT) in the neonatal rat forebrain: evidence for a physiological role of NG F in the brain? Brain Res 285:45-52.

PAGE 151

144 Gong B, Vitolo OV, Trinchese F, Liu S, Shelanski M, Arancio O (2004) Persistent improvement in synaptic and cognitive f unctions in an Alzheimer mouse model after rolipram treatment. J Clin Invest 114:1624-1634. Greenberg JH, Foidart JM, Greene RM (1980) Collagen synthesis in cultures of differentiating neural crest ce lls. Cell Differ 9:153-163. Greenough WT, Volkmar FR (19 73) Pattern of dendritic branching in occipital cortex of rats reared in complex environments. Exp Neurol 40:491-504. Gu Q (1995) Involvement of nerve growth fa ctor in visual cort ex plasticity. Rev Neurosci 6:329-351. Guilarte TR, Toscano CD, McGlothan JL Weaver SA (2003) Environmental enrichment reverses cognitive and mole cular deficits induced by developmental lead exposure. Ann Neurol 53:50-56. Hebert LE, Scherr PA, Bienias JL, B ennett DA, Evans DA (2003) Alzheimer disease in the US population: prevalence estimates using the 2000 census. Arch Neurol 60:1119-1122. Holtzman DM, Pitas RE, Kilb ridge J, Nathan B, Mahley RW, Bu G, Schwartz AL (1995) Low density lipoprotein receptor-re lated protein mediates apolipoprotein E-dependent neurite outgrowth in a cent ral nervous system-derived neuronal cell line. Proc Natl Acad Sci U S A 92:9480-9484.

PAGE 152

145 Holtzman DM, Fagan AM, Mackey B, Tenkova T, Sartorius L, Paul SM, Bales K, Ashe KH, Irizarry MC, Hyman BT (2000) Ap olipoprotein E facilitates neuritic and cerebrovascular plaque formation in an Al zheimer's disease model. Ann Neurol 47:739-747. Hone E, Martins IJ, Fonte J, Martins RN (2003) Apolipoprotein E influences amyloid-beta clearance from the murine periphery. J Alzheimers Dis 5:1-8. Hu J, LaDu MJ, Van Eldik LJ (1998) Apo lipoprotein E attenuates beta-amyloidinduced astrocyte activati on. J Neurochem 71:1626-1634. Impey S, Obrietan K, Storm DR (1999) Making new connections: role of ERK/MAP kinase signaling in neur onal plasticity. Neuron 23:11-14. Iwatsubo T (2004) The gamma-secretase complex: machinery for intramembrane proteolysis. Curr Opin Neurobiol 14:379-383. Janciauskiene S, Rubin H, Lukacs CM, Wright HT (1998) Alzheimer's peptide Abeta1-42 binds to two beta-sheets of alpha1-antichymotrypsin and transforms it from inhibitor to substrat e. J Biol Chem 273:28360-28364. Jarrett JT, Berger EP, Lansbury PT, Jr. (1 993) The carboxy te rminus of the beta amyloid protein is critical for the seeding of amyloid formation: implications for the pathogenesis of Alzheimer's disease. Biochemistry 32:4693-4697. Jordan J, Galindo MF, Miller RJ, R eardon CA, Getz GS, LaDu MJ (1998) Isoform-specific effect of apolipopro tein E on cell survival and beta-amyloid-

PAGE 153

146 induced toxicity in rat hippocampal py ramidal neuronal cultures. J Neurosci 18:195-204. Kamboh MI (1995) Apolip oprotein E polymorphism and susceptibility to Alzheimer's disease. Hum Biol 67:195-215. Kamboh MI, Sanghera DK, Ferrell RE, De Kosky ST (1995) APOE*4-associated Alzheimer's disease risk is modified by alpha 1-antichymotr ypsin polymorphism. Nat Genet 10:486-488. Kang J, Lemaire HG, Unterbeck A, Sa lbaum JM, Masters CL, Grzeschik KH, Multhaup G, Beyreuther K, Muller-Hill B ( 1987) The precursor of Alzheimer's disease amyloid A4 protein resembles a cell-surface receptor. Nature 325:733736. Kempermann G, Kuhn HG, Gage FH (1997) Mo re hippocampal neurons in adult mice living in an enriched environment. Nature 386:493-495. King DL, Arendash GW (2002) Behaviora l characterization of the Tg2576 transgenic model of Alzheimer's disease through 19 months. Physiol Behav 75:627-642. Kobayashi S, Ohashi Y, Ando S (2002) Effects of enriched environments with different durations and starting times on learning capacity during aging in rats assessed by a refined procedure of the Hebb-Williams maze task. J Neurosci Res 70:340-346.

PAGE 154

147 Koistinaho M, Lin S, Wu X, Esterman M, Koger D, Hanson J, Higgs R, Liu F, Malkani S, Bales KR, Paul SM (2004) Apolipoprotein E pr omotes astrocyte colocalization and degradat ion of deposited amyloidbeta peptides. Nat Med 10:719-726. Kuro-o M, Matsumura Y, Aizawa H, Kawagu chi H, Suga T, Utsugi T, Ohyama Y, Kurabayashi M, Kaname T, Kume E, Iwasak i H, Iida A, Shiraki-Iida T, Nishikawa S, Nagai R, Nabeshima YI (1997) Muta tion of the mouse klotho gene leads to a syndrome resembling ageing. Nature 390:45-51. Lazarov O, Robinson J, Tang YP, Hairst on IS, Korade-Mirnics Z, Lee VM, Hersh LB, Sapolsky RM, Mirnics K, Sisodia SS (2005) Environmental enrichment reduces Abeta levels and amyloid deposition in transgenic mice. Cell 120:701713. Lemaire M, Barneoud P, Bohme GA, Piot O, Haun F, Roques BP, Blanchard JC (1994) CCK-A and CCK-B receptors enhance olfactory recognition via distinct neuronal pathways. Learn Mem 1:153-164. Lemere CA, Blusztajn JK, Yamaguchi H, Wisniewski T, Saido TC, Selkoe DJ (1996) Sequence of deposition of het erogeneous amyloid beta-peptides and APO E in Down syndrome: implications for initial events in amyloid plaque formation. Neurobiol Dis 3:16-32. Letenneur L, Gilleron V, Commenges D, He lmer C, Orgogozo JM, Dartigues JF (1999) Are sex and educati onal level indepen dent predictors of dementia and

PAGE 155

148 Alzheimer's disease? Incidence data fr om the PAQUID project. J Neurol Neurosurg Psychiatry 66:177-183. Levi O, Jongen-Relo AL, Feldon J, Roses AD, Michaelson DM (2003) ApoE4 impairs hippocampal plasticity isoform-s pecifically and blocks the environmental stimulation of synaptogenesis and memory. Neurobiol Dis 13:273-282. Licastro F, Morini MC, Polazzi E, Davis LJ (1995) Increased serum alpha 1antichymotrypsin in patients with probabl e Alzheimer's disease: an acute phase reactant without the peri pheral acute phase response. J Neuroimmunol 57:71-75. Lieberman J, Schleissner L, Tachik i KH, Kling AS (1995) Serum alpha 1antichymotrypsin level as a marker for Alzheimer-type dementia. Neurobiol Aging 16:747-753. Loewenstein DA, Acevedo A, Czaja SJ, D uara R (2004) Cognit ive rehabilitation of mildly impaired Alzheimer disease pati ents on cholinesterase inhibitors. Am J Geriatr Psychiatry 12:395-402. Ma J, Brewer HB, Jr., Potter H (1996) Al zheimer A beta neurotox icity: promotion by antichymotrypsin, ApoE4; inhibiti on by A beta-related peptides. Neurobiol Aging 17:773-780. Ma J, Yee A, Brewer HB, Jr., Das S, Potter H (1994) Amyloidassociated proteins alpha 1-antichymotrypsin and apolipoprotei n E promote assembly of Alzheimer beta-protein into filam ents. Nature 372:92-94.

PAGE 156

149 Mahley RW, Weisgraber KH, Hussain MM, Greenman B, Fisher M, Vogel T, Gorecki M (1989) Intravenous infusion of apolipoprotein E accelerates clearance of plasma lipoproteins in r abbits. J Clin Invest 83:2125-2130. Mandelkow EM, Mandelkow E (1 998) Tau in Alzheimer's disease. Trends Cell Biol 8:425-427. Manelli AM, Stine WB, Van Eldik LJ LaDu MJ (2004) ApoE and Abeta1-42 interactions: effects of isoform and c onformation on structure and function. J Mol Neurosci 23:235-246. Martel CL, Mackic JB, Mats ubara E, Governale S, Miguel C, Miao W, McComb JG, Frangione B, Ghiso J, Zlokovic BV ( 1997) Isoform-specific effects of apolipoproteins E2, E3, and on cerebral capillar y sequestration and bloodbrain barrier transport of circulating Alzheimer's amyloid beta. J Neurochem 69:1995-2004. Martinez C, Smith JM, Shapiro F, Good RA (1959) Transfer of acquired immunological tolerance of skin homografts in mice joined in parabiosis. Proc Soc Exp Biol Med 102:413-417. Masliah E, Mallory M, Ge N, Alford M, Veinbergs I, Roses AD (1995) Neurodegeneration in the c entral nervous system of apoE-deficient mice. Exp Neurol 136:107-122.

PAGE 157

150 Naslund J, Haroutunian V, Mohs R, Davis KL, Davies P, Greengard P, Buxbaum JD (2000) Correlation between elevated le vels of amyloid beta-peptide in the brain and cognitive decline. Jama 283:1571-1577. Nilsson LN, Bales KR, DiCarlo G, Gor don MN, Morgan D, Paul SM, Potter H (2001) Alpha-1-antichymotr ypsin promotes beta-sheet amyloid plaque deposition in a transgenic mouse model of Alzheime r's disease. J Neurosci 21:1444-1451. Nilsson LN, Arendash GW, Leighty RE, Costa DA, Low MA, Garcia MF, Cracciolo JR, Rojiani A, Wu X, Bales KR, Paul SM, Potter H (2004) Cognitive impairment in PDAPP mice depends on ApoE and ACT-catalyzed amyloid formation. Neurobiol Aging 25:1153-1167. Patrick GN, Zukerberg L, Nikolic M, de la Monte S, Dikkes P, Tsai LH (1999) Conversion of p35 to p25 deregulates Cdk5 activity and promotes neurodegeneration. Na ture 402:615-622. Permanne B, Perez C, Soto C, Frangione B, Wisniewski T (1997) Detection of apolipoprotein E/dimeric soluble amyloid beta complexes in Alzheimer's disease brain supernatants. Biochem Biophys Res Commun 240:715-720. Poirier J (2000) Apolipoprot ein E and Alzheimer's disease. A role in amyloid catabolism. Ann N Y Acad Sci 924:81-90.

PAGE 158

151 Poirier J, Baccichet A, Dea D, Gaut hier S (1993) Cholesterol synthesis and lipoprotein reuptake during synaptic rem odelling in hippocampus in adult rats. Neuroscience 55:81-90. Potter H (1991) Review and hypothesis: Alzheimer disease and Down syndrome-chromosome 21 nondisjunction may underlie both disorders. Am J Hum Genet 48:1192-1200. Qian S, Jiang P, Guan XM, Singh G, Trum bauer ME, Yu H, Chen HY, Van de Ploeg LH, Zheng H (1998) Mutant human pr esenilin 1 protects presenilin 1 null mouse against embryonic lethality and elevates Abeta1-42/43 expression. Neuron 20:611-617. Ramirez-Amaya V, Escobar ML, C hao V, Bermudez-Rattoni F (1999) Synaptogenesis of mossy fibers induced by spatial water maze overtraining. Hippocampus 9:631-636. Robakis NK, Ramakrishna N, Wolfe G, Wisn iewski HM (1987) Molecular cloning and characterization of a cDNA encoding the cerebrovascular and the neuritic plaque amyloid peptides. Proc Natl Acad Sci U S A 84:4190-4194. Roytta M, Peltonen J, V uorinen V (1988) Schwann cells and collagen synthesis in taxol-treated nerve crush. An elec tron microscopic study. Coll Relat Res 8:123-131.

PAGE 159

152 Russo C, Angelini G, Dapino D, Piccini A, Piombo G, Schettini G, Chen S, Teller JK, Zaccheo D, Gambetti P, Tabaton M ( 1998) Opposite roles of apolipoprotein E in normal brains and in Alzheimer's disease. Proc Natl Acad Sci U S A 95:1559815602. Schafe GE, Atkins CM, Swank MW, B auer EP, Sweatt JD, LeDoux JE (2000) Activation of ERK/MAP kinase in the amygdala is required for memory consolidation of pavlovian fear c onditioning. J Neurosci 20:8177-8187. Scheuner D, Eckman C, Jensen M, Song X, Citron M, Suzuki N, Bird TD, Hardy J, Hutton M, Kukull W, Larson E, Le vy-Lahad E, Viitanen M, Peskind E, Poorkaj P, Schellenberg G, Tanzi R, Wasco W, Lannfelt L, Selkoe D, Younkin S (1996) Secreted amyloid beta-protein similar to that in the senile plaques of Alzheimer's disease is increased in vivo by the presenilin 1 and 2 and APP mutations linked to familial Alzheimer's disease. Nat Med 2:864-870. Schwarzman AL, Gregori L, Vitek MP, Ly ubski S, Strittmatter WJ, Enghilde JJ, Bhasin R, Silverman J, Weisgraber KH, Coyle PK, et al. (1994) Transthyretin sequesters amyloid beta protein and prevent s amyloid formation. Proc Natl Acad Sci U S A 91:8368-8372. Silverberg GD, Mayo M, Saul T, Rubenstein E, Mc Guire D (2003) Alzheimer's disease, normal-pressure hydroc ephalus, and senescent changes in CSF circulatory physiology: a hypothes is. Lancet Neurol 2:506-511.

PAGE 160

153 Singer W (1990) The formation of cooper ative cell assemblies in the visual cortex. J Exp Biol 153:177-197. Snowdon DA, Kemper SJ, Mortimer JA, Greiner LH, Wekstein DR, Markesbery WR (1996) Linguistic ability in early life and cognitive function and Alzheimer's disease in late life. Findings fr om the Nun Study. Jama 275:528-532. Stein TD, Anders NJ, DeCarli C, C han SL, Mattson MP, Johnson JA (2004) Neutralization of transthyretin reverses the neuroprotective e ffects of secreted amyloid precursor protein (APP) in APPSW mice resulting in tau phosphorylation and loss of hippocampal neurons: support for the amyloid hypothesis. J Neurosci 24:7707-7717. Stern Y, Gurland B, Tatemichi TK, T ang MX, Wilder D, Mayeux R (1994) Influence of education and occupation on the incidence of Alzheimer's disease. Jama 271:1004-1010. Strittmatter WJ, Roses AD (1996) Apol ipoprotein E and Alzheimer's disease. Annu Rev Neurosci 19:53-77. Strittmatter WJ, Weisgraber KH, Huang DY, Dong LM, Salvesen GS, PericakVance M, Schmechel D, Saunders AM, Goldgaber D, Roses AD (1993) Binding of human apolipoprotein E to synthetic amyloid beta peptide: isoform-specific effects and implications for late-onset Al zheimer disease. Proc Natl Acad Sci U S A 90:8098-8102.

PAGE 161

154 Strittmatter WJ, Weisgraber KH, Goedert M, Saunders AM, Huang D, Corder EH, Dong LM, Jakes R, Alberts MJ Gilbert JR, et al. (1994) Hypothesis: microtubule instability and paired helical filament forma tion in the Alzheimer disease brain are related to apolipoprotein E genotype. Ex p Neurol 125:163-171; discussion 172164. Tanzi RE, Gusella JF, Watkins PC, Bruns GA, St George-Hyslop P, Van Keuren ML, Patterson D, Pagan S, Kurnit DM, Neve RL (1987) Amyloid beta protein gene: cDNA, mRNA distribution, and genetic linkage near the Alzheimer locus. Science 235:880-884. Terai K, Iwai A, Kawabata S, Sasama ta M, Miyata K, Yamaguchi T (2001) Apolipoprotein E deposition and astrogliosi s are associated with maturation of beta-amyloid plaques in betaAPPswe tr ansgenic mouse: Implications for the pathogenesis of Alzheimer's disease. Brain Res 900:48-56. Thorngate FE, Strockbine PA, Erickson SK, Williams DL (2002) Altered adrenal gland cholesterol metabolism in the apoEdeficient mouse. J Lipid Res 43:19201926. Tsuzuki K, Fukatsu R, Yamaguchi H, Ta teno M, Imai K, Fu jii N, Yamauchi T (2000) Transthyretin binds amyloid bet a peptides, Abeta1-42 and Abeta1-40 to form complex in the autopsied human kidney possible role of transthyretin for abeta sequestration. N eurosci Lett 281:171-174.

PAGE 162

155 Tsuzuki K, Fukatsu R, Hayashi Y, Yoshid a T, Sasaki N, Takamaru Y, Yamaguchi H, Tateno M, Fujii N, Tak ahata N (1997) Amyloid beta pr otein and transthyretin, sequestrating protein colocalize in no rmal human kidney. Neurosci Lett 222:163166. Turner CA, Lewis MH, King MA (2003) Environmental enrichment: effects on stereotyped behavior and dendritic mor phology. Dev Psychobiol 43:20-27. Ueki A, Kawano M, Namba Y, Kawakami M, Ikeda K (1993) A high frequency of apolipoprotein E4 isoprotein in Japanes e patients with late-onset nonfamilial Alzheimer's disease. Neurosci Lett 163:166-168. Wallace CS, Withers GS, Weiler IJ, George JM, Cla yton DF, Greenough WT (1995) Correspondence between sites of NGFI-A induction and sites of morphological plasticity following exposu re to environmental complexity. Brain Res Mol Brain Res 32:211-220. Weisgraber KH, Innerarity TL, Mahley RW (1982) Abnormal lipoprotein receptorbinding activity of t he human E apoprotein due to cysteine-arginine interchange at a single site. J Biol Chem 257:2518-2521. Wenisch E, Stoker A, Bourre llis C, Pasquet C, Gauthier E, Corcos E, Banchi MT, De Rotrou J, Rigaud AS (2005) [A global intervention program for institutionalized demented patient s]. Rev Neurol (Paris) 161:290-298.

PAGE 163

156 Wisniewski T, Castano EM, Golabek A, Vogel T, Frangione B (1994) Acceleration of Alzheimer's fibril forma tion by apolipoprotein E in vitro. Am J Pathol 145:1030-1035. Xu PT, Gilbert JR, Qiu HL, Ervin J, Roth rock-Christian TR, Hulette C, Schmechel DE (1999) Specific regional transcripti on of apolipoprotein E in human brain neurons. Am J Pat hol 154:601-611. Young D, Lawlor PA, Leone P, Dragunow M, During MJ (1999) Environmental enrichment inhibits spontaneous apoptos is, prevents seizures and is neuroprotective. Nat Med 5:448-453. Zhang HT, Zhao Y, Huang Y, Dorairaj NR, Chandler LJ, O'Donnell JM (2004) Inhibition of the phosphodiesterase 4 (PDE 4) enzyme reverses memory deficits produced by infusion of the MEK inhibito r U0126 into the CA1 subregion of the rat hippocampus. Neuropsychopharmacology 29:1432-1439. Zhou Z, Smith JD, Greengard P, Gandy S ( 1996) Alzheimer amyloid-beta peptide forms denaturant-resistant complex with type epsilon 3 but not type epsilon 4 isoform of native apolipoprot ein E. Mol Med 2:175-180.

PAGE 164

157 APPENDIX A Source Code for High Output Morphom etry Examination Routine (HOMER) Form: IPPMACRO.f rm................................................................................................158 IpUtil32.ba s.......................................................................................................182 ModBrowse. bas................................................................................................183 ModFileO. BAS..................................................................................................185 DCmodAPI.r tf...................................................................................................187

PAGE 165

158 APPENDIX A (Continued) 'IPPMACRO.frm HOMER Version 1.0 DAVID COSTA 2003 Option Explicit 'File Management Variables Dim vstrBackgroundImage As String Dim vstrSegmentatio nImage As String Dim vstrSegmentationFile As String Dim vstrExcelFile As String Dim vstrTargetImageDirectory As String Dim vstrTargetImage As String 255 Dim vstrtxtIcComment As String Dim vintRangeMin As Integer Dim vlngRangeMax As Long Dim vsngExcelArray(10) As Single 'Dim vintTweakChoice As Integer Dim vintExcelRow As Integer 'index for r on number for Excel data export Dim vintDirectoryIndex As Integer 'in cmdCalculate loop index for IpStSearchDir() Dim vintDirectoryStatus As Integer 'in cmdCalculate return value for IpStSearchDir() 'Excel Variables Dim xlapp As Object Dim xlbook As Object Dim xlSheet As Object Dim xlRange As Object Dim n As Integer 'Index for excel export of data Dim i As Integer Dim j As Integer Dim xlData() As Single Dim xlLabels() As String Dim dInfo As IPDOCINFO 'InitExcel() initializes the OLE lin k to Excel and retu rns the Object 'in the xlApplication parameter. Retu rn value: 0 = success; -1 = failure Function InitExcel(ByRef xlApplic ation As Object) As Integer

PAGE 166

159 APPENDIX A (Continued) InitExcel = 0 Make sure the function is initialized On Error Resume Next continue if an error is generated Err.Clear clear the error flag Set xlApplication = Ge tObject(, "Excel.Application") GetObject will return an error if Excel isn't open If Err.Number <> 0 Then Start Excel with CreateObject. If this fails, we exit the macro. Err.Clear Set xlApplication = CreateObject("Excel.Application") If Er r.Number <> 0 Then MsgBox "Can't find Excel.", vbOKOnly + vbCritical, "OLE Error" InitExcel = -1 Exit Function End If End If 'show Excel (don't run in background) xlApplication.Visible = True If Err.Number <> 0 Then MsgBox "Error showing Excel!", vbOKOnly + vbCritical, "OLE Error" InitExcel = -2 End If End Function Opens Excel spreadsheet and defines rows and columns for data input Function fncOpenExcel() 'startup Excel for OLE data dump If InitExcel(xlapp) < 0 Then Exit Function End If 'create a new workbook for your data Set xlbook = xlapp.Workbooks.Add xlbook.Activate

PAGE 167

160 APPENDIX A (Continued) Set xlSheet = xlbook.ActiveSheet xlSheet.Activate xlapp.Visible = True xlapp.WindowState = vbMaximized 'Note: due to the design of OLE, it is much faster to transfer data to Excel in a 2 dimensional array, ev en if you are only filling one cell ReDim xlLabels(1 To 1, 1 To 3) As String '(row,column) 'Adds Summary sheet to excel file xlapp.worksheets.Add Set xlSheet = xlbook.ActiveSheet xlSheet.Activate xlSheet.N ame = "Result Summary" 'Adds AOI sheet to ex cel file if area is checked If chkArea.value = vbChecked Then xlapp.worksheets.Add Set xlS heet = xlbook.ActiveSheet xlSheet.Activate xlSheet.Name = "AOI" 'Labels and Formats "AOI" Sheet With xlapp .ActiveWindow.Zoom = 75 .Range("A1").Select .Activ eCell.FormulaR1C1 = "Image" .Range("B1").Select .Act iveCell.FormulaR1C1 = "Sum" .Range("C1").Select .ActiveCell.FormulaR1C1 = "Std. Deviation" .Range("D1").Select .ActiveCell.FormulaR1C 1 = "Num. of Objects" .Range("E1").Select .Activ eCell.FormulaR1C1 = "Comments" '.Columns("E:E").Select '. selection.ColumnWidth = 50 .Rows("1:1").Select With .Selection

PAGE 168

161 APPENDIX A (Continued) .Font.Underline = 2 'xlUnderlineStyleSingle .Font.Bold = True .ColumnWidth = 22 .H orizontalAlignment = 3 'xlCenter .VerticalAlignment = xlBottom .WrapText = False .Orientation = 0 .AddIndent = False .IndentLevel = 0 .ShrinkToFit = False .ReadingOrder = xlContext .MergeCells = False End With .Range("C6").Select '.Columns( "E:E").ColumnWidth = 23.57 End With End If 'Adds % Area sheet to ex cel file if area is checked If chkArea.value = vbChecked Then xlapp.worksheets.Add Set xlS heet = xlbook.ActiveSheet xlSheet.Activate xlSheet.Name = "% Area" 'Labels and Formats "% Area" Sheet With xlapp .ActiveWindow.Zoom = 75 .Range("A1").Select .Activ eCell.FormulaR1C1 = "Image" .Range("B1").Select .ActiveC ell.FormulaR1C1 = "Mean Value" .Range("C1").Select .ActiveCell.FormulaR1C1 = "Std. Deviation" .Range("D1").Select .ActiveCell.FormulaR1C1 = "Min. Measurment" .Range("E1").Select .ActiveCell.FormulaR1C1 = "Maximum Measurement" .Range("F1").Select .Act iveCell.FormulaR1C1 = "Range" .Range("G1").Select

PAGE 169

162 APPENDIX A (Continued) .Act iveCell.FormulaR1C1 = "Sum" .Range("H1").Select .ActiveCell.FormulaR1C 1 = "Num. of Objects" .Rows("1:1").Select With .Selection .Fon t.Underline = 2 'xlUnderlineStyleSingle .Font.Bold = True .ColumnWidth = 22 .H orizontalAlignment = 3 'xlCenter .VerticalAlignment = xlBottom .WrapText = False .Orientation = 0 .AddIndent = False .IndentLevel = 0 .ShrinkToFit = False .ReadingOrder = xlContext .MergeCells = False End With .Range("C6").Select .Columns(" E:E").ColumnWidth = 23.57 End With End If 'Adds Density sheet to ex cel file if Density is checked If chkDensity.value = vbChecked Then xlapp.worksheets.Add Set xlS heet = xlbook.ActiveSheet xlSheet.Activate xlSheet.Name = "Density" With xlapp .ActiveWindow.Zoom = 75 .Range("A1").Select .Activ eCell.FormulaR1C1 = "Image" .Range("B1").Select .ActiveCell.FormulaR1C1 = "Density Mean: Mean" .Range("C1").Select .ActiveCell.FormulaR1C1 = "Density Mean: Std.Dev" .Range("D1").Select

PAGE 170

163 APPENDIX A (Continued) .ActiveCell.FormulaR1C1 = "Density Mean: Sum" .Range("F1").Select .ActiveCell.FormulaR1C1 = "Density Sum: Mean" .Range("G1").Select .ActiveCell.FormulaR1C1 = "Density Sum: Std.Dev" .Range("H1").Select .ActiveCell.FormulaR1C1 = "Density Sum: Sum" .Rows("1:1").Select With .Selection .Fon t.Underline = 2 'xlUnderlineStyleSingle .Font.Bold = True .ColumnWidth = 22 .H orizontalAlignment = 3 'xlCenter .VerticalAlignment = xlBottom .WrapText = False .Orientation = 0 .AddIndent = False .IndentLevel = 0 .ShrinkToFit = False .ReadingOrder = xlContext .MergeCells = False End With .Columns("E:E").Select .S election.ColumnWidth = 5 End With End If 'Delete Unwanted sheets With xlapp .Sheet s("Sheet1").Select .SendKeys "Y" .ActiveWi ndow.SelectedShe ets.Delete .Sheet s("Sheet2").Select .ActiveWi ndow.SelectedShe ets.Delete '.S heets("Sheet3").Select '.ActiveWi ndow.SelectedShe ets.Delete

PAGE 171

164 APPENDIX A (Continued) 'End With 'Label the columns xlLabels(1, 1) = "Data 1" xlLabels(1, 2) = "Data 2" xlLabels(1, 3) = "Data 3" Set xlSheet = Nothing Set xlbook = Nothing Set xlapp = Nothing End Function Function fncExcelDataExport(vstrSheetName As String, vintYindex As Integer) Select Case vstrSheetName Case "% Area" 'declare n as index integer n = 0 'Get statistics fr om current AOI and send to Array ret = IpBlbGet(GETSTAT S, 0, BLBM_AREA, vsngExcelArray(n)) 'make % area active sheet and sends area data to excel xlapp.Sheet s(vstrSheetName).Select Set xlSheet = xlbook.ActiveSheet 'Mean, Std.Dev,Min, Max, Range, Sum For n = 0 To 5 With xlSheet .Range(.cells(vin tYindex, 2 + n), .cells(vin tYindex, 2 + n)).value = vsngExcelArray(n) End With Next n '# of Objects n = n + 2 With xlSheet .Range(.cells(vintYindex, n), .cells(vintYindex, n)).value = vsngExcelArray(n)

PAGE 172

165 APPENDIX A (Continued) End With n = 0 'nam e of current file to excel With xlSheet .Range(.cells(vintYindex, 1), .cells(vintYindex, 1)).value = vstrTargetImage End With Case "AOI" n = 0 'Get statistics fr om current AOI and send to Array ret = IpBlbGet(GETSTAT S, 0, BLBM_AREA, vsngExcelArray(n)) 'make AOI active sheet and sends area data to excel xlapp.Sheet s(vstrSheetName).Select Set xlS heet = xlbook.ActiveSheet 'Sum With xlSheet .Range(.cells(vintYindex, 2), .c ells(vintYindex, 2)).value = vsngExcelArray(5) End With 'Std.Dev With xlSheet .Range(.cells(vintYindex, 3), .c ells(vintYindex, 3)).value = vsngExcelArray(1) End With '# of Objects With xlSheet .Range(.cells(vintYindex, 4), .c ells(vintYindex, 4)).value = vsngExcelArray(8) End With If txtIcComment.text <> "" Then vstrtxtIcComment = txtIcComment.text With xlSheet .Range(.cells(vintYindex, 5), .cells(vintYindex, 5)).value = vstrtxtIcComment End With

PAGE 173

166 APPENDIX A (Continued) End If n = 0 'name of current file to excel With xlSheet .Range(.cells(vintYindex, 1), .c ells(vintYindex, 1)).value = vstrTargetImage End With Case "Density" n = 0 'Get statistics fr om current AOI and send to Array ret = IpBl bGet(GETSTATS, 0, BLBM_D ENSITY, vsngExcelArray(n)) 'make % area active sheet and sends area data to excel xlapp.Sheet s(vstrSheetName).Select Set xlS heet = xlbook.ActiveSheet 'Mean, Std. Dev For n = 0 To 1 With xlSheet .Range(.cells(vintY index, 2 + n), .cells(vintY index, 2 + n)).value = vsngExcelArray(n) End With Next n With xlSheet .Range(.cells(vintYindex, 4), .c ells(vintYindex, 4)).value = vsngExcelArray(5) End With n = 0 'Get statistics fr om current AOI and send to Array ret = IpBl bGet(GETSTATS, 0, BLBM_D ENSSUM, vsngExcelArray(n)) 'make % area active sheet and sends area data to excel xlapp.Sheet s(vstrSheetName).Select

PAGE 174

167 APPENDIX A (Continued) Set xlS heet = xlbook.ActiveSheet 'Mean, Std. Dev For n = 0 To 1 With xlSheet .Range(.cells(vintY index, 6 + n), .cells(vintY index, 6 + n)).value = vsngExcelArray(n) End With Next n With xlSheet .Range(.cells(vintYindex, 8), .c ells(vintYindex, 8)).value = vsngExcelArray(5) End With End Select End Function 'Determines percent area of segmented portions of image Function fncPercentArea() 'loads segmentation file ret = IpSegLoad( vstrSegmentationFile) ret = IpSegShow(0) ret = IpSegSet Attr(SETCURSEL, 0) ret = IpSegSetAttr(Channel, 0) ret = IpS egPreview(ALL_C_T) ret = IpSegShow(0) ret = IpBlbShow(1) 'selects manual autorange ret = IpBlbSetA ttr(BLOB_AUTORANGE, 0) 'Green Fill Color ret = IpBlbSetAtt r(BLOB_OUTLINECOLOR, 3) 'Style = Filled ret = IpBlbSetAtt r(BLOB_OUTLINEMODE, 3) 'Set Label Mode to NONE ret = IpBlbSetA ttr(BLOB_LABELMODE, 0) 'Sets Cleanborder Mode

PAGE 175

168 APPENDIX A (Continued) ret = IpBlbSetAtt r(BLOB_CLEANBORDER, 0) If chkApplyRanges.value = vbChecked Then vlngRangeM ax = txtSizeMax.text vintRangeM in = txtSizeMin.text 'set filter ranges ret = IpBl bSetFilterRange(BLBM_AREA, vintRangeMin, vlngRangeMax) 'apply filter ranges ret = IpBlbSet Attr(BLOB_FILTEROBJECTS, 1) Else vlngRangeMax = 10000 vintRangeMin = 10 'set filter ranges ret = IpBl bSetFilterRange(BLBM_AREA, vintRangeMin, vlngRangeMax) 'apply filter ranges ret = IpBlbSet Attr(BLOB_FILTEROBJECTS, 1) End If ret = IpBlbCount() ret = IpBlbUpdate(0) End Function 'Determines Integrated Optical Density Function fncDensity() End Function 'Determines total area of chosen AOI Function fncAOI() 'set large filter range vlngRangeMax = 10000000 vintRangeMin = 0 'selects manual autorange ret = IpBlbSetA ttr(BLOB_AUTORANGE, 0) 'Green Fill Color

PAGE 176

169 APPENDIX A (Continued) ret = IpBlbSetAtt r(BLOB_OUTLINECOLOR, 3) 'Style = Filled ret = IpBlbSetAtt r(BLOB_OUTLINEMODE, 3) 'Set Label Mode to NONE ret = IpBlbSetA ttr(BLOB_LABELMODE, 0) 'Sets Cleanborder Mode ret = IpBlbSetAtt r(BLOB_CLEANBORDER, 0) 'set filter ranges ret = IpBlbSetFilterRange(BL BM_AREA, vintRange Min, vlngRangeMax) 'apply filter ranges ret = IpBlbSetAtt r(BLOB_FILTEROBJECTS, 1) 'Set segmentation to Max ret = IpBlbShow(1) ret = IpSegShow(1) ret = IpSegSet Attr(SETCURSEL, 0) ret = IpSegSetAttr(Channel, 0) ret = IpS egPreview(ALL_C_T) ret = IpSegShow(2) ret = IpSegSet Attr(SETCURSEL, 0) ret = IpS egPreview(ALL_C_T) ret = IpSegShow(1) ret = IpSegSet Attr(SETCURSEL, 0) ret = IpSegSetAttr(Channel, 0) ret = IpS egPreview(ALL_C_T) ret = IpSegShow(2) ret = IpSegSet Attr(SETCURSEL, 0) ret = IpS egPreview(ALL_C_T) ret = IpSegShow(0) '1 ret = IpSegSet Attr(SETCURSEL, 0) ret = IpSegSetAttr(Channel, 0) ret = IpS egPreview(ALL_C_T) ret = IpSegS etRange(0, 0, 255) ret = IpS egPreview(ALL_C_T) ret = IpSegSetAttr(Channel, 1) ret = IpSegS etRange(1, 0, 255) ret = IpS egPreview(ALL_C_T) ret = IpSegSetAttr(Channel, 2) ret = IpSegS etRange(2, 0, 255) ret = IpS egPreview(ALL_C_T)

PAGE 177

170 APPENDIX A (Continued) ret = IpSegShow(0) ret = IpBlbCount() ret = IpBlbUpdate(0) ret = fncExcelData Export("AOI", vintExcelRow) End Function 'measurment function that calls Area, Density, and AOI functions Function fncMeasure() 'capture AOI ret = IpMacroStop("Please select Area of Interest", 0) 'open count/size dialog ret = IpBlbShow(1) 'Percent Area If chkArea.value = vbChecked Then fncPercentArea End If End Function Private Sub Capture_Click() Dim y As Integer y = 1 ret = IpAcqS how(ACQ_SNAP, 1) ret = IpAcqS how(ACQ_SNAP, 3) ret = IpAcqShow(ACQ_LIVE, 1) Load frmAutoName frmAutoName.Show

PAGE 178

171 APPENDIX A (Continued) End Sub 'MAIN SUB Private Sub cmdCalculate_Click() vintExcelRow = 3 'These 3 vars. are initialized here. vintDirectoryIndex = 0 'For Ip StSearch function Load image index for Dir vintDirectoryStatus = 1 'For IpStSearch function if 1, images still in Dir vstrSegmentationFile = txtSegmentationFile.text 's egmentation file path to variable '------------------------------Input Checks 'Checks to see if user enter ed file path in background text box If txtBackgroundImage = "" Then MsgBox ("Pl ease Enter Background Image") Exit Sub End If If txtSegment ationImage = "" Then MsgBox ("Plea se Enter Segmentation Image") Exit Sub End If If txtSegmentationFile = "" Then MsgBox ("Please Ente r or Create New Segmentation File") Exit Sub End If If txtExcelFile = "" Then MsgBox ("Please Enter Name of Excel Spreadsheet") Exit Sub End If If txtTargetIm ageDirectory = "" Then MsgBox ("Please Enter Target Image Directory") Exit Sub End If '------------------------------End Input Checks

PAGE 179

172 APPENDIX A (Continued) cmdCalculate.Visible = False lblCurrentImageLabel.Visible = True lblCurrentImage.Visible = True IPPMACRO.Height = 6855 'sends targetimagedir text box to string variable vstrTargetImageDirectory = txtTargetImageDirectory.text 'sends background text box to st ring variable and loads it using IPbasic function vstrBackgroundImage = txtBackgroundImage.text ret = IpWsLoad(vstrB ackgroundImage, "tif") ret = IpBlbEnabl eMeas(BLBM_DENSITY, 1) ret = IpBlbEnabl eMeas(BLBM_DENSSUM, 1) ret = IpBlbEn ableMeas(BLBM_AREA, 1) fncLoadDir If vintDirectoryStatus = 1 Then fncMeasure End If End Sub Function fncLoadDir() vintDirectoryStatus = IpStSear chDir(vstrTargetImageDirectory, "*.tif", vint DirectoryIndex, vstrTargetImage) If vintDirectoryStatus <> 1 Then ret = IpMacroStop("No More Images in Directory", MS_STOP) cmdIcContinue.Enabled = False cmdIcContin ue.Font.Strikethrough = True Exit Function End If lblCurrentImage.ca ption = vstrTargetImage 'sends background text box to st ring variable and loads it using IPbasic function

PAGE 180

173 APPENDIX A (Continued) vstrBackgroundImage = txtBackgroundImage.text ret = IpWsLoad(vstrB ackgroundImage, "tif") 'loads target image to IPP ret = IpWsLoad(vst rTargetImage, "tif") 'Background Correction ret = IpOpShow(3) ret = IpOpBkgndSubtract(0, 0) ret = IpOpShow(2) 'Apply Filters if Applicable 'Index (1) is for HighGauss If optFilter.Item(1) Then ret = IpFltConv olveKernel("HIGAUSS.7x7", 10, 1) 'HiGauss Filter Size:5x5; Strength:100%; Passes:1 End If 'Index (2) is for Lar's Amyloid Macro Filters If optFilter.Item(2) Then ret = IpFltFlatten(0, 20) ret = IpWsConvertToGray() ret = IpFltConv olveKernel("HIGAUSS.7x7", 2, 1) 'HiGauss Filter Size:5x5; Strength:100%; Passes:1 ret = Ip HstEqualize(EQ_BESTFIT) ret = IpS egLoad(vstrSegmentationFile) Else 'loads segmentation file ret = IpS egLoad(vstrSegmentationFile) ret = IpSegShow(0) ret = Ip SegSetAttr(SETCURSEL, 0) ret = IpSegSetAttr(Channel, 0) ret = IpSegPreview(ALL_C_T) ret = IpSegShow(0) ret = IpBlbShow(1) End If

PAGE 181

174 APPENDIX A (Continued) End Function 'Allows user to create segm entation file from segmen tation image if non existant Private Sub cmdCreateS egmentationFile_Click() ret = IpAppCloseAll 'Checks to see if user enter ed file path in background text box If txtBackgroundImage = "" Then MsgBox ("Pl ease Enter Background Image") Exit Sub End If 'Checks to see if user enter ed file path in background text box If txtSegment ationImage = "" Then MsgBox ("Plea se Enter Segmentation Image") Exit Sub End If If MsgBox("Ar e you using Lars-HiGauss Filter Option?", vbYesNoCancel, "Create Sementation File") = vbYes Then optFilter(2) = True End If 'Load Background Image vstrBackgroundImage = txtBackgroundImage.text ret = IpWsLoad(vstrB ackgroundImage, "tif") 'Load Segmentation Image vstrSegmentationImage = txtSegmentationImage.text ret = IpWsLoad(vstrS egmentationImage, "tif") 'Background Subtraction ret = IpOpShow(3) ret = IpOpBkgndSubtract(0, 0) ret = IpOpShow(2) If optFilter.Item(2) Then ret = IpBlbShow(1) ret = IpFltFlatten(0, 20) ret = IpBlbShow(0) ret = IpWsConvertImage( IMC_GRAY, CONV_SCALE 0, 0, 0, 0)

PAGE 182

175 APPENDIX A (Continued) ret = IpFltShow(1) ret = IpFltConvol veKernel("HIGAUSS.7x7", 2, 1) ret = IpFltShow(0) ret = Ip HstEqualize(EQ_BESTFIT) ret = IpBlbShow(1) ret = IpSegShow(1) ret = Ip SegSetAttr(SETCURSEL, 0) ret = IpSegSetAttr(Channel, 0) ret = Ip SegPreview(CURRENT_C_T) ret = IpMacr oStop("Adjust Segmentation and Press Continue to Save Segmentation File", 0) With CommonDialog1 .FileName = "" .Filter = "Color Segmentation Files (*.rge)|*.rge|All Files|*.*" .FilterIndex = 1 .ShowSave vstr SegmentationFile = .FileName txtS egmentationFile = .FileName End With Else ret = IpSegShow(1) ret = Ip SegSetAttr(SETCURSEL, 0) ret = IpSegSetAttr(Channel, 0) ret = IpSegPreview(ALL_C_T) ret = IpMacroStop("Adj ust Segmentation and Press OK to Save Segmentation File", 0) With CommonDialog1 .FileName = "" .Filter = "Color Segmentation Files (*.rge)|*.rge|All Files|*.*" .FilterIndex = 1 .ShowSave vstr SegmentationFile = .FileName txtS egmentationFile = .FileName End With

PAGE 183

176 APPENDIX A (Continued) End If ret = IpSegSave( vstrSegmentationFile, 0) ret = IpSegShow(0) ret = IpAppCloseAll End Sub Private Sub cmdExit_Click() Unload Me End Sub Private Sub cmdIcContinue_Click() ret = fncExcelDataExport("% Area", vintExcelRow) 'Density Measurement If chkDensity.value = vbChecked Then fncDensity ret = fncExcelDa taExport("Density", vintExcelRow) End If 'Area of Interest Total Area fncAOI 'index for directory image vintDirectoryIndex = vintDirectoryIndex + 1 'index for Excel Row vintExcelRow = vintExcelRow + 1 'Reset Comment Box txtIcComment.text = "" ret = IpDocClose() fncLoadDir

PAGE 184

177 APPENDIX A (Continued) If vintDirectoryStatus = 1 Then fncMeasure Else ret = IpAppCloseAll() End If End Sub Private Sub cmdIcDeselectItem_Click() ret = IpAoiShow(FRAME_NONE) ret = IpTemplateMode(1) ret = IpBlbHi deObject(0, 0, 0) ret = IpTemplateMode(0) ret = IpAoiShow(FRAME_IRREGULAR) End Sub Private Sub cmdIcMod ifySegmentation_Click() ret = IpSegShow(1) ret = IpSegShow(1) ret = IpSegSet Attr(SETCURSEL, 0) ret = IpSegSetAttr(Channel, 0) ret = IpS egPreview(ALL_C_T) ret = IpSegShow(2) ret = IpSegSet Attr(SETCURSEL, 0) ret = IpS egPreview(ALL_C_T) ret = IpSegShow(1) ret = IpSegSet Attr(SETCURSEL, 0) ret = IpSegSetAttr(Channel, 0) ret = IpS egPreview(ALL_C_T) ret = IpSegShow(1) ret = IpSegSet Attr(SETCURSEL, 0) ret = IpSegSetAttr(Channel, 0) ret = IpS egPreview(ALL_C_T) ret = IpMacroStop("Adjus t Segmentation and Press OK", 0) ret = IpSegShow(0) ret = IpBlbCount()

PAGE 185

178 APPENDIX A (Continued) ret = IpBlbUpdate(0) End Sub 'Opens CommonDialog1 "Open F ile" dialog and sends file to appropriate textbox Private Sub cmdOpenBackground_Click() With CommonDialog1 .Filter = "TIF|*.ti f|JPEG (.jpg)|*.jpg|All Files|*.*" .FilterIndex = 1 .ShowOpen txtBa ckgroundImage = .FileName End With End Sub 'Opens CommonDialog1 "Open F ile" dialog and sends file to appropriate textbox Private Sub cmdOpenExcel_Click() With CommonDialog1 .Filter = "Excel Spreadsheet File (. xls)|*.xls|All Files|*.*" .FilterIndex = 1 .ShowOpen txtE xcelFile = .FileName End With End Sub 'Opens "Browse Directory Structure" dial og and allows one to choose directory and 'sends chosen directory to textbox Private Sub cmdOpenImageDirectory_Click() Dim udtBrowseInfo As BROWSEINFO Dim lRet As Long Dim lPathID As Long Dim sPath As String

PAGE 186

179 APPENDIX A (Continued) Dim nNullPos As Integer txtTargetImageDirectory.SetFocus 'Specify the window handle fo r the owner of the dialog box udtBrowseInfo.hOwner = Me.hwnd 'Specify the root to start browsing from; 'if null, My Co mputer is the root udtBrowseInfo.pidlRoot = 0& 'Specify a title. This is not the caption of the dialog. Useful for 'adding any kind of additio nal information or instructions udtBrowseInfo.lpszT itle = "Select a folder" 'Specify any flags; See Declarations section udtBrowseInfo.ulFl ags = BIF_RETURNONLYFSDIRS 'Call the function. 'The return value is a point er to an item identifier list that 'specifies the locati on of the selected folder. 'If the user cancels the di alog box, the return value is 0. lPathID = SHBrowseForFolder(udtBrowseInfo) sPath = Space$(512) lRet = SHGetPathFro mIDList(lPathID, sPath) If lRet Then nNullPos = InStr(sPath, vbNullChar) txtTargetImageDirecto ry = Left(sPath, nNullPos 1) End If End Sub 'Opens CommonDialog1 "Open F ile" dialog and sends file to appropriate textbox Private Sub cmdOpenSegm entationFile_Click() With CommonDialog1 .Filter = "C olor Segmentation Files (*.r ge)|*.rge|All Files|*.*" .FilterIndex = 1

PAGE 187

180 APPENDIX A (Continued) .ShowOpen txtSegment ationFile = .FileName End With End Sub 'Opens CommonDialog1 "Open F ile" dialog and sends file to appropriate textbox Private Sub cmdOpenSegm entationImage_Click() With CommonDialog1 .Filter = "TIF|*.tif| JPEG (.jpg)|*.jpg|All Files|*.*" .FilterIndex = 1 .ShowOpen txtSegment ationImage = .FileName End With End Sub Private Sub Exit_Click() Unload Me End Sub Private Sub Form_Load() 'Both "WinExec" and "S etWindowPos" require API calls The real WinAPI file is too 'large to be included in a VB executable, so a m odified API file, DCmodAPI.rtf, is 'used instead. 'Loads IPP on start of program ret = WinExec("c :\ipwin4\ipwin32.exe", 1) 'keeps VB controls on top of IPP '-1 = HWND_TOPMOST '&H2 = SWP_NOMOVE '&H1 = SWP_NOSIZE '4 Numbers betw een are screen coordinates

PAGE 188

181 APPENDIX A (Continued) ret = SetWindowPos(IPPMACRO.hw nd, -1, 0, 0, 0, 0, &H2 + &H1) ret = IpAppCloseAll chkArea.value = vbChecked IPPMACRO.Height = 5840 'opens excel using user defined function fncOpenExcel End Sub

PAGE 189

182 APPENDIX A (Continued) 'IpUtil32.bas Function iprint(mystring As Variant) Replaces the pr int command in IP-Basic v. 3.0. Sends the stri ng argument to the output window. IpOutput (myst ring + Chr$(13) + Chr$(10)) End Function Function IpTrim(mystring As String) As String Replaces RTri m$ command in IP-Basic v.3.0. The old command used to trim zeros as well as spaces. Dim Index% Index = InStr(mystring, Chr$(0)) If Index > 0 T hen IpTrim = Trim$(Left$(mystring, Index 1)) Else IpTrim = Trim$(mystring) End Function Function IpDocActive() As Integer Returns the doc ument Id of the active image Returns -1 if no image is displayed. Dim docid As Integer ret = IpDo cGet(GETACTDOC, 0, docid) If ret < 0 Then IpDocA ctive = -1 Else IpDocActive = docid End Function

PAGE 190

183 APPENDIX A (Continued) 'ModBrowse.bas Option Explicit 'Function Declarations, type struct ure, and constants to use the 'Browse for Folder dialog box. For more information on these, 'consult the SDK, included with VB 4.0 Pro or Ent editions as 'part of the MSDN/VB Starter Kit. Public Declare Function SHGetPat hFromIDList Lib "shell32.dll" Alias "SHGetPathFromIDListA" (ByVal pidl As Long, ByVal pszPath As String) As Long Public Type BROWSEINFO hOwner As Long pidlRoot As Long pszDisplayName As String lpszTitle As String ulFlags As Long lpfn As Long lParam As Long iImage As Long End Type Public Declare Function SHBrowse ForFolder Lib "shell32.dll" Alias "SHBrowseForFolderA" (lpB rowseInfo As BROWSEINFO) As Long 'Below are the constants which can be specified in the ulFlags member 'of the BROWSEINFO structure. 'Only returns file system directorie s. If the user selects folders 'that are not part of the file system, the OK button is grayed. Public Const BIF_RETURNONLYFSDIRS = &H1

PAGE 191

184 APPENDIX A (Continued) 'Does not include network folders below the domain level in the 'tree view control. Public Const BIF_DONTGOBELOWDOMAIN = &H2 'Only returns file system ancestors If the user selects anything 'other than a file system ancesto r, the OK button is grayed. Public Const BIF_RETURNFSANCESTORS = &H8 'Only returns computers. If the user selects anything other than 'a computer, the OK button is grayed. Public Const BIF_BROWSEFORCOMPUTER = &H1000 'Only returns printers. If the user selects anything other than 'a printer, the OK button is grayed. Public Const BIF_BROWSEFORPRINTER = &H2000 'Includes a status area in the dialog box The callback function can set the status text by sending messages to the dialog box. Const BIF_STATUSTEXT = &H4

PAGE 192

185 APPENDIX A (Continued) 'ModFileO.BAS Option Explicit 'Function Declarations, type structure, and constants for the Open and Save 'common dialog boxes. For more in formation on these, consult the SDK, 'included with VB 4.0 Pro or Ent editi ons as part of t he MSDN/VB Starter 'Kit. Public Type OPENFILENAME lStructSize As Long hwndOwner As Long hInstance As Long lpstrFilter As String lpstrCustomFilter As String nMaxCust Filter As Long nFilterIndex As Long lpstrFile As String nMaxFile As Long lpstrFileTitle As String nMaxFileTitle As Long lpstrInitialDir As String lpstrTitle As String Flags As Long nFileOffset As Integer nFileExtension As Integer lpstrDefExt As String lCustData As Long lpfnHook As Long lpTemplateName As String End Type 'Functions and constants for the common dialog boxes

PAGE 193

186 APPENDIX A (Continued) Public Declare Function GetOpen FileName Lib "comdlg32.dll" Alias "GetOpenFileNameA" (pOPENFILE NAME As OPENFILENAME) As Long Public Declare Function GetSaveFileName Lib "comdlg32.dll" Alias "GetSaveFileNameA" (pOPENFI LENAME As OPENFILENAME) As Long 'These constants must be declared sinc e the Common Dialog control is not 'part of the project and, therefore, t he intrinsic constants are not 'defined. Public Const OFN_READONLY = &H1 Public Const OFN_OVERWRITEPROMPT = &H2 Public Const OFN_HIDEREADONLY = &H4 Public Const OFN_NOCHANGEDIR = &H8 Public Const OFN_SHOWHELP = &H10 Public Const OFN_ENABLEHOOK = &H20 Public Const OFN_ENABLETEMPLATE = &H40 Public Const OFN_ENABLETEMPLATEHANDLE = &H80 Public Const OFN_NOVALIDATE = &H100 Public Const OFN_ALLOWMULTISELECT = &H200 Public Const OFN_EXTENSIONDIFFERENT = &H400 Public Const OFN_PATHMUSTEXIST = &H800 Public Const OFN_FILEMUSTEXIST = &H1000 Public Const OFN_CREATEPROMPT = &H2000 Public Const OFN_SHAREAWARE = &H4000 Public Const OFN_NOREADONLYRETURN = &H8000 Public Const OFN_NOTESTFILECREATE = &H10000 Public Const OFN_NONETWORKBUTTON = &H20000 Public Const OFN_NOLONGNAMES = &H40000 force no long names for 4.x modules Public Const OFN_EXPLORER = &H80000 new look commdlg Public Const OFN_NODEREFERENCELINKS = &H100000 Public Const OFN_LONGNAMES = &H200000 force long names for 3.x modules

PAGE 194

187 APPENDIX A (Continued) 'DCmodAPI.rtf Const STARTF_USESHOWWINDOW = &H1 Declare Function WinExec Lib "kernel32" (ByVal lpCmdLine As String, ByVal nCmdShow As Long) As Long error values for ShellExecute () beyond the regular WinExec() codes Const SE_ERR_SHARE = 26 Const SE_ERR_ASSOCINCOMPLETE = 27 Const SE_ERR_DDETIMEOUT = 28 Const SE_ERR_DDEFAIL = 29 Const SE_ERR_DDEBUSY = 30 Const SE_ERR_NOASSOC = 31 ShellExecute() and ShellExecuteEx() error codes regular WinExec() codes Const SE_ERR_FNF = 2 file not found Const SE_ERR_PNF = 3 path not found Const SE_ERR_ACCESSDENIED = 5 access denied Const SE_ERR_OOM = 8 out of memory Const SE_ERR_DLLNOTFOUND = 32 Note CLASSKEY overrides CLASSNAME Const SEE_MASK_CLASSNAME = &H1 Const SEE_MASK_CLASSKEY = &H3 Note INVOKEIDLIST overrides IDLIST Const SEE_MASK_IDLIST = &H4 Const SEE_MASK_INVOKEIDLIST = &HC Const SEE_MASK_ICON = &H10 Const SEE_MASK_HOTKEY = &H20 Const SEE_MASK_NOCLOSEPROCESS = &H40 Const SEE_MASK_CONNECTNETDRV = &H80 Const SEE_MASK_FLAG_DDEWAIT = &H100 Const SEE_MASK_DOENVSUBST = &H200 Const SEE_MASK_FLAG_NO_UI = &H400 Type SHELLEXECUTEINFO cbSize As Long fMask As Long

PAGE 195

188 APPENDIX A (Continued) hwnd As Long lpVerb As String lpFile As String lpParameters As String lpDirectory As String nShow As Long hInstApp As Long Optional fields lpIDList As Long lpClass As String hkeyClass As Long dwHotKey As Long hIcon As Long hProcess As Long End Type Declare Sub WinExecError Lib "shell32.dll" Alias "WinExecErrorA" (ByVal hwnd As Long, ByVal error As Long, ByVal lpstrFil eName As String, ByVal lpstrTitle As String) Const SW_HIDE = 0 Const SW_SHOWNORMAL = 1 Const SW_NORMAL = 1 Const SW_SHOWMINIMIZED = 2 Const SW_SHOWMAXIMIZED = 3 Const SW_MAXIMIZE = 3 Const SW_SHOWNOACTIVATE = 4 Const SW_SHOW = 5 Const SW_MINIMIZE = 6 Const SW_SHOWMINNOACTIVE = 7 Const SW_SHOWNA = 8 Const SW_RESTORE = 9 Const SW_SHOWDEFAULT = 10 Const SW_MAX = 10 Old ShowWindow() Commands Const HIDE_WINDOW = 0 Const SHOW_OPENWINDOW = 1 Const SHOW_ICONWINDOW = 2 Const SHOW_FULLSCREEN = 3 Const SHOW_OPENNOACTIVATE = 4

PAGE 196

189 APPENDIX A (Continued) Identifiers for t he WM_SHOWWINDOW message Const SW_PARENTCLOSING = 1 Const SW_OTHERZOOM = 2 Const SW_PARENTOPENING = 3 Const SW_OTHERUNZOOM = 4 Const WM_SHOWWINDOW = &H18 Declare Function ShowWindow Lib "use r32" (ByVal hwnd As Long, ByVal nCmdShow As Long) As Long Const SWP_SHOWWINDOW = &H40 Declare Function ShowWindowAsync Lib "u ser32" (ByVal hwnd As Long, ByVal nCmdShow As Long) As Long '-------------------------------------------------'Set Window Pos Declare Function SetWindowPos Lib "use r32" (ByVal hwnd As Long, ByVal hWndInsertAfter As Long, ByVal x As Long, ByVal y As Long, ByVal cx As Long, ByVal cy As Long, ByVal wFlags As Long) As Long SetWindowPos Flags Const SWP_NOSIZE = &H1 Const SWP_NOMOVE = &H2 Const SWP_NOZORDER = &H4 Const SWP_NOREDRAW = &H8 Const SWP_NOACTIVATE = &H10 Const SWP_FRAMECHANGED = &H20 The frame changed: send WM_NCCALCSIZE 'Const SWP_SHOWWINDOW = &H40 Const SWP_HIDEWINDOW = &H80 Const SWP_NOCOPYBITS = &H100 Const SWP_NOOWNERZORDER = &H200 Don't do owner Z ordering Const SWP_DRAWFRAME = SWP_FRAMECHANGED Const SWP_NOREPOSITION = SWP_NOOWNERZORDER SetWindowPos() hwndInsertAfter values Const HWND_TOP = 0 Const HWND_BOTTOM = 1 Const HWND_TOPMOST = -1 Const HWND_NOTOPMOST = -2 Type DLGTEMPLATE

PAGE 197

190 APPENDIX A (Continued) style As Long dwExtendedStyle As Long cdit As Integer x As Integer y As Integer cx As Integer cy As Integer End Type Type DLGITEMTEMPLATE style As Long dwExtendedStyle As Long x As Integer y As Integer cx As Integer cy As Integer id As Integer End Type

PAGE 198

191 APPENDIX B Previous Publications Cognitive impairment in PDAPP mice depends on ApoE ans ACT-catalyzed amyloid formation....... ................ .................... .............. .............. .......192 Environmental enrichment improv es cognition in aged Alzheimer’s Transgenic mice despite stable -am yloid deposition...... .................250

PAGE 199

APPENDIX B (Continued) 192 Previously Published Neurobiology of Aging 25 (2004) 1153-1167 Cognitive impairment in PDAPP mice depends on ApoE and ACTcatalyzed amyloid formation Lars N.G.Nilsson1,6,8, Gary W. Arendash3,8, Ralph E. Leighty3, David A. Costa1, Mark A. Low3, Marcos F. Garcia3, Jennifer R. Cracciolo3,Amyn Rojiani2, Xin Wu4, Kelly R. Bales4, Steven M. Paul4 & Huntington Potter1,5,7 1Suncoast Gerontology Center and Department of Biochemistry and Molecular Biology, 1,2Moffitt Cancer Center, 2Department of Pathology, College of Medicine, MDC07, University of South Florida, 12901 Bruce B. Downs Blvd., Tampa, FL 33612, USA 3Memory and Aging Research Laboratory, Department of Biology SCA112, College of Arts and Sciences, University of South Florida, Florida, FL 33620, USA 4Neuroscience Discovery Research, Lilly Research Laboratories, Indianapolis, Indiana, 46285, USA 5The Florida Alzheimer’s Center and Research Institute 6Present address: Department of Public Health and Caring Sciences, Uppsala University, Dag Hammarskjlds Vg 20, S-75185 Uppsala, Sweden 7Correspondence: hpotter@hsc.usf.edu 8These authors contributed equally to this work Corresponding author: Dr. Huntington Potter, Department of Biochemistry and Molecular Biology, College of Medicine, MDC 7, University of South Florida, Tampa, FL 33612, tel. (813) 974-5369, fax (813) 974-7357, email. hpotter@hsc.usf.edu Acknowledgements We thank Jean-Paul Sorondo for assistance in behavioral analysis in this study, and Dr. Dave Morgan for provid ing access to a fluorescence microscope for photography. The research was supported by a grant AG09665 from the National Institute of Aging (HP), funding from USF Institute of Aging (GWA), a fellowship from the John Douglas French Alzheimer’s Foundation and grants from the Loo and Hans Osterman and the Swedish Alzheimer Foundation (LNGN). H.P. occupies the Eric Pfeiffer Chair for Research in Alzheimer’s disease at the Suncoast Gerontology Center at the University of South Florida and is acting director; Florida Alzheimer’s Center and Research Institute.

PAGE 200

APPENDIX B (Continued) 193 1.Abstract Biochemical and genetic studies indica te that the inflammatory proteins, Apolipoprotein E (ApoE) and 1 -Antichymotrypsin (ACT) are important in the pathogenesis of Alzheimer’s disease (AD). Using several lines of multiply transgenic/knockout mice we show here that murine ApoE and human ACT separately and synergis tically facilitate both diffuse A immunoreactive and fibrillar amyloi d deposition and thus also promote cognitive impairment in aged PDAPP(V717F) mice. The degree of cognitive impairment is highly co rrelated with the ApoEand ACTdependent hippocampal amyloid burden, with PDAPP mice lacking ApoE and ACT having little amyloid and little learning disability. A analysis of young mice before the onset of amyloid formation shows that steady state levels of monomeric A peptide are unchanged by ApoE or ACT. These data suggest that the process or product of amyloid formation is more critical than monomeric A for the neurological decline in AD, and that the risk factors ApoE and ACT participate primarily in disease processes downstream of APP processing. Key words: 1-Antichymotrypsin, Apolipoprotein E, Alzheimer’s disease, amyloid deposition, learning, memory, inflammation, transgenic mice, amyloid -peptide

PAGE 201

APPENDIX B (Continued) 194 1. Introduction Alzheimer’s disease (AD) is a neurodegenerative disorder characterized psychologically by progressi ve mental declin e and defined hist opathologically by parenchymal amyloid deposits and neurofibrillary tangles. Mutations within the amyloid precursor protein (APP) gen e that cause inherited AD place APP and the A peptide at the center of the disease process [52, 26, 58]. Whether A per se or the process or product of its conversion into amyloid underlies AD neurodegeneration remains unknown. Insight into the role of APP and A in AD may be provided by genetic and biochemical studies showing other proteins to be involved in the disease process. For example, A amyloid deposits also contain 1-antichymotrypsin (ACT; [1]) and apolipoprotein E (ApoE; [43, 73]), which are over-expressed in affected regions of the AD brain as part of an inflammatory process [74, 1]. ACT levels are also increased in AD serum and CSF ([15]for data and discussion). It was proposed at the time of their discoveries that ApoE and ACT might function as amyloid promoters or “Pathological Chaperones”, and both in vitro and in vivo studies support this model [36, 56, 72, 8, 7, 28, 41, 45]. As with APP, the importance of inflammation and specifically ApoE and ACT in AD is supported by genetics. Inheritance of ApoE4 is the strongest risk factor for AD besides age [13, 59], and an ACT/A signal peptide variant that increases mature glycosylated ACT available for secretion [46] increases AD

PAGE 202

APPENDIX B (Continued) 195 susceptibility and pathology [30, 75] (for discussion see [44, 51]). Similarly, polymorphisms in the IL-1 promoter greatly increase the risk of AD [40]. These genetic variations increase IL-1 expression during inflammation and would therefore be expected to promote amyloid formation by increasing the production of both APP and ACT [24, 10, 22, 14, 46, 55]. Although genetic and biochemical studies have identified key proteins in the AD pathogenic pathway, it remains unclear what role A inflammation, and amyloid formation play in the cognitive dysfunction of AD. To answer this question, cognitive studies should be performed with AD mouse models in which the inflammatory proteins and the A deposition they influence can be regulated without changing the level of monomeric A peptide. To this end, we have generated and analyzed four mouse models of AD and find that 1) ApoE and ACT independently and synergistically promote A immunoreactive and mature amyloid deposition without initially affecting A levels and 2) cognitive impairment in aged AD mice depends on the amyloid promoting effect of ApoE and/or ACT. 2. Materials and methods Construction of transgenic mice. A 1.5kbp full-length human ACT-cDNAclone was subcloned into a modified GFAP-expression vector construct as previously described [45]. Heterozygous mice from one of these founder lines (#8784, FVB/N) were crossed with homozygous PDGF-hAPP(V717F) mice [Swiss-Webster x C57BL/6 x DBA/2] to generate mice with genotypes APP+/-

PAGE 203

APPENDIX B (Continued) 196 /ApoE+/+/ACT+/and APP+/-/ApoE+/+/ACT-/-. Heterozygous ACT mice were also crossed with ApoE-/mice [Swiss-Webster x C57BL/6 xDBA/2] to generate mice with genotype ApoE+/-/ACT+/-. These mice were then crossed with APP+/+, ApoE-/mice [(Swiss-Webster x C57BL/6 x DBA/2) x C57BL/6] to generate mice with genotypes APP+/-/ApoE-/-/ACT+/and APP+/-/ApoE-/-/ACT-/-. All offspring were screened by PCR for identity of the GFAP-ACT gene [45], the PDGF-hAPP gene [23], as well as the mouse ApoE gene and neo gene [8]. The basal ACT protein expression in the transgenic mice was 12 pmol/g wet weight in the cerebral cortex and 24pmol/g wet weight in the hippocampus, which is at least ten-fold lower than the levels detected in human AD brain tissue [1, 32, 45]. The presence or absence of proper ACT protein expression in astrocytes was further verified by ACT-immunohistochemistry in all of the pathologically and behaviorally examined animals. The genetic background of the nontransgenic mice was Swiss-Webster. For the ease of reading, a single plus sign immediately following APP or ACT genotype refers to heterozygosity unless otherwise stated in the text. The presence of the normal complement of two murine ApoE genes in a strain is indicated in parenthesis as (mApoE+) and the absence of the human ACT transgene by (ACT-) for the sake of clarity. The four lines are therefore designated: APP+(mApoE+), ACT+ (heterozyg PDAP P/homozyg murine ApoE/heterozyg ACT) APP+(mApoE+)(ACT -) (heterozyg PDAPP/homozyg murine ApoE/no ACT) APP+, mApoE-KO, ACT+ (heterozyg PDAPP/murine ApoE knockout/heterozg ACT) APP+, mApoE-KO (ACT-) (hetrozyg PDAPP/murine ApoE knockout/no ACT)

PAGE 204

APPENDIX B (Continued) 197 ELISA Analysis of A levels At two months of age, animals of each genotype were sacrificed and their brains prepared for ELISA as previously described [18], with the modification that PBS rather that TBS was used to extract soluble A General Protocol and Behavioral Analysis. The four groups of transgenic mice, along with three strains of non-transgenic mice (Swiss-Webster, B6D2/F1, and FVBJ) were maintained on a 12-hr light -dark cycle. The animals were provided free access to water and rodent chow, with behavioral testing always done during the light phase and by investigators unaware of animal genotypes. Beginning at 16 months of age, all animals were tested in a 6-week battery of sensorimotor, anxiety and cognitive ta sks, which included the Morris water maze and the radial arm water maze (RAWM) as previously described [6, 38]. Two groups of animals, APP+ (mApoE+) (ACT-) and nontransgenic littermates, were behaviorally tested in the same battery of tasks at two months of age to determine whether animals destined to deposit amyloid were impaired at an early age before deposition began. Briefly, the RAWM consisted of a circular pool, 1 meter in diameter, with six swim arms (19 cm wide) radiating from an open central area (40 cm in diameter). A variety of spatial cues were present on the walls and ceiling in the immediate vicinity of the pool for RAWM testing (as well as for Morris water maze testing), in order to spatially orient the animals while swimming. The

PAGE 205

APPENDIX B (Continued) 198 submerged escape platform was positioned near the end of a different arm for each of the 9 days of testing, thus forcing the animals to use spatial working memory to decrease their escape latency to find a given days “goal” arm over 5 trials per day. Four consecutive acquisition trials were done (T1-T4), followed by a retention trial (T5) 30 minutes later. The five arms not containing the platform were designated as start arms for each of the one-minute trials, in a varied semi-randomized sequence for each day of testing. For any given trial, the animal was placed in a start arm and allowed to make arm choices, with each incorrect choice resulting in the animal being gently pulled back to the start arm for that trial. The time required to locate the submerged platform was recorded and the animal was allowed to remain on the escape platform for 30 seconds. Animals that did not find the platform within any 60 sec trial were guided to the platform and the maximal 60 sec time was recorded. RAWM latency data were analyzed over three 3-day blocks. Because unimpaired animals generally require the first two blocks to learn this task’s procedural aspects, block 3 of testing (days 7-9) is most indicative of cognitive performance. The Morris water maze consisted of an open circular pool, with a diameter of 1 meter, which was divided in four quadrants. An indiscernible 9cm platform was positioned in quadrant 2 (Q2) 1.5cm below the water surface. Acquisition testing involved four trials per day for 9 days wherein an animal was placed successively into each of the four quadrants to initiate a 60 sec trial. The time to locate the platform was recorded for each trial and the

PAGE 206

APPENDIX B (Continued) 199 average latency to find the submerged platform was calculated for these four trials. Upon locating and ascending the platform (or after 60 sec) the animal was allowed a 30 sec inter-trial resting period on the platform. Histology and Immunohistochemistry. At 18 months of age, brains were dissected from anesthetized and intracardially perfused mice and 25 m coronal tissue sections were processed as previously described [45]. The mounted sections were processed through antigen retrieval in prewarmed 25mM citrate buffer (pH 7.3) at +82 C for 5 min and further processed as previously described [45]. The sections were incubated with primary antibodies rabbit anti-ACT (AXL-145, dil 1:1000, Accurate) and mouse anti-A (6E10, dil 1:5000, Senetek) overnight at +4 C. Secondary antibodies were anti-rabbit IgG (BA-1000, 1:300, Vector) or anti-mouse IgG and developed with NovaRED substrate kit (Vector). Congo Red-staining and Thioflavine S was performed according to well-established protocols[16]. Image analysis. Data were collected from five equally spaced coronal tissue sections for both dorsal hippocampus and overlying parietal cortex (Bregma 1.06 to -2.30mm; [45]. The sections were examined with a Nikon Eclipse E600 microscope at 100X (6E10 immunostaining) and 200X (Congo Red) magnification at a constant predefined light setting and video images captured with a color CCD-camera. All images were then processed through shading correction. The 6E10 immu nostaining was converted to a grayscale image and

PAGE 207

APPENDIX B (Continued) 200 segmented with an auto threshold function (Image Pro Plus, Media Cybernetics). Diffuse immunoreactive A deposition (6E10) was estimated as area fraction (=stained areatot/measured areatot, expressed in %). Congo Red positive amyloid plaques were circled at 200X magnification, and their location noted on an anatomic atlas. The captured image was segmented with respect to threshold settings for RGB that had been specified prior to analysis so as to distinguish specific signals from background. The area occupied by amyloid within each circle, as defined by the image segmentation, was then quantitated and an area fraction calculated from the total measured area as previously described [45]. Statistical analysis. The histopathological results were analyzed with factorial ANOVA and post hoc Fisher LSD test. The behavioral data were evaluated with One-way ANOVA and post-hoc Fisher LSD test of each separate trial. Repeat-measures ANOVA across all trials yielded almost identical results. The correlations between histopathological and behavioral analysis were examined with linear regression. Discriminant analysis was performed to determine whether groups of mice with different strain backgrounds, different genotypes or varying amyloid depositing capacity, could be distinguished from one another behaviorally. The analysis, which was based on the performance of 719 measures from our behavioral test battery, was performed with the DISCRIM subroutine of the SYSTAT software package. The discriminant analysis was based on the following measures:

PAGE 208

APPENDIX B (Continued) 201 1. Open field activity (line crossings during a 5 minute trial), 2. Balance Beam (time before falling during 3 successive trials, in seconds), 3. String Agility (agility to put hindpaws and/or tail around a suspended string during a 1 minute trial), 4. Y-maze choices (number of arm choices in a single 5 minute trial), 5. Y-maze percent alternation (in a single 5 minute trial), 6. Morris maze acquisition (averaged daily latencies over 9 days of testing), 7. Morris maze acquisition (averaged latency for the last day of testing), 8. Morris maze retention (percent of time spend in former platform quadrant during a probe trial done the day following acquisition), 9. Circular platform errors (averaged over 8 days of testing), 10. Circular platform errors (last day of testing), 11. Circular platform escape latency (averaged over 8 days of testing), 12. Circular platform escape latency (last day of testing), 13. Platform recognition (averaged daily latencies over 4 days of testing), 14. Platform recognition (averaged latency for the last day of testing), 15. Radial arm water maze Trial 4 latency over all 3 blocks of testing, 16. Radial arm water maze Trial 5 latency over all 3 blocks of testing, 17. Radial arm water maze Trial 1 latency for the last block of testing, 18. Radial arm water maze Trial 4 latency for the last block of testing, 19. Radial arm water maze Trial 5 latency for the last block of testing. Discriminant analysis forms new variables called “discriminant functions”, which are linear composites of the multiple original measures that are used to evaluate the overall performance of groups of animals. The number of such functions generated always equals the number of groups analyzed minus one.

PAGE 209

APPENDIX B (Continued) 202 Thus, if at least three groups are being compared, any two “discriminant functions” can be graphed to visually illustrate the presence or absence of group discrimination, as mathematically evaluated through the statistical analysis (Pillai’s trace). 3. Results To study the effect of ApoE and ACT on A immunoreactive and mature amyloid formation and cognitive decline, we have generated and analyzed four transgenic mouse lines: APP+(mApoE+)(ACT-) (n=11), APP+,(mApoE+),ACT+ (n=17), APP+,mApoE-KO,ACT(n=8), and APP+,mApoE-KO,ACT+ (n=9). Animals from these heterozygous transgenic lines were behaviorally analyzed along with a group of nontransgenic mice (n=5) at 16 months of age and subsequently sacrificed at 18 months of age for histopathological evaluation. Other mice from these four lines were also sacrificed at two months of age to determine the effect(s) of ApoE and ACT on the steady state levels of monomeric A before amyloid deposition begins. Some were also behaviorally analyzed to determine whether animals destined to deposit amyloid were impaired at this early age. ApoE and ACT expression in the brain do not increase the steady state levels of A 1-40 or A 1-42 prior to amyloid deposition in PDAPP transgenic mice

PAGE 210

APPENDIX B (Continued) 203 All of the APP and presenilin mutations that have been found to cause familial Alzheimer’s disease with an autosomal dominant mode of inheritance appear to act by increasing the amount of the A (1-42) peptide, which displays a high propensity to polymerize into amyloid filaments [57, 20]. If ApoE or ACT were to similarly influence AD by affecting A production or clearance, one would predict that steady state levels of monomeric A should be altered in young PDAPP/ACT transgenic or PDAPP/mApoE-KO mi ce in comparison to unaltered PDAPP mice. In contrast, if the main role of ApoE and ACT is to promote the formation of amyloid deposit s, the presence or absence of these genes would not necessarily affect monomeric A levels. To investigate the potential role of ApoE and ACT in the formation/clearance of A we first used ELISA to qu antify both water soluble (defined as PBS-extractable) and me mbrane-bound (defined as guanidineextractable) A 1-40 and A 1-42 in the hippocampus and cerebral cortex of young APP transgenic mice prior to the onset of amyloid deposition (2 month old transgenic mice). No consistent, statistically significant genotype-dependent difference in either water soluble or membrane-bound A 1-40 or A 1-42 were observed between our four lines of transgenic mice in the hippocampus. In general, transgenic mice expressing ApoE tended to display lower levels of water soluble A and mice expressing ACT tend ed to show reduced levels of membrane-bound A in the hippocampus (Table 1 and Figure 1). Similar results and trends were observed in the cerebral cortex tissue extracts, where the level of water soluble A was found to be more clearly reduced by ApoE

PAGE 211

APPENDIX B (Continued) 204 expression (Table 1). The ability of ApoE to reduce soluble A levels in young APP mice has been reported in other studies [21, 18], and previous findings have shown that young APP/ACT mice ha ve unchanged or slightly increased A levels using different antibodies [41, 45]. In sum there is little evidence that either murine ApoE or human ACT increase A levels in very young APP mice, but rather there is a trend toward ApoE and ACT reducing A levels. Thus any increases in A immunoreactive and mature amyloid deposition observed in the presence of ApoE or ACT in older mice is likely to be due to amyloid-promoting mechanisms other than merely increasing production or decreasing clearance of monomeric A prior to amyloid deposition and the inflammation that it induces. Young Adult (2 month old) APP transgenic mice are not impaired in either the Morris Water Maze or the RAWM To determine whether animals genetically destined to deposit amyloid later in life show early behavioral deficits that could be attributed to expression of the transgene or the production of human A peptide, we tested APP+ (mApoE+)(ACT-) and non trangenic littermates in the same battery of tasks that would be used to assess the behavior of aged animals. The young animals showed no deficits in non-cognitive tasks. Furthermore, as shown in Figure 1e,f, the 2 month old APP-expressing and non-transgenic animals were also not significantly different in either Morris Maze acquisition or radial arm water maze (RAWM) working memory.

PAGE 212

APPENDIX B (Continued) 205 ApoE and ACT separately and in combination increase A deposition in PDAPP transgenic mice Amyloid deposition is a complex process consisting of several stages. In order to develop the most effective therapies for Alzheimer’s disease, it is important to know which proteins contribute to which stages of the amyloid cascade and which stages cause cognitive decline. The histopathological analysis of our mice was designed to distinguish the individual and synergistic contributions of ApoE and ACT to both diffuse A deposition and fibrillar amyloid formation and, together with the behavioral analysis described below, to distinguish the relative impact of these forms of A deposition on cognitive decline. Total A deposition was measured as area (in %) occupied by A immunoreactivity with the 6E10 antibody. Figure 2 shows that total A deposition was markedly reduced in the hippocampus and absent in the cerebral cortex in PDAPP transgenic mice lacking ApoE expression [APP+, mApoE-KO, (ACT-)], consistent with previous findings [8]. Furthermore there was increased deposition in APP+(mApoE+),ACT+ transgenic mice, as compared to APP+,(mApoE+)(ACT-) transgenic mice, in the cerebral cortex (4.8 0.6% vs. 2.8 0.6%, P<0.05) and a modest elevation in the hippocampus (8.7 0.6% vs. 7.4 0.7%, n.s.), due to the presence of the ACT transgene (Figure 2a-b and 2e). The difference between APP+,(mApoE+),ACT+ and APP+,(mApoE+)(ACT-) transgenic mice was greater in the lateral extension of the parietal cortex as compared to either the hippocampus or the retrosplenial

PAGE 213

APPENDIX B (Continued) 206 cortex, where immunoreactive A deposition develops at an early age (Figure 2a-b). Furthermore there was an almost fourfold higher A load in the hippocampus of the ACT-expressing/ApoE knockout (APP+,mApoE-KO,ACT+) transgenic mice as compared to the ACT minus/ApoE knockout (APP+,mApoE-KO,(ACT-)) mice (3.1 1.3% vs. 0.8 0.8%, P<0.05, Figure 2c-e). In fact, only a single animal among the transgenic mice with genotype APP+,mApoE-KO,(ACT-) showed any immunoreactive A deposition at 18 months of age. Mature, filamentous amyloid deposition is facilitated by either ApoE or ACT expression in the PDAPP transgenic mice In addition to forming diffuse, immunoreactive A deposits, as quantified with the 6E10 antibody as above, the A peptide also deposits as mature filamentous amyloid distinguishable by its strong beta sheet character. Congo Red staining, which measures such compact A deposition, was about twofold elevated in both the hippocampus and cerebral cortex of the APP+,(mApoE+),ACT+ mice (F igure 3a) as compared to APP+,(mApoE+)(ACT-) transgenic mice (Figure 3b). This increase was statistically significant both in the cerebral cortex (0.028 0.004% vs. 0.014 0.0.004%, P<0.05) and the hippocampus (0.060 0.008% vs. 0.037 0.006%, P<0.05, Figure 3c). The ACT-induced increase in congophilic amyloid load was due to increased plaque density in both the cerebral cortex (3.12 0.46 vs. 1.44 0.44 plaques/mm2, P<0.05) and the hippocampus (6.01 0.62 vs. 3.86 0.59 plaques/mm2, P<0.05, Figure 3d). Furthermore ACT

PAGE 214

APPENDIX B (Continued) 207 antibodies immunostained Congo Red positive plaques in APP+,(mApoE+),ACT+, but not APP+,mApoE+)(ACT-) mice (not shown). There were no congophilic A deposits in the hippocampus and the cerebral cortex of either the APP+,mApoE-KO, ACT+ or the APP+,mApoE-KO,(ACT-) mice at 18 months of age (but see below). ApoE-immunoreactivity has previously been demonstrated to be associated with a subset of A immunoreactive plaques, which are essentially always Thioflavine S positive [7, 62 ]. However, using a more sensitive immunostaining protocol we now can show that ApoE is localized in essentially every A immunoreactive plaque in an APP+,(mApoE+),ACT+ transgenic mouse (Fig 4a-b), while ACT-immunostaining is largely restricted to Thioflavine S positive amyloid plaques. Thioflavine S staining and ACT-immunostaining also correlate well in their relative intensity among individual amyloid plaques. These results suggest that ACT binds to A peptide or an intermediate A species, increases the extent of its beta sheet structure and thereby accelerates the formation of mature amyloid plaques (Fig 4c-d). In vitro binding and X-ray crystallography/structural modeling studies between ACT and A support such a model of ACT function in AD [50, 70, 35, 34, 29]. At an advanced age (23 months) all (6/6) homozygous APP+,mApoEKO,ACT+ mice, but only some (6/10; 60%) age-matched APP+,mApoEKO,(ACT-) mice, developed Thioflavine S positive amyloid plaques in the hippocampus. The anatomical distribution of A and Thioflavine S positive staining varied among individual animals, particularly among the

PAGE 215

APPENDIX B (Continued) 208 APP+,mApoE-KO,(ACT-) mice. The A -immunoreactive staining included the cerebral cortex in some of these mice, both with and without the ACT transgene. The Thioflavine S positive amyloid plaques in APP+,mApoEKO,ACT+ mice were always A -immunoreactive, and mostly ACT immunoreactive and were most frequently observed in the molecular layer of the dentate gyrus and the CA1 stratum lacunosum/moleulare, but were also present in the CA1 and CA3 stratum radiatum. ACT immunopositive staining was also observed in A immunoreactive deposits that were not Thioflavine S positive in these aged APP+,mApoE-KO,AC T+ mice (Figure 4e-h). Thioflavine S was used in these experiments because we found it to be a more sensitive dye than Congo Red. ApoE expression in aged PDAPP transgenic mice is associated with impaired learning ability in the Morris Water Maze Gender and age-matched animals were randomly selected from the four transgenic groups and, along with a group of non-transgenic mice, were examined in a battery of st andard behavioral tasks. As will be discussed in detail below, the non-cognit ive tasks in the battery indi cated that the mice in each of our lines were functioning well and could not be distinguished from each other in general non-cognitive abilities. For example, animals of all

PAGE 216

APPENDIX B (Continued) 209 genotypes performed at a similar level in open field activity and string agility tasks that estimate exploratory behavior and muscle strength and all exhibited similar swim speeds and were able to see. Cognitive tasks, however, did reveal differences between the transgenic lines. The most common ta sk for assessing cognition in mice is the standard Morris Water Maze which measures sp atial learning ability. The escape latencies for the four experimental and the non-transgenic control mice during the first block of testing (Days 1-3) in the Morris Water Maze were the same (F4,38 =0.636; P=0.63). This result reinforced the conclusion from the noncognitive tasks that there is no essential difference between the four transgenic groups and the non-transgenic mice in ability to perform the non-cognitive aspects of the Morris Water Maze task. However in the last block of testing (Days 7-9), significantly higher esca pe latencies were recorded for the APP+,(mApoE+)(ACT-) mice (41 5 sec) as compared to the APP+,mApoEKO,(ACT-) mice (26 3 sec, P<0.05) and the nontransgenic mice (21 3 sec, P<0.01). Similar inferior performance was displayed by the APP+,(mApoE+),ACT+ mice (36 4 sec) as compared to the APP+,mApoEKO,ACT+ mice (23 4 sec, P<0.05) and the nontransgenic mice (21 3 sec, P<0.05, Figure 5). Thus cognitive dysfunction as measured by the Morris Water Maze is associated with ApoE-catalyzed amyloid formation in APPexpressing mice. ApoE knockout mice have been reported to exhibit cognitive deficits and thigmotaxic behavior (a tendency to swim in the peripheral annulus of the pool)

PAGE 217

APPENDIX B (Continued) 210 in association with structural cytoskeletal and neurochemical deficits in the brain. However, the ApoE knockout mouse model we used shows no such defect [4, 3, 11, 64], and we did not observe any thigmotaxic behavior in our mice. The Morris Water Maze was unable to reveal any difference between, for example, APP+,mApoE-KO,(ACT -) and APP+,mApoE-KO,ACT+ mice, although these groups of mice display different extents of A immunoreactive histopathology. We surmised that a more demanding behavioral task would be more informative and sensitive in its ability to reveal subtle cognitive differences between the transgenic mice. ApoE and ACT expression in aged PDAPP transgenic mice determine the extent of cognitive impairment in the Radial Arm Water Maze The Radial Arm Water Maze (RAWM) is a more challenging version of the Morris Water Maze that is designed to test both visual-spatial and working memory in mice [6]. Previous work has shown that the RAWM provides a very sensitive behavioral measure that correlates well with the level of A immunoreactive and mature amyloid load in transgenic mouse models of AD [6, 25, 38]. Therefore all of the behaviorally-analyzed mice were examined over three blocks of testing in the RAWM (Figure 6). Each of the three blocks consists of three days in which four acquisition trials (T1-4) and a delayed retention trial (T5) are performed on each day, with the location of the submerged platform being altered daily.

PAGE 218

APPENDIX B (Continued) 211 The escape latencies in the RAWM did not differ between the five groups of mice (four experimental and one control, non-transgenic) in the first trials (T1) of the first block of testing (Days 1-3), when the animals were naive to the task (F4,31 = 0.60; P=0.67; Figure 6a), indicating that there was no inherent difference between the groups in basic swimming and other noncognitive performance (see further discriminant analysis below). However, over the full 3 blocks of testing, there devel oped group differences in both learning (Trial 1 compared to Trial 4 on each day) and memory (Trial 5). Of particular importance were group differences in average T5 performance across all three blocks, wherein APP+,(mApo E+) with or without ACT displayed substantially longer escape latencies (47 4 and 47 2 sec.) compared to APP+,mApoE-KO mice with or without ACT (36 4 and 29 2 sec., P<0.05, P<0.001) and nontransgenic controls (29 3 sec., P<0.01) (Figure 6b). In block 3, when cognitive learning processes are most distinguishable from the procedural learning that occurs during earlier testing blocks, APP+,(mApoE+) mice with or without ACT were impaired in their ability to reduce their escape latencies between trials T1 and T5. On both the final learning trial (T4) and the memory retention trial (T5), escape latencies for these two groups were significantly longer than those for APP+,mApoE-KO,(ACT-) mice and nontransgenic mice. Importantly, APP+,mApoE-KO,ACT+ mi ce displayed an intermediate RAWM performance. Although these mice were able to learn, as evidenced by their ability to reduce their escape latencies throughout the trials, their T4 escape latencies were significantly longer compared to both APP+,mApoE-

PAGE 219

APPENDIX B (Continued) 212 KO,(ACT-) mice and nontransgenic mice. This result indicates that the ACT expression not only promoted amyloid formation, but also caused cognitive decline as measured in the RAWM. ACT transgene expression in APP transgenic mice expressing ApoE further increased amyloid deposition but did not further reduce the already very poor spatial learning ability of APP+,(mApoE+) mice, which we interpret as a ceiling effect in this difficult behavioral task. Interestingly, one of the APP+,mApoEKO,ACT+ transgenic mice lacking visible A immunoreactive deposition performed poorly in the RAWM. It could be that biologically active A intermediates that are detrimental to hippocampal function had already begun to form in this mouse, but had not yet reached the threshold level necessary to precipitate detectable A immunoreactive deposits. The extent of ApoE/ACT-catalyzed A deposition correlates with cognitive performance. The data clearly show that the process and/or the product of amyloid formation found in both groups of APP+,( mApoE+) transgenic mice is of prime importance for cognitive impairment in the RAWM. This conclusion is further reinforced by a strong correlation between the T5 retention trial escape latency over all three test blocks and the amount of both immunoreactive A deposition (T5, r=0.68, P<0.001, Figure 6c) and mature congophilic amyloid (T5, r=0.60, P<0.001, Figure 6d) in the hippocampus. Furthermore, the correlation between escape latency and immunoreactive A deposition in the hippocampus remains strong when the

PAGE 220

APPENDIX B (Continued) 213 statistical analysis is restricted to the high-depositing APP+,(mApoE+) mice with or without ACT (T5, r=0.59, P<0.01). Extrapolating this linear correlation to its intercept on the y-axis (i.e. the extrapolated, theoretical, escape latency of an APP+,(mApoE+) transgenic mouse with zero A deposition) yields a predicted latency of 3112sec, which is close to the average recordings of the APP+,mApoE-KO mice (333sec). This extrapolation indicates that the main contribution of ApoE to cognitive decline is to promote A immunoreactive and mature amyloid formation. Without such deposition, there would be no cognitive difference (i.e. difference in latency) between APP+,(mApoE+) mice and either APP+,mApoE-KO mice or non-transgenic mice. Diffuse A deposition in the hippocampus is associated with cognitive impairment in PDAP P transgenic mice Thus far, the analysis has shown that ApoE and ACT catalyze A immunoreactive and mature amyloid deposition in PDAPP mice and that the extent of these deposits correlates with the cognitive dysfunction of the mice, suggesting a cause and effect relationship. We then analyzed the data further to determine the relative impact of the two major forms of A deposition, diffuse A immunoreactivity and compact/fibrillar amyloid on cognitive performance. All 18 months old PDAPP transgenic mice lacking ApoE and thus exhibiting only diffuse immunoreactive A deposits with no compact A amyloid deposition were stratified for the presence or absence of immunoreactive A deposition in the hippocampus, and their behavior in the

PAGE 221

APPENDIX B (Continued) 214 RAWM was compared to that of nontransgenic mice (Fig. 6e). In the first trial (T1) of the first block of testing (Days 1-3), when the animals were naive to the RAWM, there were no differences between any of the experimental groups (F2,18 = 3.05; P=0.072). In block 3 (Days 7-9) it was found that working memory was worse in those APP+,mApoE-KO transgenic mice with diffuse hippocampal A immunoreactive deposits as compared to APP+,mApoE-KO mice without diffuse A deposits (T5, P<0.05), which were indistinguishable from nontransgenic mice. Behavioral differences am ong the four groups of transgenic mice and the nontransgenic mice are not due to different genetic backgounds Because of the multiple crosses during breeding that were necessary to generate our four lines of transgenic/knockout mice, it was not possible to assure that the strain backgrounds we re identical short of many years of repeated backcrossing. Instead all of th e mice have a uniform, mixed genetic background that should equally assure that any consistent difference between the lines must be due to the presence or absence of ApoE or ACT, which are the only genetic trait that the members of a particular line share. To experimentally demonstrate that the differential behavior of the transgenic mice is largely due to the inheritance of the transgenic/knockout loci, we performed a comprehensive comparison of all of the behavioral data using discriminant function analysis. As previous studies have shown, this statistical tool allows for the comparison and discriminating analysis of characteristic

PAGE 222

APPENDIX B (Continued) 215 behavioral profiles of specific mouse strains [54]. We applied such discriminant analysis to the performance of our transgenic mice and of three lines of nontransgenic mice in the entire group of 19 different behavioral measures from multiple behavioral tasks in our test battery. There were consistent behavioral differences between the three groups of nontransgenic mice of different strain background, including the st rains from which our transgenic/knockout mice were derived (P<0.025, Pillai’s trace, Fi gure 7a), indicating the ability of this technique to distinguish different mouse strains. In contrast, we were not able to behaviorally discriminate our control nontransgenic and the four transgenic groups of mice based on the same 19 behavioral measures (P>0.05, Pillai’s trace, Figure 7b). When the analysis was limited to 10 cognitive-based measures, the two APP+,(mApoE+) (high A depositing) lines could be discriminated from the two APP+,mApoE-KO (low A depositing) lines, confirming that cognitive-based behavioral measures are associated with the amyloid pathology. In contrast, the high depositing lines could not be distinguished from the low depositing lines on the basis of non-cognitive behavioral measures (P>0.05, Pillai’s trace). In sum, not only are the mice within each line that we have used more similar cognitively to each other than they are to mice in any other line, indicating that their differences are due only to ApoE or ACT, but no group can be distinguished by non-cognitive measures, indicating that they are all fundamentally similar in background. 4. Discussion

PAGE 223

APPENDIX B (Continued) 216 Previously, we and others had hypothesized and then demonstrated the ability of ApoE and ACT to catalyze the polymerization of A into neurotoxic amyloid filaments [1, 73, 36, 72, 56]. Specific binding experiments, site directed mutagenesis and structural analysis then identified the molecular interactions between these pathological chaperones and A and outlined the likely mechanism of action of the amyloid promoting reaction [50, 60, 35, 29]. The present study provides the first in vivo demonstration that the increased A immunoreactive and mature plaque deposition caused by expression of the amyloid promoters ApoE and ACT, is associated with impaired spatial learnin g. Our results show 1) that ApoE and ACT separately and synergistically promote the development of both diffuse, immunoreactive A deposits and mature amyloid plaques in A overexpressing mice without affecting the levels of monomeric A itself, 2) that impaired cognitive performance in two different spatial tasks requires both A peptide and either ApoE or ACT as an amyloid promoter and 3) that both diffuse A deposits and mature amyloid plaques are correlated with impaired spatial lear ning ability. Indeed, 18 month old APP mice that expressed neither ApoE nor ACT, deposited very little A and displayed no cognitive deficits in the RAWM or Morris Water Maze, similar to nontransgenic mice. These findings imply that the process and/or product of ApoE or ACT-catalyzed amyloid formation is more critical for the cognitive decline in AD than is the amount of monomeric A -peptide. Although these experiments do not directly address which aspect or intermediate/product of amyloid formation causes cognitive decline, they

PAGE 224

APPENDIX B (Continued) 217 indicate that it must be ApoE and/or ACT-dependent and may be either some intermediate step in the process, the final, mature product, or both. Specifically, APP+,mApoE-KO,ACT+ animals accumulate diffuse, immunoreactive amyloid deposits and have impaired cognition. This finding suggests that diffuse A immunoreactive deposition, or an even earlier, ApoE/ACT-dependent, step in the process of amyloid formation, may be equally important for the cognitive decline in these mouse models of AD as is mature congophilic amyloid. This interpretation of the data is supported by previous findings that A oligomers and/or protofibrils appear to be more neurotoxic than monomeric or filamentous A both in vitro [66, 67, 27, 31] and in vivo [68, 65]. The Morris water maze and the RAWM were chosen for these studies because the place learning that they assay depends on the hippocampus [39], which is pathologically and functionally affected early in AD. It has been suggested that there is some learning deficit of PDAPP mice in a working memory version of the Morris water maze that is age-independent and therefore unrelated to A burden [12]. However, we observed no statistically significant cognitive deficits in young adult PDAPP mice (Figure 1e-f), indicating that most behavioral im pairment in the PDAPP mice depends on later-developing A deposition (and on the action of ApoE and ACT). Our data also indicate that the RAWM task is a more sensitive measure of A deposition than is the traditional reference memory version of the Morris maze.

PAGE 225

APPENDIX B (Continued) 218 Both PDAPP+,mApoE-KO and PDAP P+(mApoE+) mice have been reported to be impaired in an object recognition task as compared to nontransgenic mice. These findings were attributed to the high levels of soluble A shared by the transgenic mice and seem to contrast with our results [19]. However, the object recognition task and the RAWM measure different aspects of cognition (identification vs. working memory) that may, to a differential extent, be dependent on certain brain regions (cerebral cortex vs. hippocampus) and be sensitive to different forms of A deposition. Indeed, object recognition shows only weak correlation with A -immunoreactive deposition or mature Thioflavine S positive A deposition [19], while RAWM performance is strongly correlated with both diffuse and compact A deposition, as demonstrated here and previously [5, 25]. In addition, clearance of soluble A species from the brain through passive immunization conferred short-term cognitive benefits in the object recognition task [17]. However these were acute observations after an invasive procedure that alters the compartmentalization and equilibrium between soluble and insoluble A concentrations and are thus difficult to compare to the present findings. It is also possible that acute cognitive benefi ts of passive vaccina tion derive from the resulting reduction in the vasoconstrictor activity of A [63] or to alleviated hemodynamic responses [42], either of which could enhance cerebral blood flow and improve cognition without providing insight into the mechanism by which APP and A normally cause cognitive decline.

PAGE 226

APPENDIX B (Continued) 219 We have considered the possibility that a somewhat higher or lower proportion of the original strains in the various knockout/transgenic lines used in this study, and not differences in A deposition, could have led to the observed differences in the cognitive test results. However such an explanation is highly unlikely because 1) Different behavioral characteristics of inbred mouse strains are often lost in F1 hybrid strains [48]. For example, the protocol for generating our mice required several crosses that assured that the mice of each line contained a random mixture of backgrounds, with the only common feature being the presence or absence of APP, ApoE or ACT. Therefore, if genetic traits attributable to differences in strain background had been the major determinant of cognitive function, the individual mice within each line would have been expected to show a widely variable performance on the cognitive tests, reflecting their different genetic backgrounds. Such variation would preclude us from being able to detect differences due to mApoE or the ACT transgene. Instead the mice within each line were very similar to each other and differed as a group from the mice of the other lines. 2) There is a strong correlation between escape latency and hippocampal A immunoreactive deposition among the APP+(mApoE+) (with or without ACT) mice which all have an identical stra in background. 3) APP+, mApoE-KO (with or without ACT) mice with hippocampal A -immunoreactive deposition are impaired as compared to APP+ mApoE-KO mice without A -immunoreactive deposition. These latter two groups of mice also have an identical strain background showing that the ApoE effect on behavior cannot be due to

PAGE 227

APPENDIX B (Continued) 220 background strain difference. 4) Discriminant analysis, which can compare and reveal differences among comprehensive behavioral profiles of different mouse strains, shows that the differences among the four groups of transgenic mice and the nontransgenic mice are restricted to those behavioral measures which are the most sensitive assays of cognitive performance. In contrast, different control mouse strains can be discriminated on the basis of both cognitive and non-cognitive behavioral tests. Although ApoE and ACT clearly promote A polymerization directly, these proteins likely have additional roles to play in AD and other brain diseases. For example, it has been shown that ACT inhibits A degrading enzymes [76], and ApoE affects the level or location of APP processing so as to generate fewer A peptides and more potentially toxic C-terminal fragments (in animals slightly older than ours and of an age to show initial stages of amyloid deposition; [18]). ApoE also affects cholesterol metabolism [49]. Particularly interesting is the finding that ApoE4 is less neuroprotective than ApoE3 both in vitro and in vivo [9]. It is also important to consider the possible direct and compensatory consequences of expressing or not expressing ApoE and/or ACT during development. For example, ApoE knockout mice develop hypercholesterolemia and atherosclerosis that could have an effect on amyloid formation. However, heterozygous PD APP/ApoE+/mice have normal serum cholesterol levels and develop amyloid deposits at a rate intermediate between full PDAPP/ApoE-KO and PDAPP/ApoE+/+ animals [7] (and Nilsson and

PAGE 228

APPENDIX B (Continued) 221 Potter unpublished). This result suggest that the lower amyloid deposition induced by knocking out ApoE is not caused by high serum cholesterol due to ApoE deficiency. Similarly, the effect of long term expression of a protease inhibitor such as ACT could be de leterious or could be masked by developmental compensation that could itself have side effects. If present, such effects of ACT would affect development only late since the GFAP promoter of the ACT transgene does not become active until the very latest stages of fetal development. Furthermore, except for hypercholesterolemia in ApoE-KO mice, we have not observed gross differences in physiology or brain morphology due to knocking out ApoE or expressing ACT. In sum, our findings of a strong association between cognitive decline and both diffuse A deposits and mature amyloid plaques in transgenic mice indicate that a major pathological role of ApoE and ACT in Alzheimer’s disease is to promote A polymerization and deposition, possibly with slightly different effects in terms of time course and anatomical distribution of the pathology. Furthermore, it may be the process of amyloid formation, or an intermediate formed there in, rather than the end product (mature plaques) that causes neuronal dysfunction and cognitive decline. These results showing the essential role of ApoE and ACT in both amyloid formation and cognitive decline, together with the finding that ApoE and ACT do not alter the steady state level of A suggest that inhibition of the A /ApoE or A /ACT interactions are prime targets for therapeutic intervention in AD. Previous experiments

PAGE 229

APPENDIX B (Continued) 222 have shown that such interaction inhibitors effectively prevent ApoE and ACTcatalyzed polymerization in vitro [41]. Besides sharing the ability to promote A polymerization, ApoE and ACT also share the feature of being inflammatory proteins whose mRNAs are overexpressed in affected areas of the AD and APP mouse brain. Thus, the present findings provide an explanation for the involvement of inflammation and the prophylactic benefit of NSAIDs in AD [37, 44, 2, 53] and point to novel targets for anti-inflammatory therapy. Specifically, inhibitors of glial overproduction and release of IL-1, ApoE, or ACT should prevent amyloid formation and cognitive decline. Both in vitro and in vivo experiments indicate the likely success of such approaches [14, 35, 33]. Finally, we note that the concept of pathological chaperones playing a catalytic role in diseases of protein conformation is potentially quite general, as it appears to extend to both other AD proteins and other amyloid disorders [61, 47, 69, 71, 16].

PAGE 230

APPENDIX B (Continued) 223 5. References [1] Abraham CR, Selkoe DJ and Potter H. Immunochemical identification of the serine protease inhibitor alpha 1-antichymotrypsin in the brain amyloid deposits of Alzheimer's disease. Cell 1988; 52: 487-501. [2] Akiyama H, Barger S, Barnum S, Bradt B, Bauer J, Cole GM, Cooper NR, Eikelenboom P, Emmerling M, Fiebich BL, Finch CE, Frautschy S, Griffin WS, Hampel H, Hull M, Landreth G, Lue L, Mrak R, Mackenzie IR, McGeer PL, O'Banion MK, Pachter J, Pasinetti G, Plata-Salaman C, Rogers J, Rydel R, Shen Y, Streit W, Strohmeyer R, Tooyoma I, Van Muiswinkel FL, Veerhuis R, Walker D, Webster S, Wegrzynia k B, Wenk G and Wyss-Coray T. Inflammation and Alzheimer's disease. Neurobiol Aging 2000; 21: 383-421. [3] Anderson R, Barnes JC, Bliss TV, Cain DP, Cambon K, Davies HA, Errington ML, Fellows LA, Gray RA, Hoh T, Stewart M, Large CH and Higgins GA. Behavioural, physiological and morphological analysis of a line of apolipoprotein E knockout mouse. Neuroscience 1998; 85: 93-110.

PAGE 231

APPENDIX B (Continued) 224 [4] Anderson R and Higgins GA. Absence of central cholinergic deficits in ApoE knockout mice. Psychopharmaco logy (Berl) 1997; 132: 135-44. [5] Arendash GW, Gordon MN, Diamond DM, Austin LA, Hatcher JM, Jantzen P, DiCarlo G, Wilcock D and Morgan D. Behavioral assessment of Alzheimer's transgenic mice following long-term Abeta vaccination: task specificity and correlations between Abeta deposition and spatial memory. DNA Cell Biol 2001; 20: 737-44. [6] Arendash GW, King DL, Gordon MN, Morgan D, Hatcher JM, Hope CE and Diamond DM. Progressive, age-related behavioral impairments in transgenic mice carrying both mutant amyloid precursor protein and presenilin1 transgenes. Brain Res 2001; 891: 42-53. [7] Bales KR, Verina T, Cummins DJ, Du Y, Dodel RC, Saura J, Fishman CE, DeLong CA, Piccardo P, Petegnief V, Ghetti B and Paul SM. Apolipoprotein E is essential for amyloid deposition in the APP(V717F) transgenic mouse model of Alzheimer's disease. Proc Natl Acad Sci U S A 1999; 96: 15233-8. [8] Bales KR, Verina T, Dodel RC, Du Y, Altstiel L, Bender M, Hyslop P, Johnstone EM, Little SP, Cummins DJ, Piccardo P, Ghetti B and Paul SM. Lack of apolipoprotein E dramatically reduces amyloid beta-peptide deposition. Nat Genet 1997; 17: 263-4. [9] Buttini M, Yu GQ, Shockley K, Huan g Y, Jones B, Masliah E, Mallory M, Yeo T, Longo FM and Mucke L. Modulation of Alzheimer-Like Synaptic and Cholinergic Deficits in Transgenic Mice by Human Apolipoprotein E Depends

PAGE 232

APPENDIX B (Continued) 225 on Isoform, Aging, and Overexpression of Amyloid beta Peptides But Not on Plaque Formation. J Neurosci 2002; 22: 10539-10548. [10] Buxbaum JD, Oishi M, Chen HI, Pinkas-Kramarski R, Jaffe EA, Gandy SE and Greengard P. Cholinergic agonists and interleukin 1 regulate processing and secretion of the Alzheimer beta/A4 amyloid protein precursor. Proc Natl Acad Sci U S A 1992; 89: 10075-8. [11] Champagne D, Dupuy JB, Rochford J and Poirier J. Apolipoprotein E knockout mice display procedural deficit s in the Morris water maze: analysis of learning strategies in three versions of the task. Neuroscience 2002; 114: 64154. [12] Chen G, Chen KS, Knox J, Inglis J, Bernard A, Martin SJ, Justice A, McConlogue L, Games D, Freedman SB and Morris RG. A learning deficit related to age and beta-amyloid plaques in a mouse model of Alzheimer's disease. Nature 2000; 408: 975-9. [13] Corder EH, Saunders AM, Strittmatter WJ, Schmechel DE, Gaskell PC, Small GW, Roses AD, Haines JL and Pericak-Vance MA. Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer's disease in late onset families. Science 1993; 261: 921-3. [14] Das S and Potter H. Expression of the Alzheimer amyloid-promoting factor antichymotrypsin is induced in human astrocytes by IL-1. Neuron 1995; 14: 447-56. [15] DeKosky ST, Ikonomovic MD, Wang X, Farlow M, Wisniewski S, Lopez OL, Becker JT, Saxton J, Klunk WE, Sweet R, Kaufer DI and Kamboh MI.

PAGE 233

APPENDIX B (Continued) 226 Plasma and cerebrospinal fluid alpha1-antichymotrypsin levels in Alzheimer's disease: Correlation with cognitive impairment. Ann Neurol 2003; 53: 81-90. [16] DeMattos RB, O'Dell M A, Parsad anian M, Taylor JW, Harmony JA, Bales KR, Paul SM, Aronow BJ and Holtzman DM. Clusterin promotes amyloid plaque formation and is critical for neuritic toxicity in a mouse model of Alzheimer's disease. Proc Natl Acad Sci U S A 2002; 99: 10843-8. [17] Dodart JC, Bales KR, Gannon KS, Greene SJ, DeMattos RB, Mathis C, DeLong CA, Wu S, Wu X, Holtzman DM and Paul SM. Immunization reverses memory deficits without reducing brain Abeta burden in Alzheimer's disease model. Nat Neurosci 2002; 5: 452-7. [18] Dodart JC, Bales KR, Johnstone EM, Little SP and Paul SM. Apolipoprotein E alters the processing of the beta-amyloid precursor protein in APP(V717F) transgenic mice. Brain Res 2002; 955: 191-9. [19] Dodart JC, Mathis C, Bales KR, Paul SM and Ungerer A. Behavioral deficits in APP(V717F) tran sgenic mice deficient for the apolipoprotein E gene. Neuroreport 2000; 11: 603-7. [20] Duff K, Eckman C, Zehr C, Yu X, Prada CM, Perez-tur J, Hutton M, Buee L, Harigaya Y, Yager D, Morgan D, Gordon MN, Holcomb L, Refolo L, Zenk B, Hardy J and Younkin S. Increased amyloid-beta42(43) in brains of mice expressing mutant presenilin 1. Nature 1996; 383: 710-3. [21] Fagan AM, Watson M, Parsadanian M, Bales KR, Paul SM and Holtzman DM. Human and murine ApoE markedly alters A beta metabolism

PAGE 234

APPENDIX B (Continued) 227 before and after plaque formation in a mouse model of Alzheimer's disease. Neurobiol Dis 2002; 9: 305-18. [22] Forloni G, Demicheli F, Giorgi S, Bendotti C and Angeretti N. Expression of amyloid precursor protein mRNAs in endothelial, neuronal and glial cells: modulation by interleukin-1. Brain Res Mol Brain Res 1992; 16: 12834. [23] Games D, Adams D, Alessandrini R, Barbour R, Berthelette P, Blackwell C, Carr T, Clemens J, Donaldson T, Gillespie F and et al. Alzheimertype neuropathology in transgenic mice overexpressing V717F beta-amyloid precursor protein. Nature 1995; 373: 523-7. [24] Goldgaber D, Harris HW, Hla T, Maciag T, Donnelly RJ, Jacobsen JS, Vitek MP and Gajdusek DC. Interleukin 1 regulates synthesis of amyloid betaprotein precursor mRNA in human endothelial cells. Proc Natl Acad Sci U S A 1989; 86: 7606-10. [25] Gordon MN, King DL, Diamond DM, Jantzen PT, Boyett KV, Hope CE, Hatcher JM, DiCarlo G, Gottschall WP, Morgan D and Arendash GW. Correlation between cognitive deficits and Abeta deposits in transgenic APP+PS1 mice. Neurobiol Aging 2001; 22: 377-85. [26] Hardy J. Pathways to primary neurodegenerative disease. Ann N Y Acad Sci 2000; 924: 29-34. [27] Harper JD, Wong SS, Lieber CM and Lansbury PT. Observation of metastable Abeta amyloid protofibrils by atomic force microscopy. Chem Biol 1997; 4: 119-25.

PAGE 235

APPENDIX B (Continued) 228 [28] Holtzman DM, Bales KR, Tenkova T, Fagan AM, Parsadanian M, Sartorius LJ, Mackey B, Olney J, McKeel D, Wozniak D and Paul SM. Apolipoprotein E isoform-dependent amyloid deposition and neuritic degeneration in a mouse model of Alzheimer's disease. Proc Natl Acad Sci U S A 2000; 97: 2892-7. [29] Janciauskiene S, Rubin H, Lukacs CM and Wright HT. Alzheimer's peptide Abeta1-42 binds to two beta-sheets of alpha1-antichymotrypsin and transforms it from inhibitor to substrate. J Biol Chem 1998; 273: 28360-4. [30] Kamboh MI, Sanghera DK, Fe rrell RE and DeKosky ST. APOE*4associated Alzheimer's disease risk is modified by alpha 1-antichymotrypsin polymorphism. Nat Genet 1995; 10: 486-8. [31] Lambert MP, Barlow AK, Chromy BA, Edwards C, Freed R, Liosatos M, Morgan TE, Rozovsky I, Trommer B, Viola KL, Wals P, Zhang C, Finch CE, Krafft GA and Klein WL. Diffusible, nonf ibrillar ligands derived from Abeta1-42 are potent central nervous system neur otoxins. Proc Natl Acad Sci U S A 1998; 95: 6448-53. [32] Licastro F, Mallory M, Hansen LA and Masliah E. Increased levels of alpha-1-antichymotrypsin in brains of patients with Alzheimer's disease correlate with activated astrocytes and are affected by APOE 4 genotype. J Neuroimmunol 1998; 88: 105-10. [33] Lim GP, Yang F, Chu T, Chen P, Beech W, Teter B, Tran T, Ubeda O, Ashe KH, Frautschy SA and Cole GM. Ibuprofen suppresses plaque pathology

PAGE 236

APPENDIX B (Continued) 229 and inflammation in a mouse model for Alzheimer's disease. J Neurosci 2000; 20: 5709-14. [34] Lukacs CM and Christianson DW. Is the binding of beta-amyloid protein to antichymotrypsin in Alzheimer plaques mediated by a beta-strand insertion? Proteins 1996; 25: 420-4. [35] Ma J, Brewer HB, Jr. and Potter H. Alzheimer A beta neurotoxicity: promotion by antichymotrypsin, ApoE4; inhibition by A beta-related peptides. Neurobiol Aging 1996; 17: 773-80. [36] Ma J, Yee A, Brewer HB, Jr., Das S and Potter H. Amyloid-associated proteins alpha 1-antichymotrypsin and apolipoprotein E promote assembly of Alzheimer beta-protein into filaments. Nature 1994; 372: 92-4. [37] McGeer PL, McGeer E, Rogers J and Sibley J. Anti-inflammatory drugs and Alzheimer disease. Lancet 1990; 335: 1037. [38] Morgan D, Diamond DM, Gottscha ll PE, Ugen KE, Dickey C, Hardy J, Duff K, Jantzen P, DiCarlo G, Wilcock D, Connor K, Hatcher J, Hope C, Gordon M and Arendash GW. A beta peptide vaccination prevents memory loss in an animal model of Alzheimer's disease. Nature 2000; 408: 982-5. [39] Morris RG, Garrud P, Rawlins JN and O'Keefe J. Place navigation impaired in rats with hippocampal lesions. Nature 1982; 297: 681-3. [40] Mrak RE and Griffin WS. Interleukin-1 and the immunogenetics of Alzheimer disease. J Neuropathol Exp Neurol 2000; 59: 471-6. [41] Mucke L, Yu GQ, McConlogue L, Rockenstein EM, Abraham CR and Masliah E. Astroglial expression of human alpha(1)-antichymotrypsin enhances

PAGE 237

APPENDIX B (Continued) 230 alzheimer-like pathology in amyloid prot ein precursor transgenic mice. Am J Pathol 2000; 157: 2003-10. [42] Mueggler T, Sturchler-Pierrat C, Baumann D, Rausch M, Staufenbiel M and Rudin M. Compromised hemodynamic response in amyloid precursor protein transgenic mice. J Neurosci 2002; 22: 7218-24. [43] Namba Y, Tomonaga M, Kawasaki H, Otomo E and Ikeda K. Apolipoprotein E immunoreactivity in cerebral amyloid deposits and neurofibrillary tangles in Alzheimer's disease and kuru plaque amyloid in Creutzfeldt-Jakob disease. Brain Res 1991; 541: 163-6. [44] Nilsson L, Rogers J and Potter H. The essential role of inflammation and induced gene expression in the pathogenic pathway of Alzheimer's disease. Front Biosci 1998; 3: d436-46. [45] Nilsson LN, Bales KR, DiCarlo G, Gordon MN, Morgan D, Paul SM and Potter H. Alpha-1-antichymotrypsin promotes beta-sheet amyloid plaque deposition in a transgenic mouse model of Alzheimer's disease. J Neurosci 2001; 21: 1444-51. [46] Nilsson LN, Das S and Potter H. Effect of cytokines, dexamethasone and the A/T-signal peptide polymorphism on the expression of alpha(1)antichymotrypsin in astrocytes: si gnificance for Alzheimer's disease. Neurochem Int 2001; 39: 361-70. [47] Oda T, Wals P, Osterburg HH, Johnson SA, Pasinetti GM, Morgan TE, Rozovsky I, Stine WB, Snyder SW, Holz man TF and et al. Clusterin (apoJ) alters the aggregation of amyloid beta-peptide (A beta 1-42) and forms slowly

PAGE 238

APPENDIX B (Continued) 231 sedimenting A beta complexes that cause oxidative stress. Exp Neurol 1995; 136: 22-31. [48] Owen EH, Logue SF, Rasmussen DL and Wehner JM. Assessment of learning by the Morris water task and fear conditioning in inbred mouse strains and F1 hybrids: implications of genetic background for single gene mutations and quantitative trait loci analyses. Neuroscience 1997; 80: 1087-99. [49] Poirier J. Apolipoprotein E and Alzheimer's disease. A role in amyloid catabolism. Ann N Y Acad Sci 2000; 924: 81-90. [50] Potter H, Abraham CR and Dressler D, The two Alzheimers amyloid components -antichymotrypsin and -protein form a stab le complex in vitro in K. Iqbal, D. McLachlan, B. Winblad and H. Wisniewski, eds., Alzheimer's Disease; Basic Mechanisms, Dia gnosis and Therapuetic Strategies John Wiley and Sons Ltd., 1991. [51] Potter H, Wefes IM and Nilsson LN. The inflammation-induced pathological chaperones ACT and apo-E ar e necessary catal ysts of Alzheimer amyloid formation. Neurobiol Aging 2001; 22: 923-30. [52] Price DL and Sisodia SS. Mutant genes in familial Alzheimer's disease and transgenic models. Annu Rev Neurosci 1998; 21: 479-505. [53] Qiao X, Cummins DJ and Paul SM. Neuroinflammation-induced acceleration of amyloid deposition in the APPV717F transgenic mouse. Eur J Neurosci 2001; 14: 474-82. [54] Rogers DC, Jones DN, Nelson PR, Jones CM, Quilter CA, Robinson TL and Hagan JJ. Use of SHIRPA and discriminant analysis to characterise

PAGE 239

APPENDIX B (Continued) 232 marked differences in the behavioural phenotype of six inbred mouse strains. Behav Brain Res 1999; 105: 207-17. [55] Rogers JT, Leiter LM, McPhee J, Cahill CM, Zhan SS, Potter H and Nilsson LN. Translation of the alzheimer amyloid precursor protein mRNA is up-regulated by interleukin-1 through 5'-untranslated region sequences. J Biol Chem 1999; 274: 6421-31. [56] Sanan DA, Weisgraber KH, Russell SJ, Mahley RW, Huang D, Saunders A, Schmechel D, Wisniewski T, Frangione B, Roses AD and et al. Apolipoprotein E associates with beta amyloid peptide of Alzheimer's disease to form novel monofibrils. Isoform apoE4 associates more efficiently than apoE3. J Clin Invest 1994; 94: 860-9. [57] Scheuner D, Eckman C, Jensen M, Song X, Citron M, Suzuki N, Bird TD, Hardy J, Hutton M, Kukull W, Larson E, Levy-Lahad E, Viitanen M, Peskind E, Poorkaj P, Schellenberg G, Tanzi R, Wasco W, Lannfelt L, Selkoe D and Younkin S. Secreted amyloid beta-protein similar to that in the senile plaques of Alzheimer's disease is increased in vivo by the presenilin 1 and 2 and APP mutations linked to familial Alzheimer's disease. Nat Med 1996; 2: 864-70. [58] Selkoe DJ. Alzheimer's disease: genes, proteins, and therapy. Physiol Rev 2001; 81: 741-66. [59] Strittmatter WJ and Roses AD. Apolipoprotein E and Alzheimer disease. Proc Natl Acad Sci U S A 1995; 92: 4725-7.

PAGE 240

APPENDIX B (Continued) 233 [60] Strittmatter WJ, Saunders AM, Schmechel D, Pericak-Vance M, Enghild J, Salvesen GS and Roses AD. Apolipoprotein E: high-avidity binding to betaamyloid and increased frequency of type 4 allele in late-onset familial Alzheimer disease. Proc Natl Acad Sci U S A 1993; 90: 1977-81. [61] Telling GC, Scott M, Mastrianni J, Gabizon R, Torchia M, Cohen FE, DeArmond SJ and Prusiner SB. Prion propagation in mice expressing human and chimeric PrP transgenes implicates the interaction of cellular PrP with another protein. Cell 1995; 83: 79-90. [62] Terai K, Iwai A, Kawabata S, Sa samata M, Miyata K and Yamaguchi T. Apolipoprotein E deposition and astrogliosis are associated with maturation of beta-amyloid plaques in betaAPPswe tran sgenic mouse: Implications for the pathogenesis of Alzheimer's disease. Brain Res 2001; 900: 48-56. [63] Thomas T, Thomas G, McLendon C, Sutton T and Mullan M. betaAmyloid-mediated vasoactivity and vascular endothelial damage. Nature 1996; 380: 168-71. [64] Veinbergs I and Masliah E. Synaptic alterations in apolipoprotein E knockout mice. Neuroscience 1999; 91: 401-3. [65] Walsh DM, Klyubin I, Fadeeva JV, Cullen WK, Anwyl R, Wolfe MS, Rowan MJ and Selkoe DJ. Naturally secreted oligomers of amyloid beta protein potently inhibit hippocampal long-term potentiation in vivo. Nature 2002; 416: 535-9.

PAGE 241

APPENDIX B (Continued) 234 [66] Walsh DM, Lomakin A, Benedek GB, Condron MM and Teplow DB. Amyloid beta-protein fibrillogenesis. Detection of a protofibrillar intermediate. J Biol Chem 1997; 272: 22364-72. [67] Walsh DM, Tseng BP, Rydel RE, Podlisny MB and Selkoe DJ. The oligomerization of amyloid beta-protein begins intracellularly in cells derived from human brain. Biochemistry 2000; 39: 10831-9. [68] Wang HW, Pasternak JF, Kuo H, Ristic H, Lambert MP, Chromy B, Viola KL, Klein WL, Stine WB, Krafft GA and Trommer BL. Soluble oligomers of beta amyloid (1-42) inhibit long-term potentiation but not long-term depression in rat dentate gyrus. Brain Res 2002; 924: 133-40. [69] Webster S and Rogers J. Relative efficacies of amyloid beta peptide (A beta) binding proteins in A beta aggregation. J Neurosci Res 1996; 46: 58-66. [70] Wei A, Rubin H, Cooperman BS, Schechter N and Christianson DW. Crystallization, activity assay and prelim inary X-ray diffraction analysis of the uncleaved form of the serpin antichymotrypsin. J Mol Biol 1992; 226: 273-6. [71] Wisniewski T, Aucouturier P, Soto C and Frangione B. The prionoses and other conformational disorders. Amyloid 1998; 5: 212-24. [72] Wisniewski T, Castano EM, Golabek A, Vogel T and Frangione B. Acceleration of Alzheimer's fibril formation by apolipoprotein E in vitro. Am J Pathol 1994; 145: 1030-5. [73] Wisniewski T and Frangione B. Apolipoprotein E: a pathological chaperone protein in patients with cerebral and systemic amyloid. Neurosci Lett 1992; 135: 235-8.

PAGE 242

APPENDIX B (Continued) 235 [74] Xu PT, Gilbert JR, Qiu HL, Ervin J, Rothrock-Christian TR, Hulette C and Schmechel DE. Specific regional transcription of apolipoprotein E in human brain neurons. Am J Pathol 1999; 154: 601-11. [75] Yamada M, Sodeyama N, Itoh Y, Suematsu N, Otomo E, Matsushita M and Mizusawa H. Association of alpha1-antichymotrypsin polymorphism with cerebral amyloid angiopathy. Ann Neurol 1998; 44: 129-31. [76] Yamin R, Malgeri EG, Sloane JA, McGraw WT and Abraham CR. Metalloendopeptidase EC 3.4.24.15 is necessary for Alzheimer's amyloid-beta peptide degradation. J Biol Chem 1999; 274: 18777-84.

PAGE 243

APPENDIX B (Continued) 236 Table 1. ELISA measurements of A (1-40) and A (1-42) levels in the hippocampus of 2 months old mice. Genotype n PBS-extractabl e Guanidine-extractable (mean pg/mg tissue) (mean pg/mg tissue) A 1-40 A 1-42 A 1-40 A 1-42 Hippocampus PDAPP+ApoE+ACT+ 20 103 21 65 7 312 18 238 10 PDAPP+ApoE+ACT12 148 17 78 9 356 25 303 29* PDAPP+ApoE-ACT+ 17 167 29 91 11* 310 23 262 16 PDAPP+ApoE-ACT19 151 33 92 9* 349 39 288 27 Cerebral Cortex PDAPP+ApoE+ACT+ 20 18 2 10 1 181 12 165 12 PDAPP+ApoE+ACT12 19 3 11 1 186 12 183 12 PDAPP+ApoE-ACT+ 17 24 2 16 1** 115 5** 157 11 PDAPP+ApoE-ACT19 27 3* 16 1** 133 11** 171 18

PAGE 244

APPENDIX B (Continued) 237 Values are mean SEM; *P<0.05 as compared to PDAPP+ApoE+ACT+ transgenic mice for the hippocampal measurements. *P<0.05 and **P<0.01 as compared to PDAPP+ApoE+ACT+ as well as PDAPP+ApoE+ACTtransgenic mice for the measurements of the cerebral cortex.

PAGE 245

APPENDIX B (Continued) 238 6. Figure legends Figure 1 ELISA measurements of A (1-40) and A (1-42) levels in hippocampus and behavior analysis of cognitive ability in 2 months old mice before amyloid deposition has begun. (a-d) Quantitation of the levels of A (1-40) and A (1-42) in PDAPP+,(mApoE+),ACT+ (n=20, solid bar), PDAPP+(mApoE+)(ACT-) (n=12, open bar), PDAPP+,mApoE-KO,ACT+ (n=1 7, solid bar) and PDAPP+,mApoEKO,(ACT-) (n=19, open bar). The quantitative measurements refer to soluble (PBS-extractable) and membrane-bound (guanidine extractable) A levels. *P<0.05. (e,f) Morris Water Maze and radial arm water maze performance of young adult APP+.(mApoE+)(ACT-) mice and non-transgenic littermates. Figure 2. Total amyloid load, diffuse and compact, in four strains of 18 months old transgenic/knockout mice, as measured by 6E10 immunostaining. Representative animals with genotypes (a) PDAPP+,(mApoE+),ACT+, (b) PDAPP+,(mApoE+)(ACT-), (c) PDAPP+,mApoE-KO,ACT+ and (d) PDAPP+,mApoE-KO,(ACT-). (e) Quantitative image analysis of total amyloid load from all investigated animals with genotype PDAPP+,(mApoE+),ACT+ (n=17, solid bar) and PDAPP+,(mApoE+)(ACT-) (n=11, open bar) in the hippocampus (left) and the cerebral cortex (middle), and PDAPP+,mApoE-KO,ACT+ (n=9, solid bar) and

PAGE 246

APPENDIX B (Continued) 239 PDAPP+m,ApoE-,(ACT-) (n=8, open bar) in the hippocampus (right). *P<0.05. Scale bar measures 335 m. Figure 3 Compact amyloid deposition in four strains of 18 months old transgenic/ knockout mice as measured by Congo Red staining Representative animals of the experimental groups (a) PDAPP+,(mApoE+),ACT+ and (b) PDAPP+,(mApoE+)(ACT-). Quantitative compact amyloid load fr om all investigated animals with genotypes PDAPP+,(mApoE+),A CT+ (n=17, solid bar) and PDAPP+,(mApoE+)(ACT-) (n=11, open bar) in the hippocampus and the cerebral cortex. Congo positive amyloid load is expressed as % compact A load (c) and plaque density (d). *P<0.05. Scale bar measures 115 m. Figure 4. Amyloid pathology of an 18 months old heterozygous PDAPP+(mApoE+)ACT+ (a-d) and a 23 months old homozygous PDAPP+,mApoE-KO,ACT+ transgenic mouse (e-h). A (a and e), ApoE(b and f), and ACT-immunostaining (d and h) and Thioflavine S staining (c and g). Figure 5. Analysis of spatial learning in the Morris Water Maze Average escape latency for four daily trials during block 1 (days 1-3), block 2 (days 4-6) and block 3 (days 7-9) of testing. PDAPP+,(mApoE+),ACT+ (n=11; yellow squares), PDAPP+,(mApoE+)(ACT-) (n=10; green squares), PDAPP+,mApoE-KO,ACT+ (n=10; bl ue circles) and PDAPP+,mApoE-

PAGE 247

APPENDIX B (Continued) 240 KO,(ACT-) (n=7; red circles) and nontransgenic mice (n=5; pink triangles). Statistical significance refers to P<0.05 or higher level of significance as compared to nontransgenic mice (*), PDAPP+,ApoE-,(ACT-) mice (#) or PDAPP+,ApoE-,ACT+ mice ( ). Figure 6. Analysis of working memory in the Radial Arm Water Maze (a) Escape latency during block 1 (days 1-3), block 2 (days 4-6) and block 3 (days 7-9) of RAWM testing for the first and last acquisition trials (Trial 1, Trial 4), and the delayed memory retention trial (Trial 5). (b) Combined average escape latencies of the memory retention (T5) trials over all three blocks of testing. (c) Correlation analysis of escape laten cy for Trial 5 over all three blocks and total hippocampal amyloid deposition (6E10 immunoreactivity). (d) Correlation analysis of escape latency for T5 over all three blocks and compact amyloid deposition (Congo Red staining). (e) Escape Latency during block 1 (days 1-3), block 2 (days 4-6) and block 3 (days 7-9) of RAWM testing for PDAPP+,mApoE-KO transgenic mice stratified for the presence (n=5; green squares) or absence (n=11; blue squares) of diffuse hippocampal amyloid deposition as well as nontransgenic mice (n=5; pink triangles). Statistical significance in panels (a )and e refers to P<0.05 or higher level of significance as compared to PDAPP+,ApoE-,(ACT-) (#) or nontransgenic mice (*).

PAGE 248

APPENDIX B (Continued) 241 Statistical significance in panel (b) refers to P<0.05 (*) and P<0.01 (**). Dashed lines in panels (c) and (d) indicate 95% confidence interval of regression line. Figure 7. Discriminant analysis (a) Graphical demonstration of the ability of discriminant analysis to easily discriminate three different nontransgenic mice strains based on 19 measures of behavior. (b) Lack of ability to behaviorally discriminate the four groups of APP transgenic mice and the nontransgenic mice based on the same 19 behavioral measures.

PAGE 249

APPENDIX B (Continued) 242 0 50 100 150 200ACT+ ACTACT+ ACTACT+ ACTACT+ ACT* Soluble A (1-40) Soluble A (1-42)A (pg/mg tissue) APP+,(mApoE+) APP+ApoE-KO APP+,(mApoE+) APP+,ApoE-KO 0 100 200 300 400ACT+ ACTACT+ ACTACT+ ACTACT+ ACTAPP+,(mApoE+) APP+ApoE-KO APP+,(mApoE+) APP+,ApoE-KOMembrane Bound A (1-40)Membrane Bound A (1-42) *A (pg/mg tissue) 20 30 40 50 60Nontransgenic APP+,(mApoE+),ACTDay 1-3 Day 4-6 Day 7-9 Morris Water MazeEscape Latency (sec) 20 30 40 50 60APP+,(mApoE+),ACTNontransgenic Day 1-3 Day 4-6 Day 7-9 T1 T4 T5 T1 T4 T5 T1 T4 T5 Escape Latency (sec)a b c d e f Fig. 1

PAGE 250

APPENDIX B (Continued) 243 0.0 2.5 5.0 7.5 10.0 *ACT+ ACTACT+ ACTACT+ ACT*APP+,ApoE+ APP+,ApoE+ APP+,ApoEHippocampus Cerebral Cortex Hippocampus% 6E10 positive Amyloid load Fig. 2

PAGE 251

APPENDIX B (Continued) 244 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 **Congo-positive plaque loadACT+ ACTACT+ ACTHippocampus Cerebral Cortex APP+,ApoE+ APP+,ApoE+ *% Congo-positive Amyloid load 0 1 2 3 4 5 6 7 *Congo-positive plaque frequencyACT+ ACTACT+ ACTHippocampus Cerebral Cortex APP+,ApoE+ APP+,ApoE+Congo positive Amyloid plaques/mm2cd Fig 3

PAGE 252

APPENDIX B (Continued) 245 Fig. 4

PAGE 253

APPENDIX B (Continued) 246 20 30 40 50 60 APP+,ApoE-KO,ACTAPP+,ApoE-KO,ACT+ APP+,(mApoE),ACTAPP+,(mApoE),ACT+ Nontransgenic Day 1-3 Day 4-6 Day 7-9 *# #* Escape Latency (sec) Fig. 5

PAGE 254

APPENDIX B (Continued) 247 20 30 40 50 60 APP+,ApoE-KO,ACTAPP+,ApoE-KO,ACT+ APP+,(mApoE),ACTAPP+,(mApoE),ACT+ Nontransgenic Day 1-3 Day 4-6 Day 7-9 T1 T4 T5 T1 T4 T5 T1 T4 T5Radial Arm Water Maze* * * *# # #* * * * *# # # #*# # # # #*Escape latency (sec)a c d e 20 30 40 50 60 no amyloid diffuse amyloid nontransgenic T1 T4 T5 T1 T4 T5 T1 T4 T5 Day 1-3 Day 4-6 Day 7-9 *# #Escape Latency (sec) 20 30 40 50 60 2.5 5.0 7.5 10.0 12.5 00.02 r=0.68 P<0.001 6E10 Amyloid Load (%)Escape Latency (sec) 20 30 40 50 60 APP+,ApoE-KO,ACTAPP+,ApoE-KO,ACT+ APP+,(mApoE),ACTAPP+,(mApoE),ACT+ 0.1 0.2 0.3 0.4 0.5 0.6 0.7 Congo Amyloid Load (%) 0 r=0.60 P<0.001Escape Latency (sec) ACT+ACT-ACT+ACT-NT 0 10 20 30 40 50 ***APP+,(ApoE+) APP+,ApoE-KOEscape Latency (sec)b Fig. 6

PAGE 255

APPENDIX B (Continued) 248 Fig. 7 Previously Published NeuroReport Volume 15 Number 12 (2004)

PAGE 256

APPENDIX B (Continued) 249 Environmental Enrichment Improves Cognition in Alzheimer’s Transgenic Mice Despite Stable -Amyloid Deposition 1G.W. Arendash, 1*M.F. Garcia, 2*D.A. Costa, 1J.R. Cracchiolo, 2I.M. Wefes, and H. Potter2 1Memory and Aging Research Laboratory, SCA 110 and 2Department of Biochemistry & Molecular Biology and Suncoast Gerontology Center, MDC 7 University of South Florida, Tampa, FL 33620 These authors contributed equally to this work Correspondence should be addressed to G.W.A. ( arendash@chuma.cas.usf.edu ) Phone: (813) 974-1584 Fax: (813) 974-3263 Date of Submission: 6/23/03

PAGE 257

APPENDIX B (Continued) 250 Rodents undergoing cognitive stimulation (CS) by being placed in an “enriched environment” (e.g., socially-housed in large cages containing toys, tunnels, running wheels, etc.) exhibit improved cognitive performance1,2, as well as increased hippocampal neurogenesis3, synaptogenesis4, growth factor levels5, and cognition-linked gene expression6. Although retrospective studies in humans have similarly found that a lifelong pattern of CS generally associated with educational/occupational attainment protects against Alzheimer’s Disease (AD)7-8, it is unclear to what extent long-term CS would benefit individuals who already have AD. Here we show that long-term environmental enrichment (EE) in aged AD transgenic mice results in global, overall improvement in cognitive function without decreasing brain betaamyloid (A ) deposition, suggesting that long-term CS could provide cognitive stabilization or improvement to AD patients through mechanisms independent of A deposition and clearance. Recently, several cognitive stimulation programs have been developed in an attempt to slow or reverse the cognitive decline in AD patients9,10. These programs have generally involved a relatively short 4-12 week period of cognitive training classes and/or daily caregiver-directed mental activities, resulting in modest degree s of success. However, the therapeutic potential of intensive, “long-term” CS has not yet been evaluated in AD patients. Furthermore, retrospective/longitudinal studies cannot determine whether CS is a promotor or merely an indicator of intact cognition. Therefore, the present study sought to elucidate the cognitive-enhancing potential of long-term EE in

PAGE 258

APPENDIX B (Continued) 251 aged APPsw transgenic mice, which overexpress the “Swedish” mutant amyloid precursor protein and bear moderate cortical/hippocampal A deposition within mature AD-like neuritic plaques11. Beginning at 16 months of age, APPsw mice were put into either EE ( n = 5) or standard housing (SH; n = 4) for four months (Fig. 1). In addition to their enriched housing, EE mice were placed in novel complex environments 3 times weekly for several hours over the 4 month period. Beginning at 20 months of age, all mice were tested in four cognitive-based tasks as previously described11: Morris water maze (reference learning and memory), circular platform (reference memory), platform recognition (search/identification), and radial arm water maze (working memory). Following completion of the behavioral testing at 22 months of age, animals were euthanized and their brains processed for A load determinations12. During the course of enrichment, 2 EE mice died. Over 10 days of Morris maze acquisition (learning), EE mice performed significantly better than SH mice by ha ving lower escape latencies (Fig. 2a). During the subsequent memory retention (probe) trial (Fig. 2b), EE mice showed an exclusive preference for the former platform-containing quadrant (p = 0.01, ANOVA), while SH mice showed nominal quadrant preference (p = n.s., ANOVA). In switching from the “reference” learning/memory strategy of the Morris maze to the search/identification strategy of platform recognition, SH mice showed obvious impairment, as evidenced by their high latencies over the first 3 days of platform recognition testing (Fig. 2c). By sharp contrast, EE

PAGE 259

APPENDIX B (Continued) 252 mice effectively and immediately changed strategies by exhibiting low escape latencies even on the first day of platform recognition testing (p<0.05; Fig. 2c). Indeed, a comparison of the improved behavior of EE mice in the above tasks indicates that they had r eacquired the cognitive ability of previously-studied 16 month old non-transgenic mice with the same strain background11. To determine whether EE had an “overall” cognitive-enhancing effect encompassing all 4 tasks employed, we evaluated data from the two primary measures of each task. As shown in Fig. 3a, the means of EE mice on all 8 measures were better than those of SH mice. An analysis of these data showed that EE mice had significantly better “overall” cognitive performance spanning all 8 cognitive measures (p <0.005; Fisher Sign test). Moreover, discriminant function analysis (step-wise forward method), evaluating data from the same 8 behavioral measures, also revealed significantly better overall performance of EE vs. SH mice (F[4,2] = 33.57, p = 0.029; Wilks’ lambda = 0.0147); the Systat prog ram found four measures from 3 of the 4 tasks to collectively provide maximal discrimination between EE and SH groups. At the 22 month completion age of this study, total A loads (diffuse + compact) in both parietal cortex and hippocampus of EE and SH mice were determined by immunohistochemistry with the 6E10 antibody. Moderate A loads (5-10%) were evident in both brain areas, with no differences between EE and SH mice (Fig. 3b), indicating that the cognitive benefits of EE occurred in AD transgenic mice without an accompanying decrease in brain A deposition.

PAGE 260

APPENDIX B (Continued) 253 These results indicate that a long-term period of EE in aged AD transgenic mice results in superior overall cognitive performance encompassing multiple cognitive domains (e.g., reference learning/memory, working memory, recognition, strategy switching). Although a reduction in brain A deposition has been associated with behavioral benefit of A immunotherapy in simi lar AD transgenic mice13, the present results, together with our other study showing A -independent cognitive benefit to AD transgenic mice given a blueberry-rich diet14, indicate that mechanisms autonomous from A deposition are sufficient for behavioral benefit. Along this line, EE-induced enhancements in neurogenesis3, synaptogenesis4, growth factor levels5, and gene expression6 have all been seen in normal mice. In this first study to evaluate the potential of EE in AD transgenic mice, our results suggest that long-term intensive EE/CS (alone or in combination with cognition-stimulating drugs) could be useful in stabilizing or slowing the cognitive decline of AD and it’s predecessor, mild cognitive impairment (MCI), without needing to reduce A load. Largely because of practical considerations, longitudinal interventional studies have not been done in humans to confirm controversial retr ospective studies re porting AD risk reduction with “lifelong” education/occupation-related intellectual activity7,8. A recent 5-year longitudinal study involving non-demented 75+ year olds did find participation in cognitively-stimulating leisure activities to be associated with a lower risk of dementia15. However, a cause and effect relationship could not be established because the leisure activities were self chosen. Future studies

PAGE 261

APPENDIX B (Continued) 254 in which EE is initiated early in adulthood of AD transgenic mice should more definitively elucidate the potential protective effects of EE against development of cognitive impairment and neuropathology. Acknowledgements This research was supported by Alzheimer’s Disease and Related Diseases Association (ADRDA) grant # IIRG-02-3778. Additional funding was provided by a grant from the NIA (AG09665). H.P. occupies the Eric Pfeiffer Chair for Research in Alzheimer’s Disease at the Suncoast Gerontology Center at USF. Competing Interests statement The authors declare that they have no competing financial interests.

PAGE 262

APPENDIX B (Continued) 255 References 1. Williams, B. et al., Physiol. And Behav. 73 649-658 (2001). 2. Kobayashi, S., Ohashi, Y., and Ando, S. J. Neurosci. Res. 70 340-346 (2002). 3. Kempermann, G., Gast, D., and Gage, F. Ann. Neurol. 52 135-143 (2002). 4. Ramirez-Amaya, V. et al., J. Neuroscience 21 m7340-7348 (2001). 5. Ickes, B. et al., Exp. Neurol. 164 45-52 (2000). 6. Rampon, C. et al., Proc. Nat’l Acad. Sci. 97 12880-12884 (2000). 7. Letenneur, L. et al., J. Neurol., Neurosurg. & Psychi. 66 177-183 (1999). 8. Friedland, R. et al., Proc. Nat’l Acad. Sci 98 3440-3445 (2001). 9. Davis, R., Massman, P., and Doody, R. Alzheimer Disease & Assoc. Disorders 15 1-9 (2001). 10. de Rotrou, J. et al., Brain Aging 2 48-53 (2002). 11. Arendash, G. et al., Brain Res. 891 42-53 (2001). 12. Nilsson, L. et al., J. Neuroscience 21 1444-1451 (2001). 13. Morgan, D. et al., Nature 408 982-985 (2000). 14. Joseph, J., Denisova, N., Arendash, G.W. et al., Nut. Neurosci. 6 153-162 (2003). 15. Verghese, J. et al., New Engl. J. Med. 348 2508-2516 (2003).

PAGE 263

APPENDIX B (Continued) 256 Figure Legends Fig. 1: Standard housing of individual mice (a) and enriched housing of multiple mice in a large bin containing an inner cage (with platforms, passageways, lofts), running whee ls, toys, and novel habitats (b) Fig. 2: Environmental enrichment significantly improves reference learning, reference memory and recognition/identification abilities of AD transgenic mice. (a) Over 10 days of Morris maze acquisition (learning) at 20 months of age, EE transgenic mice had lower escape latencies compared to SH transgenic mice. Latencies to escape to a stationary submerged platform in a 100-cm diameter pool over 4 daily trials were averaged and are presented as five 2-day blocks. Group means were compared across blocks by Paired De sign [ t(4) = 4.99; p=0.007] (b) Spatial memory retention of EE transgenic mice during a 60 sec. probe trial on Day 11 was better than SH transgenic mice since EE mice had an exclusive preference for the former platform-containing quadrant (Q2) while SH mice showed only nominal quadrant preference. = significantly lower than Q2 percentage at p<0.05 or higher level of significance (ANOVA). There was no difference in swim speed between EE and SH transgenic groups. (c) For 4 days of platform recognition testing at 21 months of age, transgenic mice were given 4 swim trials daily starting from the same position, with an elevated prominentlyensigned escape platform moved to a different pool quadrant for each

PAGE 264

APPENDIX B (Continued) 257 trial. EE mice were immediately able to switch from the reference learning/memory strategy of Morris maze to the search/identification strategy of platform recognition, as evidenced by their low escape latencies even on the first day of testing. By contrast, SH mice had significantly higher escape latencies on the first 3 days of testing. = significantly higher than enriched group at p<0.05 (Mann-Whitney U test). Fig. 3: Environmental enrichment results in better “overall” cognitive performance in aged APPsw mice without affecting forebrain A deposition. (a) EE mice had higher performan ce means vs. SH mice on two primary measures from each of four cognitive-based tasks evaluated between 20 22 months of age. Significantly better overall performance of EE mice across those 8 measures was evident from both the Fisher Sign test and discriminant function analysis (see text). Abbreviations: WM acq. = Morris maze acquisition average escape latency; CP overall (x10) = circular platform average escape latency (1/10th scale); CP Final (x10) = circular platform escape latency on final day (1/10th scale); PR overall = platform recognition average escape latency; PR final = platform recognition escape latency on final day; RM overall (T5) = radial arm water maze average escape latency on the delayed retention trial; RM final (T5) = radial arm water maze escape latency on delayed retention trial in last block of testing; WM retention

PAGE 265

APPENDIX B (Continued) 258 deficit = % of time not spent in the former platform-containing quadrant during probe trial. (b) At 22 months of age, total A loads in both hippocampus and parietal cortex of APPsw mice were not affected by 4 months of EE.

PAGE 266

APPENDIX B (Continued) 259 Figure 1

PAGE 267

APPENDIX B (Continued) 260 Figure 2 a) b) Quadrant % Time in Quadrant 0 10 20 30 40 50 60 1 2 3 4 Quadrant 1234 * *Standard Housing Enriched Environment c) Da y s 1234Latency (sec) 0 10 20 30 40 *Enriched Environment*Standard Housing Blocks 12345Latency (sec) 20 30 40 50 60 Standard Housing Enriched Environment

PAGE 268

APPENDIX B (Continued) 261 Figure 3 Escape Latency (sec) 0 10 20 30 40 50 60 Standard Housing Enriched Environment W M a c q C P o v e r a l l ( x 10 ) C P F i n a l ( X 1 0 ) P R o v e r a l l P R f i n a l R M o v e r a l l ( T 5 ) R M f i n a l ( T 5 ) % Time outside platform quadrant 0 20 40 60 W M r e t e n t i o n d e f i c i t Total Ab load (% area) 0 2 4 6 8 10 Standard Housing Enriched Environment Hippocampus Parietal Cortex

PAGE 269

ABOUT THE AUTHOR David Costa was born in Connecticut, but has lived most of his life in Florida. He gradua ted, with honors, from the Univ ersity of Florida in 1999, earning a Bachelors degree in Microbiology He earned his doctoral degree in medical sciences from the University of South Florida, Co llege of Medicine, Department of Biochemistry and Molecular Biology in 2005. David is a member of the Society for Neuroscience, The Alzheimers Association, and shares affiliations with the Johnnie B. Byrd, Sr. Alzheimer's Center & Research Institute and the Suncoast Gerontology Center. David currently resides in Brandon, Florida with his wife, Heather. 262


xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001670320
003 fts
005 20051216093244.0
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 051110s2005 flu sbm s000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001179
035
(OCoLC)62253415
SFE0001179
040
FHM
c FHM
049
FHMM
090
QD415 (Online)
1 100
Costa, David Antonio.
0 245
Promoting and preventing alzheimer's disease in a transgenic mouse model
h [electronic resource] :
b apolipoprotein e and environmental enrichment /
by David Antonio Costa.
260
[Tampa, Fla.] :
University of South Florida,
2005.
500
Includes vita.
502
Thesis (Ph.D.)--University of South Florida, 2005.
504
Includes bibliographical references.
516
Text (Electronic thesis) in PDF format.
538
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
Title from PDF of title page.
Document formatted into pages; contains 269 pages.
520
ABSTRACT: Besides age, inheritance of the apoE-E4 allele is the main risk factor for late-onset AD. To determine the role of apoE in amyloid deposition, we studied mice expressing both mutant human amyloid [beta]-protein precursor (APP) and presenilin 1 (PS1) that were either normal or knocked-out for apoE. By 7 months, amorphous A[beta] deposition developed equally in both lines, indicating that A[beta] alone is sufficient for deposition to occur. In contrast, filamentous amyloid deposition was catalyzed at least 3000 fold by apoE. Electron micrographs further illustrate the filamentous nature of these plaques. These results and other, behavioral, data indicate that the primary function of apoE in AD is to promote the polymerization of A[beta] into mature, neurotoxic, amyloid. ApoE is also synthesized in the liver and is crucial in cholesterol metabolism, for mice lacking apoE exhibit hypercholesterolemia.We investigated neuropathology in mice using an uncommon technique, parabiosis, to determine whether apoE in the peripheral circulation influences brain amyloid formation. This surgical procedure allows exchange of proteins via peripheral circulation. We show that plasma apoE is found in parabiosed PS/APP/apoE-KO mice, rescuing their hypercholesterolemia. Unexpectedly, amyloid deposition is reduced in parabiosed PS/APP/apoE-KO mice compared to PS/APP controls. ApoE in the periphery seems to slightly reduce amyloid burden, by likely promoting efflux of A[beta];from the brain. These findings reinforce that the mechanisms whereby apoE affects A[beta] metabolism are complex, and the modulation of peripheral apoE metabolism is not likely to impact AD neuropathology. Since cognitive stimulation is associated with lower risk of AD, we sought to investigate the preventative potential of environmental enrichment (EE) using our mouse model.
590
Adviser: Huntington Potter.
653
Alzheimer's disease.
Transgenic mouse.
Amyloid.
Apolipoprotein e.
Parabiosis.
Environmental enrichment.
690
Dissertations, Academic
z USF
x Biochemestry and Molecular Biology
Doctoral.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.1179