<%BANNER%>

Cellular and Proteolytic Studies of Alzheimer's Disease Amyloid Beta Peptide with Microglia, Stem Cells and Mmp9

Permanent Link: http://ufdc.ufl.edu/UFE0041628/00001

Material Information

Title: Cellular and Proteolytic Studies of Alzheimer's Disease Amyloid Beta Peptide with Microglia, Stem Cells and Mmp9
Physical Description: 1 online resource (125 p.)
Language: english
Creator: Njie, Emalick
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: aging, alzheimer’s, amyloid, anisotropic, brain, cytokine, engraftment, fissure, hippocampus, il6, metalloprotease, metalloproteinase, microglia, nsc, stem, transplant
Neuroscience (IDP) -- Dissertations, Academic -- UF
Genre: Medical Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Alzheimer?s disease (AD) is a common neurodegenerative disease that primarily affects the elderly. In the brains of Alzheimer?s patients, neurons progressively die and synapses withdraw. Behaviorally, Alzheimer?s patients typically present with emotional instability and a marked depreciation in memory. In the spaces in between the cells of AD patients, one finds large aggregates of proteins. The amyloid beta (Abeta) peptide is the primary molecule within these aggregates and thus forms a hallmark pathology in AD. Genetic data from a rare set of families and from those with Down syndrome indicate that producing more Abeta leads to AD. However, the role of these aggregates in the majority of AD patients is largely unknown due to our lack of understanding of how the Abeta molecule is catabolized by the brain. We examined microglia to shed light on one of the brain?s mechanisms for regulating Abeta. We also explored mouse neuronal stem cells (NSCs) as a possible therapeutic intervention to treat Abeta pathology. Microglial cells are typically associated with the removal of extraneous materials from the brain. We find that they do not degrade Abeta. Instead, microglia appear to continually recycle Abeta, perhaps to minimize the pool of Abeta that can form aggregates. Importantly, we find that this recycling of Abeta deteriorates significantly with age. To truly determine whether Abeta is causative of AD, previously existing Abeta must be removed and a clinical improvement observed. Drug treatment regiments to remove Abeta have by and large failed. This is partly because blood vessels block most drugs from entering the brain. Drugs that are directly injected into the brain are typically broken down rapidly. One possible way to circumvent these issues is to transplant cells that continually produce drugs directly within the brain. The neuronal stem cell can live outside the brain for months before being transplanted. We demonstrate that transplants of neuronal stem cells typically settle in predefined regions within the hippocampus and are associated with reductions in Abeta aggregates. We also find that neuronal stem cells can be genetically manipulated to overexpress MMP9, a molecule that may further reduce Abeta aggregates in the brain or protect cells from Abeta toxicity. In mice modeling Alzheimer?s disease, neuronal stem cells formed larger transplants after we genetically manipulated them to express human MMP9. Together, our findings further our understanding of Alzheimer?s disease by demonstrating that microglia are less able to process Abeta with age and that neuronal stem cells may prove useful for treating Abeta pathology in Alzheimer?s patients.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Emalick Njie.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Streit, Wolfgang J.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041628:00001

Permanent Link: http://ufdc.ufl.edu/UFE0041628/00001

Material Information

Title: Cellular and Proteolytic Studies of Alzheimer's Disease Amyloid Beta Peptide with Microglia, Stem Cells and Mmp9
Physical Description: 1 online resource (125 p.)
Language: english
Creator: Njie, Emalick
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: aging, alzheimer’s, amyloid, anisotropic, brain, cytokine, engraftment, fissure, hippocampus, il6, metalloprotease, metalloproteinase, microglia, nsc, stem, transplant
Neuroscience (IDP) -- Dissertations, Academic -- UF
Genre: Medical Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Alzheimer?s disease (AD) is a common neurodegenerative disease that primarily affects the elderly. In the brains of Alzheimer?s patients, neurons progressively die and synapses withdraw. Behaviorally, Alzheimer?s patients typically present with emotional instability and a marked depreciation in memory. In the spaces in between the cells of AD patients, one finds large aggregates of proteins. The amyloid beta (Abeta) peptide is the primary molecule within these aggregates and thus forms a hallmark pathology in AD. Genetic data from a rare set of families and from those with Down syndrome indicate that producing more Abeta leads to AD. However, the role of these aggregates in the majority of AD patients is largely unknown due to our lack of understanding of how the Abeta molecule is catabolized by the brain. We examined microglia to shed light on one of the brain?s mechanisms for regulating Abeta. We also explored mouse neuronal stem cells (NSCs) as a possible therapeutic intervention to treat Abeta pathology. Microglial cells are typically associated with the removal of extraneous materials from the brain. We find that they do not degrade Abeta. Instead, microglia appear to continually recycle Abeta, perhaps to minimize the pool of Abeta that can form aggregates. Importantly, we find that this recycling of Abeta deteriorates significantly with age. To truly determine whether Abeta is causative of AD, previously existing Abeta must be removed and a clinical improvement observed. Drug treatment regiments to remove Abeta have by and large failed. This is partly because blood vessels block most drugs from entering the brain. Drugs that are directly injected into the brain are typically broken down rapidly. One possible way to circumvent these issues is to transplant cells that continually produce drugs directly within the brain. The neuronal stem cell can live outside the brain for months before being transplanted. We demonstrate that transplants of neuronal stem cells typically settle in predefined regions within the hippocampus and are associated with reductions in Abeta aggregates. We also find that neuronal stem cells can be genetically manipulated to overexpress MMP9, a molecule that may further reduce Abeta aggregates in the brain or protect cells from Abeta toxicity. In mice modeling Alzheimer?s disease, neuronal stem cells formed larger transplants after we genetically manipulated them to express human MMP9. Together, our findings further our understanding of Alzheimer?s disease by demonstrating that microglia are less able to process Abeta with age and that neuronal stem cells may prove useful for treating Abeta pathology in Alzheimer?s patients.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Emalick Njie.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Streit, Wolfgang J.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041628:00001


This item has the following downloads:


Full Text





CELLULAR AND PROTEOLYTIC STUDIES OF ALZHEIMER'S DISEASE AMYLOID
BETA PEPTIDE WITH MICROGLIA, STEM CELLS AND MMP9















By

EMALICK GOREE NJIE


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2010

































2010 eMalick Goree Njie




























I dedicate this to the half that's never been told.









ACKNOWLEDGMENTS

I thank Dr. Borchelt and Dr. Streit for giving me this opportunity to develop my

mind. I would like to extend my gratitude to members of the Streit and Borchelt labs for

their support and smiles. I would like to thank my committee for ideas and inspiration

and to my collaborators Svetlana Kantorovich, Dr. Boelen, Dr. Zheng, Dr. Steindler,

and Dr. Rowland. Most of all, I thank my sister, my mother and my extended family.

They have supported me with the strength of a strong village with one son.









TABLE OF CONTENTS

page

A C K N O W LE D G M E N T S ...................................................................................... 4

LIST O F FIG URES ............................................................................ .. ............... 8

LIST O F A B B R EV IA T IO N S .. ................................................................... ............... 10

A BSTRA CT ................................................................................................. ............... 12

CHAPTER

1 IN T R O D U C T IO N .................................................................................................... 1 4

A lzheim er's D disease and A P3 Pathology ............................................... ............... 14
The A m yloid C ascade Hypothesis................................................. ............... 14
M house M odels of A lzheim er's Disease.......................................... ............... 15
Extracellular Trafficking and Internalization of AI ........................................ 17
Lysosomal and Non-lysosomal Degradation of AP3....................................... 19
Consequences of A3 Immunization and Inflammation ................................. 22
Novel Approaches for Study of AP3 Regulation and Therapy .......................... 24

2 EX VIVO CULTURES OF MICROGLIA FROM YOUNG AND AGED RODENT
BRAIN REVEAL AGE-RELATED CHANGES IN MICROGLIAL FUNCTION.......... 26

In tro d u c tio n ......................................................................................................... .. 2 6
M e th o d s ............................................................................................................. 2 7
S o lu tio n s ...................................................................................................... 2 7
A n im a ls .......................................... .. ................... ............ ............. 2 7
Reduction of Debris Produced by Brain Homogenization................................. 28
Preparation of Discontinuous Percoll Gradients........................................... 28
Im m unochem istry ................................................................. .............. 29
C e ll V ia b ility ................................................................................................ 3 0
M icroglial Stim ulation ................................................................... .............. 30
IL-6 E LISA ........................................................................ .......... ............... 31
T N F -a E L IS A ...................................................................... .... ... ........... 3 1
G lutathione M easurem ents ........................................................... ............... 32
A P342 Fate A analysis ..................................................................................... 32
S tatistica l A na lysis .................................................................. .. .......... 34
R e s u lts ............................................................................................................... 3 4
D is c u s s io n ............... ..... .. ... ......................................................................... ... 4 1
Im provem ents on M icroglial Isolation ............................................ ............... 41
Age-related Changes in Microglial Cytokine Release................... ............... 42
Implications of Age-related Changes in Microglial Cytokine Release............. 43
Age-related Changes in Microglial Glutathione Levels................................. 44
Age-related Changes in Microglial Processing of AP3.................................... 45









Interpretation of AP3 Expulsion by Younger Microglia.................................... 46
C including C om m ents ................................................................. .. .......... 47

3 ENGRAFTMENT PATTERNS OF NSCS TRANSPLANTED INTO MOUSE
MODELS OF ALZHEIMER'S DISEASE.............................................................. 52

In tro d u c tio n ......................................................................................................... .. 5 2
M e th o d s ............................................................................................................. 5 3
Isolation of NSCs................. ..... ....... .. ... ...... ............... 53
Transplantation into Amyloid Beta AD Mice ................................................. 54
Immunochemistry ........... .. ........................ .......................... 56
Modeling Paths Of Least Resistance (PLR) ................................................. 56
R e s u lts ............................................................................................................... 5 7
D discussion ...................................... ............................ ... ............... ........... 66
Paths of Least Resistance versus migration................................................. 68
Research and Clinical Relevance of Paths of Least Resistance .................... 71
C including C om m ents ................................................................. .. .......... 73

4 THE EFFECT OF NEURONAL STEM CELLS ON AP3 PATHOLOGY AND
THEIR UTILITY AS A THERAPEUTIC DELIVERY VEHICLE FOR THE AP3
DEG RA D ING PROTEASE, M M P9 ...................................................... ............... 77

In tro d u c tio n ......................................................................................................... .. 7 7
M e th o d s ............................................................................................................. 7 8
Le ntivirus C o nstructio n ................................................................. .......... 78
Lentivirus Transduction ................................................................ .............. 79
Fluorescence A activated C ell Sorting.............................................. ............... 79
Isolation of NSCs................ ....... ............... ............... 79
Transplantation into Amyloid Beta AD Mice ................................................. 79
Im m uno che m istry ..................................................... .................... ........... 8 0
A analysis of A P3 Plaque N um ber..................... ...................... ............... 81
Reverse Transcriptase Polymerase Chain Reaction (RT-PCR) of Human
M M P 9 ......................................... ............................................... ............. 8 1
In Vitro M M P G elatinase A activity ................................................... ............... 82
In Situ M M P G elatinase A activity .................................................... ............... 82
Chemical Activation of secreted MMP9........................................................ 83
R e s u lts ............................................................................................................... 8 3
D is c u s s io n ........................................................................................................... ... 9 2
E ndogenous M M P A ctivity............................................................. .............. 94
AP3 plaque Burden is Lowered Following NSC Transplantation...................... 94
NSCs as a Platform to Deliver MMP9 and Other Candidate Therapeutics In
V iv o ............... .......................................... ... ............................................... ... 9 5
Characteristics of NSCs overexpressing transgenes ................................... 96
C including C om m ents ................................................................. .. .......... 99

5 C O N C LU S IO N S ........................................................................................... 104



6









LIST O F R E FE R E N C ES ......................................................................... ............... 107

BIO G RAPH ICAL SKETCH ..................................................................................... 125









LIST OF FIGURES

Figure page

2-1 Dispase II density centrifugation methodology. ............................................ 36

2-2 Purity, yield and viability of m icroglia ............................................ ............... 38

2-3 A dult m icroglial m orphology ........................................................... ............... 39

2-4 Neonatal microglial morphology ................................................................... 42

2-5 Microglial reaction to immunostimulation ..................................................... 45

2-6 Cytokine secretion of young and aged microglia .......................................... 47

2-7 M icroglial glutathione content ....................................................... ............... 48

2-8 Species of A342 used for microglial experiments......................................... 49

2-9 Fate of AP3 internalized by microglia.............................................................. 50

2-10 Overview of mixed glial culture (MGC) and density centrifugation
methodologies utilized to obtain microglia ................................................... 51

3-1 M orphology of in vitro N S C s .......................................................... ............... 57

3-2 In vitro characteristics of N S C s...................................................... ............... 59

3-3 MMP9 associated changes in engraftment................................................... 60

3-4 Survival and distribution of transplanted NSCs............................................. 63

3-5 Modeling paths of least resistance ............................................................... 64

3-6 Immediate distribution of NSCs in paths of least resistance............................... 67

3-7 Engraftment patterns of NSCs deposited at the ventral border of the
h ip p o ca m p u s ..................................................................................................... 6 9

3-8 lbal expression and lack of A3 migration by engrafted NSCs......................... 74

3-9 GFAP expression by engrafted NSCs .......................................................... 75

3-10 P aths of least resistance .................................... ........................ .............. 76

4-1 Transplantation of NSCs is associated with reduced amyloid burden ............. 84

4-2 Fate of A3 internalized by NSCs .................................................................. 85









4-3 NSCs express endogenous mouse MMP9.................................... ............... 89

4-4 Endogenous MMP activity in mice with AP3 plaque burden. ............................. 91

4-5 Genetic modification of NSCs for MMP9 overexpression.............................. 93

4-6 Cell type differences in transduction efficiency .............................. ............... 96

4-7 Secreted MMP9 has zymogen activity and can undergo autoactivation............. 97

4-8 Enrichm ent of NSC cultures. ...... .......... ........... ..................... 100

4-9 NSC enrichment is associated with rate of MMP9 secretion. ......................... 101

4-10 NSC overexpression and activation of MMP9 in vivo................................... 102

4-11 Transplantation of MMP9 NSCs and GFP NSCs results in similar reductions
in am yloid burden ................................................................. ........... 103











1,10 PNTL

AD

AP3

AMPA

AU

CMV

CNS

EDTA

EF1

FACS

GFP

GSH

IL-6

LPS

mRNA

MGC

MMP9

NSC

PBS

PLR

ROS

SDS

SEZ

SGV


LIST OF ABBREVIATIONS

1, 10 Phenanthroline

Alzheimer's disease

Amyloid beta

p-aminophenylmercuric acetate

Arbitrary unit

Cytomegalovirus

Central nervous system

Ethylenediaminetetraacetic acid

Elongation factor 1

Fluorescence activated cell sorting

Green fluorescent protein

Glutathione

Interleukin 6

Lippopolysaccharide

messenger ribonucleic acid

Mixed glial culture

metalloprotease 9

mouse neuronal stem cell

Phosphate buffered saline

Path of least resistance

Reactive oxygen species

Sodium dodecyl sulfate

Subependymal zone

Subgranular zone









Tumor necrosis factor-a


TNF-a









Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

CELLULAR AND PROTEOLYTIC STUDIES OF ALZHEIMER'S DISEASE AMYLOID
BETA PEPTIDE WITH MICROGLIA, STEM CELLS AND MMP9
By

eMalick Goree Njie

August 2010

Chair: Wolfgang J. Streit
Major: Medical Sciences-Neuroscience

Alzheimer's disease (AD) is a common neurodegenerative disease that primarily

affects the elderly. In the brains of Alzheimer's patients, neurons progressively die and

synapses withdraw. Behaviorally, Alzheimer's patients typically present with emotional

instability and a marked depreciation in memory. In the spaces in between the cells of

AD patients, one finds large aggregates of proteins. The amyloid beta (A3) peptide is

the primary molecule within these aggregates and thus forms a hallmark pathology in

AD. Genetic data from a rare set of families and from those with Down syndrome

indicate that producing more A3 leads to AD. However, the role of these aggregates in

the majority of AD patients is largely unknown due to our lack of understanding of how

the A3 molecule is catabolized by the brain. We examined microglia to shed light on

one of the brain's mechanisms for regulating A3. We also explored mouse neuronal

stem cells (NSCs) as a possible therapeutic intervention to treat A3 pathology.

Microglial cells are typically associated with the removal of extraneous materials from

the brain. We find that they do not degrade A3. Instead, microglia appear to continually

recycle A3, perhaps to minimize the pool of A3 that can form aggregates. Importantly,

we find that this recycling of A3 deteriorates significantly with age. To truly determine









whether AP3 is causative of AD, previously existing AP3 must be removed and a clinical

improvement observed. Drug treatment regiments to remove AP3 have by and large

failed. This is partly because blood vessels block most drugs from entering the brain.

Drugs that are directly injected into the brain are typically broken down rapidly. One

possible way to circumvent these issues is to transplant cells that continually produce

drugs directly within the brain. The neuronal stem cell can live outside the brain for

months before being transplanted. We demonstrate that transplants of neuronal stem

cells typically settle in predefined regions within the hippocampus and are associated

with reductions in AP3 aggregates. We also find that neuronal stem cells can be

genetically manipulated to overexpress MMP9, a molecule that may further reduce AP3

aggregates in the brain or protect cells from AP3 toxicity. In mice modeling Alzheimer's

disease, neuronal stem cells formed larger transplants after we genetically manipulated

them to express human MMP9. Together, our findings further our understanding of

Alzheimer's disease by demonstrating that microglia are less able to process AP3 with

age and that neuronal stem cells may prove useful for treating AP3 pathology in

Alzheimer's patients.









CHAPTER 1
INTRODUCTION

Alzheimer's Disease and Ap Pathology

Alzheimer's disease is a neurodegenerative disease that affects more people than

all other neurodegenerative diseases (Association, 2007). Half of all persons reaching

the age of 85 will be diagnosed with AD (Association, 2007). Temporally, every 72

seconds someone within the U.S is diagnosed with AD. The mean life span of patients

following diagnosis ranges from 3 years to 9 years (Brookmeyer, et al., 2002). The 150

billion dollars spent annually in healthcare costs for these individuals will only grow as

our population's overall life expectancy increases (Association, 2007). Patients with this

disease have noticeably enlarged ventricles and severe memory deficits as a result of

progressive neuronal cell death. Patients inevitably deteriorate over a period of years to

the point where they require constant assistance to manage even the most mundane of

life's activities. Although currently available therapies can slow the progression of AD

by as much as 5 years, they do not remedy the underlying cause of neurodegeneration

(Auld, et al., 2002, Parsons, et al., 2007, Terry and Buccafusco, 2003, Zandi, et al.,

2004).

The Amyloid Cascade Hypothesis

In the aged and AD brain, levels of AP3 peptide have been shown to increase

(Armstrong, et al., 1996). It has been postulated that this increase in extracellular AP3,

ranging from 50x to 1500x above normal (Farris, et al., 2007), is neurotoxic and sets the

stage for the neurodegeneration found in AD (Lewis, et al., 2001, Walsh, et al., 2002).

This belief is based on the amyloid cascade theory. The theory states that the presence

of high amyloid burden (amyloidosis), in the form of extracellular AP3 plaques or









oligomers initiates a cascade that leads to the disruption of cytoskeletal tau protein

(Tanzi and Bertram, 2005). AP3 and tau are among a class of fibril forming proteins

whose P3-spines naturally interdigitate to form highly stable dry 'steric zippers' (Sawaya,

et al., 2007). Two other notable proteins that multimerize via a dry steric zipper

mechanism are the prion protein of Bovine Spongiform Encephalitis (Mad Cow Disease)

and a-synuclein of Parkinson's disease (Sawaya, et al., 2007). Family linked mutations

that cause AD create more total AP3 or create more aggregation prone forms of AP3

(Goate, et al., 1991, Kumar-Singh, et al., 2006, Motte and Williams, 1989, Rovelet-

Lecrux, et al., 2006). This supports the position of A3 as the initiator of a neurotoxic

cascade. Secondarily, murine models that are genetically engineered to overexpress AP3

have memory and learning impairments similar to that which occurs in humans (Higgins

and Jacobsen, 2003, Hsiao, et al., 1996, Savonenko, et al., 2005, Westerman, et al.,

2002, Wong, et al., 2002).

Mouse Models of Alzheimer's Disease

Given the limitations associated with post-mortem studies on human brains,

since the mid-nineties several transgenic mouse lines have been created to model

Alzheimer's disease. These mouse models emulate with great success the deposition of

AP3 in the neocortex and the hippocampus, but not necessarily neurodegenerative

changes, such as neurofibrillary tangles. Transgenic mice typically overexpress the

amyloid precursor protein (APP) with familial AD mutations under the control of various

promoters, and although there are more than a dozen strains available, most studies

are done on the PDAPP, Tg2576, APP23 and most recently, the mo/hu

APPswe/PS1dE9 mice (Borchelt, et al., 2002, Games, et al., 1995, Hsiao, et al., 1996,

Sturchler-Pierrat, et al., 1997). The mutations carried in the hAPP transgene introduced









into these mice originated from genetic studies done on families with early onset AD. In

general, the transcription of these genes causes dense AI plaque to be deposited in the

neocortex and the hippocampus usually between six and twelve months of age. The

hAPP transgene in the PDAPP model has the V717F mutation (Indiana family origin)

(Games, et al., 1995). The dual point mutations in Tg2576, APP23 mice are at

K670N/M671 L (Swedish family origin). In addition to the Swedish mutation, mo/hu

APPswe/PS1dE9 co-expresses Presenilin 1 with a familial AD mutation (accelerates

AB42 deposition) together with the humanized form of mouse APP. Eponymous

PDAPP mice have hAPP under the control of platelet derived growth factor promoter

(PD). In the Tg2576 and mo/hu APPswe/PS1dE9 models, the APP gene is driven by

the prion protein promoter, while the APP23 model has hAPP driven by the neuron

specific Thy-1 promoter. These animals constitutively over express the APP transgene.

Recently, a inducible Tet-off mo/huAPPswe/ind transgenic mouse model was generated

using a tetracycline responsive promoter (Jankowsky, et al., 2005).

The various mouse models have demonstrable impairment in learning and

memory that typically manifest at around the same time AP3 plaques deposits become

prevalent in the limbic structures of these animals (Higgins and Jacobsen, 2003, Hsiao,

et al., 1996, Westerman, et al., 2002). Mice exhibit loss of dendritic spines and loss of

synapses in subcortical cholinergic projections, AP3 plaque associated gliosis, and

cerebrovascular abnormalities (Beckmann, et al., 2003, German, et al., 2003, Moolman,

et al., 2004, Stalder, et al., 1999, Wegiel, et al., 2004). APP23 mice exhibit

neurodegeneration (Sturchler-Pierrat, et al., 1997) while the AP3 plaques in Tet-

off/APPswe/ind mice persist for the lifetime the mouse even when APP transgene









production is halted early on (Jankowsky, et al., 2005). However, the lifespan of the

Tet-off/APPswe/ind mouse is not significantly shortened by the presence of these AP3

plaque deposits. This highlights an inherent shortcoming of modeling AD in animals

whose normal lifespan is less than 5% that of a normal human. Mouse models also do

not have hyperphosphorylated tau neurofibrillar tangles or the same level of

complement system activation found in AD (Higgins and Jacobsen, 2003, Schwab, et

al., 2004, Xu, et al., 2002). Rat models of AD have not gained widespread popularity

because rat brains are more resistant to the formation of dense A9I plaques and AD-like

learning & memory deficits (Ruiz-Opazo, et al., 2004).

The transgenic mice listed above have been mated to other mice deficient or

containing mutant proteins that are informative for AD studies. For instance, hybrid

mice were created by mating Tg2576 mice to mice deficient in expression of the

immune cell chemotaxis receptor CCR2 (El Khoury, et al., 2007). As described in the

following, these mice proved instrumental in further understanding microglial

contribution to the central nervous system (CNS) challenged with amyloidosis.

Extracellular Trafficking and Internalization of Aft

Though not a complete replicate of Alzheimer's disease, transgenic mouse

models are nonetheless useful in understanding the genetics and biochemical cascades

that lead to learning and memory deficits found in humans. As mentioned, the amyloid

AP3 plaques in these models attract glia (Stalder, et al., 1999). As the brain's

endogenous immunocompetent cells microglia are among the first cells recruited to the

AP3 plaques (El Khoury, et al., 2007, Frautschy, et al., 1998, Stalder, et al., 1999). The

AP3-protein from these plaques has been found inside microglial lysosomes indicating

that these cells actively phagocytose portions of AP3 plaques (Cole, et al., 1999,









Frautschy, et al., 1998). The process of internalizing AI is mediated by several cells

surface receptors. In the brain parenchyma, macrophage scavenger receptor Type A

(MSR-A) is only expressed by microglia. Studies by Chung and coworkers, and others

have shown that MSR-A is responsible for uptake of up to 60% of internalized fibrillar

AI9 (non-opsonized) in the brain (Chung, et al., 2001, El Khoury, et al., 1996).

Interestingly, MSR-A knock-out animals expressing hAPP with Indiana and Swedish

mutations under PD control have similar amounts of AP3 plaque burden when compared

to their littermates with normal MSR-A expression (Huang, et al., 1999). Other

receptors, such as MSR-B, and receptor for advanced glycation end products (RAGE)

are capable of internalizing AP3, and it is likely that these compensate for the loss of

MSR-A (Huang, et al., 1999, Rogers, et al., 2002, Yan, et al., 1996).

Another way in which AP3 can be internalized is as a non-covalent conjugate to

complement factors or antibodies (opsonization). Microglia have an assortment of

receptors such as Cdl 1 b and Fc gamma receptors which can mediate the phagocytosis

of opsonized AP3 (Chung, et al., 2001, Lue and Walker, 2002). A recent study showed

that APP is transported to cholesterol rich lipid rafts in neurons by low density

lipoprotein receptors like protein (LRP) (Yoon, et al., 2007), however, its AP3 cleavage

product can be carried by high density lipoprotein like protein (HDL) in the extracellular

space and then internalized by microglial LRP (Fagan, et al., 1996). Apolipoproteins E

(ApoE) and J (ApoJ) in complex with HDL-AP3 reduce the eventual degradation of AP3 in

microglia (Cole, et al., 1999).

Allelic differences in ApoE, along with mutations in APP and PS1/2 are among

the most well defined genetic risk factors for familial AD, and it is interesting that









microglia & astrocytes are the major contributors of extracellular ApoE in the brain (Xu,

et al., 2000). Release of ApoE into the extracellular space is dependent on protein

prenylation and is sensitive to station treatment (Naidu, et al., 2002). Micromolar

concentrations of A3 can induce the secretion of ApoE from microglia in vitro (Bales, et

al., 2000). Conversely, the fibrillization of AP3 is thought to be promoted by ApoE since

the ApoE can bind to AP3 (Carter, et al., 2001, Xu, et al., 2000) and mice with the Indiana

or Swedish mutations that have ApoE knocked out no longer have dense AP3 plaques or

have delayed deposition of A3 plaques, respectively (Fryer, et al., 2005, Irizarry, et al.,

2000).

Lysosomal and Non-lysosomal Degradation of Ap

Frautschy et al. quantified up to five-fold increases in microglial density

surrounding AP3 plaques in mice with the Swedish mutation under the prion promoter

(Frautschy, et al., 1998). El Khoury et al. demonstrated that elimination of CCR2

dependent microglial chemotaxis results in earlier appearance of amyloidosis, twice as

much AP342 and ~36% greater mortality in mice co-expressing the prion promoter

Swedish mutation (El Khoury, et al., 2007). A plethora of reports have provided

evidence of mouse and human microglial degradation of AP3 via the endosomal-

lysosomal pathway (Frautschy, et al., 1998, Qiu, et al., 1998, Rogers, et al., 2002).

However the rate and quantity of this degradation is a subject of great concern as the

kinetics of degradation (in relation to AP3 deposition) has direct physiological relevance

to Alzheimer's disease progression. Several articles from Maxfield's group have shown

that microglial cells from neonatal mice degrade AP3 at much slower rates compared to

blood macrophages (Majumdar, et al., 2007a, Paresce, et al., 1997). In an elegant

series of experiments, they showed that in the course of three days, neonatal microglia









in vitro degrade only 20% of the fibrillar A3 they are exposed to while peritoneal

macrophages degrade close to 80%. While microglia and macrophages are able to

make similar cleavages at the N-terminus of the fibrillar A3, the macrophages were able

to make far more thorough cuts along the A3 molecule. Neither cell type was able to

cut the C-terminal portion of the molecule. Perhaps the localization of A3's highly stable

twisted beta pleated sheet at the C-terminus confers this resistance to degradation.

This pleated sheet is the fundamental secondary structural element underlying

multimers of A3 (Sawaya, et al., 2007). The authors proposed that microglia are

hindered in their capacity to degrade A3, relative to their macrophage counterparts, due

to an incomplete set of lysosomal enzymes (Majumdar, et al., 2007a). This is indirectly

supported by the observation that microglial degradation of A3 in vitro is enhanced

when global endocytosis of lysosomal enzymes is enhanced (Majumdar, et al., 2007a).

It is worth mentioning that a recent study focusing on macrophages from the blood of

AD patients concluded that these cells are impaired in their ability to phagocytose A3

when compared to non-diseased subjects (Fiala, et al., 2005).

TGF-31 is a cytokine that attracts and activates microglia. Bigenic mice

overexpressing hAPP with Swedish and Indiana mutations under PD promoter in

addition to TGF-31 (cSJL x B6D2 background) have a 50% reduction in brain

parenchymal A3 plaque burden compared to their non- TGF-31 overexpressing

littermates (Wyss-Coray, et al., 2001). Since microglia exposed to TGF-31 in vitro

display enhanced degradation of A3, it is postulated that the marked in vivo reduction of

A3 plaque burden in these TGF-13 transgenic mice is due to microglial action. TGF-31

and similar factors that stimulate microglia most likely cause degradation of A3 through









either intracellular lysosomal degradation at the N-terminus, as already described, or

through extracellular degradation via cell surface enzymes.

Neprilysin, a zinc dependant endopeptidase and insulin degrading enzyme (IDE),

a zinc dependant metalloproteinase, are two well defined brain proteases. It is thought

that microglia express these proteases on the cell surface and also release them into

the extracellular space (Qiu, et al., 1998, Takaki, et al., 2000). A series of reports have a

built a compelling case for neprilysin being the major soluble AP3 catabolic enzyme in AD

animal models, in AD patients and in non-diseased humans (Iwata, et al., 2001,

Leissring, et al., 2003). Of note are reports quantifying 50% reductions in neprilysin

mRNA in hippocampal regions classically susceptible to amyloidosis (Yasojima, et al.,

2001) and ~48% less mouse brain neprilysin mRNA in Tg2576 mice deficient in

microglial recruitment to AP3 plaques due to CCR2 knockout (El Khoury, et al., 2007).

As mentioned previously, these CCR2 knockout mice have twice as much AP342. Even

though neprilysin is found in neurons, its major degradative function in the brain

parenchyma appears to be microglial based. On the other hand, recent studies show

that neprilysin cannot degrade the fibrillar AP3 commonly found in dense plaques (Yan,

et al., 2006) or especially neurotoxic oligomeric AP3 (EI-Amouri, et al., 2007). Thus,

neprilysin activity can modulate normal AP3 catabolism in vivo (Marr, et al., 2003) and

prevent the onset or progression of AD, however, its specificity for less multimeric forms

of AP3 will likely prevent its use as a therapy for reversing the course of symptomatic AD.

The CCR2 knockout mice mentioned (El Khoury, et al., 2007) have normal IDE

mRNA levels. This suggests that neurons and astrocytes are able to supply basal

levels of IDE when microglial function is perturbed (El Khoury, et al., 2007). IDE is









present in the cytosol where it degrades the cytoplasmic portion of APP (Edbauer, et al.,

2002), however cell surface IDE and secreted IDE are more likely the species of this

protein responsible for extracellular AP3 clearance (Qiu, et al., 1998). In three month-old

mice lacking IDE (and not expressing APP transgenes), a 64% increase in cerebral

endogenous AP340 has been found (Farris, et al., 2003). However in 16-month-old mice

with the Swedish mutation under the prion promoter, astrocytes proximal to AP3 plaques

display a two-fold increase in IDE immunoreactivity at the same time point diffuse AP3

condenses into plaques (Leal, et al., 2006). Surprisingly, there was no reported

reduction in AP3 plaque deposition (Leal, et al., 2006). In these studies, microglia

surrounding the AP3 plaques do not produce IDE at levels detectable by

immunohistochemistry.

IDE is one of the leading drug candidates for AD therapy. A recent study

suggested that chemical modifications in its active site for the purpose of keeping the

enzyme in a constitutively open state should be pursued (Shen, et al., 2006). This

could result in a forty fold increase in catalytic activity and hence a therapeutic increase

in A9I degradation. One must be prudent however as insulin is a major substrate for

IDE degradation. Unlike other tissues, the brain does not maintain energy reserves so

perturbations in sugar homeostasis caused by constitutively active IDE could result in

severe side effects. This is less of an issue with neprilysin as a therapy since its other

proteolytic substrates lie mostly outside the CNS compartment.

Consequences of AP Immunization and Inflammation

The field of AD research has experienced a significant disappointment following

the termination of the AP342 immunization (AN-1792) clinical trial due to life threatening

inflammatory side effects (Patton, et al., 2006). In contemplating the cause of









meningoencephalitis which afflicted 6% of the study's subjects, one must consider the

role of microglial interaction with A3 as a possible activating agent (Floden, et al., 2005,

Tan, et al., 1999). This is thought to be mediated via the binding of complement factor

C1 conjugated with A3 to complement receptors CR3 & CR4 on the microglial cell

surface (Heneka and O'Banion, 2007) inducing a highly cytotoxic complement cascade.

It is worth noting that 20% of the subjects in the trial had the desired antibody response

to A3 immunization. However, in this subgroup, 22% suffered from

meningoencephalitis. This suggests that the pathogenic inflammation that halted the

study likely involved antibody-induced inflammation. Perhaps the antibody response

changed the A3 to a species that is more inflammatory?

In vitro studies have shown that A3 can induce synthesis of inflammatory

cytokines via a NfKB dependant pathway (Heneka and O'Banion, 2007). Floden and

colleagues reported that there are age related differences in the ability of different forms

of A3 to induce inflammatory responses (Floden and Combs, 2006). Microglia isolated

from neonatal and adult mice (C57BL/6 non-transgenics) are able to induce the

secretion of TNFa when exposed to oligomeric A3 while fibrillar A3 can only induce

TNFa production when exposed to neonatal microglia. If this finding holds true in

humans, it could perhaps lend greater understanding to the AN-1792 trial in the gray

and white matter of AD patients with an antibody response to the A342 immunization,

post-mortem analyses intriguingly found dramatic increases in soluble A3 as a result of

antibody-dependent plaque disassembly. In fact, the quantity of soluble A3 increased

fifteen-fold in one subject. This antibody-dependent disassembly of fibrillar A3 into

soluble A3 caused an unintended increase in oligomeric species of A3. The









investigators found AP3 plaque-derived dimers, trimers, tetramers and higher order

oligomeric structures of up to 30kDa in the brains of these patients. As described by

Floden et al., oligomeric but not fibrillar species of AP3 can selectively induce adult

microglial production of proinflammatory TNFa (Floden and Combs, 2006) and directly

cause neuronal death (Floden, et al., 2005, Tan, et al., 1999).

Novel Approaches for Study of Ap Regulation and Therapy

As described above, studies on the catabolism of A3 have shed light on the brain's

mechanisms of regulating extracellular A3. Studies from neuroimmunology labs have

demonstrated that microglia, the primary immune cell of the CNS (Giulian, 1987), have

a considerable role in AP3 regulation in the AD brain (Chung, et al., 2001, El Khoury, et

al., 2007, Frautschy, et al., 1998, Rogers, et al., 2002). However, clinical trials aimed at

emulating microglial-like functions (anti-oxidant trials) or curtailing microglial functions

(anti-inflammatory trials) have largely failed at treating AD. Interestingly, these trials,

which targeted a disease of the elderly, were conspicuously carried out on knowledge

primarily from studies on in vitro neonatal mouse models. Nevertheless, it is generally

accepted that glia play a role in the regulation of the extracellular space including the

metabolism of extracellular A3.

The emerging field of stem cell research has demonstrated that the neuronal stem

cell of the subependymal zone (SEZ) is a very glial-like immature cell that may provide

a window for novel approaches to treat AD pathology (Kukekov, et al., 1997, Laywell, et

al., 2000, Raponi, et al., 2007). Our goal was to bring together the knowledge base of

neuroimmunology and stem cell biology in order to answer two questions: 1) is there a

loss of microglial AP3 catabolic functionality in the aged brain and 2) can stem cells be

used as a therapeutic approach to lessen AP3 burden in brains where AP3 homeostasis is









perturbed? Our hypotheses was that 1) there is a depreciation of AP3 degradation in the

aged and AD brain and 2) the biology of stem cells uniquely positions them as a

platform to counter AP3 burden. These hypotheses were founded on the following

observations. First, in both the aged and AD brain, there are greater levels of A3

(Armstrong, et al., 1996). This increase in AP3 cannot be fully explained by currently

known genetic risk factors since these factors confer increased AD susceptibility by

affecting AP3 anabolism. Because most AD cases are 'late onset' without any known

genetic causes, the possibility of dysregulation of A3 catabolism in AD remains

(Association, 2007). Second, histological evidence from human and rodent autopsy

brain sections show that microglia appear dystrophic with similar morphological features

in both the aged and AD brains (Flanary, et al., 2007). Thus, lose of microglial function

could contribute to increased AP3 burden with age and AD. Third, cell replacement

therapy using stem cells is a promising approach because of the physiological flexibility

inherent to stem cells. These cells naturally undergo self renewal in vivo and in vitro

and are capable of being transplanted after being expanded ex vivo (Marshall, et al.,

2006, Walton, et al., 2006a). This unique property allows for genetic introduction of

candidate anti-AP3 molecules in culture for delivery in vivo.

In the following, I shall detail our experimental findings in quantifying glial aged-

related biology as well as determining the AP3 disrupting ability of 1) transplanted stem

cells and 2) stem cells used to deliver the candidate anti-AP3 therapeutic, MMP9. Our

studies have implications in understanding catabolic pathways that contribute to

Alzheimer's disease pathology and clinical approaches to treating said pathology.









CHAPTER 2
EX VIVO CULTURES OF MICROGLIA FROM YOUNG AND AGED RODENT BRAIN
REVEAL AGE-RELATED CHANGES IN MICROGLIAL FUNCTION

Introduction

A multitude of studies have implicated microglia as important players in the

etiology of a number of age-related neurodegenerative diseases, including Alzheimer's

disease, Parkinson's disease and amyotrophic lateral sclerosis (Boillee, et al., 2006,

Chung, et al., 2001, El Khoury, et al., 2007, Frautschy, et al., 1998, Rogers, et al.,

2002). To understand how microglial cell function may change with aging, various

protocols have been developed to isolate microglia from the young and aged central

nervous system (CNS). While histological studies are essential in providing clues

regarding the cells' involvement, they are limited in terms of evaluating the functions of

living cells. In the past decade, protocols to isolate living microglia from postnatal

animals have become available (Carson, et al., 1998, de Haas, et al., 2007, Frank, et

al., 2006, Hickman, et al., 2008, Ponomarev, et al., 2005). These protocols either trap

microglia using antibodies to cell-specific antigens (Hickman, et al., 2008, Tham, et al.,

2003) or separate microglia using density centrifugation (de Haas, et al., 2007, Frank, et

al., 2006). In both cases, the rapid isolation of microglia enables ex vivo

experimentation of endogenous microglia in a controlled setting largely devoid of

neurons, oligodendrocytes and astrocytes.

Protocols utilizing density centrifugation are advantageous to those utilizing

antigen traps in terms of yield per brain (de Haas, et al., 2007). They also avoid artificial

cellular reactions caused by antigen cross linking, a risk carried with the use of

antibodies in trapping protocols. However, in our hands, significant amounts of non-

microglial, debris contaminate current density centrifugation derived cultures. In the









present study, we sought to modify density centrifugation methodology to eliminate

debris fields present in such cultures. With these modifications, microglial yields were

preserved or slightly increased.

These improvements allowed us to study microglial function with regard to

alterations during normal aging. We found that microglia from aged mice constitutively

secrete greater amounts of interleukin-6 (IL-6) and tumor necrosis factor-a (TNF-a)

relative to microglia from younger mice and are less responsive to stimulation. Also,

microglia from aged mice have reduced glutathione levels and internalize less AP3 while

microglia from mice of all ages do not retain the AP3 peptide for a significant length of

time. These studies offer further support for the idea that microglial cell function

changes with aging. They suggest that microglial AP3 phagocytosis results in AP3

redistribution rather than biophysical degradation in vivo and thereby provide

mechanistic insight to the lack of amyloid burden elimination by parenchymal microglia

in aged adults and those suffering from Alzheimer's disease.

Methods

Solutions

Dispase II (Roche, Mannheim, Germany) was reconstituted at 2U/mL in dispase

buffer (0.9% HEPES, 50mM NaCI, pH 7.4) according to manufacturer's instructions.

Percoll (GE Healthcare, St. Giles, UK) was diluted 1:10 with 10x phosphate-buffered

saline (PBS) to create an isotonic solution. 1x PBS was added to isotonic percoll to

create working solutions ranging from 75% to 25% percoll.

Animals

Debris reduction experiments were performed with non-transgenic C57BL/6 mice

and mice expressing GFP under the fractalkine-receptor promoter (Jung, et al., 2000).









Experiments were performed using young (1-2 month old) and aged (14-16 month old)

male C57BL/6 mice. The mice were housed at 22C in a controlled 12hr light/dark cycle

and provided food and water ad libitum. Animals were euthanized by exsanguination

using transcardiac perfusion with PBS under deep anesthesia with sodium pentobarbital

(50mg/kg body weight). This method of euthanasia is consistent with the

recommendations of the Panel of Euthanasia of the American Veterinary Medical

Association. After perfusion, the brain (telencephalon, cerebellum and midbrain) was

rapidly removed.

Reduction of Debris Produced by Brain Homogenization

Each brain was washed in cold 1x PBS, then minced using a small scissors. Brain

tissue was gently dissociated by immersion into 10mLs (per brain) of dispase II solution

(2U/mL), trypsin solution (0.05%) or by grinding within a tissue homogenizer (glass

Potter, Braun, Melsungen, Germany). Dissociated brain tissue was placed within a

50mL conical tube and laid horizontally in an orbital shaker set to shake for 1hr, 37C at

150rpm. Remnant tissue chunks were further homogenized by rapidly triturating with a

10mL pipette (BD Biosciences, Boston, MA) with a wide bore to prevent cell shearing.

This was carried out with a fully charged pipette aid (Drummond). Enzyme activity was

halted by diluting the resultant homogenate 1:1 with cold 10% FBS in 1x PBS.

Meninges and clumped cells were removed with 70um filtration (BD Biosciences,

Boston, MA) to obtain a suspension of single cells.

Preparation of Discontinuous Percoll Gradients

The homogenate was spun 1000 x g for 10min at 4C. The supernatant was

discarded and the pellet of an individual brain was resuspended in 6mLs of 75%

isotonic percoll (high percoll) (GE Healthcare, Buckinghamshire, U.K). Three mLs of this









mixture was then aliquoted into a 15mL polystyrene tube. Five mLs of 35% isotonic

percoll (low percoll) was layered atop the high concentration percoll at a rate of

150ul/sec to create a distinct interface between the percoll layers. To replicate

gradients described in the literature, 25% percoll was utilized for low percoll. 1x PBS

was layered atop the low concentration percoll. The resultant discontinuous gradient

was then allowed to settle on ice for 15 minutes allowing most of the homogenate to

naturally rise towards its isopycnic position. The gradient was then centrifuged at 800 x

g for 45min in a HS-4 swinging bucket rotor (Thermo Fisher Scientific, Waltham, MA)

set to 40C. We did not notice changes in microglial yields with high acceleration or the

application of the brake. However, yields were significantly diminished if the gradients

are not processed immediately following centrifugation. To process the gradients, the

volume of the PBS layer and the low concentration percoll layer were rapidly aspirated.

A band of microglia (usually 0.5-1.5mL), captured between the low concentration and

high concentration percoll layers was then collected and diluted in 50mL of 1x PBS.

This was centrifuged at 1000 x g for 10min at full acceleration and brake. The

supernatant was quickly decanted and the cell pellet resuspended in DMEM culture

media containing 10% FBS. We also added 0.15ug/mL granulocyte monocyte colony

stimulating factor (GM-CSF, R & D Systems, Minneapolis, MN), authouth this is not

required for the culturing of microglia.

Immunochemistry

Isolated cells were grown in culture media overnight. Cells were then washed,

fixed in 4% paraformaldehyde and processed for immunofluorescence of microglial

antigen lbal (1:500, Wako, Richmond, VA), microglial antigen Cd11b (1:1000, Serotec,

Raleigh, NC), astrocyte antigen GFAP (1:1000, Dako Corporation, Carpinteria, CA), and









neuronal antigen NeuN (1:500, Millipore, Bellirica, MA). Cells were rinsed and

incubated with goat anti-rabbit Alexa 488 (Invitrogen, Carlsbard, CA) and goat anti-

mouse Alexa 568 (Invitrogen, Carlsbard, CA). Cells were photographed with an

Olympus DP71 camera mounted on an Olympus BX60 microscope.

Cell Viability

Microglial mitochondrial respiratory activity, a measure of cell viability, was determined

using a colorimetric MTT (methylthiazolyldiphenyl-tetrazolium bromide) assay (Bioassay

Systems, Hayward, CA). This was compared to a reference value of HEK-293 cells, a

highly viable immortal cell line, and dying cultures treated with 1% Triton X-100, a toxic

reagent.

Microglial Stimulation

In order to compare the inflammatory reaction of microglia in young and aged

brains, cells were isolated from 2 and 14 months old mice, as described above, and

seeded in 96-well tissue culture plates (Corning Incorporated, Corning, NY) at a density

of 3 x 105 cells/well. The cultures were incubated overnight at 370C with 5% CO2 and

saturated humidity. The next day, cells were stimulated by replacing the original culture

media with media containing 2% FBS and inflammatory agents in different

concentrations. Two highly potent inflammatory stimuli were selected, i.e.

lipopolysaccharide (LPS), (Escherichia coli 055:B5) (Sigma, St. Louis, MO), a toll-like

receptor 4 (TLR4) agonist and PamCSK3 (Invitrogen, Carlsbard, CA), a TLR2 agonist.

LPS and PamCSK3 were added at a concentration of 10-100ng/mL (LPS) or 0.1-

1ug/mL (PamCSK3). Control conditions were included, containing no stimuli. After

24hrs of incubation, the media of stimulated microglia were collected and centrifugated









for 10min at 20C and 1200rpm. The supernatants were used for IL-6 and TNF-a

enzyme-linked immunosorbent assays (ELISA). For every condition, cytokine levels

were calculated in three different wells, while each experiment was performed fourfold.

IL-6 ELISA

Mouse IL-6 secreted protein levels were determined with a general sandwich

ELISA protocol. Briefly, an enhanced protein binding ELISA plate (Nunc, Rochester,

NY) was incubated overnight at 4C with the capture antibody, rat anti-mouse IL-6 (BD

Bioscience Erembodegem, Belgium). After blocking the non-specific binding for 2hrs,

standards (BD Bioscience Erembodegem, Belgium) and samples were added for 2hrs

at room temperature. Subsequently, biotinylated rat anti-mouse IL-6 (BD Bioscience

Erembodegem, Belgium) was used as a detection antibody. Following incubation with a

Streptavidin-Horseradish Peroxidase conjugate (Dako Cytomation, Heverlee,

Belgium), a TMB substrate (BD Bioscience Erembodegem, Belgium) was applied to the

plate. Finally, optical densities (OD) were read between 450-570nm, using a

spectrophotometer (Powerwave X Select) and concentrations were calculated. The

detection limit of the assay was 10pg/mL.

TNF-a ELISA

Mouse TNF-a secreted protein levels were measured using a commercially

available ELISA kit (eBioscience, San Diego, CA), according to the manufacturer's

instructions. Concentrations were determined according to the OD values, measured

using a spectrophotometer (Powerwave X Select) at a wavelength between 450-570nm.

The detection limit of the assay was 8pg/mL.









Glutathione Measurements

Total glutathione (reduced and oxidized) was measured in microglia using a

glutathione reductase enzymatic recycling assay (Cayman Chemical, Ann Arbor, MI)

that is based on the colorimetric conversion of nitrobenzoic acid to 5-thio-2-nitrobenzoic

acid (Tietze, 1969). Briefly, microglia from the brains of young or aged mice were lysed

immediately following isolation and prepared for glutathione measurements according to

manufacturer's instructions. Glutathione levels within the range of standards were

attained by combining microglia from four brains. Therefore to attain three repetitions,

12 mice per age group were assayed. All samples were normalized to total protein

using bicinchonic acid (BCA) colorimetric assay (Pierce, Rockford, IL).

Ap42 Fate Analysis

AP342 lyophilized protein with and without a FITC conjugate (rPeptide, Bogard, GA)

was resuspended to 1mg/mL in 1% NH40H and stored at -200C according to

manufacturer's directions. To visualize internalized AP3 in living cells, AP342-FITC was

diluted to 4ug/mL in DMEM and added to microglia for 3hrs. The cells were then

stained with DAPI nuclear counterstain (1:1000) for 5min. and then imaged. For

enhanced subcellular resolution of internalized AP3, cells (exposed to non-conjugated

AP342) were fixed as described above for immunocytochemistry using antibodies

against AP3 (6E10, 1:2000, Signet Laboratories, Dedham, MA) and the lysosomal

associated protein, Lamp 1 (1:2, gift from Dr. Notterpek, University of Florida). To

determine the aggregation state of A342 in stock solutions used in internalization and

fate experiments, samples were diluted in Laemmli sample buffer containing 2% sodium

dodecyl sulfate (SDS) and loaded in 4-20% TG-SDS gels (Invitrogen, Carlsbad, CA) for

standard SDS-PAGE. Immunoblots were probed with 6E10 at a dilution of 1:5000.









6E10 has affinity to individual AP342 peptides and therefore monomers, oligomers and

higher order conformers of AP342 are distinguishable by size difference. Gel blots were

photographed using a Fugi imaging system (Fugifilm Life Science, Stamford, CT).

RS chambers (Nunc, Roskilde, Denmark) that contain a hybrid of glass and

polystyrene surfaces have reduced non-specific interaction with AP3 and were therefore

chosen for fate analysis experimentation. To further reduce non-specific AP342

absorption, these chambers were blocked with 10% milk for 1hr. Microglial cells were

then isolated from 1 month old mice, 15 months old mice and mixed glial cultures

(MGC) were seeded on RS chambers at approximately 3 x 105 cells/chamber. The

cultures were incubated overnight at 370C with 5% CO2 and saturated humidity. The

next day, cultures were rinsed and given AP342 diluted in DMEM at 4ug/mL. The cells

were allowed to internalize AP342 from the media for 3hrs in 37C. The cells were then

rinsed and incubated at 37C with 1.5mLs of culture media lacking AP342. This media

and that of cells lysed immediately following rinsing were collected, as were the

conditioned media and lysate from wells incubated for 3hrs and 16hrs. The lysis buffer

consisted of NP40 (Invitrogen, Carlsbard, CA) supplemented with protease inhibitor

cocktail (1x, Sigma, St. Louis, MO) and PMSF (1mM). For each age group, lysate and

media representing 3-6 adult mice or 14 neonatal pups (2 (MGCs) were collected. To

determine the fate of A342, we employed a sandwich-style ELISA (Invitrogen,

Carlsbard, CA), configured with two capture antibodies (recognizing epitopes on the N

terminus and C terminus of human AP342) to first capture the N terminus of AP342 and

then the C terminus (Schmidt, et al., 2005). Microglial degradative activity on AP342

causes N terminal truncations (Majumdar, et al., 2007b), thus ELISA reactivity is limited









to non-degraded AP342. We were therefore able to 1) follow the loss of AP342 and 2) the

expulsion of AP342 from microglia. The ELISA was processed with duplicates of each

sample and absorbance read at 450nm using a spectrophotometer (Bio-Tek, Winooski,

VT). Data was normalized to mock treated wells that were treated as described above

but contained no cells. The detection limit of the assay was 10pg/mL.

Statistical Analysis

Average cytokine data are presented as mean SEM. Statistical analysis was

carried out using SPSS ver. 14.0 for Windows (SPSS, Chicago, IL). To analyse

differences between groups, we used unpaired, two-tailed Student's t-test or ANOVA

with a post hoc Bonferonni's test when appropriate. A p-value of <0.05 was considered

statistically significant. Average AP3 data are presented as mean SEM. For statistical

comparison of A3 internalization between age groups, we used unpaired, two-tailed

Student's t-test. For statistical comparison of A3 fate, paired, two-tailed Student's t-test

was used to compare AP3 levels immediately following internalization and 16hrs later. A

p-value of <0.05 was considered statistically significant.

Results

As reported previously (de Haas, et al., 2007, Frank, et al., 2006), centrifugation of

dissociated whole brain within discontinuous percoll gradients can separate microglia

from other brain cells. In our hands, the techniques described in the literature yielded

insufficiently pure cultures for the pulse-chase experiments we performed (Fig. 2-1 B-C).

Specifically, we observed that debris fields, which could possibly sequester AP3 peptide

(Li, et al., 2005). To address this, we utilized dispase II, an enzyme that has been

described as particularly gentle, yet capable of tissue dissociation (Borchelt, et al.,

1992, Gao, et al., 2004, McDermott, et al., 2003). Furthermore, we increased the









density of percoll by 16% from that described in the literature (de Haas, et al., 2007).

We observed that brains treated with this methodology 1) had greater separation of

dissociated microglia from tissue chunks (Fig. 2-1A-B), 2) yielded numerous adherent

microglia and 3) were largely devoid of debris (Fig. 2-1 C-D).

Altogether, the combination of these techniques resulted in the extraction of up to

3 x 106 microglia per brain. On average, 8.5 x 105 microglia per brain were extracted

from young and aged mice (Fig. 2-2B). 94% of DAPI counterstained cells were reactive

to lbal (Fig. 2-2A) as determined by 3 observers. The cells had a characteristic

amoeboid, phase bright morphology similar to previous reports of adult microglia

isolated with different methodology (Tham, et al., 2003). To further confirm that our

isolated cells were indeed microglia, we isolated cells from transgenic mice where GFP

expression is under the fractalkine receptor promoter. These mice are reported to have

microglia as the only brain cell type to express GFP (Jung, et al., 2000). Upon isolation

of cells, our cultures were reactive to antibodies specific to GFP (Fig. 2-3A). Ex vivo

cultures of adult microglia that were allowed to adhere overnight were comparable to

the HEK 293 cell line in viability (Fig. 2-2C). Recent studies have raised the possibility

that GM-CSF could push cultured microglia towards a dendritic cell fate (Esen and

Kielian, 2007). In our cultures, microglia grown in 0.15ug/ml GM-CSF or in GM-CSF

free conditions both maintained a rounded morphology (Fig. 2-3B) and had no

immunoreactivity to the dendritic cell antigen, Cdl 11 c (data not shown). Interestingly,

neonatal microglia derived from mixed glial cultures lacking GM-CSF exhibited a

ramified phenotype when cultured overnight in GM-CSF containing media (Fig. 2-4).

This morphology is similar to that observed in mixed glial cultures with prolonged









exposure to GM-CSF (Esen and Kielian, 2007). It is possible that the conditions within

mixed glial cultures prime microglia to adopt a dendritic-like morphological phenotype

upon exposure to GM-CSF.

A B


























Figure 2-1. Dispase II density centrifugation methodology. (A) Brain hemispheres that
were homogenized with disease II and loaded onto discontinuous gradients
composed of 35% 'low' percoll and 75% 'high' percoll sequestered microglia
to an isopynic density of 1.077mg/ul as determined by beads with known
densities. This configuration spatially separated unwanted brain matter to a
density 50ug/ul more buoyant. (B) Density centrifugation methodology as
described in the literature involved mechanical homogenization and reduced
percoll densities. In our hands, such methodology failed to channel unwanted
brain matter (tissue chunks, red arrowhead) to an isopynic position distal to
the microglia enriched band. (C) Phase contrast images of freshly prepared
cells under a hemacytometer demonstrate viable phase-bright cells (resistant
to trypan blue) as well as reduced particulate matter from disease II
homogenized brains. Following 24hrs of culture, these cells remain adherent
after multiple washes (D) and are thus compatible with experiments that
involve media exchanges.









Previously, histological findings of dystrophic microglia in the aged and diseased

brain have led our laboratory to suggest microglial function may deteriorate with normal

aging. Therefore, we sought to study elements of pathology that are mainly conferred

by microglia in vivo and are known to change with aging and disease (Bolmont, et al.,

2008, El Khoury, et al., 2007, Meyer-Luehmann, et al., 2008, Streit, et al., 2004, Ye and

Johnson, 1999). Recent studies have shown that mRNA copies of inflammatory

cytokines are increased in microglia from aged brains (Sierra, et al., 2007, Ye and

Johnson, 1999). However, mRNA transcripts may not necessarily translate to secreted

protein levels (Munger, et al., 1995, Storm van's Gravesande, et al., 2002), a more

ultimate measure of functional change. To determine if microglia vary their secretion of

cytokines with age, we obtained microglia from young and aged mice and measured

their cytokine levels with and without exogenous immune stimulation.

The most striking observation in this respect was the dramatic increase in IL-6

release under basal conditions (young: 211.831.7 pg/ml vs. aged: 3735.91000.2

pg/ml, p<0.001). In both young and aged microglia, a significant dose-effect relation

following either LPS or PAMCSK3 stimulation was observed (Fig. 2-5A). Moreover,

maximal release of IL-6 was significantly enhanced in aged microglial cells following

LPS (100Ong/ml) or PAMCSK3 (lug/ml) stimulation (Fig. 2-6).

As with IL-6, the amount of TNF-0 Oproduced by aged microglial cells was

significantly higher under basal conditions when compared to young microglia. While

microglia derived from young mice produced no TNF-0 under basal conditions, the

amount was significantly increased to 917.2 91.9 pg/ml in supernatants of aged

microglia cultured for 24h without any exogenous stimuli (p<0.001). This striking










difference confirms age-related higher basal levels of cytokine production previously

observed with mRNA transcript analysis (Sierra, et al., 2007) and indicates that aged

microglia are hyperactive when compared to microglia from young mice. This high

release under basal conditions may explain the lack of a significant dose-effect relation

in TNF-0 production following either LPS or PAMCSK3 stimulation (Fig. 2-5B). In

contrast, in young microglia a significant dose-effect relation was observed. Moreover,

although the maximal amount of TNF-D released by aged microglia in response to

1 0g/ml PAMCSK3 was slightly though significantly increased, responses to 100ng/ml

LPS were not different between aged or young microglia (Fig. 2-6B).













B 1250,0 C 0.6 O1 Viable culture
SDying culture
M 1.000,000 0 0.5 (Triton X treated)

S 750,000

2



Young Aged Microglia 293FT cells
n=10 n=9


Figure 2-2. Purity, yield and viability of microglia. (A) Microglia isolated with dispase II
density centrifugation methodology express lbal, a marker commonly used to
identify in vivo microglia. (B) Yields of microglia from 1 month and 15 month
old mice typically obtained using dispase II based density centrifugation
methodology show little variability with age. (C) Measurement of
mitochondrial respiratory activity indicated that isolated microglia form
cultures comparable in viability to HEK 293 cells, an immortal cell line.









In addition to quantifying microglial cytokine production as a function of age, we

were interested in whether the ability of microglia to serve as an oxidative sink and to

internalize AP3 changes with age. Glutathione acts as antioxidant by neutralizing free

radicals and peroxides and microglia are reported to be the primary glutathione

containing cells in the brain (Hirrlinger, et al., 2000, Lindenau, et al., 1998). We found a

trend indicating that microglia in aged brains have 21% less total glutathione (oxidized

and reduced) compared to microglia from young brains (Fig. 2-7). This result suggests

the reactive oxygen species (ROS) insult that can be caused by AP3 internalization

(Milton, et al., 2008) maybe more injurious to microglia in aged brains.


















Figure 2-3. Adult microglial morphology. (A) GFP-Fractalkine-Receptor transgenic mice,
where microglia are the only in vivo GFP expressing neuronal cells yielded
cultures composed of GFP positive cells. This provided independent
confirmation that our isolation methodology extracted in vivo microglia. (B)
Adult microglial cells remained rounded and did not adopt a dendritic
morphology when cultured for 24hrs with media containing GM-CSF (n=2
mice).









AP3 accumulation is a well recognized feature of AD, however extensive amyloid

deposits may be found in many aged, non-demented individuals (Bouras, et al., 1994).

This pathology may result from AD-independent deterioration of clearance processes.

Microglial scavenger activity on AP3 is proposed as a clearance process that contributes

in maintaining AP3 at physiological levels by counterbalancing constitutive AP3 secretion

by neurons. Our laboratory and others have published accounts of microglial

degeneration that is associated with age (Flanary, et al., 2007, Simmons, et al., 2007,

Streit, et al., 2008, Streit, et al., 2004). If microglia represent a major AP3 clearance

mechanism, their degeneration would result in progressively increasing AP3 levels with

age and therefore would have significant implications to the occurrence of amyloidosis

in AD and some aged individuals. We currently lack the means to isolate degenerating

microglia for experimentation. However ex vivo assessment of microglia acutely

isolated from young and aged mice likely emulates in vivo processing of A3 more so

than neonatal and 'microglial like' cell lines and may give insight to the degeneration of

microglia with age. Ex vivo cultures of microglia were given media with 4ug/mL of

AP342-FITC conjugate or non-conjugated AP342. Western blot analysis indicated that

non-conjugated AP342 preparations contained AP342 monomers, oligomers and SDS

resistant species larger than 220kDa that are likely fibrils thus reflecting in vivo amyloid

burden (Fig. 2-8A). Internalization of A342 by living microglia was visually confirmed

with cell-associated FITC fluorescence (Fig. 2-8B). AP342 observed in fixed cells was

colocalized with lysosomes (Fig. 2-8C). To quantify internalization, microglial lysates

were measured using an AP342 sensitive ELISA. Our results indicate that microglia from

aged mice internalize 53% less A342 relative to microglia from young mice (Fig. 2-9A).









We next wanted to determine if A342 phagocytosis by neonatal microglia (derived from

mixed glial cultures) is reflective of microglia from adult mice. To our surprise, neonatal

microglia internalized significantly more AP342 than microglia derived from young mice.

This suggests that microglia from mixed glial cultures may not necessarily model

microglial AP342 clearance activity in the postnatal brain. Internalization of AP342 is a

prerequisite step for intracellular clearance; however it is by no means a surrogate

marker for biophysical AP342 degradation. To more comprehensively determine the fate

of phagocytosed AP342, we bathed cells in fresh media and measured the levels of A342

in this media and within cells over 16hrs. We observed that internalized AP342 is

invariably expelled by microglial cells in an age-independent manner within 3hrs (Fig. 2-

9B). This result concurs with previous reports of lackluster microglial anti-AP3 activity in

vivo (Simard, et al., 2006) that may stem from impaired lysosomal activity (Majumdar, et

al., 2007b).

Discussion

Improvements on Microglial Isolation

In this study, we aimed to reduce debris contamination which is a feature of

microglial cultures derived from gradient centrifugation based methodology. The brains

of mice that are designated by the National Institute on Aging as an aging model were

treated successively to steps that significantly increased the purity of ex vivo microglial

cultures. Analysis of such cultures, derived from mice of various aging categories,

revealed that microglia from aged brains have markedly increased basal levels of IL-6

and TNF-a secretion, have reduced glutathione levels and have a limited capacity to

ingest A342. In contrast, microglia from younger mice are able to temporarily contain









AP342. Together, these ex vivo findings provide evidence that microglia are subject to

age-associated changes in biology.

Neonatal MGC microglia


'I


Figure 2-4. Neonatal microglial morphology. Neonatal microglia express lbal (A),
however they are distinguished from adult microglia (young and aged) by
ramified cytoplasm ic processes visible under phase contrast microscopy and
Cd11 b immunoreactivity (B). Purity of neonatal microglial cultures was
confirmed by lack of GFAP immunoreactivity, a marker for astrocytes and
immature neurons.

Age-related Changes in Microglial Cytokine Release

Prior reports from our laboratory have shown that microglia have IL-6 and TNF-a

mRNA (Streit, et al., 2000) and more recently, others have found age-related changes

in microglial IL-6 and TNF-a mRNA (Sierra, et al., 2007). We extend on these results by

measuring secreted IL-6 and TNF-a proteins (experiments performed by Ellen Boelen).

Our results indicate significantly more pronounced changes in basal cytokine production


lbal

210.









and responsiveness. It is difficult to make direct comparisons of mRNA and secreted

protein measurements. However, it is of note that the margin of change we observe in

the basal production of IL-6 between microglia from young and aged mice is

approximately 4-fold higher than that observed by mRNA analysis (A17.6x vs. A5x

respectively). We also did not observe detectable levels of basal TNF-a by microglia

from young animals in our studies. These differences can perhaps be explained by

varying sensitivities of the employed detection methodologies or by post-transcriptional

effects. The half-life of mRNA can often be rate-limiting in translation (Ross, 1995).

Secondarily, secretary pathway modulation of newly produced cytokines may also

modulate the concentration of cytokines in the extracellular milieu independent of DNA

transcription. As microRNAs are involved in regulation of gene expression at the post-

transcriptional level, possible changes in this machinery can also be mentioned to

explain the discrepancy between protein and mRNA levels. Our results, though in

agreement with previous reports, indicate that microglia from brains of various aging

groups have much greater differences in cytokine production, and responsiveness to

immune stimulation than was previously thought.

Implications of Age-related Changes in Microglial Cytokine Release

What are the possible implications of age-related changes in microglial cytokine

production? A number of authors commonly describe IL-6 and TNF-a as neurotoxic

molecules involved in AD pathogenesis (Bruunsgaard, et al., 1999, Collins, et al., 2000,

Culpan, et al., 2003, He, et al., 2002, Li, et al., 2007, Licastro, et al., 2000, McGeer and

McGeer, 2001). However, experiments presenting alternative viewpoints have been

published (Brunello, et al., 2000, Loddick, et al., 1998, Marz, et al., 1998, Streit, et al.,

2000, Tarkowski, et al., 1999, Thier, et al., 1999, Wei, et al., 1992). IL-6 may have a









role in regeneration of injured tissue in the brain (Loddick, et al., 1998, Streit, et al.,

2000, Tarkowski, et al., 1999), has known anti-apoptotic properties (Wei, et al., 2001)

and in mice that overexpress both IL-6 and its receptor, IL-6Ra, there is no evidence of

neurotoxicity (Brunello, et al., 2000). Inflammation is a component of wound healing in

the CNS (Klein, et al., 1997, Streit, et al., 2000). Recently, it has been reported that

lesions typically ascribed to cause AD dementia, are present in 20-40% of non-

diseased, aged adults (Price, et al., 2009) and both TNF-a and IL-6 increase with age

(Bruunsgaard, et al., 1999, Wei, et al., 1992). Thus the aging brain appears to exist in a

constant state of injury. Inflammatory processes, such as microglial secretion of IL-6,

maybe needed for persistent regeneration or neuroprotection. The viewpoints that

cytokine release is exclusively neurotoxic or neuroprotective could be equally

considered speculative. Further studies which consider region-specific and age-specific

differences in cytokine response are needed, amongst others, to shed light onto the role

of cytokines released by microglia on the nervous system.

Age-related Changes in Microglial Glutathione Levels

Microglial surveillance of the parenchyma involves scavenging of potentially

hazardous materials. Proteins such as LPS and AP3 induce ROS and it is thought that

the increased levels of antioxidant molecules such as glutathione found in microglia

relative to other brain cells (Hirrlinger, et al., 2000, Lindenau, et al., 1998) protect

microglia from ROS burden associated with scavenging activity (Dringen, 2005, Milton,

et al., 2008, Qin, et al., 2004, Tchaikovskaya, et al., 2005). Our findings of reduced

glutathione levels in microglia immediately analyzed after brain extraction maybe

indicative of broader microglial loss of function with age.











A IL-6 release following LPS stimulation

10000 ,-,
= 7500 -
a 5000 -
S2500 -

Control LPS10 LP5100


IL-6 release following PAMCSK3
stimulation
10000 Y
S7500 ---Aged
5000
= 2500
0 Control PAM PAM
Control PAM 0 1 PAM I


B TNF-a release following LPS stimulation TNF-a release following PAMCSK3
stimulation
3000 --You2500 s--Y
S0 -A-Younged
a2000 a 1500-
I 01000
0- 0- 500

Control LPS10 LPS100 Control PAM 0.1 PAM 1


Figure 2-5. Microglial reaction to immunostimulation. Dose dependent increases in
either IL-6 (A) or TNF-0 (B) release following different concentrations of LPS
(10-100ng/ml) or PAMCSK3 (0.1 lug/ml). Under all conditions, except for
the TNF-0 secretion by aged microglia following either LPS or PAMCSK3
stimulation, a significant effect of concentration was observed. No significant
age-dose interaction was observed under all conditions. Moreover, all figures
show the marked increase in cytokine release by aged microglia under basal
condition (Cytokine experiments performed by Ellen Boelen).

Age-related Changes in Microglial Processing of AP

Microglia are thought to participate in AP3 plaque burden regulation by sequestering


and processing of Ap. Our purpose in performing AP3 metabolism experiments with ex


vivo cultures of microglia was to further understand the relationship between microglia

and AP342 homeostasis. Individual microglia surrounding AP3 plaques become enlarged

as plaques become smaller with time (Bolmont, et al., 2008). Despite this correlative

evidence of microglial participation in AP3 plaque reduction, natural processes do not

appear to reverse amyloidosis (Jankowsky, et al., 2005). Previously, we have

suggested that microglial function may deteriorate with time. This was mainly inspired









by histological findings, demonstrating the appearance of a dystrophic microglia

phenotype with normal aging and around AP3 plaque deposits in AD brains (Miller and

Streit, 2007). The finding that microglia from young brains internalize AP342 while

microglia from aged brains do not, likely reflects a more global change in microglial

functionality with age as we do not observe dystrophic microglia in aged mice.

Interpretation of Ap Expulsion by Younger Microglia

In younger mice, the temporary sequestration of A342 suggests that microglia are

not directly involved in degradation. However there is clear potential for microglia,

which are highly motile (Bolmont, et al., 2008), to be involved in the movement of A3 in

the nervous system. The cycling of A342 through microglial endocytosis and exocytosis

could result in redistribution that modulates peptide availability for amyloid formation.

Our observations indicate that microglia remove AP3 peptide out of the extracellular miliu

at ng/ml quantities, possibly creating a dynamic cellular compartment in vivo. A 2 to 4-

fold age-related decrease in microglial AP3 internalization as observed in our

experiments could result in less transfer of A3 from the extracellular space --thereby

making available more AP3 for plaque formation. Amyloidosis occurs in a significant

amount of non-diseased aged adults (Price, et al., 2009). Therefore, it is plausible that

the dynamic redistribution of A3, a protein that is particularly aggregation prone, is

needed for homeostatic maintenance. Perturbation of this process due to age-related

changes of microglial function such as that described here could contribute to unhinged

accumulation of A3 to toxic concentrations in the extracellular space.











A IL-6 production
10000 # E Young microglia
N Aged microglia
75000



2500


LPS 100ng/ml PAMCSK3 1lig/ml




B TNF-a production
C Young microglia
3000 M Aged microglia

-S2000
Z
1--
1000


LPS 100ng/mi PAMCSK3 1pg/ml

Figure 2-6. Cytokine secretion of young and aged microglia. (A) Upon stimulation with
the biological inflammatory reagent LPS (100ng/ml) or Pam3CSK4, a
synthetic agonist of toll-like receptor 2 (1 ug/ml), IL-6 production by aged
microglia was markedly increased when compared to young microglia. (B)
LPS (100ng/ml) stimulated similar TNF-0 production between microglia
derived from young and aged mice. Yet, TNF-0 production was significantly
increased in aged microglia following Pam3CSK4 (1ug/ml) exposure. *
p<0.05; #, p<0.001.

Concluding Comments

Together, our ex vivo quantification of microglial functions of cytokine production,

glutathione levels and AP342 scavenging activity paint a complex picture of endogenous

activity. Microglia from young mice produce less cytokines, while microglia derived from

aged mice have higher basal levels of cytokine secretion than was previously thought

and have reduced glutathione levels. Microglia derived from aged mice lack

comparable AP342 internalization capacity compared to less aged microglia while

microglia from young mice and mixed glial neonatal cultures do not seem to retain










internalized AP342. These direct assessments of microglial function in ex vivo

experiments, free of confounding contributions of other brain cells and debris,

demonstrate a nuanced view of microglial function and suggest that microglial biology

may change with aging.


A 10
9


0 Neonatal microglia
* Aged microglia
n=3.4 mice/n


Figure 2-7. Microglial glutathione content. Microglia from aged mice have a 21%
reduction of glutathione antioxidant. Data is normalized to total protein and
represents oxidized and reduced forms of glutathione detected in microglia
analyzed immediately following brain extraction. p=0.27.











1 2 3
H20 + -
Naive assay medium +
Conditioned medium -





60 -




16 ,, ,
4 -


AP42-FITC DAP^I Merge


Figure 2-8. Species of A342 used for microglial experiments. (A) Western blot analysis
using 6E10, an antibody specific to the first 16 amino acids of A342, indicates
monomeric (4kDa), oligomeric (16kDa, 20kDa) and higher-order
conformations larger than 220kDa in stock preparations as well as
preparations that have been exposed to microglia. Higher-order
conformations larger than 220kDa persist following the exposure of samples
to buffer containing 2% SDS suggesting the presence of fibrillar species. 10%
serum in media overloads gel at 60kDa and may block visibility of some AP342
species. 75ng AP342/well. (B) Internalization of AP342 by microglia from adult
mice was directly observed in living microglia with FITC conjugated AP342
peptide and with 6E10 immunocytochemistry. In both cases, AP342 had a
peri-nuclear localization. Lamp1 colocalization with 6E10 immunoreactivity
suggests that some AP342 reached microglial lysosomal compartments (C).












A -
2.00

1.75 IntracellularAl
Mock
1.50 3 Ohr
m 3hrs
1.25 T 16hrs
S1.25

1.00

< -o. T T T

0.50 --

0.25 I i

Neonatal Young Aged
microglia microglia microglia
B n=6 n=5 n=7
3.0 *
** Expelled AOI
2.5 Mock
3hrs
S16hrs


1.5 -

1.0

0.5 1

0.0-
Naive Neonatal Young Aged
media microglia microglia microglia
n=6 n=5 n=7


Figure 2-9. Fate of A3 internalized by microglia. Microglia extracted from mice of various
ages were exposed to AP342 preparations containing monomeric, oligomeric
and SDS-resistant fibrillar species (reflecting in vivo amyloid diversity) in
pulse-chase experiments. (A) Neonatal and young microglia respectively
internalized 74% and 53% more AP342 relative to aged microglia. (B)
Invariably, internalized AP342 was expelled by neonatal and young microglia
within 3hrs of ingestion, suggesting disengagement from biophysical
degradation following phagocytosis. Mock data (gray) represents experiments
without the presence of cells to control for non-specific adherence of AP3 to
culture wells. Detection of AP342 requires the presence of both NH2 and
COOH terminals of A342, thus only intact AP342 peptides are quantified in the
above experiments. *, p<0.05; **, p<0.01.













Neonate Single cell
Strong nturation suspension
130ui filtration
400 x g/l1min
40urn filtration





Ix PI
Adult Enzyme digestion Single cell 750%
Strong nturation suspension 000>
70u( fltration
1n00 xr/rmin


.. .. I .- -' -: Adherent
culture "' : .. culture


Percol l BSwash
gradient DMEMw/ 0%FBS
DM EM w;10%FBS


Adherent
culture


1hr-37-C Experiment









1hrS37C--- Experiment


Figure 2-10. Overview of mixed glial culture (MGC) and density centrifugation
methodologies utilized to obtain microglia. MGC derived microglia were
harvested after 1 to 4 weeks of growing atop a multi-cell feeder layer
comprised mostly of astrocytes. Homogenized brains of neonatal origin form
MGCs while adult brains do not, thus MGC microglia are of neonatal origin.
Density centrifugation based isolation offered the advantages of immediate
culture creation from young and aged brains. This was important because
microglia from brains of aging models were accessible and ex vivo analysis
utilized cells temporally proximate to in vivo microglia.









CHAPTER 3
ENGRAFTMENT PATTERNS OF NSCS TRANSPLANTED INTO MOUSE MODELS
OF ALZHEIMER'S DISEASE

Introduction

The brains of AD patients are characterized by AP3 pathology as well as neuronal

cell death. This has led to interest in the use of neuronal stem cells (NSCs) for cell

replacement or delivery of therapeutic molecules such as proteases that reduce AP3

pathology. Before such studies can be carried out on humans, several fundamental

questions must first be addressed. First, do NSCs survive in brains with the pathologies

of AD? Second, how does the complex architecture of the brain effect the engraftment

of transplanted NSCs? Third, does the region of engraftment affect integration and

migration within the parenchyma? And finally, does transgene expression in genetically

modified NSCs effect engraftment?

The question of NSC survival in AD or models of AD has been explored in in vitro

and in in vivo (non-transplant) settings. Unfortunately, these studies present

contradictory data on how NSCs are affected by AP3 pathology. Specifically, in vitro and

post mortem studies yield data contradictory to studies performed on mouse models of

AD. Lopez et al., and Calafiore et al., report that AP3 oligomers induce isolated

subependymal zone (SEZ) and subgranular zone (SGZ) mouse NSCs to differentiate

into neurons. They also report that AP3 concentrations that are toxic to neurons do not

induce apoptosis in SGZ and SEZ stem cells or hinder their proliferation (Calafiore, et

al., 2006, Lopez-Toledano and Shelanski, 2004). These observations are consistent

with research which quantified increases in neurogenesis in humans with AD (Jin, et al.,

2004, Ziabreva, et al., 2006). Conversely, multiple publications report evidence of

decreased and abnormal neurogenesis in the hippocampus of several mouse models of









AD (Dong, et al., 2004, Donovan, et al., 2006, Wang, et al., 2004). Our lab recently

reported that the amount of A3 deposition in AD mice reflects negatively on the survival

of new born SGZ neurons as they reach functional maturity.

Because the current literature fails to explicate how AP3 burden affects endogenous

NSCs, how transplanted NSCs will behave in AD patients remains an open question.

To shed light on this topic, we performed NSC transplant studies in mice modeling AD

AP3 pathology. Our specific aim was to define engraftment patterns in the hippocampus.

The pathology in the hippocampus is linked to dementia (Braak and Braak, 1995, Braak,

et al., 1996). Therefore, the hippocampus is a region of great clinical interest for cell

restoration and drug delivery therapies. Our primary findings in this study demonstrate

that physical forces relating to hippocampal anisotropic architecture dictate the

distribution of NSCs more so than cell migration. We also find that genetic

overexpression of MMP9 is associated with significant enhancement of NSC graft size.

Methods

Isolation of NSCs

The protocols for isolating NSCs are contained in the literature (Marshall, et al.,

2006, Zheng, et al., 2006). In brief, a rectangular forebrain block containing the

subependymal zone was isolated from neonatal (P4-P9) green fluorescence protein

transgenic mice (003116, The Jackson Laboratory, Bar Harbor, MI) or from non

transgenic B6 mice (bred in house). This was done by removing the OB, cerebellum,

hippocampus, lateral portions of the striatum, and lateral and dorsal cerebral cortex.

This block was minced with a razor blade, incubated in 0.25% trypsin/EDTA (Atlanta

Biologicals, Lawrenceville, GA) and dissociated into a single cell suspension by

triturating through a diametrically descending series of glass pipettes. Cells were then









pelleted and washed several times before plating in NSC media (DMEM/F12 with 5%

FBS, penicillin (100U/ml), streptomycin (100ug/ml), Bovine Pituitary Extract (35ug/ml),

Fungizone (250ng/ml). NSC monolayer's were kept in an immature state within tissue

culture flasks by supplementing the media with 20ng EGF and 20ng FGF every two to

three days (Walton, et al., 2006a, Zheng, et al., 2006).

Transplantation into Amyloid Beta AD Mice

The Line 85 and the Line 107xtTa mouse models, which constitutively overexpress

and selectively express human AP3, respectively, have been previously described to

model the AP3 physiology that is a hallmark AD pathology. Host and donor mice are

immune-matched due to their shared B6 background. Mice were induced to a state of

deep anesthesia with 1-5% isoflurane. The hair on their scalps was shaved and the

surgical area sterilized with betadine antiseptic and 70% ethanol. The mice were

securely mounted with ear bars and a nose bar to a stereotaxic apparatus. The

anesthesia mixture was delivered through an inlet within the nose bar enclosure for the

duration of the surgery. A sterile scalpel was used to make a small incision into the skin

above the skull. The skin was reflected in order to expose Bregma. A Hamilton 10ul

syringe with a 33 gauge needle (Hamilton Company, Reno, NV) was then loaded with

NSCs prepared at 1 x 105 cells/ul for lateral ventricle injections and ~5 x 104 cells/ul in

1x dPBS for multi-deposit hippocampal injections. The cells were derived from a

trypsinized and pelleted monolayer of NSCs, washed with 200ul 1x dPBS and diluted to

the appropriate volume using a reference cell count done on a hemacytometer. The

following coordinates relative to Bregma were used to target the lateral ventricle:

anterior/posterior (AP): -0.2mm, medial/lateral (ML): +/-1.2mm, dorsal/ventral (DV):-

2.5mm from the dura. Occasionally, needle tracks using these coordinates were found









in the base of the hippocampus where it meets the thalamus. This border region is not

termed in modern atlases (Paxinos Atlas and Allen Online Atlas) and vaguely referred to

in a study by Nagaraja and colleagues (Nagaraja, et al., 2005). Cerebrospinal fluid

research 2005). There may be cerebrospinal fluid in this border in the posterior brain

midbrainn). However, this is not the case in the rest of the brain. Therefore, we refer to

this region here as the 'hippothalamic' fissure rather than the 'hippothalamic' cistern. To

target the corpus callosum, we used: AP: +1.2mm, ML: +/-0.5mm, DV: 2.5mm from the

dura. We used multiple depth coordinates in an attempt to maximally disperse cells in

the hippocampus. The regions targeted were the ventral hippocampus dentatee gyrus,

SGZ region), the medial hippocampus (molecular cell layer, hippocampal fissure region)

and the dorsal hippocampus (CA3). The coordinates used were: AP: +2mm, ML: +/-

2mm's, DV: -2.0mm, -2.3mm, -2.5mm from the skull surface. 1.25 x 105 cells were

deposited at -2.0mm and -2.3mm, while 2.5 x 105 cells were deposited at -2.5mm. In all

injections, 4ul-8ul total volume was deposited at the rate of 0.25ul per 15 seconds

dependantt on cell concentration). Subsequently, the needle was left alone for 5

minutes to allow for the diffusion of cells from the injection tract. It was then retracted

slowly to minimize damage along the injection tract. The incision was closed with a

staple and the mouse placed in a warm, dark recovery area. To avoid or minimize the

discomfort, distress and pain associated with this procedure, 0.1 mg/kg of buprenophine

was administered as the animals recovered on the day of surgery and the subsequent

day. The mice were checked for signs of abnormal recovery during the survival periods.

At the end of the survival periods, the mice were deeply anesthetized in an isoflurane

induction chamber, euthanized by Beuthanasia administration and then perfused with









cold 1x PBS. Whole brains were quickly dissected out and placed in cold 4%

paraformaldyhyde fixative overnight. The brains were immersed in 30% sucrose before

sectioning within a cryostat. 20um sections were stored in anti-freeze media at -20C

until further processing.

Immunochemistry

4% paraformaldehyde fixed cells or tissue sections processed for

immunofluorescence in solutions containing 0.1% Triton-X, 10% goat serum in 1x PBS.

Primary antibodies used in this study include copGFP (1:2000, Evrogen, Moscow,

Russia), anti human MMP9 Clone 56-2A4 (Abcam, Cambridge, MA), anti-Ap3 6E10

(1:2000, Signet, Dedham, MA), microglial antigen lbal (1:1000, Wako, Richmond, VA),

astrocyte antigen GFAP (1:1000, Dako Corporation, Carpinteria, CA), and neuronal

antigen NeuN (1:500, Millipore, Bellirica, MA), neuronal antigen Bill Tubulin (1:500,

Covance, Princeston, NJ). Cells were rinsed and incubated with goat secondary

antibodies Alexa 488, 568 (Invitrogen, Carlsbard, CA). Cells were photographed with

an Olympus DP71 camera mounted on an Olympus BX60 microscope.

Modeling Paths Of Least Resistance (PLR)

To simulate PLR's that may occur within the hippocampal formation, 3ml's of 1%

agarose (w/v) in PBS was allowed to solidify in a clear 15ml polystyrene tube. Addition

of 100ul of H20 atop this solidified block formed a layer ~1mm to 3mm in height. 3ml's

of warm 1 % agarose was then added. As the two layers of agarose fused, a region of

weakness developed due to an abrupt decrease in agarose density at the H20 layer. 5-

10ul of crysyl violet solution was then injected at various regions in the agarose block.









Results

NSCs robustly propagate in vitro (Reynolds and Weiss, 1992, Walton, et al.,

2006a). In our hands, cells from the brains of two neonatal mice have been passage

>10x for a duration of six months producing a conservative estimate of approximately

one hundred million cells. NSCs formed monolayers of elongated cells (Fig. 3-1A, C)

similar to that of human NSCs (Walton, et al., 2006a). NSCs were also similar to

human NSCs in their immunoreactivity for the neuronal progenitor markers GFAP and

13111 Tub (Fig. 3-2A-B). lbal reactivity was observed in some cells in NSC cultures (Fig.

3-2C), indicating the presence of microglia. However, these cells were much larger

than adult ex vivo microglia (compare Fig. 3-1 B-C & 3-2C) and had a morphology

similar to neonatal microglia from mixed glial cultures (Fig. 2 .4, see Chapter 2). It has

been suggested that microglia derived soluble factors maintain the self-renewal capacity

of NSCs in vitro (Walton, et al., 2006b).


GFP NSC's


Figure 3-1. Morphology of in vitro NSCs. NSCs form monolayers of elongated cells (A)
and are significantly larger than microglia acutely isolated from adult mice (B).


DAPI GFP





A

Microglia NSC





B Mum c lNum









We initially transplanted NSCs from GFP transgenic mice into the lateral ventricle

(LV) in order to recapitulate previous reports of rostral migration into the olfactory bulb

(Marshall, et al., 2006, Zheng, et al., 2006). In our AD models, this is a region of high

AP3 plaque burden and in humans, the loss of smell is purportedly amongst the first

signs of AD (Fusetti, et al.). Therefore, the migratory potential of NSCs suggests the

possibility of wide-spread dispersion of candidate therapeutics to relevant regions of the

brain. We performed transplants of GFP cells on non transgenic mice and transgenic

mice symptomatic for AP3 pathology using previously published stereotactic coordinates

(Zheng, et al., 2006). Similar to previous studies with SEZ NSCs (Walton, et al., 2006a,

Zheng, et al., 2006), mice were sacrificed 4-8 weeks later. In transplants where the

needle track was confirmed to have entered the lateral ventricle, we did not observe

ventricular wall colonization or extravasation towards the olfactory bulb (non transgenic;

n=3, TG=4), (data not shown). This result was not due to the inability of NSCs to form

grafts as some surviving cells were observed alongside needle tracks that had entered

the lateral ventricle.

These results motivated us to modify stereotactic coordinates to directly target the

hippocampus. Our goal was to achieve maximal cell dispersion within the

hippocampus. We therefore chose 3-point depth coordinates for simultaneous injection

into the dorsal hippocampus proximall to CA3), the medial hippocampus hippocampall

fissure, molecular cell layer and lateral arm of the dentate gyrus) and the ventral

hippocampus (subgranular zone of the dentate gyrus). The accuracy of these

coordinates was verified with injections of the tracer crysl violet (n=4/depth coordinate,

data not shown) on freshly deceased mice. A total of 5 x 105 cells were injected, a









value corresponding with previous animal work (Parl et al., Exp Neurol 2006) and

proportional to human NSC clinical trials (Stem Cells Inc; American Association of

Neurological Surgeons Annual Meeting 2010).


Figure 3-2. In vitro characteristics of NSCs. NSC cultures were characterized by GFAP
(A) and Bill Tubulin immunoreactivity (B). Some cells were immunoreactive
for the microglial marker, lbal (C). n=4 cultures per immunostain.

We then injected NSCs with lentiviral directed expression of 1) MMP9 and GFP or

2) GFP only. The GFP used here is from the plankton copepod. It is similar in size to

EGFP (26kDA), but has more fluorescence output (Shagin, et al., 2004). GFP NSCs

and MMP9 NSCs had equal fluorescence output of A.U>103 as determined by

fluorescence activated cell sorting (FACS) (see Chapter 4).

After a month of survival, brains were confirmed to have hippocampal needle

track penetration in several mice. MMP9 NSCs formed graft cores that were 82.4%


P111-Tub (TuJl) DAPI





B











larger than GFP NSCs (n=4; GFP NSC, n=5; MMP9 NSC, Fig. 3-3A-C). Interestingly,

the effect of MMP9 on survival is diminished with increasing amounts of proximal AP3

plaques (R2=0.701), (Fig. 3-3D). This data suggests that AP3 amyloidosis is toxic to

NSCs and that MMP9 genetic modification partly rescues cell survival. No other

differences were observed between GFP NSCs and MMP9 NSCs. The engraftment

patterns described below apply to both cell types.


10000


8000
Cu

* 6000

o 4000
U

0 2000


GFP Only MMP9


0 GFP Only
M MMP9


DR==00162


50 100
Plaque # (Contra)


Figure 3-3. MMP9 associated changes in engraftment. MMP9 NSC grafts (n=5) were
82.4% larger than GFP NSC grafts (n=4) (A-C). The fluorescence intensity of
both cell types were matched with FACS analysis (A.U>103). The size of
MMP9 GFP grafts was negatively correlated with the amount of nearby AP3
plaques (R2=0.701) (D). As seen here, the horizontal distribution of GFP
NSCs and MMP9 NSCs appears similar. **, p<0.001

Both GFP NSCs and MMP9 NSCs formed grafts in subcortical and corpus

callosum white matter tracks despite injection within the interior of the hippocampus

(Fig. 3-3A-B, Fig. 3-4A,C). This distribution was observed in all mice (n=9). NSCs


8000
7000
6000
5000
4000
o 3000
5 2000
1000
0









formed a smaller, secondary graft within the hippocampal fissure in 33% of mice (Fig. 3-

3-4B). This distribution pattern was confirmed in a separate cohort of mice that survived

two months following surgery (n=7), (Fig. 3-9A). NSC engraftments in the white matter

and the hippocampal fissure had a consistent disc-shaped morphology that extended

horizontally; ~1 mm on the anterior/posterior plane and ~1mm on the medial/lateral

plane in needle track containing sections and adjacent sections.

NSCs injected directly into the corpus callosum and allowed to graft for 3 days in

non transgenic mice (n=2) resulted in engraftment that was strikingly similar to the

distribution found in AP3 transgenic mice (compare Fig. 3-4A & D). Furthermore, NSCs

that were engrafted for 1 year in the fimbria white matter of non transgenic mice were

similarly distributed horizontally at the hippothalamic fissure (n=3), Fig. 3-4E-F).

Recently, several investigators have suggested directional migration of NSCs as

an explanation of why cellular transplantation into the hippocampal gray matter

paradoxically results in engraftment of cells in the white matter (Blurton-Jones, et al.,

2009, Pihlaja, et al., 2008, Radojevic and Kapfhammer, 2009, Raedt, et al., 2009, Tang,

et al., 2008). As described above, our transplant studies were characterized by 1)

white matter grafts in AP3 transgenic mice receiving NSCs in the hippocampus and 2)

white matter grafts in non transgenic mice receiving NSCs in the white matter of the

fimbria and corpus callosum. With the exception of NSCs closely associated with the

hippocampal fissure, transplanted cells did not form graft cores in gray matter. If cell

migration were a major factor determining the distribution of transplanted NSCs, one

would expect a range of hippocampal distribution reflecting NSCs in transit. However,

our long and short term survival studies resulted in a predictable distribution of NSCs









along horizontal fissures of the hippocampus and surrounding white matter.

Consequently, we hypothesized that the distribution of transplanted cells was largely

determined by PLR's that exist between anisotropic regions of fissures, white matter

and densely packed layers of neurons.

To provide support for this hypothesis, we decided to simulate a region of

horizontal weakness using 1 % agarose. We placed a thin layer of water (~1 to 3mm) on

top of already solidified agarose. This created a non-uniform (anisotropic) region of

agarose density when a new layer of agarose was added. Crysl violet tracer that was

injected above or below this region resulted in a vertical distribution of tracer (n=5), (Fig.

3-5A-B). However, a horizontal distribution was observed upon injecting into the region

of weakness (n=3), (Fig. 3-5C). These results demonstrate a change in agarose density

creates anisotropic forces that direct movement of infusate along a horizontal path.

Such forces may occur in the hippocampus. If this is the case, anisotropic forces that

distribute NSCs during surgery may explain why transplantation of NSCs into

hippocampal gray matter results in distal white matter engraftment.

To gain insight on this question, we injected GFP NSCs proximal to the corpus

callosum, hippocampal fissure, the dentate, the hippothalamic fissure and sacrificed

mice immediately following surgery. As a control, we also injected GFP NSCs into

regions of the brain we suspected to have a more uniform isotropicc) distribution of

tissue. These included the cortex, the hippocampus and the striatum. We found that

injection of NSCs into the hippocampus at various dorsoventral depths resulted in

immediate NSC distribution along the horizontal features of fissures and white matter

(Fig. 3-6). Specifically, NSCs that were injected into the dorsal aspect of the









hippocampus yielded distributions throughout the subcortical white matter tracks above

the hippocampus (corpus callosum, cingulum bundle, alveus) and to a lesser extent,

CA1 (n=8), (Fig. 3-6A). NSCs targeted to the ventral aspect of the hippocampus

distributed mediolaterally at the base of the hippocampus (i.e., hippothalamic fissure)

(n=8), (Fig. 3-6B). The horizontal distribution of the NSCs extended hundreds of

micrometers; from the 3rd ventricle to lateral regions such as above the dorsal lateral

geniculate nucleus. NSCs targeted to the dorsoventral center of the hippocampus

hippocampall fissure coordinates) distributed within the corpus callosum, the

hippocampal fissure and the base of the hippocampus (n=6), (Fig 3-6C).


Figure 3-4. Survival and distribution of transplanted NSCs. Deposition of NSCs into
hippocampal gray matter of A3 transgenic mice resulted in horizontal
engraftments within subcortical and corpus callosum white matter (n=9) (A,C).
In 33% of mice, secondary engraftments were observed in the hippocampal
fissure (B, arrows). Deposition of NSCs into the corpus callosum (n=2) or
fimbria (n=3) of non transgenic animals resulted in PLR engraftments with
similar horizontal distributions (D-E, arrows indicate needle track, arrowheads
indicate hippocampal fissure). The separation of tissue observed in panel F
(double arrows) provides evidence of a fissure with natural weakness into
which NSCs selectively distributed into. AP3 transgenic mice were sacrificed a
month following NSC transplantation (D-F). Non transgenic mice were
sacrificed 3 days (A) and 1 year (B,C) following NSC transplantation.


C[TT
Ill :1 ;,[: cc

bd~r~CAI


lc HlIPPO









This result agrees with previous studies investigating the influence of hippocampal

structure on infusate distribution patterns (Astary, et al., 2010). As noted in Astary et al,

distribution profile and shape of our infusions were also dependent on neuroanatomical

and cytoarchitectonic structure. Therefore, in contrast to NSCs injected into anisotropic

white matter or cell-free CSF-filled regions, NSCs that were injected into the thalamus

(n=6) (Fig. 3-6D), cortex (n=2, data not shown) or striatum (n=6), (Fig 3-6E) had circular

or vertical distributions. Notably, NSC striatal distribution bifurcated along the gray

matter and radiating fibers of white matter (Fig. 3-6E).

Non-uniform
Uniform (Isotropic) Uniform (Isotropic) (anisotropic)
resistance resistance resistance






A B C

Figure 3-5. Modeling paths of least resistance. A 1% agarose block, with structural
weakness in the region denoted by arrows and crosses was injected with the
tracer, crysl violet. Deposition above (n=5) (A) or below (n=5) (B) the arrow
resulted in a vertical distribution of the tracer. Deposition at the area of
structural weakness resulted in a horizontal distribution of the tracer (n=3) (C).

This data together with our long term survival experiments suggests that the initial

distribution of transplanted NSCs is the main determinant of engraftment pattern. Our

long term survival experiments also demonstrated that transplanted NSCs are not

characterized by wide-spread penetration into the hippocampal parenchyma. However

we have encouraging results that suggest an alternate route may be used to attain such

a result. Extensive intra-hippocampal grafts occurred in transplants that deposited

NSCs in the point of the hippothalamic fissure that is proximal to the dentate, (Fig. 3-7).









Similar to transplants that result in subcortical white matter engraftment, NSCs

distributed horizontally at the hippothalamic fissure. However, engraftment in this

region uniquely resulted in significant representation of cells around and within the

dentate. In AP3 transgenic mice, these cells extended processes into the surrounding

gray matter at two weeks (Fig 3-7A) and appear to be migratory at two months (Fig. 3-

7B). In a non transgenic animal that survived five months, NSC grafts were found

within the dentate of the ipsilateral hippocampus (Fig. 3-7C). However, in the

contralateral hippocampus, NSCs were found in the hippothalamic fissure (Fig 3-7C, 2nd

and 3rd panels). This suggests cells were deposited at this PLR. In both hemispheres,

cells appear to be migrating dorsally from the graft core (Fig. 3-7C). We did not find

cells in the thalamus in these animals. This suggests vertical directional migration away

from the thalamus possibly due to factors associated with the endogenous NSC niche

that exists in the SGZ. The robustness of these engraftments prompted us to perform

further transplants to repeat these results. These studies are ongoing.

Previous work from our lab has shown that the survival of newborn neurons (NSC

progeny) is negatively correlated with the presence of AP3 pathology (Verret, et al.,

2007). However, the survival of astrocytes, which were more than 50% of the newborn

population, was not affected. The differentiation of transplanted NSCs has been well

characterized in multiple studies. In general, 30-90% of endogeneous and transplanted

NSCs express glial markers in multiple studies using various NSCs in different disease

models (Hattiangady, et al., 2007, Lundberg, et al., 1997, Shetty, et al., 2008,

Svendsen, et al., 1996, Tang, et al., 2008, Verret, et al., 2007). A minority of

transplanted NSCs become neurons (Shetty, et al., 2008). However, a few studies









describe neurons as the major phenotype transplanted NSCs differentiate into (Bennett,

et al., 2010, Lu, et al., 2007, Park, et al., 2002). Due to this ambiguity, we were

interested in determining the differentiation state of NSCs we transplanted in AP3

transgenic mice. As mentioned above, these cells are immunoreactive for GFAP and

1311 Tubulin (Tujl) in vitro. A subpopulation of cells in NSC cultures express lbal. We

stained AP3 transgenic mice tissue sections containing NSC white matter engraftments

with antibodies against these markers and NeuN, a marker for post-mitotic neurons.

We did not observe ll31-Tubilin or NeuN immunoreactivity in any sections (n=2 mice,

4x20um sections), (data not shown). However, there was strong graft associated

immunoreactivity for lbal and GFAP (n=3), (Fig. 3-8, 3-9). The elongated morphology

of lbal and GFAP positive NSCs in vitro matches that of transplanted cells in vivo

(compare Fig. 3-2 to 3-8 & 3-9, arrows). This morphology contrasts that of endogenous

quiescent glia or glia reacting to nearby AP3 plaques (Fig. 3-8, 3-9, arrowheads).

As described above, transplant cells largely remained within a graft core. Amyloid

beta plaques proximal to graft cores (<50um or ~3 cell lengths) did not attract NSCs

(Fig. 3-4A, 3-8B, 3-9) thus confirming prior reports (Burton-Jones, PNAS 2009). It is

possible that white matter tracks support a unique NSC niche that is defined by a glial-

like phenotype. However, transplanted cells in this niche do not respond to or do not

have access to the cues that direct chemotaxis of endogenous glia towards AP3 plaques.

Discussion

Newly discovered properties of NSCs, namely long-term in vitro expansion and

engraftment potential (Walton, et al., 2006a, Zheng, et al., 2006), have generated

significant interest for application towards novel therapies such as cell replacement and













Ohr
survival







Ohr
survival


Ohr
survival


Natlie GF


[=ll;J'.CrX




cc|.I=
HIPPOM
A


Figure 3-6. Immediate distribution of NSCs in paths of least resistance. 5 x 105 NSCs
were injected into various regions and analyzed immediately following
surgery. Deposition of NSCs in the hippocampus which has variably dense
layers of tissue resulted in horizontal distribution along anisotropic paths of
least resistance (A-C). These included the corpus callosum (n=8) (A), the
hippothalamic fissure (n=8) (B), the dentate (n=6) (not shown) and the
hippocampal fissure (n=6) (C). Interestingly, NSCs whose target was the
hippocampal fissure were found hundreds of microns away from the site of
injection (C). Because directional migration occurs in the order of hours to
days, this distribution can only be explained by anisotropic forces directing
cell distribution. In contrast to anisotropic horizontal distribution observed in
A-C, a vertical or circular distribution was observed with NSCs injected into
isotropic regions such as the thalamus (n=6) (D), the cortex (n=2) (data not
shown) and the striatum (n=6) (E). In the striatum, NSCs distributed in a
radial pattern that likely reflects white matter that is striated within gray matter.

drug delivery. However, a series of clinical questions regarding how NSC grafts behave

in diseased brains motivated us to study transplanted NSCs in mouse models of AD. In

this study, we aimed to determine the engraftment patterns of NSCs in mouse brains

with AP3 pathology. Our studies indicate that NSCs can survive in an environment with


3A IIF


AIT1 "./



GFPJl r


DAPIIGFP
LV




STRATUM
E









AP3 pathology, however these cells do not migrate towards AP3 plaques. The localization

of cell engraftment is largely dependent on previously uncharacterized paths of least

resistance. These paths distribute cells in a horizontal pattern along both lateral/medial

and anterior/posterior planes. Process extension and migration appear most robust in

NSCs deposited proximal to the hippothalamic fissure and dentate. Graft size is

significantly enhanced by overexpression of MMP9. Because MMP9 NSC graft size is

inversely correlated to AP3 plaque number, MMP9 may be a novel factor for protecting

NSCs in pathologic amyloid environments.

Paths of Least Resistance versus migration

Our observations of the pattern of NSC engraftment are also observed in studies

that focused on a myriad of pathologies including ischemia, epilepsy and Alzheimer's

disease (Blurton-Jones, et al., 2009, Olstorn, et al., 2007, Pihlaja, et al., 2008,

Prajerova, et al., Radojevic and Kapfhammer, 2009, Raedt, et al., 2009, Tang, et al.,

2008, Watson, et al., 2006). In these studies, NSCs (including those from humans)

(Olstorn, et al., 2007), formed graft cores at sites distal to where they were targeted. To

explain this paradox, several authors suggest transplanted cells directionally migrate

(Blurton-Jones, et al., 2009, Pihlaja, et al., 2008, Raedt, et al., 2009, Tang, et al., 2008).

If migration is the primary force determining why transplanted NSCs are frequently

found in white matter, one would expect a trail of in transit cells orienting away from the

site of deposition. We not observe NSCs in the hippocampal fissure migrating dorsally

towards the white matter or ventrally towards the hippothalamic fissure. Migration

undoubtedly occurs in vivo, especially towards lesions. However as observed in Olstorn

et al., migration to a site of infarct is secondary to a PLR distribution (Olstorn, et al.,

2007). In this study, transplant human NSCs in a non injured brain distributed hundreds










of micrometers in the corpus callosum white matter. These NSCs then migrated tens of

micrometers to the infarct zone in CA1. Within the infarct zone, a PLR-like distribution

was maintained. This suggests a pre-existing PLR distribution was a significant factor

in determining the final distribution of migrating cells.


2 week
survival
TG








2 month
survival
TG






5 month
survival
nonTG


Figure 3-7. Engraftment patterns of NSCs deposited at the ventral border of the
hippocampus. Needle tracks (A, arrow, left panel) which deposited NSCs
proximal to the hippothalamic fissure resulted in NSC distribution along this
horizontal border and the dentate (A-C). Process extension and directional
migration from the point of cell deposition was observed in AP3 transgenic
animals (n=2) (A,B) and in a non transgenic animal (n=l) (C). Unilateral
injection resulted in bilateral distribution of NSCs along the hippothalamic
fissure (C, middle, right panels). No cells were found in the thalamus in these
studies.

Our experiments on mice sacrificed immediately after surgery provide definitive

proof that NSCs distribute hundreds of micrometers along horizontal PLR's. Depending

on which region of the hippocampus was targeted, NSCs distributed horizontally in 1)


Sagital iewSagitalviewlagttalvie

HIPP-O;[o
I~ll;HIPPO
HIPPiO,5
V 1Ill:IlT,]HLMr

A THLM '100-

SagittalI 1[, view Saitlviw T iFS A )Sgitlve

[;[l-HIPPO

H ] i;IPPOm,.









the subcortical and corpus callosum white matter, 2) the hippocampal fissure or 3) the

hippothalamic fissure or in all three. Horizontal distribution along PLR's was also seen

in MRI studies of tracers infused into the hippocampus (Astary, et al.). Our observation

of NSCs horizontally spread in graft cores hundreds of micrometers apart within minutes

of being injected in the hippocampal parenchyma provides a mechanistic demonstration

of engraftment patterns observed in multiple studies (Blurton-Jones, et al., 2009,

Olstorn, et al., 2007, Pihlaja, et al., 2008, Prajerova, et al., Radojevic and Kapfhammer,

2009, Raedt, et al., 2009, Tang, et al., 2008).

In studies were less than 1 x 105 NSCs were transplanted, PLR distribution is

infrequently found (Hattiangady, et al., 2007, Ryu, et al., 2009). PLR distribution is also

not found in studies where NSCs are compacted into neurospheres (Shetty, et al.,

2008) or restricted in distribution by nature of being enclosed in a capsule (Imitola, et

al., 2004, Park, et al., 2002). An exception to this observation is a neurosphere study

by Radojevic and colleagues (Radojevic and Kapfhammer, 2009). The amount of NSCs

injected in many rodent studies equals or exceeds 1 x 105 cells. In our study, we chose

to inject 5 x 105 NSCs because this amount is proportional by weight to the amount of

NSCs injected in the first and currently only FDA-approved clinical trial for transplanting

human stem cells into humans (Stem Cells Inc). In these studies, 1 to 2 billion NSCs

were injected into children suffering from Batten's disease. This amount recently

passed phase 1 safety trials (American Association of Neurological Surgeons

Annual Meeting 2010). Another reason to inject greater than 1 x 105 NSCs is that brain

transplants in rodents and humans are generally characterized by 40-97% loss of

transplanted cells (Bjorklund, et al., 2003, Lundberg, et al., 1997, Olstorn, et al., 2007,









Raedt, et al., 2009, Tang, et al., 2008). Therefore, the amount of NSCs to be injected in

future studies that explore stem cell potential in mice and humans will likely yield PLR

distributions. The importance of PLR's in these future studies is further underscored by

the persistence of PLR distributions in two month engraftments within mice modeling

Alzheimer's disease and in one year engraftments within non diseased mice.

Research and Clinical Relevance of Paths of Least Resistance

The PLR's described here inherently restrict cell distribution. This presents is a

possible pitfall in the exploration of NSCs for use in cell replacement or drug delivery.

For instance, studies by Prajerovaet et al., resulted in corpus colossal PLR distribution

despite transplantation with cortical coordinates. Alternatively, one can view PLR's as

an opportunity to attain consistent engraftment patterns in the course of multiple

transplants. The horizontal nature of hippocampal PLR's favor precision by buffering

against stereotaxic error. In our experience, deposition across a range of hundreds of

micrometers in the medial/lateral and anterior/posterior plane yielded similar

hippocampal engraftments.

We envision targeting of various PLR's for unique research and clinical aims. The

PLR of the hippothalamic fissure appears to support dorsal directionality of cell

migration. The extensive process extension by cells in this region suggests integration

into the dentate architecture.

On the other hand, cells in the white matter PLR appear stationary for at least two

months in AD mice. Genetically modified NSCs stationed in this PLR have access to

both the cortex and hippocampus for long-term infusion of therapeutic molecules.

However the dense layer of CA1/CA3 neurons may restrict movement of infusate. More









studies are needed to understand the effect of dense cell layers on the distribution of

candidate therapeutic molecules.

To further understand the movement of cells injected into the hippocampus, MRI

technology may be used to visualize quantum dot labeled NSCs (T. Zheng, personal

communication). This technology can possibly reveal coordinates for bilateral

engraftment with unilateral cell injection. Our observation in one mouse provides

evidence for a pathway, perhaps through the 3rd ventricle, for contralateral colonization.

This pathway may allow for less invasive brain surgery on humans without sacrificing

therapeutic distribution.

MMP9 Associated Changes in Graft Size

We find larger NSC graft sizes are associated with overexpression of MMP9. This

result demonstrates that ex vivo genetic modification of NSCs can result in changes in

engraftment pattern. To our knowledge, this finding is novel and may be an important

observation to consider in future studies that aim to deliver other candidate therapeutics

through NSC overexpression and transplantation. It is possible that MMP9 is making

the graft site more survivable by degrading toxic molecules such as A3. The negative

correlation of graft size to surrounding AP3 plaques suggests this. However, it may be

that MMP9 is modifying the ECM in a manner that enhances cell survival. Studies of

MMP9 have largely concentrated on its association with cancer (Chambers and

Matrisian, 1997, Lubbe, et al., 2006, Moll, et al., 1990). However, a growing body of

work demonstrates the function of MMP9 in important biological functions. Of particular

relevance to this study is recent work demonstrating the need for MMP9 in the migration

of endogenous NSCs in vitro (Wang, et al., 2006) and to sites of ischemia (Kang, et al.,









2008). It is likely that there exists a spectrum of MMP9 physiology. At one end of this

spectrum, MMP9 enables metastasis of cancer cells (Chambers and Matrisian, 1997,

Lubbe, et al., 2006, Moll, et al., 1990). At the other end, MMP9 facilitates endogenous

NSC function in stroke (Kang, et al., 2008), modulates AP3 (Yin, et al., 2006), and is

associated with larger NSC grafts in a pathological environment, as demonstrated here.

As mentioned above, multiple laboratories report 40% to 97% loss of NSCs following

transplantation in animals and humans (Bjorklund, et al., 2003, Lundberg, et al., 1997,

Olstorn, et al., 2007, Raedt, et al., 2009, Tang, et al., 2008). MMP9 overexpression

may be a novel method to address cell loss.

We aim to perform further studies to determine whether MMP9 associated

changes in engraftment occur in non transgenic mice. A positive result would mean that

the effect of MMP9 on NSC survival is more associated with tissue remodeling than

mitigating AP3 pathology.

Concluding Comments

In summary, the data presented here provides evidence for consistent NSC

engraftment in three regions within the hippocampus. These regions are characterized

by a change in neural density and therefore provide paths of least resistance for the

flow of cells at the moment of injection. The tightly, bound and layered nature of the

hippocampus is not unique within the brain; the olfactory bulb and cerebellum also

share similar structural organization. Therefore further studies aimed at characterizing

putative PLR's may yield more specific targeting of NSCs for cell restoration or drug

delivery throughout the brain. We find that MMP9 is associated with significant

increases in the size of transplanted NSCs. This data suggests that MMP9 may be

used to enhance the robustness of grafts for cell restoration and drug delivery studies.










Contra Ipsi
[DT_, 11.baI,"(' i co la [DAT-PJll'~I/b l [(microg1a]






A.' B: 25um
1AP11a ".,l m."~ i [ ro] l ia)~ DAPr/lbal"?ll [mirog ]a
i[ Graft

A rels10

IAr"




APplqu


CTX








Corpus
Callosum







HIPPO


Figure 3-8. lbal expression and lack of A3 migration by engrafted NSCs.
Representative images of microglia on both hemispheres indicate increased
lbal reactivity in the region of GFP NSC engraftment (n=3) (D, arrow). Cells
have an elongated phenotype that is similar to GFP immunostained cells (Fig
3-3F). This morphology is distinct from that of ramified, quiescent microglia
dorsal to the graft site (B) or AP3 reactive glia (D, arrowheads, E, F). We did
not observe clustering of NSCs around AP3 plaques such as the one located
less than 20um from this engraftment (D, arrowheads). Control images of
adjacent sections stained with only 20 antibody indicated that native GFP was
not a source of significant green fluorescence background (data not shown).










6E10 (AP) GFAP (astrocyte)
CTX
CTX


cc

HIPPO


Figure 3-9. GFAP expression by engrafted NSCs. GFAP immunoreactivity is
colocalized with NSC engraftment (n=3). Cells have an elongated
morphology similar to that seen with lbal immunoreactivity (Fig 3-8D). NSCs
did not migrate to nearby AP3 plaques. AP3 transgenic mice shown here were
sacrificed two months following NSC transplantation. New AP3 production was
genetically halted in these mice during the survival period. Control images of
adjacent sections stained with only 20 antibody indicated that native GFP was
not a source of significant green fluorescence background (data not shown).




































Figure 3-10. Paths of least resistance. Cells injected into d were consistently found
along a horizontal stretch defined by arrows: a, b, c (A). Cells here appear
stationary. This result is different from injections into a,' b,' c' because
deposition of cells here can result in engraftments in d' (B). Coordinates that
target a, may yield bilateral engraftment through the 3rd ventricle. Images
modified from Paxinos Mouse Brain Atlas.









CHAPTER 4
THE EFFECT OF NEURONAL STEM CELLS ON AB PATHOLOGY AND THEIR
UTILITY AS A THERAPEUTIC DELIVERY VEHICLE FOR THE AB DEGRADING
PROTEASE, MMP9

Introduction

The use of stem cells that are engineered to produce molecules of therapeutic

value holds much promise for the treatment of AD. The goal of this study was to

explore their use for long-term delivery of candidate therapeutic molecules into mouse

models of AD. Our aim was essentially to establish a foundation for ex vivo modified

stem cells in the emerging clinical field of cell replacement therapy for AD.

We used SEZ derived NSCs because they are the only somatic cell that we know

of that is endogenous to the brain, can be cultured for extended periods of time and

may migrate millimeter distances when transplanted (Scheffler, et al., 2005, Walton, et

al., 2006a, Zheng, et al., 2006). Other CNS cells such as neurons and glia are not

suitable for ex vivo manipulation because they can only be cultured for a few weeks.

Notably, direct injection of isolated microglia does not result in long-term grafts (G.

Marshall, personal communication).

To show the feasibility of genetic manipulation of NSCs to overexpress AP3

disrupting factors, we transduced them with lentiviruses carrying transgenes for

secreted Metalloprotease 9, membrane bound Heparanase and membrane bound

Neprilysin. Because cultured NSCs are mitotic cells, lentiviral transduction was

particularly suitable for this study. This is because these viruses integrate their genetic

load into the host's chromosome and thus obviate [episomal] gene dilution with each

cell division. The high efficiency of lentiviral transduction results in a transgene









expressing cells (Blits, et al., 2005). These cells were further purified using fluorescent

activated cell sorting (FACS).

We focused most of our studies on MMP9 over other competing anti-Ap candidate

molecules because 1) it is naturally secreted, 2) it is overexpressed by astrocytes

reactive to Ap plaques, and 3) it has previously been demonstrated to degrade Ap

plaques in situ (Yan, et al., 2006, Yin, et al., 2006).

Methods

Lentivirus Construction

3rd generation self-inactivating lentiviruses (Dull, et al., 1998, Englund, et al., 2000)

containing cDNA's for human MMP9, HPSE & Nep were created using the pCDH

cloning and expression system (SBI, Mountain View, CA). Briefly, MMP9 (2.54kB

insert) and HPSE (1.8kB) cDNA's contained within pCMV6-XL4 plasmid vectors (4.7kB

empty vector) were purchased from Origene (Rockville, MD). We already had

Neprilysin cDNA within the pCDNA 3.1 vector. All three constructs were amplified to

provide enough material for removing and then transferring the cDNA insert's into 3rd

generation GFP containing lentiviral vectors. The pCMV6 vector has Not1 restriction

sites flanking the MMP9 and HPSE cDNA's, so we isolated MMP9 and HPSE inserts

using Not1 restriction enzyme digest (New England Biolabs, Ipswich, MA). Nep cDNA

was isolated from the pCDNA 3.1 vector using Nhel and Not1 restriction enzyme

digests. These inserts were ligated into the pCDH lentiviral vector (SBI, Mountain View,

CA). Viruses were packaged by transiently co-transfecting HEK293 cells with the pCDH

construct (containing MMP9, HPSE, Nep inserts) along with plasmids for the creation of

lentiviral structural and integration proteins, and VSV-G pseudotype (courtesy of S.S.

Rowland). The VSV-G envelope protein enables lentiviruses to transduce a broad









range of mammalian cells. Viruses packaged by the 293 cells were concentrated to ~1

x 1011 particles/ul by centrifugation and minimal dilution.

Lentivirus Transduction

1 x 105 trypsinized NSCs and 293FT cells from confluent cultures were

resuspended in 100ul dPBS. These cells were transduced by exposure to 6ul volume

of virus concentrate for 1 hr in 370C, with agitation every 10-15 min. This ratio of virus

to cells was found to most optimally and consistently yield NSCs expressing the GFP

reporter.

Fluorescence Activated Cell Sorting

Transduced cells were grown to confluent cultures in T25 flasks. Cells were

trypsinized and resuspended in 5mls of PBS with 2% fetal bovine serum. GFP intensity

cut-off points of A.U. 103 and 104 were used to obtain cells varying in transgene

expression.

Isolation of NSCs

The protocols for isolating NSCs are contained in the literature (Marshall, et al.,

2006, Zheng, et al., 2006). NSCs were cultured as described in Chapter 3.

Transplantation into Amyloid Beta AD Mice

The Line 85, which constitutively overexpress human AP3, have been previously

described to model the AP3 physiology that is a hallmark AD pathology. Surgeries were

performed as described in Chapter 3. Briefly, a Hamilton 33 gauge needle (Hamilton

Company, Reno, NV) was then loaded with NSCs prepared at ~5 x 104 cells/ul in lx

dPBS. The cells were derived from a trypsinized and pelleted monolayer of NSCs,

washed twice with 200ul lx dPBS and diluted to the appropriate volume using a

reference cell count done on a hemacytometer. Cells were deposited at -2.0mm, -









2.3mm and -2.5mm into the hippocampus. 1.25 x 105 cells were deposited at -2.0mm

and -2.3mm, while 2.5 x 105 cells were deposited at -2.5mm. Typically, 4-8ul total

volume was deposited at the rate of 0.25ul per 15 seconds dependantt on cell

concentration). 20um coronal sections were processed from whole brains and stored in

anti-freeze media at -20C until further processing.

Immunochemistry

4% paraformaldehyde fixed cells or tissue sections processed for

immunofluorescence in solutions containing 0.1% Triton-X, 10% goat serum in 1x PBS.

Primary antibodies used in this study include copGFP (1:2000, Evrogen, Moscow,

Russia), anti human MMP9 Clone 56-2A4 (Abcam, Cambridge, MA), anti human MMP9

Clone 6-6B (EMD, Gibbstown, NJ), anti human MMP9 Clone 7-11 c (Santa Cruz

Biotech, Santa Cruz, CA), anti mouse MMP9 (courtesy of R. Senior, Washington

University, St. Louis, MO), anti-Ap3 6E10 (1:2000, Signet, Dedham, MA), microglial

antigen lbal (1:1000, Wako, Richmond, VA), astrocyte antigen GFAP (1:1000, Dako

Corporation, Carpinteria, CA). Cells were rinsed and incubated with goat secondary

antibodies Alexa 488, 568 (Invitrogen, Carlsbard, CA). Cells were photographed with

an Olympus DP71 camera mounted on an Olympus BX60 microscope. For western

blot analysis, samples were diluted in Laemmli sample buffer containing 2% sodium

dodecyl sulfate and loaded in 4-20% TG-SDS gels (Invitrogen, Carlsbad, CA) for

standard SDS-PAGE. Immunoblots were probed with anti mouse/human MMP9 Clone

38898 (1:5000 Abcam) and Abcam anti human MMP9 Clone 56-2A4 (1:500 (Abcam).

Gel blots were photographed using a Fugi imaging system (Fugifilm Life Science,

Stamford, CT).









Analysis of AP Plaque Number

Image Pro Plus software (MediaCybernetics, Bethesda, MD) was used to quantify

intensity over background for images of plaques in coronal sections (Dolev and

Michaelson, 2004, Podoly, et al., 2008). Because transplanted cells largely settled in the

subcortical and corpus callosum white matter tracks, we focused on a region of interest

(ROI A) that included the dorsal hippocampus below the site of engraftment and the

cortex above the site of engraftment (1.25mm x 2mm. To be consistent across mice,

the vertical boundary for ROI A was the meeting of the dentate arms. The lateral blade

of the dentate gyrus was the horizontal boundary. Only sections containing needle

tracks were analyzed. Comparison was done on the equivalent region on the

contralateral side. For statistical comparison, contralateral areas were analyzed using

paired, two-tailed Student's t-test using Microsoft Excel. An unpaired, two-tailed

Student's t-test was used for comparison between transplants of GFP NSC and MMP9

NSC transplants. A p-value of <0.05 was considered statistically significant.

Reverse Transcriptase Polymerase Chain Reaction (RT-PCR) of Human MMP9

Human immortalized 293FT cells, mouse immortalized 3T3 cells and NSCs were

transduced in simultaneous 100ul reactions that contained 1 x 105 cells and 6ul of

MMP9 virus concentrate as described above. Cells were allowed to grow for 6 days

before being lysed with Trizol (Invitrogen, Carlsbard, CA). Because there were cell-type

specific differences in growth rate, lysates were normalized for total protein using a

bicinchoninic acid assay (BCA) (Peierce, Rockford, IL). RT-PCR reactions utilized

primers specific to P3-actin (400bp) and human MMP9 (600bp). Reagents from a

Superscript One-Step RT-PCR System were used according to manufacturer









instructions (Invitrogen, Carlsbard, CA). Because of limited supplies of virus, we could

not repeat transduction of cells. However, RT-PCR reactions were repeated 3x.

In Vitro MMP Gelatinase Activity

Conditioned media (C.M.) was collected from transduced and mock transduced

NSCs 6 days after transduction (~75% confluent). To eliminate contamination by

floating cells in our experiments, C.M. was centrifuged at 1000 x g for 10 min before

being applied to 50ug/ml DQ gelatin (Invitrogen, Carlsbard, CA) at a 1:4 dilution.

Reactions were incubated overnight at room temperature. Fluorescence increases

upon degradation of DQ gelatin due to release of quenched fluorescein. To show

gelatin degradation is due to MMP activity, the general MMP inhibitor 1, 10,

phenanthroline (1, 10 PNTL) was applied at a 0.8mM concentration. Collagenase

activity was used as a positive control. Triplicate reactions of each sample were

measured with the excitation/emission spectra of 495nm/515nm using a

spectrophotometer (Bio-Tek, Winooski, VT).

In Situ MMP Gelatinase Activity

Fixed brain sections were treated with a protocol modified from Yan et al. (Yan, et

al., 2006). Briefly, a PBS solution with 0.1% Triton-X100 and 1% agarose (w/v) was

made homogeneous by 2 to 4 rounds of short 10 sec microwave pulses and brief

vortexing. The following additions were made to this solution in order to visualize or

inhibit MMP activity: DQ gelatin (100ug/ml, warmed to 37C), DAPI nuclear counterstain

(1:1000), 1,10 PNTL (0.8mM), and EDTA (20mM). 300ul of DQ gelatin mixture was

quickly added per section. After an overnight incubation in room temperature, green

fluorescence of sections was captured with an Olympus DP71 camera mounted on an

Olympus BX60 microscope.









Chemical Activation of secreted MMP9

Organomercurial compounds such as p-am inophenylmercuric acetate (AMPA)

induce autoactivation of MMPs including MMP9 (Ramos-DeSimone, et al., 1999, Yan,

et al., 2006). A 1mM concentration of AMPA was added to conditioned media from

transduced and mock transduced NSCs. Commercially available MMP9 proenzyme

(Perkin Elmer, Waltham, MA) was similarly treated as a positive control. The reaction

volume was incubated overnight at 370C before Western blot analysis with antibodies

specific to human MMP9.

Results

GFP NSCs were transplanted into the hippocampus of four mice symptomatic for

AP3 pathology using coordinates that resulted in primary corpus callosum engraftment

and in some mice, secondary hippocampal fissure engraftment. This pattern of

engraftment allowed us to ask whether NSCs have effects in the hippocampus and the

cortex. After a month survival period, mice were harvested and their brains processed

with antibodies reactive to AP3 and GFP. Needle track containing sections contained the

largest engraftments. We analyzed five to seven such sections per animal for amyloid

burden (AP3 plaque number) with Image Pro Plus software previously used in a similar

capacity (Dolev and Michaelson, 2004, Podoly, et al., 2008). A region of interest (ROI

A) proximal to white matter engrafted cells was studied (see Chapter 3). This area

included the cortex above the white matter tracks and the dorsal hippocampus below

(Fig. 4-1A). Compared to the equivalent contralateral region, we observed a 26.4%

(p=0.04, paired t-test) reduction in AP3 plaque number in this ROI. Though there was

variability in the number of AP3 plaques found per section (Fig. 4-1 C), all mice had

reduced AP3 plaque numbers (Fig. 4-1 D).























B C 120 D 100

100* 0
Contra lOOi
o


6 HIPP 50
0 30
00




C ontra Ip Contra Ips
GFP Only GFP Only
CTX & HIPPO CTX & HIPPO

Figure 4-1. Transplantation of NSCs is associated with reduced amyloid burden.
Example of A3 immunostained section shows reduced A(3 plaque numbers in
the region of engraftment of 103 GFP NSCs a month post-transplantation (A).
Representative image demonstrating the sensitivity of software used to count
Ad3 plaques in sections (B). Plot of results where each dot indicates the
number of Ai3 plaques on a section. Compared to the contralateral side, the
number of Ap3 plaques was reduced in ROI A by 26.4% (p=0.03, paired t-test)
(C). All mice had reduced Ap3 plaque numbers (n=4) (D). *, p<0.05

To understand this result, we have performed preliminary experiments to

determine whether 1) NSCs were directly clearing Af3, 2) NSCs were producing

diffusible factors that cleared Ap3 or 3) NSCs were interacting with host cells to clear A3.

Pulse-chase experiments similar to those described in Chapter 2 determined that NSCs

were capable of internalizing 0.52ng/ml of A342 in a period of 3 hours (n=4), (Fig. 4-2A).

In comparison, neonatal mouse microglia internalized about twice as much A342











(1.1 ng/ml), but then expelled most of what was internalized (see Chapter 2) (Njie, et al.,


2010). Interestingly, NSCs appear to process AP3 differently as internalization of AP3 was


not followed by expulsion (Fig. 4-2B). This suggests that NSCs biophysically degrade


AP3 in vitro and possibly in vivo.


A ** **
Intracellular AOi
1.00 Mock
0.90 0 Ohr
i 3hrs
0.80 16hrs
0.70
E60
E 0.60
.E. 0.50
0.40
< 0.30 T
0.20
0.10
0.00 NSC
NSC
n=4
B
1.00 **
0.90 Expelled A|3
0.80 Mock
S3hrs
0.70 m 16hrs
"- 0.60
S 0.50
'" 0.40
< 0.30
0.20
0.10
0.00
Naive NSC
media n=4


Figure 4-2. Fate of A3 internalized by NSCs. GFP NSCs were exposed to AP342 in
pulse-chase experiments (n=4). GFP NSCs internalized 0.52ng/ml within
3hrs (A). In contrast to microglia (Fig. 2-9), GFP NSCs did not expel
internalized AP342 (B). This suggests NSCs engage in biophysical
degradation of AP342 following phagocytosis. The ability of GFP NSCs to
degrade AP342 provides a possible mechanism for reductions in in vivo AP3
plaques following transplantation (see Fig. 4-1). Mock data (gray) represents
experiments without the presence of cells to control for non-specific
adherence of AP3 to culture wells. Detection of AP342 requires the presence of
both NH2 and COOH terminals of A342, thus only intact AP342 peptides are
quantified in the above experiments. *, p<0.05; **, p<0.01.











We found that NSC monolayers are reactive to an antibody against mouse MMP9

(n=3), (Fig. 4-3A). In vivo, astrocytes reacting to AP3 plaques overexpress mouse MMP9

as part of a general glial reaction to AP3 pathology (Fig. 4-3B). It may be that MMP9 and

other NSC derived diffusible factors contribute to the clearance of A3. The lack of

MMP9 specific chemical inhibitors makes this difficult to confirm in vivo. However,

infusion of antibodies known to stop MMP9 gelatinase activity is a possible alternate

approach that we may explore in the future.

Previously, degradation of DQ gelatin has been used to show MMP activity around

AP3 plaques (Yan, et al., 2006). We verified this finding in mice that have continuous

overexpression of AP3 (Fig. 4-4). This result suggests endogenous pathways to regulate

AP3 plaques are active. However, recent work from our lab has shown that once formed,

AP3 plaques are not removed by endogenous processes (Jankowsky, et al., 2005) -

which include MMPs. We therefore wondered whether this is due to an overwhelming

rate of amyloid formation. To shed light on this question, we stopped new AP3

production in mice with amyloid precursor protein under a tetracycline regulatory

element. Following a month, we found similar levels of MMP activity around AP3 plaques

(n=3), (Fig. 4-4B, C). This result suggests that endogenous processes continue to

regulate actively AP3 plaque burden in the absence of new deposition. Since this activity

does not result in eventual AP3 plaque clearance, it is perhaps the level of MMP anti-Ap3

activity rather than the rate of new amyloid deposition that may be responsible for the

permanence of AP3 plaques.









We decided to genetically modify NSCs to increase the amount of MMP9 in mice

expressing AP3 pathology. To do so, we transduced NSCs with VSV-G serotype

lentiviruses carrying the human MMP9 transgene. These lentiviruses, which carry GFP

under a separate promoter (Fig. 4-5A) and are able to transduce mammalian cells,

conferred GFP fluorescence to NSCs ~3 days post-transduction and 293FTs within 2

days post-transduction. No remnant virus genome was detected in NSC cultures

following 2 washes (Fig. 4-5C). Enrichment of NSCs (described below), yielded

cultures reactive to two antibodies specific to human MMP9 (Fig 4-5D). 293FT cells

produced noticeably more GFP compared to NSCs (Fig. 4-6A, B). This result occurred

in 3 independent transductions with independent lots of virus. Several possibilities were

considered to explain this observation. These included a species effect, an immortal

cell line effect, NSC quiescence, transcription versus translation and lack of uniformity in

transduction conditions. To shed light on some of these possibilities, human [immortal]

293FT cells, mouse NSCs as well as mouse [immortal] 3T3 cells were transduced in

parallel and analyzed for mRNA transcript production. We were unable to include

human NSCs in this experiment due to the limited availability of these cells. This

experiment was performed once due to a limited supply of virus, however RT-PCR

reactions were repeated three times to reveal variability in analysis. Our results

indicate that mouse 3T3 immortal cells had similar levels of transgene mRNA as NSCs

(Fig. 4-6C, D). This result suggests that our observations in (A) are not unique to

NSCs. Perhaps translational properties unique to 293FT's or human cells in general,

may account for enhanced transgene overexpression in 293FT cells.









Though transduced NSCs do not match 293FT cells in transgene production, they

nonetheless express human MMP9 which complements the endogenous mouse MMP9

they produce. These cells therefore overexpress total MMP9. Similar to previous

reports (Ramos-DeSimone, et al., 1999), transgene directed MMP9 mRNA is translated

to a secreted 92kDa zymogen (proenzyme). Mock treated NSCs have intrinsic MMP

activity on gelatin that is chemically inhibited by 1,10 phenanthroline (Fig. 4-7A). The

secretion of MMP9 zymogen by transduced NSCs is associated with a 3.5x increase in

gelatinase activity (Fig. 4-7A). We did not observe degradation of A342 in conditioned

media from 103 MMP9 NSCs (103 cells, data not shown). However, the 39kDa catalytic

subunit of MMP9 (ctMMP9) reduced AP3 detectable by Western blot by approximately

75% in overnight reactions (Fig. 4-7B). The amount of ctMMP9 needed to achieve this

degradation was less than ideal: ~20:1 ratio of enzyme to substrate. However the

observation that truncated MMP9 is the isoform capable of degrading AP3 is consistent

with previous reports (Yan, et al., 2006). This suggests in vivo autoactivation of NSC

produced MMP9 is required for anti-AP3 activity.

To determine if NSC produced MMP9 is capable of undergoing such

autoactivation, conditioned media was treated to p-aminophenylmercuric acetate

(AMPA). This organomercurial compound disrupts the interaction of an unpaired, N-

terminal cysteine with the zinc ion of the MMP9 catalytic center. Such disruption

initiates autoproteolytic N-terminal shedding that generates the MMP9 isoform shown to

degrade AP3 in the literature (Yan, et al., 2006). We applied a 1mM concentration of

AMPA to conditioned media and detected only the truncated (active) form of MMP9

following 24hrs (100% conversion) (Fig. 4-7C). We therefore concluded that transgene









directed MMP9 secreted by NSCs is capable of undergoing autoactivation. There are a

number of proteases thought to mimic the action of AMPA in vivo (Ramos-DeSimone, et

al., 1999). On the other hand, tissue inhibitors of metalloproteases (TIMP1-4) inhibit

MMP9 but are limited by stoichiometric binding (Ramos-DeSimone, et al., 1999). TIMP

levels have been shown to increase in AD (Lorenzl, et al., 2003, Peress, et al., 1995)

and in one mouse model of AD (Hoe, et al., 2007). However, the concentration of

TIMPs in mice with AP3 pathology has not been quantified. We therefore aimed to

introduce as much MMP9 into the brain as possible in order to saturate endogenous

activators of MMP9 and overcome inhibitors of MMP9.


NSC








Transgenic
mouse


6E1 (A) OM


Figure 4-3. NSCs express endogenous mouse MMP9. NSCs were reactive to
antibodies against mouse MMP9 (n=3) (A). It is possible that transplant
derived mouse MMP9 adds to the concentration of MMP9 secreted by
endogenous astrocytes reacting to AP3 plaques (n=2) (B, arrows).

Cultures of transduced NSCs had a non-uniform distribution of GFP. This

suggested that in our population of transduced cells, transgene copies were varied from

cell to cell. Peripheral drug treatment paradigms typically tailor dosage to optimally

balance between desired effects and side effects. To emulate this preclinically, we


MOMMP9









used FACS technology to take advantage of the differential transduction of NSCs by

selecting for cells whose GFP intensity was A.U>103 or 104 (Fig. 4-8). The conditioned

media from the resulting cultures was analyzed with Western blot using an antibody that

detects commercially available human proMMP9 and verified with an antibody cross-

reactive to both human and mouse MMP9. In vitro, a confluent layer of 103 cells

secreted 0.51 ug/ml of human MMP9 in a T25 flask with 5mls of media over 3 weeks.

104 cells treated similarity, secreted 0.66ug/ml over only 1 week (Fig. 4-9A). Non-

transduced and non FACS sorted cells had no detectable secretion of human MMP9.

We project that over 4 weeks, 103 and 104 cells will secrete 0.7ug/ml and 2.7ug/ml of

MMP9, respectively (Fig. 4-9B). This demonstrates that we can create NSC cultures

with low and high doses of MMP9 with FACS technology.

This result, though promising, is limited by yield. A confluent T25 flask holds a

monolayer of ~5 x 106 NSCs. The average yield of 103 NSCs is 8.9 x 105 cells/flask.

This is reduced 6-fold, or 1.4 x 105 cells/flask for 104 cells (n=7) (Fig. 4-8E). 104 cells

are only 4.5% +/-0.8% (n=8) of the original population of transduced cells (Fig. 4-8D).

Further selection exponentially diminishes yield. Therefore, we are confident that our

enrichment protocol approached theoretical limits of selection. We reasoned that these

highly enriched MMP9 expressing NSCs stand the best chance of overcoming

endogenous MMP9 repressors and acting against A3. However, MMP9 expression can

lead to negative consequences (Chambers and Matrisian, 1997, Lubbe, et al., 2006,

Moll, et al., 1990). Therefore, we first worked with 103 cells in transplant studies in

hopes of establishing safety as well as efficacy.









MMP9 expressing 103 NSCs were injected into five AP3 transgenic mice. We

confirmed that NSCs continued MMP9 overexpression a month post-surgery (Fig. 4-10).

Graft derived MMP9 had N-terminal shedding that is associated with activation (Fig. 4-

10E). We have ongoing studies to determine the endogenous proteins that activated

graft derived-MMP9. Analyses of in vivo MMP9 overexpression utilized antibodies

specific to human MMP9 for immunocytochemistry (n=3), 1 um z-plane confocal

histology (n=2), and western blot (n=3).

Constitutive A A halted 1 month A halted 1 month















N n trans enlc l EDTA 1 10 PNTL* general MMP inhibitor)








Figure 4-4. Endogenous MMP activity in mice with AP3 plaque burden. Sections
incubated overnight with DQ gelatin demonstrate persistent MMP activity on
AP3 plaques 4 weeks after cessation of AP3 production (n=3) (A-B). Controls
show DQ gelatin fluorescence is specific to AP3 plaques (C), does not occur in
non transgenic mice (D), and is dependent on MMP activity (E, F).

Mice receiving MMP9 NSC transplants were examined for AP plaque burden with

methodology described above for GFP NSCs. A sham injection on the contralateral









side was performed in order to control for the effect of needle injury on the cell injected

side. Compared to the needle injury on the contralateral side, we did not notice

additional death of neuronal cells or otherwise abnormal cytoarchitecture in the side of

the brain containing MMP9 NSCs. We observed a 28.6% (p=0.03, paired t-test)

reduction in AP3 plaque number within ROI A (Fig. 4-11A, B).

MMP9 NSC and GFP NSC transplants had comparatively similar percent

reductions in AP3 plaque number (Fig. 4-11 C). Since AP3 plaque number reductions were

statistically significant in both groups, we asked whether there was change in the

amount of AP3 plaques cleared in mice receiving GFP NSCs and MMP9 NSCs. We

subtracted the averaged contralateral AP3 plaque number in each animal from the

averaged ipsilateral AP3 plaque number and then compared across the two transplant

groups. MMP9 NSC transplants were associated with clearance of 31% more AP3

plaques compared to GFP NSC transplants (p=0.47, unpaired t-test), (Fig 4-11 E). This

trend, though non-statistically significant, is promising. It should be noted that mice

hosting MMP9 NSC transplants were a month older and consequently had 25% greater

AP3 plaque numbers (Fig. 4-11 D). The lack of a uniform baseline of AP3 plaque number

due to age-related changes is confounding. The increased engraftment of MMP9 NSCs

relative to GFP NSCs further complicates interpretation. Though our data indicates a

trend of more AP3 plaque clearance in 103 MMP9 NSCs, further experimentation with 104

cells that more robustly express MMP9 is needed with special attention to address the

confounding variables described above.

Discussion

In this study, we aimed to determine whether NSCs could be used to reduce AP3

pathology either without modification or as a vehicle to deliver molecules that have















PCMV6-XL4
470 byp


RSV 5'ITR



human pCDH-CMV-MCS- P r
MNP9 PC ----O EF I -(Puro/copCFP) CMV
transgene
SV40 O J 1 MCS
SV4o poi A promoter
S (+) hMMP9 P op GFP


POW (+/-) huMMP9 PERu copGFP


- 2.5kB


1
..m~


2


3 4


Figure 4-5. Genetic modification of NSCs for MMP9 overexpression. Vector diagram of
MMP9 transgene illustrates subcloning of human MMP9 cDNA from a pCMV
vector (4.7kB) to a pCDH vector (7.5kB). The pCDH vector contains plankton
derived GFP reporter gene and components necessary for the packaging of
self-inactivating lentiviruses. The dual promoter design of pCDH meant
MMP9 was independently driven by the CMV promoter while copGFP was
driven by the EF1 promoter (A). Restriction enzyme digests yielded MMP9
transgene of correct size (2.5kB) at the end of the cloning process (B, Lanes
3 & 4). Real-time PCR showed no evidence of viral genome in media in
transduced cultures washed with dPBS (C). Transduction resulted in GFP
reporter gene expression (D, E middle panels). GFP NSCs (D) and MMP9
NSCs (E) differed only in MMP9 expression. GFP and MMP9
immunoreactivity did not always colocalize (D, compare left and middle
panels).

promising anti-AP3 qualities. Unlike our results with microglia (see Chapter 2), NSCs in

vitro biophysically degrade AP3. Transplantation of NSCs is associated with close to a

one-third reduction in AP3 plaque numbers in the cortex and hippocampus. Genetically









modifying NSCs to overexpress MMP9 is associated with a trend of more clearance of

AP3 plaques; however a statistically significant difference was not found. It is possible

that greater overexpression of MMP9 may yield more robust reductions in AP3 plaque

numbers. We developed a method to enrich for MMP9 production in NSCs cultures by

utilizing FACS selection to take advantage of the plurality of gene dosage associated

with lentiviral transduction. This positive selection scheme approaches the theoretical

limit of enrichment and has resulted in improvement from initially undetectable levels of

MMP9 in NSC conditioned media to detection of ug/ml amounts of MMP9. We were

also able to demonstrate enrichment of heparanase overexpressing NSCs. Therefore,

this method of enrichment is generally applicable to emerging therapies using NSCs.

Our future studies will concentrate on NSCs four-fold more enriched that those utilized

for the transplant studies described in this report.

Endogenous MMP Activity

Our current findings show that the brain continues to use MMPs to regulate pre-

existing AP3 plaques suggesting that endogenous processes do not cease actively

mitigating AP3 plaque burden. These processes appear insufficient as AP3 plaques

persist six months after genetic cessation of new AP3 production (Jankowsky, et al.,

2005). However, once formed, AP3 plaques resist biophysical removal despite

persistence of these endogenous efforts. This underscores the need for exogenous

intervention to clear AP3 plaques in mice and perhaps in humans.

Ap plaque Burden is Lowered Following NSC Transplantation

We introduced GFP NSCs and MMP9 NSCs into the brain and found both cell

types reduced AP3 plaque numbers. This result suggests that transplanted NSCs or

uncharacterized biology associated with their engraftment, reduces the prevalence of









otherwise permanent AP3 plaque burden. Our findings contradict recent work by Blurton-

Jones and colleagues (Blurton-Jones, et al., 2009). In their observations, AP3 content

was not reduced in hippocampal injections of NSCs. The discrepancy with our findings

may perhaps be explained by unknown factors unique to their model --a triple-

transgenic mouse with tauopathy, versus our models which strictly model amyloidosis

(Oddo, et al., 2003). Secondly, the cells used by Blurton-Jones et al., were created

without region-specificity; they were derived from neurospheres obtained from whole

brain homogenates. It is possible that the SEZ origin of our NSCs enabled unique anti-

AP3 effects. If true, our finding may mean that some but not all populations of NSCs

exhibit anti-Ap3 activity.

NSCs as a Platform to Deliver MMP9 and Other Candidate Therapeutics In Vivo

We hypothesized that constitutive MMP9 secretion by NSCs proteolytically will act

on fibrillar AP3 and confer further decreases in AP3 plaque burden. Since ex vivo

manipulation of NSCs for use as delivery vehicles of therapeutics is a field still in its

infancy, our work sheds light on several important questions regarding NSCs in this

paradigm. These questions include: do NSCs retain stem cell qualities if genetically

manipulated in vitro? Do NSCs express active, transgene directed proteins of interest?

Are such proteins produced in sufficient amounts? Can NSCs be genetically modified

with a repertoire of candidate molecules? New molecules with interesting anti-Ap3

effects continue to be published. For instance MT1-MMP has recently been shown to

degrade AP3 in vitro (Liao and Van Nostrand, 2010). Our observations on genetically

manipulating and purifying NSCs and the in vivo effects of MMP9 NSCs demonstrate

potential positive effects as well as pitfalls that may apply to emerging candidate

molecules.














Transduced
293FT






Transduced
NSC




C





Human MMP9-_.
P-actin--


/ 2 3 4 5/ 6

A// ^A 4 ^A


_____ _-__ s"" ]IIH
Mouse Human 0
NSC 3T3 293FT


Figure 4-6. Cell type differences in transduction efficiency. 293FT cells produced
significantly more GFP protein relative to equivalently treated NSCs (n=2) (A,
B). Semi-quantitative RT-PCR analysis of transduced cells indicated mouse
3T3 immortal cells had comparable amounts of MMP9 mRNA as NSCs,
suggesting low levels of transgene expression is not NSC specific (C). This
comparative transduction experiment was not investigated further due to the
limited supply of lentiviral stocks. Error bars reflective RT-PCR analyses (D).

Characteristics of NSCs overexpressing transgenes

Regarding the maintenance of 'stemness' following genetic manipulation, our

transgene receiving NSCs continued expression of BIII-Tubulin (data not shown), a

immunological marker commonly associated with precursor cells (Walton, et al., 2006a).

Interestingly, transplanted NSCs in white matter tracks do not stain for BIII-Tubulin (see

Chapter 3). Studies have shown that microglia are associated with the maintenance of


* Transduced
0 Mock T


1400
1200
- 1000
S800
X 600
> 400
0


DIC Native GFP






B











the self-renewal capacity of NSC cultures (Walton, et al., 2006b). We do not see


obvious changes in the microglial subpopulation following MMP9 genetic modification.


[] NSC
M NSC+0.3mM 1,10
PNTL'


proMMP9 [+] ctrl + +
Transduced C.M. -
AMPA* +
1



100 kDa
80 kDa


ctMMP9 [+] ctrl -
Transduced C.M. +
1


+ +
+ + +
2 3 4


Figure 4-7. Secreted MMP9 has zymogen activity and can undergo autoactivation.
Conditioned media (C.M.) from transduced NSCs has 3-fold more gelatinase
activity compared to media from mock treated NSCs (A). MMP9 NSC
conditioned media was not associated with degradation of A3 (data not
shown). Commercially available catalytic subunit of MMP9 (39kDa) was able
to degrade AP3 (B). Application of p-aminophenylmercuric acetate (AMPA) to
conditioned media results in conversion (autoactivation) of transgene derived
92kDa MMP9 zymogen to ~84kDa isoform that is associated with AP3
degradation (C). Mouse MMP9 (105kDa) not detected here. Human specific
MMP9 antibody used.

In vivo, NSCs transition between states of quiescence and proliferation (Chambers


and Matrisian, 1997, Lubbe, et al., 2006, Moll, et al., 1990). In a quiescent state, it is


likely that the metabolic rate of NSCs is reduced. This may effect NSC drug delivery as


transgene transcription or translation may be affected negatively. In our experience,


NSC consistently secreted appropriately folded MMP9 that is capable of N-terminal


shedding and gelatinase activity. Following transduction, we found an assortment of


1000
900
800
3 700
S600
U 500
U 400
. 300
200
LL 100
0


Transduced Mock









GFP expression in a population of cells that did not yield detectable levels of MMP9

(cell associated antibody reactivity). We attribute this variability in reporter gene

expression to the randomness of lentiviral infection, regions of transgene integration

and unknown promoter effects. However, NSC quiescence may contribute to the

assorted expression of GFP. Nonetheless, we were able to select for the brightest cells

and improve our yields of transgene protein to ug/ml quantities. The success of our

selection suggests that in a clinical setting, similar methodology may be used to obtain

populations of cells that yield appropriate doses of candidate drugs.

Whether transgene expression changes once NSCs are transplanted remains

unanswered. Further exploration of temporal changes in transgene expression as well

as studies to understand endogenous activators and inhibitors are needed. MMP9

NSCs form significantly larger grafts than GFP NSCs (see Chapter 3). Yet, both cell

types are associated with similar reductions in amyloid burden. It is possible that

subpopulations of NSCs that are responsible for anti-Ap3 activity are also prone to AP3

toxicity. Previous studies have shown AP3 to be selectively toxic to certain populations of

the hippocampal stem cell niche (Verret, et al., 2007). If a similar scenario applies to

transplanted NSCs, it may be that a subpopulation within MMP9 NSC grafts was

ablated after a bolus of anti-Ap3 activity commensurate with that of GFP NSCs.

Remaining MMP9 NSCs lacking anti-Ap3 activity may then have continued surviving due

to MMP9 overexpression. Determining whether this hypothesis is true is complicated by

the fact that large numbers of transplanted NSCs do not display common immunological

markers (T. Zheng, personal communication), (Verret, et al., 2007) and what

researchers know about transplanted NSCs is largely based on snapshots (histology,









etc) of dynamic graft environments. Temporal studies of graft survival and better

markers of NSCs would facilitate more granular understanding of NSC grafts.

NSCs showed reporter gene expression when transduced with neprilysin and

heparanase. We discontinued our work with neprilysin because NSC cultures

repeatedly died days after transduction. However, heparanase NSCs have been

passage multiple times. These cells are reactive to an antibody against heparanase

(data not shown), suggesting transgene expression. MMP9 and heparanase NSCs

both display two populations of GFP intensity when analyzed with FACS.

Consequently, we have been able to isolate 103 and 104 heparanase NSCs. This data

suggests that neprilysin expression may be uniquely toxic to NSC cultures. However

our experience with MMP9 and heparanase expression indicates that NSCs are

capable of producing a diverse pallet of anti-AP3 molecules.

Concluding Comments

In this preclinical study, we took advantage of interesting stem cell properties in a

disease setting. Our experiments demonstrate that NSCs can be genetically

manipulated ex vivo and yield robust grafts when transplanted into the hippocampus of

diseased mice. Such grafts are associated with reductions of AP3 plaques in a mouse

model of AD. This result provides evidence of a novel attribute of transplanted NSCs.

We attempted to further enhance NSC anti-AP3 activity by inducing MMP9

overexpression. However, only a trend towards further reductions in AP3 plaques was

observed. It is possible that enriched NSCs will produce enough MMP9 to overcome

endogenous tissue inhibitors of MMP9. We describe an enrichment method that has

yielded NSCs cultures able to produce ug/ml concentrations of MMP9. This method

has been generalized to NSCs expressing other candidate molecules. Our future










studies aim to find the right dose of MMP9 or alternatively, the right drug to maximally

reduce AP3 pathology. Together, data presented here demonstrates that NSCs can be

used as a platform for testing novel therapeutic molecules in vivo.


7 %NSC: A.U.<104
[ %NSC:A.U.>103
* % NSC: A.U.>10Q


OFPFfTC-A

1.200.000.00
1,000.000.00
, 800.000.00
600,000.00
1:400.000.00
200.000.00
0.00--


OFPFfTC-A


* Yield per100%
confluent T25 flask


Figure 4-8. Enrichment of NSC cultures. FACS analysis demonstrated two populations
of cells in transduced cultures that vary in reporter gene expression (A, C).
Within the population of brightest cells (B), we selected for intensity greater
than103 or 104 (C, GFP low, GFP high). The resultant cells represent 16%
and 4.5% of the overall transduced population, respectively (n=8) (D). These
cells are referred to as 103 NSCs and 104 NSCs. Yields of103 NSCs and 104
NSCs attained per T25 flask holding ~5 x 106 NSCs (n=7) (E). The
methodology used here resulted in highly enriched cultures (compare G, I,
equally exposed images of similarly confluent cultures) and can be used for
NSCs expressing other transgenes: Chromatogram in (C) is from heparanase
overexpressing NSCs. Characteristic double hump is also exhibited by GFP
NSCs and MMP9 NSCs (data not shown).


100












MMP9 MMP9 Mock 103 103
[+] ctrl [+] ctrl C.M. C.M. C.M.





"9ng lUn_1 C.M. C. c. r.


MMP9 MMP9 104 3
[+] ctrl [+] ctrl C.M.
(, 2.5





*M 1


1 mon. huMMP9 cone.
High Dose [-5 x 106 NSCs in 5ml]







Low Dose



103 104 Mock
Transduced


Figure 4-9. NSC enrichment is associated with rate of MMP9 secretion. MMP9 from 103
NSCs is detectable at 3 weeks while MMP9 from similarly confluent 104 NSCs
is detectable at 1 week (n=3) (A). Projected concentration of MMP9 that a
confluent layer of NSCs will secrete in a 5ml volume during the course of a
month (B) is derived from densitometric analysis of data in (A).


101
























1 2 3 4 5
C D E iPz _- `9

--m--- .... ,-- .. "








Figure 4-10. NSC overexpression and activation of MMP9 in vivo. 1 um z-plane depth
confocal images indicate NSCs overexpressed huMMP9 a month following
transplantation (A-B). MMP9 was observed in cell bodies (arrowheads) as
well as processes (arrows) extended by NSCs within the white matter of A3
transgenic mice (n=3) and not in the contralateral [non-injected] hemisphere
(not shown) (A). To gain insight on the distribution of MMP9 secreted by
NSCs, the brains of non transgenic mice with one month MMP9 NSC
engraftments were sectioned at 1mm intervals (C). Cortical and hippocampal
regions were then dissected (D) and analyzed with western blot using
antibodies against human MMP9. Immunoreactivity was strongest in the
hippocampus of the hemisphere where NSCs were injected into (n=2). NSCs
in vitro secrete proMMP9 (Fig. 4-9A). However, MMP9 in vivo is of a lower
molecular weight indicating activation by endogenous factors. Mouse MMP9
is >100kDa and not detected by antibodies used here.


102










A 200 B 160
180 140
160
C r-120
140 .
120 "100

M 01
100 80




S40 4

Contra Ipsi Contra Ipsi
MMP9 MMP9
CTX & HIPPO CTX HIPPO

c U Contra
F GFP Only D 35
SMMP9 120 CTX & HIPPO CTX& HIPPO
100 30

-o e 25
80 E- -80
0 m MM 9 20
I60
-.07 15

|40 40 I A0
'a 10
R" 20 0 20 -

0
0 CTX&HIPPO GFPOnly MMP9 GFPOnly MMP9
Age(month): 13 14 p-o0 47


Figure 4-11. Transplantation of MMP9 NSCs and GFP NSCs results in similar
reductions in amyloid burden. Compared to the contralateral side, the number
of A3 plaques in AP3 transgenic mice with one month 103 MMP9 NSC
engraftments was reduced by 28.5% (p=0.04, paired t-test) (A). All mice
receiving MMP9 NSCs had reduced AP3 plaque numbers (n=5) (B). GFP NSC
and MMP9 NSCs had similar reductions of A3 plaque numbers (C), (see Fig.
4-1). Due to age-related effects, contralateral AP3 plaque burden was 25%
greater in mice receiving MMP9 NSCs (D). Therefore, percentage reduction
is not indicative of total AP3 plaques cleared between the two groups of mice.
To determine if MMP9 overexpression is associated with greater clearance of
AP3 plaques, we compared the absolute number of A3 plaques cleared across
groups. The results indicate MMP9 overexpression by NSCs is associated
with a non-statistically significant trend of 31 % more AP3 plaques cleared
(p=0.47, unpaired t-test) (E). *, p<0.05


103









CHAPTER 5
CONCLUSIONS

Beyond the treatment of Alzheimer's disease, our studies may hold relevance to

other brain diseases that feature protein aggregation. For instance, Amyloid Lateral

Sclerosis (ALS) and Parkinson's disease have aggregates of superoxide dismutase

(SOD1) and a-synuclein, respectively. Recently, elegant studies by Don Cleveland and

others show that microglia expressing mutated SOD1 play an important role in the end

stage of ALS. This evidence fits with our findings: the perturbation of microglial

functionality with age may contribute to neurodegeneration in mouse models and

perhaps in diseased humans. However, therapeutic manipulation of endogenous

microglia is difficult. This is evidenced by the lack of robust outcomes in clinical trials

that aimed to reduce glial activity (ex. non steroidal anti-inflammatory drug trials)

(Sabbagh, 2009) or enhance glial activity (AP3 immunization trials) (Patton, et al., 2006).

Stem cell transplants present an alternative avenue to directly target amyloid

burden. In work presented in here, we were unable to modify NSCs to reduce A3. It is

likely that we haven't found the right dose or the right drug to elicit a change in

pathology. However, this negative result shouldn't detract from other relevant

observations. First, transplantation of NSCs is associated with reduced AP3 burden.

Second, injection of NSCs into the hippocampus results consistently in engraftment in

defined regions of the hippocampus and surrounding white matter. Finally, transgene

overexpression modified NSC engraftment patterns. These findings have significant

implications for proposals for the use of stem cells to restore dying cells or to deliver

drugs to sites of injury.


104









In our studies, we focus on the AP3 pathology of AD. Currently, there is intense

debate regarding the importance of AP3 in AD. The different schools of thought have

noteworthy evidence to cite. For instance, a recent drug trial with an anti-tau compound

has yielded perhaps the most promising clinical outcomes in AD drug trials history

(TauRx, Rember trials). This suggests a causative role for tau in AD etiology. On the

other hand, research published this month shows fibrillar A3 in non demented

individuals who have parents with late-onset (sporadic) AD (Mosconi, et al.). This

buttresses already strong genetic data from Down's syndrome cases and mutations in

the amyloid precursor protein and presenilin loci. The ubiquity of data to support the

different schools of thought of how AD arises is remarkable and unique amongst

neurodegenerative diseases. This ubiquity also highlights the fact that success has not

been had with removal of any of the pathologies found in AD. Until this occurs, and

clinical outcome is observed, arguments on the role of A3 versus tau are inherently

academic. They are based on correlation, not causation.

Removing the pathologies of AD has proven difficult partly because of difficulties

with delivering drugs into the brain. The use of NSCs, as demonstrated here, presents

a novel approach to overcome this difficulty. The scope of this approach is not limited

to AD, as research on treatments for most diseases of the brain is hindered by the

inability to deliver drugs into the brain.

In the course of human history, medical practice on the brain has typically involved

the removal of tissue. At this junction in history, NSCs transplants represent a

bifurcation point. What happens after the introduction of new tissue into the brain is

largely unknown. The work presented here adds to our understanding of the behavior


105









of implanted brain tissue and the manipulation of said tissue to counter disease

pathology.


106









LIST OF REFERENCES


Armstrong, R.A., Cairns, N.J., Myers, D., Smith, C.U., Lantos, P.L., Rossor, M.N. 1996.
A comparison of beta-amyloid deposition in the medial temporal lobe in sporadic
Alzheimer's disease, Down's syndrome and normal elderly brains.
Neurodegeneration 5(1), 35-41.

Association, A.s. 2007. Alzheimer's Disease Facts and Figures 2007. Alzheimer's
Association.

Astary, G.W., Kantorovich, S., Carney, P.R., Mareci, T.H., Sarntinoranont, M. Regional
convection-enhanced delivery of gadolinium-labeled albumin in the rat
hippocampus in vivo. J Neurosci Methods 187(1), 129-37.

Astary, G.W., Kantorovich, S., Carney, P.R., Mareci, T.H., Sarntinoranont, M. 2010.
Regional convection-enhanced delivery of gadolinium-labeled albumin in the rat
hippocampus in vivo. J Neurosci Methods 187(1), 129-37.

Auld, D.S., Kornecook, T.J., Bastianetto, S., Quirion, R. 2002. Alzheimer's disease and
the basal forebrain cholinergic system: relations to beta-amyloid peptides,
cognition, and treatment strategies. Prog Neurobiol 68(3), 209-45.

Bales, K.R., Du, Y., Holtzman, D., Cordell, B., Paul, S.M. 2000. Neuroinflammation and
Alzheimer's disease: critical roles for cytokine/Abeta-induced glial activation, NF-
kappaB, and apolipoprotein E. Neurobiol Aging 21(3), 427-32; discussion 51-3.

Beckmann, N., Schuler, A., Mueggler, T., Meyer, E.P., Wiederhold, K.H., Staufenbiel,
M., Krucker, T. 2003. Age-dependent cerebrovascular abnormalities and blood
flow disturbances in APP23 mice modeling Alzheimer's disease. J Neurosci
23(24), 8453-9.

Bennett, L.B., Cai, J., Enikolopov, G., lacovitti, L. 2010. Heterotopically transplanted
CVO neural stem cells generate neurons and migrate with SVZ cells in the adult
mouse brain. Neurosci Lett 475(1), 1-6.

Bjorklund, A., Dunnett, S.B., Brundin, P., Stoessl, A.J., Freed, C.R., Breeze, R.E.,
Levivier, M., Peschanski, M., Studer, L., Barker, R. 2003. Neural transplantation
for the treatment of Parkinson's disease. Lancet Neurol 2(7), 437-45.

Blits, B., Kitay, B.M., Farahvar, A., Caperton, C.V., Dietrich, W.D., Bunge, M.B. 2005.
Lentiviral vector-mediated transduction of neural progenitor cells before
implantation into injured spinal cord and brain to detect their migration, deliver
neurotrophic factors and repair tissue. Restor Neurol Neurosci 23(5-6), 313-24.


107









Blurton-Jones, M., Kitazawa, M., Martinez-Coria, H., Castello, N.A., Muller, F.J., Loring,
J.F., Yamasaki, T.R., Poon, W.W., Green, K.N., LaFerla, F.M. 2009. Neural stem
cells improve cognition via BDNF in a transgenic model of Alzheimer disease.
Proc Natl Acad Sci U S A 106(32), 13594-9.

Boillee, S., Yamanaka, K., Lobsiger, C.S., Copeland, N.G., Jenkins, N.A., Kassiotis, G.,
Kollias, G., Cleveland, D.W. 2006. Onset and progression in inherited ALS
determined by motor neurons and microglia. Science 312(5778), 1389-92.

Bolmont, T., Haiss, F., Eicke, D., Radde, R., Mathis, C.A., Klunk, W.E., Kohsaka, S.,
Jucker, M., Calhoun, M.E. 2008. Dynamics of the microglial/amyloid interaction
indicate a role in plaque maintenance. J Neurosci 28(16), 4283-92.

Borchelt, D.R., Lee, M.K., Gonzales, V., Slunt, H.H., Ratovitski, T., Jenkins, N.A.,
Copeland, N.G., Price, D.L., Sisodia, S.S. 2002. Accumulation of proteolytic
fragments of mutant presenilin 1 and accelerated amyloid deposition are co-
regulated in transgenic mice. Neurobiol Aging 23(2), 171-7.

Borchelt, D.R., Taraboulos, A., Prusiner, S.B. 1992. Evidence for synthesis of scrapie
prion proteins in the endocytic pathway. J Biol Chem 267(23), 16188-99.

Bouras, C., Hof, P.R., Giannakopoulos, P., Michel, J.P., Morrison, J.H. 1994. Regional
distribution of neurofibrillary tangles and senile plaques in the cerebral cortex of
elderly patients: a quantitative evaluation of a one-year autopsy population from
a geriatric hospital. Cereb Cortex 4(2), 138-50.

Braak, H., Braak, E. 1995. Staging of Alzheimer's disease-related neurofibrillary
changes. Neurobiol Aging 16(3), 271-8; discussion 8-84.

Braak, H., Braak, E., Bohl, J., Reintjes, R. 1996. Age, neurofibrillary changes, A beta-
amyloid and the onset of Alzheimer's disease. Neurosci Lett 210(2), 87-90.

Brookmeyer, R., Corrada, M.M., Curriero, F.C., Kawas, C. 2002. Survival following a
diagnosis of Alzheimer disease. Arch Neurol 59(11), 1764-7.

Brunello, A.G., Weissenberger, J., Kappeler, A., Vallan, C., Peters, M., Rose-John, S.,
Weis, J. 2000. Astrocytic alterations in interleukin-6/Soluble interleukin-6 receptor
alpha double-transgenic mice. Am J Pathol 157(5), 1485-93.

Bruunsgaard, H., Andersen-Ranberg, K., Jeune, B., Pedersen, A.N., Skinhoj, P.,
Pedersen, B.K. 1999. A high plasma concentration of TNF-alpha is associated
with dementia in centenarians. J Gerontol A Biol Sci Med Sci 54(7), M357-64.

Calafiore, M., Battaglia, G., Zappala, A., Trovato-Salinaro, E., Caraci, F., Caruso, M.,
Vancheri, C., Sortino, M.A., Nicoletti, F., Copani, A. 2006. Progenitor cells from
the adult mouse brain acquire a neuronal phenotype in response to beta-amyloid.
Neurobiol Aging 27(4), 606-13.


108









Carson, M.J., Reilly, C.R., Sutcliffe, J.G., Lo, D. 1998. Mature microglia resemble
immature antigen-presenting cells. Glia 22(1), 72-85.

Carter, D.B., Dunn, E., McKinley, D.D., Stratman, N.C., Boyle, T.P., Kuiper, S.L.,
Oostveen, J.A., Weaver, R.J., Boiler, J.A., Gurney, M.E. 2001. Human
apolipoprotein E4 accelerates beta-amyloid deposition in APPsw transgenic
mouse brain. Ann Neurol 50(4), 468-75.

Chambers, A.F., Matrisian, L.M. 1997. Changing views of the role of matrix
metalloproteinases in metastasis. J Natl Cancer Inst 89(17), 1260-70.

Chung, H., Brazil, M.I., Irizarry, M.C., Hyman, B.T., Maxfield, F.R. 2001. Uptake of
fibrillar beta-amyloid by microglia isolated from MSR-A (type I and type II)
knockout mice. Neuroreport 12(6), 1151-4.

Cole, G.M., Beech, W., Frautschy, S.A., Sigel, J., Glasgow, C., Ard, M.D. 1999.
Lipoprotein effects on Abeta accumulation and degradation by microglia in vitro.
J Neurosci Res 57(4), 504-20.

Collins, J.S., Perry, R.T., Watson, B., Jr., Harrell, L.E., Acton, R.T., Blacker, D., Albert,
M.S., Tanzi, R.E., Bassett, S.S., Mclnnis, M.G., Campbell, R.D., Go, R.C. 2000.
Association of a haplotype for tumor necrosis factor in siblings with late-onset
Alzheimer disease: the NIMH Alzheimer Disease Genetics Initiative. Am J Med
Genet 96(6), 823-30.

Culpan, D., MacGowan, S.H., Ford, J.M., Nicoll, J.A., Griffin, W.S., Dewar, D., Cairns,
N.J., Hughes, A., Kehoe, P.G., Wilcock, G.K. 2003. Tumour necrosis factor-alpha
gene polymorphisms and Alzheimer's disease. Neurosci Lett 350(1), 61-5.

de Haas, A.H., Boddeke, H.W., Brouwer, N., Biber, K. 2007. Optimized isolation
enables ex vivo analysis of microglia from various central nervous system
regions. Glia 55(13), 1374-84.

Dolev, I., Michaelson, D.M. 2004. A nontransgenic mouse model shows inducible
amyloid-beta (Abeta) peptide deposition and elucidates the role of apolipoprotein
E in the amyloid cascade. Proc Natl Acad Sci U S A 101(38), 13909-14.

Dong, H., Goico, B., Martin, M., Csernansky, C.A., Bertchume, A., Csernansky, J.G.
2004. Modulation of hippocampal cell proliferation, memory, and amyloid plaque
deposition in APPsw (Tg2576) mutant mice by isolation stress. Neuroscience
127(3), 601-9.

Donovan, M.H., Yazdani, U., Norris, R.D., Games, D., German, D.C., Eisch, A.J. 2006.
Decreased adult hippocampal neurogenesis in the PDAPP mouse model of
Alzheimer's disease. J Comp Neurol 495(1), 70-83.


109









Dringen, R. 2005. Oxidative and antioxidative potential of brain microglial cells. Antioxid
Redox Signal 7(9-10), 1223-33.

Dull, T., Zufferey, R., Kelly, M., Mandel, R.J., Nguyen, M., Trono, D., Naldini, L. 1998. A
third-generation lentivirus vector with a conditional packaging system. J Virol
72(11), 8463-71.

Edbauer, D., Willem, M., Lammich, S., Steiner, H., Haass, C. 2002. Insulin-degrading
enzyme rapidly removes the beta-amyloid precursor protein intracellular domain
(AICD). J Biol Chem 277(16), 13389-93.

EI-Amouri, S.S., Zhu, H., Yu, J., Gage, F.H., Verma, I.M., Kindy, M.S. 2007. Neprilysin
protects neurons against Abeta peptide toxicity. Brain Res 1152, 191-200.

El Khoury, J., Hickman, S.E., Thomas, C.A., Cao, L., Silverstein, S.C., Loike, J.D. 1996.
Scavenger receptor-mediated adhesion of microglia to beta-amyloid fibrils.
Nature 382(6593), 716-9.

El Khoury, J., Toft, M., Hickman, S.E., Means, T.K., Terada, K., Geula, C., Luster, A.D.
2007. Ccr2 deficiency impairs microglial accumulation and accelerates
progression of Alzheimer-like disease. Nat Med 13(4), 432-8.

Englund, U., Ericson, C., Rosenblad, C., Mandel, R.J., Trono, D., Wictorin, K.,
Lundberg, C. 2000. The use of a recombinant lentiviral vector for ex vivo gene
transfer into the rat CNS. Neuroreport 11(18), 3973-7.

Esen, N., Kielian, T. 2007. Effects of low dose GM-CSF on microglial inflammatory
profiles to diverse pathogen-associated molecular patterns (PAMPs). J
Neuroinflammation 4, 10.

Fagan, A.M., Bu, G., Sun, Y., Daugherty, A., Holtzman, D.M. 1996. Apolipoprotein E-
containing high density lipoprotein promotes neurite outgrowth and is a ligand for
the low density lipoprotein receptor-related protein. J Biol Chem 271(47), 30121-
5.

Farris, W., Mansourian, S., Chang, Y., Lindsley, L., Eckman, E.A., Frosch, M.P.,
Eckman, C.B., Tanzi, R.E., Selkoe, D.J., Guenette, S. 2003. Insulin-degrading
enzyme regulates the levels of insulin, amyloid beta-protein, and the beta-
amyloid precursor protein intracellular domain in vivo. Proc Natl Acad Sci U S A
100(7), 4162-7.

Farris, W., Schutz, S.G., Cirrito, J.R., Shankar, G.M., Sun, X., George, A., Leissring,
M.A., Walsh, D.M., Qiu, W.Q., Holtzman, D.M., Selkoe, D.J. 2007. Loss of
neprilysin function promotes amyloid plaque formation and causes cerebral
amyloid angiopathy. Am J Pathol 171(1), 241-51.


110









Fiala, M., Lin, J., Ringman, J., Kermani-Arab, V., Tsao, G., Patel, A., Lossinsky, A.S.,
Graves, M.C., Gustavson, A., Sayre, J., Sofroni, E., Suarez, T., Chiappelli, F.,
Bernard, G. 2005. Ineffective phagocytosis of amyloid-beta by macrophages of
Alzheimer's disease patients. J Alzheimers Dis 7(3), 221-32; discussion 55-62.

Flanary, B.E., Sammons, N.W., Nguyen, C., Walker, D., Streit, W.J. 2007. Evidence that
aging and amyloid promote microglial cell senescence. Rejuvenation Res 10(1),
61-74.

Floden, A.M., Combs, C.K. 2006. Beta-amyloid stimulates murine postnatal and adult
microglia cultures in a unique manner. J Neurosci 26(17), 4644-8.

Floden, A.M., Li, S., Combs, C.K. 2005. Beta-amyloid-stimulated microglia induce
neuron death via synergistic stimulation of tumor necrosis factor alpha and
NMDA receptors. J Neurosci 25(10), 2566-75.

Frank, M.G., Wieseler-Frank, J.L., Watkins, L.R., Maier, S.F. 2006. Rapid isolation of
highly enriched and quiescent microglia from adult rat hippocampus:
immunophenotypic and functional characteristics. J Neurosci Methods 151(2),
121-30.

Frautschy, S.A., Yang, F., Irrizarry, M., Hyman, B., Saido, T.C., Hsiao, K., Cole, G.M.
1998. Microglial response to amyloid plaques in APPsw transgenic mice. Am J
Pathol 152(1), 307-17.

Fryer, J.D., Simmons, K., Parsadanian, M., Bales, K.R., Paul, S.M., Sullivan, P.M.,
Holtzman, D.M. 2005. Human apolipoprotein E4 alters the amyloid-beta 40:42
ratio and promotes the formation of cerebral amyloid angiopathy in an amyloid
precursor protein transgenic model. J Neurosci 25(11), 2803-10.

Fusetti, M., Fioretti, A.B., Silvagni, F., Simaskou, M., Sucapane, P., Necozione, S.,
Eibenstein, A. Smell and preclinical Alzheimer disease: study of 29 patients with
amnesic mild cognitive impairment. J Otolaryngol Head Neck Surg 39(2), 175-81.

Games, D., Adams, D., Alessandrini, R., Barbour, R., Berthelette, P., Blackwell, C.,
Carr, T., Clemens, J., Donaldson, T., Gillespie, F., et al. 1995. Alzheimer-type
neuropathology in transgenic mice overexpressing V717F beta-amyloid
precursor protein. Nature 373(6514), 523-7.

Gao, C.Y., Stepp, M.A., Fariss, R., Zelenka, P. 2004. Cdk5 regulates activation and
localization of Src during corneal epithelial wound closure. J Cell Sci 117(Pt 18),
4089-98.

German, D.C., Yazdani, U., Speciale, S.G., Pasbakhsh, P., Games, D., Liang, C.L.
2003. Cholinergic neuropathology in a mouse model of Alzheimer's disease. J
Comp Neurol 462(4), 371-81.


111









Giulian, D. 1987. Ameboid microglia as effectors of inflammation in the central nervous
system. J Neurosci Res 18(1), 155-71, 32-3.

Goate, A., Chartier-Harlin, M.C., Mullan, M., Brown, J., Crawford, F., Fidani, L., Giuffra,
L., Haynes, A., Irving, N., James, L., et al. 1991. Segregation of a missense
mutation in the amyloid precursor protein gene with familial Alzheimer's disease.
Nature 349(6311), 704-6.

Hattiangady, B., Shuai, B., Cai, J., Coksaygan, T., Rao, M.S., Shetty, A.K. 2007.
Increased dentate neurogenesis after grafting of glial restricted progenitors or
neural stem cells in the aging hippocampus. Stem Cells 25(8), 2104-17.

He, B.P., Wen, W., Strong, M.J. 2002. Activated microglia (BV-2) facilitation of TNF-
alpha-mediated motor neuron death in vitro. J Neuroimmunol 128(1-2), 31-8.

Heneka, M.T., O'Banion, M.K. 2007. Inflammatory processes in Alzheimer's disease. J
Neuroimmunol 184(1-2), 69-91.

Hickman, S.E., Allison, E.K., El Khoury, J. 2008. Microglial dysfunction and defective
beta-amyloid clearance pathways in aging Alzheimer's disease mice. J Neurosci
28(33), 8354-60.

Higgins, G.A., Jacobsen, H. 2003. Transgenic mouse models of Alzheimer's disease:
phenotype and application. Behav Pharmacol 14(5-6), 419-38.

Hirrlinger, J., Gutterer, J.M., Kussmaul, L., Hamprecht, B., Dringen, R. 2000. Microglial
cells in culture express a prominent glutathione system for the defense against
reactive oxygen species. Dev Neurosci 22(5-6), 384-92.

Hoe, H.S., Cooper, M.J., Burns, M.P., Lewis, P.A., van der Brug, M., Chakraborty, G.,
Cartagena, C.M., Pak, D.T., Cookson, M.R., Rebeck, G.W. 2007. The
metalloprotease inhibitor TIMP-3 regulates amyloid precursor protein and
apolipoprotein E receptor proteolysis. J Neurosci 27(40), 10895-905.

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(5284), 99-102.

Huang, F., Buttini, M., Wyss-Coray, T., McConlogue, L., Kodama, T., Pitas, R.E.,
Mucke, L. 1999. Elimination of the class A scavenger receptor does not affect
amyloid plaque formation or neurodegeneration in transgenic mice expressing
human amyloid protein precursors. Am J Pathol 155(5), 1741-7.


112









Imitola, J., Raddassi, K., Park, K.I., Mueller, F.J., Nieto, M., Teng, Y.D., Frenkel, D., Li,
J., Sidman, R.L., Walsh, C.A., Snyder, E.Y., Khoury, S.J. 2004. Directed
migration of neural stem cells to sites of CNS injury by the stromal cell-derived
factor 1alpha/CXC chemokine receptor 4 pathway. Proc Natl Acad Sci U S A
101(52), 18117-22.

Irizarry, M.C., Cheung, B.S., Rebeck, G.W., Paul, S.M., Bales, K.R., Hyman, B.T. 2000.
Apolipoprotein E affects the amount, form, and anatomical distribution of amyloid
beta-peptide deposition in homozygous APP(V717F) transgenic mice. Acta
Neuropathol (Berl) 100(5), 451-8.

Iwata, N., Tsubuki, S., Takaki, Y., Shirotani, K., Lu, B., Gerard, N.P., Gerard, C., Hama,
E., Lee, H.J., Saido, T.C. 2001. Metabolic regulation of brain Abeta by neprilysin.
Science 292(5521), 1550-2.

Jankowsky, J.L., Slunt, H.H., Gonzales, V., Savonenko, A.V., Wen, J.C., Jenkins, N.A.,
Copeland, N.G., Younkin, L.H., Lester, H.A., Younkin, S.G., Borchelt, D.R. 2005.
Persistent amyloidosis following suppression of Abeta production in a transgenic
model of Alzheimer disease. PLoS Med 2(12), e355.

Jin, K., Peel, A.L., Mao, X.O., Xie, L., Cottrell, B.A., Henshall, D.C., Greenberg, D.A.
2004. Increased hippocampal neurogenesis in Alzheimer's disease. Proc Natl
Acad Sci U S A 101(1), 343-7.

Jung, S., Aliberti, J., Graemmel, P., Sunshine, M.J., Kreutzberg, G.W., Sher, A.,
Littman, D.R. 2000. Analysis of fractalkine receptor CX(3)CR1 function by
targeted deletion and green fluorescent protein reporter gene insertion. Mol Cell
Biol 20(11), 4106-14.

Kang, S.S., Kook, J.H., Hwang, S., Park, S.H., Nam, S.C., Kim, J.K. 2008. Inhibition of
matrix metalloproteinase-9 attenuated neural progenitor cell migration after
photothrombotic ischemia. Brain Res 1228, 20-6.

Klein, M.A., Moller, J.C., Jones, L.L., Bluethmann, H., Kreutzberg, G.W., Raivich, G.
1997. Impaired neuroglial activation in interleukin-6 deficient mice. Glia 19(3),
227-33.

Kukekov, V.G., Laywell, E.D., Thomas, L.B., Steindler, D.A. 1997. A nestin-negative
precursor cell from the adult mouse brain gives rise to neurons and glia. Glia
21(4), 399-407.

Kumar-Singh, S., Theuns, J., Van Broeck, B., Pirici, D., Vennekens, K., Corsmit, E.,
Cruts, M., Dermaut, B., Wang, R., Van Broeckhoven, C. 2006. Mean age-of-
onset of familial alzheimer disease caused by presenilin mutations correlates
with both increased Abeta42 and decreased Abeta40. Hum Mutat 27(7), 686-95.


113









Laywell, E.D., Rakic, P., Kukekov, V.G., Holland, E.C., Steindler, D.A. 2000.
Identification of a multipotent astrocytic stem cell in the immature and adult
mouse brain. Proc Natl Acad Sci U S A 97(25), 13883-8.

Leal, M.C., Dorfman, V.B., Gamba, A.F., Frangione, B., Wisniewski, T., Castano, E.M.,
Sigurdsson, E.M., Morelli, L. 2006. Plaque-associated overexpression of insulin-
degrading enzyme in the cerebral cortex of aged transgenic tg2576 mice with
Alzheimer pathology. J Neuropathol Exp Neurol 65(10), 976-87.

Leissring, M.A., Farris, W., Chang, A.Y., Walsh, D.M., Wu, X., Sun, X., Frosch, M.P.,
Selkoe, D.J. 2003. Enhanced proteolysis of beta-amyloid in APP transgenic mice
prevents plaque formation, secondary pathology, and premature death. Neuron
40(6), 1087-93.

Lewis, J., Dickson, D.W., Lin, W.L., Chisholm, L., Corral, A., Jones, G., Yen, S.H.,
Sahara, N., Skipper, L., Yager, D., Eckman, C., Hardy, J., Hutton, M., McGowan,
E. 2001. Enhanced neurofibrillary degeneration in transgenic mice expressing
mutant tau and APP. Science 293(5534), 1487-91.

Li, J.P., Galvis, M.L., Gong, F., Zhang, X., Zcharia, E., Metzger, S., Vlodavsky, I.,
Kisilevsky, R., Lindahl, U. 2005. In vivo fragmentation of heparan sulfate by
heparanase overexpression renders mice resistant to amyloid protein A
amyloidosis. Proc Natl Acad Sci U S A 102(18), 6473-7.

Li, L., Lu, J., Tay, S.S., Moochhala, S.M., He, B.P. 2007. The function of microglia,
either neuroprotection or neurotoxicity, is determined by the equilibrium among
factors released from activated microglia in vitro. Brain Res 1159, 8-17.

Liao, M.C., Van Nostrand, W.E. 2010. Degradation of soluble and fibrillar amyloid beta-
protein by matrix metalloproteinase (MT1-MMP) in vitro. Biochemistry 49(6),
1127-36.

Licastro, F., Pedrini, S., Caputo, L., Annoni, G., Davis, L.J., Ferri, C., Casadei, V.,
Grimaldi, L.M. 2000. Increased plasma levels of interleukin-1, interleukin-6 and
alpha-1-antichymotrypsin in patients with Alzheimer's disease: peripheral
inflammation or signals from the brain? J Neuroimmunol 103(1), 97-102.

Lindenau, J., Noack, H., Asayama, K., Wolf, G. 1998. Enhanced cellular glutathione
peroxidase immunoreactivity in activated astrocytes and in microglia during
excitotoxin induced neurodegeneration. Glia 24(2), 252-6.

Loddick, S.A., Turnbull, A.V., Rothwell, N.J. 1998. Cerebral interleukin-6 is
neuroprotective during permanent focal cerebral ischemia in the rat. J Cereb
Blood Flow Metab 18(2), 176-9.

Lopez-Toledano, M.A., Shelanski, M.L. 2004. Neurogenic effect of beta-amyloid peptide
in the development of neural stem cells. J Neurosci 24(23), 5439-44.
114









Lorenzl, S., Albers, D.S., LeWitt, P.A., Chirichigno, J.W., Hilgenberg, S.L., Cudkowicz,
M.E., Beal, M.F. 2003. Tissue inhibitors of matrix metalloproteinases are
elevated in cerebrospinal fluid of neurodegenerative diseases. J Neurol Sci
207(1-2), 71-6.

Lu, W.G., Chen, H., Wang, D., Li, F.G., Zhang, S.M. 2007. Regionspecific survival and
differentiation of mouse embryonic stem cell-derived implants in the adult rat
brain. Sheng Li Xue Bao 59(1), 51-7.

Lubbe, W.J., Zhou, Z.Y., Fu, W., Zuzga, D., Schulz, S., Fridman, R., Muschel, R.J.,
Waldman, S.A., Pitari, G.M. 2006. Tumor epithelial cell matrix metalloproteinase
9 is a target for antimetastatic therapy in colorectal cancer. Clin Cancer Res
12(6), 1876-82.

Lue, L.F., Walker, D.G. 2002. Modeling Alzheimer's disease immune therapy
mechanisms: interactions of human postmortem microglia with antibody-
opsonized amyloid beta peptide. J Neurosci Res 70(4), 599-610.

Lundberg, C., Martinez-Serrano, A., Cattaneo, E., McKay, R.D., Bjorklund, A. 1997.
Survival, integration, and differentiation of neural stem cell lines after
transplantation to the adult rat striatum. Exp Neurol 145(2 Pt 1), 342-60.

Majumdar, A., Chung, H., Dolios, G., Wang, R., Asamoah, N., Lobel, P., Maxfield, F.R.
2007a. Degradation of fibrillar forms of Alzheimer's amyloid beta-peptide by
macrophages. Neurobiol Aging.

Majumdar, A., Cruz, D., Asamoah, N., Buxbaum, A., Sohar, I., Lobel, P., Maxfield, F.R.
2007b. Activation of microglia acidifies lysosomes and leads to degradation of
Alzheimer amyloid fibrils. Mol Biol Cell 18(4), 1490-6.

Marr, R.A., Rockenstein, E., Mukherjee, A., Kindy, M.S., Hersh, L.B., Gage, F.H.,
Verma, I.M., Masliah, E. 2003. Neprilysin gene transfer reduces human amyloid
pathology in transgenic mice. J Neurosci 23(6), 1992-6.

Marshall, G.P., 2nd, Laywell, E.D., Zheng, T., Steindler, D.A., Scott, E.W. 2006. In vitro-
derived "neural stem cells" function as neural progenitors without the capacity for
self-renewal. Stem Cells 24(3), 731-8.

Marz, P., Cheng, J.G., Gadient, R.A., Patterson, P.H., Stoyan, T., Otten, U., Rose-John,
S. 1998. Sympathetic neurons can produce and respond to interleukin 6. Proc
Natl Acad Sci U S A 95(6), 3251-6.

McDermott, J.R., Bartram, R.E., Knight, P.A., Miller, H.R., Garrod, D.R., Grencis, R.K.
2003. Mast cells disrupt epithelial barrier function during enteric nematode
infection. Proc Natl Acad Sci U S A 100(13), 7761-6.


115









McGeer, P.L., McGeer, E.G. 2001. Polymorphisms in inflammatory genes and the risk
of Alzheimer disease. Arch Neurol 58(11), 1790-2.

Meyer-Luehmann, M., Spires-Jones, T.L., Prada, C., Garcia-Alloza, M., de Calignon, A.,
Rozkalne, A., Koenigsknecht-Talboo, J., Holtzman, D.M., Bacskai, B.J., Hyman,
B.T. 2008. Rapid appearance and local toxicity of amyloid-beta plaques in a
mouse model of Alzheimer's disease. Nature 451(7179), 720-4.

Miller, K.R., Streit, W.J. 2007. The effects of aging, injury and disease on microglial
function: a case for cellular senescence. Neuron Glia Biol 3(3), 245-53.

Milton, R.H., Abeti, R., Averaimo, S., DeBiasi, S., Vitellaro, L., Jiang, L., Curmi, P.M.,
Breit, S.N., Duchen, M.R., Mazzanti, M. 2008. CLIC1 function is required for
beta-amyloid-induced generation of reactive oxygen species by microglia. J
Neurosci 28(45), 11488-99.

Moll, U.M., Youngleib, G.L., Rosinski, K.B., Quigley, J.P. 1990. Tumor promoter-
stimulated Mr 92,000 gelatinase secreted by normal and malignant human cells:
isolation and characterization of the enzyme from HT1080 tumor cells. Cancer
Res 50(19), 6162-70.

Moolman, D.L., Vitolo, O.V., Vonsattel, J.P., Shelanski, M.L. 2004. Dendrite and
dendritic spine alterations in Alzheimer models. J Neurocytol 33(3), 377-87.

Mosconi, L., Rinne, J.O., Tsui, W.H., Berti, V., Li, Y., Wang, H., Murray, J., Scheinin, N.,
Nagren, K., Williams, S., Glodzik, L., De Santi, S., Vallabhajosula, S., de Leon,
M.J. Increased fibrillar amyloid-{beta} burden in normal individuals with a family
history of late-onset Alzheimer's. Proc Natl Acad Sci U S A.

Motte, J., Williams, R.S. 1989. Age-related changes in the density and morphology of
plaques and neurofibrillary tangles in Down syndrome brain. Acta Neuropathol
(Berl) 77(5), 535-46.

Munger, J.S., Haass, C., Lemere, C.A., Shi, G.P., Wong, W.S., Teplow, D.B., Selkoe,
D.J., Chapman, H.A. 1995. Lysosomal processing of amyloid precursor protein to
A beta peptides: a distinct role for cathepsin S. Biochem J 311 (Pt 1), 299-305.

Nagaraja, T.N., Patel, P., Gorski, M., Gorevic, P.D., Patlak, C.S., Fenstermacher, J.D.
2005. In normal rat, intraventricularly administered insulin-like growth factor-1 is
rapidly cleared from CSF with limited distribution into brain. Cerebrospinal Fluid
Res 2, 5.

Naidu, A., Xu, Q., Catalano, R., Cordell, B. 2002. Secretion of apolipoprotein E by brain
glia requires protein prenylation and is suppressed by stations. Brain Res 958(1),
100-11.


116









Njie, E.G., Boelen, E., Stassen, F.R., Steinbusch, H.W., Borchelt, D.R., Streit, W.J.
2010. Ex vivo cultures of microglia from young and aged rodent brain reveal age-
related changes in microglial function. Neurobiol Aging.

Oddo, S., Caccamo, A., Shepherd, J.D., Murphy, M.P., Golde, T.E., Kayed, R.,
Metherate, R., Mattson, M.P., Akbari, Y., LaFerla, F.M. 2003. Triple-transgenic
model of Alzheimer's disease with plaques and tangles: intracellular Abeta and
synaptic dysfunction. Neuron 39(3), 409-21.

Olstorn, H., Moe, M.C., Roste, G.K., Bueters, T., Langmoen, I.A. 2007. Transplantation
of stem cells from the adult human brain to the adult rat brain. Neurosurgery
60(6), 1089-98; discussion 98-9.

Paresce, D.M., Chung, H., Maxfield, F.R. 1997. Slow degradation of aggregates of the
Alzheimer's disease amyloid beta-protein by microglial cells. J Biol Chem
272(46), 29390-7.

Park, K.I., Teng, Y.D., Snyder, E.Y. 2002. The injured brain interacts reciprocally with
neural stem cells supported by scaffolds to reconstitute lost tissue. Nat
Biotechnol 20(11), 1111-7.

Parsons, C.G., Stoffler, A., Danysz, W. 2007. Memantine: a NMDA receptor antagonist
that improves memory by restoration of homeostasis in the glutamatergic system
too little activation is bad, too much is even worse. Neuropharmacology 53(6),
699-723.

Patton, R.L., Kalback, W.M., Esh, C.L., Kokjohn, T.A., Van Vickle, G.D., Luehrs, D.C.,
Kuo, Y.M., Lopez, J., Brune, D., Ferrer, I., Masliah, E., Newel, A.J., Beach, T.G.,
Castano, E.M., Roher, A.E. 2006. Amyloid-beta peptide remnants in AN-1792-
immunized Alzheimer's disease patients: a biochemical analysis. Am J Pathol
169(3), 1048-63.

Peress, N., Perillo, E., Zucker, S. 1995. Localization of tissue inhibitor of matrix
metalloproteinases in Alzheimer's disease and normal brain. J Neuropathol Exp
Neurol 54(1), 16-22.

Pihlaja, R., Koistinaho, J., Malm, T., Sikkila, H., Vainio, S., Koistinaho, M. 2008.
Transplanted astrocytes internalize deposited beta-amyloid peptides in a
transgenic mouse model of Alzheimer's disease. Glia 56(2), 154-63.

Podoly, E., Bruck, T., Diamant, S., Melamed-Book, N., Weiss, A., Huang, Y., Livnah, 0.,
Langermann, S., Wilgus, H., Soreq, H. 2008. Human recombinant
butyrylcholinesterase purified from the milk of transgenic goats interacts with
beta-amyloid fibrils and suppresses their formation in vitro. Neurodegener Dis
5(3-4), 232-6.


117









Ponomarev, E.D., Novikova, M., Maresz, K., Shriver, L.P., Dittel, B.N. 2005.
Development of a culture system that supports adult microglial cell proliferation
and maintenance in the resting state. J Immunol Methods 300(1-2), 32-46.

Prajerova, I., Honsa, P., Chvatal, A., Anderova, M. Neural stem/progenitor cells derived
from the embryonic dorsal telencephalon of D6/GFP mice differentiate primarily
into neurons after transplantation into a cortical lesion. Cell Mol Neurobiol 30(2),
199-218.

Price, J.L., McKeel, D.W., Jr., Buckles, V.D., Roe, C.M., Xiong, C., Grundman, M.,
Hansen, L.A., Petersen, R.C., Parisi, J.E., Dickson, D.W., Smith, C.D., Davis,
D.G., Schmitt, F.A., Markesbery, W.R., Kaye, J., Kurlan, R., Hulette, C., Kurland,
B.F., Higdon, R., Kukull, W., Morris, J.C. 2009. Neuropathology of nondemented
aging: presumptive evidence for preclinical Alzheimer disease. Neurobiol Aging
30(7), 1026-36.

Qin, L., Liu, Y., Wang, T., Wei, S.J., Block, M.L., Wilson, B., Liu, B., Hong, J.S. 2004.
NADPH oxidase mediates lipopolysaccharide-induced neurotoxicity and
proinflammatory gene expression in activated microglia. J Biol Chem 279(2),
1415-21.

Qiu, W.Q., Walsh, D.M., Ye, Z., Vekrellis, K., Zhang, J., Podlisny, M.B., Rosner, M.R.,
Safavi, A., Hersh, L.B., Selkoe, D.J. 1998. Insulin-degrading enzyme regulates
extracellular levels of amyloid beta-protein by degradation. J Biol Chem 273(49),
32730-8.

Radojevic, V., Kapfhammer, J.P. 2009. Directed fiber outgrowth from transplanted
embryonic cortex-derived neurospheres in the adult mouse brain. Neural Plast
2009, 852492.

Raedt, R., Van Dycke, A., Waeytens, A., Wyckhuys, T., Vonck, K., Wadman, W., Boon,
P. 2009. Unconditioned adult-derived neurosphere cells mainly differentiate
towards astrocytes upon transplantation in sclerotic rat hippocampus. Epilepsy
Res 87(2-3), 148-59.

Ramos-DeSimone, N., Hahn-Dantona, E., Sipley, J., Nagase, H., French, D.L., Quigley,
J.P. 1999. Activation of matrix metalloproteinase-9 (MMP-9) via a converging
plasmin/stromelysin-1 cascade enhances tumor cell invasion. J Biol Chem
274(19), 13066-76.

Raponi, E., Agenes, F., Delphin, C., Assard, N., Baudier, J., Legraverend, C., Deloulme,
J.C. 2007. S100B expression defines a state in which GFAP-expressing cells
lose their neural stem cell potential and acquire a more mature developmental
stage. Glia 55(2), 165-77.


118









Reynolds, B.A., Weiss, S. 1992. Generation of neurons and astrocytes from isolated
cells of the adult mammalian central nervous system. Science 255(5052), 1707-
10.

Rogers, J., Strohmeyer, R., Kovelowski, C.J., Li, R. 2002. Microglia and inflammatory
mechanisms in the clearance of amyloid beta peptide. Glia 40(2), 260-9.

Ross, J. 1995. mRNA stability in mammalian cells. Microbiol Rev 59(3), 423-50.

Rovelet-Lecrux, A., Hannequin, D., Raux, G., Le Meur, N., Laquerriere, A., Vital, A.,
Dumanchin, C., Feuillette, S., Brice, A., Vercelletto, M., Dubas, F., Frebourg, T.,
Campion, D. 2006. APP locus duplication causes autosomal dominant early-
onset Alzheimer disease with cerebral amyloid angiopathy. Nat Genet 38(1), 24-
6.

Ruiz-Opazo, N., Kosik, K.S., Lopez, L.V., Bagamasbad, P., Ponce, L.R., Herrera, V.L.
2004. Attenuated hippocampus-dependent learning and memory decline in
transgenic TgAPPswe Fischer-344 rats. Mol Med 10(1-6), 36-44.

Ryu, J.K., Cho, T., Wang, Y.T., McLarnon, J.G. 2009. Neural progenitor cells attenuate
inflammatory reactivity and neuronal loss in an animal model of inflamed AD
brain. J Neuroinflammation 6, 39.

Sabbagh, M.N. 2009. Drug development for Alzheimer's disease: where are we now
and where are we headed? Am J Geriatr Pharmacother 7(3), 167-85.

Savonenko, A., Xu, G.M., Melnikova, T., Morton, J.L., Gonzales, V., Wong, M.P., Price,
D.L., Tang, F., Markowska, A.L., Borchelt, D.R. 2005. Episodic-like memory
deficits in the APPswe/PS1dE9 mouse model of Alzheimer's disease:
relationships to beta-amyloid deposition and neurotransmitter abnormalities.
Neurobiol Dis 18(3), 602-17.

Sawaya, M.R., Sambashivan, S., Nelson, R., Ivanova, M.I., Sievers, S.A., Apostol, M.I.,
Thompson, M.J., Balbirnie, M., Wiltzius, J.J., McFarlane, H.T., Madsen, A.O.,
Riekel, C., Eisenberg, D. 2007. Atomic structures of amyloid cross-beta spines
reveal varied steric zippers. Nature 447(7143), 453-7.

Scheffler, B., Walton, N.M., Lin, D.D., Goetz, A.K., Enikolopov, G., Roper, S.N.,
Steindler, D.A. 2005. Phenotypic and functional characterization of adult brain
neuropoiesis. Proc Natl Acad Sci U S A 102(26), 9353-8.

Schmidt, S.D., Nixon, R.A., Mathews, P.M. 2005. ELISA method for measurement of
amyloid-beta levels. Methods Mol Biol 299, 279-97.

Schwab, C., Hosokawa, M., McGeer, P.L. 2004. Transgenic mice overexpressing
amyloid beta protein are an incomplete model of Alzheimer disease. Exp Neurol
188(1), 52-64.


119









Shagin, D.A., Barsova, E.V., Yanushevich, Y.G., Fradkov, A.F., Lukyanov, K.A., Labas,
Y.A., Semenova, T.N., Ugalde, J.A., Meyers, A., Nunez, J.M., Widder, E.A.,
Lukyanov, S.A., Matz, M.V. 2004. GFP-like proteins as ubiquitous metazoan
superfamily: evolution of functional features and structural complexity. Mol Biol
Evol 21(5), 841-50.

Shen, Y., Joachimiak, A., Rosner, M.R., Tang, W.J. 2006. Structures of human insulin-
degrading enzyme reveal a new substrate recognition mechanism. Nature
443(7113), 870-4.

Shetty, A.K., Rao, M.S., Hattiangady, B. 2008. Behavior of hippocampal stem/progenitor
cells following grafting into the injured aged hippocampus. J Neurosci Res
86(14), 3062-74.

Sierra, A., Gottfried-Blackmore, A.C., McEwen, B.S., Bulloch, K. 2007. Microglia derived
from aging mice exhibit an altered inflammatory profile. Glia 55(4), 412-24.

Simard, A.R., Soulet, D., Gowing, G., Julien, J.P., Rivest, S. 2006. Bone marrow-
derived microglia play a critical role in restricting senile plaque formation in
Alzheimer's disease. Neuron 49(4), 489-502.

Simmons, D.A., Casale, M., Alcon, B., Pham, N., Narayan, N., Lynch, G. 2007. Ferritin
accumulation in dystrophic microglia is an early event in the development of
Huntington's disease. Glia 55(10), 1074-84.

Stalder, M., Phinney, A., Probst, A., Sommer, B., Staufenbiel, M., Jucker, M. 1999.
Association of microglia with amyloid plaques in brains of APP23 transgenic
mice. Am J Pathol 154(6), 1673-84.

Storm van's Gravesande, K., Layne, M.D., Ye, Q., Le, L., Baron, R.M., Perrella, M.A.,
Santambrogio, L., Silverman, E.S., Riese, R.J. 2002. IFN regulatory factor-1
regulates IFN-gamma-dependent cathepsin S expression. J Immunol 168(9),
4488-94.

Streit, W.J., Hurley, S.D., McGraw, T.S., Semple-Rowland, S.L. 2000. Comparative
evaluation of cytokine profiles and reactive gliosis supports a critical role for
interleukin-6 in neuron-glia signaling during regeneration. J Neurosci Res 61(1),
10-20.

Streit, W.J., Miller, K.R., Lopes, K.O., Njie, E. 2008. Microglial degeneration in the aging
brain--bad news for neurons? Front Biosci 13, 3423-38.

Streit, W.J., Mrak, R.E., Griffin, W.S. 2004. Microglia and neuroinflammation: a
pathological perspective. J Neuroinflammation 1(1), 14.


120









Sturchler-Pierrat, C., Abramowski, D., Duke, M., Wiederhold, K.H., Mistl, C., Rothacher,
S., Ledermann, B., Burki, K., Frey, P., Paganetti, P.A., Waridel, C., Calhoun,
M.E., Jucker, M., Probst, A., Staufenbiel, M., Sommer, B. 1997. Two amyloid
precursor protein transgenic mouse models with Alzheimer disease-like
pathology. Proc Natl Acad Sci U S A 94(24), 13287-92.

Svendsen, C.N., Clarke, D.J., Rosser, A.E., Dunnett, S.B. 1996. Survival and
differentiation of rat and human epidermal growth factor-responsive precursor
cells following grafting into the lesioned adult central nervous system. Exp Neurol
137(2), 376-88.

Takaki, Y., Iwata, N., Tsubuki, S., Taniguchi, S., Toyoshima, S., Lu, B., Gerard, N.P.,
Gerard, C., Lee, H.J., Shirotani, K., Saido, T.C. 2000. Biochemical identification
of the neutral endopeptidase family member responsible for the catabolism of
amyloid beta peptide in the brain. J Biochem (Tokyo) 128(6), 897-902.

Tan, J., Town, T., Paris, D., Mori, T., Suo, Z., Crawford, F., Mattson, M.P., Flavell, R.A.,
Mullan, M. 1999. Microglial activation resulting from CD40-CD40L interaction
after beta-amyloid stimulation. Science 286(5448), 2352-5.

Tang, J., Xu, H., Fan, X., Li, D., Rancourt, D., Zhou, G., Li, Z., Yang, L. 2008.
Embryonic stem cell-derived neural precursor cells improve memory dysfunction
in Abeta(1-40) injured rats. Neurosci Res 62(2), 86-96.

Tanzi, R.E., Bertram, L. 2005. Twenty years of the Alzheimer's disease amyloid
hypothesis: a genetic perspective. Cell 120(4), 545-55.

Tarkowski, E., Blennow, K., Wallin, A., Tarkowski, A. 1999. Intracerebral production of
tumor necrosis factor-alpha, a local neuroprotective agent, in Alzheimer disease
and vascular dementia. J Clin Immunol 19(4), 223-30.

Tchaikovskaya, T., Fraifeld, V., Urphanishvili, T., Andorfer, J.H., Davies, P., Listowsky,
I. 2005. Glutathione S-transferase hGSTM3 and ageing-associated
neurodegeneration: relationship to Alzheimer's disease. Mech Ageing Dev
126(2), 309-15.

Terry, A.V., Jr., Buccafusco, J.J. 2003. The cholinergic hypothesis of age and
Alzheimer's disease-related cognitive deficits: recent challenges and their
implications for novel drug development. J Pharmacol Exp Ther 306(3), 821-7.

Tham, C.S., Lin, F.F., Rao, T.S., Yu, N., Webb, M. 2003. Microglial activation state and
lysophospholipid acid receptor expression. Int J Dev Neurosci 21(8), 431-43.

Thier, M., Marz, P., Otten, U., Weis, J., Rose-John, S. 1999. Interleukin-6 (IL-6) and its
soluble receptor support survival of sensory neurons. J Neurosci Res 55(4), 411 -
22.


121









Tietze, F. 1969. Enzymic method for quantitative determination of nanogram amounts of
total and oxidized glutathione: applications to mammalian blood and other
tissues. Anal Biochem 27(3), 502-22.

Verret, L., Jankowsky, J.L., Xu, G.M., Borchelt, D.R., Rampon, C. 2007. Alzheimer's-
type amyloidosis in transgenic mice impairs survival of newborn neurons derived
from adult hippocampal neurogenesis. J Neurosci 27(25), 6771-80.

Walsh, D.M., Klyubin, I., Fadeeva, J.V., Cullen, W.K., Anwyl, R., Wolfe, M.S., Rowan,
M.J., Selkoe, D.J. 2002. Naturally secreted oligomers of amyloid beta protein
potently inhibit hippocampal long-term potentiation in vivo. Nature 416(6880),
535-9.

Walton, N.M., Sutter, B.M., Chen, H.X., Chang, L.J., Roper, S.N., Scheffler, B.,
Steindler, D.A. 2006a. Derivation and large-scale expansion of multipotent
astroglial neural progenitors from adult human brain. Development 133(18),
3671-81.

Walton, N.M., Sutter, B.M., Laywell, E.D., Levkoff, L.H., Kearns, S.M., Marshall, G.P.,
2nd, Scheffler, B., Steindler, D.A. 2006b. Microglia instruct subventricular zone
neurogenesis. Glia 54(8), 815-25.

Wang, L., Zhang, Z.G., Zhang, R.L., Gregg, S.R., Hozeska-Solgot, A., LeTourneau, Y.,
Wang, Y., Chopp, M. 2006. Matrix metalloproteinase 2 (MMP2) and MMP9
secreted by erythropoietin-activated endothelial cells promote neural progenitor
cell migration. J Neurosci 26(22), 5996-6003.

Wang, R., Dineley, K.T., Sweatt, J.D., Zheng, H. 2004. Presenilin 1 familial Alzheimer's
disease mutation leads to defective associative learning and impaired adult
neurogenesis. Neuroscience 126(2), 305-12.

Watson, D.J., Walton, R.M., Magnitsky, S.G., Bulte, J.W., Poptani, H., Wolfe, J.H. 2006.
Structure-specific patterns of neural stem cell engraftment after transplantation in
the adult mouse brain. Hum Gene Ther 17(7), 693-704.

Wegiel, J., Imaki, H., Wang, K.C., Wegiel, J., Rubenstein, R. 2004. Cells of
monocyte/microglial lineage are involved in both microvessel amyloidosis and
fibrillar plaque formation in APPsw tg mice. Brain Res 1022(1-2), 19-29.

Wei, J., Xu, H., Davies, J.L., Hemmings, G.P. 1992. Increase of plasma IL-6
concentration with age in healthy subjects. Life Sci 51(25), 1953-6.

Wei, L.H., Kuo, M.L., Chen, C.A., Chou, C.H., Cheng, W.F., Chang, M.C., Su, J.L.,
Hsieh, C.Y. 2001. The anti-apoptotic role of interleukin-6 in human cervical
cancer is mediated by up-regulation of Mcl-1 through a PI 3-K/Akt pathway.
Oncogene 20(41), 5799-809.


122









Westerman, M.A., Cooper-Blacketer, D., Mariash, A., Kotilinek, L., Kawarabayashi, T.,
Younkin, L.H., Carlson, G.A., Younkin, S.G., Ashe, K.H. 2002. The relationship
between Abeta and memory in the Tg2576 mouse model of Alzheimer's disease.
J Neurosci 22(5), 1858-67.

Wong, P.C., Cai, H., Borchelt, D.R., Price, D.L. 2002. Genetically engineered mouse
models of neurodegenerative diseases. Nat Neurosci 5(7), 633-9.

Wyss-Coray, T., Lin, C., Yan, F., Yu, G.Q., Rohde, M., McConlogue, L., Masliah, E.,
Mucke, L. 2001. TGF-betal promotes microglial amyloid-beta clearance and
reduces plaque burden in transgenic mice. Nat Med 7(5), 612-8.

Xu, G., Gonzales, V., Borchelt, D.R. 2002. Abeta deposition does not cause the
aggregation of endogenous tau in transgenic mice. Alzheimer Dis Assoc Disord
16(3), 196-201.

Xu, Q., Li, Y., Cyras, C., Sanan, D.A., Cordell, B. 2000. Isolation and characterization of
apolipoproteins from murine microglia. Identification of a low density lipoprotein-
like apolipoprotein J-rich but E-poor spherical particle. J Biol Chem 275(41),
31770-7.

Yan, P., Hu, X., Song, H., Yin, K., Bateman, R.J., Cirrito, J.R., Xiao, Q., Hsu, F.F., Turk,
J.W., Xu, J., Hsu, C.Y., Holtzman, D.M., Lee, J.M. 2006. Matrix
metalloproteinase-9 degrades amyloid-beta fibrils in vitro and compact plaques in
situ. J Biol Chem 281(34), 24566-74.

Yan, S.D., Chen, X., Fu, J., Chen, M., Zhu, H., Roher, A., Slattery, T., Zhao, L.,
Nagashima, M., Morser, J., Migheli, A., Nawroth, P., Stern, D., Schmidt, A.M.
1996. RAGE and amyloid-beta peptide neurotoxicity in Alzheimer's disease.
Nature 382(6593), 685-91.

Yasojima, K., Akiyama, H., McGeer, E.G., McGeer, P.L. 2001. Reduced neprilysin in
high plaque areas of Alzheimer brain: a possible relationship to deficient
degradation of beta-amyloid peptide. Neurosci Lett 297(2), 97-100.

Ye, S.M., Johnson, R.W. 1999. Increased interleukin-6 expression by microglia from
brain of aged mice. J Neuroimmunol 93(1-2), 139-48.

Yin, K.J., Cirrito, J.R., Yan, P., Hu, X., Xiao, Q., Pan, X., Bateman, R., Song, H., Hsu,
F.F., Turk, J., Xu, J., Hsu, C.Y., Mills, J.C., Holtzman, D.M., Lee, J.M. 2006.
Matrix metalloproteinases expressed by astrocytes mediate extracellular
amyloid-beta peptide catabolism. J Neurosci 26(43), 10939-48.

Yoon, I.S., Chen, E., Busse, T., Repetto, E., Lakshmana, M.K., Koo, E.H., Kang, D.E.
2007. Low-density lipoprotein receptor-related protein promotes amyloid
precursor protein trafficking to lipid rafts in the endocytic pathway. Faseb J.


123









Zandi, P.P., Anthony, J.C., Khachaturian, A.S., Stone, S.V., Gustafson, D., Tschanz,
J.T., Norton, M.C., Welsh-Bohmer, K.A., Breitner, J.C. 2004. Reduced risk of
Alzheimer disease in users of antioxidant vitamin supplements: the Cache
County Study. Arch Neurol 61(1), 82-8.

Zheng, T., Marshall, G.P., 2nd, Laywell, E.D., Steindler, D.A. 2006. Neurogenic
astrocytes transplanted into the adult mouse lateral ventricle contribute to
olfactory neurogenesis, and reveal a novel intrinsic subependymal neuron.
Neuroscience 142(1), 175-85.

Ziabreva, I., Perry, E., Perry, R., Minger, S.L., Ekonomou, A., Przyborski, S., Ballard, C.
2006. Altered neurogenesis in Alzheimer's disease. J Psychosom Res 61(3),
311-6.


124









BIOGRAPHICAL SKETCH

eMalick Njie is from The Gambia and has always dreamed of being a scientist.


125





PAGE 1

1 BETA PEPTIDE WITH MICROGLIA, STEM CELLS AND MMP9 By EMALICK G OREE NJIE A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIA L FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2010

PAGE 2

2 2010 eMalick Goree Njie

PAGE 3

3 I

PAGE 4

4 ACKNOWLEDGMENTS I thank Dr. Borchelt and Dr. Streit for giving me this opportunity to develop my mind. I would like to extend my gratitude to members of the Streit and Borchelt labs for their support and smiles. I would like to thank my committee for ideas and inspiration and to my collaborators Svetlana Kantorovich Dr. Boelen, Dr. Zheng, Dr. Steindler, and Dr. Rowland Most of all, I thank my sister, my mother and my extended family They have supported me with the strength of a strong village with one son.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF FIGURES ................................ ................................ ................................ .......... 8 LIST OF ABBREVIATIONS ................................ ................................ ........................... 10 ABSTRAC T ................................ ................................ ................................ ................... 12 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 14 ................................ ................................ .. 14 The Amyloid Cascade Hypothesis ................................ ................................ .... 14 ................................ ............................. 15 Extracellular Trafficking and Internalization of A ................................ ............ 17 Lysosomal and Non ................................ ........... 19 ................................ ..... 22 Novel Approaches for Study of A ............................ 24 2 EX VIVO CULTURES OF MICROGLIA FROM YOUNG AND AGED RODENT BRAIN REVEAL AGE RELATED CHANGES IN MICROGLIAL FUNCTION .......... 26 Introduction ................................ ................................ ................................ ............. 26 Methods ................................ ................................ ................................ .................. 27 Solutions ................................ ................................ ................................ ........... 27 Animals ................................ ................................ ................................ ............. 27 Reduction of Debris Produced by Brain Homogenization ................................ 28 Preparation of Discontinuous Percoll Gradients ................................ ............... 28 Immunochemistry ................................ ................................ ............................. 29 Cell Viability ................................ ................................ ................................ ...... 30 Microglial Stimulation ................................ ................................ ....................... 30 IL 6 ELISA ................................ ................................ ................................ ........ 31 TNF ELISA ................................ ................................ ................................ .... 31 Glutathione Measurements ................................ ................................ .............. 32 ................................ ................................ .......................... 32 Statistical Analysis ................................ ................................ ............................ 34 Results ................................ ................................ ................................ .................... 34 Di scussion ................................ ................................ ................................ .............. 41 Improvements on Microglial Isolation ................................ ............................... 41 Age related Changes in Microglial Cytokine Release ................................ ....... 42 Implications of Age related Changes in Microglial Cytokine Release ............... 43 Age related Changes in Microglial Glutathione Levels ................................ ..... 44 Age ................................ ........ 45

PAGE 6

6 ................................ ........ 46 Concluding Comments ................................ ................................ ..................... 47 3 ENGRAFTMENT PATTERNS OF NSCS TRANSPLANTED INTO MOUSE ................................ ................................ .. 52 Introduction ................................ ................................ ................................ ............. 52 Methods ................................ ................................ ................................ .................. 53 Isolation of NSCs ................................ ................................ .............................. 53 Transplantation into Amyloid Beta AD Mice ................................ ..................... 54 Immunochemistry ................................ ................................ ............................. 56 Modeling Paths Of Least Resistance (PLR) ................................ ..................... 56 Results ................................ ................................ ................................ .................... 57 Discussion ................................ ................................ ................................ .............. 66 Paths of Least Resistance versus migration ................................ ..................... 68 Research and Clinical Relevance of Paths of Least Resistance ...................... 71 Concluding Comments ................................ ................................ ..................... 73 4 PATHOLOGY AND DEGRADING PROTEASE, MMP9 ................................ ................................ ......... 77 Introduction ................................ ................................ ................................ ............. 77 Methods ................................ ................................ ................................ .................. 78 Lentivirus Construction ................................ ................................ ..................... 78 Lentivirus Transduction ................................ ................................ .................... 79 Fluorescence Activat ed Cell Sorting ................................ ................................ 79 Isolation of NSCs ................................ ................................ .............................. 79 Transplantation into Amyloid Beta AD Mice ................................ ..................... 79 Immunochemistry ................................ ................................ ............................. 80 ................................ ................................ ......... 81 Reverse Transcriptase Polymerase Chain Reaction (RT PCR) of Human MMP9 ................................ ................................ ................................ ............ 81 In Vitro MMP Gelatinase Activity ................................ ................................ ...... 82 In Situ MMP Gelatinase Activity ................................ ................................ ....... 82 Chemical Activation of secreted MMP9 ................................ ............................ 83 Results ................................ ................................ ................................ .................... 83 Discussion ................................ ................................ ................................ .............. 92 Endogenous MMP Activity ................................ ................................ ................ 94 ........................ 94 NSCs as a Platform to Deliver MMP9 and O ther Candidate Therapeutics In Vivo ................................ ................................ ................................ ............... 95 Characteristics of NSCs overexpressing transgenes ................................ ....... 96 Concluding Comments ................................ ................................ ..................... 99 5 CONCLUSIONS ................................ ................................ ................................ ... 104

PAGE 7

7 LIST OF REFERENCES ................................ ................................ ............................. 10 7 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 125

PAGE 8

8 LIST OF FIGURES Figure page 2 1 Dispase II density centrifugation methodology. ................................ .................. 36 2 2 Purity, yield and viability of microglia ................................ ................................ 38 2 3 Adult microglial morphology. ................................ ................................ ............... 39 2 4 Neonatal microglial morphology. ................................ ................................ ........ 42 2 5 Microglial reaction to immunostimulation ................................ ........................... 45 2 6 Cytokine secretion of young and aged microglia. ................................ ............... 47 2 7 Microgli al glutathione content ................................ ................................ ............ 48 2 8 ................................ .............. 49 2 9 ................................ ................................ ... 50 2 10 Overview of mixed g lial culture (MGC) and density centrifugation methodologies utilized to obtain microglia. ................................ ......................... 51 3 1 Morphology of in vitro NSCs. ................................ ................................ .............. 57 3 2 In vitro characteristics of NSCs. ................................ ................................ .......... 59 3 3 MMP9 associated changes in engraftment ................................ ......................... 60 3 4 Survival and distribution of tr ansplanted NSCs. ................................ .................. 63 3 5 Modeling paths of least resistance. ................................ ................................ .... 64 3 6 Immediate distribution of NSCs in paths of least resistance .............................. 67 3 7 Engraftment patterns of NSCs deposited at the ventral border of the hippocampus. ................................ ................................ ................................ ..... 69 3 8 Iba1 expression and lack of ........................... 74 3 9 GFAP expression by engrafted NSCs. ................................ ............................... 75 3 10 Paths of least resistance. ................................ ................................ ................... 76 4 1 Transplantation of NSCs is associated with reduced amyloid burden ................ 84 4 2 ................................ ................................ ....... 85

PAGE 9

9 4 3 NSCs express endogenous mouse MMP9. ................................ ........................ 89 4 4 Endogenous MMP activity in mice with plaque burden ................................ 91 4 5 Genetic modification of NSCs for MMP9 overexpression. ................................ .. 93 4 6 Cell type differences in transduction efficiency. ................................ .................. 96 4 7 Secreted MMP9 has zymogen activity and can undergo autoactivation. ............ 97 4 8 Enrichment of NSC cultures. ................................ ................................ ............ 100 4 9 NSC enrichment is associated with rate of MMP9 secretion. ........................... 101 4 10 NSC overexpression and activation of MMP9 in vivo ................................ ...... 102 4 11 Transplantation of MMP9 NSCs and GFP NSCs results in similar reductions in amyloid burden ................................ ................................ ............................. 103

PAGE 10

10 LIST OF ABBREVIATION S 1,10 PNTL 1, 10 Phenanthroline AD Amyloid beta AMPA p aminophenylmercuric acetate AU Arbitrary unit CMV Cytomegalovirus CNS Central nervous system EDTA Ethylenediaminetetraacetic acid EF1 Elongation factor 1 FACS Fluorescence activated cell sorting GFP Green fluorescent protein GSH Glut athione IL 6 Interleukin 6 LPS Lippopolysaccharide mRNA messenger ribonucleic acid MGC Mixed glial culture MMP9 metalloprotease 9 NSC mouse n euronal s tem c ell PBS Phosphate buffered saline PLR Path of least resistance ROS Reactive oxygen species SDS Sodium dodecyl sulfate SEZ Subependymal zone SGV Subgranular zone

PAGE 11

11 TNF Tumor necrosis factor

PAGE 12

12 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy CELLULAR AND PROTE BETA PEPTIDE WITH MICROGLIA, STEM CELLS AND MMP9 By eMalick G oree Njie August 2010 Chair: Wolfgang J. Streit Major: Medical Sciences Neuroscience e that primarily affects the elderly. synapses withdraw. instability and a marked depreciation in memory. In the sp aces in between the cells of AD patients, one finds large aggregates of proteins. The amyloid beta peptide is the primary molecule within these aggregates and thus forms a hallmark pathology in AD Genetic data from a rare set of families and from those with Down syndrome indicate that producing more leads to AD However, t he role of these aggregates in the majority of AD patients is largely unknown due to our lack of understanding of how the molecule is catabolized by the brain. W e examine d microglia to shed light on one of the mechanisms for regulating We also explored mouse neuronal stem cells ( NSC s ) as a possible therapeutic intervention to treat pathology Microglial cells are typically associated with the removal of ext raneous materials from the brain. We find that they do not degrade Instead, microglia appear to continually recycle perhaps to minimize the pool of that can form aggregates. Importantly, we find that this recycling of deteriorates signific antly with age. To truly determine

PAGE 13

13 improvement observed. D rug treatment regiments to remove have by and large failed. This is partly because blood vessels block most drugs from entering the brain Drugs that are directly injected into the brain are typically broken down rapidly. One possible way to circumvent these issues is to transplant cells that continually produce drugs directly within the brain. T he neuronal stem cell can live outside the brain for months before being transplanted W e demonstrate that transplants of neuronal stem cell s typically settle in predefined regions within the hippocampus and are associated with reductions in aggregates We also find that neuronal stem cells can be genetically manipulated to over express MMP9, a aggregates in the brain or protect cells from toxicity In mice modeling Alzheimer s disease, n euronal stem cells formed larger transp lants after we g enetically manipulated them to express human MMP9 Together, our findings further our understanding of emonstrating that microglia are less able to process with age and that neuronal stem cells may prove useful for treating pa thology in

PAGE 14

14 CHAPTER 1 INTRODUCTION Disease and Pathology neurodegenerative disease that affects more people than all other neurodegenerative diseases (Association, 2007) Half of all persons reaching the age of 85 will be diagnosed with AD (Association, 2007) Temporally, every 72 seconds someone within the U.S is dia gnosed with AD The mean life span of patients following diagnosis ranges from 3 years to 9 years (Brookmeyer, et al., 2002) The 150 billion dollar s spent annually in healthcare costs for these individuals will only grow as (Association, 2007) Patients with this disease have noticeably enlarged ventricles and severe memory deficits as a result of progressiv e neuronal cell death. Patients inevitably deteriorate over a period of years to the point where they require constant assistance to manage even the most mundane of Although currently available therapies can slow the progression of AD by as much as 5 years, they do not remedy the underlying cause of neurodegeneration (Auld, et al., 2002, Parsons, et al., 2007, Terry and Buccafusco, 2003, Zandi, et al., 2004) The Amyloid Cascade Hypothesis In the aged and AD brain, levels of pept ide have been shown to increase (Armstrong, et al., 1996) It has been postulated that this increase in extracellul ranging from 50x to 1500x above normal (Farris, et al., 2007) is neurotoxic and sets the stage for the neurodegeneration found in AD (Lewis, et al., 2001, Walsh, et al., 2002) This belief is based on th e amyloid cascade theory The theory states that the presence of high amyloid burden (a myloidosis), in the form of extracellular s or

PAGE 15

15 oligomers initiates a cascade that leads to the disruption of cytoskeletal tau protein (Tanzi and Bertram, 2005) (Sawaya, et al., 2007) Two other notable proteins that multimerize via a dry steric zipper mechanism are the prion protein of Bovine Spongiform Encephalitis (Mad Cow Disease) (Sawaya, et al., 2007) Family linked mutations (Goate, et al., 1991, Kumar Singh, et al., 2006, Motte and Williams, 1989, Rovelet Lecrux, et al., 2006) This supports have memory and learning impairments similar to that which occurs in humans (Higgins and Jacobsen, 2003, Hsiao, et al., 1996, Savonenko, et al., 2005, Westerman, et al., 2002, Wong, et al., 2002) Mouse Models Disease Given the limitations associated with post mortem studies on human brains, since the mid nineties several transgenic mouse lines hav e been created to model These mouse models emulate with great success the deposition of changes, such as neurofibrillary tangles. Transgenic mice typically overexpress the amyloid precursor protein (APP) with familial AD mutations under the control of various promoters, and a lthough there are more than a dozen strains available most studies are done on the PDAPP, Tg2576, APP23 and most recently, the mo/hu A PPswe/PS1dE9 mice (Bo rchelt, et al., 2002, Games, et al., 1995, Hsiao, et al., 1996, Sturchler Pierrat, et al., 1997) The mutations carried in the hAPP transgene introduced

PAGE 16

16 in to these mice originated from genetic studies done on families with early onset AD. In general, the transcription of these genes causes dense A plaque to be deposited in the neocortex and the hippocampus usually between six and twelve months of age. The hAPP transgene in the PDAPP model has the V717F mutation (Indiana family origin) (Games, et al., 1995) The dual point mutations in Tg2576, APP23 mice are at K670N/M671L (Swedish family origin). In addition to the Swedish mutation, mo/hu APPswe/PS1dE9 co expresses Presenilin 1 with a familial AD mutation (accelerates A42 deposition) together with the humanized form of mouse APP. Eponymous PDAPP mice have hAPP under the control of platelet derived growth factor promoter (PD). In the Tg2576 and mo/hu APPswe/PS1dE9 models, the APP ge ne is driven by the prion protein promoter, while the APP23 model has hAPP driven by the neuron specific Thy 1 promoter. These animals constitutively over express the APP transgene. Recently, a inducible Tet off mo/huAPPswe/ind transgenic mouse model was generated using a tetracycline responsive promoter (Jankowsky, et al., 2005) The various mouse models have demonstrable impairment in learning and memory that typically manifest at around the same time s deposits become prevalent in the limbic structures of these animals (Higgins and Jacobsen, 2003, Hsiao, et al., 1996, Westerman, et al., 2002) Mice exhibit loss of dendritic spines and loss of synapses in subcortical cholinergic projections, associated gliosis, and cerebrovascular abnormalities (Beckmann, et al., 2003, Germ an, et al., 2003, Moolman, et al., 2004, Stalder, et al., 1999, Wegiel, et al., 2004) APP23 mice exhibit neurodegeneration (Sturchler Pierrat, et al., 1997) while the s in Tet off/APPswe/ind mice persist for the lifetime the mouse even when APP transgene

PAGE 17

17 production is halted early on (Jankowsky, et al., 2005) However, the lifes pan of the Tet off/APPswe/ind mouse is not significantly shortened by the presence of these plaque deposits. This highlights an inherent shortcoming of modeling AD in animals whose normal lifespan is less than 5% that of a normal human. Mouse models also do not have hyperphosphorylated tau neurofibrillar tangles or the same level of compleme nt system activation found in AD (Higgins and Jacobsen, 2003, Schwab, et al., 2004, Xu, et al., 2002) Rat models of AD have not gained widespread popularity because rat brains are more resistant to the formation of dense A plaques and AD like learning & memory deficits (Ruiz Opazo, et al., 2004) The transgenic mice listed above have been mated to other mice deficient or containing mutant proteins that are informative for AD studies. For instance, hybrid mice were created by mating Tg2576 mice to mice deficient in expression of the immune cell chemotaxis receptor CCR2 (El Khoury, et al., 2007) As described in t he following, these mice proved instrumental in further understanding microglial contribution to the central nervous system ( CNS ) challenged with amyloidosis. Extracellular Trafficking and Internalization of A Though not a complete replicate of Alzheim transgenic mouse models are nonetheless useful in understanding the genetics and biochemical cascades that lead to learning and memory deficits found in humans. As mentioned the amyloid s in these models attract glia (Stalder, et al., 1999) As the endogenous immunocompetent cells microglia are among the first cells recruited to the s (El Khoury, et al., 2007, Frautschy, et al., 1998, Stalder, et al., 1999) The protein from these plaques has been found inside microglial lysosomes indicating that these cells actively phagocytose portions of plaques (Col e, et al., 1999,

PAGE 18

18 Frautschy, et al., 1998) The process of internalizing A is mediated by several cells surface receptors. In the brain parenchyma, macrophage scavenger receptor Type A (MSR A) is only expressed by microglia. Studies by Chung and cowork ers and others have shown that MSR A is responsible for uptake of up to 60% of internalized fibrillar A (non opsonized) in the brain (Chung, et al., 2001, El Khoury, et al., 1996) Interestingl y, MSR A knock out animals expressing hAPP with Indiana and Swedish mutations under PD control have similar amounts of burden when compared to their littermates with normal MSR A expression (Huang, et al., 1999) Other receptors such as MSR B, and receptor for advanced glycation end products (RAGE) are capable of internalizing and it is likely that these compensate for the loss of MSR A (Huang, et al., 1999, Rogers, et al., 2002, Yan, et al., 1996) Another way in w hich can be internalized is as a non covalent conjugate to complement factors or antibodies (opsonization). Microglia have an assortment of receptors such as Cd11b and Fc gamma receptors which can mediate the phagocytosis of opsonized (Chung, et al., 2001, Lue and Walker, 2002) A recent study showed that APP is transported to cholesterol rich lipid rafts in neurons by low density lipoprotein receptors like protein (LRP) (Yoon, et al., 2007) however, its cleavage product can be carried by high density lipoprotein like protein (HDL) in the extracellular space and then internalized by microglial LRP (Fagan, et al., 1996) Apolipoproteins E ( ApoE ) and J ( ApoJ ) in complex with HDL reduce the eventual degradation of in microglia (Cole, et al., 1999) Allelic differences in ApoE, along with mutations in APP and PS1/2 are among the most well defined genetic risk factors for familial AD, and it is interesting that

PAGE 19

19 microglia & astrocytes are the major contributors of extracellular ApoE in the brain (Xu, et al., 2000) Release of ApoE into the extracellular space is dependent on protein prenylation and is sensitive to statin treatment (Naidu, et al., 2002) Micromolar concentrations of can induce the secretion of ApoE from microglia in vitro (Bales, et al., 2000) Conversely, the fibrillization of is thought to be promoted by ApoE since the ApoE can bind to (Carter, et al., 2001, Xu, et al., 2000) and mice with the Indiana or Swedish mutations that have ApoE knocked out no longer have dense s or have delayed deposition of s, respectively (Fryer, et al., 2005, Irizarry, et al., 2000) Lysosomal and Non lysosomal Degradation of Frautschy et al. quantified up to five fold increases in microglial density surrounding s in mice with the Swedish mutation under the prion promoter (Frautschy, et al., 1998) El Khoury et al. demonst rated that elimination of CCR2 dependent microglial chemotaxis results in earlier appearance of amyloidosis, twice as much and ~36% greater mortality in mice co expressing the prion promoter Swedish mutation (El Khoury, et al., 2007) A plethora of reports have provided evidence of mouse and human microglial degradation of via the endosomal lysosomal pathway (Frautschy, et al., 1998, Qiu, et al., 1998, Rogers, et al., 2002) However the rate and quantity of this degradation is a subject of great concern as the kinetics of degradation (in relation to deposition) has direct physiological relevance that microglial cells from neonatal mice degrade at much slower rates compared to blood macrophages (Majumdar, et al., 2007a, Paresce, et al. 1997) In an elegant series of experiments, they showed that in the course of three days, neonatal microglia

PAGE 20

20 in vitro degrade only 20% of the fibrillar they are exposed to while peritoneal macrophages degrade close to 80% While microglia and macrop hages are able to make similar cleavages at the N terminus of the fibrillar the macrophages were able to make far more thorough cuts along the molecule. Neither cell type was able to cut the C terminal portion of the molecule. Perhaps the localizat ion of highly stable twisted beta pleated sheet at the C terminus confers this resistance to degradation. This pleated sheet is the fundamental secondary structural element underlying multimers of (Sawaya, et al., 2007) The authors proposed that microglia are hindered in their capacity to degrade relative to their macrophage counterparts, due to an incomplete set of lysosomal enzymes (Majumdar, et al., 2007a) This is indirectly supported by the observation that microglial degradation of in vitro is enhanced when global endocytosis of lysosomal enzymes is enhanced (Majumdar, et al., 2007a) It is worth mentioning that a recent study focusing on macrophages from the blood of AD patients concluded that these cells are impaired in their ability to phagocytose when compare d to non diseased subjects (Fiala, et al., 2005) TGF is a cytokine that attracts and activates microglia. Bigenic mice overexpressing hAPP with Swedish and Indiana mutations under PD promoter in addition to TGF (cSJL x B6D2 background) have a 50% reduction in brain parenchymal burden comp ared to their non TGF overexpressing littermates (Wyss Coray, et al., 2001) Since microglia exposed to TGF in vitro display enhanced degradation of it is postulated that the marked in vivo reduction of plaque burden in these TGF 1 transgenic mice is due to mi croglial action. TGF and similar factors that stimulate microglia most likely cause degradation of through

PAGE 21

21 either intracellular l ysosomal degradation at the N terminus as already described or through extracellular degradation via cell surface enzym es Neprilysin, a zinc dependant endopeptidase and insulin degrading enzyme (IDE), a zinc dependant metalloproteinase, are two well defined brain proteases. It is thought that microglia express these proteases on the cell surface and also release them into the extracellular space (Qiu, et al., 1998, Takaki, et al., 2000) A series of reports have a built a compelling case for neprilysin being the maj or soluble catabolic enzyme in AD animal models, in AD patients and in non diseased humans (Iwata, et al., 2001, Leissring, et al., 2003) Of note are reports quantifying 50% reductions in neprilysin mRNA in hippocampal regions classically susceptible to amyloidosis (Yasojima, et al., 2001) and ~48% less mouse brain neprilysin mRNA in Tg2576 mice deficient in microglial recruitment t o plaques due to CCR2 knockout (El Khoury, et al., 2007) As mentioned previously, these CCR2 knockout mice have twice as much Even though neprilysin is found in neurons, its major degradative function in the brain parenchyma appears to be microglial based. On the other hand, recent studies show that neprilysin cannot degrade the fibrillar commonly found in dense plaques (Yan, et al., 2006) or especially neurotoxic oligomeric (El Amouri, et al., 2007) Thus, neprilysin activity can modulate normal catabolism in vivo (Marr, et al., 2003) and prevent the onset or progression of AD, however, its specificity for less multimeric forms of will likely prevent its use as a therapy for reversing the course of symptomatic AD. The CCR2 knockout mice me ntioned (El Khoury, et al., 2007) have normal IDE mRNA levels. This suggests that neurons and astrocytes are able to supply basal levels of IDE when microglial function is perturbed (El Khoury, et al., 2007) IDE is

PAGE 22

22 present in the cytosol where it degrades the cytoplasmic portion of APP (Edbauer, et al., 2002) however cell surface IDE and secreted IDE are more likely the species of this (Qiu, et al., 1998) In three month old mice lacking IDE (and not expressing APP transgenes), a 64% inc rease in cerebral (Farris, et al., 2003) However in 16 month old mice with the Swedish mutation under the prion promoter, astrocytes proximal to s display a two condenses into plaques (Leal, et al., 2006) Surprisingly, there was no reported reduction in deposition (Leal, et al., 2006) In these studies, microglia surrounding the s do not produce IDE at levels detectable by immunohistochemistry. IDE is one of the leading drug candidates for AD therapy. A recent study suggested that chemical modifications in its active site for the purpose of keeping the enzyme in a con stitutively open state should be pursued (Shen, et al., 2006) This could result in a forty fold increase in catalytic activity and hence a therapeutic increase in A degradation. On e must be prudent however as insulin is a major substrate for IDE degradation. Unlike other tissues, the brain does not maintain energy reserves so perturbations in sugar homeostasis caused by constitutively active IDE could result in severe side effects. This is less of an issue with neprilysin as a therapy since its other proteolytic substrates lie mostly outside the CNS compartment Consequences of Immunization and Inflammation The field of AD research has experienced a significant disappointment fo llowing the termination of the immunization (AN 1792) clinical trial due to life threatening inflammatory side effects (Patton, et al., 2006) In contemplating the cause of

PAGE 23

23 meningoencephalitis bjects, one must consider the role of microglial interaction with as a possible activating agent (Floden, et al., 2005, Tan, et al., 1999 ) This is thought to be mediated via the binding of complement factor C1 conjugated with to complement receptors CR3 & CR4 on the microglial cell surface (Hene ka and O'Banion, 2007) inducing a highly cytotoxic complement cascade. It is worth noting that 20% of the subjects in the trial had the desired antibody response to immunization. However, in this subgroup, 22% suffered from meningoencephalitis. This suggests that the pathogenic inflammation that halted the study likely involved antibody induced inflammation. Perhaps the antibody response changed the to a spec ies that is more inflammatory? In vitro studies have shown that can induce synthesis of inflammatory (Heneka and O'Banion, 2007) Floden and colleagues reported that there are age related differences in the ability of different forms of to induce inflammatory responses (Floden and Combs, 2006) Microglia isolated from neonatal and adult mice (C57BL/6 non transgenics) are able to induce the while fibrillar can only induce croglia. If this finding holds true in humans, it could perhaps lend greater understanding to the AN 1792 trial in the gray and white matter of AD patients with an antibody response to the immunization, post mortem analyses intriguingly found drama tic increases in soluble as a result of antibody dependent plaque disassembly. In fact, the quantity of soluble increased fifteen fold in one subject. This antibody dependent disassembly of fibrillar into soluble caused an unintended increase in oligomeric species of The

PAGE 24

24 investigators found derived dimers, trimers, tetramers and higher order oligomeric structures of up to 30kDa in the brains of these patients. As described by Floden et al., oligomeric but not fibrillar species of can selectively induce adult (Floden and Combs, 2006) and directly cause neuronal death (Floden, et al., 2005, Tan, et al., 1999) Novel Approaches for Study of Regulation and Therapy As described above, S tudies from neuroimmunology labs have demonstrated that mi croglia, the primary immune cell of the CNS (Giu lian, 1987) have regulation in the AD brain (Chung, et al., 2001, El Khoury, et al., 2007, Frautschy, et al., 1998, Rogers, et al., 2002) However, clinical trials aimed at emulating microglial like functions ( anti oxidant trials) or curtailing microglial functions (anti inflammatory trials) ha ve largely failed at treating AD Interestingly, these trials which targeted a disease of the elderly, were conspicuously carried out on knowledge primarily from studies on in vitro neonatal mouse models Nevertheless, it is generally accepted that glia play a role in the regulation of the extracellular space including the metabolism of extracellular The emerging field of s tem cell research has demonstrated that the neuronal stem cell of the subependymal zone (SEZ) is a very glial like immature cell that may provide a window for novel approaches to treat AD pathology (Kukekov, et al., 1997, Laywell, et al., 2000, Raponi, et al., 2007) Our goal was to bring together the knowledge base of neuroimmunology and stem cell biology i n order to answer two questions:1 ) is there a used as a therapeutic approach to

PAGE 25

25 aged and AD brain and 2) the biology of stem cells uniquely position s them as a platform to counter e hypotheses were founded on the following observations. (Armstrong, et al., 1996) known genetic risk factors since these factors confer increased AD susceptibility by ny known (Association, 2007) Second, histological evidence from human and rodent autopsy brain sections show that microglia appear dystrophic with similar morphological features in bo th the aged and AD brains (Flanary, et al., 2007) Thus, lose of microglial function therapy using stem cells is a promising approach becaus e of the physiological flexibility inherent to stem cells. These cells naturally undergo self renewal in vivo and in vitro and are capable of being transplanted after being expanded ex vivo (Marshall, et al., 2006, Walton, et al., 2006a) This unique property allows for genetic introduction of candidate anti A molecules in culture for delivery in vivo In the following, I shall detail o ur experimental findings in quantifying glial aged related biology cells and 2) stem cells used to del iver the candidate anti A therapeutic, MMP9. Our studies have implications in understanding catabolic pathway s that contribute to to treating said pathology.

PAGE 26

26 CHAPTER 2 EX VIVO CULTURES OF MICROGLI A FROM YOUNG AND AGE D RODENT BRAIN REVEAL AGE RELATED CHANGES IN M ICROGLIAL FUNCTION Introduction A multitude of studies have implicated microglia as important players in the etiology of a number of age related neurodegenerative diseases, including Alzheim and amyotrophic lateral sclerosis (Boillee, et al., 2006, Chung, et al., 2001, El Khoury, et al., 2007, Frautschy, et al., 1998, Rogers, et al., 2002) To understand how microglial cell function may change with aging, various protocols have been developed to isolate microglia from the young and aged central nervous system (CNS). While histol ogical studies are essential in providing clues living cells. In the past decade, protocols to isolate living microglia from postnatal animals have become available (Carson, et al., 1998, de Haas, et al., 2007, Frank, et al., 2006, Hickman, et al., 2008, Ponomarev, et al., 2005) These protocols either trap microglia using antibodies to cell s pecific antigens (Hickman, et al., 2008, Tham, et al., 2003) or separate microglia using dens ity centrifugation (de Haas, et al., 2007, Frank, et al., 2006) In both cases, the rapid isolation of microglia enables ex vivo experimentation of endogenous microglia in a controlled setting largely devoid of neurons, oligodendrocytes and astrocytes. Protocols utilizing density centrifugation are advantageous to those utilizing antigen traps in terms o f yield per brain (de Haas, et al., 2007) They also avoid artificial cellular reactions caused by antigen cross linking, a risk carried with the use of antibodies in trapping protocols. However, in our hands, significant amounts of non microglial, debris co ntaminate current density centrifugation derived cultures. In the

PAGE 27

27 present study, we sought to modify density centrifugation methodology to eliminate debris fields present in such cultures. With these modifications, microglial yields were preserved or sl ightly increased. These improvements allowed us t o study microglial function with regard to alterations during normal aging. We found that microglia from aged mice constitutively secrete greater amounts of interleukin 6 (IL 6) and tumor necrosis factor (TNF relative to microglia from younger mice and are less responsive to stimulation. Also, microglia from aged mice have reduced glutathione levels and internalize less while microglia from mice of all ages do not retain the peptide for a signif icant length of time. These studies offer further support for the idea that microglial cell function redistribution rather than biophysical degradation in vivo and thereby pro vide mechanistic insight to the lack of amyloid burden elimination by parenchymal microglia Methods Solutions Dispase II (Roche, Mannheim, Germany) was reconstituted at 2U/mL in dispase buffer (0 Percoll (GE Healthcare, St. Giles, UK) was diluted 1:10 with 10x phosphate buffered saline (PBS) to create an isotonic solution. 1x PBS was added to isotonic percoll to create workin g solutions ranging from 75% to 25% percoll. Animals Debris reduction experiments were performed with non transgenic C57BL/6 mice and mice expressing GFP under the fractalkine receptor promoter (Jung, et al., 2000)

PAGE 28

28 Experiments were performed using young (1 2 month old) and aged (14 16 month old) and provided food and water ad libitum. Animals were euthanized by exsanguination u sing transcardiac perfusion with PBS under deep anesthesia with sodium pentobarbital (50mg/kg body weight). This method of euthanasia is consistent with the recommendations of the Panel of Euthanasia of the American Veterinary Medical Association. After perfusion, the brain (telencephalon, cerebellum and midbrain) was rapidly removed. Reduction of Debris Produced by Brain Homogenization Each brain was washed in cold 1x PBS, then minced using a small scissors. Brain tissue was gently dissociated by imme rsion into 10mLs (per brain) of dispase II solution (2U/mL), trypsin solution (0.05%) or by grinding within a tissue homogenizer (glass Potter, Braun, Melsungen, Germany). Dissociated brain tissue was placed within a 50mL conical tube and laid horizontal ly in an orbital shaker set to shake for 1hr, 37C at 150rpm. Remnant tissue chunks were further homogenized by rapidly triturating with a 10mL pipette (BD Biosciences, Boston, MA) with a wide bore to prevent cell shearing. This was carried out with a fu lly charged pipette aid (Drummond). Enzyme activity was halted by diluting the resultant homogenate 1:1 with cold 10% FBS in 1x PBS. Meninges and clumped cells were removed with 70 um filtration (BD Biosciences, Boston, MA) to obtain a suspension of sing le cells. Preparation of Discontinuous Percoll Gradients The homogenate was spun 1000 x g for 10min at 4C. The supernatant was discarded and the pellet of an individual brain was resuspended in 6mLs of 75% isotonic percoll (high percoll) (GE Healthcar e, Buckinghamshire, U.K). Three mLs of this

PAGE 29

29 mixture was then aliquoted into a 15mL polystyrene tube. Five mLs of 35% isotonic percoll (low percoll) was layered atop the high concentration percoll at a rate of 150 ul /sec to create a distinct interface betwee n the percoll layers. To replicate gradients described in the literature, 25% percoll was utilized for low percoll. 1x PBS was layered atop the low concentration percoll. The resultant discontinuous gradient was then allowed to settle on ice for 15 minu tes allowing most of the homogenate to naturally rise towards its isopycnic position. The gradient was then centrifuged at 800 x g for 45min in a HS 4 swinging bucket rotor (Thermo Fisher Scientific, Waltham, MA) set to 4C. We did not notice changes in microglial yields with high acceleration or the application of the brake. However, yields were significantly diminished if the gradients are not processed immediately following centrifugation. To process the gradients, the volume of the PBS layer and the low concentration percoll layer were rapidly aspirated. A band of microglia (usually 0.5 1.5mL), captured between the low concentration and high concentration percoll layers was then collected and diluted in 50mL of 1x PBS. This was centrifuged at 1000 x g for 10min at full acceleration and brake. The supernatant was quickly decanted and the cell pellet resuspended in DMEM culture media containing 10% FBS. We also added 0.15 ug /mL granulocyte monocyte colony stimulating factor (GM CSF, R & D Systems, Mi nneapolis, MN), authouth this is not required for the culturing of microglia. Immunochemistry Isolated cells were grown in culture media overnight. Cells were then washed, fixed in 4% paraformaldehyde and processed for immunofluorescence of microglial an tigen Iba1 (1:500, Wako, Richmond, VA), microglial antigen Cd11b (1:1000, Serotec, Raleigh, NC), astrocyte antigen GFAP (1:1000, Dako Corporation, Carpinteria, CA), and

PAGE 30

30 neuronal antigen NeuN (1:500, Millipore, Bellirica, MA). Cells were rinsed and incubat ed with goat anti rabbit Alexa 488 (Invitrogen, Carlsbard, CA) and goat anti mouse Alexa 568 (Invitrogen, Carlsbard, CA). Cells were photographed with an Olympus DP71 camera mounted on an Olympus BX60 microscope. Cell Viability Microglial mitochondrial re spiratory activity, a measure of cell viability, was determined using a colorimetric MTT (methylthiazolyldiphenyl tetrazolium bromide) assay (Bioassay Systems, Hayward, CA). This was compared to a reference value of HEK 293 cells, a highly viable immortal cell line, and dying cultures treated with 1% Triton X 100, a toxic reagent. Microglial Stimulation In order to compare the inflammatory reaction of microglia in young and aged brains, cells were isolated from 2 and 14 months old mice, as described abov e, and seeded in 96 well tissue culture plates (Corning Incorporated, Corning, NY) at a density of 3 x 10 5 cells/well. The cultures were incubated overnight at 37 C with 5% CO 2 and saturated humidity. The next day, cells were stimulated by replacing the original culture media with media containing 2% FBS and inflammatory agents in different concentrations. Two highly potent inflammatory stimuli were selected, i.e. lipopolysaccharide (LPS ) ( Escherichia coli 055:B5) (Sigma, St. Louis, MO), a toll like rec eptor 4 (TLR4) agonist and PamCSK3 (Invitrogen, Carlsbard, CA), a TLR2 agonist. LPS and PamCSK3 were added at a concentration of 10 100ng/mL (LPS) or 0.1 1 ug /mL (PamCSK3). Control conditions were included, containing no stimuli. After 24hrs of incubati on, the media of stimulated microglia were collected and centrifugated

PAGE 31

31 for 10min at 20C and 1200rpm. The supernatants were used for IL 6 and TNF enzyme linked immunosorbent assays (ELISA). For every condition, cytokine levels were calculated in three di fferent wells, while each experiment was performed fourfold. IL 6 ELISA Mouse IL 6 secreted protein levels were determined with a gene ral sandwich ELISA protocol. Briefly an enhanced protein binding ELISA plate (Nunc, Rochester, NY) was incubated overni ght at 4C with the capture antibody, rat anti mouse IL 6 (BD Bioscience Erembodegem, Belgium). After blocking the non specific binding for 2hrs, standards (BD Bioscience Erembodegem, Belgium) and samples were added for 2hrs at room temperature. Subseque ntly, biotinylated rat anti mouse IL 6 (BD Bioscience Erembodegem, Belgium) was used as a detection antibody. Following incubation with a Streptavidin Horseradish Peroxidase conjugate (Dako Cytomation, Heverlee, Belgium), a TMB substrate (BD Bioscience Er embodegem, Belgium) was applied to the plate. Finally, optical densities (OD) were read between 450 570nm, using a spectrophotometer (Powerwave X Select) and concentrations were calculated. The detection limit of the assay was 10pg/mL. TNF ELISA Mouse T NF secreted protein levels were measured using a commercially instructions. Concentrations were determined according to the OD values, measured using a spectrophotometer ( Powerwave X Select) at a wavelength between 450 570nm. The detection limit of the assay was 8pg/mL.

PAGE 32

32 Glutathione Measurements Total glutathione (reduced and oxidized) was measured in microglia using a glutathione reductase enzymatic recycling assay (Cayman Chemical, Ann Arbor, MI) that is based on the colorimetric conversion of nitrobenzoic acid to 5 thio 2 nitrobenzoic acid (Tietze, 1969) Briefly, microglia from the brains of young or aged mice were lysed immediately following isolation and prepared for glutathione measurements according to s were attained by combining microglia from four brains. Therefore to attain three repetitions, 12 mice per age group were assayed. All samples were normalized to total protein using bicinchonic acid (BCA) colorimetric assay (Pierce, Rockford, IL). 2 Fate Analysis was resuspended to 1mg/mL in 1% NH 4 OH and stored at 20C according to FITC was diluted to 4 ug /mL in DMEM and added to microglia for 3hrs. The cells were then stained with DAPI nuclear counterstain (1:1000) for 5min. and then imaged. For enhanced subcellular resolution of internalized cells (exposed to non conjugated fixed as described above for immunocytochemistry using antibodies against (6E10, 1:2000, Signet Laboratories, Dedham, MA) and the lysosomal associated protein, Lamp 1 (1:2, gift from Dr. Notterpek, University of Florida). To determine the aggregation fate experiments, samples were diluted in Laemmli sample buffer containing 2% sodium dodecyl sulfate (SDS) and loaded in 4 20% TG SDS gels (Invitrogen, Carlsbad, CA) for standard SDS PAGE. Immu noblots were probed with 6E10 at a dilution of 1:5000.

PAGE 33

33 photographed using a Fugi im aging system (Fugifilm Life Science, Stamford, CT). RS chambers (Nunc, Roskilde, Denmark) that contain a hybrid of glass and polystyrene surfaces have reduced non chosen for fate analysis experimentation. T o further reduce non absorption, these chambers were blocked with 10% milk for 1hr. Microglial cells were then isolated from 1 month old mice, 15 months old mice and mixed glial cultures (MGC) w ere seeded on RS chambers at approximately 3 x 10 5 cells/chamber. The cultures were incubated overnight at 37 C with 5% CO 2 and saturated humidity. The ug /mL. The cells The cells were then and that of cells lysed immediately following rinsing were collected, as were the conditioned media and lysate from wells incubated for 3hrs and 16hrs The lysis buffer consisted of NP40 (Invitrogen, Carlsbard, CA) supplemented with protease inhibitor cocktail (1x, Sigma, St. Louis, MO) and PMSF (1mM). For each age group, lysate and media representing 3 6 adult mice or 14 neonatal pups (2 (M GCs ) were collected. To style ELISA (Invitrogen, Carlsbard, CA), configured with two capture antibodies (recognizing epitopes on the N and then the C terminus (Schmidt, et al., 2005) causes N terminal truncations (Majumdar, et al., 2007b) thus ELISA reactivity is limited

PAGE 34

34 to non sample and absorbance read at 450nm using a spectrophotom eter (Bio Tek, Winooski, VT). Data was normalized to mock treated wells that were treated as described above but contained no cells. The detection limit of the assay was 10pg/mL. Statistical Analysis Average cytokine data are presented as mean SEM. Sta tistical analysis was carried out using SPSS ver. 14.0 for Windows (SPSS, Chicago, IL). To analyse differences between groups, we used unpaired, two test or ANOVA value of <0.05 wa s considered tailed ta test p value of <0.05 was considered statistically significant. Results As reported previously (de Haas, et al., 2007, Frank, et al., 2006) centrifugation of dissociated whole brain within discontinuous percoll gradients can separate microglia from other brain cells. In our hands, the techniques described in the literature yielded insufficiently pure cultures for the pulse chase experiments we performed ( Fig. 2 1 B C). Specifically, we observed that debris fields, which could possibly sequester p eptide (Li, et al., 2005) To address this, we utilized dispase II, an enzyme that has been described as particularly gentle, yet capable of tissue dissociation (Borchelt, et al., 1992, Gao, et al., 2004, McDermott, et al., 2003) Furthermore, we increased the

PAGE 35

35 density of percoll by 16% from that described in the literature (de Haas, et al., 20 07) We observed that brains treated with this methodology 1) had greater separation of dissociated microglia from tissue chunks ( Fig. 2 1 A B), 2 ) yielded numerous adherent microglia and 3 ) were largely devoid of debris ( Fig. 2 1 C D). Altogether, the c ombination of these techniques resulted in the extraction of up to 3 x 10 6 microglia per brain. On average, 8.5 x 10 5 microglia per brain were extracted from young and aged mice ( Fig. 2 2 B ). 94% of DAPI counterstained cells were reactive to Iba1 ( Fig. 2 2 A ) as determined by 3 observers. The cells had a characteristic amoeboid, phase bright morphology similar to previous reports of adult microglia isolated with different methodology (Tham, et al., 2003) To further confirm that our isolated cells were indeed microglia, we isolated cells from transgenic mice where GFP expression is under the fractalkine receptor promoter. These mice are reported to have microglia as the only brain cell type to express GFP (Jung, et al., 2000) Upon isolation of ce lls, our cultures were reactive to antibodies specific to GFP ( Fig. 2 3 A). Ex vivo cultures of adult microglia that were allowed to adhere overnight were comparable to the HEK 293 cell line in viability (Fig. 2 2 C). Recent studies have raised the possibi lity that GM CSF could push cultured microglia towards a dendritic cell fate (Esen and Kielian, 2007) In our cultures, microglia grown in 0.15 ug /ml GM CSF or in GM CSF free conditions both maintained a rounded morphology (Fi g. 2 3 B) and had no immunoreactivity to the dendritic cell antigen, Cd11c (data not shown). Interestingly, neonatal microglia derived from mixed glial cultures lacking GM CSF exhibited a ramified phenotype when cultured overnight in GM CSF containing medi a (Fig. 2 4 ). This morphology is similar to that observed in mixed glial cultures with prolonged

PAGE 36

36 exposure to GM CSF (Esen and Kielian, 2007) It is possible that the conditions within mixed glial cultures prime microglia to adopt a dendritic like morphological phenotype upon exposure to GM CSF. Figure 2 1. Dispase II density centrifugation methodology. (A) Brain hemispheres that were homogenized with dispase II and loaded onto discontinuous gradients to an isopynic density of 1.077mg/ul as determined by beads with known densities. This configuration spatially separated unwanted brain matter to a density 50ug/ul more buoyant. (B) Density centrifugati on methodology as described in the literature involved mechanical homogenization and reduced percoll densities. In our hands, such methodology failed to channel unwanted brain matter (tissue chunks, red arrowhead) to an isopynic position distal to the mic roglia enriched band. (C) Phase contrast images of freshly prepared cells under a hemacytometer demonstrate viable phase bright cells (resistant to trypan blue) as well as reduced particulate matter from dispase II homogenized brains. Following 24hrs of culture, these cells remain adherent after multiple washes (D) and are thus compatible with experiments that involve media exchanges.

PAGE 37

37 Previously, histological findings of dystrophic microglia in the aged and diseased brain have led our laboratory to sugge st microglial function may deteriorate with normal aging. Therefore, we sought to study elements of pathology that are mainly conferred by microglia in vivo and are known to change with aging and disease (Bolmont, et al., 2008, El Khoury, et al., 2007, Meyer Luehmann, et al., 2008, Streit, et al., 2004, Ye and Johnson, 1999) Recent studies have shown that mRNA copies of inflammatory cytokines are increased in microglia from aged brains (Sierra, et al., 2007, Ye and Johnson, 1999) However, mRNA transcripts may not necessarily translate to secreted protein levels (Munger, et al., 1995, Storm van's Gravesande, et al., 2002) a more ultimate measure of functional change. To determine if microglia vary their secret ion of cytokines with age, we obtained microglia from young and aged mice and measured their cytokine levels with and without exogenous immune stimulation. The most striking observation in this respect was the dramatic increase in IL 6 release under basa l conditions (young: 211.831.7 pg/ml vs. aged: 3735.91000.2 pg/ml, p<0.001). In both young and aged microglia, a significant dose effect relation following either LPS or PAMCSK3 stimulation was observed ( Fig. 2 5 A). Moreover, maximal release of IL 6 was significantly enhanced in aged microglial cells followi ng LPS (100ng/ml) or PAMCSK3 (1u g/ml) stimulation ( Fig. 2 6 ). As with IL 6, the amount of TNF significantly higher under basal conditions when compared to young microglia. While microglia derived from young mice produced no TNF amount was significa ntly increased to 917.2 91.9 pg/ml in supernatants of aged microglia cultured for 24h without any exogenous stimuli (p<0.001). This striking

PAGE 38

38 difference confirms age related higher basal levels of cytokine production previously observed with mRNA transcri pt analysis (Sierra, et al., 2007) and indicates that aged microglia are hyperactive when compared to microglia f rom young mice. This high release under basal conditions may explain the lack of a significant dose effect relation in TNF Fig. 2 5 B). In contrast, in young microglia a significant dose effect relat ion was observed. Moreover, although the maximal amount of TNF LPS were not different between aged or young microglia ( Fig. 2 6 B). Figure 2 2 Purity, yield and viability of microglia. (A) Microglia isolated with dispase II density centrifugation methodology express Iba1, a marker commonly used to identify in vivo microglia. (B) Yields of microglia from 1 month and 15 month old mi ce typically obtained using dispase II based density centrifugation methodology show little variability with age. (C) Measurement of mitochondrial respiratory activity indicated that isolated microglia form cultures comparable in viability to HEK 293 cells an immortal cell line.

PAGE 39

39 In addition to quantifying microglial cytokine production as a function of age we were interested in whether the ability of microglia to serve as an oxidative sink and to internalize changes with age. Glutathione acts as antio xidant by neutralizing free radicals and peroxides and microglia are reported to be the primary glutathione containing cells in the brain (Hirrlinger, et al., 2000, Lindenau, et al., 1998) We found a trend indicating that microglia in aged brains have 21% less tota l glutathione (oxidized and reduced) compared to microglia from young brains ( Fig. 2 7 ). This result suggests the reactive oxygen species (ROS) insult that can be caused by internalization (Milton, et al., 2008) maybe more injurious to microglia in aged brains. Figure 2 3. Adult microglial morphology (A) GFP F ractalkine Receptor transgenic mice, where microglia are the only in vivo GFP expressing neuronal cells yielded cultures composed of GFP positive cells. This provided independent confirmation that our isolation methodology extracted in vivo microglia. (B ) Adult microglial cells remained rounded and did not adopt a dendritic morphology when cultured for 24hrs with media containing GM CSF (n=2 mice).

PAGE 40

40 accumulation is a well recognized feature of AD, however extensive amyloid deposits may be found in man y aged, non demented individuals (Bouras, et al., 1994) This pathology may result from AD independent deterioration of clearance processes. Microglial scavenger activity on is prop osed as a clearance process that contributes in maintaining at physiological levels by counterbalancing constitutive secretion by neurons. Our laboratory and others have published accounts of microglial degeneration that is associated with age (Flanary, et al., 2007, Simmons, et al., 2007, Streit, et al., 2008, Streit, et al., 2004) If microglia represent a major clearance mechanism, their degeneration would result in progressively increasing levels with age and therefore would have significant implications t o the occurrence of amyloidosis in AD and some aged individuals. We currently lack the means to isolate degenerating microglia for experimentation. However ex vivo assessment of microglia acutely isolated from young and aged mice likely emulates in vivo processing of more so microglia with age. Ex vivo cultures of microglia were given media with 4 ug /mL of FITC conjugate or non cated that non resistant species larger than 220kDa that are likely fibrils thus reflecting in vivo amyloid burden ( Fig. 2 8 y confirmed with cell associated FITC fluorescence ( Fig. 2 8 colocalized with lysosomes ( Fig. 2 8 C). To quantify internalization, microglial lysates at microglia from Fig. 2 9 A).

PAGE 41

41 by neonatal microglia (derived from mixed glial cultures) is reflective of microglia from adult mice. To our surprise, neonatal microglia internalized significantly croglia derived from young mice This suggests that microglia from mixed glial cultures may not necessarily model prerequisite step for intracellular clearance; however it is by no means a surrogate invariably expelled by microglial cells in an age independent manner within 3hrs ( Fig. 2 9 B ). This result concurs with previous reports of lackl uster microglial anti in vivo (Simard, et al., 2006) that may stem from impaired lysosomal activity (Majumdar, et al., 2007b) Discussion Improvements on Microglial Isolation In this study we aimed to reduce debris contamination which is a feature of microglial cultures derived from gradient centri fugation based methodology. The brains of mice that are designated by the National Institute on Aging as an aging model were treated successively to steps that significantly increased the purity of ex vivo microglial cultures. Analysis of such cultures, derived from mice of various aging categories, revealed that microglia from aged brains have markedly increased basal levels of IL 6 and TNF om younger mice are able to temporarily contain

PAGE 42

42 ex vivo findings provide evidence that microglia are subject to age associated changes in biology. Figure 2 4. Neonatal microglial morphology. Neonatal microglia express Iba1 (A), how ever they are distinguished from adult microglia (young and aged) by ramified cytoplasmic processes visible under phase contrast microscopy and Cd11b immunoreactivity (B). Purity of neonatal microglial cultures was confirmed by lack of GFAP immunoreactivi ty, a marker for astrocytes and immature neurons. Age related Changes in Microglial Cytokine Release Prior reports from our laboratory have shown that microglia have IL 6 and TNF mRNA (Streit, et al., 2000) and more recently, others have found age rel ated changes in microglial IL 6 and TNF (Sierra, et al., 2007) We extend on these results by measuring s ecreted IL 6 and TNF (experiments performed by Ellen Boelen) Our results indicate significantly more pronounced changes in basal cytokine production

PAGE 43

43 and responsiveness. It is difficult to make direct comparisons of mRNA and secreted protein m easurements. However, it is of note that the margin of change we observe in the basal production of IL 6 between microglia from young and aged mice is approximately 4 respectively). We also did not observe detectable levels of basal TNF from young animals in our studies. These differences can perhaps be explained by varying sensitivities of the employed detection methodologies or by post transcriptional effects. The half life of mRNA can often be rate limiting in translation (Ross, 1995) Secondarily, secretory pathway modulation of newly produced cytokines may also modulate the concentration of cytokines in the extracellular milieu independent of DNA transcription. As microRNAs are involved in regulation of gene expression at the post transcriptional level, possible changes in this machinery can also be mentioned to explain the discrepancy between protein and mRNA levels. Our results, though in ag reement with previous reports, indicate that microglia from brains of various aging groups have much greater differences in cytokine production, and responsiveness to immune stimulation than was previously thought. Implications of Age related Changes in Mi croglial Cytokine Release What are the possible implications of age related changes in microglial cytokine production? A number of authors commonly describe IL 6 and TNF molecules involved in AD pathogenesis (Bruunsgaard, et al., 1999, Collins, et al., 2000, Culpan, et al., 2003, He, et al., 2002, Li, et a l., 2007, Licastro, et al., 2000, McGeer and McGeer, 2001) However, experiments presenting alternative viewpoints have been published (Brunello, et al., 2000, Lodd ick, et al., 1998, Marz, et al., 1998, Streit, et al., 2000, Tarkowski, et al., 1999, Thier, et al., 1999, Wei, et al., 1992) IL 6 may have a

PAGE 44

44 role in regeneration of injured tissue in the brain (Loddick, et al., 1998, Streit, et al., 2000, Tarkowski, et al., 1999) has known anti apoptotic properties (Wei, et al., 2001) and in mice that overexpress both IL 6 and its receptor, IL neurotoxicity (Brunello, et al., 2000) Inflammation is a component of wound healing in the CNS (Klein, et al., 1997, Streit, et al., 2000) Recently, it has been reported that lesions typically ascribed to cause AD dementia, are present in 20 40% of non diseased, aged adults (Price, et al., 2009) and both TNF 6 increase with age (Bruunsgaa rd, et al., 1999, Wei, et al., 1992) Thus the aging brain appears to exist in a constant state of injury. Inflammatory processes, such as microglial secretion of IL 6, maybe needed for persistent regeneration or neuroprotection. The viewpoints that cy tokine release is exclusively neurotoxic or neuroprotective could be equally considered speculative. Further studies which consider region specific and age specific differences in cytokine response are needed, amongst others, to shed light onto the role o f cytokines released by microglia on the nervous system. Age related Changes in Microglial Glutathione Levels Microglial surveillance of the parenchyma involves scavenging of potentially the increased levels of antioxidant molecules such as glutathione found in microglia relative t o other brain cells (Hirrlinger, et al., 2000, Lindenau, et al., 1998) protect microglia from ROS burden associated with scavenging activity (Dringen, 2005, Milton, et al., 2008, Qin, et al., 2004, Tchaikovskaya, et al., 2005) Our findings of reduced glutathione levels in microglia immediately analyzed after brain extraction maybe indicative of broader microglial loss of function with age.

PAGE 45

45 Figure 2 5. Microglial reaction to immunostimulation. Dose dependent increases in either IL 6 (A) or TNF (B) release following different concentrations of LPS (10 100ng/ml) or PAMCSK3 (0.1 1ug/ml). Under all conditions, except for the TNF secretion by aged microglia following either LPS or PAMCSK3 stimulation, a significant effect of concentration was observe d. No significant age dose interaction was observed under all conditions. Moreover, all figures show the marked increase in cytokine release by aged microglia under basal condition (Cytokine experiments performed by Ellen Boelen). Age related Changes in Mi croglial Processing Microglia are thought to participate in burden regulation by sequestering and processing of A ex vivo cultures of microglia was to further understand the relationship between micro glia as plaques become smaller with time (Bolmont, et al., 2008) Despite this correlative evidence of mi appear to reverse amyloidosis (Jankowsky, et al., 2005) Previously, we have suggested that microglial function may deteriorate with time. This was mainly inspired

PAGE 46

46 by histological findings, demonstrating the appearance of a dystrophic microglia phenotype with normal aging and around deposits in AD brains (Miller and Streit, 2007) The fi microglia from aged brains do not, likely reflects a more global change in microglial functionality with age as we do not observe dystrophic microglia in aged mice. Interpretation of Expuls ion by Younger Microglia not directly involved in degradation. However there is clear potential for microglia, which are highly motile (Bolmont, et al., 2008) through microglial endocytosis and exocytosis could result in redistribution that modulates peptide availability for amylo id formation. at ng/ml quantities, possibly creating a dynamic cellular compartment in vivo A 2 to 4 fold age observed in our -thereby amount of non diseased aged adults (Price, et al., 2009) T herefore, it is plausible that needed for homeostatic maintenance. Perturbation of this process due to age related changes of microglial function such as that described here could contribute to unhinged

PAGE 47

47 Figure 2 6. Cytokine secretion of young and aged microglia. (A) Upon stimulation with the biological inflammatory reagent LPS (100ng/ml) or Pam3CSK 4, a synthetic agonist of toll like receptor 2 (1 ug /ml), IL 6 production by aged microglia was markedly increased when compared to young microglia. (B) LPS (100ng/ml) stimulated similar TNF production between microglia derived from young and aged mice. Y et, TNF production was significantly increased in aged microglia following Pam3CSK4 (1 ug /ml) exposure. *, p<0.05; #, p<0.001. Concluding Comments Together, our ex vivo quantification of microglial functions of cytokine production, glutathione levels an activity. Microglia from young mice produce less cytokines, while microglia derived from aged mice have higher basal levels of cytokine secretion than was previously thought and have reduced glutathione levels. Microglia derived from aged mice lack microglia from young mice and mixed glial neonatal cultures do not seem to retain

PAGE 48

48 as sessments of microglial function in ex vivo experiments, free of confounding contributions of other brain cells and debris, demonstrate a nuanced view of microglial function and suggest that microglial biology may change with aging. Figure 2 7. Microgl ial glutathione content. Microglia from aged mice have a 21% reduction of glutathione antioxidant. Data is normalized to total protein and represents oxidized and reduced forms of glutathione detected in microglia analyzed immediately following brain ext raction. p=0.27

PAGE 49

49 Figure 2 8 (A) Western blot analysis monomeric (4kDa), oligomeric (16kDa, 20kDa) and higher order conf ormations larger than 220kDa in stock preparations as well as preparations that have been exposed to microglia. Higher order conformations larger than 220kDa persist following the exposure of samples to buffer containing 2% SDS suggesting the presence of fibrillar species. 10% pepti peri nuclear localization. Lamp1 colocalization with 6E10 immunoreactivity

PAGE 50

50 Fig ure 2 9 roglia Microglia extracted from mice of various and SDS resistant fibrillar species (reflecting in vivo amyloid diversity) in pulse chase experiments. (A) Neonatal and young microgli a respectively within 3hrs of ingestion, suggesting disengagement from biophysical degradation following phagocyt osis. Mock data (gray) represents experiments without the presence of cells to control for non antified in the above experiments. *, p<0.05; **, p<0.01.

PAGE 51

51 Figure 2 10. Overview of mixed glial culture (MGC) and density centrifugation methodologies utilized to obtain microglia. MGC derived microglia were harvested after 1 to 4 weeks of growing atop a multi cell feeder layer comprised mostly of astrocytes. Homogenized brains of neonatal origin form MGCs while adult brains do not, thus MGC microglia are of neonatal origin. Density centrifugation based isolation offered the advantages of immediate cult ure creation from young and aged brains. This was important because microglia from brains of aging models were accessible and ex vivo analysis utilized cells temporally proximate to in vivo microglia.

PAGE 52

52 CHAPTER 3 ENGRAFTMENT PATTERNS OF NSCS TR ANSPLANTED INTO MOUSE MODELS Introduction The brains of AD patients are characterized by pathology as well as neuronal cell death. This has led to interest in the use of neuronal stem cells (NSC s ) for cell replacement or delivery of therapeutic molecules such as proteases that reduce pathology. Before such studies can be carried out on humans several fundamental questions must first be addressed. First, do NSCs survive in br ains with the pathologies of AD? Second, how does the complex architecture of the brain effect the engraftment of transplanted NSCs ? Third, does the region of en graftment affect integration and migration within the parenchyma ? And finally, does transgene expression in genetically modified NSCs effect engraftment? The question of NSC survival in AD or mode ls of AD has been explored in in vitro and in in vivo (non transplant) settings Unfortunately, these studies present contradictory data on how NSCs are a ffect ed by pathology Specifically, in vitro and post mortem studies yield data contradictory to studies performed on mouse models of AD. Lopez et al., and subependymal zone (SEZ) and subgranular zone (SGZ ) mouse NSCs to differentiate s do not induce apoptosis in SGZ and S EZ stem cells or hinder their proliferation (Calafiore, et al., 2006, Lopez Toledano and Shelanski, 2004) These observations are consistent with research which quantifi ed increases in neurogenesis in humans with AD (Jin, et al., 2004, Ziabreva, et al., 2006) Conversely, multiple publicatio ns report evidence of decreased and abnormal neurogenesis in the hippocampus of several mouse models of

PAGE 53

53 AD (Dong, et al., 2004, Donovan, et al., 2006, Wang, et al., 2004) Our lab recently of new born SGZ neurons as they reach functional maturity. endogenous NSC s how transplanted NSCs will behave in AD patients remains an open question. To shed light on this topic we performed NSC transplant studies in mice modeling AD pathology Our specific aim was to define engraftment patterns in the hippocampus The pathology in the hippocampus is linked to dementia (Braak and Braak, 1995, Braak, et al., 1996) Therefore the hippocampus is a region of great clinica l interest for cell restoration and drug delivery therap ies Our primary findings in this study demonstrate that physical forces relating to hippocampal anisotropic architecture dictate the distribution of NSCs more so than cell migration We also find t hat genetic overexpression of MMP9 is associated with significant enhancement of NSC graft size Methods Isolation of NSCs The protocols for isolating NSCs are contained in the literature (Marshall, et al., 2006, Zheng, et al., 2006) In brief, a rectangular forebrain block containing the subependymal zone was isolated from neonatal (P4 P9) green fluorescence protein transgenic mice (003116, The Jackson Laboratory, Bar Harbor, MI) or from non transgenic B6 mice (bred in house). This was done by removing the OB, cerebellum, hippocampus, lateral portions of the striatum, and lateral and dorsal cerebral cortex. This blo ck was minced with a razor blade, incubated in 0.25% trypsin/EDTA (Atlanta Biologicals, Lawrenceville, GA) and dissociated into a single cell suspension by triturating through a diametrically descending series of glass pipettes. Cells were then

PAGE 54

54 pelleted a nd washed several times before plating in NSC media ( DMEM/F12 with 5% FBS, penicillin (100U/ml), streptomycin (100ug/ml), Bovine Pituitary Extract (35ug/ml), Fungizone (250ng/ml). NSC culture flasks by supplementing the media with 20ng EGF and 20ng FGF every two to three days (Walton, et al., 2006a, Zheng, et al., 2006) Transplantation into Amyloid Beta AD Mice The Line 85 and the Line 107xtTa mouse models, which constitutively overexpress an d selectively express human respectively, have been previously described to model the physiology that is a hallmark AD pathology. Host and donor mice are immune matched due to their shared B6 background. M ice were induced to a state of deep anesthesia with 1 5% isoflurane. The hair on their scalps was shaved and the surgical area sterilized with betadine antiseptic and 70% ethanol. The mice were securely mounted with ear bars and a nose bar to a stereotaxic apparatus. The anesthesia mixture was delivered through an inlet within the nose bar enclosure for the duration of the surgery. A sterile scalpel was used to make a small incision into the skin above the skull. The skin was reflected in order to expose Bregma. A Hamilton 10ul syringe with a 33 gauge needle (Hamilton Company, Reno, NV) was then loaded with NSCs prepared at 1 x 10 5 cells/ul for lateral ventricle injections and ~ 5 x 10 4 cells/ul in 1x dPBS for multi deposit hippocampal injections The cells were derived from a trypsinized and pelleted monolayer of NSCs washed with 200ul 1x dPBS and diluted to the appropriate volume using a reference cell count done on a hemacytometer. The following coordinates relative to Bregma were used to target the lateral ventricle : anterior/posterior ( AP ) : 0.2mm medial/ lateral ( ML ) : +/ 1.2 mm, dorsal /ventral ( DV ) : 2.5mm from the dura Occasionally, needle tracks using these coordinates were found

PAGE 55

55 in the base of the hippocampus where it meets the thalamus. This border region is not termed in modern atlases (Paxinos Atlas and Al len Online Atlas) and vaguely referred to in a study by Nagaraja and colleagues (Nagaraja, et al., 200 5) Cerebrospinal fluid research 2005) There may be cerebrospinal fluid in this border in the posterior brain (midbrain). However, this is not the case in the rest of the brain. Therefore, we refer to this region here as the hippothalamic fissure r ather than the hippothalamic cistern. To target the corpus callosum we used: AP: +1.2mm, ML: +/ 0.5mm, DV: 2.5mm from the dura We used multiple depth coordinates in an attempt to maximally disperse cells in the hippocampus. The regions targeted wer e the ventral hippocampus (dentate gyrus, SGZ region), the medial hippocampus (molecular cell layer, hippocampal fissure region) and the dorsal hippocampus (CA3) The coordinates used were : AP: +2mm, ML: +/ 2.0mm, 2.3mm, 2.5mm from the skull surface. 1.25 x 10 5 cells were deposited at 2.0mm and 2.3mm, while 2.5 x 10 5 cells were deposited at 2.5mm. In all injections, 4ul 8ul total volume was deposited at the rate of 0.25ul per 15 seconds (dependant on cell concentration) Subsequently, t he needle was left alone for 5 minutes to allow for the diffusion of cells from the injection tract. It was then retracted slowly to minimize damage along the injection tract. The incision was closed with a staple and the mouse placed in a warm, dark reco very area. To avoid or minimize the discomfort, distress and pain associated with this procedure, 0.1mg/kg of buprenophine was administered as the animals recover ed on the day of surgery and the subsequent day The mice were checked for signs of abnormal recovery during the survival periods. At the end of the survival periods, the mice were deeply anesthetized in an isoflurane induction chamber, euthanized by Beuthanasia administration and then perfused with

PAGE 56

56 cold 1x PBS. Whole brains were quickly disse cted out and placed in cold 4% paraformaldyhyde fixative overnight. The brains were immersed in 30% sucrose before sectioning within a cryostat. 20 um sections were stored in anti freeze media at 20C until further processing. Immunochemistry 4% parafor maldehyde fixed c ells or tissue sections proc essed for immunofluorescence in solutions containing 0.1% Triton X, 10% goat serum in 1x PBS. Primary antibodies used in this study include copGFP (1:20 00, Evrogen Moscow, Russia ), anti human MMP9 Clone 56 2A4 (Abcam, Cambridge, MA ) anti 6E10 ( 1:2000, Signet, Dedham, MA) microglial antigen Iba1 (1:1000, Wako, Richmond, VA ), astrocyte antigen GFAP (1:1000, Dako Corporation, Carpinteria, CA), and neuronal antigen NeuN (1:500, Millipore, Bellirica, MA) neuronal antigen BIII Tubulin (1:50 0, Covance, Princeston, NJ ) Cells were rinsed and incubated with goat secondary antibodies Alexa 488, 568 (Invitrogen, Carlsbard, CA). Cells were photographed with a n Olympus DP71 camera mounted on an Olympus BX60 microscope Modeling Paths Of Least Res istance (PLR) To simulate that may occur within the hippocampal formation agarose (w/v) in PBS was allowed to solidify in a clear 15ml polystyrene tube. Addition of 100ul of H 2 O atop this solidified block formed a layer ~1mm to 3mm in h of warm 1% agarose was then added. As the two layers of agarose fused, a region of weakness developed due to an abrupt decrease in agarose density at the H 2 O layer. 5 10ul of crysyl violet solution was then injected at various regions in th e agarose block.

PAGE 57

57 Results NSCs robustly propagate in vitro (Reynolds and Weiss, 1992, Walton, et al., 2006a) In our hands, cells from the brains of two neonatal mice have been p assage d >10x for a duration of six months producing a conservative estimate of approximately one hundred million cells. NSCs form ed monolayers of elongated cells ( Fig. 3 1 A, C ) similar to that of human NSCs (Walton, et al., 2006a) NSCs were also similar to human NSCs in their immunoreactivity for the neuronal progenitor markers GFAP and III Tub ( Fig. 3 2 A B ). Iba1 reactivity was observed in some cells in NSC cultures ( Fig. 3 2 C ), indicating the presence of microglia. However, these cells were much larger than adult ex vivo microglia (compare Fig. 3 1 B C & 3 2 C) and had a morphology simi lar to neonatal microglia from mixed glial cultures ( Fig. 2 .4, see Chapter 2). It has been suggested that microglia derived soluble factors maintain the self renewal capacity of NSCs in vitro (Walton, et al., 2006b) Figure 3 1. Morphology of in vitro NSCs. NSCs form monolayers of elongated cells (A) and are signific antly larger than microglia acutely isolated from adult mice (B).

PAGE 58

58 We initially transplanted NSCs from GFP transgenic mice into the lateral ventricle (LV) in order to recapitulate previous reports of rostral migration into the olfactory bulb (Marshall, et al., 2006, Zheng, et al., 2006) In our AD models, this is a region of high burden and in humans, the loss of smell is purportedly amongst the first signs of AD (Fusetti, et al.) Therefore, the migratory potential of NSCs suggests the possibility of wide spread dispersion of candidate therapeutics to relevant regions of the brain. We performed trans plants of GFP cells on non transgenic mice and transgenic coordinates (Zheng, et al., 2006) Similar to previous studies with SEZ NSCs (Walton, et al., 2006a, Zheng, et al., 2006) mice were sacrificed 4 8 weeks later In transplants where the needle track was confirmed to have entered the lateral ventricle we did not observe ventricular wall colonization or extravasation towards the olfactory bulb (non transgenic ; n=3, TG=4 ) ( data not shown ). This result was not due to the inability of NSCs to form grafts as some surviving cells were observed along side n eedle track s that ha d entered the lateral ventricle These results motivated us to modify stereotactic coordinates to directly target the hippocampus. Our goal was to achieve maximal cell di spersion within the hippocampus. We therefore chose 3 point de pth coordinates for simultaneous injection into the dorsal hippocampus (proximal to CA3), the medial hippocampus (hippocampal fissure, molecular cell layer and lateral arm of the dentate gyrus ) and the ventral hippocampus (subgranular zone of the dentate g yrus) The accuracy of these coordinates was verified with injections of the tracer crysl v iolet (n=4/depth coordinate data not shown) on freshly deceased mice A total of 5 x 10 5 cells were injected, a

PAGE 59

59 value corresponding with previous animal work (Par l et al., Exp Neurol 2006) and proportional to human NSC clinical trials (Stem Cells Inc; American Association of Neurological Surgeons Annual Meeting 2010 ). Figure 3 2. In vitro characteristics of NSCs. NSC cultures were characterized by GFAP (A) and BIII Tubulin immunoreactivity (B). Some cells were immunoreactive for the microglial marker, Iba1 (C). n=4 cultures per immunostain. We then injected NSCs with lentiviral directed expression of 1) MMP9 and GFP or 2) GFP only The GFP used here is from t he plankton copepod. It is similar in size to EGFP (26kDA), bu t has more fluorescence output (Shagin, et al., 2004) GFP NSCs and MMP9 NSCs had equal fluorescence output of A. U> 10 3 as determined by fluorescence activated cell sorting (FACS) ( see Chapter 4). After a month of survival, brains were confirmed to have hi ppocampal needle track penetration in several mice MMP9 NSCs formed graft cores that were 82.4%

PAGE 60

60 larger than GFP NSCs (n=4; GFP NSC, n=5; MMP9 NSC, Fig 3 3A C ). Interestingly, the effect of MMP9 on survival is diminished with increasing amounts of prox imal A plaques (R 2 =0.701 ) ( Fig 3 3D ). This data suggests that amyloidosis is toxic to NSCs and that MMP9 genetic modification partly rescues cell survival. No ot her differences were observed between GFP NSCs and MMP9 NSCs The engraftment patterns described below apply to both cell types. Figure 3 3 MMP9 associated changes in engraftment. MMP9 NSC grafts (n=5) were 82.4% larger than GFP NSC grafts (n=4) (A C). The fluorescence intensity of both cell types were matched with FACS analysis (A.U >10 3 ). The size of plaques ( R 2 =0.701 ) (D). As seen here, the horizontal distribution of GFP NSCs and MMP9 NSCs appears similar. *, p<0.001 Both GFP NSCs and MMP9 NSCs formed graft s in su bcortical and corpus callosum white matter tracks despite injection within the interior of the hippocampus (Fig. 3 3A B, Fig. 3 4 A,C ) This distribution was observed in all mice (n= 9 ). NSCs

PAGE 61

6 1 formed a smaller, secondary graft within the hippocampal fissure in 33% of mice ( Fig. 3 3 4B ) This distribution pattern was confirmed in a separate cohort of mice that survived two months following surgery (n=7), (Fig 3 9A ) NSC engraftments in the white matter and the hippocampal fissure had a consistent disc shap ed morphology that extended horizontally; ~1mm on the anterior/posterior plane and ~1mm on the medial/lateral plane in needle track containing sections and adjacent sections. NSCs injected directly into the corpus callosum and allowed to graft for 3 days in non transgenic mice (n=2) result ed in engraftment that was strikingly similar to the distribution found in ( compare Fig. 3 4A & D ) Furthermore, NSCs that were engrafted for 1 year in the fimbria white matter of non transgenic mice were similarly distributed horizontally at the hippothalamic fissure (n=3) Fig. 3 4E F ) Recently, several investigators have suggested directional migration of NSCs as an explanation of why cellular transplantation into the hippocampal gray matter paradoxically results in engraftment of cells in the white matter (Blurton Jones, et al., 2009, Pihlaja, et al., 2008, Radojevic and Kapfhammer, 2009, Raedt, et al., 2009, Tang, et al., 2008) As described above, o ur transplant studies were characterized by 1) white matter graft s in 2) white matter graft s in non transgenic mice receiving NSCs in the white matter of the fimbria and corpus callosum. With the exception of NSCs closely associated with the hippocampal fissure tran splanted cells did not form graft cores i n gray matter. If cell migration were a major factor determining the distribu tion of transplanted NSCs, one would expect a range of hippocampal distribution reflecting NSCs in transit. However, our long and short term survival studies resulted in a predictable distribution of NSCs

PAGE 62

62 along horizontal fissure s of th e hippocampus and surrounding white matter Consequently, we hypothesized that the distribution of transplanted cells was largely determined by t exist between anisotropic regions of fissures, white matter and densely packed layers of neurons To provide support for this hypothesis, we decided to simulate a region of horizontal weakness using 1% agarose. We placed a thin layer of water (~1 to 3 mm) on top of already solidified agarose This created a non uniform (anisotropic) region of agarose density when a new layer of agarose was added C rysl violet tracer that was injected above or below this region resulted in a vertical distribution of tr acer (n=5 ), ( Fig. 3 5 A B) However, a horizontal distribution was observed upon injecting into the region of weakness (n=3 ) ( Fig. 3 5 C ). These results demonstrate a change in agarose density creates anisotropic forces that direct movement of i n fusate al ong a horizontal path. Such forces may occur in the hippocampus If this is the case anisotropic forces that distribute NSCs during surgery may explain why transplantation of NSCs into hippocampal gray matter results in distal white matter engraftment. To gain insight on this question, we injected GFP NSCs proximal to the corpus callosum, h ippocampal fissure, the dentate, the hippothalamic fissure and sacrificed mice immediately following surgery. As a control, w e also injected GFP NSCs into regions of the brain we suspected to have a more uniform (isotropic) distribution of tissue. These included the cortex, the hippocampus and the striatum. We found that injection of NSCs into the hippocampus at various dorsoventral depths resulted in immediate NS C distribution along the horizontal features of fissures and white matter (Fig 3 6 ). Specifically, NSCs that were injected into the dorsal aspect of the

PAGE 63

63 hippocampus yielded distributions throughout the subcortical white matter tracks above the hippocampu s (corpus callosum, cingulum bundle, alveus) and to a lesser extent, CA1 (n=8), (Fig. 3 6A). NSCs targeted to the ventral aspect of the hippocampus distributed mediolaterally at the base of the hippocampus ( i.e., hippothalamic fissure) (n=8 ) ( Fig 3 6B ). The horizontal distribution of the NSCs extended hundreds of micrometers; from the 3 rd ventricle to lateral regions such as above the dorsal lateral geniculate nucleus. NSCs targeted to the dorsoventral center of the hippocampus (hippocampal fissure coo rdinates ) distributed within the corpus callosum, the hippocampal fissure and the base of the hippocampus (n=6 ) ( Fig 3 6C ). Figure 3 4. Survival and distribution of transplanted NSCs. Deposition of NSCs into engraftments within subcortical and corpus callosum white matter (n=9) (A,C). In 33% of mice, seconda ry engraftments were observed in the hippocampal fissure (B, arrows). Deposition of NSCs into the corpus callosum (n=2) or fimbria (n=3) of non transgenic animals resulted in PLR engraftments with similar horizontal distributions (D E, arrows indicate nee dle track arrowheads indicate hippocampal fissure ). The separation of tissue observed in panel F (double arrows) provides evidence of a fissure with natural weakness into which NSCs selectively distributed into. month following NSC transplantation (D F). Non transgenic mice were sacrificed 3 days (A) and 1 year (B,C ) following NSC transplantation

PAGE 64

64 This result agrees with previous studies investigating the influence of hippocam pal structure on infusate distribution patterns (Astary, et al., 2010) As noted in Astary et al, distribution profile and shape of our infusions were also depend ent on neuroanatomical and cytoarchitectonic structure. Therefore, in contrast to NSCs injected into anisotropic white matter or cell free CSF filled regions, NSCs that were injected into the thalamus (n=6) (Fig. 3 6D) cortex (n=2 data not shown ) or stri atum (n=6) (Fig 3 6E) had circular or vertical distributions. Notably, NSC striatal distribution bifurcated along the gray matter and radiating fibers of white matter (Fig 3 6E ). Figure 3 5. Modeling paths of least resistance. A 1% agarose block, w ith structural weakness in the region denoted by arrows and crosses was injected with the tracer, crysl violet. Deposition above (n=5) (A) or below (n=5) (B) the arrow resulted in a vertical distribution of the tracer. Deposition at the area of structura l weakness resulted in a horizontal distribution of the tracer (n=3) (C). This data together with our long term survival experiments suggests that the initial distribution of transplanted NSCs is the main determinant of eng raftment pattern Our long ter m survival experiments also demonstrated that transplanted NSCs are not characterized by wide spread penetration into the hippocampal parenchyma. However we have encouraging results that suggest an alternate route may be used to attain such a result. E xt ensive intra hippocampal graft s occurred in transplants that deposited NSCs in the point of the hippothalamic fissure that is proximal to the dentate ( Fig. 3 7 ).

PAGE 65

65 Similar to transplants that result in subcor tical white matter engraftment, NSCs distributed horizontally at the hippothalamic fissure However, engraftment in this region uniquely result ed in significant representation of cells ar ound and within the dentate In transgenic mice, t hese cells extended processes into the surrounding gray matter at two weeks (Fig 3 7 A) and appear to be migratory at two months (Fig. 3 7 B) In a non t ransgenic animal that survived five month s, NSC graft s were found within the denta te of the ipsilateral hippocampus (Fig. 3 7 C) However, in the contralateral hippocampus, NSCs were found in the hippothalamic fissure (Fig 3 7 C, 2 nd and 3 rd panels) This suggests cells were deposited at this PLR In both hemisphere s, cells appear to b e migrating dorsally from the graft core (Fig. 3 7 C) We did not find cells in the thala mus in these animals. This suggests vertical directional migration away from the thalamus possibly d ue to factors associated with the endogenous NSC niche that exists in the SGZ The robustness of these engraftment s prompted us to perform further transplants to repeat these results. These studies are ongoing. Previous work from our lab has shown that the s urvival of newborn neurons ( NSC progeny ) is negatively correl ated with the presence of pathology (Verret, et al., 2007) However, the survival of astrocytes, which were more than 50% of the newborn populati on, was not affected. The differentiation of transplanted NSCs has been well characterized in multiple studies. In general, 3 0 90% of endogeneous and transplanted NSCs express glial markers in multiple studies using various NSCs in different disease mode ls (Hattiangady, et al., 2007, Lundberg, et al., 1997, Shetty, et al., 2008, Svendsen, et al., 1996, Tang, et al., 2008, Verret, et al., 2007) A minority of transplanted NSCs become neurons (Shetty, et al., 2008) However, a few studies

PAGE 66

66 describe neurons as the major phenotype transplanted NSCs differentiate into (Bennett, et al., 2010, Lu, et al., 2007, Park, et al., 2002) Due to this ambigui ty, w e were interested in determining the differentiation state of NSCs transgenic mice As mentioned above, these cells are immunoreactive for GFAP and Tubulin (Tuj1) in vitro. A subpopulation of cells in NSC cultures express Iba1. We stained tissue sections containing NSC whi te matter engraftments with antibodies against these markers and NeuN, a marker for post mitotic neurons. We did not observe Tubilin or NeuN immunoreactivity in any sections (n= 2 mice 4 x20 u m sections ), ( data not shown). However, there was strong gr aft associated immuno reactivity for Iba1 and GFAP ( n=3), ( Fig. 3 8 3 9 ) T he elonga ted morphology of Iba1 and GFAP positive NSCs in vitro m atches that of transplant ed cells in vivo (compare Fig 3 2 to 3 8 & 3 9 arrow s ) This morphology contrast s that of endogenous quiescent glia or glia reacting to n earby s (Fig 3 8, 3 9 arrow heads ) As described above, transplant cells largely remained within a graft core. Amyloid beta plaques proximal to graft cores (<50um or ~3 cell lengths) did not attract NSCs (Fig 3 4A 3 8B, 3 9 ) thus confirming prior r eports (Burton Jones, PNAS 2009). It is possible that white matter tracks support a unique NSC niche that is defined by a glial like phenotype. However, transplanted cells in this niche do not respond to or do not have access to the cues that direct chem Discussion Newly discovered properties of NSCs, namely long term in vitro expansion and engraftment potential (Walton, et al., 2006a, Zheng, et al., 2006) have generated significant interest for applic ation towards novel therapies such as cell replacement and

PAGE 67

67 Figure 3 6. Immediate distribution of NSCs in paths of least resistance. 5 x 10 5 NSCs were injected into various regions and analyzed immediately following surgery. Deposition of NSCs in the hi ppocampus which has variably dense layers of tissue resulted in horizontal distribution along anisotropic paths of least resistance (A C). These included the corpus callosum (n=8) (A), the hippothalamic fissure (n=8) (B), the dentate (n=6) (not shown) and the hippocampal fissure (n=6) (C). Interestingly, NSCs whose target was the hippocampal fissure were found hundreds of microns away from the site of injection (C). Because directional migration occurs in the order of hours to days, this distribution can only be explained by anisotropic forces directing cell distribution. In contrast to anisotropic horizontal distribution observed in A C, a vertical or circular distribution was observed with NSCs injected into isotropic regions such as the thalamus (n=6) (D), the cortex (n=2) (data not shown) and the striatum (n=6) (E). In the striatum, NSCs distributed in a radial pattern that likely reflects white matter that is striated within gray matter. drug delivery However, a series of clinical questions regard ing how NSC grafts behave in disease d brain s motivated us to study transplanted NSCs in mouse models of AD In this study, we aimed to determine the engraftment patterns of NSCs in mouse brains Our studies indicate that NSCs can survive in an environment with

PAGE 68

68 pathology however these cells do not migrate towards plaques The localization of cell engraftment is largely dependent on previously uncharacterized path s of lea st resistance. These paths distribute cells in a horizontal pattern along both lateral/medial and anterior / posterior planes. Process extension and migration appear most robust in NSCs deposited proximal to the hippothalamic fissure and dentate Graft si ze is significantly enhanced by overexpression of MMP9. Because MMP9 NSC graft size is inversely correlated to plaque number MMP9 may be a novel factor for protecting NSCs in pathologic amyloid environments Paths of Least Resistance versus migration Our observations of the pattern of NSC engraftment are also observed in studies that focus e on a myriad of pa thologies including ischemia, epilepsy and disease (Blurton Jones, et al., 2009, Olstorn, et al., 2007, Pihlaja, et al., 2008, Prajerova et al., Radojevic and Kapfhammer, 2009, Raedt, et al., 2009, Tang, et al., 2008, Watson, et al., 2006) In these studies, NSCs ( including those from human s) (Olstorn, et al., 2007) formed graft cores at sites distal to where they were targeted. To explain this paradox, several authors suggest transplanted cells directionally migrate (Blurton Jones, et al., 2009, Pihlaja, et al., 2008, Raedt, et al., 2009, Tang, et al., 2008) If migration is the primary force determining why transplanted NSCs are frequently found in white matter, one would expect a trail of in transit cells orienting away from t he site of deposition. We not observe NSCs in the hippocampa l fissure migrating dorsally towards the white matter or ventrally towards the hippothal amic fissure Migration undoubtedly occurs in vivo especially to wards lesions. H owever as observed in Olstorn et al., migration to a site of infarct is secondary to a PLR distribution (Olstorn, et al., 2007) In this study, transplant human NSCs in a non injured brain distributed hundreds

PAGE 69

69 of micrometers in the corpus callosum white matter. These NSCs then migrated tens of microm eters to the infarct zone in CA1. Within the infarct zone, a PLR like distributio n was maintained. This suggests a pre existing PLR distribution was a significant factor in determining the final distribution of migrating cells Figure 3 7 Engraftment patterns of NSCs deposited at the ventral border of the hippocampus. Needle tracks (A, arrow, left panel) which deposited NSCs proximal to the hippothalamic fissure resulted in NSC distribution along this horizonta l border and the dentate (A C). Process extension and directional animals (n=2) (A,B) and in a non transgenic animal (n=1) (C). Unilateral injection resulted in bilateral distribution of NSCs along the hippothalamic fissu re (C, middle, right panels). No cells were found in the thalamus in these studies. Our experiments on mice sacrificed immediately after surgery provide definitive proof that NSCs distribute hundreds of micrometers along horizontal Depending on whi ch region of the hippocampus was targeted, NSCs distribute d horizontally in 1)

PAGE 70

70 the subcortical and corpus callosum white matter, 2) the hippocampal fissure or 3) the hippothal amic fissure or in all three. Horizontal distribution along was also seen in MRI studies of tracers infused into the hippocampus (Astary, et al.) Our observation of NSCs horizontally spread in graft cores hundreds of micrometers apart within minutes of being injected in the hippocampal parenchyma provides a mechanistic demonstration of engraftment patterns observed in multiple studies (Blurton Jones, et al., 2009, Olstorn, et al., 2007, Pihlaja, et al., 2008, Prajerova, et al., Radojevic and Kapfhammer, 2009, Raedt, et al., 2009, Tang, et al., 2008) In studies were less than 1 x 10 5 NSCs were transplanted, PLR distribution is infrequently found (Hattiangady, et al., 2007, Ryu, et al., 2009) PLR distribution is also not fou nd in studies where NSCs are compacted into neurospheres (Shetty, et al., 2008) or restricted in distribution by nature of being enclosed in a capsule (Imitola, et al., 2004, Park, et al., 2002) An exception to this obser vation is a neurosphere study by Radojevic and colleagues (Radojevic and Kapfhammer, 2009) The amount of N SCs injected in many rodent studies equals or exceeds 1 x 10 5 cells In our study, we chose to inject 5 x 10 5 NSCs because this amount is proportional by weight to the amount of NSCs injected in the first and currently only FDA approved clinical trial for transplanting human stem cells into humans ( Stem Cells Inc ) In these studies, 1 to 2 billion NSCs were injected into children suffering from Batten s disease This amount recently passed phase 1 saf e ty trials ( American Association of Neurological Surgeo ns Annual Meeting 2010 ) Another reason to inject greater than 1 x 10 5 NSCs is that brain transplants in rodents and humans are generally characterized by 40 97% loss of transplanted cells (Bjorklund, et al., 2003, Lundberg, et al., 1997, Olstorn, et al., 2007,

PAGE 71

71 Raedt, et al., 2009, Tang et al., 2008) Therefore, the amount of NSCs to be injected in future studies that explore stem cell potential in mice and huma n s will likely yield PLR distributions. T he importance of in these future studies is further underscored by the persis tence of PLR di stribution s in two month engraftments within mice modeling in one year engraftments with in non diseased mice Research and Clinical Relevance of Paths of Least Resistance The PLR s described here inherently restrict c ell distribution. This presents is a possible pitfall in the exploration of NSCs for use in cell replacement or drug delivery. For instance, studies by Prajerovaet et al., resulted in corpus colossal PLR distribution despite transplantation with cort ical coordinates Alternatively, one can view as an opportunity to attain consistent engraftment patterns in the course of multiple transplants. The h orizontal nature of hippocampal favor precision by buffering against stereotaxic error In our experience deposition across a range of hundred s of micrometers in the m edial/lateral and anterior/posterior plane yi eld ed similar hippocampal engraftments. We envision targeting of v arious for uniq ue research and clinical aims. The PLR of the hi ppothalamic fissure appears to support dorsal directionality of cell migration. The extensive process extension by cells in this region suggests integration into the dentate architecture On the other hand, cells in the white matter PLR appear stationa ry for at least two month s in AD mice Genetically modified NSCs stationed in this PLR have access to both the cortex and hippocampus for long term infusion of therapeutic molecules However the dense layer of CA1/CA3 neurons may restrict movement of inf usate. More

PAGE 72

72 studies are needed to understand the effect of dense cell layers on the distribution of candidate therapeutic molecules. To further under stand the movement of cells injected into the hippocampus MRI technology may be used to visualize quant um dot labeled NSCs (T. Zheng, personal communication) This technology can possibly reveal coordinates for bilateral engraftment with unilateral cell injection. Our observation in one mouse provides evidence for a pathway, perhaps through the 3 rd ventri cle for contralateral colonization This pathway may allow for less invasive brain surgery on humans without sacrificing therapeutic distribution. MMP9 Associated Changes in Graft Size We find larger NSC graft sizes are associated with overexpression of MMP9 T his result demonstrates that ex vivo genetic modification of NSCs can result in changes in engraftment pattern To our knowl edge, this finding is novel and may be an important observation to consider in future studies that aim to deliver other ca ndidate therapeutics through NSC overexpression and transplantation It is possible that MMP9 is making the graft site more survivable by degrading toxic molecules such as The negative correlation of graft this However, it may be that MMP9 is modifying the ECM in a manner that enhances cell survival Studies of MMP9 have largely concentrated on it s association with cancer (Chambers and Matrisian, 1997, Lubbe, et al., 2006, Moll, et al., 1990) However, a growin g body of work demonstrates the function of MMP9 in important biological functions. Of particular relevance to this study is recent work d emonstrating th e need for MMP9 in the migration of endogenous NSCs in vitro (Wang, et al., 2006) and to sites of ischemia (Kang, et al.,

PAGE 73

73 2008) It is likely that there exists a spectrum of MMP9 physiology. At one end of this spectrum, MMP9 enables metastasis of can cer cel ls (Chambers and Matrisian, 1997, Lubbe, et al., 2006, Moll, et al., 1990) At the other end, MMP9 facilitates endogenous NSC function in stroke (Kang, et al., 2008) modulate s (Y in, et al., 2006) and is associated with larger NSC graft s in a pathological environment, as demonstrated here As mentioned above, multiple laboratories report 40% to 97% loss of NSCs following transplantation in animals and humans (Bjorklund, et al., 2003, Lundberg, et al., 1997, Ols torn, et al., 2007, Raedt, et al., 2009, Tang, et al., 2008) MMP9 overexpression may be a novel method to address cell loss. We ai m to perform further s tudies to determine whether MMP9 associated changes in engraftment occur in non transgenic mice. A positive result would mean that the effect of MMP9 on NSC survival is more associated with tissue remodeling than mitigating pathology. Concluding Comments In summary, t he data presented here provide s evidence for consistent NSC engraftment in three regions within the hippocampus T hese regions are characterized by a change in neural density and therefore provide paths of least resistance for the flow of cells at the moment of injection. The tightly, bound and layered nature of the hippocampus is no t unique within the brain; the olfactory bulb and cerebellum also share similar structural organization. T herefore further studies aimed at characterizing putative may yield more specific targeting of NSCs for cell restoration or drug delivery throu ghout the brain We find that MMP9 is associated with significant increases in the size of transplanted NSCs This data suggests that MMP9 may be used to enhance the robustness of graft s for cell restorat ion and drug delivery studies.

PAGE 74

74 Figure 3 8. Iba1 express Representative images of microglia on both hemispheres indicate increased Iba1 reactivity in the region of GFP NSC engraftment (n=3) (D, arrow). Cells have an elongated phenotype that is similar to GFP immu nostained cells (Fig 3 3F). This morphology is distinct from that of ramified, quiescent microglia les s than 20um from this engraftment (D, arrowheads). Control images of adjacent sections stained with only 2 antibody indicated that native GFP was not a source of significant green fluorescence background (data not shown).

PAGE 75

75 Figure 3 9. GFAP expressi on by engrafted NSCs GFAP immunoreactivity is colocalized with NSC engraftment (n=3) Cells have an elongated morphology similar to that seen with Iba1 immunoreactivity (Fig 3 8 D). NSCs shown here were sacrificed two month s following NSC transplantation N was genetically halted in these mice during the survival period Control images of adjacent sections sta ined with only 2 antibody indicated that native GFP was not a source of significant green fluorescence background (data not shown).

PAGE 76

76 Figure 3 10 Paths of least resistance. Cells injected into d were consistently found along a horizontal stretch defi ned by arrows: a, b, c (A) Cells here appear stationary. This result is different from injections into a because deposition of cells here can result in engraftments in (B). Coordinates that target a may yield bilateral engraftment through t he 3 rd ventricle. Images modified from Paxinos Mouse Brain Atlas.

PAGE 77

77 CHAPTER 4 THE EFFECT OF NEURON PATHOLOGY AND THEIR UTIL ITY AS A THERAPEUTIC DELIVERY VE RADING PROTEASE, MMP9 Introduction The use of stem cells that are engineered to produce molecules of therapeutic value holds much promise for the treatment of AD. The goal of this study was to explore their use for long term delivery of candidate therapeutic molecules into mouse models of AD. Our aim was esse ntially to establish a foundation for ex vivo modified stem cells in the emerging clinical field of cell replacement therapy for AD. We used SEZ derived NSCs because they are the only somatic cell that we know of that is endogenous to the brain, can be c ultured for extended periods of time and may migrate millimeter distances when transplanted (Scheffler, et al., 2005, Walton, et al., 2006a, Zheng, et al., 2006) Other CNS cells such as neurons and glia are not suitable for ex vivo manipulation because they can only be cultured for a few weeks. Notably, direct injection of isolated microglia does not result in long term grafts (G. Marshall, personal communication). To show the feasibility of genetic manipulation of NSCs to disrupting factors, we transduced them with lentiviruses carrying transgenes for secreted Metalloprotease 9, membrane bound Heparanase and membrane bound Neprilysin. Because cultured NSCs are mitotic cells, lentiviral transduction was parti cularly suitable for this study. This is because these viruses integrate their genetic cell division. The high efficiency of lentiviral transduction results in a transgen e

PAGE 78

78 expressing cells (Bl its, et al., 2005) These cells were further purified using f l uorescent activated cell s orting (FACS) We focused most of our studies on MMP9 over other competing anti candidate molecules because 1) it is naturally secreted, 2) i t is overexpressed b y astrocytes reactive to plaques, and 3) it has previously been demonstrated to degrade plaques in situ (Yan, et al., 2006, Yin, et al., 2006) Methods Lentivirus Construction 3 rd generation self inactivating lentiviruses (Dull, et al., 1998, Englund, et al., 2000) cloning and expression system (SBI, Mountain View, CA). Briefly, MMP9 (2.54kB insert) and H XL4 plasmid vectors (4.7kB empty vector) were purchased from Origene (Rockville, MD). We already had Neprilysin cDNA within the pCDNA 3.1 vector. All three constructs were amplified to provide enough material for rd generation GFP containing lentiviral vectors. The pCMV6 vector has Not1 restriction using Not1 restriction enzyme dige st (New England Biolabs, Ipswich, MA). Nep cDNA was isolated from the pCDNA 3.1 vector using Nhe1 and Not1 restriction enzyme digests. These inserts were ligated into the pCDH lentiviral vector (SBI, Mountain View, CA). Viruses were packaged by transien tly co transfecting HEK293 cells with the pCDH construct (containing MMP9, HPSE, Nep inserts) along with plasmids for the creation of lentiviral structural and integration proteins, and VSV G pseudotype (courtesy of S.S. Rowland) The VSV G envelope prote in enables lentiviruses to transduce a broad

PAGE 79

79 range of mammalian cells. Viruses packaged by the 293 cells were concentrated to ~ 1 x 10 11 particles/ul by centrifugation and minimal dilution. Lentivirus Transduction 1 x 10 5 trypsinized NSCs and 293FT cells from confluent cultures were resuspended in 100ul dPBS These cells were transduced by exposure to 6ul volume of virus concentrate for 1 hr in 37C with agitation every 10 15 min This ratio of virus to cells was found to most optimally and consistentl y yield NSCs expressing the GFP reporter. Fluorescence Activated Cell Sorting Transduced cells were grown to confluent cultures in T25 flasks. Cells were trypsinized and resuspended in 5mls of PBS with 2% fetal bovine serum. GFP intensity cut off points of A.U. 10 3 and 10 4 were used to obtain cells varying in transgene expression. Isolation of NSCs The protocols for isolating NSCs are contained in the literature (Marshall, et al., 2006, Zheng, et al., 2006) NSCs were culture d as described in Chapter 3. Transplantation into Amyloid Beta AD Mice have been previ ously S urgeries were performed as described in Chapter 3 Briefly, a Hamilton 33 gauge needle (Hamilton Company, Reno, NV) was then loaded with NSCs prepared at ~5 x 10 4 cells/ul in 1x dPBS. The cells were derived from a trypsinized and pelleted monolayer of NSCs washed twice with 200ul 1x dPBS and diluted to the appropriate volume using a reference cell count done on a hemacytometer. Cells were deposited at 2.0mm,

PAGE 80

80 2.3mm and 2.5mm into the hippocampus. 1.25 x 10 5 cells were deposited at 2.0mm and 2.3mm, while 2.5 x 10 5 cells were deposited at 2.5mm. Typically, 4 8ul total volume was deposited at the rate of 0.25ul per 15 seconds (dependant on cell concentration). 20 um coronal sections were processed from whole brains and stored in anti freeze media at 20C until further processing. Immunochemistry 4% paraformaldehyde fixed cells or tissue sections processed for immunofluorescence in solutions containing 0.1% Triton X, 10% goa t serum in 1x PBS. Primary antibodies used in this study include copGFP (1:2000, Evrogen, Moscow, Russia), anti human MMP9 Clone 56 2A4 (Abcam, Cambridge, MA), anti human MMP9 Clone 6 6B (EMD, Gibbstown, NJ), anti human MMP9 Clone 7 11c (Santa Cruz Biotec h, Santa Cruz, CA), anti mouse MMP9 (courtesy of R. Senior, Washington University, St. Louis, MO), anti antigen Iba1 (1:1000, Wako, Richmond, VA), astrocyte antigen GFAP (1:1000, Dak o Corporation, Carpinteri a, CA). Cells were rinsed and incubated with goat secondary antibodies Alexa 488, 568 (Invitrogen, Carlsbard, CA). Cells were photographed with an Olympus DP71 camera mounted on an Olympus BX60 microscope. For western blot analysis, samples were diluted in Laemmli sample buffer containing 2% sodium dodecyl sulfate and loaded in 4 20% TG SDS gels (Invitrogen, Carlsbad, CA) for standard SDS PAGE. Immunoblots were probed with anti mouse/human MMP9 Clone 38898 (1:5000 Abcam) and Abcam anti human MMP9 Clone 56 2A4 (1:500 (Abcam). Gel blots were photographed using a Fugi imaging system (Fugifilm Life Science, Stamford, CT).

PAGE 81

81 Image Pro Plus software (MediaCybernetics, Bethesda, MD) was used to quantify intensity over background for images of plaques in coronal sections (Dolev and Michaelson, 2004, Podoly, et al., 2008) Because transplanted cells largely settled in the subcortical and corpus callosum whi te matter tracks, we focused on a region of interest (ROI A) that included the dorsal hippocampus below the site of engraftment and the cortex above the site of engraftment (1.25mm x 2mm To be consistent across mice, the vertical boundary for ROI A was t he meeting of the dentate arms. The lateral blade of the dentate gyrus was the horizontal boundary. Only sections containing needle tracks were analyzed. Comparison was done on the equivalent region on the contralateral side. For statistical comparison contralateral areas were analyzed using paired, two test using Microsoft Excel. An unpaired, two tailed test was used for comparison between transplants of GFP NSC and MMP9 NSC transplants A p value of <0.05 was consider ed statistically significant. Reverse Transcriptase Polymerase Chain Reaction (RT PCR) of Human MMP9 Human immortalized 293FT cells, mouse immortalized 3T3 cells and NSCs were transduced in simultaneous 100ul reactions that contained 1 x 10 5 cells and 6 ul of MMP9 virus concentrate as described above. Cells were allowed to grow for 6 days before being lysed with Trizol (Invitrogen, Carlsbard, CA). Because there were cell type specific differences in growth rate, lysates were normalized for total protein using a bicinchoninic acid assay (BCA) (Peierce, Rockford, IL). RT PCR reactions utilized actin (400bp) and human MMP9 (600bp). Reagents from a Superscript One Step RT PCR System were used according to manufacturer

PAGE 82

82 instructions (In vitrogen, Carlsbard, CA). Because of limited supplies of virus, we could not repeat transduction of cells. However, RT PCR reactions were repeated 3x. In Vitro MMP Gelatinase Activity Conditioned media (C.M.) was collected from transduced and mock transdu ced NSCs 6 days after transduction (~75% confluent). To eliminate contamination by floating cells in our experiments, C.M. was centrifuged at 1000 x g for 10 min before being applied to 50ug/ml DQ gelatin (Invitrogen, Carlsbard, CA) at a 1:4 dilution. Re actions were incubated overnight at room temperature. Fluorescence increases upon degradation of DQ gelatin due to release of quenched fluorescein. To show gelatin degradation is due to MMP activity, the general MMP inhibitor 1, 10, phenanthroline (1, 10 PNTL) was applied at a 0.8mM concentration. Collagenase activity was used as a positive control. Triplicate reactions of each sample were measured with the excitation/emission spectra of 495nm/515nm using a spectrophotometer (Bio Tek, Winooski, VT). In Situ MMP Gelatinase Activity Fixed brain sections were treated with a pro tocol modified from Yan et al. (Yan, et al., 2006) Briefly, a PBS solution with 0.1% Triton X100 and 1% agarose (w/v) was made homogeneous by 2 to 4 rounds of short 10 sec microwave pulses and brief vortexing. The following additions were made to this solution in order to visualize or inhibit MMP activity: DQ gelatin (100 ug /ml, warmed to 37C), DAPI nuclear counterstain (1:1000), 1,10 PNTL (0.8mM), and EDTA (20mM). 300ul of DQ gelatin mixture was quickly added per section. After an overnight incubation in room te mperature, green fluorescence of sections was captured with an Olympus DP71 camera mounted on an Olympus BX60 microscope.

PAGE 83

83 Chemical Activation of secreted MMP9 Organomercurial compounds such as p aminophenylmercuric acetate (AMPA) induce autoacti vation of M MPs including MMP9 (Ramos DeSimone, et al., 1999, Yan, et al., 2006) A 1mM concentration of AMPA was added to conditioned media from tra nsduced and mock transduced NSCs Commercially available MMP9 proenzyme (Perkin Elmer, Waltham, MA) was similarly treated as a positive control. The reaction volume was incubated overnight at 37C before Western blot analysis with antibodies specific to human MMP9. Results GFP NSCs were transplanted into the hippocampus of four mice symptomatic for and in some mice, secondary hippocampal fissure engraftment. This pattern of engraftment allowed us to ask whether NSCs have effects in the hippocampus and the cortex. After a month survival period, mice were harvested and their brains processed largest engraftments. We analyzed five to seven such sections per animal for amyloid with Image Pro Plus software pr eviously used in a similar capacity (Dolev and Michaelson, 2004, Podoly, et al., 2008) A region of interest (ROI A) proximal to white matter engrafted cells was studied ( see Chapter 3) This area includ ed the cortex above the white matter tracks and the dorsal hippoc ampus below ( Fig. 4 1 A). Compared to the equivalent contralateral region, we observed a 26.4% (p=0.04, paired t test) reduction in number in this ROI. Though there was variability in the number of s found per section ( Fig. 4 1 C ), all mice had reduced numbers ( Fig. 4 1 D ).

PAGE 84

84 Figure 4 1. Transplantation of NSCs is associated with reduced amyloid burden. E the region of engraftment of 10 3 GFP NSCs a month post transplantation (A). Representative image demonstrating the sensitivity of software used to count t of results where each dot indicates the test) *, p<0. 05 To understand this result, w e have performed preliminary experiments to determine whether 1) NSCs were directly clearing 2) NSCs were producing diffusible factors that clear ed or 3) NSCs were interacting with host cells to clear Pulse chase experiments similar to those described in Chapter 2 determined that NSCs were capable of internalizing 0.52ng/ml of A 42 in a period of 3 hours (n=4), (Fig 4 2A ). In comparison, neonatal mouse microglia internalize d about twice as much 42

PAGE 85

85 (1.1ng/ml), bu t then expelled most of what was internalized (see Chapter 2) (Njie, et al., 2010) Interestingly, NSCs appear to process differently as internaliza tion of was not followed by expulsion (Fig 4 2B ). This suggests that NSCs biophy si cally degrade in vitro and possibly in vivo Figure 4 2 GFP pulse chase experiments (n=4). GFP NSCs internalized 0.52ng/ml within 3hrs (A). In contrast to microglia (Fig. 2 9), GFP NSCs did not expel GFP NSCs to des a possible mechanism for reductions in in vivo plaques following transplantation (see Fig. 4 1). Mock data (gray) represents experiments without the presence of cells to control for non specific requires the presence of quantified in the above experiments. *, p<0.05; **, p<0.01.

PAGE 86

86 We found that NSC monolayers are reactive to an antibody against mouse MMP9 (n=3 ) ( Fig. 4 3 A) I n vivo 3B). It may be that MMP9 and other NSC derived diffusible factors contribute to the clearance of The lack of MMP9 specific ch emical inhibitors makes this difficult to confirm in vivo However, infusion of antibodies known to stop MMP9 g elatinase activity is a possible alternate approach that we may explore in the future Previously, degradation of DQ gelatin has been used to sh ow MMP activity around plaques (Yan, et al., 2006) We verified this finding in mice that have continuous ov erexpression of ( Fig. 4 4 ). This result suggests endogenous pathways to regulate plaques are active. However, recent work from our lab has shown that once formed, s are not removed by endogenous processes (Jankowsky, et al., 2005) which include MMPs We therefore wondered whether this is due to an overwhelming rate of amyloid formation. To shed light on this question, we stopped new A production in mice with amyloid precursor protein under a tetracycline regulatory element. Following a month, we found similar levels of MMP activity around plaques (n=3 ), ( Fig. 4 4 B, C). This result suggests that endogenous processes continue to re gulate actively burden in the absence of new deposition. Since this activity does not result in eventual clearance, it is perhaps the level of MMP anti activity rather than the rate of new amyloid deposition that may be responsible f or the permanence of plaques.

PAGE 87

87 We decided to genetically modify NSCs to increase the amount of MMP9 in mice NSCs with VSV G serotype lentiviruses carrying the human MMP9 transgene. These lentiviruses, wh ich carry GFP under a separate promot er ( Fig. 4 5 A) and are able to transduce mammalian cells, conferred GFP fluorescence to NSCs ~3 days post transduction and 293FTs within 2 days post transduction. No remnant virus genome was detected in NSC cultures fo llowing 2 washes (Fig. 4 5C) Enrichment of NSCs (described below), yielded cultures reactive to two antibodies specific to human MMP9 (Fig 4 5 D). 293FT cells produced noticeably more GFP compared to NSCs ( Fig. 4 6 A, B). This result occurred in 3 indepe ndent transductions with independent lots of virus. Several possibilities were considered to explain this observation. These included a species effect, an immortal cell line effect, NSC quiescence, transcription versus translation and lack of uniformity in transduction conditions. To shed light on some of these possibilities, human [immortal] 293FT cells, mouse NSCs as well as mouse [immortal] 3T3 cells were transduced in parallel and analyzed for mRNA transcript production. We were unable to include h uman NSCs in this experiment due to the limited availability of these cells. This experiment was performed once due to a limited supply of virus, however RT PCR reactions were repeated three times to reveal variability in analysis. Our results indicate that mouse 3T3 immortal cells had similar levels of transgene mRNA as NSCs ( Fig. 4 6 C D ). This result suggests that our observations in (A) are not unique to NSCs Perhaps may accoun t for enhanced transgene overexpression in 293FT cells.

PAGE 88

88 Though transduced NSCs do not match 293FT cells in transgene production, they nonetheless express human MMP9 which complements the endogenous mouse MMP9 they produce. These cells therefore overexpress total MMP9. Similar to previous reports (Ramos DeSimone, et al., 1999) transgene directed MMP9 mRNA is translated to a secreted 92kDa zymogen (proenzyme). Mock treated NSCs have intrinsic MMP activity on gelatin that is chemically inhibited by 1,10 phenant hroline ( Fi g. 4 7 A). The secretion of MMP9 zymogen by transduced NSCs is associated with a 3.5x increase i n gelatinase activity ( Fig. 4 7 A). We did not observe degradation of 42 in conditioned media from 10 3 MMP9 NSCs (10 3 cells, data not shown). However, the 39kDa catalytic subunit of MMP9 (ctMMP9) reduced detectable by Western blot by approximately 75% i n overnight reactions ( Fig. 4 7 B). The amount of ctMMP9 needed to achieve this degradation was less than ideal: ~20:1 ratio of enzyme to substrate. However the observation that truncated MMP9 is the isoform capable of degrading is consistent with previous reports (Yan, et al., 2006) This suggests in vivo autoactivation of NSC produced MMP9 is required for anti activity. To determine if NSC produced MMP9 is capable of undergoing such autoactivation, conditioned media was treated to p aminophenylmercuric acetate (AMPA). This organomercurial compound disrupts the interaction of an unpaired, N terminal cystein e with the zinc i on of the MMP9 catalytic center. Such disruption initiates autoproteolytic N terminal shedding that generates the MMP9 isoform shown to degrade in the literature (Yan, et al., 2006) We applied a 1mM concentration of AMPA to conditioned media and detected only the truncated (active) form of MMP9 following 24hrs (100% co nversion) ( Fig. 4 7 C). We therefore concluded that transgene

PAGE 89

89 directed MMP9 secreted by NSCs is capable of undergoing autoactivation. There are a number of proteases thought to mimic the action of AMPA in vivo (Ramos DeSimone, et al., 1999) On the other hand, tissue inhibito rs of metalloproteases (TIMP 1 4 ) inhibit MMP9 but are limited by stoichiometric binding (Ramos DeSimone, et al., 1999) TIMP levels hav e been shown to increase in AD (Lorenzl, et al., 2003, Peress, et al., 1995) and in one mouse model of AD (Hoe, et al., 2007) However, the concentration of TIMPs in mice with introduce as much MMP9 into the brain as possible in order to saturate endogenous activators of MMP9 and overcome inhibitors of MMP9. Figure 4 3. NSCs express endogenous mouse MMP9. NSCs wer e reactive to antibodies against mouse MMP9 (n=3) (A). It is possible that transplant derived mouse MMP9 adds to the concentration of MMP9 secreted by Cultures of transduced NSCs had a non un iform distribution of GFP. This suggested that in our population of transduced cells, transgene copies were varied from cell to cell. Peripheral drug treatment paradigms typically tailor dosage to optimally balance between desired effects and side effect s. To emulate this preclinically, we

PAGE 90

90 used FACS technology to take advantage of the differential transduction of NSCs by selecting for cells whose GFP intensity was A.U >10 3 or 10 4 ( Fig. 4 8 ). The conditioned media from the resulting cultures was analyzed with Western blot using an antibody that detects commercially available human proMMP9 and verified with an antibody cross reactive to both human and mouse MMP9. In vitro a confluent layer of 10 3 cells secreted 0.51ug/ml of human MMP9 in a T25 flask with 5mls of media over 3 weeks. 10 4 cells treated similarity, secreted 0.66u g/ml over only 1 week ( Fig. 4 9 A). Non transduced and non FACS sorted cells had no detectable secretion of human MMP9. We project that over 4 weeks, 10 3 and 10 4 cells will secrete 0.7 ug/ml and 2.7ug/ml of MMP9, respectively ( Fig. 4 9 B) This demonstrates that we can create NSC cultures with low and high doses of MMP9 with FACS technology. This result, though promising, is limited by yield. A confluent T25 flask holds a monolayer of ~5 x 10 6 NSCs The average yield of 10 3 NSCs is 8.9 x 10 5 cells/flask. This is reduced 6 fold, or 1.4 x 10 5 cells/flask for 10 4 cells (n=7) ( Fig. 4 8 E). 10 4 cells are only 4.5% +/ 0.8% (n=8) of the original populatio n of transduced cells ( Fig. 4 8 D). F urther selection exponentially diminishes yield. Therefore, we are confident that our enrichment protocol approached theoretical limits of selection. We reasoned that these highly enriched MMP9 expressing NSCs stand the best chance of overcoming endogenou s MMP9 repressors and acting against However, MMP9 expression can lead to negative consequences (Chambers and Matrisian, 1997, Lubbe, et al., 2006, Moll, et al., 1990) Ther efore, we first worked with 10 3 cells in transplant studies in hopes of establishing safety as well as efficacy.

PAGE 91

91 MMP9 expressing 10 3 NSCs were injected into five mice We confirmed that NSCs continued MMP9 overexpression a month post surger y (Fig 4 10 ). Graft derived MMP9 had N terminal shedding that is associated with activation (Fig 4 10E ). We have ongoing studies to determine the endogenous proteins that activated graft derived MMP9. Analyses of in vivo MMP9 overexpression utilized a ntibodies specific to human MMP9 for immunocytochemistry (n=3), 1 u m z plane confocal histology (n=2), and western blot (n=3). Figure 4 4. Endogenous MMP activity in mice with plaque burden. Sections incubated overnight with DQ gelatin demonstrate persistent MMP activity on B). Controls non transgenic mice (D), and is dependent on MMP activity (E, F). Mice receiving MM P9 NSC transplants were examined for burden with methodology described above for GFP NSCs A sham injection on the contralateral

PAGE 92

92 side was performed in order to co ntrol for the effect of needle injury on the cell injected side. Compared to the needle injury on the contralateral side, we did not notice additional death of neuronal cells or otherwise abnormal cytoarchitecture in the side of the brain containing MMP9 NSCs We observed a 28.6% (p=0.03, paired t test) reduction in number within ROI A ( Fig. 4 11A B ). MMP9 NSC and GFP NSC transplants had comparatively similar percent reductions in nu mber ( Fig. 4 1 1C ). Since number reductions were statistically significant in both groups, we asked whe ther there was change in the amount of s cleared in mice receiving GFP NSCs and MMP9 NSCs We subtracted the averaged contralateral number in each animal from the averaged ipsilateral number and then compared across the two tr ansplant groups. MMP9 NSC transplants were associated with clearance of 31% more plaque s compared to GFP NSC transplants (p=0.47, unpaired t test), (Fig 4 1 1E ). This trend, though non statistically significant, is promising. It should be noted that m ice hosting MMP9 NSC transplants were a month older and consequently had 25% greater numbers ( Fig. 4 1 1D ) The lack of a uniform baseline of number due to age related changes is confoundin g. The increased engraftment of MMP9 NSCs relat ive to GFP NSCs further complicates interpretation. Though our data indicates a trend of more clearance in 10 3 MMP9 NSCs further experimentation with 10 4 cells that more robustly express MMP9 is needed with special attention to address the conf ounding variables described above. Discussion In this study, we aimed to determine whether NSCs pathology either without modification or as a vehicle to deliver molecules that have

PAGE 93

93 Figure 4 5. Genetic modification of NSCs f or MMP9 overexpression. Vector diagram of MMP9 transgene illustrates subcloning of human MMP9 cDNA from a pCMV vector (4.7kB) to a pCDH vector (7.5kB). The pCDH vector contains plankton derived GFP reporter gene and components necessary for the packaging of self inactivating lentiviruses. The dual promoter design of pCDH meant MMP9 was independently driven by the CMV promoter while copGFP was driven by the EF1 promoter (A). Restriction enzyme digests yielded MMP9 transgene of correct size (2.5kB) at the e nd of the cloning process (B, Lanes 3 & 4). Real time PCR showed no evidence of viral genome in media in transduced cultures washed with dPBS (C). Transduction resulted in GFP reporter gene expression (D, E middle panels). GFP NSCs (D) and MMP9 NSCs (E) differed only in MMP9 expression. GFP and MMP9 immunoreactivity did not always colocalize (D, compare left and middle panels) promising anti qualities. Unlike our results with microglia (see Chapter 2), NSCs in vitro Transplantation of NSCs is associated with close to a one third reduction in numbers in the cortex and hippocampus. Genetically

PAGE 94

94 modifying NSCs to overexpress MMP9 is associated with a trend of more clearance of plaques; however a statistically significant difference was not found. It is possible that greater overexpression of MMP9 may yield more robust reductions in plaque numbers. W e developed a method to enrich for MMP9 production in NSCs cultures by utilizing FACS selection to take advantage of the plurality of gene dosage associated with lentiviral transduction. This positive selection scheme approaches the theoretical limit of e nrichment and has resulted in improvement from initially undetectable levels of MMP9 in NSC conditioned media to detection of u g /ml amounts of MMP9. We were also able to demonstrate enrichment of heparanase overexpressing NSCs Therefore, t his method of enrichment is generally applicable to emerging therapies using NSCs Our future studies will concentrate on NSCs four fold more enriched that those utilized for the transplant studies described in this report. Endogenous MMP Activity Our current findings show that the brain continues to use MMPs to regulate pre existing s suggesting that endogenous processes do not cease actively mitigating burden. These processes appear insufficient as s persist six (Jankowsky, et al., 2005) However, once formed, s resist biophysical remov al despite persistence of these endogenous efforts. This underscores the need for exogenous Burden is Lowered Following NSC Transplantation We introduced GFP NSCs and MMP9 NSCs i n to the brain and found both cell types reduced numbers. This result suggests that transplanted NSCs or uncharacterized biology associated with their e ngraftment reduces the prevalence of

PAGE 95

95 otherwise permanent burden. Our findings contr adict recent work by Blurton Jones and colleagues (Blurton Jones, et al., 2009) was not reduced in hippocampal injections of NSCs The discrepancy with our findings may perhaps be explained by unknown factors unique to their model -a triple transgenic mouse with tauopathy, versus our models which strictly model amyloidosis (Oddo, et al., 2003) Secondly, the cells used by Blurton Jones et al., were created without region specificity; they were derived from neurospheres obtained from whole brain homogenates. It is possible that the SEZ origin of our NSC s enabled unique anti A NSCs exhibit anti A NSCs as a Platform to Deliver MMP9 and Other Candidate Therapeutics In Vivo We hypothesized that constitutive MMP9 secret ion by NSCs proteolytically will act burden. Since ex vivo manipulation of NSCs for use as delivery vehicles of therapeutics is a field still in its infancy, our work sheds light on several i mporta nt questions regarding NSCs in this paradigm These questions include: do NSCs retain stem cell qualities if genetically manipulated in vitro? Do NSCs express active, transgene directed proteins of interest? Are such proteins produced in sufficient amou nts? Can NSCs be genetically modified with a repertoire of candidate m olecules? New m olecules with interesting anti effects continue to be published. For instance M T1 MMP has recently been shown to in vitro (Liao and Van Nostrand, 2010) Our observations on genetic ally manipulating and purifying NSCs and the in vivo effec ts of MMP9 NSCs demonstrate potential positive effects as well as pitfalls that may apply to emerging candidate molecules

PAGE 96

96 Figure 4 6. Cell type differences in transduction efficiency. 293FT cells produced significantly more GFP protein relative to equi valently treated NSCs (n=2) (A, B). Semi quantitative RT PCR analysis of transduced cells indicated mouse 3T3 immortal cells had comparable amounts of MMP9 mRNA as NSCs, suggesting low levels of transgene expression is not NSC specific (C). This comparati ve transduction experiment was not investigated further due to the limited supply of lentiviral stocks. Error bars reflective RT PCR analyses (D). Characteristics of NSCs overexpressing transgenes Regarding the maintenance tic manipulation, our transgene receiving NSCs continued expression of BIII Tubulin (data not shown), a immunological marker commonly associated with precursor cells (Walton, et al., 2006a) Interestingly, transplanted NSCs in white matter tracks do not stain for BIII Tubulin ( see Chapter 3). Studies have shown that microglia are ass ociated with the maintenance of

PAGE 97

97 the self re newal capacity of NSC cultures (Walton, et al., 2006b) We do not see obvious changes in the microglial subpopulation following MMP9 genetic modification Figure 4 7 Secreted MMP9 has zymogen activity and can undergo autoactivation. Conditioned media (C.M.) from transduce d NSCs has 3 fold more gelatinase activity compared to media from mock treated NSCs (A). MMP9 NSC shown). Commercially available catalytic subunit of MMP9 (39kDa) was able (B). Application of p aminophenylmercuric acetate (AMPA) to conditioned media results in conversion (autoactivation) of transgene derived degradation (C). Mouse MMP9 (105kDa) not detected h ere. Human specific MMP9 antibody used. In vivo, NSCs transition between states of quiescen ce and proliferation (Chambers and Matrisian, 1997, Lubbe, et al., 2006, Moll, et al., 1990) In a quiescent state, it is likely that the metabolic rate of NSCs i s reduced. This may e ffect NSC drug delivery as transgene transcription or translation may be affected negatively In our experience NSC con sistently secreted appropriately folded MMP9 that is capable of N terminal shedding and gelatinase activity. Following transduction, we found an assortment of

PAGE 98

98 GFP expression in a population of cells that did not yield detectable levels of MMP9 (cell assoc iated antibody reactivity) We attribute this variability in reporter gene expression to the randomness of lentiviral infection, region s of transgene integration and unknown promoter effects However, NSC quiescence may contribute to the assorted express ion of GFP. Nonetheless we were able to select for the brightest cells and improve our yields of transgene protein to ug /ml quantities. The success of our selection suggests that in a clinical setting, similar methodology may be used to obtain populatio ns of cells that yield appropriate doses of candidate drugs Whether transgene expression changes once NSCs are transplanted remains unanswered. Further exploration of temporal changes in transgene expression as well as studies to understand endogenous activators and inhibitors are needed MMP9 NSCs form significantly larger graft s than GFP NSCs (see Chapter 3) Yet, both cell types are associated with similar reductions in amyloid burden. It is possible that subpopulations of NSCs that are responsible for anti toxicity the hippocampal stem cell niche (Verret, et al., 2007) If a similar scenario applies to transplanted NSCs it may be that a subpopulation within MMP9 NSC grafts was ablated after a bolus of anti R emaining MMP9 NSCs lack ing anti may then have continue d surviving due to MMP9 overexpression Determining whether this hypothesis is true is complicated by the fact that large numbers of transplanted NSCs do not display common immunological markers (T. Zheng, personal communicat ion ) (Verret, et al., 2007) and what researchers know about transplanted NSCs is la rgely based on snapshots (histology,

PAGE 99

99 etc) of dynamic graft environments T emporal studies of graft survival and better markers of NSCs would facilita te more granular understanding of NSC grafts NSCs showed reporter gene expression when transduced with neprilysin and heparanase. We discontinued our work with neprilysin because NSC cultures repeatedly died days after transduction. However, heparanase NSCs have been passaged multiple times. These cells are reactive to an antibody against heparanase (da ta not shown) suggesting transgene expression MMP9 and heparanase NSCs both display two populations of GFP intensity when analyzed with FACS. Consequently, we have been able to isolate 10 3 and 10 4 heparanase NSCs This data suggests tha t neprilysin ex pression may be uniquely toxic to NSC cultures. However our experience with MMP9 and heparanase expression indicates that NSCs are capable of producing a diverse pallet of anti A les. Concluding Comments In this preclinical study, we took advanta ge of interesting stem cell properties in a disease setting. Our experiments demonstrate that NSCs can be genetically manipulated ex vivo and yield robust grafts when transplanted into the hippocampus of diseased mice. Such graft s are associated with red uction s of plaques in a mouse model of AD This result provides evidence of a novel attribute of transplanted NSCs We attempted to further enhance NSC anti activity by inducing MMP9 overexpression However, only a trend towards further reductions in plaques was observed. It is possible that enriched NSCs will produce enough MMP9 to overcome endogenous tissue inhibitors of MMP9 We describe an enrichment method that has yielded NSCs cultures able to produce ug/ml concentrations of MMP9. This method has been generalized to NSCs expressing other candidate molecules. Our future

PAGE 100

100 studies aim to find the right dose of MMP9 or alternatively, the right drug to maximally reduce pathology. To gether, data presented here demonstrates that NSCs can be used as a platform for testing novel therapeutic molecules in vivo Fig ure 4 8 Enrichment of NSC cultures. FACS analysis demonstrated two populations of cells in transduced cultures that vary in reporter gene expression (A C ). Within the populatio n of brightest cells (B), we selected for intensity greater than10 3 or 10 4 (C GFP low, GFP high ). The resultant cells represent 16% and 4.5% of the overall transduced population respectively (n=8) (D) These cells are referred to as 10 3 NSCs and 10 4 NS Cs Yields of10 3 NSCs and 10 4 NSCs attained per T25 flask holding ~5 x 10 6 NSCs (n=7) (E) The methodology used here result ed in highly enriched cultures (compare G, I equally exposed images of similarly confluent cultures ) and can be used for NSCs expre ssing other transgenes : Chromatogram in (C) is from heparanase overexpressing NSCs Characteristic double hump is also exhibited by GFP NSCs and MMP9 NSCs (data not shown).

PAGE 101

101 Figure 4 9 NSC enrichment is associated with rate of MMP9 secretion. MMP9 from 10 3 NSCs is detectable at 3 weeks while MMP9 from similarly confluent 10 4 NSCs is detectable at 1 week (n=3) (A) Projected concentration of MMP9 that a confluent layer of NSCs will secrete in a 5 ml volume during the course of a month (B) is derived from densitometric analysis of data in (A).

PAGE 102

102 Figure 4 1 0 NSC overexpression and activation of MMP9 in vivo. 1um z plane depth confocal images indicate NSCs overexpressed huMMP9 a month following transplantation (A B). MMP9 was observed in cell bodies (arrowheads) as transgenic mice (n=3) and not in the contralateral [non injected] hemisphere (not shown) (A). To gain insight on the distribution of MMP9 secreted by NSCs, the brain s of non transgenic mice with one month MMP9 NSC engraftments were sectioned at 1mm intervals (C). Cortical and hippocampal regions were then dissected (D) and analyzed with western blot using antibodies against human MMP9. Immunoreactivity was strongest in the hippocampus of the hemisphere where NSCs were injected into (n=2). NSCs in vitro secrete proMMP9 (Fig. 4 9A ). However, MMP9 in vivo is of a lower molecular weight indicating activation by endogenous factors. Mouse MMP9 is >100kDa and not detected by antibodies used here.

PAGE 103

103 Fig ure 4 1 1 Transplantation of MMP9 NSCs and GFP NSCs results in similar reductions in amyloid burden Compared to the contralateral side, the number 3 MMP9 NSC engraftmen ts was reduced by 2 8.5 % ( p=0.0 4 paired t test ) ( A ). All mice receiving MMP9 NSCs ( n=5 ) ( B ). GFP NSC and MMP9 NSCs C ) (see Fig. 4 1) Due to age related effects, c ontralateral A burden was 25% greater in mice receiving MMP9 NSCs ( D ). T herefore, percentage reduction leared between the two groups of mice To determine if MMP9 overexpression is associated with greater clearance of A plaque s, we compared the absolute groups T he results indicate MMP9 overexpression by NSCs is associated with a non statistically significant trend of 31 % more s cleared (p=0.47 unpaired t test ) (E ) *, p<0.0 5

PAGE 104

104 CHAPTER 5 CONCLUSIONS other brain diseases that feature protein aggre gation. For instance, Amyloid L ateral S oxide dismutase synuclein, respectively. Recently, elegant studies by Don Cleveland and others show that microglia expressing mutated SOD1 play an important role in the end stage of ALS. This evidence fits with our findings: the perturbatio n of microglial functionality with age may contribute to neurodegeneration in mouse models and perhaps i n diseased humans. However, therapeutic manipulation of en dogenous microglia is difficult. This is evidenced by the lack of robust outcomes in clinica l trials that aimed to reduce glial activity ( ex. non steroid al anti inflammatory drug trials) (Sabbagh, 2009) or enhance g lial activity ( immunization trials) (Patton, et al., 2006) S tem cell transplants present an alternative avenue to directly target amyloid burden. In work presented in here, we were unable to modify NSCs It is like pathology. other relevant observations. First, transplantation of NSCs Second, inj ection of NSCs into the hippocampus results consistently in engraftment in defined regions of the hippocampus and surrounding white matter. Finally, transgene overexpression modified NSC engraftment patterns. These findings have significant implications f or proposals for the use of stem cells to restore dying cells or to deliver drugs to sites of injury.

PAGE 105

105 The different schools of th ought have noteworthy evidence to cite. For instance, a recent drug trial with an anti tau compound ha s yielded perhaps the most promising clinical outcomes in AD drug trials history ( TauRx, Rember trials ). This suggests a causative role for tau in AD et iology. On the other hand, research published t in non demented individuals who have parents with late onset (sporadic) AD (Mosconi, et al.) This the amy loid precu rsor protein and presen ilin loci. The ubiquity of data to support the different schools of thought of how AD arises is remarkable and unique amongst neuro de generative diseases. This ubiquity also highlights the fact that success has not been had with re moval of any of the pathologies found in AD. Until this occurs and academic. They are based on correlation, not causation. Removing the pathologies of AD has proven diff icult partly because of difficulties with delivering drugs into the brain. The use of NSCs as demonstrated here, presents a novel approach to overcome this difficulty. The scope of this approach is not limited to AD, as research on treatments for most d iseases of the brain is hindered by the inability to deliver drugs into the brain. In the course of human history, medical practice on the brain has typically involved the removal of tissue. At this junction in history, NSCs transplants represent a bifu rcation point. What happens after the introduction of new tissue in to the brain is largely unknown. The work presented here adds to our understanding of the behavior

PAGE 106

106 of implanted brain tissue and the manipulation of said tissue to counter disease patholo gy.

PAGE 107

107 LIST OF REFERENCES Armstrong, R.A., Cairns, N.J., Myers, D., Smith, C.U., Lantos, P.L., Rossor, M.N. 1996. A comparison of beta amyloid deposition in the medial temporal lobe in sporadic Alzheimer's disease, Down's syndrome and normal elderly brains. Neurodegeneration 5(1), 35 41. Association, A.s. 2007. Alzheimer's Disease Facts and Figures 2007. Alzheimer's Association. Astary, G.W., Kantorovich, S., Carney, P.R., Mareci, T.H., Sarntinoranont, M. Regional convection enhanced de livery of gadolinium labeled albumin in the rat hippocampus in vivo. J Neurosci Methods 187(1), 129 37. Astary, G.W., Kantorovich, S., Carney, P.R., Mareci, T.H., Sarntinoranont, M. 2010. Regional convection enhanced delivery of gadolinium labeled albumin in the rat hippocampus in vivo. J Neurosci Methods 187(1), 129 37. Auld, D.S., Kornecook, T.J., Bastianetto, S., Quirion, R. 2002. Alzheimer's disease and the basal forebrain cholinergic system: relations to beta amyloid peptides, cognition, and treatment strategies. Prog Neurobiol 68(3), 209 45. Bales, K.R., Du, Y., Holtzman, D., Cordell, B., Paul, S.M. 2000. Neuroinflammation and Alzheimer's disease: critical roles for cytokine/Abeta induced glial activation, NF kappaB, and apolipoprotein E. Neurobiol Agi ng 21(3), 427 32; discussion 51 3. Beckmann, N., Schuler, A., Mueggler, T., Meyer, E.P., Wiederhold, K.H., Staufenbiel, M., Krucker, T. 2003. Age dependent cerebrovascular abnormalities and blood flow disturbances in APP23 mice modeling Alzheimer's disease J Neurosci 23(24), 8453 9. Bennett, L.B., Cai, J., Enikolopov, G., Iacovitti, L. 2010. Heterotopically transplanted CVO neural stem cells generate neurons and migrate with SVZ cells in the adult mouse brain. Neurosci Lett 475(1), 1 6. Bjorklund, A., Dunn ett, S.B., Brundin, P., Stoessl, A.J., Freed, C.R., Breeze, R.E., Levivier, M., Peschanski, M., Studer, L., Barker, R. 2003. Neural transplantation for the treatment of Parkinson's disease. Lancet Neurol 2(7), 437 45. Blits, B., Kitay, B.M., Farahvar, A., Caperton, C.V., Dietrich, W.D., Bunge, M.B. 2005. Lentiviral vector mediated transduction of neural progenitor cells before implantation into injured spinal cord and brain to detect their migration, deliver neurotrophic factors and repair tissue. Restor Ne urol Neurosci 23(5 6), 313 24.

PAGE 108

108 Blurton Jones, M., Kitazawa, M., Martinez Coria, H., Castello, N.A., Muller, F.J., Loring, J.F., Yamasaki, T.R., Poon, W.W., Green, K.N., LaFerla, F.M. 2009. Neural stem cells improve cognition via BDNF in a transgenic model of Alzheimer disease. Proc Natl Acad Sci U S A 106(32), 13594 9. Boillee, S., Yamanaka, K., Lobsiger, C.S., Copeland, N.G., Jenkins, N.A., Kassiotis, G., Kollias, G., Cleveland, D.W. 2006. Onset and progression in inherited ALS determined by motor neurons and microglia. Science 312(5778), 1389 92. Bolmont, T., Haiss, F., Eicke, D., Radde, R., Mathis, C.A., Klunk, W.E., Kohsaka, S., Jucker, M., Calhoun, M.E. 2008. Dynamics of the microglial/amyloid interaction indicate a role in plaque maintenance. J Neuros ci 28(16), 4283 92. Borchelt, D.R., Lee, M.K., Gonzales, V., Slunt, H.H., Ratovitski, T., Jenkins, N.A., Copeland, N.G., Price, D.L., Sisodia, S.S. 2002. Accumulation of proteolytic fragments of mutant presenilin 1 and accelerated amyloid deposition are co regulated in transgenic mice. Neurobiol Aging 23(2), 171 7. Borchelt, D.R., Taraboulos, A., Prusiner, S.B. 1992. Evidence for synthesis of scrapie prion proteins in the endocytic pathway. J Biol Chem 267(23), 16188 99. Bouras, C., Hof, P.R., Giannakopoulo s, P., Michel, J.P., Morrison, J.H. 1994. Regional distribution of neurofibrillary tangles and senile plaques in the cerebral cortex of elderly patients: a quantitative evaluation of a one year autopsy population from a geriatric hospital. Cereb Cortex 4(2 ), 138 50. Braak, H., Braak, E. 1995. Staging of Alzheimer's disease related neurofibrillary changes. Neurobiol Aging 16(3), 271 8; discussion 8 84. Braak, H., Braak, E., Bohl, J., Reintjes, R. 1996. Age, neurofibrillary changes, A beta amyloid and the ons et of Alzheimer's disease. Neurosci Lett 210(2), 87 90. Brookmeyer, R., Corrada, M.M., Curriero, F.C., Kawas, C. 2002. Survival following a diagnosis of Alzheimer disease. Arch Neurol 59(11), 1764 7. Brunello, A.G., Weissenberger, J., Kappeler, A., Vallan, C., Peters, M., Rose John, S., Weis, J. 2000. Astrocytic alterations in interleukin 6/Soluble interleukin 6 receptor alpha double transgenic mice. Am J Pathol 157(5), 1485 93. Bruunsgaard, H., Andersen Ranberg, K., Jeune, B., Pedersen, A.N., Skinhoj, P., Pedersen, B.K. 1999. A high plasma concentration of TNF alpha is associated with dementia in centenarians. J Gerontol A Biol Sci Med Sci 54(7), M357 64. Calafiore, M., Battaglia, G., Zappala, A., Trovato Salinaro, E., Caraci, F., Caruso, M., Vancheri, C., Sortino, M.A., Nicoletti, F., Copani, A. 2006. Progenitor cells from the adult mouse brain acquire a neuronal phenotype in response to beta amyloid. Neurobiol Aging 27(4), 606 13.

PAGE 109

109 Carson, M.J., Reilly, C.R., Sutcliffe, J.G., Lo, D. 1998. Mature microglia r esemble immature antigen presenting cells. Glia 22(1), 72 85. Carter, D.B., Dunn, E., McKinley, D.D., Stratman, N.C., Boyle, T.P., Kuiper, S.L., Oostveen, J.A., Weaver, R.J., Boller, J.A., Gurney, M.E. 2001. Human apolipoprotein E4 accelerates beta amyloid deposition in APPsw transgenic mouse brain. Ann Neurol 50(4), 468 75. Chambers, A.F., Matrisian, L.M. 1997. Changing views of the role of matrix metalloproteinases in metastasis. J Natl Cancer Inst 89(17), 1260 70. Chung, H., Brazil, M.I., Irizarry, M.C., Hyman, B.T., Maxfield, F.R. 2001. Uptake of fibrillar beta amyloid by microglia isolated from MSR A (type I and type II) knockout mice. Neuroreport 12(6), 1151 4. Cole, G.M., Beech, W., Frautschy, S.A., Sigel, J., Glasgow, C., Ard, M.D. 1999. Lipoprotein effects on Abeta accumulation and degradation by microglia in vitro. J Neurosci Res 57(4), 504 20. Collins, J.S., Perry, R.T., Watson, B., Jr., Harrell, L.E., Acton, R.T., Blacker, D., Albert, M.S., Tanzi, R.E., Bassett, S.S., McInnis, M.G., Campbell, R.D. Go, R.C. 2000. Association of a haplotype for tumor necrosis factor in siblings with late onset Alzheimer disease: the NIMH Alzheimer Disease Genetics Initiative. Am J Med Genet 96(6), 823 30. Culpan, D., MacGowan, S.H., Ford, J.M., Nicoll, J.A., Griffin W.S., Dewar, D., Cairns, N.J., Hughes, A., Kehoe, P.G., Wilcock, G.K. 2003. Tumour necrosis factor alpha gene polymorphisms and Alzheimer's disease. Neurosci Lett 350(1), 61 5. de Haas, A.H., Boddeke, H.W., Brouwer, N., Biber, K. 2007. Optimized isolatio n enables ex vivo analysis of microglia from various central nervous system regions. Glia 55(13), 1374 84. Dolev, I., Michaelson, D.M. 2004. A nontransgenic mouse model shows inducible amyloid beta (Abeta) peptide deposition and elucidates the role of apol ipoprotein E in the amyloid cascade. Proc Natl Acad Sci U S A 101(38), 13909 14. Dong, H., Goico, B., Martin, M., Csernansky, C.A., Bertchume, A., Csernansky, J.G. 2004. Modulation of hippocampal cell proliferation, memory, and amyloid plaque deposition in APPsw (Tg2576) mutant mice by isolation stress. Neuroscience 127(3), 601 9. Donovan, M.H., Yazdani, U., Norris, R.D., Games, D., German, D.C., Eisch, A.J. 2006. Decreased adult hippocampal neurogenesis in the PDAPP mouse model of Alzheimer's disease. J Co mp Neurol 495(1), 70 83.

PAGE 110

110 Dringen, R. 2005. Oxidative and antioxidative potential of brain microglial cells. Antioxid Redox Signal 7(9 10), 1223 33. Dull, T., Zufferey, R., Kelly, M., Mandel, R.J., Nguyen, M., Trono, D., Naldini, L. 1998. A third generation lentivirus vector with a conditional packaging system. J Virol 72(11), 8463 71. Edbauer, D., Willem, M., Lammich, S., Steiner, H., Haass, C. 2002. Insulin degrading enzyme rapidly removes the beta amyloid precursor protein intracellular domain (AICD). J B iol Chem 277(16), 13389 93. El Amouri, S.S., Zhu, H., Yu, J., Gage, F.H., Verma, I.M., Kindy, M.S. 2007. Neprilysin protects neurons against Abeta peptide toxicity. Brain Res 1152, 191 200. El Khoury, J., Hickman, S.E., Thomas, C.A., Cao, L., Silverstein, S.C., Loike, J.D. 1996. Scavenger receptor mediated adhesion of microglia to beta amyloid fibrils. Nature 382(6593), 716 9. El Khoury, J., Toft, M., Hickman, S.E., Means, T.K., Terada, K., Geula, C., Luster, A.D. 2007. Ccr2 deficiency impairs microglial ac cumulation and accelerates progression of Alzheimer like disease. Nat Med 13(4), 432 8. Englund, U., Ericson, C., Rosenblad, C., Mandel, R.J., Trono, D., Wictorin, K., Lundberg, C. 2000. The use of a recombinant lentiviral vector for ex vivo gene transfer into the rat CNS. Neuroreport 11(18), 3973 7. Esen, N., Kielian, T. 2007. Effects of low dose GM CSF on microglial inflammatory profiles to diverse pathogen associated molecular patterns (PAMPs). J Neuroinflammation 4, 10. Fagan, A.M., Bu, G., Sun, Y., Dau gherty, A., Holtzman, D.M. 1996. Apolipoprotein E containing high density lipoprotein promotes neurite outgrowth and is a ligand for the low density lipoprotein receptor related protein. J Biol Chem 271(47), 30121 5. Farris, W., Mansourian, S., Chang, Y., Lindsley, L., Eckman, E.A., Frosch, M.P., Eckman, C.B., Tanzi, R.E., Selkoe, D.J., Guenette, S. 2003. Insulin degrading enzyme regulates the levels of insulin, amyloid beta protein, and the beta amyloid precursor protein intracellular domain in vivo. Proc Natl Acad Sci U S A 100(7), 4162 7. Farris, W., Schutz, S.G., Cirrito, J.R., Shankar, G.M., Sun, X., George, A., Leissring, M.A., Walsh, D.M., Qiu, W.Q., Holtzman, D.M., Selkoe, D.J. 2007. Loss of neprilysin function promotes amyloid plaque formation and c auses cerebral amyloid angiopathy. Am J Pathol 171(1), 241 51.

PAGE 111

111 Fiala, M., Lin, J., Ringman, J., Kermani Arab, V., Tsao, G., Patel, A., Lossinsky, A.S., Graves, M.C., Gustavson, A., Sayre, J., Sofroni, E., Suarez, T., Chiappelli, F., Bernard, G. 2005. Ineff ective phagocytosis of amyloid beta by macrophages of Alzheimer's disease patients. J Alzheimers Dis 7(3), 221 32; discussion 55 62. Flanary, B.E., Sammons, N.W., Nguyen, C., Walker, D., Streit, W.J. 2007. Evidence that aging and amyloid promote microglial cell senescence. Rejuvenation Res 10(1), 61 74. Floden, A.M., Combs, C.K. 2006. Beta amyloid stimulates murine postnatal and adult microglia cultures in a unique manner. J Neurosci 26(17), 4644 8. Floden, A.M., Li, S., Combs, C.K. 2005. Beta amyloid stimu lated microglia induce neuron death via synergistic stimulation of tumor necrosis factor alpha and NMDA receptors. J Neurosci 25(10), 2566 75. Frank, M.G., Wieseler Frank, J.L., Watkins, L.R., Maier, S.F. 2006. Rapid isolation of highly enriched and quiesc ent microglia from adult rat hippocampus: immunophenotypic and functional characteristics. J Neurosci Methods 151(2), 121 30. Frautschy, S.A., Yang, F., Irrizarry, M., Hyman, B., Saido, T.C., Hsiao, K., Cole, G.M. 1998. Microglial response to amyloid plaqu es in APPsw transgenic mice. Am J Pathol 152(1), 307 17. Fryer, J.D., Simmons, K., Parsadanian, M., Bales, K.R., Paul, S.M., Sullivan, P.M., Holtzman, D.M. 2005. Human apolipoprotein E4 alters the amyloid beta 40:42 ratio and promotes the formation of cere bral amyloid angiopathy in an amyloid precursor protein transgenic model. J Neurosci 25(11), 2803 10. Fusetti, M., Fioretti, A.B., Silvagni, F., Simaskou, M., Sucapane, P., Necozione, S., Eibenstein, A. Smell and preclinical Alzheimer disease: study of 29 patients with amnesic mild cognitive impairment. J Otolaryngol Head Neck Surg 39(2), 175 81. Games, D., Adams, D., Alessandrini, R., Barbour, R., Berthelette, P., Blackwell, C., Carr, T., Clemens, J., Donaldson, T., Gillespie, F., et al. 1995. Alzheimer ty pe neuropathology in transgenic mice overexpressing V717F beta amyloid precursor protein. Nature 373(6514), 523 7. Gao, C.Y., Stepp, M.A., Fariss, R., Zelenka, P. 2004. Cdk5 regulates activation and localization of Src during corneal epithelial wound closu re. J Cell Sci 117(Pt 18), 4089 98. German, D.C., Yazdani, U., Speciale, S.G., Pasbakhsh, P., Games, D., Liang, C.L. 2003. Cholinergic neuropathology in a mouse model of Alzheimer's disease. J Comp Neurol 462(4), 371 81.

PAGE 112

112 Giulian, D. 1987. Ameboid microglia as effectors of inflammation in the central nervous system. J Neurosci Res 18(1), 155 71, 32 3. Goate, A., Chartier Harlin, M.C., Mullan, M., Brown, J., Crawford, F., Fidani, L., Giuffra, L., Haynes, A., Irving, N., James, L., et al. 1991. Segregation of a missense mutation in the amyloid precursor protein gene with familial Alzheimer's disease. Nature 349(6311), 704 6. Hattiangady, B., Shuai, B., Cai, J., Coksaygan, T., Rao, M.S., Shetty, A.K. 2007. Increased dentate neurogenesis after grafting of glial r estricted progenitors or neural stem cells in the aging hippocampus. Stem Cells 25(8), 2104 17. He, B.P., Wen, W., Strong, M.J. 2002. Activated microglia (BV 2) facilitation of TNF alpha mediated motor neuron death in vitro. J Neuroimmunol 128(1 2), 31 8. Heneka, M.T., O'Banion, M.K. 2007. Inflammatory processes in Alzheimer's disease. J Neuroimmunol 184(1 2), 69 91. Hickman, S.E., Allison, E.K., El Khoury, J. 2008. Microglial dysfunction and defective beta amyloid clearance pathways in aging Alzheimer's di sease mice. J Neurosci 28(33), 8354 60. Higgins, G.A., Jacobsen, H. 2003. Transgenic mouse models of Alzheimer's disease: phenotype and application. Behav Pharmacol 14(5 6), 419 38. Hirrlinger, J., Gutterer, J.M., Kussmaul, L., Hamprecht, B., Dringen, R. 2 000. Microglial cells in culture express a prominent glutathione system for the defense against reactive oxygen species. Dev Neurosci 22(5 6), 384 92. Hoe, H.S., Cooper, M.J., Burns, M.P., Lewis, P.A., van der Brug, M., Chakraborty, G., Cartagena, C.M., Pa k, D.T., Cookson, M.R., Rebeck, G.W. 2007. The metalloprotease inhibitor TIMP 3 regulates amyloid precursor protein and apolipoprotein E receptor proteolysis. J Neurosci 27(40), 10895 905. Hsiao, K., Chapman, P., Nilsen, S., Eckman, C., Harigaya, Y., Younk in, S., Yang, F., Cole, G. 1996. Correlative memory deficits, Abeta elevation, and amyloid plaques in transgenic mice. Science 274(5284), 99 102. Huang, F., Buttini, M., Wyss Coray, T., McConlogue, L., Kodama, T., Pitas, R.E., Mucke, L. 1999. Elimination o f the class A scavenger receptor does not affect amyloid plaque formation or neurodegeneration in transgenic mice expressing human amyloid protein precursors. Am J Pathol 155(5), 1741 7.

PAGE 113

113 Imitola, J., Raddassi, K., Park, K.I., Mueller, F.J., Nieto, M., Te ng, Y.D., Frenkel, D., Li, J., Sidman, R.L., Walsh, C.A., Snyder, E.Y., Khoury, S.J. 2004. Directed migration of neural stem cells to sites of CNS injury by the stromal cell derived factor 1alpha/CXC chemokine receptor 4 pathway. Proc Natl Acad Sci U S A 1 01(52), 18117 22. Irizarry, M.C., Cheung, B.S., Rebeck, G.W., Paul, S.M., Bales, K.R., Hyman, B.T. 2000. Apolipoprotein E affects the amount, form, and anatomical distribution of amyloid beta peptide deposition in homozygous APP(V717F) transgenic mice. Act a Neuropathol (Berl) 100(5), 451 8. Iwata, N., Tsubuki, S., Takaki, Y., Shirotani, K., Lu, B., Gerard, N.P., Gerard, C., Hama, E., Lee, H.J., Saido, T.C. 2001. Metabolic regulation of brain Abeta by neprilysin. Science 292(5521), 1550 2. Jankowsky, J.L., S lunt, H.H., Gonzales, V., Savonenko, A.V., Wen, J.C., Jenkins, N.A., Copeland, N.G., Younkin, L.H., Lester, H.A., Younkin, S.G., Borchelt, D.R. 2005. Persistent amyloidosis following suppression of Abeta production in a transgenic model of Alzheimer diseas e. PLoS Med 2(12), e355. Jin, K., Peel, A.L., Mao, X.O., Xie, L., Cottrell, B.A., Henshall, D.C., Greenberg, D.A. 2004. Increased hippocampal neurogenesis in Alzheimer's disease. Proc Natl Acad Sci U S A 101(1), 343 7. Jung, S., Aliberti, J., Graemmel, P., Sunshine, M.J., Kreutzberg, G.W., Sher, A., Littman, D.R. 2000. Analysis of fractalkine receptor CX(3)CR1 function by targeted deletion and green fluorescent protein reporter gene insertion. Mol Cell Biol 20(11), 4106 14. Kang, S.S., Kook, J.H., Hwang, S. Park, S.H., Nam, S.C., Kim, J.K. 2008. Inhibition of matrix metalloproteinase 9 attenuated neural progenitor cell migration after photothrombotic ischemia. Brain Res 1228, 20 6. Klein, M.A., Moller, J.C., Jones, L.L., Bluethmann, H., Kreutzberg, G.W., Ra ivich, G. 1997. Impaired neuroglial activation in interleukin 6 deficient mice. Glia 19(3), 227 33. Kukekov, V.G., Laywell, E.D., Thomas, L.B., Steindler, D.A. 1997. A nestin negative precursor cell from the adult mouse brain gives rise to neurons and glia Glia 21(4), 399 407. Kumar Singh, S., Theuns, J., Van Broeck, B., Pirici, D., Vennekens, K., Corsmit, E., Cruts, M., Dermaut, B., Wang, R., Van Broeckhoven, C. 2006. Mean age of onset of familial alzheimer disease caused by presenilin mutations correlate s with both increased Abeta42 and decreased Abeta40. Hum Mutat 27(7), 686 95.

PAGE 114

114 Laywell, E.D., Rakic, P., Kukekov, V.G., Holland, E.C., Steindler, D.A. 2000. Identification of a multipotent astrocytic stem cell in the immature and adult mouse brain. Proc Nat l Acad Sci U S A 97(25), 13883 8. Leal, M.C., Dorfman, V.B., Gamba, A.F., Frangione, B., Wisniewski, T., Castano, E.M., Sigurdsson, E.M., Morelli, L. 2006. Plaque associated overexpression of insulin degrading enzyme in the cerebral cortex of aged transgen ic tg2576 mice with Alzheimer pathology. J Neuropathol Exp Neurol 65(10), 976 87. Leissring, M.A., Farris, W., Chang, A.Y., Walsh, D.M., Wu, X., Sun, X., Frosch, M.P., Selkoe, D.J. 2003. Enhanced proteolysis of beta amyloid in APP transgenic mice prevents plaque formation, secondary pathology, and premature death. Neuron 40(6), 1087 93. Lewis, J., Dickson, D.W., Lin, W.L., Chisholm, L., Corral, A., Jones, G., Yen, S.H., Sahara, N., Skipper, L., Yager, D., Eckman, C., Hardy, J., Hutton, M., McGowan, E. 2001. Enhanced neurofibrillary degeneration in transgenic mice expressing mutant tau and APP. Science 293(5534), 1487 91. Li, J.P., Galvis, M.L., Gong, F., Zhang, X., Zcharia, E., Metzger, S., Vlodavsky, I., Kisilevsky, R., Lindahl, U. 2005. In vivo fragmentati on of heparan sulfate by heparanase overexpression renders mice resistant to amyloid protein A amyloidosis. Proc Natl Acad Sci U S A 102(18), 6473 7. Li, L., Lu, J., Tay, S.S., Moochhala, S.M., He, B.P. 2007. The function of microglia, either neuroprotecti on or neurotoxicity, is determined by the equilibrium among factors released from activated microglia in vitro. Brain Res 1159, 8 17. Liao, M.C., Van Nostrand, W.E. 2010. Degradation of soluble and fibrillar amyloid beta protein by matrix metalloproteinase (MT1 MMP) in vitro. Biochemistry 49(6), 1127 36. Licastro, F., Pedrini, S., Caputo, L., Annoni, G., Davis, L.J., Ferri, C., Casadei, V., Grimaldi, L.M. 2000. Increased plasma levels of interleukin 1, interleukin 6 and alpha 1 antichymotrypsin in patients with Alzheimer's disease: peripheral inflammation or signals from the brain? J Neuroimmunol 103(1), 97 102. Lindenau, J., Noack, H., Asayama, K., Wolf, G. 1998. Enhanced cellular glutathione peroxidase immunoreactivity in activated astrocytes and in microg lia during excitotoxin induced neurodegeneration. Glia 24(2), 252 6. Loddick, S.A., Turnbull, A.V., Rothwell, N.J. 1998. Cerebral interleukin 6 is neuroprotective during permanent focal cerebral ischemia in the rat. J Cereb Blood Flow Metab 18(2), 176 9. L opez Toledano, M.A., Shelanski, M.L. 2004. Neurogenic effect of beta amyloid peptide in the development of neural stem cells. J Neurosci 24(23), 5439 44.

PAGE 115

115 Lorenzl, S., Albers, D.S., LeWitt, P.A., Chirichigno, J.W., Hilgenberg, S.L., Cudkowicz, M.E., Beal, M .F. 2003. Tissue inhibitors of matrix metalloproteinases are elevated in cerebrospinal fluid of neurodegenerative diseases. J Neurol Sci 207(1 2), 71 6. Lu, W.G., Chen, H., Wang, D., Li, F.G., Zhang, S.M. 2007. Regionspecific survival and differentiation o f mouse embryonic stem cell derived implants in the adult rat brain. Sheng Li Xue Bao 59(1), 51 7. Lubbe, W.J., Zhou, Z.Y., Fu, W., Zuzga, D., Schulz, S., Fridman, R., Muschel, R.J., Waldman, S.A., Pitari, G.M. 2006. Tumor epithelial cell matrix metallopro teinase 9 is a target for antimetastatic therapy in colorectal cancer. Clin Cancer Res 12(6), 1876 82. Lue, L.F., Walker, D.G. 2002. Modeling Alzheimer's disease immune therapy mechanisms: interactions of human postmortem microglia with antibody opsonized amyloid beta peptide. J Neurosci Res 70(4), 599 610. Lundberg, C., Martinez Serrano, A., Cattaneo, E., McKay, R.D., Bjorklund, A. 1997. Survival, integration, and differentiation of neural stem cell lines after transplantation to the adult rat striatum. Ex p Neurol 145(2 Pt 1), 342 60. Majumdar, A., Chung, H., Dolios, G., Wang, R., Asamoah, N., Lobel, P., Maxfield, F.R. 2007a. Degradation of fibrillar forms of Alzheimer's amyloid beta peptide by macrophages. Neurobiol Aging. Majumdar, A., Cruz, D., Asamoah, N., Buxbaum, A., Sohar, I., Lobel, P., Maxfield, F.R. 2007b. Activation of microglia acidifies lysosomes and leads to degradation of Alzheimer amyloid fibrils. Mol Biol Cell 18(4), 1490 6. Marr, R.A., Rockenstein, E., Mukherjee, A., Kindy, M.S., Hersh, L.B ., Gage, F.H., Verma, I.M., Masliah, E. 2003. Neprilysin gene transfer reduces human amyloid pathology in transgenic mice. J Neurosci 23(6), 1992 6. Marshall, G.P., 2nd, Laywell, E.D., Zheng, T., Steindler, D.A., Scott, E.W. 2006. In vitro derived "neural stem cells" function as neural progenitors without the capacity for self renewal. Stem Cells 24(3), 731 8. Marz, P., Cheng, J.G., Gadient, R.A., Patterson, P.H., Stoyan, T., Otten, U., Rose John, S. 1998. Sympathetic neurons can produce and respond to inte rleukin 6. Proc Natl Acad Sci U S A 95(6), 3251 6. McDermott, J.R., Bartram, R.E., Knight, P.A., Miller, H.R., Garrod, D.R., Grencis, R.K. 2003. Mast cells disrupt epithelial barrier function during enteric nematode infection. Proc Natl Acad Sci U S A 100( 13), 7761 6.

PAGE 116

116 McGeer, P.L., McGeer, E.G. 2001. Polymorphisms in inflammatory genes and the risk of Alzheimer disease. Arch Neurol 58(11), 1790 2. Meyer Luehmann, M., Spires Jones, T.L., Prada, C., Garcia Alloza, M., de Calignon, A., Rozkalne, A., Koenigskne cht Talboo, J., Holtzman, D.M., Bacskai, B.J., Hyman, B.T. 2008. Rapid appearance and local toxicity of amyloid beta plaques in a mouse model of Alzheimer's disease. Nature 451(7179), 720 4. Miller, K.R., Streit, W.J. 2007. The effects of aging, injury and disease on microglial function: a case for cellular senescence. Neuron Glia Biol 3(3), 245 53. Milton, R.H., Abeti, R., Averaimo, S., DeBiasi, S., Vitellaro, L., Jiang, L., Curmi, P.M., Breit, S.N., Duchen, M.R., Mazzanti, M. 2008. CLIC1 function is requi red for beta amyloid induced generation of reactive oxygen species by microglia. J Neurosci 28(45), 11488 99. Moll, U.M., Youngleib, G.L., Rosinski, K.B., Quigley, J.P. 1990. Tumor promoter stimulated Mr 92,000 gelatinase secreted by normal and malignant h uman cells: isolation and characterization of the enzyme from HT1080 tumor cells. Cancer Res 50(19), 6162 70. Moolman, D.L., Vitolo, O.V., Vonsattel, J.P., Shelanski, M.L. 2004. Dendrite and dendritic spine alterations in Alzheimer models. J Neurocytol 33( 3), 377 87. Mosconi, L., Rinne, J.O., Tsui, W.H., Berti, V., Li, Y., Wang, H., Murray, J., Scheinin, N., Nagren, K., Williams, S., Glodzik, L., De Santi, S., Vallabhajosula, S., de Leon, M.J. Increased fibrillar amyloid {beta} burden in normal individuals with a family history of late onset Alzheimer's. Proc Natl Acad Sci U S A. Motte, J., Williams, R.S. 1989. Age related changes in the density and morphology of plaques and neurofibrillary tangles in Down syndrome brain. Acta Neuropathol (Berl) 77(5), 535 4 6. Munger, J.S., Haass, C., Lemere, C.A., Shi, G.P., Wong, W.S., Teplow, D.B., Selkoe, D.J., Chapman, H.A. 1995. Lysosomal processing of amyloid precursor protein to A beta peptides: a distinct role for cathepsin S. Biochem J 311 (Pt 1), 299 305. Nagaraja, T.N., Patel, P., Gorski, M., Gorevic, P.D., Patlak, C.S., Fenstermacher, J.D. 2005. In normal rat, intraventricularly administered insulin like growth factor 1 is rapidly cleared from CSF with limited distribution into brain. Cerebrospinal Fluid Res 2, 5. Naidu, A., Xu, Q., Catalano, R., Cordell, B. 2002. Secretion of apolipoprotein E by brain glia requires protein prenylation and is suppressed by statins. Brain Res 958(1), 100 11.

PAGE 117

117 Njie, E.G., Boelen, E., Stassen, F.R., Steinbusch, H.W., Borchelt, D.R., St reit, W.J. 2010. Ex vivo cultures of microglia from young and aged rodent brain reveal age related changes in microglial function. Neurobiol Aging. Oddo, S., Caccamo, A., Shepherd, J.D., Murphy, M.P., Golde, T.E., Kayed, R., Metherate, R., Mattson, M.P., A kbari, Y., LaFerla, F.M. 2003. Triple transgenic model of Alzheimer's disease with plaques and tangles: intracellular Abeta and synaptic dysfunction. Neuron 39(3), 409 21. Olstorn, H., Moe, M.C., Roste, G.K., Bueters, T., Langmoen, I.A. 2007. Transplantati on of stem cells from the adult human brain to the adult rat brain. Neurosurgery 60(6), 1089 98; discussion 98 9. Paresce, D.M., Chung, H., Maxfield, F.R. 1997. Slow degradation of aggregates of the Alzheimer's disease amyloid beta protein by microglial ce lls. J Biol Chem 272(46), 29390 7. Park, K.I., Teng, Y.D., Snyder, E.Y. 2002. The injured brain interacts reciprocally with neural stem cells supported by scaffolds to reconstitute lost tissue. Nat Biotechnol 20(11), 1111 7. Parsons, C.G., Stoffler, A., Da nysz, W. 2007. Memantine: a NMDA receptor antagonist that improves memory by restoration of homeostasis in the glutamatergic system too little activation is bad, too much is even worse. Neuropharmacology 53(6), 699 723. Patton, R.L., Kalback, W.M., Esh, C.L., Kokjohn, T.A., Van Vickle, G.D., Luehrs, D.C., Kuo, Y.M., Lopez, J., Brune, D., Ferrer, I., Masliah, E., Newel, A.J., Beach, T.G., Castano, E.M., Roher, A.E. 2006. Amyloid beta peptide remnants in AN 1792 immunized Alzheimer's disease patients: a bio chemical analysis. Am J Pathol 169(3), 1048 63. Peress, N., Perillo, E., Zucker, S. 1995. Localization of tissue inhibitor of matrix metalloproteinases in Alzheimer's disease and normal brain. J Neuropathol Exp Neurol 54(1), 16 22. Pihlaja, R., Koistinaho, J., Malm, T., Sikkila, H., Vainio, S., Koistinaho, M. 2008. Transplanted astrocytes internalize deposited beta amyloid peptides in a transgenic mouse model of Alzheimer's disease. Glia 56(2), 154 63. Podoly, E., Bruck, T., Diamant, S., Melamed Book, N., W eiss, A., Huang, Y., Livnah, O., Langermann, S., Wilgus, H., Soreq, H. 2008. Human recombinant butyrylcholinesterase purified from the milk of transgenic goats interacts with beta amyloid fibrils and suppresses their formation in vitro. Neurodegener Dis 5( 3 4), 232 6.

PAGE 118

118 Ponomarev, E.D., Novikova, M., Maresz, K., Shriver, L.P., Dittel, B.N. 2005. Development of a culture system that supports adult microglial cell proliferation and maintenance in the resting state. J Immunol Methods 300(1 2), 32 46. Prajerova, I., Honsa, P., Chvatal, A., Anderova, M. Neural stem/progenitor cells derived from the embryonic dorsal telencephalon of D6/GFP mice differentiate primarily into neurons after transplantation into a cortical lesion. Cell Mol Neurobiol 30(2), 199 218. Price J.L., McKeel, D.W., Jr., Buckles, V.D., Roe, C.M., Xiong, C., Grundman, M., Hansen, L.A., Petersen, R.C., Parisi, J.E., Dickson, D.W., Smith, C.D., Davis, D.G., Schmitt, F.A., Markesbery, W.R., Kaye, J., Kurlan, R., Hulette, C., Kurland, B.F., Higdon, R. Kukull, W., Morris, J.C. 2009. Neuropathology of nondemented aging: presumptive evidence for preclinical Alzheimer disease. Neurobiol Aging 30(7), 1026 36. Qin, L., Liu, Y., Wang, T., Wei, S.J., Block, M.L., Wilson, B., Liu, B., Hong, J.S. 2004. NADPH ox idase mediates lipopolysaccharide induced neurotoxicity and proinflammatory gene expression in activated microglia. J Biol Chem 279(2), 1415 21. Qiu, W.Q., Walsh, D.M., Ye, Z., Vekrellis, K., Zhang, J., Podlisny, M.B., Rosner, M.R., Safavi, A., Hersh, L.B. Selkoe, D.J. 1998. Insulin degrading enzyme regulates extracellular levels of amyloid beta protein by degradation. J Biol Chem 273(49), 32730 8. Radojevic, V., Kapfhammer, J.P. 2009. Directed fiber outgrowth from transplanted embryonic cortex derived neu rospheres in the adult mouse brain. Neural Plast 2009, 852492. Raedt, R., Van Dycke, A., Waeytens, A., Wyckhuys, T., Vonck, K., Wadman, W., Boon, P. 2009. Unconditioned adult derived neurosphere cells mainly differentiate towards astrocytes upon transplant ation in sclerotic rat hippocampus. Epilepsy Res 87(2 3), 148 59. Ramos DeSimone, N., Hahn Dantona, E., Sipley, J., Nagase, H., French, D.L., Quigley, J.P. 1999. Activation of matrix metalloproteinase 9 (MMP 9) via a converging plasmin/stromelysin 1 cascad e enhances tumor cell invasion. J Biol Chem 274(19), 13066 76. Raponi, E., Agenes, F., Delphin, C., Assard, N., Baudier, J., Legraverend, C., Deloulme, J.C. 2007. S100B expression defines a state in which GFAP expressing cells lose their neural stem cell p otential and acquire a more mature developmental stage. Glia 55(2), 165 77.

PAGE 119

119 Reynolds, B.A., Weiss, S. 1992. Generation of neurons and astrocytes from isolated cells of the adult mammalian central nervous system. Science 255(5052), 1707 10. Rogers, J., Stro hmeyer, R., Kovelowski, C.J., Li, R. 2002. Microglia and inflammatory mechanisms in the clearance of amyloid beta peptide. Glia 40(2), 260 9. Ross, J. 1995. mRNA stability in mammalian cells. Microbiol Rev 59(3), 423 50. Rovelet Lecrux, A., Hannequin, D., Raux, G., Le Meur, N., Laquerriere, A., Vital, A., Dumanchin, C., Feuillette, S., Brice, A., Vercelletto, M., Dubas, F., Frebourg, T., Campion, D. 2006. APP locus duplication causes autosomal dominant early onset Alzheimer disease with cerebral amyloid ang iopathy. Nat Genet 38(1), 24 6. Ruiz Opazo, N., Kosik, K.S., Lopez, L.V., Bagamasbad, P., Ponce, L.R., Herrera, V.L. 2004. Attenuated hippocampus dependent learning and memory decline in transgenic TgAPPswe Fischer 344 rats. Mol Med 10(1 6), 36 44. Ryu, J. K., Cho, T., Wang, Y.T., McLarnon, J.G. 2009. Neural progenitor cells attenuate inflammatory reactivity and neuronal loss in an animal model of inflamed AD brain. J Neuroinflammation 6, 39. Sabbagh, M.N. 2009. Drug development for Alzheimer's disease: wher e are we now and where are we headed? Am J Geriatr Pharmacother 7(3), 167 85. Savonenko, A., Xu, G.M., Melnikova, T., Morton, J.L., Gonzales, V., Wong, M.P., Price, D.L., Tang, F., Markowska, A.L., Borchelt, D.R. 2005. Episodic like memory deficits in the APPswe/PS1dE9 mouse model of Alzheimer's disease: relationships to beta amyloid deposition and neurotransmitter abnormalities. Neurobiol Dis 18(3), 602 17. Sawaya, M.R., Sambashivan, S., Nelson, R., Ivanova, M.I., Sievers, S.A., Apostol, M.I., Thompson, M. J., Balbirnie, M., Wiltzius, J.J., McFarlane, H.T., Madsen, A.O., Riekel, C., Eisenberg, D. 2007. Atomic structures of amyloid cross beta spines reveal varied steric zippers. Nature 447(7143), 453 7. Scheffler, B., Walton, N.M., Lin, D.D., Goetz, A.K., Eni kolopov, G., Roper, S.N., Steindler, D.A. 2005. Phenotypic and functional characterization of adult brain neuropoiesis. Proc Natl Acad Sci U S A 102(26), 9353 8. Schmidt, S.D., Nixon, R.A., Mathews, P.M. 2005. ELISA method for measurement of amyloid beta l evels. Methods Mol Biol 299, 279 97. Schwab, C., Hosokawa, M., McGeer, P.L. 2004. Transgenic mice overexpressing amyloid beta protein are an incomplete model of Alzheimer disease. Exp Neurol 188(1), 52 64.

PAGE 120

120 Shagin, D.A., Barsova, E.V., Yanushevich, Y.G., Fr adkov, A.F., Lukyanov, K.A., Labas, Y.A., Semenova, T.N., Ugalde, J.A., Meyers, A., Nunez, J.M., Widder, E.A., Lukyanov, S.A., Matz, M.V. 2004. GFP like proteins as ubiquitous metazoan superfamily: evolution of functional features and structural complexity Mol Biol Evol 21(5), 841 50. Shen, Y., Joachimiak, A., Rosner, M.R., Tang, W.J. 2006. Structures of human insulin degrading enzyme reveal a new substrate recognition mechanism. Nature 443(7113), 870 4. Shetty, A.K., Rao, M.S., Hattiangady, B. 2008. Behav ior of hippocampal stem/progenitor cells following grafting into the injured aged hippocampus. J Neurosci Res 86(14), 3062 74. Sierra, A., Gottfried Blackmore, A.C., McEwen, B.S., Bulloch, K. 2007. Microglia derived from aging mice exhibit an altered infla mmatory profile. Glia 55(4), 412 24. Simard, A.R., Soulet, D., Gowing, G., Julien, J.P., Rivest, S. 2006. Bone marrow derived microglia play a critical role in restricting senile plaque formation in Alzheimer's disease. Neuron 49(4), 489 502. Simmons, D.A. Casale, M., Alcon, B., Pham, N., Narayan, N., Lynch, G. 2007. Ferritin accumulation in dystrophic microglia is an early event in the development of Huntington's disease. Glia 55(10), 1074 84. Stalder, M., Phinney, A., Probst, A., Sommer, B., Staufenbiel, M., Jucker, M. 1999. Association of microglia with amyloid plaques in brains of APP23 transgenic mice. Am J Pathol 154(6), 1673 84. Storm van's Gravesande, K., Layne, M.D., Ye, Q., Le, L., Baron, R.M., Perrella, M.A., Santambrogio, L., Silverman, E.S., Ri ese, R.J. 2002. IFN regulatory factor 1 regulates IFN gamma dependent cathepsin S expression. J Immunol 168(9), 4488 94. Streit, W.J., Hurley, S.D., McGraw, T.S., Semple Rowland, S.L. 2000. Comparative evaluation of cytokine profiles and reactive gliosis s upports a critical role for interleukin 6 in neuron glia signaling during regeneration. J Neurosci Res 61(1), 10 20. Streit, W.J., Miller, K.R., Lopes, K.O., Njie, E. 2008. Microglial degeneration in the aging brain -bad news for neurons? Front Biosci 13, 3423 38. Streit, W.J., Mrak, R.E., Griffin, W.S. 2004. Microglia and neuroinflammation: a pathological perspective. J Neuroinflammation 1(1), 14.

PAGE 121

121 Sturchler Pierrat, C., Abramowski, D., Duke, M., Wiederhold, K.H., Mistl, C., Rothacher, S., Ledermann, B., B urki, K., Frey, P., Paganetti, P.A., Waridel, C., Calhoun, M.E., Jucker, M., Probst, A., Staufenbiel, M., Sommer, B. 1997. Two amyloid precursor protein transgenic mouse models with Alzheimer disease like pathology. Proc Natl Acad Sci U S A 94(24), 13287 9 2. Svendsen, C.N., Clarke, D.J., Rosser, A.E., Dunnett, S.B. 1996. Survival and differentiation of rat and human epidermal growth factor responsive precursor cells following grafting into the lesioned adult central nervous system. Exp Neurol 137(2), 376 88 Takaki, Y., Iwata, N., Tsubuki, S., Taniguchi, S., Toyoshima, S., Lu, B., Gerard, N.P., Gerard, C., Lee, H.J., Shirotani, K., Saido, T.C. 2000. Biochemical identification of the neutral endopeptidase family member responsible for the catabolism of amyloi d beta peptide in the brain. J Biochem (Tokyo) 128(6), 897 902. Tan, J., Town, T., Paris, D., Mori, T., Suo, Z., Crawford, F., Mattson, M.P., Flavell, R.A., Mullan, M. 1999. Microglial activation resulting from CD40 CD40L interaction after beta amyloid sti mulation. Science 286(5448), 2352 5. Tang, J., Xu, H., Fan, X., Li, D., Rancourt, D., Zhou, G., Li, Z., Yang, L. 2008. Embryonic stem cell derived neural precursor cells improve memory dysfunction in Abeta(1 40) injured rats. Neurosci Res 62(2), 86 96. Tan zi, R.E., Bertram, L. 2005. Twenty years of the Alzheimer's disease amyloid hypothesis: a genetic perspective. Cell 120(4), 545 55. Tarkowski, E., Blennow, K., Wallin, A., Tarkowski, A. 1999. Intracerebral production of tumor necrosis factor alpha, a local neuroprotective agent, in Alzheimer disease and vascular dementia. J Clin Immunol 19(4), 223 30. Tchaikovskaya, T., Fraifeld, V., Urphanishvili, T., Andorfer, J.H., Davies, P., Listowsky, I. 2005. Glutathione S transferase hGSTM3 and ageing associated neu rodegeneration: relationship to Alzheimer's disease. Mech Ageing Dev 126(2), 309 15. Terry, A.V., Jr., Buccafusco, J.J. 2003. The cholinergic hypothesis of age and Alzheimer's disease related cognitive deficits: recent challenges and their implications for novel drug development. J Pharmacol Exp Ther 306(3), 821 7. Tham, C.S., Lin, F.F., Rao, T.S., Yu, N., Webb, M. 2003. Microglial activation state and lysophospholipid acid receptor expression. Int J Dev Neurosci 21(8), 431 43. Thier, M., Marz, P., Otten, U ., Weis, J., Rose John, S. 1999. Interleukin 6 (IL 6) and its soluble receptor support survival of sensory neurons. J Neurosci Res 55(4), 411 22.

PAGE 122

122 Tietze, F. 1969. Enzymic method for quantitative determination of nanogram amounts of total and oxidized gluta thione: applications to mammalian blood and other tissues. Anal Biochem 27(3), 502 22. Verret, L., Jankowsky, J.L., Xu, G.M., Borchelt, D.R., Rampon, C. 2007. Alzheimer's type amyloidosis in transgenic mice impairs survival of newborn neurons derived from adult hippocampal neurogenesis. J Neurosci 27(25), 6771 80. Walsh, D.M., Klyubin, I., Fadeeva, J.V., Cullen, W.K., Anwyl, R., Wolfe, M.S., Rowan, M.J., Selkoe, D.J. 2002. Naturally secreted oligomers of amyloid beta protein potently inhibit hippocampal lon g term potentiation in vivo. Nature 416(6880), 535 9. Walton, N.M., Sutter, B.M., Chen, H.X., Chang, L.J., Roper, S.N., Scheffler, B., Steindler, D.A. 2006a. Derivation and large scale expansion of multipotent astroglial neural progenitors from adult human brain. Development 133(18), 3671 81. Walton, N.M., Sutter, B.M., Laywell, E.D., Levkoff, L.H., Kearns, S.M., Marshall, G.P., 2nd, Scheffler, B., Steindler, D.A. 2006b. Microglia instruct subventricular zone neurogenesis. Glia 54(8), 815 25. Wang, L., Zhan g, Z.G., Zhang, R.L., Gregg, S.R., Hozeska Solgot, A., LeTourneau, Y., Wang, Y., Chopp, M. 2006. Matrix metalloproteinase 2 (MMP2) and MMP9 secreted by erythropoietin activated endothelial cells promote neural progenitor cell migration. J Neurosci 26(22), 5996 6003. Wang, R., Dineley, K.T., Sweatt, J.D., Zheng, H. 2004. Presenilin 1 familial Alzheimer's disease mutation leads to defective associative learning and impaired adult neurogenesis. Neuroscience 126(2), 305 12. Watson, D.J., Walton, R.M., Magnitsky S.G., Bulte, J.W., Poptani, H., Wolfe, J.H. 2006. Structure specific patterns of neural stem cell engraftment after transplantation in the adult mouse brain. Hum Gene Ther 17(7), 693 704. Wegiel, J., Imaki, H., Wang, K.C., Wegiel, J., Rubenstein, R. 2004 Cells of monocyte/microglial lineage are involved in both microvessel amyloidosis and fibrillar plaque formation in APPsw tg mice. Brain Res 1022(1 2), 19 29. Wei, J., Xu, H., Davies, J.L., Hemmings, G.P. 1992. Increase of plasma IL 6 concentration with age in healthy subjects. Life Sci 51(25), 1953 6. Wei, L.H., Kuo, M.L., Chen, C.A., Chou, C.H., Cheng, W.F., Chang, M.C., Su, J.L., Hsieh, C.Y. 2001. The anti apoptotic role of interleukin 6 in human cervical cancer is mediated by up regulation of Mcl 1 th rough a PI 3 K/Akt pathway. Oncogene 20(41), 5799 809.

PAGE 123

123 Westerman, M.A., Cooper Blacketer, D., Mariash, A., Kotilinek, L., Kawarabayashi, T., Younkin, L.H., Carlson, G.A., Younkin, S.G., Ashe, K.H. 2002. The relationship between Abeta and memory in the Tg25 76 mouse model of Alzheimer's disease. J Neurosci 22(5), 1858 67. Wong, P.C., Cai, H., Borchelt, D.R., Price, D.L. 2002. Genetically engineered mouse models of neurodegenerative diseases. Nat Neurosci 5(7), 633 9. Wyss Coray, T., Lin, C., Yan, F., Yu, G.Q. Rohde, M., McConlogue, L., Masliah, E., Mucke, L. 2001. TGF beta1 promotes microglial amyloid beta clearance and reduces plaque burden in transgenic mice. Nat Med 7(5), 612 8. Xu, G., Gonzales, V., Borchelt, D.R. 2002. Abeta deposition does not cause the aggregation of endogenous tau in transgenic mice. Alzheimer Dis Assoc Disord 16(3), 196 201. Xu, Q., Li, Y., Cyras, C., Sanan, D.A., Cordell, B. 2000. Isolation and characterization of apolipoproteins from murine microglia. Identification of a low density lipoprotein like apolipoprotein J rich but E poor spherical particle. J Biol Chem 275(41), 31770 7. Yan, P., Hu, X., Song, H., Yin, K., Bateman, R.J., Cirrito, J.R., Xiao, Q., Hsu, F.F., Turk, J.W., Xu, J., Hsu, C.Y., Holtzman, D.M., Lee, J.M. 2006. Matri x metalloproteinase 9 degrades amyloid beta fibrils in vitro and compact plaques in situ. J Biol Chem 281(34), 24566 74. Yan, S.D., Chen, X., Fu, J., Chen, M., Zhu, H., Roher, A., Slattery, T., Zhao, L., Nagashima, M., Morser, J., Migheli, A., Nawroth, P., Stern, D., Schmidt, A.M. 1996. RAGE and amyloid beta peptide neurotoxicity in Alzheimer's disease. Nature 382(6593), 685 91. Yasojima, K., Akiyama, H., McGeer, E.G., McGeer, P.L. 2001. Reduced neprilysin in high plaque areas of Alzheimer brain: a possible relationship to deficient degradation of beta amyloid peptide. Neurosci Lett 297(2), 97 100. Ye, S.M., Johnson, R.W. 1999. Increased interleukin 6 expression by microglia from brain of aged mice. J Neuroimmunol 93(1 2), 139 48. Yin, K.J., Cirrito, J.R., Y an, P., Hu, X., Xiao, Q., Pan, X., Bateman, R., Song, H., Hsu, F.F., Turk, J., Xu, J., Hsu, C.Y., Mills, J.C., Holtzman, D.M., Lee, J.M. 2006. Matrix metalloproteinases expressed by astrocytes mediate extracellular amyloid beta peptide catabolism. J Neuros ci 26(43), 10939 48. Yoon, I.S., Chen, E., Busse, T., Repetto, E., Lakshmana, M.K., Koo, E.H., Kang, D.E. 2007. Low density lipoprotein receptor related protein promotes amyloid precursor protein trafficking to lipid rafts in the endocytic pathway. Faseb J

PAGE 124

124 Zandi, P.P., Anthony, J.C., Khachaturian, A.S., Stone, S.V., Gustafson, D., Tschanz, J.T., Norton, M.C., Welsh Bohmer, K.A., Breitner, J.C. 2004. Reduced risk of Alzheimer disease in users of antioxidant vitamin supplements: the Cache County Study. Arch Neurol 61(1), 82 8. Zheng, T., Marshall, G.P., 2nd, Laywell, E.D., Steindler, D.A. 2006. Neurogenic astrocytes transplanted into the adult mouse lateral ventricle contribute to olfactory neurogenesis, and reveal a novel intrinsic subependymal neuron. Neuro science 142(1), 175 85. Ziabreva, I., Perry, E., Perry, R., Minger, S.L., Ekonomou, A., Przyborski, S., Ballard, C. 2006. Altered neurogenesis in Alzheimer's disease. J Psychosom Res 61(3), 311 6.

PAGE 125

125 BIOGRAPHICAL SKETCH eMali ck Njie is from The Gambia and has always dreamed of being a scientist.