/

University of Southern California Dissertations and Theses

/

Non-apoptotic involvement of caspases in astrogliosis

(USC Thesis Other)

Non-apoptotic involvement of caspases in astrogliosis

PDF

Download

Open document

Flip pages

Contact Us

Copy asset link

Request this asset

Transcript (if available)

Content NON-APOPTOTIC INVOLVEMENT OF CASPASES IN ASTROGLIOSIS
by
Radha V. Aras
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
(NEUROSCIENCE)
December 2006
Copyright 2006 Radha V. Aras
ii
Dedication
For Amma, Abbu, Raghav, Raman
&
Aamir
iii
Acknowledgements
I would like to thank my advisor Dr Christian Pike for his guidance and support. His
mentorship has shaped my scientific curiosity and the skills and knowledge that I
gathered under his tutelage have equipped me with tools to answer those questions
that excite me.
I am also thankful to my committee members Drs Caleb Finch, Chieng-Ping Ko
Michel Baudry, and John Tower for taking the time from their busy schedules to
offer important suggestions and critiques of my work. Their opinions and ideas were
key to the completion of this thesis.
I would never have gone through the trials and tribulations of graduate school
without the help and support from my all my past and present collegues, Sarah
Stolzner, Allison Quaglino, Dr Mingzhong Yao, Dr Martin Ramsden, Emily Rosario,
Vi Nguyen, and Jenna Carroll. These are people I have spend most part of graduate
life with; who I grew to look upon as a surrogate family. I am also thankful to the
hundreds of students I have interacted with as a teaching assistant over the years.
With their eager minds and thirst for knowledge they re-invigorated my jaded palate
and kept my scientific temper alive.
iv
The technical help and use of equipment and lab supplies from the Finch lab
members provided me with the required infrastructure to carry out a large part of my
work, for which I am grateful. I would especially like to thank Dr Irina Rozovsky,
and Drs Min Wei, and Nilay Patel, past members. I am also grateful for Dr Akopian
from the Walsh lab for providing me with valuable technical knowledge, which was
necessary for completion of this dissertation. The Davies lab members also must be
recognized for kindly offering me the occasional but necessary use of their lab
equipment.
And finally, deepest thanks to Prasad and Athena and my friends at USC Raechelle,
Sandra, Nitin, Amol, Purnima, Doney and Mitali. Interacting with these talented
people from diverse academic backgrounds contributed to my development as a truly
educated scientist.
v
Table of Contents
Dedication ii
Acknowledgements iii
List of Tables vii
List of Figures viii
Abstract ix
Chapter 1 Introduction
What is astrogliosis? 2
Biochemical markers of astrocyte reactivity 4
Signal transduction mediating astrogliosis 16
Functional significance of reactive astrocytes 25
Factors underlying the contradictory effects of astrogliosis 30
Introduction to caspases 34
Non-apoptotic caspase functions 40
Proposed mechanisms underlying non-apoptotic caspase 47
functions
Thesis objectives 53
Chapter 2 Non-apoptotic caspase involvement in astrogliosis
Abstract 57
Introduction 59
Materials and Methods 62
Results 69
Discussion 85
Chapter 3 Functional significance of caspase activation in reactive astrocytes
with respect to neuronal survival
Abstract 92
Introduction 94
Materials and Methods 97
Results 101
Discussion 112
vi
Chapter 4 Conclusion and future directions 118
Possible reasons for caspase activation in reactive astrocytes
to be non-apoptotic 119
Sub-threshold caspase activation in reactive astrocytes is a
mechanism for eliminating highly inflammed astrocytes 123
Mediators of non-apoototic caspase functions in astrogliosis 125
Placing caspases in the signaling cascade underlying
astrogliosis: A hypothesis 130
Caspase involvement in astrogliosis: future directions of
investigation 132
Bibliography 134
vii
List of Tables
1. Molecules up regulated in reactive astrocytes: an update 12
2. Signal transduction in astrogliosis 23
3. Non-apoptotic caspase functions and molecular targets 43
4. Potential mediators of caspase function in astrogliosis 127
viii
List of Figures
1. Figure 1: In vitro model of astrogliosis 70
2. Figure 2: Significant caspase activation is observed in astrocytes
made reactive by dBcAMP and Aβ, in the absence of cell death 74
3. Figure 3: Caspase inhibition partially attenuates increase in GS
and FGF-2 expression 76
4. Figure 4: Caspase inhibition does not reverse stellate morphology 77
5. Figure 5: Specific inhibition of caspase11 and caspase-3
confers maximal attenuation of GS and FGF-2 79
6. Figure 6: Specific inhibition of caspases 1,6,8 and 9 has no
effect on GS and FGF-2 expression 80
7. Figure 7: Ex-vivo model of astrogliosis 82
8. Figure 8: Significant caspase activation observed in astrocytes
from kainate lesioned animals in absence of cell death 83
9. Figure 9: Caspase inhibition partially attenuates increase in GS
and FGF-2 expression in KA astrocytes 84
10. Figure 10: Neurite growth is stunted on astrocytes from kainate
lesioned animals 102
11. Figure 11: Caspase inhibition partially improves neurite extension
and number of neurites per neuron 105
12. Figure 12: Neuron survival is not hampered on astrocytes from kainate
lesioned animals 107
13. Figure 13: Neurons are more susceptible to 3-NP in presence of
astrocytes from kainate lesioned animals as compared to those from
sham-lesioned 109
14. Figure 14: Increased neuron susceptibility to 3-NP is not
caspase-dependent 111
ix
Abstract
Astrogliosis is a characteristic response of astrocytes to almost all forms of brain injury.
Despite being studied extensively, the molecular basis of astrogliosis remains largely
unknown. This dissertation investigates caspases as potential signaling molecules
involved in mediating astrocyte reactivity. Chapter 2 documents data from two different
experimental models of astrogliosis supporting the hypothesis of a non-apoptotic
involvement of caspases in astrogliosis. Our results show astrocytes made reactive by
treatment with dibutryl cAMP and Aβ
25-35
, demonstrate an increase in total caspase
activity with a corresponding increase in the expression of active pro-apoptotic caspase-
3 in the absence of cell death. In addition, caspase inhibition by zVAD, a broad-
spectrum caspase inhibitor resulted in a partial attenuation of the increased expressions
of two known markers of reactivity, glutamine synthetase and fibroblast growth factor-
2, thus suggesting a non-apoptotic role for caspases in mediating astrogliosis. We
further extended the study to an ex-vivo model of astrogliosis, comprising of adult
hippocampal astrocyte cultures generated from kainate-lesioned rats. Astrocytes from
the kainate-lesioned animals exhibited reactive features in culture, and a non-apoptotic
caspase activation similar to that observed in the in vitro model. Further, a preliminary
analysis of the caspase family using specific inhibitors suggests caspase-11 and 3 might
contribute to the caspase function in astrogliosis. Reactive astrocytes are known to up-
regulate expression of several different proteins with important functional consequences
x
on the injured brain. Chapter 3 investigates two major outcomes: neurite outgrowth and
neuronal survival following injury and how caspase activation in reactive astrocytes
might influence them. Our findings suggest inhibition of neurite growth on reactive
astrocytes might be partly caspase-dependent.
1
Chapter 1
Introduction
Astrocytes are the most abundant of cell types in the brain responsible for executing
important functions. Damage to the brain elicits a robust astrocytic reaction termed
astrogliosis, which is instrumental in restoring homeostasis in the in brain following
injury. The functional consequences of astrogliosis on neuronal recovery could be
beneficial or detrimental depending on the type and duration of injury and even the
age of the organism. Owing to the fact that this cellular response forms such an
integral part of any form of CNS damage, the characteristic changes attributed to
astrocyte reactivity have been extensively studied over the past years. However it is
still unclear what molecular cues govern this reactive response. Emerging studies
have investigated some common transcription factors and signaling molecules as
potential mediators of astrocyte reactivity. The common rationale behind studying
these proteins has been their increased expression in the brain following injury and
pleiotropic nature. This dissertation focuses on caspases a family of signaling
molecules, also commonly activated in the damaged brain and as recent data suggest
capable of executing multiple cellular events. Caspases, predominantly associated
with the execution of apoptosis and processing of pro-inflammatory cytokines are
now also known to regulate non-apoptotic cellular events like proliferation and
differentiation. Caspase activation has been observed in astrocytes following damage
2
to the brain. Interestingly, in some of the studies the caspase activation was
demonstrated without an overlap of cell death markers. This dissertation explores the
possibility of a non-apoptotic function for caspases, in astrocytes in the damaged
brain. Since the predominant astrocytic response to brain injury is to become
reactive, this non-apoptotic caspase function could be a role in regulating
astrogliosis. Chapter one of this dissertation, is a review of current literature that
helps provide the rationale for this hypothesis. The subsequent two chapters discuss
the studies, which test the hypothesis. The final chapter four analyzes our findings
with respect to current literature on signal transduction governing astrogliosis and the
possible mechanisms of non-apoptotic caspase activation in reactive astrocytes.
What is astrogliosis?
Any source of damage to the brain, be it disease, mechanical trauma, or even normal
physiological aging elicits a characteristic response from neuroglia which is termed
as “reactive gliosis”. This reaction, predominantly demonstrated by astrocytes and
microglia and to a lesser extent by oligodendroglia consists of various morphological
and biochemical changes. Although microglia undoubtedly have an important role in
the events following CNS injury, this dissertation focuses on reactive astrocytes and
hence our introduction will emphasize more on the literature regarding astrocyte
reactivity or “astrogliosis”. Astrocyte reactivity is associated with the following
characteristic changes. Stellate morphology: This morphological change although
characteristic of the reactive state, is also observed during astrocyte development. An
3
astroblast, which is a proliferating astrocyte precursor, is flattened and amoeboid, but
upon its differentiation into a fibrous astrocyte aquires long well defined processes
and a distinct cell body. Similarly, quiescent or non-reactive astrocyte displays a
flattened, polygonal morphology, which is transformed into the highly branched
stellate shape characteristic of the reactive state when there is any form of damage to
the brain. Adult astrocytes when observed in vivo are highly process bearing and
might appear stellate even under non-reactive conditions. However, the stellate
morphology associated with astrocyte reactivity is much more pronounced and
accompanied with hypertrophy, and thus an easily distinguishable phenotype.
Hypertrophy: Reactive astrocytes enlarge in size or display hypertrophy due to
increase in intermediate filament content. In certain forms of injury this feature of
astrocyte reactivity is very prominent and is even thought to be the major indicator of
the progression of certain kinds of injury (Ludowyk, Hughes et al. 1993; Rogers,
Peters et al. 2003). It precedes most other features of reactive astrocytes and unlike
stellation is typically seen only in the reactive state. Astrocyte proliferation: Some
forms of injury like mechanical trauma or excitotoxic lesions which result in a
compromised blood brain barrier and / or extensive neuron loss, are characterized by
the formation of a glia scar. Although other glia and fibroblasts could be present, the
glial scar tissue is mainly composed of proliferating reactive astrocytes. This
characteristic feature of astrogliosis is often referred to as “reactive astrocytosis”.
Whether reactive astrocytes proliferate or simply migrate from distal parts to the
lesion area to form the glial scar is still a much-debated topic. The present consensus
4
seems to be that both astrocyte proliferation and migration might be occurring, and
may involve different sub-populations of reactive astrocytes (Lee, Pappas et al. 2003;
Wang, Bekar et al. 2004; Tatsumi, Haga et al. 2005). Although the glial scar serves
to contain the damage done to the brain, it also constitutes a major impediment to
axonal regeneration following injury (McKeon, Schreiber et al. 1991; McKeon,
Hoke et al. 1995). Biochemical changes: The most crucial change that occurs during
astrogliosis is the altered expression of a wide variety of proteins. Although
expression of most proteins is up regulated, a recent study examining the gene
profile of reactive astrocytes from the striatum of OHDA-lesioned rats revealed a
down regulation of various different genes also occurred (Nakagawa, Yabe et al.
2005). The next section will discuss these biochemical changes in more detail and
emphasize some of those commonly used to identify reactive astrocytes.
Biochemical markers of astrocyte reactivity
One of the characteristic features of reactive astrocytes is the increased expression of
a wide variety of proteins. These biochemical changes have important functional
impact on neurons as well as glia and hence have been extensively investigated.
Several of them contribute to both beneficial as well as detrimental astrocytic effects
on neuron recovery. This section will discuss in detail, selected markers of astrocyte
reactivity including those relevant to this dissertation, followed by a review of the
current list of various categories of proteins up regulated in reactive astrocytes.
5
Classical markers of reactive astrocytes
Glial Fibrillary Acidic Protein (GFAP): This 50Kd protein is a major constituent of
the numerous 10nm intermediate filaments that form the cytoskeleton of astrocytes.
Reactive astrocytes display hypertrophy, thus increasing their intermediate filament
and hence GFAP content (Eng and Ghirnikar 1994; Eng, Ghirnikar et al. 2000).
GFAP knockout mice have provided important insights into its importance in
astrocyte reactivity as well as the consequences on neuron function. Mice deficient in
GFAP possess astrocytes with impaired intermediate filaments, resulting in improper
formation of the glial scar. Possibly due to the reduced intermediate content, these
astrocytes can display stellate morphology only when co-cultured with neurons
(Pekny, Eliasson et al. 1998). Astrocytes from these animals, also display increased
expression of extracellular matrix proteins, which promotes neurite growth as well as
neuron survival (Pekny, Eliasson et al. 1998; Menet, Gimâenez et al. 2000; Larsson,
Wilhelmsson et al. 2004). The neuroprotective effect of GFAP deficiency has been
attributed to increased GDNF levels (Hanbury, Ling et al. 2003), but the mechanisms
underlying other effects of GFAP are not understood completely (Levison,
Ducceschi et al. 1996). One postulated reason contributing to neurite growth
inhibition is GFAP-mediated suppression of growth promoting molecules expressed
by astrocytes. Eliminating GFAP expression is associated with a up regulation of
laminin, a neurite growth promoting ECM molecule in reactive astrocytes although
the exact mechanism for this concurrent increase is not known (Lefran∂cois, Fages et
al. 1997; Costa, Planchenault et al. 2002).
6
Glutamine synthetase (GS): This is an enzyme that catalyzes the synthesis of
glutamine from glutamate, ATP and ammonia. Although GS is widely distributed in
mammalian tissues, within the CNS it is found solely in astrocytes. Glutamine
produced by astrocytes is utilized by neurons to synthesize glutamate. On the other
hand, any excess glutamate secreted by neurons, which is toxic is reconverted to
non-toxic glutamine by astroglial GS. It is one of the first proteins expressed in fully
differentiated astrocytes and hence was one of the first markers used to identify
reactive astrocytes. There is some evidence to believe certain kinds of damage to the
brain might not raise the expression level of GS while others might cause a sustained
increase for an extended period. During aging even though astrocytes display several
features of reactivity, GS expression is not increased with age (Wu, Zhang et al.
2005) and might even be down regulated. The principal reason being, increased
accumulation of reactive oxygen species (ROS) with age, and sensitivity of GS to
ROS-mediated oxidation (Castegna, Aksenov et al. 2002). GS is also one of the
astrocytic proteins prone to tyrosine nitration-mediated inactivation following
chronic exposure to pro-inflammatory cytokines (Gèorg, Bidmon et al. 2006), which
is probably the reason GS expression is reduced during experimental allergic
encephalomyelitis (EAE), an experimental model of multiple sclerosis (Hardin-
Pouzet, Krakowski et al. 1997). Localized injury such as that caused by a mechanical
lesion, as well is not associated with increased GS expression (Condorelli,
Dell'Albani et al. 1990). On the other hand sustained upregulation of GS expression
is observed following β-amyloid toxicity (Pike, Ramezan-Arab et al. 1996)
7
excitotoxic lesions (Ong, Leong et al. 1996), and transient forebrain ischemia
(Tanaka, Araki et al. 1992). Thus overall GS can be considered an excellent gauge of
astrocyte function.
Vimentin: Astrocytes form intermediate filaments from three different proteins with
GFAP being the most important, and vimentin and nestin being the other two.
Astrocytes lose vimentin expression upon maturity but re-express it when they
transform into a reactive state, thus making vimentin a very selective marker for
identifying reactive astrocytes. A recent study reported that transfection of antisense
vimentin cDNA in astrocyte cultures curbed their proliferation, thus suggesting a key
role for vimentin in astrocyte proliferation contributing to glial scar formation (Lin,
Cai et al. 2004). Vimentin immunoreactivity is found more prominently in reactive
astrocytes on the periphery of the lesion site. Hence it may be more important as a
marker in injuries resulting in anisomorphic astrogliosis, used to differentiate
between proximal and distal reactive astrocytes.
Ciliary neurotrophic factor (CNTF): CNTF is a pleiotropic cytokine with very low
level in the adult brain, but following injury is up regulated several fold in astrocytes.
Overexpression of CNTF in transgenic mice as well as its injection into the uninjured
brain is known to mimic the glial response to CNS lesion, implicating CNTF as an
inducer of reactive gliosis (Winter, Saotome et al. 1995), (Levison, Ducceschi et al.
1996). Furthermore, exogenous treatment of spinal cord injury with CNTF
8
ameliorated neuron loss and hastened recovery, despite sustaining the astrocyte
reactivity (Ye, Cao et al. 2004), making CNTF one of the key astrocytic factors
responsible for beneficial effects of astrogliosis on neuronal recovery.
S100β: S100-β is a calcium binding protein found predominantly in astrocytes and
up regulated in response to a wide variety of insults. In some studies S100 β
expression has been demonstrated predominantly in reactive astrocytes forming the
glial scar (Bendotti, Guglielmetti et al. 2000) and transgenic mice with an over-
expressed S100-β gene display pronounced astrocytosis (Reeves, Yao et al. 1994);
suggesting it might contribute to their proliferative properties. Functionally S100-β
can have both beneficial and harmful effects on neuronal recovery. It can exert
potent neuritotrophic effects either independently or through interactions with other
trophic factors like FGF-2 (Gomide, Chadi et al. 1999). On the other hand it also
functions as a pro- inflammatory (Matsui, Mori et al. 2002) cytokine in enhancing
the expression of iNOS resulting in nitric oxide mediated death of co-cultured
neurons (Matsui, Mori et al. 2002). Co-localization of S100 β immunoreactivity with
iNOS in the periphery of lesion site suggests similar situation may exist even in vivo
(Yasuda, Tateishi et al. 2004). Specific inhibition of S100 β production by reactive
astrocytes is known to reduce infarct volume following ischemia (Asano, Mori et al.
2005) further stressing its importance in regulating neuronal death following injury.
9
Fibroblast growth factor-2 (FGF-2): Fibroblast growth factor (FGF) is synthesized
and stored by astroglial cells and regulates their proliferation and differentiation in
vitro. This growth factor is known to induce features of reactivity in astrocyte
cultures (Bolego, Ceruti et al. 1997) (Eclancher, Kehrli et al. 1996) when applied
exogenously and yet is also secreted by reactive astrocytes (Eddleston and Mucke,
1993). Increased expression of FGF-2 as well as its receptor FGFR has been
observed upto 2 weeks following excitoxic injury in rats (Ballabriga, Pozas et al.
1997) suggesting a long-term influence of FGF on astrocyte reactivity and possibly
its consequences on neuron function following injury. It has a potent neurotrophic
and neuritrotrophic properties and yet under certain situations can also contribute to
the detrimental effects associated with astrogliosis. The neurite growth promoting
ability of astrocyte conditioned medium, has been attributed solely to the presence of
FGF-2 (Le, Esquenazi et al. 2002). On the other hand, FGF-2 is known to up regulate
the expression of neurite growth inhibitory protein tenascin in astrocytes (Meiners,
Marone et al. 1993) as well as inhibit signaling molecules used by neurite growth
promoting cell adhesion molecules (Williams, Mittal et al. 1995). The beneficial
effects of FGF-2 could also be antagonized by pro-inflammatory cytokines like
interferon-gamma (DiProspero, Meiners et al. 1997; Powell, Meiners et al. 1997)
secreted by reactive astrocytes. Down-regulation of glutamine synthetase expression,
due to endogenous release of FGF-2 following ischemia or treatment with FGF-2
suggests that FGF-2 may exacerbate glutamate mediated neurotoxicity (Kruchkova,
10
Ben-Dror et al. 2001), in addition to its reported neuroprotective effects. FGF-2 is
thus among the important mediators of astrocytic response to neuronal injury.
Update on the list of markers for astrogliosis
Previous reviews by Eddleston and Mucke in 1993 (Eddleston and Mucke 1993) and
later in 1997 by Ridet and Gage (Ridet, Malhotra et al. 1997) have provided an
exhaustive list of proteins upregulated in reactive astrocytes. Since then several more
proteins with varied functional properties have been added to the list further
contributing to the complexity of astrogliosis. Previous studies on astrocyte reactivity
seeked to investigate factors that might execute the functions attributed to reactive
astrocytes following injury. In the recent years however the relevance of astrogliosis
in the etiology of disease has grown, making reactive astrocytes potential therapeutic
targets. Hence the attention has rapidly shifted to a less understood aspect of
astrogliosis, which is the signal transduction mechanism that governs astrocyte
reactivity. As a result the latest additions to the list of known markers for identifying
reactive astrocytes include several signaling molecules and transcription factors.
Reactive astrocytes are also known to re-express certain proteins unique to immature
developing astrocytes. This has given rise to the idea that astrogliosis might be a
process of ‘de-differentiation’ possibly re-iterating some of the developmental
changes. Among such proteins are previously known intermediate filament proteins
vimentin and IFAP (Abd-el-Basset EM, Kalnins VI et al. 1989), and recent additions
like bystin, and transcription factors: n-myc, Id helix-loop helix proteins, and c-myb
11
normally associated with cellular differentiation. Table 1 is an update on these
biochemical markers of reactive astrocytes. It also includes important examples in
each category of up-regulated summarized from previous reviews. The signal
transduction molecules implicated in astrogliosis are listed in table 2 and their
relevance with respect to astrocyte reactivity as well as neuronal recovery is
discussed in later section.
12
Table 1- Molecules up regulated in reactive astrocytes: an update
Key to In vivo/In vitro injury models: 1=kainate, 2=transient focal ischemia,
3=entorhinal cortex lesion, 4=axotomy, 5= spinal cord stab injury, 6= thermal
ablation injury, 7= MPTP, 8=cortical aspiration lesion, 9=morphine, 10=quinolinic
acid, 11=EAE, 12=cryo-induced injury, 13=trauma, 14=hydrodopamine, C=cytokine
treatment
Abbreviations: GDNF: Glial derived nerve growth factor; IL12/23: interleukin12/23;
AD: Alzheimer’s disease, PD: Parkinson’s disease, MS: multiple sclerosis, BACE:
beta secretase enzyme, Cys-LTs: Cysteinyl leukotrienes
Category Up regulated
molecule
In vivo In vitro Human
brain
tissue
Reference
Adhesion /
ECM
molecules
Cd44, CS-PG
E-NCAM
E-Selectin
GHAP, HNK-1
HSPG
ICAM-1, IG9
Laminin
Tenascin
Thrombospondi
n
VCAM-1
VLA-1, VLA-2
VLA-6
PSA-NCAM
N-cadherin
H-CAM, G-
CAM, integrin
Slit / glypican
Neurotractin/ki
lon
Syndecans
1,2,3,4
13
1
13,1
4
1
3
serum
TNF-α
PDGF
TNF-α
IL-1β
INF-γ
Review
(Eddleston and
Mucke 1993)
Review
(Ridet, Malhotra
et al. 1997;
Hagino, Iseki et
al. 2003)
(Hagino, Iseki et
al. 2003)
(Schèafer,
Brèauer et al.
2005)
(Iseki, Hagino et
al. 2002)
13
Table 1 continued
Category Up regulated
molecule
In vivo In vitro Human
brain
tissue
Reference
Antigen
presentation
MHC class I
MHC class II
B7-1, B7-2
(co-stimulatory
molecules)
Thymidilate
kinase
C
INF-γ +
IL-1β
Review
(Eddleston and
Mucke 1993)
(Zeinstra,
Wilczak et al.
2003)
(Falsig,
Pèorzgen et al.
2006)
Apoptosis
regulators
bis (bcl-2
binding
protein)
Bcl-w
1,2
2
Lee et.al,
2000, 2002
(Yan, Chen et
al. 2000)
Calcium-
binding
proteins
S100 β
Calcineurin
Calsenilin
Aβ
Aβ
AD
Aging
AD
Review
(Eddleston and
Mucke 1993)
(Norris, Kadish
et al. 2005)
(Jin, Choi et al.
2005)
Cytokines /
growth
factors
FGF-1, FGF-2
BDNF
endothelin
G-CSF
GM-CSF
IFN-α, IFN-β
IGF-1
IL-1, IL-1α,β,
IL-8, NGF
TGF-α, β
TNF-α
GDNF
IL-12/IL23
13, 2
13
10,2
11
IL-1β,
IL-6
EGF
DA
LPS, PMA
LPS,
TNF-α
AD
MS
Review
(Eddleston and
Mucke 1993)
(Bresjanac,
Antauer et al.
2000),
(Ikeda, Koo et
al. 2002)
(Constantinescu,
Tani et al. 2005)
14
Table 1 continued
Category Up regulated
molecule
In vivo In vitro Human
brain
tissue
Reference
Cell cycle
proteins
Cyclin A
SHP-1
(tyrosine
Phosphatase)
2
Scratch
wound
(Koguchi,
Nakatsuji et al.
2002)
(Wishcamper,
Brooks et al.
2003)
Early
response
genes
AP-1
c-fos
hsp 68/70/72
Fyn (src-
related protein
tyrosine
kinase)
Hsp90
GRP94
C/ EBP-β and
δ
Fra-1 (fos
family)
Hsp 27
1
1, 3
1
1
1, 10
1
8
TPA,
FGF,
EGF,
Heat
shock
Aβ-
plaques
AD, PD
patients
Review
(Eddleston and
Mucke 1993)
(Chun, Crispino
et al. 2004)
(Jeon, Park et al.
2004)
Cardinaux,
Allaman et al.
2000)
(Pozas, Aguado
et al. 1999)
(Renkawek,
Bosman et al.
1994; Sanz,
Acarin et al.
2001) (Akbar,
Wells et al.
2001)
Eicosanoids Cys-LTs C
Cicarelli et.al
2004
Enzymes GS, Calpain
PKC α/β
Review
15
Table 1 continued
Category Up regulated
molecule
In vivo In vitro Human
brain
tissue
Reference
Cathepsin B, D,
APP
NADPH
oxidase
Cystatin
(cysteine
protease
Inhibitor)
BACE-1
1
1,2
Aβ
(Eddleston and
Mucke 1993)
(Abramov,
Jacobson et al.
2005)
(Aronica, van
Vliet et al. 2001)
(Hartlage-
Rèubsamen,
Zeitschel et al.
2003)
Receptors Death receptors
(FAS/ CD95,
FAS-L)
3 C
(Saas, Boucraut
et al. 1999;
Bechmann,
Lossau et al.
2000)
Misc bystin
Chaperone
protein 14-3-3
p 53
14
2
2
(Sheng, Yang et
al. 2004)
(Kawamoto,
Akiguchi et al.
2006)
(Chung, Shin et
al. 2002)
16
Signal transduction mediating astrogliosis
The molecular cues that govern the different changes characteristic of astrogliosis
have only recently started to emerge and even now the information is far from being
complete. This section will review the studies investigating signal transduction
pertaining specifically to the various morphological and biochemical changes.
Stellate morphology: The molecular basis of stellation has been largely studied in
the context of astrocyte differentiation or ‘astrocyte spreading’ (Abe and Saito 1997)
(Suidan, Nobes et al. 1997; Ramakers and Moolenaar 1998). Whether the molecular
pathways regulating astrocyte stellation are different during differentiation versus
astrogliosis is not clear. In vitro studies indicate cAMP-activated, and protein kinase
C-mediated signaling cascades could be the two major signaling pathways that
regulate stellate morphology in astrocytes (Abe and Saito 1997, 2000). Glycans,
which are carbohydrate moieties located on the astrocyte membrane, can also trigger
stellation via downstream tyrosine de-phosphorylation (Sasaki, Endo et al. 2000).
Adenosine-mediated stellation in astrocytes (Abe and Saito 1998) is also through
downstream tyrosine de-phosphorylation. Stellate morphology can be easily reversed
unlike other reactive features of astrocytes. Increase in glutamate levels (Abe and
Saito 2001), activation of serine protease thrombin (Pindon, Festoff et al. 1998), and
endothelins (Koyama, Ishibashi et al. 1993) are some of the known signaling
pathways involved in reversal of stellation.
17
Astrocyte hypertrophy: Astrocyte hypertrophy is one of the first changes to occur
during astrogliosis and is the result of an accumulation of intermediate filaments
leading to the increase in size. Hypertrophy of astrocytic processes in response to a
variety of CNS insults is also accompanied by an up-regulation of endothelin B
receptors (Hama, Kasuya et al. 1997; Ishikawa, Takemura et al. 1997; Koyama,
Takemura et al. 1999; Peters, Rogers et al. 2003), making them potential candidates
as regulators of astrocyte hypertrophy. One study suggests, exposure to endothelin-1
might result in the hypertrophic response in astrocytes due to uncoupling of the
astrocytic gap junctions leading to the increase in intermediate filament protein
content (Hasselblatt, Bunte et al. 2003). A recent study by Norris et.al, demonstrated,
that over-expression of activated calcineurin, a calcium dependent phosphatase, in
primary astrocyte cultures, resulted in thickening of the cell bodies and processes
closely mimicking the hypertrophic phenotype (Norris, Kadish et al. 2005). This
report provides the first evidence of a signaling cascade directly regulating
hypertrophic changes in astrocytes. Astrocytes from mice deficient in GFAP, the
main component of intermediate filaments, fail to display the hypertrophic
phenotype (Pekny, Eliasson et al. 1998). This underscores the importance of GFAP
over other intermediate filament proteins vimentin and nestin, in the development of
the hypertrophic morphology.
Astrocyte proliferation and migration: Several studies using BrDU and GFAP
double labeling have confirmed that, although mature astrocytes are considered
18
‘amitotic’, in the injured brain they re-enter cell cycle and divide, contributing to the
formation of the glial scar (Lee, Pappas et al. 2003; Tatsumi, Haga et al. 2005).
Manipulating the expression of cell-cycle proteins in reactive astrocytes can
influence their proliferation following injury. The over-expression of cyclin-
dependent protein kinase p27
kip1
in reactive astrocytes is accompanied by a down-
regulation of cyclin A, which inhibits their proliferation following a scratch wound
(Koguchi, Nakatsuji et al. 2002). Studies examining in-vivo injury models as well, as
tissue from human epilepsy patients have observed that the transcription factor
CREB, known for its importance in cellular proliferation, displays sustained
activation specifically in reactive astrocytes, and hence has been implicated in the
formation of the glial scar (Ong, Lim et al. 2000; Park, Kim et al. 2003).
There is evidence to support that the initial response to injury might also involve
migration of reactive astrocytes from distal parts to the lesion. Temporal mapping of
vimentin immunoreactivity following an injury suggested migration of vimentin
positive astrocytes from distal regions to the periphery of the lesion site immediately
following the damage (Wang, Bekar et al. 2004). S100A4, a calcium binding protein,
is another molecule implicated in the migration of astrocytes. Following injury, its
expression is downregulated, preventing further movement of reactive astrocytes and
thus possibly contributing to the rigidity of the glial scar. Induced over-expression of
this molecule on the other hand, increased the migration rate of reactive astrocytes
by several fold (Takenaga, Kozlova et al. 2006).
19
Even though evidence exists to support both astrocyte proliferation and migration of
reactive astrocytes to the lesion site, it is still a debated topic. Precisely, which sub-
populations of reactive astrocytes proliferate as opposed to migrating to the lesion
site and how these events might be temporally regulated has not been easy to
decipher.
Biochemical changes: The biochemical change consisting of an up regulation of
several different proteins is by far the most defining feature of reactive astrocytes
with important functional implications. Several common signaling molecules and
transcription factors have been investigated as potential inducers of astrocyte
reactivity owing to the fact that they are multifunctional and activated in the injured
brain across all cell types. However most of this information is based on
immunohistochemcical staining, which indicates expression of a particular signaling
molecule but not its functional significance in regulating astrogliosis. One way to
investigate a possible function would be manipulate the expression of the protein in
question and observe the consequence on astrocyte reactivity. Such an analysis has
been attempted by only a handful of studies investigating signal transduction in
reactive astrocytes.
One of the first signaling molecules to be analyzed for a role in astrocyte reactivity
was the extracellular signal-regulated kinase (ERK/MAPK) pathway since several
candidate inducers of astrogliosis are capable of activating this cascade. In vitro as
20
well as in vivo injury models have demonstrated an early and transient activation of
the ERK pathway associated with induction of astrocyte reactivity (Mandell and
VandenBerg 1999; Mandell, Gocan et al. 2001). The study also implicates the
pathway in mediating a paracrine signal that might be responsible for the spread of
reactivity to astrocytes distal from the site of injury. Inhibition of the ERK pathway
using specific inhibitors could block the induction as well as the spread of reactivity
as assessed by changes in GFAP expression. Interestingly, the activation of the ERK
pathway was also found in reactive astrocytes from brain tissue obtained specifically
from early-stage Alzheimer’s disease patients (Webster, Hansen et al. 2006). It is
thus reasonable to conclude the ERK pathway might one of the first signaling events
occurring following brain injury that might mediate astrocyte reactivity. The ability
of ERK to regulate a wide variety of cellular processes and downstream signaling
cascades further makes it a good candidate to be considered an ‘initiator’ signaling
molecule in astrogliosis.
The JNK and JAK/STAT (Janus kinase/ signal transduction and activator of
trascription) signaling cascades are also commonly used by a variety of cytokines
and chemokines known to induce reactivity in astrocytes. Early, transient activation
of STAT3 has been associated with kainate excitotoxicity (Acarin, Gonzâalez et al.
2000). Similarly, in the 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) model
of neurodegeneration, rapid activation of STAT3 via JAK2-mediated
phosphorylation was observed, prior to hypertrophy and the increase in GFAP
21
expression. Furthermore, inhibition of JAK2, prevented the activaton of STAT3 and
the resulting astroglial changes (Sriram, Benkovic, et al. 2004). An extended
activation of the JAK-STAT pathway lasting for more than 10 days has also been
associated with astrocyte reactivity, following kainate excitoxicty and transient
forebain ischemia in rat brains although this study does not suggest how the pathway
might be contributing to astrogliosis (Choi, Kim et al. 2003).
A recent study implicated the epidermal growth factor receptor (EGFR) mediated
pathway as one of the early signaling events invovlved in transforming quiescent
astrocytes into a reactive state. The activation of EGF receptor was found to regulate
not only the morphogical changes associated with astrocyte reactivity, but as
detected by microarray analysis, expression of a variety of genes. Since the
activation of this pathway was observed prior to induction of any features of
reactivity, it could be another important upstream signal triggering astrogliosis after
neural injury (Liu, Chen et al. 2006)
The p38 mitogen activated protein kinase (p38MAPK) pathway has been implicated
in pathological changes associated with inflammatory and apoptotic processes in
several cell types including neurons and as a recent study demonstrates also in
reactive astrocytes, following kainate excitotoxicity. Interestingly the activation of
p38MAPK in reactive astrocytes seems to be a delayed reaction and is sustained for
more than 10 days (Che, Yu et al. 2001). Furthermore, upon inhibition of p38 MAPK
22
significant suppression of astrocyte reactivity is observed as evidenced by decrease
in GFAP expression (Kim, Yu et al. 2004). Given its role in regulating inflammatory
responses, p38MAPK might be playing an essential role in mediating astrocyte
inflammation.
As table 2 indicates a number of signaling molecules and transcription factors are
activated in astrogliosis but there is limited information on how they might regulate
astrocyte reactivty. More detailed studies that can provide further downstream targets
of the signaling cascades or provide a functional significance need to be carried out.
23
Table 2- Signal transduction in astrogliosis
Key for In vivo injury models: 1=kainate, 2=transient focal ischemia, 3=entorhinal
cortex lesion, 4=axotomy, 5= spinal cord stab injury, 6= thermal ablation injury, 7=
MPTP, 8=cortical aspiration lesion, 9=morphine, 10=quinolinic acid, 11=EAE,
12=cryo-induced injury, 13=trauma, 14=hydrodopamine injection
Abbreviations: SW: scratch wound, EP: epilepsy, AD: Alzheimer’s disease, PD:
Parkinson’s disease, HD: Huntington’s disease, PKC: protein kinase C
Induced by Up
regulated
molecule
Physiological
function
In
vivo
In
vitro
Human
Proposed
role in
astrogliosis
Ref
ERK (p42
/44
MAPK)
Cell survival,
apoptosis
1
(3 to
12h)
SW
(4 -
8h)
Early
AD
cases
Paracrine
signal,
inducing
spread
of reactivity
(Choi, Kim
et al.
2003),
(Mandell
and
VandenBer
g 1999;
Mandell,
Gocan et
al. 2001;
Webster,
Hansen et
al. 2006)
p38MAPK Apoptosis 1,
(>10d)
Induction of
inflammatory
genes
(Che, Yu et
al. 2001)
c-PL2
(calcium-
dependent
phospholi
pase-2)
Cell survival
Apoptosis
2,
(24-72
h, 7d)
ATP
(72h)
Stellation,
proliferation
(Stephenso
n, Rash et
al. 1999;
Brambilla,
Ceruti et al.
2000;
Brambilla
and
Abbracchio
2001)
Secretory
Phospholi
pase-2
2,
(12h-
14d)
(Lin, Wang
et al. 2004)
PKC-γ Cell survival,
Apoptosis
Stellation,
proliferation
(Narita,
Suzuki et
al. 2004)
24
Table 2 continued
Induced by Up
regulated
molecule
Physiological
function In
vivo
In
vitro
Human
Proposed
role in
astrogliosis
Ref
calcineuri
n
Inflammatory
response
Aβ aging Hypertrophy
(Norris,
Kadish et
al. 2005)
CREB Cell survival 1,
(1-4
wks)
EP Proliferation
(Park, Kim
et al. 2003)
(Ferrer,
Blanco et
al. 1996;
Ong, Lim
et al. 2000)
c-myb differentiation 1,
(3-7d)
Induction of
reactivity
(Jeon, Shin
et al. 2004)
N-myc, c-
myc
differentiation AD,
PD,
HD
Induction of
reactive
phenotype
(Ferrer,
Blanco et
al. 2000)
gp130-
STAT1/S
TAT3
Inflammatory
response
1,2
(3h-3d)
GFAP
increase,
hypertrophy
(Choi, Kim
et al. 2003)
gp130-
JAK2-
STAT3
Inflammatory
response
7,
(10d)
Inflammatory
response
(Sriram,
Benkovic
et al. 2004)
Id family differentiation
Proliferation
1 AD Induction of
reactivity
(Aronica,
Vandeputte
et al. 2001)
NF-κ B Cell survival,
Apoptosis
6, (4-8
d)
2, 4 d
1, 2
wks
Inflammatory
response
genes,
Neurite
growth
(Nomoto,
Yamamoto
et al. 2001)
(Perez_Ota
no,
McMillian
et al. 1996;
Pennypack
er, Kassed
et al. 2000)
25
Functional significance of reactive astrocytes
Astrogliosis can be considered as an attempt by the injured brain to regain
homeostasis. Reactive astrocytes are hence equipped with a large repertoire of
proteins that allows them to influence various aspects of neuron function following
injury. It still remains controversial, if the induced changes are beneficial or
detrimental in nature. Every function attributed to reactive astrocytes seems to have a
positive as well as a negative effect on neuronal recovery depending on various
factors. Thus understanding the factors that influence the functional outcome of
astrogliosis is as important as possessing a thorough knowledge about how astrocyte
reactivity is triggered. This section will briefly discuss the different functions
performed by reactive astrocytes and the factors that execute these functions. The
information is summarized from earlier reviews on functional signifance of
astrogliosis (Eddleston and Mucke 1993; Giulian 1993) with inclusions of relevant
recent findings.
Glial scar formation and neurite growth: The glial scar formed by reactive
astrocytes helps to isolate areas of tissue necrosis. The various proteases and protease
inhibitors secreted by reactive astrocytes further help in tissue remodeling by
eliminating the debris of degenerating cells. On the other hand, the glial scar also
hinders neurite outgrowth, an important regenerative process. However, reactive
astrocytes are known to express neurite growth promoting factors as well besides the
26
inhibitory molecules. Hence under some circumstances, reactive astrocytes can be
permissive substrates and promote axon regeneration. There is further evidence to
suggest reactive astrocytes might differentially regulate dendritic extension as
opposed to axon growth following injury. While not permitting axonal extension,
dendrite growth might actually be enhanced in presece of reactive astrocytes
(Le_Roux and Reh 1994; Le_Roux and Reh 1995; Le_Roux and Reh 1995;
Le_Roux and Reh 1996; Monnerie, Esquenazi et al. 2005).
Astrocytic factors regulating neurite growth: Astrocytes express a range of proteins
both neurite outgrowth promoting and inhibitory in nature, which could in part
explain the contradictory influence on neurite growth (reviews: (Nieto_Sampedro
1999; Sandvig, Berry et al. 2004). The largest group of neurite growth inhibitory
molecules are the chondroitin sulfate proteoglycans (CSPG), which are extracellular
matrix proteins comprising of both transmembrane and secreted forms. Structurally
they have a central core protein attached to chondroitin sulfate and glycosaminglycan
side chains (GAG). The GAG side chains can be of varied lengths with different
sulfation patterns contributing to the diversity and neurite growth inhibitory
properties of CSPGs. Some examples of CSPG are neurocan, phosphacan, brevican,
and versican. There are no specific receptors known for most of the CSPG and hence
their mechanism of action is not very clear. Several studies using brain injury as well
as in vitro paradigms have documented the key role of CSPG in making reactive
astrocytes non-permissive to neurite growth. Furthermore, down-regulating CSPG
27
expression either with anti-bodies or enzymes significantly improves the
regenerative properties of reactive astrocytes. Another key inhibitory molecule is
GFAP, the princple cytoskeletal protein of astrocytes. Astrocytes from mice deficient
in GFAP retain their permissive nature and promote neurite growth following injury.
The mechanism by which GFAP exerts its neurite growth inhibitory effect is not
completely understood. However, there is a concurrent increase in laminin, a neurite
growth-promoting factor observed in GFAP deficient mice. This seems to suggest, a
GFAP-mediated suppression of neurite growth-promoting factors may be responsible
for its inhibitory effects. Another ECM molecule secreted by reactive astrocytes is
tenascin, which has many different isoforms, some neurite growth promoting and
several inhibitory. Tenascins and other secreted ECM molecules like hyaluronic
acid, Sema 3A, lam, fibronectin, and collagens through various homophilic and
heterophilic interactions form an extra-cellular matrix which forms an axon growth
inhibitory substratum. Selective degradation of this matrix also promotes axon
growth. The last group of inhibitory molecules is ephrins, initially seen during
development, aiding in axon guidance. Following injury, reactive astrocytes
especially in the spinal cord, express several members of this family. There is
evidence to suggest the expression pattern of these molecules may reflect that
observed during development, alluding to a possibility that functional reconnection
might be possible at least in some areas.
Astrocytes also secrete a variety of growth factors and express various cell
adhesion molecules that can promote neurite outgrowth. The fact that several of
28
these growth-promoting factors are also up-regulated in reactive astrocytes following
brain injury contributes to the confusion regarding when astrogliosis can retard or
promote axon regeneration. Among the most important are the cytokine S100-beta
and the growth factor FGF-2. Cell adhesion molecules are divided into the cadherins,
integrins, and the CAMs (cell adhesion molecules). N-cadherin, β1 integrin, N-
CAM, and ICAM are some members of these groups of proteins that are activated in
various forms of brain damage. Recently, neurotractin / kilon a member of the
immunoglobulin superfamily has also been found to be up regulated in reactive
astrocytes, possibly contributing to axonal sprouting observed in denervated dentate
gyrus. Increased expression of growth promoting molecules within subpopulations of
reactive astrocytes could be an explanation for the abnormal axonal sprouting
observed in the dentate gyrus following denervation as well as excitotoxicity (Liu,
Ying et al. 2005; Schèafer, Brèauer et al. 2005)
Neuronal survival: Following injury the two major sources of neurotoxicity are high
levels of excitatory neurotransmittors, and reactive oxygen species. Neurons
themselves have limited protective mechanisms against these toxins and require the
presence of astrocytes to cope. Reactive astrocytes offer significant neuronal
protection by up-regulating expression of enzymes like GS, involved in metabolism
of excitatory neurotransmittors. The increased expression of various growth factors
especially FGF-2, CNTF, NGF and IGF is also known to provide significant trophic
support and may serve to ameliorate neuron loss following injury (Giulian, Vaca et
29
al. 1993). Selective ablation of astrocytes from the brain prior to injury, resulted in a
dramatic down regulation of glutamate transporters thus hindering the astrocyte
mediated removal of excess glutamate, an excitotoxin, leading to excessve neuron
loss (Cui, Allen et al. 2001). This study provides the first in vivo report on the
importance of reactive astrocytes on neuronal survival. Reactive astrocytes also offer
protection against reactive oxygen species via up-regulation of antioxidant enzymes
like catalase and superoxide dismutase and free radical scavengers like glutathione
(GSH). Specifically following injury, neurons are known to downregulate the
expression of antioxidant enzymes like Cu/Zn superoxide dismutase (SOD), with a
concurrent up regulation of the enzyme in reactive astrocytes (Peluffo, Acarin et al.
2006).
On the other hand studies indicate that in some forms of injury, the factors secreted
by reactive astrocytes might serve as neurotoxins themselves and thus prove to be
detrimental to neuronal survival. Such astrocytic factors include various pro-
inflammatory cytokines and inducible nitric oxide synthetase (iNOS), the key
enzyme involved in the synthesis of nitric oxide, a key reactive oxygen species
(review(Brown, Bal-Price et al. 2003). Conditioned medium from astrocytes made
reactive by β-amyloid, can induce significant death in primary neuron cultures
(Paradisi, Sacchetti et al. 2004). Even under circumstances when they are not directly
toxic themselves, reactive astrocytes can make neurons susceptible to other toxins or
insults (Pâerez-Capote, Serratosa et al. 2004)(Mander Borutaite et al.2005)
30
Reactive astrocytes as antigen presenting cells: Although microglia are considered
the primary immune cells of the brain, there is accumulating evidence suggesting
reactive astrocytes might also have an important role in regulating immune reponses
within the CNS. Besides expressing antigens belonging to the major
histocompatibility complex (MHC I and II) reactive astrocytes are now also known
to express co-stimulatory molecules B7-1 and 2 (Zeinstra, Wilczak et al. 2003), other
essential components required for antigen presentation. The cytokine IL-12/IL-23
especially important in immune reponse observed in multiple sclerosis and other
inflammatory lesions is also up regulated in reactive astrocytes (Constantinescu, Tani
et al. 2005). Microarray analysis of the transcriptional response of astrocytes to a
mixture of inflammatory cytokines, revealed an upregulation of a variety of genes
usually associated with inflamed macrophages (Falsig, Pèorzgen et al. 2006).
However, in vivo studies still fail to support a prominent role for reactive astrocytes
as major antigen presenting cells. This gives rise to the possibility that they might be
more important in other regulatory immune functions.
Factors underlying the contradictory effects of astrogliosis
Following injury, the two principle aspects of neuronal recovery influenced by
reactive astrocytes are neurite growth and neuronal survival. As discussed above the
proteins capable of having both beneficial as well as detrimental effects on these
parameters of neuronal health, can be expressed by reactive astrocytes. However,
31
which subset of proteins dominates and drives the nature of astrocytic response is
influenced by the following factors.
Age of organism: Astrocytes from animals in a neonatal stage of development
display a milder reactive response to at least some forms of brain injury. The
postulated reason for a milder reactive response in neonatal astrocytes might be the
relative immaturity of the astrocytes, with lower cytokine responsiveness and a
molecular profile more permissive to neurite growth. While a simple stab wound to
the neonatal rat brain displays a mild astrogliosis, microinjection of a mixture of
cytokines at the time of injury resulted a much more robust astrocyte reactivity
(Balasingam, Tejada-Berges et al. 1994). It has also been suggested the limited
astrogliosis observed in neonatal organisms might be because only a certain sub-
population of astrocytes can become reactive. There is also some evidence to
believe that during this developmental stage, other non-astrocytic factors might be
more important in regulating neurite growth. Lesions to the corpus callosum of
embryonic rat pups (E17-E20), did not result in a reactive response from the
astrocytes and yet axons failed to regenerate (Ajtai, Kâalmâan et al. 2000). It has
been proposed that, there might be a critical period of development during which,
inflicting brain injury is associated with no or limited astrogliosis, consequently still
permitting axonal regeneration. Beyond that stipulated critical period, age of the
organism ceases to be the main factor effecting astrogliosis and neurite growth and
32
other parameters might be more influential in deciding the consequences of the
astroglial response (Kâalmâan, Ajtai et al. 2000) (Varga, Schwab et al. 1995).
Type and extent of injury: Although characteristic features of astrocytic reactivity
remain the same regardless of the type of insult or injury to the brain, there could be
subtle differences in their molecular profile, which in turn might dictate the
consequences on neuronal recovery. Even in neonatal animals, a stab wound which
is more localized results in a milder reactive response from astrocytes, as compared
to implantation of nitrocellulose fibers. The same form of stab wound in an adult
animal however elicits robust astrocyte reactivity (Balasingam, Tejada-Berges et al.
1994). Neonatal astrocytes treated with dibutryl cAMP, a syntheic analog of cAMP,
display several features of reactivity and are hence a commonly used culture model
of astrogliosis. However, despite showing common features of reactivity, they retain
the permissive nature allowing, even enhancing neurite growth (Wandosell,
Bovolenta et al. 1993; Miller, Tsatas et al. 1994). There is evidence to suggest that
the extent of time the astrocyte remains reactive, might influence which proteins are
expressed and in turn determine the consequences on neuronal function. Even
neonatal astrocytes, which normally form a permissive substratum for neurite
growth, when allowed to age in culture, over a period of time will display up-
regulation of inhibitory molecules resulting in neurite growth retardation (Smith,
Rutishauser et al. 1990). Although, prolonged astrocyte reactivity may not always
result in detrimental effects. For example, the reactive astrocytes constituting the
33
glial scar formed as a result of spinal cord hemisection in mouse display over time
higher ratio of growth promoting molecules (Camand, Morel et al. 2004). Thus, the
increasing permissivity of the glial scar over time seems to be unique to the spinal
cord astrocytes. It is hence possible that the sub-population of astrocytes in the
specific brain region targeted by the injury might influence the nature of their
response and subsequent effect on neuronal recovery.
Heterogeneity of reactive astrocytes: Reactive gliosis as classically defined has a
similar sequence of events associated with it regardless of the brain region. However,
there is heterogeneity in the astrocyte populations from the various parts of the brain,
which might influence their response to injury. For example, it has been shown that
astrocytes from various regions from the brain respond differently to toxins like ß-
amyloid with respect to the ratio of chondroitin sulfates (CSPGs), which are
inhibitory to the outgrowth promoting laminin (Canning, McKeon et al. 1993;
Hèoke, Canning et al. 1994; Hèoke, Silver et al. 1994). In general cortical and
hippocampal astrocytes, and cerebellar and spinal cord astrocytes were found to have
similar molecular profiles of neurite growth molecules in reponse to injury. Targeted
elimination of reactive astrocytes from the spinal cord actually exacerbated neuronal
loss and prevented axon regeneration following contusive injury to the spinal cord
(Faulkner, Herrmann et al. 2004). However, a similar ablation of reactive astrocytes
in a cortical stab wound, improved neurite growth and neuronal survival (Bush,
Puvanachandra et al. 1999). These observations further support the idea of a
34
functional difference in the sub-populations of reactive astrocytes from various parts
of the brain. The variety in the astrocyte populations might also be responsible for
the differences observed in the molecular response of reactive astrocytes surrounding
the lesion site and those distal from it (Review: Ridet and Gage, 1997). Some of
these differences may be responsible for the contradictory functional consequences
observed, especially on neurite extension. While reactive astrocytes surrounding the
lesion might be part of the glial scar and hence non-permissive for neurite growth,
those distal from the injury site are generally less reactive and more likely to
promote axon regeneration.
Thus astrogliosis can be influenced by multiple factors resulting in functionally very
opposite outcomes. A thorough understanding of the signal transduction pathways
regulating astrocyte reactivity might also provide an insight into why these
contradictory consequences exist. This dissertation investigates the role of caspases
as potential signaling molecules contributing to astrocyte reactivity. The following
sections discuss the characteristic features and mechanism of action of the caspase
family followed by the rationale for their involvement in astrogliosis.
Introduction to caspases
The following introductory information about caspases and their mechanism of
action has been assimilated from various review articles (Earnshaw, Martins et al.
1999; Fuentes-Prior, Salvesen et al. 2004). Caspases are cysteine proteases with a
35
characteristic aspartate cleavage site. Currently there are fourteen known members of
the caspase family. Most caspases are synthesized as dormant proenzyme, ranging in
size from 32 to 55 kDa. The inactive zymogen is then cleaved into an N-terminal
‘pro-arm’ and the active form of the enzyme composed of two sub-units one 20 kDa
(p20) and one 10 kDa (p10). Majority of caspases are found in the cytosol, with the
exception of caspase2, which resides in the Golgi, caspase 13, which is
predominantly located in the ER and caspase 9, which is predominantly
mitochondrial.
Main functional groups: The process of apoptosis as well as the processing of
proinflammatory cytokines requires the sequential activation of an initiator caspase,
which in turn activates an executioner caspase. Thus functionally caspases are
divided into the initiator and executioner caspases. The initiator caspases have a
characteristic long N-terminal or ‘pro-arm’ and are further divided into those with a
CARD (Caspase-Recruitement-Domain) or a DED (Death-Effector-Domain) at the
N-terminal. Apoptotic initiator caspases with a DED can either be triggered by cell
surface receptors and include caspase 8 and 10 and 12. On the other hand caspases
with a CARD include apoptotic caspases 2 and 9 as well as the inflammatory
caspases 1, 4 and 5 and are activated by environmental insults. The effector caspases
3, 6 and 7 as well as caspase 14 which is entirely non-apoptotic in function, have a
very short N-terminal with no recognizable interaction domains.
36
Basic mechanism of action: The quarternary structure of an active caspase
comprises of two procaspase zymogens forming a tetrameric complex. The substrate
cleavage site for all caspases is a tetrapeptide motif with the stringent requirement of
an aspartate residue in the P1 position. However the residue at P4 position forms a
determinant of caspase-substrate specificity. The catalytic reaction is hydrolysis of
the peptide bonds on the carboxyl side of the aspartate residue and is carried out by
the three catalytic residues Cys 285, His237 and Gly 238 on the active site, which
appear to be conserved in all human caspases. The process of converting a
procaspase to its active form involves two cleavages, one separating the prodomain
from the large subunit and another separating the large and small subunits. The
activation of initiator caspases is termed ‘scaffold mediated caspase activation’,
which involves an assembly of other molecules that either aid in increasing the net
concentration and/or induce a conformation change of the procaspase that facilitates
their self-activation.
Two distinct types of apoptotic cascades are known based on the molecular assembly
used to activate the initiator caspase. The first pathway is the extrinsic pathway,
consisting of cell surface receptor binding to their cognate ligands and recruit
adaptor molecules capable of forming homotypic interactions with procaspases
containing the DED motifs (caspases 2,8,10). These cell surface receptors capable of
inducing apoptosis are termed death receptors, with CD95 (Fas) and TRAIL
receptors DR4 and DR5 being the most widely known. Common characteristic of all
37
death receptors is a conserved stretch of amino acids termed the death domain, which
can interact with procaspases containing the DED motif. Stimulation of these
receptors with their respective ligands, results in recruitment of the adaptor molecule
FADD (Fas-interacting death domain) via the death domain as well as the
procaspase. The resulting complex activates the effector caspases. The second
pathway is the intrinsic caspase cascade, activated by environmental insults that
induce mitochondria to release cytochrome c. Release of cytochrome-c from
mitochondria is regulated by members of the bcl-2 family. Upon stimulation by an
apoptotic insult, pro-apoptotic bcl-2 protein Bax translocates to the mitochondria and
either forms pores or induces the opening of pre-existing pores, resulting in the
release of cytochrome c. Cytochrome-c along with the adaptor molecule, apoptosis
protein activating factor-1 (Apaf-1) and the initiator caspase-9 to forms an
apoptosome complex. This complex can then activate caspase-9 can then initiate the
caspase cascade by processing the effector caspase into their active forms. Certain
proteins of the bcl-2 family can act as points of interaction between the extrinsic and
intrinsic apoptotic pathways. Caspase-8 mediated cleavage of Bid results in
activation of Bax, which in turn can initiate the intrinsic apoptotic pathway (reviews:
(Earnshaw, Martins et al. 1999; Fuentes-Prior, Salvesen et al. 2004)
Mechanisms of caspase inhibition: The tight regulation of apoptosis is extremely
important since uncontrolled cell proliferation as well as premature cell death can
both lead to detrimental consequences. Various classes of proteins exist in the cell
38
that can protect the cell against untimely caspase-mediated cell death. Among these
are included, bcl-2 proteins, calpains, kinases and heat shock proteins. As emerging
studies suggest, these proteins might an important insight into why caspase
activation might not always result in apoptosis.
1) Bcl-2 family proteins : Members of the Bcl-2 family include pro as well as
anti-apoptotic proteins and the balance between them is critical for proper regulation
of apoptosis. The pro-apoptotic members like Bax are responsible for release of
cytochrome c from the mitochondria and initiation of the intrinsic apoptotic pathway.
At the same time, other anti-apoptotic members like bcl-2 and bcl-xl, have the
potential to inhibit caspases directly, via activation of endogenous caspase inhibitors.
There is contradictory evidence of a direct interaction between bcl-2 and caspases,
underlying caspase inhibition (review, Allen, Cluck et al 1998). However, recent
studies have uncovered adaptor proteins that might facilitate an interaction between
the anti-apoptotic members of the bcl-2 family and caspases, in turn leading to
caspase inhibition.
2) Enzymatic modifications : Caspase activation can also be regulated by post-
translational enzymatic modification. Phosphorylation of procaspase-9 via
serine/threonine kinase Akt (Cardone, Roy et al. 1998), or protein kinase A (Martin,
Allan et al. 2005) is known to inhibit its activation thus preventing initiation of the
intrinsic apoptotic cascade. Recent study reported Src kinase mediated
phosphorylation resulted in down-regulation, but not complete inactivation of
caspase-8, suppressing Fas-mediated apoptosis (Cursi, Rufini et al. 2006). Activity
39
of caspase 9 and caspase 3 can also be suppressed through nitrosylation of the
cysteine residues present in the active site, by nitric oxide (review (Mannick,
Schonhoff et al. 2001). Interestingly, inhibition of caspase-3 via nitrosylation is
known to be reversible (Kim, Talanian et al. 1997). Although prolonged
overproduction of nitric oxide can function as a pro-apoptotic agent, sustained, low
levels of nitric oxide can protect cells from several apoptotic insults. This dual nature
of nitric oxide makes it an important regulator of caspase function and might
contribute to some of the unconventional consequences of caspase function.
3) Calpains: These are calcium dependent proteases known to be important for
apoptosis. Several reports indicate the relevance of calpain mediated caspase
activation in apoptosis. In particular it is essential for activation of caspase-12
(Nakagawa, Yuan et al. 2000; Oubrahim, Chock et al. 2002). Calpain activation also
occurs upstream of caspase-3 resulting in cell death (Chen, Yap et al. 2006).
Interestingly, calpain-mediated cleavage of procaspase-7, 8, 9 that is different from
the regular cleavage resulting in caspase activation has been reported (Chua, Guo et
al. 2000). Thus, calpain-mediated caspase cleavage can result in inactive fragments
leading to caspase inhibition.
4) Endogenous inhibitors: A specific group of proteins termed inhibitors of
apoptosis proteins (IAPs) are important endogenous caspase inhibitors capable of
interacting with active caspases and inhibiting them or tagging them for a
proteasome-mediated degradation (reviews: (Ekert, Silke et al. 1999; Richter and
Duckett 2000; LeBlanc and Department of Neurology 2003). Mammalian IAPs are
40
thus far known to inhibit only caspase 3, 7 and 9. The IAPs contain a characteristic
Baculovirus IAP repeat (BIR) comprising approximately of 70 amino acids, which
can interact with the caspase active site thus preventing caspase function. The IAPs
that mediate their anti-apoptotic function through the BIR motif are X-linked IAP
(XIAP), cellular inhibitor of apoptosis protein-1 and 2 (c-IAP 1, c-IAP 2) and
neuronal apoptosis inhibitory protein (NAIP). Some of these IAPs (c-IAP 1,2, XIAP,
and livin) also contain a C-terminal zinc-binding motif called RING that can recruit
ubiquitin, thus priming the bound caspase for proteasome degradation. As discussed
in the following section, IAPs might play an important role in pushing caspase
activation into non-apoptotic consequences.
5) Heat shock proteins: Heat shock proteins are chaperone proteins induced by
a variety of chemical, environmental, and physiological stresses. Several members of
the heat shock protein family can directly interact with initiator procaspases and
prevent their conversion to the active form. These include Hsp 27, Hsp 70, Hsp72,
Hsp90 and α B crystallin (review: (Beere and St Jude Children's Research Hospital
2005).
Non-apoptototic caspase functions
The basis for the opinion that caspases might have functions beyond executing
apoptosis and cytokine processing has been the increasing number of reports
implicating caspases in various non-apoptotic cellular processes in the recent years
(Reviews: (Fadeel, Orrenius et al. 2000). It has been hypothesized that caspases
41
might control cell number by not only mediating cell death, but also influencing its
proliferation and / or differentiation. Thus, the two major non-apoptotic functions
attributed to caspases are their involvement in cellular differentiation and
proliferation. The first evidence of non-apoptotic caspase involvement in cellular
differentiation was observed in cell types that undergo a terminal differentiation
involving enucleation and loss of organelles, a process similar to final stages of
apoptosis. These cell types include erythroblasts (De Maria, Zeuner et al. 1999;
Zermati, Garrido et al. 2001), keratinocytes (Weil, Raff et al. 1999) and lens
epithelial cells (Ishizaki, Jacobson et al. 1998), which lose their nucleus and sub-
cellular organelles upon differentiation to their mature stage, and yet are
metabolically active. Caspase activation is a required event in the signaling cascade
that regulates these changes. The non-apoptotic role of caspase is best understood in
erythroblast differentiation, with caspase-mediated cleavage of target transcription
factors underlying postive as well as negative regulation of the process (review:
(Testa, Department of et al. 2004). Keratinocyte differentiation on the other hand
involves the predominantly non-apoptotic, caspase-14. Non-apoptotic caspase
activation is also associated with cell differentiation without enucleation, the best-
studied example being skeletal muscle cells (Fernando, Kelly et al. 2002; Barnoy,
Kosower et al. 2003). In Drosophila melanogaster not only multiple caspases, but
also other proteins acting upstream of caspases in the apoptotic cascasde are
involved in sperm differentiation (Arama, Agapite et al. 2003; Huh, Vernooy et al.
2004). One of the traditional roles of caspases is processing of pro-inflammatory
42
cytokines to their active form, thus forming an integral part of the innate and
inflammatory immune responses.
However, in the recent years there is evidence also implicating caspases in
adaptive immune reponses. Caspase activation in the absence of cell death in
proliferating activated T-cells was one of the first reported non-apoptotic functions
attributed to caspases (Wilhelm, Wagner et al. 1998; Alam, Cohen et al. 1999;
Kennedy, Kataoka et al. 1999). Recently caspase-mediated regulation of the cell
cycle of B-cells has also been reported (Woo, Hakem et al. 2003). Non-apoptotic
caspase involvement in cell proliferation is not restricted to immune cells and has
been reported in other cell types, which include neuronal precursor cells (Yan,
Najbauer et al. 2001; Oomman, Finckbone et al. 2004), radial glial cells (Oomman,
Strahlendorf et al. 2005; Oomman, Strahlendorf et al. 2006), and tumor cell lines
(Frost, Al-Mehairi et al. 2001). One study reports both apoptotic and non-apoptotic
caspase activation occurs in different cell types of cancerous pancreatic tissue
(Gansauge, Gansauge et al. 1998; Ramadani, Gansauge et al. 2001; Ramadani, Yang
et al. 2001). Their observations further support the premise that caspases might
regulate both cell proliferation and differention as well as cell death.
Among other non-apoptotic functions attributed to caspases are, maintenance
of cytoskeletal integrity and cell migration although these roles are not well defined
or studied. Casapse-3 like activity was detected in adherent cells, but not in the
suspension, and the general caspase inhibitor zVAD was able to prevent the cell
spreading (Watanabe and Akaike 1999). Recently another novel apoptotic function
43
in cell protection was assigned to caspase 2. Under certain circumstances caspase-2
instead of triggering apoptosis can sensitize the cell to genotoxin induced damage
and induce activation of the DNA repair machinery instead (Tinel, Tschopp et al.
2004). Table 3 reviews these studies, on the involvement of caspases in the various
cellular events in different peripheral cell types. The table also provides an update on
the current knowledge of the signaling cascades targeted by caspases in executing
these non-apoptotic functions.
Table 3- Non-apoptotic caspase functions and molecular targets
Function Cell type Activated
caspase
Ref Downstream
target
Ref
Proliferation T-cell Caspase-3
Alam
et.al, 1997
PARP
Wilhelmson
et.al, 1999
Caspase-8
Kennedy
et.al, 1998
Pancreatic
carcinoma
duct cells
Caspase-1
Ramadani
et.al, 2001
B-cells Caspase-3
Woo et.al,
2003
p21
cip 1/waf 1
Neural
progenitor
cells
Caspase-3
Caspase-
8, 9
Yan et.al,
2001
Ooman
et,al 2002
Radial glial
cells
Caspase-3
Ooman
et.al, 2005
Astrocytes Caspase-3
Zhu et.al,
2005
Differentiation Erythroblasts Caspase-3
DeMaria
et.al, 1998
GATA-1
Raf-1
SCF/tal
DeMaria,
1999,
Zeuner, 2003
Myoblasts Caspase-3
Fernando
et.al, 2002
MST1
Caspase-1
Barnoy
and
Kosower,
2003
Calpastatin
44
Table 3- continued
Lens fiber
cells
Caspase-3
Ishzaki
et.al,
2000,
Zandy
et.al, 2005
Bcl-2, IAP,
Weber et.al,
2005
Keratinocyte
s
Caspase-
14
Eckhart
et.al, 2000
PC12 to
neurons
Caspase-1
Vaisid
et.al, 2005
tau
Rohn et.al,
2005
Cell spreading Hsk cell line Caspase-3
Akaike
et.al, 1999
Focal
adhesion
kinase
Besides the ‘classical’ non-apoptotic roles in cell proliferation and differentiation
seen in a variety of cell types, recent studies have implicated caspases in important
CNS-related non-apoptotic processes. These include synaptic plasticity and axonal
guidance as well as contributing to neuroprotective pathways activated during
ischemic preconditioning.
Synaptic plasticity: The apoptotic cascade can be independently activated in the
synaptic terminals resulting in localized functions. Also, among the caspase
substrates present in the pre- and post synaptic compartments of the neurons (Chan,
Mattson et al. 1999) are glutamate receptor sub-units that are essential for synaptic
plasticity. Signal transduction proteins implicated in long-term potentiation (LTP)
for example iPLA2, PKC are also known caspase substrates, with synaptic
localization. Hence it is not surprising to find more studies suggesting caspase
function in modulating synaptic function, come to light. It has also been proposed
45
that a lower caspase activation might serve as an adaptive response to certain
stressors, and whereas higher levels of activation mediate synaptic degeneration and
cell death (Mattson, Keller et al. 1998; Mattson, Duan et al. 1999). Studies
investigating the non-apoptotic role of caspases in synaptic function have chosen to
do so in a widely used model of synaptic function, the NMDA-receptor dependent
long-term potentiation (LTP) in the CA1 region of the hippocampus. A significant
impairment of LTP was observed when rat hippocampal slices were incubated with
an irreversible cell permeable caspase-3 inhibitor DEVD for 1.5hr, with a complete
inhibition after incubation for 3.5 hrs (Kudryashov, Yakovlev et al. 2004). The
dependence of the magnitude of LTP impairment with caspase inhibition, on the time
of incubation with the caspase inhibitor suggests possible involvement of caspase
substrates and downstream pathways activated with their cleavage that might be
important for LTP. The first mechanistic evidence explaining caspase-mediated
regulation of LTP comes from a recent study by Lu et.al, which demonstrates a
caspase-1 dependent suppression of AMPA receptor-mediated calcium influx
leading to inhibition of LTP (Lu, Wang et al. 2006).
The use of caspase inhibitors in vivo also agrees with the idea that caspases might
have a role in synaptic function and hence possibly in memory and learning. The first
such report was by Dash et.al, (Dash, Blum et al. 2000) where they demonstrated a
caspase-dependent impairment of long-term spatial memory. Intra-hippocampal
infusion of caspase-3 inhibitor in rats has recently been shown to impair other forms
46
of learning and memory like contextual fear conditioning, spatial memory associated
with water maze and shuttle box learning (Stepanichev, Kudryashova et al. 2005). A
recent study also implicated caspase-1 in memory tasks, when they reported chronic
infusion of a specific caspase-1 inhibitor in aged rats ameliorated age-related
increases in hippocampal IL-1beta and improved memory for context (Gemma,
Fister et al. 2005).
Axon guidance and neurite growth: Growth cones are dynamic structures present at
the tips of growing axons and dendrites consisting of finger-like projections. They
can almost function as independent entities owing to key organelles like
mitochondria and ER situated just below the plasma membrane as well as the
presence of various receptors and ion channels. Besides other signaling pathways
even the apoptotic machinery can be independently activated in the growth cone.
Actin and several other proteins involved in regulating cytoskeletal dynamics that
contribute to growth cone motility are known caspase substrates, giving rise to the
notion that caspases might be involved in neurite extension. A preliminary study by
Mattson’s group reported an increase in the outgrowth rate of axons and dendrites
when cultured embryonic rat hippocampal neurons were treated with zVAD a broad-
spectrum caspase inhibitor (Gilman, Mattson et al. 2002). A recent study reported
caspase-3 activation to be essential for extension of retinal growth cones in response
to netrin-1 during development (Campbell, Holt et al. 2003).
47
Neuroprotection: Preconditioning is a phenomenon where a sub-lethal dose of a
toxin or an milder insult protects the cell from further more potent dose or extent of
the same insult. With respect to neurons this protective mechanism is frequently
found to be activated following a mild ischemic insult and is termed ischemic
preconditiong. Recent studies have found caspase-3 activation to be important for
the neuro-protective effect of ischemic preconditioning (Garnier, Ying et al. 2003;
McLaughlin, Hartnett et al. 2003).
Proposed mechanisms underlying non-apoptotic caspase activation
Sub-lethal activation of caspase: Several studies have consistently reported a sub-
threshold caspase activity associated with the non-apoptotic functions. Hence level
of caspase activity has been put forward as a putative mechanism underlying its non-
apoptotic consequences. These studies propose various mechanisms by which sub-
threshold caspase activation may lead to non-apoptotic consequences. A more
accepted theory is the differential cleavage of a substrate depending on the level of
caspase activation. RasGAP, a regulator of Ras- and Rho-dependent pathways, is the
first caspase substrate known whose caspase-mediated cleavage pattern was
dependent on the level of caspase activation. A sub-threshold level of caspase
activation in cells results in the first cleavage of RasGAP generating fragments that
can function as anti-apoptotic signals. However as the caspase activity is increased
the fragments are cleaved further potentiating an apoptotic response (Yang and
48
Widmann 2001; Yang and Widmann 2002). A recent study suggests similar
regulation of the transcription factor STAT3, which also possess multiple caspase
cleavage sites. Cleavage of STAT3 by low caspase activation can generate fragments
with non-apototic consequences (Darnowski, Goulette et.al. 2006). Several caspase
substrates have multiple cleavage sites making this an attractive regulatory
mechanism. A low level of caspase activity could be non-apoptotic also because the
cleavage pattern of the targeted substrate is changed making the substrate less
efficient or preventing it from reaching the proper apoptotic targets. Poly ADP
(ribose) polymerase (PARP) is a common apoptotic caspase-3 substrate. However its
cleavage has been observed even during non-apoptotic events like T-cell
differentiation. Interestingly, PARP cleavage occurring as a result of the sub-lethal
caspase-3 activation observed during ischemic preconditioning, is incomplete
resulting in a less efficient protein, which is turn is responsible for the
neuroprotective caspase effect (Garnier, Ying et al. 2003). Under some
circumstances, sub-lethal caspase activation might activate those substrates that are
normally inactivated upon caspase cleavage during apoptosis. This premise is based
on the example of NF-κ B, a transcription factor involved in mediating both adaptive
and innate immune responses, as well as regulating cell survival pathways and
apoptosis. During apoptosis, NF-κ B is cleaved and inactivated by caspase-3, thus
preventing the activation of any cell survival pathways mediated by it. However,
recent studies indicate low levels of both initiator as well as executioner caspases can
49
independently activate NF-κ B leading specifically to inflammatory response but not
apoptosis (Lamkanfi, Declercq et al. 2006).
Alternate splice variants: The protein-coding regions in human genes or exons are
interspersed with long non-coding regions called introns, which are selectively
removed from the precursor mRNA to obtain the mature transcript. Since each gene
can have several coding and non-coding regions, alternate splicing allows the
removal and joining of various combinations, of these exons and introns from the
precursor transcript. Hence each gene has the potential to generate several splice
variants, which could behave as discrete iso-forms with distinct functions in cellular
events. Alternate splicing can also result in transcripts that do not code for anything
and hence lose the original function serving as endogenous inhibitors of the main
protein. This mechanism has been used to regulate cellular processes like apoptosis,
axon guidance, sex determination, and cell excitation and contraction (Review:
(Schwerk, Schulze-Osthoff et al. 2005).
Caspases are part of the apoptotic machinery that can be regulated by alternate
splicing and almost all members of the caspase family have several isoforms (Jiang,
Wu et al. 1999). Interestingly while some isoforms perform the normal function of
executing apoptosis, some of them encode for inactive proteins or those with
antagonistic functions during cell death. Caspase 2 and 9 in particular are known to
generate shorter versions of the protein, Casp 2S (Ich-1S) and Casp 9S respectively,
50
which can inhibit cell death. A novel calcium binding protein termed Ich-1S binding
protein or ISBP was found to partially suppress processing of the longer version of
caspase 2 in vitro, and hence might be instrumental in increasing the production of
Casp 2S and mediating its cell survival pathways (Ito, Uehara et al. 2000). Casp 9S
lacks the catalytic domain and suppresses apoptosis by competing with full-length
caspase-9 for the same binding sites on Apaf-1 resulting in a non-functional
apoptosome formation (Seol, Billiar et al. 1999). Furthermore, cell lines stably
transfected with casp 9S, were not only more resistant to apoptosis but also
prevented degradation of the anti-apoptotic protein Bcl-XL (Yi, Wang et al. 2006).
Caspase-8, another up-stream caspase involved in the receptor-dependent apoptosis,
is also implicated in T-cell proliferation (Kennedy et.al, 1999, Alam et.al, 1999).
Targeted disruption of the gene for caspase-8 has also revealed its importance in
heart muscle development and hematopoiesis (Varfolomeev, Schuchmann et al.
1998). Several splice variants of caspase-8 exist (termed casp 8/a through casp-8/h)
of which, casp-8/a and 8/b are known to be primarily apoptotic (Eckhart, Henry et al.
2001) . Unlike caspase 2 or 9, splice variants of caspase-8 generated during human
leukocyte differentiation are non-functional and might serve to prevent accumulation
of caspase-8 from reaching critical levels, and hence apoptosis (Eckhart, Henry et al.
2001). However a recent study also reported a large percentage of human
neuroblastoma cell lines that were screened for sensitivity to Fas-mediated apoptosis,
expressed a distinct short version of caspase 8. This splice variant was a result of the
removal of exon 3, giving rise to a shorter death effector domain or DED making it
51
less sensitive to Fas treatment probably via interfering with further downstream
events in death signal transduction pathway (Kisenge, Toyoda et al. 2003). Splice
variants of caspase-1, 3, 4, 6,7, 10 and 14 also exist, but are not known to encode for
proteins with distinct functions.
The involvement of alternate splicing in regulating caspase expression is less likely
to be important in situations of acute trauma, where a fast cell death is favorable
(Schwerk, Schulze-Osthoff et al. 2005). On the other hand it makes a very attractive
mechanism for suppressing caspase activation or restricting it to a sub-lethal level,
preventing cell death or possibly divert it to the non-apoptotic functions like
differentiation and proliferation.
Substrate specificity: Some of the non-apoptotic caspase functions seem to require
the cleavage of a specific caspase substrate. For example erythroblast differentiation
is executed by the caspase-cleavage of GATA-1, a transcription factor specific to
erythrocytes. Similarly caspase-mediated cleavage of another transcription factor
Mst-1, unique to skeletal muscle cells is the underlying mechanism for myoblast
differentiation to mature skeletal muscle cells (Fernando, 2002). Caspases promote
cell proliferation by inactivating cell cycle inhibitors. Proliferating B-cells in
particular require caspase-mediated cleavage of p21
cip1/waf 1
(Woo, Hakem et al.
2003). There is however, no effect on p27
kip1
, another cell cycle protein and a
caspase substrate. Thus the non-apoptotic caspase function in B-cell proliferation
52
seems to be mediated specifically, via one particular substrate, even though other
caspase substrates might be expressed. As indicated in Table 3, the specific caspase
substrates mediating non-apoptotic functions in the various cell types are still not
known entirely. Further studies might provide more information on whether such a
non-apoptotic caspase function is indeed susbstrate specific. Substrate specificity
might not be a requirement underlying all non-apoptotic caspase functions, or might
be influenced by other factors like level of caspase activation. It is also possible
different subsets of proteins are targeted by caspases at sub-lethal activation or the
cleavage pattern might change resulting in non-apoptotic functions. Unfortunately
there is no information about the level of caspase activation in differentiating
erythroblasts or myoblasts, which utilize specific transcription factors that are
caspase substrates.
Regulation by molecules upstream of the caspase cascade: Certain molecules
upstream of the caspase cascade can also influence the consequences of caspase
activation steering it towards non-apoptotic functions. These molecules include the
adaptor proteins recruited by the death receptors (Park, Schickel et al. 2005) as well
as proteins that don’t normally interact with caspases but possess the CARD domain
(Hong, Jung et al. 2002). As discussed earlier mitochondrial initiator caspases are
activated through interactions with adaptor proteins that contain the CARD motif at
the N-terminal. Interestingly, recent studies have uncovered various CARD
containing molecules that are involved in cell survival pathways and inflammation
53
via NF-κ B activation, giving rise to the theory that they might function as a cellular
switch resulting in apoptosis or cell survival depending on the cellular context. One
example is Nod 1 a member of the Apaf-1 family, which can activate procaspase-9
through its CARD domain. However, unlike Apaf-1, Nod-1 also has another domain,
which also makes it capable of activating NF-κ B and the associated downstream
inflammatory or cell survival pathways (Inhara et.al. 1999). Certain CARD-
containing proteins like Psuedo-ICE and ICEBERG have a significant sequence
homology to the inflammatory caspase-1 (Druilhe, Srinivasula et al. 2001), which
also contains the CARD motif and hence can inhibit activation of this caspase by
directly interacting with the CARD domain. Death receptors Fas (CD95) and
members of the TRAIL family can also steer caspase activation towards non-
apoptotic consequences.
Thesis objectives
Caspase activation is a widespread event following any form of damage to the brain.
Apoptosis and inflammation, which involves cytokine processing, are two major
cellular responses to CNS damage largely exhibited by neurons and microglia
respectively. Hence, active caspases have been demonstrated in neurons as well as glia
(Beer, Franz et al. 2000), Rossiter 2002, (Krupinski, Lopez et al. 2000), (Ferrer, Lopez
et al. 2000). In contrast, several studies have demonstrated that astrocytic expression of
active caspases is not always associated with cell death (Beer et.al, 2000, Ferrer et.al,
2000, Narkilahti et.al., 2003). In fact the recent most study by Narkilahti et.al observed
54
caspase-3 immunoreactivity predominantly in astrocytes in the absence of overlap with
cell death marker TUNEL. However, significance of this non-apoptotic caspase
activation in astrocytes has not been investigated fully. Since the main astrocytic
response to injury is astrogliosis, it is possible that the observed non-apoptotic caspase
activation in astrocytes might also play a role in astrogliosis. Increased expression of
active caspase-3 in astrocytes has been shown to co-localize with increased expression
of inducible nitric oxide synthetase (iNOS), a known marker of reactivity in astrocytes
(Acarin, Peluffo et al. 2005). Another recent study demonstrated caspase-3 activation in
proliferating astrocytes, another feature of astrogliosis, following kainate excitotoxicity
(Zhu, Dahlstrèom et al. 2005). Although, these studies rely only on
immunohistochemical data they offer some circumstantial evidence for a possible
caspase involvement in astrogliosis.
Astrogliosis is often thought, to be a process of ‘de-differentiation’ wherein reactive
astrocytes revert back to a more immature state by undergoing transcriptional changes
and subsequent increase in expression of various proteins. As discussed previously
reactive astrocytes express transcription factors like Id (Andres-Barquin, Hernandez et
al. 1998; Tzeng, Kahn et al. 1999) (Aronica, Vandeputte et al. 2001) c-myb (Jeon, Shin
et al. 2004), c-myc (Ferrer, Blanco et al. 2000) and STAT (Acarin, Gonzâalez et al.
1998; Acarin, Gonzâalez et al. 2000) that are predominantly associated with cellular
differentiation. This theory that a subset of astrocytes could have been differentiated
into a reactive state is also supported by the ex-vivo models of astrogliosis where
55
reactive features of astrocytes that are generated from injured brains are not only
retained in culture but also persist through multiple cell division cycles (Wu, Nishiyama
et al. 1998; Rozovsky, Wei et al. 2005) Given the fact that known non-apoptotic
caspase functions involve mediating cellular proliferation and differentiation and since
the reactive state of astrocytes displays features of both processes it further suggests the
possibility of a caspase involvement in astrogliosis. The studies in this dissertation hope
to investigate this hypothesis of a possible caspase function in astrogliosis.
Study 1 (Chapter 2) comprises of studies investigating this hypothesis in an in vitro
model of astrogliosis as well as an ex-vivo paradigm. If caspases had a non-apoptotic
function in astrogliosis, we would expect a significant caspase activity in reactive
astrocytes in the absence of cell death. Furthermore, inhibition of caspase activity would
influence parameters defining astrocyte reactivity. Stellate morphology and the
expression levels of glutamine synthetase (GS) and fibroblast growth factor-2 (FGF-2)
were the chosen criteria to assess astrocyte reactivity in this study. As discussed
previously, GS is one of the classic markers of astrocyte reactivity and FGF-2 is
important as an inducer of astrocyte reactivity as well as a mediator of its functions. I
use an in vitro culture model of astrogliosis for initial investigation of the hypothesis
followed by an ex-vivo paradigm of adult astrocyte cultures from kainate and saline
injected animals. The study also includes a preliminary investigation into which
members of the caspase family might be contributing to astrogliosis.
56
Study 2 (Chapter 3) discusses studies investigating a potential involvement of caspases
in astrocyte-mediated effect on neuron function. Reactive astrocytes have important
consequences on neuron recovery following injury. Hence if caspases were involved in
regulating astrocyte reactivity, they might also contribute to functions attributed to
reactive astrocytes. My studies are focused on neurite growth and neuronal survival, the
two important parameters of neuron function influenced by reactive astrocytes.
57
Chapter 2
Non-apoptotic involvement of caspases in astrogliosis
Abstract
Astrogliosis is a characteristic response of astrocytes to almost all forms of brain injury.
Despite being studied extensively, the molecular basis of astrogliosis remains largely
unknown. Some of the signaling molecules and transcription factors commonly known,
to be activated in the injured brain are also implicated in astrogliosis. Caspases, a family
of cysteine proteases is one such group of signaling molecules with extensive activation
in the damaged brain. While caspases are, predominantly involved in execution of
apoptosis and cytokine processing, they are now known to have non-apoptotic functions
in several cell types. The present study investigates a potential non-apoptotic
involvement of caspases in astrogliosis using an in vitro model. Our results show
astrocytes made reactive by treatment with dibutryl cAM P and Aβ
25-35
, demonstrate an
increase in total caspase activity with a corresponding increase in the expression of
active pro-apoptotic caspase-3 in the absence of cell death. In addition, caspase
inhibition by zVAD, a broad-spectrum caspase inhibitor resulted in a partial attenuation
of the increased expressions of two known markers of reactivity, glutamine synthetase
and fibroblast growth factor-2, thus suggesting a non-apoptotic role for caspases in
mediating astrogliosis. We further extended our study to an ex-vivo model of
astrogliosis, comprising of adult hippocampal astrocyte cultures generated from kainate-
lesioned rats. Astrocytes from the kainate-lesioned animals exhibited reactive features
58
in culture, and a non-apoptotic caspase activation similar to that observed in the in vitro
model. Together, the data from these two models of astrogliosis support the hypothesis
of a non-apoptotic involvement of caspases in astrogliosis. Analysis of the caspase
family using specific inhibitors suggests caspase-11 and 3 might contribute to the
caspase function in astrogliosis.
59
Introduction
An important astrocytic function is to respond to injury and restore homeostasis in the
damaged brain. This reaction is called astrogliosis and it occurs in response to diverse
forms of neurological insults as well as normal aging. In general, astrogliosis consists of
morphological changes including hypertrophy, stellation and proliferation, as well as
biochemical changes involving increase in expression of numerous proteins, several of
which have important functions (reviews: (Malhotra, Shnitka et al. 1990; Eddleston and
Mucke 1993; Ridet, Malhotra et al. 1997). Although it is known that various signaling
molecules and transcription factors are up regulated in reactive astrocytes, there is a
lack of cohesive understanding about the molecular cascades underlying astrogliosis.
Reactive astrocytes have important functional consequences on neurons as well as other
glia following injury to the brain. Hence a better knowledge of the molecular pathways
that govern this process of reactivity might contribute greatly to our understanding of
brain injury and disease.
Previous studies investigating the molecular pathways associated with astrocyte
reactivity have focused on signaling factors that are commonly activated in the injured
brain. One such group of signaling molecules is the caspase family of cysteine
proteases, which are primarily involved in the execution of apoptosis and pro-
inflammatory cytokine processing (review: (Earnshaw, Martins et al. 1999),(Chang,
Yang et al. 2000; Grèutter and Institute of Biochemistry 2000) (Fischer, Jèanicke et al.
2003)). Apoptosis and inflammation, which involves cytokine processing, are two
60
major cellular responses to CNS damage largely exhibited by neurons and microglia
respectively. Hence, active caspases have been demonstrated in neurons as well as glia
(Beer, Franz et al. 2000), Rossiter 2002, (Krupinski, Lopez et al. 2000), (Ferrer, Lopez
et al. 2000)). Interestingly, caspase activation in astrocytes has also been demonstrated
in the absence of cell death. In a traumatic brain injury model increased caspase-3
immunoreactivity was observed in astrocytes 6-72 hrs following injury (Beer, Franz et
al. 2000). Another study by Ferrar et.al., also observed increased activation of several
caspases in astrocytes 24 hrs following kainate-induced excitotoxicity (Ferrer, Lopez et
al. 2000). Both studies reported a limited or no co-localization with cell death markers.
A recent study investigating the delayed effects of kainate-induced seizures observed
non-apoptotic caspase-3 activation localized predominantly in astrocytes, 7 days
following injury (Narkilahti, Nissinen et al. 2003). Some studies attribute caspase
activation in astrocytes to glial degeneration (Su, Nichol et al. 2000). However
considering the fact that cell death is not the principal astrocytic response to injury, it is
possible that caspase activation in astrocytes represents other non-apoptotic functions.
This hypothesis of a possible non-apoptotic caspase function in astrocytes is consistent
with the emerging data, which suggest that caspases might mediate various non-
apoptotic functions including regulation of cell cycle progression and proliferation,
survival pathways, differentiation, and cell motility. Such non-apoptotic caspase
functions have been described in a variety of non-neuronal systems (review: (Zeuner,
Eramo et al. 1999), (Schwerk, Schulze-Osthoff et al. 2003), (McLaughlin and
61
Department of Pharmacology 2004). Recent studies have extended the role of activated
caspases in normal cell function to the CNS (review: (McLaughlin and Department of
Pharmacology 2004), (Gulyaeva, Department of Functional Biochemistry of the
Nervous System et al. 2003). A role for caspase-3 has been suggested in neuronal
differentiation during forebrain development (Yan, Najbauer et al. 2001), in synaptic
plasticity and growth cone guidance (Campbell, Holt et al. 2003), (Gilman, Mattson et
al. 2002), (Chan, Mattson et al. 1999), (Gulyaeva, Kudryashov et al. 2003), as well as
contributing to cell survival pathways activated during ischemic preconditioning
(McLaughlin, Hartnett et al. 2003), (Garnier, Ying et al. 2003). Thus caspases could
regulate other functions along with mediating apoptosis in the injured brain. The present
study investigates a possible non-apoptotic function for caspases in astrogliosis.
62
Materials and Methods
Materials: Dibutryl cAMP, lipopolysaccharide, kainate and staurosporine were
purchased from Sigma-Aldrich (St. Louis, MO). β-amyloid (Aβ) peptide 25-35 was
acquired from Bachem (Torrance, CA). Caspase inhibitors- zVAD-fmk, DEVD-fmk,
WEHD-fmk, AEVD-fmk, YVAD-fmk, LEHD-fmk, and LETD-fmk were obtained from
Enzyme systems (MP biomedicals, Irvine, CA).
Neonatal astrocyte cultures: Astrocytes were cultured using the method of McCarthy
and deVellis (1980) with modifications as described previously (Pike, Vaughan et al.
1996). Briefly, cortices were dissected from postnatal (P0-P1) Sprague-Dawley rat
pups, then enzymatically dissociated by incubating for 5-7 min at 37
o
C, in 0.25%
trypsin solution in Ca
+2
/Mg
+2
free Hank’s balanced salt solution (supplemented with
20mM HEPES. 4.2 mM bicarbonate, 1mM pyruvate). Trypsin was quenched by adding
DMEM / FBS (26.2mM bicarbonate, 20 mM HEPES, 1mM pyruvate and
supplemented with 10% fetal bovine serum). Dissociated cells were centrifuged and the
pellet resuspended in DMEM/FBS. The cell suspension is triturated by passing through
a constricted flame polished pasteur pipette and filtered using a 40µM cell strainer
(Falcon). The cells are plated into 75 cm
2
flasks containing DMEM/10% FBS and
placed in a humidified incubator at 37
o
C with 5% CO
2
. The cells are fed with fresh
medium every three days and allowed to grow to confluence over a period of 10-14
days. After the cells are confluent, they are subjected to overnight shaking at 37
o
C at
63
280 RPM on an orbital shaker, to eliminate non-astroglial cells. Purified cultures are
trypsinized and then re-plated onto multi-well poly-lysine coated plates in DMEM/
10%FBS, at a density of 12,000/cm
2
. The cultures are maintained for a period ranging
from 4d-4wks for various experimental procedures. The cultures are shifted to serum
free medium for a period of 1-3 days prior to use in experiments.
Adult astrocyte cultures: Adult astrocytes are cultured using the method of Schwartz
and Wilson (Schwartz and Wilson 1992). Hippocampus and cortex are dissected from
adult Sprague-Dawley female rats (90day), followed by enzymatic dissociation with
trypsin, at 37
o
C for 7 minutes. Dissociated cells are centrifuged and the pellet
resuspended in DMEM/F12 supplemented with 20% fetal bovine serum. The cells are
plated into 50 cm
2
flasks containing DMEM/20% FBS and placed in a humidified
incubator at 37
o
C with 5% CO
2
. The cells are fed with fresh medium every three days
and allowed to divide over a period of 10-14 days, till at least 15 separate colonies of
astrocytes are detected. At this stage the cultures are passaged once and allowed to
become confluent. After the cells reach confluence they are subjected to overnight
shaking at 37
o
C at 120 RPM on an orbital shaker, to eliminate non-astroglial cells.
Purified cultures are trypsinized then re-plated onto multi-well poly-lysine coated plates
in DMEM/F12+ 20%FBS. The cultures are shifted to serum free medium (DMEM-F12)
for a period of 1-3 days prior to use in experiments.
64
Cell viability : Assessment of cell viability was by measuring lactate dehydrogenase
(LDH) activity and cell counts. For the in vitro experiments, cell viability was assessed
4 days after treatment with the reactive stimuli, dBcAMP and Aβ
25-35
.
In the adult ex
vivo cultures, from kainate and sham-lesioned animals, viability was assessed 48 hours
after plating in multi-well plates. LDH ASSAY: LDH was measured using a 96 multi-
well plate spectrophotometric assay (Koh and Choi, 1987). Conditioned medium was
added to each well, with triplicates for each condition. Further, β-NADH at a
concentration of 0.5 mg/ml in potassium phosphate buffer was added to each well. After
incubation at room temperature for 30 min., the substrate pyruvate (12 mM) was added
just before reading the plate. The plate was read at 340 nm on a Molecular Devices
spectrophotometer. SOFTmax PRO was used to analyze the data. CELL COUNT: A
cell viability kit from Molecular Probes (Eugene, OR) was used to count live cells
stained with calcein acetoxymethyl ester (calcein AM) as well as dead cells stained with
ethidium bromide. For each independent experiment, 4 fields of view per well were
counted, and data was collected from 3 independent experiments. In both the viability
assays, astrocytes treated with calcium ionophore A23187 (5nM, 12hrs) were used as a
standard for maximum cell death. Data were statistically examined by ANOVA,
followed by between group comparisons using Fisher’s LSD test. For graphical
representation, cell viability is expressed as percentage of dead cells counted or LDH
release in the calcium-ionophore, treated maximum death condition.
65
Western Blot: Western blotting was done using a previously described protocol (Pike
et.al. 1996). Cell lysates were obtained by addition of boiling reducing buffer (62.5mM
tris-HCl, 1% sodium dodecyl sulfate (SDS), 2.5% b-mercaptoethanol), then collected
and boiled for 5 min, followed by shearing the samples by passing them through 26-
gauge needle and centrifugation for 5 min. Protein separation was done by SDS-PAGE.
Equivalent lysate volumes were loaded into SDS 12% polyacrylamide gels and
electrophoresed at constant 135 V, then transferred to 0.45µM PVDF membrane at a
constant 100V. The membranes were then blocked by TTBS/5% non-fat milk (10mM
tris, 100mM NaCl. 0.1% tween-20, 5% dry non-fat milk) followed by sequential
incubation with the primary and secondary antibodies in TTBS/5% non-fat milk with
6x5min rinses with TTBS following incubation with each antibody. Protein-antibody
conjugates were visualized on film (Hyperfilm, Amersham) by enhanced
chemiluminescence method (ECL, Amersham). For data quantification, images from
scanned films were analyzed by NIH image (1.61). Results from at least 3 independent
experiments were combined and statistically compared by ANOVA. The different anti-
bodies used were: Glutamine synthetase (1:1000), Fibroblast growth factor-2 (1:500)
BD Biosciences, (San Diego, CA, USA), GFAP (1:15000) from DAKO, Caspase-3 (1:
50) EMD Biosciences. To confirm equal amount of protein in all conditions, blots were
stripped and re-probed with ß-tubulin. The procedure for stripping of blots involved
treatment with 100 mM glycine at pH 2.5, followed by the strip solution consisting of
62.5 mM Tris, 2% SDS and 0.7% β-mercaptoethanol and probing with β-tubulin
(1:500, Chemicon).
66
Caspase Activity Assay: Astrocytes were grown in 6-well plates for the caspase assays
with 2 wells for each condition. In order to obtain a cell suspension, cells were collected
by mechanical scrapping, followed by centrifugation at 3000 rpm for 30 minutes and re-
suspension of the pellet in 400µl of growth medium. The resulting cell suspension was
assayed for caspase activity except for 100µl, which was used to measure protein levels.
A fluorimetric assay kit (Calbiochem, San Diego, CA) was used to measure total
caspase activity, and the assays were performed following the manufacturer’s
instructions. Briefly, after correcting for protein content, cell suspension from each
condition was incubated with 1 µl of FITC-conjugated caspase substrate for 45 minutes.
This was followed by centrifugation at 3000rpm for 10 minutes, and re-suspension of
the pellet in wash buffer provided with the kit. The cells were washed twice in this
manner, before the final re-suspension in100µl of wash buffer, which was loading onto
a 96-well plate, and read on a fluorimetric plate reader (Molecular Devices) at emission
wavelength of 345 nm and excitation wavelength of 548nm.
Stereotaxic kainate injections: Nine female Sprague-Dawley (90 day old) rats weighing
approximately 230g were used for the study. Rats were anaesthetized with an
intraperitoneal injection of sodium pentobarbital (50mg/kg body weight). Kainic acid
(1.5µl of a 0.5mg/ml solution) or sterile saline solution (2µl) was injected into the
hippocampus of both hemispheres (co-ordinates: -5.2 bregma, 4.5 lateral to the mid-
line, 4.0 mm from the surface of the cortex) using a Hamilton syringe. The needle was
inserted at a rate of 1mm/min, and withdrawn 10 min after the injection at a rate of
67
1mm/min. Injection was followed by suturing the scalp, and allowing the animal to
recuperate. Seizure behavior was not observed after the rats had revived. The animals
were sacrificed 2 weeks following the injections, by decapitation. The left hemisphere
was fixed with 4% paraformaldehyde, and preserved for immunohistochemistry
analysis. Astrocyte cultures were generated from the hippocampus of the right
hemisphere, to establish the ex-vivo paradigm of astrogliosis.
Immmunohistochemistry: The left hemisphere from each animal was fixed using 4%
paraformaldehye. The fixed hemi-brains were mounted in agarose and sectioned (40
µm) exhaustively in the horizontal plane using a vibratome. Every tenth brain section
was processed with the initial section being randomly selected from the first ten
sections. The free-floating sections were first permeabilized using 0.02% triton-X and
then incubated with 3% BSA (in 0.1M tris) to block non-specific binding. This was
followed by overnight incubation with primary anti-body for neuron-marker NeuN
(1:500) or astrocyte marker GFAP (1:15000) in 0.1M tris /0.02% Triton-X (NeuN
antibody: Chemicon, Temecula CA; GFAP antibody: ICN). The following day the
sections were sequentially incubated for 4 hours with the secondary peroxide-
conjugated antibody (anti-mouse for neuN and anti-rabbit for GFAP) and avidin-biotin-
peroxidase complex (ABC) for I hour. The immunostaining was visualized using 3,3’
diaminobenzidine (DAB) in absence of nickel treatment. Tissue was then rinsed,
mounted on glass slides and cover-slipped. Sub-sets of sections from each animal were
also stained with cresyl violet using a standard protocol.
68
Immunocytochemistry: Immunostaining with GFAP was performed for both the in vitro
and ex-vivo astrocyte cultures. The cells were fixed using 4% paraformaldehyde and
permeabilized using 0.01% Triton-X. After incubation with 3% BSA to block non-
specific binding, the cells were incubated with the primary GFAP antibody (1:15000,
Dako) for 2 hours. This was followed by sequential incubation with anti-rabbit
peroxide-conjugated antibody and avidin-biotin-peroxidase complex (ABC) each for 1
hour. The staining was visualized using DAB in the absence of nickel treatment.
69
Results
In vitro model of astrogliosis:
A frequently used method of studying astrogliosis in culture is to induce reactivity in
neonatal astrocyte cultures by various chemical or physical treatments. Common
examples of factors used to induce reactivity are dibutryl cAMP (dBcAMP), a synthetic
analogue of cyclic AMP, and β-amyloid (Aß). Previous work in our laboratory has
established the efficacy of these factors in inducing reactivity in astrocytes. Treatment
of primary neonatal astrocyte cultures (P0-P1) with 1mM dBcAMP or 25µM Aß
25-35
resulted in induction of reactivity demonstrated by stellate morphology (Fig1 B, C) and
increased expression of glutamine synthetase (GS) and fibroblast growth factor-2 (FGF-
2) two known markers of astrogliosis (Fig1 D, E). Significant increase in the expression
of GS and FGF-2 was seen as early as 24 hours (data not shown) after exposure to
dBcAMP and Aβ, with maximum increase at day 4 (Fig1 D, E). GS and FGF-2
expression varied from a three-fold to a six-fold increase in cultures from independent
experiments. Cultures treated with lipopolysaccharide (LPS) for 4 days failed to
demonstrate any features of reactivity. LPS treatment requires the presence of microglia
to induce reactivity in astrocytes and thus was used to demonstrate purity of the
cultures.
70
Figure 1
Figure 1: primary astrocyte cultures (A), display stellate morphology when treated
with 1mM dBcAMP (B), or 25µM Aβ (C). Western blots D and E demonstrate
increased expression of GS and FGF-2 in astrocytes treated with dBcAMP, Aβ, or
vehicle for 4 days. Data are representative of at least five independent experiments
71
Astrogliosis is associated with a non-apoptotic activation of caspases
If there were indeed a non-apoptotic caspase involvement in astrogliosis, then we would
predict an absence of cell death in the reactive astrocytes despite caspase activation.
Cell viability of the reactive astrocyte cultures was assessed by counting astrocytes
stained with vital dye calcein AM and the cell death marker ethidium bromide under
various conditions, (Fig2 A, C). Concurrently, lactate dehydrogenase (LDH) activity
assays, another method of assessing cell viability, were also performed (Fig2B). In
order to rule out a delayed apoptotic response, both the procedures were performed after
a maximum exposure of 4 days to the reactive stimuli. LDH assays (Fig2C) as well as
cell counts (Fig2B) indicated an absence of cell death. To examine the possibility of
non-apoptotic caspase activation in reactive astrocytes, caspase activity assays
concurrent with the cell viability assays were performed. At 48 hours following
treatment with either dBcAMP or Aβ
25-35
, a robust increase in reactivity is observed in
the astrocyte cultures, hence total caspase activity was measured at this time point.
Caspase activity in reactive astrocytes was increased approximately two-fold as
compared to non-reactive control astrocytes (Fig2D). We further used western blotting
to determine if the increased caspase activity in reactive astrocytes corresponded to an
up-regulation of the cleaved active fragment of caspase-3, the most common effector
caspase. Higher expression of the active cleaved 20 kD fragment of caspase-3 in
reactive astrocytes was observed compared to control astrocytes (Fig 2E). Interestingly,
the increase in total caspase activity (Fig2D) and cleavage of caspase-3 (Fig2E)
72
associated with reactivity was modest as compared to that observed in astrocyte cultures
subjected to an apoptotic insult, staurosporine (1µM) for 24hrs.
73
Figure 2: Astrocytes do not undergo cell death even after 4 days of treatment with
the reactive stimuli. A, astrocytes treated with vehicle, 1mM dBcAMP or 25 µM
Aβ
25-35
are stained green with the vital dye calcein AM demonstrating an absence of
cell death. Dead or dying astrocytes resulting from treatment with 1µM of calcium
ionophore A23187 are stained red with the cell death marker ethidium bromide. B,
Cell counts and C, LDH activity assays support the absence of cell death in reactive
astrocytes. D, total caspase activity in astrocytes following treatment with vehicle,
dBcAMP (1mM), Aβ
25-35
(25 µM) or vehicle for 48 hours; or with 1µM
staurosporine for 24 hours. E, western blot showing active caspase-3 expression in
control and reactive astrocytes. Data represent at least four independent experiments,
*p< 0.001.
74
Figure 2
75
Caspase activation in reactive astrocytes has functional significance
If caspases have a non-apoptotic function in astrogliosis, inhibiting caspases would
affect parameters of astrogliosis. To examine this, a broad-spectrum caspase inhibitor
zVAD was used to inhibit total caspase activity. Astrocyte cultures were pre-treated for
an hour with 50µM zVAD before inducing reactivity using dBcAMP and Aβ
25-35
Caspase inhibition resulted in a partial attenuation of the increased expression of GS
and FGF-2 (Fig3A, B). The other parameter of reactivity, stellate morphology was not
reversed by caspase inhibition. Since thrombin is a factor known to reverse stellate
morphology of astrocytes (Pindon, Festoff et al. 1998), we examined the effect on
stellation astrocytes pre-treated with thrombin (10nM) or with zVAD (50µM).
Following pre-treatment with thrombin or zVAD, cells were treated with the reactive
stimuli for 48 hours, then immunolabeled for GFAP (Fig4 A) and immunolabeled
stellate cells for each condition were counted. Thrombin caused reversal of
approximately 75% of the stellate astrocytes, whereas zVAD treatment had no effect
(Fig 4B). These data suggest caspases might play a role in the biochemical, but not
morphological changes associated with astrogliosis.
76
Figure 3
Figure 3: A, C western blots demonstrateGS and FGF-2 expression following dBcAMP
or Aβ treatment for 48 hours, with or without zVAD pre-treatment. Graphs B, D show
mean (+ / - SEM) densitometric values of GS and FGF-2 levels respectively from 4
independent experiments normalized to levels observed in vehicle treated astrocytes (*p
< 0.005)
77
Figure 4
Figure 4: Figure 4: A, astrocyte cultures fixed with 4% paraformaldehye and immuno-
stained with GFAP after pre-treatment with zVAD or thrombin prior to treatment with
vehicle or the reactive stimuli. B, cell counts of stellate versus non-stellate cells of
GFAP-labeled cultures. Data are representative of four independent experiments (* p <
0.001).
Specific inhibition of caspases-11 and 3 results in the maximum attenuation of
increased GS and FGF-2.
Since caspases typically function as part of a signaling cascade, it is likely that multiple
caspases are activated during astrogliosis. If astrogliosis involves multiple caspases,
specific inhibition of various caspases would have a differential effect on parameters of
astrogliosis. Inhibition of caspases more important in astrogliosis would have greater
attenuating effect on the expressions of GS and FGF-2. On the other hand, inhibition of
caspases not involved in astrogliosis would have little or no effect. Synthetic peptide
78
inhibitors specific for different caspases were used for a preliminary screening of the
caspase family. Neonatal astrocyte cultures were treated with increasing concentrations
(1,10, 30, 60, 100 µM) of the inhibitor for an hour prior to inducing reactivity with
1mM dBcAMP. Effect of inhibiting various caspases on increased expression of GS and
FGF-2 was examined. For a particular inhibitor effect of only the lower concentrations
(1, 10 and 30 µM) on GS and FGF-2 expression was considered relevant. At higher
concentrations the inhibitors can become promiscuous and reflect secondary
specificities. Therefore, those caspases whose inhibition attenuated GS and/or FGF-2
expression only at higher doses of the inhibitor (60 or 100µ M) were not considered
important for astrogliosis.
Based on this criterion, inhibition of caspase-11 with WEHD resulted in the
maximum attenuation of GS and FGF-2 (Fig5 A, B). Active caspase-3 expression was
observed in reactive astrocytes. And as expected, specific inhibition of caspase-3 with
DEVD also significantly attenuated increase in GS and FGF-2, although to a lesser
extent than that achieved by inhibition of caspase-11 (Fig5 C, D). These data suggest
that caspase-11 and 3 are more likely to be involved in astrogliosis, and caspase-11
might function upstream of caspase-3. AEVD and LETD, inhibitors for caspases 6 and
8 respectively, attenuated just the GS expression and only at the highest concentration
of 100 µM (Fig6 A, B). Inhibition of caspases 1 and 9 by YVAD and LEHD
respectively, did not attenuate either GS or FGF-2 expression (Fig6 A, C). Thus,
caspase 1,6,8 and 9 are not likely to be involved in astrogliosis.
79
Figure 5
Figure 5: Specific inhibition of caspases 11 and 3 resulted in maximum attenuation of
GS and FGF-2 expression in reactive astrocytes. A, C western blots of astrocyte
cultures were pre-treated with either caspase 11 inhibitor WEHD or caspase-3 inhibitor
DEVD at concentrations of 1, 10, 30, 60 and 100µM for 2 hours prior to treatment with
vehicle or dBcAMP (1mM) for 48 hours. Graphs B, D show mean (+/- SEM)
densitometric values of GS and FGF-2 levels from 4 independent experiments
normalized to levels observed in vehicle treated astrocytes *P < 0.05).
80
Figure 6
Figure 6: Specfic inhibition of caspases 1, 6,8 and 9 had no effect on GS and FGF-2
expression. A, C western blots of astrocyte cultures pre-trested with caspase-1inhibiter
YVAD, caspase-6 inhibitor AEVD, caspase-8 inhibitor LETD or caspase-9 inhibitor
LEHD at concentrations of 1, 10, 30, 60 or 100 µM for 2 hours prior to treatment with
vehicle or dBcAMP (1mM) for 48 hrs. Graphs B, D show mean(+ / -SEM)
densitometric values of GS and FGF-2 levels respectively from 4 independent
experiments normalized to levels observed in vehicle treated astrocytes (*, # p <
0.005).
81
Ex-vivo model of astrogliosis
To substantiate the significance of the in vitro studies, we further investigated caspase
involvement in astrogliosis in an in-vivo excitotoxicity model of injury. In order to
study caspase activation in astrocytes specifically, astrocytes from the hippocampus of
kainate lesioned animals were compared to those from saline-injected sham controls.
Ex-vivo astrocyte cultures from injury models are known to retain, the biochemical
changes associated with reactivity even in culture, thus permitting the study of reactive
astrocytes in isolation (Wu et.al., 1999; Rozovsky et.al., 2005). Kainate treatment
resulted in neuronal loss in the CA3 region of the hippocampus and robust astrogliosis
as seen by immunohistochemical staining with NeuN (Fig7 B), and GFAP respectively
(Fig7 C, D). Astrocytes from kainate-lesioned animals were reactive in culture as
demonstrated by an increased expression of the markers glutamine synthetase (GS) and
fibroblast growth factor-2 (FGF-2) (Fig9). However, no morphological difference was
observed between astrocytes cultured from kainate versus saline injected animals (Fig8
A, B). Lactate dehydrogenase activity assays were performed on cultures from kainate
and sham-lesioned animals. As seen in figure 8C there was no cell death in either
condition.
82
Figure 7
Figure 7: Kainate excitotoxicity results in neuronal loss and astrogliosis. A-F, 40
micron brain sections, immuno-stained with neuronal marker NeuN (A,B) and
astrgolial marker GFAP (C-F). Figures are representative of sham: N=4 and kainate:
N=5
83
Figure 8
Figure 8: Hippocampal astrocyte cultures from sham (A) and kainate-lesioned (B)
animals fixed with 4% paraformaldehyde and immunostained with GFAP. C, Graph
showing LDH release in astrocytes from sham, kainate animals and sham astrocytes
treated with apoptotic stimulus staurosporine. D, Graph showing total caspase
activity analyzed as a measure of FITC release. E, western blot of active caspase-3
expression with or without treatment with zVAD for 24 hours. Astrocytes from
sham-lesioned animals treated with 800nM staurosporine (STS) for 16 hours were
used as positive control for total caspase activity and active caspase-3 expression.
Data are representative of at least 4 independent experiments (* p < 0.05).
84
Figure 9
Figure 9: A, C representative western blots showing increase in GS and FGF-2
expression respectively in astrocytes from kainate lesioned (KA1, KA2) as compared to
those from sham controls. Graphs, B, D show densitometric analysis of GS and FGF-2
levels in sham and kainate astrocytes with or without zVAD treatment (sham N=4, KA
N=5, * p< 0.05).
85
Discussion
In this study, we investigated a possible non-apoptotic role for caspases in astrogliosis
using both an in-vitro and an ex-vivo model of astrogliosis. In both experimental
paradigms, significant caspase activation was observed in reactive astrocytes in the
absence of cell death. Consistent with the increase in total caspase activity, an up-
regulation of the cleaved active fragment of caspase-3, a common effecter caspase was
observed. Inhibition of total caspase activity using a broad-spectrum caspase inhibitor,
zVAD resulted in partial attenuation of the increased expression of glutamine
synthetase and fibroblast growth factor-2. However, since there was no effect on the
morphology of the reactive astrocytes, caspases might have a non-apoptotic role in
mediating the biochemical changes but not the morphological changes associated with
astrogliosis. Preliminary investigation of the various caspases suggested caspase-11
might also play a role in astrogliosis possibly upstream of caspase-3.
Caspase activation in astrocytes
Following injury, caspase activation in neurons is predominantly associated with
neuronal apoptosis. In contrast, several studies have demonstrated that astrocytic
expression of active caspases is not always associated with cell death (Beer et.al, 2000,
Ferrer et.al, 2000, Narkilahti et.al., 2003). However, significance of this non-apoptotic
caspase activation in astrocytes has not been investigated fully. Since the main
astrocytic response to injury is astrogliosis, it is possible that the observed non-
apoptotic caspase activation in astrocytes contributes to regulation of astrogliosis.
86
Increased expression of active caspase-3 in astrocytes has been shown to co-localize
with increased expression of inducible nitric oxide synthetase (iNOS), a known marker
of reactivity in astrocytes (Acarin, Peluffo et al. 2005). Another recent study
demonstrated caspase-3 activation in proliferating astrocytes, another feature of
astrogliosis, following kainate excitotoxicity (Zhu, Dahlstreom et al. 2005). However,
both these studies rely only on immunohistochemical data and thus offer only partial
proof for a possible caspase involvement in astrogliosis. For the first time, our study
presents direct evidence of a functional significance of non-apoptotic caspase activation
in astrocytes, wherein caspase inhibition influences the reactive state of astrocytes.
The fact that the reactive state of astrocytes can involve proliferation and has
features of cellular differentiation further supports the possibility of a caspase
involvement in astrogliosis, since the known non-apoptotic caspase functions involve
mediating cellular proliferation and differentiation. Astrogliosis can involve astrocyte
proliferation in certain conditions and is often thought to be a process of ‘de-
differentiation’ wherein reactive astrocytes revert back to a more immature state by
undergoing transcriptional changes and subsequent increase in expression of various
proteins. Reactive astrocytes express transcription factors like Id (Andres-Barquin et al.
1998; Tzeng et al. 1999; Aronica et al. 2001) c-myb (Jeon et al. 2004), c-myc (Ferrer et
al. 2000) and STAT (Acarin, Gonzâalez et al. 1998; Acarin, Gonzâalez et al. 2000) that
are predominantly associated with cellular differentiation. This theory that a subset of
astrocytes could have been differentiated into a reactive state is also supported by the
ex-vivo models of astrogliosis where reactive features of astrocytes that are generated
87
from injured brains are not only retained in culture but also persist through multiple cell
division cycles (Wu, Nishiyama et al. 1998; Rozovsky, Wei et al. 2005).
Caspase activation may be involved in mediating the biochemical but not
morphological changes occurring in reactive astrocytes
The change in astrocyte morphology from a flat polygonal to stellate shape is not
restricted to astrogliosis and is observed in astrocytes even during differentiation. The
molecular basis of stellation has been largely studied in the context of astrocyte
differentiation or astrocyte spreading (Abe and Saito 1997) (Suidan, Nobes et al. 1997;
Ramakers and Moolenaar 1998). There is some evidence to suggest that the
morphology and the biochemical changes associated with reactivity are differentially
regulated. Treatment of astrocyte cultures with PMA, a chemical agent activating the
MAPK signaling pathway results in stellation without causing the increase in different
bio-active factors (Abe and Saito 2000). We observed that astrocyte stellation induced
by dBcAMP is reversed when the growth medium is replaced without affecting the
increased glutamine synthetase expression (data not shown). Previous studies as well as
our findings using the ex-vivo paradigm of astrogliosis have found that, astrocytes from
the injured brain in culture exhibit only the biochemical and not the morphological
changes associated with reactivity (Fig 8; (Wu, Nishiyama et al. 1998; Rozovsky, Wei
et al. 2005). Furthermore, in-vivo studies investigating long-term effects of kainate
excitotoxicity on astrocytes have shown elevated GFAP levels which persist long after
reactive morphology is lost (Dusart, Marty et al. 1991). All these observations suggest
88
that the increased protein expression in reactive astrocytes is a more stable event
persisting through cell division whereas the change in morphology is more transient
and, both processes are possibly mediated by different signaling cascades. Our data
indicate that caspase activation might be a part of the signaling process that regulates
only the biochemical changes.
Caspase effects on astrogliosis may be mediated by caspase 11 and caspase3
The well-established caspase functions involve sequential activation of the various
initiator and effector caspases. Therefore we analyzed the effects of inhibiting specific
caspases, on the selected markers of reactivity (i.e. glutamine synthetase and fibroblast
growth factor-2) and we observed a differential effect of inhibiting various caspases.
Down-regulation of caspase-11, an upstream inflammatory caspase, resulted in
maximum attenuation of GS and FGF-2. Inhibition of caspase-3 also resulted in
significant attenuation of the two markers, although not to the extent achieved by
selectively inhibiting caspase-11, suggesting that caspase-11 might work upstream of
caspase-3 in mediating astrogliosis. Caspase-11 is predominantly known as an
inflammatory initiator caspase, essential for activation of caspase-1 (Wang, Miura et al.
1998). However recent studies have demonstrated that under certain pathological
conditions, caspase-11 could also activate caspase-3, either directly or via activation of
caspase-1 (Kang, Wang et al. 2000). These cascades involving caspases 11,1 and 3 are
known to be particularly important for inflammation induced glial apoptosis (Hisahara,
Yuan et al. 2001; Hur, Kim et al. 2001; Suk, Kim et al. 2002). Specific inhibition of
89
other initiator caspases 8 and 9, the effector caspase 6 and the inflammatory caspase 1
all resulted in either no attenuation or attenuation only at higher concentrations of the
inhibitor, when they lose specificity. Therefore, our data thus imply a caspase cascade
involving direct activation of caspase-3 by caspase-11, might be involved in mediating
astrogliosis. However, further studies would be needed to confirm if caspase-11 and 3
are indeed sequentially activated during astrogliosis.
Potential mechanisms of caspase function in mediating astrogliosis
Execution of apoptosis by caspases involves cleavage of various substrates by effector
caspases leading to transcription and protein synthesis. Nuclear translocation of the
common effector caspase-3 is indicative of its proteolytic activity and associated
transcriptional changes occurring during apoptosis (Kamada, Kikkawa et al. 2005).
Certain studies have demonstrated activated caspases to have nuclear localization, even
under non-apoptotic circumstances (Noyan-Ashraf, Brandizzi et al. 2005), (Dash, Blum
et al. 2000), (Yan, Najbauer et al. 2001; Oomman, Finckbone et al. 2004) thus implying
that caspase function might involve substrate cleavage and transcriptional changes even
during non-apoptotic events. Caspase cleavage of substrates is known to underlie
several non-apoptotic functions. Differentiation of erythrocytes and muscle cells
involves non-apoptotic caspase-mediated cleavage and activation of transcription
factors specific to the cell type (De_Maria, Zeuner et al. 1999; Fernando, Kelly et al.
2002). Several of the transcription factors and signaling molecules activated in reactive
astrocytes are known caspase substrates and hence could be potential down-stream
90
effectors of caspase activation in reactive astrocytes. For example, common
transcription factors like CREB and NF-κ B activated in reactive astrocytes (de Freitas,
Spohr et al. 2002; Nikaido, Iseki et al. 2002) are both known caspase-substrates
(Fran∂cois, Godinho et al. 2000; Qin, Wang et al. 2000). NF-κ B, while earlier thought
to be inactivated upon caspase cleavage is now known to be activated even by low
levels of caspase activation in turn leading to activation of inflammatory pathways
(Lamkanfi, Declercq et al. 2006). Proliferation of reactive astrocytes, following a
cortical stab wound is known to be associated with a down-regulation of the cell-cycle
arrest protein p27
kip1
(Koguchi, Nakatsuji et al. 2002). Caspase-mediated cleavage and
inactivation of p27
kip1
is also known to be essential for proliferation of transformed
lymphoid cell lines (Frost, Al-Mehairi et al. 2001). Whether caspases mediates various
aspects of astrogliosis via these or other signaling molecules remains to be seen and
forms an important topic for further investigation.
Other signaling molecules implicated in astrogliosis like ERK/MAPK
(Mandell and VandenBerg 1999; Mandell, Gocan et al. 2001) and NF-kappa B
(Perez_Otano, McMillian et al. 1996) freitas et.al,2002) also have diverse functions in
the brain, like differentiation, proliferation, and mediating apoptosis or cell survival
pathways depending on the cellular context. As recent studies suggest, so can caspases.
Hence, our study suggests a novel non-apoptotic function for caspase activation
following brain injury where caspases might be involved in mediating astrogliosis along
with their known function of executing apoptosis. This idea of performing opposing
functions agrees with reports of other signaling cascades such as p38 MAPK commonly
91
activated in the injured brain and contributing to neuronal apoptosis, as well as
astrogliosis. (Che, Yu et al. 2001),(Che, Piao et al. 2001). Although further studies are
needed to evaluate the precise role of caspases, our study opens an exciting avenue of
research into the molecular basis of astrogliosis.
92
Chapter 3:
Functional significance for the caspase activation in reactive astrocytes with
respect to neuronal recovery
Abstract
The previous studies indicate a role for caspases in the molecular mechanisms
regulating astrocyte reactivity. Several of the signaling molecules implicated in
astrogliosis, also contribute to the consequences of reactive astrocytes on neuronal
recovery following injury. Hence the current study investigates the potential
involvement of caspases in regulating the reactive astrocyte-mediated effects on
neurite growth and neuronal survival. Primary hippocampal neurons were plated
onto adult astrocyte cultures generated from rats injected intrahippocampally with
either kainate (KA astrocytes) or saline (sham astrocytes) injections. Co-cultures
were immuno-stained after 3days, with the neurite marker NeuN and assessed for
neurite growth and neuronal survival. Neurite growth, as assessed by neurite length
and number of neurites per neuron, was significantly hampered on KA astrocytes,
which displayed reactive features in culture, as compared to sham astrocytes. There
was no difference in neuronal survival on either culture. Caspase inhibition, upon
treatment with the pharmacological inhibitor zVAD (50µM), significantly improved
neurite growth on the KA astrocytes. Neurons displayed increased susceptibility to
3-nitropropionic acid, in presence of astrocytes from kainate lesioned animals as
compared to when cultured with astrocytes from sham lesioned animals. However,
93
this increase in neuron loss was not ameliorated when caspase activation was
suppressed in KA astrocytes. Thus caspase activation in reactive astrocytes might
contribute to some of the associated functions.
94
Introduction
A characteristic feature of reactive astrocytes is the up-regulation of several different
groups of proteins. Assuming that the biological functions of reactive astrocytes are
reflected in these proteins they express, there are beneficial as well as detrimental
consequences (Eddleston and Mucke 1993; Giulian, Vaca et al. 1993; Ridet,
Malhotra et al. 1997). Studies have shown factors like type and duration of injury
(Heales, Lam et al. 2004) as well as age (Balasingam, Tejada-Berges et al. 1994;
Nakagawa, Schwartz et al. 2004) of the organism can influence the reactive response
of astrocytes and in turn their effect on the injured brain. There is evidence to
suggest reactive astrocytes might actually exacerabate the injury or disease
conditions by contributing to neurotoxicity and sustaining gliosis (Hu, Ferreira et al.
1997; Pâerez-Capote, Serratosa et al. 2004; Mander, Borutaite et al. 2005). Since
reactive astrocytes also play a major role in the recovery process following injury
(Ye, Sontheimer et al. 1998; Cui, Allen et al. 2001; Srebro and Dziobek 2001;
Swanson, Ying et al. 2004), it is of interest to retain the useful functions while
eliminating the deleterious effects. If there was a better knowledge of the signaling
cascades that drive astrogliosis it might be easier to understand when its
consequences would be beneficial or harmful.
One avenue of investigation is to analyze signaling molecules that are not only
important for the reactive response of the astrocytes but also influence these
functional consequences on neurons. There are several examples of such common
95
biochemical pathways. Activation of p38MAPK implicated in astrocyte reactivity,
might also be contributing to increased neuronal death via cytokine production (Kim
Yu et al. 2004). The NF-kappa B- Rel C signaling cascade is involved in the
inflammatory response of astrocytes made reactive by treatment with ß-amyloid, and
hence is implicated in the adverse influence of reactive astrocytes on neuronal
survival (Bales, Du et al. 1998). Inhibition of S100-ß, a calcium-activated multi-
functional cytokine secreted by reactive astrocytes, also known to induce reactivity
in astrocytes, resulted in not only improved neurite growth, but also neuronal
survival (Kato, Kurosaki et al. 2004). Certain signaling cascades implicated in
astrogliosis are also known to play a role in neurite growth and neuronal survival.
The Rho GTPase-ROCK pathway is implicated in astrogliosis induced via cytokine
treatment (John, Chen et al. 2004). Interestingly Rho GTPases are commonly used
second messengers by several of the neurite growth regulating molecules secreted by
reactive astrocytes (Sandvig, Berry et al. 2004). Another second messenger system
utilized by neurite growth regulatory proteins is protein kinase C (PKC), which is
also known to be involved in ATP and fibroblast growth factor-2 induced
astrogliosis. Thus regulating signaling molecules that are implicated in inducing the
reactive features in astrocytes, might also contribute to the effects of reactive
astrocytes on neuronal recovery.
Previous studies in our laboratory have demonstrated non-apoptotic caspase
activation in reactive astrocytes. Treatment with caspase inhibitors resulted in a
96
significant attenuation of the reactive parameters studied, thus suggesting a role for
caspases in the signaling cascade regulating astrocyte reactivity. Hence in the current
study we investigate whether caspase activation in reactive astrocytes has other
function besides being involved in up regulation of certain markers of astrocyte
reactivity. If that were the case, then inhibiting caspases in reactive astrocytes might
also influence parameters of neuronal function following damage to the brain.
Pharmacological inhibitors of caspases have been used to reduce neuronal death
following various forms of injury and are viewed as potential therapeutic agents
against brain trauma (Wiessner, Sauer et al. 2000; Graczyk 2002). Although the
caspase inhibitors might very well have a direct influence on neuronal recovery, it is
possible their protective effects are due to their role in mediating astrocyte reactivity.
The present study hopes to clarify whether indeed caspases can influence neuronal
recovery indirectly, via their role in reactive astrocytes.
97
Materials and Methods
Materials: 3-Nitropropionic acid, hydrogen peroxide, kainate were obtained from
Sigma-Aldrich (St.Louis, MO). Caspase inhibitor zVAD was purchased from MP
Biomedicals (Irvine, CA)
Adult astrocyte cultures: Adult astrocytes are cultured using the method of Schwartz
and Wilson (Schwartz and Wilson 1992). Hippocampi are dissected from adult
Sprague-Dawley female rats (90day), followed by enzymatic dissociation with
trypsin, at 37
o
C for 15 minutes. Dissociated cells are pelleted by centrifugation and
the pellet resuspended in DMEM/F12 supplemented with 20% fetal bovine serum.
The cells are plated into 50 cm2 flasks containing DMEM/20% FBS and placed in a
humidified incubator at 37
o
C with 5% CO2. The cells are fed with fresh medium
every three days and allowed to divide over a period of 10-14 days, till at least 15
separate colonies of astrocytes are detected. At this stage the cultures are passaged
once, and allowed to become confluent. After the cells reach confluency they are
subjected to overnight shaking at 37
o
C at 120 RPM on an orbital shaker, to eliminate
non-astroglial cells. Purified cultures are trypsinized then re-plated onto multi-well
poly-lysine coated plates in DMEM/F12+ 20%FBS. The cultures are shifted to
serum free medium (DMEM-F12) for a period of 1-3 days prior to use in
experiments.
98
Neuron-astrocyte co-cultures: Cortical neurons are generated from embryonic (E17)
Sprague-Dawley rats using a previously established method. Briefly, cortices are
dissected from the embryonic pups, and subjected to mechanical dissociation by
passaging through fire polished Pasteur pipets of increasingly narrow bores.
Dissociated cells are subjected to centrifugation and the pellet re-suspended in serum
free DMEM supplemented with N2, followed by further mechanical dissociation.
The cells are plated onto confluent astrocyte cultures at a density of 5000 cells/cm
3
for the neurite growth study and 25 cells/cm
3
for the neuroprotection study. The co-
cultures were maintained in serum free DMEM.
Immunocytochemistry: Immunostaining with the neurite marker Map1b or neuron
specific marker NeuN was performed 24-48 hours after co-cultures were established.
The cells were fixed using 4% paraformaldehyde and permeabilized using 0.01%
triton-X. After incubation with 3% BSA to block non-specific binding, the cells were
incubated with the primary Map1b antibody (1:5000, Sigma Aldrich) for 8 hours,
NeuN antibody (1:1000, Chemicon) for 4 hours. This was followed by sequential
incubation with anti-mouse peroxide-conjugated antibody and avidin-biotin-
peroxidase complex (ABC) each for 1 hour. The staining was visualized using 3,3’
diaminobenzidine (DAB) in the absence of nickel treatment.
Neurite growth assessment: Neurite growth was assessed by analyzing two
parameters 1) neurite length and 2) number of primary neurites 1) Neurite length: a)
99
IP Lab: neurite length was measured using an assay developed by Rozovsky et.al
(Rozovsky, Wei et al. 2005). Digital images of cultures immuno-stained with Map1b
were captured and immuno-positive cells counted in at least 20 images per condition.
Neurite growth was measured as the area covered by Map1b immuno-positive
neurites defined with a modified Sobel edge detection algorithm (IPLab,
Scanalytics). This is a semi-automated program using IPLab imaging software that
can define the edge profile of neurites without taking into account the cell bodies.
This filtered image is further skeletonized to a single pixel line, which represents
neurite length in arbitary units. This data was expressed as % control of map1b
immunopositive neurites. b) NIH Image: In order the ensure reliability of the results
obtained using the semi-automated IP lab software in our experimental system,
neurite length was also measured by manually tracing individual neurites using the
NIH image software. Per condition, at least 150 neurons were selected for
assessment, the criteria being: 1) cell body clearly visible in the field 2) cell bodies
are distinguishable and not part of a group of more than three. In order for a neurite
to be selected for measurement: 1) it had to be at least the length of the cell body 2)
the cell body from which is originated had to be visible in the field. Neurite length
was measured in micrometers. Total neurite length was the combined neurite length
of all the neurites of the selected neurite. Average neurite length, which is the total
neurite length divided by the number of primary neurites was also calculated. 2)
Number of primary neurites: Primary neurites were defined as the neurites
originating from the soma. The primary neurites were analyzed with the following
100
parameters: a) Number of neurites per neuron: This parameter was analyzed for each
condition by manually counting the number of neurites for each neuron in each field.
b) Neurite density per field: The neurons in each field were divided into those devoid
of neurites, those with a single primary neurite and those with two or more primary
neurites.
Neuron viability: Neuronal survival was assessed by manual cell counts of neurons
immuno-positive for NeuN, in neuron-astrocyte co-cultures under various
experimental conditions. For each experimental condition a total of at least 20 fields
across 3 wells, were analyzed. Data was statistically examined by ANOVA, followed
by between group comparisons using Fisher’s LSD test. Data are expressed as
average count of cells stained with NeuN, and represent at least four independent
experiments.
101
Results
Neurite growth is significantly reduced on astrocytes from kainate lesioned
animals
Reactive astrocytes can behave as a permissive or an inhibitory substratum for
neurite growth depending on different factors like type and duration of the injury as
well as the age of the organism. Hence before examining a potential function for
caspases in regulating neurite growth, it was necessary to analyze whether astrocytes
from the kainate lesion model inhibited or promoted neurite growth. Hippocampal
neurons form embryonic (E17) Sprague-dawley rat pups were plated on hippocampal
astrocytes from kainate or sham-leshioned animals (KA and sham astrocytes
respectively). The co-cultures were maintained for three days and later fixed using
4% paraformaldehyde, followed by immuno-staining for the neurite marker Map1b.
The neurite growth assay was performed on images of the co-cultures stained with
Map1b using the IPlab spectrum software. Figure 10 shows representative images of
co-cultures of neurons and sham (A) or KA (B) astrocytes. Significant reduction in
neurite growth was observed on KA astrocytes as compared to that on sham
astrocytes. Figure 10C is the quantification of neurite growth using the IPlab
spectrum software, which demonstrates an approximate 40% reduction in neurite
growth on KA astrocytes.
102
Figure 10
Figure 10: Co-cultures of hippocampal neurons with hippocampal astrocytes from
sham lesioned animals (A), and kainate animals (B), stained with Map1b after 3days
of growth. C, quantification of Map1b immunoreactivity, expressed as percentage of
sham control, * p < 0.005. Data represent five independent experiments (KA: N=5,
Sham:N=4)
103
The reduced neurite growth on the KA astrocytes can be partially improved by
caspase inhibition.
In order to better investigate the role of caspases, the time frame of neurite growth
analysis was reduced to 24 hours. The shortened time-point facilitated inhibition of
caspases using a pharmacological inhibitor zVAD. As observed in Figure 11A
displaying representative images of neuron-astrocyte co-cultures at 24 hour time
period, even at 24 hours, significant reduction in neurite growth was observed on KA
astrocytes as compared to that on sham astrocytes. Interestingly, neurons that were
plated on the KA astrocytes pre-treated with the caspase inhibitor, zVAD for 48
hours demonstrated an improvement in neurite growth. Figure 11B is quantification
of neurite growth on all conditions using the IPlab spectrum software, expressed as a
measure of Map1b immunoreactivity. Figure 11Cis a quantification of total neurite
length calculated by manual measurement of all neurites originating from a neuron.
Both measures of quantification demonstrate a reduction in neurite growth on KA
astrocytes at 24 hours, which is significantly improved with the zVAD treatment.
There is no significant effect of zVAD on neurite growth on the sham astrocytes.
104
Figure 11: Assessment of neurite outgrowth at 24 hrs, in neuron co-cultured with
astrocytes from kainate and sham lesioned animals. A displays representative images of
co-cultures, with or without zVAD pre-treatment, immunostained with neurite marker
Map1b. B: quantification of Map1b imunoreactivity; C: total neurite length
measurement; D: neurite number analysis and E: average neurite length measurement.
Data represent five independent experiments (Sham: N=3; KA: N=5; *p<0.001, #
p<0.05)
105
Figure 11
106
The number of neurites per neuron is also reduced on reactive (KA) astrocytes in a
caspase-dependent manner
Neurite growth can be influenced by length of neurite extension as well as number of
neurites originating from a neuron. To analyze the affect on number of neurites,
primary neurites were counted for each neuron, in each condition. As seen in figure
11D the average number of neurites per neuron was reduced to less than one on KA
astrocytes thus demonstrating that there were a significant number of neurons
without neurites. Treatment with zVAD resulted in a significant increase in the
average number of neurites per neuron thus indicating more neurons now possessed
neurites.
This effect on the neurite number was also reflected in the neurite length
analysis, when we estimated the average neurite length. Figure 11E shows
quantification of the average neurite length which is total neurite length divided by
the number of neurites possessed by that neuron. Since number of neurites is reduced
on KA astrocytes the average neurite length is also shortened. The improvement in
neurite number observed due to the caspase inhibition is also accompanied by an
increase in the average neurite length.
Neuronal survival is not significantly reduced on KA astrocytes
Since reactive astrocytes are known to secrete both neurotoxic as well as neuronal
survival promoting factors, it was necessary to analyze how KA astrocytes affected
neuronal survival. This study was also important in order to ensure that the reduced
107
neurite growth observed was not due to a reduced neuronal survival. Cell counts
were performed on co-cultures from the 3-day neurite growth experiment. Map1b
also stains cell bodies along with neurites thus facilitating the analysis of cell
viability as well. As seen in Figure 12 there is no significant difference in neuron
survival on KA astrocytes as compared to sham astrocytes. Treatment with zVAD
did contribute to any effect on neuronal survival.
Figure 12
Figure 12: Three day, neuron-astrocyte co-cultures, with or without zVAD treatment,
immunostained with neurite marker Map1b, and cell bodies counted. Mean count of
neuron cell bodies expressed as percent of sham control, data represent six
independent experiments
Neuronal survival on reactive astrocytes in presence of a toxin
It is possible that by virtue of the factors they secrete, even though KA astrocytes by
themselves do not affect neuronal survival their presence makes neurons more
susceptible to other toxins. To investigate this premise we used two insults 3-
nitropropionic acid (3-NP) and hydrogen peroxide (H
2
O
2
), which cause a caspase-
108
independent cell death. The toxic doses of both 3-NP and H
2
O
2
, required to kill
astrocytes are ten-fold higher, thus allowing treatment of the co-cultures with these
toxins without risking astrocyte damage. Co-cultures were maintained for 48 hours
and then treated with increasing concentrations of the 3-NP or H
2
O
2
. If neurons were
indeed more vulnerable to these toxins in presence of KA astrocytes as compared to
when cultured with sham astrocytes we would expect significant death at lower
doses of the toxins. As seen in figure 13A there was a non-significant trend seen
towards increased neuronal death on KA astrocytes compared to that on sham
astrocytes when co-cultures were treated with hydrogen peroxide but only at higher
doses of the toxin. However with 3-NP treatment, there was a significant shift in the
dose-response curve on KA astrocytes as compared to the sham astrocytes. As seen
in figure 13B significant neuron death is seen at the 3-NP dose of 0.5mM on KA
astrocytes as compared to 1 mM on sham astrocytes. Figure 13C shows
representative images of co-cultures treated with different doses of 3-NP. As seen in
the images, at 1mM 3-NP even though neuron death is significant on both KA and
sham astrocytes, it is significantly higher on KA astrocytes. By 5mM 3-NP almost no
neurons survive on KA astrocytes whereas there is still a significant number of live
neurons on the sham astrocytes.
109
Figure 13
Figure 13: Figure 4: A, astrocyte-neuron co-cultures treated with 3-NP, immuno-
stained with live neuron marker NeuN. B, C, representative cell counts of live
neurons surviving on astrocytes from sham versus kainate lesioned animals
following treatment with 3-NP and H
2
O
2
respectively. Data represent at least three
independent experiments (* , # p <0.001)
110
Effect of caspase inhibition on neuronal survival on reactive astrocytes in presence
of a toxin
We observed neurons were more vulnerable to cell death following treatment with 3-
NP in presence of KA astrocytes as compared to sham astrocytes. If activated
caspases in reactive astrocytes were contributing to this effect, then inhibiting them
would influence the neuron susceptibility to the toxin. Astrocyte cultures were
treated with the caspase inhibitor zVAD for 24 hours followed by plating of neurons.
The co-cultures were maintained for 24 hours during which zVAD was added back
into the medium. At, 24 hrs the co-cultures were treated with toxin. As a control
some wells were pre-treated with zVAD only for 2 hours followed by treatment with
the toxin. This was to ensure there was no direct protective effect of zVAD and any
effect observed was via astrocytes. We found no significant effect of caspase
inhibition on the neuronal response to toxins. Figure 14A and B are representative
experiments of hippocampal neurons co-cultured with sham or KA astrocytes
respectively.
111
Figure 14
Figure 14: A, representative cell count of live neurons growing on astrocytes from
sham versus kainate lesioned animals, pretreated with zVAD. B, representative neuron
cell count under various conditions, following treatment with 3-NP (1mM, 24hr). Data
represent 5 independent experiments (* p < 0.001, ns = non-significant trend).
112
Discussion
Summary of results
Previous findings in our lab suggest a caspase imvolvement in the signaling cascade
governing astrocyte reactivity. The present study investigates the effect of
suppressing astrocyte reactivity using the caspase inhibitor zVAD, on neuronal
recovery following kainate excitotoxicity. We found neurite growth is hampered on
astrocytes from kainate-lesioned animals (KA astrocytes) as compared to those from
sham-lesioned animals (sham astrocytes). Caspase inhibition resulted in significant
improvement in neurite growth on KA astrocytes. Neuronal survival was not
significantly affected on KA astrocytes as compared to sham astrocytes. However,
neurons were more susceptible to toxins in presence of KA astrocytes, although it
did not seem to be mediated by caspase activation in reactive astrocytes
Neurite growth on astrocytes from kainate lesioned animals is stunted.
Astrocytic influence on neurite growth is a result of the combination of proteins
expressed by them, which in turn is dictated by factors like type and duration of
injury, age of the organism and even region of the brain. Various in vivo studies have
looked at growth promoting and growth-retarding factors expressed in reactive
astrocytes following kainate excitotoxicity. Adult rats after KA injury express lower
levels of certain neurotrophins BDNF, NGF and NT-3 (Shetty, Rao et al. 2004). KA
induced seizures cause prolonged changes in the expression of neurite growth
113
inhibitory proteoglycan molecules like neurocan and phospocan, (Okamoto,
Sakiyama et al. 2003) and tenascin, (Nakic, Mitrovic et al. 1996; Brenneke,
Schachner et al. 2004). In vitro co-cultures of neurons and hippocampal astrocytes
treated with kainate are also associated with decreased adhesiveness of astrocytes
(Mahler, Ben-Ari et al. 1997). Astrocytes from in vivo injury models are known to
retain in culture increased gene expression associated with reactivity (Wu,
Nishiyama et al. 1998; Rozovsky, Wei et al. 2005). Glutamine synthetase and
fibroblast growth factor-2 are two markers of reactive astrocytes also up regulated
following KA injury (Ong, Leong et al. 1996) (Ballabriga, Pozas et al. 1997) and as
demonstrated in our previous studies retained in culture. Thus astrocytes from KA-
lesioned animals in our study could be expressing a subset of neurite growth
regulating molecules similar to that observed in various in vivo injury studies
making them a less permissive substrate even in culture. Further studies would be
required to analyze which neurite-growth regulating molecules are retained in culture
by these astrocytes.
Caspase-mediated effect on neurite growth may not be through FGF-2
Our previous studies indicated that the increased expression of FGF-2; a potent
neuritotrophic factor, in reactive astrocytes could be controlled by a caspase-
mediated signaling pathway. Hence we predicted caspase inhibition, which resulted
in a partial attenuation of the up regulated FGF-2 expression, could make reactive
astrocytes even less permissive to neurite growth. Instead our results indicate a
114
completely opposite effect, a significant improvement in neurite growth upon
caspase inhibition. Although, fibroblast growth factor-2 is known to have robust
neurite-growth promoting properties as demonstrated by several studies, there is
ample evidence to suggest it might also function as an inhibitory molecule.
Exogenous treatment with FGF-2 might have a potent beneficial effect on neurite
growth, but the indirect effect via astrocytes might be more complicated. FGF-2 is
known to up regulate the expression of neurite growth inhibitory protein tenascin in
astrocytes (Meiners, Marone et al. 1993) as well as inhibit signaling molecules used
by neurite growth promoting cell adhesion molecules (Williams, Mittal et al. 1995).
The beneficial effects of FGF-2 could also be antagonized by pro-inflammatory
cytokines like interferon-gamma (DiProspero, Meiners et al. 1997) secreted by
reactive astrocytes. Thus although FGF-2 is an important trophic factor, in our
experimental system it may not be contributing to neurite growth regulation or could
have a more indirect mode of action.
Caspases could be exerting their effects on neurite growth by directly targeting
the growth-regulating molecules or by interacting with other astrocytic factors
contributing to the synthesis of these growth inhibitory molecules. Several inhibitory
molecules including members of the CSPG family, ephrins and ECM proteins are
known to activate the caspase cascade, but there is no evidence to suggest a converse
caspase-mediated regulation of synthesis of these factors. Another potential caspase
substrate that might contribute to observed effects on neurite growth is the
transcription factor NF-κB. Recent study reported suppressing NF-κB specifically in
115
reactive astrocytes reduced production of two neurite growth inhibitory molecules of
the CSPG family, neurocan and phosphacan (Brambilla et.al, 2005). Glial fibrillary
acidic protein (GFAP) is another important astrocytic protein that inhibits neurite
growth (McKeon, Schreiber et al. 1991; Menet, Gimenez_Y_Ribotta et al. 2000) and
is also known to be cleaved by caspases (Mouser, Head et al. 2006). The idea of
caspase-mediated cleavage of cytoskeletal proteins regulating neurite growth is
supported by in vitro studies using retinal explant cultures (Gilman and Mattson,
2002). Unfortunately, these studies fail to address the possibility that the caspase-
mediated effect on neurite growth could be due to the presence of astrocytes in the
explant cultures and not as a result of direct influence on neurons. The
pharmacological broad-spectrum caspase inhibitor used in our study zVAD has been
reported to also inhibit the calcium activated protease calpain in some instances
(Bizat, Galas et al. 2005). Calpain is also known to play an important role in neurite
growth, and given the cross-talk between the two protease systems (Nakagawa, Yuan
et al. 2002) we cannot rule the possibility that our data might reflect a calpain
activation. Further studies are needed to address this issue. Whether these or other
yet unknown factors execute the caspase-mediated effect on neurite growth remains
to be investigated and forms an important topic of future research.
Effect of KA astrocytes on neuronal survival may be toxin-dependent
Astrocytes can improve neuron survival and tolerance to toxins, however their
reactive state can be detrimental. Studies have shown reactive astrocytes to either
116
have direct neurotoxic effects or increase neuronal vulnerability to further insult.
Astrocytes made reactive using β-amyloid (Paradisi, Sacchetti et al. 2004) or the pro-
inflammatory cytokine IL-1 beta (Deshpande Zheng et al. 2005) directly caused
increased neuronal death. On the other hand glial cultures made reactive with
lipopolysacchride or inflammatory cytokines were not directly neurotoxic but
increased neuronal susceptibility to glutamate (Pâerez-Capote, Serratosa et al. 2004)
or hypoxia (Mander, Borutaite et al. 2005). Our results indicated KA astrocytes,
which were reactive in culture, although not neurotoxic directly, increased neuronal
susceptibility to 3-NP. The extent of time that the astrocytes have been in the
reactive stage could be one of the reasons why under some circumstances they cause
neuron death, but otherwise merely prime the neurons to be more vulnerable to other
insults. Even in our study the neuron loss on KA astrocytes was much lower when
co-cultures were maintained only for 24 hours (Figure 5B) as compared to 48 hours
(figure 4B) and exposed to the same dose of the toxin.
Increased neuron susceptibility to 3-NP may not be caspase-dependent
Caspase inhibition in KA astrocytes did not have any effect on neuron susceptibility
to 3-NP. It is possible we did not see an effect because the neuron death was not
sufficient at the time of analysis and dose of 3-NP selected. Co-cultures for these
experiments were maintained only for 24 hours, which might not be sufficient time
for astrocytes to exert their full effect on neuronal survival. Alternately it is possible
caspase activation in KA astrocytes may not contribute to the increased neuron loss
117
upon 3-NP treatment. However, the experimental design used in our study to
investigate astrocyte-specific caspase-mediated neuronal survival, had certain
restrictions that might have prevented correct interpretation. In order to observe the
effects of caspase inhibition using zVAD, the neuron-astrocyte co-cultures were
maintained only for a short time prior to treatment with the toxin. Hence the resulting
neuron death probably wasn’t extensive enough to trigger a caspase-mediated
protective mechanism. Transfection of astrocyte cultures with viral caspase
inhibitors might allow prolonged caspase inhibition in reactive astrocytes, in turn
permitting the astrocyte-neuron co-cultures to be maintained for longer time periods
required to cause sufficient neuron death. Such an experimental design might be help
reach a more conclusive decision about caspase involvement in neuron survival
pathways regulated by reactive astrocytes.
In conclusion we observed that neurite growth inhibiton on reactive astrocytes
following kainate exitotoxicity could be at least partially dependent on caspase
activation. Neuronal survival, which is another parameter of neuronal recovery, was
not influenced by caspase activation. Further studies are needed to elucidate the
precise molecular pathways executing caspase-mediated regulation of neurite
growth.
118
Chapter 4
Conclusion and future directions
Summary of results: The work included within this thesis demonstrates for the first
time a non-apoptotic role for caspase activation in reactive astrocytes, possibly as a
part of the signaling machinery governing the process of astrogliosis. The fact that
we could achieve similar results in two disparate models of astrogliosis seems to
suggest non-apoptotic caspase activation in reactive astrocytes may not be entirely
injury dependent. However, it may be of interest to examine other injury models to
make this statement with more conviction. Although caspase inhibition significantly
attenuated the increased expression of the markers of astrocyte reactivity, it had no
effect on the stellate morphology. Thus, suggesting caspases might preferentially
regulate the biochemical changes associated with reactivity. Preliminary
investigation of the caspase family revealed caspases 11 and 3 but not caspase 1, 6, 8
or 9 might be the ones contributing to astrocyte reactivity. Caspase activity in
reactive astrocytes compared to that in astrocytes treated with apoptotic stimulus
staurosporine although significant is several fold lower. This sub-threshold caspase
might be a possible mechanism of action underlying its non-apoptotic functions in
astrogliosis. However, further studies need to be done to examine this premise in
more detail. Analyzing the impact of caspase activation in reactive astrocytes on the
functional consequences on neurons revealed a possible function in regulation of
119
neurite extension. However, no protective effect of caspase inhibition was seen on
neuronal survival or the increased susceptibility of neurons to a toxin like 3-
nitropropionic acid. Thus, caspase function in reactive astrocytes might not be solely
playing a part in driving the reactive response, but might also contribute to some of
the functions attributed to reactive astrocytes.
Possible reasons for caspase activation to be non-apoptotic in reactive astrocytes
In Chapter 2, we demonstrated the activation of caspases in reactive astrocytes in the
absence of cell death. As discussed in the introduction caspase activation in absence
of cell death, has now been demonstrated in the various stages of the ‘life-cycle’ of
an astrocyte; during development, in proliferating astrocytes and in those with a
reactive phenotype under brain injury conditions. Why non-apoptotic caspase
activation is observed in reactive astrocytes and yet upon exposure to certain stimuli
apoptotic death can be induced, forms an interesting question.
As discussed in the introduction several non-apoptotic functions attributed to
caspases are associated with a sub-lethal level of activation. Similarly caspase
activation observed in reactive astrocytes, although significant is also sub-lethal.
Hence the level of activation might be one possible mechanism that decides when
caspase have non-apoptotic versus apoptotic consequences in astrocytes. One can
speculate as to why a sub-lethal activation of caspases occurs in reactive astrocytes.
A posssible theory is caspase activation in reactive astrocytes reflects incomplete
120
apoptosis, prevented from being completed due to a yet unknown astrocytic factor.
Another possible reason for failure of apoptosis could be the specific activation of a
‘reactive caspase cascade’ but not other apoptotic initiator caspases. Preliminary
analysis of the caspase members suggests caspase activity in reactive astrocytes
reflects activation of caspases 11 and 3, but not other initiator caspases. Such an
inherent inability to activate classical caspase cascades, might also explain why the
initial astrocytic response to injury is to become reactive but not undergo apoptosis,
making astrocytes a more resiliant cell type. There is some evidence to support an
initial astrocytic resistance to activation of apoptotic cascades. For example although
astrocytes express the death receptors Fas / CD95 known to activate the death-
receptor mediated apoptotic pathway, they are normally resistant to Fas-mediated
death when exposed the Fas ligand (Saas, Boucrat et al. 1999). Since apoptotic
caspase activation and subsequent cell death can easily be induced in astrocytes by a
wide variety of stimuli, there must be a tightly regulated mechanism that keeps
caspase activation low in reactive astrocytes steering it towards non-apoptotic
functions. Since caspase function in astrogliosis is not merely prevention of
apoptosis, but as our results suggest, a part of the signaling cascade governing
astrocyte reactivity. Exploring factors capable of restricting of caspase activity at a
sub-lethal level without complete inhibition that might regulate caspase activation
should be.
Putative candidates that can perform such a fine tuned regulation of caspase
function include endogenous caspase inhibitors. Members of the inhibitor of
121
apoptosis protein (IAP) family can bind to caspases inhibiting their enzymatic
activity and can also target them for degradation. Cellular Flice inhibitory proteins or
cFLIPs are caspase homologues and endogenous inhibitors. A recent study found
upregulation of cFLIP-delta in a subset of reactive astrocytes, 48 to 72 hours
following traumatic injury in rat brains, as well as in brain tissue from human
epilepsy patients (Hainsworth, Bermpohl et al. 2005). Interestingly cFLIP being a
caspase-8 homologue competes with caspase-8 for binding to the adaptor protein
FADD that triggers the death-receptor mediated apoptotic pathway, resulting in its
inhibition. Survivin, a member of the inhibitor of apoptosis (IAP) protein family is
another endogenous caspase inhibitor that was recently found to have an increased
expression in reactive astrocytes, following cortical stab injury (Johnson, Svetlov et
al. 2004). Investigating a possible role of survivin or any other IAP, or caspase
homologues in regulating caspase activity in reactive astrocytes forms an interesting
question needing further investigation. Isoforms of certain caspases derived from
alternate splicing have limited or no function and hence can also be considered
potential candidates to restrict level of caspase activation. However, currently there
is no known evidence of the expresson of such splice variants in astrocytes.
As discussed in the introduction, other possible mechanisms, which can
restrict caspase activity is sequestration by chaperone proteins or enzymatic post-
translational modifications. Reactive astrocytes are known to up-regulate several
members of the heat shock protein family. Some are known to cleave and inhibit
caspase activity even downstream of caspase-3 activation. Another putative
122
mechanism could be nitrosylation of caspases by nitric oxide released from reactive
astrocytes. A consistent release of NO is commonly observed in reactive astrocytes,
and while being non-toxic to astrocytes themselves is known to contribute to the
death of neurons in the vicinity (Hu, Ferreira et al 1997; Hu, Akama et al. 1998).
Thus, might serve as a regulatory mechanism in reactive astrocytes to restrict caspase
activation. An overlap of caspase-3 and iNOS, observed in astrocytes with a reactive
phenotype following kainate excitotoxicity (Acarin, Peluffo et al. 2005), further
lends support to the idea that caspase activation might not be completely inhibited by
NO, but merely curbed.
The analysis of non-apoptotic caspase function in lens fiber differentiation,
revealed a novel mechanism of action, which the authors termed apoptosis-related
Bcl-2 and caspase dependent (ABC) differentiation (Weber, Menko et al. 2005). This
process involves the initiation of the mitochondrial apoptotic cascade along with
concurrent activation of the anti-apoptotic bcl-2 and IAPs, which in turn prevent the
caspase activation from reaching apoptotic levels. IAPs are thus likely to play an
important regulatory role during non-apoptotic cellular events holding the caspase
activation at a sub-lethal level, enough to allow them to execute important functions,
but preventing apoptosis from occurring Although the process of lens fiber
differentiation is not similar to astrogliosis, such concurrent activation of cell
survival pathways can be a potential regulatory mechanism. Fibroblast growth
factor-2 (FGF-2), which induces the expression of bcl-2 as well as IAPs during lens
fiber differentiation, is a known marker of astrocyte reactivity. Interestingly, our
123
results suggest a possible regulation of FGF-2 expression by caspases. I speculate a
putative mechanism whereby caspase-mediated FGF-2 activation in turn induces bcl-
2 related cell survival pathways that might contribute to restricting the caspase
activation to a sub-lethal level.
Sub-threshold caspase activation in reactive astrocytes is a mechanism of
eliminating highly inflammed astrocytes.
Caspase activation has been observed in degenerating astrocytes, following damage
to the brain. Thus, reactive astrocytes either eventually undergo apoptosis or this
response is restricted to a sub-population of astrocytes. In a situation of chronic brain
injury, the prolonged astrocyte reactivity is usually associated with increased
cytokine production and free radicals, contributing to increased neuronal death. In
such a situation the sub-threshold caspase activation might be a protective
mechanism to eliminate highly reactive and potentially neurotoxic astrocytes. The
low level of caspase activity could function as a feed back loop amplifying the
caspase cascade, pushing the reactive astrocyte towards apoptosis sooner. The
amplified caspase pathway could be the one implicated in astrogliosis and / or other
upstream caspases resulting in enough accumulation of the active effector caspase-3
leading to astrocyte death. There is evidence to suggest that such mechanisms that
allow astrocytes to switch from a highly inflammatory state to apoptosis do exist.
Astrocytes usually resistant to Fas-mediated apoptosis despite possessing the
required receptors, upon exposure to certain stimuli like pro-inflammatory cytokines
124
become responsive to Fas and undergo apoptosis. Downstream effectors of caspase
activation in reactive astrocytes might serve as potential amplifiers of the caspase
cascade. Caspase-3 substrates include several members of the bcl-2 family as well as
other apoptotic regulators, which can serve as potential candidates contributing to a
positive feedback amplification loop of apoptotic signals. Bim, a pro-apoptotic
member of the bcl-2 family is subjected to caspase cleavage at an early stage of
apoptosis induced by stimuli that activate death receptor or mitochondria activated
apoptotic pathways. The cleaved form of Bim is very effective in amplifying the
mitochondria driven apoptotic pathway (Chen, Zhou et al. 2004). Caspase-3
mediated cleavage of bcl-2 during genotoxic stress-induced apoptosis, can in turn
induce further release of cytochrome c from the mitochondria, resulting in
amplification of the cascade (Chen Gong, Almasan et.al, 2000). Whether caspase
activation in reactive astrocytes is indeed a mechanism to tag them for degeneration
thus forms an interesting avenue of investigation. A potential study to test this
hypothesis would be to compare the susceptibility of reactive versus non-reactive
astrocytes to additional stresses. If indeed the low level of caspase activation in
reactive astrocytes primed them for a quicker death, they would likely be more
vulnerable to further insults and succumb to cell death at lower doses of the toxin as
compared to non-reactive astrocytes.
125
Mediators of non-apoptotic caspase function in astrogliosis
Our findings suggest caspase activation in reactive astrocytes is not just a passive
occurrence but plays a role in the signaling mechanism underlying astrogliosis. My
data suggest sub-lethal caspase activation level might be the putative mechanism
underlying the non-apoptotic caspase functions associated with astrogliosis.
However, the precise astrocytic factors and associated signaling pathways that are
targeted by this sub-lethal caspase activation to regulate astrocyte reactivity remain
to be elucidated. The critical influence of caspases on cell survival is reflected in vast
number of signaling molecules that are caspase substrates. Apoptosis is executed
through caspase-mediated cleavage of several different substrates leading to either
their activation or inactivation (Review: Fischer, Janicke 2003). As newer studies
emerge it becomes evident that, even non-apoptotic caspase functions might involve
similar cleavage-mediated regulation of multiple downstream effectors. For example,
caspases are known to mediate erythrocyte differentiation via cleavage of
transcription factors GATA-1 (DeMaria, Zeuner et al, 1999) and SCL/Tal-1 (Zeuner,
Eramo et al. 2003). Multiple caspases as well as downstream effectors calpastatin
(Barnoy and Kosower 2003) and the transcription factor MST-1 (Fernando, Kelly et
al. 2002) are implicated in myoblast differentiation. Our results demonstrate an
attenuating effect of caspase inhibition on two very disparate markers of astrogliosis,
glutamine synthetase and fibroblast growth factor-2. It is hence reasonable to
suppose caspases influence astrocyte reactivity by targeting one or more downstream
126
transcription factor or signaling pathway capable of regulating the expression of a
variety of genes. Our data indicate caspase activity in astrocytes only after a robust
reactive response is observed. Hence, the most likely candidates to execute caspase
function in astrogliosis would be those signaling molecules that are activated after
the initial reactive response have been induced in astrocytes and can thus work
downstream in the caspase-mediated signaling cascade.
As seen in table 4, among known caspase-3 substrates are, several
transcription factors and signaling molecules whose activation is also observed in
reactive astrocytes. Although, most of these caspase substrates have been studied in
an apoptotic context, there could be other consequences of their cleavage when
caspase activation is low. Sub-threshold caspase activation is known to mediate non-
apoptotic functions either because it induces a different cleavage pattern of substrates
or it might target a different subset of substrates not usually cleaved during
apoptosis.
127
Table 4: Potential mediators of caspase function in in astrogliosis
Caspase
substrate
Physiological
function
Caspase-
cleavage
effect
Reference Proposed
function of
substrate in
astrogliosis
Transcription factors
inactivation
Francois et.al,
2000
CREB Cell survival,
apoptosis
activation
Lee et,al
2003
Proliferation
Glial scar
formation
inactivation
Ravi et.al
1998, Levkau,
1999
NF-κB immune
response,
Cell survival
activation
Lamkanfi
et.al, 2006
Inflammatory
response,
neurite
growth
inhibition
STAT1 Cell survival
differentiation
inactivation
Kinag et.al,
1998
GFAP
expression
STAT 3 Cell survival
differentiation
inactivation,
activation
Darnowski
et.al, 2006
GFAP
expression,
stellation
Signaling molecules
fyn Tyrosine protein
kinase
activated
Luciano et.al,
2001
Contributes
to neuronal
damage
ROCK-1 Rho-associated
kinase
activated
Coleman et.al,
2001
Inhibition of
stellate
morphology
calcineurin Calmodulin-
dependent
phosphatase
activated
Mukerjee
et.al,
2000,2001
Hypertrophy
cPLA(2)α Cytosolic
phospholipase
A2
inactivated
Adam-Klages
et.al, 1998
Specific
function not
proposed
The transcription factor NF-κB, can probably be considered the most promising
candidate for undertaking caspase-mediated functions in astrogliosis. NF-κB, is
involved in various cellular functions like proliferation, differentiation, apoptosis,
128
immune response and carcinogenesis. Delayed activation of NF-κB has been
associated with astrocyte reactivity triggered by a variety of injury-induced factors
(Pennypacker Kassad et al. 1999). Furthermore, suppressing NF-κB specifically in
reactive astrocytes was found to reduce production of pro-inflammatory chemokines
as well as neurite growth inhibitory molecules neurocan and phosphacan (Brambilla,
Bracchi-Ricard, et al. 2005). Caspases cleave and inactivate NF-κB expression
during apoptosis. However recent studies have revealed sub-lethal level of caspase
activation could up regulate NF-κB expression leading to non-apoptotic
consequences. Non-apoptotic activation of caspase-8 is involved in T-cell
proliferation, and further requires the cleavage of NF-κB. However, there is evidence
to show caspase-8 in this situation is not processed and hence is less active (Su,
Bidere, Zheng et.al, 2005), as compared to during apoptosis. NF-κB activation can
also be initiated by a limited activation of effecter caspases. Caspase-3 mediated
cleavage of poly (ADP-ribose) polymerase-1 is a characteristic feature of apoptosis.
Recent evidence from mice expressing caspase-resistant PARP suggests caspase-3
cleavage of PARP might be essential for NF-κB-mediated gene transcription
(Petrilli, Herceg, Hassa, et.al, 2004). Thus, considering its functional importance in
astrocyte reactivity, it might be worthwhile to investigate the caspase-mediated
cleavage of NF-κB as one of the putative effectors of non-apoptotic caspase function
in astrogliosis.
CREB, or cAMP response element binding protein is another important
transcription factor regulating gene expression in response to diverse stimuli. Target
129
genes of CREB are those possessing the cAMP recognition elements or CREs and
include early response genes like c-fos, and AP-1. These early response genes can in
turn activate various different genes including factors like FGF-2 (Butyaert et.al,
2001) a potent mitogen and important in astrogliosis. Caspase-mediated cleavage can
both activate as well as inactivate CREB depending on the cellular context, and type
of apoptotic stimuli. If caspase-mediated cleavage of CREB occurs during
astrogliosis, one can speculate that it might result in its activation, since CREB
activation in reactive astrocytes can be sustained for upto 4 weeks post injury.
STAT-1, signal transducer and activator of transcription-1 is another signaling
molecule capable of regulating expression of various genes relevant to astrogliosis.
However, its activation occurs soon after brain injury and thus seems to be
associated with induction of early reactive response in astrocytes. Hence even though
it is a known caspase substrate, it may not be a downstream factor in caspase-
mediated signaling involved in astrogliosis. On the other hand, STAT-3 another
member of the STAT family, has multiple caspase cleavage sites. Which cleavage
site is targeted by caspases depends on the level of caspase activity. Interestingly,
cleavage fragments of STAT3 generated from low level of caspase activity
specifically contribute to non-apoptotic functions (Darnowski, Goulette et al. 2006).
Apoptosis is highly regulated by caspase-mediated proteolysis of various signaling
molecules. Some of these, as listed in Table 4 are relevant to astrocyte reactivity. The
consequences of caspase-mediated cleavage of these molecules can depend on the
cellular context. Furthermore the downstream factors activated by caspase-cleavage
130
products of these signaling molecules are not completely known. Hence it is more
difficult to speculate how caspase-mediated activation of some of these signaling
molecules might be involved in regulation of astrogliosis.
Placing caspases in the signaling cascade underlying astrogliosis: a hypothesis
Considering the diversity of insults and toxins that can induce reactivity in
astrocytes, it is unlikely there is common signaling pathway that governs this
response. A more probable scenario is activation of a few signaling molecules being
common to the various stimuli capable of triggering astrocyte reactivity but not the
entire cascade. It is also feasible multiple signaling cascades are sequentially
activated, contributing to various aspects of astrocyte reactivity. Astrogliosis can be
perceived as a very dynamic process where the molecular profile of reactive
astrocytes can change depending on the type and duration of injury. Caspases are
widely activated in the injured brain and thus form good candidates as potential
signaling molecules common to astrogliosis associated with a variety of injuries.
I hypothesize a general signaling cascade, involving caspases that might
contribute to astrocyte reactivity. Depending on the injury, several factors capable of
inducing reactivity in astrocytes are released into the brain, either due to a
compromised blood brain barrier or from reactive microglia or dying neurons. These
include cytokines, extracellular ATP, β-amyloid and viral proteins, among others.
Several of these molecules can activate upstream signaling pathways like
ERK/MAPK, STAT/gp130 or the JAK/STAT pathways. There is evidence to show
131
activation of these pathways in reactive astrocytes occurs very early, and is transient,
prior to increase in GFAP expression, an essential marker of reactivity. These
proteins are capable of activating as well as inhibiting caspases. Our studies suggest
significant caspase activation occurs in reactive astrocytes only after there is a
pronounced reactivity, meaning they display a stellate morphology and there is a
several fold increase in the expression of the reactive markers glutamine synthetase
and fibroblast growth factor-2. It is hence possible ERK or other ‘initiator’ signaling
molecules activate the caspase cascade either directly or via other signaling
molecules that are first induced by the injury. Also, according to our results, once
reactivity is already induced in astrocyte cultures either by Aβ or dibutryl cAMP, the
reactive astrocyte cultures need to be treated with zVAD for at least 24 hours in
order to achieve caspase inhibition and see the attenuating effect on GS and FGF-2.
This observation suggests not the caspase cascade directly, but signaling events
occurring downstream of it, might be responsible for bringing about the reactive
changes in astrocytes. As discussed in the previous section caspases have several
known substrates, and some of them can be considered putative downstream
effectors of caspase function in regulating not only astrocyte reactivity, but also the
effects on neuron function. Concurrent activation of other signaling pathways might
serve to restrict the caspase activation at a sub-lethal level during the course of
astrogliosis. On the other hand, if the injury and hence astrocyte reactivity became
chronic, there could be other mechanisms that might de-repress the block on caspase
132
activity allowing it to reach apoptotic levels, thus eliminating a highly inflammatory
and hence potentially neurotoxic astrocyte.
Caspase involvement in astrogliosis: future directions of investigation
The observations arising from this dissertation open an exciting avenue of investigation
into the molecular mechanism governing astrogliosis. It establishes the importance of
caspase activation in astrogliosis. However, the extent of the relevance and functional
significance of this sub-threshold, non-apoptotic caspase activation in reactive
astrocytes is far from being understood. In my opinion, elucidating the caspase-
mediated signaling pathway involved in astrogliosis should be next direction of this
investigation. As discussed earlier, there are several caspase substrates known to be
activated in reactive astrocytes, which might function as putative executors of caspase
mediated signaling pathways. Another line of experimentation that might be crucial to
gaining a complete understanding of non-apoptotic caspase involvement in astrogliosis
is the significance of the level of caspase activation. The approach used by several
studies including the present one, was to compare the level of caspase activation during
non-apoptotic events, to that observed during apoptosis. This does not provide
conclusive evidence for a stringent requirement of a low level of activation for non-
apoptotic events to occur. It could also mean apoptosis is somehow being prevented or
delayed. A better approach would be to see if inducing low level of capsase activation
in astrocytes gave rise to reactive features. My initial attempts at this approach using
low levels the apoptotic insult staurosporine however gave disappointing results. This
133
could however be due to the specific activation of the mitochondrial caspase cascade by
staurosporine, which may not be involved in astrocyte reactivity. Microinjection of
active caspases, especially effector caspase-3 directly into astrocytes at various
concentrations might be a better method to use. If consequences of caspase activation
were indeed dependent on the level, then I would predict, reactive features would be
induced at low concentration, and as the level of concentration increased, the astrocyte
would get increasingly prone to apoptosis. Investigating this problem might thus also
provide an important mechanistic detail about non-apoptotic caspase function. Caspases
regulate cell proliferation and caspase activation has also been demonstrated in
proliferating lymphoid tumor cell-lines (review, Schwerk, Schutlz-Osthoff 2003),
giving rise to the premise that caspases are important in regulating cell number, and
might even be relevant to tumor formation. Non-apoptotic caspase activitation has been
observed in proliferating astrocytes (Zhu, Dahlstrom, 2005). Reactive astrocytes are
often found surrounding and sometimes interspersed within astroglial tumors, and share
common biomarkers including GFAP (Tan, Magdelene et al.2006). Expression of
certain proteins in both reactive and neoplastic astrocytes, has given rise to the idea that
the two phenotypes might share a common lineage (Nishigawa, Suzuki et al, 2002).
Thus, it might be interesting to investigate the relevance of caspase activity in
proliferating astrocytes using tissue from human astrocytomas or glioma cell lines. This
area of investigation might provide valuable clinical relevance to the findings from this
disseration.
134
Bibliography
Abd-el-Basset EM, Kalnins VI et al. (1989) “A 48 kilodalton ntermediate filament
associated protein (IFAP) in reactive-like astrocytes induced by dibutyryl
cyclic AMP in culture and in reactive astrocytes in situ”. J Neuropathol Exp
Neurol. 48(3): 245-54
Abe, K. and H. Saito (1997). "Developmental changes in cyclic AMP-stimulated
stellation of cultured rat cortical astrocytes." Japanese Journal of
Pharmacology 75(4): 433-8.
Abe, K. and H. Saito (1998). "Adenosine stimulates stellation of cultured rat cortical
astrocytes." Brain Research 804(1): 63-71.
Abe, K. and H. Saito (2000). "The p44/42 mitogen-activated protein kinase cascade
is involved in the induction and maintenance of astrocyte stellation mediated
by protein kinase C." Neuroscience Research 36(3): 251-7.
Abe, K. and H. Saito (2001). "L-glutamate suppresses astrocyte stellation induced by
actin breakdown in culture." Biological and Pharmaceutical Bulletin 24(4):
347-50.
Abramov, A. Y., J. Jacobson, et al. (2005). "Expression and modulation of an
NADPH oxidase in mammalian astrocytes." J Neurosci 25(40): 9176-84.
Acarin, L., B. Gonzâalez, et al. (1998). "Stat3 and NFkappaB glial expression after
excitotoxic damage to the postnatal brain." Neuroreport. 9(12): 2869-73.
Acarin, L., B. Gonzâalez, et al. (2000). "STAT3 and NFkappaB activation precedes
glial reactivity in the excitotoxically injured young cortex but not in the
corresponding distal thalamic nuclei." Journal of neuropathology and
experimental neurology. 59(2): 151-63.
135
Acarin, L., H. Peluffo, et al. (2005). "Astroglial nitration after postnatal excitotoxic
damage: correlation with nitric oxide sources, cytoskeletal, apoptotic and
antioxidant proteins." Journal of neurotrauma. 22(1): 189-200.
Ajtai, B. M., M. Kâalmâan, et al. (2000). "Axon growth failure following corpus
callosum lesions precedes glial reaction in perinatal rats." Anat Embryol
(Berl) 202(4): 313-21.
Akbar, M. T., D. J. Wells, et al. (2001). "Heat shock protein 27 shows a distinctive
widespread spatial and temporal pattern of induction in CNS glial and
neuronal cells compared to heat shock protein 70 and caspase 3 following
kainate administration." Brain Res Mol Brain Res 93(2): 148-63.
Alam, A., L. Y. Cohen, et al. (1999). "Early activation of caspases during T
lymphocyte stimulation results in selective substrate cleavage in nonapoptotic
cells." Journal of Experimental Medicine 190(12): 1879-90.
Allen, R.T. Cluck, M.W et al (1998). "Mechanisms controlling cellular suicide: role
of Bcl-2 and caspases". Cellular and Molecular Life Sciences 54 (5): 427-45
Andres-Barquin, P. J., M. C. Hernandez, et al. (1998). "Injury selectively down-
regulates the gene encoding for the Id4 transcription factor in primary
cultures of forebrain astrocytes." Neuroreport 9(18): 4075-80.
Arama, E., J. Agapite, et al. (2003). "Caspase activity and a specific cytochrome C
are required for sperm differentiation in Drosophila." Dev Cell 4(5): 687-97.
Aronica, E., E. A. van Vliet, et al. (2001). "Cystatin C, a cysteine protease inhibitor,
is persistently up-regulated in neurons and glia in a rat model for mesial
temporal lobe epilepsy." Eur J Neurosci 14(9): 1485-91.
Aronica, E., D. A. Vandeputte, et al. (2001). "Expression of Id proteins increases in
astrocytes in the hippocampus of epileptic rats." Neuroreport 12(11): 2461-5.
136
Asano, T., T. Mori, et al. (2005). "Arundic acid (ONO-2506) ameliorates delayed
ischemic brain damage by preventing astrocytic overproduction of S100B."
Curr Drug Targets CNS Neurol Disord 4(2): 127-42.
Balasingam, V., T. Tejada-Berges, et al. (1994). "Reactive astrogliosis in the
neonatal mouse brain and its modulation by cytokines." J Neurosci 14(2):
846-56.
Ballabriga, J., E. Pozas, et al. (1997). "bFGF and FGFR-3 immunoreactivity in the
rat brain following systemic kainic acid administration at convulsant doses:
localization of bFGF and FGFR-3 in reactive astrocytes, and FGFR-3 in
reactive microglia." Brain Res 752(1-2): 315-8.
Barnoy, S., N. S. Kosower, et al. (2003). "Caspase-1-induced calpastatin degradation
in myoblast differentiation and fusion: cross-talk between the caspase and
calpain systems." FEBS Lett 546(2-3): 213-7.
Bechmann, I., S. Lossau, et al. (2000). "Reactive astrocytes upregulate Fas (CD95)
and Fas ligand (CD95L) expression but do not undergo programmed cell
death during the course of anterograde degeneration." Glia 32(1): 25-41.
Beer, R., G. Franz, et al. (2000). "Temporal profile and cell subtype distribution of
activated caspase-3 following experimental traumatic brain injury." Journal
of Neurochemistry 75(3): 1264-73.
Beere, H. M. and D. o. I. M. T. U. S. A. H. B. s. o. St Jude Children's Research
Hospital (2005). "Death versus survival: functional interaction between the
apoptotic and stress-inducible heat shock protein pathways." J Clin Invest
115(10): 2633-9.
Bendotti, C., F. Guglielmetti, et al. (2000). "Differential expression of S100beta and
glial fibrillary acidic protein in the hippocampus after kainic acid-induced
lesions and mossy fiber sprouting in adult rat." Exp Neurol 161(1): 317-29.
137
Bolego, C., S. Ceruti, et al. (1997). "Characterization of the signalling pathways
involved in ATP and basic fibroblast growth factor-induced astrogliosis." Br
J Pharmacol 121(8): 1692-9.
Brambilla, R. and M. P. Abbracchio (2001). "Modulation of cyclooxygenase-2 and
brain reactive astrogliosis by purinergic P2 receptors." Annals of the New
York Academy of Sciences 939: 54-62.
Brambilla, R., S. Ceruti, et al. (2000). "A novel gliotic P2 receptor mediating
cyclooxygenase-2 induction in rat and human astrocytes." Journal of the
Autonomic Nervous System 81(1-3): 3-9.
Brenneke, F., M. Schachner, et al. (2004). "Up-regulation of the extracellular matrix
glycoprotein tenascin-R during axonal reorganization and astrogliosis in the
adult rat hippocampus." Epilepsy research. 58(2-3): 133-43.
Bresjanac, M., G. Antauer, et al. (2000). "Reactive astrocytes of the quinolinic acid-
lesioned rat striatum express GFRalpha1 as well as GDNF in vivo." Exp
Neurol 164(1): 53-9.
Brown, G. C., A. Bal-Price, et al. (2003). "Inflammatory neurodegeneration
mediated by nitric oxide, glutamate, and mitochondria." Mol Neurobiol
27(3): 325-55.
Bush, T. G., N. Puvanachandra, et al. (1999). "Leukocyte infiltration, neuronal
degeneration, and neurite outgrowth after ablation of scar-forming, reactive
astrocytes in adult transgenic mice." Neuron 23(2): 297-308.
Camand, E., M. P. Morel, et al. (2004). "Long-term changes in the molecular
composition of the glial scar and progressive increase of serotoninergic fibre
sprouting after hemisection of the mouse spinal cord." Eur J Neurosci 20(5):
1161-76.
138
Campbell, D. S., C. E. Holt, et al. (2003). "Apoptotic pathway and MAPKs
differentially regulate chemotropic responses of retinal growth cones."
Neuron. 37(6): 939-52.
Campbell, D. S., C. E. Holt, et al. (2003). "Apoptotic pathway and MAPKs
differentially regulate chemotropic responses of retinal growth cones."
Neuron 37(6): 939-52.
Canning, D. R., R. J. McKeon, et al. (1993). "beta-Amyloid of Alzheimer's disease
induces reactive gliosis that inhibits axonal outgrowth." Experimental
Neurology 124(2): 289-98.
Cardone, M. H., N. Roy, et al. (1998). "Regulation of cell death protease caspase-9
by phosphorylation." Science 282(5392): 1318-21.
Castegna, A., M. Aksenov, et al. (2002). "Proteomic identification of oxidatively
modified proteins in Alzheimer's disease brain. Part II: dihydropyrimidinase-
related protein 2, alpha-enolase and heat shock cognate 71." J Neurochem
82(6): 1524-32.
Chan, S. L., M. P. Mattson, et al. (1999). "Caspase and calpain substrates: roles in
synaptic plasticity and cell death." Journal of neuroscience research. 58(1):
167-90.
Chan, S. L., M. P. Mattson, et al. (1999). "Caspase and calpain substrates: roles in
synaptic plasticity and cell death." J Neurosci Res 58(1): 167-90.
Chang, H. Y., X. Yang, et al. (2000). "Proteases for cell suicide: functions and
regulation of caspases." Microbiology and molecular biology reviews :
MMBR. 64(4): 821-46.
Che, Y., C. S. Piao, et al. (2001). "Delayed induction of alpha B-crystallin in
activated glia cells of hippocampus in kainic acid-treated mouse brain."
Journal of Neuroscience Research 65(5): 425-31.
139
Che, Y., Y. Yu, et al. (2001). "Delayed induction of p38 MAPKs in reactive
astrocytes in the brain of mice after KA-induced seizure." Brain Research.
Molecular Brain Research 94(1-2): 157-65.
Chen, Q.Gong, B et al. (2000). "Distinct stages of cytochrome c release from
mitochondria: evidence for a feedback amplification loop linking caspase
activation to mitochondrial dysfunction in genotoxic stress induced
apoptosis" Cell Death Differ 7(2): 227-33
Chen, M. J., Y. W. Yap, et al. (2006). "Early induction of calpains in rotenone-
mediated neuronal apoptosis." Neurosci Lett 397(1-2): 69-73.
Chen, D. Zhou, Q. et al. (2004). "Caspase cleavage of BimEL triggers a positive
feedback amplification of apoptotic signaling" Proc Natl Acad Sci U S A
101(5) 1235-40
Choi, J. S., S. Y. Kim, et al. (2003). "Upregulation of gp130 and differential
activation of STAT and p42/44 MAPK in the rat hippocampus following
kainic acid-induced seizures." Brain Res Mol Brain Res 119(1): 10-8.
Chua, B. T., K. Guo, et al. (2000). "Direct cleavage by the calcium-activated
protease calpain can lead to inactivation of caspases." J Biol Chem 275(7):
5131-5.
Chun, J. T., M. Crispino, et al. (2004). "The dual response of protein kinase Fyn to
neural trauma: early induction in neurons and delayed induction in reactive
astrocytes." Exp Neurol 185(1): 109-19.
Chung, Y. H., C. M. Shin, et al. (2002). "Enhanced expression of p53 in reactive
astrocytes following transient focal ischemia." Neurol Res 24(3): 324-8.
Condorelli, D. F., P. Dell'Albani, et al. (1990). "Glial fibrillary acidic protein
messenger RNA and glutamine synthetase activity after nervous system
injury." J Neurosci Res 26(2): 251-7.
140
Constantinescu, C. S., M. Tani, et al. (2005). "Astrocytes as antigen-presenting cells:
expression of IL-12/IL-23." J Neurochem 95(2): 331-40.
Costa, S., T. Planchenault, et al. (2002). "Astroglial permissivity for neuritic
outgrowth in neuron-astrocyte cocultures depends on regulation of laminin
bioavailability." Glia 37(2): 105-13.
Cui, W., N. D. Allen, et al. (2001). "Inducible ablation of astrocytes shows that these
cells are required for neuronal survival in the adult brain." Glia. 34(4): 272-
82.
Cursi, S., A. Rufini, et al. (2006). "Src kinase phosphorylates Caspase-8 on Tyr380:
a novel mechanism of apoptosis suppression." Embo J 25(9): 1895-905.
Dash, P. K., S. Blum, et al. (2000). "Caspase activity plays an essential role in long-
term memory." Neuroreport 11(12): 2811-6.
De Freitas, M. S., T. C. Spohr, et al. (2002). "Neurite outgrowth is impaired on
HSP70-positive astrocytes through a mechanism that requires NF-kappaB
activation." Brain research. 958(2): 359-70.
De Maria, R., A. Zeuner, et al. (1999). "Negative regulation of erythropoiesis by
caspase-mediated cleavage of GATA-1." Nature 401(6752): 489-93.
De_Maria, R., A. Zeuner, et al. (1999). "Negative regulation of erythropoiesis by
caspase-mediated cleavage of GATA-1." Nature 401(6752): 489-93.
DiProspero, N. A., S. Meiners, et al. (1997). "Inflammatory cytokines interact to
modulate extracellular matrix and astrocytic support of neurite outgrowth."
Exp Neurol 148(2): 628-39.
Druilhe, A., S. M. Srinivasula, et al. (2001). "Regulation of IL-1beta generation by
Pseudo-ICE and ICEBERG, two dominant negative caspase recruitment
domain proteins." Cell Death Differ 8(6): 649-57.
141
Dusart, I., S. Marty, et al. (1991). "Glial changes following an excitotoxic lesion in
the CNS--II. Astrocytes." Neuroscience 45(3): 541-9.
Earnshaw, W. C., L. M. Martins, et al. (1999). "Mammalian caspases: structure,
activation, substrates, and functions during apoptosis." Annual Review of
Biochemistry 68: 383-424.
Earnshaw, W. C., L. M. Martins, et al. (1999). "Mammalian caspases: structure,
activation, substrates, and functions during apoptosis." Annual review of
biochemistry. 68.
Eckhart, L., M. Henry, et al. (2001). "Alternative splicing of caspase-8 mRNA
during differentiation of human leukocytes." Biochem Biophys Res Commun
289(4): 777-81.
Eclancher, F., P. Kehrli, et al. (1996). "Basic fibroblast growth factor (bFGF)
injection activates the glial reaction in the injured adult rat brain." Brain
Research 737(1-2): 201-14.
Eddleston, M. and L. Mucke (1993). "Molecular profile of reactive astrocytes--
implications for their role in neurologic disease." Neuroscience 54(1): 15-36.
Ekert, P. G., J. Silke, et al. (1999). "Caspase inhibitors." Cell Death and
Differentiation 6(11): 1081-6.
Eng, L. F. and R. S. Ghirnikar (1994). "GFAP and astrogliosis." Brain Pathology
(Zurich, Switzerland) 4(3): 229-37.
Eng, L. F., R. S. Ghirnikar, et al. (2000). "Glial fibrillary acidic protein: GFAP-
thirty-one years (1969-2000)." Neurochemical Research 25(9-10): 1439-51.
Fadeel, B., S. Orrenius, et al. (2000). "The most unkindest cut of all: on the multiple
roles of mammalian caspases." Leukemia 14(8): 1514-25.
142
Falsig, J., P. Pèorzgen, et al. (2006). "The inflammatory transcriptome of reactive
murine astrocytes and implications for their innate immune function." J
Neurochem 96(3): 893-907.
Faulkner, J. R., J. E. Herrmann, et al. (2004). "Reactive astrocytes protect tissue and
preserve function after spinal cord injury." The Journal of neuroscience : the
official journal of the Society for Neuroscience. 24(9): 2143-55.
Fernando, P., J. F. Kelly, et al. (2002). "Caspase 3 activity is required for skeletal
muscle differentiation." Proceedings of the National Academy of Sciences of
the United States of America 99(17): 11025-30.
Ferrer, I., R. Blanco, et al. (1996). "CREB-1 and CREB-2 immunoreactivity in the
rat brain." Brain Research 712(1): 159-64.
Ferrer, I., R. Blanco, et al. (2000). "N-myc and c-myc expression in Alzheimer
disease, Huntington disease and Parkinson disease." Brain Res Mol Brain Res
77(2): 270-6.
Ferrer, I., E. Lopez, et al. (2000). "Differential c-Fos and caspase expression
following kainic acid excitotoxicity." 99(3): 245-56.
Fischer, U., R. U. Jèanicke, et al. (2003). "Many cuts to ruin: a comprehensive
update of caspase substrates." Cell death and differentiation. 10(1): 76-100.
Fran∂cois, F., M. J. Godinho, et al. (2000). "CREB is cleaved by caspases during
neural cell apoptosis." FEBS letters. 486(3): 281-4.
Frost, V., S. Al-Mehairi, et al. (2001). "Exploitation of a non-apoptotic caspase to
regulate the abundance of the cdkI p27(KIP1) in transformed lymphoid
cells." Oncogene 20(22): 2737-48.
143
Fuentes-Prior, P., G. S. Salvesen, et al. (2004). "The protein structures that shape
caspase activity, specificity, activation and inhibition." Biochem J 384(Pt) 2:
201-32.
Gansauge, S., F. Gansauge, et al. (1998). "Interleukin 1beta-converting enzyme
(caspase-1) is overexpressed in adenocarcinoma of the pancreas." Cancer Res
58(13): 2703-6.
Garnier, P., W. Ying, et al. (2003). "Ischemic preconditioning by caspase cleavage of
poly(ADP-ribose) polymerase-1." The Journal of neuroscience : the official
journal of the Society for Neuroscience. 23(22): 7967-73.
Garnier, P., W. Ying, et al. (2003). "Ischemic preconditioning by caspase cleavage of
poly(ADP-ribose) polymerase-1." J Neurosci 23(22): 7967-73.
Gemma, C., M. Fister, et al. (2005). "Improvement of memory for context by
inhibition of caspase-1 in aged rats." Eur J Neurosci 22(7): 1751-6.
Gèorg, B., H. J. Bidmon, et al. (2006). "Inflammatory cytokines induce protein
tyrosine nitration in rat astrocytes." Arch Biochem Biophys 449(1-2): 104-14.
Gilman, C. P., M. P. Mattson, et al. (2002). "Do apoptotic mechanisms regulate
synaptic plasticity and growth-cone motility?" Neuromolecular medicine.
2(2).
Gilman, C. P., M. P. Mattson, et al. (2002). "Do apoptotic mechanisms regulate
synaptic plasticity and growth-cone motility?" Neuromolecular Med 2(2).
Giulian, D. (1993). "Reactive glia as rivals in regulating neuronal survival." Glia
7(1): 102-10.
Giulian, D., K. Vaca, et al. (1993). "Brain glia release factors with opposing actions
upon neuronal survival." J Neurosci 13(1): 29-37.
144
Gomide, V. C., G. Chadi, et al. (1999). "The trophic factors S-100beta and basic
fibroblast growth factor are increased in the forebrain reactive astrocytes of
adult callosotomized rat." Brain Res 835(2): 162-74.
Graczyk, P. P. (2002). "Caspase inhibitors as anti-inflammatory and antiapoptotic
agents." Progress in Medicinal Chemistry 39: 1-72.
Grèutter, M. G. and U. o. Z. W. Z. S. g. b. u. c. Institute of Biochemistry (2000).
"Caspases: key players in programmed cell death." Current opinion in
structural biology. 10(6): 649-55.
Gulyaeva, N. V., I. o. H. N. A. Department of Functional Biochemistry of the
Nervous System, et al. (2003). "Non-apoptotic functions of caspase-3 in
nervous tissue." Biochemistry. Biokhimiia. 68(11): 1171-80.
Gulyaeva, N. V., I. E. Kudryashov, et al. (2003). "Caspase activity is essential for
long-term potentiation." Journal of neuroscience research. 73(6): 853-64.
Hagino, S., K. Iseki, et al. (2003). "Expression pattern of glypican-1 mRNA after
brain injury in mice." Neurosci Lett 349(1): 29-32.
Hama, H., Y. Kasuya, et al. (1997). "Role of endothelin-1 in astrocyte responses
after acute brain damage." J Neurosci Res 47(6): 590-602.
Hanbury, R., Z. D. Ling, et al. (2003). "GFAP knockout mice have increased levels
of GDNF that protect striatal neurons from metabolic and excitotoxic
insults." J Comp Neurol 461(3): 307-16.
Hardin-Pouzet, H., M. Krakowski, et al. (1997). "Glutamate metabolism is down-
regulated in astrocytes during experimental allergic encephalomyelitis." Glia
20(1): 79-85.
145
Hartlage-Rèubsamen, M., U. Zeitschel, et al. (2003). "Astrocytic expression of the
Alzheimer's disease beta-secretase (BACE1) is stimulus-dependent." Glia
41(2): 169-79.
Hasselblatt, M., M. Bunte, et al. (2003). "Effect of endothelin-1 on astrocytic protein
content." Glia 42(4): 390-7.
Heales, S. J., A. A. Lam, et al. (2004). "Neurodegeneration or neuroprotection: the
pivotal role of astrocytes." Neurochem Res 29(3): 513-9.
Hèoke, A., D. R. Canning, et al. (1994). "Regional differences in reactive gliosis
induced by substrate-bound beta-amyloid." Experimental neurology. 130(1):
56-66.
Hèoke, A., J. Silver, et al. (1994). "Heterogeneity among astrocytes in reactive
gliosis." Perspectives on developmental neurobiology. 2(3).
Hisahara, S., J. Yuan, et al. (2001). "Caspase-11 mediates oligodendrocyte cell death
and pathogenesis of autoimmune-mediated demyelination." The Journal of
experimental medicine. 193(1): 111-22.
Hong, G. S., Y. K. Jung, et al. (2002). "Caspase recruitment domain (CARD) as a bi-
functional switch of caspase regulation and NF-kappaB signals." J Biochem
Mol Biol 35(1): 19-23.
Hu, J., A. Ferreira, et al. (1997). "S100beta induces neuronal cell death through nitric
oxide release from astrocytes." Journal of Neurochemistry 69(6): 2294-301.
Huh, J. R., S. Y. Vernooy, et al. (2004). "Multiple apoptotic caspase cascades are
required in nonapoptotic roles for Drosophila spermatid individualization."
PLoS Biol 2(1): E15.
Hur, J., S. Y. Kim, et al. (2001). "Induction of caspase-11 by inflammatory stimuli in
rat astrocytes: lipopolysaccharide induction through p38 mitogen-activated
protein kinase pathway." Febs Letters 507(2): 157-62.
146
Ikeda, T., H. Koo, et al. (2002). "Bimodal upregulation of glial cell line-derived
neurotrophic factor (GDNF) in the neonatal rat brain following
ischemic/hypoxic injury." Int J Dev Neurosci 20(7): 555-62.
Iseki, K., S. Hagino, et al. (2002). "Increased syndecan expression by pleiotrophin
and FGF receptor-expressing astrocytes in injured brain tissue." Glia 39(1):
1-9.
Ishikawa, N., M. Takemura, et al. (1997). "Endothelins promote the activation of
astrocytes in rat neostriatum through ET(B) receptors." Eur J Neurosci 9(5):
895-901.
Ishizaki, Y., M. D. Jacobson, et al. (1998). "A role for caspases in lens fiber
differentiation." Journal of Cell Biology 140(1): 153-8.
Ito, A., T. Uehara, et al. (2000). "Isolation of Ich-1S (caspase-2S)-binding protein
that partially inhibits caspase activity." Febs Letters 470(3): 360-4.
Jeon, G. S., S. W. Park, et al. (2004). "Glial expression of the 90-kDa heat shock
protein (HSP90) and the 94-kDa glucose-regulated protein (GRP94)
following an excitotoxic lesion in the mouse hippocampus."Glia 48(3): 250-8
Jeon, G. S., D. H. Shin, et al. (2004). "Induction of transcription factor c-myb
expression in reactive astrocytes following intracerebroventricular kainic acid
injection in mouse hippocampus." Neurosci Lett 360(1-2): 13-6.
Jiang, Z. H., J. Y. Wu, et al. (1999). "Alternative splicing and programmed cell
death." Proc Soc Exp Biol Med 220(2): 64-72.
Jin, J. K., J. K. Choi, et al. (2005). "Expression of calsenilin in neurons and
astrocytes in the Alzheimer's disease brain." Neuroreport 16(5): 451-5.
147
Kâalmâan, M., B. M. Ajtai, et al. (2000). "Characteristics of glial reaction in the
perinatal rat cortex: effect of lesion size in the 'critical period'." Neural Plast
7(3).
Kamada, S., U. Kikkawa, et al. (2005). "Nuclear translocation of caspase-3 is
dependent on its proteolytic activation and recognition of a substrate-like
protein(s)." The Journal of biological chemistry. 280(2): 857-60.
Kang, S. J., S. Wang, et al. (2000). "Dual role of caspase-11 in mediating activation
of caspase-1 and caspase-3 under pathological conditions." The Journal of
cell biology. 149(3): 613-22.
Kato, H., R. Kurosaki, et al. (2004). "Arundic acid, an astrocyte-modulating agent,
protects dopaminergic neurons against MPTP neurotoxicity in mice." Brain
research. 1030(1): 66-73.
Kawamoto, Y., I. Akiguchi, et al. (2006). "Upregulated expression of 14-3-3 proteins
in astrocytes from human cerebrovascular ischemic lesions." Stroke 37(3):
830-5.
Kennedy, N. J., T. Kataoka, et al. (1999). "Caspase activation is required for T cell
proliferation." Journal of Experimental Medicine 190(12): 1891-6.
Kim, S. W., Y. M. Yu, et al. (2004). "Inhibition of delayed induction of p38
mitogen-activated protein kinase attenuates kainic acid-induced neuronal loss
in the hippocampus." Brain Res 1007(1-2): 188-91.
Kim, Y. M., R. V. Talanian, et al. (1997). "Nitric oxide inhibits apoptosis by
preventing increases in caspase-3-like activity via two distinct mechanisms."
J Biol Chem 272(49): 31138-48.
Kisenge, R. R., H. Toyoda, et al. (2003). "Expression of short-form caspase 8
correlates with decreased sensitivity to Fas-mediated apoptosis in
neuroblastoma cells." Cancer Sci 94(7): 598-605.
148
Koguchi, K., Y. Nakatsuji, et al. (2002). "Modulation of astrocyte proliferation by
cyclin-dependent kinase inhibitor p27(Kip1)." Glia. 37(2): 93-104.
Koguchi, K., Y. Nakatsuji, et al. (2002). "Modulation of astrocyte proliferation by
cyclin-dependent kinase inhibitor p27(Kip1)." Glia 37(2): 93-104.
Koyama, Y., T. Ishibashi, et al. (1993). "Endothelins modulate dibutyryl cAMP-
induced stellation of cultured astrocytes." Brain Research 600(1): 81-8.
Koyama, Y., M. Takemura, et al. (1999). "BQ788, an endothelin ET(B) receptor
antagonist, attenuates stab wound injury-induced reactive astrocytes in rat
brain." Glia 26(3): 268-71.
Kruchkova, Y., I. Ben-Dror, et al. (2001). "Basic fibroblast growth factor: a potential
inhibitor of glutamine synthetase expression in injured neural tissue." J
Neurochem 77(6): 1641-9.
Krupinski, J., E. Lopez, et al. (2000). "Expression of caspases and their substrates in
the rat model of focal cerebral ischemia." Neurobiology of Disease 7(4): 332-
42.
Kudryashov, I. E., A. A. Yakovlev, et al. (2004). "Inhibition of caspase-3 blocks
long-term potentiation in hippocampal slices." Neurosci Behav Physiol 34(9):
877-80.
Lamkanfi, M., W. Declercq, et al. (2006). "Caspases leave the beaten track: caspase-
mediated activation of NF-kappaB." J Cell Biol 173(2): 165-71.
Larsson, A., U. Wilhelmsson, et al. (2004). "Increased cell proliferation and
neurogenesis in the hippocampal dentate gyrus of old GFAP(-/-)Vim(-/-)
mice." Neurochem Res 29(11): 2069-73.
149
Le, R., S. Esquenazi, et al. (2002). "Astrocytes mediate cerebral cortical neuronal
axon and dendrite growth, in part, by release of fibroblast growth factor."
Neurol Res 24(1): 81-92.
LeBlanc, A. C. and T. B. C. f. R. i. A. H. T. E. M. Q. C. a. l. m. c. Department of
Neurology (2003). "Natural cellular inhibitors of caspases." Prog
Neuropsychopharmacol Biol Psychiatry 27(2): 215-29.
Lee, C. Y., G. D. Pappas, et al. (2003). "Proliferation of a subpopulation of reactive
astrocytes following needle-insertion lesion in rat." Neurol Res 25(7): 767-
76.
Lefran∂cois, T., C. Fages, et al. (1997). "Neuritic outgrowth associated with
astroglial phenotypic changes induced by antisense glial fibrillary acidic
protein (GFAP) mRNA in injured neuron-astrocyte cocultures." J Neurosci
17(11): 4121-8.
Levison, S. W., M. H. Ducceschi, et al. (1996). "Acute exposure to CNTF in vivo
induces multiple components of reactive gliosis." Experimental Neurology
141(2): 256-68.
Le_Roux, P. D. and T. A. Reh (1994). "Regional differences in glial-derived factors
that promote dendritic outgrowth from mouse cortical neurons in vitro." The
Journal of Neuroscience : the Official Journal of the Society For
Neuroscience 14(8): 4639-55.
Le_Roux, P. D. and T. A. Reh (1995). "Astroglia demonstrate regional differences in
their ability to maintain primary dendritic outgrowth from mouse cortical
neurons in vitro." Journal of Neurobiology 27(1): 97-112.
Le_Roux, P. D. and T. A. Reh (1995). "Independent regulation of primary dendritic
and axonal growth by maturing astrocytes in vitro." Neuroscience Letters
198(1): 5-8.
150
Le_Roux, P. D. and T. A. Reh (1996). "Reactive astroglia support primary dendritic
but not axonal outgrowth from mouse cortical neurons in vitro."
Experimental Neurology 137(1): 49-65.
Lin, J., W. Cai, et al. (2004). "Effect of vimentin on reactive gliosis: in vitro and in
vivo analysis." J Neurotrauma 21(11): 1671-82.
Lin, T. N., Q. Wang, et al. (2004). "Induction of secretory phospholipase A2 in
reactive astrocytes in response to transient focal cerebral ischemia in the rat
brain." J Neurochem 90(3): 637-45.
Liu, X., G. Ying, et al. (2005). "Entorhinal deafferentation induces upregulation of
SPARC in the mouse hippocampus." Brain Res Mol Brain Res 141(1): 58-65.
Liu B, Chen H, et al (2006)."Epidermal growth factor receptor activation: an
upstream signal for transition of quiescent astrocytes into reactive astrocytes
after neural injury." J Neurosci. 26(28) :7532-40.
Lu, C., Y. Wang, et al. (2006). "Evidence that caspase-1 is a negative regulator of
AMPA receptor-mediated long-term potentiation at hippocampal synapses." J
Neurochem 97(4): 1104-10.
Ludowyk, P. A., W. Hughes, et al. (1993). "Astrocytic hypertrophy: an important
pathological feature of chronic experimental autoimmune encephalitis in
aged rats." J Neuroimmunol 48(2): 121-34.
Mahler, M., Y. Ben-Ari, et al. (1997). "Differential expression of fibronectin,
tenascin-C and NCAMs in cultured hippocampal astrocytes activated by
kainate, bacterial lipopolysaccharide or basic fibroblast growth factor." Brain
Res 775(1-2): 63-73.
Malhotra, S. K., T. K. Shnitka, et al. (1990). "Reactive astrocytes--a review."
Cytobios 61(246-247): 133-60.
151
Mandell, J. W., N. C. Gocan, et al. (2001). "Mechanical trauma induces rapid
astroglial activation of ERK/MAP kinase: Evidence for a paracrine signal."
Glia 34(4): 283-95.
Mandell, J. W. and S. R. VandenBerg (1999). "ERK/MAP kinase is chronically
activated in human reactive astrocytes." Neuroreport 10(17): 3567-72.
Mander, P., V. Borutaite, et al. (2005). "Nitric oxide from inflammatory-activated
glia synergizes with hypoxia to induce neuronal death." Journal of
neuroscience research. 79(1-2): 208-15.
Mannick, J. B., C. Schonhoff, et al. (2001). "S-Nitrosylation of mitochondrial
caspases." J Cell Biol 154(6): 1111-6.
Martin, M. C., L. A. Allan, et al. (2005). "Protein kinase A regulates caspase-9
activation by Apaf-1 downstream of cytochrome c." J Biol Chem 280(15):
15449-55.
Matsui, T., T. Mori, et al. (2002). "Astrocytic activation and delayed infarct
expansion after permanent focal ischemia in rats. Part I: enhanced astrocytic
synthesis of s-100beta in the periinfarct area precedes delayed infarct
expansion." J Cereb Blood Flow Metab 22(6): 711-22.
Mattson, M. P., W. Duan, et al. (1999). ""Apoptotic" biochemical cascades in
synaptic compartments: roles in adaptive plasticity and neurodegenerative
disorders." J Neurosci Res 58(1): 152-66.
Mattson, M. P., J. N. Keller, et al. (1998). "Evidence for synaptic apoptosis." Exp
Neurol 153(1): 35-48.
McKeon, R. J., A. Hoke, et al. (1995). "Injury-induced proteoglycans inhibit the
potential for laminin-mediated axon growth on astrocytic scars."
Experimental Neurology 136(1): 32-43.
152
McKeon, R. J., R. C. Schreiber, et al. (1991). "Reduction of neurite outgrowth in a
model of glial scarring following CNS injury is correlated with the
expression of inhibitory molecules on reactive astrocytes." The Journal of
Neuroscience : the Official Journal of the Society For Neuroscience 11(11):
3398-411.
McKeon, R. J., R. C. Schreiber, et al. (1991). "Reduction of neurite outgrowth in a
model of glial scarring following CNS injury is correlated with the
expression of inhibitory molecules on reactive astrocytes." J Neurosci 11(11):
3398-411.
McLaughlin, B. and V. K. C. f. R. o. H. D. V. U. N. T. N. U. S. A. b. m. v. e.
Department of Pharmacology (2004). "The kinder side of killer proteases:
caspase activation contributes to neuroprotection and CNS remodeling."
Apoptosis : an international journal on programmed cell death. 9(2): 111-21.
McLaughlin, B., K. A. Hartnett, et al. (2003). "Caspase 3 activation is essential for
neuroprotection in preconditioning." Proceedings of the National Academy of
Sciences of the United States of America. 100(2): 715-20.
McLaughlin, B., K. A. Hartnett, et al. (2003). "Caspase 3 activation is essential for
neuroprotection in preconditioning." Proc Natl Acad Sci U S A 100(2): 715-
20.
Meiners, S., M. Marone, et al. (1993). "Regulation of astrocytic tenascin by basic
fibroblast growth factor." Dev Biol 160(2): 480-93.
Menet, V., Y. R. M. Gimâenez, et al. (2000). "GFAP null astrocytes are a favorable
substrate for neuronal survival and neurite growth." Glia 31(3): 267-72.
Menet, V., M. Gimenez_Y_Ribotta, et al. (2000). "GFAP null astrocytes are a
favorable substrate for neuronal survival and neurite growth." Glia 31(3):
267-72.
153
Miller, C., O. Tsatas, et al. (1994). "Dibutyryl cAMP, interleukin-1 beta, and
macrophage conditioned medium enhance the ability of astrocytes to promote
neurite growth." J Neurosci Res 38(1): 56-63.
Monnerie, H., S. Esquenazi, et al. (2005). "Beta-amyloid-induced reactive astrocytes
display altered ability to support dendrite and axon growth from mouse
cerebral cortical neurons in vitro." Neurol Res 27(5): 525-32.
Mouser, P. E., E. Head, et al. (2006). "Caspase-mediated cleavage of glial fibrillary
acidic protein within degenerating astrocytes of the Alzheimer's disease
brain." Am J Pathol 168(3): 936-46.
Nakagawa, T., J. P. Schwartz, et al. (2004). "Gene expression patterns in in vivo
normal adult astrocytes compared with cultured neonatal and normal adult
astrocytes." Neurochem Int 45(2-3): 203-42.
Nakagawa, T., T. Yabe, et al. (2005). "Gene expression profiles of reactive
astrocytes cultured from dopamine-depleted striatum." Neurobiol Dis 20(2):
275-82.
Nakagawa, T., J. Yuan, et al. (2000). "Cross-talk between two cysteine protease
families. Activation of caspase-12 by calpain in apoptosis." J Cell Biol
150(4): 887-94.
Nakic, M., N. Mitrovic, et al. (1996). "Kainic acid activates transient expression of
tenascin-C in the adult rat hippocampus." J Neurosci Res 44(4): 355-62.
Narita, M., M. Suzuki, et al. (2004). "Neuronal protein kinase C gamma-dependent
proliferation and hypertrophy of spinal cord astrocytes following repeated in
vivo administration of morphine." Eur J Neurosci 19(2): 479-84.
Narkilahti, S., J. Nissinen, et al. (2003). "Administration of caspase 3 inhibitor
during and after status epilepticus in rat: effect on neuronal damage and
epileptogenesis." Neuropharmacology. 44(8): 1068-88.
154
Nieto_Sampedro, M. (1999). "Neurite outgrowth inhibitors in gliotic tissue."
Advances in Experimental Medicine and Biology 468: 207-24.
Nikaido, T., K. Iseki, et al. (2002). "Expression of OASIS, a CREB/ATF family
transcription factor, in CNS lesion and its transcriptional activity." Brain
research. Molecular brain research. 108(1-2): 129-38.
Nomoto, Y., M. Yamamoto, et al. (2001). "Expression of nuclear factor kappaB and
tumor necrosis factor alpha in the mouse brain after experimental thermal
ablation injury." Neurosurgery 48(1): 158-66.
Norris, C. M., I. Kadish, et al. (2005). "Calcineurin triggers reactive/inflammatory
processes in astrocytes and is upregulated in aging and Alzheimer's models."
J Neurosci 25(18): 4649-58.
Noyan-Ashraf, M. H., F. Brandizzi, et al. (2005). "Constitutive nuclear localization
of activated caspase 3 in subpopulations of the astroglial family of cells."
Glia. 49(4): 588-93.
Okamoto, M., J. Sakiyama, et al. (2003). "Kainic acid-induced convulsions cause
prolonged changes in the chondroitin sulfate proteoglycans neurocan and
phosphacan in the limbic structures." Exp Neurol 184(1): 179-95.
Ong, W. Y., S. K. Leong, et al. (1996). "An immunocytochemical study of glutamate
receptors and glutamine synthetase in the hippocampus of rats injected with
kainate." Exp Brain Res 109(2): 251-67.
Ong, W. Y., H. M. Lim, et al. (2000). "Kainate-induced neuronal injury leads to
persistent phosphorylation of cAMP response element-binding protein in
glial and endothelial cells in the hippocampus." Experimental Brain
Research. Experimentelle Hirnforschung. Experimentation Cerebrale 131(2):
178-86.
155
Oomman, S., V. Finckbone, et al. (2004). "Active caspase-3 expression during
postnatal development of rat cerebellum is not systematically or consistently
associated with apoptosis." The Journal of comparative neurology. 476(2):
154-73.
Oomman, S., V. Finckbone, et al. (2004). "Active caspase-3 expression during
postnatal development of rat cerebellum is not systematically or consistently
associated with apoptosis." J Comp Neurol 476(2): 154-73.
Oomman, S., H. Strahlendorf, et al. (2006). "Bergmann glia utilize active caspase-3
for differentiation." Brain Res 1078(1): 19-34.
Oomman, S., H. Strahlendorf, et al. (2005). "Non-lethal active caspase-3 expression
in Bergmann glia of postnatal rat cerebellum." Brain Res Dev Brain Res
160(2): 130-45.
Oubrahim, H., P. B. Chock, et al. (2002). "Manganese(II) induces apoptotic cell
death in NIH3T3 cells via a caspase-12-dependent pathway." J Biol Chem
277(23): 20135-8.
Pâerez-Capote, K., J. Serratosa, et al. (2004). "Glial activation modulates glutamate
neurotoxicity in cerebellar granule cell cultures." Glia. 45(3): 258-68.
Pâetrilli, V. Herceg, Z. et al (2004) "Noncleavable poly(ADP-ribose) polymerase-1
regulates the inflammation response in mice". J Clin Invest 114(8) :1072-81
Paradisi, S., B. Sacchetti, et al. (2004). "Astrocyte modulation of in vitro beta-
amyloid neurotoxicity." Glia 46(3): 252-60.
Park, S. A., T. S. Kim, et al. (2003). "Chronic activation of CREB and p90RSK in
human epileptic hippocampus." Exp Mol Med 35(5): 365-70.
156
Park, S. M., R. Schickel, et al. (2005). "Nonapoptotic functions of FADD-binding
death receptors and their signaling molecules." Curr Opin Cell Biol 17(6):
610-6.
Pekny, M., C. Eliasson, et al. (1998). "GFAP-deficient astrocytes are capable of
stellation in vitro when cocultured with neurons and exhibit a reduced
amount of intermediate filaments and an increased cell saturation density."
Experimental Cell Research 239(2): 332-43.
Peluffo, H., L. Acarin, et al. (2006). "Neuroprotection from NMDA excitotoxic
lesion by Cu/Zn superoxide dismutase gene delivery to the postnatal rat brain
by a modular protein vector." BMC Neurosci 7.
Pennypacker, K. R., C. A. Kassed, et al. (2000). "Brain injury: prolonged induction
of transcription factors." 60(4): 515-30.
Perez_Otano, I., M. K. McMillian, et al. (1996). "Induction of NF-kB-like
transcription factors in brain areas susceptible to kainate toxicity." Glia 16(4):
306-15.
Peters, C. M., S. D. Rogers, et al. (2003). "Endothelin receptor expression in the
normal and injured spinal cord: potential involvement in injury-induced
ischemia and gliosis." Exp Neurol 180(1): 1-13.
Pike, C. J., N. Ramezan-Arab, et al. (1996). "beta-Amyloid increases enzyme
activity and protein levels of glutamine synthetase in cultured astrocytes."
Exp Neurol 139(1): 167-71.
Pike, C. J., P. J. Vaughan, et al. (1996). "Thrombin attenuates neuronal cell death
and modulates astrocyte reactivity induced by beta-amyloid in vitro." Journal
of Neurochemistry 66(4): 1374-82.
Pindon, A., B. W. Festoff, et al. (1998). "Thrombin-induced reversal of astrocyte
stellation is mediated by activation of protein kinase C beta-1." European
Journal of Biochemistry 255(3): 766-74.
157
Powell, E. M., S. Meiners, et al. (1997). "Mechanisms of astrocyte-directed neurite
guidance." Cell Tissue Res 290(2): 385-93.
Pozas, E., F. Aguado, et al. (1999). "Fra-1 immunoreactivity in the rat brain during
normal postnatal development and after injury in adulthood." Neurosci Res
33(2): 137-45.
Qin, Z., Y. Wang, et al. (2000). "A caspase-3-like protease is involved in NF-kappaB
activation induced by stimulation of N-methyl-D-aspartate receptors in rat
striatum." Brain research. Molecular brain research. 80(2): 111-22.
Ramadani, M., F. Gansauge, et al. (2001). "Overexpression of caspase-1 in
pancreatic disorders: implications for a function besides apoptosis." J
Gastrointest Surg 5(4): 352-8.
Ramadani, M., Y. Yang, et al. (2001). "Overexpression of caspase-1 (interleukin-
1beta converting enzyme) in chronic pancreatitis and its participation in
apoptosis and proliferation." Pancreas 22(4): 383-7.
Ramakers, G. J. and W. H. Moolenaar (1998). "Regulation of astrocyte morphology
by RhoA and lysophosphatidic acid." Experimental Cell Research 245(2):
252-62.
Reeves, R. H., J. Yao, et al. (1994). "Astrocytosis and axonal proliferation in the
hippocampus of S100b transgenic mice." Proc Natl Acad Sci U S A 91(12):
5359-63.
Renkawek, K., G. J. Bosman, et al. (1994). "Expression of small heat-shock protein
hsp 27 in reactive gliosis in Alzheimer disease and other types of dementia."
87(5): 511-9.
Richter, B. W. and C. S. Duckett (2000). "The IAP proteins: caspase inhibitors and
beyond." 2000(44): PE1.
158
Ridet, J. L., S. K. Malhotra, et al. (1997). "Reactive astrocytes: cellular and
molecular cues to biological function." Trends in Neurosciences 20(12): 570-
7.
Rogers, S. D., C. M. Peters, et al. (2003). "Endothelin B receptors are expressed by
astrocytes and regulate astrocyte hypertrophy in the normal and injured
CNS." Glia 41(2): 180-90.
Rozovsky, I., M. Wei, et al. (2005). "Reversible age impairments in neurite
outgrowth by manipulations of astrocytic GFAP." Neurobiology of aging.
26(5): 705-15.
Saas, P., J. Boucraut, et al. (1999). "CD95 (Fas/Apo-1) as a receptor governing
astrocyte apoptotic or inflammatory responses: a key role in brain
inflammation?" Journal of Immunology (Baltimore, Md. : 1950) 162(4):
2326-33.
Sandvig, A., M. Berry, et al. (2004). "Myelin-, reactive glia-, and scar-derived CNS
axon growth inhibitors: expression, receptor signaling, and correlation with
axon regeneration." Glia 46(3): 225-51.
Sanz, O., L. Acarin, et al. (2001). "Expression of 27 kDa heat shock protein (Hsp27)
in immature rat brain after a cortical aspiration lesion." Glia 36(3): 259-70.
Sasaki, T., T. Endo, et al. (2000). "Both cell-surface carbohydrates and protein
tyrosine phosphatase are involved in the differentiation of astrocytes in
vitro." Glia 32(1): 60-70.
Schèafer, M., A. U. Brèauer, et al. (2005). "Neurotractin/kilon promotes neurite
outgrowth and is expressed on reactive astrocytes after entorhinal cortex
lesion." Mol Cell Neurosci 29(4): 580-90.
159
Schwartz, J. P. and D. J. Wilson (1992). "Preparation and characterization of type 1
astrocytes cultured from adult rat cortex, cerebellum, and striatum." Glia
5(1): 75-80.
Schwerk, C., K. Schulze-Osthoff, et al. (2003). "Non-apoptotic functions of caspases
in cellular proliferation and differentiation." Biochemical pharmacology.
66(8): 1453-8.
Schwerk, C., K. Schulze-Osthoff, et al. (2005). "Regulation of apoptosis by
alternative pre-mRNA splicing." Mol Cell 19(1): 1-13.
Seol, D. W., T. R. Billiar, et al. (1999). "A caspase-9 variant missing the catalytic
site is an endogenous inhibitor of apoptosis." J Biol Chem 274(4): 2072-6.
Sheng, J., S. Yang, et al. (2004). "Bystin as a novel marker for reactive astrocytes in
the adult rat brain following injury." Eur J Neurosci 20(4): 873-84.
Shetty, A. K., M. S. Rao, et al. (2004). "Hippocampal neurotrophin levels after
injury: Relationship to the age of the hippocampus at the time of injury." J
Neurosci Res 78(4): 520-32.
Smith, G. M., U. Rutishauser, et al. (1990). "Maturation of astrocytes in vitro alters
the extent and molecular basis of neurite outgrowth." Dev Biol 138(2): 377-
90.
Srebro, Z. and K. Dziobek (2001). "Neuroprotection: the role of neuroglia." Folia
Medica Cracoviensia 42(3): 113-21.
Sriram, K., S. A. Benkovic, et al. (2004). "Induction of gp130-related cytokines and
activation of JAK2/STAT3 pathway in astrocytes precedes up-regulation of
glial fibrillary acidic protein in the 1-methyl-4-phenyl-1,2,3,6-
tetrahydropyridine model of neurodegeneration: key signaling pathway for
astrogliosis in vivo?" J Biol Chem 279(19): 19936-47.
160
Stepanichev, M. Y., I. V. Kudryashova, et al. (2005). "Central administration of a
caspase inhibitor impairs shuttle-box performance in rats." Neuroscience
136(2).
Stephenson, D., K. Rash, et al. (1999). "Cytosolic phospholipase A2 is induced in
reactive glia following different forms of neurodegeneration." Glia 27(2):
110-28.
Su, J. H., K. E. Nichol, et al. (2000). "DNA damage and activated caspase-3
expression in neurons and astrocytes: evidence for apoptosis in
frontotemporal dementia." Experimental Neurology 163(1): 9-19.
Suidan, H. S., C. D. Nobes, et al. (1997). "Astrocyte spreading in response to
thrombin and lysophosphatidic acid is dependent on the Rho GTPase." Glia
21(2): 244-52.
Suk, K., S. Y. Kim, et al. (2002). "Essential role of caspase-11 in activation-induced
cell death of rat astrocytes." Journal of neurochemistry. 80(2): 230-8.
Swanson, R. A., W. Ying, et al. (2004). "Astrocyte influences on ischemic neuronal
death." Current molecular medicine. 4(2): 193-205.
Takenaga, K., E. N. Kozlova, et al. (2006). "Role of intracellular S100A4 for
migration of rat astrocytes." Glia 53(3): 313-21.
Tanaka, H., M. Araki, et al. (1992). "Reaction of astrocytes in the gerbil
hippocampus following transient ischemia: immunohistochemical
observations with antibodies against glial fibrillary acidic protein, glutamine
synthetase, and S-100 protein." Exp Neurol 116(3): 264-74.
Tan KB, Magdalene Koh et al. (2006)."Double Immunofluorescence Shows
Coexpression of Bcl-x with GFAP in a Variety of Glial Lesions."
J Neurooncol. Jun 14; [Epub ahead of print]
161
Tatsumi, K., S. Haga, et al. (2005). "Characterization of cells with proliferative
activity after a brain injury." Neurochem Int 46(5): 381-9.
Testa, U., H. Department of, et al. (2004). "Apoptotic mechanisms in the control of
erythropoiesis." Leukemia 18(7): 1176-99.
Tinel, A., J. Tschopp, et al. (2004). "The PIDDosome, a protein complex implicated
in activation of caspase-2 in response to genotoxic stress." Science
304(5672): 843-6.
Tzeng, S. F., M. Kahn, et al. (1999). "Tumor necrosis factor-alpha regulation of the
Id gene family in astrocytes and microglia during CNS inflammatory injury."
Glia 26(2): 139-52.
Varfolomeev, E. E., M. Schuchmann, et al. (1998). "Targeted disruption of the
mouse Caspase 8 gene ablates cell death induction by the TNF receptors,
Fas/Apo1, and DR3 and is lethal prenatally." Immunity 9(2): 267-76.
Varga, Z. M., M. E. Schwab, et al. (1995). "Myelin-associated neurite growth-
inhibitory proteins and suppression of regeneration of immature mammalian
spinal cord in culture." Proc Natl Acad Sci U S A 92(24): 10959-63.
Wandosell, F., P. Bovolenta, et al. (1993). "Differences between reactive astrocytes
and cultured astrocytes treated with di-butyryl-cyclic AMP." J Neuropathol
Exp Neurol 52(3): 205-15.
Wang, K., L. K. Bekar, et al. (2004). "Vimentin-expressing proximal reactive
astrocytes correlate with migration rather than proliferation following focal
brain injury." Brain Res 1024(1-2): 193-202.
Wang, S., M. Miura, et al. (1998). "Murine caspase-11, an ICE-interacting protease,
is essential for the activation of ICE." Cell. 92(4): 501-9.
162
Watanabe, Y. and T. Akaike (1999). "Possible involvement of caspase-like family in
maintenance of cytoskeleton integrity." Journal of Cellular Physiology
179(1): 45-51.
Weber, G. F. Menko, A. S (2005) "The canonical intrinsic mitochondrial death
pathway has a non-apoptotic role in signaling lens cell differentiation" J Biol
Chem 280(23) :22135-45
Webster, B., L. Hansen, et al. (2006). "Astroglial activation of extracellular-
regulated kinase in early stages of Alzheimer disease." J Neuropathol Exp
Neurol 65(2): 142-51.
Weil, M., M. C. Raff, et al. (1999). "Caspase activation in the terminal
differentiation of human epidermal keratinocytes." Current Biology 9(7):
361-4.
Wiessner, C., D. Sauer, et al. (2000). "Protective effect of a caspase inhibitor in
models for cerebral ischemia in vitro and in vivo." 46(1): 53-62.
Wilhelm, S., H. Wagner, et al. (1998). "Activation of caspase-3-like enzymes in non-
apoptotic T cells." European Journal of Immunology 28(3): 891-900.
Williams, E. J., B. Mittal, et al. (1995). "FGF inhibits neurite outgrowth over
monolayers of astrocytes and fibroblasts expressing transfected cell adhesion
molecules." J Cell Sci 108(Pt 11): 3523-30.
Winter, C. G., Y. Saotome, et al. (1995). "A role for ciliary neurotrophic factor as an
inducer of reactive gliosis, the glial response to central nervous system
injury." Proc Natl Acad Sci U S A 92(13): 5865-9.
Wishcamper, C. A., D. M. Brooks, et al. (2003). "Focal cerebral ischemia
upregulates SHP-1 in reactive astrocytes in juvenile mice." Brain Res 974(1-
2): 88-98.
163
Woo, M., R. Hakem, et al. (2003). "Caspase-3 regulates cell cycle in B cells: a
consequence of substrate specificity." Nat Immunol 4(10): 1016-22.
Wu, V. W., N. Nishiyama, et al. (1998). "A culture model of reactive astrocytes:
increased nerve growth factor synthesis and reexpression of cytokine
responsiveness." Journal of Neurochemistry 71(2): 749-56.
Wu, Y., A. Q. Zhang, et al. (2005). "Age related changes of various markers of
astrocytes in senescence-accelerated mice hippocampus." Neurochem Int
46(7): 565-74.
Yan, C., J. Chen, et al. (2000). "Overexpression of the cell death suppressor Bcl-w in
ischemic brain: implications for a neuroprotective role via the mitochondrial
pathway." Journal of Cerebral Blood Flow and Metabolism 20(3): 620-30.
Yan, X. X., J. Najbauer, et al. (2001). "Expression of active caspase-3 in mitotic and
postmitotic cells of the rat forebrain." The Journal of Comparative Neurology
433(1): 4-22.
Yang, J. Y. and C. Widmann (2001). "Antiapoptotic signaling generated by caspase-
induced cleavage of RasGAP." Molecular and Cellular Biology 21(16): 5346-
58.
Yang, J. Y. and C. Widmann (2002). "The RasGAP N-terminal Fragment Generated
by Caspase Cleavage Protects Cells in a Ras/PI3K/Akt-dependent Manner
That Does Not Rely on NFkappa B Activation." The Journal of Biological
Chemistry 277(17): 14641-6.
Yasuda, Y., N. Tateishi, et al. (2004). "Relationship between S100beta and GFAP
expression in astrocytes during infarction and glial scar formation after mild
transient ischemia." Brain Res 1021(1): 20-31.
Ye, J., L. Cao, et al. (2004). "The effects of ciliary neurotrophic factor on
neurological function and glial activity following contusive spinal cord injury
in the rats." Brain Res 997(1): 30-9.
164
Ye, Z. C., H. Sontheimer, et al. (1998). "Astrocytes protect neurons from neurotoxic
injury by serum glutamate." Glia. 22(3): 237-48.
Yi, X., J. Wang, et al. (2006). "Characterization of cell clones stably transfected with
short form caspase-9: apoptotic resistance and Bcl-XL expression." Mol Cell
Biochem 282(1-2): 1-12.
Zeinstra, E., N. Wilczak, et al. (2003). "Reactive astrocytes in chronic active lesions
of multiple sclerosis express co-stimulatory molecules B7-1 and B7-2." J
Neuroimmunol 135(1-2): 166-71.
Zermati, Y., C. Garrido, et al. (2001). "Caspase activation is required for terminal
erythroid differentiation." Journal of Experimental Medicine 193(2): 247-54.
Zeuner, A., A. Eramo, et al. (1999). "Caspase activation without death." 6(11): 1075-
80.
Zhu, H., A. Dahlstrèom, et al. (2005). "Characterization of cell proliferation in the
adult dentate under normal conditions and after kainate induced seizures
using ribonucleotide reductase and BrdU." Brain Res 1036(1-2): 7-17.

Abstract (if available)

Abstract Astrogliosis is a characteristic response of astrocytes to almost all forms of brain injury. Despite being studied extensively, the molecular basis of astrogliosis remains largely unknown. This dissertation investigates caspases as potential signaling molecules involved in mediating astrocyte reactivity. Chapter 2 documents data from two different experimental models of astrogliosis supporting the hypothesis of a non-apoptotic involvement of caspases in astrogliosis. Our results show astrocytes made reactive by treatment with dibutryl cAMP and Aß[superscript 25-35], demonstrate an increase in total caspase activity with a corresponding increase in the expression of active pro-apoptotic caspase-3 in the absence of cell death. In addition, caspase inhibition by zVAD, a broadspectrum caspase inhibitor resulted in a partial attenuation of the increased expressions of two known markers of reactivity, glutamine synthetase and fibroblast growth factor-2, thus suggesting a non-apoptotic role for caspases in mediating astrogliosis. We further extended the study to an ex-vivo model of astrogliosis, comprising of adult hippocampal astrocyte cultures generated from kainate-lesioned rats. Astrocytes from the kainate-lesioned animals exhibited reactive features in culture, and a non-apoptotic caspase activation similar to that observed in the in vitro model. Further, a preliminary analysis of the caspase family using specific inhibitors suggests caspase-11 and 3 might contribute to the caspase function in astrogliosis. Reactive astrocytes are known to upregulate expression of several different proteins with important functional consequences on the injured brain. Chapter 3 investigates two major outcomes: neurite outgrowth and neuronal survival following injury and how caspase activation in reactive astrocytes might influence them. Our findings suggest inhibition of neurite growth on reactive astrocytes might be partly caspase-dependent.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Progesterone receptors in the rat brain and their role in steroidal regulation of neurite outgrowth

PDF

Studies on plasticity and neurodegeneration in rat hippocampus

PDF

Glial support of neurite outgrowth: effects of aging and estrogen

PDF

The role of Schwann cells in the development of the neuromuscular junction

PDF

The roles of TRPM8 in cold sensation: the six sides of TRPM8

PDF

Estrogen and progesterone interactions in neurons: implications for Alzheimer's disease-related pathways

PDF

The role of glial cells in spinal muscular atrophy

PDF

New understanding of RIC-3, a regulator of alpha7 nicotinic acetylcholine receptor expression

PDF

The selective role of GFRα3 in cold pain

PDF

Role of neuronal nitric oxide synthase in aging and neurodegeneration

PDF

Role of oxidative stress in age-associated mild cognitive impairment and Alzheimer's disease

PDF

The role of endoplasmic reticulum protein GRP78 in normal hematopoeises and PTEN-null leukemogenesis

PDF

Synthetic small molecules and protein secondary structure mimetics as modulators of hypoxia-inducible transcription and integrin receptors function

PDF

The role of endoplasmic reticulum chaperones in adipogenesis, liver cancer and mammary gland development

PDF

Detecting joint interactions between sets of variables in the context of studies with a dichotomous phenotype, with applications to asthma susceptibility involving epigenetics and epistasis