University of Southern California Dissertations and Theses

The response of Allograft inflammatory factor-1 to neurotoxic injury, and its role as a secreted protein

(USC Thesis Other)

The response of Allograft inflammatory factor-1 to neurotoxic injury, and its role as a secreted protein

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content
THE RESPONSE OF ALLOGRAFT INFLAMMATORY FACTOR-1 TO
NEUROTOXIC INJURY, AND ITS ROLE AS A SECRETED PROTEIN.

by

Kristina M. Nowitzki

A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(BIOCHEMISTRY AND MOLECULAR BIOLOGY)

December 2008

Copyright 2008 Kristina M. Nowitzki

ii
Dedication

This thesis is dedicated to my parents who have always been the highest-ranking officials
in my fan club.
iii
Acknowledgements
The old saying it takes a village to raise a child translates easily into the rearing of a
graduate student. The completion of this thesis required the support of not only two
departments but also an extensive network of mentors and colleagues and a loving and
accommodating family. In addition to those overseeing my training in the department of
Biochemistry and Molecular Biology, I would like to acknowledge several members of
the Neurobiology department. I would like to thank Dr. Leslie P. Weiner for his belief in
my ability and ideas, his constant support, and his perpetual encouragement. The
completion of this work also relied heavily on the guidance and excellent trouble-
shooting skills of Dr. Brett Lund. Work involving the MPTP mouse model was done in
cooperation with Dr. Mike Jakowec and Dr. Giselle Petzinger and with the assistance of
several members of their lab group including Liz Hogg, Pablo Arevalo, and Marta
Vuckovic. Molecular biology was done with the assistance of Kathy Burke. Many other
members of the neurology department including Dr. Wendy Gilmore, Dr. Roel Van der
Veen, and Therese Dietlin contributed both insight and technical assistance.
My journey started long before graduate school and for that reason I would also like to
thank my parents who always encouraged the self-pursuit of answers and engrained in me
the value of hard work and perseverance and several mentors along the way including
Ms. Jane Schaffer, Mr. Jim Peabody, Dr. Ronald Hart, and especially Dr. Eric Stemp.
iv
Table of Contents
Dedication ii
Acknowledgements iii
List of Tables vi
List of Figures vii
Abstract ix
Introduction 1
Chapter 1: Background 5
Allograft Inflammatory Factor-1 5
Microglia 12
Other CNS Macrophages 13
Development, Maturation, and Maintenance of Microglia 14
Microglia Activation 17
Neuronal-Microglia Interactions 18
MPTP Neurotoxicity 19
Parkinson’s Disease 23
Chapter 2: The AIF-1 response to MPTP neurotoxicity 26
Introduction 26
Materials and Methods 28
Results 32
Discussion 48
Chapter 3: Secretion of AIF-1 by inflammatory cells 51
Introduction 51
Materials and Methods 52
Results 55
Discussion 63
Chapter 4: Production of a recombinant AIF-1 protein using a mammalian
expression system 68
Introduction 68
Materials and Methods 69
Results 76
Discussion 86
v
Table of Contents, Continued
Chapter 5: Extracellular AIF-1 as a modulator of inflammation 92
Introduction 92
Materials and Methods 93
Results 97
Discussion 111
Chapter 6: Conclusion 116
Bibliography 123
vi
List of Tables
Table 2.1 CD45 expression in AIF-1
+
cells isolated from MPTP-lesioned mice 45
Table 2.2 CD45 expression in AIF-1
hi
cells isolated from MPTP-lesioned mice 46
Table 2.3 MFI of AIF-1 expression in AIF-1
+
cells isolated from MPTP-lesioned mice 47
Table 5.1 Real-time RT-PCR Primers 96

vii
List of Figures

Figure 2.1 AIF-1 expression in the SNpc following MPTP lesioning. 33
Figure 2.2 Area and average density are increased in AIF-1 bright cells. 35
Figure 2.3 Number of AIF-1 bright cells in SNpc following MPTP lesioning. 37
Figure 2.4 AIF-1 expression in the striatum following MPTP lesioning 38
Figure 2.5 Loss of TH staining in the striatum of MPTP-lesioned animals. 39
Figure 2.6 AIF-1 bright cells express CD11b, but not GFAP or TH. 41
Figure 2.7 Flow cytometric analysis of mononuclear cells isolated from MPTP-
lesioned mice.
44
Figure 3.1 Western blot analysis of AIF-1 levels in supernatants of THP-1 cells
stimulated in vitro..
57
Figure 3.2 Western blot analysis of AIF-1 levels of supernatants of murine
splenocytes.
60
Figure 3.3 Adult murine microglia secrete AIF-1. 62
Figure 3.4 Microglia and splenocytes isolated from mice treated with systemic
LPS stimulation do not alter secretion of AIF-1.
64
Figure 4.1 Potential phosphorylation sites predicted in the coding sequence of
murine AIF-1.
77
Figure 4.2 PCR design of AIF-1 insert. 78
Figure 4.3 pTT3 expression vector used in the production of recombinant AIF-1. 80
Figure 4.4 EcoRV restriction digest of pTT3 and pAIF-1. 81
Figure 4.5 DNA sequencing of pAIF-1. 83
Figure 4.6 Detection and purification of recombinant AIF-1. 84
Figure 4.7 Identification of recombinant AIF-1 by mass spectrophotometery. 85
viii
List of Figures, continued

Figure 4.8 Actin bundling activity of AIF-1. 87
Figure 5.1 Increased glial activation surrounding intracranial injections of rAIF-1. 99
Figure 5.2 rAIF-1 has no effect on cell survival of BV2 microglia. 101
Figure 5.3 rAIF-1 does not autoregulate AIF-1 expression in BV2 microglia 102
Figure 5.4 rAIF-1 does not alter iNOS expression in BV2 microglia 104
Figure 5.5 Microglia isolated from adult mice secrete low levels of inflammatory
cytokines ex vivo.
105
Figure 5.6 rAIF-1 stimulates secretion of IL-6 and TNFa from human PBMC. 107
Figure 5.7 rAIF-1 does not stimulate secretion of IL-10, IL-1, or IL-4 from human
PBMC
109
Figure 5.8 rAIF-1 stimulates the secretion of MCP-1, RANTES, and IL-8 from
human PBMC.
110
.
ix
Abstract
Allograft inflammatory factor-1 (AIF-1) is an evolutionary conserved protein important
to inflammatory responses throughout the body including that of microglia in the central
nervous system (CNS). In addition to critical intracellular roles in the activation of
microglia and macrophages, AIF-1 can be secreted by these cells in response to
inflammatory signals as well as soluble signals released by dying neurons. In response to
the neurotoxin 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine, we found increased levels
of AIF-1 expression in cells clustered in the substantia nigra pars compacta (SNpc), the
site of dopaminergic cell death. The number of these AIF-1 bright cells continued to
increase even after neuronal cell death was complete. This increased expression of AIF-1
was restricted to resident microglia; flow cytometric analysis showed that infiltrating
CD45
hi
leukocytes did not express high levels of AIF-1. Analysis of microglia ex vivo
demonstrated the secretion of AIF-1 by these cells, lending support to a role for this
molecule as an extracellular participant in the microglia response to neurotoxicity. To
define a functional role for extracellular AIF-1 in CNS inflammation, we produced a
recombinant mouse AIF-1 protein using a mammalian expression system and tested the
functional activity of this protein both in vitro and in vivo. Intracerebral injection of
recombinant AIF-1 into both the mouse striatum and midbrain led to activation of
microglia and astrocytes near the site of injection, which was more prolonged and
widespread compared to controls. Interestingly, ex vivo analysis of microglia showed no
significant changes in inflammatory cytokine secretion whereas human peripheral blood
mononuclear cells showed altered patterns in the secretion of TNF, IL-6, MCP-1, IL-8,
x
and RANTES. This work shows for the first time that AIF-1 can be secreted by adult
murine microglia and describes a new role for AIF-1 as a secreted molecule that can
peripherally and locally influence inflammatory responses, opening up new areas of
research into the initiation and propagation of the inflammation response in the CNS.
1
Introduction
Microglia are immunocompent cells of the central nervous system (CNS) that become
activated in response to several types of CNS injury including neurodegenerative diseases
such as Alzheimer’s disease and Parkinson’s disease (PD). In PD and a well-studied
mouse model of PD using the neurotoxin 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine
(MPTP), microglia activation is prominent in regions of dopaminergic injury and is
believed to occur secondary to neuronal death. The ways in which microglia activation
contribute to the pathology of PD remain unclear. In vitro studies of microglia cell
biology and mouse models of CNS injury reveal that microglia can contribute to the
resolution of injury through the phagocytosis of cell debris, secretion of growth factors,
and scavenging of excitotoxic molecules. However, activated microglia can also secrete
several neurotoxic factors including pro-inflammatory cytokines and reactive oxygen
species. Thus, although initially triggered by neuronal death, activated microglia may
participate in additional neuronal damage in both the MPTP mouse model and PD.
Because there are no current treatments to stop or slow neurodegeneration in patients
with PD, understanding exactly how microglia respond to neuronal degeneration may
lead to therapeutic strategies aimed at promoting neuroprotective measures and inhibiting
neurotoxic activities of microglia.
The allograft inflammatory factor-1 (AIF-1) protein is rapidly gaining popularity as a
specific marker for microglia and macrophages. In vitro studies have demonstrated the
functional importance of AIF-1 in several microglia and macrophage activities including
2
migration, phagocytosis, and the secretion of cytokines. In addition, the secretion of
AIF-1 by both microglia and macrophages has been reported. In vitro, microglia not only
increase expression of AIF-1 but also induce secretion of AIF-1 in the presence of
apoptotic cerebellar granule neurons. The function of extracellular AIF-1 in the CNS
remains unknown; AIF-1 was not neuroprotective in the cerebellar granule neuron model.
Studies in the periphery suggest that extracellular AIF-1 can alter the secretion of
cytokines by peripheral inflammatory cells.
The secretion of AIF-1 by microglia cells in response to neuronal death and the strong
association of AIF-1 with inflammation provides the backdrop for the hypothesis of this
work, which is that AIF-1 may be an important modulator of microglia activation in PD
and the MPTP mouse model of PD. Specifically, secretion of AIF-1 by microglia in
response to dopaminergic cell death may act in an autocrine fashion to regulate basic
microglia functions such as proliferation, gene expression, and cytokine secretion.
The secretion of AIF-1 in response to dopaminergic cell death has not been established.
In fact, AIF-1 expression has not been reported in PD brains and information regarding
the expression of AIF-1 in the MPTP mouse model is limited. One of the specific goals
of this thesis was to first determine the extent of the AIF-1 response to dopaminergic
neurooxicity in vivo. This work is described in Chapter 2, which covers an
immunohistochemical time course of AIF-1 expression in MPTP-lesioned mice.
Increased AIF-1 expression was seen at early time points (days 1-3) in the striatum,
where MPTP lesioning leads to terminal damage and dopamine loss. A more pronounced
3
increase in AIF-1 expression was seen in the substantia nigra pars compacta (SNpc), the
site of dopaminergic cell death, with increased numbers of cells exhibiting greatly
increased expression of AIF-1 (AIF-1 bright cells) over the first 10 days. Co-staining
experiments confirmed AIF-1 expression was specific to microglia/macrophage cells and
flow cytometric analysis was used to determine the contribution of infiltrating peripheral
leukocytes to the increase in AIF-1 expression seen in the SNpc.
The AIF-1 secretion study mentioned above used microglia cells cultured from newborn
rats, and recent data has shown that these microglia cultures, which have been regularly
used to study microglia cell biology, are generally more reactive than microglia isolated
from mature animals. Due to the specific interest of this thesis in neurodegenerative
disease, the secretion of AIF-1 by adult microglia was an important distinction not
previously covered by the literature. Chapter 3 describes secretion of AIF-1 by a human
mononuclear cells line, murine splenocytes, and ex vivo cultures of adult murine
microglia.
The next specific goal of this thesis was to understand the role of extracellular AIF-1 in
the microglia response to CNS injury. The production and purification of a murine
recombinant AIF-1 protein using a mammalian expression system is described in Chapter
4. Functional studies of recombinant AIF-1 are described in Chapter 5. Steriotactic
injections of recombinant AIF-1 into the brains of normal adult mice led the activation of
both microglia and astrocytes that was more diffuse and prolonged than controls. In vitro
experiments were aimed at defining specific roles of AIF-1 on microglia, but no activity
4
of recombinant AIF-1 was seen in the studies outlined. Experiments were extended to
the peripheral immune system to validate activities previously reported in the literature,
and recombinant AIF-1 produced as a part of this thesis did increase the secretion of
interleukin-6 (IL-6), tumor necrosis factor-alpha (TNF!), monocyte chemotactic protein-
1 (MCP-1), RANTES, and interleukin-8 (IL-8) by human peripheral mononuclear cells
(hPBMC).
A general review of the literature pertinent to what is currently known about AIF-1,
microglia cell biology, and the mechanisms of MPTP neurotoxicity are given in Chapter
1, along with evidence of microglia activation in both MPTP and PD. Finally,
concluding remarks on this work including recommendations for future study and a
discussion on the clinical relevance of this work is presented in Chapter 6.
5
Chapter 1: Background
Allograft Inflammatory Factor-1
Allograft inflammatory factor-1 (AIF-1) is a 17 kDa, Ca
2+
-binding protein, quickly
establishing itself as an important participant in the inflammatory response in both the
central nervous system (CNS) and peripheral tissues. aif-1 was originally cloned from rat
and human cardiac allografts and expression was localized to infiltrating mononuclear
cells (Utans et al. 1995; Utans et al. 1996). Shortly after, similar gene products were
identified in other model systems: ionized calcium binding protein (iba1) was identified
in a screen of rat mixed brain primary cultures as a microglia-specific transcript (Imai et
al. 1996), microglia response factor-1 (mrf-1) was localized to microglia in cultures of rat
cerebellar granule neurons undergoing apoptosis (Tanaka et al. 1998). These cDNA
sequences are nearly identical, showing only some variability in the 5’ end, consistent
with alternative transcription start sites or incomplete 5’ extension (Sibinga et al. 2002).
In addition, several AIF-1 splice variants have been identified including irt-1 (Autieri and
Agrawal, 1998), bart-1 (Autieri et al. 1996), and two unnamed human splice variants
(Hara et al. 1999). A recent review called for a more unified nomenclature for these
transcripts (Deininger et al. 2002). However, the Iba1 name persists in part due to the
popularity of the antibody contributed by this group (Imai et al. 1996). For simplicity, I
will consistently refer to this full-length 17kDa protein as AIF-1, even when reporting
data from studies using other nomenclature.
6
AIF-1 is encoded within the HLA class III region of the major histocompatibility
complex, a region containing genes for several important immune components including
cytokines and members of the complement cascade. The importance of AIF-1 is also
suggested by a high sequence identity across species. The mouse nucleotide sequence
shares 92.1% and 89% identity with rat and human, respectively (Watano et al. 2001). In
addition to several mammals, AIF-1 has been cloned in more phylogenetically distant
species including the carp (Fujiki et al. 1999), and marine sponge (Kruse et al. 1999). In
fact, the similarity between sponge and vertebrates reaches 70%. The role of AIF-1 in
the immune response seems to be conserved as well. Though primitive metazoans,
marine sponges have a highly evolved histocompatibilty system and mount a complex
rejection process when genetically different sponge tissue is introduced. Using an in
vitro model of sponge graft rejection, Müller et. al. showed an increase in aif-1 transcript
levels in an allogenic mixed sponge cell reaction (Muller et al. 2002).
Motif analysis of the 147 amino acid sequence encoded by mouse or human AIF-1
reveals two EF-hand motifs, putting it in the same family as the classical Ca
2+
binding
proteins calmodulin and troponin C (Lewit-Bentley and Rety 2000). In fact, within the
EF-hand region, AIF-1 shares 33% and 32% homology with troponin C and calmodulin,
respectively (Imai and Kohsaka 2002). Outside of the EF-hand region, AIF-1 has no
homology with any other identified proteins (Imai and Kohsaka 2002). At least in mice,
the first EF-hand is able to bind calcium and crystallization studies have shown that the
binding of Ca
2+
results in a conformational shift in the protein structure (Yamada et al.
7
2004). With this shift comes the ability to form a protein dimer, at least in the crystal
form. Another highly conserved motif is a 44-amino acid segment with a –KR-KK-
GKR- pattern characteristic of peptide hormone precursor proteins (Chen et al. 1997).
Sequence analysis of mouse AIF-1 revealed potential phosphorylation sites for casein
kinase II (38-41, 45-48, 82-85, 95-98, 102-105), tyrosine kinase (29-37), protein kinase A
(133-136), protein kinase C (69-72) (Watano et al. 2001).
Consistent with how these transcripts were originally described, in most models AIF-1
has been found to be specific for microglia/macrophages. In fact, the Iba1 antibody has
been used widely as a microglia/macrophage marker in several models of CNS injury and
infection. Increased staining for AIF-1/Iba1 has been noted in experimental autoimmune
encephalomylitis, uveitis, and neuritis (Schluesener et al. 1998; Schluesener et al. 1999);
mouse facial nerve axotomy (Ito et al. 1998); focal human brain infarctions (Ito et al.
2001); rat and human traumatic brain injury (Beschorner et al. 2000); spinal cord injury
in rats (Schwab et al. 2001); neurovirulent influenza A virus infection (Mori et al. 2001);
infection of the trigeminal nucleus with herpes virus (Mori et al. 2003); and human
gliomas (Deininger et al. 2000). Increased AIF-1 expression has also been localized to
infiltrating macrophages in a variety of systemic injury models including muscle
devascularization (Kuschel et al. 2000), inflammatory skin disorders (Orsmark et al.
2007), systemic sclerosis (Del Galdo et al. 2006), endometriosis (Koshiba et al. 2005),
rheumatoid arthritis (Kimura et al. 2007), and balloon angioplasty (Autieri 1996).
8
There are specific exceptions to the microglia/macrophage expression that should be
noted. First, AIF-1 shows high expression in rodent testis (Imai et al. 1996) that is not
seen in humans and can be localized to developing germ cells (Watano et al. 2001). In
addition, AIF-1 expression has been noted in human CD3+ T-lymphocytes, where it may
play a role in T-cell activation (Kelemen and Autieri 2005; Del Galdo and Jimenez
2007). AIF-1 expression in rodent T-lymphocytes has not been described. In both rat
and human, AIF-1 expression in the kidney can be localized to podocytes, epithelial cells
involved in glomerular filtration (Tsubata et al. 2006). When stimulated with
inflammatory cytokines or T-cell conditioned media, human vascular smooth muscle
cells (VSMC) can express AIF-1 (Autieri et al. 2000). One study in Creutzfeld-Jakob
patients reported the expression of AIF-1 in both neurons near prion plaques and
neuroblastoma cells (Deininger et al. 2003).
In microglia/macrophages, AIF-1 is upregulated by a number of inflammatory factors
including IFN", IL-1#, TNF (Utans et al. 1995; Yang et al. 2005). Meanwhile,
expression can be inhibited in macrophages by the anti-inflammatory drug sodium
salicylate (Yang et al. 2005). A few studies showed that products of oxidation could also
stimulate AIF-1 expression. Oxidized LDL was found to stimulate expression of AIF-1
in macrophages (Tian et al. 2006), and docosahexaenoic acid hydroperoxide, a
photoreceptor membrane fatty acid that undergoes peroxidation during autoimmune
uveoretinitis, increased expression of AIF-1 in microglia (Saraswathy et al. 2006).
Regulation can come at the level of transcription. The mouse AIF-1 promoter has been
9
sequenced and has many potential sites for the binding of transcription factors (Sibinga et
al. 2002). Both basal and IFN" inducible expression in murine macrophages requires
Pu.1 binding to an ETS site within the promoter and binding of unknown factor to an
interferon-gamma responsive elements (Sibinga et al. 2002).
AIF-1 has been well studied as an intracellular protein in cultures of microglia,
macrophages, or VSMC, and has been found to effect many fundamental
microglia/macrophage functions. First, AIF-1 interacts at several steps in the regulatory
pathway involved in dynamic remodeling of the actin cytoskeleton, which is central to
cellular processes like migration and phagocytosis. This area has been developed in
detail in studies using the microglia cell line MG5 and is substantiated by reports from
another group using VSMC. When MG5 cells are stimulated to form membrane ruffles
or phagocytic cups, AIF-1 co-localizes with actin and L-fimbrin, an actin-bundling
protein (Ohsawa et al. 2000) (Sasaki et al. 2001) (Ohsawa et al. 2004). AIF-1 can form
functional interactions with both these proteins. AIF-1 also co-localizes with Rac, a
member of the Rho family of small GTPases that tightly controls cytoskeleton
remodeling (Ohsawa et al. 2000). AIF-1 can lead to the activation of Rac and may
directly bind to Rac proteins. Expression of a dominant-active Rac in MG5 microglia
causes formation of membrane ruffles without stimulation, but this phenomenon is
suppressed in a pathway specific manner when AIF-1 mutants are co-expressed . In
addition, experiments dissecting the signaling involved in PDGF stimulation of
membrane ruffling revealed a novel signaling pathway in which the activation of Rac
10
through phospholipase C-gamma is dependent on AIF-1 (Kanazawa et al. 2002). Thus,
through many sites of interaction, AIF-1 can regulate cytoskeleton remodeling in these
cells.
Limited studies have also suggested the involvement of AIF-1 in other cellular processes.
For instance, AIF-1 may be important to cell survival. siRNA silencing of AIF-1 in a
macrophage cell line caused increased apoptosis accompanied by increases in the
cleavage of caspase 3 and expression of Bax (Yang et al. 2005), while siRNA silencing in
a different macrophage cell line led to a decrease in proliferation that was not due to
apoptosis (Tian et al. 2006). In addition, over-expression of AIF-1 in VSMC leads to an
increase in proliferation, mainly through the secretion and autocrine effects of G-CSF
(Chen et al. 2004). AIF-1 can also contribute to the generation of reactive oxygen
species. In a macrophage cell line, over-expression of AIF-1 led to both an increase in
expression of iNOS and production of nitric oxide (NO), while siRNA silencing of AIF-1
expression led to a decrease in iNOS levels (Yang et al. 2005). In addition to actin
remodeling, Rac is also one of the regulatory subunits involved in formation of the
NADPH oxidase complex at the inner cell membrane of phagocytic cups, responsible for
super oxide production (Imai and Kohsaka 2002). So the effects of AIF-1 activation of
Rac can contribute to the generation of super oxide, though this has not been clearly
demonstrated in vitro. Lastly, studies have also suggested involvement of AIF-1 in the
regulation of cytokine and chemokine production. In RAW 264.7 macrophages, stable
transfection with AIF-1 enhanced expression and secretion of IL-6, IL-10 and IL-12p40
11
upon stimulation with LPS (Watano et al. 2001). Overexpression of AIF-1 in another
macrophage cell line caused an increase in migration associated with an increase in
expression of the chemokine MCP-1, while siRNA silencing of AIF-1 decreased MCP-1
transcript levels (Yang et al. 2005).
Though AIF-1 does not have a classical signal sequence, its secretion has been described
in many studies. Levels of AIF-1 in the peritoneal fluid of women with endometriosis
were higher than that from normal women and correlated with clinical scores of disease
severity (Koshiba et al. 2005). In addition, peritoneal macrophages isolated from women
with endometriosis secreted more AIF-1 which was stimulated with addition of IL-1# or
IFN". Similarly, AIF-1 levels were elevated in the synovial fluid of patients with
rheumatoid arthritis (Kimura et al. 2007). AIF-1 has also been found in the sera of mice
with EAE. Of particular interest to this study, AIF-1 is both upregulated and secreted by
microglia in co-culture with cerebellar granule cells undergoing apoptosis (Tanaka and
Koike 2002). Secretion could also be increased by ATP and PAF, which may be released
by dying neurons. In this model, secreted AIF-1 had no effect on the survival of neurons,
and the function of AIF-1 extracellularly remains unclear. In the mixed sponge system
addition of recombinant AIF-1 stimulated the expression of Tcf-like transcription factor
(Muller et al. 2002). In addition, select studies have shown extracellular AIF-1 can
regulate proliferation and differentiation of muscle satellite cells (Kuschel et al. 2000)
and alter glucose-stimulated insulin secretion (Chen et al. 1997). In the rheumatoid
arthritis study, addition of AIF-1 lead to an increase in IL-6 secretion from both synovial
12
cells and human PBMCs (Kimura et al. 2007). No role for extracellular AIF-1 in the
CNS has been suggested and no receptor for AIF-1 has been identified.
Microglia
This thesis examines the expression and secretion of AIF-1 by microglia cells,
specialized immunocompetent cells of the CNS. The blood-brain barrier prevents regular
surveillance by immune cells of the peripheral circulation, normally protecting the CNS
from the extensive tissue damage associated with a full immune response. Instead, the
CNS is equipped with specialized cells of myeloid origin called microglia that respond to
injury and infection through a process termed activation. An average mouse brain
contains about 3.5 million microglia, which vary in density throughout regions of the
brain (Lawson et al. 1990). One particularly dense area is the substantia nigra, with
microglia representing 12% of cells (Lawson and Perry 1995).
In the normal CNS, microglia are described as being in a resting state and have a unique
morphology with a small nucleus of 5-7 µm, little cytoplasm, and several highly
branched processes that maximize cell surface coverage of the surrounding parenchyma
(Perry et al. 1985). In further contrast to tissue macrophages of other organs, resting
microglia have greatly attenuated phagocytic activity and low expression of
hematopoeitic markers. Despite the terminology used, resting microglia are not inactive.
The branched processes are undergoing constant reorganization and movement, scanning
the extracellular environment (Davalos et al. 2005; Nimmerjahn et al. 2005). In the
13
normal brain this scanning is random but becomes more targeted in the direction of small
lesions. There is evidence that the resting state is actively maintained by signals from
neighboring neurons, and not just the result of an absence of activating stimuli. This is
discussed in more detail in a following section on microglia-neuronal interactions.
Other CNS Macrophages
Microglia should be distinguished from other myeloid cells associated with the CNS.
Perivascular macrophages are long, flattened cells that lie next to the vasculature,
completely surrounded by the basal membrane (Graeber and Streit 1990). Other
macrophage populations are found in the choroid plexus and the meninges (McMenamin
1999). Functionally, these cells behave much more like other tissue macrophages in their
ability to phagocytose and serve as antigen presenting cells (Hickey and Kimura 1988).
These CNS macrophages also differ from resting microglia in their immunophenotype.
In particular, CNS macrophages express both MHC Class II (Streit 1989, Exp Neurol)
and the macrophage scavenger receptor, CD163, while resident microglia do not express
these in normal conditions. Activated microglia can express these markers (Honda et al.
1998; Fabriek et al. 2005). The only well established method to distinguish between
activated microglia and CNS-associated macrophages (as well as infiltrating
macrophages) is based on flow cytometric analysis of CD45 levels. CD45 is a common
leukocyte antigen that is expressed by all hematopoeitic cells. Peripheral leukocytes and
CNS-associated macrophages express high levels of CD45, wheras resident microglia
14
express low levels of CD45. Activated microglia can upregulate CD45 expression in
some instances, but levels do not reach that of other macrophages.
In contrast to parenchymal microglia, these macrophage cells are readily replenished by
circulating monocytes (Bechmann et al. 2001; Bechmann et al. 2001). Due to their rapid
turnover and strategic locations at the blood brain barrier, vascular macrophages are
considered as the entryway for many CNS infections including HIV (Vallat et al. 1998).
Infected macrophages have an increased ability to migrate into the neural tissue and
infect parenchymal microglia. Perivascular macrophages may also enable the spread of
systemic inflammation signals into the CNS. For instance, intraperitoneal injection of the
bacterial endotoxin lypopolysacharide (LPS) leads to an increase in COX-2 expression in
perivascular macrophages , which is key to promoting the febrile response (Elmquist et
al. 1997).
Development, Maturation, and Maintenance of Microglia
The origin and development of microglia has remained controversial since the subject
was first addressed by Pio del Rio-Hortega in 1932 (del Rio-Hortega 1932). A thorough
history of the theories surrounding the origin of ramified microglia was provided by Ling
and Wong (Ling and Wong 1993). It is now generally accepted that, unlike other glia of
the CNS which arise from the neuroectoderm, microglia are derived from mesodermal
hematopoietic cells (Perry et al. 1985; Hickey and Kimura 1988; Hickey et al. 1992; Ling
and Wong 1993; Eglitis and Mezey 1997; Kurz and Christ 1998). Microglia with mature
morphology are not seen in the brain until the middle of the second trimester in humans
15
(Monier et al. 2006), at approximately 14 days postpartum in the mouse (Hirasawa et al.
2005), and 18 days postpartum in the rat (Dalmau et al. 2003). However, a population of
ameboid microglia cells that are immunoreactive for many myeloid markers are found as
early as embryonic day 6 in rodents (Hirasawa et al. 2005) and 5.5 gestational weeks in
humans (Monier et al. 2006). These microglia precursors begin to populate the early
neural folds and are believed to have migrated from the blood pools of the yolk sac
(Alliot et al. 1999). There is a dramatic increase in the number of ameboid microglia
during neurogenesis and these cells not only clear apoptotic cells (Ashwell 1990; Ferrer
et al. 1990; Alliot et al. 1991) and aberrant axons (Innocenti et al. 1983; Ashwell 1990),
but also participate in decisions of cell fate (Thery et al. 1991; Lang et al. 1994; Mallat
and Chamak 1994) (Nagata et al. 1993) (Cammer and Zhang 1996; Miller and Kaplan
2001) (Cuadros and Navascues 1998). Secretion of trophic factors by these cells can
promote neurite and axonal growth (David et al. 1990; Stolz et al. 1991; Farinas et al.
2002) (Markus et al. 2002) as well as promote the proliferation of astrocytes and the
myelination and vascularization of the CNS (Giulian et al. 1988; Hamilton and Rome
1994). Controversy is now focused on the origin and turnover of adult microglia cells.
Many groups believe ameboid microglia differentiate into resident microglia and the
rapid colonization of these cells and the appearance of cells with an intermediate
morphology support this theory. However, a second wave of cells infiltrating the
postnatal rodent CNS from the bone marrow may also be responsible for mature
microglia. Much of the work showing the ability of cells from the bone marrow to take
on mature microglia morphology uses bone marrow chimeras. Two recent studies have
16
revealed certain biases in this experimental set-up. First, irradiation used to deplete the
bone marrow cells of the recipient can also lead to an increase in cytokines and
chemokines that directly recruit bone marrow cells (Mildner et al. 2007). When the brain
is protected from irradiation, engraftment is not seen. Another bias of chimeric mice is
the introduction of cells into the peripheral circulation that may not normally leave the
bone marrow. A study using parabiotic model of donation found no engraftment of
donor bone-marrow cells even with irradiation and injury (Ajami et al. 2007). Thus,
populations of bone marrow cells may have the ability to take on a microglia phenotype,
other sources may be responsible for microglia turnover in the normal brain.
Due to the limited number of microglia isolated from adult animals, much of the in vitro
work on microglia biology has been done using microglia harvested from mixed brain
cultures of neonatal rodents. However, based on the studies discussed above, these cells
are clearly different than those encountered in the adult brain. The response of ameboid
microglia and mature microglia differ in in vivo models of injury (Milligan et al. 1991)
and inflammation (Lawson and Perry 1995). There are also noted differences in the
antigen presenting abilities and cell surface phenotypes of cells isolated from postnatal
and adult rodents (Carson et al. 1998). Although microglia cells isolated from adults do
not readily proliferate, they may provide a more accurate portrayal of microglia activity
in the adult brain and in neurodegenerative disease.
17
Microglia Activation
In response to brain injury and disease, adult microglia shift from their ramified, resting
state to a hypertrophic, activated state characterized by the retraction of distal processes
and an enlarged cell body (Kreutzberg 1996; Streit 2002). In general, microglia can
respond to inciting stimuli with changes in cell surface markers and the secretion of a
wide variety of molecules including pro- and anti-inflammatory cytokines, chemokines,
and growth factors as well as reactive oxygen species and exitotoxic molecules (Streit et
al. 1988). Activated microglia can also exhibit an increased proliferative ability and high
motility along chemotactic gradients, resulting in rapid focal accumulations of microglia
at sites of injury (Streit et al. 1988). Once characterized by a programmed, stereotypic
response, microglia activation is now seen as a more complex and adaptive process,
responding to various stimuli and microenvironments in distinctive ways (Schwartz et al.
2006).
This study focuses on the specific activation of microglia in response to MPTP, which is
thought to be due to neuronal cell death and is not associated with a florid immune
response. Another well studied in vivo model of microglia activation in response to
neuronal damage is facial nerve axotomy. This model involves the transection of the
facial nerve at the level of the stylomastoid foramen, which leads to both degeneration of
motor nuclei and a pronounced glial response at the level of the brainstem (Streit and
Kreutzberg 1988). Thus, neurodegeneration is achieved without direct trauma to the
CNS and disruption of the blood brain barrier. Microglia activation in this model is
18
characterized by a grade response with early increases in expression of immunogenic
markers like complement receptor-3 and MHC Class I, rapid proliferation of microglia
(peaking at 2 to 4 days), and nodules of phagocytic microglia surrounding degenerating
neurons (Raivich et al. 1999). CNS injury consisting of a full immune response like
infection, trauma, and auroimmune disease exhibits a considerable increase in MHC class
II and iNOS expression that is thought to be mediated by the secretion of IFN" by
infiltrating lymphocytes (Raivich et al. 1999).
Neuronal-Microglia Interactions
Part of the molecular mechanisms regulating microglia activation are signals received
from neurons. And, in the MPTP model used in this study, microglia activation is
believed to be in response to neuronal degeneration (Gao et al. 2003). van Rossum and
Hanisch introduced the concept of ‘On’ and ‘Off’ signals for neuronal inputs on
microglia activity; in this paradigm, the resting microglia state is actively maintained by
‘Off’ signals constitutively expressed by healthy neurons, while separate neuronal ‘On’
signals engage microglia in an activated role (van Rossum and Hanisch 2004). Important
ligand-receptor pairs have been recently identified as candidates for neuronal ‘Off’
signals include both CD200/CD200R (Hoek et al. 2000; Wright et al. 2000) and
CX3CL1(fractalkine)/CXCR1 (Cardona et al. 2006; Bessis et al. 2007). Both receptors
are found on microglial membranes, and disruption of ligand-receptor pairing (either in
ligand-deficient mice or by antibody blocking) leads to a more activated morphology of
microglia and higher levels of microglia activity in disease models. Microglia also have
19
various neurotransmitter receptors including those for glutamate, GABA, noradrenaline,
purines and dopamine that they may use to monitor neuronal activity (Pocock and
Kettenmann 2007). This is supported by reports that ion channel activity can influence
activation in microglia (Farber and Kettenmann 2006).
‘On’ signals serve as neuronal requests for microglia assistance and so far include
purines, chemokines, glutamate, and matrix metalloproteinases (MMPs). Purines like
ATP and UTP can be released by damaged neurons (Wang et al. 2004; Koizumi et al.
2007). Both in vitro and in vivo studies show that ATP activates microglia mainly
through P2Y12 receptors which are actually down-regulated with activation, suggesting
the response to ATP occurs early after neuronal injury (Inoue 2006). Neurons can also
release chemokines like CCL21 which is able to induce microglia migration in vitro (de
Jong et al. 2005). MMPs are proteolytic enzymes that degrade extracellular
macromolecules and can be leaked from damaged neurons. MMP-3 in particular has
been implicated in the activation of microglia in response to dopaminergic cell death and
can activate they can be secreted by neurons and activate the secretion of TNF!, IL-1#,
IL-1r! in microglia cultures (Kim et al. 2007).
MPTP Neurotoxicity
The mouse MPTP model was used in this study as a model of AIF-1 response to
neurotoxicity in vivo. The toxic effects of MPTP were originally discovered in a small
population of drug abusers who had taken a street synthetic opioid contaminated with
20
MPTP and began suffering from rapidly progressive Parkinsonian symptoms (Langston
et al. 1983). MPTP causes relatively selective dopaminergic neurodegeneration, and since
its discovery has been used as an important animal model for Parkinson’s disease in both
mice (Heikkila et al. 1984) and nonhuman primates (Burns et al. 1983; Kolata 1983).
The MPTP mouse model is the most widely used model for studying the cellular and
molecular events of dopaminergic cell death. Disease progression is dependent on mouse
strain (Sundstrom et al. 1987), age, (Sershen et al. 1985), and sex (Freyaldenhoven et al.
1996), as well as route of administration (Gerlach and Riederer 1996), and dosing
regimen. The regimen used in our studies (four i.p. injections of 20mg/kg, each 2 hours
apart) has been well established in C57Bl mice (Jackson-Lewis et al. 1995; Jakowec et al.
2004). This is often referred to as an ‘acute’ regimen, as other “chronic” regimens utilize
lower daily doses over an extended period of time. In this regimen, neuronal cell death is
seen as soon as 12 hours after the last injection and continues up to 4 days as visualized
through combinations of with loss of staining for the enzyme tyrosine hydroxylase (TH),
Nissl staining for neuronal material and silver staining to identify degenerating neurons
(Jakowec et al.). The MPTP mouse model consists of many features seen in human PD
including oxidative stress, mitochondrial dysfunction, and a pronounced glia response. In
addition, mice can exhibit motor deficits similar to those seen in human PD patients
including akinesia and gait disturbances (Sedelis et al. 2001). Of note, the protein
aggregates called Lewy bodies seen in PD are not seen in MPTP mice (Shimoji et al.
2005).
21
MPTP is highly lipophilic and can freely cross the blood brain barrier (Markey et al.
1984). Oxidation of MPTP by monoamine oxidase B (MAO-B) results in the production
of 1-methyl-4-phenyl-2,3-dihydropyridium (MPDP+) which is then further oxidized to
the active toxin 1-methyl-4-phenylpyridium (MPP+), probably by spontaneous oxidation
(Chiba et al. 1984; Langston et al. 1984; Nicklas et al. 1985). In the CNS, MAO-B is
found in astrocytes and serotonergic neurons (Sundstrom and Jonsson; Westlund et al.
1985; Brooks et al. 1989). MPP+ is a high specificity ligand for the dopamine transporter
and entry into dopaminergic cells is an active process through this transmembrane
receptor (Javitch et al. 1985).
The exact sequence of events leading to neuronal death in the MPTP model are still being
worked out, as several of the factors implicated overlap in cause and effects including
mitochondrial dysfunction, oxidative stress, and inflammation. Upon entering the cell,
MPP+ can accumulate in mitochondria (Ramsay et al. 1986; Ramsay and Singer 1986)
leading to the inhibition of complex I, depletion of ATP, and the generation of reactive
oxygen species (ROS), especially superoxide anions (Ramsay et al. 1987; Hasegawa et
al. 1990; Chan et al. 1991; Tipton and Singer 1993). MPP+ can also lead to the release of
dopamine from cellular stores and auto-oxidation of free dopamine can lead to the
formation of hydrogen peroxide and reactive quinones (Asanuma et al. 2004). Another
major source of ROS is microglia, which, upon activation upregulate iNOS and NADPH
whose products (nitric oxide and superoxide, respectively) lead to the formation of
peroxynitrite, another major culprit of oxidative damage.
22
Several in vivo studies have used changes in morphology in combination with increased
staining with either isolectin-B4 (Francis et al. 1995; Kurkowska-Jastrzebska et al. 1999;
Wu et al. 2003) or antibodies against CD11b/Mac-1/CR-3 (Langston et al. 1999) to
demonstrate microglial activation in response to MPTP intoxication in the rodent.
Microglia activation in the MPTP model occurs early. Liberatore et al followed Mac-1
levels in the midbrain by immunoblot and found an increase as soon as 12 hours after the
last of four MPTP injections, with a 4-fold peak at 2 days and no difference from controls
by Day 7 (Liberatore et al. 1999). They described the striatum as having a “similar
trend”, though that data was not shown. Immunohistochemistry showed a robust
microglia activation throughout the SNpc at 24 hours with increased Mac-1 staining and
an activated morphology. In their study, the response of astrocytes (measured by
increased GFAP expression) occurred later than microglial activation and persisted
longer. Wu et al looked at transcript levels of Mac-1 from 0-14 days after injection and
found that expression peaked at Day 2 and returned to baseline levels by Day 7 (Wu et al.
2003). IHC at Day 2 show similar staining to ours with Mac-1 antibody. In mixed
cultures, microglia are not directly reactive to MPTP/MPP+ and only become activated
after substantial neuronal damage (Liberatore et al. 1999; Jakowec et al. 2004). In vivo,
neuronal degeneration have been described as early as 12 hours after MPTP intoxication.
The mechanisms by which dopaminergic neuronal damage triggers activation in
microglia are largely unknown, although activated matrix metalloproteinase-3 has been
recently suggested (Sriram et al. 2002) and lack of signals as described above may also
be responsible. However microglia become activated, there is substantial evidence for
23
their further contribution to neuronal damage through the generation of reactive oxygen
species such as nitric oxide (Gao et al. 2003) and superoxide (Kim et al. 2000). In fact,
microglial activation via lipopolysaccharide can lead directly to neuronal loss in the
dopaminergic system (Czlonkowska et al. 1996,; Kohutnicka et al. 1998) and potentates
MPTP lesioning.
Parkinson’s Disease
The overall aim of this study is to identify the role of AIF-1 as an extracellular molecule,
particularly in the context of the microglia activation seen in neurodegenerative diseases
like PD. PD is a progressive neurodegenerative disease characterized by bradykinesia,
resting tremor, and rigidity that was first described in 1817 by James Parkinson
(Parkinson 2002). It is the second most common neurodegenerative disease behind
Alzheimer’s with a prevalence nearing 1% of people over 60 years of age (de Lau and
Breteler 2006). Incidence rates are estimated at 20/100,00 overall, rising to 120/100,000
in people over 70 years old (Dauer and Przedborski 2003). PD is distinguished from
other Parkinsonian Syndromes when there is no identifiable cause like drugs or trauma
(Secondary Parkinsonism) and the symptoms are not part of broader syndromes
(Parkinson-plus Syndromes) (Stacy and Jankovic 1992). A minority of PD patients show
familial inheritance, currently estimated at 10%, (de Lau and Breteler 2006), and
mutations in a handful of genes have been identified in these families. However, the
initiating cause of sporadic PD has not been identified and may be due to complex
reactions between environmental exposures and genetic predispositions.
24
The motor symptoms seen in PD are the result of a dramatic loss of dopaminergic
neurons in the nigrostriatal system, which is involved in the production of movement.
Neurons in the nigrostriatal system are located within the SNpc and project to the
striatum. Loss of nigrostriatal neurons associated with PD is easily identifiable on gross
examination of postmortem brains due to a loss of pigmented melanin-containing cell
bodies in the SNpc (Marsden 1983). Terminals in the striatum are lost to an even greater
extent than cell bodies, accompanied by a large decline in striatal dopamine.
Neurodegeneration is progressive and onset of symptoms begins only after a 60% loss of
neurons and an 80% loss of striatal dopamine (Dauer and Przedborski 2003). The
mesolimbic system, the other main dopaminergic system whose cell bodies lie in the
adjacent ventral tegmental area (VTA), is relatively spared (Uhl et al. 1985).
Replenishment of striatal dopamine via oral administration of the dopamine precursor
levodopa is still the mainstay of clinical management of PD (Jankovic 2008). Dopamine
agonists can also alleviate motor symptoms and, although they are not as potent as
levodopa, they are not associated with the motor fluctuations and dyskenisias developed
with levodopa treatment. With progression of the disease these treatments are not well
tolerated. Thus, current emphasis is being placed on therapies to slow, stop, or reverse
neuronal degeneration. Although some neuroprotective trials have been promising, none
have clearly shown neuroprotection at the cellular level (Marras and Lang 2008).
Activated microglia have been identified in post mortem samples from patients with
Parkinson’s disease (PD) (Mogi et al. 1994; Imamura et al. 2003) and in humans with
25
Parkinsonian symptoms caused by accidental self-administration of the MPTP neurotoxin
(Muller et al. 1998). Interestingly, in activated microglia have been seen up to 16 years
after MPTP-intoxication in these patients, suggesting a persistent microglia activation
that may be further contributing to neuronal death (Langston et al. 1999). Numerous
cytokines are also found to be elevated in the brains and cerebral spinal fluids of PD
patients including TNF-!, interleukin-1#, -2, -4, -6, transforming growth factor-!, and
transforming growth factor- #1 and #2; and microglia are thought to be the primary
source of these cytokines (Mogi et al. 1994; Mogi et al. 1994; Mogi et al. 1995; Nagatsu
et al. 2000). However, the exact role of microglia in the progression of PD is still being
defined, mostly in animal models like the MPTP mouse model.
26
Chapter 2: The AIF-1 response to MPTP neurotoxicity
Introduction
Resident microglia change from a resting state to a hypertrophic, activated state in
response to brain injury and disease (Streit et al. 1988). Microglia activation is
characterized by proliferation, expression of immunogenic cell surface markers, and
secretion of cytokines, chemokines, and growth factors (Streit et al. 1988; Banati et al.
1993). Activated microglia have been identified in postmortem samples from patients
with PD (McGeer et al. 1988; Banati et al. 1998; Mirza et al. 2000; Imamura et al. 2003;
Nagatsu and Sawada 2007) and from patients with MPTP-induced Parkinsonism
(Langston et al. 1999). However, the exact role of microglia in the progression of PD
and related disorders is still being defined. MPTP is a drug contaminant that selectively
targets dopaminergic neurons of the nigrostriatal pathway. Microglia activation has been
noted in both the SNpc and striatum of mice in the MPTP mouse model. The well-
established acute regimen of i.p. MPTP administration results in 60-70% loss of
dopaminergic neurons in the SNpc within four days (Jakowec et al. 2004). The maximal
microglia response using this regimen was also reported to occur within four days
(Kurkowska-Jastrzebska et al. 1999; Liberatore et al. 1999). In mixed cell cultures,
microglia are not directly reactive to MPTP or its active metabolite 1-methyl-4-
phenylpyridium (MPP+), and only become activated after substantial neuronal damage
(Gao et al. 2003). Thus activation in this model is considered to be in response to
neuronal death and not due to any effects of the neurotoxin directly on microglia.
27
Allograft inflammatory factor-1 (AIF-1) is a 17 kDa Ca
2+
-binding protein that is
constitutively expressed by microglia and macrophages, and is markedly upregulated in
response to inflammatory stimuli (Utans et al. 1995) and dying neurons (Tanaka et al.
1998). AIF-1 is identical in sequence to ionized calcium-binding adapter molecule 1
(Iba1), which is commonly used as a marker for microglia and macrophages (Imai et al.
1996). Increased staining for AIF-1/Iba1 has been noted in several models of injury and
infection in the CNS (Schluesener et al. 1998; Schluesener et al. 1999; Deininger et al.
2000; Ito et al. 2001; Mori et al. 2003), including rat facial nerve axotomy, a well-studied
in vivo model of the microglia response to neuronal degeneration (Ito et al. 1998). In
vitro evidence also supports a role for intracellular AIF-1 in the cytoskeletal remodeling
required for microglial migration and phagocytosis (Ohsawa et al. 2000; Imai and
Kohsaka 2002; Kanazawa et al. 2002; Autieri et al. 2003; Ohsawa et al. 2004).
Intracellular AIF-1 may also contribute to the ability of microglia/macrophage to
proliferate, secrete cytokines, and generate reactive oxygen species.
In this study we examined the AIF-1 response to neuronal death in vivo using the well-
established MPTP neurotoxicity mouse model. AIF-1 expression was analyzed over an
extended time course and cells with greatly increased immunoreactivity for AIF-1 were
seen in the SNpc of lesioned animals in increasing numbers in the first 10 days. Because
AIF-1 is also expressed by peripheral macrophages, which are known to be recruited to
sites of neural injury, we carried out a thorough series of immunohistochemical co-
28
staining and flow cytometric analyses to determine that resident microglia were the
primary source of this increased AIF-1 expression in the SNpc of MPTP-lesioned mice.
Materials & Methods
Animals and MPTP Lesioning. Young adult (8 weeks) male C57BL/6 mice, supplied by
Jackson Laboratory (Ben Harbor, ME), were used for all experiments. Neurotoxicity was
induced with four intraperitoneal injections of 20mg/kg MPTP (Sigma, St. Louis, MO),
each two hours apart. This regimen leads to approximately 60-70% loss of nigrostriatal
neurons within four days following MPTP-lesioning (Jackson-Lewis et al., 1995). Saline
injected mice served as controls. Whole brains of mice were collected 1, 2, 3, 5, 7, 10, 21
or 28 days after MPTP-lesioning. First, mice were anesthetized by intraperitoneal
administration of phenobarbital (100µL of 10mg/ml) and then mice were transcardially
perfused with 50ml of ice-cold 0.9% saline. For immunohistochemistry, mice were then
perfused with 50ml of 4% paraformaldehyde in phosphate buffered saline, pH 7.3 (4%
PFA/PBS).
Immunohistochemistry. Whole brains were immersion-fixed in 4% PFA/PBS overnight
at 4°C, cryoprotected in 20% sucrose for 24 to 48 hours, frozen in ice cold isopentane
and stored at -80°C until analyzed. Coronal sections (30µm thick) were cut using a Leica
CM1900 microtome. A set of every sixth section from each mouse was stained with
rabbit polyclonal antibody specific for AIF-1/Iba1 (1:000; Wako Chemicals, Richmond,
VA). A separate set of every sixth section was stained with rabbit monoclonal antibody
specific for tyrosine hydroxylase (1:1000; Chemicon, Temecula, CA) to evaluate
29
dopaminergic loss. First, free-floating sections were rinsed in 0.1M Tris-buffered saline
(TBS) and then quenched with 10% methanol and 3% H202 in TBS. Nonspecific
staining was blocked with 4% normal goat serum (Vector Laboratories, Burlingame, CA)
for 1 hour at room temperature. After staining with primary antibody for 48 hours at
4°C, sections were washed repeatedly in TBS and then incubated with either biotinylated
goat anti-rabbit IgG (1:500; Vector Laboratories) or Alexa Fluor 488 goat anti-rabbit IgG
(1:1000; Invitrogen, Eugene, OR). For visualization of biotinylated secondary
antibodies, incubation with Avidin-HRP enzyme (BD Biosciences; San Jose, CA) at a
1:1000 dilution was followed by processing using the Vectastain DAB kit (Vector
Laboratories). Sections were set on subbed slides and fluorescently stained sections were
mounted with Vectorshield Hard Mount with DAPI (Vector Laboratories) while
remaining slides were dehydrated, cleared in xylene, and mounted using Permamount.
Staining was visualized using an Olympus BX51 microscope. Dual staining experiments
were performed on remaining sections from mice sacrificed 7 or 10 days after MPTP-
lesioning using methods similar to those described above. For these studies, sections
were simultaneously incubated with rabbit polyclonal AIF-1 (1:1000) and a mouse
monoclonal antibody specific for either tyrosine hydroxylase (TH) (1:1000; Millipore,
Billerica, MA), glial fibrillary acidic protein (GFAP) (1:1000; Sigma) or CD11b (1:000;
AbD Serotec, Raleigh, NC). Secondary antibodies were also incubated together and
consisted of Alexa Fluor 488 goat anti-rabbit IgG (1:1000, Invitrogen) and Alexa Fluor
594 goat anti-mouse IgG (1:1000; Invitrogen).
30
Microglia Isolation. Single cell suspensions of adult mouse microglia were prepared by a
protocol adapted from Frank, et al (Frank et al. 2006). Briefly, mononuclear cells were
purified from mechanically homogenized suspensions of brain by density gradient
centrifugation. All steps were performed using asceptic techniques. Dissected regions of
striatum, ventral midbrain or cortex were homogenized in Hank’s Buffered Salt Solution
(HBSS, Mediatech, Herndon, VA) supplemented with 25mM HEPES using a Tenbroeck
homogenizer on ice. 100% isotonic Percol was made by diluting stock Percol
(Amersham Biosciences, Uppsala, Sweden) with 1/10th the volume of 10xPBS. Further
dilutions of 100% Percol were made with HBSS. Tissue homogenate was collected by
centrifugation at 300g for 7min at room temperature, washed once in HBSS, and then
resuspended in 70% isotonic Percol. Whole brains were resuspended in a total of 10ml of
70% Percol and split into two 15ml tubes, while dissected regions were resuspended in
5ml 70% Percol and transferred to a fresh 15ml tube. This cell mixture was then
carefully overlayed with 4ml of 50% Percol, followed by an additional overlay of 2ml of
HBSS. This density gradient was centrifuged for 40min at 200g at 18°C with no brake.
Mononuclear cells were collected from the interface between the 70% and 50% isotonic
Percol layers, washed with HBSS, and prepared for flow cytometry or ex vivo cultures.
Microglia for ex vivo cultures were resuspended in either Macrophage SFM (Invitrogen)
or culture medium consisting of RMPI medium 1640 supplemented with 10% FBS,
2.5mM L-glutamine, 1mM sodium pyruvate, 0.1mM non-essential amino acid, 50U/ml
streptomycin, 50U/ml penicillin and 0.25µg/ml amphotercin B. and seeded at 1x10
6

31
cells/ml in 96 well plates. All reagents used for tissue culture were obtained from
GIBCO® Invitrogen unless otherwise stated.
Flow cytometry. Three-color flow cytometry was performed according to standard
protocols. Isolated microglia were washed three times in FACS buffer (PBS
supplemented with 2% normal calf serum and 0.1% sodium azide) and blocked for 15min
at 4°C with both 5% normal goat serum to block nonspecific binding and CD16/32 (BD
Biosciences) to block Fc receptors. Cells were washed again in FACS buffer and then
incubated with combinations of FITC labeled anti-CD45 (BD Biosciences), and PE-Cy5
labeled anti-CD11b (BD Biosciences) for 15min at 4°C in the dark. A fraction of the
samples were incubated with equal concentrations of isotype matched control antibodies
to account for nonspecific binding and autofluorescence. Cells were fixed in 4%
PFA/PBS for 15min at 4°C, washed once in FACS buffer and stored overnight at 4°C.
For intracellular staining with AIF-1, cells were permeabalized the following day using
the BD Pharmingen Cytofix/Cytoperm kit (BD Biosciences) according to manufacturer’s
protocol. Another blocking step exactly as described above was included after
permeabalization. Cells were then incubated with unlabelled rabbit polyclonal antibody
specific for Iba1/AIF1 (Wako Chemicals) for 60min at 4°C, washed twice in
permeablization buffer and then incubated with PE labeled anti-rabbit IgG for 20min at
4°C. Washed cells were analyzed using a FACSCalibur flow cytometer (Becton
Dickinson, San Jose, CA). Data was further analyzed using the WinMDI software.
32
Results
To characterize the AIF-1 response to dopaminergic neuronal loss, AIF-1 expression was
assessed immunohistochemically both at various early time points following MPTP
lesioning (1, 2, 5, 7, and 10 days) and also at later time points characterized by repair and
regeneration (21 and 28 days) (Bezard et al. 2000; Jakowec et al. 2004). In saline control
animals, AIF-1 immunoreactivity was regularly distributed throughout the brain in
parenchymal cells possessing morphologic features characteristic of microglia including
small cell bodies and thin extensions (figure 2.1A & 2.1B) and also in ameboid cells
located in the meninges and choroid plexus; sites of CNS-associated macrophage
populations (data not shown). In contrast, following MPTP lesioning, cells with high
intensity staining for AIF-1 (termed AIF-1 bright cells) could be continually observed in
the SNpc throughout the first ten days (figure 2.1C-1I). These AIF-1 bright cells were
not observed on day 21 (figure 2.1J), day 28, or elsewhere in the brain (data not shown).
In contrast to saline treated mice (figure 2.1B), the cell bodies of AIF-1 bright cells in the
SNpc of MPTP-treated mice were hypertrophied with short, thick processes and were
often found in clusters (figure 2.1H). The area (figure 2.2A) and average density (figure
2.2B) of the cell bodies of AIF-1 bright cells in the SNpc were measured using Bioquant
software and compared to AIF-1 positive cells in the SNpc of saline control animals. The
cell bodies of AIF-1 bright cells were significantly larger than those of resting microglia
(figure 2.2A: AIF-1 bright cells: 1831±254 units, resting: 500±90 units; p<0.001).
Concordant with this increase in cell body size was a significant increase in fluorescence
33
Figure 2.1 AIF-1 expression in the SNpc following MPTP lesioning. Coronal sections
of SNpc from mice sacrificed after MPTP injection at the time points indicated were
stained with antibody specific for AIF-1 and visualized with a FITC-conjugated
secondary antibody. Specific AIF-1 stained cells are indicated with the small arrow and
AIF-1 bright cells are indicated with the large arrow in C. The white squares in A and G
outline the fields shown in the higher magnification images shown in B and H,
respectively. Scale bars represent 200mm (A,C-G,I, J) or 20mm (B,H). Images are
representative of 2-3 animals for each time point.
34
Figure 2.1, Continued

35

Figure 2.2 Area and average density are increased in AIF-1 bright cells. The area
(A) and average density (B) of cell bodies of individual AIF-1 bright cells in the SNpc of
saline control mice, were measured using Bioquant software. Open circles represent the
values for individual cells (n=15). Error bars represent the mean ± standard cells in SNpc
of day 7 MPTP mice, and AIF-1+ deviation. Simple student t-test of saline versus MPTP
was carried out for each group.
36
intensity of AIF-1 staining in MPTP treated animals (figure 2.2B: AIF-1 bright
cells:104±4 units, resting: 54±3 units; p<0.001). In addition to assessing the change in
size and intensity of staining of AIF-1 bright cells we also enumerated these cells in
coronal sections containing the SNpc. there was a progressive significant increase in the
number of AIF-1 bright cells in the SNpc over the first 10 days as compared to saline
(figure 2.3). Only a few isolated AIF-1 bright cells were observed in saline control mice
(0.111±0.111 cells) or mice 21-28 days after MPTP lesioning (0.667±0.667 cells) as
compared to 34±4 cells in the first 2 days, 56±12 cells at days 3-5, and 91±10 cells at
days 7-10.
We also examined AIF-1 expression in the striatum, where MPTP-lesioning results in
loss in terminal projections of dopaminergic neurons. In contrast to our observations in
the SNpc, AIF-1 bright cells were not seen in the striatum. Rather, there was a more
subtle increase in AIF-1 immunoreactivity in parenchymal microglia at early time points
(figure 2.4).
So as to ensure the integrity of our data, all mice used for this study were simultaneously
assessed for MPTP-induced nigrostriatal damage by immunohistochemical analyses of
striatal TH, a common biomarker for dopaminergic neuron integrity. Saline-treated mice
(figure 2.5A) were always shown to have high levels of TH-immunoreactivity in contrast
to MPTP mice which showed progressive reductions in the levels of TH staining at 1 day
(figure 2.5B), 7 days (figure 2.5C) and 28 days (figure 2.5D) after lesioning.
37

Figure 2.3 Number of AIF-1 bright cells in substantia nigra following MPTP
lesioning. AIF-1 bright cells were counted with Bioquant software in the SNpc in fields
of view as shown in Fig 2. A minimum of three individual fields were counted for each
animal and 2-3 animals were analyzed per time point. One way ANOVA analysis was
followed by pairwise multiple comparison (Holm-Sidak) between time points.
Significant differences between the different groups are shown by the following symbols:
* = p<0.05 vs. saline; # = p<0.05 vs. Days 21-28; † = p<0.05 vs. Days 1-2; ‡ = p<0.05
vs. Days 3-5.
38

Figure 2.4 AIF-1 expression in the striatum following MPTP lesioning. Coronal
sections of striatum from mice sacrificed at given time points after MPTP injection were
stained with antibody for AIF-1 and visualized with a FITC-conjugated secondary
antibody. Images shown are representative of 2-3 animals analyzed for each time point;
scale bars represent 50!m.
39

Figure 2.5 Loss of TH staining in the striatum of MPTP lesioned animals. Coronal
sections containing the midstriatum of MPTP lesioned mice were stained with a
polyclonal antibody specific for TH. Staining was visualized with DAB substrate and is
clearly visible in figure 5A as the dark stain in the center of the section. Images shown
are representative of 2-3 animals per time point; scale bars represent 50!m.
40
Limited studies have shown cell types other than microglia/macrophages are capable of
expressing AIF-1 in pathological conditions (Autieri et al. 2000; Deininger et al. 2003).
Thus, we carried out dual staining of SNpc of mice taken 10 days after MPTP lesioning.
Glial fibrillary acidic protein (GFAP) was used as a marker for astrocytes and TH as a
marker for dopaminergic neurons to ensure these cell types were not expressing AIF-1.
The SNpc showed increased staining for GFAP (figure 2.6B; red fluorescence); these
cells were similar in distribution to AIF-1 bright cells (figure 2.6A; green fluorescence),
but represented a distinct population in a merged image (figure 2.6B) as clearly seen at
higher magnification (figure 2.6D). Similarly, TH staining of surviving dopaminergic
neurons (figure 2.6F; red fluorescence) clearly identified that neurons were also distinct
from AIF-1 bright cells (figure 2.6G; green fluorescence) as seen in the merged image at
both low (figure 2.6G) and high magnifications (figure 2.6H). In fact, TH-positive
neurons and AIF-1 bright cells appeared to occupy different subregions of the SNpc. In
contrast, AIF-1 bright cells found in mice 10 days after treatment with MPTP (figure
2.6I,K,L; green fluorescence) did co-stain with antibodies for CD11b (figure 2.6J,K,L;
red fluorescence), a widely used marker for microglia and macrophages. Thus, in
response to MPTP lesioning AIF-1 expression is specific to cells of myeloid origin.
Because both activated microglia and infiltrating macrophages or neutrophils can express
CD11b and AIF-1 (Springer et al. 1978; Utans et al. 1995), flow cytometric analysis was
used to determine whether the increase in AIF-1 bright cells indicated an increase of AIF-
1 expression by resident microglia or the infiltration of peripheral mononuclear cells.
Peripheral leukocytes have higher CD45 staining (CD45
hi
) compared to resident
41
Figure 2.6 AIF-1 bright cells express CD11b, but not GFAP or TH. Coronal sections
of SNpc from mice sacrificed 10 days after MPTP injection were co-stained with
antibody specific for AIF-1 (figures A, E & I) and either GFAP (figure B), TH (figure
F), or CD11b (figure J). Double stained images are shown at both low and high
magnification respectively for AIF-1 & GFAP (figures C & D); for AIF-1 & TH (figures
G & H) and for AIF-1 & CD11b (figures K & L). Scale bars represent 200!m (figures
A-C, E-G & I-K) or 50!m (figures E, H & L).
42

Figure 2.6, Continued

43
microglia (CD45
lo
), which can also upregulate their CD45 expression in response to
inflammatory signals or infection (CD45
int
) (Sedgwick et al. 1991). To determine the
extent of infiltrating macrophages in MPTP-lesioned animals, mononuclear cells were
isolated from the ventral midbrain, which contains the SNpc, of MPTP-treated mice 10
days after lesioning. Mononuclear cells were also isolated from the striatum and cortex
of the same mice to serve as control regions without AIF-1 bright cells visible by
immunohistochemistry. Cells were stained with antibodies to CD45, CD11b, and AIF-1
and analyzed by three-color flow cytometry. Representative dot plots of cells isolated
from the midbrain are shown in figures 2.7A-C; similar profiles were seen in all three
regions (data not shown). Analysis of forward scatter and side scatter dot plots (figure
2.7A) showed two distinct populations of mononuclear cells which were gated away from
other debris prior to analyses of phenotype as shown (figure 2.7A; R1). The vast majority
of cells recovered from the SNpc expressed low levels of CD45 (CD45
lo
) and co-
expressed AIF-1 (figure 2.7B; box i). A small population of cells expressing high levels
of CD45 (CD45
hi
) and co-expressing AIF-1 could also be seen in each region (figure
2.7B; box iii). Also, consistent with what was seen by immunohistochemistry, almost all
(>97%) AIF-1
+
cells also stained with antibodies to CD11b (figure 2.7C). The
expression of CD45 in AIF-1 positive cells was compared across the three brain regions
and the percentage of AIF-1
+
cells that exhibited CD45
lo
, CD45
int
and CD45
hi
expression
levels are given in Table 2.1. Regardless of the brain region, CD45
lo
microglia made up
the greatest percentage of AIF-1 positive cells (cortex: 97.6%±0.5%; striatum:
96.6%±1.1%; midbrain: 93.6%±1.1%). However, the percentage of CD45
lo
cells isolated
44

Figure 2.7 Flow cytometric analysis of mononuclear cells isolated from MPTP lesionined mice. Mononuclear
cells were isolated from the midbrain of mice 10 days after MPTP lesioning and analyzed by three-color flow
cytometry with anti-AIF-1, anti-CD11b, and anti-CD45. Mononuclear cells were gated (R1) from cell debris based on
forward scatter (FSC) side scatter (SSC) profiles (A) and then gated into subsets based on CD45 expression (B). A
small population of cells expressing higher levels of AIF-1 were gated based on CD45 vs. CD11b dot plot (C) and
analyzed separately. Percentage of all AIF-1 positive cells and AIF-1hi cells that were CD45lo, CD45int or CD45hi
are given in Tables 1 and Table 2, respectively. Plots are representative for three animals.

45
from the midbrain was significantly lower than that of cells isolated from the cortex
(p<0.05), but not significantly different than the percentage of CD45
lo
cells isolated from
the striatum. There was also a significant increase in the percentage of CD45
hi
peripheral
mononuclear cells isolated from the midbrain (3.8%±0.8%) compared to the cortex
(1.5%±0.5%, p<0.05). There were no significant differences in the percentage of CD45
int

activated microglia isolated from each region. Since there was no significant increase in
the percentage of CD45
hi
cells isolated from the midbrain compared to the striatum,
which had no AIF-1 bright cells, the infiltration of peripheral leukocytes could not fully
explain the presence of AIF-1 bright cells in the midbrain of MPTP-lesioned mice.

Table 2.1 CD45 expression in AIF-1
+
cells isolated from MPTP-lesioned mice

CD45
lo
CD45
int
CD45
hi

Cortex 97.6% ±0.5% 0.9% ±0.1% 1.5% ±0.5%
Striatum 96.6% ±1.1% 1.1% ±0.2% 2.3% ±0.9%
Midbrain 93.6% ±1.1%* 2.2% ±0.8% 3.8% ±0.8%*

Mononuclear cells isolated from the cortex, striatum, and midbrain of mice 10 days after
MPTP-lesioning were analyzed for AIF-1 and CD45 expression using flow cytometry.
All AIF-1 positive cells were gated and the percentage of cells with CD45
lo
, CD45
int
, or
CD45
hi
expression levels were calculated. Data are presented as mean ± SEM (n=3).
*p<0.05 compared to cortex region.

A small population of AIF-1
+
/CD11b
+
cells showed slightly elevated levels of AIF-1
staining (figure 2.7C; box iv). The CD45 expression levels of these cells were analyzed
separately and the percentage of AIF-1
hi
cells with CD45
lo
, CD45
int
, and CD45
hi

46
expression levels are given in Table 2.2. The vast majority of AIF-1
hi
cells were also
CD45
lo
(cortex: 93.8%±0.2%;; striatum: 92.6%±2.7%; midbrain: 88.9%±3.6%),
indicative of resting microglia. There were no significant differences in the percentage of
the three CD45 subtypes when different regions were compared, although for all three
regions, there was a greater percentage of CD45
int
cells (cortex: 3.9%±2.6%; striatum:
4.9%±2.1%; midbrain: 9.3%±5.5%) than CD45
hi
cells (cortex: 2.4%±1.2%; striatum:
3.1%±2.1%; midbrain: 2.2%±0.6%), which is in contrast to what was seen when all AIF1
positive cells were analyzed, suggesting that CD45
int
cells may express higher AIF-1
levels.

Table 2.2 CD45 expression in AIF-1
hi
cells isolated from MPTP-lesioned mice
CD45
lo
CD45
int
CD45
hi

Cortex 93.8% ±0.2% 3.9% ±2.6% 2.4% ±1.2%
Striatum 92.6% ±2.7% 4.9% ±2.1% 3.1% ±2.1%
Midbrain 88.9% ±3.6% 9.3% ±5.5% 2.2% ±0.6%

Mononuclear cells isolated from the cortex, striatum, and midbrain of mice 10 days after
MPTP-lesioning were analyzed for AIF-1, CD11b, and CD45 expression using flow
cytometry. AIF-1
hi
cells were gated based on CD11b and AIF-1 dot plots as shown in
figure 2.7c and the percentage of cells with CD45
lo
, CD45
int
, or CD45
hi
expression levels
were calculated. Data are presented as mean ± SEM (n=3).

The mean fluorescence intensity (MFI) of AIF-1 staining was also evaluated for the three
different CD45 subsets and values for all three brain regions are given in Table 2.3. No
significant differences were seen in AIF-1 expression between the three regions.
However, for all three regions, CD45
hi
peripheral mononuclear cells (cortex:
47
379.44±6.87; striatum: 361.56±21.25; midbrain: 401.47±68.63) showed a lower MFI
than CD45
lo
(cortex: 474.51±38.17; 492.05±70.6; midbrain: 571.01±99.66) or CD45
int

(cortex: 666.44±236.91; striatum: 705.20±322.77; midbrain: 906.38±174.12) microglia.
In fact, CD45
int
cells (activated microglia) did show the highest MFI for AIF-1, and,
when data from all three brain regions were combined, the MFI for AIF-1 expression in
CD45
int
(759.34±114.16) cells was significantly higher than in CD45
lo
(512.52±39.60)
and CD45
hi
cells (380.82±21.62, p<0.05). Taken together, these data suggest that an
increased expression of AIF-1 by resident microglia was primarily responsible for the
AIF-1 bright cells seen by immunohistochemistry. Infiltrating peripheral cells were not
found in significantly greater numbers in areas with AIF-1 bright cells (midbrain) than in
regions without (striatum), and infiltrating peripheral cells actually expressed lower
levels of AIF-1.
Table 2.3 MFI of AIF-1 expression in AIF-1
+
cells isolated from MPTP-lesioned
mice
CD45
lo
CD45
int
CD45
hi

Cortex 474.51 ±38.17 666.44 ±236.91 379.44 ±6.87
Striatum 492.05 ±70.26 705.20 ±232.77 361.56 ±21.25
Midbrain 571.01 ±99.66 906.38 ±174.12 401.47 ±68.63
Combined 512.52 ±39.60 759.34 ±114.16* 380.82 ±21.62

Mononuclear cells isolated from the cortex, striatum, and midbrain of mice 10 days after
MPTP-lesioning were analyzed for AIF-1, CD11b, and CD45 expression by flow
cytometry. AIF-1 positive cells were gated based on CD45 expression levels (CD45
lo
,
CD45
int
, CD45
hi
) and the mean fluorescence intensity for each group was calculated.
Data from all three brain regions were combined to show statistical significance increase
of AIF-1 expression levels in CD45
int
cells. Data are presented as mean ± SEM (n=3). *,
p<0.05 vs. CD45
lo
and CD45
hi
groups.
48
Discussion
The Iba1 antibody has been used in previous studies simply to identify the activation of
microglia in the MPTP mouse model (Furuya et al. 2004; Yasuda et al. 2007). Here we
report for the first time an extended time course of AIF-1 expression in response to
MPTP-lesioning. Following AIF-1/Iba1 levels by immunohistochemistry, we found a
significant number of cells with greatly increased AIF-1 expression in the SNpc of
lesioned animals (figure 2.1). The number of these AIF-1 bright cells continually
increased in the SNpc through the first 10 days following lesioning (figure 2.3), but were
not seen at later time points. Microglia activation in the acute MPTP model has been
considered to occur early after MPTP-lesioning, and to peak during active neuronal
degeneration (Teismann et al. 2003) which occurs in the first four days (Jackson-Lewis et
al. 1995). Indeed, others have shown by Western blot analysis that Mac-1 protein levels
in the ventral midbrain peaked 2 days after MPTP-lesioning, returning to baseline by 7
days (Liberatore et al. 1999). There is however support in the literature for activation
extending beyond the period of active neuronal degeneration. Using isolectin B4
staining, another group found an increase in the density of microglia that was maximal in
both the striatum and the SNpc in the first four days following lesioning but extended
through day 7-14 in the SNpc (Czlonkowska et al. 1996; Kohutnicka et al. 1998). Each
of these studies relied on a different method for evaluating microglia activation, and this
may be the primary reason for apparent discrepancies. Because we looked solely at AIF-
1 bright cells, and not the total number of AIF-1 immunoreactive cells in the SNpc, the
further increases we see at 7-10 days may reveal an important role for AIF-1 at these later
49
time points. At days 7-10, AIF-1 bright cells were found in clusters (figure 2.1h) which
is highly suggestive of phagocytosing cells (Streit and Kreutzberg 1988). Indeed, in
MPTP-lesioned mice, Koohutnicka et al found microglia containing large, electron dense
lysosomes, indicative of phagosomes, and the number of microglia with these inclusions
greatly increased between day 4 and day 7 (Kohutnicka et al. 1998). AIF-1 has been
shown to be important in the cytoskeletal remodeling required for phagocytosis (Imai and
Kohsaka 2002). An association of increased AIF-1 expression with active phagocytosis
may also explain the lack of AIF-1 bright cells in the striatum where there is no need for
clearance of degenerated cell bodies (figure 2.4). It should be noted that AIF-1 bright
cells in the SNpc also expressed CD11b (which recognizes the same integrin as Mac-1),
thus we are not identifying a new subset of activated microglia/macrophages with this
marker.
Another important objective of this study was to identify which cells were responsible for
the increased expression of AIF-1 seen in the SNpc. Limited studies have shown that
neurons can express AIF-1 under pathological conditions (Deininger et al. 2003), and
there is substantial evidence for the infiltration of peripheral cells in MPTP lesioned
SNpc (Kurkowska-Jastrzebska et al. 1999). Co-staining experiments confirmed neither
astrocytes nor dopaminergic neurons expressed AIF-1 (figure 2.6), but neither AIF-1 nor
CD11b can distinguish between microglia and peripheral cells such as macrophages and
neutrophils. The only well-established method for distinguishing between these cell
types is based on differences in CD45 expression by flow cytometric analysis (Sedgwick
et al. 1991; Ford et al. 1995). Isolated peripheral mononuclear cells express higher levels
50
of CD45 (CD45
hi
) than parenchymal microglia (CD45
lo
). Although activated microglia
can increase expression of CD45 (CD45
int
) levels do not reach that of CD45
hi
peripheral
cells. Analysis of mononuclear cells isolated from the midbrains of mice 10 days after
MPTP-lesioning showed a small population which expressed higher levels of AIF-1
(figure 2.7c). Most of these AIF-1
hi
cells were CD45
lo
or CD45
int
, indicative of
parenchymal microglia. The few CD45
hi
/AIF-1
+
cells that were found in the midbrain
had a lower mean fluorescence intensity of AIF-1 expression than CD45
lo
or CD45
int

cells. Thus, the AIF-1 bright cells that were detected by immunohistochemistry are
resident microglia and not infiltrating leukocytes.
In conclusion, this study demonstrates a clear AIF-1 response in microglia to MPTP
neurotoxicity that, in the SNpc, extended beyond the period of neuronal death, but not
into later stages characterized by repair and regeneration. These kinetics and the
expression in microglia exhibiting clustering activity suggest an important role of AIF-1
in phagocytosis of cell debris.
51
Chapter 3: Secretion of AIF-1 by inflammatory cells
Introduction
The previous chapter demonstrated that AIF-1 expression is greatly increased in
microglia near the cell bodies of dying neurons following MPTP neurotoxicity. A
previous study demonstrated that, in addition to being upregulated, AIF-1 is also secreted
by microglia in co-culture with apoptotic neurons and secretion can be stimulated by
soluble factors released by dying neurons (Tanaka and Koike 2002). When this thesis
was initiated, further evidence for the secretion of AIF-1 was limited. AIF-1 had been
found in the sera of rats with experimental autoimmune neuritis (Pashenkov et al. 2000).
AIF-1 was also found in the supernatants of neuroblastoma cells stressed with hydrogen
peroxide (Deininger et al. 2003). There were only a few studies that explored activities
of purified or recombinant AIF-1 proteins added to cultures or injected into mice, which
are discussed in more detail in following chapters. More recently, AIF-1 has also been
found in the peritoneal fluid of women with endometriosis (Koshiba et al. 2005) and in
the synovial fluid of patients with rheumatoid arthritis (Kimura et al. 2007). In vitro
experiments in these studies showed that both human peritoneal macrophages and
synovial cells could secrete AIF-1. This growing evidence of the secretion of AIF-1
throughout the body is somewhat surprising given that it does not have a classical signal
sequence. Also, the function of AIF-1 intracellularly has been studied in detail (See
background chapter for full review). However, a growing number of proteins have
distinct functions both inside and outside of the cell, and a growing body of literature is
52
detailing the secretion of proteins by nonclassical methods. Both groups include proteins
involved in the inflammatory response of myeloid cells: members of the IL-6 cytokine
family have intracellular functions in addition to their extracellular signaling (Kong et al.
1996; Haines et al. 2000) and IL-1# is secreted by a nonclassical pathway (Qu et al.
2007).
There is then sufficient interest in examining AIF-1 as an extracellular molecule in the
CNS. I decided to approach the question of AIF-1 secretion ex vivo. Experiments were
initiated in both splenocytes and the THP-1 cell line as a means of establishing methods
for measuring AIF-1 levels and strategies for inducing AIF-1 secretion. Supernatants
were monitored for AIF-1 levels by western blotting as ELISA using available antibodies
did not provide adequate sensitivity. This chapter is the first reporting of the secretion of
AIF-1 from adult murine microglia.
Materials and Methods
Animals. Young adult (6-8 weeks) male C57Bl6 mice were used for all experiments.
For cardiac perfusion, mice were anesthetized with pentobarbital and cardiac perfused
with ice cold 0.9% saline. Whole brains and spleens were removed aseptically for
purification of microglia or splenocytes, respectively.
Cell culture of THP-1 cells. THP-1 cells were purchased from ATCC and maintained in
culture medium consisting of RMPI 1640 with 10% FBS, supplemented with 2.5mM L-
glutamine, 1mM sodium pyruvate, 0.1mM non-essential amino acid 50U/ml
53
streptomycin, 50U/ml penicillin and 0.25µg/ml amphotercin B. All cell culture reagents
were obtained from Gibco.
Mouse splenocyte culture. Splenocytes were isolated from whole spleens of saline
perfused mice. Spleens were first washed with RPMI containing 10x
antibiotics/antimycotics (500U/ml streptomycin, 500U/ml penicillin, and 250ng/ml
amphotercin B). A single cell suspension was made in RPMI with 10x
antibiotics/antimycotics using a sterile nylon mesh and the plunger of a 5ml syringe..
Cells were collected by centrifugation at 800 x g for 5 minutes and resuspended in 5ml of
fresh culture medium (as descried for THP-1 cell culture). The cell suspension was
carefully underlayed with 2ml of Lympholyte M (Cedarlane Labs, Ontario, Canada) and
centrifuged at 1200 xg for 20 minutes. Splenocytes were collected from the interface,
washed twice in culture medium, and seeded at 1x10
6
to 4x10
6
cells/ml in 12-well tissue
culture plates.
Microglia cell culture. Microglia were isolated from saline perfused mice using methods
similar to those described in Chapter 2. Because these cells were incubated overnight,
particular attention was paid to aseptic techniques. In addition, microglia were isolated
from whole brains and not from dissected regions. The homogenate from each brain was
resuspended in 7ml of 100% isotonic percoll and made up to 10ml with PBS for a final
70% percoll suspension. Cell homogenates from each brain were split into two 15ml
conical tubes, carefully overlayed with 4ml of 50% isotonic percoll, and again with a
layer of 2ml 1xPBS. After centrifugation, microglia cells were collected from the
54
50%/70% interface and cells from the same brains were pooled into a single 50ml tube.
Cells were washed once with culture medium and then transferred to a 1.5ml
microcentrifuge tube and washed twice more in culture medium. Cells were counted
using a hemacytometer and plated at 1x10
6
cells/ml in 96-well plates.
FACS analysis. Splenocytes or microglia were analyzed by three-color cytometry as
described in Chapter 2 using anti-CD45, anti-CD11b, and anti-AIF-1.
LPS injections. For in vivo stimulation of microglia, mice were injected with a single
dose of 20ug LPS (Sigma) diluted in 0.2ml of sterile 0.9% saline. Mice were sacrificed
the next day and microglia were isolated for FACS analysis or cell culture as described.
AIF-1 Secretion. Splenocytes, THP-1 cells, or microglia cells were plated at a density
ranging from 1x10
6
to 4x10
6
cells/ml and supernatants were collected 4, 24, 48, or 72
hours later. Supernatants were cleared of cell debris by low speed centrifugation (300 xg
for 10 minutes) and then diluted with 5x Laemmli buffer. Levels of AIF-1 in
supernatants were analyzed by western immunoblotting, loading equal volumes of
supernatant for each sample. Lysate from THP-1 cells was included as a positive control.
AIF-1 secretion is expressed relative to controls as specified in each experiment. For in
vitro stimulation, cells were incubated with either culture medium alone (untx) or culture
medium supplemented with either (1) 100ng LPS (Sigma) and 10ng/ml IFN" (R&D
Systems, Minneapolis, MN), (2) 4µg/ml concavalin A (Con A), or (3) 10% medium
preconditioned by rat splenocytes stimulated with 4µg/ml concavalin A for 48 hours (Con
A sups). We discovered that Con A sups contain significant levels of AIF-1. So, culture
55
medium with 10% Con A sups was run in a separate lane as a control and band intensities
were subtracted as background from supernatants collected from cells stimulated with
Con A sups.
Western Immunoblotting. Equal volumes of supernatant were separated on 15% Tris-
Glycine gels (Cambrex, Rockland, ME) and transferred to nitrocellulose by
electroblotting in Towbin’s buffer (50mM Tris, 38mM glycine, 4% SDS, 20% methanol)
for 2 hours at 400mA. After blocking with Blocking Buffer for Near Infra Red
Fluorescent Western Blotting (Rockland, Gibertsville, PA), blots were incubated
overnight at 4°C with antibody specific for AIF-1 (1:1000, Wako) diluted in blocking
buffer supplemented with 0.1% Tween-20. Blots were then washed five times with
phosphate buffered saline with Tween-20 (PBST; 0.1mM phosphate, 0.9% NaCl, 0.05%
Tween-20) and then incubated with Alexa Fluor$ 688 goat anti-mouse IgG (1:1000;
Invitrogen) for 30 minutes at room temperature. Washed blots were visualized and band
densinometries were measured with an Odyssey Infrared Imager (LI-COR
Biotechnology, Lincoln, NE).
Results
Analysis of AIF-1 Protein Sequence. With growing evidence of a secreted role for AIF-1,
I decided to analyze the amino acid sequence specifically for predictors of secretion. As
reported by others, analysis of mouse AIF-1 protein sequence with the SignalP 3.0 Server
(http://www.cbs.dtu.dk/services/SignalP) revealed no classical N-terminal signal
sequence. The sequence did contain two possible sites for GalNAc O-glycosylation at
56
positions 135 and 144, as predicted by the NetOGlyc 3.1 Server
(http://www.cbs.dtu.dk/services/NetOGlyc). However, without a signal sequence, it is
unlikely that AIF-1 will encounter the necessary machinery for glycosylation during post-
translational processing. A propeptide cleavage site was predicted by ProP 1.0 Server
(http://www.cbs.dtu.dk/services/ProP) at position 133 with a predictive value of 0.552
that was just above threshold (threshold = 0.500). The SecretomeP 2.0 Server
(http://www.cbs.dtu.dk/services/SecretomeP) predicted the secretion of AIF-1 by a non-
classical route with a N-N score of 0.872 (threshold = 0.5).
Secretion of AIF-1 by murine splenocytes. Because low yields of primary microglia cells
can be obtained from adult mice (~2 x 10
5
cells), murine splenocytes (which yield 2-5 x
10
7
cells per mouse) were used first to confirm methods for detection and to characterize
AIF-1 secretion. Splenocytes isolated from untreated adult mice were plated at 3x10
6

cells/ml in culture medium and supernatants were collected 24 hours later and analyzed
for AIF-1 by western immunoblot. Unconditioned culture medium was not
immunoreactive for AIF-1 (data not shown). As shown in figure 3.1A, two bands in the
supernatants collected from splenocytes were immunoreactive for AIF-1. One band was
of the predicted size for AIF-1 (17kDa) as previously reported (Watano et al. 2001).
Another smaller band of approximately 15kDa was also seen in all samples. Lysate from
THP-1 cells was regularly used as a positive control because THP-1 cells have high basal
levels of AIF-1 (Imai et al. 1996; Utans et al. 1996). Multiple bands were also seen in
THP-1 lysate (figure 3.1A, lane T): a large band at 17kDa and two smaller bands that
were not seen in previous reports (Imai et al. 1996). Because the smaller band seen in
57

Figure 3.1 Western blot analysis of AIF-1 levels in supernatants of THP-1cells
stimulated in vitro. THP-1 cells were plated at 1x10
6
cells/ml in culture medium alone,
with 100ng/ml LPS and 10ng/ml IFN" (LPS/IFN), or with 10% Con A sups (Con A
Sups). Supernatants were collected after 4, 24, or 48 hours and AIF-1 levels were
measured by western immunoblotting. (A) Representative immunoblot of THP-1
supernatants; arrows indicate 17kDa and 15kDa bands. M=protein ladder, T=THP-1
lysate (B) Band densitometries of the larger 17kDa band were measured for untreated and
IFN/LPS stimulated cells. AIF-1 levels are given relative to untreated controls at 24
hours and represent the mean ! SEM (n=3) for 24 and 48 hour time points. Mean of 4
hour time point is included to show trend (n=1). The difference between untreated and
LPS/IFN groups were analyzed by paired t-test. *, p<0.05.
58
supernatants was of unknown identity, the band running at 17kDa was used to measure
AIF-1 levels. AIF-1 levels in the supernatants of murine splenocytes were dependent on
cell concentration (figure 3.1B). Cells were plated at 1, 2, or 4x10
6
cells/ml.
Densitometries of the 17kDa band were analyzed and used to calculate AIF-1 levels
relative to those seen in the supernatants of cells plated at 1x10
6
cells/ml. AIF-1 levels in
spleen supernatants also increased with time in culture (figure 3.1C). The first
supernatants taken at 4 hours were compared to those taken at 24, 48 and 72 hours. AIF-
1 levels increased by 50% from 4 hours to 24 hours and then remained steady through 72
hours. Lastly, we attempted to increase AIF-1 secretion by stimulating the cells in vitro
with inflammatory stimulants. Concanavalin A (Con A) is a non-specific lymphocyte
mitogen and the supernatants of splenocytes stimulated with Con A (Con A sups) contain
several cytokines including IFN" and IL1#. In fact, high levels of AIF-1 were found in
Con A sups prepared from rat splenocytes (data not shown). So, Con A itself was used as
a potential stimulator of AIF-1 secretion. We also used the powerful combination of
IFN" and LPS, which both initiate pro-inflammatory responses from myeloid cells.
Figure 3.1D shows that none of these treatments had any significant effects on AIF-1
levels in the supernatant. AIF-1 levels were increased with Con A sups slightly above
control cells (with the background level of AIF-1 in the sups themselves accounted for).
The Con A sups used in this study were prepared from rat splenocytes grown at a
concentration of 4x10
6
cells/ml, so high AIF-1 levels in Con A sups may be a result of
splenocyte concentration as we did not find any increase in splenocyte secretion of AIF-1
with Con A stimulation.
59
Secretion of AIF-1 by THP-1 cells. The THP-1 cell line was established from human
acute monocytic leukemia cells and has high basal levels of intracellular AIF-1 (Imai et
al. 1996). The supernatants of THP-1 cells contained two bands immunoreactive for
AIF-1 that were the same size as the bands seen in the supernatants of splecnocytes
(figure 3.2A). Unlike splenocytes, the stimulation of THP-1 cells with inflammatory
factors resulted in measurable increases in AIF-1 secretion (figure 3.2B). AIF-1 levels in
the supernatants of THP-1 cells stimulated with a combination of LPS and IFN" were
double that of controls at 4, 24, and 48 hours. Although only one replicate was analyzed
at 4 hours, the results were included to show the trend is consistent at earlier time points.
Stimulation with Con A sups had no effect on the secretion of AIF-1 by THP-1 cells.
This data suggests that AIF-1 secretion is inducible in THP-1 cells.
Secretion of AIF-1 by adult murine microglia. The main objective of this chapter was to
show secretion of AIF-1 from adult microglia cells. Microglia were isolated from the
whole brains of adult normal mice. Flow cytometric analysis showed that cultured cells
were comprised of predominantly resting microglia (98%±2%) that co-expressed CD45
lo

and CD11b (figures 3.3A & 3.3B). AIF-1 was present in the supernatants of cells
isolated from normal mice when grown in culture medium with 10% FBS (figure 3.3C),
but not when cells were cultured ex vivo in serum-free medium (data not shown). Similar
to what was seen in splenocytes and THP-1 cells, two bands were seen in the
supernatants of microglia, a 17kDa protein and a 15kDa protein.
60
Figure 3.2 Western blot analysis of AIF-1 levels in supernatants of murine
splenocytes. (A) Representative immunoblot of splenocyte supernatants. Cells were
isolated from normal, untreated C57Bl6 mice and plated in culture medium at
3x106cells/ml; supernatants were collected after 24 hours. L= protein ladder T= THP-1
lysate,1-4 = supernatants from spleens of 4 different mice. (B) Splenocytes were plated
in culture medium alone at varying cell concentrations and AIF-1 levels were measured
in supernatants collected after 24 hours. Band intensities of the 17kDa protein are given
relative to measurements at 1x10
6
cells/ml and represent mean ! SEM (n=4). One-way
ANOVA analysis (p<0.001) was followed by pairwise comparison (Holm-Sidak). *,
p<0.05 vs 1x10
6
cells/ml. (C) AIF-1 levels were measured in supernatants of splenocytes
plated at 3x10
6
cells/ml taken after 4, 24, 48, and 72 hours in culture. Band intensities of
the 17kDa protein are given relative to measurements at 1x10
6
cells/ml and represent
mean ! SEM (n=3). (D) Splenocytes were plated at 3x106cells/ml in culture medium
alone (untx), with 10ng/ml LPS and 100ng/ml IFN (LPS/IFN), with 10% Con A (Con A),
or with 10% medium conditioned by Con A stimulated rat splenocytes (Con A sups).
AIF-1 levels were measured in supernatants collected after 4 hours or 24 hours. Mean
band densitometries of the 17kDa protein (n=3) are given relative to untreated controls at
each time point.
61
Figure 3.2, Continued

62

Figure 3.3 Adult murine microglia secrete AIF-1. Microglia cells were isolated from
normal C57Bl6 mice as described in Materials and Methods. Samples of cell
preparations were analyzed by two-color flow cytometry for expression of CD11b and
CD45 to confirm purity. Figure A shows a representative dot plot of side scatter versus
forward scatter of the cell preparation. Figure B shows the relative expression of CD45
and CD11b on these purified cells. Purified microglia were cultured for 24 hours and
supernatants analyzed by western immunoblot for secreted AIF-1 (figure C). Lane 1
shows the positive control THP-1 lysate; lanes 2, 3, 4 show supernatants from microglia
isolated from 3 different mice. The arrows indicate the molecular weight of the protein
bands shown.
63
Peripheral injections with LPS have been shown to stimulate microglia throughout the
CNS (Chakravarty and Herkenham 2005; Czapski et al. 2007). To determine if microglia
activated in this manner would secrete increased levels of AIF-1 ex vivo, microglia and
splenocytes were isolated from mice 24 hours after a single injection of 1mg/kg LPS and
AIF-1 levels were analyzed in supernatants collected from these cells after 24 hours
(figure 3.4). AIF-1 levels were not significantly different from levels in the supernatants
of normal mice. Similarly, systemic LPS stimulation did not alter the secretion of AIF-1
from splenocytes.
Discussion
Secretion of AIF-1 has been demonstrated in cultures of microglia stimulated by soluble
factors released by apoptotic cerebellar granule neurons (Tanaka and Koike 2002). This
work used cultures of microglia prepared from newborn rats, which represent a different
physiological state than microglia isolated from adult rodents, as distinguished by
differences in cell phenotype and APC function (Carson et al. 1998) and age-based
differences in microglia function in in vivo models of infection and injury (Milligan et al.
1991; Lawson and Perry 1995) . Since this thesis is focused on the involvement of AIF-1
secretion during the microglia response to neurodegenerative disease in adult animals, the
demonstration of secretion of AIF-1 by microglia from adult mice was an important
distinction. Significant levels of AIF-1 were detected in the supernatants of freshly
isolated adult microglia after 24 hours incubation in culture medium containing 10%
64
Figure 3.4 Microglia and splenocytes isolated from mice treated with systemic LPS
stimulation do not alter secretion of AIF-1. Splenocytes and microglia were isolated
24 hours after animals were given a single i.p. injection of LPS (10mg/kg) and plated in
culture medium at 1x10
6
cells/ml and 2x10
6
cells/ml, respectively. Band densitometries
for 17kDa protein are given relative to untreated microglia and represent the mean !
SEM for 3 animals.
0
0.5
1
1.5
2
2.5
3
3.5
4
Microglia Spleen
Relative AIF-1 Secretion
untx
LPSx1
65
serum (figure 3.3C). No bands immunoreactive for AIF-1 were found in unconditioned
culture medium or in the supernatants of microglia grown in serum-free conditions (data
not shown). It may be that serum components induced the secretion of AIF-1 in our ex
vivo cultures. Because the secretion of AIF-1 was not induced in mouse splenocytes by
the addition of inflammatory stimuli in vitro, an in vivo stimulation strategy was used to
test for the inducible secretion of AIF-1 by microglia. We chose to induce widespread
inflammation in the brain using i.p. injections of LPS instead of SNpc localized
activation with MPTP-lesioning, so that microglia from whole brains could be used. In
vivo stimulation with i.p. LPS did not stimulate the secretion of AIF-1 by microglia or
splenocytes ex vivo (figure 3.4).
Of interest was the observation of two distinct bands of similar size that were
immunoreactive for AIF-1. In addition to the predicted 17 kDa band, a smaller band
(~15kDa) of unknown identity was detected in the supernatants of microglia. These same
two bands were also detected in the lysate and supernatants of THP-1 cells (figure 3.2A)
and in the supernatants of murine splenocytes (figure 3.1A). This smaller band may be a
cleaved AIF-1 product or a splice site variant. At least three AIF-1 isoforms, generated
by alternative splicing, have been identified in humans (Hara et al. 1999). One isoform is
generated from a alternate splicing that results in a truncated protein (~9.5 kDa) with no
EF hand and a C-terminal region with no homology to full length AIF-1. By RT-PCR
they found this isoform to be strongly expressed in both human PBMC and THP-1 cells.
The second splice variant results in an in frame-deletion that codes for a protein lacking
14 amino acids (~13kDa) resulting in an incomplete EF hand. Other possible splice
66
variants described include BART-1, G1, and IRT-1 (Deininger et al. 2002). It is also
possible that the smaller band is the result of posttranslational processing. Our analysis
predicted a cleavage site at the C-terminus that would remove 14 amino acids from the C-
terminus, resulting in a protein of approximatey 15 kDa. Others have described a
prohormone cleavage site, that could lead to a truncated protein (Chen et al. 1997). No
cleaved proteins have been described in cell lystates or supernatants. Further studies into
the identity of this protein may be helpful in distinguishing mechanisms of AIF-1
secretion.
The secretion of AIF-1 has also been demonstrated by peripheral inflammatory cells.
Recently, AIF-1 has been found in the peritoneal fluid of women with endometriosis
(Koshiba et al. 2005) and in the synovial fluid of patients with rheumatoid arthritis
(Kimura et al. 2007). In vitro experiments in these studies showed that both human
peritoneal macrophages and synovial cells could secrete AIF-1. No receptor has been
identified for AIF-1, but growing evidence supports an extracellular function for AIF-1.
Recombinant AIF-1 can alter proliferation and differentiation of muscle satellite cells in
vitro (Kuschel et al. 2000). Intravenous injection of AIF-1 alters gluscose-stimulated
insulin secretion in rats (Chen et al. 1997). Despite the popularity of AIF-1/Iba1 as a
microglia marker, no role for AIF-1 in the CNS has been suggested; indeed AIF-1 had no
effect on neuronal survival in mixed cell cultures demonstrating secretion (Tanaka and
Koike 2002) However, in rheumatoid arthritis patients it was shown that recombinant
AIF-1 could stimulate the secretion of IL-6 in both synovial cells and peripheral blood
mononuclear cells (Kimura et al.). If AIF-1 had a similar effect in the adult brain, it
67
would most likely play a significant role in regulating the response of microglia cells or
infiltrating mononuclear cells.
This work demonstrates the ability of adult microglia to secrete AIF-1, but more work is
needed to determine if the secretion would be increased in models of neurodegeneration
like the MPTP neurotoxicity described in the previous chapter. In addition to possible
signals from dying dopaminergic neurons, several immunoregulatory molecules are
elevated in the brains of MPTP-lesioned mice including MCP-1, IL-6, TNF$ and IFN"
(Kohutnicka et al. 1998; Sriram et al. 2006; Pattarini et al. 2007). The ability of these
substance to stimulate the secretion of AIF-1 in microglia should be assessed in this ex
vivo model. It may also be interesting to determine if activated microglia isolated from
the midbrain of MPTP–lesioned mice would show increased secretion of AIF-1
compared to microglia isolated from normal controls or other areas of the brain.
68
Chapter 4: Production of a recombinant AIF-1 protein using a mammalian
expression system
Introduction
Previous chapters have shown a robust upregulation of AIF-1 in response to MPTP
neurotoxicity (Chapter 2) and secretion of AIF-1 from adult murine microglia ex vivo
(Chapter 3). This presents many questions as to what role AIF-1 may serve as an
extracellular molecule in the microglial response to neuronal death. Although
intracellular functions of this protein have been examined in detail (see Chapter 1 for full
review), information regarding the extracellular actions of AIF-1 is limited to only a few
studies, all in systems outside the CNS. It was demonstrated that intraperitoneal
injections of AIF-1 protein isolated from porcine intestine can alter glucose-stimulated
insulin secretion in mice (Chen et al. 1997). Recombinant rat AIF-1 can stimulate muscle
satellite cell differentiation when added to tissue culture preparations of rat muscle
(Kuschel et al. 2000). In addition, human AIF-1 can stimulate proliferation of synovial
cells and secretion of cytokines by both synovial cells and human peripheral blood
mononuclear cells (Kimura et al. 2007).
To begin to assess possible roles for extracellular AIF-1 in the CNS, large quantities of
protein were needed for the activity studies described in the next chapter. My strategy
involved the production of a His-tagged AIF-1 recombinant protein using a mammalian
expression system. Recombinant human, rat, and mouse AIF-1 have been generated
69
using bacterial expression systems (Imai et al. 1996; Autieri et al. 2003; Yamada et al.
2006). Most were used for antibody production, where bioactivity is not critical. Two
studies that did use recombinant AIF-1 for functional studies include one study that
generated both a full length recombinant AIF-1 and a mutant AIF-1 with a C-terminal
truncation to further study actin binding activity in vitro (Sasaki et al. 2001) and the rat
muscle study previously mentioned (Kuschel et al. 2000).
Analysis of the amino acid sequence reveals putative sites for the post-translational
modification of AIF-1, including several phosphorlyation sites (Watano et al. 2001). I
chose to produce rAIF-1 using a mammalian expression system to provide for any post-
translational modifications or protein folding necessary for the full processing and
optimum activity of extracellular AIF-1. The secretion of AIF-1 by transfected 293T
cells allowed for the assessment of differences between the secreted and intracellular
proteins by mass spectrophotometery.
Materials and Methods
PCR Amplification. Amplification of DNA sequences was achieved using the Platinum
Taq DNA Polymerase (Invitrogen) according to manufacturer’s instructions. Each
reaction mix consisted of final concentrations of 1xPCR buffer, 0.2mM dNTP mixture,
1.5mM MgCl
2
, 0.2mM of each primer, and 1.0 unit of Platinum Taq polymerase.
Thermal cycling included an initial step of 10min at 95°C to activate the enzyme,
followed by 35 cycles consisting of 30s at 94°C, 30s at 54°C, and 1min at 72°C. A final
extension step at 72°C for 10min was also included before sample were kept at 4°C until
70
analyzed by agarose gel electrophoresis. Where necessary, PCR products were purified
using the QIAquick PCR Purification Kit (QIAGEN) or excised from agarose gels and
purified using the QIAquick Gel Extraction Kit (QIAGEN).
Agarose Gel Electrophoresis. DNA samples were diluted in 10x BlueJuice Gel Loading
Buffer (Invitrogen) and separated on a 0.9%-2% agarose gel made in Tris acetate buffer
(0.04M Tris, 1mM EDTA, pH 8.0) with 0.5mg/ml ethidium bromide. Samples were run
alongside 1µg of either the 1kb DNA ladder (Gibco) or 100bp DNA ladder (Invitrogen).
Gels were run at a constant voltage of 3-5 V/cm for 1-2 hours using the EasyCast
Electrophoresis System (Owl Scientific) and EC105 power supply (E-C Apparatus
Corporation). Bands were visualized on UV transilluminator and photographed using the
Kodak DC3200 digital camera.
Ligation. The GeneChoice Rapid Ligation Kit was used for ligation reactions according
to instructions. Briefly, vector and insert were combined in a 3:1 molar ratio in supplied
ligation buffer. 1ml of DNA ligase was added and the mixture was incubated for 5
minutes at room temperature. This was then used immediately in the transformation of
competent E coli. Control reactions with vector alone were done in parallel.
Plasmid preparation. The pTT3 plasmid was provided by the lab of Dr. W. French
Anderson. Plasmids were amplified in One Shot ® TOP10 chemically competent E. coli
(Invitrogen). Cells were transformed with 2.0ml of ligation reaction according to the
rapid chemical transformation procedure provided by Invitrogen and initially plated on
LB-ampicillin agar. After an overnight incubation at 37°C, individual clones were
71
selected and grown in LB broth with ampicillin overnight at 37°C with shaking.
Plasmids were purified using Qiagen plasmid mini kits or Qiagen plasmid maxi kits
according to manufacturers instructions. For quantification, plasmids were diluted in
50mM Tris-HCl pH 7.4 and absorbance was measured at 260nm and 280nm. Purified
plasmids with a A
260
/A
280
ratio of 1.7 to 2.0 were used for transfection.
Restriction Digest. All restriction enzymes were obtained from New England Biolabs
and digests were carried out with their recommended buffers. Plasmids or PCR products
were double digested with EcoRI and XbaI in 1xEcoRI buffer and 100µg/ml BSA for 2
hours at 37°C. Digestion with EcoRV was done under the same conditions, but with NE2
buffer and BSA.
DNA Sequencing. DNA sequencing was performed by the USC/Norris Microchemical
Core Facility, which uses a BigDye terminator cycling sequencing (Applied Biosystems).
Sequencing of the pAIF-1 vector was done in both directions using the same primers used
for PCR cloning.
Cell Culture. 293T were obtained from the Anderson lab. Culture medium consisted of
Dulbecco’s modified Dulbecco's Modified Eagle's Medium (Gibco) with 10% FBS
(Gibco), supplemented with 2.5mM L-glutamine (Gibco), 1mM sodium pyruvate, 0.1mM
non-essential amino acid and cells were kept at 37°C, 5% CO
2
. When cells reached 90%
confluency, they were passaged by trypsinizing and replating in fresh culture medium.
72
Transfection. Transfection of 293T cells with the pTT3 empty vector or pTT3 vector
containing AIF-1 insert (pAIF-1) was performed using Lipofectamine 2000 reagent
according to the manufacturer’s instructions (Invitrogen). Transfection was first
optimized on a small scale using 24 well plates. Cells were plated at 90% confluency and
cells were transfected using DNA: lipofectamine ratios of 1:0.5 to 1:5. The amount of
AIF-1 in cell lysates was compared by Western Blot and the best yield was found using a
1:5 ratio with 0.8mg pAIF-1. For the production of rAIF-1, 293T cells were plated in
150cm
3
tissue culture flasks and grown to 90% confluency in culture medium. The next
day, culture medium was removed and replaced with 10ml of OptiMEM media (Gibco).
Plasmids (30%g per flask) were incubated at room temperature for 20 min with
lipofectamine (90%l per flask) in OptiMEM medium and then added to cultures. After 3
hours, FBS was added to a final concentration of 4%. The next day, medium was
replaced with 293 SFM II medium (Invitrogen). At 72 hours after transfection,
supernatants were collected and cleared of cell debris by low speed centrifugation. Cells
were collected from culture flasks by rinsing with DPBS. Media was removed by
centrifugation and cells were lysed in modified RIPA buffer (20mM Tris-HCl pH 7.5, 1%
triton X-100, 0.05% SDS, 5mg/ml sodium deoxycholate, 150mM NaCl, 1mM PMSF)
supplemented with both a protease and a phosphatase inhibitors cocktail (Roche
Diagnostics, Mannheim, Germany). Lysates were cleared of cell debris by centrifugation
for 10 minutes at 10,000 xg at 4°C. Lysates were immediately used for purification of
rAIF-1 or stored at -80°C.
73
SDS-PAGE. Cells lysates or supernatants were diluted in 5x gel loading dye and resolved
on 15% or 4-20% Tris-Glycine gels (Cambrex, Rockland, ME) Gels were run at a
constant voltage of 70mV until the leading dye reached the interface of the loading gel
and then kept at 120mV until approximately 1cm from the end of the gel. Protein ladders
were run in at least one well to approximate sizes of identified bands. Gels were either
transferred to nitrocellulose for immunoblotting or stained for total protein with
Coommassie blue. Staining consisted of incubation in staining buffer (0.1% brilliant
blue, 10% acetic acid and 40% ethanol) for 1 hour at room temperature and then
destaining overnight in destaining buffer. Stained gels were imaged and band
densinometries were measured using the Odyssey Infrared Imager.
Western Immunoblotting. Proteins separated by SDS-PAGE were transferred to
nitrocellulose by electroblotting in Towbin’s buffer (50mM Tris, 38mM glycine, 4%
SDS, 20% methanol). To assess transfer efficiency, total protein was visualized with
Ponceu’s stain (Sigma) and then reversed with 0.1M NaOH. After blocking with
Blocking Buffer for Near Infra Red Fluorescent Western Blotting (Rockland,
Gibertsville, PA), blots were incubated with antibody specific for AIF-1 (1:1000, Wako )
or RGS-His (1:1000, QIAGEN) overnight at 4°C. Blots were washed five times with
PBST and then incubated with Alexa Fluor$ 688 anti-mouse IgG (1:1000; Invitrogen) or
Alexa Fluor$ 688 anti-mouse IgG (1:1000; Invitrogen) for 30 minutes at room
temperature before a second round of washing. Blots were visualized and band
densitometries were measured with the Odyssey Infrared Imager (LI-COR
Biotechnology, Lincoln, NE).
74
Protein purification. His-tagged proteins were purified from supernatants or cell lysates
using standard Ni
+
-NTA affinity purification techniques. Briefly, 10X binding buffer
was added to lysates or supernatants containing His-tagged proteins or serving as mock
controls. Meanwhile, Ni-sepharose beads (GE Healthcare) were washed once in sterile
water and then two times in 1x binding buffer. 1ml of 1x binding buffer was added to
1ml of washed beads to make a 50% slurry, and this slurry was then added to prepared
samples and rocked overnight at 4°C. The next day, beads were collected by
centrifugation 500 xg for 5 minutes, resuspended in 1x binding buffer, and transferred to
disposable PD-10 column (GE Healthcare). Initial flow through from the column was
collected as the first wash (W1). Beads were washed two times more with 5ml of 1X
binding buffer (W2, W3). His-tagged proteins were then released from the columns in
five elutions of 0.5ml elution buffer each and collected in five separate tubes (E1-E5).
Samples from all washes (W1-3) and elutions (E1-5) were assayed for AIF-1 by SDS-
PAGE and Coommassie staining. Elution containing rAIF-1 was combined and salts
were removed using PD-10 desalting columns according to instructions (GE Healthcare)
Briefly, PD-10 columns were equilibrated with 15ml of DPBS. Combined elutions were
diluted to 2.5ml with DPBS and applied to equilibrated columns. The first 2.5ml was
saved for analysis. Then, 3.5ml of DPBS were added to the columns and the 3.5ml of
flow through was collected and tested for presence of rAIF-1. Protein was concentrated
using a Centriprep centrifugal filter device with a 3kDa cutoff (Millipore), involving
centrifugation at 3000rpm, 4°C until the desired volume was reached.
75
Protein Concentration. Protein concentration was measured using the Total Protein Kit
from Sigma. Samples were diluted 1:10 or 1:100 in DPBS. DPBS was used as a blank
and 0.3mg/ml BSA (included in the kit) was used as a standard. Absorbance was
measured at 595nm and concentrations were calculated using the following formula.
Test conc. = standard conc. % A
595
test ÷ A
595
standard
Mass Spectrophotometry. Mass spectrophotometry analysis was done at the USC School
of Pharmacy Proteomic Core Facility. rAIF-1 purified from supernatants or lysates in the
presence of phosphatase inhibitors cocktail II and III (Sigma) were resolved by SDS-
PAGE, excised from the gel, and supplied to the core facility. Following an in gel tryptic
digest, LC MS-MS tandem mass spectrometric analysis was performed using a
Thermofinnigan LCQ DecaXP. Results provided included protein identification based on
mass matching algorithms and a coverage map of the sequence.
Endotoxin Testing. Samples were sent out for quantitative limulus amebocyte lysate
(LAL) assay by Focus Diagnostics. Purified protein products or mock preparations were
diluted 1:10 in deionized water and kept at -20°C prior to pick-up. Levels <0.05 EU/ml
are considered negative.
Actin Polymerization. In vitro actin polymerization was assayed using a kit purchased
from Cytoskeleton, Inc (Denver, CO). Briefly, rAIF-1 and actin were incubated in F-
actin buffer containing ATP and Ca
2+
for 30min. The mixture was then centrifuged at
8000 xg for 60min. At this speed, free actin remains in the supernatant, while cross-
76
linked actin will pellet. Protein in the pellet and supernatant were resolved independently
by SDS-PAGE with Coommassie staining of total protein. Densitometry was used (using
Odyssey software) to measure ratios of precipitated actin to total actin (ppt actin/total
actin). BSA was tested as a negative control as it does not bind to or affect the
polymerization of actin. &-actinin, an actin-binding and -polymerizing protein, served as
a positive control.
Results
Phosphorylation Sites of murine AIF-1. The amino acid sequence for AIF-1 was
analyzed using the NetPhos2.0 Server (http://www.cbs.dtu.dk/services/NetPhos/) in order
to predict phosphorylation sites (figures 4.1A & 4.1B). AIF-1 contains twelve serine
residues, and phosphorylation was predicted for seven of those residues (positions 38, 48,
69, 94, 95, 97, 102) with scores ranging from 0.841 to 0.994. AIF-1 also contains three
threonine sites and phosphorylation was predicted for the threonine at position 135 with a
score of 0.651. Three of the four tyrosine residues at positions 37, 54, and 103 were
predicted to be phosphorylated with scores of 0.977, 0.933, and 0.791, respectively.
Cloning of AIF-1. The coding region for AIF-1 (444bp, NM_019467) was PCR cloned
from mouse spleen cDNA. Primers (figure 4.2A) were designed to amplify the coding
region for AIF-1 with the addition of an EcoRI restriction site at the 5’ end and an XbaI
site at the 3’ end of the coding region. In addition, the original TGA stop codon was
omitted and residues coding for the inclusion of six histidines at the C-terminus of the
protein were included in the antisense primer upstream of a new TTA stop codon. This
77
Figure 4.1. Potential phosphorylation sites predicted in the coding sequence for murine AIF-1. The murine AIF-1
amino acid sequence (NP_6) was analyzed using the NetPhos 2.0 server for potential phosphorylation sites. The
predictive phosphorylation score and sequence position of all serine, threonine , and tyrosine residues is listed both in
tabular (A) and graphical(B) forms. Threshold was set at 0.50 and is represented by the horizontal line on the graph

78

Figure 4.2. PCR design of AIF-1 insert. (A) Primer sequences used to amplify the AIF-
1 coding region (top = sense primer, bottom = antisense primer). On the sense primer,
sequence for the EcoRI restriction site is shown in italics. The remainder of the sequence
is the first 25 residues of the murine AIF-1 CD (acquisition number AF109719). The
antisense primer contains sequence complementary to the last 19 residues of the murine
AIF-1 CD (with the exception of the TGA stop codon), sequence coding for six histidines
(bold) followed by a new TTA stop codon (grey), and a restriction site for XbaI (italics).
(B) PCR of mouse spleen cDNA with these primers resulted in a single DNA sequence of
approximately 500 bp. PCR product was resolved on a 1% agarose gel with ethidium
bromide and run alongside a 1 kb DNA ladder (arrowhead indicates 600 kb ladder
fragment). The arrow indicates the 600 bp band in the 1 kb DNA ladder. (C) Schematic
of AIF-1 insert as designed by the primers listed in 1A. The AIF-1 coding sequence was
modified to include codons for six histidines (6xHis) before the stop codon. This new
coding sequence is flanked with restriction sites for EcoRI and XbaI (light grey boxes) to
allow for insertion into the pTT3 expresssion vector. Bold lines indicate sequences
covered by the primers including all added elements and complimentary sequence at the
ends of the AIF-1 coding region.
79
6xHis tag allowed for easier purification of the recombinant protein as described below.
PCR with these primers, using mouse spleen cDNA as a template, resulted in a product of
approximately 500bp, as visualized by agarose gel electrophoresis (figure 4.2B). A
schematic of the insert showing the orientation of elements added by PCR design is given
in figure 4.2C.
The flanking restriction sites allowed for insertion of the sequence into the multiple
cloning site of the pTT3 vector. pTT3 is a modified pCEP4 vector with a CMV
expression cassette and an Epstein-Barr Virus oriP that allows for episomal replication
(Durocher et al. 2002). Elements of the pTT3 vector, including recognized restriction
sites, are given in figure 4.3. Following purification of the PCR product, both the PCR
product and the pTT3 vector were cut using EcoRI and XbaI restriction enzymes and
combined at a 3:1 (vector: insert) molar ratio in the presence of DNA ligase enzyme.
This mixture was then used to transform competent E. Coli. Plasmids were isolated from
selected clones and insertion of the AIF-1 coding region was confirmed by digestion with
EcoRV (figure 4.4). Both uncut plasmids ran at approximately 4kb (figure 4.4, pAIF-
1=lane 1, pTT3=lane 2). The pTT3 vector itself contains a single EcoRV restriction site
(bp 2083) and digestion with EcoRV resulted in a linear product of 6008 bp (figure 4.4,
lane 4). Insertion of the PCR product adds 457bp to the plasmid between the EcoRI (bp
1297) and XbamI (bp 1305) sites and includes and additional EcoRV site. With the
insert. digestion of the plasmid results in two bands of 5,414bp and 1,051bp (figure 4.4,
lane 3). DNA sequencing in two directions using the same primers used for subcloning
was used to confirm the correct sequence and insertion site of the AIF-1 coding region.
80

Figure 4.3 pTT3 expression vector used in the production of recombinant AIF-1.
pTT3 contains an enhanced CMV expression cassette and an Epstein-Barr Virus OriP site
allowing for episomal replication. EcoRI and XbaI restriction sites in the multiple
cloning site were used for insertion of the AIF-1 coding region. One EcoRV srestriction
site is located in the OriP. Taken from supplemental materials provided by the NRC
Biotechnology Research Intsitute.
81

Figure 4.4 EcoRV restriction digest of pTT3 and pAIF-1. To check for insertion of
the PCR cloned AIF-1 sequence into the pTT3 expression vector, plasmids were digested
with the EcoRV restriction enzyme and fragments were resolved on 1% agarose gel
alongside a 1 kb DNA ladder (lane 5). pTT3 vector contains one EcoRV restriction site
in the OriP and the AIF-1 insert also contains a single EcoRV site. Both uncut plasmids
(lane 1 = pAIF-1, lane 2 = pTT3) ran at approximately 4 kb, with the pAIF-1 plasmid
containing the insert running a bit higher. As expected, two bands were found in pAIF-1
samples cut with EcoRV (lane 3), one at approximately 5 kb and another at 1 kb, while
pTT3 cut with EcoRV resulted in a single linearized product running at 6 kb (lane 4).

82
Figure 4.5 gives the sequence covered by the DNA sequencing and the corresponding
amino acid sequence of the coding region. Two single base pairs differed from the
published sequence at 216bp and 234bp into the coding region, but both of these were
located in the last base pair of a codon and did not alter the predicted amino acid
sequence.
Generation and Purification of rAIF-1. 293T cells were transfected with either empty
pTT3 vector or pTT3 containing the coding region for AIF-1 (pAIF-1). Western
immunoblotting with a polyclonal antibody specific for AIF-1 detected a 17kDa band in
both the lysate and supernatant of 293T cells transfected with pAIF-1 but not in those
transfected with vector alone (figure 4.6A). Corresponding immunoblots demonstrated
that this 17kDa protein (rAIF-1) was also recognized by antibody to the histidine tag
(figure 4.6B). Purification of rAIF-1 from the lysate or supernatant took advantage of the
6xHis tag. Mock preparations using supernatants or lysates from 293T cells transfected
with empty pTT3 vector were processed in parallel during all purification steps to serve
as a negative control in activity experiments. Purity of rAIF-1 and mock protein
preparations were examined by SDS-PAGE and visualized using Coommassie blue
staining in combination with highly sensitive infrared imaging (figure 4.6C). A single
band at 17kDa was seen in rAIF-1 preparations, but no bands were seen in mock
preparations. The sequence of the recombinant protein was confirmed by mass
spectrometry with 50% coverage of the protein sequence for murine AIF-1 (NP_062340;
figure 4.7A). rAIF-1 purified from the lysate or from the supernatant of transfected 293T
cells were indistinguishable by SDS-PAGE or mass spectrometry (figure 4.7B). No post-
83
…GTTTAAACGGATCTCTAGC|GAATTC|ATG AGC CAA AGC AGG GAT TTG CAG
M S Q S R D L Q
GGA GGA AAA GCT TTT GGA CTG CTG AAG GCC CAG CAG GAA GAG AGG
G G K A F G L L K A Q Q E E R
CTG GAG GGG ATC AAC AAG CAA TTC CTC GAT GAT CCC AAA TAC AGC
L E G I N K Q F L D D P K Y S
AAT GAT GAG GAT CTG CCG TCC AAA CTT GAA GCC TTC AAG GTG AAG
N D E D L P S K L E A F K V K
TAC ATG GAG TTT GAT CTG AAT GGA AAT GGA GAT ATC GAT ATT ATG
Y M E F D L N G N G D I D I M
TCC TTG AAG CGA ATG CTG GAG AAA CTT GGG GTT CCC AAG ACC CAC
S L K R M L E K L G V P K T H
CTA GAG CTG AAG AGA TTA ATT AGA GAG GTG TCC AGT GGC TCC GAG
L E L K R L I R E V S S G S E
GAG ACG TTC AGC TAC TCT GAC TTT CTC AGA ATG ATG CTG GGC AAG
E T F S Y S D F L R M M L G K
AGA TCT GCC ATC TTG AGA ATG ATT CTG ATG TAT GAG GAG AAA AAC
R S A I L R M I L M Y E E K N
AAA GAA CAC AAG AGG CCA ACT GGT CCC CCA GCC AAG AAA GCT ATC
K E H K R P T G P P A K K A I
TCC GAG CTG CCC|CAT CAT CAC CAT CAC CAT|TTA|ATCTAGA|GGCCG…
S E L P H H H H H H *
Figure 4.5 DNA Sequencing of pAIF-1. Full coverage of the PCR cloned insert was
achieved by sequencing in two directions using either the sense primer or antisense
primer used in the original cloning PCR. The covered sequence is given with predicted
amino acid code indicated below in one letter code. Sequence of the flanking pTT3
backbone (pTT3) is also given, including restriction sites utilized for insertion (EcoRI,
XbaI). Sequencing with the sense primer gave residues starting about 25 bp into the
coding region for AIF-1 (right hand arrow) and continued through into the pTT3
backbone. Sequencing with the antisense primer gave complimentary sequence that
started 25 bp from the end of the AIF-1 coding region (left hand arrow) and continued
through the entire insert into the pTT3 backbone. Two single base pairs differed from
that of the documented mouse AIF-1 sequence and are marked in red. Both of these
mutations occurred at the third base pair of separate codons, and the predicted amino acid
sequence remained the same.
pTT3
3

EcoRI

G

A

XbaI

pTT3

84
"

Figure 4.6 Detection of AIF-1 in Transfected Cells and Purification. (A) 48 hours
after transfection , supernatants and lysates from 293T cells were examined by Western
immunoblot for AIF-1. A 17kDa protein was detected in both the lysate and supernatant
of cells transfected with pAIF-1 but not in cells transfected with empty vector (pTT3).
(B) Western blot analysis showed that this protein was also recognized by RGS-His, an
antibody specific for histidine tags. (C) rAIF-1 was purified to homogeneity from the
lysate of transfected cells using Ni-NTA sepharose beads. Mock preparations using the
cell lysates from 293T cells transfected with empty vector were purified in parallel as a
negative control. Both were separated by SDS-PAGE and analyzed for purity using
infrared imaging of Coommassie stained gels. Purified rAIF-1 preparations contained a
single band at 17kDa. Mock preparations contained no detectable bands. M = protein
ladder.

85

A)
gi:9506379 mass: 16900 total score: 338 peptides matched: 24
Nominal mass (Mr): 16900 calculated pI value: 8.71
Sequence coverage: 50%
"
1 MSQSRDLQGG KAFGLLKAQQ EERLEGINKQ FLDDPKYSND EDLPSKLEAF
51 KVKYMEFDLN GNGDIDIMSL KRMLEKLGVP KTHLELKRLI REVSSGSEET
101 FSYSDFLRMM LGKRSAILRM ILMYEEKNKE HKRPTGPPAK KAISELP

B)
gi:9506379 mass: 16900 total score: 367 peptides matched: 43
Nominal mass (Mr): 16900 calculated pI value: 8.71
Sequence coverage: 48%
"
1 MSQSRDLQGG KAFGLLKAQQ EERLEGINKQ FLDDPKYSND EDLPSKLEAF
51 KVKYMEFDLN GNGDIDIMSL KRMLEKLGVP KTHLELKRLI REVSSGSEET
101 FSYSDFLRMM LGKRSAILRM ILMYEEKNKE HKRPTGPPAK KAISELP

Figure 4.7 Identification of recombinant protein by mass spectrophotometry.
Recombinant protein purified from either the supernatant (A) or lysate (B) of transfected
293T cells were analyzed by tryptic digestion and mass spectrophotometry. Results
were compared to databases of known proteins and both were identified as murine AIF-1
(gi: 9586379). Amino acids covered by individual fragments are higjlighted in grey.
Both gave almost identical coverage maps at 50% and 48% respectively.
86
translational modifications could be identified by mass spectrophotometry. Endotoxin
levels in both rAIF-1 and mock preparations were below the level of detection.
In vitro bioactivity of rAIF-1. Previous studies using bacterial systems to generate
recombinant AIF-1 demonstrated the ability of AIF-1 to promote actin bundling (Sasaki
et al. 2001). Every batch of rAIF-1 was tested in vitro to ensure this basic bioactivity was
preserved in the preparation of the recombinant protein. When incubated with actin and a
polymerization buffer, rAIF-1 can increase the polymerization of actin, which can then
be pelleted by high-speed centrifugation and visualized by SDS-PAGE (figure 4.8A).
Neither the mock protein preparation nor bovine serum albumin (BSA) led to the shift of
actin from the supernatant to pellet. Combined results across multiple batches (figure
4.8B) showed a significant increase in the ratio of precipitated actin to total actin (ppt
actin/total actin) in the presence of rAIF-1 (0.31±0.13) compared to mock (0.07±0.06,
p<0.05) and BSA (0.03±0.02, p<0.05). !-actinin was used as a positive control and
resulted in a higher ratio of ppt actin/total actin (0.49±0.12), though not significantly
different than values obtained with rAIF-1.
Discussion
This chapter describes the successful cloning of murine AIF-1 from mouse spleen cDNA
and the production a His-tagged recombinant AIF-1 using the pTT3 vector and
expression in 293T cells. Identity of the recombinant product was confirmed by size
(figure 4.6A), antibody recognition, (figure 4.6A &4.6B), and mass spectrophotometry
(figure 4.7) with 50% coverage of the putative mouse AIF-1 sequence. The final rAIF-1
87
!"

#"

Figure 4.8 Actin bundling activity of rAIF-1. Preparations of rAIF-1 and mock controls were tested for actin-
bundling activity using a kit available from Cytoskeleton, Inc. Proteins were incubated with F-actin in polymerization
buffer and bundled actin was separated from actin monomers by centrifugation. Supernatants (S) and pellets (P) were
resolved by SDS-PAGE and stained with Coommassie Blue (A). Incubation of actin with mock or BSA (negative
control) resulted in most actin remaining in the supernatant. However, incubation of actin with rAIF-1 causes actin to
be pelleted, suggesting actin-polymerization activity. Band densiometries were measured and expressed as a ratio of
precipitated (ppt) actin over total actin. Combined results across multiple batches show a clear trend of increased
pelleted actin upon incubation with rAIF-1 or $-actinin (positive control) compared to mock or BSA controls. Data
represents means ± SEM.

88
preparation was of high purity (figure 4.6C) and free from endotoxin. Mock preparations
using supernatants or lysates from 293T cells transfected with empty vector were
subjected to all purification steps in parallel with rAIF-1 to serve as a negative control in
activity experiments.
Actin bundling activity was used as an in vitro assay to ensure individual batches of
recombinant protein were bioactive (figure 4.8). This same assay was previously used by
Sasaki et al to show that intracellular AIF-1 is an actin-regulating protein important to the
Rac signaling pathway (Sasaki et al. 2001). Their in vitro results showed a clear dose-
dependency on AIF-1 concentration and inhibition with a mutant AIF-1 (1-120) and were
confirmed with fluorescence and electron microscopy of the actin fibrils. The importance
of AIF-1 actin-binding and -polymerizing activities in the extracellular environment are
probably minimal, but were used here as an important quality control assay in the
production of the recombinant protein. The only other reported in vitro bioactivity of
AIF-1 is the reduced proliferation and differentiation of satellite cells in rat skeletal
muscle, but the in vitro model used by Kuschel et al, required techniques too involved for
the purpose of a bioactivity control (Kuschel et al. 2000).
In addition to the generation of a recombinant, bioactive AIF-1, the secretion of AIF-1 by
293T cells allowed for the examination of differences in structure of secreted AIF-1 and
intracellular AIF-1, which has not been reported in the literature. Abundant full length
(17kDa) AIF-1 was found in both the lysate and supernatant of 293T cells transfected
with pAIF-1 (figure 4..6A). As others have reported, the amino acid sequence of AIF-1
89
contains an internal 44 amino acid -KK-KR-GKR- sequence that is characteristic of
peptide hormone precursors (Chen et al. 1997). The double basic amino acids are
potential sites for activation cleavage, with the final site containing an additional signal
for formation of C-terminal amidation. These sites are conserved through most
mammalian AIF-1 coding regions sequenced including human, rat, and pig. We were
unable detect any cleaved AIF-1 products in the supernatant of transfected 293T cells.
No cleaved products have been reported in other studies showing secretion of AIF-1.
Since cleavage would result in the removal of the C-terminal His tag and possibly disturb
the epitope recognized by AIF-1 antibodies, detection of a cleaved product may not be
possible with available antibodies. Also, due to the use of a C-terminal His tag for
purification, purification would not include any products with C-terminal cleavage.
Thus, the rAIF-1 activity outlined in the next chapter is likely due to full length AIF-1
and not to any splice site variants or post-translationally cleaved products. Previous
reports of AIF-1 extracellular function have all used full-length protein (Chen et al. 1997;
Kuschel et al. 2000; Kimura et al. 2007). In addition, C-terminal deletions of AIF-1
showed decreased actin bundling activity in vitro (Sasaki et al. 2001), and transfection
with C-terminal deleted AIF-1 suppressed membrane ruffling in microglia (Ohsawa et al.
2000). However, it is not known if extracellular actions of AIF-1 are dependent on the C-
terminal sequence.
Intracellular and secreted rAIF-1 were examined independently by mass
spectrophotometry and no residues with post-translational modifications including
phosphorylation and glycosylation were identified on either. As described in Chapter 3,
90
AIF-1 amino acid sequence contains two putative O-linked glycosylation sites (135,144),
but is probably not exposed to glycosylating enzymes due to lack of a signal peptide and
entrance into the ER and Golgi. (Hansen et al. 1995). The phosphorylation of AIF-1 has
been previously reported (Imai and Kohsaka 2002) (Kohsaka et al. 2004), although
details on the phosphorylation site or method of detection were not described. Original
cloning of the mouse AIF-1 cDNA predicted potential phosphorylation sites for protein
kinase A (132–135), protein kinase C (69–72), casein kinase II (38–41, 45–48, 82–85,
95–98, 102–105), and tyrosine kinases (29–37) (Watano et al. 2001); all sites are highly
conserved across species (Watano et al. 2001). Analysis with the NetPhos2.0 server also
predicted several phosphorylation sites, most in agreement with the literature, and many
with high predictive values (figure 4.1). Further enrichment may be needed in order to
detect phosphorylated residues using mass spectrophotometry techniques if they occur at
a low-frequency. It is also possible that 293T cells may not be properly equipped for the
phosphorylation of AIF-1 or that stimulation of the cells may be necessary for
phosphorylation. Attempts to express rAIF-1 in the BV2 microglia cell line were not
successful using lipofectamine (data not shown). Future work examining the
phosphorylation of intracellular and secreted AIF-1 could include using electroporation
or lentiviral vectors for stable transfection of microglial or macrophage cell lines.
Although more costly and labor intensive, production of recombinant AIF-1 using a
mammalian expression system gave the greatest chance at uncovering the effects of
extracellular AIF-1 in the CNS. The His-tag allowed for the purification of rAIF-1 from
the supernatants of 293T cells in enough quantity to examine differences in
91
posttranslational modification of secreted versus intracellular AIF-1, although none were
seen. Because no differences in size or posttranslational modifications were found and
because rAIF-1 was much more abundant in the lysate, we used rAIF-1 isolated from the
lysate of 293T cells to examine extracellular activities as reported in the next chapter.
92
Chapter 5: Extracellular AIF-1 as a modulator of inflammation
Introduction
A major objective of this thesis was to determine how extracellular AIF-1 might
participate in neurodegenerative disease like MPTP neurotoxicity. Original studies that
showed secretion of AIF-1 by microglia cells in response to cerebellar granule cell death,
reported that addition of AIF-1 had no effect on neuronal survival (Tanaka and Koike
2002). The only clues as to the function of AIF-1 have come from the periphery, where
the secretion of AIF-1 is strongly associated with inflammation. Increased AIF-1 levels
have been found in the peritoneal fluid of women with endometriosis and AIF-1 levels
correlated with clinical scores in these patients (Koshiba et al. 2005). AIF-1 levels were
also increased in synovial fluid of patients with rheumatoid arthritis compared to patients
with osteoarthritis (Kimura et al. 2007). In vitro studies showed that secretion of AIF-1
by peritoneal macrophages was stimulated by pro-inflammatory molecules like IL-1# and
IFN" (Koshiba et al. 2005) and that AIF-1 could stimulate the secretion of pro-
inflammatory cytokines IL-1# and TNF! by synovial cells and the secretion of IL-6 by
both synovial cells and hPBMC (Kimura et al.).
Since no receptor has been identified for the activity of extracellular AIF-1 on peripheral
inflammatory cells or other cell types, receptor expression cannot be used to detect the
cellular targets of extracellular AIF-1 in the CNS. However, based on the similarities of
receptor expression by activated microglia and peripheral myeloid cells, it is reasonable
93
to suggest that extracellular AIF-1 in the CNS may be acting in an autocrine fashion,
modulating the inflammatory response in microglia themselves.
This study took a functional approach to determine the activity or recombinant AIF-1 in
the CNS and on microglia specifically. Intracerebral injections of rAIF-1 into the
striatum and SNpc of mice resulted in a more prolonged and diffuse activation of both
microglia and astrocytes compared to mock controls. Initial in vitro functional studies
used the BV2 microglia cell line and showed no effect of rAIF-1 on cell survival or the
expression of AIF-1 itself or iNOS in these cells. While cultures of mouse adult
microglia or splenocytes did not increase secretion of inflammatory cytokines in response
to rAIF-1, significant increases in the secretion of IL-6, TNF!, MCP-1, RANTES, and
IL-8 were seen with the addition of rAIF-1 to cultures of human PBMCs. Work
described in this chapter identifies new roles for AIF-1 in modulating inflammation in the
periphery; roles that may translate to increased inflammation in the CNS either through a
similar action of extracellular AIF-1 on microglia directly or on the function of
infiltrating peripheral cells.
Materials and Methods
AIF-1 injections. Adult male C57Bl6 mice (30-40g) were anesthetized with
tribromomethanol (250mg/kg i.p.) and positioned in a stereotaxic apparatus. Each mouse
was injected a total of 4 times: in the left hemisphere, AIF-1 protein was delivered to
both the striatum (AP +0.62mm, ML +2.0mm, DV +2.5mm from bregma) and the ventral
midbrain (AP -3.08mm, ML +1.25mm, DV +4.25mm from bregma); in the right
94
hemisphere, mock was delivered to both the striatum and ventral midbrain using the same
coordinates. Each injection consisted of 3µl of a 0.4mg/ml preparation of AIF-1 or 3µl of
mock and was delivered over the course of 5 minutes with a 26-guage Hamilton syringe.
After injection, the needle was left in place for 3 minutes before being slowly retracted.
Whole brains were isolated from mice 1, 2, 3, 5, or 7 days after injection and prepared for
immunohistochemistry as described in Chapter 2.
Immunohistochemistry. Coronal sections (30µm thick) were prepared from PFA-fixed
brains. Sections near each injection site (striatum and SNpc) were identified by taking
the last section with a visible needle track along with the next three sections and were
then stained as free-floating sections simultaneously with antibodies to AIF-1 and GFAP
as described previously (Chapter 2).
Cell Culture. The BV2 cell line was originally generated by the transformation of
primary microglia cell cultures with infection by the J2 retrovirus carrying a v-raf/v-myc
oncogene (Blasi et al. 1990). BV2 cells were obtained from Dr. Caleb Finch and were
maintained in DMEM/F12 supplemented with 10% FBS. Splenocytes and microglia
were isolated from untreated adult mice as described previously (Chapter 3) and
maintained in culture medium consisting of RMPI medium 1640 supplemented with 10%
FBS, 2.5mM L-glutamine, 1mM sodium pyruvate, 0.1mM non-essential amino acid,
50U/ml streptomycin, 50U/ml penicillin and 0.25µg/ml amphotercin B. Human PBMCs
were isolated from whole blood from normal healthy volunteers by density gradient
centrifugation using Ficoll-Paque (GE Healthcare).
95
MTS Proliferation. Cell survival was measured in BV2 cells using the CellTiter 96®
AQueous Non-Radioactive Cell Proliferation Assay (Promega, Madison, WI). This is a
colorimetric assay based on the conversion of a tetrazolium compound (MTS) into a
colored formazan product by living cells. Absorbance at 490nm is directly proportional
to the number of living cells in culture. The linear range of cell viability was calculated
for BV2 cells and was determined to be 2x10
4
cells/well to 1x10
5
cells/well in a 96-well
plate. BV2 cells were plated at 5 x 10
3
cells/well in 100µl of culture medium and various
concentrations of rAIF-1 or mock were added.

After 24 hours, 20µl of MTS solution was
added to cell cultures and incubated from 1 to 4 hours at 37°C in a humidified, 5% CO
2

atmosphere. Wells with culture medium and MTS solution but no cells were used as a
blank. Absorbance at 490nm was measured using a plate reader and % cell survival was
calculated as a ratio between A
490
of control and test samples.
RNA isolation. Total RNA was isolated from BV2 cells by Trizol reagent (Gibco) or
RNeasy Kit (QIAGEN) according to the supplier’s protocol. To measure concentration,
RNA samples were diluted 1:75 in TE buffer (10mM Tris-HCl, 1mM EDTA, pH 7.0) and
absorbance at 260nm and 280 nm was measured using a UV spectrophotometer. RNA
samples with A
260
/A
280
ratio of 1.6 to 2.0 were used for quantitative RT-PCR analysis.
Quantitative RT-PCR. One-step quantitative real time polymerase chain reaction (RT-
PCR) was performed using the QuantiTect SYBR Green RT-PCR Kit (QIAGEN) and the
Eppendorf Mastercycler (Eppendorf; Hamburg, Germany). The following thermal
program was used: stage 1, 95°C, 15 min; stage 2, 45 cycles through 95°C for 15s, 58°C
96
for 30s, and 72°C for 30 s, stage 3, 95°C for 15 s. Primers are given in Table 5.1.
Standard conditions for PCR were 0.4µg template, 1xSYBR green master mix, forward
and reverse primers (2.5µM), 2ul RT mix, and RNAse-free water to a final volume of
50µl. Target and reference transcripts were detected in separate tubes. Reactions without
template were included for each primer pair. Relative expression of transcripts was
calculated using 2
-&&Ct
which compares the difference in crossing points (Ct, the PCR
cycle number when product can first be detected) between the target and reference gene
(&Ct) between treated and untreated controls (&&Ct). These values give an estimate of
target mRNA present relative to untreated groups.
Table 5.1 Real-time RT-PCR Primers
Gene Direction Primer Amplicon (bp)
AIF-1
forward
reverse
ATCAACAAGCAATTCCTCGATGA
CAGCATTCGCTTCAAGGACATA
144
iNOS
forward
reverse
GTTCTCAGCCCAACAATACAAGA
GTGGACGGGTCGATGTCAC
127
GAPDH
forward
reverse
AGGTCGGTGTGAACGGATTTG
TGTAGACCATGTAGTTGAGGTCA
123

Measurement of Cytokine Levels. For cytokine secretion studies, hPBMC and
splenocytes were plated at 1x10
6
cells/ml in 12-well plates, while microglia were plated
at 1x10
6
cells/ml in 96-well plates. Microglia and splenocytes were plated in culture
medium with 10% FBS and hPBMCs were plated in RPMI 1640 alone. After 24 or 48
hours, cell supernatants were collected and cleared of cell debris by low speed
centrifugation. Samples were stored at -20°C until analysis. Concentrations of cytokines
and chemokines were measured using Cytometric Bead Array (CBA) kits available from
97
BD Biosciences and analyzed using a FACSCalibur. CBA is a multiplexed, bead-based
immunoassay which allows for the measurement of up to eight cytokines simultaneously
in as little as 20µl of sample. The principle of CBA kits is similar to that of a capture
ELISA, except it is complete in suspension. Capture beads coated with specific
antibodies for each chemokine or cytokine to be measured can be distinguished by flow
cytometric analysis on a single channel (FL3). A mixture of these beads are incubated
with sample and PE-conjugated detection antibodies. Thus, while beads are
distinguished based on intensity on the FL-3 channel, levels of bound protein are
indicated by the FL-2 channel. Concentrations can be calculated by comparing samples
to a standard curve using recombinant protein provided by the kit. For mouse microglia
and splenocytes supernatants, the Mouse Inflammation Kit, which measures mouse IL-6,
IL-10, TNF$, IFN", MCP-1, and IL-12p70 was used. For human hPBMCs a total of 3
kits were used. The Human Th1/Th2 Cytokine kit measures IL-2, IL-4, IL-5, IL-10,
TNF$, IFN". The Human Inflammation Cytokine kit measures IL-8, IL-1b, IL-6, IL-10,
TNF$, and IL-12p70. The Human ChemokinekKit measures IL-8, RANTES, MCP-1,
IP-10 and MIG-1. In the assays reported here, standard curves were low for IFN" in the
Mouse Inflammation Kit and IP-10 and MIG-1 in the Human Chemokine kit, so these
results were not analyzed.
Results
Intracerebral injection of rAIF-1. To examine the in vivo effects of rAIF-1 on glial
activation, mice were injected with AIF-1 in both the striatum and ventral midbrain of
98
one hemisphere and mock was injected in both the striatum and ventral midbrain of the
opposite hemisphere. The activation of microglia and astrocytes was analyzed by
immunohistochemical analysis of AIF-1 and GFAP, respectively. At early time points (1
and 2 days) after injection, both AIF-1 and mock injected hemispheres showed extensive
amounts of glial activation surrounding injection sites in both the striatum and SNpc, and
no clear difference could be distinguished between mock and rAIF-1 injected
hemispheres (data not shown). Mice were also examined at later time points (5 and 7
days) and both microglial immunoreactivity for AIF-1 and astrocytic immunoreactivity
for GFAP were reduced throughout the brain. However, at 7 days after injection, a clear
difference in the breadth of glial activation could be distinguished between the two
treatments/hemispheres (figure 5.1). Microglia activation, demonstrated by increased
AIF-1 immunoreactivity) in the striatum injected with rAIF-1 remained diffuse and
extended out from the injection site (figure 5.1A), whereas fewer microglia surrounding
the mock injection site showed increased levels of AIF-1 immunoreactivity (figure 5.1B).
Analysis for GFAP staining by astrocytes showed a similar pattern: at the rAIF-1 injected
striatum, astrocytes with increased immunoreactivity for GFAP extended well into the
surrounding area (figure 5.1C), while in the mock injected striatum GFAP
immunoreactive cells were concentrated around the site of injection (figure 5.1D). Lower
magnification of GFAP stained striatum shows this difference even more clearly (figure
5.1E,F). Thus, in vivo, rAIF-1 acts to prolong the activation of microglia and astrocytes
after stereotactic injection.
99

Figure 5.1 Increased glial activation surrounding intracranial injections of rAIF1. Adult mice were injected with
both rAIF-1 in the striatum (A,C,E) and ventral midbrain of the left hemisphere and with mock in the striatum (B,D,F) and
ventral midbrain of the right hemishere. Animals were sacrificed 7 days later and activation of microglia (A,B) and
astrocytes (B-F) was analyzed by immunohistochemisitry with Iba1 and GFAP, respectively. Lower magnification of
GFAP stained sections are shown in E & F. Scarring indicating the injection site is highlighted by the dashed circle.
Images are representative of three different experiments. Scale bars represent 50#m (A-D), 50#m (E,F).

100
rAIF-1 effects on Cell Survival. Recombinant AIF-1 has been shown to stimulate the
proliferation of synovial cells (Kimura et al. 2007) and decrease proliferation of muscle
satellite cells (Kuschel et al. 2000). rAIF-1 was added to cultures of BV2 microglia to
determine if rAIF-1 had any effects on the proliferation of microglia. BV2 is a murine
microglia cell line that retains many of the phenotypical and functional properties
described for freshly isolated microglia (Blasi et al. 1990). As shown in figure 5.2, the
addition of rAIF-1 or mock preparations at concentrations ranging from 1fg/ml to 1µg/ml
did not alter the BV2 cell survival from that of untreated controls.
rAIF-1 effects on gene expression. Cytokines and chemokines that act in an autocrine
fashion can often regulate their own expression (Hensel et al. 1987; Browning et al.
2000). Quantitative RT-PCR was used to measure AIF-1 transcript levels in cultured
BV2 cells. As previously demonstrated in myeloid cell lines and bone marrow derived
macrophages (Utans et al.), AIF-1 transcript levels in BV2 cells were increased after just
24 hours of stimulation with either LPS (10.02±0.89 times control, p<0.001) or IFN!
(130.52±0.43 times control, p<0.001) (figure 5.3A). rAIF-1 was first added to
unstimulated BV2 cells to determine if it could alter basal levels of AIF-1 mRNA.
Addition of recombinant AIF-1 to unstiumlated BV2 cells did not alter AIF-1 transcript
levels after 4 hours or 24 hours (figures 5.3B, & 5.3C). At 48 hours (figure 5.3D), AIF-1
transcript levels were decreased in cells treated with 500ng/ml rAIF-1 (0.64±0.07 times
control), but a similar decrease was also seen in mock treated cells at this concentration
(0.46±0.13 times control, p=0.185), suggesting that this reduction in AIF-1 transcript
101

Figure 5.2 rAIF-1 has no effect on cell survival of BV2 microglia. BV2 cells were
cultured in medium alone (control) or in culture medium containing the indicated
concentrations of rAIF-1 or a corresponding dilution of mock. Cell survival was
measured after 24 hours using the MTS proliferation assay as described in Materials and
Methods. Values were standardized to cells cultured in medium alone and represent the
mean ± SEM. Results are representative of three experiments performed in triplicate.
102
Figure 5.3 rAIF-1 does not autoregulate AIF-1 expression in BV2 microglia. BV2 cells were stimulated
with inflammatory mediators for 24 hours (A; LPS=100ng/ml; IFN"=10ng/ml) or with rAIF-1 or mock controls
for 4 (A), 24(C) or 48 (D) hours. Total RNA was isolated and AIF-1 transcripts were measured using real time
RT-PCR as described in Materials and Methods. Data represent the mean ± SEM (n=3). *, p<0.001

103
levels was not specific. rAIF-1 was also added to cells in combination with LPS or IFN"
to determine if rAIF-1 could moderate the upregulation of AIF-1 transcript levels seen
with LPS or IFN" stimulation, but AIF-1 expression was not different from controls (data
not shown). If extracellular AIF-1 does not act to autoregulate its own expression levels
it may still regulate other genes associated with activation. Yang et al showed that
overexpression of AIF-1 in a macrophage cell line resulted in increased iNOS transcript
levels while siRNA silencing led to decreased iNOS transcript levels; this study did not
address the possibility of AIF-1 acting in a secreted fashion (Yang et al. 2005). Addition
of rAIF-1 at 20ng/ml or 100ng/ml or corresponding dilutions of mock preparations for 24
hours did not result in any significant changes in iNOS transcript levels in BV2 cells
(figure 5.4).
rAIF-1 effects on cytokine secretion. One of the only identified functions of extracellular
AIF-1 is the ability to stimulate secretion of cytokines from peripheral inflammatory
cells. To establish whether rAIF-1 could also stimulate the secretion of cytokines from
microglia, cytokine levels were measured in supernatants collected from ex vivo cultures
of microglia isolated from adult normal mice. Microglia cultures were grown overnight
in culture medium alone or in culture medium supplemented with 100ng/ml rAIF-1 or a
corresponding dilution of mock. Microglia were also stimulated with LPS as a positive
control. Supernatants were collected after 24 hours and cytokine levels were measured
using the Mouse Inflammation CBA kit. Even with LPS stimulation, levels of TNF$ or
MCP-1 were below detection limits (data not shown). Only low levels of IL-6, IL-10,
and IL-12 were found in supernatants of ex vivo microglia cultures (figure 5.5), and the
104

Figure 5.4 rAIF-1 does not alter iNOS expression in BV2 microglia. BV2 cells were
cultured for 24 hours in medium alone (Untx) or in medium containing 20ng/ml or
100ng/ml of rAIF-1 or mock. Total RNA was isolated and iNOS transcripts were
measured using real time RT-PCR as described in Materials and Methods. Data represent
the mean ± SEM (n=3).

105

Figure 5.5 Microglia isolated from adult mice secrete low levels of inflammatory
cytokines ex vivo. Microglia cells were isolated from normal young adult mice and
incubated overnight in culture medium alone (Control), mock control (Mock), 100ng/ml
rAIF-1 (AIF-1), or 100ng/mL LPS (LPS). Cytokine levels in supernatants collected after
24 hours were analyzed using a Mouse Inflammation CBA Kit. Levels of TNF$ and
MCP-1 were below detection limits and not included. Data represent the mean ± SEM
(n=3).

106
addition of rAIF-1 did not significantly increase levels of these cytokines at 24 hours.
LPS stimulation produced slightly increased levels of IL-10, but had no effect on
secretion of other cytokines, suggesting that ex vivo cultures of adult microglia may not
be refractory to stimulation.
The stimulation of cytokines by AIF-1 may be a phenomenon reserved for the peripheral
immune system. To determine if murine recombinant AIF-1 might act on peripheral
immune cells, the secretion of cytokines from murine splenocytes was measured using
the same Mouse Inflammation CBA kit. No secretion of IL-6, IL-10, TNF$, MCP-1, or
IL-12 could be detected in splenocyte preparations after 24 or 48 hours (data not shown).
Although the recombinant AIF-1 used in this study demonstrates actin bundling activity
(figure 4.8), the in vitro functional studies described so far in this chapter were unable to
detect functional effects of recombinant AIF-1. There is high homology between the
mouse and human protein sequences, and so the effects of our recombinant AIF-1 was
tested on the secretion of cytokines by hPBMCs. In agreement with previous reports
using a recombinant protein based on the human AIF-1 sequence (Kimura et al.), the
addition of 100ng/ml rAIF-1 to hPBMC grown in serum-free conditions did significantly
increase IL-6 levels (125.79±19.77pg/ml) compared to controls (9.802±6.77pg/ml,
p<0.01) (figure 5.6A). IL-6 levels were below the level of detection in supernatants of
cells treated with mock preparations. LPS was used as a positive control and greatly
increased the secretion of IL-6 in these cells (3867.65±1173.49pg/ml; p<0.05 vs. control).
107

Figure 5.6 rAIF-1 stimulates secretion of IL-6 and TNF! from human PBMC.
PBMC from healthy individuals were incubated in RPMI alone and stimulated for 24
hours with mock, 100ng/mL AIF-1, or 100ng/mL LPS. IL-6 and TNF$ levels were
measured in collected supernatants using the Human Inflammation Cytokine CBA kit.
Data represent the mean ± SEM (n=3). * p< 0.05 vs. untreated controls; * p<0.05 vs.
control; #, p<0.01 vs. control; ‡, p<0.001 vs. control.

108
A similar pattern was seen in the levels of TNF! in these supernatants (figure 5.6B).
TNF! levels were below the level of detection in the supernatants of untreated cells and
those treated with mock controls. Treatment with rAIF-1 resulted in significantly
increased levels of TNFa (18.19±1.78pg/ml, p<0.001), and LPS treatment resulted in a
dramatic increase in the secretion of TNF! (1904.96±416.40pg/ml, p<0.05 vs. control).
The increases in IL-6 and TNF! levels were particular as other cytokines were measured
in these same supernatants as part of the Human Th1/Th2 Cytokine and Human
Inflammation cytokine CBA kits and did not show effects of stimulation with rAIF-1.
IL-2, IL-5, IL-12, and IFN" were below the level of detection for all treatments including
LPS stimulation (data not shown). IL-10, IL-1#, and IL-4 levels were all significantly
increased with LPSstimulation (IL-10: 20.72±5.30, p<0.01; IL-1#: 745.66±109.14pg/ml,
p<0.05; IL-4: 46.26±2.074pg/ml; p<0.001), but rAIF-1 did not stimulate the secretion of
these cytokines (figure 5.7).
Levels of chemokines were also measured in the same supernatants and the secretion of
MCP-1, RANTES, and IL-8 were all increased by treatment with rAIF-1 (figure 5.8).
MCP-1 levels (figure 5.8A) were increased to similar levels by treatment with either
rAIF-1 (213.3±15.71pg/ml) or LPS (194.97±11.97), and although both were significantly
different than supernatants of untreated cells (123.03±31.97pg/ml, p<0.01), treatment
with rAIF-1 was not significantly different than mock controls (101.93±31.974pg/ml,
p=0.064). Significant increases in RANTES (figure 5.8B) were seen with rAIF-1
(4,999.63±327.79pg/ml) and LPS treatment (6,020.93±327.79pg/ml) compared to both
109

Figure 5.7 rAIF-1 does not stimulate the secretion of IL-10, IL-1#, or IL-4 from
human PBMC. PBMC from healthy individuals were incubated in RPMI alone or
stimulated for 24 hours with mock, 100 ng/mL AIF-1, or 100 ng/mL LPS. Cytokines
were measured using the Human Inflammation or Th1/Th2 CBA kits. Other cytokines
measured that were below the level of detection for all treatments included IL-2, IL-5, IL-
12, and IFN". Data represent the mean ± SEM (n=3).
110

Figure 5.8 rAIF-1 stimulates the secretion of MCP-1, RANTES, and IL-8 from
human PBMC. PBMC from healthy individuals were incubated in RPMI alone
(Control) or stimulated for 24 hours with mock, 100ng/ml AIF-1, or 100ng/ml LPS.
MCP-1 (A), RANTES (B) and IL-8 (C) levels were measured in collected supernatants as
part of the Human Chemokine CBA kit. Data represent the mean ± SEM (n=3). *,
p<0.01 vs. control; #, p<0.001 vs control
111
untreated (3,007.53±63.78pg/ml, p<0.01 vs. rAIF-1, p<0.001 vs. LPS) and mock
(3,564.73±63.43pg/ml, p<0.05) controls. Finally, significant increases in IL-8 levels
(figure 5.8C) were also seen in the supernatants of PBMCs treated with rAIF-1
(3,220.33±205.36pg/ml) or LPS (14,680.67±333.28pg/ml) as compared to untreated
(1,085.6±41.79pg/ml, p<0.001) or mock controls (1,529.83±426.08pg/ml, p<0.05).
Although no effects were seen on the secretion of cytokines by microglia or splenocytes,
the effects of rAIF-1 on the secretion of cytokines and chemokines by hPBMC confirm
functional activity of the murine recombinant AIF-1 used in this study. This data not
only supports functions of extracellular AIF-1 already described in the literature, but also
extends the functional significance of extracellular AIF-1 to the secretion of select
chemokines.
Discussion
In vivo injection of rAIF-1 into the brains of normal mice resulted in a more prolonged
and diffuse glial reaction than mock controls (figure 5.1), suggesting that AIF-1 may be
important in the persistence of microglial and astrocytic activation in response to injury.
This may be especially important to neurodegenerative diseases like PD as chronic
microglial activation is suspected as a culprit leading to amplification of neuronal
degeneration past initial insult. In fact, activated microglia are found in the brains of
mice and non-human primates several months after MPTP treatment (McGeer et al. 2003;
Barcia et al. 2004; Yasuda et al. 2007). This was also seen in humans who inadvertently
self-administered MPTP; microglia activation was seen up to 16 years after MPTP
112
intoxication (Langston et al. 1999). Thus, if AIF-1 is secreted in response to neuronal
signals or inflammatory factors and then acts to further amplify glial reactivity, this may
be a contributing factor to the progressive neuronal damage seen in the disease.
In vitro work in this study examined the activity of recombinant AIF-1 on cultures of
BV2 microglia cells, murine primary microglia cells, murine splenocytes, and human
PBMCs. rAIF-1 was not toxic to BV2 cells nor did it show any effects on proliferation of
cells in culture (figure 5.2) or the expression of transcripts associated with microglia
activation including AIF-1 itself (figure 5.3) and iNOS (figure 5.4). These experiments
demonstrated clearly the importance of the mock control, as some activity that could be
seen with addition of rAIF-1 could also be seen with addition of mock. Mock control
was prepared by transfecting 293T cells with empty vector and purifying the lysate from
these cells in parallel with lysates from 293T cells transfected with pAIF-1 (Chapter 4).
Thus, mock not only serves as a buffer control because it is prepared in the same DPBS
buffer as our recombinant protein, but mock would also contain any trace amounts of
lysate proteins that may nonspecifically bind to the Ni-sepharose and endure the
purification steps that follow. It is unclear why mock would have any effect on the
proliferation or gene transcription in these cells, but it is useful in limiting focus to
activities specific to AIF-1.
While the focus of functional studies of extracellular AIF-1 as they pertain to this thesis
is on its role in the CNS, extending this study to peripheral immune cells allowed for the
identification of clear functional activities of the mouse recombinant protein. As others
113
have shown previously for human recombinant AIF-1 (Kimura et al.), mouse
recombinant protein stimulated the secretion of IL-6 and TNF! by hPBMC (figures 5.6A
& 5.6B). Mouse AIF-1 and human AIF-1 share 89% homology in their amino acid
sequence (Watano et al.), so it is not surprising that the mouse protein would share
functional characteristics with the human protein and be recognized by receptors for AIF-
1 on human cells (which remain to be identified). In addition to stimulation of the
secretion of cytokines, stimulation of hPBMCs with rAIF-1 also led to increases in the
secretion of the chemokines MCP-1, RANTES, and IL-8 (figure 5.8). This extracellular
AIF-1 activity may translate function in CNS injury either by a similar effect on cytokine
and chemokine secretion by microglia or may be limited to altering the secretory profile
of infiltrating peripheral mononuclear cells. rAIF-1 did not stimulate the secretion of IL-
6, TNF!, or MCP-1 in ex vivo cultures of adult microglia (figure 5.5; RANTES and IL-8
were not measured). However, further studies should be designed to examine the effect
of extracellular AIF-1 on cytokine secretion in the brain in vivo since clear effects of
intracerebral injections of AIF-1 on microglia activity were seen (figure 5.1). In fact,
increased cytokine and chemokine secretion by microglia or infiltrating cells in response
to extracellular AIF-1 could explain the perisistant activation seen with intracerebral
injection of rAIF-1.
The cytokines and chemokines stimulated by extracellular AIF-1 may contribute to the
pathogenesis of MPTP-lesioning by promoting inflammatory responses. TNF!, IL-6,
and MCP-1 are found at increased levels in the CNS of MPTP-lesioned mice and PD
patients (Mogi et al.; Hunot et al. 1999) and have all been examined for their role in the
114
progression of neuronal degeneration. Animals deficient in either TNF! or both
receptors for TNF! (TNFR1 and TNFR2) were more resistant to neuronal damage
caused by MPTP-lesioning (Sriram et al.; Ferger et al. 2004; Sriram et al.). Data
regarding the contributions of IL-6 in MPTP lesioning are not as clear. One group found
increased vulnerability of dopaminergic neurons to MPTP in IL-6 knockout mice (Bolin
et al. 2002), while another group observed impaired microgliosis and an increase in
dopaminergic cell death (Cardenas and Bolin 2003). Interestingly, IL-6 levels are higher
in female MPTP-lesioned mice, which are also more resistant to dopaminergic loss than
males (Ciesielska et al. 2007). MCP-1 is a chemokines involved in the recruitment and
activation of monocytes and microglia (Hayashi et al. 1995; Gu et al. 1999). Mice
deficient in MCP-1, its receptor CCR2, or both ligand and receptor did not differ in
dopaminergic damage due to MPTP (Kalkonde et al. 2007) .RANTES is a chemokines
that promotes the migration of many types of leukocytes and promotes the infiltration of
these cells to sites of inflammation (Appay and Rowland-Jones 2001). The involvement
of RANTES in MPTP-lesioning has not been described, but slightly increased levels have
been found in PD patients taking levodopa (Gangemi et al. 2003). The importance of
AIF-1 stimulation of secretion of these cytokines may be limited to the periphery which
is also important in PD. PBMCs from PD patients increased basal and LPS stimulated
secretion of MCP-1, RANTES, MIP-1alpha, IL-8, IFN", IL-1# and TNF! (Reale et al.
2008).
Most of the in vitro work described in this chapter focuses on the direct activity of
extracellular AIF-1 on microglia and peripheral inflammatory cells, but AIF-1 may also
115
act in the CNS through astrocytes as also suggested by the in vivo injection of rAIF-1
(figure 5.1). In fact, astrocytes are thought to be the primary source of IL-6 and MCP-1
in the MPTP-lesioned mouse (Bolin et al. 2002; Kalkonde et al. 2007), so the affects of
AIF-1 on the cytokine secretion by astrocytes should also be examined.
116
Chapter 6: Conclusion
The work presented in this thesis calls for increased consideration of the functional
import of AIF-1 in microglia activation, beyond its usefulness as a specific marker for
microglia and macrophages in CNS injury. In addition to establishing the popular
Iba1/AIF-1 antibody, Imai and colleagues have devoted much work to establishing the
importance of this protein as an intracellular participant in the cytoskeleton remodeling
required for migration and phagocytosis (Imai and Kohsaka 2002). One of the first
studies describing the secretion of AIF-1 demonstrated its secretion by microglia in
response to apoptotic neurons (Tanaka and Koike 2002). However, interest in AIF-1 as
an extracellular molecule in the CNS was not pushed forward from these initial reports.
The little that has subsequently been learnt about AIF-1 secretion has come from studies
in the periphery (Chen et al. 1997; Kuschel et al. 2000; Koshiba et al. 2005; Kimura et al.
2007).
The hypothesis of this thesis was that extracellular AIF-1 could modulate microglia
activities important in the pathogenesis of the MPTP mouse model of dopaminergic
neurotoxicity. Because little information has been reported regarding the expression of
AIF-1 in the MPTP model, a time course of AIF-1 expression in the striatum and SNpc
was performed using immunohistochemistry. Interestingly, increased expression of AIF-
1 by microglia extended well beyond the 4 day window previously reported for microglia
activation (Kohutnicka et al. 1998; Liberatore et al. 1999; Wu et al. 2003). In our study,
cells with greatly increased AIF-1 staining were not only present at time points after the
117
first four days, but also increased in number up to day 10. Reasons for the discrepancy
between the kinetics of AIF-1 expression and other time course evaluations of microglia
activation reported using Mac-1 and isolectin-B4 as markers are unclear, but may
represent a greater importance of AIF-1 expression in the days following active neuronal
death. The increased AIF-1 expression was also associated with clustering of cells which
is suggestive of active phagocytosis (Streit and Kreutzberg 1988), a known functional
activity of intracellular AIF-1.
Based on the early report of AIF-1 secretion in response to apoptotic neurons and soluble
factors leaked by dying neurons such as ATP (Tanaka and Koike 2002), the increased
AIF-1 expression in response to MPTP neurotoxicity is highly suggestive of AIF-1
secretion. Although stimulation of AIF-1 in response to MPTP neurotoxicity was not
explicitly undertaken, this report demonstrates for the first time the secretion of AIF-1 in
ex vivo cultures of adult microglia (figure 3.3). Recent studies demonstrating
phenotypical and functional differences between microglia cultures prepared from
newborn rodents and mature microglia isolated from adults made this an important
distinction.
The rest of the work presented here was focused on determining the function of
extracellular AIF-1 in the CNS, specifically on regulating functions of microglia
activation. Because both the differences between the secreted protein and intracellular
AIF-1 and the mechanisms of AIF-1 secretion were unknown, recombinant AIF-1 was
using a mammalian cell line. It should be noted that this is the first report of the
118
production of murine AIF-1 and of any AIF-1 using a mammalian cell line. Secretion of
AIF-1 by the 293T cell line used for AIF-1 production allowed for the analysis of
differences between the secreted protein and intracellular proteins, but no differences in
size, sequence, or post-translational modifications were identified by mass
spectrophotometery.
When this recombinant protein was injected directly into the striatum or SNpc of normal
mice, a more prolonged and diffuse activation of both microglia and astrocytes was seen
near the injection site, compared to controls (figure 5.1). In vitro studies focused on the
effects seen on microglia activation, as the activity of this cell is more implicated in the
pathology of MPTP-lesioning and PD (Vila et al. 2001) Early studies with BV2 cell line
showed no activity (figures 5.2-5.4). BV2 cells are a transformed cell line, and although
AIF-1 expression was stimulated with either LPS or IFN" (figure 5.3A), these cells have
limitations as a model of microglia and may represent a more activated state than primary
adult microglia. Although primary microglia isolated from adult mice may be the most
accurate in vitro model for this study, the yield of these cells (2-5 x 10
5
cells per animal)
was a major limiting factor. The only functional studies undertaken with primary
microglia was the effects of AIF-1 on secretion of cytokines by these cells based on
recent reports on the stimulation of cytokine secretion in human synovial cells and
PBMCs (Kimura et al. 2007). Addition of rAIF-1 and even LPS did not stimulate the
secretion of inflammatory cytokines by these cells (figure 5.5). It is unclear whether
these results represent limitations of the specific assay or resistance of these cultures to
stimulation, as not many functional studies of these cultures have been reported. Due to
119
the limitation of cell yields, it may be worthwhile to return to microglia cultures prepared
from newborn mice in terms of cell yield as many more studies into the effects of AIF-1
on microglia function including phagocytosis and migration would be worthwhile.
Extension of the study of recombinant AIF-1 activity into human PBMC was initiated as
a means of validating the activity of the recombinant protein. Although recombinant
AIF-1 was able to increase actin bundling in vitro (figure 4.8), all functional studies at the
cellular level failed to demonstrate activity of recombinant AIF-1. The study examining
AIF-1 in the synovial fluids of rheumatoid arthritis and demonstrating the activity of
extracellular AIF-1 on synovial cells and human PBMC was published a few years after
this thesis was initiated (Kimura et al. 2007). Given the high homology of the protein
across species (Watano et al. 2001), this seemed like a relatively easy way to validate
activity of our recombinant protein. The availability of CBA technology allowed for the
simultaneous measurement of several cytokines and chemokines. In addition to the IL-6
secretion originally reported, AIF-1 also stimulated the secretion of TNF!, MCP-1,
RANTES, and IL-8. Several of these molecules are no strangers to the study of
inflammationin MPTP-lesioning and PD, and if stimulated by extracellular AIF-1 in the
CNS, either in microglia or infiltrating peripheral cells, would have potential
consequences on the amplification and propagation of the inflammatory response as well
as neuronal death. Thus, several important findings were included in this work, and
published results can hopefully act as a springboard for future studies validating the
importance of extracellular AIF-1 in the MPTP neurotoxicity model of injury.
120
In fact, continuation of the work of this thesis could take two directions. First, many
questions remain regarding the secretion of AIF-1. While secretion was demonstrated in
ex vivo cultures of microglia isolated from adult normal mice for the first time (figure
3.3), further work could identify factors that stimulate AIF-1 secretion in these cells. In
vivo stimulation with a single i.p. injection of LPS did not result in increased AIF-1
secretion by ex vivo cultures of microglia (figure 3.4), but in vitro stimulation with factors
known to induce AIF-1 secretion such as ATP (Tanaka and Koike 2002) and IL-1#
(Koshiba et al. 2005) should be explored. Ideally, the demonstration of increased
secretion of AIF-1 by activated microglia isolated from MPTP-lesioned mice would
establish the presence of extracellular AIF-1 in this model, but this may be technically
difficult considering potential cell yields and the need for demonstrating this explicitly is
debatable.
Perhaps a more interesting question regarding the secretion of AIF-1 is the identity of the
AIF-1-immunoreactive 15kDa band identified in western blots of cell supernatants of not
only microglia but also murine splenocytes and THP-1 cells (figure 3.1A, 3.2A, 3.3C).
This band had not been previously identified in studies reporting secretion of AIF-1, but
many of these studies used ELISA as a means of measuring AIF-1 in supernatants. Mass
spectrophotometery could be used to confirm this band is a truncated AIF-1 product, and
may provide insight into how exactly AIF-1 is secreted. The mechanisms of AIF-1
secretion remain a bit of a mystery as it has no classical N-terminal signal sequence,
however a growing number of proteins are being identified as having a nonclassical route
121
to secretion (Nickel 2003), and our analysis of the AIF-1 sequence using the SecretomeP
server gave a high score for prediction of the nonclassical secretion of AIF-1.
A much more important outcome of continuation of this work would be the
documentation of direct functions of AIF-1 in the CNS. As mentioned previously, the
cell yields of primary microglia isolated from adult mice was a major limitation in this
study, so returning to cultures prepared from newborn mice may be worthwhile until
specific functions are identified. These functions could then be substantiated in adult
microglia cultures or in vivo. In addition, although this thesis chose to focus on the
effects of extracellular AIF-1 on microglia cells, stereotactic injections also showed
effects on the activation of astrocytes. Effects of AIF-1 on astrocyte functions such as
proliferation and cytokine secretion should be examined in vitro.
Due to the robust response of extracellular AIF-1 on human PBMCs and the ease of
access to these cells, a good strategy in identifying the role of AIF-1 in the CNS would be
to identify the receptor for AIF-1 extracellular activity in this system. Receptor
expression could then be used to determine the cellular target of AIF-1 in the CNS and to
explore the signaling pathways involved. Although this work would be tedious and time-
consuming, it has the potential of firmly substantiating the work of extracellular AIF-1
which would also be applicable to interests in AIF-1 peripherally including rheumatoid
arthritis (Kimura et al. 2007), endometriosis (Kohsaka et al. 2004), systemic sclerosis
(Del Galdo et al. 2006), and vascular disease (Kelemen and Autieri 2005).
122
Although potentially significant to all the clinical syndromes just mentioned, the work of
this thesis was focused on questions pertinent to the establishment of new therapies for
treatment of Parkinson’s disease and related disorders. Current therapeutic interventions
are still mainly symptomatic, and are also implicated in participating in progression of the
disease (Marras and Lang 2008). Therapies are need to stop the progression of
neurodegeneration. There is mounting evidence of the cytotoxic potential of microglia
activation in Parkinson’s disease, especially the deleterious effects of nitric oxide
production by these cells (Hunot et al. 1996). Identifying targets that can modulate the
immune response could translate into therapeutic strategies to stop the chronic microglia
activation seen PD or at least restrain cytotoxic activities of microglia and/or promote
neuroprotective functions of microglia. With this in mind, the stereotactic injection of
recombinant AIF-1, which resulted in a more prolonged and diffuse activation of glia at
the site of injection compared to controls, may have yielded some of the most exciting
results of this thesis. If extracellular AIF-1 is proven to be involved in the chronic
activation of microglia seen in these disorders (perhaps through the stimulation of pro-
inflammatory cytokine secretion by either microglia, astrocytes, or infiltrating cells), this
could lead to therapeutic treatments to hamper microglia activation.
In conclusion, the work presented here provides strong evidence for a closer examination
of the extracellular effects on AIF-1 in the CNS as it may have direct effects on local glia
and infiltrating peripheral inflammatory cells in neurodegenerative disease.
123
Bibliography
Ajami, B., J. L. Bennett, C. Krieger, W. Tetzlaff and F. M. Rossi (2007). Local self-
renewal can sustain CNS microglia maintenance and function throughout adult
life. Nat Neurosci 10(12): 1538-43.
Alliot, F., I. Godin and B. Pessac (1999). Microglia derive from progenitors, originating
from the yolk sac, and which proliferate in the brain. Brain Res Dev Brain Res
117(2): 145-52.
Alliot, F., E. Lecain, B. Grima and B. Pessac (1991). Microglial progenitors with a high
proliferative potential in the embryonic and adult mouse brain. Proc Natl Acad
Sci U S A 88(4): 1541-5.
Appay, V. and S. L. Rowland-Jones (2001). RANTES: a versatile and controversial
chemokine. Trends Immunol 22(2): 83-7.
Asanuma, M., I. Miyazaki, F. J. Diaz-Corrales and N. Ogawa (2004). Quinone formation
as dopaminergic neuron-specific oxidative stress in the pathogenesis of sporadic
Parkinson's disease and neurotoxin-induced parkinsonism. Acta Med Okayama
58(5): 221-33.
Ashwell, K. (1990). Microglia and cell death in the developing mouse cerebellum. Brain
Res Dev Brain Res 55(2): 219-30.
Autieri, M. V. (1996). cDNA cloning of human allograft inflammatory factor-1: tissue
distribution, cytokine induction, and mRNA expression in injured rat carotid
arteries. Biochem Biophys Res Commun 228(1): 29-37.
Autieri, M. V., C. Carbone and A. Mu (2000). Expression of allograft inflammatory
factor-1 is a marker of activated human vascular smooth muscle cells and arterial
injury. Arterioscler Thromb Vasc Biol 20(7): 1737-44.
Autieri, M. V., S. E. Kelemen and K. W. Wendt (2003). AIF-1 is an actin-polymerizing
and Rac1-activating protein that promotes vascular smooth muscle cell migration.
Circ Res 92(10): 1107-14.
124
Autieri, M. V., M. B. Prystowsky and E. H. Ohlstein (1996). Isolation and
characterization of BART-1: A novel balloon angioplasty responsive transcript in
rat carotid arteries. DNA Cell Biol 15(4): 297-304.
Banati, R. B., S. E. Daniel and S. B. Blunt (1998). Glial pathology but absence of
apoptotic nigral neurons in long-standing Parkinson's disease. Mov Disord 13(2):
221-7.
Banati, R. B., J. Gehrmann, P. Schubert and G. W. Kreutzberg (1993). Cytotoxicity of
microglia. Glia 7(1): 111-8.
Barcia, C., A. Sanchez Bahillo, E. Fernandez-Villalba, V. Bautista, Y. P. M. Poza, A.
Fernandez-Barreiro, E. C. Hirsch and M. T. Herrero (2004). Evidence of active
microglia in substantia nigra pars compacta of parkinsonian monkeys 1 year after
MPTP exposure. Glia 46(4): 402-9.
Bechmann, I., E. Kwidzinski, A. D. Kovac, E. Simburger, T. Horvath, U. Gimsa, U.
Dirnagl, J. Priller and R. Nitsch (2001). Turnover of rat brain perivascular cells.
Exp Neurol 168(2): 242-9.
Bechmann, I., J. Priller, A. Kovac, M. Bontert, T. Wehner, F. F. Klett, J. Bohsung, M.
Stuschke, U. Dirnagl and R. Nitsch (2001). Immune surveillance of mouse brain
perivascular spaces by blood-borne macrophages. Eur J Neurosci 14(10): 1651-8.
Beschorner, R., S. Engel, M. Mittelbronn, D. Adjodah, K. Dietz, H. J. Schluesener and R.
Meyermann (2000). Differential regulation of the monocytic calcium-binding
peptides macrophage-inhibiting factor related protein-8 (MRP8/S100A8) and
allograft inflammatory factor-1 (AIF-1) following human traumatic brain injury.
Acta Neuropathol (Berl) 100(6): 627-34.
Bessis, A., C. Bechade, D. Bernard and A. Roumier (2007). Microglial control of
neuronal death and synaptic properties. Glia 55(3): 233-8.
Bezard, E., S. Dovero, C. Imbert, T. Boraud and C. E. Gross (2000). Spontaneous long-
term compensatory dopaminergic sprouting in MPTP-treated mice. Synapse
38(3): 363-8.
125
Blasi, E., R. Barluzzi, V. Bocchini, R. Mazzolla and F. Bistoni (1990). Immortalization
of murine microglial cells by a v-raf/v-myc carrying retrovirus. J Neuroimmunol
27(2-3): 229-37.
Bolin, L. M., I. Strycharska-Orczyk, R. Murray, J. W. Langston and D. Di Monte (2002).
Increased vulnerability of dopaminergic neurons in MPTP-lesioned interleukin-6
deficient mice. J Neurochem 83(1): 167-75.
Brooks, W. J., M. F. Jarvis and G. C. Wagner (1989). Astrocytes as a primary locus for
the conversion MPTP into MPP+. J Neural Transm 76(1): 1-12.
Browning, D. D., W. C. Diehl, M. H. Hsu, I. U. Schraufstatter and R. D. Ye (2000).
Autocrine regulation of interleukin-8 production in human monocytes. Am J
Physiol Lung Cell Mol Physiol 279(6): L1129-36.
Burns, R. S., C. C. Chiueh, S. P. Markey, M. H. Ebert, D. M. Jacobowitz and I. J. Kopin
(1983). A primate model of parkinsonism: selective destruction of dopaminergic
neurons in the pars compacta of the substantia nigra by N-methyl-4-phenyl-
1,2,3,6-tetrahydropyridine. Proc Natl Acad Sci U S A 80(14): 4546-50.
Cammer, W. and H. Zhang (1996). Carbonic anhydrase II in microglia in forebrains of
neonatal rats. J Neuroimmunol 67(2): 131-6.
Cardenas, H. and L. M. Bolin (2003). Compromised reactive microgliosis in MPTP-
lesioned IL-6 KO mice. Brain Res 985(1): 89-97.
Cardona, A. E., E. P. Pioro, M. E. Sasse, V. Kostenko, S. M. Cardona, I. M. Dijkstra, D.
Huang, G. Kidd, S. Dombrowski, R. Dutta, J. C. Lee, D. N. Cook, S. Jung, S. A.
Lira, D. R. Littman and R. M. Ransohoff (2006). Control of microglial
neurotoxicity by the fractalkine receptor. Nat Neurosci 9(7): 917-24.
Carson, M. J., C. R. Reilly, J. G. Sutcliffe and D. Lo (1998). Mature microglia resemble
immature antigen-presenting cells. Glia 22(1): 72-85.
Chakravarty, S. and M. Herkenham (2005). Toll-like receptor 4 on nonhematopoietic
cells sustains CNS inflammation during endotoxemia, independent of systemic
cytokines. J Neurosci 25(7): 1788-96.
126
Chan, P., L. E. DeLanney, I. Irwin, J. W. Langston and D. Di Monte (1991). Rapid ATP
loss caused by 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine in mouse brain. J
Neurochem 57(1): 348-51.
Chen, X., S. E. Kelemen and M. V. Autieri (2004). AIF-1 expression modulates
proliferation of human vascular smooth muscle cells by autocrine expression of
G-CSF. Arterioscler Thromb Vasc Biol 24(7): 1217-22.
Chen, Z. W., B. Ahren, C. G. Ostenson, A. Cintra, T. Bergman, C. Moller, K. Fuxe, V.
Mutt, H. Jornvall and S. Efendic (1997). Identification, isolation, and
characterization of daintain (allograft inflammatory factor 1), a macrophage
polypeptide with effects on insulin secretion and abundantly present in the
pancreas of prediabetic BB rats. Proc Natl Acad Sci U S A 94(25): 13879-84.
Chiba, K., A. Trevor and N. Castagnoli, Jr. (1984). Metabolism of the neurotoxic tertiary
amine, MPTP, by brain monoamine oxidase. Biochem Biophys Res Commun
120(2): 574-8.
Ciesielska, A., I. Joniec, I. Kurkowska-Jastrzebska, A. Przybylkowski, G. Gromadzka, A.
Czlonkowska and A. Czlonkowski (2007). Influence of age and gender on
cytokine expression in a murine model of Parkinson's disease.
Neuroimmunomodulation 14(5): 255-65.
Cuadros, M. A. and J. Navascues (1998). The origin and differentiation of microglial
cells during development. Prog Neurobiol 56(2): 173-89.
Czapski, G. A., M. Cakala, M. Chalimoniuk, B. Gajkowska and J. B. Strosznajder
(2007). Role of nitric oxide in the brain during lipopolysaccharide-evoked
systemic inflammation. J Neurosci Res 85(8): 1694-703.
Czlonkowska, A., M. Kohutnicka, I. Kurkowska-Jastrzebska and A. Czlonkowski (1996).
Microglial reaction in MPTP (1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine)
induced Parkinson's disease mice model. Neurodegeneration 5(2): 137-43.
Dalmau, I., J. M. Vela, B. Gonzalez, B. Finsen and B. Castellano (2003). Dynamics of
microglia in the developing rat brain. J Comp Neurol 458(2): 144-57.
127
Dauer, W. and S. Przedborski (2003). Parkinson's disease: mechanisms and models.
Neuron 39(6): 889-909.
Davalos, D., J. Grutzendler, G. Yang, J. V. Kim, Y. Zuo, S. Jung, D. R. Littman, M. L.
Dustin and W. B. Gan (2005). ATP mediates rapid microglial response to local
brain injury in vivo. Nat Neurosci 8(6): 752-8.
David, S., C. Bouchard, O. Tsatas and N. Giftochristos (1990). Macrophages can modify
the nonpermissive nature of the adult mammalian central nervous system. Neuron
5(4): 463-9.
de Jong, E. K., I. M. Dijkstra, M. Hensens, N. Brouwer, M. van Amerongen, R. S. Liem,
H. W. Boddeke and K. Biber (2005). Vesicle-mediated transport and release of
CCL21 in endangered neurons: a possible explanation for microglia activation
remote from a primary lesion. J Neurosci 25(33): 7548-57.
de Lau, L. M. and M. M. Breteler (2006). Epidemiology of Parkinson's disease. Lancet
Neurol 5(6): 525-35.
Deininger, M. H., R. Meyermann and H. J. Schluesener (2002). The allograft
inflammatory factor-1 family of proteins. FEBS Lett 514(2-3): 115-21.
Deininger, M. H., K. Seid, S. Engel, R. Meyermann and H. J. Schluesener (2000).
Allograft inflammatory factor-1 defines a distinct subset of infiltrating
macrophages/microglial cells in rat and human gliomas. Acta Neuropathol (Berl)
100(6): 673-80.
Deininger, M. H., T. Weinschenk, R. Meyermann and H. J. Schluesener (2003). The
allograft inflammatory factor-1 in Creutzfeldt-Jakob disease brains. Neuropathol
Appl Neurobiol 29(4): 389-99.
Del Galdo, F. and S. A. Jimenez (2007). T cells expressing allograft inflammatory factor
1 display increased chemotaxis and induce a profibrotic phenotype in normal
fibroblasts in vitro. Arthritis Rheum 56(10): 3478-88.
128
Del Galdo, F., G. G. Maul, S. A. Jimenez and C. M. Artlett (2006). Expression of
allograft inflammatory factor 1 in tissues from patients with systemic sclerosis
and in vitro differential expression of its isoforms in response to transforming
growth factor beta. Arthritis Rheum 54(8): 2616-25.
del Rio-Hortega, P. (1932). Microglia. Cytology and Cellular Pathology of the Nervous
System 2: 483-534.
Durocher, Y., S. Perret and A. Kamen (2002). High-level and high-throughput
recombinant protein production by transient transfection of suspension-growing
human 293-EBNA1 cells. Nucleic Acids Res 30(2): E9.
Eglitis, M. A. and E. Mezey (1997). Hematopoietic cells differentiate into both microglia
and macroglia in the brains of adult mice. Proc Natl Acad Sci U S A 94(8): 4080-
5.
Elmquist, J. K., C. D. Breder, J. E. Sherin, T. E. Scammell, W. F. Hickey, D. Dewitt and
C. B. Saper (1997). Intravenous lipopolysaccharide induces cyclooxygenase 2-
like immunoreactivity in rat brain perivascular microglia and meningeal
macrophages. J Comp Neurol 381(2): 119-29.
Fabriek, B. O., E. S. Van Haastert, I. Galea, M. M. Polfliet, E. D. Dopp, M. M. Van Den
Heuvel, T. K. Van Den Berg, C. J. De Groot, P. Van Der Valk and C. D. Dijkstra
(2005). CD163-positive perivascular macrophages in the human CNS express
molecules for antigen recognition and presentation. Glia 51(4): 297-305.
Farber, K. and H. Kettenmann (2006). Purinergic signaling and microglia. Pflugers Arch
452(5): 615-21.
Farinas, I., M. Cano-Jaimez, E. Bellmunt and M. Soriano (2002). Regulation of
neurogenesis by neurotrophins in developing spinal sensory ganglia. Brain Res
Bull 57(6): 809-16.
Ferger, B., A. Leng, A. Mura, B. Hengerer and J. Feldon (2004). Genetic ablation of
tumor necrosis factor-alpha (TNF-alpha) and pharmacological inhibition of TNF-
synthesis attenuates MPTP toxicity in mouse striatum. J Neurochem 89(4): 822-
33.
129
Ferrer, I., E. Bernet, E. Soriano, T. del Rio and M. Fonseca (1990). Naturally occurring
cell death in the cerebral cortex of the rat and removal of dead cells by transitory
phagocytes. Neuroscience 39(2): 451-8.
Ford, A. L., A. L. Goodsall, W. F. Hickey and J. D. Sedgwick (1995). Normal adult
ramified microglia separated from other central nervous system macrophages by
flow cytometric sorting. Phenotypic differences defined and direct ex vivo antigen
presentation to myelin basic protein-reactive CD4+ T cells compared. J Immunol
154(9): 4309-21.
Francis, J. W., J. Von Visger, G. J. Markelonis and T. H. Oh (1995). Neuroglial
responses to the dopaminergic neurotoxicant 1-methyl-4-phenyl-1,2,3,6-
tetrahydropyridine in mouse striatum. Neurotoxicol Teratol 17(1): 7-12.
Frank, M. G., J. L. Wieseler-Frank, L. R. Watkins and S. F. Maier (2006). Rapid
isolation of highly enriched and quiescent microglia from adult rat hippocampus:
immunophenotypic and functional characteristics. J Neurosci Methods 151(2):
121-30.
Freyaldenhoven, T. E., J. L. Cadet and S. F. Ali (1996). The dopamine-depleting effects
of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine in CD-1 mice are gender-
dependent. Brain Res 735(2): 232-8.
Fujiki, K., D. H. Shin, M. Nakao and T. Yano (1999). Molecular cloning of carp
(Cyprinus carpio) CC chemokine, CXC chemokine receptors, allograft
inflammatory factor-1, and natural killer cell enhancing factor by use of
suppression subtractive hybridization. Immunogenetics 49(10): 909-14.
Furuya, T., H. Hayakawa, M. Yamada, K. Yoshimi, S. Hisahara, M. Miura, Y. Mizuno
and H. Mochizuki (2004). Caspase-11 mediates inflammatory dopaminergic cell
death in the 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine mouse model of
Parkinson's disease. J Neurosci 24(8): 1865-72.
Gangemi, S., G. Basile, R. A. Merendino, A. Epifanio, G. Di Pasquale, B. Ferlazzo, V.
Nicita-Mauro and L. Morgante (2003). Effect of levodopa on interleukin-15 and
RANTES circulating levels in patients affected by Parkinson's disease. Mediators
Inflamm 12(4): 251-3.
130
Gao, H. M., B. Liu, W. Zhang and J. S. Hong (2003). Critical role of microglial NADPH
oxidase-derived free radicals in the in vitro MPTP model of Parkinson's disease.
Faseb J 17(13): 1954-6.
Gerlach, M. and P. Riederer (1996). Animal models of Parkinson's disease: an empirical
comparison with the phenomenology of the disease in man. J Neural Transm
103(8-9): 987-1041.
Giulian, D., J. Woodward, D. G. Young, J. F. Krebs and L. B. Lachman (1988).
Interleukin-1 injected into mammalian brain stimulates astrogliosis and
neovascularization. J Neurosci 8(7): 2485-90.
Graeber, M. B. and W. J. Streit (1990). Perivascular microglia defined. Trends Neurosci
13(9): 366.
Gu, L., S. C. Tseng and B. J. Rollins (1999). Monocyte chemoattractant protein-1. Chem
Immunol 72: 7-29.
Haines, B. P., R. B. Voyle and P. D. Rathjen (2000). Intracellular and extracellular
leukemia inhibitory factor proteins have different cellular activities that are
mediated by distinct protein motifs. Mol Biol Cell 11(4): 1369-83.
Hamilton, S. P. and L. H. Rome (1994). Stimulation of in vitro myelin synthesis by
microglia. Glia 11(4): 326-35.
Hansen, J. E., O. Lund, J. Engelbrecht, H. Bohr, J. O. Nielsen and J. E. Hansen (1995).
Prediction of O-glycosylation of mammalian proteins: specificity patterns of
UDP-GalNAc:polypeptide N-acetylgalactosaminyltransferase. Biochem J 308 (Pt
3): 801-13.
Hara, H., M. Ohta, K. Ohta, M. Nishimura, H. Obayashi and T. Adachi (1999). Isolation
of two novel alternative splicing variants of allograft inflammatory factor-1. Biol
Chem 380(11): 1333-6.
131
Hasegawa, E., K. Takeshige, T. Oishi, Y. Murai and S. Minakami (1990). 1-Methyl-4-
phenylpyridinium (MPP+) induces NADH-dependent superoxide formation and
enhances NADH-dependent lipid peroxidation in bovine heart submitochondrial
particles. Biochem Biophys Res Commun 170(3): 1049-55.
Hayashi, M., Y. Luo, J. Laning, R. M. Strieter and M. E. Dorf (1995). Production and
function of monocyte chemoattractant protein-1 and other beta-chemokines in
murine glial cells. J Neuroimmunol 60(1-2): 143-50.
Heikkila, R. E., F. S. Cabbat, L. Manzino and R. C. Duvoisin (1984). Effects of 1-
methyl-4-phenyl-1,2,5,6-tetrahydropyridine on neostriatal dopamine in mice.
Neuropharmacology 23(6): 711-3.
Hensel, G., D. N. Mannel, K. Pfizenmaier and M. Kronke (1987). Autocrine stimulation
of TNF-alpha mRNA expression in HL-60 cells. Lymphokine Res 6(2): 119-25.
Hickey, W. F. and H. Kimura (1988). Perivascular microglial cells of the CNS are bone
marrow-derived and present antigen in vivo. Science 239(4837): 290-2.
Hickey, W. F., K. Vass and H. Lassmann (1992). Bone marrow-derived elements in the
central nervous system: an immunohistochemical and ultrastructural survey of rat
chimeras. J Neuropathol Exp Neurol 51(3): 246-56.
Hirasawa, T., K. Ohsawa, Y. Imai, Y. Ondo, C. Akazawa, S. Uchino and S. Kohsaka
(2005). Visualization of microglia in living tissues using Iba1-EGFP transgenic
mice. J Neurosci Res 81(3): 357-62.
Hoek, R. M., S. R. Ruuls, C. A. Murphy, G. J. Wright, R. Goddard, S. M. Zurawski, B.
Blom, M. E. Homola, W. J. Streit, M. H. Brown, A. N. Barclay and J. D.
Sedgwick (2000). Down-regulation of the macrophage lineage through
interaction with OX2 (CD200). Science 290(5497): 1768-71.
Honda, M., H. Akiyama, Y. Yamada, H. Kondo, Y. Kawabe, M. Takeya, K. Takahashi,
H. Suzuki, T. Doi, A. Sakamoto, S. Ookawara, M. Mato, P. J. Gough, D. R.
Greaves, S. Gordon, T. Kodama and M. Matsushita (1998).
Immunohistochemical evidence for a macrophage scavenger receptor in Mato
cells and reactive microglia of ischemia and Alzheimer's disease. Biochem
Biophys Res Commun 245(3): 734-40.
132
Hunot, S., F. Boissiere, B. Faucheux, B. Brugg, A. Mouatt-Prigent, Y. Agid and E. C.
Hirsch (1996). Nitric oxide synthase and neuronal vulnerability in Parkinson's
disease. Neuroscience 72(2): 355-63.
Hunot, S., N. Dugas, B. Faucheux, A. Hartmann, M. Tardieu, P. Debre, Y. Agid, B.
Dugas and E. C. Hirsch (1999). FcepsilonRII/CD23 is expressed in Parkinson's
disease and induces, in vitro, production of nitric oxide and tumor necrosis factor-
alpha in glial cells. J Neurosci 19(9): 3440-7.
Imai, Y., I. Ibata, D. Ito, K. Ohsawa and S. Kohsaka (1996). A novel gene iba1 in the
major histocompatibility complex class III region encoding an EF hand protein
expressed in a monocytic lineage. Biochem Biophys Res Commun 224(3): 855-
62.
Imai, Y. and S. Kohsaka (2002). Intracellular signaling in M-CSF-induced microglia
activation: role of Iba1. Glia 40(2): 164-74.
Imamura, K., N. Hishikawa, M. Sawada, T. Nagatsu, M. Yoshida and Y. Hashizume
(2003). Distribution of major histocompatibility complex class II-positive
microglia and cytokine profile of Parkinson's disease brains. Acta Neuropathol
106(6): 518-26.
Innocenti, G. M., S. Clarke and H. Koppel (1983). Transitory macrophages in the white
matter of the developing visual cortex. II. Development and relations with axonal
pathways. Brain Res 313(1): 55-66.
Inoue, K. (2006). The function of microglia through purinergic receptors: neuropathic
pain and cytokine release. Pharmacol Ther 109(1-2): 210-26.
Ito, D., Y. Imai, K. Ohsawa, K. Nakajima, Y. Fukuuchi and S. Kohsaka (1998).
Microglia-specific localisation of a novel calcium binding protein, Iba1. Brain
Res Mol Brain Res 57(1): 1-9.
Ito, D., K. Tanaka, S. Suzuki, T. Dembo and Y. Fukuuchi (2001). Enhanced expression
of Iba1, ionized calcium-binding adapter molecule 1, after transient focal cerebral
ischemia in rat brain. Stroke 32(5): 1208-15.
133
Jackson-Lewis, V., M. Jakowec, R. E. Burke and S. Przedborski (1995). Time course
and morphology of dopaminergic neuronal death caused by the neurotoxin 1-
methyl-4-phenyl-1,2,3,6-tetrahydropyridine. Neurodegeneration 4(3): 257-69.
Jakowec, M. W., K. Nixon, E. Hogg, T. McNeill and G. M. Petzinger (2004). Tyrosine
hydroxylase and dopamine transporter expression following 1-methyl-4-phenyl-
1,2,3,6-tetrahydropyridine-induced neurodegeneration of the mouse nigrostriatal
pathway. J Neurosci Res 76(4): 539-50.
Jankovic, J. (2008). Parkinson's disease: clinical features and diagnosis. J Neurol
Neurosurg Psychiatry 79(4): 368-76.
Javitch, J. A., R. J. D'Amato, S. M. Strittmatter and S. H. Snyder (1985). Parkinsonism-
inducing neurotoxin, N-methyl-4-phenyl-1,2,3,6 -tetrahydropyridine: uptake of
the metabolite N-methyl-4-phenylpyridine by dopamine neurons explains
selective toxicity. Proc Natl Acad Sci U S A 82(7): 2173-7.
Kalkonde, Y. V., W. W. Morgan, J. Sigala, S. K. Maffi, C. Condello, W. Kuziel, S. S.
Ahuja and S. K. Ahuja (2007). Chemokines in the MPTP model of Parkinson's
disease: absence of CCL2 and its receptor CCR2 does not protect against striatal
neurodegeneration. Brain Res 1128(1): 1-11.
Kanazawa, H., K. Ohsawa, Y. Sasaki, S. Kohsaka and Y. Imai (2002).
Macrophage/microglia-specific protein Iba1 enhances membrane ruffling and Rac
activation via phospholipase C-gamma -dependent pathway. J Biol Chem
277(22): 20026-32.
Kelemen, S. E. and M. V. Autieri (2005). Expression of allograft inflammatory factor-1
in T lymphocytes: a role in T-lymphocyte activation and proliferative
arteriopathies. Am J Pathol 167(2): 619-26.
Kim, W. G., R. P. Mohney, B. Wilson, G. H. Jeohn, B. Liu and J. S. Hong (2000).
Regional difference in susceptibility to lipopolysaccharide-induced neurotoxicity
in the rat brain: role of microglia. J Neurosci 20(16): 6309-16.
134
Kim, Y. S., D. H. Choi, M. L. Block, S. Lorenzl, L. Yang, Y. J. Kim, S. Sugama, B. P.
Cho, O. Hwang, S. E. Browne, S. Y. Kim, J. S. Hong, M. F. Beal and T. H. Joh
(2007). A pivotal role of matrix metalloproteinase-3 activity in dopaminergic
neuronal degeneration via microglial activation. Faseb J 21(1): 179-87.
Kimura, M., Y. Kawahito, H. Obayashi, M. Ohta, H. Hara, T. Adachi, D. Tokunaga, T.
Hojo, M. Hamaguchi, A. Omoto, H. Ishino, M. Wada, M. Kohno, Y. Tsubouchi
and T. Yoshikawa (2007). A critical role for allograft inflammatory factor-1 in
the pathogenesis of rheumatoid arthritis. J Immunol 178(5): 3316-22.
Kohsaka, S., K. Ohsawa, Y. Sasaki, S. Honda, Y. Imai and K. Inoue (2004). Activation
of microglia by extracellular ATP through GI/O-coupled P2Y12 receptor. Journal
of Neurochemistry 88(Suppl. 1): 11.
Kohutnicka, M., E. Lewandowska, I. Kurkowska-Jastrzebska, A. Czlonkowski and A.
Czlonkowska (1998). Microglial and astrocytic involvement in a murine model
of Parkinson's disease induced by 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine
(MPTP). Immunopharmacology 39(3): 167-180.
Koizumi, S., Y. Shigemoto-Mogami, K. Nasu-Tada, Y. Shinozaki, K. Ohsawa, M. Tsuda,
B. V. Joshi, K. A. Jacobson, S. Kohsaka and K. Inoue (2007). UDP acting at
P2Y6 receptors is a mediator of microglial phagocytosis. Nature 446(7139):
1091-5.
Kolata, G. (1983). Monkey model of Parkinson's disease. Science 220(4598): 705.
Kong, B., T. Isozaki and S. Sasaki (1996). IL-6 antisense-mediated growth inhibition of
a choriocarcinoma cell line: an intracellular autocrine growth mechanism.
Gynecol Oncol 63(1): 78-84.
Koshiba, H., J. Kitawaki, M. Teramoto, Y. Kitaoka, H. Ishihara, H. Obayashi, M. Ohta,
H. Hara, T. Adachi and H. Honjo (2005). Expression of allograft inflammatory
factor-1 in human eutopic endometrium and endometriosis: possible association
with progression of endometriosis. J Clin Endocrinol Metab 90(1): 529-37.
Kreutzberg, G. W. (1996). Microglia: a sensor for pathological events in the CNS.
Trends Neurosci 19(8): 312-8.
135
Kruse, M., R. Steffen, R. Batel, I. M. Muller and W. E. Muller (1999). Differential
expression of allograft inflammatory factor 1 and of glutathione peroxidase during
auto- and allograft response in marine sponges. J Cell Sci 112 ( Pt 23): 4305-13.
Kurkowska-Jastrzebska, I., A. Wronska, M. Kohutnicka, A. Czlonkowski and A.
Czlonkowska (1999). MHC class II positive microglia and lymphocytic
infiltration are present in the substantia nigra and striatum in mouse model of
Parkinson's disease. Acta Neurobiol Exp (Wars) 59(1): 1-8.
Kurkowska-Jastrzebska, I., A. Wronska, M. Kohutnicka, A. Czlonkowski and A.
Czlonkowska (1999). The inflammatory reaction following 1-methyl-4-phenyl-
1,2,3, 6-tetrahydropyridine intoxication in mouse. Exp Neurol 156(1): 50-61.
Kurz, H. and B. Christ (1998). Embryonic CNS macrophages and microglia do not stem
from circulating, but from extravascular precursors. Glia 22(1): 98-102.
Kuschel, R., M. H. Deininger, R. Meyermann, A. Bornemann, Z. Yablonka-Reuveni and
H. J. Schluesener (2000). Allograft inflammatory factor-1 is expressed by
macrophages in injured skeletal muscle and abrogates proliferation and
differentiation of satellite cells. J Neuropathol Exp Neurol 59(4): 323-32.
Lang, R., M. Lustig, F. Francois, M. Sellinger and H. Plesken (1994). Apoptosis during
macrophage-dependent ocular tissue remodelling. Development 120(12): 3395-
403.
Langston, J. W., P. Ballard, J. W. Tetrud and I. Irwin (1983). Chronic Parkinsonism in
humans due to a product of meperidine-analog synthesis. Science 219(4587):
979-80.
Langston, J. W., L. S. Forno, J. Tetrud, A. G. Reeves, J. A. Kaplan and D. Karluk (1999).
Evidence of active nerve cell degeneration in the substantia nigra of humans years
after 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine exposure. Ann Neurol 46(4):
598-605.
Langston, J. W., I. Irwin, E. B. Langston and L. S. Forno (1984). 1-Methyl-4-
phenylpyridinium ion (MPP+): identification of a metabolite of MPTP, a toxin
selective to the substantia nigra. Neurosci Lett 48(1): 87-92.
136
Lawson, L. J. and V. H. Perry (1995). The unique characteristics of inflammatory
responses in mouse brain are acquired during postnatal development. Eur J
Neurosci 7(7): 1584-95.
Lawson, L. J., V. H. Perry, P. Dri and S. Gordon (1990). Heterogeneity in the
distribution and morphology of microglia in the normal adult mouse brain.
Neuroscience 39(1): 151-70.
Lewit-Bentley, A. and S. Rety (2000). EF-hand calcium-binding proteins. Curr Opin
Struct Biol 10(6): 637-43.
Liberatore, G. T., V. Jackson-Lewis, S. Vukosavic, A. S. Mandir, M. Vila, W. G.
McAuliffe, V. L. Dawson, T. M. Dawson and S. Przedborski (1999). Inducible
nitric oxide synthase stimulates dopaminergic neurodegeneration in the MPTP
model of Parkinson disease. Nat Med 5(12): 1403-9.
Ling, E. A. and W. C. Wong (1993). The origin and nature of ramified and amoeboid
microglia: a historical review and current concepts. Glia 7(1): 9-18.
Mallat, M. and B. Chamak (1994). Brain macrophages: neurotoxic or neurotrophic
effector cells? J Leukoc Biol 56(3): 416-22.
Markey, S. P., J. N. Johannessen, C. C. Chiueh, R. S. Burns and M. A. Herkenham
(1984). Intraneuronal generation of a pyridinium metabolite may cause drug-
induced parkinsonism. Nature 311(5985): 464-7.
Markus, A., T. D. Patel and W. D. Snider (2002). Neurotrophic factors and axonal
growth. Curr Opin Neurobiol 12(5): 523-31.
Marras, C. and A. Lang (2008). Invited article: changing concepts in Parkinson disease:
moving beyond the decade of the brain. Neurology 70(21): 1996-2003.
Marsden, C. D. (1983). Neuromelanin and Parkinson's disease. J Neural Transm Suppl
19: 121-41.
137
McGeer, P. L., S. Itagaki, B. E. Boyes and E. G. McGeer (1988). Reactive microglia are
positive for HLA-DR in the substantia nigra of Parkinson's and Alzheimer's
disease brains. Neurology 38(8): 1285-.
McGeer, P. L., C. Schwab, A. Parent and D. Doudet (2003). Presence of reactive
microglia in monkey substantia nigra years after 1-methyl-4-phenyl-1,2,3,6-
tetrahydropyridine administration. Ann Neurol 54(5): 599-604.
McMenamin, P. G. (1999). Distribution and phenotype of dendritic cells and resident
tissue macrophages in the dura mater, leptomeninges, and choroid plexus of the
rat brain as demonstrated in wholemount preparations. J Comp Neurol 405(4):
553-62.
Mildner, A., H. Schmidt, M. Nitsche, D. Merkler, U. K. Hanisch, M. Mack, M.
Heikenwalder, W. Bruck, J. Priller and M. Prinz (2007). Microglia in the adult
brain arise from Ly-6ChiCCR2+ monocytes only under defined host conditions.
Nat Neurosci 10(12): 1544-53.
Miller, F. D. and D. R. Kaplan (2001). Neurotrophin signalling pathways regulating
neuronal apoptosis. Cell Mol Life Sci 58(8): 1045-53.
Milligan, C. E., P. Levitt and T. J. Cunningham (1991). Brain macrophages and
microglia respond differently to lesions of the developing and adult visual system.
J Comp Neurol 314(1): 136-46.
Mirza, B., H. Hadberg, P. Thomsen and T. Moos (2000). The absence of reactive
astrocytosis is indicative of a unique inflammatory process in Parkinson's disease.
Neuroscience 95(2): 425-32.
Mogi, M., M. Harada, T. Kondo, H. Narabayashi, P. Riederer and T. Nagatsu (1995).
Transforming growth factor-beta 1 levels are elevated in the striatum and in
ventricular cerebrospinal fluid in Parkinson's disease. Neurosci Lett 193(2): 129-
32.
Mogi, M., M. Harada, T. Kondo, P. Riederer, H. Inagaki, M. Minami and T. Nagatsu
(1994). Interleukin-1 beta, interleukin-6, epidermal growth factor and
transforming growth factor-alpha are elevated in the brain from parkinsonian
patients. Neurosci Lett 180(2): 147-50.
138
Mogi, M., M. Harada, T. Kondo, P. Riederer, H. Inagaki, M. Minami and T. Nagatsu
(1994). Interleukin-1[beta], interleukin-6, epidermal growth factor and
transforming growth factor-[alpha] are elevated in the brain from parkinsonian
patients. Neuroscience Letters 180(2): 147-150.
Mogi, M., M. Harada, P. Riederer, H. Narabayashi, K. Fujita and T. Nagatsu (1994).
Tumor necrosis factor-alpha (TNF-alpha) increases both in the brain and in the
cerebrospinal fluid from parkinsonian patients. Neurosci Lett 165(1-2): 208-10.
Monier, A., P. Evrard, P. Gressens and C. Verney (2006). Distribution and
differentiation of microglia in the human encephalon during the first two
trimesters of gestation. J Comp Neurol 499(4): 565-82.
Mori, I., F. Goshima, T. Koshizuka, Y. Imai, S. Kohsaka, N. Koide, T. Sugiyama, T.
Yoshida, T. Yokochi, Y. Kimura and Y. Nishiyama (2003). Iba1-expressing
microglia respond to herpes simplex virus infection in the mouse trigeminal
ganglion. Brain Res Mol Brain Res 120(1): 52-6.
Mori, I., M. J. Hossain, K. Takeda, H. Okamura, Y. Imai, S. Kohsaka and Y. Kimura
(2001). Impaired microglial activation in the brain of IL-18-gene-disrupted mice
after neurovirulent influenza A virus infection. Virology 287(1): 163-70.
Muller, T., D. Blum-Degen, H. Przuntek and W. Kuhn (1998). Interleukin-6 levels in
cerebrospinal fluid inversely correlate to severity of Parkinson's disease. Acta
Neurol Scand 98(2): 142-4.
Muller, W. E., A. Krasko, A. Skorokhod, C. Bunz, V. A. Grebenjuk, R. Steffen, R. Batel
and H. C. Schroder (2002). Histocompatibility reaction in tissue and cells of the
marine sponge Suberites domuncula in vitro and in vivo: central role of the
allograft inflammatory factor 1. Immunogenetics 54(1): 48-58.
Nagata, K., N. Takei, K. Nakajima, H. Saito and S. Kohsaka (1993). Microglial
conditioned medium promotes survival and development of cultured
mesencephalic neurons from embryonic rat brain. J Neurosci Res 34(3): 357-63.
Nagatsu, T., M. Mogi, H. Ichinose and A. Togari (2000). Cytokines in Parkinson's
disease. J Neural Transm Suppl(58): 143-51.
139
Nagatsu, T. and M. Sawada (2007). Biochemistry of postmortem brains in Parkinson's
disease: historical overview and future prospects. J Neural Transm Suppl(72):
113-20.
Nickel, W. (2003). The mystery of nonclassical protein secretion. A current view on
cargo proteins and potential export routes. Eur J Biochem 270(10): 2109-19.
Nicklas, W. J., I. Vyas and R. E. Heikkila (1985). Inhibition of NADH-linked oxidation
in brain mitochondria by 1-methyl-4-phenyl-pyridine, a metabolite of the
neurotoxin, 1-methyl-4-phenyl-1,2,5,6-tetrahydropyridine. Life Sci 36(26): 2503-
8.
Nimmerjahn, A., F. Kirchhoff and F. Helmchen (2005). Resting microglial cells are
highly dynamic surveillants of brain parenchyma in vivo. Science 308(5726):
1314-8.
Ohsawa, K., Y. Imai, H. Kanazawa, Y. Sasaki and S. Kohsaka (2000). Involvement of
Iba1 in membrane ruffling and phagocytosis of macrophages/microglia. J Cell
Sci 113 ( Pt 17): 3073-84.
Ohsawa, K., Y. Imai, Y. Sasaki and S. Kohsaka (2004). Microglia/macrophage-specific
protein Iba1 binds to fimbrin and enhances its actin-bundling activity. J
Neurochem 88(4): 844-56.
Orsmark, C., T. Skoog, L. Jeskanen, J. Kere and U. Saarialho-Kere (2007). Expression
of allograft inflammatory factor-1 in inflammatory skin disorders. Acta Derm
Venereol 87(3): 223-7.
Parkinson, J. (2002). An essay on the shaking palsy. 1817. J Neuropsychiatry Clin
Neurosci 14(2): 223-36; discussion 222.
Pashenkov, M., S. Efendic, J. Zhu, L. P. Zou, C. G. Ostenson and M. Mustafa (2000).
Augmented expression of daintain/allograft inflammatory factor-1 is associated
with clinical disease: dynamics of daintain/allograft inflammatory factor-1
expression in spleen, peripheral nerves and sera during experimental autoimmune
neuritis. Scand J Immunol 52(2): 117-22.
140
Pattarini, R., R. J. Smeyne and J. I. Morgan (2007). Temporal mRNA profiles of
inflammatory mediators in the murine 1-methyl-4-phenyl-1,2,3,6-
tetrahydropyrimidine model of Parkinson's disease. Neuroscience 145(2): 654-
68.
Perry, V. H., D. A. Hume and S. Gordon (1985). Immunohistochemical localization of
macrophages and microglia in the adult and developing mouse brain.
Neuroscience 15(2): 313-26.
Pocock, J. M. and H. Kettenmann (2007). Neurotransmitter receptors on microglia.
Trends Neurosci 30(10): 527-35.
Qu, Y., L. Franchi, G. Nunez and G. R. Dubyak (2007). Nonclassical IL-1 beta secretion
stimulated by P2X7 receptors is dependent on inflammasome activation and
correlated with exosome release in murine macrophages. J Immunol 179(3):
1913-25.
Raivich, G., M. Bohatschek, C. U. Kloss, A. Werner, L. L. Jones and G. W. Kreutzberg
(1999). Neuroglial activation repertoire in the injured brain: graded response,
molecular mechanisms and cues to physiological function. Brain Res Brain Res
Rev 30(1): 77-105.
Ramsay, R. R., A. T. Kowal, M. K. Johnson, J. I. Salach and T. P. Singer (1987). The
inhibition site of MPP+, the neurotoxic bioactivation product of 1-methyl-4-
phenyl-1,2,3,6-tetrahydropyridine is near the Q-binding site of NADH
dehydrogenase. Arch Biochem Biophys 259(2): 645-9.
Ramsay, R. R., J. I. Salach and T. P. Singer (1986). Uptake of the neurotoxin 1-methyl-
4-phenylpyridine (MPP+) by mitochondria and its relation to the inhibition of the
mitochondrial oxidation of NAD+-linked substrates by MPP+. Biochem Biophys
Res Commun 134(2): 743-8.
Ramsay, R. R. and T. P. Singer (1986). Energy-dependent uptake of N-methyl-4-
phenylpyridinium, the neurotoxic metabolite of 1-methyl-4-phenyl-1,2,3,6-
tetrahydropyridine, by mitochondria. J Biol Chem 261(17): 7585-7.
Reale, M., C. Iarlori, A. Thomas, D. Gambi, B. Perfetti, M. Di Nicola and M. Onofrj
(2008). Peripheral cytokines profile in Parkinson's disease. Brain Behav Immun.
141
Saraswathy, S., G. Wu and N. A. Rao (2006). Retinal microglial activation and
chemotaxis by docosahexaenoic acid hydroperoxide. Invest Ophthalmol Vis Sci
47(8): 3656-63.
Sasaki, Y., K. Ohsawa, H. Kanazawa, S. Kohsaka and Y. Imai (2001). Iba1 is an actin-
cross-linking protein in macrophages/microglia. Biochem Biophys Res Commun
286(2): 292-7.
Schluesener, H. J., K. Seid, J. Kretzschmar and R. Meyermann (1998). Allograft-
inflammatory factor-1 in rat experimental autoimmune encephalomyelitis,
neuritis, and uveitis: expression by activated macrophages and microglial cells.
Glia 24(2): 244-51.
Schluesener, H. J., K. Seid and R. Meyermann (1999). Effects of autoantigen and
dexamethasone treatment on expression of endothelial-monocyte activating
polypeptide II and allograft-inflammatory factor-1 by activated macrophages and
microglial cells in lesions of experimental autoimmune encephalomyelitis,
neuritis and uveitis. Acta Neuropathol (Berl) 97(2): 119-26.
Schwab, J. M., E. Frei, I. Klusman, L. Schnell, M. E. Schwab and H. J. Schluesener
(2001). AIF-1 expression defines a proliferating and alert microglial/macrophage
phenotype following spinal cord injury in rats. J Neuroimmunol 119(2): 214-22.
Schwartz, M., O. Butovsky, W. Bruck and U. K. Hanisch (2006). Microglial phenotype:
is the commitment reversible? Trends Neurosci 29(2): 68-74.
Sedelis, M., R. K. Schwarting and J. P. Huston (2001). Behavioral phenotyping of the
MPTP mouse model of Parkinson's disease. Behav Brain Res 125(1-2): 109-25.
Sedgwick, J. D., S. Schwender, H. Imrich, R. Dorries, G. W. Butcher and V. ter Meulen
(1991). Isolation and direct characterization of resident microglial cells from the
normal and inflamed central nervous system. Proc Natl Acad Sci U S A 88(16):
7438-42.
Sershen, H., M. F. Mason, A. Hashim and A. Lajtha (1985). Effect of N-methyl-4-
phenyl-1,2,3,6-tetrahydropyridine (MPTP) on age-related changes in dopamine
turnover and transporter function in the mouse striatum. Eur J Pharmacol 113(1):
135-6.
142
Shimoji, M., L. Zhang, A. S. Mandir, V. L. Dawson and T. M. Dawson (2005). Absence
of inclusion body formation in the MPTP mouse model of Parkinson's disease.
Brain Res Mol Brain Res 134(1): 103-8.
Sibinga, N. E., M. W. Feinberg, H. Yang, F. Werner and M. K. Jain (2002).
Macrophage-restricted and interferon gamma-inducible expression of the allograft
inflammatory factor-1 gene requires Pu.1. J Biol Chem 277(18): 16202-10.
Springer, T., G. Galfre, D. S. Secher and C. Milstein (1978). Monoclonal xenogeneic
antibodies to murine cell surface antigens: identification of novel leukocyte
differentiation antigens. Eur J Immunol 8(8): 539-51.
Sriram, K., J. M. Matheson, S. A. Benkovic, D. B. Miller, M. I. Luster and J. P.
O'Callaghan (2002). Mice deficient in TNF receptors are protected against
dopaminergic neurotoxicity: implications for Parkinson's disease. Faseb J
16(11): 1474-6.
Sriram, K., J. M. Matheson, S. A. Benkovic, D. B. Miller, M. I. Luster and J. P.
O'Callaghan (2006). Deficiency of TNF receptors suppresses microglial
activation and alters the susceptibility of brain regions to MPTP-induced
neurotoxicity: role of TNF-alpha. Faseb J 20(6): 670-82.
Sriram, K., D. B. Miller and J. P. O'Callaghan (2006). Minocycline attenuates microglial
activation but fails to mitigate striatal dopaminergic neurotoxicity: role of tumor
necrosis factor-alpha. J Neurochem 96(3): 706-18.
Stacy, M. and J. Jankovic (1992). Differential diagnosis of Parkinson's disease and the
parkinsonism plus syndromes. Neurol Clin 10(2): 341-59.
Stolz, B., S. D. Erulkar and D. P. Kuffler (1991). Macrophages direct process elongation
from adult frog motorneurons in culture. Proc Biol Sci 244(1311): 227-31.
Streit, W. J. (2002). Microglia as neuroprotective, immunocompetent cells of the CNS.
Glia 40(2): 133-9.
Streit, W. J., M. B. Graeber and G. W. Kreutzberg (1988). Functional plasticity of
microglia: a review. Glia 1(5): 301-7.
143
Streit, W. J. and G. W. Kreutzberg (1988). Response of endogenous glial cells to motor
neuron degeneration induced by toxic ricin. J Comp Neurol 268(2): 248-63.
Sundstrom, E. and G. Jonsson (1985). Pharmacological interference with the neurotoxic
action of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) on central
catecholamine neurons in the mouse. Eur J Pharmacol 110(3): 293-9.
Sundstrom, E., I. Stromberg, T. Tsutsumi, L. Olson and G. Jonsson (1987). Studies on
the effect of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) on central
catecholamine neurons in C57BL/6 mice. Comparison with three other strains of
mice. Brain Res 405(1): 26-38.
Tanaka, S. and T. Koike (2002). Selective inflammatory stimulations enhance release of
microglial response factor (MRF)-1 from cultured microglia. Glia 40(3): 360-71.
Tanaka, S., K. Suzuki, M. Watanabe, A. Matsuda, S. Tone and T. Koike (1998).
Upregulation of a new microglial gene, mrf-1, in response to programmed
neuronal cell death and degeneration. J Neurosci 18(16): 6358-69.
Teismann, P., K. Tieu, O. Cohen, D. K. Choi, D. C. Wu, D. Marks, M. Vila, V. Jackson-
Lewis and S. Przedborski (2003). Pathogenic role of glial cells in Parkinson's
disease. Mov Disord 18(2): 121-9.
Thery, C., B. Chamak and M. Mallat (1991). Cytotoxic Effect of Brain Macrophages on
Developing. Eur J Neurosci 3(11): 1155-1164.
Tian, Y., S. E. Kelemen and M. V. Autieri (2006). Inhibition of AIF-1 expression by
constitutive siRNA expression reduces macrophage migration, proliferation, and
signal transduction initiated by atherogenic stimuli. Am J Physiol Cell Physiol
290(4): C1083-91.
Tipton, K. F. and T. P. Singer (1993). Advances in our understanding of the mechanisms
of the neurotoxicity of MPTP and related compounds. J Neurochem 61(4): 1191-
206.
144
Tsubata, Y., M. Sakatsume, A. Ogawa, B. Alchi, Y. Kaneko, T. Kuroda, H. Kawachi, I.
Narita, T. Yamamoto and F. Gejyo (2006). Expression of allograft inflammatory
factor-1 in kidneys: A novel molecular component of podocyte. Kidney Int
70(11): 1948-54.
Uhl, G. R., J. C. Hedreen and D. L. Price (1985). Parkinson's disease: loss of neurons
from the ventral tegmental area contralateral to therapeutic surgical lesions.
Neurology 35(8): 1215-8.
Utans, U., R. J. Arceci, Y. Yamashita and M. E. Russell (1995). Cloning and
characterization of allograft inflammatory factor-1: a novel macrophage factor
identified in rat cardiac allografts with chronic rejection. J Clin Invest 95(6):
2954-62.
Utans, U., W. C. Quist, B. M. McManus, J. E. Wilson, R. J. Arceci, A. F. Wallace and M.
E. Russell (1996). Allograft inflammatory factory-1. A cytokine-responsive
macrophage molecule expressed in transplanted human hearts. Transplantation
61(9): 1387-92.
Vallat, A. V., U. De Girolami, J. He, A. Mhashilkar, W. Marasco, B. Shi, F. Gray, J. Bell,
C. Keohane, T. W. Smith and D. Gabuzda (1998). Localization of HIV-1 co-
receptors CCR5 and CXCR4 in the brain of children with AIDS. Am J Pathol
152(1): 167-78.
van Rossum, D. and U. K. Hanisch (2004). Microglia. Metab Brain Dis 19(3-4): 393-
411.
Vila, M., V. Jackson-Lewis, C. Guegan, D. C. Wu, P. Teismann, D. K. Choi, K. Tieu and
S. Przedborski (2001). The role of glial cells in Parkinson's disease. Curr Opin
Neurol 14(4): 483-9.
Wang, X., G. Arcuino, T. Takano, J. Lin, W. G. Peng, P. Wan, P. Li, Q. Xu, Q. S. Liu, S.
A. Goldman and M. Nedergaard (2004). P2X7 receptor inhibition improves
recovery after spinal cord injury. Nat Med 10(8): 821-7.
145
Watano, K., K. Iwabuchi, S. Fujii, N. Ishimori, S. Mitsuhashi, M. Ato, A. Kitabatake and
K. Onoe (2001). Allograft inflammatory factor-1 augments production of
interleukin-6, -10 and -12 by a mouse macrophage line. Immunology 104(3): 307-
16.
Westlund, K. N., R. M. Denney, L. M. Kochersperger, R. M. Rose and C. W. Abell
(1985). Distinct monoamine oxidase A and B populations in primate brain.
Science 230(4722): 181-3.
Wright, G. J., M. J. Puklavec, A. C. Willis, R. M. Hoek, J. D. Sedgwick, M. H. Brown
and A. N. Barclay (2000). Lymphoid/neuronal cell surface OX2 glycoprotein
recognizes a novel receptor on macrophages implicated in the control of their
function. Immunity 13(2): 233-42.
Wu, D. C., P. Teismann, K. Tieu, M. Vila, V. Jackson-Lewis, H. Ischiropoulos and S.
Przedborski (2003). NADPH oxidase mediates oxidative stress in the 1-methyl-4-
phenyl-1,2,3,6-tetrahydropyridine model of Parkinson's disease. Proc Natl Acad
Sci U S A 100(10): 6145-50.
Yamada, M., Y. Imai, S. Kohsaka and S. Kamitori (2004). Crystallization and
preliminary X-ray crystallographic analysis of macrophage/microglia-specific
calcium-binding protein Iba1. Acta Crystallogr D Biol Crystallogr 60(Pt 3): 569-
71.
Yamada, M., K. Ohsawa, Y. Imai, S. Kohsaka and S. Kamitori (2006). X-ray structures
of the microglia/macrophage-specific protein Iba1 from human and mouse
demonstrate novel molecular conformation change induced by calcium binding. J
Mol Biol 364(3): 449-57.
Yang, Z. F., D. W. Ho, C. K. Lau, C. T. Lam, C. T. Lum, R. T. Poon and S. T. Fan
(2005). Allograft inflammatory factor-1 (AIF-1) is crucial for the survival and
pro-inflammatory activity of macrophages. Int Immunol 17(11): 1391-7.
Yasuda, Y., R. Shinagawa, M. Yamada, T. Mori, N. Tateishi and S. Fujita (2007). Long-
lasting reactive changes observed in microglia in the striatal and substantia nigral
of mice after 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine. Brain Res 1138:
196-202.

Abstract (if available)

Abstract Allograft inflammatory factor-1 (AIF-1) is an evolutionary conserved protein important to inflammatory responses throughout the body including that of microglia in the central nervous system (CNS). In addition to critical intracellular roles in the activation of microglia and macrophages, AIF-1 can be secreted by these cells in response to inflammatory signals as well as soluble signals released by dying neurons. In response to the neurotoxin 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine, we found increased levels of AIF-1 expression in cells clustered in the substantia nigra pars compacta (SNpc), the site of dopaminergic cell death. The number of these AIF-1 bright cells continued to increase even after neuronal cell death was complete. This increased expression of AIF-1 was restricted to resident microglia

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

The contribution of arachidonate 5-lipoxygenase to inflammatory response in coronary artery disease

PDF

Role of GRP78 and its novel cytosolic isoform GRP78va in regulating the unfolded protein response in cancer

PDF

A novel role for hypoxia-inducible factor-1alpha (HIF-1alpha) in the regulation of inflammatory chemokines and leukotriene expression in sickle cell disease

PDF

Fibroblast growth factors and notch signaling in a diethoxycarbonyl dihydrocollidine-induced hepatic progenitor cell liver injury model

PDF

A novel construct to study the pulsatility of insulin secretion in single cells, islets and whole pancreas

PDF

Role of endoplasmic reticulum chaperones GRP78 and GRP94 in regulating unfolded protein response and cell survival

PDF

Interleukin-11: a study of its effects in glioblastoma multiforme

PDF

Characteristics of hydrogen peroxide inducible clone-5 and its potential role as a nuclear receptor coactivator

PDF

Investigating the role of SASH1 gene located on chromosome 6 in ovarian cancer

PDF

Studies on the role of a novel protein, TMEM 56 in tumorigenic growth for MCF-7 cells

PDF

Construction of plasmids for constitutive and induced expression of secreted alkaline phosphatase in mammalian ovarian carcinoma cells

PDF

Placenta growth factor-miRNAs-lncRNAs axis in the regulation of ET-1 gene involved in pulmonary hypertension in sickle cell disease

PDF

Examining perillyl alcochol in glioblastoma treatment and relating its effect to endoplasmic reticulum stress response to further advance its usage in combination treatment with dimethyl-celecoxib

PDF

Alcohol mediated expression of cyto-protective enzyme - NQO-1 and its post translational regulation

PDF

Studies on the expression and function of the human TMEM56 protein

PDF

The transcriptional regulation of the pro-survival protein Grp78 by activating transcription factors and chromatin-modifying enzymes: its upregulation in response to HDAC-inhibitor treatment and ...

PDF

Roles of three domains of CCAR1 in transcription activation by nuclear receptors

PDF

Ethanol induced modulation of microglial P2X7 receptor expression and its role in neuroinflammation

PDF

A functional genomic approach based on shRNA-mediated gene silencing to delineate the role of NF-κB and cell death proteins in the survival and proliferation of KSHV associated primary effusion l...

PDF

Neuroendocrine regulation of the transcription factor SKN-1/Nrf2 in oxidative stress response

Asset Metadata

Creator Nowitzki, Kristina M. (author)

Core Title The response of Allograft inflammatory factor-1 to neurotoxic injury, and its role as a secreted protein

School Keck School of Medicine

Degree Doctor of Philosophy

Degree Program Biochemistry and Molecular Biology

Degree Conferral Date 2008-12

Publication Date 12/16/2008

Defense Date 06/27/2008

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag AIF-1,Allograft inflammatory factor-1,migroglia,MPTP,OAI-PMH Harvest

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Tokes, Zoltan A. (committee chair), Garner, Judy A. (committee member), Mosteller, Raymond (committee member), Weiner, Leslie P. (committee member)

Creator Email kristina.nowitzki@gmail.com,nowitzki@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m1931

Unique identifier UC1148750

Identifier etd-Nowitzki-2366 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-147701 (legacy record id),usctheses-m1931 (legacy record id)

Legacy Identifier etd-Nowitzki-2366.pdf

Dmrecord 147701

Document Type Dissertation

Rights Nowitzki, Kristina M.

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu