Pathological and biochemical consequences of chronicneuroinflammation in the basal forebrain: An investigation

of the mechanisms underlying neuroprotection

Item Type text; Dissertation-Reproduction (electronic)

Authors Willard, Lauren Baker

Publisher The University of Arizona.

Rights Copyright © is held by the author. Digital access to this materialis made possible by the University Libraries, University of Arizona.Further transmission, reproduction or presentation (such aspublic display or performance) of protected items is prohibitedexcept with permission of the author.

Download date 01/07/2018 23:30:36

Link to Item http://hdl.handle.net/10150/284068

INFORMATION TO USERS

This manuscript has t>een reproduced from the microfilm master. UMI films

the text directly from the original or copy submitted. Thus, some thesis and

dissertation copies are In typewriter face, while others may be from any type of

computer printer.

The quality of this reproduction is dependent upon the quality of the

copy submitted. Broken or indistinct print, colored or poor quality illustrations

and photographs, print bleedthrough, substandard margins, and Improper

alignment can adversely affect reproduction.

In the unlikely event that the author dkJ not send UMI a complete manuscript

and there are missing pages, these will be noted. Also, if unauthorized

copyright material had to be removed, a note will indicate the deletkxi.

Oversize materials (e.g., maps, drawings, charts) are reproduced by

sectioning the original, beginning at the upper left-hand comer and continuing

from left to right in equal sections with small overiaps.

Photographs included in the original manuscript have been reproduced

xerographically in this copy. Higher quality 6' x 9* black and white

photographic prints are available for any photographs or illustrations appearing

in this copy for an additional charge. Contact UMI directty to order.

Bell & Howell Information and Leaming 300 North Zeeb Road, Ann Arbor, Ml 48106-1346 USA

800-521-0600

PATHOLOGICAL AND BIOCHEMICAL CONSEQUENCES OF CHRONIC

NEUROINFLAMMATION IN THE BASAL FOREBRAIN: AN INVESTIGATION OF

THE MECHANISMS UNDERLYING NEUROPROTECTION

by

Lauren Baker Willard

A Dissertation Submitted to the Faculty of the

INTERDISCIPLINARY GRADUATE PRO(»AM IN NEUROSCIENCE

In Partial Fulfillment of the Requirements

For the Degree of

DOCTOR OF PHILOSOPHY

In the Graduate College

THE UNIVERSITY OF ARIZONA

1 9 9 9

UMI Number 9960272

®

UMI UMI Microfomi9960272

Copyright 2000 by Bell & Howell Infomiatlon and Leaming Company. All rights reserved. This microfonn edition is protected against

unauthorized copying under Title 17, United States Code.

Bell & Howell Information and Leaming Company 300 North Zeeb Road

P.O. Box 1346 Ann Ari3or,MI 48106-1346

2

THE UNIVERSITY OF ARIZONA ®

GRADUATE COLLEGE

As members of the Final Examination Committee, we certify that we have

read the dissertation prepared by Lauren Baker Wlllard

entitled Pathological and Biochemical Consequences of Chronic

Neurolnflammatlon in the Basal Forebrain: An Investigation

of the Mechanisms underlying Neuroprotection

and recommend that it be accepted as fulfilling the dissertation

requirement for the Degree of Doctor of Philosophv

Garv/Wenk^^h.D. ^ Date ? 1 -/r

Naomi Ranee, M.D., Ph.D. Ph.D. Dattfe '

Marchalonls, Ph.D. Date

lAra, Ph.D. DaU ' J ' Henrv xama

Robert Slovlter, Ph.D. Date

Final approval and acceptance of this dissertation Is contingent upon

the candidate's submission of the final copy of the dissertation to the Graduate College.

I hereby certify that I have read this dissertation prepared under my direction and recommend that it be accepted as fulfilling the dissertation

requirement.

^ ̂ ^ Diss^tatioX ̂ irector Gary Wenk, Ph.D. Date

3

STATEMENT BY THE AUTHOR

This dissertation has been submitted in partial fulfillment of requirements for an advanced degree at The University of Arizona and is deposited in the University Library to be made available to borrowers under rules of the Library.

Brief quotations from this dissertation are allowable without special permission, provided that accurate acknowledgement of source is made. Requests for permission for extended quotation from or reproduction of this manuscript in whole or in part may be granted by the head of the major department or the dean of the Graduate College when in his or her judgment the proposed use of the material is in the interests of scholarship. In all other instances, however, permission must be obtained from the author.

SIGNED:

4

ACKNOWLEDGEMENTS

Any work of science is usually the product of many and this work is no exception.

Special thanks to the members of my committee: Drs. John Marchalonis, Naomi

Ranee, Robert Sloviter, Gary Wenk and Henry Yamamura. It has been my privilege to

work with these distinguished scientists and educators, and all of them have made

important contributions to my education and training. I would also like to separately

thank my advisor. Dr. Gary Wenk for his continual feedback and direction, and Dr.

Beatrice Hauss-Wegrzyniak for sharing her histological expertise and her passion for

science.

My mentor Dr. James Naime, who, though at another university, has been a

constant source of guidance and encouragement.

I would like to thank my husband, Chad, not only for the unending support and

patience necessary for enduring the graduate school process, but also for rescuing this

document from certain digital oblivion.

Finally, I would like to thank the 419 rats that provided the data.

5

TABLE OF CONTENTS

LISTOFEiUSTRATIONS 8

UST OF TABLES 9

ABSTRACT 10

1. INTRODUCTION 12

Role of Inflammatory Processes in the Degeneration of Basal

Forebrain Cells in AD 16

Role of Prostaglandins 19

Role of Glutamate and NMDA Receptors 21

2. PATHOLOGICAL AND BIOCHEMICAL CONSEQUENCES OF ACUTE

AND CHRONIC NEUROINFLAMMATION WITHIN THE BASAL

FOREBRAIN CHOLINERGIC SYSTEM OF RATS 23

Abstract 23

Introduction 24

Materials and Methods 25

Subjects 25

Surgical Procedures 25

Biochemistry 27

Histology 28

6

TABLE OF CONTENTS - Continued

Results 29

Acute Administration Studies 29

Chronic LPS Administration Studies 30

Discussion 31

3. THE CYTOTOXICITY OF CHRONIC NEUROESIFLAMMATION UPON

BASAL FOREBRAIN CHOLINERGIC NEURONS OF RATS CAN BE

ATTENUATED BY GLUTAMATERGIC ANATAGONISM OR

CYCLOOXYGENASE-2 INfflBITION 44

Abstract 44

Introduction 45

Materials and Methods 48

Subjects 48

Surreal Procedures 48

Chronic LPS Administration 49

Drug Treatment 50

NMDA open channel antagonist - 37 day infusion length 50

COX2/Lipoxygenase inhibitor - 74 day infusion length 50

7

TABLE OF CONTENTS - Continued

Biochemistry 50

Histology 51

Results 53

Biochemical Studies 53

Histological Studies 54

Memantine 54

CI987 55

Camera Lucida 56

Radioimmunoassay 56

Silver 56

Discussion 57

4. SUMMARY AND DISCUSSION. 72

Inflammation and AD - From past to present 72

Time dependent role of inflammation in cholinergic cell death 77

Mechanisms of neurotoxicity and protection 78

Issues of selectivity 84

Conclusions 86

5. APPENDIX A: HUMAN/ANIMAL SUBJECTS 88

6. REFERENCES 89

8

LIST OF ILLUSTRATIONS

Figure 2.1 Effects of unilateral injections of LPS into the BF upon cortical levels of

ChAT activity 37

Figure 2.2 Camera lucida reconstruction of the location of ChAT-inununoreactive

cells within the NBM of rats given a single acute injection of LPS 39

Figure 2.3 The effects of chronic infusion of LPS or aCSF upon GFAP-positive

reactive astrocytes in the BF 41

Figure 2.4 OX-6-positive reactive microglia and p7S-immunoreactive neurons

increased within the NBM of rats following the chronic infusion of LPS or

aCSF 43

Figure 3.1 The effects of unilateral infusions of LPS into the BF on cortical levels of

ChAT activity 63

Figure 3.2 ChAT immunoreactivity or reactive microglia within the BF of rats

infused chronically with aCSF, LPS alone or LPS + memantine 65

Figure 3.3 ChAT immunoreactivity or reactive microglia within the BF of rats

infused chronically with aCSF, LPS alone or LPS + CI987 67

Figure 3.4 The number of OX-6 positive cells within the BF 69

Figure 3.5 Silver impregnation staining within the NBM region of the BF of rats

given aCSF for 74 days, or LPS for 37 days or 74 days 71

9

LIST OF TABLES

Table 1 Effects of acute unilateral lipopolysaccharide injections upon levels of

neuropeptides within the basal fordirain 35

Table 2 Effects of chronic 37- and 74-day unilateral lipopolysaccharide infusion

upon levels of neuropeptides within the basal forebrain 61

10

ABSTRACT

Alzheimer's disease (AD) is a disorder of aging that is characterized by

progressive loss of cognitive functioning, eventually culminating in complete dementia.

Many lines of evidence, i.e., epidemiological, pathological and biochemical, suggest

inflammatory processes play a role in neurodegenerative disease processes responsible

for AD. Recent reports suggest that patients who chronically take anti-inflammatory

drugs to treat the symptoms of inflanmiatory diseases, such as rheumatoid arthritis, have

a reduced risk of developing AD. The inflammation found in the brains of AD patients is

characterized by activated microglia and astrocytes in brain regions that demonstrate

extensive neuronal degeneration. In addition, the activated glia are closely associated

with amyloid plaques and may contribute to their formation. Various molecules involved

in inflanmiatoty reactions, including cytokines, complement, arachidonic acid and

prostaglandins are found to be elevated in the brain and CSF of AD patients. These

molecules may also contribute to the evolution of the pathology in this disorder.

The purpose of the first study was to determine if inflanmiatoty processes are

toxic to neurons. The results will show that both acute injections and chronic infusions of

the proinflammagen lipopolysaccharide imo the basal forebrain of rats was toxic to

cholinergic cells. Chronic infiisions of LPS decreased the level of activity of the

acetylcholine synthetic enzyme choline acetyhransferase (ChAT). bi addition, these

infosions decreased the number of neurons that demonstrated immunoieactivity for this

11

enzyme. Furthermore, low dose, long-term infusions of LPS produced a greater decline

in ChAT activity than a single injection of high dose LPS. These results suggest that the

toxic efifects of neuroinflammation are more time-dependent than dose-dependent and

support the hypothesis that long-term inflammatory processes can impair cholinergic

neuronal function.

The purpose of the second study was to examine the mechanism by which the

chronic LPS infusions were neurotoxic to cholinergic neurons within the basal forebrain

of young rats. This study investigated the hypothesis that increased activity of glutamate

and prostt^andin neurons might underlie the toxic actions of chronic inflammation

induced by LPS. First, this study investigated whether chronic administration of a non­

competitive N-methyl-D-aspartate (NMDA) sensitive receptor antagonist, memantine,

could provide neuroprotection for forebrain cholinergic neurons. Second, this study also

investigated whether a cyclooxygenase-2 (C0X2)/lipoxygenase inhibitor, CI987, could

provide significant neuroprotection from the cytotoxic effects of LPS-induced

neuroinflanunation. Daily peripheral administration of memantine (s.c.) or CI987 (s.c.)

significamly attenuated the cytotoxic effects of the chronic inflammatory processes upon

cholinergic cells within the basal foFd)rain. However, only CI987 attenuated the

neuroinflammation produced by LPS and the subsequent changes in microglial

activation. These results indicate that the ^otoxic effects of chronic neuroinflammation

may involve prostanoid synthesis and may operate through NMDA receptors, and that the

effects of prostaglandins occur upstream to NMDA receptor activation.

12

Charter One

INTRODUCTION

One of the most feared aspects of aging is dementia, the deterioration of memory,

and the decline of higher cognitive function. Dementia occurs among the elderly with

increasing frequency in advancing years. Alzheimer's disease (AD) is the prototype and

the most conmion cause of dementia; it accounts for (4)proximately 70% (Terry and

Katzman, 1983; Tomlinson et al., 1970) of the instances of dementia in the aged

population and contributes substantially to the dementia in another 15-20% of the general

population (Roses, 1994). While there are many elderly people that are mentally alert

and productive in their 7^, 8***, and even 9^ decade of life, a significant proportion of

otherwise healthy persons show a significant decline in mental function in later life. It

has been estimated that 10% of the population over 65 years old and 47% for those over

85 sufifers from AD (Pfeffer et al., 1987; Schoenbeig et al., 1987; Evans et al., 1989).

AD may affect as many as 4 million people in the US alone (Geldmacher and

Whitehouse, 1997). The total cost of patient care has recently been estimated at nearly

$90 billion annually (Ernst and Hay, 1994; Gualtieri et al., 1995). Because the proportion

of our population 80 and older will double by 2010, the financial and emotional burden of

AD for patients, families, caregivers, and society will increase to enormous proportions.

13

AD has an insidious onset and the cognitive changes are difficult to notice at first.

Short-term memory deficits, dyscalculia, constructional q>raxia characterize early stages

of AD. As the disease progresses, the patients lose other cognitive abilities and exhibit

dysphasia, agnosia, disorientation in time and place, severe mental confusion, personality

changes and behavioral disturbances. The patients cannot feed or dress themselves, get

lost only a short distance from home and do not recognize their own families. In

advanced stages of the disease, the patients are unresponsive, bedridden, incontinent,

unable to care for themselves, and ultimately do not even know their own names

(Hasegawa and Aoba, 1994).

Currently, the true diagnosis of AD must be confirmed at autopsy. A definitive

diagnosis requires a pathological examination of the brain to identify a defined set of

characteristic morphological features. Gross postmortem examination of the brains of

AD patients reveals a marked atrophy of the cortical gyri as compared to age-matched

non-demented controls. This tissue loss is thought to result from n^onal shrinkage and

loss of fibers and synapses. The classic histopathological features of AD are the presence

of abnormal structures called neuritic plaques and neurofibrillary tangles. Neuritic

plaques are so-called because th^ possess dystrophic and degenerating neuronal

processes along with reactive microglia and astrocytes. They also contain beta-amyloid

filamerns and a dense extracellular core made up largely of an insoluble amyloid deposit.

While many brain regions show pathology, brain regions associated with the medial

14

temporal lobe seem to be more affected. These structures include the entorhinal cortex,

uncal cortex, parahippocampal gyrus, hippocampus, and corticomedial amygdala.

In addition to the anatomical changes, there are also abnormalities in various

neurochemical systems. Deterioration in neural pathways involves loss of ascending

cholinergic projections from the basal forebrain, ascending serotonergic projections from

the dorsal raphe and ascending noradrenergic tracts from the locus coeruleus. While

many neurotransmitter systems are affected, the most striking and consistent

neurochemical disruption is seen in the cholinergic system in the basal forebrain.

The basal forebrain choluiergic system is located ventral to the anterior

commisure and globus pallidus and lateral to the hypothalamus. It consists of large

neurons that extend from the medial septum through the diagonal band of Broca to the

nucleus basalis of Meynert. The nucleus basalis of Meynert is larger in humans and

primates than to its homologous structure, the nucleus basalis magnocellularis (NBM),

found in other species such as rodents and is the major component of the substantia

innominata. Neurons situated more rostrally in the medial septum and the vertical limb

of the diagonal band predominately innervate the hippocampus and the cingulate cortex

respectively, while neurons of the NBM in the more caudal aspect of the BF project

widely to innervate the neocortex and amygdala.

15

The cholinergic deficit is one of the most clearly defined neurochemical

deficiencies that has been identified in the brains of AD patients. One of the earliest

findings of changes in the integrity of this system in AD was the report of a significant

decrease in choline acetyltransferase (ChAT) activity in the postmortem samples of tissue

from the brains of AD patients (Davies and Maloney, 1976). ChAT is a presynaptic

enzyme necessary for the synthesis of acetylcholine. It has also been used as a

biochemical marker for cholinergic neurons. The degree of the decline in ChAT activity

has been found to be correlated to the degree of imellectual impairment in AD.

Subsequently, others (Whitehouse et al., 1981) demonstrated a significant loss of

cholinergic neurons in the basal forebrain of AD patients. These and other findings led to

the concept that memory loss seen in AD might be attributable to deficiencies in

cholinergic function. This finding came to be known as the cholinergic hypothesis of AD

(Bartus et al., 1982). The cholinergic hypothesis was appealing because it offered a

schema analogous to the dopaminergic model of Parkinson's disease, and therefore

offered a basis for trials of neurotransmitter replacement therapy. This lead to the

development of pharmacotherapies aimed at augmenting the loss of cholinergic neurons

with the expectation that supplying precursors to acetylcholine synthesis might relieve

some of the behavioral symptoms. More recently, this approach has been used with the

expectation that cholinergic drug then4)y might actually halt the progression of the

disease. Thus cholinomimetic agents and cholinesterase inhibitors were used to promote

cholinergic transmitter function (Knapp et al., 1994; Mohr et al., 1994). Most of these

drug trials have had disappointing results (Wagstaf and McTavish, 1994). The major

limitation in the use of cholinergic precursors or cholinesterase inhibitors is that intact

cholinergic neurons are required to synthesize ACh.

16

It became increasingly evident that AD is not simply a condition of

hypocholinergic function. While disruption of central cholinergic function can produce

attention and memory deficits similar to that seen in AD, the symptoms of dementia are

due to more complicated neural changes than just cholinergic cell loss. It has become

increasingly clear that pathology exists in other neural systems and that these changes

might underlie aspects of the dementia. Even if one could compensate for the cholinergic

system deficiency and provide temporary clinical benefit for the AD patients, one would

still have to confront the problem of the cause of the loss of cholinergic neurons. In

contrast to the futile efforts to treat AD by enhancing cholinergic neuronal function, long­

term treatment with anti-inflammatory drugs has been shown to decrease the incidence

(Rich et al., 1995) and slow progression (Rogers et al., 1993) of AD. The apparent

positive benefit of long-term therapy with anti-inflammatories had led to the speculation

that the inflammatory changes first described by Alzheimer in 1907 (Alzheimer, 1907)

might be involved in the degeneration seen in this disorder.

Role of Inflammatory Processes in the Degeneration of Basal Forebrain Cells in AD

Chronic inflammatory processes may play an important role in the etiology or

pathogenesis of AD (Aisen and Davis, 1994; Frederickson and Brunden, 1994; McGeer

17

et al., 1989; Mirak et al., 1995; Rogers et al., 1992). Many biomarkers associated with

inflammatory activity are elevated in AD brains. For example, AD is associated with

increased levels of the inflammatory cytokine interleukin-1 (IL-1) in brain tissue (Grifiin

et al., 1989). Reactive gliosis, a conspicuous pathologic feature in AD, is a prominent

component of neuritic plaques (Delacourte, 1990; Griffin et al., 1989; Mancardi et al.,

1983; Mandybur and Chuirazzi, 1990; Perry et al., 1993a; Schechter et al., 1981).

Neuritic plaques contain both ^-amyloid and reactive glial cells that overexpress

inflanmiatory cytokines, including IL-1, IL-6, and tumor necrosis factor-a (Benveniste et

al., 1990, Chung and Benveniste, 1990; Frei et al., 1989; Sparacio et al., 1992, Mrak et

al., 1995). IL-1 promotes the gene expression of amyloid precursor protein (APP) in

astrocytes (Blume and Vitek, 1989) and microglia (Kreutzberg, 1996) and glia-producing

cytokines such as IL-1 appear to be involved in the generation of amyloid (Vandenabeeie

and Fiers, 1992). In addition, elevated IL-la can induce the expression of SIOOP (Griffin

et al., 1989; Sheng et al., 1996) and apolipoprotein E (Mrak et al., 1995) in astrocytes.

Therefore, reactive glia and elevations in inflammatory cytokines, such as IL-1, may

contribute to the pathology seen in AD (Frederickson, 1992; Frederickson and Brunden,

1994; Grififm et al., 1989).

The primary function of microglia and astrocytes is to protect the int^rity of the

brain during the developmental period of the nervous system and throughout the lifetime

of the healthy adult. They function as surveillance cells to maintain homeostasis and to

repair small traumas in acute destabilizing conditions. Microglia are involved in

18

producing neuronal survival and grouting fiictors and produce fiictors which influence

the brain's immune response. Astrocytes provide trophic and survival fiictors for neurons

and themselves produce fiutors which influence the proliferation and migration of

microglia. As protectors of the nervous system, microglia clear the brain of inducers of

inflammation such as apoptotic cells and opportunistic pathogens through phagocytosis

and pinocytosis, while astrocytes buffer concentrations of ions, excitatory amino acids

and biogenic amines in the extracellular space. However, activation of microglia and

astrocytes by viruses, chemicals, or trauma may result in a chronic series of interactions

among glia, neurons, blood-borne inflanmiatoiy cells, cytokines and neuropeptides that

can lead to a pathological environment and subsequent nervous system damage (Perry

and Gordon, 1997; Norenberg, 1997).

Cytokine-mediated processes produce a variety of responses in the brain that

could be neurotoxic (Frederickson, 1992; Griffin et al., 1989). Chronic, regional

inflanmiation can be produced experimentally by infusion of the proinflanmiagen

lipopolysaccharide (LPS), a component of the cell wall of Gram-negative bacteria. LPS

can stimulate the endogenous production of IL-1 and other inflammatory cytokines by

activated astrocytes and microglia (Bluthe et al., 1992; Quan et al., 1994). IL-1 can in

turn stimulate the production of a variety of other inflammatory mediators, such as

prostaglandins (Katsuura et al., 1989; Weidenfeld et al., 1992). Prostaglandins induce the

release of glutamate from astrocytes (Bezzi et al., 1998) leading to the stimulation of

ghitamate receptcvs, the depolarizatioii-dqiendent unblocking of NMDA receptors by

19

magnesium, and the entry of toxic amounts of calcium into neurons. Prostaglandins may

also indirectly elevate the extracellular concentration of glutamate by inhibiting its

reuptake by astrocytes (Robinson et al., 1993; Rothstein et al., 1993). Activation of

NMDA receptors could result in the influx of toxic amounts of calcium into the cell and

the release of nitric oxide that may be toxic to nearby neurons. The end result of excess

glutamatergic stimulation would be neuronal death. A potential cascade of biochemical

steps that might lead to cholinergic cell death:

LPS ~IL-l ~Prostaglandins~ t[Glutamate]ext ~ NMDArecpt ~ Cell Death

LPS was chosen to induce the proposed inflammatory cascade in the BF of young rats

because it is a potent proinflammagen, its mechanism of action is well known and its

effects have been well characterized centrally and peripherally (Espey et al., 1995; Fujii

et al., 1996).

Role of Prostaglandins

LPS ~IL-l~ Prostaglandins~ t[Glutamate]ext ~ NMD~~ Cell Death

Activated glia can release other cytokines and initiate an inflammatory cascade

which can be detrimental to cellular function (Bluthe et al., 1992; Quan et al., 1994).

During inflammatory processes in the brain, a marked rise of arachidonic acid and

20

increased synthesis of prostaglandins are noted (Gaudet et al., 1980; Bemheim et al.,

1980). bi a nonstimulated cell, arachidonic acid is esterified to membrane glycoUpids.

Upon stimulation, phosphoUpase M (PLAj) is activated in a Ca^ dependent manner and

frees arachidonic acid from the membrane. Once arachidonic acid is released it can

either serve its own messenger functions in certain pathways or be metabolized into

prostaglandins by the enzyme cyclooxygenase (CO^. There are two isoforms of the

COX enzyme. The constitutively expressed COX-1 acts in a variety of settings to

produce homeostatic or maintenance levels of prostaglandins while COX-2 is inducible

upon inflammatory stimulation. Therefore, prostaglandins may be involved in the brain's

inflammatory response. Evidence for this includes the increased gene expression for

PLA] in rat astrocytes by inflammatory mediators TNF and IL-1 as well as LPS (Oka and

Arita, 1991). Cultured astrocytes and microglia produce and release PGE2 in response to

stimulants such as LPS (Sawada et al., 1993). In addition, increased prostaglandin

production has been implicated in other dementia states such as HTV dementia (Grif!in et

al., 1994) and chronic use of anti-inflammatory treatments that inhibit the synthesis of

prostaglandins correlates with a decreased incidence of AD (McGeer et al., 1990).

Because this evidence implicates an important role of prostaglandins in inflammatory

mediated neurotoxicity, the ability of a prostaglandin synthesis inhibitor to provide

neuroprotection from the effects of LPS administration will be investigated.

21

Role of Glutamate and NMDA Receptors

LPS--+ IL-l--+ Prostaglandins--+ t[Giutamate]en--+ NMDArecpt--+ Cell Death

The cytotoxic mechanisms of inflammatory processes may be due in part to

excitatory amino acids. Astrocytes play a critical role in the extracellular buffering of

glutamate and other excitatory amino acids. Astrocytes may mitigate neuronal

excitotoxicity under normal conditions (Sugiyama et al., 1989). Neurons in astrocyte

poor cultures are more vulnerable to glutamate excitotoxicity (Rosenberg and Aizenman,

1989). Cholinergic neurons within the BF are vulnerable to excess stimulation of

glutamatergic NMDA receptors (Wenk et al., 1995) possibly due to the fact that they

receive a dense glutamatergic projection from the pedunculopontine tegmentum

(Rasmusson et al., 1994).

In culture, activated microglia can produce significant amounts of quinolinic acid

and other excitotoxins, which may destroy nerves cells through an NMDA-receptor

mediated pathway (Giulian et al., 1993; Lipton et al., 1994; Espey et al., 1995). This

mechanism may play a significant role in the induction of neuronal loss in AD.

lmmunostimulated glia release an NMDA-like substance (Boje and Arora, 1992; Chao et

al., 1992) that could activate excitotoxic pathways independent of extracellular buffering

mechanisms. In addition, both microglia and astrocytes have been suggested to play a

22

role in the generation of cognitive dysfiinction in HIV dementia by releasing cytokines

and activating NMDA receptors (Epstein and Gendelman, 1993; Lipton, 1992).

Microglia respond to P-amyloid by releasing IL-1 (Araujo and Cotman, 1992)

which can induce an increased release of arachidonic acid and prostaglandins from

neuronal and nonneuronal cells (Gaudet et al., 1980; Bemheim et a!., 1980). An

increased release of arachidonic acid has been shown to inhibit reuptake of glutamate and

other excitatory amino acids by astrocytes (Barbour et al., 1989; Volterra et al., 1992).

Arachidonic acid has also been reported to enhance NMDA-evoked currents (Miller et

al., 1992). Therefore, arachidonic acid could contribute to neurotoxicity not only by

inhibiting the reuptake of glutamate but also by increasing the effectiveness of glutamate

in evoking NMDA receptor-mediated processes. Arachidonic acid metabolites have also

been reported to have direct and indirect effects on glutamate buffering mechanisms by

astrocytes. Prostaglandins induce the release of glutamate from astrocytes (Bezzi et al.,

1998) and can indirectly elevate the extracellular concentration of glutamate by inhibiting

its reuptake by astrocytes (Robinson et al., 1993; Rothstein et al., 1993). Because this

evidence implicates a significant role of >]MDA receptors in inflammatory mediated

neurotoxicity, the ability of an NMDA receptor antagonist to provide neuroprotection

from the effects of LPS administratioii will be investigated.

23

Chapter Two

PATHOLOGICAL AND BIOCHEMICAL CONSEQUENCES OF ACUTE AND

CHRONIC NEUROINFLAMMATION WITHIN THE BASAL FOREBRAIN

CHOLINERGIC SYSTEM OF RATS

Abstract

Inflammatory processes may play a critical role in the degeneration of basal

forebrain cholinergic cells that underlies some of the cognitive impairments associated

with Alzheimer's Disease. In the present study, the proinflammagen lipopolysaccharide,

from the cell wall of Gram-negative bacteria, was used to produce inflammation within

the basal forebrain of rats. The effects of acute, high-dose injections of

lipopolysaccharide (2, 20 or 40 J..lg) upon basal forebrain chemistry and neuronal integrity

were compared with the effects of chronic, low-dose lipopolysaccharide infusions (0.9,

0.25, 9.0, or 25.0 J..lg/hr) for either 14, 37, 74 or 112 days. Acute exposure to

lipopolysaccharide decreased cortical choline acetyltransferase activity and the number of

immunoreactive choline acetyltransferase-positive cells within a small region of the basal

forebrain. Regional levels of five different neuropeptides were unchanged by acute,

high-dose lipopolysaccharide injections. Chronic lipopolysaccharide infusions produced

1) a primarily time-dependent, rather than dose-dependent, decrease in cortical choline

acetyltransferase activity that paralleled a decline in the number of choline

acetyltransferase- and p75-immunoreactive cells within the basal forebrain, and 2) a

dense distribution of reactive astrocytes and microglia within the basal forebrain.

24

Chronic neuroinflammation might underlie the genesis of some neuropathological

changes associated with normal aging or Alzheimer's Disease.

Introduction

A decline in the number, size and function of cholinergic cells in the basal

forebrain (BF) may be responsible for the cognitive impairments associated with aging

and Alzheimer's Disease in humans (McGeer et al., 1984; Rinne et al., 1987; Swaab et

al., 1994; Terry et al., 1991; Vogels et al., 1990). The neural mechanisms that underlie

the degeneration of these BF cholinergic cells are unknown. The current study

investigated the potential role of inflammation as a mechanism of cell loss in this brain

region. Both acute and chronic regional inflammation was produced by injection of the

proinflammagen lipopolysaccharide (LPS), a component of the cell wall of Gram­

negative bacteria. LPS can stimulate the endogenous production ofinterleukin-la. (IL­

Ia.), IL-1(3, IL-6 and tumor necrosis factor-a. (TNFa.) by both glia and neurons (Bluthe et

al., 1992; Quan et al., 1994), as well as f3-amyloid precursor protein (Brugg et al., 1995),

complement proteins (Morgan, 1990) and prostanoids such as PG~ (Katsuura et al.,

1989; Weindenfeld et al., 1992). LPS was chosen for the current study because our

preliminary studies had found that at very low doses it can produce a strong inflammatory

response by local glia cells. The present study compared the neurobiological effects of

acute, high-dose injections ofLPS into the BF of young rats with the effects of chronic,

low-dose infusions ofLPS for either 14, 37, 74 or 112 days.

Materials and Methods

Subjects

25

Eighty young (3 mo.) male Fisher-344 rats (Harlan Sprague-Dawley) were housed

in pairs in a colony room with a 12:12 dark:light cycle with lights off at 10:00 and food

and water provided ad libitum.

Surgical Procedures

Each rat was anesthetized with pentobarbital (50 mg!kg, i.p.), given 0.3 ml of

atropine methylbromide (5 mglml, i.p.), and placed in a stereotaxic instrument with the

incisor bar set 3 mm below the earbars (i.e. flat skull). The scalp was incised and

retracted and holes were drilled in appropriate locations in the skull with a dental drill.

For the acute administration studies, LPS (2, 20 or 40 J,.lg in 1.0 J.ll) was injected

unilaterally (left side) into the BF at these coordinates: AP: -0.6 mm, ML: +2.65 mm, and

DV: -7.9 mm. These rats were sacrificed two weeks later.

26

For the chronic administration studies, an Alzet (Palo Alto, CA) osmotic

minipump containing LPS (Sigma, St. Louis, MO, E. coU, serotype 026:B6, TCA

octraction) was implanted into the dorsal abdomen and attached via Tygon tubing (0.060"

O.D.) to a chronic indwelling cannula (Model 3280P, osmotic pump connect, 28 gauge.

Plastics One, Inc., Roanoke, VA) that had been positioned stereotaxically so that the

cannula tip extended into the BF (same coordinates as above). Controls were infused

with the artificial cerebrospinal fluid (aCSF) vehicle: (in mM) 140 NaCl; 3.0 KCl; 2.5

CaCb; 1.0 MgCb; 1.2 Na2HP04; pH 7.4. Two dififerent osmotic minipumps were used in

the chronic administration studies: model 2002 to deliver 2.5 ^1/hr for 14 days or model

2004 to deliver 0.2S ̂ 1/hr for 37 days (as estimated by information provided by Alzet).

The model 2002 osmotic minipumps were prefilled with either aCSF or LPS (to deliver

0.9 ̂ g/hr, 9.0 ̂ g/hr, or 25.0 ^g/hr). The model 2004 osmotic minipumps were prefilled

with either aCSF or LPS (to deliver 0.25 ̂ g/hr).

Thirty-seven days later, twelve rats that had been implanted with model 2004

osmotic minipumps were anesthetized with metophane and their indwelling minipumps

were replaced with identical pumps (i.e., model 2004) containing either LPS or aCSF.

Thirty-seven days later still, sue of these rats had their indwelling osmotic minipumps

replaced once again. Ultimately, therefore, rats were administered either LPS (0.25

l^/hr)oraCSFforatotalofeither37,74 or 112 days. A volume overload to the CSF

^)ace was discounted because the 0.5 ̂ l/h ̂administered contributed to only about 0.4%

of the CSF volume produced by the rat each hour and was only 0.25% of the rat's total

27

CSF volume ̂ bnisch et ai., 1997). For all rats, post-operative care included

chloramphenicol (1% solution) applied to the exposed skull and scalp prior to closure,

lidocaine applied locally to the scalp to lessen pain, and S.O ml of sterile isotonic saline

injected (s.c.) to prevent dehydration. All efforts were made to minimize discomfort or

pain to the rats and to reduce the number of rats required for these studies. The rats were

closely monitored during recovery and kept under a heat lamp or on a heating pad until

they were awake and active.

Biochemistry

Rats used for biochemical analyses were first lightly anesthetized and then

quickly decapitated. The nucleus basalis magnocellularis (NBM) within the BF was

isolated by two cuts made through the ventral pallidum and ventral substantia innominata

and a vertical cut made near the lateral border of the hypothalamus. This NBM sample

was rapidly frozen (-70*'C) and later analyzed for endogenous levels of galanin,

neurokinin B, neurotensin, neuropeptide Y and somatostatin by radioimmunoassay using

standard methods (Wenk et al., 1992; Wenk et al., 1994). Tissue samples were taken

bilaterally from the frontolateral sensorimotor cortex and assayed for choline

acetyltransferase (ChAT) activity (Fonnum, 1969). The ChAT enzyme is q)ecific for

cholinergic cells; its decline is used as a standard measure of cholinergic cell loss in the

NBM (Wenk et al., 1992; Wenk et al., 1994). This biochemical measure also provided

independent and quantitative confirmation of the immuno^ochemical studies, desoibed

28

below, ofthe relative number ofthese cholinergic cells. Protein content was determined

by a standard method (Lowry et al., 1951). ChAT enzymatic data were obtained from

every rat in all of the groups and were analyzed by ANOV A.

Histology

Rats used for the histological analyses were given a lethal injection of

pentobarbital and prepared by in situ transcardial perfusion of the brain with cold saline

containing heparin. Small sections of anterior cortex were first removed (for ChAT

enzymatic analysis) from each rat prior to being perfused with (filtered) 4%

paraformaldehyde in 0.1 M sodium phosphate buffer, pH 7.3. The brain was removed,

post-fixed in 4% paraformaldehyde/0.1 M sodium phosphate buffer for one hour, and

cryoprotected in 0.1 M sodium phosphate buffer containing 20% sucrose at 4°C for 24

hours, snap-frozen by transfer into isopentane ( -50°C) and stored ( -70°C). Blocks

including both hemispheres were cut (30 J.lm) on a freezing-sliding microtome.

Activated microglia and astrocytes rapidly upregulate their expression of tissue

antigens following inflammation (Finsen et al., 1993; Perry et al., 1993b) and can release

potentially toxic levels of cytokines (Finsen et al., 1993). To visualize activated

microglia, a monoclonal antibody (OX-6, 1/100 dilution, PharMingen, San Diego, CA)

directed against the Class II major histocompatibility complex was used (Flaris et al.,

1993; Perry et al., 1993b). Activated astrocytes were visualized using a polyclonal

29

antibody against GFAP (1/100 dilution, Sigma). Sections that included the NBM were

also immunostained using a polyclonal antibody against ChAT to label cholinergic cells

(1/100 dilution, Chemicon, Temecula, CA), and p75-positive (1/100 dilution, Chemicon)

cells. NBM cholinergic cells express high levels of the low-affinity p75 nerve growth

factor receptor (Yan and Johnson, 1989). The destruction ofNBM cholinergic cells is

associated with the loss ofboth ChAT- and p75- immunoreactive staining (Berger­

Sweeney et al., 1994; Wenk et al., 1992; Wenk et al., 1994).

Results

Acute Administration Studies

A single injection ofLPS significantly [F(3,21)=11.9, p<0.001] decreased cortical

ChAT activity at the two higher doses (20 or 40 J.Lg , see Fig 2.1 ). The 40 J.Lg dose did not

produce a significantly greater decline in cortical ChAT activity than a 20 J.Lg dose.

Immunocytochemical staining for ChAT -positive cells found a decrease in cell number

(Fig. 2.2) near the site of the injection (20 or 40 J.1.8 LPS), as compared to the effects of an

aCSF injection. Radioimmunoassays of the NBM tissue samples taken from rats injected

with either 20 or 40 J.1.8 ofLPS did not find significant changes in endogenous levels of

galanin, neurokinin B, neurotensin, neuropeptide Y or somatostatin (Table 2.1 ).

30

^r«yniy LPS AdmiiiMtratioB Studica

The chronic infusion of LPS into the BF did not induce seizures or cause any

significant changes in the rats' health. After an initial decline following surgery, all rats

gained weight normally during the infusion period. Although chronic infusion of LPS for

14 days did not significantly alter cortical ChAT activity, an ANOVA of all of the ChAT

data found a significant main effect of duration of infusion [F(io4i)=6.77, p<0.001], no

main effect of dose and no significant interaction. Chronic infusion of LPS (0.25 ̂ g/hr)

for 37, 74 or 112 days significantly (all p's<O.OS) decreased cortical ChAT activity (Fig.

2.1), as compared to the effects of acute LPS injections or a 14-day chronic LPS

injection. Infusion of LPS for 112 days produced a significantly greater decline (p<O.OS)

in cortical ChAT activity than did the 37- or 74-day LPS infusions.

Chronic infusion of LPS for 14 days (9.0 ̂ ^hr), 37, 74 or 112 days (0.2S ug/hr)

produced a dense distribution of GFAP-positive reactive astrocytes (Fig 2.3 A, B).

Chronic infusion of LPS for 37,74 or 112 days (0.2S ̂ g/hr) produced a decline in the

number of ChAT-immunoreactive neurons (Fig 2.3 C, D) throughout the NBM. In rats

exposed to LPS for 112 days, GFAP-positive reactive astrocytes were also seen scattered

throughout the BF region, as well as in the overlying neocortex and hippocampus.

Chronic infusion of LPS for 14,37, or 112 days produced a dense distribution of OX-6-

positive reactive microglia within the BF and throughout nearby brain r^ons (Fig 2.4

A), as compared to rats chronically infused with aCSF (Fig 2.4 B). Chronic infusion of

LPS (0.25 J.Lg/hr) for 112 days also produced a decline in the number ofp75-

immunoreactive neurons (Fig 2.4 C) within the NBM, as compared to rats chronically

infused with aCSF (Fig 2.4 D). In contrast, chronic infusion ofaCSF for 112 days

produced only a modest increase in gliosis near the site of the injection.

Discussion

31

This study demonstrated that chronic inflammation produced a primarily time­

dependent, rather than dose-dependent, cholinergic cell death within the BF. The LPS

infusions produced an extensive activation of astrocytes and microglia following either

acute or chronic exposures. Activated glia are the main source of IL-l a and IL-l~ after

an injection ofLPS; IL-Ia can in tum activate astrocytes to produce a variety of other

interleukins as well as TNFa, ~-amyloid precursor and complement proteins (Bluthe et

al., 1992; Brugg et al., 1995; Morgan, 1990; Quan et al., 1994). The microglia-derived

cytokines can induce the release of prostaglandins (Morgan, 1990), which in tum

elevates, directly or indirectly, the extracellular concentration of glutamate (Frederickson

and Brunden, 1994; Katsuura et al., 1989). This would lead to the stimulation of

glutamate receptors, the depolarization-dependent unblocking ofN-methyl-D-aspartate­

sensitive glutamate receptors by magnesium, and the entry of toxic amounts of calcium

into the neurons (Mrak et al., 1995; Relton and Rothwell, 1992; Rothwell et al., 1996;

Skaper et al., 1995). Widespread increased activation ofN-methyl-D-aspartate-sensitive

glutamate receptors would also lead to the release of cytotoxic amounts of nitric oxide

and to the death of nearby neurons (Rothwell, et al., 1996). Nitric oxide and its

metabolites are toxic molecules that can impair respiratory enzymes (Brosnan et al.,

1994; Dn^)ier and Hibbs, 1988), oxidize proteins and nitrosylate nucleic acids (Radi et

al., 1991). There is recent evidence that LPS can induce a calcium-dependent form of

nitric oxide synthase in astroglial and macrophage cultures (Swierkosz et al., 1995).

Consistent with these findings is the report that inflammation-related neurotoxicity can be

attenuated by nitric oxide synthase inhibitors (Rothwell et al., 1996). Therefore, various

cytokine-, glutamatergic- and nitric oxide-dependent mechanisms might all underiie the

destruction of NBM cholinergic neurons in the brains of rats in the present study.

The results of our histological and biochemical studies are consistent with the

hypothesis that the cytotoxic effects of chronic neuroinflammation are selective for

cholinergic cells, at least within the BF region. Similarly, an immune-mediated process

was shown to selectively reduce septal ChAT activity and decrease hippocampal

acetylcholine release following the injection of hybridized cholinergic cells into the

medial septal area of guinea pigs (K^min et al., 1997). A recent study also isolated

antibodies from the sera of Alzheimer's Disease patients that selectively recognized BF

cholinergic cell groups in the rat brain and impaired cholinergic function (Foley et al.,

1988). It should be noted that a decline in cortical ChAT activity is consistent with the

death of BF cholinergic cells but does not refiite the possibility that these cholinergic

cells have stopped expressing the ChAT proteiiL In addition to LPS, our preliminary

studies have shown that TNFa and IL-1 also reduce cortical ChAT activity (uiqMibUshed

33

observations) when injected into the BF. Both TNFa and LPS can induce astrocytes to

mediate cell death in cocultures of cerebellar and hippocampal neurons (Skaper et al,

1995). Interestingly, the TNFa receptor has a high degree of sequence homology with

the low affinity p75 nerve growth factor receptor that is expressed by all NBM

cholinergic cells. Under conditions of enhanced release TNFa might bind to the low

affinity nerve growth factor receptor and induce apoptosis in cholinergic cells (Van

Antwerp et al., 1996; VanderZee et al., 1996). Therefore, in the presence of a chronic

inflammatory state associated with elevated levels ofTNFa, the selective expression of

the p75 receptor upon NBM cholinergic cells might underlie their apparent selective

vulnerability.

The release of cytokines from activated glia might initiate a self-propagating

series of steps that involve the production of ~-amyloid, apolipoprotein E, and

S100~ (Goldgaber et al., 1989; Mrak et al., 1995; Sheng et al., 1996). These steps are

followed by the ~-amyloid-induced stimulation of microglia to release additional

interleukins, complement factors (Mrak et al., 1995; Rogers et al., 1992) and a potentially

neurotoxic phenolic amine (Giulian et al., 1995). The chronic presence of elevated levels

of cytokines, associated with increases in intracellular calcium and extracellular nitric

oxide, may facilitate neurodegenerative changes throughout the brain.

In conclusion, in AD, the brain develops a complicated pattern of changes that

include, but are not limited to, the degeneration, atrophy or dysfunction of specific neural

34

systems that influence learning, memory and attention (Swaab et al., 1994; Terry et al.,

1991). Recent evidence suggests that inflammatory processes play an important role in

the pathogenesis of AD-related pathology (Aisen and Davis, 1994, McGeer et al., 1984,

Mrak et al., 1995, Rogers et al., 1992). A complete understanding of the mechanisms

that underlie NBM c~olinergic cell death in Alzheimer's Disease will lead to effective

pharmacotherapies to prevent this neuronal degeneration. Recent studies have provided

evidence that chronic anti-inflammatory therapy can provide neuroprotection from the

degenerative processes that lead to Alzheimer's Disease (Aisen and Davis, 1994; Breitner

et al., 1994; Rogers et al., 1993).

35

Table 1 Effects of Acute Unilateral Lipopolysaccharide Injection Upon Levels of

Neuropeptides within the Basal Forebrain

20 JJ.g LPS

Galanin

N eurotensin

Somatostatin

Neuropeptide Y

Neurokinin B

40 JJ.g LPS

Galanin

Neurotensin

Somatostatin

Neuropeptide Y

Neurokinin B

Right Side Left Side

13.8±5.4

6.0±0.9

3.9±0.9

36.4±6.4

20.3±5.2

13.7±3.4

6.7±0.6

2.6±0.3

26.4±1.8

16.3±3.4

11.9±2.3

5.8±1.1

3.8±0.7

32.7±7.4

20.6±5.6

11.4±2.8

6.3±0.8

2.5±0.7

27.6±4.5

15.5±2.3

Values are Mean±SD nanograms/mg protein. Lipopolysaccharide was injected into the

left basal forebrain of all rats. The lipopolysaccharide injections did not significantly

alter any of these neuropeptide levels, as compared to the control side.

36

Fig. 2.1. Effects of unilateral injections of LPS into the BF upon coitical levels of ChAT

activity. Acute administration of high doses of LPS (20 ̂ g & 40 |ig) significantly

(p<0.001) decreased cortical ChAT activity two weeks after the injection. Chronic

administration of much lower doses of LPS produced a time-dependent, but not dose-

dependent, decrease in cortical ChAT activity. I>=Days of infusion, chronic doses are

expressed as ^g of LPS that were infused per hour.

~ ·-.2::: 35 .... C.) c( 1- =- 30 <C~ .c ... 0 5 25 n;o -~ LL 20 t::(/) cO 0 ftS 15 C I

·- c CD.2 .5 tn 10 -CD (J_J Cl) ......... c 5 ~ 0

0

37

Acute LPS

38

Fig. 2.2. Camera lucida reconstruction of the location of ChAT-immunoreactive cells

within the NBM of rats given a single acute injection of LPS (20 \x%/\ ^1). The number

of cholinergic cells are decreased on the side of the injection (right side), as compared to

the control side (left side). Precise quantitative estimation of cholinergic cell loss was

provided by the ChAT biochemical assay. Each dot represents a single ChAT-positive

cell. ChAT-immunoreactive cells within the cortex are not shown. The cholinergic cell

loss was almost complete within the sight of the injection (between the arrowheads). The

needle artifact can be seen rising just above this site. The section shown is located 0.8

mm posterior to Bregma.

39

• • • • • . • .

• • . • • •• .. :.• . . .

• .. • • • •••

• I ~· .... . . I

. ..... . .. : •

. . . . .. •

. , ... • fi

!• • .,, I

• •

. - .... , , ..

• •• ·• • ·.f. • . . -· '

•

'~· .. , .· •

··~~·· . -.~ ~· -.· .... • iJ

•

•

40

Fig. 2.3. The effects of chronic infiision of LPS (A, 9.0 ̂ ^hr x 14 days) or aCSF (B)

upon GFAP-positive reactive astrocytes in the BF. Chronic LPS administration, for

either 14, 37 or 112 days, equally increased the number and level of activation of local

astrocytes. Injection of LPS (C, 9.0 ̂ g/hr x 37 days), but not aCSF (D), produced a

decline in the number, and staining intensity, of ChAT-immunoreactive magnocellular

neurons throughout the NBM. Magnifications are 2SX.

41

42

Fig. 2.4. OX-6-positive reactive microglia increased within the NBM of rats following

the chronic infusion of LPS (A, 0.25 ̂ g/hr, 112 days), but not aCSF (B). p7S-

immunoreactive neurons decreased in number within the NBM of rats chronically

infused with LPS (C, 0.2S ̂ g/hr, 112 days), but not aCSF (D). Magnifications are 16X

43

44

Chapter Three

THE CYTOTOXICITY OF CHRONIC NEUROINFLAMMATION UPON BASAL

FOREBRAIN CHOLINERGIC NEURONS OF RATS CAN BE ATTENUATED

BY GLUTAMATERGIC ANATAGONISM OR CYCLOOXYGENASE-2

INHIBmON

Abstract

Inflammatory processes may play an important role in the pathogenesis of the

degeneration of basal forebrain cholinergic cells that underlies some of the cognitive

impairments associated with Alzheimer's Disease. In the present study, the

proinflammagen lipopolysaccharide (LPS) was infused into the basal forebrain of rats.

The current study determined whether the chronic administration of either a non­

competitive N-methyl-D-aspartate (NMDA) receptor antagonist, memantine, or a

cyclooxygenase-2 (COX2)/lipoxygenase inhibitor, CI987, could provide significant

neuroprotection from the cytotoxic effects ofLPS-induced neuroinflammation. Chronic

LPS infusions decreased cortical choline acetyltransferase activity that paralleled a

decline in the number of choline acetyltransferase immunoreactive-cells within the basal

forebrain as well as the number of activated resident microglia. The infusions appeared

to be selective for cholinergic neurons. Peripheral administration ofmemantine (s.c.) or

CI987 (s.c.) significantly attenuated the cytotoxic effects of the chronic inflammatory

processes upon cholinergic cells within the basal forebrain. However, only CI987

attenuated the neuroinflammation produced by LPS and the subsequent changes in

microglial activation. These results indicate that the cytotoxic effects of chronic

neuroinflammation may involve prostanoid synthesis and may operate through NMDA

receptors, and that the effects of prostaglandins occur upstream to NMDA receptor

activation.

Introduction

45

Alzheimer's disease (AD) is a progressive, age-related neurologic disorder that

eventually leads to dementia and death (Rogers, 1995). AD is characterized by a distinct

pattern of neuropathological lesions, such as J3-amyloid plaques, neurofibrillary tangles,

cortical deficiency in acetylcholine, and a dense distribution of activated astrocytes and

microglia (Frederickson, 1992; Farlow, 1998). A decline in the number of cholinergic

cells in the basal forebrain (BF) may underlie the cognitive impairments associated with

AD (McGeer et al., 1984; Muir, 1997; Rinne et al., 1987; Swaab, et al., 1994; Terry et al.,

1991; Vogels et al., 1990). The mechanisms underlying the degeneration ofBF

cholinergic cells are unknown.

Chronic inflammatory processes may play an important role in the etiology or

pathogenesis of AD (Aisen and Davis, 1994; Frederickson and Brunden, 1994; Griffin et

al., 1998; McGeer et al., 1989; Mrak et al., 1995; Rogers et al., 1992). Many biomarkers

46

associated with inflammatory activity are elevated in AD brains. For example, AD is

associated with increased levels of inflammatory cytokines, such as interleukin-1 (IL-l),

several acute-phase proteins (ll..-6, al-antichymotrypsin, a2-macroglobulin), and

complement proteins C3 and C4 that are not seen in the brain tissue of age-matched

controls (Griffin et al., 1989; Bauer et al., 1991; McGeer and McGeer, 1998). Reactive

gliosis is a prominent component of neuritic plaques (Delacourte, 1990; Griffin et al.,

1989; Mancardi et al., 1983; Mandybur and Chuirazzi, 1990; Schechter et al., 1981).

Reactive glia and elevations in inflammatory cytokines may contribute to the pathology

seen in AD (Frederickson, 1992; Frederickson and Brunden, 1994; Griffin et al., 1989;

Griffin et al., 1998).

Chronic, regional inflammation can be produced experimentally by infusion of

the proinflammagen lipopolysaccharide (LPS), a component of the cell wall of Gram­

negative bacteria. LPS can stimulate the endogenous production of IL-l and other

inflammatory cytokines by activated astrocytes and microglia (Bluthe et al., 1992; Quan

et al., 1994). IL-l can in tum stimulate the production of a variety of other inflammatory

mediators such as prostaglandins (Fiebich et al., 1996; Katsuura et al., 1989; Minghetti

and Levi, 1995; Weidenfeld et al., 1992). Prostaglandins induce the release of glutamate

from astrocytes (Bezzi et al., 1998) leading to the stimulation of glutamate receptors, the

depolarization-dependent unblocking ofNMDA receptors by magnesium, and the entry

of toxic amounts of calcium into neurons. Prostaglandins may also indirectly elevate the

extracellular concentration of glutamate by inhibiting its reuptake by astrocytes

47

(Robinson et al., 1993; Rothstein et al., 1993). Cholinergic neurons within the BF are

vulnerable to excess stimulation of glutamatergic NMDA receptors (Wenk et al., 1995).

The end result of excess glutamatergic stimulation would be neuronal death. A potential

cascade of biochemical steps that might lead to cholinergic cell death:

LPS ~IL-l~ Prostaglandins~ t[Glutamate]ext ~ NMDAcecpt~ Cell Death

Previous studies from our lab have shown that a chronic, low dose, LPS infusion

can destroy cholinergic cells (Willard et al., 1999). The present study was designed to

pharmacologically block the cascade at different points. Since inflammatory mediators

are toxic to cholinergic cells, then inhibiting their production should provide

neuroprotection. If cytokine toxicity due to increased prostaglandin synthesis or excess

excitatory stimulation through NMD A channels, then the administration of an antagonist

at either of these steps should provide neuroprotection to cholinergic cells. In addition, if

the order of the cascade is correct, then the administration of a prostaglandin synthesis

inhibitor should also decrease the inflammatory response brought about by cytokines

while administration of an NMDA receptor antagonist should not interfere with this

response. We used the non-competitive NMDA receptor antagonist, memantine, and the

selective cyclooxygenase-2/lipoxygenase inhibitor, CI987, to address this issue.

48

Materials and Methods

Subjects

Sixty-nine young (3 mo.; 250g at beginning of experiment) male Fisher-344 rats

(Harlan Sprague-Dawley) were housed in pairs in a colony room with a 12:12 dark:light

cycle with lights off at 10:00 and food and water provided ad libitum.

Surgical Procedures

Each rat was anesthetized with pentobarbital (50 mg/kg, i.p.), given 0.3 ml of

atropine methylbromide (5 mglml, i.p.), and placed in a stereotaxic instrument with the

incisor bar set 3 mm below the earbars (i.e. flat skull). The scalp was incised and

retracted and holes were drilled in appropriate locations in the skull with a dental drill.

For all rats, post-operative care included chloramphenicol (1% solution) applied to the

exposed skull and scalp prior to closure, lidocaine applied locally to the scalp to lessen

pain, and 5.0 ml of sterile isotonic saline injected (s.c.) to prevent dehydration. All

efforts were made to minimize discomfort or pain to the rats. The rats were closely

monitored during recovery and kept under a heat lamp or on a heating pad until they were

awake and active.

49

Chronic LPS Administration

For the chronic administration ofLPS (n=41), an Alzet (Palo Alto, CA) osmotic

minipump (model2004; 0.25 JJ.]/hr) containing LPS (Sigma, St. Louis, MO, E. coli,

serotype 055:B5, TCA extraction, 1.0 J..lg/JJl) was implanted into the dorsal abdomen and

attached via Tygon tubing (0.060" O.D.) to a chronic indwelling cannula (Model3280P,

osmotic pump connect, 28 gauge, Plastics One, Inc., Roanoke, VA) that had been

positioned stereotaxically so that the cannula tip extended into the BF at the following

coordinates: AP: -0.6 mm, ML + 2.8 mm, and DV: -7.9 mm (from the skull). Controls

(n=28) were infused with artificial cerebrospinal fluid (aCSF) vehicle: (in mM) 140

NaCl; 3.0 KCl; 2.5 CaCh; 1.0 MgCh; 1.2 Na2HP04; pH 7.4. The 2004 minipump

delivered 0.25 J.1]/hr of either LPS or aCSF for 37 days (as estimated by information

provided by Alzet). After the infusion was complete, thirty-seven rats were anesthetized

with metophane and their indwelling minipumps were replaced with identical pumps

containing either LPS (n=21) or aCSF (n=16) for an additional37 days (74 day total

infusion length) as done previously in this lab (Willard et al., 1999). A volume overload

to the CSF space was discounted because the 0.25 J,J.llhr administered contributed only

about 0.4% of the CSF volume produced by the rat each hour and only 0.25% of the rat's

total CSF volume (Hanisch et al., 1997). During the infusion period, each animal's body

weight was determined daily and their general behavior was monitored for seizures or

indications of drug toxicity.

Drug Treatment

NMDA open channel antagonist - 37 day infusion length

Eight rats that were given LPS for 37 days were also implanted (i.p.) with an

additional minipump (model2ML4) that administered memantine (20 mg!kglday).

COX2/Iipoxygenase inhibitor- 74 day infusion length

Eight rats that were given LPS for 74 days were injected daily (s.c.) with Cl987

(30 mg/kglday) for 74 days. Six rats with aCSF infusions were also given daily

injections (s.c.) ofCI987 (30 mg/kg/day) as a control.

Biochemistry

50

Rats to be used for biochemical analysis were deeply anesthetized and sacrificed

by decapitation. Tissue samples were taken from right and left anterior sensorimotor

cortex and assayed for ChAT activity. ChAT activity in the cortex was measured by the

formation of[14C]-acetylcholine formed from [I_14C]-acetyl-coenzyme-A and choline

(Fonnum, 1969). The ChAT enzyme is specific for cholinergic cells; its decline is used

as a standard measure of cholinergic cell loss in the BF (Wenk et al., 1992; Wenk et al.,

1994). This biochemical measure also provided independent and quantitative

51

confirmation of the immunocytochemical studies, described below, on the relative

number of these cholinergic cells. Protein content was determined for each sample by a

standard method (Lowry et al., 1951). All assays were performed in triplicate. ChAT

enzymatic data were analyzed by ANOV A using Student-Newman-Keuls post-hoc

analyses.

In order to determine an indication of the degree of specificity ofLPS-induced

inflammation, tissue punches of the nucleus basalis magnocellularis (NBM) within the

BF was isolated for radioimmunoassays (RIAs). The BFINBM region was dissected by

making three scalpel cuts: a vertical cut near the lateral border of the hypothalamus

(inferior from the lateral ventricle), a horizontal cut made through the ventral pallidum

just above the level of the anterior commisure and a diagonal cut to remove the most

ventral edge of cortex. Frozen tissue samples were sonicated in 500 ,.U of ice-cold l.OM

HCl and centrifuged at lO,OOOxg for 10 minutes (4 °C). Aliquots of the supernatant were

frozen ( -70 °C), lyophilized and assayed for specific neuropeptides including galanin,

leu-enkephalin, neurokinin B, neurotensin, somatostatin and substance P using kits

obtained from Peninsula Laboratories (San Carlos, CA).

Histology

Each rat was deeply anesthetized with a combination of metophane and an

injection of pentobarbital until pain reflexes were absent. Each rat was then prepared for

52

analysis by in situ perfusion of the brain with cold saline containing heparin. Next, the

brain was perfused with filtered 4% paraformaldehyde in 0.1 M sodium phosphate buffer,

pH 7.4. The brains were then post-fixed in 4% paraformaldehyde/0.1 M sodium

phosphate buffer overnight, cryoprotected in 0.1 M sodium phosphate buffer containing

20% sucrose at 4°C for 24 hours, snap-frozen by transfer into isopentane (-50°C), and

stored at -70°C. Coronal sections (40 J.Uil) that included the entire extent of the BF of

both hemispheres were cut on a cryostat for histological analysis. The location of

immunoreactive or histochemically-defined neurons was examined by light microscopy.

To label cholinergic cells, sections were immunostained using a polyclonal antibody

against ChAT (1/1 00 dilution, Chemicon, Temecula, CA). Activated microglia and

astrocytes rapidly upregulate their expression of tissue antigens following inflammation

(Finsen et al., 1993; Perry et al., 1993b). To visualize activated microglia, a monoclonal

antibody (OX-6, 1/400 dilution, PharMingen, San Diego, CA) directed against the Class

IT major histocompatibility complex (MHC IT) was used (Flaris et al., 1993; Perry et al.,

1993b). The location ofOX-6 positive cells was plotted by camera Iucida. The number

ofOX-6 positive cells in the BF of 14 rats was counted using a comparable region from

each rat as defined by the presence of the anterior commisure. In addition, sections were

examined for evidence of necrotic cell death using a standard silver impregnation stain

(Colicos et al., 1996)

53

Rcsulti

All of the rats gained weight normally during the drug treatment period. None of

the rats was observed to have seizure activity or demonstrate any other indication of

toxicity or poor health.

Biochemical Studies

Chronic infusion of LPS into a BF region for 37- or 74-days significantly

decreased cortical ChAT activity on the side ipsilateral to the infusion [F4^4==8.46,

P<0.0001]. Chronic administration (s.c. via osmotic minipump) of memantine provided

significant (P<O.OS) neuroprotection from the effects of LPS upon cholinergic cells.

Likewise, chronic administration of CI987 (s.c., daily injection) provided significant

(P<0.05) neuroprotection from the effects of LPS upon cholinergic cells. (See Figure

3.1).

54

Histological Studies

Memantine

Immunohistochemical staining for ChAT within the BF revealed a normal

distribution of immunoreactive cells in the BF of control rats and in BF contralateral to

the infusion side (Figure 3.2 A). Rats infused with LPS had a total loss of ChAT­

immunoreactive cells near the injection site (Figure 3.2 B). The tissue was cavitated and

numerous injection artifacts were also visible near the injection site (see arrows in 3.2 B).

In rats treated with LPS and memantine (Figure 3.2 C), numerous ChAT-immunoreactive

cells were observed in the BF region ventral to the injection site. However, no

immunoreactive cells were observed within 0.5 mm from the tip of the injection cannula.

Immunohistochemical staining for the MHC TI (OX-6 antibody) revealed that the

microglia within the BF developed the greatest inflammatory response following chronic

exposure to LPS, however, activated microglia were observed within the entire coronal

section of the brain. Rats treated with aCSF had very few reactive microglia (Figure 3.2

D). In the absence of activated microglia, the antibody shows cross-reactivity with local

neurons. In LPS treated rats, the microglia were characterized by a contraction of their

highly ramified processes giving them a bushy morphology (Figure 3.2 E). Rats treated

with LPS and memantine did not appear to have fewer nor less reactive microglia as

compared to the LPS-treated rats (Figure 3.2 F).

55

Immunohistochemical staining for ChAT within the BF also revealed a normal

distribution of immunoreactive cells in the BF of control rats and in BF contralateral to

the injection side (Figure 3.3 A). Rats infused with LPS had an almost total loss of

ChAT-immunoreactive cells near the infusion site (Figure 3.3 B). In rats treated with LPS

and CI987 (Figure 3.3 C), numerous ChAT-immunoreactive cells were observed in the

BF region ventral to the infusion site.

Immunohistochemical staining for the MHC IT (OX-6 antibody) revealed that the

microglia within the BF developed the greatest inflammatory response following chronic

exposure to LPS. Rats treated with aCSF had very few reactive microglia (Figure 3.3 D).

In LPS treated rats, the microglia were numerous and characterized by the contraction of

their highly ramified processes giving them a bushy morphology (Figure 3.3 E). Rats

treated with LPS and CI987 appeared to have fewer and less reactive microglia as

compared to the LPS-treated rats (Figure 3.3 F).

56

Camera Lucida

The number ofOX-6 positive cells within the BF was significantly [Fs,s=12.0,

P=O.Ol5] attenuated by treatment with Cl987, the COX-2 inhibitor, but not by memantine

(See Figure 3.4). This result quantifies the apparent lack of microglia activation in the

LPS + CI987 treated animals versus the LPS + memantine treated animals.

Radioimmunoassay

Radioimmunoassays of the NBM tissue samples taken from rats infused with LPS

for either 37 or 74 days did not find significant changes in endogenous levels of galanin,

leu-enkephalin, neurokinin B, neurotensin, somatostatin and substance P (See Table 3.1)

Within the BF region of37- and 74-day LPS treated rats, there were many large,

dark, densely silver-stained neurons suggestive of an active necrotic process. The

morphology ofthese cell bodies is consistent with cholinergic neurons. In contrast, the

BF region ofaCSF-treated rats contained very few·silver-stained cells (See Figure 3.5).

57

Discussion

In the present study, infusion ofLPS into the BF of young rats produced extensive

inflammation and a significant loss of cholinergic cells. These results are consistent with

our previous findings, i.e. both acute and chronic infusions ofLPS destroyed cholinergic

cells and activated astrocytes and microglia (Willard et al., 1999). The current study

extended our understanding of the mechanisms by which chronic neuroinflammation

destroys BF cholinergic cells. The present study determined that peripheral

administration of memantine or CI987 significantly attenuated the cytotoxic

consequences of chronic neuroinflammation upon basal forebrain cholinergic cells.

However, only CI987 attenuated the glial response to the neuroinflammation produced by

LPS. The results of our studies are consistent with the hypothesized role for both

prostaglandins and NMD A receptors in the cascade of biochemical processes that lead to

BF cholinergic cell loss in AD. These results are also consistent with the proposed order

of the cascade presented above.

Inflammatory processes that activate glia can lead to the release of other cytokines

and initiate an inflammatory cascade which can be ·detrimental to cellular function

(Bluthe et al., 1992; Quan et al., 1994). IL-l is a principal participant in the

inflammatory reaction and it can further stimulate the induction of other inflammatory

molecules such as prostaglandins and cytokines (Gaudet et al., 1980; Bernheim et al.,

58

1980). During inflammatory processes in the brain, a marked rise of arachidonic acid and

increased synthesis of prostaglandins are noted (Oka and Arita, 1991). Increased

prostaglandin production has been implicated in other dementia states that are associated

with neuroinflammation, such as HIV dementia (Griffin et al., 1994)

Increased levels of prostaglandins can directly and indirectly increase the level of

glutamate in the extracellular space (Bezzi et al., 1998; Robinson et al., 1993; Rothstein

et al., 1993). Increased levels of glutamate will then lead to the stimulation of glutamate

receptors, the depolarization-dependent unblocking ofNMDA receptors by magnesium,

and the entry of toxic amounts of calcium into neurons. Recently, the cytokine IL-6 was

shown to increase NMDA-mediated calcium influx and potentiate its neurotoxicity (Qiu

et al., 1998). Likewise, both TNFa. and LPS can induce astrocytes to mediate cell death

in co-cultures of cerebellar and hippocampal neurons (Skaper et al., 1995). The astrocytic

involvement in glutamate metabolism has important implications in states of

excitotoxicity. For example, impaired astrocytic control of glutamate uptake may

underlie certain aspects of neuronal excitotoxicity (Sugiyama et al., 1989). Furthermore,

neurons in astrocyte-poor cultures are more vulnerable to glutamate excitotoxicity

(Rosenberg and Aizenman, 1989) and blocking astrocytic uptake of glutamate results in

neurodegeneration (Robinson et al., 1993; Rothstein et al., 1993). Cholinergic neurons

within the BF receive a dense glutamatergic projection from the pedunculopontine

tegmentum (Rasmusson et al., 1994) thus are potentially vulnerable to excess stimulation

of glutamatergic NMDA receptors (Wenk et al., 1995). Therefore, the end result of

59

excess prostaglandin induction might be increased activation of glutamate receptors and

neuronal death.

The results of our histological and biochemical analyses are consistent with the

hypothesis that the cytotoxic effects of chronic neuroinflammation within the BF are

selective for cholinergic neurons. The radioimmunoassays found no significant

difference in the endogenous level of various neuropeptides within the BF. The

histological analyses found a dense distribution of silver-stained cells within the BF that

had a general morphology and size that was consistent with cholinergic magnocellular

neurons. When considered together with the decrease in neocortical ChAT activity and

the reduced number of ChAT-immunoreactive cells, the results strongly suggests that the

inflammatory processes reduced cholinergic cell death number and not simply

cholinergic cellular function.

The mechanism by which chronic neuroinflammation destroys cholinergic BF

cells is unknown. TNFa. may play a significant role. For example, over-expression of

TNFa. in the brain of transgenic mice decreases ChAT activity (Aloe et al., 1999). The

TNFa. receptor has a high degree of sequence homology with the low affinity p75

receptor for nerve growth factor that is expressed by all cholinergic neurons within the

BF (Book et al., 1992). During periods of increased inflammatory activity when elevated

amounts of cytokines are being released, enhanced TNFa. release could lead to over

stimulation of the p75 nerve growth factor receptor on cholinergic neurons and induce

60

apoptosis (Majdan et al., 1997; Van Antwerp et al., 1996; VanderZee et al., 1995). This

mechanism may explain the selective wlnerability ofBF cholinergic cells to chronic

neuroinflammation. In addition, we have found that cholinergic BF cells are wlnerable

to chronic infusion of TNFa..

In conclusion, to the extent that chronic neuroinflammation underlies aspects of

the pathological changes that occur in the brains of AD patients, the results of the present

study suggest that combined chronic administration of selective COX-2 and NMDA

receptor channel antagonists might provide significant nelu-oprotection. The long-term,

low-dose administration of such antagonists might delay the onset of AD, or slow its

progression once diagnosed.

61

Table 2 Effects of Chronic 37- and 74-day unilateral Lipopolysaccharide Infusion

upon Levels ofNeuropeptides within the Basal Forebrain

37-dayLPS

Galanin

Leu-Enkephalin

Neurokinin B

Neurotensin

Somatostatin

SubstanceP

74-dayLPS

Galanin

Leu-Enkephalin

Neurokinin B

Neurotensin

Somatostatin

SubstanceP

Control Side

4.8±2.4

2.6±1.1

5.6±2.3

4.5±1.6

22.3±9.0

2.6+1.0

7.6±2.2

5.0±2.3

12.0±4.0

5.2±1.3

17.0±8.0

6.3±2.4

Lesion Side

4.6±0.9

2.6±0.5

3.6±1.4

3.2±1.0

15.3±6.1

2.6+1.2

6.9±2.6

4.0±2.0

6.2±3.3

3.8±1.6

15.8±8.0

3.5±2.2

Values are Mean+ SO nanograms/mg protein. The LPS infusions did not significantly

alter any of these neuropeptide levels, as compared to the control side.

62

Fig. 3.1. The e£fects of unilateral infusions of LPS into the BF on cortical levels of

ChAT activity. Chronic 37-day LPS infusion significantly (P<0.0001) decreased cortical

ChAT activity. Chronic administration of memantine significantly (P<O.OS) attenuated

the efifects of the 37-day infusion of LPS on cholinergic cells. Chronic 74-day LPS

infusion significantly ^<0.0001) decreased cortical ChAT activity. Chronic

administration of CI987 significantly attenuated the effects of the 74-day infusion of LPS

on cholinergic cells.

% D

eclin

e in

Cor

tical

ChA

T A

ctiv

ity

W

(A

o\

64

Fig. 3.2. ChAT immunoreacdvity within the BF of rats infused chronically with aCSF

(A), LPS alone (B) or LPS + memantine (C). Chronic LPS administration led to the total

loss of ChAT-positive cells. The tissue is cavitated and numerous injection artifacts are

visible near the injection site (see arrows, 2B). In rats treated with LPS + memantine,

numerous ChAT-immunoreactive cells were observed ventral to the infusion site.

Reactive microglia within the BF in rats infused chronically with aCSF (D), LPS alone

(E) or LPS + memantine (F). Original magnification A-C (S4x) D-F (66x)

65

66

Fig. 3.3. ChAT inununoreactivity within the BF of rats infused chronically with aCSF

(A), LPS alone or LPS + CI987(C). Chronic LPS administration led to the almost

total loss of ChAT positive cells. In rats treated with LPS + CI987 numerous ChAT

immunoreactive cells were observed within Smm from the infusion site. Reactive

microglia within the BF in rats infused chronically with aCSF (D), LPS alone (E) or LPS

+ CI987 (F). Original magnification A-C (S4x) D-F (66x)

67

68

Fig. 3.4 The number of OX-6 positive cells within the BF. The number of OX-6 positive

cells within the BF was significantly attenuated by treatment with CI987, a COX-2

inhibitor, but not by memantine, an NMDA-receptor antagonist.

69

240

(0 220 • mam

"5 2 o

o 200

^ 180

g> 160 -

•"S 140-O Q- 120 -

9 X 100 -o M- 80 -o 0 60 -

E 40 -

Z 20 -

0 -

-k p<0.05vsaCSF

-I- p < 0.05 vs LPS

% •'S^

'9. •»

70

Fig. 3.5. Silver impregnation staining within the NBM region of the BP of rats given

aCSF for 74 days (A) or LPS for 37 days (B) or 74 days (C). In both the 37- and 74-day

infusions, numerous large cells were densely stained with silver grains suggesting an

active necrotic process within this region.

71

72

Chapter Four

SUMMARY AND DISCUSSION

The present studies have important implications for several broad areas of

investigation, including inflammatory toxicity, selective vuhierability and

neuroprotection. In this section, these topics are briefly reviewed, emphasizing the

contributions of the present work.

Inflammation A AD - From PMt to Preient

Alzheimer's disease (AD) is a disorder of aging that is characterized by

progressive loss of cognitive functioning, eventually culminating in complete dementia.

Alzheimer's disease takes its name from a German neurologist named Alois Alzheimer,

b 1907, Alzheimer published a landmark case study describing a set of previously

unidentified histopathological features m the brain of a SI year old woman who had been

demented at the time of her death (Alzheimer, 1907). In 1899, previous to Alzheimer's

discovery of the disorder that bears his name, his colleague Franz NissI had identified

unique cells within the central nervous system (CNS) which he named '̂ stabchenzellerf'

or rod cells. He considered them to be reactive neuroglia and suggested they had the

cq[>acity for proliferation, migration and phagocytosis. While the role of neuroglia was

advanced by del Rio-Hoitegi, glia cells htve primarily regarded as have bystander roles

73

in the brain, functions such as scafifolding, insulation, and removing cellular debris. It

wasn't until the 1980's that the role that glia play in inflammatory processes came to be

appreciated.

It has been well documented that AD has a robust inflammatory response (Aisen

and Davis, 1994; Frederickson and Brunden, 1994; McGeer et al., 1989; Mrak et al.,

1995; Rogers et al., 1992). Kficroglia are known to be activated in the CNS of patients

with AD at sites of neuronal degeneration and microglia are closely associated with

amyloid plaques in AD. The key question of this observation was whether the microglia

were simply responding to the degeneration of noirons and were activated as they

phagocytose the debris, or whether they were exacerbating the neuronal injury when they

became activated.

Although a variety of genetic, environmental, socioeconomic, and dietary factors

may be related to the pathogenesis of AD, immune-mediated inflammation within the

brain remains a common correlate for the disease. Many biomarkers associated with

inflammatory activity are elevated in AD brains. For example, AD is associated with

increased levels of the inflammatory cytokine interieukin-1 (IL-1) in brain tissue (Griffin

et al., 1989). Reactive gliosis, a conspicuous pathologic feature in AD, is a prominent

component of neuritic plaques (Delacourte, 1990; Griffin et al., 1989; Mancardi et al.,

1983; Mandybur and Chuirazzi, 1990; Schechter et al., 1981). Neuritic plaques contain

both 3-amyloid and reactive glial cells that overexpress inflammatory cytokines.

74

including IL-l, IL-6, and tumor necrosis factor-a (Benveniste et al., 1990, Chung and

Benveniste, 1990; Frei et al., 1989; Mrak et al., 1995; Sparacio et al., 1992). IL-l

promotes the gene expression of amyloid precursor protein (APP) in astrocytes (Blume

and Vitek, 1989) and microglia (Kreutzberg, 1996) and glia-producing cytokines such as

IL-l appear to be involved in the generation of amyloid (V andenabeele and Fiers, 1992).

In addition, elevated IL-Ia can induce the expression ofSIOOf3 (Griffin et al., 1989;

Sheng et al., 1996) and apolipoprotein E (Mrak et al., 1995) in astrocytes. Therefore,

reactive glia and elevations in inflammatory cytokines, such as IL-l, may contribute to

the pathology seen in AD (Frederickson, 1992; Frederickson and Brunden, 1994; Griffin

et al., 1989).

Inflammation involves stereotypic responses that do not distinguish between a

beneficial effect form a harmful one. The primary function of microglia and astrocytes is

to protect the integrity of the brain during the developmental period of the nervous

system and throughout the lifetime of the healthy adult. They function as surveillance

cells to maintain homeostasis and to repair small traumas in acute destabilizing

conditions. Microglia are involved in producing neuronal survival and sprouting factors

and produce factors which influence the brain's immune response. Astrocytes provide

trophic and survival factors for neurons and themselves produce factors which influence

the proliferation and migration of microglia. As protectors of the nervous system,

microglia clear the brain of inducers of inflammation such as apoptotic cells and

opportunistic pathogens through phagocytosis and pinocytosis, while astrocytes buffer

75

concentrations of ions, excitatory amino acids and biogenic amines in the extracellular

^mce. However, activation of miax)glia and astrocytes by viruses, chemicals, or trauma

may result in a chronic series of interactions among glia, neurons, blood-bome

inflammatory cells, cytokines and neuropeptides that can lead to a pathological

environment and subsequent nervous system damage (Perry and Gtordon, 1997;

Norenberg, 1997).

Clinical data provide additional evidence for the involvement of inflammatoiy

changes in the pathogenesis of AD. Recent reports suggest that patients who chronically

take anti-inflammatory drugs to treat the symptoms of inflanunatory diseases, such as

riieumatoid arthritis, have a reduced risk of developing AD (McGeer and McGeer, 1995;

McGeer et al., 1996; Breitner, 1996; Stewart et al., 1997). Among twins and siblings,

there is a negative association between previous use of anti-inflammatory drugs and

clinical AD ̂ reitner et al., 1994; Breitner et al., 1995). While clinical trials using anti­

inflammatory drugs as a treatment for AD is limited, one study (Rogers et al., 1993)

administered indomethacin or placebo over a 6-month period to patients and found a

significant cognitive improvement whereas placebo-treated patients declined.

It can be suggested that if these changes were benign and secondary to the

pathogenesis of AD, then ami-inflammatocy treatment would not have any effect on the

course of the disease, and in fact might exacerbate the condition by slowing down the

removal of cellular debris. Thisdoesnotiq)peartobethe8ituati(»L However, the

76

specific target for the anti-inflammatories (if there is one) in protecting against AD still

remains to be discovered. The question then becomes: Are inflammatory processes toxic

to neuronal cell functioning?

We have concentrated attention on the basal forebrain cholinergic system to

answer this question. The basal forebrain (BF) is a structure with an established role in

attention and memory processes (Robbins et al., 1989; Stoehr et al., 1997; Voytko et al.,

1994; Wenk et al., 1994). A decline in the number, size and function of cholinergic cells

in the BF may be responsible for the cognitive impairments associated with aging and

Alzheimer's Disease in humans (McGeer et al, 1984; Rinne et al, 1987; Swaab et al,

1994; Terry et al., 1991; Vogels et al., 1990). The basal forebrain cholinergic system is

located ventral to the anterior commisure and globus pallidus and a lateral to the

hypothalamus. It consists of large neurons that extend from the medial septum through

the diagonal band of Broca to the nucleus basalis of Meynert. The nucleus basalis is

larger in humans and primates than to its homologous structure, the nucleus basalis

magnocellularis (NBM), found in other species such as rodents and is the major

component of the substantia innominata. Neurons situated more rostrally in the medial

septum and the vertical limb of the diagonal band predominately innervate the

hippocampus and the cingulate cortex respectively, while neurons of the NBM in the

more caudal aq)ect of the BF project widely to innervate the neocortex and amygdala.

The cholinergic deficit is one of the most dearly defined neurochemical deficiencies that

has been identified in the brains of AD patients. Therefore, basal forebrain cholinergic

(BFC) neurons were chosen as a model system for 2 reasons:

(1) these neurons degenerate in AD and may be responsible for the cognitive deficits

seen in the disorder, therefore making it an important area to study and

77

(2) the BF is a large, compact collection of neurons whose anatomy has been thoroughly

described and the modulation of which can be readily studied.

Time Dependent Role of Inflammation in Cholinergic Cell Death

The purpose of Chapter 2 was to determine if inflammatory processes are toxic to

neurons. The results showed that both acute injections and chronic infusions of the

proinflammagen lipopolysaccharide into the basal forebrain of rats was toxic to

cholinergic cells. LPS was chosen to induce the inflammation in the BF of young rats

because its mechanism is well known and its effects have been well characterized

centrally and peripherally (Espey et al., 1995; Fujii et al., 1996). Furthermore, our

preliminary studies had found that at very low doses it produced a strong inflammatory

response by local glia cells.

Chronic infusions ofLPS decreased the level of activity of the acetylcholine

synthetic enzyme choline acetyltransferase (ChAT). In addition, these infusions

decreased the number of neurons that demonstrated immunoreactivity for this enzyme.

Furthermore, low dose, long-term infusions ofLPS produced a greater decline in ChAT

78

activity than a single injection of high dose LPS. These results suggest that the toxic

effects of neuroinflammation are more time-dependent than dose-dependent and support

the hypothesis that long-term inflammatory processes can impair cholinergic neuronal

function. It should be noted that a decline in cortical ChAT activity is consistent with the

death ofBF cholinergic cells but does not refute the possibility that these cholinergic

cells have stopped expressing the ChAT protein. In addition, the LPS infusions produced

an extensive activation of astrocytes and microglia following either acute or chronic

exposures.

Mechanisms of Neurotoxicity and Protection

The purpose of Chapter 3 was to investigate whether treatment with a

prostaglandin synthase inhibitor and a NMDA receptor antagonist could provide

neuroprotection from the cytotoxic effects ofLPS-induced neuroinflammation upon basal

forebrain cholinergic neurons.

Inflammatory processes that activate glia can lead to the release of other cytokines

and initiate an inflammatory cascade which can be detrimental to cellular function

(Bluthe et al., 1992; Quan et al., 1994). IL-l is a principal participant in the

inflammatory reaction and it can further stimulate the induction of other inflammatory

molecules such as prostaglandins and cytokines (Gaudet et al., 1980; Bernheim et al.,

1980). Below is a hypothesized cascade of biochemical steps that might lead to

cholinergic cell death:

LPS ~ IL-l ~Prostaglandins ~ t[Glutamate ]ext ~ NMD.Acecpt ~ Cell Death

79

During inflammatory processes in the brain, a marked rise of arachidonic acid and

increased synthesis of prostaglandins are noted (Gaudet et al., 1980; Bernheim et al.,

1980). In a nonstimulated cell, arachidonic acid is esterified to membrane glycolipids.

Upon stimulation, phospholipase A2 (PLA2) is activated in a Ca ++ dependent manner and

frees arachidonic acid from the membrane. Once arachidonic acid is released it can

either serve its own messenger functions in certain pathways or be metabolized into

prostaglandins by the enzyme cyclooxygenase (COX). There are two isoforms of the

COX enzyme. The constitutively expressed COX-I acts in a variety of settings to

produce homeostatic or maintenance levels of prostaglandins while COX-2 is inducible

upon inflammatory stimulation. Therefore, prostaglandins may be involved in the brain's

inflammatory response. Evidence for this includes the increased gene expression for

PLA2 in rat astrocytes by inflammatory mediators TNF and IL-l as well as LPS (Oka and

Arita, 1991). Cultured astrocytes and microglia produce and release PG~ in response to

stimulants such as LPS (Sawada et al., 1993). In addition, increased prostaglandin

production has been implicated in other dementia states such as mv dementia (Griffin et

al., 1994) and chronic use of anti-inflammatory treatments that inhibit the synthesis of

prostaglandins correlates with a decreased incidence of AD (McGeer et al., 1990).

80

The cytotoxic mechanisms of inflammatory processes may also be due in part to

excitatory amino acids. Astrocytes play a critical role in the extracellular buffering of

glutamate and other excitatory amino acids. Astrocytes may mitigate neuronal

excitotoxicity under normal conditions (Sugiyama et al., 1989). Neurons in astrocyte

poor cultures are more vuhierable to glutamate excitotoxicity (Rosenberg and Aizenman,

1989). Cholinergic neurons within the BF are vulnerable to excess stimulation of

glutamatergic NMDA receptors (Wenk et al., 199S) possibly due to the fact that they

receive a dense glutamatergic projection from the pedunculopontine tegmentum

(Rasmusson et al., 1994).

In culture, activated microglia can produce significant amounts of quinolinic acid

and other excitotoxins, which may destroy nerves cells through an NMDA-receptor

mediated pathway (Giulian et al., 1993; Lipton et al, 1994; Espey et al., 1995). This

mechanism may play a significant role in the induction of neuronal loss in AD.

Immunostimulated glia release an NMDA-like substance (Boje and Arora, 1992; Chao et

al., 1992) that could activate excitotoxic pathways independent of extracellular buffering

mechanisms. In addition, both microglia and astrocytes have been suggested to play a

role in the generation of cognitive dysfunction in HIV dementia by releasing cytokines

and activating NMDA receptors Epstein and Gendelman, 1993; Lipton, 1992).

81

Microglia respond to 13-amyloid by releasing IL-l (Araujo and Cotman, 1992)

which can induce an increased release of arachidonic acid and prostaglandins from

neuronal and nonneuronal cells (Gaudet et al., 1980; Bernheim et al., 1980). An

increased release of arachidonic acid has been shown to inhibit reuptake of glutamate and

other excitatory amino acids by astrocytes (Barbour et al., 1989; Volterra et al., 1992).

Arachidonic acid has also been reported to enhance NMDA-evoked currents (Miller et

al., 1992). Therefore, arachidonic acid could contribute to neurotoxicity not only by

inhibiting the reuptake of glutamate but also by increasing the effectiveness of glutamate

in evoking NMDA receptor-mediated processes. Arachidonic acid metabolites have also

been reported to have direct and indirect effects on glutamate buffering mechanisms by

astrocytes. Prostaglandins induce the release of glutamate from astrocytes (Bezzi et al,

1998) and can indirectly elevate the extracellular concentration of glutamate by inhibiting

its reuptake by astrocytes (Robinson et al., 1993; Rothstein et al., 1993).

As mention in above, there are two isoforms of COX enzyme - COX-I is

constitutively expressed for homeostatic and maintenance purposes and COX-2 is

inducible upon inflammatory stimulation. Traditional anti-inflammatory drugs (NSAIDs)

compete with arachidonate for binding to the cyclooxygenase active site of the enzyme

and are not specific for the isoform. The unpleasa.Iit side effects of gastric ulcers exist

after the chronic treatment of high doses ofNSAIDs because of this lack of specificity.

NSAIDs target the homeostatic isoform as well as the inflammatory isofonn. The

prostaglandin synthesis inhibitor CI987 (Parke-Davis) is a selective COX-2 inhibitor

82

currently undergoing Phase m clinical trials. CI987 was chosen because of its specificity

for the inflammatory isofonn, lack of side effects and clinical relevance.

If cytokine toxicity is mediated through NMDA-type glutamate receptors as

described above, then it should be possible to antagonize the activation of these receptors

and attenuate cytotoxic effects of neuroinflammation. Most NMDA antagonists are not

clinically tolerated (Lipton, 1993; Lipton and Rosenberg, 1994). The antagonist chosen

in these studies was memantine (Merz + Co., Frankfurt, Germany), a low affinity, non­

competitive, open-channel NMDA receptor antagonist. Blockade of the NMDA receptor

at the glutamate binding site with drugs such as APS leads to the undesirable side effect

of amnesia at extracellular glutamate concentrations well below toxic. Memantine,

however, binds to the PCP site inside the NMDA receptor-associated ion channel and is

therefore use-dependent by blocking the channel only when it is open. Increasing

concentrations of glutamate or other NMDA agonists that are hypothesized to be released

in the inflammatory cascade cause NMDA channels to remain open for a greater period

of time, on average. Under such conditions, an open-channel blocking drug has a better

chance to enter the cannel and block it. Unlike other NMDA open-channel blockers,

such as MK-801, memantine has a low-affinity for the binding site and therefore does not

remain in the channel for long periods of time. Because of its mechanism of action, the

cytotoxic effects of pathologic concentrations of glutamate are prevented to a greater

extent than the effects of physiologic concentrations (Lipton and Rosenberg, 1994). In

addition, memantine has been used clinically in Europe for many years with few

83

significant side-effects at clinically effective doses (Chen et al, 1992; Parsons et al., 1993;

Danyz et al., 1997).

In this study, infusion ofLPS into the BF of young rats produced extensive

inflammation and a significant loss of cholinergic cells. These results are consistent with

the findings of Chapter 2, i.e. both acute and chronic infusions ofLPS destroyed

cholinergic cells and activated astrocytes and microglia (see also Willard et al., 1999).

This study extends our understanding of the mechanisms by which chronic

neuroinflammation destroys BF cholinergic cells. We determined that daily peripheral

administration of memantine or Cl987 significantly attenuated the cytotoxic

consequences of chronic neuroinflammation upon basal forebrain cholinergic cells.

However, only CI987 attenuated the glial response to the neuroinflammation produced by

LPS.

The results of our studies are consistent with the hypothesized role for both

prostaglandins and NMDA receptors in the cascade of biochemical processes that lead to

BF cholinergic cell loss in AD. These results are also consistent with the proposed order

of the cascade presented above, i.e., that the effects of prostaglandins occur upstream to

NMDA receptor activation.

Issues of Selectivity

The purpose of these studies found in Chapters 2 and 3 was to investigate the

biochemical and histological changes of acute and chronic LPS administration upon BF

cholinergic and noncholinergic neuronal populations to determine the selectivity and

specificity of neuroinflammation.

The results of our histological and biochemical studies were consistent with the

hypothesis that the cytotoxic effects of chronic neuroinflammation are selective for

cholinergic cells, at least within the BF region as discussed within Chapters 2 and 3.

The radioimmunoassays found no significant difference in the endogenous level of

various neuropeptides within the BF. The histological analyses found a dense

distribution of silver-stained cells within the BF that had a general morphology and size

that was consistent with cholinergic magnocellular neurons. When considered together

with the decrease in neocortical ChAT activity and the reduced number of ChAT­

immunoreactive cells, the results strongly suggests that the inflammatory processes

reduced cholinergic cell number and not simply cholinergic cellular function.

The mechanism by which chronic neuroinflammation destroys cholinergic BF

cells is unknown. TNFa. may play a significant role. For example, over-expression of

TNFa. in the brain of transgenic mice decreases ChAT activity (Aloe et al., 1999). The

TNFa receptor has a high degree of sequence homology with the low affinity p75

84

85

receptor for nerve growth factor that is expressed by all cholinergic neurons within the

BF (Book et al., 1992). During periods of increased inflammatory activity when elevated

amounts of cytokines are being released, enhanced TNFa release could lead to over­

stimulation of the p75 nerve growth factor receptor on cholinergic neurons and induce

apoptosis (Majdan et al., 1997; Van Antwerp et al, 1996; VanderZee et al., 1995). This

mechanism may explain the selective vulnerability ofBF cholinergic cells to chronic

neuroinflammation. In addition, we have found that cholinergic BF cells are vulnerable

to chronic infusion ofTNFa.

Similarly, an immune-mediated process was shown to selectively reduce septal

ChAT activity and decrease hippocampal acetylcholine release following the injection of

hybridized cholinergic cells into the medial septal area of guinea pigs (Kalman et al,

1997). A recent study also isolated antibodies from the sera of Alzheimer's Disease

patients that selectively recognized BF cholinergic cell groups in the rat brain and

impaired cholinergic function (Foley et al, 1988). Likewise, both TNFa and LPS can

induce astrocytes to mediate cell death in co-cultures of cerebellar and hippocampal

neurons (Skaper et al., 1995).

In addition, if inflammatory processes induce toxicity via an NMDA receptor

mediated process, then cholinergic neurons within the BF are doubly vulnerable due to

the dense glutamatergic projection from the pedunculopontine tegmentum (Rasmusson et

al., 1994) as mentioned previously.

86

Conclusions

AD is a relentlessly progressive and fatal neurodegenerative disease. As the aged

population increases there is an ever-growing need for effective treatment strategies.

During the past decade, the knowledge of the molecular and environmental factors

contributing to the pathogenesis of AD has substantially increased. Many models of AD

have been proposed that have given profound insights to components of this disorder and

the workings of brain. However, despite these research advances, therapies that will

reverse or augment the dementing process have not emerged.

I have demonstrated how a small persistent insult can have enormous toxic

consequences over time. By infusing an inflammogen at low doses over a long period, I

have demonstrated selective cholinergic cell death within the basal forebrain of young

rats. Then identified a mechanism by which these cells might be impaired and provided

neuroprotection. The studies contained in this dissertation have components of AD that

no other model has been able to reproduce, that is, selective destruction of a neuronal

population over time set against a backdrop of widespread toxicity. The results obtained

from these studies have broad implications for not only the field of AD but for research

on aging and other degenerative disease states.

Finally, I wish to close this dissertation by stating what I believe has been the

guiding principal of these studies: to the degree that the value of a model of any disease is

determined by the ther^ies that are derived from it, then this woric is highly valuable

because the current model clearly predicts treatment strategies that have already been

shown to be effective.

88

APPENDIX A

HUMAN/ANIMAL SUBJECTS

This project is exempt from Human Subject Approval. Animal treatments were

carried out in compliance with federal and state laws, standards of the Department of

Health and Biman Services, and the guidelines of the Institutional Animal Care and Use

Conmiittee at the University of Arizona, protocol number 98-175.

REFERENCES

Aisen PS and Davis KL (1994) Inflammatory mechanisms in Alzheimer's disease: implications for therapy. Am. J. Psychiat. 151: 1105-1113.

Aloe L, Fiore M, Probert L, Turrini P and Tirassa P (1999) Overexpression of tumor necrosis factor-alpha in the brain of transgenic mice differentially alters nerve growth factor levels and choline acetyltransferase activity. Cytokine 11:45-54.

Alzheimer A (1907) About a peculiar disease ofthe cerebral cortex. Allgemeine Zeitschrift fur Psychiatrie und Psychischt. Gerichtliche Medizin 64: 16.

89

Araujo DM and Cotman CW (1992) J3-amyloid stimulates glial cells in vitro to produce growth factors that accumulate in senile plaques in Alzheimer's disease. Brain Res. 569: 141-145.

Barbour B, Szatkowski M, Ingledew N and Attwell D (1989) Arachidonic acid induces a prolonged inhibition of glutamate uptake into glial cells. Nature 342:918-920.

Bartus RT, Dean RL, Beer Band Lippa AS (1982) The cholinergic hypothesis of geriatric memory dysfunction. Science 217:408-417.

Bauer J; Strauss S, Schreiter-Gasser U, Ganter U, Schlegel P, Witt I, Yolk Band Berger M (1991) Interleukin-6 and a.-2-macroglobulin indicate an acute-phase state in Alzheimer's disease cortices. FEES Lett. 285:111-114.

Benveniste EN, Sparacio SM, Norris JG, Grenett HE and Fuller GM (1990) Induction and regulation of interleukin-6 gene expression in rat astrocytes. J. Neuroimmunol. 25:209-219.

Berger-Sweeney J, Heckers S, Mesulam M-M, Wiley RG, Lappi DA and Sharma M (1994) Differential effects on spatial navigation ofimmunotoxin-induced cholinergic lesions of the medial septal area and nucleus basalis magnocellularis. J. Neurosci. 14:4507-4519.

Bernheim HA, Gilbert TM and Stitt IT (1980) Prostaglandin E levels in third ventricular cerebrospinal fluid of rabbits during fever and changes in body temperature. J. Physio/. 301:69-78.

90

Bezzi P, Carmignoto G, PastiL, Vesce S, Rossi D, Rizzini BL, Pozzan T and Volterra A (1998) Prostaglandins stimulate calcium-dependent glutamate release in astrocytes. Nature 391:281-285.

Blume AJ and Vitek MP (1989) Focusing on IL-l promotion of f3-amyloid precursor protein synthesis as an early event in Alzheimer's disease. Neurobiol. Aging 10:406-408.

Bluthe R-M, Dantzer Rand Kelley KW (1992) Effects ofinterleukin-1 receptor antagonist on the behavioral effects of lipopolysaccharide in rat. Brain Res. 573:318-320.

Boje KM and Arora PK (1992) Microglial-produced nitric oxide and reactive nitrogen oxides mediate neuronal death. Brain Res. 587:250-256.

Book AA, Wiley RG and Schweitzer JB (1992) Specificity of 192 IgG-saporin for NGF receptor-positive cholinergic basal forebrain neurons in the rat. Brain Res. 590:350-355.

Breitner JC, Gav BA, Welsh KA, Plassman BL, McDonald WM and Helms MJ (1994) Inverse association of anti-inflammatory treatments and Alzheimer's disease: initial results of a co-twin control study. Neurol. 44:227-232.

Breitner JCS, Welsh KA, Helms MJ, Gaskell PC, Gau BA, Roses AD, Pericak-Vance MA and Saunders AM (1995) Delayed onset of Alzheimer's disease with nonsteroidal antiinflammatory and histamine H2 blocking drugs. Neurobiol. Aging 16:523-530.

Breitner JCS (1996) Inflammatory processes and antiinflammatory drugs in Alzheimer's disease: a current appraisal. Neurobiol. Aging 17:789-794.

Brosnan CF, Battistini L, Raine CS, Dickson DW, Casadevall A and Lee SC (1994) Reactive nitrogen intermediates in human neuropathology: an overview. Dev. Neurosci. 16:152-161.

Brugg B, Dubreuil YL, Huber G, Wollman EE, Delhaye-Bouchaud Nand Mariani J (1995) Inflammatory processes induce f3-amyloid precursor protein changes in mouse brain. Proc. Nat/. Acad. Sci. USA 92:3032-3035.

Chao CC, Hu A, Molitor TW, Shaskan EG and Peterson PK (1992) Activated microglia mediate neuronal injury via a nitric oxide mechanism. J. Immunol 149:2736-2741.

Chen H-SV, Pellegrini JW, Aggarwal SK, Lei SZ, Warach S, Jensen FE and Lipton SA (1992) Open-channel block ofNMDA responses by memantine: Therapeutic advantage against NMDA receptor-mediated neurotoxicity. J. Neurosci. 12:4427-4436.

Chung IY and Benveniste EN (1990) Tumor necrosis factor-a production by astrocytes: induction by lipopolysaccharide, IFN-y, and IL-1f3. J. /mmunol 144:2999-3007.

91

Colicos MA, Dixon CE and Dash PK (1996) Delayed, selective neuronal death following experimental cortical impact injury in rats: possible role in memory deficits. Brain Res. 739:111-119.

Danysz W, Parsons CG, Komhuber J, Schmidt WJ, and Quack G (1997) Aminoadamantanes as NMDA receptor antagonists and antiparkinsonian agents -preclinical studies. Neurosci. Biobehav. Rev. 21:455-468.

Davies P and Maloney AJF (1976) Selective loss of central cholinergic neurons in Alzheimer's disease. Lancet 2:1403.

Delacourte A (1990) General and dramatic glial reaction in Alzheimer brains. Neurology 40:33-37.

Drapier J-C and Hibbs m Jr. (1988) Differentiation of mouse macrophages to express nonspecific cytotoxicity for tumor cells results in L-arginine-dependent inhibition of mitochondrial iron-sulfur enzymes in the macrophage effector cells. J. Immunol. 140:2829-2838.

Epstein LG and Gendelman HE (1993) Human immunodeficiency virus type-1 infection of the nervous system: Pathogenic mechanisms. Ann. Neurol. 33:419-436.

Ernst RL and Hay JW (1994) The U.S. economic and social costs of Alzheimer's disease revisited. Am. J. Pub. Health 84:1261-1264.

Espey MG, Moffett JR and Namboodiri MA (1995) Temporal and spatial changes of quinolinic acid immunoreactivity in the immune system of lipopolysaccharide-stimulated mice. J. Leukoc. Bioi. 57:199-206.

Evans DA, Funkenstein HH, Albert MS, Scherr P A, Cook NR, Chown MJ, Hebert LE, Hennekens CHand Taylor 10(1989) Prevalence of Alzheimer's disease in a community population of older persons higher than previously reported. JAMA 262:2551-2556.

Farlow MR (1998) Etiology and pathogenesis of Alzheimer's disease. Am. J. Health­Syst. Pharm. 55 (Suppl2):S5-S10.

Fiebich BL, Biber K, Lieb K, Van Calker D, Berger M, Bauer J and Gebicke-Haerter PJ (1996) Cyclooxygenase-2 expression in rat microglia is induced by adenosine A2a-receptors. G/ia 18:152-160.

Finsen BR, Jorgensen MB and Zimmer J (1993) Microglia MHC antigen expression after ischemic and kainic acid lesions of the adult rat hippocampus. G/ia 7:41-49.

Flaris NA. Densmore TL, Molleston MC and Hickey WF (1993) Characterization of microglia and macrophages in the central nervous system of rats: definition ofthe differential expression of molecules using standard and novel monoclonal antibodies in normal CNS and in four models of parenchymal reaction. G/ia 7:34-40.

92

Foley P, Bradford HF, Dochart M, Fillet H, Luine VN, McEwen B, Buch G, Winblad B and Hardy J (1988) Evidence for the presence of antibodies to cholinergic neurons in the serum of patients with Alzheimer disease. J. Neurol. 235:466-471.

Fonnum F (1969) Radiochemical micro assays for the determination of choline acetyltransferase and acetylcholinesterase activities. Biochem. J. 115:465-472.

Frederickson RCA (1992) Astroglia in Alzheimer's disease. Neurobiol. Aging 13:239-253.

Frederickson RCA and Brunden KR (1994) New opportunities in AD research- Roles of immunoinflammatory responses and glia. Alz. Dis. Assoc. Dis. 8:159-165.

Frei K, Malipier UV, Leist TP, Zinkernagel RM, Schwab ME and Fontanna A (1989) On the cellular source and function of interleukin-6 produced in the central nervous system in viral diseases. Eur. J. Immunol. 19:689-694.

Fujii E, Irie K, Ogawa A. Ohba K and Muraki T (1996) Role of nitric oxide and prostaglandins in lipopolysaccharide-induced increase in vascular permeability in mouse skin. Eur. J. Pharmaco/. 297:257-263.

Gaudet RJ, Alam I and Levine L (1980) Accumulation ofcyclooxygenase products of arachidonic acid metabolism in gerbil brain during reperfusion after bilateral common carotid artery occlusion. J. Neurochem. 35:653-658.

Geldmacher DS and Whitehouse PJ Jr. (1997) Differential diagnosis of Alzheimer's disease. Neurology 48(suppl6):S2-S9.

Giulian D, Haverkamp LJ, Li J, Karshin WL, Yu J, Tom D, Li X and Kirkpatrick JB (1995) Senile plaques stimulate microglia to release a neurotoxin found in Alzheimer brain. Neurochem. Int .. 27:119-137.

Giulian D, Wendt E, Vaca K and Noonan CA (1993) The envelope glycoprotein of human immunodeficiency virus type-1 stimulates and releases neurotoxins from monocytes. Proc. Nat/. Acad. Sci. USA 90:2769-2773.

Goldgaber D, Harris HW, Hla T, Maciag T, Donnelly RJ, Jacobsen JS, Vitek MP and Gajdusek DC (1989) Interleukin-1 regulates synthesis of amyloid beta-protein

precursor mRNA in human endothelia cells. Proc. Nat/. Acad. Sci. USA 86:7606-7610.

Griffin DE, Wesselingh SL and McArthur JC (1994) Elevated central nervous system prostaglandins in human immunodeficiency virus-associated dementia. Ann. Neurol. 35:592-597

Griffin WST, Stanley LC, Ling C, White L, MacLeod V, Perrot LJ, White CL III and Arora C (1989) Brain interleukin-1 and S-100 immunoreactivity are elevated in Down Syndrome and Alzheimer disease. Proc. Nat/. Acad. Sci. USA 86:7611-7615.

93

Griffin WS, Sheng JG, Royston MC, Gentleman SM, Mckenzie JE, Graham DI, Roberts GW and Mrak RE (1998) Glial-neuronal interactions in Alzheimer's disease: the potential role of a 'cytokine cycle' in disease progression. Brain Pathology 8:65-72.

Gualtieri F, Dei S, Manetti D, Romanelli MN, Scapecchi Sand Teodori E (1995) The medicinal chemistry of Alzheimer's and Alzheimer-like diseases with emphasis on the cholinergic hypothesis. Farmaco 50:489-503

Hasegawa K and Aoba A (1994) Symptoms, diagnosis and epidemiology of dementia. In Anti-Dementia Agents, p 13-31. Academic Press Ltd.

Hanisch U-K, Neuhaus J, Rowe W, VanRossum D, Moller D, Kettenmann Hand Quirion R (1997) Neurotoxic consequences of central long-term administration of interleukin-2 in rats. Neurosci. 79:799-818.

Kalman J, Engelhardt ll, Le WD, Xie W, Kovacs I, Kasa P and Appel SH (1997) Experimental immune-mediated damage of septal cholinergic neurons. J. Neuroimmunol. 77:63-77.

Katsuura G, Gottschall PE, Dahl RR and Airmura A (1989) Interleukin-1J3 increases prostaglandin E2 in rat astrocyte cultures: modulatory effect of neuropeptides. Endocrinol. 124:3125-3127.

Kreutzberg GW (1996) Microglia: a sensor for pathological events in the CNS. TINS 19:312-318.

Knapp MJ, Knopman DA, Solomon PR, Pendlebury WW, Davis CS and Gracon SI (1994) A 30-week, randomized controlled trial of high-dose Tacrine in patients with Alzheimer's disease. JAMA 271:985-991.

Lipton SA and Rosenberg PA (1994) Mechanisms of disease: Excitatory amino acids as a final common pathway for neurologic disorders. N. Engl. J. Med 330:613-622.

Lipton SA, Yeh M and Dreyer EB (1994) Update on current models ofHIV -related neuronal injury: platelet-activating factor, arachidonic acid and nitric oxide. Adv. Neuroimmunol. 4:181-188.

Lipton SA (1993) Prospects for clinically tolerated NMDA antagonists: Open-channel blockers and alternative redox states of nitric oxide. 11NS 16:527-532.

Lipton SA (1992) Models of neuronal injury in AIDS: Another role for the NMDA receptor? 11NS 15:75-79

94

Lowry OH, Rosenbrough NJ, Farr ALand Randall RJ (1951) Protein measurement with folin phenol reagent. J. Biol Chem. 193:265-275.

Majdan M, Lachance C, Gloster A, Aloyz R, Zeindler C, Bamji S, Bhakar A, Belliveau D, Fawcett J, Miller FD and Barker P A (1997) Transgenic mice expressing the intracellular domain of the p75 neurotrophin receptor undergo neuronal apoptosis. J. Neurosci. 17:6988-6998.

Mancardi GL, Liwnicz BH and Mandybur TI (1983) Fibrous astrocytes in Alzheimer's disease and senile dementia of Alzheimer's type. Acta Neuropathol .. 61 :7 6-80.

Mandybur TI and Chuirazzi CC (1990) Astrocytes and the plaques of Alzheimer's disease. Neurol. 40:635-639.

McGeer EG and McGeer PL (1998) The importance of inflammatory mechanisms in Alzheimer disease. Experimental Gerontology 33:371-378.

McGeer PL, Akiyama H, Itagaki Sand McGeer EG (1989) Immune system response in Alzheimer's disease. Can. J. Neurol. Sci. 16:516-527.

McGeer PL and McGeer EG (1995) The inflammatory response system of the brain: implications for therapy of Alzheimer and other neurodegenerative diseases. Brain Res. Rev. 21:195-218.

McGeer PL, McGeer EG, Rogers J and Sibley J (1990) Anti-inflammatory drugs and Alzheimer's disease. Lancet 335:1037.

McGeer PL, McGeer EG, Suzuki J, Dolman CE and Nagai T (1984) Aging, Alzheimer's disease and the cholinergic system ofthe basal forebrain. Neurol. 34:741-745.

McGeer PL, Schulzer M and McGeer EG (1996) Arthritis and anti-inflammatory agents as possible protective factors for Alzheimer's disease: a review of 17 epidemiologic studies. Neurology 47:425-432.

95

Miller B, Sarantis M, Traynelis SF and Attwell D (1992) Potentiation ofNMDA receptor currents by arachidonic acid. Nature 355:722-725.

Minghetti L and Levi G (1995) Induction of prostanoid biosynthesis by bacterial lipopolysaccharide and isoproterenol in rat microglial cultures. J. Neurochem. 65:2690-2698.

Mohr E, Mendis T, Rusk IN and Grimes JD (1994) Neurotransmitter replacement therapy in Alzheimer's disease. J. Psychiatry Neurosci. 19: 17-23.

Morgan BP (1990) Complement: Clinical aspects and relevance to disease. p215 -225. Academic Press Ltd.

Mrak RE, Sheng JG and Griffin WST (1995) Glial cytokines in Alzheimer's disease: review and pathogenic implications. Human Pathol. 26:816-823.

Muir JL (1997) Acetylcholine, aging, and Alzheimer's disease. Phamacol. Biochem. Behav. 56:687-696.

Norenberg MD (1997) Astrocytes: normal aspects and response to CNS injury. p173-199. In Immunology of the Nervous System. (Keane RW & Hickey WF eds.) Oxford University Press.

Oka S and Arita H ( 1991) Inflammatory factors stimulate expression of group TI phospholipase A2 in rat cultured astrocytes. J. Biol Chem. 266:9956-9960.

Parsons CG, Gruner R, Rozental., Millar J and LodgeD (1993) Patch clamp studies on the kinetics and selectivity ofN-methyl-D-aspartate receptor antagonism by memantine (1-amino-3,5-dimethyladamantan). Neuropharm .. 32:1337-1350.

Perry VH, Andersson P-B.and Gordon S (1993a) Macrophages and inflammation in the central nervous system. TINS 16:268-273.

Perry VH and Gordon S (1997) Microglia and macrophages. In Immunology of the Nervous System (Keane RW & Hickey WF eds.) p155-172. Oxford University Press.

Perry VH, Matyszak MK and Fearn S (1993b) Altered antigen expression of microglia in the aged rodent CNS. Glia 7:60-67.

Pfeffer RI, Afifi A and Chance JM (1987) Prevalence of Alzheimer's disease in a retirement community. Am. J. Epidemiol. 125:420-436.

Qiu Z, Sweeney DD, Netzeband JG and Gruol DL (1998) Chronic interleukin-6 alters NMDA receptor-mediated membrane responses and enhances neurotoxicity in developing CNS neurons. J. Neuroscience 10445-10456.

Quan N, Sundar SK and Weiss JM (1994) Induction ofinterleukin-1 in various brain regions after peripheral and central injections of lipopolysaccharide. J. Neuroimmunol. 49:125-134.

Radi R, Beckman JS, Bush KM, and Freeman BA (1991) Peroxynitrite oxidation of sulfhydryls. The cytotoxic potential of superoxide and nitric oxide. J. Bioi. Chem. 266:4244-4250.

96

Rasmusson DD, Clow K and Szerb JC (1994) Modification of neocortical acetylcholine release and EEG desynchronization due to brainstem stimulation by drugs applied to the basal forebrain. Neurosci. 60:665-677.

Relton JK and Rothwell NJ (1992) Interleukin-1 receptor antagonist inhibits ischaemic and excitotoxic neuronal damage in the rat. Brain Res. Bull. 29:243-246.

Rich ffi, Rasmusson DX. Folstein MF, Carson K, Kawas C and Brandt J (1995) Nonsteroidal anti-inflammatory drugs in Alzheimer's disease. Neuro/45:51-55.

Rinne JO, Paljarvi Land Rinne K (1987) Neuronal size and density in the nucleus basalis ofMeynert in Alzheimer's disease. J. Neuronal. Sci. 79:67-76.

Robbins TW, Everitt BJ, Marston HM, Wilkinson J, Jones GH and Page KJ (1989) Comparative effects of ibotenic acid- and quisqualic acid-induced lesions of the substantia innominata on attentional function of the rat: Further implications for the role of the cholinergic neurons of the nucleus basalis in cognitive processes. Behav. Brain Res. 35:221-240.

Robinson MB, Djali S and Buchhalter JR (1993) Inhibition of glutamate uptake with L­transpyrrolidine-2,4-dicarboxylate potentiates glutamate neurotoxicity in primary hippocampal cultures. J. Neurochem. 61:2099-2103.

Rogers J, Kirby LC, Hempelman SR, Berry DL, McGeer PL, Kaszniak A W, Zalinski J, Cofield M, Mansukhani L, Willson P and Kogan F (1993) Clinical trial of indomethacin in Alzheimer's disease. Neurol. 43: 1609-1611.

Rogers J, Cooper N, Webster S, Schultz J, McGeer PL, Styren SD, Civin WH, Brachova L, Bradt B, Ward P and Lieberburg I (1992) Complement activation by J3-amyloid in Alzheimer's disease. Proc. Nat/. Acad. Sci. USA 89:10016-10020.

97

Rogers J (1995) Inflammation as a pathogenic mechanism in Alzheimer's disease. Drug Res. 45:439-442.

Rosenberg PA and Aizenman E (1989) Hundred-fold increase in neuronal vulnerability to glutamate toxicity in astrocyte-poor cultures of rat cerebral cortex. Neurosci. Lett. 103:162-168.

Roses A (1994) The Alzheimer's diseases. In Current Neurology (S Appel, ed.) pill. Mosby-Year Book.

Rothstein JD, Jin L, Dykes-Hoberg M and Kuncl RW (1993) Chronic inhibition of glutamate uptake produces a model of slow neurotoxicity. Proc. Nat/. Acad. Sci. USA 90:6591-6595.

Rothwell NJ, Luheshi G and Toulmond S (1996) Cytokines and their receptors in the central nervous system: physiology, pharmacology and pathology. Pharmacol. Ther. 69:85-95.

Sawada M, Suzumura A, Ohno K and Marunouchi T (1993) Regulation of astrocyte proliferation by prostaglandin E2 and the alpha subtype of protein kinase C. Brain Res. 613:67-73.

Schechter R, Yen S-HC, and Teny RD (1981) Fibrous astrocytes in senile dementia of the Alzheimer's type. J. Neuropath. Exp. Neurol. 40:95-100.

Schoenberg BS, Kokmen E and Okazaki H (1987) Alzheimer's disease and other dementing illnesses in a defined United States population: incidence rates and clinical features. Ann. Neurol. 22:724-729.

Sheng JG, Ito K, Skinner RD, Mrak RE, Rovnaghi CR, Van Eldik LJ and Griffin WST (1996) In vivo and in vitro evidence supporting the role for the inflammatory cytokine interleukin-1 as a driving force in Alzheimer pathogenesis. Neurobiol. Aging 17:761-766.

Skaper SD, Facci Land Leon A (1995) Inflammatory mediator stimulation of astrocytes and meningeal fibroblasts induces neuronal degeneration via the nitridergic pathway. J. Neurochem. 64:266-276.

Sparacio SM, Zhang Y, Vilcek J and Benveniste EN (1992) Cytokine regulation of interleukin-6 gene expression in astrocytes involves activation of an NF -lCB -like nuclear protein. J. Neuroimmunol. 39:231-242.

Stewart WF, Kawas C, Corrada M and Metter EJ (1997) Risk of Alzheimer's disease and the duration ofNSAID use. Neurology 48:626-632.

98

Stoehr JD, Mobley SL, Roice D, Brooks R, Baker LM, Wiley RG and Wenk GL (1997) The effects of selective cholinergic basal forebrain lesions and aging upon expectancy in the rat. Neurobio. Learn. Mem. 67:214-227.

Sugiyama KA, Brunori A and Mayer ML (1989) Glial uptake of excitatory amino acids influences neuronal survival in cultures of mouse hippocampus. Neuroscience 32:779-791.

Swaab DF, Hofman MA, Lucassen PJ, Salehi A and Uylings HBM (1994) Neuronal atrophy, not cell death, is the main hallmark of Alzheimer's disease. Neurobiol. Aging 15:369-371.

Swierkosz TA, Mitchell JA, Warner TD, Botting RM and Vane IR (1995) Co­induction of nitric oxide synthase and cyclo-oxygenase: interactions between nitric oxide and prostanoids. Br. J. Pharmacal. 114:1335-1342.

Terry RD and Katzman R (1983) Senile dementia of the Alzheimer type. Ann. Neurol. 14:497-506.

Terry RD, Masliah E, Salmon DP, Butters N, DeTeresa R, Hill R, Hansen LA and Katzman R (1991) Physical basis of cognitive alterations in Alzheimer's disease: Synapse loss is the major correlate of cognitive impairment. Ann. Neurol. 30:572-580.

Tomlinson BE, Blessed G and Roth M (1970) Observations on the brains of demented old people. J. Neurol. Sci. 11 :205-242.

Van Antwerp DJ, Martin SJ, Kafri T, Green DR and Verman 1M (1996) Suppression ofTNF-a-induced apoptosis byNF-xB. Science 274:787-789.

Van der Zee CEEM, Ross GM, Riopelle RJ and Hagg T (1996) Survival of cholinergic forebrain neurons in developing p75 deficient mice. Science 274:1729-1732.

Vandenabeele P and Fiers W (1991) Is amyloidogenesis during Alzheimer's disease due to an TI...,-1/a-6-mediated 'acute phase' in the brain? lmmunol. Today 12:217-219.

Vogels OJM, Broere CAJ, Ter Laak IU, Ten Donkelaar IU, Nieuwenhys Rand Schulte BPM (1990) Cell loss and shrinkage in the nucleus basalis Meynert complex in Alzheimer's disease. Neurobiol. Aging 11:3-13.

Volterra A, Trotti D, Cassutti P, Tromba C, Salvaggio A, Melcangi RC and Racagni G (1992) High sensitivity of glutamate uptake to extracellular free arachidonic acid levels in rat cortical synaptosomes and astrocytes. J. Neurochem. 59:600-606.

Voytko ML, Olton DS, Richardson RT, Gorman LK, Tobin IT and Price DL (1994) Basal forebrain lesions in monkeys disrupt attention but not learning and memory. J. Neurosci. 14:167-186.

Wagstaff AJ and McTavish D (1994) Tacrine. A review of its pharmacodynamic and pharmacokinetic properties, and therapeutic efficacy in Alzheimer's Disease. Drugs Aging 4:510-540.

99

Weidenfeld J, Kahbha K, Reches A and Shohami E (1992) Role of the central adrenergic system in the regulation of prostaglandin biosynthesis in rat brain. J. Neurochem .. 58:694-699.

Wenk GL, Harrington CA, Walker LC, Tucker DA and Rance NE (1992) Basal forebrain neurons and memory: A biochemical, histological and behavioral study of differential wlnerability to ibotenate and quisqualate. Behav. Neurosci. 106:909-923.

Wenk GL, Stoehr ID, Quintana G, Mobley SL and Wiley R.G. (1994) Behavioral, biochemical, histological and electrophysiological effects of 192 IgG-saporin injections into the basal forebrain of rats. J. Neurosci. 14:5986-5995.

Wenk GL, Danysz Wand Mobley SL (1995) MK-801, memantine and amantadine show potent neuroprotective activity against NMDA toxicity in the NBM - A dose response study. Eur. J. Pharmacol. 293:267-270.

Whitehouse PJ, Price DL, Clark AW, Coyle IT and DeLong MR (1981) Alzheimer's disease: Evidence for selective loss of cholinergic neurons in the nucleus basalis. Ann. Neurol. 10:122-126.

Willard LB, Hauss-Wegrzyniak Band Wenk GL (1999) The pathological and biochemical consequences of acute and chronic neuroinflammation within the basal forebrain ofrats. Neurosci. 88:193-200.

Wink DA, Kasprzak KS, Maragos CM, Elespuru RK, Misra M, Dunams TM, Cebula T A, Koch WH, Andrews AW, Allen JS, et al. (1991) DNA deaminating ability of genotoxicity of nitric oxide and its progenitors. Science 254:1001-1003.

Yan Q and Johnson EM Jr. (1989) Immunohistochemical localization and biochemical characterization of nerve growth factor receptor in adult rat brain. J. Comp. Neurol. 290:585-598.