USF Libraries
USF Digital Collections

Recombinant AAV gene therapy and delivery for Alzheimer's disease

MISSING IMAGE

Material Information

Title:
Recombinant AAV gene therapy and delivery for Alzheimer's disease
Physical Description:
Book
Language:
English
Creator:
Carty, Nikisha Christine
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
Alzheimer Disease -- metabolism   ( mesh )
Alzheimer Disease -- pathology   ( mesh )
Alzheimer Disease -- therapy   ( mesh )
Amyloid Precursor Protein Secretases -- metabolism   ( mesh )
Blotting, Western   ( mesh )
Dependovirus -- genetics   ( mesh )
Gene Therapy   ( mesh )
Gene Transfer Techniques   ( mesh )
Genetic Vectors   ( mesh )
Hippocampus   ( mesh )
Immunoenzyme Techniques   ( mesh )
Presenilin-1 -- metabolism   ( mesh )
Aspartic Acid Endopeptidases -- genetics   ( mesh )
Metalloendopeptidases -- genetics   ( mesh )
Senile Plaques   ( mesh )
Mice   ( mesh )
Mice, Transgenic   ( mesh )
Beta amyloid
Amyloid degrading enzyme
Convection enhanced delivery
Mannitol
Adeno-associated viral vector
Transgenic mice
Dissertations, Academic -- Molecular Pharmacology and Physiology -- Doctoral -- USF   ( lcsh )
Genre:
non-fiction   ( marcgt )

Notes

Summary:
ABSTRACT: Alzheimer's disease (AD), first characterized in the early 20th century, is a common form of dementia which can occur as a result of genetic mutations in the genes encoding presenilin 1, presenilin 2, or amyloid precursor protein (APP). These genetic alterations can accelerate the pathological characteristics of AD, including the formation of extracellular neuritic plaques composed of amyloid beta peptides and the formation of intracellular neurofibrillary tangles consisting of hyperphosphorylated tau protein. Ultimately, AD results in gross neuron loss in the brain which is evidenced clinically as a progressive decline in mental capacity. A strong body of scientific evidence has previously demonstrated that the driving factor in the pathogenesis of AD is potentially the accumulation of Aβ peptides in the brain. Thus, reduction of Aβ deposition is a major therapeutic strategy in the treatment of AD.Recently it has been suggested that Aβ accumulation in the brain is modulated, not only by Aβ production, but also by its degradation. Several important studies have demonstrated that Aβ degradation is modulated by several endogenous zinc metalloproteases shown to have amyloid degrading capabilities. These endogenous proteases include neprilysin (NEP), endothelin converting enzyme (ECE), insulin degrading enzyme (IDE) and matrix metalloprotease 9 (MMP9). In this investigation we study the effects of upregulating expression of several of these proteases through administration of recombinant adeno-associated viral vector (rAAV) containing both endogenous and synthetic genes for ECE and NEP on amyloid deposition in amyloid precursor protein (APP) plus presenilin-1 (PS1) transgenic mice. rAAV administration directly into the brain resulted in increased expression of ECE and NEP and a substantial decrease in amyloid pathology.We were able to significantly increase the area of viral distribution by using novel delivery methods resulting in increased gene expression and distribution. These data support great potential of gene therapy as a method of treatment for neurological diseases. Optimization of gene transfer methods aimed at a particular cell type and brain region in the CNS can be accomplished using AAV serotype specificity and novel delivery techniques leading to successful gene transduction thus providing a promising therapeutic avenue through which to treat AD.
Thesis:
Dissertation (Ph.D.)--University of South Florida, 2009.
Bibliography:
Includes bibliographical references.
System Details:
Mode of access: World Wide Web.
System Details:
System requirements: World Wide Web browser and PDF reader.
Statement of Responsibility:
by Nikisha Christine Carty.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 193 pages.
General Note:
Includes vita.

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 - 002063793
oclc - 557379496
usfldc doi - E14-SFE0003026
usfldc handle - e14.3026
System ID:
SFS0027343:00001


This item is only available as the following downloads:


Full Text

PAGE 1

Recombinant AAV Gene Therapy and Delivery for Alzheimer’s Disease Nikisha Christine Carty A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Molecular P harmacology and Physiology College of Medicine University of South Florida Major Professor: David Morgan, Ph.D. Marcia N. Gordon, Ph.D. Keith Pennypacker, Ph.D. Edwin Weeber, Ph.D. Amyn Rojiani, M.D., Ph.D. Date of Approval May 19, 2009 Keywords: Beta amyloid, amyloid degr ading enzyme, convection enhanced delivery, mannitol, adeno-associat ed viral vector, transgenic mice Copyright 2009, Nikisha Christine Carty

PAGE 2

i TABLE OF CONTENTS LIST OF FIGURES iii LIST OF ABBREVIATIONS v ABSTRACT vi INTRODUCTION Alzheimer’s Disease 1 Amyloid Precursor Protein (APP) and Presenilin 1 and 2 3 Transgenic Mouse Models of Amyloid Deposition 6 Zinc Metalloproteases/ Endogenous A Degrading Enzymes 8 Gene Therapy and Viral Vectors 13 Adeno-associated Viral Vectors and Serology 16 Enhancing Distribution of Gene Product 20 PAPER 1: ADENO-ASSOCIATED VI RAL (AAV) SEROTYPE 5 VECTOR MEDIATED GENE DELIVERY OF ENDOTHELIN CONVERTING ENZYME REDUCES A DEPOSITS IN APP +PS-1 TRANSGENIC MICE 23 Abstract 24 Introduction 25 Results 28 Discussion 31 Materials and Methods 36 Generation of ECE Constr ucts and rAAV Production Western Blot Analysis 37 Transgenic Mice Surgical Procedure 38 Immunohistochemistry 39 Enzyme Activity Assay 41 PAPER 2: ADENO-ASSOCIATED VI RAL VECTOR MEDIATED GENE DELIVERY OF SECRETED NEPRILYSIN REDUCES -AMYLOID DEPOSITION IN APP + PS1 TRANSGENIC MICE 55 Abstract 57 Introduction 58 Materials and Methods 61 Generation of NEP Gene Constructs and rAAV Production Transgenic Mice 62 Surgical Procedure 63 Study 1

PAGE 3

ii Step Design Cannula 64 Study 2 Immunohistochemistry Enzyme Activity Assay 67 Results 68 Discussion 75 PAPER 3: CONVECTION-ENHANCED DELIVERY AND MANNITOL AS A METHOD TO INCREASE DISTRIBUTION OF AAV VECTORS 5, 8, AND 9 AND INCREASE GENE PRODUCT IN THE ADULT MOUSE BRAIN 104 Abstract 105 Introduction 107 Materials and Methods 110 Animals Step Design Cannula 111 GFP Expression Using CED GFP Expression with Serotypes AAV 5, 8, and 9 114 GFP Expression with Serotypes AAV 5 or 9 and Mannitol 115 Quantification and Statistical Analysis Results 116 Discussion 123 CONCLUSIONS 150 REFERENCES 176 ABOUT THE AUTHOR End Page

PAGE 4

iii LIST OF FIGURES PAPER 1 Figure 1: Diagram of ECE constr ucts and Western analysis of ECE expression in HEK 293 cells. 44 Figure 2 : ECE enzymatic activity obtained from mice hippocampal r egions after injection of rAAV viru s expressing ECE protein. 45 Figure 3: Examinati on of ECE expression using anti-HA immunoreactivity in the hippocampus and ant erior cortex. 46 Figure 4: ECE and GFP expression profiles in mice brains following r AAV administration 47 Figure 5: Total amyloid load is reduced following intracranial administrati on of rAAV-ECE-HA vector. 48 Figure 6: Congophilic compact pl aque load is reduced following intracranial admini stration of ECE-HA rAAV vector. 49 PAPER 2 Figure 1: Diagrammatic representation of rAAV constructs expressing t he NEP gene under the cont rol of the chicken -actin (CBA) promoter. 87 Figure 2: Examination of NEP expression in transduced cells of the hippocampus contralateral and ipsilateral to injection site. 88 Figure 3: Examination of expression levels of NEP after intracranial administration of vira l vectors into the right anterior cortex of mice. 89 Figure 4: Quantification of Neprilysin specific activitiy in HEK293 transfected cells. 90 Figure 5: Total amyloid load is reduced following intracranial administration of NEP-s or NEP-n AAV vector. 91 Figure 6: Quantification total am yloid load in the hippocampus and cortical regions following intracranial administration of rAAV. 92

PAGE 5

iv Figure 7: Congophilic compact pl aque load is reduced following intracranial admi nistration of NEP-n AAV vector. 94 Figure 8: Quantification of to tal congophilic load in the hippocampus and cortical regions following intracranial administration of rAAV. 95 Figure 9: Distribution of NEP 5 mo after in jections of rAAV vectors in old mice APP + PS1 mice 15 months of age. 97 Figure 10: Total abeta load is r educed following intracranial administration of rAAV-s in aged mice. 98 Figure 11: Total congophilic staining is reduced following intracranial administration of rAAV-s in aged mice. 99 PAPER 3 Figure 1: AAV mediated GFP expressi on in the mouse CNS following intracrani al administration using c onvection enhanced delivery. 134 Figure 2: CED Method does not result in neuron loss or significant increase in CD45 expression. 135 Figure 3: Comparison of different AAV serotypes 5, 8, and 9 expressing GFP followi ng intracranial administration into the hippocampus. 136 Figure 4: GFP expression is signific antly increased in the right and left hi ppocampus in 9 month old mice following CED delivery of AAV5 and syst emic mannitol pretreatment. 137 Figure 5: GFP expression is signific antly increased in the thalamus and entorhinal cortex in 9 month ol d mice following CED delivery of AAV9. 138 Figure 6: Quantificatio n of anti-GFP staini ng following AAV 9 administration. 139 Figure 7: GFP expression patterns in different cell types of the hippocam pus following AAV 5 and AAV9 administration. 140 Figure 8: Transduction of GFP in different regions of the mouse brain following AAV 5 and AAV9 adm inistration into the hippocampus. 142

PAGE 6

v LIST OF ABBREVIATIONS AAV: Adeno-associated virus BACE: Beta site APP cleavi ng enzyme (beta secretase) A : Amyloid Beta AD: Alzheimer’s disease APP: Amyloid precursor protein CA: Cornu ammonis CBA: Chicken -actin CED: Convection enhanced delivery DG: Dentate gyrus GFAP: Glial fibrillary acidic protein GFP: Green Fluorescent Protein ECE: Endothelin converting enzyme FAD: Familial Alzheimer’s disease HA: Heamagluttinin IDE: Insulin degrading enzyme ITR: Inverted terminal repeat MMP: Matrix metalloprotease NEP: Neprilysin NeuN: Neuronal nuclei NGF: Nerve Growth Factor PS: Presenilin rAAV: Recombinant adeno-associated virus

PAGE 7

vi Recombinant AAV Gene Therapy and Delivery for Alzheimer’s Disease Nikisha Christine Carty ABSTRACT Alzheimer’s disease (AD), first characterized in the early 20th century, is a common form of dementia which can occur as a result of genetic mutations in the genes encoding presenilin 1, presenilin 2, or amyloid precursor protein (APP). These genetic alterations can accelerate the pathological characteristics of AD, including the formation of extracellular neuritic plaques composed of amyloid beta peptides and the formation of intracellu lar neurofibrillary tangles consisting of hyperphosphorylated tau protein. Ultim ately, AD results in gross neuron loss in the brain which is evidenced clinica lly as a progressive decline in mental capacity. A strong body of scientific evidence has previously demonstrated that the driving factor in the pathogenesis of AD is potentially the accumulation of A peptides in the brain. Thus, reduction of A deposition is a major therapeutic strategy in the treatment of AD. Recently it has been suggested that A accumulation in the brain is modulated, not only by A production, but also by its degradation. Several important studies have demonstrated that A degradation is modulated by several endogenous zi nc metalloproteases shown to have

PAGE 8

vii amyloid degrading capabilitie s. These endogenous proteases include neprilysin (NEP), endothelin converting enzyme (E CE), insulin degrading enzyme (IDE) and matrix metalloprotease 9 (MMP9). In th is investigation we study the effects of upregulating expression of several of these proteases th rough administration of recombinant adeno-associated vira l vector (rAAV) containing both endogenous and synthetic genes for ECE an d NEP on amyloid deposition in amyloid precursor protein (APP) plus pr esenilin-1 (PS1) transgenic mice. rAAV administration directly into the brain resu lted in increased expression of ECE and NEP and a substantial decrease in amyl oid pathology. We were able to significantly increase the area of vira l distribution by using novel delivery methods resulting in increased gene expression and distribution. These data support great potential of gene therapy as a method of treatment for neurological diseases. Optimization of gene transfer methods aimed at a particular cell type and brain region in the CNS can be accomplished using AAV serotype specificity and nov el delivery techniques leading to successful gene transduction thus prov iding a promisin g therapeutic avenue through which to treat AD.

PAGE 9

1 INTRODUCTION ALZHEIMER’S DISEASE Alzheimer’s disease (AD) is a debilitating and devastating neurodegenerative disorder that leads to a progressive decline in memory, the ability to learn and reason, and loss of overa ll brain function ultimately resulting in death. AD is the most common form of senile dementia (a group of conditions that gradually destroy brain cells leadi ng to progressive decline in mental capacity) affecting approxim ately 4.5 million individual s in the United States consequently causing a great financia l burden on the U.S. economy. The estimated annual costs of caring for indivi duals with Alzheimer’s disease are at least $100 billion, according to the Al zheimer’s Association and the National Institute on Aging. The molecular mechanisms underlying Alzheimer’s disease (AD) have been extensively investigated. AD can occur as a result of genetic autosomal-dominant mutations in the genes encoding presenilin 1, presenilin 2, or amyloid precursor protein (APP) ( Levy et al., 1990), (Goate et al., 1991), (Hardy et al., 1992),(Scheuner et al.,1996) Interestingly, these autosomal dominant genetic mutations only account for 1-2% of AD cases which are inherited and commonly referred to as early onset Alzheimer’s disease or familial AD (FAD)(Fidani and Goate, 1992). The caus e of AD accounting for the majority of the cases remains unknow n, although Apolipopr otein E4 is a known risk factor.

PAGE 10

2 These genetic alterations have been found to accelerate the pathological characteristics of AD which include the fo rmation of extracellular amyloid plaques and the formation of intracellular neur ofibrillary tangles consisting of hyperphosphorylated tau The accumulation of thes e amyloid plaques are not only a crucial factor in the pathology of AD (Selkoe, 2001), (Bard et al., 2003), but have been argued to contribute to the distinctive clinical symptoms of AD such as progressive cognitive decli ne, loss of memory and decreased mental capacity (Cummings et al., 1996a), (Nic oll et al., 2003), (Masliah et al., 2005). The genetic mutations that result in AD pathology all cause alterations in the normal processing of the amyloid prec ursor protein (APP) by altering the proteolytic cleavage sites which c ause the production and release of A peptides in the brain. Accumulation of thes e peptides leads to the formation of extracellular deposits commonly referred to as neuritic plaques (Cummings et al., 1996b), (Selkoe, 2001). Another hallmar k pathological feat ure of AD are intraneuronal neurofibrillary tangles which can occur independent of the neuritic plaque. The tau aggregates which co mpose these tangles are nonmembrane bound fibers arranged as pairs of helical f ilaments. Although the mechanism by which neuritic plaques and neurofibrillary tangles can eventually lead to neuron loss is debated among researchers, it has been repeatedly demonstrated that reducing amyloid deposits in the brain can significantly improve cognitive deficits (Janus et al., 2000), (Morgan et al., 2000), (Dodart et al., 2002). Consequently, elucidating novel methods of decreasing or preventing amyloid accumulation has been a primary focus in the treatm ent of Alzheimer’s disease.

PAGE 11

3 AMYLOID PRECURSOR PROTEIN (APP) AND PRESENILIN 1 AND 2 The genetic etiology of AD provides important evidence supporting the “Amyloid Cascade Hypothesis” first put forth by John Hardy in 1991 (Hardy and Allsop, 1991), (Hardy and Selk oe, 2002). Specifically, a ll genetic mutations that cause early onset or familial AD alter the production of A in the brain (Kimberly et al., 2000). The amyloid precurso r protein (APP) gene was the first AD susceptibility gene to be identifie d (George-Hyslop, 2000). The APP gene encodes for a transmembrane glycoprotein that contains 770 amino acids, but has several isoforms which vary in length (George-Hyslop, 2000). The extracellular domain of the protein is t he critical portion that normally undergoes processing through a series of proteolyt ic cleavages. These cleavage pathways involve an -secretase, -secretase, and -secretase that give rise to several different peptides. The normal processing of the APP molecule only involves the -secretase cleavage which readily cleaves the APP695 isoform of APP between the Lys687 and Leu688 residues in t he extracellular domain. This cleavage is subsequently followed by a -secretase cleavage which yields a soluble APP molecule and a smaller P3 fragment. The abnormal processing of APP involves a -secretase cleavage followed by a -secretase cleavage yielding A peptide species A 1-40 and A 1-42 (Selkoe, 1998), (Sinha and Lieberburg, 1999), (George-Hysl op, 2000). Cells that contain the APP mutation readily produce lar ger amounts of the A 1-42 peptide, the pathogenic species which is slightly more hydrophobic caus ing it to aggregate, as opposed to the A 1-40 species (Wang et al., 1999). This results in a change in the ratio of the

PAGE 12

4 particular species which are normally prod uced by the cell, causing the peptides to abnormally accumulate forming plaques which are primarily made up the A 142 peptide and deposited within the brain pa ranchyma (Cummings et al., 1992), (Suzuki et al., 1994), (Cummings et al ., 1996b), (Skovronsky et al., 1998). The aberrant accumulation of A peptides in the brain has been recognized as an essential factor in the pathol ogy of Alzheimer’s disease (A D). Due to the fact that -secretase catalyzes the final step in the production of A peptides and because it determines the final length of the A peptide variants, understanding this protease has become an important target not only in the molecular genetics of AD but has emerged as an important tar get in the development of potential AD therapy. The continued study of the molecular genetics of AD eventually revealed the genes known as presenilin 1 (PS1) and presenilin 2 (PS2) (57 and 55 kDa polypeptides, respectively) which were initially discovered in 1995. The presenilins play an integral role in t he molecular mechanisms leading to the development of Alzheimer’s disease (Anw ar et al., 1996), (Hutton et al., 1996), (Hutton and Hardy, 1997), (Dewji and Sing er, 1998), (Siman et al., 2000). Several different functions have been credite d to presenilin including a role in promoting or reducing a neurons susc eptibility to apoptosis, regulating intracellular calcium signaling, trafficki ng of membrane proteins, processing of the APP molecule, and involvement in the cleavage of other substrates (Ray et al., 1999), (Siman et al., 2000), (Brunkan and Goate, 2005), (Dewji, 2005). The presenilin 1 mutations account for the ear liest and most aggressive forms of AD,

PAGE 13

5 while presenilin 2 mutations have a later and more variable age of onset (George-Hyslop, 2000), (Ishii et al., 2001). Over one h undred different mutations have been reported in the PS1 gene, most of which are missense mutations, and approximately 6 mutations found in PS2. The presenilins have been found to be in part expressed at the cell surface with the majority of the protein being expressed in the intracellular compartm ent. Reports have also indicated that PS1 and PS2 are expressed in the endoplasmic reticulum and the golgi apparatus as well as in nuclear memb ranes (Dewji, 2005). Ultimately, the presenilin mutations lead to an increased production of cerebral A 1-42 peptides and a 1.5 to 3 fold increase in A 42 containing plaques in human brain tissue of FAD human tissue as opposed to sporadic ca ses of AD (Lamere et al., 1996). The finding that presenilin mutations cause specific increases in A 1-42 production lead to the idea that mutant PS must play a role in the modulation of -secretase activity and are essential for the -secretase cleavage of APP (Ray et al., 1999). Results from transgenic mouse models with PS mutations and APP mutations clearly indicate that PS muta tions not only increase production and deposition of A species in the brain but also impl y that presenilin may act as the -secretase enzyme or an essential cofactor for the -secretase cleavage of APP (Duff et al., 1996), (Ishii et al., 2001), (Brunkan and Goate, 2005). Despite the breakthrough discovery of genes responsi ble for early onset AD this accounts for only 1-2% of all AD cases and ther efore investigation to identify AD susceptibility genes have become an ar ea of intense research.

PAGE 14

6 TRANSGENIC MOUSE MODELS OF AMYLOID DEPOSITION Almost ten years after the discovery of the FAD mutations in the APP gene and the introduction of transgenic te chnology, researchers have used knowledge of the numerous mutations on APP, PS1, and PS2 to create murine models that have lead to an increas ed understanding of AD pathogenesis. The first generation of an AD transgenic mouse model was based on the over expression of the wild type human (h APP) cDNAs and in most cases these efforts failed to produce mice with the hallmark pathogenicity found in human AD cases with the exception of the NSEA PP (APP751) mouse produced by Quon et. al. in 1991. Investigators eventually di scovered that the overexpression of the hAPP in addition to either the Swedish mutation (APPK670N,M671L) or the London mutation (APPV717F) was sufficient to produce enough A peptides to cause significant amount of am yloid deposition (Quon et al., 1991), (Higgins and Jacobsen, 2003). This gave rise to t he PDAPP mouse first presented by Games et. al. in 1995 that contains the London mutation (APPV717F) under control of the platelet derived growth factor promot er (PDGF). The Tg2576 mouse first described by Hsiao et. al. in 1996 contai ns the APP695 isoform with the Swedish mutation (APPK670N, M671L) under the prion protein prom oter (PrP) which ensures high CNS expression (Hsiao et al., 1996, Higgins and Jacobsen, 2003). These mice exhibit a similar positive corre lation between increased age and amyloid deposition as well as selectiv e distribution of amyloid wit hin specific brain regions such as cortex and hippocampus.

PAGE 15

7 Another single transgenic mouse m odel of amyloid deposition (APP23 mouse) was developed by Novartis and des cribed by Sturchler-Pierrat et. al., in 1997 and contains the hAPP751 isoform with the Swedish mutation under the control of the Thy-1 promotor to enhance neuronal expression. These mice also emulate the PDAPP and T g2576 mice in their region specific development of amyloid deposits but unlike the two previous transgenic models the APP23 mouse has been shown to demonstrate neur odegeneration in the CA1 region of the hippocampus which seems to correlate with increased plaque load (between 14-18 months of age) (Calhoun et al., 1998) (Higgins and Jacobsen, 2003). In more recent years double transgenic m ouse models of AD which express both mutant APP and human presenilin with FA D mutations have been successfully created (Higgins and Jacobsen, 2003). These mice have significantly enhanced production of the A 1-42 peptide and also begin to develop amyloid deposits at a much earlier age compared to their singl e transgenic counterparts (Borchelt et al., 1997), (Holcomb et al., 1998). T he PSAPP mouse overexpresses the PS1 gene with the M146L mutation (Duff et al ., 1996) and the Swedish mutation (Hsiao et al., 1996) and show significant ly increased plaque pathology, beginning around 3 months of age, which is associated with gliosis (Holcomb et al., 1998). Interestingly, transgenic mice that expr ess only FAD presenilin mutations alone show no signs of amyloid deposition (D uff et al., 1996). The CRND8 transgenic model expresses a double mutated form of the APP695 (APPV717F and APPK670N, M671L) under the PrP promoter. This model also results in accelerated plaque deposition which is clearly evident at 3 m onths of age. PDAPP mice have also

PAGE 16

8 been shown to exhibit some behavioral defic its as early as three months of age even before the development of amyloid plaque depos its could be detected (Higgins and Jacobsen, 2003). The T g2576 transgenic model differ from the PDAPP animals in that they begin to develop cognitive deficits between 9-10 months of age and this correlates more closely with the development of amyloid deposits in the brain (Hsiao et al ., 1996). In early 2002 Westerman and collegues proposed that cognitive deficits mo re likely are the result of increases in soluble forms of A (or small protofibrils of A ) rather than plaque deposition (Westerman et al., 2002) More recent repor ts argue that large A aggregates may in fact play a neuroprotective role providing a mehcanims in which smaller soluble A moieties are sequestered into larger insoluble aggregates preventing them from causing destru ction. Specifically, t he soluble oligomeric A assemblies have been implicated as the drivi ng factor in AD pathogenisis as they exhibit potent toxicity in their capacity to significantly decrease LTP, contribute to learning and memory deficits in vivo and induce neuronal cell death in vitro (Lambert et al., 1998), (El-Agnaf et al ., 2000), (Klein, 2002; Glabe, 2005), (Malaplate-Armand et al., 2006), (Towns end et al., 2006), (Cerpa et al., 2008), (Varvel et al., 2008) ZINC METALLOPROTEASES/ENDOGENOUS A DEGRADING ENZYMES In recent years a growing number of endogenous proteases have been found to have A degrading capabilities in the brain and other tissues both in vivo

PAGE 17

9 and in vitro Several of these enzymes are members of zinc metalloprotease family and include neprilysin (NEP), insu lin degrading enzyme (IDE; insulysin), endothelin converting enzymes (ECE-1 and ECE-2), and angiotensin converting enzyme (ACE). Recently, a number of other proteases have been shown to degrade A in vitro, including matrix metall oproteases 2, 3, and 9, and, the serine protease, plasmin but the degree to whic h they may be involved in the normal catabolism of A remains uncertain (Yan et al., 2006), (Turner and Nalivaeva, 2007). Down regulation of these enzymes within the brain could potentially contribute to A accumulation and lead to development of AD. NEP, a 92-kDa glycosylated ectoenzyme, was originally found as a kidney enzyme and is capable of degrading multiple peptide hormones (Iwata et al., 2000a). Also known as enkephalinase, it is capable of cleaving enkephalins and termination of peptidergic neurotransmissi on. It is membrane bound and its active site is oriented outside the cell. NEP has a very large range of regulatory activity as it is capable of degrading numer ous bioactive peptides including opioid peptides, tackykinins, substance P, atri al natriuretic peptides, chemotactic peptides and adrenocorticotropin hormone (ACTH) (Matsas et al., 1984a), (Turner et al., 1996). NEP is expre ssed in the CNS and in the periphery particularly in epithelial cells of the inte stines, kidney and lungs (Fulcher et al., 1982), (Matsas et al., 1984b), (Kenny et al., 1985). In the CNS NEP is expressed in neurons, mainly pyramidal neurons, and glial cells types (Turner et al., 2003). Therefore, NEP is involved in regulation of several different physiological processes, such as cardio vascular activity, cell migration and

PAGE 18

10 proliferation, inflammati on, and neuropeptide regulation (T urner, 2003). NEP, is a highly conserved protein and the N EP gene in located on chromosome 3 spaning more the 80 kb (Roques et al., 1993) in humans. Several homologues of NEP have also been identified most of which remain uncharacterized (Turner and Nalivaeva, 2007), (Nalivaeva et al., 2008). Although, one homologue named NEP2 (also known as Sep and NL1) has been shown to also degrade A unlike NEP it is expressed only in specific neuronal populations in the brain and in the spinal cord. Furthermore, NEP2 is expr essed in a membrane bound form in the endoplasmic reticulum and a soluble secr eted form but very little is known regarding its function or whether it plays a role in A metabolism. Iwata et. al., in 2000 first demonstrat ed the ability of NEP to degrade A peptides in the brain parenchyma and also illustrated that suppressing NEP would lead to an increase in A deposition (Iwata et al., 2000). In 2002 Iwata et. al.. also demonstrated that NEP activi ty decreased with age in regions where A accumulated such as the cortex and hippoc ampus, however in areas such as the striatum where A did not accumulate NEP was increased (Yasojima et al., 2001b). Evidence from transgenic ani mal models demonstrate that NEP deficient mice show a significant incr ease in amyloid depos ition as opposed to their wild type counterparts clearly indicati ng that NEP activity plays a significant role in endogenous A degradation (Iwata et al., 2 002), (Iwata et al., 2004), (Marr et al., 2004). Another important A degrading enzyme is known as endothelin converting enzyme (ECE). ECE, like NEP is a membrane bound

PAGE 19

11 metalloprotease approximately 87 kd in size and exists in several isoforms including ECE-1a, ECE-1b, ECE-1c and ECE-2. ECE-1 has been found localized in the Golgi and vesicles. Its primary function is to catalyze the conversion of big endothelin (big ET) in to vasoactive endothelins(Eckman et al., 2003). ECE-1 has also been reported to hy drolyse other biologically active peptides including bradykini n, neurotensin and substance P. ECE is a homologue protein of NEP, yet an import ant structural difference between the two is that ECE-1 exists as a disulfi de linked dimer. ECE-1 is expressed in endothelial cells as well as several ot her cell types including neurons and glia within the brain that are particularly relev ant to AD disease pathology. Although several other isoforms of ECE exist, ECE-1c is the predom inantly expressed mRNA in humans. Each isoform has distin ct subcellular localizations; specifically ECE-1a, ECE-1c, and ECE1d are primarily localized at the cell surface, while ECE-1b is found in the intracellular compartment. To date there are no significant differences in the catalytic pr operties of the different isoforms (Turner et al., 2004). The A degrading capabilities of ECE were discovered when Eckman et. al. observed that the meta lloprotease inhibitor, phosphoramidon, caused a rapid accumulation of A in a cell line which expressed ECE and did not cause this increase in cells not expressing ECE (Eckman and Eckman, 2005). Subsequent studies revealed that when ECE was overexpressed in the cell lines the A accumulation was reduced by appr oximately 90%. In vivo data from ECE (+/-) transgenic mice which show about 27% reductions in ECE-1

PAGE 20

12 activity also show significant increases in both A 40 and A 42 levels in the brain (Eckman and Eckman, 2005). In addition to NEP and ECE, IDE (insulysin) has been suggested as another A degrading enzyme that may play an important role in the dynamic process of A degradation in the brain. Unlike NEP and ECE-1, IDE is primarily localized in the cytosol but it can also be located, at low levels, in peroxisomes as well as in the plasma membrane and in a secreted form in the extracellular compartment (Turner et al., 2004). ID E is expressed in the human CNS in cortical and subcortical neurons as we ll as in endothelial cells, pericytes and smooth muscle cells (Morelli et al., 2004). In vitro, IDE is capable of degrading a variety of substrates includi ng insulin and its role in A degradation in the brain was first proposed by Kurochkin et. al., (Kurochkin and Goto, 1994), (Kurochkin, 2001), (Turner et al., 2004), (Farris et al ., 2005). Several studies support the importance of IDE activity in A catabolism, including a correlation study presented by Perez et. al. in 2000 revealed that IDE activity which correlated with a decrease in soluble A peptide degradation in the AD brain compared to normal aged matched control brains (Per ez et al., 1999). Furthermore, in transgenic mice with IDE (-/-) deficiencies results in almost a 50% decrease in A degradation in not only the brai n but in primary neuron cult ures as well; while the overexpression of IDE significantly reduces brain A levels in APP mice (Farris et al., 2003), (Leissring et al., 2003), (Farri s et al., 2005). The in vivo data from multiple studies involving NEP, ECE, and IDE transgenic mice provides clear evidence t hat the activity of these specific

PAGE 21

13 proteases are essential for the regulation of A accumulation in the brain. Therefore, the regul ation of these proteases th rough novel methods such as gene therapy provides a unique avenue through which novel therapeutic treatments can be explored. GENE THERAPY AND VIRAL VECTORS Gene therapy has a fairly short and controversial history and was first envisioned as a potentially vi able scientific tool for t he treatment of hereditary single-gene defects. Beginning in the late 1970’s and 80’s gene therapy made its unofficial clinical debut with two unapproved clinical trails. In the first trials, two young boys with arginase defici ency syndrome were treated using in vivo gene therapy with a wild-type Shope papilloma virus in an effort to replace the missing enzyme with the viral arginase. Anot her trial involved the use of ex vivo gene therapy in which a bone marrow tr ansplant was performed using bone marrow treated with a -globin-containing plasmid in order to treat two patients with -thalassemia. The results were neither efficacious nor deleterious (Scollay, 2001). During the 1990’s the use of gene therapy in approved gene therapy trials steadily increased and by t he mid 2000 nearly 4000 patients have been treated with gene therapy in more than 500 tria ls (Scollay, 2001). The majority of the trials in the 1990’s were unsuccessf ul and in 1999 the first published death resulting from gene therapy using an adeno virus vector was highly publicized lending to a bad public perception of gene t herapy and it’s potential as a viable therapeutic avenue in the treatment of di sease and revealing the need to further

PAGE 22

14 understand the viral vector interacti ons with the human immune system. Fortunately, by 2000 the first successful procedures were conducted in three individuals suffering from severe comb ined immunodeficiency (SCID), (Scollay, 2001). This progress was the result of improvements in understanding different types of viral vectors and in our under standing of genetics and the biology of transducing cells. There are several different mechanisms by which gene therapy can be achieved and an important fact or in the success of the method involves the type of vehicle used to delivery the therapeutic gene to the target tissue (Thomas et al., 2003). Viral vectors utilized in the firs t gene therapy attempts were extremely inefficient and did not persist in the host cells and the transgene expression was in most cases for a shor t time. A thorough understanding of the molecular basis of how viruses and viral vectors intera ct with the host has been a major focal point in the area of gene therapy. As a result over the past few years, several vectors have been developed aimed at impr oved efficiency, specificity, and safety which have also resulted in succe ssful treatments (Thomas et al., 2003). The diversity of diseases that gene therapy research has aimed at targeting is expanding from the initial hereditary single gene defect treatment to include acquired diseases such as cancer, cardiovascular disease, infectious disease and neurodegenerative disease (M andel and Burger, 2004). This means that no one vector method will be the most efficacious for all disease categories. Thus understanding and utilizing unique characteristics of differen t viral vector types is essential for the success of gene transfe r and persistence of the transgene.

PAGE 23

15 There are five major categories of viral vectors which can be categorized into two main groups (Thomas et al., 2003). The groups include viruses whose genome integrates into the host cellular genome known as oncoretroviruses or those which persist apart from the cellular host g enome as an episome, or extrachomosomal unit, within the cell nucleus. Retroviruses, such as human T cell leukemia virus, sarcom a virus and HIV, are a class of viruses that can create double-stranded DNA copies of their RNA genomes. These copies of its genome can be integrated into the chromosomes of ho st cells. Lentivirus is characterized by a long incubation period. Human i mmunodeficiency virus (HIV) is an example of a lentivirus. Adenoviruses are a cl ass of viruses with double-stranded DNA genomes that cause respirator y, intestinal, and eye infe ctions in humans. The virus that causes the common cold is an adenovirus. Adeno-associated viruses are a class of small non-pathogenic human parvovirus, single-stranded DNA viruses that can insert their genetic ma terial which can persist as an episomal plasmid. Finally, herpes simplex viru ses are a class of double-stranded DNA viruses that infect a particular cell type, specifically neurons. Herpes simplex virus type 1 is a common human pathogen that causes cold sores (Thomas et al., 2003). There are advantages and disa dvantages to each type of vector which determines it’s suitabilit y for specific treatments. Particularly, vectors such as adeno-associated virus (AAV), adenovir us and herpes virus, which have nonintegrating vectors, usua lly have a persistent transgene expression only in non proliferating cells whereas retroviruses with integrating genomes have persistent transgene expression in dividing cells. The si ze of the viral vehicle can also be a

PAGE 24

16 determining factor mediating the type of gene that can be expressed. The table below, published by Thomas et. al.. in 2003 summarized the advantages and disadvantages of the five different vira l vectors (Thomas et al., 2003). Adeno-associated viruses are members of the parvoviridae family and are subsequently a member of the helper vi rus genus which indicates that they need a helper virus, in most cases adenovirus or herpes simplex virus, in order to support successful infection and replicatio n. When the helper virus is absent AAVs will generate a latent infection wit hin the cell and in 90% of cases this means the AAV genome will persist as an episomal unit (Thomas et al., 2003), (Tenenbaum et al., 2004a), (Wu et al., 2006). ADENO-ASSOCIATED VIRA L VECTORS AND SEROLOGY AAVs are one of the smallest viruses, approximately 25 nm in diameter, and their DNA genome is less than 5 kb which contains two large open reading frames with inverted terminals repeats loca ted at either end (Samulski et al., 1982), (Tenenbaum et al., 1994), (Tenenbaum et al., 2004b), (Wu et al., 2006). AAVs must accomplish several major steps to achieve successful gene expression which include attachment to the cell surface receptors, endocytosis, trafficking to the nucleus, uncoating of virus which releases viral genome, and finally conversion of the genome to double stranded DNA which is transcribed in the nucleus. The efficiencies of rAAV transduction intricately depend on the successful implementation of each step. AAV vectors arrive in a variety of flavors otherwise known as serotypes. T he AAV serotype refers to the efficiency

PAGE 25

17 by which the specific AAV can infect a particular cell type through attachment to specific cell surface receptors. A new AAV serotype is defined by the inability of an antibody that is reactive to the vi ral capsid protein of one serotype in neutralizing those of another serotype (Choi et al., 2005). Through the modifications to the viral genome and the capsids rAAV can be designed to be more efficient at transducing specific cells or tissue types to optimize gene therapy (Choi et al., 2005). AAV2 serotype was the first to be cloned into bacterial plasmids by Samulski et. al. in 1982 and to date it is the most characte rized of all the serotypes of which a total of 10 have been discovered, and has been shown to effectively transducer neurons in the CNS in a number of animal studies (Samulski et al., 1982), (Fu et al., 2003), (Burger et al., 2004), (Tenenbaum et al., 2004b). AAV5 was originally discovered in a human clinical sample in 1984 as a contaminant in adenovirus stock and it contai ns ITRs (inverted terminal repeats) that are not unlike the struct ures of AAV2 ITRs and is the most divergent of all of the serotypes (Choi et al., 2005), (Wu et al., 2006). AAV1 and AAV3 were also initially found as contaminants of simian adenovirus 15, while AAV6 was identified as a contaminant of adenovirus type 5 by Rutledge et. al. in 1998. AAV7 and AAV8 were originally isolated fr om rhesus monkey tissue in 2002 by Gao et. al.. AAV9, like AAV5, was originally discovered in human tissue, while AAV10 and AAV11 are the most recently identified serotypes and initially were isolated from cynomolgus monkey tissue in 2004 by Mori et. al. but they are not well characterized (Mori et al ., 2004), (Wu et al., 2006).

PAGE 26

18 The fairly recent identification of different AAV serotypes had advanced the study of rAAV vectors which has qui ckly become a major dominant focus in the field of gene therapy. T here are a number of studies aimed at characterizing the different AAV serotypes with respec t to transduction efficiency, tissue tropism, cell surface receptors, intrace llular processing, and c apsid structure. AAV2 vectors are the most extensivel y characterized and due to their well established safety profile and range of infect ivity, approximately 20 clinical trails have been conducted using the AAV2 vector serotypes in numerous patients (Wu et al., 2007). In or der to effectively study transduction efficiencies of different serotypes of AAV vectors in different tissue types, scientists have utilized the “cross-packaging” strategy which essentially enables an unbiased direct comparison of the transducti on rates without influence of ITRs on transgene expression (Grieger and Samul ski, 2005), (Wu et al., 2007). To accomplish this the Cap genes of different serotypes are plac ed downstream of the AAV2 Rep genes which ultimately allows the generation of serotypes specific capsids while the packaged genomes within the capsids are identical, these are also referred to as hybrid viruses (Rabi nowitz et al., 2002), (Rabinowitz et al., 2004), (Wu et al., 2006). In addition to transcapsidation first described by Rabinowitz et. al. in 2002 there are seve ral other techniques to accomplish the formation of a hybrid virus armed wit h specific modifications to enhance efficiency of gene uptake, transfer, and ex pression in a specified therapeutic avenue. These include absorption modifi cations where the capsid surface is modified to carry a foreign antibody that will bind to the cell surface receptor of

PAGE 27

19 interest to increase absorption efficiency. Mosaic capsids are another method to increase transduction efficiency. This te chnique involves creat ing a mixture of viral capsid proteins from different seroty pes at various ratios in order to combine tissue tropisms of interest (Xiao et al., 1999), (Rabinowitz and Samulski, 2000). Finally, chimeric capsids can also provi de a means of increasing transduction of a particular specificity and this techniqu e involves the packaging of capsid proteins with foreign peptide sequences, su ch as a hemagglutinin (HA) tag fused at either the N or C terminus of the capsid coding sequence to alter tissue tropism (Yang et al., 1999), (Bowle s et al., 2003), (Wu et al., 2006). Although the data regarding AAV seroty pe specific tissue tropism is subject to different interpretation due to va riations in vector titer, promoters, transgenes between studies, in general the transduction efficienc ies for most of the AAV serotypes have been determined. In the CNS serotype characterization studies have revealed that AAV1, and 5 have higher transduction efficiencies than AAV2 throughout the all r egions and cell types of the CNS (Alisky et al., 2000), (Burger et al., 2004), (Burger et al., 2005) while AAV4 will efficiently transduce specific cell types such as astrocytes within the subventricular zone (Davidson et al., 2000), (Weber et al., 2003) (Wu et al., 2006). Studies by Wolfe et. al. also reveal that AAV7, 8, 9 and Rh10 expressing cDNA for lysosomal enzyme are also capable of transducing n eurons within specific regions in the mouse brain. AAV9 and AAVRh10 appeared to have the highest transduction efficiencies and were found to undergo ve ctor genome transport through axonal transport pathways (Cearley and Wolfe, 2006).

PAGE 28

20 Despite the rapidly growing body of knowledge in the field of AAV vector serology there is still much to be done. In addition to the 10 AAV serotypes that have been found over 100 AAV variants have also been discovered and with the technological advances in engineering hybr id vectors, including mosaic AAVs, chimeric AAVs, absorbtion modifications and transcapsidation AAV vectors appear to have a promising future in the field of gene therapy. (AAV serotypes 6-11 are not currently commercially available). ENHANCING DISTRIBUTION OF GENE PRODUCT AAV vectors are a novel mean by which transgenes can be delivered to potentially treat several types of degenerativ e brain diseases. Unfortunately, one of the major disadvantages of a single in tracranial injection into the brain parenchyma using simple diffusion does not allow for efficient uptake of the transgene or significant distribution of t he AAV macromolecules to significantly large areas of the affect ed regions within the brain. Convection enhanced delivery (CED), first described by Bobo et al in 1994, is a method of delivering clinically relevant volumes of therapeutic agents to signi ficantly larger areas of the brain in a direct intracranial inje ction procedure in comparison to simple diffusion methods (Bobo et al., 1994). T he CED technique is designed to utilize the phenomenon of bulk flow and positive pr essure to distribute macromolecules to a large area within solid tissue. The CED technique was originally proposed by scientists in the early 1990s as a met hod of delivering drugs directly to the parenchyma that would not normally cr oss the blood brain barrier and that

PAGE 29

21 consisted of large macromolcules too large to diffuse throughout the tissue (Raghavan et al., 2006). Due to the lack of approved drugs that can be directly intracranially administered to the brai n and the difficulty in predicting methods that ensure delivery of the t herapeutic agent to its target si te in spite of its use in clinical trails CED remains an experi mental procedure and research of CED delivery devices is under current in vestigation by several researchers (Bankiewicz et al., 2000), (Raghavan et al., 2006). This CED method has been investigated in gene therapy studies as a way to increase the distribution of AAV vector s in the brain. Studies conducted by Bankiewicz et. al. in 2000 revealed that CE D can significantly increase in gene transfer and distribution of AAV expre ssing AADC in the st riatum of MPTPtreated monkeys. The AAV vector was f ound to be safely distributed throughout the entire region of the stri atum compared to the simple injection method where the distribution was severely limited (B ankiewicz et al., 2000). Similar results were replicated in the rat brain by Cunningham et. al. in 2000 with AAV2 expressing thymidine kinase (TK) wher e the CED method showed robust gene transfer and increased distribution area wit hin the putamen. CED injections in the striatum were found distributed t he AAV-TK throughout the striatum after a single injection into this region and TK immunoreactive cells were also found outside the striatum, in the globus palli dus, subthalamic nucleus, thalamus, and substantia nigra (Hadaczek et al., 2006). One of the mechanistic limitations of the CED method as well as the simple injection method is the reflux of the injected material from the injection

PAGE 30

22 hole upon the removal of the cannula. In 2005 Krauze et. al. developed a reflux free step cannula design which effectively eliminates reflux by placing silicone coated tubing within the cannula creating a step that prevent s the backflow of fluid (Krauze et al., 2005b), (Krauze et al., 2005a). The optimization of more efficient cannula designs coupled with the encouraging results from studies showing enhanced gene trans fer and distribution emph asizes the therapeutic potential of the CED method in helping overcome some of the mechanical disadvantages of gene delivery in regards to gene therapy (Krauze et al., 2005a). The use of osmotic agents such as mannitol is an addition method that can be used to increase the area of dist ribution of macromolecules throughout the CNS. Mannitol is a blood brain barrier interruptive reagent and is also known to temporarily increase vascular pressu re subsequently reducing intracranial pressure. High concentrations of manni tol intravenously infused are currently used in patients with traumatic brain dise ase to reduce intracranial pressure. This osmotic agent pulls fluid from t he CNS by increasing vascular osmotic pressure. Several studies have also show n that with intra-arterial infusion of mannitol the blood brain barrier can be opened to enhance the distribution of chemotherapeutics throughout the CNS in both rats and humans (Nilaver et al., 1995), (Rapoport, 2001), (Fu et al., 2003).

PAGE 31

23 PAPER 1: ADENO-ASSOCIATED VIRAL (AAV) SEROTYPE 5 VECTOR MEDIATED GENE DE LIVERY OF ENDOTHELIN CONVERTTNG ENZYME REDUCES A DEPOSITS IN APP+PS-1 TRANSGENIC MICE. Carty N.1, Nash K.2, Lee D.1, Mercer M.1, Gottschall P.E.3, Meyers C.2, Muzyczka N.2, Gordon M.N.1, Morgan D.1. 1Alzheimer’s Research Laboratory, Departm ent of Molecular Pharmacology and Physiology, School of Biomedical Sciences University of South Florida College of Medicine, 12901 Bruce B Downs Blvd, Tampa, FL 33612, USA. 2 Department of Molecular Genetics and Microbiology, University of Florida College of Medicine, Box 100266, Gainesville FL 32610 3 University of Arkansas for Medical Sciences Department of Pharmacology and Toxicology, Slot #611, 4301 West Markham Street Little Rock, Arkansas 722057199. Correspondence should be addressed to D. M. (scientist.dave@gmail.com). Corresponding Author: Dave Morgan, Alzheimer’s Research Laboratory, University of South Florida College of Medicine, 12901 Bruce B Downs Blvd

PAGE 32

24 MDC Box 9, Tampa, FL 33612, USA Phone: +1 (813) 974-3949; Fax: +1 (813) 974-3079. Email: scientist.dave@gmail.com Short title: rAAV delivery of ECE reduces A in APP + PS1 mice. ABSTRACT Reduction of A deposition is a major therapeut ic strategy in Alzheimer's disease (AD). The concentration of A in the brain is modulated, not only by A production, but also by its degradati on. One protease involved in the degradation of A peptides is endothelin conver ting enzyme (ECE). In the current study, we investigated the effect s of an intracranial administration of a recombinant adeno-associated viral ve ctor (rAAV) cont aining the ECE-1 synthetic gene on amyloid deposition in am yloid precursor protein (APP) plus presenilin-1 (PS1) transgenic mice. T he recombinant AAV vector was injected unilaterally into the right anterior cort ex and hippocampus of six-month-old mice while control mice received an AAV vector expressing GFP. Immunohistochemistry for the haemagglutinin tag appended to ECE revealed strong expression in areas surrounding the in jection sites but minimal expression in the contralateral regions. Immunohi stochemistry for A decreased in the anterior cortex and hippocampus of mice receiving ECE synthetic gene. Further, decreases in Congo red positive deposits we re also observed in both regions. These results indicate that increasing the expression of -amyloid degrading

PAGE 33

25 enzymes through gene therapy is a promising therapeut ic avenue through which to treat AD. Key Words: Alzheimer’s disease; Beta Amyloid, Gene Therapy, Viral Vector, Amyloid Degrading Enzyme, Zinc Metalloprotease. INTRODUCTION AD is the most common form of sen ile dementia (a group of conditions that gradually destroy brain cells leadi ng to progressive decline in mental capacity) affecting approximat ely 4.5 million individuals in the United States. The molecular mechanisms underlying AD have been extensively investigated (Hardy and Selkoe, 2002). Although the mechani sm by which neuritic plaques and neurofibrillary tangles can eventually l ead to neuron loss is debated, it has been repeatedly demonstrated that reducing amyloid deposits in the brain can significantly improve cognitive deficits in amyloid depositing transgenic mice (Janus et al., 2000),(Morgan et al., 2000),(We sterman et al., 2002),(Wilcock et al., 2006). Consequently, elucidating novel methods of decreasing or preventing amyloid accumulation has been a primary focus in the treatment of AD. In recent years several endogen ous proteases have been found which degrade A in the brain and other tissues both in vivo and in vitro These zinc metalloproteases, include neprilysin (N EP), insulin degrading enzyme (IDE; insulysin), and endothelin converting enzymes (ECE-1 and ECE-2). Other

PAGE 34

26 proteases that appear to play a role in A metabolism include matrix metalloproteinase-9 (Yan et al., 2006), cat hepsin B (Mueller-Steiner et al., 2006) and plasmin (Turner et al., 2004). The overall accumulation of A in the brain is attributed to an imbalance between it s production and degradation/clearance. Down-regulation of these degrading enzymes within the brain during aging could potentially contribute to A accumulation eventually leading to development of AD (Caccamo et al., 2005),(Yasojima et al., 2001). Several current studies have implic ated ECE as an important enzyme in the degradation of A and preventing its accumulation (Eckman and Eckman, 2005). ECE is a membrane-bound metall oprotease with an N-terminal cytosolic domain and an extracellular catalytic domai n. ECE-1 is expressed in endothelial cells in all organs as well as other cell types including neurons, glia and neuroendocrine cells. Two different genes encoding for ECE-1 and ECE-2 have been identified, however ECE-1 is pr edominately expressed. ECE was discovered for its ability to catalyze the conversion of big endothelin into vasoactive endothelins (Shimada et al ., 1996), (Eckman et al., 2003), (Leissring et al., 2002), (Muller et al., 2003),(Hunter and Turner, 2006). ECE-1 has now been reported to hydrolyse other biologically active peptides including bradykinin, neurotensin and substance P. In humans ECE-1 exists as four different isoforms including ECE-1a, ECE-1b, EC E-1c,and ECE-1d (Muller et al., 2003), (Hunter and Turner, 2006), (Turner et al., 2004), (Jackson et al., 2006). Expression patterns have shown that thes e isoforms have distinct subcellular localizations with slight variations depending on the species and cell type (Muller

PAGE 35

27 et al., 2003), (Hunter and Turner, 2006). To date, there are no significant differences reported in the catalytic properti es of the different isoforms (Turner et al., 2004). The A degrading capability of ECE was discovered when Eckman (Eckman and Eckman, 2005) observed that the ECE inhibitor, phosphoramidon, caused a rapid accumulation of A in a cell line expressing ECE but not in cells lacking ECE expression. Subsequent studi es revealed that overexpression of ECE in cell lines reduced A accumulation by approximately 90%. In vivo data from ECE (+/-) transgenic mice, which show 25% reductions in ECE-1 activity, also show significant increases in both A 40 and A 42 levels in the brain (Eckman and Eckman, 2005). Moreover, intraventricular injections of phosphoramidon, increase A in wild ty pe and APP transgenic mice (Eckman et al., 2006). These studies indicate a clear correlation between ECE-1 activity and A load in the brain. In the present study we investigated the effects of upr egulating the ECE-1 enzyme by using a rAAV vector, serotype 5, on A load in the brain. rAAV vectors are desirable candidates for gene t herapy in the central nervous system because AAV is a nonpathogenic virus; it has low immunogenicity and it is deficient for replication due to the re moval of all the viral encoded genes. Further, rAAV vectors have been shown to be very efficient in infecting neuronal cells and maintaining long term expression (Burger et al., 2005), (Mandel et al., 2006). We observed a significant reduction in the levels of A in the mice injected with the ECE virus which shows t hat this method may offer a promising therapeutic avenue through wh ich to treat AD.

PAGE 36

28 RESULTS The ECE-1 synthetic gene within the rAAV was under the control of the hybrid chicken -actin cytomegalovirus (CBA) promoter and was tagged with haemagglutinin (HA) peptide sequence for detection within the brain and discrimination from endogenous ECE. Pr ior to virus production the rAAV vector was tested in HEK 293 cells to evaluate t he expression cassette. Cell lysate and conditioned media from transfected and untransfected cells were examined by Western blot analysis to determine ECE ex pression. Untransfected cell lysate and media were negative for ECE-HA prot ein expression. ECE-HA protein expression was only detected in the cell lyate fraction and not the media (Fig 1b). The rAAV ECE-HA vector was test ed for enzyme activity in vivo. Nontransgenic mice (n=12) aged 9 months were injected bilaterally into the right and left hippocampus. Group one received the rAAV-ECE-HA treatment and the second group served as the control group and received no treatment (n=6 per group). All injections were performed as described in Methods with each site of injection receiving a total volume of 2 l of the vector at a concentration of 1 x 1012 vg/ml. Six weeks post injection the animals were sacrificed and frozen hippocampal tissue from the animals was homogenized and separated into membrane and soluble fractions by centrifugat ion. The ECE specific activity was determined as the nmole of MOCA/ mi n/g protein. Within the homogenate and in the membrane fractions of the homogeni zed tissue, animals receiving the ECE treatment had significantly higher ECE-like specific activity than the control animals (Fig 2). Error bar s indicate standard error of the mean value from

PAGE 37

29 triplicates for each sample (specific activity/min/ g protein) for each treatment group (n=6 per group). rAAV was initially injected unilaterally into the right hippocampus and right anterior cortical regions of six month old APP + PS1 mice. Each injection site received 2 l volume and a concentration of 1 x 1012 vg/ml of material at a flow rate of 0.5 l/min. The expression of ECE was evaluated six weeks following treatment. Immunostaining of the tissue with an anti-HA antibody revealed ECEHA expression in most animals. The expression patterns for the ECE protein were confined to the areas surrounding the injection site. ECE-HA was detected in CA4 neurons in the hilus and CA3 neur ons of the hippocampus pyramidal cell layer. In the dentate molecular laye r, a number of r ound cells, possibly, oligodendrocytes or astrocytes also expr essed the label (Fig. 3b). In animals receiving the rAAV-GFP control vector t here was no positive HA staining in the hippocampus, but the GFP expression patte rn was comparable to that of the ECE vector (Fig 4). In animals receiving the r AAV expressing ECE-HA there were low levels of positive staining in the contra-lateral dentate gyrus (Fig. 3a). When the cortical regions were analyz ed for expression, there was a large amount of ECE-HA expression which was al so detected over a larger area of the cortex. ECE positive expression was concentr ated in the anterior cortex (Fig 3c), but was observed also in the striatum corpus callosum and septum along the midline (Fig 3d) in the ipsilateral and contralateral hemispheres. In animals receiving rAAV-GFP, GFP was detected in a similar expression patterns as ECEHA. GFP positive cells were limited to the areas immediately surrounding the

PAGE 38

30 injection site in both the cortex and the hippocampus and no GFP positive cells were seen in the contralateral hemisphere (Fig. 4b, 4d and 4f respectively). This data suggests that the rAAV vector sero type 5 is effective at transducing neuronal cell types in vivo and expressi ng significant amounts of recombinant ECE-1. Our next goal was to evaluate the e ffects of a single intracranial administration of the rAAV-ECE-HA ve ctor in APP + PS1 transgenic mice, to determine the effect of over expression of the ECE protein on amyloid deposition. rAAV was injected unilaterally into the ri ght anterior cortex and hippocampus of six month old mice while the left untr eated hemisphere. T he control group was treated with rAAV containing GFP. Total A load was ascertained six weeks after intracranial injections by imm unohistochemical methods using a polyclonal anti-A antiserum which primarily recognize s the N-terminal domain of A, and thus labels both A 1-40 and A 1-42 (gift of Gottschall, PE, Univ of Arkansas). The regional A distribution and density in APP + PS1 transgenic mice were similar to those reported by Gordon et al. (Gor don et al., 2002). I mmunohistochemistry revealed darkly stained compact plaques an d more lightly stained diffuse plaque deposits in the APP + PS1 animal tissue. Plaque deposition was distributed throughout the cortical regions as well as in the hippocampus (although most concentrated in the molecular layers of the dentate gyrus and the CA1 region, surrounding the hippocampal fissure). Animals injected with the control rAAVGFP showed A immunohistochemical staining patterns throughout the cortex and hippocampus comparable to those of untreated APP + PS1 transgenic mice

PAGE 39

31 of the same age (Fig 5a and 5b). A notable decrease in the amount of hippocampal A staining was observed in an imals injected with the rAAV expressing ECE-HA six weeks after the time of injection when compared to animals injected with the control rAAV-GFP vector (Fig. 5c, 5d). The reductions in A deposition were limited to the areas surrounding the cortical and hippocampal injection sites. ANOVA anal ysis of total A was significantly decreased by 50% in mice receiving r AAV-ECE-HA in both the hippocampus and the anterior cortex (Fig. 5e). Congophilic plaque load was also analyzed following intracranial injections of rAAV-ECE-HA vector. The area of congophilic labeling was substantially less than A immunochemistry, staining only compacted fibrillar A deposits, as expected (Gordon et al., 2002). Figures 6c and 6d indicate that the positive congophilic staining for the mice receiv ing the ECE vector was visibly less, especially in the hippocampus, compared to the control animals. The reductions in congophilic deposition were limited to t he site of injection in cortex and hippocampus. Quantificati on of the Congo red staining by ANOVA analysis revealed that animals receiving the rAAV expressing ECE-HA showed significant reductions in the hippocampus (62%) as well as significant reductions in the anterior cortex (46%; Fig 6e). DISCUSSION There have been several gene therapy approaches examined for the treatment of Alzheimer’s disease. Initial experiments were based on the

PAGE 40

32 observation that application of neuronal growth factors medi ated neuroprotection and reduced the loss of cholinergic neurons in experimental lesion models (Hefti and Weiner, 1986);(Williams et al., 1986). Th erefore, NGF (ner ve growth factor) genes have been delivered to the brain using recombinant viral vectors, such as AAV and lentivirus, or delivered via an ex vivo approach, with transformed fibroblasts or neuronal stem cells (Tu szynski, 2007),(Cenciarelli et al., 2006). These studies have shown rescue of c holinergic neurons and are being further examined as a potential therapy. However, it is unlik ely that NGF would cure AD because the widespread neurona l loss that occurs in the later stages of the disease would not be compensated by the early rescue of cholinergic neurons. NGF rescue of neurons is currently being ex amined in clinical trials because it offers a substantial decline in the ra te of neuronal loss and could slow the progression of symptoms of Alzheimer’s. Other growth factor s that are also of potential interest in AD include br ain-derived neurotrophic factor and neurotrophin-4/5 to address the neuronal loss in the cortex and hippocampus. A second approach to treat Alzheimer’s pathology involves the reduction of A through immunization. It was obs erved that immunization against A would result in a significant reduction in A deposits and improve learning and memory deficits (Morgan et al., 2000),(Janus et al., 2000). Peripheral passive immunization against A with antibody injections has been successful in mouse models and more recently viral vectors have been used as a gene therapy approach to deliver A antibodies (Fukuchi et al., 2006), (Levites et al., 2006). This approach has demonstrated some pr omising reduction (25-50%) in the

PAGE 41

33 levels of A deposition. AAV vectors have also been used as an A vaccine, with the over expression of Abeta42. This method has shown reduction in A deposition and some reduction in cognitive impairment in Alzheimer Tg mice (Mouri et al., 2007), (Hara et al ., 2004), (Zhang et al., 2003). It has been demonstrated that the apolipoprotein E ( apoE) gene is a major risk factor for late-onset AD with t he ApoE2 allele decreasing and the ApoE4 allele increasing the morbid risk for dev eloping AD. Therefore to develop a gene therapy approach to Alzheimer’s, Dodart et al. (2005) over expressed ApoE2 using a lentiviral vector. The aut hors observed a dramatically reduced hippocampal A burden in Tg mice. It is as ye t unclear how ApoE2 is reducing the A but it may be increasing the A clearance (Dodart et al., 2005). A final gene therapy approach for the treatment of Alzheimer’s has been the over expression of A degrading enzymes such as nepr ilysin. Either using a direct transduction of neurons using a vira l vector or an ex vivo approach with transformed fibroblasts, the increase in NEP expression has been shown to significantly reduce the overall A load on Alzheimer Tg mice (Marr et al., 2003), (Iwata et al., 2004),(Hong et al., 2006), (He mming et al., 2007). In this report we are examining another A degrading enzyme, ECE, for its ability to reduce the A load on Alzheimer APP + PS1 mice using rAAV for increased neuronal expression of ECE. Recombinant AAV has become widely used for the transduction of neuronal cells. AAV is nonpathogenic, has low immunogenicity, lacks all viral genes and is capable of long term expre ssion in neurons. This makes rAAV a

PAGE 42

34 good candidate for the use as a gene therapy vector fo r neurological disorders (Mandel et al., 2006), (Burger et al., 2005). Further, rAAV is currently being examined in a number of neurological clin ical trials (Mandel and Burger, 2004). Here we are using rAAV for the over ex pression of the ECE protease which had been previously shown to cleave A peptide (Eckman and Eckman, 2005). We believed that the over ex pression of ECE using rAAV would enable the clearance of A deposition within the mouse brain and ameliorate the pathogenesis of Alzheimer’s. rAAV serotype-5 ECE-HA vector was in jected unilaterally into the mouse hippocampus and anterior cortex. Examinat ion of the expression profile of the expressed ECE has shown that the rAAV of ECE can transduce several different neuronal cell types within the mouse brai n, and possible other cell types in the dentate molecular layer. This is cons istent with the pr eviously published serological specificity of this vector (Alisky et al., 2000) (Burger et al., 2004),(Choi et al., 2005). The expression profile of ECE was similar to that observed with our rAAV5 vector expressing GFP. No noticeable toxic effects were s een in mice receiving the ECE-HA vector compared to control animals. No neuron loss or gross morphometric changes were observed in fixed brain tissue nor were there any signs of general toxicity as evidence by abnormal behavior in the mice. Only one animal died throughout the course of the st udy but this did not appear to be a result of toxicity from the over expression of the ECE protein. All animals were weighed before treatment and immediately bef ore sacrifice and no signific ant changes in weight

PAGE 43

35 were noted indicating that up regulati on of ECE does not appear to have adverse effects or cause general toxicity in t he mouse model, yet further investigation must be done to confirm it s safety profile. ECE has many endogenous peptides including opioid peptides, tachykinin, at rial naturetic peptides, and other small regulatory vasoactive peptides. It is uncertain whether interactions between other potential substrates in the brain may limit the ability of ECE to degrade A or cause potential problems by reducing the levels of these other peptides. Previous studies have revealed that in t he AD brain as well as in the animal model of AD, ECE (in addition to other A degrading enzymes) are down regulated specifically in areas that are prone to plaque formati on (Iwata et al., 2002), (Yono et al., 2004). Therefore, upregulation of ECE to restore normal levels of this endogenous protease should have minimal adverse effects. In addition, ECE has been shown to have higher affinity for larger peptides such as bradykinin, substance P, and neurotensin hydrolyzing them at amino acid hydrophobic residues (Johnson et al., 1999) (Lo et al., 1999). Monitoring changes in these peptides between c ontrol and treatment groups may help identify other potential toxic effects result ing from increases in ECE that could occur in this model. Our results from the over expr ession of ECE suggest that the upregulation of ECE through r AAV vectors can provide a viable method to decrease total amyloid deposition in the brain. The activity level determined using an ECE specific assay demonstrated that we achieved a several fold fold increase in ECE activity in homogenates, most of which wa s membrane associated. This increase

PAGE 44

36 in ECE activity was able to significantly reduce total A deposition in the anterior cortex and hippocampal sites of injection (Fig. 5). Similarly, when Congo red stained compacted deposits were measured; the ECE-HA vector significantly reduced staining (Fig. 6). We were able to achieve a 50% reduction in the total A present in the cortex and hippocampal regions. Similarly, we reduced the congophilic compact plaque load to ~50% of that of the controls. We have yet to determine if this reduction in the A levels in the mice with the ECE will lead to significant improvement in behavioral test s such as the Morris water maze. We are currently testing more injections of ECE to see if we can create a significant reduction of whole brain A levels and ameliorate the memory deficits that are observed in the APP + PS1 mice. Our data are consistent with reports that ECE can degrade A in vitro, and that partial knockdown of the ECE gene leads to more rapid accumulation of A (Eckm an et al., 2001), (Eckman et al., 2003). The present work adds to the evidence that ECE plays an important role in A deposition by demonstrating that local over expression of the enzyme activity can dramatically reduce the deposition of amyloid in the brains of APP + PS1 transgenic mice. Thus, regulation of EC E may be used as therapeutic target for the treatment of Alzheimer's disease. MATERIALS AND METHODS Generation of ECE Constr ucts and rAAV Production The ECE-1 coding sequence (GI:4503442) was cloned using polymerase chain reaction (PCR) from a GenePool cDNA li brary obtained from Invitrogen. The

PAGE 45

37 primers used for the full length ECE were GAGGAATTCACCGGTCCACCATGC GGGGCGTGTGGCCGCCCCCGGTGTC and GAGATCGATTACCAGACTTCACACTTGT GAGGCGG. The PCR product was cloned into pBluescript and sequenced to confirm sequence identity. The ECE was then cloned into the ve ctor called pTR5-MCS at the Eco RI and Cla I cloning sites. This vector contains the AAV terminal repeats for AAV virus production and the CBA promot er for ECE mRNA transcription. A Hemagglutinin (HA) tag was added to the C-terminus of ECE synthetic gene using the following oligonucleotide GTGTGAAGTCTGGATGGCTTCTAGCTA TCTTATGACGTGCCTGACTATGCCA TGTAA and its complement. The recombinant viruses were generated and purified using the method of Zolotukhin et al. (2002). Infectious rAAV particles are expressed as vector genomes (v/g )/mL. Vector genomes were quantitated using the dot plot protocol, with a probe for the CBA promoter, as described by (Zolotukhin et al., 2002). Western Blot Analysis Protein samples were boiled in Lamaelli sample buffer prior to loading on a 10% polyacrylamide gel. The proteins on t he gel were transferred to Millipore Nitrocellulose membrane and probed with an ti-HA antibody (Santa Cruz) and antimouse-HRP conjugate (Amersham). Millipore Immobilon detection reagent was used to visualize bands. Transgenic Mice

PAGE 46

38 APP + PS1 mice (Holcomb et al., 1998) were acquired from the breeding colonies at the University of South Flor ida. Multiple mice were housed together whenever possible until the time of the experiment. Mice were then singlyhoused 1 week before surgical procedures un til the time of sacrifice. Study animals were given water and food ad libitum and maintained on a twelve hour light/dark cycle and standard vivarium condi tions. Two cohorts of mice were used, the first cohort consisted of APP + PS1 mice aged 6 months ( n =16) and the second cohort consisted of nontransgenic mice aged 9 months ( n =12). Animals in each cohort were assigned to two groups. Group one received a control vector expressing GFP (first cohort n =8; second cohort n=6). Group two received AAV vector expressing t he membrane bound endothelin converting enzyme sequence (first cohort n =8; second cohort n=6). All groups were sacrificed after six weeks post intracranial injection. Surgical Procedure Immediately before sur gery mice were weigh ed then anesthetized using isoflurane. Surgery was performed usi ng a stereotaxic apparat us. The cranium was exposed using an incision through the skin along the medi an sagittal plane, and two holes were drilled through the cranium over the right cort ex injection site and the right hippocampal injection site. Previously determined coordinates for burr holes, taken from bregma were as fo llows; frontal cortex, anteroposterior, 1.5 mm; lateral, -2.0 mm, vertical, 3.0 mm, hippocampus, anteroposterior, 2.7mm; lateral -2.5 mm, vertical, 3.0 mm Burr holes were drilled using a dental drill bit (SSW HP-3, SSWhite Burs Inc. Lakewood, NJ). Injections of 2 l of total

PAGE 47

39 volume of each of the viral vectors in sterile PBS at a concentration of 1 X 1012 vg/ml were dispensed into hippocampus and cortex over a period of 4 min using a 26 gauge needle attached to a 10 l syringe (Hamilton Co., Reno, NV). The incision was then cleaned and closed with su rgical staples. Animals were recovered within 10 minutes and housed si ngly until time of sacrifice. Immunohistochemistry Six weeks post surgery, mice were weighed, overdosed with pentobarbital (200 mg/kg;) and perfused with 25 ml of 0.9% normal saline solution then 50 ml of freshly prepared 4% paraformaldehyde. Brains were collected from the animals immediately following perfus ion and immersion fixed in 4% paraformaldehyde for 24 h. Mouse brains were cryoprotected in successive incubations in 10%, 20%, 30% solutions of sucrose; 24 h in each solution. Subsequently, brains were frozen on a co ld stage and sectioned in the horizontal plane (25 m thickness) on a sliding micr otome and stored in Dulbecco’s phosphate buffered saline (DPBS) with 0. 2% sodium azide solution at 4 C. Six sections 100 m apart spanning the site of injection were chosen and free-floating immunochemical and hist ological analysis was performed to determine ECE expression using anti-HA biotinylated rabbit polyclonal antibody at a concentration of 1:1000 (Roc he, Indianapolis, IN), total A using a rabbit primary anti-A serum at a concentr ation 1:10,000 and a secondary anti-rabbit antibody (Serotec, Raleigh, NC). Another series of sections were mounted on slides and stained with Congo red to asse ss compact congophilic positive plaque load. Immunohistochemical procedural methods are analogous to those

PAGE 48

40 described by Gordon et al. 2002, for each ma rker. Six to eight sections from each animal were placed in multisample staining tray and endogenous peroxidase was blocked (10% methanol, 3% H202, in PBS). Tissue samples were permeabilized (with 0.2% lysine, 1% Triton X-100 in PBS solution), and incubated overnight in appropr iate primary antibody. Sections were washed in PBS then incubated in corresponding biotinylated secondary antibody (Vector Laboratories, Burlingame, CA). T he tissue was again washed after a 2 h incubation period and then in cubated with Vectastin Elite ABC kit (Vector Laboratories, Burlingame, CA ) for enzyme conjugation. Finally, sections were stained using 0.05% diam inobenzidine and 0.3% H202. For anti-HA 0.5% nickelous ammonium sulf ate was added for color enhanc ement. Tissue sections were then mounted onto slides, dehydr ated, and cover slipped. Each immunochemical assay omitted some sect ions from primary antibody incubation period to evaluate nonspecific r eaction of the secondary antibody. Congo red histology was performed usi ng sections that were premounted on slides then air dried for a minimum of 24 hours. The sections were rehydrated for 30 seconds before beginning staining protocol. For Congo red, hydrated sections were incubated in an freshl y prepared alkaline alcoholic saturated sodium chloride solution (2.5mM NaOH in 80% alcohol) for 20 min, then incubated in 0.2% congo red in alkaline alcoholic saturated sodium chloride solution for 30 minutes. Slides with sections were rinsed through three changes of 100% ethanol, and cleared through three changes of xylene and finally coverslipped with DPX. Histological sect ions from control animals treated with

PAGE 49

41 rAAV expressing GFP were mounted on slides and dehydrated through a series of increasing concentrations of ethanol. The mounted sections were then cleared in three changes of histoc lear and coverslipped with DPX. Stained sections were imaged using an Evolution MP digital camera mounted on an Olympus BX51 microscope at 100X final magnification (10 X objective). Six horiz ontal brain sections (100m apart; every 4th section) were taken from each animal and four nonoverlapp ing images near the si te of injection from each of these sections were capt ured (24 measurements per mouse). All images were taken from the same locati on in all animals. Quantification of positive staining product surrounding and in cluding the injection sites in the right anterior cortex and the right hippocam pus and the corresponding regions in the left hemisphere were determined using Image-Pro Plus (Media Cybernetics, Silver Springs, MD). Q uantifications of the right regions, ispilateral to the injection site, were calculated and ANOVA statistical analysis was performed using StatView version 5.0.1 (S AS Institute, Raleigh, NC). Enzyme Activity Assay ECE activity was characterized and adapt ed from a previous fluorometric assay method (Johnson and Ahn, 2000) for a 96 well plate format with slight modifications. Six weeks post injection brain tissue was removed then dissected immediately following sacrifice and froz en at -80 C. The tissue from each animal was rapidly thawed and homogenize d with the Ultra-Turrax T8 motordriven homogenizer (IKA-Labortechnik, Ge rmany) in solubilizing buffer containing 20 mM Tris-HCl, pH 7.4, 250 mM sucrose immediately prior to

PAGE 50

42 assaying (Iwata et al., 2002). Samples we re then centrifuged at 100,000 x g, 4 C for 45 min using an Avanti J-30I Centri fuge (Beckman Instruments, Inc., Palo Alto, CA) to obtain a soluble fraction and a membrane fraction. The membrane fraction was resuspended in buffer containing 20 mM Tris-HCl, 250 mM sucrose, pH 7.4. Protein concentration was determined from the homogenate, membrane fraction and soluble fractions from eac h sample using a general BCA assay (Pierce). Aliquots of the cell lysate (5 g) were incubated with 10 M (final) of the fluorogenic peptide (MOCA Arg-Pro-Pro-Gly-Phe-Ser-Ala-Phe-Lys(Dnp)OH), (R&D Systems) in MES buffer (s odium phosphate pH 6.8 containing 0.1M NaCl). MOCAArg-Pro-Pr o-Gly-Phe-Ser-Ala-Phe-Lys(D np)-OH is efficiently quenched by resonance energy transfer to the dinitrophenyl group and the continuous fluorescent intensity is increased upon internal cleavage of the peptide (ECE-1 cleavage between t he Ala-Phe bond). The increased fluorescence produced from cleavage of the substrate was measured using a Molecular Devices fMax spectrofluorom eter plate reader (MDS Analytical Technologies, Sunnyvale, CA) with a 60 mi n time point to normalize independent experiments. A standard curve of (7-m ethoxycoumarin-4-yl) acetyl (MOCA) was analyzed along with each assay and used to convert the relative fluorescence units (RFU) to the moles of MOCA produced by the respective cell lysates. Parallel assay reactions of all samples (in triplicate) were carried out in the presence of an ECE specific inhibi tor (SM19712, Sigma Aldrich) at a concentration of 20 M (Umekawa et al., 2000), (Matsumura et al., 2000). ECE specific activity was considered to be the difference in RFU between samples

PAGE 51

43 including SM19712 from the total activity samples (no SM19712). Values were calculated and expressed as nanomoles MCA/min/ g protein. Acknowledgments. Supported by The J ohnnie Byrd Center for Alzheimer's Research, NIH grants AG -25509, AG 15490, AG 18478, AG 04418, AG 25711

PAGE 52

44 FIGURE 1: (A) DIAGRAM OF ECE CONSTRUCT AND (B) WESTERN ANALYSIS OF ECE EXPRESSION IN HEK 293 CELLS. (A) Diagrammatic representation of the rAAV construct expressing the ECE synthetic gene under the control of the chicken -actin (CBA) promoter. (B ) Western analysis of ECE expressed in 293 cells using an anti-HA ant ibody. Lane 1, 293 cell lysate; lane 2, ECE transfected cell lysate, lane 3, condi tioned media from untransfected cells, lane 4 media from cells transfected with ECE.

PAGE 53

45 FIGURE 2: ECE ENZYMATIC ACTIVITY OBTAINED FROM MOUSE HIPPOCAMPAL REGIONS AFTER IN JECTION OF rAAV VIRUS EXPRESSING ECE PROTEIN ECE-like specific activity is higher in the membrane fraction and homogen ate of the hippocampa l tissue in the ECE treatment group compared to the control group. The star (*) indicates significance with a p-value < 0.05. Erro r bars indicate standard error of the mean value from triplicates for each sample (specific activity/min/ g protein) for each treatment group (n=6 per group).

PAGE 54

46 FIGURE 3: EXAMINATION OF ECE EXPRESSION USING ANTI-HA IMMUNOREACTIVITY IN THE HIPPO CAMPUS AND ANTERIOR CORTEX Panel A shows slight ECE expression in gr anule cells of the dentate gyrus of the left contra lateral hippocampus following in tracranial administration of ECE into the right hippocampus. Panel B shows st rong ECE expression detected at the site of vector injection. Panel C shows strong ECE expression in the anterior cortex. Panel D shows slight ECE expression along the midline and some expression in the lateral septum of left (c ontra lateral) side following intracranial administration of ECE into t he right anterior cortex. fi = fimbtria; LS = lateral septum; CC = corpus callosum; dg = dent ate gyrus; CA1 = Cajal Area 1; CA3 = Cajal area 3; Magnification = 40X and Scale bar = 120 m

PAGE 55

47 FIGURE 4: ECE AND GFP EXPRESSION PROFILES IN MICE BRAINS FOLLOWING rAAV ADMINISTRATION Panels A shows strong ECE expression in the right anter ior cortex following intracr anial administration of ECE vector. Panel B shows GFP expressi on in the anterior cortex following intracranial administration of cont rol GFP vector. Panel C shows ECE expression in pyramidal cells in CA2 r egion of the right hi ppocampus. Panel D shows GFP expression in the CA3 region of the hippocampus. Panel E shows slight ECE expression in CA4 neurons of the dentate gyrus of the left contra lateral hippocampus. Panel F shows no pos itive expression in the left uninjected hippocampus following intracranial inje ction of GFP vector into the right hippocampus. Scale bar = 50 m (panels A, C, E). Scale bar = 25 m (panels B, D, F).

PAGE 56

48 FIGURE 5: TOTAL AMYLOID LOAD IS REDUCED FOLLOWING INTRACRANIAL ADMINISTRAT ION OF rAAV-ECE-HA VECTOR Panels A and B show A immunostaining in the right co rtex and hippocampus respectively of animals receiving intracranial injection of control vector, GFP. Panels C and D show A immunostaining in right cortex and hippocampal regions respectively of mice receiving intracranial inje ction of ECE. Scale bar = 120 m. Panel G shows the percent area of positive staining, no rmalized to the control, for both cortex (left) and hippocampus (right) after AAV-ECE injection. The asterisk (*) indicates significant reduction compared with contro l values with p-values < 0.05.

PAGE 57

49 FIGURE 6: CONGOPHILIC COMPACT PL AQUE LOAD IS REDUCED FOLLOWING INTRACRANIAL ADMIN ISTRATION OF ECE-HA rAAV VECTOR Panels A and B show total congophilic staining in the right cortex and hippocampus respectively of animals receiv ing intracranial injection of control vector, GFP. Panels C and D show congophi lic positive staining in right cortex and hippocampal regions respectively of mi ce receiving intracranial injection of ECE. Magnification = 40X, scale bar = 120 m. Panel E shows the percent area of positive staining, normalized to the control, for the cortical (right) and hippocampal (left) injected regions. The star (*) indicates significance with a pvalue < 0.05.

PAGE 58

50 REFERENCE LIST Alisky JM, Hughes SM, Sauter SL, Jo lly D, Dubensky TW, Jr., Staber PD, Chiorini JA, Davidson BL (2000) Transduc tion of murine ce rebellar neurons with recombinant FIV and AAV5 vect ors. Neuroreport 11:2669-2673. Burger C, Nash K, Mandel RJ (2005) Recombinant adeno-associated viral vectors in the nervous syst em. HumGene Ther 16:781-791. Burger C, Gorbatyuk OS, Velardo MJ, P eden CS, Williams P, Zolotukhin S, Reier PJ, Mandel RJ, Muzyczka N (2004) Reco mbinant AAV viral vectors pseudotyped with viral capsids from serotypes 1, 2, and 5 display differential efficiency and cell tropism after delivery to different regions of the central nervous system. MolTher 10:302-317. Caccamo A, Oddo S, Sugarman MC, Akbar i Y, LaFerla FM (2005) Ageand region-dependent alterations in Abetadegrading enzymes: implications for Abeta-induced disorders. NeurobiolAging 26:645-654. Cenciarelli C, Budoni M, Mercanti D, Fer nandez E, Pallini R, Aloe L, Cimino V, Maira G, Casalbore P (2006) In vitr o analysis of mouse neural stem cells genetically modified to stably express human NGF by a novel multigenic viral expression system. NeurolRes 28:505-512. Choi VW, McCarty DM, Samulski RJ (2005) AAV hybrid serotypes: improved vectors for gene delivery. CurrGene Ther 5:299-310. Dodart JC, Marr RA, Koistinaho M, Greger sen BM, Malkani S, Verma IM, Paul SM (2005) Gene delivery of human apolipoprotein E alters brain Abeta burden in a mouse model of Alzheimer's diseas e. Proc Natl Acad Sci U S A 102:12111216. Eckman EA, Eckman CB (2005) Abeta-degrading enzymes: modulators of Alzheimer's disease pathogenesis and tar gets for therapeutic intervention. BiochemSocTrans 33:1101-1105. Eckman EA, Reed DK, Eckman CB (2001) Degradation of the Alzheimer's amyloid beta peptide by endothelin-c onverting enzyme. JBiolChem 276:2454024548. Eckman EA, Watson M, Marlow L, Sambamurti K, Eckman CB (2003) Alzheimer's disease beta-amyloid pepti de is increased in mice deficient in endothelin-converting enzyme. JBiolChem 278:2081-2084. Eckman EA, Adams SK, Troendle FJ, Stodol a BA, Kahn MA, Fauq AH, Xiao HD, Bernstein KE, Eckman CB (2006) Regulatio n of steady-state beta-amyloid levels

PAGE 59

51 in the brain by neprilysin and endothelin -converting enzyme but not angiotensinconverting enzyme. JBiolChem 281:30471-30478. Fukuchi K, Tahara K, Kim HD, Maxwell JA Lewis TL, Accavitti-Loper MA, Kim H, Ponnazhagan S, Lalonde R (2006) Anti-Abet a single-chain antibody delivery via adeno-associated virus for treatment of Alzheimer's disease. NeurobiolDis 23:502-511. Gordon MN, Holcomb LA, Jantzen PT, DiCar lo G, Wilcock D, Boyett KW, Connor K, Melachrino J, O'Callaghan JP, Mor gan D (2002) Time course of the development of Alzheimer-like pathol ogy in the doubly transgenic PS1+APP mouse. Exp Neurol 173:183-195. Hara H, Monsonego A, Yuasa K, Adachi K, Xiao X, Takeda S, Takahashi K, Weiner HL, Tabira T (2004) Development of a safe oral Abeta vaccine using recombinant adeno-associated virus ve ctor for Alzheimer's disease. JAlzheimersDis 6:483-488. Hardy J, Selkoe DJ (2002) The amyloid hypothesis of Alzheimer's disease: progress and problems on the road to ther apeutics. Science %19;297:353-356. Hefti F, Weiner WJ (1986) Nerve gr owth factor and Alzheimer's disease. AnnNeurol 20:275-281. Hemming ML, Patterson M, Reske-Nielsen C, Lin L, Isacson O, Selkoe DJ (2007) Reducing amyloid pl aque burden via ex vivo g ene delivery of an Abetadegrading protease: a novel therapeutic approach to Alzheimer disease. PLoSMed 4:e262. 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. NatMed 4:97-100. Hong CS, Goins WF, Goss JR, Burton EA, Glorioso JC (2006) Herpes simplex virus RNAi and neprilysin gene transfer vectors reduce accumulation of Alzheimer's disease-related amyloidbeta peptide in vivo. Gene Ther 13:10681079. Hunter AR, Turner AJ (2006) Expr ession and localization of endothelinconverting enzyme-1 isoforms in human end othelial cells. ExpBiolMed(Maywood) 231:718-722. Iwata N, Takaki Y, Fukami S, Tsubuk i S, Saido TC (2002) Region-specific reduction of A beta-degradi ng endopeptidase, neprilysin, in mouse hippocampus upon aging. JNeurosc iRes 70:493-500.

PAGE 60

52 Iwata N, Mizukami H, Shirotani K, Taka ki Y, Muramatsu S, Lu B, Gerard NP, Gerard C, Ozawa K, Saido TC (2004) Presynaptic localization of neprilysin contributes to efficient clearance of amyloid-beta peptide in mouse brain. JNeurosci 24:991-998. Jackson CD, Barnes K, Homer-Vanniasinkam S, Turner AJ (2006) Expression and localization of human endothelin-c onverting enzyme-1 isoforms in symptomatic atherosclerotic disease and saphenous vein. ExpBiolMed(Maywood) 231:794-801. Janus C, Pearson J, McLaurin J, Mathew s PM, Jiang Y, Schmidt SD, Chishti MA, Horne P, Heslin D, French J, Mount HT Nixon RA, Mercken M, Bergeron C, Fraser PE, St GeorgeHyslop P, Westaway D (2000) A beta peptide immunization reduces behavioural impai rment and plaques in a model of Alzheimer's disease. Nature 408:979-982. Johnson GD, Stevenson T, Ahn K (1999) Hydrolysis of peptide hormones by endothelin-converting enzyme-1. A comparison with nepr ilysin. J Biol Chem 274:4053-4058. Leissring MA, Murphy MP, Mead TR, Ak bari Y, Sugarman MC, Jannatipour M, Anliker B, Muller U, Saftig P, De Strooper B, Wolfe MS, Golde TE, LaFerla FM (2002) A physiologic signaling role for the gamma -secretase-derived intracellular fragment of APP. ProcNa tlAcadSciUSA 99:4697-4702. Levites Y, Jansen K, Smithson LA, Daki n R, Holloway VM, Das P, Golde TE (2006) Intracranial adeno-associated viru s-mediated delivery of anti-pan amyloid beta, amyloid beta40, and amyloid bet a42 single-chain variable fragments attenuates plaque pathology in amyloid precursor protein mice. JNeurosci 26:11923-11928. Lo WD, Qu G, Sferra TJ, Clark R, Chen R, Johns on PR (1999) Adeno-associated virus-mediated gene transfer to the brain: duration and modulatio n of expression. Hum Gene Ther 10:201-213. Mandel RJ, Burger C (2004) Clinical trials in neurological disorders using AAV vectors: promises and challe nges. CurrOpinMolTher 6:482-490. Mandel RJ, Manfredsson FP, Foust KD, Risi ng A, Reimsnider S, Nash K, Burger C (2006) Recombinant adeno-associated vi ral vectors as therapeutic agents to treat neurological disorder s. Mol Ther 13:463-483. Marr RA, Rockenstein E, Mukherjee A, Kindy MS, Hersh LB, Gage FH, Verma IM, Masliah E (2003) Neprilysin gene tr ansfer reduces hum an amyloid pathology in transgenic mice. JNeurosci 23:1992-1996.

PAGE 61

53 Matsumura Y, Kuro T, Kobayashi Y, Um ekawa K, Ohashi N, Takaoka M (2000) Protective effect of SM-19712, a nov el and potent endothelin converting enzyme inhibitor, on ischemic acute renal failu re in rats. Jpn J Pharmacol 84:16-24. Morgan D, Diamond DM, Gottsch all PE, Ugen KE, Dickey C, Hardy J, Duff K, Jantzen P, DiCarlo G, Wilcock D, Connor K, Hatcher J, Hope C, Gordon M, Arendash GW (2000) A beta peptide vaccina tion prevents memory loss in an animal model of Alzheimer's disease. Nature 408:982-985. Mouri A, Noda Y, Hara H, Mizoguchi H, Tabira T, Nabeshima T (2007) Oral vaccination with a viral vector containi ng Abeta cDNA attenuates age-related Abeta accumulation and memory deficit s without causing inflammation in a mouse Alzheimer model. FASEB J 21:2135-2148. Mueller-Steiner S, Zhou Y, Arai H, Rober son ED, Sun B, Chen J, Wang X, Yu G, Esposito L, Mucke L, Gan L (2006) Antiamyloidogenic and neuroprotective functions of cathepsin B: implications for Alzheimer's disease. Neuron 51:703714. Muller L, Barret A, Etienne E, Meidan R, Valdenaire O, Corvol P, Tougard C (2003) Heterodimerization of endothelin-c onverting enzyme-1 isoforms regulates the subcellular distribution of this metalloprotease. JBiolChem 278:545-555. Shimada K, Takahashi M, Turner AJ, Tanzawa K (1996) Rat endothelinconverting enzyme-1 forms a dimer thr ough Cys412 with a similar catalytic mechanism and a distinct substrate binding mechanism compared with neutral endopeptidase-24.11. Bi ochemJ 315:863-867. Turner AJ, Fisk L, Nalivaeva NN (2004) Targeting amyloid-degrading enzymes as therapeutic strategies in neurodegenerat ion. AnnNYAcadSci 1035:1-20.:1-20. Tuszynski MH (2007) Nerve growth factor gene therapy in Alzheimer disease. Alzheimer DisAssocDisord 21:179-189. Umekawa K, Hasegawa H, Ts utsumi Y, Sato K, Matsum ura Y, Ohashi N (2000) Pharmacological characterization of a novel sulfonylureid-p yrazole derivative, SM-19712, a potent nonpeptidic inhibitor of endothelin conver ting enzyme. Jpn J Pharmacol 84:7-15. Westerman MA, Cooper-Blacketer D, Marias h A, Kotilinek L, Kawarabayashi T, Younkin LH, Carlson GA, Younkin SG, As he KH (2002) The relationship between Abeta and memory in the Tg2576 mouse model of Alzheimer's disease. J Neurosci 22:1858-1867. Wilcock DM, Alamed J, Gottschall PE, Gri mm J, Rosenthal A, Pons J, Ronan V, Symmonds K, Gordon MN, Morgan D ( 2006) Deglycosylated anti-amyloid-beta antibodies eliminate cognitive deficit s and reduce parenchymal amyloid with

PAGE 62

54 minimal vascular consequences in aged am yloid precursor protein transgenic mice. JNeurosci 26:5340-5346. Williams LR, Varon S, Pete rson GM, Wictorin K, Fisch er W, Bjorklund A, Gage FH (1986) Continuous infusion of nerve growth factor prevents basal forebrain neuronal death after fimbri a fornix transection. ProcNatlAcadSciUSA 83:92319235. Yan P, Hu X, Song H, Yin K, Bateman RJ, Cirrito JR, Xiao Q, Hsu FF, Turk JW, Xu J, Hsu CY, Holtzman DM, Lee JM (2006) Matrix metalloproteinase-9 degrades amyloid-beta fibrils in vitro and compact plaques in situ. JBiolChem 281:24566-24574. Yasojima K, Akiyama H, McGeer EG, McGeer PL (2001) Reduced neprilysin in high plaque areas of Alzhei mer brain: a possible relationship to deficient degradation of beta-amyloid pept ide. NeurosciLett 297:97-100. Yono M, Latifpour J, Takahashi W, Poures mail M, Afiatpour P, Weiss RM (2004) Age-related changes in the properties of the endothelin rec eptor system at protein and mRNA levels in the rat va s deferens. J Recept Signal Transduct Res 24:53-66. Zhang J, Wu X, Qin C, Qi J, Ma S, Zhang H, Kong Q, Chen D, Ba D, He W (2003) A novel recombinant adeno-asso ciated virus vaccine reduces behavioral impairment and beta-amyloid plaques in a mouse model of Alzheimer's disease. Neurobiol Dis 14:365-379. Zolotukhin S, Potter M, Zolo tukhin I, Sakai Y, Loiler S, Fraites TJ, Jr., Chiodo VA, Phillipsberg T, Muzyczka N, Hauswirth WW, Flotte TR, Byrne BJ, Snyder RO (2002) Production and purificat ion of serotype 1, 2, and 5 recombinant adenoassociated viral vector s. Methods 28:158-167.

PAGE 63

55 PAPER 2: RECOMBINANT ADENO-ASSOCIATED VI RAL VECTOR MEDIATED GENE DELIVERY OF SECRETED NEPRILYSIN REDUCES -AMYLOID DEPOSITION IN APP + PS1 TRANSGENIC MICE. Carty N. 1, Wilcock, D.M. 1, Lee D.C. 1, Gottschall P.E.3, Burger, C.4, Mandel, R.J.5, Gordon M.N.1, Muzyczka N.,2 Morgan D1, Nash K.1 1Alzheimer’s Research Laboratory, Departm ent of Molecular Pharmacology and Physiology, School of Biomedical Sciences University of South Florida College of Medicine, 12901 Bruce B Downs Blvd, Tampa, FL 33612, USA. 2 Department of Molecular Genetics and Microbiology, University of Florida College of Medicine, Box 100266, Gainesville FL 32610 USA. 3 University of Arkansas for Medical Sciences Department of Pharmacology and Toxicology, Slot #611, 4301 West Markham Street Little Rock, Arkansas 722057199 4 Department of Neurology, Medical Sciences Center, 1300 University Av, Rm 73 Bardeen, Madison WI 53706 USA. 5 Department of Neuroscience, McKnight Br ain Institute, University of Florida College of Medicine, Box 100244, Gainesville FL 32610 USA. Niki C Carty: ncarty@hsc.usf.edu ; Donna M. Wilcock: ; Daniel Lee: dlee1@hsc.usf.edu; Mary Mercer: mmercer@mail.usf.edu ; Paul E. Gottschall:

PAGE 64

56 pegottschall@uams.edu ; C. Burger: burger@neurology.wisc.edu ; R. Mandel: rmandel@mbi.ufl.edu; Marcia N Gordon: mgordon@hsc.usf.edu ; Nicholas Muzyczka: NMuzyczka@ufl.edu; Dave Morgan: dmorgan@hsc.usf.edu ; Kevin Nash: nash@ufl.edu Please address correspondence to: Dave Morgan Director, Basic Neuroscience Research University of South Florida 12901 Bruce B Downs Blvd, MDC Box 8 Tampa, FL 33612 USA Key Words: Alzheimer’s disease; Beta Amyloid, Gene Therapy, Viral Vector, Amyloid Degrading Enzyme, Zinc Metalloprotease.

PAGE 65

57 ABSTRACT The accumulation of -amyloid (A) peptides in the brain has been recognized as an instigating factor in Alzheimer’s disease (AD) pathology. Recently, it has been argued that clearance of A is partially dependent upon several endogenous zinc metalloprot eases. Gene therapy using adenoassociated viral (AAV) vect ors are an effective means of delivering transgenes encoding for these specific proteases which include neprilysin, endothelin converting enzyme, and insulin degrading enzym e to regions in the brain affected by AD pathology. In this study the convection enhanced delivery method using the step-design cannula (Bankiewicz, 2005) was used to deliver recombinant AAV (rAAV) vectors expressi ng either a native, membrane bound form of human neprilysin gene (NEP-n) or an engineered, se creted form of th e neprilysin gene (NEP-s) into the right hippoc ampus and right frontal cortic al regions of the mouse brain. The control group was treated wit h an rAAV vector expressing a mutant neprilysin gene (NEP-m) with a single amino acid substitution in the active site rendering it inactive. Six weeks after injection, immunohistochemistry for NEP revealed strong expression throughout the hippocampus in animals treated with the NEP-n and NEP-m vectors. Anim als treated with the NEP-s showed expression in a smaller portion of t he hippocampus compared to the NEP-n treated group. Immunohistochemistry for total A was significantly decreased in animals receiving the NEP-n and NEP-s viral vectors when compared to control animals in both the hippocampus and co rtex. Congo red staining followed a similar trend revealing significant decreas es in the hippocampal fissure and CA1

PAGE 66

58 regions and the cortex for NEP-n and NE P-s treatment groups compared to mice given control treatments. These data suggest that in creasing the expression of certain endogenous -amyloid degrading enzymes through gene therapy using the CED step cannula provides and effi cient means of increasing transgene distribution and ultimately may provi de a promising therapeutic avenue through which to treat AD. INTRODUCTION Alzheimer’s disease (AD) is the most common form of senile dementia that can occur sporadically or as a re sult of genetic mutations in the genes encoding presenilin 1, presenilin 2, or am yloid precursor protein (APP). These mutations result in the overproduction of A peptides which ultimately lead to the formation of extracellular amyloid plaques and intracellular neur ofibrillary tangles consisting of hyperphosphorylated tau The AD cases with these autosomal dominant mutations, commonly referred to as early onset AD or familial AD (FAD) (Fidani and Goate1992), account fo r only 1-2% of all AD cases. Interestingly, the more common late onset AD cases do not seem to over produce A peptide species suggesting that t here is a deficit in the clearance mechanisms which are normally involved in re gulation of A levels in the brain. Therefore, the accumulation of A in late onset AD is attributed to an imbalance between its production and degradation/clearance.

PAGE 67

59 Several endogenous proteases have been shown to degrade A in the brain and other tissues both in vivo and in vitro This family of zinc metalloproteases, include neprilysin (N EP), insulin degrading enzyme (IDE; insulysin), and endothelin converting enzymes (ECE-1 and ECE-2). Other proteases that appear to play a role in A metabolism include matrix metalloproteinase-9 (Yan et al., 2006) and cathepsin B (Mueller-Steiner et al., 2006) and plasmin (Turner et al., 2002). Down-regulation of these degrading enzymes within the brain during aging could potentially contribute to A accumulation eventually leading to develop ment of AD pathology (Caccamo et al., 2005); (Yasojima et al., 2001). NEP is a membrane bound a 92-kDa glycosylated ectoenzyme whose active site is oriented outside the cell. NEP belongs to the M13 sub family of neutral endopeptidases, and has a range of regulatory activities as it degrades numerous bioactive peptides (Turner et al ., 2000), (Iwata et al ., 2000a). Iwata et. al. in 2000 first demonstrated t he ability of NEP to degrade A peptides in the brain parenchyma and also ill ustrated that suppressing NEP would lead to an increase in A deposition (Iwata et al., 2000a). NEP activity decreases with age in amyloid depositing regions, such as the cortex and hippocampus, yet remained unchanged in areas such as t he striatum (Iwata et al., 2002), (Yasojima et al., 2001). NEP deficient mice showed a significant increase in amyloid deposition as opposed to their wild type counterparts indicating that NEP activity plays significantly impacts A pathology (Iwata et al., 2000c), (Marr et al., 2004).

PAGE 68

60 Recently, viral vector gene therapy has emerged as a viable approach to the delivery of potentially therapeutic genes. Several vectors and serotypes have been developed with improved efficiency, spec ificity, and safety (Thomas et al., 2003). In the present study we investi gate the effects of upregulating the NEP enzyme by using a recombinant adeno-associated viral (rAAV) vector, serotype 5, on A load in the brain using a nov el convection enhanced delivery administration technique to further incr ease area of distribution and potentially increase transduction efficiency of the trans gene (a limitation of viral vector gene therapy). rAAVs are parvovirus es that require a helper vi rus, typically Adenovirus or herpes simplex virus, in order to support viral rep lication (Berns and Parrish, 1996). rAAV vectors are desirable candidat es for gene therapy in the central nervous system because AAV is a nonpathogenic virus; it has low immunogenicity and it is defici ent for replication due to the removal of all the viral encoded genes. Further, rAAV serotype 5 vect ors are very efficient in infecting neuronal cells and maintaining long term expression (Burger et al., 2005), (Mandel et al., 2006). Theref ore we have chosen to ex amine the efficacy of a rAAV serotype 5 vector to deliver the recombinant expressed NEP cDNA. The rAAV vectors used in this study express either a membrane bound native form of the NEP cDNA, or a modified form of the NEP cDNA that is secreted into the extracellular compartment. We observed a significant reduction in the levels of A in the mice injected with the NEP viruses, both the native NEP and the secreted NEP appeared to reduce the levels of A load significantly compared to the mutated NEP in both the hi ppocampus and the anterior cortex.

PAGE 69

61 MATERIALS AND METHODS Generation of NEP Gene Constructs and rAAV Production: The NEP gene (GI:4503442) was cloned usi ng polymerase chain reaction (PCR) from a GenePool cDNA library obtained fr om Invitrogen. The primers used for the full length NEP were GAGGAATTCACCGGTCCACCATGC GGGGCGTGTGGCCGCCCCCGGTGTC (contains an Eco RI and Age I restriction sites, 5' primer) and GAGATCGATTACCAGACTTCAC ACTTGTGAGGCGG (contains a Cla I and Ale I restriction sites, 3' pr imer). The PCR product was cloned into pBluescript and sequenced to confirm sequence identity. The NEP was then cloned into the vector called pTR5-MCS at the Eco RI and Cla I cloning sites. This vector contains the AAV terminal repeats for AAV virus production and the CBA promoter for NEP mRNA tr anscription. A Hemagglutinin (HA) tag was added to the C-terminus of NEP gene at the Ale I restriction site. The following oligonucleotides GTGTGAAGTCTGGATGGCTTCTAGCTA TCTTATGACGTGCCTGACTATGCCA TGTAA and its compliment were annealed and ligated to the vector. Generating the soluble truncated form of NEP (232-2313 bp): The 5' primer that was used for this PCR was GAGGAATTCACCGGTGCAGGACTGGTGG CCTGCTTGGGCAGC, and the 3' primer was the same as above. The PCR product was cloned as described for the full length NEP including the addition of a HA-tag. To add the secretion signal sequence the following oligonucleoti de and its compliment were annealed and

PAGE 70

62 ligated at the Age I site of the construct [CCGGTCCACCATGAAGTTATGGGATG TCGTGGCTGTCTGCCTGGTGCTGC TCCACACCGGCGTCCG]. The sequence was conf irmed to be in frame with the NEP gene coding frame. This signal peptide sequence was derived from the GDNF gene. The NEP-m was generated vi a site directed mutagenesis of the NEP-n construct. The codon for glutamic acid 585, GAA, was changed to encode for valine, GTA. The recombinant viruses were generated and purified using the method of Zolotukhin et al. (2002). Infe ctious rAAV particles are expressed as vector genomes (v/g)/mL. Vector genomes were quantitated using the dot plot protocol, with a probe for the CBA promoter, as described by (Zolotukhin et al., 2002). Transgenic Mice APP + PS1 mice (Holcomb et al., 1998) were acquired from the breeding colonies at the University of South Flor ida. Multiple mice were housed together whenever possible until the time of the experiment; mice were then singlyhoused 1 week before surgical procedures un til the time of sacrifice. Study animals were given water and food ad libitum and maintained on a twelve hour light/dark cycle and standard vivarium conditions. Mice in this colony have been interbred for the last 8 years and hav e had the retinal degeneration mutation ( rd1 ) removed by selective breeding. In study 1, two cohorts of mice were used, the first cohort consisted of APP + PS1 mice aged 6 months ( n =32) and a second cohort 6 mo old APP + PS1 mice ( n =18) were treated at separate time s. Animals in each cohort were

PAGE 71

63 assigned to one of three groups. Group one received a control vector expressing mutated NEP (first cohort n =9; second cohort n=6). Group two received rAAV vector expressing native nepr ilysin transgene (first cohort n =12; second cohort n=6). Group three receiv ed the rAAV vector expressi ng the secreted neprilysin transgene (first cohort n = 11; second cohort n=6). All groups were sacrificed after six weeks post intracranial in jection at 7.5 mo of age. In study 2, APP + PS1 mice were 15 months of age at the time of treatment and were assigned to one of tw o groups. Group 1 received the NEP-s rAAV vector while group two received the control vector which was rAAV expressing GFP. Mice were sacrificed at 20 mo of age. Surgical Procedure Study 1 Immediately before sur gery mice were weigh ed then anesthetized using isoflurane. Surgery was performed usi ng a stereotaxic apparatus. The cranium was exposed using an incision through the skin along the medi an sagittal plane, and two holes were drilled through the cr anium over the right anterior cortex injection site and the right hippocampal injection site. Previously determined coordinates for burr holes, taken from bregma were as follows; anterior cortex, anteroposterior, 1.5mm; lateral, -2.0 mm, vertical, 3.0mm, hippocampus, anteroposterior, -2.7mm; lateral -2.5mm, vert ical, 3.0mm. Burr holes were drilled using a dental drill bit (SSW HP-3, SSWhite Burs Inc., Lakewood, NJ). Injections of 2 l of total volume of each of the viral vectors in sterile PBS at a concentration of 1.5 X 1011 vg/ml were dispensed into hippoc ampus and cortex over a period of

PAGE 72

64 2 min. using a 27 gauge step design cannula needle (see below) attached to a 10 l syringe (Hamilton Co., Reno, NV). The incision was then cleaned and closed with surgical staples. Animal s were recovered within 10 minutes and housed singly until time of sacrifice. Step Design Cannula The step design cannula was used for all intracranial surgeries. Fused silica tubing (polymicro technologies, Pheonix, AZ) was inserted into a 27 gauge Hamilton blunt ended needle and fixed in pl ace with super glue (Krauze et al., 2005). The end of the silica tubing was cut leaving 1mm of tubing protruding from the end of t he Hamilton needle. Study 2 Immediately before surgery fifteen m onth old mice were weighed then anesthetized using isoflurane. Sur gery was performed in a similar manner mentioned previously. Four holes were drilled through the crani um over the right and left cortex injection site and the right and left hippocampal injection sites with predetermined coordinates listed prev iously. NEP rAAV vectors were administered intracranially using a 26 gauge beveled Hamiliton needle (Hamilton Co., Reno NV). Injections of 2 l were given over a four min period at a rate of 0.5l/min. Immunohistochemistry In study 1, 6 weeks post surgery; mice were weighed, overdosed with pentobarbital (200 mg/kg) and perfused with 25 ml of 0.9% normal saline solution then 50 ml of freshly prepared 4% paraformaldehyde. Brains were collected from

PAGE 73

65 the animals immediately following per fusion and immersion fixed in 4% paraformaldehyde for 24 hrs. Mouse brai ns were cryoprotected in successive incubations in 10%, 20%, 30% solutions of sucrose; 24hrs in each solution. Subsequently, brains were frozen on a co ld stage and sectioned in the horizontal plane (25 m thickness) on a sliding micr otome and stored in Dulbecco’s phosphate buffered saline (DPBS) with 0. 2% sodium azide solution at 4 C. Eight sections 100 m apart spanning the site of injection were chosen and free-floating immunochemical and histol ogical analysis was performed to determine transgene expression using antiHA biotinylated rabbit polyclonal antibody at a concentration of 1:1000 (Roche, Indianapolis, IN), total A using a rabbit primary anti-A serum at a conc entration 1:10,000 and a secondary antirabbit antibody (Serotec, Raleigh, NC). A nother series of sections were mounted on slides and stained with Congo red to assess compact congophilic positive plaque load. Immunohistochemical proc edural methods are analogous to those described by Gordon et al. 2002, for each ma rker. Six to eight sections from each animal were placed in multisample staining tray and endogenous peroxidase was blocked (10% methanol, 3% H202, in PBS). Tissue samples were permeabilized (with 0.2% lysine, 1% Triton X-100 in PBS solution), and incubated overnight in appropr iate primary antibody. Sections were washed in PBS then incubated in corresponding biotinylated secondary antibody (Vector Laboratories, Burlingame, CA). T he tissue was again washed after a 2 h incubation period and inc ubated with Vectastin Elite ABC kit (Vector Laboratories, Burlingame, CA ) for enzyme conjugation. Finally, sections were

PAGE 74

66 stained using 0.05% diam inobenzidine and 0.3% H202. For anti-HA 0.5% nickelous ammonium sulf ate was added for color enhanc ement. Tissue sections were then mounted onto slides, dehydr ated, and coverslipped. Each immunochemical assay omitted some sect ions from primary antibody incubation period to evaluate nonspecific r eaction of the secondary antibody. Congo red histology was performed usi ng sections that were premounted on slides then air dried for a minimum of 24 hours. The sections were rehydrated for 30 seconds before beginning the staini ng protocol. For Congo red, hydrated sections were incubated in an freshly prepared alkaline alcoholic saturated sodium chloride solution (2.5mM NaOH in 80% alcohol) for 20 min, then incubated in 0.2% Congo red in alkaline alcoholic saturated sodium chloride solution for 30 minutes. Slides with sections were rinsed through three changes of 100% ethanol, and cleared through three changes of xylene and finally coverslipped with DPX. Histological sect ions from control animals treated with rAAV expressing GFP were mounted on slides and dehydrated through a series of increasing concentrations of ethanol. The mounted sections were then cleared in three changes of histoc lear and coverslipped with DPX. Stained sections were imaged using an Evolution MP digital camera mounted on an Olympus BX51 microscope at 100X final magnification (10 X objective). Eight horizontal br ain sections (100m apart; every 4th section) were taken from each animal and four nonoverlapp ing images near the si te of injection from each of these sections were capt ured (32 measurements per mouse). All images were taken from the same locati on in all animals. Quantification of

PAGE 75

67 positive staining product surrounding and in cluding the injection sites in the right frontal cortex and the right hippocampus and the corresponding regions in the left hemisphere were determined using ImagePro Plus (Media Cybernetics, Silver Springs, MD. Quantif ication of percent area of pos itive stain for cortex and hippocampus for both right and left hemis pheres was determined. Data were analyzed using ANOVA statistical analysis followed by Fisher’s LSD test for individual means differences as re commended by the computer software program StatView vers ion 5.0.1 (SAS Institut e, Raleigh, NC). Enzyme Activity Assay NEP activity of the three different NEP constructs was characterized and adapted from a previous fluorometric assay method (Johnson and Ahn, 2000) and the NEP ELISA kit (R&D Systems) fo r a 96 well plate format with slight modifications. HEK 293 cells were tr ansfected with NEP-n, NEP-m, and NEP-s plasmids and control cells were transfected with a GFP plasmid using lipofectamine 2000 per invitrogen protocol. Cells were harvested after 72 hrs. and samples were centrifuged at 1,000 x g 4 C for 45 min. using Beckman J6HC Centrifuge (Beckman Instrum ents, Inc., Palo Alto, CA) to obtain cell media fraction and a cell pellet containing the membrane fraction. The membrane fraction was resuspended in M-PER mamma lian protein extraction reagent buffer (Thermo Scientific, Rockford, IL) to obtai n a cell lysate. Protein concentration was determined from the cell media and cell lysate containing the membrane fractions from each sample using a general BCA assay (Per Pierce Protocol). Aliquots of the cell media and membrane fractions were (100 g) were incubated

PAGE 76

68 in NEP ELISA plate containing NEP c apture antibody. The plate was washed with PBS buffer then incubated with 20 M (final) of the fluorogenic peptide (MOCAArg-Pro-Pro-Gly-Phe-Ser-Ala-PheLys(Dnp)-OH), (R&D Systems) in Tris-HCL buffer (sodium phosphate pH 7.4 containing 0.1M NaCl) MOCAArgPro-Pro-Gly-Phe-Ser-Ala-Phe-Lys(Dnp )-OH is efficiently quenched by resonance energy transfer to the di nitrophenyl group and the continuous fluorescent intensity is increased upon internal cleavage of the peptide (NEP cleavage between the Ala-Phe bond). The increased fluorescence produced from cleavage of the substrate was measur ed using a Molecular Devices fMax spectrofluorometer plate reader (MDS Analytical Technologies, Sunnyvale, CA) with a 60 min time point to normaliz e independent experim ents. A standard curve of (7-methoxycoumarin-4-yl) acet yl (MOCA) was analyzed along with each assay. Values were calculated and expres sed as RFU/ min/ ml protein. Data were analyzed using ANOVA statistical analysis test. RESULTS Three rAAV vectors expressing differ ent versions of human NEP were developed. cDNAs were packaged in AAV ve ctors to generate ei ther the native, membrane bound, form of the enzyme (denot ed NEP-n), a secreted form of the enzyme (denoted NEP-s) or an enzymatically-deficient mutant enzyme (denoted NEP-m). The secreted form of the enzym e contained a signal peptide in place of the transmembrane domain of the nativ e enzyme to direct the enzyme's secretion into the extracellular compar tment (Fig 1). The gene sequences are

PAGE 77

69 under the control of the hybrid chicken -actin cytomegalovirus (CBA) promoter sequence and were tagged with haemagglutinin (HA) peptide sequence for detection within the brain and discrimi nation from endogenous NEP. The control group received the rAAV expressing NEP-m cDNA, which contains a single point mutation in the active site, E585V, render ing the expressed enzyme inactive. Prior to virus production the NEP construc ts were tested in HEK 293 cells to evaluate the expression cassettes and t he effectiveness of the signaling peptide in directing secretion of the NEPs gene product. Cell lysate and conditioned media from transfected and untransfected cells were examined by Western blot analysis to determine NEP expression. Untransfected cell lysate and media were negative for NEP protein expression. NEP-n and NEP-m protein expression was only detected in the cell lysate fracti on and not the media. In contrast, NEP-s protein was detected in bot h the cell lysate and the media indicating that the signaling peptide did effectively direct secretion of the NEP protein. rAAV vectors for NEP-m, NEP-n and NEPs were injected unilaterally into the right hippocampus and right anterior cortic al regions of six month old APP + PS1 mice. Animals received 2 l of virus at a flow rate of 2.5 l/min and at a concentration of 1.5 x 1011 vg/mL. The transgene expression of NEP was evaluated six weeks after injection. Gene expression in the mice which received either the NEP-n or the N EP-s was compared to control animals which received NEP-m or were untreated. Immunostaining of the tissue with an anti-haemagglutinin (HA) antibody revealed NEP expression in all animals for NEP-n, NEP-m and NEP-s. The gene

PAGE 78

70 expression patterns filled virtually all of the hippocampus. The staining for NEP-n and NEP-m was greater in intensity than the staining for NEP-s. NEP-n was detected in CA4 neurons in the hi lus and CA2 and CA3 neurons of the hippocampus pyramidal cell layer (Fig. 2A). Some expression of NEP-n was even noticeable in the entorhi nal cortex. Interestingl y, it appears that the cell bodies in the dentate granule cell layer were not stained as intensely as the corresponding neuropil, suggesting some dendr itic localization of the transfected protein. NEP-s was detected in cell bod ies of CA4 neurons in the hilus, some CA3 neurons as well as a few cells in t he molecular layers of hippocampus. NEP-s expression was of noticeably le ss intense and limited in area of distribution compared to that of NEP-n and NEP-m expressi on (Fig 2C). It is uncertain the degree to which this reflects reduced expression, or diffusion of the soluble enzyme in vivo or during tissue processing. NEP-m staining pattern was very similar to the pattern with NEP-n (F ig. 2B) revealing expression throughout the hippocampus in the dentate gyrus as well as in all CA regions. When the cortical regions were anal yzed for expression, there was again a greater amount of NEP-n and NEP-m expr ession than NEP-s (greater intensity of HA staining; Fig 3). NEP-n as well as NEP-m positive expression was concentrated in the anterior cortex (Fig 3A and B), but was observed also in the striatum and corpus callosum and to small ex tent in the contralateral hemisphere. Again, NEP-m and NEP-n expression profiles were very similar, with expression largely confined to the neuropil and little staining of the neuronal somata. NEP-s cortical HA staining was less intense t han that of NEP-n HA staining and was

PAGE 79

71 confined to largely to somatic areas of neurons, conceivably in vesicular compartments in preparation fo r secretion (Fig. 3C). Overall, intensity of HA immunohistochemical staining in the cort ical region of the brain appeared less intense than in the hippocampus possibl y indicating a lower level of the gene expression of NEP in this region. Hi ppocampal and anterior cortical regions of uninjected animals showed no noticeable HA staining (Fig 2D and 3D). The rAAV NEP-s, NEP-m and NEP-n vect or constructs were tested for enzyme activity in vitro using HEK 293 cells. Cells were transfected using NEPn, NEP-m and NEP-s plasmids and contro l cells were transfected with the GFP plasmid. All cells were harvested 72 hr s after transfection using lipofectamine 2000. The NEP specific activity was determined as RFU/min/ml protein. HEK 293 cells transfected with NEP-n had significant ly higher NEP specific activity in the cell lysate containing the membrane fraction than the cells transfected with the NEP-m, NEP-s and the control GFP trans fected cells (Fig 4). NEP specific activity of the lysate from NEP-s transfe cted cells was also significantly greater than NEP-m and GFP transfected cell lysates. In addition, the NEP-s specific activity in the cell media was significant ly greater than the NEP-n, NEP-m, and GFP transfected cells (Fig 4). Our next goal was to evaluate the e ffects of a single intracranial administration of a rAAV vect or containing either the secreted or native form of the NEP gene in APP + PS1 transgenic mice to determine the effect of over expression of the transgene on amyloid dep osition. In study 1 rAAV vectors were injected unilaterally into the right anterior cortex and hippocampus of six

PAGE 80

72 month old mice. The control group wa s treated with rAAV containing a NEP-m gene. Total A load was ascertained six weeks afte r intracranial injections by immunohistochemical methods. The regional A distribution and density in APP + PS1 transgenic mice that were contro l animals were similar to aged matched untreated animals reported by us previously (Gordon et al., 2002). Immunohistochemistry for A revealed both darkly stained compact plaques and more lightly stained diffuse plaque deposits in the APP+PS1 animal tissue (Fig 5A and 5D). Plaque deposition was distribut ed throughout the cortical regions as well as in the hippocampus (although most concentrated in the molecular layers of the dentate gyrus and the CA1 region, surrounding the hippocampal fissure). Animals injected with the c ontrol rAAV-NEP-m showed A immunohistochemical staining patterns throughout the cortex and hippocampus comparable to those of untreated APP transgenic mice of the sa me age. A notable decrease in the amount of hippocampal A staining was observed in an imals injected with either the rAAV expressing NEP-n or NEP-s six weeks after the time of injection when compared to animals injected with the cont rol rAVV-NEP-m vector (Fig. 5A-5F). The reductions in A deposition were not only limit ed to the areas surrounding the cortical and hippocampal injection site s. Significant reductions were also notable in corresponding areas contralateral to the site of injection (although only a few cells showed very faint positive HA staining in the contralateral anterior cortex and hippocaompus). ANOVA analysis of total A in the ipsilateral hemisphere revealed significant decreases of, 78% and 65% in the anterior cortex and hippocampus respectively, in mice receiving NEP-n injections.

PAGE 81

73 Decreases in total A were also observed in mice receiving the NEP-s rAAV with significant declines of 60% in the ant erior cortex and 56% in the hippocampus (Fig. 6A). Quantificati on of total A in the contralateral hemisphere showed significant reductions of 56% and 36% in the anterior cortex and hippocampus respectively, of mice receiving NEP-n. Similarly, significant decreases of 53% and 44% in total A of the contralateral anter ior cortex and contralateral hippocampus respectively, were seen in mice receiving NEP-s (Fig. 6B). Congophilic plaque load was ana lyzed following intracranial injections of rAAV vectors. The density of congophilic labeling was substantially less than A immunohistochemistry, staining only fibrillar A deposits as expected (Gordon et al., 2002). Animals injected with the control rAAV-NEP-m showed positive congophilic staining patterns thr oughout the cortex and hippocampus comparable to those of unt reated APP transgenic mice of the same age (Fig. 7A and 7D). Figures 6B and 6E show that the presence of congophilic staining for the mice receiving the NEP-n vector was visibly less, especially in the hippocampus, compared to both the cont rol animals and the animals that received the NEP-s vector (Fig 7C and 7F). ANOVA analysis revealed that animals receiving the NEP-n rAAV s howed significant reductions in the hippocampal region (56%) and in the anterior cortex (56%; Fig 8A) ipsilateral to the site of injection. rAAV expressing NEP-s also had significant decreases in the compact plaque load in both the hi ppocampus (51%) and cortical regions (57%;Fig 8A). When compact plaque load was calculated in the contralateral hemisphere and compared to the control group, significant decreases were

PAGE 82

74 observed only in the hippocampus (49%) of animals receiving the NEP-s vector treatment (Fig 8B). Animals receivi ng the NEP-n showed a decreased trend but no statistical difference in congophilic stai ning from the NEP-m. No significant decreases of Congo red staining were noted in the contralateral anterior cortex of animals receiving either treatment (Fig 8B). Study 2 In study 1 we demonstrated that eit her NEP-s or NEP-n could reduce amyloid accumulation when tested in a pr evention type of study. Study 2 was designed to test NEP-s, which appeared s lightly superior to NEP-n at sites distant from the injection, in mice with large amounts of preexisting amyloid deposits, what some would term a therapeut ic type of study design. Aged APP + PS1 transgenic mice (15 months of age) were injected into hippocampus and anterior cortex bilaterally. Tissues we re collected 5 mo later when mice were 20 mo old. Total A load was reduced following intrac ranial administration of rAAVNEP-s in aged mice. Panels A and C of fig. 10 show positive immunohistochemical staining of total A in the hippocampus and cortex respectively of aged APP+PS1 mice tr eated with rAAV-NEP-s. Panels B and D of Fig. 9 showed positive A staining in the hippocampu s and cortical regions of untreated aged mice. Panel E shows quantification of A immunostaining of NEP-s treated and untreated 20 mo old APP+PS1 mice. Total A was reduced significantly in the left anterior cortex as well as in the left and right hippocampus following treatment with the NEP-s vector compared to control untreated animals.

PAGE 83

75 Total congophilic staining is reduced fo llowing intracranial administration of rAAV-NEP-s in aged 20 month old mice. Panels A and C of Fig. 11 show total positive congophilic staining in the hippocampus and cortex respectively of APP+PS1 mice treated with rAAV-NEP-s. Panels B and D of Fig. 11 show positive congophilic staining of the hippocampus and cortex respectively of untreated 20 month old mice. Panel E of Fig. 11 shows quantification of congophilic staining in NEP-s and untr eated 20 Mo old APP+PS1 mice. Total congophilic staining was signifi cantly reduced in the left and right anterior cortex as well as the left and right hippocampal regions in animals receiving treatment with the NEP-s vector compared to untreat ed control animals. DISCUSSION Several recent findings have clearly implicated the important role of endogenous proteases such as nepril ysin in the catabolism of A peptides in the brain. Also known as enkephalinase and CD10, it is capable of cleaving enkephalins and terminating peptidergic neurotrans mission. In addition to NEP, other endogenous A degrading proteases maintain a conserved catalytic domain which includes a zinc binding moti f, HEXXH (Turner et al., 2000). Other proteases in addition to the zinc metalloprot eases that appear to play a role in A metabolism include matrix metalloprotei nase-9 (Yan et al., 2006) and cathepsin B (Mueller-Steiner et al., 2006) and plasmin (Turner et al., 2001). Downregulation of these degr ading enzymes within the brain during aging could

PAGE 84

76 potentially contribute to A accumulation eventually leading to development of AD pathology (Caccamo et al., 2005), (Yasojima et al., 2001). Deficiency in the expression of N EP and other metallopeptidases in the aged brain and in animal knockout models co rrelates with plaque accumulation in a region specific manner (Iwata et al ., 2000), (Eckman et al., 2001), (Fukami et al., 2002), (Yasojima et al., 2001b), (Fa rris et al., 2004), (Sai to et al., 2003), (Caccamo et al., 2005), (Miner s et al., 2006). Enzyme kinetic studies using specific NEP inhibitors have shown that neprilysin can efficiently degrade numerous peptides on the N-terminus of hydrophobic amino acid residues, which is essential for the efficient catabolism of the A peptide (Marie-Claire et al., 1997), (Shimada et al., 1996), (Hoang et al., 1997), (Turner et al., 2001), (Leissring et al., 2003), (Hersh 2003). Furthermore, in vivo studies show that an increase in the enzyme activity of amyloid degrading enzymes specifically neprilysin, endothelin conver ting enzyme, and insulin degrading enzyme, have a significant effect on both in tracellular and extracellular A levels in the brain (Hama et al., 2001), (Zou et al ., 2006), (Eckman et al., 2006), (Guan et al., 2008). Most recently, Farris et al., demons trated that when the expression of endogenous NEP is partially or co mpletely inhibited both A deposition and cerebral amyloid angiopathy are significant ly exacerbated (Farris et al., 2007b). Subsequently, peptidases hav e become popular targets for Alzheimer’s disease therapies. Previously, we examined the effects of the overexpres sion of ECE-1 gene on amyloid load in the APP + PS1 mouse model (Carty et al., 2006). In the

PAGE 85

77 current study, we examine the effect iveness of rAAV to deliver NEP into hippocampus and anterior cortex of the mouse brain to determine its effects on amyloid burden. In particula r, we are interested in developing a delivery method that increases the disseminat ion of therapeutic levels of NEP protein. Previous studies involving gene therapy approaches for the treatment of Alzheimer’s pathology have used viral vectors to reduce A deposition by overexpressing A degrading enzymes, including NEP, ECE a nd IDE and more recently Cathepsin B, a cysteine protease also implicated in the lysosomal degratory pathway of A peptides (Marr et al., 2003), (I wata et al., 2004), (Mueller-S teiner et al., 2006). A comparison of gene transfer studies indica te that overexpression of neprilysin appears to have the most pronounced effect on plaque load in the brain (Farris et al., 2007a), (Eckman et al., 2003), (Hersh et al 2008), (Spencer et al., 2008). Marr et al (2003) used lentiviral vector and were successful in reducing amyloid levels in APP Tg mice but only moderat ely peptides (Marr et al., 2003). In addition, others have shown that the tr ansgenic overexpression of NEP and IDE crossed with transgenic APP mice significant ly reduces amyloid accumulation in the brain (Leissring et al., 2003), (Poirier et al., 2006), (Meilandt et al 2009). Iwata et al.used AAV, as in this st udy, to express NEP-n and were also successful in reducing A levels in the hippocampus of Tg2576 mice. In this paper, we further explore t he use of different rAAV expressing different forms of the NEP gene. Particularly, we exam ine a secreted form of the NEP in addition to using the CED deliv ery method to increase distribution and ultimately total area of gene expression. We also determine the effect of both the

PAGE 86

78 native and secreted versions of NEP ex pression in the anterior cortex and hippocampus comparing the relative distribution and differences in NEP gene expression in specific cell types in t he brain. Unlike previous gene transfer studies with these proteases we use the double transgenic APP + PSI mouse model which have a very aggressive am yloid deposition profile and exhibit memory deficits at a younger age t han single transgenic amyloid depositing models. Recombinant AAV has become widely used for the transduction of neuronal cells. rAAV is nonpathogenic, has low immunogenicity, lacks all viral genes and is capable of long term expre ssion in neurons. This profile makes rAAV a good candidate for the use as a gene therapy vector for neurological disorders (Mandel et al 2006), (Mandel and Burger 2004). One limitation with a gene therapy approach for a disease such as Alzheimer’s is that the therapeutic protein must be delivered to the whole brain. Unfort unately, the intracranial injections into the brain parenchyma us ing simple diffusion does not allow for efficient uptake of the transgene or signifi cant dispersion of the AAV vectors to significantly large areas of the affected r egions within the brain. Thus if only a small area of the brain is transduced by the viral vector, multiple injections would be required to cover the entire brain. Theref ore, in an attempt to overcome this problem we engineered the NEP gene to contain a signal peptide sequence in the hopes of creating a nucleus from which the expressed protein could diffuse to greater regions of the brain; thus eliminating the requirement for a large number of injections. Additionally we implem ented a novel infusion technique using the CED method allowing us to obtain a lar ger area of distribution of the rAAV

PAGE 87

79 vectors upon delivery with a single injecti on in the brain parachyma both in the hippocampus and cortical regions of the br ain. CED is a method of delivering clinically relevant volumes of therapeutic agents to signi ficantly larger areas of the brain in comparison to simple di ffusion methods. The CED technique is designed to utilize the phenomenon of bulk flow and positive pressure to distribute macromolecules to a large ar ea within solid tissue. Macromolecules can therefore be administered to significantly larger areas of specific brain regions by utilizing the advantage of fl uid convection within and throughout the interstitial space in the brain (Leiberman et al. 1995, Sanftner et al 2005). The greatest area of distribution is theref ore achieved by using an increased and optimal pressure gradient, flow rate, and vo lume of liquid material (Sanftner et al., 2005). One of the mechanist ic limitations of the CE D method as well as the simple injection method is t he reflux of the injected mate rial up the injection tract. In 2002 Krauze et. al. developed a reflux free step cannula design which effectively minimizes reflux by placing silicone coated tubing within the stainless steel blunt end cannula creating a step that prevents the backflow of fluid. The optimization of more efficient cannul a designs coupled with the encouraging results from studies showing enh anced gene transfer and distribution emphasizes the therapeutic potential of the CED met hod in helping overcome some of the mechanical disadvantages of gene delivery in regards to gene therapy (Krauze et al., 2005b). Implementing these techniques, rAAV serotype-5 NEP vectors were injected unilaterally into the mous e hippocampus and anterior cortex.

PAGE 88

80 Examination of the expressi on profile, using anti-HA i mmunohistochemisty, of the expressed NEP reveals that the rAAV constructs of both NEP-m and NEP-n can transduce several different neuronal and glia l cell types within the mouse brain. Interestingly, it appears that the cell bodies in the dent ate molecular layer were not stained as intensely as the correspondi ng axonal projections of these cells which were very darkly stained. This may suggest that the expressed NEP is processed in the endoplasmic reticulum and golgi apparatus of the cell body, then transported along the ax on via vesicles to the synaptic terminal. These results are consistent wit h data published in 2006 by Hu ang et al. describing the normal neuronal metabolism of NEP (Huang et al, 2006) and by Iwata expressing NEP-n with rAAV (Iwata et al 2004). The expression profiles ar e also consistent with the previously published serological specificity of the AAV vectors (Alisky et al., 2000),(Burger et al., 2004),(Choi et al., 2005). The AAV 5 serotype has previously been shown to effectively transduce non-dividing neuronal cells as well as glial cells in the mouse brain. The NEP-s expression profile revealed less intensity of HA staining when compared to the NEP-m and NEP-n staining patterns. Cells within the hilus of the molecular layer were stained as were as cells in the CA3 region of the hippocampus. We believe that more diffuse staining of the NEP-s construct does not necessarily mean that this vector had lower transduction efficiency. It is likely due to the diffusion of the NEP proteas e throughout the extracellular milieu, making it more difficult to detect us ing anti-HA immunohistochemisty compared to the membrane bound NEP-n protease. However, another explanation for the

PAGE 89

81 decreased staining of the NEP-s protease may be that once the NEP-s is secreted the NEP protein or at least t he HA-tag is more prone to degradation by other extracellular proteases. We plan to conduct further investigation of the expression and activity profile using in vitro studies to delineate the metabolism of both the NEP-n and NEPs proteases. Additionally, the NEP-s staining was more intense in the cortical region s compared to the hippocampus which may indicate that the signal peptide is proc essed more effectively in the cortex compared to the hippocampus. Despite t he differences in staining intensity of the different NEP proteins our in vitr o data confirmed that both NEP-n and NEP-s proteases shown activity, while the NEP -m version did not. Additionally, only the NEP-s protein showed enzyme activity once secreted into the cell media. We also plan to re-examine the levels of soluble protease activity in the cortex compared to the hippocampus of injected animal to see if this is t he case in vivo. Our results demonstrate that the up -regulation of NEP through rAAV vectors can provide a viable method to decrease the total amyloid deposition in the brain of amyloid de positing Tg mice. NEP-n and NEP-s were able to significantly reduce total A deposition in the anterior cortex and hippocampus at the site of injection. Similarly, both NEP-n and NEP-s significantly reduced the level of congophilic deposits. Interestingly, when the area of positive staining for A was analyzed in the hippocampus and cort ical regions of the contralateral hemisphere, both the NEP-s and NEP-n rAAV were able to decrease total A (in anterior cortex) load compared to the cont rol group. Only NEP-s treated animals showed significant reductions in total A load in the contralateral hippocampus

PAGE 90

82 compared to control animals. However, examination of the congophilic deposits on the contralateral anterior cortex show ed no reduction. Interestingly, on the contralateral hippocampal side there was a 40-50% reduction of congophilic staining, but only the reduction with NEP-s was statistically significant compared to the control group. The reduction on the contralateral hemisphere observed here is likely due to retrograde transpor t of the rAAV and anterograde transport of the NEP protein. Retrograde transport of rAAV has previously been reported (Burger et al 2004). The lower number of neuronal connections between the ipsilateral and contralateral sides in the anterior cortex compared to those present in the hippocampus could explain why there is a reduction in congophilic staining in the contralateral hippocam pus and not the contralateral anterior cortex. Further, since the contralateral anterior cortex showed reduction of total A load but not congophilic staining it woul d seem to suggest that a critical amount of NEP expression is required in order to reduce the insoluble A plaque formation. As previously mentioned our findings showed that by implementing the use of a secreted form of the NEP protease in addition to using the CED method for intracranial administration, we were able to successfully improve the area of gene expression to further reduce amyloid load in the brain. An alternative method which would potentially overcome th e limitations of a direct intracranial injection is to attempt to express the NEP protease in a peripheral fashion. Previous studies using anti-amyloi d vaccines have clearly demonstrated that effective therapy does not have to cr oss the blood brain barrier to have a

PAGE 91

83 significant effect on amyloid load in the brain. It is well es tablished that both passive and active A immunotherapy can effectively decrease amyloid load from the brain as well as improve c ognitive impairment in transgenic animals (Morgan et al., 2000), (Janus et al., 2000), (Dodart et al., 2002), (Masliah et al., 2005), (Wilcock et al., 2004),(Solomon, 2005),(Maier et al., 2006). The mechanism underlying this phenomenon referr ed to as the ‘peripheral sink’ hypothesis asserts that amyloid is in a state of equilibrium th roughout the body. Therefore, sequestering and removing am yloid in the periphery will change the equilibrium such that amyl oid will move from the brain into the periphery (DeMattos et al., 2002), (Mats uoka et al., 2003), (Lemere et al., 2003), (Deane et al., 2005). Most recently, ex vivo gene tr ansfer studies using cell mediated over expression of amyloid degrading peptidas es have shown initial success and may provide a useful alternative to passive immunotherapy. Hemming et al in 2007 showed that over expression of secr eted neprilysin in primary fibroblasts reintroduced peripherally into transgenic APP mice showed a robust reduction in plaque load (Hemming et al., 2007). A dditionally, other methods implementing cell mediated gene expression using both leukocytes and erthrocytes have been used enhance A degradation in the periphery and in the CNS (Guan et al., 2008), Lui et al 2008). We also wanted to examine the effect of increasing NEP expression in aged animals where there is alr eady significant deposition of A Evidence has shown that rAAV viral vectors can successf ully transfect neurons in the CNS and result in sustained and stable long term gene expression in the brain (McCown et

PAGE 92

84 al., 1996), (McCown, 2005), (Royo et al., 2008), (Spencer et al., 2008). Our results demonstrate that the rAAV NEP-s vector was able to significantly reduce both total A load and fibrillar A in aged animals. These data are consistant with evidence from Spencer et al (2008). As expected, the NEP-s showed strong expression in the brain 5 m onths following treat ment in older animals 20 months of age (treatment was at 15 mo. of age) demonstrati ng persistence and stability of the transgene expressi on. Upon observation of congophilic positive plaque load in 20 mo. old animals compared to plaque load in 15 mo.old animals it appears that treatment with NEP in the old animals with a large amount of preexisting plaques simply halts the progr ession of plaque build up rather than actively removing compact plaques est ablished well before treatment onset. Further investigation must be done to ve rify the aforementioned observation. Gordon et al. in 2001 showed that c ongophilic positive staining, primarily composed of A 40, increases until about 12 m onths of age in the double transgenic APP+PS1 mouse and remains rela tively stable up until 18 months of age (Gordon et al, 2001). Although we have yet to perform a detailed exami nation, initial observations showed no noticeable toxic effects in mice receiving the NEP vectors. No neuron loss or gross morphometric changes were observed in fixed brain tissue. No significant changes in total body weight were noted, indicating that up regulation of NEP did not appear to have adverse effects or cause general toxicity in the mouse model during our study period. NEP has been shown to have other endogenous peptide subs trates, including atrial natriuretic

PAGE 93

85 peptide, substance P, endot helin and enkephalin. It is uncertain whether interactions between NEP and these other potential substrates in the brain may cause potential problems although one recent study suggests that high levels of enkephalins can contribute to memory and c ognitive deficits in amyloid producing mice, further implicating t he importance of NEP activity in AD (Meilandt et al., 2008). Monitoring changes in these NEP p eptide substrates between control and treatment groups may help id entify any potentially harmful side effects resulting from increases in NEP activity. Previous studies have revealed that in the AD brain as well as in the animal model of AD, NEP are down regulated specifically in areas that are prone to plaque formati on (Yasojima et al., 2001a), (Fukami et al 2002). Therefore, up-r egulation of NEP to restor e normal levels of this endogenous protease should have minimal advers e effects. In fu ture studies, we could regulate the levels of NEP expres sion by using an i nducible promoter system which would enable the levels of NEP to be managed as required per individual and potentially r educe unwanted side effects of continuous high levels of NEP expression. Future studies whic h implement a combination of peripheral and central gene transfer studies can also have additive effect on total amyloid load in the brain. Another therapeutic gene therapy technique could implement a cell mediated gene transfer technique in an in vivo fashion. Rivest et al in 2008 demonstrated that injection of lentivir us expressing the TLR2 gene resulted in successful expression of the gene in bone ma rrow cells. Additionally, these cells successfully migrated to the CNS of am yloid depositing transgenic mice ensuing in a decrease in amyloid pathology (Ric hard et al., 2008). These findings further

PAGE 94

86 demonstrate that viral vectors and cell mediated therapy can be combined to provide a novel therapeutic technique in the treatment of neurological disease. We have yet to determine if this reduction in the A levels with our NEP-n or NEP-s rAAV vectors will lead to signific ant improvement in behavioral tests such as the Morris water maze in both the young and old populations of transgenic mice. However, these data are consistent with repo rts that NEP can degrade A in vitro, and that partial k nockdown of the NEP gene leads to more rapid accumulation of A (Farris et al., 2007), (Eckman et al., 2003). The present work adds to the evidence that NEP plays an important role in A deposition by demonstrating t hat local overexpression NEP enzyme activity can dramatically reduce the deposition of am yloid in the brains of APP transgenic mice. Thus, regulation of NEP through methods such as gene therapy may be used as a potential therapeutic target for the treatment of Alzheimer's disease warranting further study into the use of gene therapy techniques in a combinatory fashion to increase effectiveness while reducing adverse events. Acknowledgments. Supported by The J ohnnie Byrd Center for Alzheimer's Research, NIH grants AG -25509, AG 15490, AG 18478, AG 04418, AG 25711

PAGE 95

87 FIGURE 1: DIAGRAMMATIC REPRESENTATION OF rAAV CONSTRUCTS EXPRESSING THE NEP GENE UNDER THE CONTROL OF THE CHICKEN ACTIN (CBA) PROMOTER Note the C-terminal fusion HA tag in both sequences and the secretion si gnaling peptide in the NEPH A-s construct. Panel A shows a gene map of the recombi nant AAV NEP-n and NEP-s constructs under control of the hybrid CBA chicken -actin promoter. A hemagluttinin tag has been appended to the NEP encoding s equence to allow for easy detection. HA tag HA tag

PAGE 96

88 FIGURE 2: EXAMINATION OF NEP EXPRESSION IN TRANSDUCED CELLS OF THE HIPPOCAMPUS CONTRALATER AL AND IPSILATERAL TO VECTOR INJECTION. Strong NEP-n and NEP-m expre ssion is detected using anti-HA immunostaining in the right hippocampus fo llowing intracranial administration of NEP-n and NEP-m vector (panels A and B, respectively). Panel C shows less intense diffuse NEP-s expression in granule cells of the dentate gyrus in the right hippocampus following intracranial adminis tration of NEP-s. Panels D shows no positive staining in the uninject ed hippocampus of an untreated age matched animal. Magnification = 40X for all panel s. dg = dentate gyrus; CA1 = Cajal Area 1; CA3 = Cajal area 3. NEP-n NEP-m NEP-s CA1 untreatedCA3 dg

PAGE 97

89 FIGURE 3: EXAMINATION OF EXPRESSION LEVELS OF NEP AFTER INTRACRANIAL ADMINISTRATION OF rAAV VECTORS INTO THE RIGHT ANTERIOR CORTEX OF MICE Panels E and F show strong NEP-n and NEPm expression as detected by anti-HA i mmunostaining in the anterior cortex and striatum ipsilateral to the injection site Panel G shows slightly less intense diffuse NEP-s expression in the anterior cortex and striatum ipsilateral to the injection site. Panels H shows no positiv e staining in the uninjected cortex of untreated aged matched animals. Str = stri atum; CC = corpus callosum; CX= cortex. Magnification = 40X. NEP-n NEP-m NEP-s untreated CX Str cc

PAGE 98

90 FIGURE 4: QUANTIFICATION OF NEP SPECIFIC ACTIVITY IN HEK293 TRANSFECTED CELLS Neprilysin specific activity is significantly higher in the HEK293 cells transfected with NEP-n in the cell lysate (membrane fraction) compared to NEP-m, NEP-s, and GF P transfected cell lysates. NEP-s transfected cell lysates (cell media fr action) showed signif icantly higher NEP activity compared to NEP-n, NEP-m and GF P expressing cells. The asterisk (*) indicates significance with a p-value < 0.05. The number (#) indicates significance with a p-value < 0.001. Note differences in Y axis scaling for each panel.

PAGE 99

91 FIGURE 5: TOTAL AMYLOID LOAD IS REDUCED FOLLOWING INTRACRANIAL ADMINISTRATION OF NEP-s OR NEP-n rAAV VECTORS A immunostaining is observed in mice throughout both hippocampus (panels A, B and C) and anterior cortex (panel D, E and F). A staining in the ipsilateral hippocampus of animals receiving intracrani al injection NEP-n (panel B) or NEP –s (panel C) is reduced compared to control vector NEP-m (panel A). A staining in the right anterior cortex of mice receiv ing intracranial inje ction of NEP-s (panel D) or NEP-n (panel F) is reduced compar ed to control vector NEP-m (panel B). Scale bar = 120 m. dg CA1 CA3 CX

PAGE 100

92 FIGURE 6: QUANTIFICATION TOTAL AMYLOID LOAD IN THE HIPPOCAMPUS AND CORTICAL RE GIONS FOLLOWING INTRACRANIAL ADMINISTRATION OF rAAV Panel A shows percent area of positive staining for total Abeta load of the right cortex and hippocampus ip silateral to the injection site. Panel B shows quantification left co rtical and hippocampal percent area of positive staining for total abeta load of the contralateral hemisphere to the injection site. The star (*) indicates significance with a p-value < 0.05; (**) indicates p-value <.001. A beta L oad I sp il ate r a l A B Abeta Load Contralateral

PAGE 101

93 FIGURE 7: CONGOPHILIC COMPACT PL AQUE LOAD IS REDUCED FOLLOWING INTRACRANIAL ADMINIS TRATION OF NEP rAAV VECTORS Total congophilic staining is observ ed in mice throughout both hippocampus (panels A, B and C) and anterior cortex (panel D, E and F). Positive congophilic staining in the ipsilateral hippocampus of animals receiving intracranial injection NEP-n (panel B) or NEP-s (panel C) is reduced compared to control vector NEPm (panel A). Congophilic staining in the ri ght anterior cortex of mice receiving intracranial injection of NEP-s (panel D) or NEP-n (panel F) is reduced compared to control vector NEP-m (panel B). Scale bar = 120 m. dg CA1 CA3 CX

PAGE 102

94 FIGURE 8: QUANTIFICATION OF TOTAL CONGOPHILIC LOAD IN THE HIPPOCAMPUS AND CORTEX FOLLOWING INTRACRANIAL ADMINISTRATION OF rAAV Panel A shows quantification of percent area of positive staining for total congophilic staining of the cort ex and hippocampus ipsilateral to the injection site in the hippocampus and cortex. Panel B shows quantification of percent area of positive stain of tota l congophilic staining in the contralateral hemisphere to the injection site. The star (*) i ndicates significance with a p-value < 0.05 Total Con g oContralateral B A TotalCong oispilateral

PAGE 103

95 FIGURE 9: DISTRIBUTION OF NEP EXPRESS ION 20 WEEKS AFTER rAAV NEP-S ADMINISTRATION IN AGED APP + PS1 MICE AT 15 MONTHS Brains sections were immunostained for NEP with an antibody recognizing both rodent and human NEP. Panels A, C, and E show NEP expression from mice treated with rAAV-NEP-s. Panels B, D, and E show no NEP expression from control animals treated with rAAV-GFP. CX CA1 CA3 dg thal

PAGE 104

96 FIGURE 10: TOTAL ABETA LOAD IS REDUCED FOLLOWING INTRACRANIAL ADMINISTRATION OF rAAV NEP-S IN AGED MICE Panels A and C show positive immunohistochemic al staining of total abeta in the hippocampus and cortex respectively of APP+PS1 mice treated with rAAV-NEPs. Panels B and D show positive abeta st aining of the hippocampus and cortex of untreated mice. Panel E shows quantif ication of A immunostaining in NEP-s and untreated 20 Mo old APP+PS1 mice. Mice were injected at 15 mo of age in hippocampus and cortex of both hemispheres RCX = right anterior cortex, LCX= left anterior cortex, RHPC = right hippocam pus, LHPC=left ippocampus. Y axis is A load (percent area occupied by reaction product). ** P < 0.01 NEP-s untreated dg CA3 CA1 CX

PAGE 105

97 FIGURE 11: TOTAL CONGOPHILIC STAINI NG IS REDUCED FOLLOWING INTRACRANIAL ADMINISTRATION OF rAAV NEP-s IN AGED MICE Panels A and C show positive immunohistochemical staining of total congo in the hippocampus and cortex respectively of APP+PS1 mice treated with rAAV-NEPs. Panels B and D show positive abeta st aining of the hippocampus and cortex of untreated mice. Panel E shows quantif ication of A immunostaining in NEP-s and untreated 20 Mo old APP+PS1 mice. Mice were injected at 15 mo of age in hippocampus and cortex of both hemispheres RCX = right anterior cortex, LCX= left anterior cortex, RHPC = right hippoc ampus, LHPC=left hippocampus. Y axis is A load (percent area occupied by reaction product). ** P < 0.01 NEP-s untreated dg CA3 CA1 CX

PAGE 106

98 REFERENCE LIST Alisky JM, Hughes SM, Sauter SL, Jo lly D, Dubensky TW, Jr., Staber PD, Chiorini JA, Davidson BL (2000) Transduc tion of murine ce rebellar neurons with recombinant FIV and AAV5 vect ors. Neuroreport 11:2669-2673. Berns, K. I., Parrish, C. R., 1996. Parvoviridae: The Viruses and Their Replication. In: D. M. Kn ipe, P. M. Howley, (Eds.), Fields Virology. Lippincott Williams and Wilkins, Philade lphia, PA, pp. 2437-2477. Burger C, Gorbatyuk OS, Velardo MJ, P eden CS, Williams P, Zolotukhin S, Reier PJ, Mandel RJ, Muzyczka N (2004) Reco mbinant AAV viral vectors pseudotyped with viral capsids from serotypes 1, 2, and 5 display differential efficiency and cell tropism after delivery to different regions of the central nervous system. MolTher 10:302-317. Caccamo A, Oddo S, Sugarman MC, Akbar i Y, LaFerla FM (2005) Ageand region-dependent alterations in Abetadegrading enzymes: implications for Abeta-induced disorders. Neurobiol Aging 26:645-654. Carty NC, Wilcock DM, Rosenthal A, Grim m J, Pons J, Ronan V, Gottschall PE, Gordon MN, Morgan D (2006) Intracranial administration of deglycosylated Cterminal-specific anti-Abeta antibody effi ciently clears amyloid plaques without activating microglia in amyloid-deposit ing transgenic mice. JNeuroinflammation 3:11.:11. Choi VW, McCarty DM, Samulski RJ (2005) AAV hybrid serotypes: improved vectors for gene delivery. CurrGene Ther 5:299-310. Deane R, Sagare A, Hamm K, Parisi M, LaRue B, Guo H, Wu Z, Holtzman DM, Zlokovic BV (2005) IgG-assisted agedependent clearance of Alzheimer's amyloid beta peptide by the blood-brain ba rrier neonatal Fc receptor. J Neurosci 25:11495-11503. DeMattos RB, Bales KR, Cummins DJ, P aul SM, Holtzman DM (2002) 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. Dodart JC, Bales KR, Gannon KS, Greene SJ DeMattos RB, Mathis C, DeLong CA, Wu S, Wu X, Holtzman DM, Paul SM (2002) Immunization reverses memory deficits without reducing br ain Abeta burden in Alzheimer's disease model. NatNeurosci 5:452-457.

PAGE 107

99 Eckman EA, Reed DK, Eckman CB (2001) Degradation of the Alzheimer's amyloid beta peptide by endothelin-conver ting enzyme. J Biol Chem 276:2454024548. Eckman EA, Watson M, Marlow L, Sambamurti K, Eckman CB (2003) Alzheimer's disease beta-amyloid pepti de is increased in mice deficient in endothelin-converting enzyme. JBiolChem 278:2081-2084. Eckman EA, Adams SK, Troendle FJ, St odola BA, Kahn MA, Fauq AH, Xiao HD, Bernstein KE, Eckman CB (2006) Regulatio n of steady-state beta-amyloid levels in the brain by neprilysin and endothelin -converting enzyme but not angiotensinconverting enzyme. JBiolChem 281:30471-30478. Farris W, Mansourian S, Leissring MA Eckman EA, Bertram L, Eckman CB, Tanzi RE, Selkoe DJ (2004) Partial lo ss-of-function mutations in insulindegrading enzyme that induce diabetes also impair degradation of amyloid betaprotein. Am J Pa thol 164:1425-1434. Farris W, Schutz SG, Cirrito JR, Shankar GM, Sun X, George A, Leissring MA, Walsh DM, Qiu WQ, Holtzman DM, Selkoe DJ (2007a) Loss of neprilysin function promotes amyloid plaque formation and caus es cerebral amyloid angiopathy. Am J Pathol 171:241-251. Fukami S, Watanabe K, Iwata N, Haraoka J, Lu B, Gerard NP, Gerard C, Fraser P, Westaway D, St G eorge-Hyslop P, Saido TC (2002) Abeta-degrading endopeptidase, neprilysin, in mouse brain: synaptic and axonal localization inversely correlating with Abeta pathology. Neurosci Res 43:39-56. Guan H, Liu Y, Daily A, Police S, Ki m MH, Oddo S, Laferla FM, Pauly JR, Murphy MP, Hersh LB (2008) Peripherally expressed neprilysin reduces brain amyloid burden: A novel appr oach for treating Alzheimer's disease. J Neurosci Res. Hama E, Shirotani K, Masumoto H, Se kine-Aizawa Y, Aizawa H, Saido TC (2001) Clearance of extracellular and cell-associated amyloid beta peptide through viral expression of neprilysin in primary neurons. JBiochem(Tokyo) 130:721-726. Hemming ML, Selkoe DJ, Farris W (2007) Effects of prolonged angiotensinconverting enzyme inhibitor treatment on amyloid beta-protein metabolism in mouse models of Alzheimer disease. Neurobiol Dis 26:273-281. Hersh LB (2003) Peptidases, proteases and amyloid beta-peptide catabolism. Curr Pharm Des 9:449-454.

PAGE 108

100 Hoang MV, Sansom CE, Turner AJ ( 1997) Mutagenesis of Glu403 to Cys in rabbit neutral endopeptidase-24. 11 (neprilysin) creates a disulphide-linked homodimer: analogy with endothelin-conv erting enzyme. BiochemJ 327:925-929. Iwata N, Mizukami H, Shirotani K, Taka ki Y, Muramatsu S, Lu B, Gerard NP, Gerard C, Ozawa K, Saido TC (2004) Presynaptic localization of neprilysin contributes to efficient clearance of amyloid-beta peptide in mouse brain. JNeurosci 24:991-998. Iwata N, Tsubuki S, Takaki Y, Watanabe K, Sekiguchi M, Hosoki E, KawashimaMorishima M, Lee HJ, Hama E, Sekine-Aiza wa Y, Saido TC ( 2000) Identification of the major Abeta1-42-degrading catabo lic pathway in br ain parenchyma: suppression leads to biochemical and pathological deposit ion. NatMed 6:143150. Janus C, Pearson J, McLaurin J, Mathew s PM, Jiang Y, Schmidt SD, Chishti MA, Horne P, Heslin D, French J, Mount HT Nixon RA, Mercken M, Bergeron C, Fraser PE, St GeorgeHyslop P, Westaway D (2000) A beta peptide immunization reduces behavioural impai rment and plaques in a model of Alzheimer's disease. Nature 408:979-982. Krauze MT, Saito R, Noble C, Tamas M, Bringas J, Park JW, Berger MS, Bankiewicz K (2005) Reflux-free cannul a for convection-enhanced high-speed delivery of therapeutic agent s. JNeurosurg 103:923-929. Leissring MA, Farris W, Chang AY, Walsh DM Wu X, Sun X, Frosch MP, Selkoe DJ (2003) Enhanced proteolysis of bet a-amyloid in APP transgenic mice prevents plaque formation, secondary pat hology, and premature death. Neuron 40:1087-1093. Lemere CA, Spooner ET, LaFrancois J, Malester B, Mori C, Leverone JF, Matsuoka Y, Taylor JW, DeMattos RB, Holtzman DM, Clements JD, Selkoe DJ, Duff KE (2003) Evidence for peripheral cl earance of cerebr al Abeta protein following chronic, active Abeta immuni zation in PSAPP mi ce. Neurobiol Dis 14:10-18. Maier M, Seabrook TJ, Lazo ND, Jiang L, Das P, Janus C, Lemere CA (2006) Short amyloid-beta (Abeta) immunogens reduce cerebral Abeta load and learning deficits in an Alzheimer's di sease mouse model in the absence of an Abeta-specific cellular immune response. JNeurosci 26:4717-4728. Marie-Claire C, Ruffet E, Antonczak S, Beaumont A, O'Donohue M, Roques BP, Fournie-Zaluski MC (1997) Evidence by site-directed mutagenesis that arginine 203 of thermolysin and arginine 717 of neprilysin (neutral endop eptidase) play

PAGE 109

101 equivalent critical roles in substrate hy drolysis and inhibitor binding. Biochemistry 36:13938-13945. Marr RA, Rockenstein E, Mukherjee A, Kindy MS, Hersh LB, Gage FH, Verma IM, Masliah E (2003) Neprilysin gene tr ansfer reduces hum an amyloid pathology in transgenic mice. JNeurosci 23:1992-1996. Masliah E, Hansen L, Adame A, Crews L, Bard F, Lee C, Seubert P, Games D, Kirby L, Schenk D (2005) Abeta vaccinatio n effects on plaque pathology in the absence of encephalitis in Alzheime r disease. Neurology 64:129-131. Matsuoka Y, Saito M, LaFrancois J, Sait o M, Gaynor K, Ol m V, Wang L, Casey E, Lu Y, Shiratori C, Lemere C, Duff K (2003) Novel therapeut ic 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. McCown TJ (2005) Adeno-associated virus (AAV) vectors in the CNS. Curr Gene Ther 5:333-338. McCown TJ, Xiao X, Li J, Breese GR Samulski RJ (1996) Differential and persistent expression patterns of CNS gene transfer by an adeno-associated virus (AAV) vector. Brain Res 713:99-107. Meilandt WJ, Yu GQ, Chin J, Roberson ED Palop JJ, Wu T, Scearce-Levie K, Mucke L (2008) Enkephalin elevations contribute to neuronal and behavioral impairments in a transgenic mouse model of Alzheimer's disease. J Neurosci 28:5007-5017. Miners JS, Van Helmond Z, Chalmers K, Wilcock G, Love S, Kehoe PG (2006) Decreased expression and activity of neprilysin in Alzheimer disease are associated with cerebral amyloid angio pathy. J Neuropathol Exp Neurol 65:10121021. Morgan D, Diamond DM, Gottsch all PE, Ugen KE, Dickey C, Hardy J, Duff K, Jantzen P, DiCarlo G, Wilcock D, Connor K, Hatcher J, Hope C, Gordon M, Arendash GW (2000) A beta peptide vaccina tion prevents memory loss in an animal model of Alzheimer's disease. Nature 408:982-985. Mueller-Steiner S, Zhou Y, Arai H, R oberson ED, Sun B, Chen J, Wang X, Yu G, Esposito L, Mucke L, Gan L (2006) Antiamyloidogenic and neuroprotective functions of cathepsin B: implications for Alzheimer's disease. Neuron 51:703714.

PAGE 110

102 Poirier R, Wolfer DP, Welzl H, Tracy J, Galsworthy MJ, Nit sch RM, Mohajeri MH (2006) Neuronal neprilysin overexpression is associated with attenuation of Abeta-related spatial memory deficit. NeurobiolDis 24:475-483. Richard KL, Filali M, Prefontaine P, Rivest S (2008) Toll-like receptor 2 acts as a natural innate immune receptor to clear amyloid beta 1-42 and delay the cognitive decline in a mouse model of Alzheimer's disease. J Neurosci 28:57845793. Royo NC, Vandenberghe LH, Ma JY, Hauspurg A, Yu L, Maronski M, Johnston J, Dichter MA, Wilson JM, Watson DJ (2008) Specific AAV serotypes stably transduce primary hippocampal and cortical cultures with high efficiency and low toxicity. Brain Res 1190:15-22. Saito T, Takaki Y, Iwata N, Trojanowski J, Saido TC (2003) Alzheimer's disease, neuropeptides, neuropeptidase, and amyloid-beta peptide metabolism. SciAging KnowledgeEnviron 2003:E1. Sanftner LM, Sommer JM, Suzuki BM, Smit h PH, Vijay S, Vargas JA, Forsayeth JR, Cunningham J, Bankiewicz KS, Kao H, Bernal J, Pierce GF, Johnson KW (2005) AAV2-mediated gene delivery to m onkey putamen: evaluation of an infusion device and delivery param eters. Exp Neurol 194:476-483. Schenk D, Hagen M, Seubert P (2004) Cu rrent progress in beta-amyloid immunotherapy. Curr Opin Immunol 16:599-606. Schenk D, Barbour R, Dunn W, Gordon G, Grajeda H, Guido T, Hu K, Huang J, Johnson-Wood K, Khan K, Kholodenko D, Lee M, Liao Z, Lieberbu rg I, Motter R, Mutter L, Soriano F, Shopp G, Vasquez N, Vandevert C, Walker S, Wogulis M, Yednock T, Games D, Seubert P (1999) Immunization with amyloid-beta attenuates Alzheimer-disease-like pathology in the PDAPP mouse. Nature 400:173-177. Shimada K, Takahashi M, Turner AJ Tanzawa K (1996) Rat endothelinconverting enzyme-1 forms a dimer thr ough Cys412 with a similar catalytic mechanism and a distinct substrate binding mechanism compared with neutral endopeptidase-24.11. Biochem J 315 ( Pt 3):863-867. Solomon B (2005) Generation of anti-bet a-amyloid antibodies via phage display technology towards Alzheimer's dis ease vaccination. Vaccine 23:2327-2330. Spencer B, Marr RA, Rockenstein E, Crew s L, Adame A, Potkar R, Patrick C, Gage FH, Verma IM, Masliah E (2008) Long-term neprilysin gene transfer is associated with reduced levels of intr acellular Abeta and behavioral improvement in APP transgenic mice. BMC Neurosci 9:109.

PAGE 111

103 Turner AJ, Isaac RE, Coates D (2001) The neprilysin (NEP) family of zinc metalloendopeptidases: genomics and function. Bioessays 23:261-269. Turner AJ, Brown CD, Carson JA, Bar nes K (2000) The neprilysin family in health and disease. Adv Ex p Med Biol 477:229-240. Wilcock DM, Rojiani A, Rosenthal A, Subbarao S, Freeman MJ, Gordon MN, Morgan D (2004) Passive immunotherapy against Abeta in aged APP-transgenic mice reverses cognitive deficits and depletes parenchymal amyloid deposits in spite of increased vascular amyloid and microhemorrhage. J Neuroinflammation 1:24. Yasojima K, McGeer EG, McGeer PL (2001a) Relationship between beta amyloid peptide generating molecules and neprilysin in Alzheimer disease and normal brain. Brain Res 919:115-121. Yasojima K, Akiyama H, McGeer EG, McGeer PL (2001b) Reduced neprilysin in high plaque areas of Alzhei mer brain: a possible relationship to deficient degradation of beta-amyloid pept ide. NeurosciLett 297:97-100. Zou LB, Mouri A, Iwata N, Saido TC, W ang D, Wang MW, Mizoguchi H, Noda Y, Nabeshima T (2006) Inhibition of neprilysi n by infusion of thiorphan into the hippocampus causes an accumulation of amyloid Beta and impairment of learning and memory. JP harmacolExpTher 317:334-340.

PAGE 112

104 PAPER 3: CONVECTION-ENHANCED DELIVERY AND MANNITOL AS A METHOD TO INCREASE DISTRIBUTION OF AAV VE CTORS 5, 8, AND 9 AND INCREASE GENE PRODUCT IN THE ADULT MOUSE BRAIN. Carty N.1, Nash K.2, Dickey C.1, Muzyczka N.2, Gordon M.N.1, Morgan D1, 1Alzheimer’s Research Laboratory, Departm ent of Molecular Pharmacology and Physiology, School of Biomedical Sciences University of South Florida College of Medicine, 12901 Bruce B Downs Blvd, Tampa, FL 33612, USA. 2 Department of Molecular Genetics and Microbiology, University of Florida College of Medicine, Box 100266, Gainesville FL 32610 Niki C Carty: ncarty@hsc.usf.edu Kevin Nash: Knash@health.usf.edu Chad Dickey: dickey.chad@gmail.com Nicholas Muzyczka: NMuzyczka@ufl.edu Marcia N Gordon: mgordon@hsc.usf.edu Dave Morgan: dmorgan@hsc.usf.edu Please address correspondence to: Dave Morgan Director, Basic Neuroscience Research University of South Florida

PAGE 113

105 12901 Bruce B Downs Blvd, MDC Box 8 Tampa, FL 33612 USA ABSTRACT The use of recombinant adeno-associ ated viral (rAAV) vectors as a means of gene delivery to the central nervous system has emerged as a viable method of gene therapy for t he treatment of several ty pes of degenerative brain diseases. A major disadvantage of typical in tracranial injections into the brain parenchyma is a limited distribution of the rAAV macromolecules to all brain regions where therapies may be needed. Optimizing specific parameters of the administration techniques with t he purpose of obtaining maximal gene distribution and gene uptake is an import ant obstacle to overcome for gene therapy studies. Convection enhanced deliv ery (CED) is a method of delivering clinically relevant volumes of therapeutic agents to signi ficantly larger areas of the brain. The CED technique is des igned to utilize the phenomenon of bulk flow and positive pressure to distribute macrom olecules to a lar ge area within solid tissue in a direct intracranial injection procedure. In the present study the CED method using the step-design cannula (Kra uze et al., 2005a) was used to deliver AAV vector serotype 5 expressing GFP into the hippocampus and cortical regions of the mouse brain. Regions of the hippocampus and cortex receiving the CED injection showed significantly more robust expression of GFP and an

PAGE 114

106 increased area of distribution than the simple diffusion method. We also compare transduction efficiency of AAV serotypes 5, 8, and 9 using CED administration. Following a single CED injection AAV 9 resulted in the largest area of distribution in the mouse CNS. Mo reover, the CED method, in addition to systemic administration of m annitol, was used to deliver rAAV vectors (serotypes 5 and 9) expressing GFP into the hippocampus of 11 month old mice. Mannitol is a blood brain barrier interruptive reagent which also induces hyperosmolarity thereby reducing intracranial pressure and facilitating the movement of particles through the interstitial space. Mice were injected intracranially into the right hippocampus using the CED method with (n =8) or without (n=8) pre treatment with mannitol delivered intraper itoneally. Animals receiving systemic injections of mannitol followed by the intracranial CE D injection of rAAV into the hippocampus showed an increased area of distribution compared to animals that did not receive pre treatment with mannitol. In addition, GFP expression was also detected in regions distant fr om the initial site of inje ction, in animals receiving the CED method with mannitol, includi ng significant expression in the contralateral hemisphere to the inject ion site. These data suggest that by optimizing injection techniques such as CED in addition to mannitol induced hyperosmolarity, provides an efficient means of increasing viral vector distribution and transgene uptake.

PAGE 115

107 INTRODUCTION AAV (adeno associated viral) vectors have recently emerged as a promising and novel mean by which transgenes can be delivered to different tissue types successfully in a large range of animal species including humans. AAV have a unique profile with a number of advantageous characteristics which identify them as one of the mo st feasible gene transfer ve ctors in the treatment of a variety of diseases, including neurologi cal diseases, compared to other viral vectors. AAVs are one of t he smallest viruses, approxim ately 25 nm in diameter, and their DNA genome is less than 5 kb which contains two large open reading frames with inverted terminals repeats lo cated at either end Thomas (Berns, 1990), (Thomas et al., 2003); (Wu et al., 2006). The most attractive characteristics of AAV is their lack if pathogenicity, persistence of the transgene as an episome, and long term gene expression. AAV vectors arrive in a variety of flavors otherwise known as seroty pes. The AAV serotype refers to the efficiency by which the specific AAV can infect a particular cell type through attachment to specific cell surface recept ors. A new AAV serotype is defined by the inability of an antibody t hat is reactive to the vi ral capsid protein of one serotype in neutralizing thos e of another serotype (Choi et al., 2005) The availability of different AAV serotypes is another major advantage that lends to its feasibility as a therapeutic in the tr eatment of neurodegenerative disease. Through the modification of serotype diversit y, a variety different capsid proteins can be incorporated to cr eate pseudotypes of recombinant AAV (mainly derived from AAV 2) which can ultimately medi ate tropism and transduction efficiencies

PAGE 116

108 targeted towards specified cell and tiss ue types throughout the body to optimize gene therapy (Choi et al., 2005), (Akac he et al., 2006), (Cearley and Wolfe, 2006), (Kwon and Schaffer, 2008). Unfortunately, one of the major disadv antages of a single intracranial injection of AAV vectors into the brain parenchyma using simple diffusion does not allow for efficient uptake of the tr ansgene or significant di stribution of the AAV macromolecules to significantly large areas of the affect ed regions within the brain. Convection enhanced delivery (CED ) is a method of delivering clinically relevant volumes of therapeutic agents to si gnificantly larger ar eas of the brain in a direct intracranial injection proc edure in comparison to simple diffusion methods. The CED technique is designed to utilize the phenomenon of bulk flow and positive pressure to distribute macrom olecules to a lar ge area within solid tissue. The CED technique was origina lly proposed by scientists in the early 1990s as a method of delivering drugs, or macromolecules, directly to the parenchyma that would not normally cr oss the blood brain barrier (Raghavan et al., 2006). Due to the lack of approved drugs that can be directly intracranially administered to the brain and the difficult y in predicting methods that ensure delivery of the therapeutic agent to its ta rget site, CED remains an experimental procedure. Furthermore, research of CED delivery devices is under current investigation by several researchers (Bankiewicz et al., 2000), (Raghavan et al., 2006). This CED method has been investigated in gene therapy studies as a way to increase the distribution of AAV vector s in the brain. Studies conducted by

PAGE 117

109 Bankiewicz et. al. in 2000 revealed that CED can significantly increase gene transfer and distribution of AAV expre ssing AADC in the st riatum of MPTPtreated monkeys. The AAV vector was f ound to be safely distributed throughout the entire region of the stri atum compared to the simple injection method where the distribution was severely limited (B ankiewicz et al., 2000). Similar results were replicated in the rat brain by Cunningham et. al. in 2000 with AAV2 expressing thymidine kinase (TK) wher e the CED method showed robust gene transfer and increased distribution area wit hin the putamen. CED injections in the striatum were found distribute the AAV-TK throughout the striatum after a single injection into this region and TK immunoreactive cells were also found outside the striatum, in the globus palli dus, subthalamic nucleus, thalamus, and substantia nigra (Cunningham et al., 2000), (Hadaczek et al., 2006). One of the mechanistic limitations of the CED method as well as the simple injection method is the reflux of the injected material from the injection hole upon the removal of the cannula. In 2002 Krauze et. al. developed a reflux free step cannula design which effectively eliminates reflux by placing silicone coated tubing within the cannula creating a step that prevent s the backflow of fluid (Krauze et al., 2005a) The optimiz ation of more efficient cannula designs coupled with the encouraging results from studies showing enhanced gene transfer and distribution emphasizes the therapeutic potential of the CED method in helping overcome some of the mechani cal disadvantages of gene delivery in regards to gene therapy (Krauze et al., 2005b).

PAGE 118

110 The use of osmotic agents such as mannitol is another method that can be used to increase the area of distribut ion of macromolecules throughout the CNS. Mannitol is a blood brain barrier in terruptive reagent and is also known to temporarily increase vascular pressure subsequently reducing intracranial pressure. High concentrations of manni tol intravenously infused are currently used in patients with traumatic brain dise ase to reduce intracranial pressure. This osmotic agent pulls fluid from t he CNS by increasing vascular osmotic pressure. Several studies have also show n that with intra-arterial infusion of mannitol the blood brain barrier can be opened to enhance the distribution of chemotherapeutics throughout the CNS in both rats and humans (Nilaver et al., 1995), (Rapoport, 2001), (Fu et al., 2003). In this particular study we anticipate that by optimization of these mechanistic strategies in addition to AAV serotype tissue specificity we can optimize transduc tion efficiency and increase distribution area of AAV in the brain. MATERIALS AND METHODS Animals Non transgenic C57BL6 mice were acquire d from the breeding colonies at the University of South Florida. Mu ltiple mice are housed together whenever possible until the time of use for the study; mice were then singly housed just before surgical procedures until the time of sacrifice. Study animals were given water and food ( ad libitum ) and maintained on the twelve hour light/dark cycle

PAGE 119

111 and standard vivarium conditions. Mice aged 9 to 11 months were used for all experimental procedures with an n = 4 6 for each experimental group. Step Design Cannula The step design cannula was used for all in tracranial surgeries. Fused silica tubing (polymicro technologies, P heonix, AZ) was inserted into a 27 gauge Hamilton blunt ended needle and fixed in pl ace with super glue (Krauze et al., 2005a). The end of the silica tubing was cu t leaving 1mm of tubing protruding from the end of t he Hamilton needle. GFP Expression Using CED Part A; animals were assigned to one of two cohorts, group one received AAV vector expressing GFP ( n =6). These animals rece ived a single intracranial injection of the AAV-GFP vector using the CED injection method (5ul/min) or the traditional injection method (0.5ul/min) into either the right or left hippocampus and into the right or left fr ontal cortex. I mmediately before su rgery mice were weighed then anesthetized usi ng isoflurane. Surgery was performed on animals using a stereotaxic apparatus, injecti ons using the CED method were used to inject into hippocampus and frontal cortex at a flow rate of 5ul /min over a total period of 2 min. (infusion time of 0.4min) using the CED method described earlier. The traditional injection method wa s used to inject into the hippocampus and frontal cortex at a flow ra te of .5ul/min over a tota l period of 4 min. (infusion time of 2 min.) also described earlier. The surgical procedure was performed by exposing the cranium using an incision th rough the skin along the median sagittal plane, and two holes were drilled through the cranium over the right frontal cortex

PAGE 120

112 injection site and the right hippocampal in jection site. Previously determined coordinates for burr holes, taken from bregma were as follows; frontal cortex, anteroposterior, 1.5mm; lateral, -2.0 mm, vertical, 3.0mm, hippocampus, anteroposterior, -2.7mm; lateral -2.5mm, vert ical, 3.0mm. Burr holes were drilled using a dental drill bit (SSW HP-3, SSWhi te Burs Inc., Lakewood, NJ). Each animal received both injection types in opposite hemispheres to control for mouse to mouse variability in gene uptake and expression. Six weeks post surgery mice were weighed, over dosed with pentobarbital (200 mg/kg) and perfused with 25 ml of 0.9% normal sali ne solution then 50 ml of freshly prepared 4% paraformaldehyde. Brains were co llected from the animals immediately following perfusion and immersion fixed in 4% paraformaldehyde for 24hrs. Mouse brains were cryoprotected in successive incubations in 10%, 20%, 30% solutions of sucrose; 24hrs in each soluti on. Subsequently, brains were frozen on a cold stage and sectioned in the horizontal plane (25 m thickness) on a sliding microtome and stored in Dulbe cco’s phosphate buffered saline (DPBS) with 0.2% sodium azide solution at 4 C. Six to eight sections 100 m apart spanning the site of injection were chosen and free-floating immunochemical and histological analysis was performed to determine gene expression using an anti-GFP antibody at a concentration 1:3,000 (Chemicon; Teme cula, CA). Immunohistochemical procedural methods were analogous to thos e described by Gordon et al. 2002 for each marker. Animal tissue was plac ed in multisample staining tray and endogenous peroxidase block ed (10% methanol, 30% H202, in PBS). Tissue

PAGE 121

113 samples were then permeabilized (with lysine 0.2%, 1% Triton X-100 in PBS solution), and incubated overnight in appropr iate primary antibody. Sections were washed in PBS then incubated in corresponding biotinylated secondary antibody (Vector Laboratories, Burlingame CA). The tissue was again washed after a 2hr. incubation period and incubat ed with Vectastin Elite ABC kit (Vector Laboratories, Burlingam e, CA) for enzyme conjugation. Finally, sections were stained using 0.05% diaminobenzidine and 0.3% H202 (for CD45 and Fc R 0.5% nickelous ammonium sulfate wa s added for color enhancement). Tissue sections were mounted onto slid es, dehydrated, and coverslipped. Part B of the experiment was to a ccess the safety profile of the CED injection method immediately following the injection procedure. Animals received a single injection of saline ( n =6) using either the CED method or the traditional injection method into the right or le ft hippocampus and frontal cortex. Each animal received both injection methods into opposite hemispheres to control for variations between animals with regard to microglial activation and neurotoxicity to determine the safety profile of the CE D technique. Animals were sacrificed 4 days post surgery as described earlier to determine whether the CED technique causes an acute significant increase in microglial activation or mechanical tissue damage using an immunohistochemi cal procedure described earlier. Six sections 100 m apart spanning the site of injection were chosen from each animal for part B of the experiment and free-floating immunochemical and histological analysis was performed for analysi s of the safety profile of the CED method in comparison to the traditional injection method. A series of sections

PAGE 122

114 from experiment part B were immunosta ined for CD45 to determine relative microglial activation. Each immunochemic al assay omitted some sections from primary antibody incubation period to evaluate nonspecific reaction of the secondary antibody. Another series of sections from part B were mounted on slides and stained with fluoro-jade stain (0.001%, Sigma Aldrich, St. Louis, MO) to assess the degree of neurotoxicity. Fluoro-jade is an anionic fluorochrome that selectively stains degenerating neurons effectively detecting neuropathic lesions by fluorescent microscopy (S chmued et al., 1997). A subsequent series of sections from part B (6 per animal) were mounted on slides and stained with the Cresyl Violet (Nissle) stain (0.05%, Si gma Aldrich, St. Louis, MO). The cresyl violet stains all Nissl bodies of the rough endoplasmic reticulum and other acidic components in all neuronal cytoplasm t herefore an absence of staining will indicate neuron loss. For experiments wh ich were analyzed to determine acute effects of the CED technique all i mmunohistochemical staining for CD45 and fluoro-jade and cresyl violet (Nissl) st aining all procedures followed the same protocol aforementioned 4 da ys post surgery rather than six weeks post surgery. GFP Expression with Serotypes AAV 5, 8, and 9 Study animals were assigned to one of three treatment groups receiving AAV serotype 5, 8, or 9. All AAV vector s contain a coding sequence for green fluorescent protein (GFP). All animals re ceived an intracranial injection into the right hippocampus and all injections we re performed using the CED technique described earlier. Each group received a singl e intracranial injection of 2ul of an each respective AAV vector expressing GFP (1.5 x 1011vg/ul) into the right

PAGE 123

115 hippocampus. All intracranial surger ies were performed on a stereotaxic apparatus using predetermined hippocam pal coordinates for the mouse hippocampus described earlier. Six weeks post injection animals were sacrificed and brain tissue was collected as described earlier. GFP Expression with Serotypes AAV 5 or 9 and Mannitol Animals were assigned to one of f our groups. Group 1 and 2 received a single intracranial injection of 2ul of AAV serotypes 5 and 9 respectively expressing GFP (1.5 x 1011vg/ul concentration) into t he right frontal cortex and right hippocampus using the CED injecti on method. Group 3 and 4 also received a single intracranial 2ul injection of the AAV-GFP serotypes 5 and 9 (1.5 x 1011vg/ul concentration) into the right fr ontal cortex and hippocampus in addition to a 200 ul single systemic intraperitoneal in jection of 25% mannitol administered 15 minutes before the intracr anial injection. The le ft untreated hemisphere in all animals remained untreated and used as an in ternal control. Surgery was again performed on animals using a stereotaxic apparatus using burr hole coordinates previously described. Six weeks post in jection animals were sacrificed and brain tissue was collected as described earlier. Sections were immunohistochemically stained using anti-GFP antibody (1:3000; Chemicon; Teme cula, CA) to test for gene distribution and expression in addition to the markers listed in part A using tissue animals in each group. Quantification and Statistical Analysis All immunostained sections were im aged using an Evolution MP digital camera mounted on an Olympus BX51 micr oscope at 100X final magnification

PAGE 124

116 (10 X objective). Immunof lourescent double labeled images were taken using a confocal microscope. Six to eight horizontal brain se ctions (100m apart; every 4th section) were taken from each ani mal and four nonove rlapping images near the site of injection from each of these sections were captured (24 measurements per mouse). All images were taken from the same location in all animals. Quantification of positiv e staining surrounding and including the injection sites in the right frontal cortex and the right hippocampus and the corresponding regions in the left hemis phere were determined using Image-Pro Plus (Media Cybernetics, Silver Springs MD). ANOVA statistical analysis was performed using StatView ve rsion 5.0.1 (SAS Institut e, Raleigh, NC). RESULTS Non transgenic mice, 9 months old, were injected bilaterally into the hippocampus using either the CED injecti on method or the traditional injection method. The CED injection method was performed as previ ously stated using the step design cannula with a flow rate of 5l/min with a total injection time of approximately 2 min. including the time t he cannula remained in the injection site to further prevent backflow of the AAV vect or material. The si mple injection was done at a flow rate of 0.5l/min with a tota l injection time of approximately 4 min. Each animal received both the CE D method in one hemisphere and the traditional injection method in the opposite he misphere to control for variations in gene uptake between individual animals. The animals were survived 4 weeks post surgery and histology was perfo rmed to assess gene distribution and

PAGE 125

117 expression. Regions of the hippocampus and cortex receiving the CED injection showed more robust expression of GFP and an increased area of distribution than the traditional in jection method (figure 1). In addition, GFP expression was also detected in regions distant from the in itial site of injection in the hemisphere which received the CED injection method. Areas with positive GFP staining included the entorhinal cortex, as well as some neurons in the striatum and thalamus. The striatum and entorrhinal cortex showed a significantly increased area of distribution in the animals rece iving the CED injection as opposed to the traditional injection method. The hippocam pal and cortical regions in which the CED method was implemented also showed an increased intensity of staining immediately surrounding the cannula tip and in some distal areas in regions surrounding the injection site. The diffe rence in staining intensity appears to indicate an increase in gene product (GFP) and potentially indicate that the CED technique increases the transduction effici ency of the AAV vector by increasing the susceptibility of the cell to infection. Part B of the experiment was to dete rmine if safety profile of the CED method on mechanical damage of parenchyma l brain tissue. Brain samples were analyzed for acute effects of t he CED method in a second cohort of animals. Non transgenic mice (n=6) were b ilaterally injected in the same manner described above (injections were of PBS) using both the CED and traditional injection methods. The mice were sacrificed and tissue was analyzed 4 days post injection and histology performed. Neurotoxity was determined using the fluoro-jade stain which was only positive in a very small area in 2 animals which

PAGE 126

118 was not significant. Microglial expr ession was also analyzed to assess the increase in microglial activation due to the mechanical technique. CD45 is a protein tyrosine phosphatase that is only expressed when microglial cells become activated, in this case, as a result of tissue damage (figure 2). Immunohistochemistry revealed that CD45 staining was increased but not significantly following the CED injections but this increase was limited to the area immediately surround the cannul a tract (figure 2). In addition the tissue was also stained for cresyl violet or Ni ssl. The cresyl violet stai ns all Nissl bodies of the rough endoplasmic reticulum and other ac idic components in all neuronal cytoplasm, therefore an abs ence of staining indicates neuron loss. The results from the tissue did not reveal any signi ficant areas devoid of Nissl staining signifying that there was no appr eciable neuron loss (Figure 2). To examine whether AAV serotype wa s able to affect transduction efficiency as well as area of distribution three different vector serotypes were administered into right hippocampus of 11 month old non transgenic mice using the CED method. A comparison of area of distribution as well as a comparison of transduced neurons in the hippocampus following administration of AAV serotypes 5, 8 and 9 each expressing GFP revealed that serotype 9 was distributed to a significantly greater area than either 8 or 5. Six weeks post injection, serotypes 8 and 9 not onl y transduced neurons throughout a large region of the entire hippocam pus but they were also successful in transducing neurons in the left hippocampus contralateral to the site of injection. Conversely, AAV serotype 5 was primarily expressed in the neurons of the hippocampus

PAGE 127

119 ipsilateral to the injection site. GF P positive neurons were limited to the pyramidal and molecular cell layers dentate gyrus as well as cells in the hilus and CA1 regions. A few GFP positive cells were observed in the entorhinal cortex in addition to a few GFP positive cells in the lateral thalamic region directly adjacent to the medial hippocampal regions but thes e were insignificant in comparison to serotypes 8 and 9. AAV 8 and 9 seroty pes revealed intens e staining of GFP positive neurons in all th ree CA regions of the hippocampus as well as very intense staining of pyramidal cells in t he molecular layer of the dentate gyurs and a large number of positive cells in t he hilus in the ipsilateral hippocampus. A large number of intensely stained cells can also be observed in the entorhinal cortex of the right hemispher e. It appears that the majo rity of cells that have been transduced and are positive for GFP have intense staining not only of the cell body but the axonal and den dritic projections (although slightly less intense). These axonal projections are positiv e throughout the hippocampus and corpus callosum (which contains axonal connections between the right and left hemisphere). The GFP positive axons ar e clearly stained at midbrain level ventral from the top of t he cortex. These axons are positive at the cells bodies from which they originate in the right contralateral hemisphere to the injections site, through the midline to the axonal te rmination site in the left hippocampus. Very few cells bodies were positively st ained in the left uninjected hemispheres of mice receiving the AAV 8 and 9 serotypes. Observations of the left hemisphere contralateral to the injections revealed positive staining of the axon terminals of neurons in the right dentate gy rus and CA regions. Intensely stained

PAGE 128

120 neurons can also be observed in the right entorhinal cortex for AAV 8 and 9 treatment groups, while the left entorhinal cortex was less intense and as in the left hippocampus, only axons were positivel y stained for GFP. Interestingly, GFP expression in animals receiving ei ther AAV 8 or AAV 9 was observed not only in regions proximal to the right hi ppocampal injection site (as with AAV 5), or the left hippocampus resulting from re trograde axonal transport, but was also evident in neurons distal to the site of injection. Following a single injection of AAV 8 or AAV9 GFP positive neurons in these groups were observed in the anterior cortex, thalamic nuclei, striat um, septal nuclei, superior colliculus, subiculum and cerebellum of the ipsilateral hemisphere. Quantification of the pe rcent area of positive st ain revealed that AAV9 showed a 5 fold significant increase in the area of distribution in the right (ipsilateral) hippocampus with the left (contralateral) hippocampus showed an approximate 6 fold increase in the area of distribution (figure 3). While, AAV serotype 8 was able to transduce a signifi cantly greater area and number of neurons than AAV5, it was significantly less effective, therefor e distributed to a significantly smaller region than AAV 9. Finally, we analyzed whether the osmo tic agent, mannitol, could improve total area of distribution thus enhancing tr ansduction efficiencies of the different AAV serotypes in addition to the utilization of the CED method for the intracranial administration. AAV serotypes 5 and 9 (the highest and lowest transducing vectors in our study respectively) were giv en via direct intracranial injection using the CED method to 11 month old non trans genic animals with or without systemic

PAGE 129

121 pre treatment with mannitol. Animals receiving a systemic injection of 25% mannitol followed by CED administrati on into the right hippocampus of AAV 5 showed significantly greater area of GFP expression 6 weeks post injection compared to animals only receiving AAV 5 administration alone. GFP expression in animals receiving systemic mannitol was throughout the entire right hippocampal region and was also obser ved in the left (contralateral) hippocampus (figure 4). All brain s lices (25m) in thickness through the hippocampus showed positive GFP staining in the mannitol treated group while the animals that did not receive mannitol did not have staining in the most ventral and dorsal portions of the hippocampus (Figur e 4). A 3 fold significant increase in the percent area of posit ive GFP stain in the right hippocampus was observed compared to animals that did not rece ive mannitol. In the contralateral hemisphere we observed a 2 fold significant increase in GFP expression in mannitol treated animals compared to ani mals without mannitol pretreatment (figure 4). All animals receiving AAV 9 adminis tered using CED with or without systemic mannitol showed much higher GF P expression than animals receiving AAV 5 similar to our previous findings In our comparison between animals receiving AAV 9 with systemic mannitol pretreatment and those that did not receive mannitol we did not observe signifi cant differences of GFP expression in the right hippocampus (at the site of injection) or in the left hippocampus (contralateral to the injection site). GFP expression in the right hemisphere was within the cell bodies of pyramidal neurons in the CA regions as well as granule

PAGE 130

122 cells in the dentate gyrus, mossy fibers and cells within the hilus. Axonal projections making up the perforant pat hway of the hippocampus were also positive for GFP expression. All cell bod ies and axonal proj ections in these regions were intensely stained revealing a significant amount of GFP expression. GFP positive neurons were observed in the subiculum and ent orhinal cortex which complete the hippocampal circuit. GFP expression in the left hemisphere was limited to axonal projec tions which originate in t he right hippocampal regions (the site of injection). In the left hi ppocampus a few cells in the hilus were positive for GFP expression, but most staining was observed in mossy fibers as well as axonal projections of CA regions presumably of those originating in the right hippocampus. The left entorhinal cortex was also positive for GFP expression but significantly less than the ri ght entorhinal cortex (figure 5). This was also true for cells within the thalamus (figure 5). GFP positive neurons were also observed in the right thalamic nuclei which receives input from the subiculum of the hippocampus, and int ense staining was observed in white matter tracts corpus collusum which connect the right and left hemispheres as well as the hippocampal commissure or comm issure of fornix adjoining the right and left hippocampus. Neurons positive for GFP expression were also observed in the anterior cortex, caudate putamen, st riatum, lateral and dorsal septal nuclei, superior colliculus, and cerebellum of the ipsilateral hemisphere in both groups receiving AAV9 independent of whether t hey received mannito l pretreatment. To further characterize vector se rotype tropism and whether tropism and transduction efficiency of AAV 5 and AAV9 was modified by mannitol the tissue

PAGE 131

123 was double labeled using fluorescent i mmunostaining. Our results, following injection with either AAV5 or AAV9, s howed no co-localization of astrocytes (GFAP which stains astrocytes) and GFP positive labelling in the dentate gyrus of the hippocampus (or in any other regions in the brain) (figure 8). This was noted in all animals receiving either serotype with or without mannitol pretreatment. GFP labeled cells only co -localized with Neu-N (which stains neurons) positive staining (figure 8). Posi tive labelling in the left uninjected hemisphere was in a few cells in the hilus of the DG but the ma jority of positive labelling was in the axons surrounding the cell bodies (figure 8, panel E-H). DISCUSSION The recent identification of different AAV serotypes has advanced the study of rAAV vectors which has quickly become a major dominant focus in the field of gene therapy. There are a number of studies aimed at characterizing the different AAV serotypes with respect to transduction efficiency, tissue tropism, cell surface receptors, intracellular pr ocessing, and capsid structure. AAV2 serotype was the first to be cloned into bacte rial plasmids by Sa mulski et. al. in 1982 and to date it is the most characterized of all the serotypes of which a total of 11 have been discovered (Samulski et al., 1982). AAV5 was originally discovered in a human clinical sample in 1984 as a contaminant in adenovirus stock and it contains ITRs that are not unlike the structures of AAV2 ITRs and is the most divergent of all of the serotypes (Choi et al ., 2005), (Wu et al., 2007). AAV8 was originally isolated from rhesus monkey tissue in 2002 by Gao and

PAGE 132

124 collegues (Gao et al., 2002). AAV9, like AAV5, was originally discovered in human tissue, while AAV10 and AAV11 ar e the most recently identified serotypes, initially isolated from cynomol gus monkey tissue in 2004 by Mori et. al., to date have not been fully characterized (Mori et al., 2006), (Wu et al., 2006), (Wu et al., 2007). Several recent studies have shown that different serotypes have distinct variations in their trans duction efficiencies depending on the specific cell or tissue type, yet the molecular mechanisms which determine the preferred target cell for many serotypes remains unknown. Despite what is currently known concerning different AAV transducti on profiles, the data regarding AAV serotype specific tissue tropism is subj ect to different interpretation due to variations in vector titer, promoter s, and transgenes between studies. In the CNS, serotype characterization studies have revealed that AAV1, and 5 have higher transduction efficiencies than AAV2 throughout all the regions and cell types of the CNS (Alisky et al., 2000); (Bur ger et al., 2004), (Burger et al., 2005) while AAV4 will efficiently transduce specific cell types such as astrocytes within the subventricular zone (Davidson et al., 2000), (Weber et al., 2003), (Wu et al., 2006). Studies by Wolfe and collegues al so reveal that AAV7, 8, 9 and Rh10 expressing cDNA for lysosomal enzyme ar e also capable of transducing neurons within specific regions in the mous e brain. AAV9 and AAVRh10 appeared to have the highest transduction efficiencies and were found to undergo vector genome transport through axonal transport pathways (Cearley and Wolfe, 2006, 2007)(Cearley and Wolfe, 2006).

PAGE 133

125 AAV2 vectors are the most extensivel y characterized and due to their well established safety profile and range of infe ctivity, approximately 20 clinical trails have been conducted using the AAV2 vector serotypes in numerous patients (Herzog and Hagstrom, 2001), (Grieger and Samulski, 2005b), (Wu et al., 2006), (Cai et al., 2009). Phase I/II clinic al trails have been initiated utilizing recombinant AAV 2 vectors for the treatment of human diseases such as cystic fibrosis, -1 anti-trypsin deficiency, Parkin son’s disease, Batten’s disease, muscular dystrophy, hemophilia B, (Robbi ns and Ghivizzani, 1998) (Zhong et al., 2008). Unfortunately, one major disadvant age of using AAV in the treatment of brain disorders is the limitation in the dispersion of specific therapeutic gene or protein to relevant regions or the entire brain (Cearley and Wolfe, 2007). In this particular study we aim to characterize and compare transduction efficiencies in the mouse CNS for AAV serotypes 5, 8, and 9 after a single intracranial administration and improve distribution of the AAV as well as increase gene expression in brain regions distal to t he injection site using optimal injection techniques in an innovative manner. Prev ious data from comparison studies of different AAV serotypes have shown AAV 2, 5, 7 ,8 and 9 efficiently transduce neurons in the murine CNS (Fu et al., 2003) (Burger et al., 2004) (Burger et al., 2005) (Taymans et al., 2007). We hy pothesized that we would be able to sucessfully transduce neurons in the mous e brain after a single intracranial injection of AAV 5, 8, or 9 (that expre ss GFP) and we will be able to increase the area of vector distribution using a CED adm inistration technique in addition to systemic mannitol treatment in a novel combinatory fashion.

PAGE 134

126 Following a single intracranial inject ion we successfully demonstrated that AAV9 was the most efficient serotype in transducing a larger number of neurons in a larger area of the brain than both AAV 5 and AAV 8. AAV 8 was only slightly less efficient than AAV 9 and much more efficient that AAV5 at transducing neurons in the mouse CNS. Our data in consistent with others who have also shown that both AAV 8 and 9 can more effi ciently transduce a larger region of brain than AAV 5 in a murine animal model and yet others have shown this is pattern is consistent in other species as well (Cearley and Wolfe, 2006). AAV 9 not only provides superior transduction compared to serotypes 5 and 8 in the CNS but in other organs such as heart, liver and lung as well as global transduction of several cell and tissue ty pes in the periphery when systemically injected (Bish et al., 2008), (Zincarelli et al., 2008). Furthermore, studies have clearly shown AAV serotype 8 to have si gnificantly superior transduction efficiency than AAV2 or 5 in murine and nonhuman primate animal models which is also consistent with our data showi ng serotype 8 to be more efficient at transducing neurons in the mouse CNS than serotype 5 (Davidoff et al., 2005). There are several explanations which may account for the differences in transduction efficiencies between the different serotypes. AAV mu st first bind to the cell surface receptor in order to successfully transduce the neurons once injected into the brain (Cearley and Wolfe, 2006), (Kwon and Schaffer, 2008), (Daya and Berns, 2008). This is followed by a series of essential steps; viral uptake through endocytosis, intracellular tr afficking and translocation of the particle to the nucleus, virion uncoating, synthesis of double stranded DNA and

PAGE 135

127 finally viral gene expression (Rabinowitz et al., 2004), (Nash et al., 2007), (Kwon and Schaffer, 2008), (Nash et al., 2009); (C hoi et al., 2005); (Cearley and Wolfe, 2006), (Cearley et al., 2008); (Daya and Bern s, 2008). Each of these steps is essential to successful transgene expr ession and also determines the unique tropism of the AAV serotype. Although t he cell surface binding receptor for AAV 1, 2, 3, 4, and 5 have been previously de scribed, receptors for more recently discovered serotypes remains unclear. In 1998 Samulski et al determined that AAV 2 utilizes the heparin su lfate proteoglycan receptor as its binding target and also uses secondary binding coreceptor targets (including the V 5 integrin and fibroblast growth factor receptor) whic h help stabilize viral binding thereby enhancing transduction efficiency (S ummerford and Samulski, 1998), (Summerford et al., 1999), (Qiao et al ., 2002), (Qing et al., 2003). AAV 1 has been shown to bind to the same cell surfac e receptor and co-receptors as AAV 2, while AAV 5 has been shown to interact wit h the sailic acid receptor and platelet growth factor receptor and AAV 4, simila r to AAV5, will bind to the sailic acid receptor but needs specific carbohydrate linkage to successfully bind to the cell surface (Walters et al., 2000), (Walters et al., 2001), (Kaludov et al., 2003), (Akache et al., 2006). Most recently, Akache et al. have demonstrated that AAV 8 uses the 37/67 kD lamini n receptor (LamR) to bind to mammalian cells and overexpression of the LamR receptor results in increased binding of AAV 2,3 8 and 9 (expressing GFP) in vitro and a 2.5 fold increase in GFP expression (Akache et al., 2006). These data indicate that the differences in expression patterns between the different AAV serotypes used in the present study is

PAGE 136

128 primarily mediated by the cells surface re ceptor that the AAV serotype binds to as well the co-receptors and the expression patterns of these receptors in the mouse CNS. Other factor s such as those previously mentioned (intracellular trafficking and translocation of the particl e to the nucleus, virion uncoating, and synthesis of double stranded DNA and gene expression) may play a secondary role in transduction efficiencies. Finally the capacity to neutralize viral infection particles with neutralizing antibodies, gener ated from a previous exposure, also can affect the transduction profile of each serotype. Our findings as well as others confirm that AAV 9 has significant ly higher transduction efficiency in neurons in the CNS compared to AAV 8 and AAV 5. The differences in cell specific transduction efficiencies are likely a consequence of the effective cellular entry through cell surface receptor m odulation as well as un-packaging and postentry properties of each serotype (Hau ck et al., 2004), (Wang et al., 2007). We speculate that in this study AAV 9 was the most efficient likely due to its effective and stable binding to specific cell surfac e receptors possibly the LamR receptor or other potential receptors in addition to their expression profile within the murine CNS. AAV 9 may be taken up by the cell more efficiently followed by more efficient translocation and viral uncoating and un-packaging within the nucleus of the neuronal cell and effici ent DNA replication and expression properties. The relative importance of eac h step in the process of transduction in the mouse CNS remains unclear and warra nts more study which would lend to efforts in creating tailored recombi nant AAV vectors with unique and disease modifying specificity.

PAGE 137

129 Our next goal was to determine an optim al injection protocol aimed at improving distribution proper ties of the AAV vectors wit hin the brain parenchyma. All of the intracranial injections into the hippocampus were performed using the CED technique with a modified version of the step design cannula (Bankiewicz et al., 2000) (Krauze et al., 2005a) which has previously been shown by us and others to further increase the area of distribution of material in the brain parenchyma in murine, canine, nonhuman primate as well as in humans (Bankiewicz et al., 2000), (Krauze et al., 2005b), (Szerlip et al., 2007), (Cunningham et al., 2008), (Dickinson et al., 2008), (Fiandaca et al., 2008). In the present study we successfully further increase the area of distribution and gene expression by combining the CED me thod with administra tion of mannitol systemically approximately 15 min prior to t he intracranial injection of AAV. Our data are consistent with previous studies using AAV 2, in which mannitol further increased the area of di stribution within the st riatum of the rat br ain (Burger et al., 2004). We clearly demonstrated that mannitol, by osmotically lowering intracranial pressure, could significantly increase the area of distribution of AAV 5 after a single CED administ ration into the DG of the right hippocampus. This distribution was evident throughout the ent ire right hippocampus allowing viral particles to travel to areas throughout the right hippocampus as well as other areas adjacent to the injection site such as the entorhinal cortex and thalamus (although the expression was limited to a few cells in these regions). In addition a significant amount of GFP gene expres sion was noted in the left uninjected hemisphere most likely due to axonal retrograde transport which was not as

PAGE 138

130 apparent without mannitol pret reatment. The increase in GFP expression in the left hippocampus we attribute to greatly increasing the number of cells in the injected hemisphere that were exposed an d therefore infect ed and expressing significantly more GFP that was then transported axonally to the left hemisphere. Mannitol also increased GFP gene expressi on in the AAV 9 treatment group in the entorhinal cortex and thal amus. We did not see significant changes in the hippocampus due to the preexisting hi gh transduction efficiency of AAV 9 throughout the right and left hippocampi. Despite the rapidly growing popul arity of AAV as a method of gene delivery, much remains unknown with r egard to cellular mechanisms that are responsible for different AAV characte ristics. Although AAV demonstrate longterm gene expression in vivo and a good safety profile, potential risks due to the host response to these vectors need furt her study. Unlike adenovirus vectors which induce expression of chemoki nes and acute toxicity, AAV have not been previously associated with an inflammatory response or significant toxcity (Beuler et al., 1999), (Carter et al., 2000), (Muruve et al., 2004). Some studies have shown repeated administration of AAV in the periphery results in a hummoral immune response, generated in the first administration, which dramatically lowers transduction efficiency upon se cond administration (Halbert et al., 1997),(Halbert et al., 1998). This problem can be overcome by host immunosuppression (Halbert et al., 1998), (Manning et al., 1998). The majority of the human population (an estimated 80%) has been naturally previously exposed to wt AAV 2 and demonstrates presence of neutralizing anticapsid antibodies

PAGE 139

131 (Chirmule et al., 1999), (Erl es et al., 1999) which coul d be a potential problem for clinical success of AAV 2. Recently studies have demonstrated that generating recombinant AAV can effectively overcome wt AAV neutralizing antibodies (Mandell et al., 2004). Through modificati on of wt AAV and the creation of a variety of rAAV with uniquely tailored spec ificity, there is immeasurable potential for the use of AAV as a successful gene therapy technique. In the present study we investigate not only differences in transduction profiles of specific AAV serotypes but also optimization of administration techniques with regard to AAV in the CNS. In recent years, scientists have conducted several studies aimed at impr oving transduction efficiencies of AAV vectors in different tissue types thr ough modification of the AAV genome. The utilization of a “cross-packaging” strat egy which essentially enables an unbiased direct comparison of the transducti on rates without influence of ITRs on transgene expression has recently become a novel way to create recombinant AAV vectors (Rabinowitz and Samulski, 2000) (Rabinowitz et al., 2004), (Grieger and Samulski, 2005a), (Wu et al., 2006). To accomplish this the Cap genes of different serotypes are placed downst ream of the AAV2 Rep genes which ultimately allows the generation of seroty pes specific capsids while the packaged genomes within the capsids are identical, these are also referred to as hybrid viruses (Rabinowitz et al., 2002), (Wu et al., 2006). In addition to transcapsidation first described by Rabino witz et. al. in 2002 there are several other techniques to accomplish the forma tion of a hybrid virus armed with specific modifications to enhance effi ciency of gene uptake, transfer, and

PAGE 140

132 expression in a specif ied therapeutic avenue. These include absorption modifications where the capsid surface is modified to carry a foreign antibody that will bind to the cell surface recept or of interest to increase absorption efficiency. Mosaic capsids are anot her method to increase transduction efficiency. This technique involves creat ing a mixture of viral capsid proteins from different serotypes at various ratios in order to combine tissue tropisms of interest (Xiao et al., 1999), (Rabinowit z and Samulski, 2000), (Kruger et al., 2008). Finally, chimeric capsids can also provide a means of increasing transduction of a particular specificity and this technique involves the packaging of capsid proteins with foreign peptide s equences, such as a hemagglutinin (HA) tag fused at either the N or C terminus of the capsid coding sequence to alter tissue tropism (Yang et al., 1999), (Bo wles et al., 2003), (Wu et al., 2006). Despite the rapidly growing body of knowledge in the field of AAV vector serology there is still much to be done. In addition to the 11 AAV serotypes that have been found, over 100 AAV variants have al so been discovered. Delineating differences in transduction properties, and defining distinct characteristics of each virus is a burgeoning area of investigation. The technological advances in recombinant molecular biology have permitted investigat ors to construct a myriad of recombinant AAV vectors that c an be customized to take advantage of differences in transduction profiles which can be targeted for several different cell types. In addition, optimization of admin istration techniques that increase distributive properties of AAV such as CED and mannitol have improved gene delivery capacity of these vectors sugges ting near limiteless potential they may

PAGE 141

133 provide as a superior gene transfer vehicl e through which to treat a variety of neurological diseases. Acknowledgments. Supported by The J ohnnie Byrd Center for Alzheimer's Research, NIH grants AG -25509, AG 15490, AG 18478, AG 04418, AG 25711 We also acknowledge Todd Golde for prov iding the rAAV–GFP vector serotypes 8 and 9. FIGURE 1: GFP EXPRESSION IS INCREASED FOLLOWING INTRACRANIAL ADMINISTRATION OF AAV5 USING CO NVECTION ENHANCED DELIVERY. Percent positive area stained for anti-GF P is increased in the hippocampus, cortex and thalamus in 9 mont h old mice using CED method (5 l/min flow rate). The percent positive staining of GFP is si gnificantly increased in the striatum and entorhinal cortex using the CED met hod compared to traditional injection method. The star (*) indicates significanc e with a p-value < 0.05. Magnification 100X on Olympus BX51.

PAGE 142

134 Ent CX CA 1 CED NO CED DG B A C D E F G

PAGE 143

135 FIGURE 2: CED METHOD DOES NOT RESULT IN NEURON LOSS OR SIGNIFICANT INCREASE IN CD45 EXPRESSION. Nissl staining showed no obvious loss of neurons in the hippocam pus following either the CED injection method (panel B) or the traditi onal injection method (panel A). CD45 immunostaining is increased (panel E) but not significantly in animals receiving the CED injection (panel D) compared to animals receiving the traditional injection (panel C). Quantification of CD45 immunostaining is represented as % Area stain (panel E). Magni fication 40X on Olympus BX51. E

PAGE 144

136 FIGURE 3: COMPARISON OF AAV SEROTYPE S 5, 8, AND 9 EXPRESSING GFP. Expression of GFP in the right and left hippocampus following a single CED intracranial injection into t he right hippocampus of AAV5, 8, and 9 serotypes. AAV 9 was distributed to a significantly larger area of the hippocampus in both the left and right hi ppocampus (panels A and B) following treatment than both AAV serotype 8 and 5 ( panel C-F). The star (*) indicates significance with a p-value < 0.05. Magnification 40X on Olympus BX51. AAV 9 AAV 8 AAV 5 Right Left HPC B A C D E F G

PAGE 145

137 FIGURE 4: GFP EXPRESSION IS INCREASED FOLLOWING ADMINISTRATION OF AAV5 USING CED AND SYSTEMIC MANNITOL. Percent positive area stained fo r anti-GFP is significantly increased in the right and left hippocampus in 11 month old mice following CED delivery of AAV5 and systemic mannitol pretreatm ent (panel C and D) compar ed to animals injected with AAV 5 and did not receive systemic mannitol (panel A and B). Panel E shows quantification of percent area positive stain for anti-GFP immunohistochemistry. The star (*) indica tes significance with a p-value < 0.05. Magnification 40X on Olympus BX51 (panels A-D). E No Mannitol Mannitol

PAGE 146

138 FIGURE 5: GFP EXPRESSION IS INCREASED IN THE THALAMUS AND ENTORINAL CORTEX FOLLOWING ADM INISTRATION OF AAV9 USING CED AND SYSTEMIC MANNITOL. Positive area stained for anti-GFP was intense throughout the entire hippocampus for all 11 month old mice following CED delivery of AAV9 wit h (panels A and B) or wit hout systemic mannitol pretreatment (panels C and D) A larger area of pos itive area staining was evident in the both the left and right entor hinal cortex and the right thalamus following CED delivery of AAV9 in anima ls pretreated with systemic mannitol (panels G and H) compared to animals th at did not receive systemic mannitol (panels E and F). Magnification 40 X on Olympus BX51 (panels A-H). No Mannitol Mannitol

PAGE 147

139 FIGURE 6: QUANTIFICATION OF ANTI-GFP STAINING FOLLOWING AAV 9 ADMINISTRATION. Quantification of Percent posit ive area stained for anti-GFP is significantly increased in the thalamus and entorhinal cortex in 11 month old mice following CED delivery of AAV9 and syst emic mannitol pretreatment (panel B) No significant difference in anti-GF P staining was noted in the right or left hippocampus following AAV 9 admini stration and systemic mannitol pretreatment. The star (*) indicates significance with a p-value < 0.05. B A

PAGE 148

140 FIGURE 7: TRANSDUCTION EFFICIENCY IN HIPPOCAMPUS FOLLOWING AAV5 AND AAV9 ADMINISTRATION. Transduction of GFP following AAV 5 and AAV9 administration into the hippocampus. GFP is expressed in the dentate gyrus granule and pyramidal cell bodies and axons in CA regions in the hippocampus primarily in the right hem isphere following a single intracranial injection of AAV 5 using CED method and syst emic mannitol administration into the right hippocampus and in axonal projec tions in the left hippocampus (panels C and D; and E-H). GFP is expressed in neuronal cell bodies and axonal projections in all regions of right hi ppocampus following AAV 9 injections using CED and systemic mannitol administrati on (panels A and B). GFP expression is primarily in axonal projecti ons eminating from the right hippocampus as well as in some neuronal cell bodies in the hilus of the dentate gyrus (panels I-K). Magnification 40X panels A-D; m agnification 100X panels E-K.

PAGE 149

141

PAGE 150

142 FIGURE 8: TRANSDUCTION EFFICIENCY AND GFP EXRESSION IN CELL TYPES FOLLOWING AAV 5 AND AAV9 ADMINISTRATION INTO HIPPOCAMPUS. GFP is expressed in the granule and pyramidal cell layers in the hippocampus as well as in neurons in en torhinal cortex primarily in the right hemisphere following a single intracranial in jection into the right hippocampus. Panels A-D shows no co-localization (panel D) of astrocytic staining (GFAP) (panel B) and GFP positive labeling (panel A) in the dentate gyrus of the hippocampus (or in any other regions in t he brain). This was noted in all animals receiving either serotype. Positive l abelling in the left uninjected hemisphere was in a few cells in the hilus (panel E) of the DG. Co-localization of NeuN and GFP positive staining indicate that mainly neurons were transfected in this region (panel E-H). The majority of positive l abelling was in the axons surrounding the cell bodies (panel E). Panels I-L show staining of the ento rhinal cortex ipsilateral to the injection site. Co-localization of GFP positive cells and Neu-N (panel L) indicates that AAV transfect mainly neurons in this region. Magnification 100X.

PAGE 151

143 In j ected Side Dentate G y rus GFP GFAP DAPI DAPI GFP NeuN MERGED MERGED E I F B J C G K D H L Unin j ected Side Dentate G y rus In j ected Side Entorhinal A GFP NeuN DAPI MERGED

PAGE 152

144 REFERENCE LIST Akache B, Grimm D, Pandey K, Yant SR, Xu H, Kay MA (2006) The 37/67kilodalton laminin receptor is a recept or for adeno-associated virus serotypes 8, 2, 3, and 9. J Virol 80:9831-9836. Alisky JM, Hughes SM, Sauter SL, Jo lly D, Dubensky TW, Jr., Staber PD, Chiorini JA, Davidson BL (2000) Transduc tion of murine ce rebellar neurons with recombinant FIV and AAV5 vect ors. Neuroreport 11:2669-2673. Bankiewicz KS, Eberling JL, Kohutnicka M, Jagust W, Pivirotto P, Bringas J, Cunningham J, Budinger TF, Harvey -White J (2000) Convection-enhanced delivery of AAV vector in parkinsonian monkeys; in vivo detection of gene expression and restoration of dopaminergic function us ing pro-drug approach. ExpNeurol 164:2-14. Berns KI (1990) Parvovirus replic ation. Microbiol Rev 54:316-329. Bish LT, Morine K, Sleeper MM, Sanm iguel J, Wu D, Gao G, Wilson JM, Sweeney L (2008) AAV9 Provides Global Cardiac Gene Transfer Superior to AAV1, AAV6, AAV7, and AAV8 in the Mouse and Rat. Hum Gene Ther. Bowles DE, Rabinowitz JE, Samulski RJ (2003) Marker rescue of adenoassociated virus (AAV) capsid mutant s: a novel approach for chimeric AAV production. JVirol 77:423-432. Burger C, Nash K, Mandel RJ (2005) Recombinant adeno-associated viral vectors in the nervous syst em. HumGene Ther 16:781-791. Burger C, Gorbatyuk OS, Velardo MJ, P eden CS, Williams P, Zolotukhin S, Reier PJ, Mandel RJ, Muzyczka N (2004) Reco mbinant AAV viral vectors pseudotyped with viral capsids from serotypes 1, 2, and 5 display differential efficiency and cell tropism after delivery to different regions of the central nervous system. MolTher 10:302-317. Cai X, Conley SM, Naash MI (2009) R PE65: role in the visual cycle, human retinal disease, and gene therapy Ophthalmic Genet 30:57-62. Cearley CN, Wolfe JH (2006) Transduction characteristics of adeno-associated virus vectors expressing cap serotypes 7, 8, 9, and Rh10 in the mouse brain. MolTher 13:528-537. Cearley CN, Wolfe JH (2007) A single injection of an adeno-associated virus vector into nuclei with di vergent connections result s in widespread vector distribution in the brain and global co rrection of a neurogenetic disease. J Neurosci 27:9928-9940.

PAGE 153

145 Cearley CN, Vandenberghe LH, Parente MK, Carnish ER, Wilson JM, Wolfe JH (2008) Expanded repertoire of AAV vector serotypes me diate unique patterns of transduction in mouse brai n. Mol Ther 16:1710-1718. Choi VW, McCarty DM, Samulski RJ (2005) AAV hybrid serotypes: improved vectors for gene delivery. CurrGene Ther 5:299-310. Cunningham J, Oiwa Y, N agy D, Podsakoff G, Colosi P, Bankiewicz KS (2000) Distribution of AAV-TK following intracr anial convection-enhanced delivery into rats. Cell Transplant 9:585-594. Cunningham J, Pivirotto P, Bringas J, Suzuki B, Vijay S, Sanftner L, Kitamura M, Chan C, Bankiewicz KS (2008) Biodistribution of adenoassociated virus type-2 in nonhuman primates after convection-en hanced delivery to brain. Mol Ther 16:1267-1275. Davidoff AM, Gray JT, Ng CY, Zhang Y, Zhou J, Spence Y, Bakar Y, Nathwani AC (2005) Comparison of the abilit y of adeno-associated viral vectors pseudotyped with serotype 2, 5, and 8 capsid proteins to mediate efficient transduction of the liver in murine and nonhuman primate models. Mol Ther 11:875-888. Davidson BL, Stein CS, Heth JA, Martin s I, Kotin RM, Derksen TA, Zabner J, Ghodsi A, Chiorini JA (2000) Recombinant adeno-associated virus type 2, 4, and 5 vectors: transduction of variant cell types and regions in the mammalian central nervous system. ProcNatl AcadSciUSA 97:3428-3432. Daya S, Berns KI (2008) Gene therapy using adeno-associated virus vectors. Clin Microbiol Rev 21:583-593. Dickinson PJ, LeCouteur RA, Higgins RJ, Bringas JR, Roberts B, Larson RF, Yamashita Y, Krauze M, Noble CO, Drumm ond D, Kirpotin DB, Park JW, Berger MS, Bankiewicz KS (2008) Canine model of convection-enhanced delivery of liposomes containing CPT-11 monitor ed with real-time magnetic resonance imaging: laboratory investigat ion. J Neurosurg 108:989-998. Fiandaca MS, Forsayeth JR, Dickinson PJ Bankiewicz KS (2008) Image-guided convection-enhanced delivery platform in th e treatment of neurol ogical diseases. Neurotherapeutics 5:123-127. Fu H, Muenzer J, Samulski RJ, Breese G, Sifford J, Zeng X, McCarty DM (2003) Self-complementary adeno-asso ciated virus serotype 2 vector: global distribution and broad dispersion of AAV-mediated tr ansgene expression in mouse brain. MolTher 8:911-917.

PAGE 154

146 Grieger JC, Samulski RJ (2005a) Adenoassociated virus as a gene therapy vector: vector development, production and clinical applicatio ns. AdvBiochemEng Biotechnol 99:119-45.:119-145. Grieger JC, Samulski RJ (2005b) Adenoassociated virus as a gene therapy vector: vector development, production and clinical applications. Adv Biochem Eng Biotechnol 99:119-145. Hadaczek P, Kohutnicka M, Krauze MT, Br ingas J, Pivirotto P, Cunningham J, Bankiewicz K (2006) Convection-enhanced delivery of adeno-associated virus type 2 (AAV2) into the striatum and tr ansport of AAV2 within monkey brain. HumGene Ther 17:291-302. Hauck B, Zhao W, High K, Xiao W (2004) Intracellular viral processing, not single-stranded DNA accumulation, is cruc ial for recombinant adeno-associated virus transduction. J Virol 78:13678-13686. Herzog RW, Hagstrom JN (2001) Gene therapy for hereditary hematological disorders. Am J Pharmacogenomics 1:137-144. Kaludov N, Padron E, Govindasamy L, McKenna R, Chiorini JA, AgbandjeMcKenna M (2003) Production, purification and preliminary X-ra y crystallographic studies of adeno-associated virus serotype 4. Virology 306:1-6. Krauze MT, Saito R, Noble C, Tamas M, Bringas J, Park JW, Berger MS, Bankiewicz K (2005a) Refl ux-free cannula for convection-enhanced high-speed delivery of therapeutic agent s. JNeurosurg 103:923-929. Krauze MT, McKnight TR, Yamashita Y, Bringas J, Noble CO, Saito R, Geletneky K, Forsayeth J, Berger MS, Jackson P, Park JW, Bankiewicz KS (2005b) Real-time visualizati on and characterization of li posomal delivery into the monkey brain by magnetic resonance im aging. Brain ResBrain ResProtoc 16:2026. Kruger L, Eskerski H, Dinsart C, Cor nelis J, Rommelaere J, Haberkorn U, Kleinschmidt JA (2008) Augmented trans gene expression in transformed cells using a parvoviral hybrid vector Cancer Gene Ther 15:252-267. Kwon I, Schaffer DV (2008) Designer gene delivery vectors: molecular engineering and evolution of adeno-associated viral vectors for enhanced gene transfer. Pharm Res 25:489-499. Mori S, Takeuchi T, Enomoto Y, Kondo K, Sato K, Ono F, Iwata N, Sata T, Kanda T (2006) Biodistribution of a lo w dose of intravenously administered AAV2, 10, and 11 vectors to cynomolgus monkeys. JpnJInfectDis 59:285-293.

PAGE 155

147 Nash K, Chen W, Salganik M, Muzyczka N (2009) Identification of cellular proteins that interact with the adeno-associated virus rep protein. J Virol 83:454469. Nash K, Chen W, McDonald WF, Zhou X, Muzyczka N (2007) Purification of host cell enzymes involved in adeno-associated virus DNA replication. J Virol 81:5777-5787. Nilaver G, Muldoon LL, Kroll RA, Pagel MA, Breakefield XO, Davidson BL, Neuwelt EA (1995) Delivery of herpesvirus and adenovirus to nude rat intracerebral tumors after osmotic blood-brai n barrier disruption. ProcNatlAcadSciUSA 92:9829-9833. Qiao C, Li J, Skold A, Zhang X, Xiao X (2002) Feasibility of generating adenoassociated virus packaging cell lines containing inducible adenovirus helper genes. J Virol 76:1904-1913. Qing K, Li W, Zhong L, Tan M, Hansen J, Weigel-Kelley KA, Chen L, Yoder MC, Srivastava A (2003) Adeno-associated viru s type 2-mediated gene transfer: role of cellular T-cell protein tyrosine phos phatase in transgene expression in established cell lines in vitro and trans genic mice in vivo. J Virol 77:2741-2746. Rabinowitz JE, Samulski RJ (2000) Buildi ng a better vector: th e manipulation of AAV virions. Virology %20;278:301-308. Rabinowitz JE, Bowles DE, Faust SM, Ledford JG, Cunningham SE, Samulski RJ (2004) Cross-dressing the virion: t he transcapsidation of adeno-associated virus serotypes functionally defin es subgroups. JVirol 78:4421-4432. Rabinowitz JE, Rolling F, Li C, Conrath H, Xiao W, Xiao X, Samulski RJ (2002) Cross-packaging of a single adeno-associ ated virus (AAV) type 2 vector genome into multiple AAV serotypes enables trans duction with broad specificity. JVirol 76:791-801. Raghavan R, Brady ML, Rodriguez-Ponce MI Hartlep A, Pedain C, Sampson JH (2006) Convection-enhanced delivery of therapeutics for brain disease, and its optimization. Neuros urgFocus 20:E12. Rapoport SI (2001) Advances in osmoti c opening of the blood-brain barrier to enhance CNS chemotherapy. ExpertOp inInvestigDrugs 10:1809-1818. Robbins PD, Ghivizzani SC (1998) Vi ral vectors for gene therapy. Pharmacol Ther 80:35-47.

PAGE 156

148 Samulski RJ, Berns KI, Tan M, Muzyczka N (1982) Cloning of adeno-associated virus into pBR322: rescue of intact viru s from the recombinant plasmid in human cells. Proc Natl Acad Sci U S A 79:2077-2081. Schmued LC, Albertson C, Slikker W, Jr. (1997) Fluoro-Jade: a novel fluorochrome for the sensitiv e and reliable histochemical localization of neuronal degeneration. Brain Res 751:37-46. Summerford C, Samulski RJ (1998) Me mbrane-associated heparan sulfate proteoglycan is a receptor for adeno-asso ciated virus type 2 virions. J Virol 72:1438-1445. Summerford C, Bartlett JS, Samulski RJ (1999) AlphaVbeta5 integrin: a coreceptor for adeno-associated virus ty pe 2 infection. Nat Med 5:78-82. Szerlip NJ, Walbridge S, Yang L, Morris on PF, Degen JW, Jarrell ST, Kouri J, Kerr PB, Kotin R, Oldfield EH, Lons er RR (2007) Real-t ime imaging of convection-enhanced delivery of viruses and virus-sized particles. J Neurosurg 107:560-567. Taymans JM, Vandenberghe LH, Haute CV, Thir y I, Deroose CM, Mortelmans L, Wilson JM, Debyser Z, Baekelandt V (2007) Comparative analysis of adenoassociated viral vector serotypes 1, 2, 5, 7, and 8 in mouse brain. Hum Gene Ther 18:195-206. Thomas CE, Ehrhardt A, Kay MA (2003) Progress and problems with the use of viral vectors for gene therapy. NatRevGenet 4:346-358. Walters RW, Duan D, Engelhardt JF, We lsh MJ (2000) Inco rporation of adenoassociated virus in a calcium phosphate coprecipitate improves gene transfer to airway epithelia in vitro and in vivo. J Virol 74:535-540. Walters RW, Yi SM, Keshav jee S, Brown KE, Welsh MJ, Chiorini JA, Zabner J (2001) Binding of adeno-associated virus type 5 to 2,3-linked sialic acid is required for gene transfer. J Biol Chem 276:20610-20616. Wang J, Xie J, Lu H, Chen L, Hauck B, Samulski RJ, Xiao W (2007) Existence of transient functional doublestranded DNA intermediates during recombinant AAV transduction. Proc Natl Acad Sci U S A 104:13104-13109. Weber M, Rabinowitz J, Prov ost N, Conrath H, Folliot S, Briot D, Cherel Y, Chenuaud P, Samulski J, Moullier P, Rolling F (2003) Recombinant adenoassociated virus serotype 4 medi ates unique and exclusive long-term transduction of retinal pigmented epithel ium in rat, dog, and nonhuman primate after subretinal delivery. Mol Ther 7:774-781.

PAGE 157

149 Wu J, Zhao W, Zhong L, Han Z, Li B, Ma W, Weigel-Kelley KA, Warrington KH, Srivastava A (2007) Self-complement ary recombinant adeno-associated viral vectors: packaging capacity and the role of rep proteins in vector purity. Hum Gene Ther 18:171-182. Wu Z, Asokan A, Samulski RJ (2006) Ad eno-associated virus serotypes: vector toolkit for human gene t herapy. MolTher 14:316-327. Xiao W, Chirmule N, Berta SC, McCu llough B, Gao G, W ilson JM (1999) Gene therapy vectors based on adeno-associated virus type 1. JVirol 73:3994-4003. Yang J, Zhou W, Zhang Y, Zidon T, Ritchie T, Engelhardt JF (1999) Concatamerization of adeno-associated vi rus circular genomes occurs through intermolecular recombination. JVirol 73:9468-9477. Zhong L, Li B, Mah CS, Govindasamy L, Agbandje-McKenna M, Cooper M, Herzog RW, Zolotukhin I, Warrington KH, Jr., Weigel-Van Aken KA, Hobbs JA, Zolotukhin S, Muzyczka N, Srivasta va A (2008) Next generation of adenoassociated virus 2 vectors: point mutations in tyrosines lead to high-efficiency transduction at lower doses. Proc Natl Acad Sci U S A 105:7827-7832. Zincarelli C, Soltys S, R engo G, Rabinowitz JE (2008) A nalysis of AAV serotypes 1-9 mediated gene expression a nd tropism in mice after systemic injection. Mol Ther 16:1073-1080.

PAGE 158

150 CONCLUSIONS Alzheimer’s disease (AD) is the sixth leading cause of deat h in the United States and has been known as one of t he most pervasive and devastating forms of dementia afflicting the aged population worldwide a ccounting for 70% of all dementia cases. Although the pathologic al etiology which causes a slow progressive decline in normal brain function is well characterized, we have yet to identify novel therapeutics that result in the cessation of the disease process and a marked overall significant and longterm improvement in cognitive function. Unfortunately, with the steady increase in life expectancy of our aged population, the number of individuals ov er the age of 65 affected with this disease is rapidly increasing and will continue to increase by approximatly 50% over the next two decades. Current estimations predict that 10 million baby boomers will develop AD in the coming years. AD causes emotional trauma to family members and caregivers in addition to generating a growing economic burden as healthcare costs continue to rise in accordance with the increasing num ber of individuals who suffer from the disease. Finding a novel therapeutic to delay onset of symptoms, improve cognitive function and potentially prevent further mental decline is essential to relieve both ec omonic and emotional burdens which may plague many individuals in the near future.

PAGE 159

151 The pathological disease process of AD that instigates the destruction of neurons in the CNS is thought to ultimately result in the progressive decline in mental function and cognitive abilities that manifest as a general loss of memory that is severe enough to interfere with daily life. This decline in memory can be an early clinical symptom of AD. Bec ause there are several other types of dementia that have overlappi ng symptom patterns, a true diagnostic confirmation of AD is the presence of neuritic plaques composed of beta-amyloid (A ) and intracellular neurofibrillar y tangles composed of hy perphosphorylated tau protein identified in post-mortem brain tissue. The A peptide which was first discovered in 1984 has for decades been identified as the central pathologic al feature from which all other neuropathological insults that occur in AD disease pr ogression stem (Glenner et al., 1984). The Amyloid Cascade Hypothesis, first described by John Hardy and David Allsop circa 1991 suggests that it is the the overproduction and ensuing accumulation of A peptides, resulting from the abber ant processing of APP that is the fundamental event which then dr ives the formation of phosphorylated tau into neurofibrillary tangles which then re sult in disruption and disregulation of synaptic transmission, eventually causing the death of neurons and manifestation of dementia (Hardy and Allsop, 1991), (S elkoe, 1991a, 2001), (Tanzi, 2005). This theory has provided a roadmap of AD pathology for decades and is supported by a number of epidemiologica l genetic studies, molecular in vitro studies as well as in vivo studies using an imal models of amyloid accumulation.

PAGE 160

152 The role of APP protein met abolism in AD pathogenesis has been extensively studied, although it s native biological role is still only speculative. Some studies suggest that APP is involved in synaptic formation and repair as well as neuronal plasticity (Wang et al., 2006), (Priller et al., 2006). Recent studies provide substantial evidence that APP expression is upregulated during neuronal differentiation following neural injury. Additionally, APP has been implicated to participate in cell signal ing, long-term pot entiation, and cell adhesion but only a limited number of st udies have been done in this area to date (Zheng et al., 2006), (Priller et al., 2006). APP undergoes numerous post translation modifications (De Strooper et al., 2000) as well as many types of proteolyic processing. A number of genet ic studies clearly illustrate that mutations in APP can cause the over production of A peptides resulting in an increased ratio of the more fibrillogenic A 1-42/1-40 (Wang et al., 1999), (Ling et al., 2003). The APP gene has been localized to chromosome 21 and thus as a result AD-like neuropathology is consist ently observed in Down’s syndrome or trisomy 21, due to the increased APP expression and thus higher A levels (Glenner and Wong, 1984). Unfortunately, the autosomal dominant gen etic mutations that result in AD (originally identified during the 1990’s) which include the APP mutations and the presenilin mutations (PS1 and PS2), occur infrequently and account for a miniscule percentage of the current popul ation (approximatel y 1-2 %) which suffers from the disease (Selkoe, 1991b), (Nunan and Small, 2002), (Saito et al., 2003). These FAD (familial AD) mutations al so result in very aggressive disease

PAGE 161

153 progression and clinical symptoms can become apparent as early as 30 to 40 years of age but most commonly before t he age of 65 (Selkoe, 1991c). Thus the overwhelming majority of AD cases o ccur in an age dependent manner at a rate that increases exponentially with age. Recently, a more thorough investigation into the cellular processes that mediat e the sporadic late onset AD cases has become a dominant approach in the identification of s pecific targets in the treatment of AD. The identification of the FAD mutations and the role they play in the production of the toxic A peptides (due to abberant changes in APP processing) gave rise to the first generation of A overexpressing mouse models of AD. The PDAPP mouse model of AD was the fi rst transgenic mouse containing the overexpressed mutant human APP and pr ovided a new animal model through which to investigate mechanisms of am yloid deposition and removal (but did not contain tau pathology nor neuron loss) (G ames et al., 1995). Other transgenic mouse models followed includ ing the Tg2576 which was the first mouse model to show learning memory deficits that were closely correlated with age dependent amyloid deposition in a reliable fashi on (Hsiao et al., 1 996), (Morgan et al., 2000), (Ashe, 2006). Additional trans genic mouse models of AD include the double transgenic APP + PS1 mutations t hat harbor an increased aggressive pathology that begins at a younger age t han that of the afor-mentioned single transgenic animals. These mice also s how learning and memo ry deficits at an early age (which again correlated with par anchymal amyloid load) compared to

PAGE 162

154 the mouse models with only the APP mutations (Borchelt et al., 1997), (Citron et al., 1998), (Holcomb et al., 1998), (Gordon et al., 2002). The development of mouse models of A accumulation and deposition paved the way for the development of severa l therapeutic target strategies that have evolved over recent years and ye t unfortunately an effective therapeutic that can permanently reverse cognitive def icits suffered by AD patients remains elusive. Currently, there are two types of medications approved for the treatment of AD which include choli nesterase inhibitors and N –methyl –D aspartate (NMDA) antagonist. Although some patients respond initially to these drugs they do not modify the disease pr ocess, only having a transient moderate effect on clinical symptoms with a host of unappealing side effects. The lack of disease modifying drugs for AD has fueled research efforts focused on development of novel ther apeutics that have the potentia l to halt its progression and improve clinical symptoms. The supposition that A is the primary toxi c pathogenic agent that accumulates subsequently inducing a num ber of other pathogical abnormalities has provided the catalyst driving se veral therapeutic approaches aimed at lowering A levels in the brain to treat AD (Hardy and Allsop, 1991), (Turner et al., 2004). Although A toxicity has been extensively studied and is generally accepted, the specific form of A aggregate that is the pr imary caustic moiety contributing to neuron dysfunction is still debat ed. Initial reports identified fibrillar A as the major toxic species which s ubsequently resulted in decreased synaptic function and neuronal death. Cu rrent reports have refuted these initial findings

PAGE 163

155 revealing that large A deposits or plaques may in fact play a neuroprotective role. Their formation may provide a mechanism in which smaller soluble A moieties are sequestered into larger in soluble aggregates preventing them from causing destruction. Specific ally, the soluble oligomeric A assemblies have been implicated as the driving factor in AD pathogenesis as they exhibit potent toxicity in their capacity to signific antly decrease long term potentiation (LTP), contribute to learning and memory deficits in vivo and induce neuronal cell death in vitro (Lambert et al., 1998), (Nas lund et al., 2000), (Klein, 2002), (Glabe, 2005), (Townsend et al., 2006), (Selkoe, 20 08), (Varvel et al., 2008). Under normal physiological conditions A is constitutively produced and catabolized in the brain which suggests that it may have a role as a physiological metabolite of APP processing. Furthermore, recent studies have shown that picomolar concentrations of A containing both monomers and oligomers cause an increase in hippocampal long term potent iation and improved memory whereas higher concentrations in the nanomolar range, A conversely causes a reduction in potentiation by modifiying the activi ty of voltage-dependent Ca + channels and GTPase activity in neurons (Cirrito et al., 2003), (Puzzo et al., 2008); (Koudinov and Berezov, 2004). These results sugges t the production of low levels of A is a normal physiological process involved in modulating neurotransmission that occurs during learning and memory. Conversely, high levels of A leads to deficits in neurotransmission eventually le ading cognitive deficits manifested as dementia (Cirrito et al., 2003), (Haass and Selkoe, 2007), (Puzzo et al., 2008). The link between A levels and its influence on le arning and memory has been

PAGE 164

156 supported by several in vivo studies in volving APP knockout mice that show impairments in memory and LTP (Seabrook et al., 1999). These studies have provided one focus of disease modification strategies pertaining to drugs that can potentially inhibit A formation, a process requiring and secretase activity (Selkoe, 1991b), (Hardy and Selkoe, 2002). The identification of the essential enzyme BACE 1, a transmembrane aspar tyl protease that is believed to contribute to the majority of -secretase activity, requi red for the production of A in the brain initially provided a novel ther apeutic target for the treatment of AD. Thus finding effective inhibitors of BACE 1 was an early disease modification approach in the development of AD drug ta rgets. The first BACE 1 knockout animal models were initially characteriz ed in 2001, and fortuitously did not show any phenotypic deleterious consequenc e while showing a decrease in A in the brain (Roberds et al., 2001), (Dominquez et al., 2001), (Turner et al., 2004). Unfortunately, BACE 1 inhibitors have s hown little success which may be due to the fact that BACE 1 has recently been found to have several endogenous substrates. Upon further investigat ion it has been revealed that BACE 1 knockout mice have several deleterious phenotypes includi ng premature death, cognitive deficits, and hypermyelination (Dominguez et al., 2001), (Turner et al., 2004). Targeting -secretase activity provided yet another disease modifying therapeutic approach initially receiving a great deal of attention and support in the initial stages of develop ing drug targets for the treat ment of AD. In vivo studies involving -secretase and PS1 knockout models reveal the essential role

PAGE 165

157 it plays Notch signaling ( an important transmembrane pr otein that is essential during development) rendering these anima ls non viable (Shen, 1998). Although -secretase inhibitors can successfully reduce A synthesis, accumulation and deposition, they have also been shown to im prove cognition. Blocking BACE 1 activity also prevents Notch signaling and ch ronic dosing of this class of inhibitors has been shown to have several deleterious effects including changes in the spleen, thymus, and gastrointestinal chan ges (Searfoss et al., 2003), (Milano et al., 2004), (Wong et al., 2004). The expression of both and -secretase is ubiquitous and evidence that they maintain important activity in several essential physiological processes has limited their pot ential as viable therapeutic targets for AD drug development. Another therapeutic strategy that has recently gained more momentum in the AD research community as a possible drug target aimed at modifying the AD disease process, involves modifying mechanisms of A degradation (as opposed to targeting its synthesis through secretas e inhibition). The normal synthesis and catabolism of the A peptide has been demonstrated to be a dynamic process involving several endogenous -amyloid degrading enzymes and include a family of zinc metalloprotease expre ssed in the periphery and in t he brain. In contrast to inherited FAD, the accumulation of A in late onset sporadic AD, has been implicated to be the result of an imbalance between A production and removal. This deficiency in the removal process has been attributed to abnormal accumulation and formation of A deposits in the brain (Sel koe, 2001), (Saito et al., 2003), (Turner et al., 2004). Deficits in the normal A removal is mediated

PAGE 166

158 through several contributing cellular proce sses, including drainage by diffusion of A into the extracellular matrix, low-densit y lipoprotein related receptor protein 1 (LRP-1) mediated transport across blood vesse ls into the circulation acting as a peripheral sink (Sagare et al., 2007) (Deane et al., 2009), and enzymatic degradation through endogenously expressed proteases (Iwata et al., 2001), (Saito et al., 2001), (Sai to et al., 2003). Alt hough a growing number of endogenous proteases have been implicated in several studies as having amyloid degrading capabilities in vitro, only a small mi nority of these have been identified as major contributing factors in the pathological accumulation of A peptides in the brain. Under normal ph ysiologic conditions the catabolism of A is mediated through the activation of neprilysin, endothelin converting enzyme, insulin degrading enzyme, angiotensin converting enzyme and matrix metalloproteases 2, 6 and 9, all of which belong to a family of zinc metalloproteinases (Mentlein et al., 1998), (Carson and Turner, 2002), (Mukherjee and Hersh, 2002), (Saito et al ., 2003), (Turner et al., 2004), (Mouri et al., 2006), (Turner and Nalivaeva, 2007), (Mi ners et al., 2008). More recently, additional proteases that have been show n to have several cleavage locations along the A peptide thus may contribute to maintaining homeostatic A levels, include plasmin, mitochondrial pept idasome (PreP) and cathepsins B and D (Turner and Nalivaeva, 2007). A growing body of evidence from both in vivo and in vitro data, including data presented in papers 1 and 2, support the importance of these proteases in preventing the aberrant accumulation of A in the brain. A series of in vitro

PAGE 167

159 studies first identified insulin degradi ng enzyme and neutral endopeptidase (later recognized as neprilysin) as proteoly tic enzymes capable of cleaving the A peptide at multiple sites. Subsequently a series of studies published in 2001 and 2002 by Saido and Iwata et al clearly established NEP as a major modulator of A metabolism in the brain (Iwata et al., 2000), (Iwata et al., 2001), (Saito et al., 2001), (Iwata et al., 2002). Furthe rmore, post mortem analysis of human AD brain tissue showed a distinct correlation between increases in A deposition and decreases in NEP mRNA levels as well as NEP and IDE protein levels in a region specific manner compared to contro l individuals (Yasojima et al., 2001b), (Yasojima et al., 2001a), (Caccamo et al., 2005). Further investigation into the link between NEP and A under normal physiological conditions has attempted to identify cellular mechanisms that may modify NEP gene expression in the AD brain environment. Pardossi-Piquard et al, in 2005, proposed that NEP gene transcription under normal physiological c onditions can be upregulated by the APP intracellular domain (ACID), the cytosolic fragment generated following cleavage of APP by -secretase (Pardossi-Piquard et al., 2005). These data support the speculation that A production/degradation ma y be closely linked but does not explain decreased NEP expression in the post mortem AD tissue or the net accumulation and deposition of amyloid in late onset AD cases. Interestingly, Jiang and collegues in 2008 provide additi onal molecular mechanisms that contribute to A metabolism, reporting that ApoE a known risk factor in late onset AD, can bind A and promote its degradation (J iang et al., 2008). Once bound to A it is postulated that the ApoE can acts as a chaperone allowing

PAGE 168

160 efflux out of the brain ac ross the blood brain barrier though specific receptors LRP1 or VLDL (Poirier, 2000), (Deane et al., 2004), (Zlokovic et al., 2005), (Deane and Zlokovic, 2007) or facilitate its degradation by NEP and IDE in the late endosome of microglia (Jiang et al., 2008). The efficiency with which ApoE binds soluble A either facilitating or slo wing its degradation, is dependent on the lipidation state and spec ific isoform. Unfortunately the underlying molecular mechanisms modulating this process have yet to be clarified. The overall accumulation and deposition of A particularly in late onset sporadic AD, is likely the result of the additive effects of multip le deleterious events occurring over time which influence production, accumulation and fibrillogenesis, and degradation. These data are supported by a large number of in vivo studies with transgenic animal models to further understand the roles of specific enzymes in the catabolism of A Initial reports of mice deficient of these enzymes, specifically NEP, ECE, and IDE demonstrated significant increases in steady state levels of A in the brain. Eckman and co llegues reported that the dual knockdown of NEP and ECE resulted in an additive cumulative effect on amyloid load in the brain (Eckman et al., 2006). Additional studies using the direct transduction of neurons resulting in the increased expression in a variety of amyloid degrading enzymes including NEP, ECE, IDE, MMP-2 and 9, plasmin and cathepsin B were accompanied by the si gnificant reductions in the overall A in vitro (Marr et al., 2004), (Mueller-Steiner et al., 2006) (Yan et al., 2006). In vivo studies involving gene transfer met hods for the treatment of Alzheimer’s pathology using viral vectors hav e had some success in reducing A deposition

PAGE 169

161 by overexpressing A degrading enzymes, includi ng NEP, ECE and IDE and more recently Cathepsin B, a cysteine prot ease also implicated in the lysosomal degratory pathway of A peptides (Marr et al., 2004) (Mueller-Steiner et al., 2006). The degree to which each enzyme c ontributes to the overall catabolism of A is not entirely delineated. Some r eports identify neprilysin as having the most pronounced effect on plaque load in t he brain, while others identify MMP-9 as the most efficient proteolyt ic enzyme in the degradation of A compared to others, due to its ability to degrade both fibrillar and monomeric forms of A in vitro (Eckman et al., 2003) (Yan et al., 2006), (Farri s et al., 2007), (Hersh and Rodgers, 2008), (Spencer et al., 2008). In paper 1 we investigate the e ffectiveness of overexpessing ECE on amyloid load in the more aggressive model of A deposition double transgenic APP+PS1 mice using rAAV vectors as a method of gene therapy. Recombinant AAV has become widely used for the tr ansduction of neuronal cells. AAV is nonpathogenic, has low immunogenicity, la cks all viral genes and is capable of long term expression in neurons. Thus maki ng rAAV an attractive method for the use as a gene therapy vector for neurologi cal disorders (Burger et al., 2005), (Mandel et al., 2006). Moreover, rAAV is currently being examined in a number of neurological clinical trials (Ma ndel and Burger, 2004). Six weeks post intracranial injection of rAAV-ECEHA se rotype 5, into the right hippocampus and anterior cortical regions, we demonstr ated successful transduction of several different neuronal cell types in the m ouse CNS including CA4 neurons in the hilus and CA3 neurons of the hippocampus pyramidal cell layer (paper 1, figure

PAGE 170

162 3) We also were able to show trans duction of neurons in the dentate molecular layer, as well as a number of round cells possibly olig odendrocytes. The positive transduction of cells in the CNS although co ncentrated at the injection sites was also evident in regions distal from these regions and was detected in the striatum, entorhinal cortex, subiculum, and corpus callosum along the midline of the ipsilateral hemisphere. Additi onally, some positive neurons in the contralateral left hippocampus were also detected. These findings are consistent with Burger and collegues, and are believed to result from retrograde transport of the virus along axonal projections orig inating in the right hippocampus and synapsing in the left hippocampus (Burger et al., 2004). Positi ve transduction of neurons expressing ECE were evidenced by anti HA immunohistochemistry. Not only were we able to show an increase ECE expression following rAAV treatment, but were successfully demonstr ate that the trans duced neurons show a 70% increase in ECE activity com pared to control rAAV-GFP animals. Furthermore, this increase in ECE activity resulted in significant decreases in total amyloid load and congophilic plaque load in these regions (paper 1, figure 5 and 6). Our data are consistent with reports that ECE can degrade A in vitro, and that partial knockdown of the ECE gen e leads to more rapid accumulation of A in transgenic mouse model s (Eckman and Eckman, 2005). Our results from paper 1 showed that modification of ECE expression through the use of rAAV vectors contributes to degradation of A ultimately decreasing both total A and congophilic load in the brain. In paper 2 we investigate whether overex pression of neprilysin will have an equal or greater

PAGE 171

163 effect on A levels. In vivo data demons trating NEP contributes to A degradation, has been proven by results that show an increase in endogenous A levels in a gene dose dependent manner. Decreased NEP protein expression thus contributes to plaque fo rmation and increase CAA levels in NEP knockout animal studies (Farris et al., 2007). Similar results were obtained with NEP specific inhibitors that dramatically decrease NEP activity also resulting in significant increases in endogenous A levels in mice (I wata et al., 2001), (Eckman and Eckman, 2005), (Iwata et al 2001), ( Eckman et al. 2005). Previously reported gene transfer studies de monstrated that tr eatment of amyloid depositing APP transgenic mice using l enti virus and AAV to overexpress NEP could significantly decelerate/decrease am yloid accumulation (Marr et al., 2003), (Iwata et al., 2004); (Iwata et al 2004), (Marr et al 2003). In Paper 2 we investigate both the e ffectiveness of rAAV to deliver NEP to neurons in the mouse brain as well as novel means of increasing dispersion of the virus and gene product to therapeutically relevent regions within the mouse CNS. Specifically, we examine transducti on efficiency and area of distribution of two different NEP constructs (a membrane bound or an engineered secreted form; NEP-n and NEP-s respectively) in the hippocampus and anterior cortex. Conclusively, we compared the effects of overexpression of both NEP enzymes on amyloid burden. Unlike previous gene tr ansfer studies aimed at modification of AD pathology, we administered rAAV by convection enhanced delivery method, initially described by Krauze and co llegues, in an effort to increase the dispersion of the rAAV to larger regions in the brai n parenchyma to improve gene

PAGE 172

164 transduction efficiency and increase levels of NEP protein expression to obtain a larger global effect on total amyloid burden. Six weeks post injections of rAAV vect ors administration into the right hippocampal and anterior cortical regions we were able to obtain successful gene transfer NEP constructs which includ e those previously mentioned, NEP-n and NEP-s as well as an inactive mutant NEP (NEP-m) construct. We were able to transduce a variety of neuronal cell types similar to those reported in paper 1 due to the fact that we used the same AAV serotype 5. Positive NEP expression revealed by anti-HA immunohistochemistr y showed staining concentrated in areas at the site of injection in the hi ppocampus and anterior cortex as well as in sites distal from the injection site. NEP protein expression was evident thoughout the entire hippocam pus and some staining was observed in the entorhinal cortex in the right hemisphere (ipsilateral to the injection site). Miminal staining was also observed in the contrala teral hippocampus of animals receiving the NEP-n and NEP-m rAAV. Staining was also seen in the striatum and corpus callosum and to small extent in the ipsi lateral hemisphere along the midline. The staining for NEP-n and NEP-m was gr eater in intensity covering a larger region both in the hippocampus and anterior cortex compared to the staining for NEP-s (Paper 2, Fig. 2A and 2C). T he NEP-s construct was engineered, through modification of the membrane binding sequence to instead encoding a signal peptide sequence in the hopes of creating a nucleus from which the expressed protein could diffuse to greater regions of the brain; t hus eliminating the requirement for a large number of inje ctions. The NEP-s expression profile

PAGE 173

165 revealed less intensity of HA staini ng when compared to the NEP-m and NEP-n staining patterns. Cells within the hilus of the molecular layer were stained as were as cells in the CA3 region of the hippocampus. Although the NEP-s showed less intense staining this may have been due to the diffusion of the NEP protease throughout the extracellular matrix Diffuse secret ed NEP protein may not be concentrated in a spec ified region, making it more difficult to detect using anti-HA immunohistochemisty com pared to the membrane bound NEP-n protease An alternative explanation for the NEP-s staining pattern may be that once the NEP-s is secreted t he NEP protein or possibly the HA-tag is more prone to degradation by other extracellular pr oteases that is would not otherwise be exposed to upon secretion not necessarily t hat this vector had lower transduction efficiency. Although NEP-s staining wa s significantly less intense (compared to the other constructs), the anterior cortex s howed significantly more intense staining compared to the hippocampus. This resu lt may indicate t hat the si gnal peptide is processed more effectively in the cortex compared to the hippocampus. Despite the differences in staining intensit y of the different NEP proteins, in vitro data confirmed that both NEP-n and NEP-s proteases showed significant levels of enzyme activity in the cell pelle t, while the NEP-m version did not. Additionally, only the media from NEPs transfected cells showed significant enzyme activity. The in vitro activity data was further co rroborated with results from our examination of amyloid burden in APP+PS1 mice following treatment with the rAAV NEP constructs. NEP-n and NEP-s were able to significantly

PAGE 174

166 reduce total A deposition and congophilic deposits in the anterior cortex and hippocampus at the site of inje ction. Examination of A staining in the contralateral hemisphere revealed that NEP-s rAAV significantly decreased total A load in hippocampus and cortex compar ed to the control group. NEP-n treated animals showed significant decreases in total A in the contralateral anterior cortex but not the hippocampus. More, interestingly, examination of congophilic staining on the contralatera l hippocampal side revealed there was approximately 40% and 50% reductions with the NEP-n and NEP-s treatments respectively, but only the reduction wit h NEP-s was statistically significant compared to the control gr oup. No reductions in congophilic staining were observed in the contralateral anterior co rtex with either NEP constructs. The reduction on the contralateral hemis phere observed here is likely due to retrograde transport of the rAAV which has been previous ly reported (Burger et al., 2004; Iwata et al., 2004) and anterograde transport of the NEP protein. No expression was detected in the cont ralateral cortex which may be due a decreased number of neuronal connections between the right and left cortices compared to those in the hippocampus. Th is may explain our results which show a reduction in congophilic staining in the contralateral hippocampus and not the contralateral anterior cortex. We were also able to demonstrate decreases in amyloid load in aged 20 month old APP+PS1 mice following overexpression of NEP-s for five months (injections were administered bilaterally at 15 months). Total A was reduced significantly in the left ant erior cortex as well as in the left and right hippocampus followi ng treatment with the NEP-s vector compared to

PAGE 175

167 control untreated animals. Total congophil ic staining was also significantly reduced in both contralateral and ipsilatera l hemisphere in the hippocampi and cortices following intracranial administrati on of rAAV-NEP-s. Our results reported here in paper 2 are consist ent with data previously repo rted by Iwata et al in 2004 using rAAV to overexpress NEP in young and old APP transgenic mice (Iwata et al., 2004). Our results also rev ealed that we were able to show more substaintial decreases in A load in both the ispilateral and contralateral hemisphere’s following NEP overexpressi on in the more aggressive amyloid depositing double transgenic mouse model in both young and old animals. Additionally, we demonstrate reductions were not limited to soluble forms of A but signifcant reductions in congophili c load as well. An explanation for decreases in congophilic staining ma y be that NEP degrades oligomeric and monomeric soluble A aggregates (Kanemitsu et al., 2003), (Yan et al., 2006), generally believed to be the toxic precur sors to fibrillar insoluble aggregates preventing accumulation rather than ac tively degrading existing deposits. Ultimately, the engineered N EP-s construct was more efficient in decreasing fibrillar A load in the contralateral hemis phere compared to the NEP-n and control vectors, while the NEP-n more efficiently decreased total A load, in the ipsilateral hemisphere. As previously mentioned our findings showed that by implementing the use of a secreted form of the NEP protease in addition to using the CED method for intracranial administr ation, we were able to successfully improve the area of gene expression and further reduce amyloid load in the brain. We speculate that these findings result from the ability of NEP-s to diffuse

PAGE 176

168 readily throughout the extracellular matr ix upon secretion t herefore obtaining access (more readily than membrane bound NEP-n) to both intracellular and extracellular pools of A aggregates. The AAV 5 serotype has previously been shown to effectively transduce non-dividing neuronal cells as well as glia l cells in the mouse brain. Our data support this finding, demonstrating that by using rAAV serotype 5 expressing different human gene constructs of NEP or ECE; we can transduce several different neuronal and glial cell types. The expression profiles for the gene transfer experiments in papers 1 and 2 are al so consistent with the previously published serological specificity of the AAV vectors (Alisky et al., 2000), (Burger et al., 2004), (Choi et al., 2005). Although the transducti on of specific cell types were similar between study animals receiving ECE constructs and NEP constructs the staining patterns within the neurons were differen t. Unlike ECE, staining patterns for NEP revealed less inte nse staining in the cell bodies in the dentate molecular layer compared to t he corresponding axonal projections of these cells which were very darkly stai ned. These results are consistent with data published by Huang et al. and Iwata et al., describing the normal neuronal metabolism of NEP processing first in the endoplasmic reticulum then in golgi apparatus of the cell body and finally trans ported along the axon via vesicles to the synaptic terminal (Iwata et al., 2004); (Iwata et al 2004), (Huang et al., 2006). Although we demonstrate that both ECE and NEP significantly decrease amyloid, due to the differences in injection me thods we cannot directly compare the efficiency by which ECE and NEP decrease A load in the APP+PS1 aminal

PAGE 177

169 model. Despite this caveat, we can conclude that by implementing novel the CED delivery method in addition to modifica tion of the NEP construct directing its secretion allowed significantly increased area of vector distribution thus exposing more cells to viral infection and ultima tely enhancing gene expression levels following a single intracranial injection. Due to the differences in localization of endogenous amyloid degrading enzymes, a list wh ich is continually growing, it is likely that they each contribute to the normal catabolism of A maintaining its basal levels in the brain. In vitro studies have identifie d that each of the proteases have distinctive A cleavage sites that yiel d unique fragments (Iwata et al., 2001)(Yan et al., 2006), (Nalivaeva et al., 2008). Additionally, some proteases, including NEP, IDE, ECE and ACE have been shown to only degrade monomeric soluble A NEP can also degrade toxic oligomer forms and still others including plasmin and MMP-9 have been shown to degrade fibrillar A It is therefore likely that the di fferent proteases play an ad ditive role in maintaining homeostatic endogenous levels of A in the brain degrading different pools of A aggregates ultimately prev enting deposition. Although we have identified rAAV as a potential gene transfer method through which we can upregul ate amyloid degrading en zymes to modify AD pathology in vivo, there are limitations and obstacle that must be overcome to make gene therapy a more attractive ther apeutic approach in the clinic. One of the current limitations of gene therapy is obtaini ng reliable gene transfer techniques that will provide therapeutically relevant levels of gene product. Specifically, with the treatm ent of neurological disor ders, methods involving a

PAGE 178

170 direct intracranial injection are lim ited due to gene transduction and expression being confined to a very small region primar ily at the site of injection. This limitation would require num erous injections to overcome the site specific expression of the gene of interest adding to potential complications. We address this issue in paper 3 through the invest igation of several novel mechanisms aimed at enhancing gene transduction effici ency and improving distribution of viral vectors within the parachyma to increase gene product. Specifically, we investigated the different transduction charac teristics of three different serotypes, AAV5, AAV8, and AAV 9 each of which expr ess GFP (green fluorescent protein) in the mouse CNS. Additionally, we implemented a novel “convection enhanced delivery technique” using a slightly modi fied reflux free cannula design (Krauze et al., 2005a) to dispense AAV directly into the brain parenchyma. Finally, we attempt to optimize the CED injecti on method and distribution parameters by using a reflux free cannula design in addi tion to pre-treatment with hyperosmotic mannitol, an agent that decreases intr acranial pressure, administered by intraperitoneal injection. Previous reports have demonstrated CED, which employs positive pressure and a high flow rate, to dramatically facilitate widespread distribution of AAV within the parachyma following a single intracranial injection (Bankiewicz et al., 2000), (Cunningham et al., 2000). In addition we have modified the CED cannula using a step design (Krauze et al., 2005a) which we have customized to provide a more efficient method to prevent reflux of AAV suspension following adm inistration (paper 3, materials and methods). A comparis on of AAV vectors revealed that each AAV serotype

PAGE 179

171 showed distinctly different transduction patterns with AAV 9 being the most efficient transducer of neurons in seve ral brain regions in the ipsilateral hemisphere as well as the contralateral hemisphere (paper 3; figure 2). AAV 5 was the least efficient and showed the sma llest region of vector distribution and GFP expression, while AAV 8 showed sim ilar patterns of transduction to AAV 9 but showed a decreased amount of GFP ex pression. Our findings as well as others confirm that AAV 9 has signific antly higher transduction efficiency in neurons in the CNS compared to AAV 8 and AAV 5 (Cearley and Wolfe, 2006). The differences in transduction patterns may be a result of different binding affinity to specific cell surface receptors as well as un-packaging and post-entry properties of each serotype (Hauck et al., 2004), (Wang et al., 2007). We speculate that in this study the effici ency of AAV 9 transduction of neurons may be due to its effective and stable binding to cell surface receptors possibly the LamR receptor or other potentia l receptors. It may also be that specific capsid proteins allow AAV 9 to be taken up by the cell more efficiently followed by more efficient translocation and viral uncoati ng and un-packaging within the nucleus of the neuronal cell and efficient DNA rep lication and expression properties. Additionally, anti-GFP immunohistochem istry revealed that AAV 9 treated animals had the most intense darkly st ained pattern of both cell bodies and axonal projections, compared to animals treated with AAV 5 or 8. These results may indictate that cells tranduced with AAV9 may be producing more GFP which is later transported via vesicles within the cell body and along axons to regions that have connectivity with the injecti on site in both the ipsilateral and

PAGE 180

172 contralateral hemispheres. Another possible explanation for the high transduction efficiency of AAV 9 recently proposed by several research groups, may be that more than one AAV 9 virion can transduce a single cell and that once a viral particle gains entry into t he cell it may be easily transported axonally to other regions of the br ain (Passini et al., 2002), (B urger et al., 2004), (Cearley and Wolfe, 2006). Despite differences in vector distribution and GFP expression, all three serotypes appear to primarily transduce neurons, which was proven by co-localization of double labeled NeuN, a neuronal marker, and GFP in the mouse CNS. Double labeling of GFP and GFAP, a marker for astrocytes, showed no co-localization of the two pr oteins indicating that AAV did not tranduce astrocytes. These dat a are consistent with previous reports of AAV 5, 8 and 9, transduction patterns in the murine CNS (Burger et al., 2004), (Davidoff et al., 2005), (Broekman et al., 2006), (C earley and Wolfe, 2006). A clear understanding of AAV characteristics t hat mediate cell ular mechanisms underlying viral infectivity of specific ce ll and tissue type would advance efforts to create tailored recombinant AAV vectors with disease m odifying specificity. Once AAV 9 was identified in our study as having the highest transduction efficiency in the mouse CNS following a si ngle CED intracranial injection, we investigated the effects that mannitol had on vector distribution. We determined that systemic pre-treatment with hyperosmotic mannitol significantly improved vector distribution of AAV. Prev ious reports have demonstrated improved distribution of AAV 2 following direct intr acranial injection into the striatum (Burger et al., 2004) and thalamus (F u et al., 2003) with intravascular

PAGE 181

173 administration of mannitol or coadmini stration with rAAV into the parachyma (Mastakov et al., 2002). The effects of hyperosmotic manni tol in previous studies has been reported to cause an efflux of wa ter out of surrounding tissue or brain cells and increasing the porosity of the extracellular matrix (Rapoport, 2001), (Neeves et al., 2007). Additional ly, in 2002, Sandberg and collegues investigated the affects of mannitol adm inistration on CED delivery in the rat brain stem and did not find significant differences or improved distribution parameters for CED with either systemic administration or co-administration of mannitol. We assert that this ma y have been due to the slow infusion rate, 0.1l/min, or the large molecula size of the molecule deliv ered (FITC-dextran) (Sandberg et al., 2002). In c ontrast, we use a high flow rate (2.5 l/min) to administer very small AAV vector particl es to the hippocampus resulting in widespread distribution of AAV throughout t he entire hippocampal region and to regions distal to the site of injecti on in both the ipsilateral and contralateral hemispheres (paper 3; figures 3 & 4). The efficacy and safety profile of the CED technique has been well established in both animal studies and in human trials (Laske et al., 1997), (Chen et al., 1999), (Bankiewicz et al., 2000). In recent years the CED method has gained acceptance and popularity as an ex tremely effective method to enhance delivery of an expand ing list of different molecu les with wide range molecular sizes including therapeutics agents, proteins, macromolec ules, viral vectors, and nanoparticles in the brain (Cunningham et al., 2008), (Kikuchi et al., 2008), (Perlstein et al., 2008), (Song and Lonser 2008), (Patel et al., 2009). Most

PAGE 182

174 recently, this novel technique has been impl emented in several clinical trials to delivery therapeutics agents for the treatment of Parkinson’s disease and glioblastoma multiform (Lopez et al., 2006), (Vogelbaum, 2007), (Vogelbaum et al., 2007). Despite its clinical effica cy, the parameters under which CED can be optimized providing further distribution are still under investigation. Our data are consistent with previous r eports describing significantly improved distribution of rAAV in the parenchyma. We also demons trate that by comb ining the CED with mannitol further enhanced distribution param eters for CED administration of AAV particles. In addition to CED and mannito l, investigators have used co-infusions of heparin with rAAV 2 (Mastakov et al., 2002) (Nguyen et al., 2001) (Hadaczek et al., 2006) or bFGF (basic fibroblast growth factor) (Hadaczek et al., 2004) to significantly increase viral distribution in the CNS. Presumably, these molecules increase viral spread by preventing or lim iting the interaction between the virion particle and its cellular receptor or co -receptor thus masking the a significant portion of the potential viral receptors, thereby increasin g the number of cells that can potentially be infected by decreasi ng the cells that undergo infection by multiple virion particles (Mastakov et al., 2002). Unfortunately, due to the potential risks of these agent s they may have little clinical appeal but identifying an alternative method of improving viral spread in the CNS by mediating cell surface receptor interactions is a viable approach to improving r AAV efficiency. Overall we provide convincing evidence that A levels can be modified through regulation of endogenous -amyloid degrading enzymes in the brain, which suggests that these enzymes cont ribute to the normal catabolism of A

PAGE 183

175 peptides preventing its accumulation and dep osition. Moreover, we show that significant up-regulation of these enzym es specifically NEP and ECE can be achieved through gene therapy using rAAV vectors that provid e reliable, stable gene expression and viable gene product with significant enzyme activity. Furthermore, we demonstrate the potential for the use of rAAV as a clinical therapeutic in the treatment of AD by pr oviding novel methods that improve viral distribution to larger regions of the brain parenchyma, a major obstacle that must be overcome in the field of gene therapy. We demonstrate that optimal injection techniques in addition to serotype specific AAV characteristics can significantly enhance transduction efficacy and overall dis ease modifying capacity. Further investigation must be done to fully understand and characterize cellular mechanisms underlying viral infectivity of AAV thus unlocking the true potential of their use in the treatment of neurological diseases.

PAGE 184

176 REFERENCES CITED Alisky JM, Hughes SM, Sauter SL, Jo lly D, Dubensky TW, Jr., Staber PD, Chiorini JA, Davidson BL (2000) Transduc tion of murine ce rebellar neurons with recombinant FIV and AAV5 vect ors. Neuroreport 11:2669-2673. Anwar R, Moynihan TP, Ardley H, Brindle N, Coletta PL, Cairns N, Markham AF, Robinson PA (1996) Molecular analysis of the presenilin 1 (S182) gene in "sporadic" cases of Alzheimer's disease: identification and characterisation of unusual splice variants. JNeurochem 66:1774-1777. Bankiewicz KS, Eberling JL, Kohutnicka M, Jagust W, Pivirotto P, Bringas J, Cunningham J, Budinger TF, Harvey -White J (2000) Convection-enhanced delivery of AAV vector in parkinsonian monkeys; in vivo detection of gene expression and restoration of dopaminergic function us ing pro-drug approach. ExpNeurol 164:2-14. Bard F, Barbour R, Cannon C, Carretto R, Fox M, Games D, Guido T, Hoenow K, Hu K, Johnson-Wood K, Khan K, Kholodenko D, Lee C, Lee M, Motter R, Nguyen M, Reed A, Schenk D, Tang P, Vasquez N, Seubert P, Yednock T (2003) Epitope and isotype specificities of antibodies to beta -amyloid peptide for protection against Alzheimer's disease-lik e neuropathology. Proc Natl Acad Sci U S A 100:2023-2028. Bobo RH, Laske DW, Akbasak A, Morrison PF, Dedrick RL, Oldfield EH (1994) Convection-enhanced delivery of macromolecul es in the brain. Proc Natl Acad Sci U S A 91:2076-2080. Borchelt DR, Ratovitski T, van Lare J, Lee MK, Gonzales V, Jenkins NA, Copeland NG, Price DL, Sisodia SS (1997) Accelerated amyloid deposition in the brains of transgenic mice coexpre ssing mutant presenilin 1 and amyloid precursor proteins. Neuron 19:939-945. Bowles DE, Rabinowitz JE, Samulski RJ (2003) Marker rescue of adenoassociated virus (AAV) capsid mutant s: a novel approach for chimeric AAV production. JVirol 77:423-432. Broekman ML, Comer LA, Hyman BT, S ena-Esteves M (2006) Adeno-associated virus vectors serotyped with AAV8 capsid are more efficient than AAV-1 or -2 serotypes for widespread gene deliver y to the neonatal mouse brain. Neuroscience 138:501-510.

PAGE 185

177 Brunkan AL, Goate AM (2005) Presenilin function and gamma-secretase activity. JNeurochem 93:769-792. Burger C, Nash K, Mandel RJ (2005) Recombinant adeno-associated viral vectors in the nervous syst em. HumGene Ther 16:781-791. Burger C, Gorbatyuk OS, Velardo MJ, P eden CS, Williams P, Zolotukhin S, Reier PJ, Mandel RJ, Muzyczka N (2004) Reco mbinant AAV viral vectors pseudotyped with viral capsids from serotypes 1, 2, and 5 display differential efficiency and cell tropism after delivery to different regions of the central nervous system. MolTher 10:302-317. Caccamo A, Oddo S, Sugarman MC, Akbar i Y, LaFerla FM (2005) Ageand region-dependent alterations in Abetadegrading enzymes: implications for Abeta-induced disorders. NeurobiolAging 26:645-654. Calhoun ME, Wiederhold KH, Abramowski D, Phinney AL, Probst A, SturchlerPierrat C, Staufenbiel M, Sommer B, Jucker M (1998) Neuron loss in APP transgenic mice. Nature 395:755-756. Carson JA, Turner AJ (2002) Beta-amyloid catabolism: roles for neprilysin (NEP) and other metallopeptidases? JNeurochem 81:1-8. Cearley CN, Wolfe JH (2006) Transducti on characteristics of adeno-associated virus vectors expressing cap serotypes 7, 8, 9, and Rh10 in the mouse brain. MolTher 13:528-537. Cerpa W, Dinamarca MC, Ines trosa NC (2008) Structure-f unction implications in Alzheimer's disease: effect of Abet a oligomers at central synapses. Curr Alzheimer Res 5:233-243. Chen MY, Lonser RR, Morrison PF, Governal e LS, Oldfield EH (1999) Variables affecting convection-enhanced delivery to the striatum: a system atic examination of rate of infusion, cannula size, in fusate concentration, and tissue-cannula sealing time. J Neurosurg 90:315-320. Choi VW, McCarty DM, Samulski RJ (2005) AAV hybrid serotypes: improved vectors for gene delivery. CurrGene Ther 5:299-310. Cirrito JR, May PC, O'Dell MA, Taylor JW, Pars adanian M, Cramer JW, Audia JE, Nissen JS, Bales KR, Paul SM, DeMa ttos RB, Holtzman DM (2003) In vivo assessment of brain interstitial fluid wit h microdialysis reveals plaque-associated changes in amyloid-beta metabolism and half-life. J Neurosci 23:8844-8853.

PAGE 186

178 Citron M, Eckman CB, Diehl TS, Corcor an C, Ostaszewski BL, Xia W, Levesque G, St George Hyslop P, Younkin SG, Selkoe DJ (1998) Additive effects of PS1 and APP mutations on secretion of t he 42-residue amyloid beta-protein. Neurobiol Dis 5:107-116. Cummings BJ, Pike CJ, Shankle R, Cotman CW (1996a) Beta-amyloid deposition and other measures of neuropathology predict cognitive status in Alzheimer's disease. NeurobiolAging 17:921-933. Cummings BJ, Su JH, Geddes JW, Van Nostrand WE, Wagner SL, Cunningham DD, Cotman CW (1992) Aggregation of t he amyloid precursor protein within degenerating neurons and dystrophic neurit es in Alzheimer's disease. Neuroscience 48:763-777. Cummings BJ, Satou T, Head E, Milgra m NW, Cole GM, Savage MJ, Podlisny MB, Selkoe DJ, Siman R, Greenberg BD, Cotman CW (1996b) Diffuse plaques contain C-terminal A beta 42 and not A beta 40: evidence from cats and dogs. NeurobiolAging 17:653-659. Cunningham J, Oiwa Y, N agy D, Podsakoff G, Colosi P, Bankiewicz KS (2000) Distribution of AAV-TK following intracr anial convection-enhanced delivery into rats. Cell Transplant 9:585-594. Cunningham J, Pivirotto P, Bringas J, Suzuki B, Vijay S, Sanftner L, Kitamura M, Chan C, Bankiewicz KS (2008) Biodistribution of adenoassociated virus type-2 in nonhuman primates after convection-en hanced delivery to brain. Mol Ther 16:1267-1275. Davidoff AM, Gray JT, Ng CY, Zhang Y, Zhou J, Spence Y, Bakar Y, Nathwani AC (2005) Comparison of the abilit y of adeno-associated viral vectors pseudotyped with serotype 2, 5, and 8 capsid proteins to mediate efficient transduction of the liver in murine and nonhuman primate models. Mol Ther 11:875-888. Davidson BL, Stein CS, Heth JA, Martins I, Kotin RM, Derksen TA, Zabner J, Ghodsi A, Chiorini JA (2000) Recombinant adeno-associated virus type 2, 4, and 5 vectors: transduction of variant cell types and regions in the mammalian central nervous system. ProcNatl AcadSciUSA 97:3428-3432. De Strooper B, Umans L, Van Leuven F, Van Den Berghe H (1993) Study of the synthesis and secretion of normal and ar tificial mutants of murine amyloid precursor protein (APP): cleavage of APP occurs in a late compartment of the default secretion pathway. J Cell Biol 121:295-304.

PAGE 187

179 Deane R, Zlokovic BV (2007) Role of the blood-brain barrier in the pathogenesis of Alzheimer's disease. Curr Alzheimer Res 4:191-197. Deane R, Bell RD, Sagare A, Zlokovic BV (2009) Clearance of amyloid-beta peptide across the blood-brain barrier: imp lication for therapies in Alzheimer's disease. CNS Neurol Disord Drug Targets 8:16-30. Deane R, Wu Z, Sagare A, Davis J, Du Y an S, Hamm K, Xu F, Parisi M, LaRue B, Hu HW, Spijkers P, G uo H, Song X, Lenting PJ, Van Nostrand WE, Zlokovic BV (2004) LRP/amyloid beta-pept ide interaction mediates differential brain efflux of Abeta isoforms. Neuron 43:333-344. Dewji NN (2005) The structure and functions of the presenilins Cell MolLife Sci 62:1109-1119. Dewji NN, Singer SJ (1998) Specific inte rcellular binding of the beta-amyloid precursor protein to the presenilins induces intercellular signaling: its significance for Alzheimer's disease. Pr ocNatlAcadSciUSA 95:15055-15060. Dodart JC, Bales KR, Gannon KS, Greene SJ DeMattos RB, Mathis C, DeLong CA, Wu S, Wu X, Holtzman DM, Paul SM (2002) Immunization reverses memory deficits without reducing br ain Abeta burden in Alzheimer's disease model. NatNeurosci 5:452-457. Dominguez DI, De Strooper B, Annaert W (2001) Secretases as therapeutic targets for the treatment of Alzheimer's disease. Amyloid 8:124-142. 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. Eckman EA, Eckman CB (2005) Abeta-degrading enzymes: modulators of Alzheimer's disease pathogenesis and tar gets for therapeutic intervention. BiochemSocTrans 33:1101-1105. Eckman EA, Watson M, Marlow L, Sambamurti K, Eckman CB (2003) Alzheimer's disease beta-amyloid pepti de is increased in mice deficient in endothelin-converting enzyme. JBiolChem 278:2081-2084. Eckman EA, Adams SK, Troendle FJ, St odola BA, Kahn MA, Fauq AH, Xiao HD, Bernstein KE, Eckman CB (2006) Regulatio n of steady-state beta-amyloid levels in the brain by neprilysin and endothelin -converting enzyme but not angiotensinconverting enzyme. JBiolChem 281:30471-30478.

PAGE 188

180 El-Agnaf OM, Mahil DS, Patel BP, Austen BM (2000) Oligomeriz ation and toxicity of beta-amyloid-42 implicated in Alzh eimer's disease. Biochem Biophys Res Commun 273:1003-1007. Farris W, Leissring MA, Hemming ML, C hang AY, Selkoe DJ (2005) Alternative splicing of human insulin-degrading en zyme yields a novel isoform with a decreased ability to degrade insulin and am yloid beta-protein. Biochemistry 44:6513-6525. Farris W, Mansourian S, C hang Y, Lindsley L, Eckman EA, Frosch MP, Eckman CB, Tanzi RE, Selkoe DJ, Guenette S (2003) Insulin-degrading enzyme regulates the levels of insulin, am yloid beta-protein, and the beta-amyloid precursor protein intracellular domain in vivo. Proc Natl Acad Sci U S A 100:4162-4167. Farris W, Schutz SG, Cirrito JR, Shankar GM, Sun X, George A, Leissring MA, Walsh DM, Qiu WQ, Holtzman DM, Selkoe DJ (2007) Loss of neprilysin function promotes amyloid plaque formation and caus es cerebral amyloid angiopathy. Am J Pathol 171:241-251. Fidani L, Goate A (1992) Mutations in APP and their role in beta-amyloid deposition. ProgClinBio lRes 379:195-214.:195-214. Fu H, Muenzer J, Samulski RJ, Breese G, Sifford J, Zeng X, McCarty DM (2003) Self-complementary adeno-asso ciated virus serotype 2 vector: global distribution and broad dispersion of AAV-mediated tr ansgene expression in mouse brain. MolTher 8:911-917. Fulcher IS, Matsas R, Turner AJ, K enny AJ (1982) Kidney neutral endopeptidase and the hydrolysis of enkephalin by synaptic membranes show similar sensitivity to inhibitors. Biochem J 203:519-522. 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. George-Hyslop PH (2000) Genet ic factors in the genesis of Alzheimer's disease. AnnNYAcadSci 924:1-7.:1-7. Glabe CC (2005) Amyloid accumulation and pathogensis of Alzheimer's disease: significance of monomeric, oligomeric and fibrillar Abeta. Subcell Biochem 38:167-177.

PAGE 189

181 Glenner GG, Wong CW, Quaranta V, Eanes ED (1984) The amyloid deposits in Alzheimer's disease: their nature and pathogenesis. Appl Pathol 2:357-369. Goate A, Chartier-Harlin MC, Mullan M, Brow n J, Crawford F, Fi dani L, Giuffra L, Haynes A, Irving N, James L, et al. ( 1991) Segregation of a missense mutation in the amyloid precursor prot ein gene with familial Alzheimer's disease. Nature 349:704-706. Gordon MN, Holcomb LA, Jantzen PT, DiCar lo G, Wilcock D, Boyett KW, Connor K, Melachrino J, O'Callaghan JP, Mo rgan D (2002) Time course of the development of Alzheimer-like pathol ogy in the doubly transgenic PS1+APP mouse. Exp Neurol 173:183-195. Grieger JC, Samulski RJ (2005) Adenoassociated virus as a gene therapy vector: vector development, production and clinical applicatio ns. AdvBiochemEng Biotechnol 99:119-45.:119-145. Haass C, Selkoe DJ (2007) Soluble pr otein oligomers in neurodegeneration: lessons from the Alzheimer's amyloid bet a-peptide. Nat Rev Mol Cell Biol 8:101112. Hadaczek P, Mirek H, Bringas J, C unningham J, Bankiewicz K (2004) Basic fibroblast growth factor enhances transducti on, distribution, and axonal transport of adeno-associated virus type 2 vector in rat brain. Hum Gene Ther 15:469-479. Hadaczek P, Kohutnicka M, Krauze MT, Br ingas J, Pivirotto P, Cunningham J, Bankiewicz K (2006) Convection-enhanced delivery of adeno-associated virus type 2 (AAV2) into the striatum and tr ansport of AAV2 within monkey brain. HumGene Ther 17:291-302. Hardy J, Allsop D (1991) Amyloid deposition as the central event in the aetiology of Alzheimer's disease. Trends PharmacolSci 12:383-388. Hardy J, Selkoe DJ (2002) The amyloid hypothesis of Alzheimer's disease: progress and problems on the road to t herapeutics. Scienc e 19;297:353-356. Hardy JA, Higgins GA (1992) Alzheime r's disease: the amyloid cascade hypothesis. Science 256:184-185. Hauck B, Zhao W, High K, Xiao W (2004) Intracellular viral processing, not single-stranded DNA accumulation, is cruc ial for recombinant adeno-associated virus transduction. J Virol 78:13678-13686. Hersh LB, Rodgers DW (2008) Neprilysin an d amyloid beta peptide degradation. Curr Alzheimer Res 5:225-231.

PAGE 190

182 Higgins GA, Jacobsen H (2003) Transgen ic mouse models of Alzheimer's disease: phenotype and applicatio n. BehavPharmacol 14:419-438. 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. NatMed 4:97-100. Hsiao K, Chapman P, Nilsen S, Eckman C, Harigaya Y, Younkin S, Yang F, Cole G (1996) Correlative memory deficits, Abeta elevation, and amyloid plaques in transgenic mice. Science 274:99-102. Huang SM, Mouri A, Kokubo H, Nakajima R, Suemoto T, Higuchi M, Staufenbiel M, Noda Y, Yamaguchi H, Nabeshima T, Saido TC, Iwata N (2006) Neprilysinsensitive synapse-associated amyloidbeta peptide oligomer s impair neuronal plasticity and cognitive f unction. JBiolChem 281:17941-17951. Hutton M, Hardy J (1997) The preseni lins and Alzheimer's disease. HumMolGenet 6:1639-1646. Hutton M, Busfield F, Wragg M, Crook R, Perez-Tur J, Clark RF, Prihar G, Talbot C, Phillips H, Wright K, Baker M, Lendon C, Duff K, Martinez A, Houlden H, Nichols A, Karran E, Roberts G, Roques P, Rossor M, Venter JC, Adams MD, Cline RT, Phillips CA, Goate A (1996) Comp lete analysis of the presenilin 1 gene in early onset Alzheimer's disease. Neuroreport 7:801-805. Ishii K, Lippa C, Tomiyama T, Miyatake F, Ozawa K, Tamaoka A, Hasegawa T, Fraser PE, Shoji S, Nee LE Pollen DA, St George-Hysl op PH, Ii K, Ohtake T, Kalaria RN, Rossor MN, Lantos PL, Ca irns NJ, Farrer LA, Mori H (2001) Distinguishable effects of presenilin1 and APP717 mutations on amyloid plaque deposition. Neurobiol Aging 22:367-376. Iwata N, Takaki Y, Fukami S, Tsubuk i S, Saido TC (2002) Region-specific reduction of A beta-degradi ng endopeptidase, neprilysin, in mouse hippocampus upon aging. JNeurosc iRes 70:493-500. Iwata N, Tsubuki S, Takaki Y, Shirotani K, Lu B, Gerard NP, Ge rard C, Hama E, Lee HJ, Saido TC (2001) Metabolic regul ation of brain Abeta by neprilysin. Science 292:1550-1552. Iwata N, Mizukami H, Shirotani K, Taka ki Y, Muramatsu S, Lu B, Gerard NP, Gerard C, Ozawa K, Saido TC (2004) Presynaptic localization of neprilysin

PAGE 191

183 contributes to efficient clearance of amyloid-beta peptide in mouse brain. JNeurosci 24:991-998. Iwata N, Tsubuki S, Takaki Y, Watanabe K, Sekiguchi M, Hosoki E, KawashimaMorishima M, Lee HJ, Hama E, Sekine-Aiza wa Y, Saido TC ( 2000) Identification of the major Abeta1-42-degrading catabo lic pathway in br ain parenchyma: suppression leads to biochemical and pathological deposit ion. NatMed 6:143150. Janus C, D'Amelio S, Amit ay O, Chishti MA, Strome R, Fraser P, Carlson GA, Roder JC, St George-Hyslop P, Westaw ay D (2000) Spatial learning in transgenic mice expressing human pres enilin 1 (PS1) transgenes. Neurobiol Aging 21:541-549. Jiang Q, Lee CY, Mandrekar S, Wilkinson B, Cramer P, Zelcer N, Mann K, Lamb B, Willson TM, Collins JL, Richardson JC Smith JD, Comery TA, Riddell D, Holtzman DM, Tontonoz P, Landreth GE ( 2008) ApoE promotes the proteolytic degradation of Abeta. Neuron 58:681-693. Kanemitsu H, Tomiyama T, Mori H ( 2003) Human neprilysin is capable of degrading amyloid beta peptide not only in the monomeric form but also the pathological oligomeric fo rm. NeurosciLett 350:113-116. Kenny AJ, Bowes MA, Gee NS, Matsas R (1985) Endopeptidase-24.11: a cellsurface enzyme for metabolizing regul atory peptides. Biochem Soc Trans 13:293-295. Kikuchi T, Saito R, Sugiyama S, Yamash ita Y, Kumabe T, Kr auze M, Bankiewicz K, Tominaga T (2008) Convection-enhanc ed delivery of polyethylene glycolcoated liposomal doxorubicin: characteri zation and efficacy in rat intracranial glioma models. J Neurosurg 109:867-873. Kimberly WT, Xia W, Rahmati T, Wolfe MS, Selkoe DJ (2000) The transmembrane aspartates in presenili n 1 and 2 are obligatory for gammasecretase activity and amyloid beta-p rotein generation. JBiolChem 275:31733178. Klein WL (2002) Abeta toxicity in Al zheimer's disease: globular oligomers (ADDLs) as new vaccine and drug ta rgets. Neuroche m Int 41:345-352. Koudinov AR, Berezov TT (2004) Alzhei mer's amyloid-beta (A beta) is an essential synaptic protein, not neurotoxic junk. Acta Neurobiol Exp (Wars) 64:7179.

PAGE 192

184 Krauze MT, Saito R, Noble C, Tamas M, Bringas J, Park JW, Berger MS, Bankiewicz K (2005a) Refl ux-free cannula for convection-enhanced high-speed delivery of therapeutic agent s. JNeurosurg 103:923-929. Krauze MT, McKnight TR, Yamashita Y, Bringas J, Noble CO, Saito R, Geletneky K, Forsayeth J, Berger MS, Jackson P, Park JW, Bankiewicz KS (2005b) Real-time visualizati on and characterization of li posomal delivery into the monkey brain by magnetic resonance im aging. Brain ResBrain ResProtoc 16:2026. Kurochkin IV (2001) Insulin-degr ading enzyme: embarking on amyloid destruction. Trends BiochemSci 26:421-425. Kurochkin IV, Goto S (1994) Alzheime r's beta-amyloid pept ide specifically interacts with and is degraded by insuli n degrading enzyme. FEBS Lett 345:3337. Lambert MP, Barlow AK, Chromy BA, Edward s C, Freed R, Liosatos M, Morgan TE, Rozovsky I, Trommer B, Viola KL, Wa ls P, Zhang C, Finch CE, Krafft GA, Klein WL (1998) Diffusible, nonfibrillar li gands derived from Abeta1-42 are potent central nervous system neurotoxins. Proc Natl Acad Sci U S A 95:6448-6453. Laske DW, Youle RJ, Oldfield EH (1997) Tumor regression with regional distribution of the target ed toxin TF-CRM107 in pati ents with malignant brain tumors. Nat Med 3:1362-1368. Leissring MA, Farris W, Chang AY, Walsh DM Wu X, Sun X, Frosch MP, Selkoe DJ (2003) Enhanced proteolysis of bet a-amyloid in APP transgenic mice prevents plaque formation, secondary pat hology, and premature death. Neuron 40:1087-1093. Lemere CA, Lopera F, Kosik KS, Lendon CL, Ossa J, Sa ido TC, Yamaguchi H, Ruiz A, Martinez A, Madrigal L, Hi ncapie L, Arango JC, Anthony DC, Koo EH, Goate AM, Selkoe DJ, Ar ango JC (1996) The E280A presenilin 1 Alzheimer mutation produces increased A beta 42 deposition and severe cerebellar pathology. Nat Med 2:1146-1150. Lesne S, Koh MT, Kotilinek L, Kayed R, Glabe CG, Yang A, Gallagher M, Ashe KH (2006) A specific amyloid-beta protein assembly in the brain impairs memory. Nature 440:352-357. Levy E, Carman MD, Fernandez-Madrid IJ, Power MD, Lieberburg I, van Duinen SG, Bots GT, Luyendijk W, Frangione B ( 1990) Mutation of the Alzheimer's disease amyloid gene in hereditary cer ebral hemorrhage, Dutch type. Science 248:1124-1126.

PAGE 193

185 Ling Y, Morgan K, Kalsheker N (2003) Am yloid precursor protein (APP) and the biology of proteolytic processing: relevanc e to Alzheimer's disease. Int J Biochem Cell Biol 35:1505-1535. Lopez KA, Waziri AE, Canoll PD, Bruce JN (2006) Convection-enhanced delivery in the treatment of malignant glioma. Neurol Res 28:542-548. Malaplate-Armand C, Florent -Bechard S, Youssef I, Koziel V, Sponne I, Kriem B, Leininger-Muller B, Olivier JL, Oster T, Pillot T (2006) Solu ble oligomers of amyloid-beta peptide induce neuronal apoptosis by activating a cPLA2dependent sphingomyelinase-ceramide pathway. Neurobiol Dis 23:178-189. Mandel RJ, Burger C (2004) Clinical trials in neurological disorders using AAV vectors: promises and challe nges. CurrOpinMolTher 6:482-490. Mandel RJ, Manfredsson FP, Foust KD, Risi ng A, Reimsnider S, Nash K, Burger C (2006) Recombinant adeno-associated vi ral vectors as therapeutic agents to treat neurological disorder s. Mol Ther 13:463-483. Marr RA, Rockenstein E, Mukherjee A, Kindy MS, Hersh LB, Gage FH, Verma IM, Masliah E (2003) Neprilysin gene tr ansfer reduces hum an amyloid pathology in transgenic mice. JNeurosci 23:1992-1996. Marr RA, Guan H, Rockenstein E, Kindy M, Gage FH, Verma I, Masliah E, Hersh LB (2004) Neprilysin regulates amyloid Be ta peptide levels. JMolNeurosci 22:511. Masliah E, Hansen L, Adame A, Crews L, Bard F, Lee C, Seubert P, Games D, Kirby L, Schenk D (2005) Abeta vaccinatio n effects on plaque pathology in the absence of encephalitis in Alzheime r disease. Neurology 64:129-131. Mastakov MY, Baer K, Kotin RM, During MJ (2002) Recombinant adenoassociated virus serotypes 2and 5-m ediated gene transfer in the mammalian brain: quantitative analysis of heparin co-infusion. MolTher 5:371-380. Matsas R, Kenny AJ, Turner AJ (1984) The metabolism of neuropeptides. The hydrolysis of peptides, including enkephalin s, tachykinins and their analogues, by endopeptidase-24.11. Bi ochem J 223:433-440. Mentlein R, Ludwig R, Martensen I (1998) Proteolytic degradation of Alzheimer's disease amyloid beta-peptide by a meta lloproteinase from microglia cells. J Neurochem 70:721-726.

PAGE 194

186 Milano J, McKay J, Dagenais C, Foster-B rown L, Pognan F, Gadient R, Jacobs RT, Zacco A, Greenberg B, Ciaccio PJ (2004) Modulation of notch processing by gamma-secretase inhibitors causes intestinal goblet cell metaplasia and induction of genes known to specify gut secretory lineage differentiation. Toxicol Sci 82:341-358. Miners JS, Ashby E, Van Helmond Z, Chal mers KA, Palmer LE, Love S, Kehoe PG (2008) Angiotensin-converting enzym e (ACE) levels and activity in Alzheimer's disease, and relationship of perivascular ACE-1 to cerebral amyloid angiopathy. Neuropathol Appl Neurobiol 34:181-193. Morelli L, Llovera RE, Mathov I, Lue LF, Frangione B, Ghiso J, Castano EM (2004) Insulin-degrading enzym e in brain microvessels: proteolysis of amyloid {beta} vasculotropic variants and reduced acti vity in cerebral amyloid angiopathy. J Biol Chem 279:56004-56013. Morgan D, Diamond DM, Gottsch all PE, Ugen KE, Dickey C, Hardy J, Duff K, Jantzen P, DiCarlo G, Wilcock D, Connor K, Hatcher J, Hope C, Gordon M, Arendash GW (2000) A beta peptide vaccina tion prevents memory loss in an animal model of Alzheimer's disease. Nature 408:982-985. Mori S, Wang L, Takeuchi T, Ka nda T (2004) Two novel adeno-associated viruses from cynomolgus monkey: pseudotyping characterization of capsid protein. Virology 20;330:375-383. Mouri A, Zou LB, Iwata N, Saido TC, Wang D, Wang MW, Noda Y, Nabeshima T (2006) Inhibition of neprilysin by thiorphan (i.c.v.) causes an accumulation of amyloid beta and impairment of learni ng and memory. BehavBrain Res 168:8391. Mueller-Steiner S, Zhou Y, Arai H, R oberson ED, Sun B, Chen J, Wang X, Yu G, Esposito L, Mucke L, Gan L (2006) Antiamyloidogenic and neuroprotective functions of cathepsin B: implications for Alzheimer's disease. Neuron 51:703714. Mukherjee A, Hersh LB (2002) Regulati on of amyloid beta-peptide levels by enzymatic degradation. JA lzheimersDis 4:341-348. Nalivaeva NN, Fisk LR, Belyaev ND, Turner AJ (2008) Amyloid-degrading enzymes as therapeutic targets in Alzheimer's disease. Curr Alzheimer Res 5:212-224. 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 dec line. JAMA 283:1571-1577.

PAGE 195

187 Neeves KB, Sawyer AJ, Foley CP, Saltzman WM, Olbricht WL (2007) Dilation and degradation of the brain extracellular matrix enhances penetration of infused polymer nanoparticles. Br ain Res 1180:121-132. Nguyen JB, Sanchez-Pernaute R, C unningham J, Bankiewicz KS (2001) Convection-enhanced delivery of AAV-2 combined with heparin increases TK gene transfer in the rat brai n. Neuroreport 12:1961-1964. Nicoll JA, Wilkinson D, Holmes C, St eart P, Markham H, Weller RO (2003) Neuropathology of human Alzheimer diseas e after immunization with amyloidbeta peptide: a case report. NatMed 9:448-452. Nilaver G, Muldoon LL, Kroll RA, Pagel MA, Breakefield XO, Davidson BL, Neuwelt EA (1995) Delivery of herpesvirus and adenovirus to nude rat intracerebral tumors after osmotic blood-brai n barrier disruption. ProcNatlAcadSciUSA 92:9829-9833. Nunan J, Small DH (2002) Prot eolytic processing of the amyloid-beta protein precursor of Alzheimer's disease. Essays Biochem 38:37-49. Pardossi-Piquard R, Petit A, Kawarai T, Sunyach C, Alves da Costa C, Vincent B, Ring S, D'Adamio L, Shen J, Muller U, St George Hyslop P, Checler F (2005) Presenilin-dependent transcriptional cont rol of the Abeta-degrading enzyme neprilysin by intracellular domains of betaAPP and APLP. Neuron 46:541-554. Passini MA, Lee EB, Heuer GG, Wolfe JH (2002) Distribution of a lysosomal enzyme in the adult brain by axonal transpor t and by cells of t he rostral migratory stream. J Neurosci 22:6437-6446. Patel MM, Goyal BR, Bhadada SV, Bhatt JS, Amin AF (2009) Getting into the brain: approaches to enhance brain dr ug delivery. CNS Drugs 23:35-58. Perez RG, Soriano S, Hayes JD, Ostasze wski B, Xia W, Selkoe DJ, Chen X, Stokin GB, Koo EH (1999) Mutagenesis iden tifies new signals for beta-amyloid precursor protein endocytosis, tur nover, and the generation of secreted fragments, including Abet a42. JBiolChem 274:18851-18856. Perlstein B, Ram Z, Daniels D, Ocherashv illi A, Roth Y, Ma rgel S, Mardor Y (2008) Convection-enhanced delivery of maghemite nanoparticles: Increased efficacy and MRI monitoring. Neuro Oncol 10:153-161. Poirier J (2000) Apolipoprot ein E and Alzheimer's disease. A role in amyloid catabolism. Ann N Y Acad Sci 924:81-90.

PAGE 196

188 Priller C, Bauer T, Mi tteregger G, Krebs B, Kret zschmar HA, Herms J (2006) Synapse formation and function is modulated by the amyloid precursor protein. J Neurosci 26:7212-7221. Puzzo D, Privitera L, Leznik E, Fa M, Staniszewski A, Palmeri A, Arancio O (2008) Picomolar amyloid-beta positive ly modulates synaptic plasticity and memory in hippocampus. J Neurosci 28:14537-14545. Quon D, Wang Y, Catalano R, Scardina JM, Murakami K, Cordell B (1991) Formation of beta-amyloid protein deposits in brains of transgenic mice. Nature 352:239-241. Rabinowitz JE, Samulski RJ (2000) Buildi ng a better vector: th e manipulation of AAV virions. Virology 20;278:301-308. Rabinowitz JE, Bowles DE, Faust SM, Ledford JG, Cunningham SE, Samulski RJ (2004) Cross-dressing the virion: t he transcapsidation of adeno-associated virus serotypes functionally defin es subgroups. JVirol 78:4421-4432. Rabinowitz JE, Rolling F, Li C, Conrath H, Xiao W, Xiao X, Samulski RJ (2002) Cross-packaging of a single adeno-associ ated virus (AAV) type 2 vector genome into multiple AAV serotypes enables trans duction with broad specificity. JVirol 76:791-801. Raghavan R, Brady ML, Rodriguez-Ponce MI Hartlep A, Pedain C, Sampson JH (2006) Convection-enhanced delivery of therapeutics for brain disease, and its optimization. Neuros urgFocus 20:E12. Rapoport SI (2001) Advances in osmoti c opening of the blood-brain barrier to enhance CNS chemotherapy. ExpertOp inInvestigDrugs 10:1809-1818. Ray WJ, Yao M, Mumm J, Schroeter EH Saftig P, Wolfe M, Selkoe DJ, Kopan R, Goate AM (1999) Cell surface pres enilin-1 participates in the gammasecretase-like proteolysis of Notch. JBiolChem 274:36801-36807. Roberds SL, Anderson J, Basi G, Bi enkowski MJ, Branstetter DG, Chen KS, Freedman SB, Frigon NL, Games D, Hu K, Johnson-Wood K, Kappenman KE, Kawabe TT, Kola I, Kuehn R, Lee M, Liu W, Motter R, Nichols NF, Power M, Robertson DW, Schenk D, Schoor M, Shopp GM, Shuck ME, Sinha S, Svensson KA, Tatsuno G, Tintrup H, Wijsman J, Wright S, McConlogue L (2001) BACE knockout mice are healthy despite lacking the primary beta-secretase activity in brain: implications for Alzheimer' s disease therapeutics. Hum Mol Genet 10:1317-1324.

PAGE 197

189 Roques BP (1993) Zinc metallopeptidases: active site structure and design of selective and mixed inhibitors: new approac hes in the search for analgesics and anti-hypertensives. Biochem So c Trans 21 ( Pt 3):678-685. Sagare A, Deane R, Bell RD, Johnson B, Hamm K, Pendu R, Marky A, Lenting PJ, Wu Z, Zarcone T, Goate A, Mayo K, Perlmutter D, Coma M, Zhong Z, Zlokovic BV (2007) Clearance of amyloi d-beta by circulating lipoprotein receptors. Nat Med 13:1029-1031. Saito T, Kijima H, Kiuchi Y, Isobe Y, Fukushima K (2001) beta-amyloid induces caspase-dependent early neurotoxic chang e in PC12 cells: correlation with H2O2 neurotoxicity. Neurosci Lett 305:61-64. Saito T, Takaki Y, Iwata N, Trojanowski J, Saido TC (2003) Alzheimer's disease, neuropeptides, neuropeptidase, and amyloid-beta peptide metabolism. SciAging KnowledgeEnviron 2003:E1. Samulski RJ, Berns KI, Tan M, Muzyczka N (1982) Cloning of adeno-associated virus into pBR322: rescue of intact viru s from the recombinant plasmid in human cells. Proc Natl Acad Sci U S A 79:2077-2081. Sandberg DI, Edgar MA, Souweidane MM (2002) Effect of hyperosmolar mannitol on convection-enhanced delivery into the rat brain stem. J Neurooncol 58:187-192. 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. Scollay R (2001) Gene therapy: a brief ov erview of the past, present, and future. AnnNYAcadSci 953:26-30.:26-30. Seabrook GR, Smith DW, Bowery BJ, Easter A, Reynolds T, Fitzjohn SM, Morton RA, Zheng H, Dawson GR, Sirinathsinghj i DJ, Davies CH, Collingridge GL, Hill RG (1999) Mechanisms contributing to the deficits in hippocampal synaptic plasticity in mice lacking amyloid prec ursor protein. Neuropharmacology 38:349359. Searfoss GH, Jordan WH, Calligaro DO, Galb reath EJ, Schirtzinger LM, Berridge BR, Gao H, Higgins MA, May PC, Ryan TP (2003) Adipsin, a biomarker of gastrointestinal toxicity mediated by a functional gamma-secretase inhibitor. J Biol Chem 278:46107-46116.

PAGE 198

190 Selkoe DJ (1991a) Alzheimer's disease. In the beginning. Nature 354:432-433. Selkoe DJ (1991b) The molecular pathology of Alzheimer's disease. Neuron 6:487-498. Selkoe DJ (1998) The cell biology of beta-amyloid precursor protein and presenilin in Alzheimer's diseas e. Trends Cell Biol 8:447-453. Selkoe DJ (2001) Alzheimer's disease: genes, proteins, and therapy. Physiol Rev 81:741-766. Selkoe DJ (2008) Soluble ol igomers of the amyloid bet a-protein impair synaptic plasticity and behavior. B ehav Brain Res 192:106-113. Shen MM (1998) Cell signaling in grow th and development: the 11th Annual CABM Symposium, October 8-9, 1997. Biochim Biophys Acta 1377:R55-62. Shimada K, Takahashi M, Turner AJ, Tanzawa K (1996) Rat endothelinconverting enzyme-1 forms a dimer thr ough Cys412 with a similar catalytic mechanism and a distinct substrate bi nding mechanism compared with neutral endopeptidase-24.11. Bi ochemJ 315:863-867. Siman R, Reaume AG, Savage MJ, Trusko S, Lin YG, Scott RW, Flood DG (2000) Presenilin-1 P264L knock-in muta tion: differential effects on abeta production, amyloid deposition, and neuronal vulnerability. J Neurosci 20:87178726. Sinha S, Lieberburg I (1999) Cellular mechanisms of beta-amyloid production and secretion. Proc Natl Acad Sci U S A 96:11049-11053. Skovronsky DM, Doms RW, Lee VM (1998) Detection of a novel intraneuronal pool of insoluble amyloid beta protein that accumulates with time in culture. J Cell Biol 141:1031-1039. Song DK, Lonser RR (2008) Convection-en hanced delivery for the treatment of pediatric neurologic disorder s. J Child Neurol 23:1231-1237. Spencer B, Marr RA, Rockenstein E, Crew s L, Adame A, Potkar R, Patrick C, Gage FH, Verma IM, Masliah E (2008) Long-term neprilysin gene transfer is associated with reduced levels of intr acellular Abeta and behavioral improvement in APP transgenic mice. BMC Neurosci 9:109. Suzuki N, Iwatsubo T, Odaka A, Ishibashi Y, Kitada C, Ihara Y (1994) High tissue content of soluble beta 1-40 is linked to cerebral am yloid angiopathy. AmJPathol 145:452-460.

PAGE 199

191 Tanzi RE (2005) The synaptic Abeta hy pothesis of Alzheimer disease. Nat Neurosci 8:977-979. Tenenbaum L, Darling JL, Hooghe-Pete rs E (1994) Adeno-associated virus (AAV) as a vector for gene transfer into glial cells of the human central nervous system. Gene Ther 1 Suppl 1:S80.:S80. Tenenbaum L, Chtarto A, Lehtonen E, Velu T, Brotchi J, Levivier M (2004a) Recombinant AAV-mediated gene delivery to the central nervous system. JGene Med 6 Suppl 1:S212-22.:S212-S222. Tenenbaum L, Peschanski M, Melas C, Rodesh F, Lehtonen E, Stathopoulos A, Velu T, Brotchi J, Levivier M (2004b) Effi cient early and sustained transduction of human fetal mesencephalon using adeno-asso ciated virus type 2 vectors. Cell Transplant 13:565-571. Thomas CE, Ehrhardt A, Kay MA (2003) Progress and problems with the use of viral vectors for gene therapy. NatRevGenet 4:346-358. Townsend M, Shankar GM, Mehta T, Wa lsh DM, Selkoe DJ (2006) Effects of secreted oligomers of amyl oid beta-protein on hippocampal synaptic plasticity: a potent role for trimers. J Physiol 572:477-492. Turner AJ (2003) Exploring the structur e and function of zinc metallopeptidases: old enzymes and new discoveries. Biochem Soc Trans 31:723-727. Turner AJ, Nalivaeva NN (2007) New insights into the roles of metalloproteinases in neurodegeneration and neurop rotection. Int Rev Neurobiol 82:113-135. Turner AJ, Fisk L, Nalivaeva NN (2004) Targeting amyloid-degrading enzymes as therapeutic strategies in neurodegenerat ion. AnnNYAcadSci 1035:1-20.:1-20. Varvel NH, Bhaskar K, Patil AR, Pimplik ar SW, Herrup K, Lamb BT (2008) Abeta oligomers induce neuronal cell cycle events in Alzheimer's disease. J Neurosci 28:10786-10793. Vogelbaum MA (2007) Convection enhanced delivery for treating brain tumors and selected neurological disorders: symposium review. J Neurooncol 83:97109. Vogelbaum MA, Sampson JH, Kunwar S, Chang SM, Shaffrey M, Asher AL, Lang FF, Croteau D, Parker K, Grahn AY, Sherman JW, Husain SR, Puri RK (2007) Convection-enhanced delivery of ci ntredekin besudotox (interleukin-13PE38QQR) followed by radiation therapy with and without temozolomide in newly

PAGE 200

192 diagnosed malignant gliomas: phase 1 study of final safe ty results. Neurosurgery 61:1031-1037; discu ssion 1037-1038. Wang J, Dickson DW, Trojanowski JQ, Lee VM (1999) The levels of soluble versus insoluble brain Abeta distinguish Alzheimer's disease from normal and pathologic aging. Exp Neurol 158:328-337. Wang J, Xie J, Lu H, Chen L, Hauck B, Samulski RJ, Xiao W (2007) Existence of transient functional doublestranded DNA intermediates during recombinant AAV transduction. Proc Natl Acad Sci U S A 104:13104-13109. Wang R, Zhang YW, Zhang X, Liu R, Zhang X, Hong S, Xia K, Xia J, Zhang Z, Xu H (2006) Transcripti onal regulation of APH-1A and increased gammasecretase cleavage of APP and Notch by HIF-1 and hypoxia. Faseb J 20:12751277. Weber M, Rabinowitz J, Prov ost N, Conrath H, Folliot S, Briot D, Cherel Y, Chenuaud P, Samulski J, Moullier P, Rolling F (2003) Recombinant adenoassociated virus serotype 4 medi ates unique and exclusive long-term transduction of retinal pigmented epithel ium in rat, dog, and nonhuman primate after subretinal delivery. Mol Ther 7:774-781. Westerman MA, Cooper-Blacketer D, Marias h A, Kotilinek L, Kawarabayashi T, Younkin LH, Carlson GA, Younkin SG, As he KH (2002) The relationship between Abeta and memory in the Tg2576 mouse model of Alzheimer's disease. J Neurosci 22:1858-1867. Wong GT, Manfra D, Poul et FM, Zhang Q, Josien H, Bara T, Engstrom L, Pinzon-Ortiz M, Fine JS, Lee HJ, Zhang L, Higgins GA, Parker EM (2004) Chronic treatment with the gamma-secreta se inhibitor LY-411,575 inhibits betaamyloid peptide producti on and alters lymphopoiesis and intestinal cell differentiation. J Bi ol Chem 279:12876-12882. Wu J, Zhao W, Zhong L, Han Z, Li B, Ma W, Weigel-Kelley KA, Warrington KH, Srivastava A (2007) Self-complement ary recombinant adeno-associated viral vectors: packaging capacity and the role of rep proteins in vector purity. Hum Gene Ther 18:171-182. Wu Z, Asokan A, Samulski RJ (2006) Ad eno-associated virus serotypes: vector toolkit for human gene t herapy. MolTher 14:316-327. Xiao W, Chirmule N, Berta SC, McCull ough B, Gao G, Wils on JM (1999) Gene therapy vectors based on adeno-associated virus type 1. JVirol 73:3994-4003.

PAGE 201

193 Yan P, Hu X, Song H, Yin K, Bateman RJ, Cirrito JR, Xiao Q, Hsu FF, Turk JW, Xu J, Hsu CY, Holtzman DM, Lee JM (2006) Matrix metalloproteinase-9 degrades amyloid-beta fibrils in vitro and compact plaques in situ. JBiolChem 281:24566-24574. Yang J, Zhou W, Zhang Y, Zidon T, Ritchie T, Engelhardt JF (1999) Concatamerization of adeno-associated vi rus circular genomes occurs through intermolecular recombination. JVirol 73:9468-9477. Yasojima K, McGeer EG, McGeer PL (2001a) Relationship between beta amyloid peptide generating molecules and neprilysin in Alzheimer disease and normal brain. Brain Res 919:115-121. Yasojima K, Akiyama H, McGeer EG, McGeer PL (2001b) Reduced neprilysin in high plaque areas of Alzhei mer brain: a possible relationship to deficient degradation of beta-amyloid pept ide. NeurosciLett 297:97-100. Zheng H, Koo EH (2006) The amyloid pr ecursor protein: beyond amyloid. Mol Neurodegener 1:5. Zlokovic BV, Deane R, Sallstrom J, Chow N, Miano JM (2005) Neurovascular pathways and Alzheimer amyloid bet a-peptide. Brain Pathol 15:78-83.

PAGE 202

ABOUT THE AUTHOR Nikisha Christine Carty received her Bac helor’s of Arts degree in Biology with a concentration in Neuroscience fr om Cornell University in 2000 and her Masters degree in Medical Sciences from th e University of South Florida in 2005. Before entering the graduate program at USF, Nikisha began working as a research assistant and lab manager in t he department of Pediatric Hemotology at the Weill Cornell Medical College in New York, NY. Looking to gain more experience and knowledge in the field of Neuroscience, in 2002, Nikisha began work as a neuro-imager research assistant at the New York State Psychiatric Institute of Columbia Univer sity. Enthusiastic and enthralled by her research in the field of neuroscience, Nik isha looking to increase her knowledge of molecular basis of neurological disease processe s, entered the Alzheimer’s Disease Research Laboratory under t he mentorship of Dave Morgan, Ph.D., and Marcia Gordon, Ph.D. Nikisha’s research focused on a novel gene therapy approach for the treatment of Alzheimer’s disease ai med at decreasing amyloid burden using adeno-associated viral vectors ov erexpressing endogenous and modified amyloid degrading enzymes. She successfu lly defended her doctoral dissertation May 19, 2009 at the Universi ty of South Florida.


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 2200421Ka 4500
controlfield tag 001 002063793
005 20100419135703.0
007 cr mnu|||uuuuu
008 100317s2009 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0003026
035
(OCoLC)557379496
040
FHM
c FHM
049
FHMM
090
RM101 (Online)
1 100
Carty, Nikisha Christine.
0 245
Recombinant AAV gene therapy and delivery for Alzheimer's disease
h [electronic resource] /
by Nikisha Christine Carty.
260
[Tampa, Fla] :
b University of South Florida,
2009.
500
Title from PDF of title page.
Document formatted into pages; contains 193 pages.
Includes vita.
502
Dissertation (Ph.D.)--University of South Florida, 2009.
504
Includes bibliographical references.
516
Text (Electronic dissertation) in PDF format.
538
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
590
Advisor: David Morgan, Ph.D.
520
ABSTRACT: Alzheimer's disease (AD), first characterized in the early 20th century, is a common form of dementia which can occur as a result of genetic mutations in the genes encoding presenilin 1, presenilin 2, or amyloid precursor protein (APP). These genetic alterations can accelerate the pathological characteristics of AD, including the formation of extracellular neuritic plaques composed of amyloid beta peptides and the formation of intracellular neurofibrillary tangles consisting of hyperphosphorylated tau protein. Ultimately, AD results in gross neuron loss in the brain which is evidenced clinically as a progressive decline in mental capacity. A strong body of scientific evidence has previously demonstrated that the driving factor in the pathogenesis of AD is potentially the accumulation of A peptides in the brain. Thus, reduction of A deposition is a major therapeutic strategy in the treatment of AD.Recently it has been suggested that A accumulation in the brain is modulated, not only by A production, but also by its degradation. Several important studies have demonstrated that A degradation is modulated by several endogenous zinc metalloproteases shown to have amyloid degrading capabilities. These endogenous proteases include neprilysin (NEP), endothelin converting enzyme (ECE), insulin degrading enzyme (IDE) and matrix metalloprotease 9 (MMP9). In this investigation we study the effects of upregulating expression of several of these proteases through administration of recombinant adeno-associated viral vector (rAAV) containing both endogenous and synthetic genes for ECE and NEP on amyloid deposition in amyloid precursor protein (APP) plus presenilin-1 (PS1) transgenic mice. rAAV administration directly into the brain resulted in increased expression of ECE and NEP and a substantial decrease in amyloid pathology.We were able to significantly increase the area of viral distribution by using novel delivery methods resulting in increased gene expression and distribution. These data support great potential of gene therapy as a method of treatment for neurological diseases. Optimization of gene transfer methods aimed at a particular cell type and brain region in the CNS can be accomplished using AAV serotype specificity and novel delivery techniques leading to successful gene transduction thus providing a promising therapeutic avenue through which to treat AD.
2 650
Alzheimer Disease
x metabolism.
Alzheimer Disease
pathology.
Alzheimer Disease
therapy.
Amyloid Precursor Protein Secretases
metabolism.
Blotting, Western.
Dependovirus
genetics.
Gene Therapy.
Gene Transfer Techniques.
Genetic Vectors.
Hippocampus.
Immunoenzyme Techniques.
Presenilin-1
metabolism.
Aspartic Acid Endopeptidases
genetics.
Metalloendopeptidases
genetics.
Senile Plaques.
Mice.
Mice, Transgenic.
653
Beta amyloid
Amyloid degrading enzyme
Convection enhanced delivery
Mannitol
Adeno-associated viral vector
Transgenic mice
690
Dissertations, Academic
z USF
Molecular Pharmacology and Physiology
Doctoral.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.3026