University of Southern California Dissertations and Theses

Studies on plasticity and neurodegeneration in rat hippocampus

(USC Thesis Other)

Studies on plasticity and neurodegeneration in rat hippocampus

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

STUDIES ON PLASTICITY AND NEURODEGENERATION IN RAT
HIPPOCAMPUS

by

Miou Zhou

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 2008

Copyright 2008 Miou Zhou
ii
Acknowledgements
First I want to thank Dr. Michel Baudry. Dr. Baudry is always ready to help
me when I am lost in my experiments and cannot find the correct direction. Instead
of detailed instructions, he gives me related references and makes me think and work
independently. I really enjoy working under the direction of Dr. Baudry and have
learned a lot from his advanced research experience. Dr. Baudry is such a nice
person to work with that I feel I am really lucky to have him as my advisor.
I also want to thank all my committee members for their support and help in
my research. Dr Thompson and Dr Ko are always there to help me when I have
problems in my research, and they gave me very helpful suggestions for my poster
and oral qualification exam. Dr Liman taught me patch clamp and molecular
biological techniques when I was doing a rotation in her lab. Dr Qin opened his lab
to me and let me use his instrument freely. Without the help from my committee
members, I will not have been able to finish my thesis successfully.
I want to say thanks to all the members in Dr. Baudry lab. Sam is very
knowledgeable in science, and has never hesitated to share his ideas with me and all
other labmates. Wei Xu gave me many useful suggestions for my experiments and
set a good example for doing scientific research. Aaron Clausen showed me all the
detailed methods he developed for the measurement of oxidative stress. Anna Knize
provided me with healthy and beautiful hippocampal slice cultures, and the various
delicious cakes she made for all our lab member birthday parties made me feel our
lab was a big family. Reymundo Dominguez helped me operate the confocal
microscope. Yutien Hsu, Maggie Chou, Karoline Rostamiani, Claudia Aguirre, and
iii
Alexis Seegan also helped me a lot and their help made my experiments and my life
much easier in the lab.
I am so grateful to my parents who are always with me no matter how old I
am and where I go, even though I have had little amount of time to be with them in
the past several years. Last and most importantly, I want to thank my wife Shumin
for everything she has done for me. With her in my life, pain is less painful and
sweet is sweeter. She brings me happiness and is an inseparable part of my life.
iv
Table of Contents

Acknowledgements ii

List of Figures vii

Abstract xi

1. Chapter 1: Introduction 1

2. Chapter 2 10
2.1. Introduction 10
2.1.1. Cerebral ischemia 10
2.1.2. Different ischemic models 11
2.1.3. Reactive oxygen species (ROS) and superoxide dismutase/ 12
catalase mimetics
2.1.4. ATP depletion, necrosis and apoptosis in ischemia 18
2.1.5. MAPK 19
2.2. Materials and Methods 20
2.2.1. Acute hippocampal slice preparation 21
2.2.2. OGD treatment for acute hippocampal slices 21
2.2.3. Organotypic hippocampal slice culture 22
2.2.4. OGD treatment for cultured hippocampal slices 22
2.2.5. Cytosolic and nuclear proteins preparation 23
2.2.6. Western blots 23
2.2.7. Cell viability assays - LDH assay 24
2.2.8. Cell viability assays - PI uptake assay 24
2.2.9. ATP measurement 26
2.2.10. Evaluation of ROS production in slices 26
2.2.11. TBARS assay 27
2.2.12. Statistical analysis 27
2.3. Results and Discussion 28
2.3.1. acute hippocampal slices 28
2.3.1.1. EUK-189 and EUK-207 attenuate OGD-induced cell death 28
in acute hippocampal slices from 2-month-old rats
2.3.1.2. EUK-189 and EUK-207 partly blocked OGD-induced ATP 31
depletion in acute hippocampal slices
2.3.1.3. EUK-189 and EUK-207 completely blocked OGD-induced 33
ROS generation in acute hippocampal slices
2.3.1.4. The protective effect of EUK-207 against OGD-induced cell 36
death did not involve the ERK or the p38 pathway
2.3.1.5. Discussion 43
2.3.2. Organotypic hippocampal slice cultures 49
2.3.2.1. EUK-207 attenuated OGD-induced cell death in organotypic 49
hippocampal slice cultures
v
2.3.1.2. EUK-207 protects organotypic hippocampal slice cultures 53
from OGD-induced cell death by attenuation of OGD-
induced ROS generation, lipid peroxidation and AIF
translocation
2.3.1.3. Discussion 59

3. Chapter 3 65
3.1. Introduction 65
3.1.1. NMDA receptors 65
3.1.2. Calpain 67
3.1.3. PSD-95 and NMDA receptors 68
3.2. Materials and Methods 70
3.2.1. NMDA treatment 70
3.2.2. OGD treatment and LDH assay 70
3.2.3. PI uptake assay 71
3.2.4. Nissl staining 72
3.3. Results and Discussion 72
3.3.1. NMDA treatment elicits rapid excitotoxicity in acute hippo- 72
campal slices of young but not adult rats
3.3.2. NMDA-induced excitotoxicity in acute hippocampal slices from 76
1-week-old rats is mediated through NR2B-containing NMDA
receptors
3.3.3. Calpain activation is necessary for NMDA-induced excito- 79
toxicity in acute hippocampal slices of young rats
3.3.4. The lack of NMDA-induced toxicity in slices from adult rats is 83
not due to a priming effect
3.3.5. Oxygen-glucose deprivation (OGD) induces excitotoxicity in 84
acute slices from both young and adult rats
3.3.6. NMDA treatment induces calpain-mediated PSD95 truncation 88
3.4. Discussion 88

4. Chapter 4 94
4.1. Introduction 94
4.1.1. AMPA receptor positive modulaters 94
4.1.2. Brain-derived neurotrophic factor (BDNF) 96
4.1.2.1. Neurotropins 96
4.1.2.2. BDNF, protein translation and synaptic plasticity 97
4.2. Methods 100
4.2.1. Acute hippocampal slice preparation 100
4.2.2. Primary cortical neuronal cultures 101
4.2.3. Western blots 101
4.3. Results and Discussion 102
4.3.1. CNQX blocked CX614-induced increase in TrkB, mTOR or 102
4EBP1 phosphorylation in acute hippocampal slices

vi
4.3.2. K252a blocked BDNF- or CX614- induced increase in TrkB, 106
mTOR or 4EBP1 phosphorylation in acute hippocampal slices
4.3.3. Rapamycin blocked BDNF- or CX614- induced increase in 111
4EBP1 phosphorylation in acute hippocampal slices
4.3.4. CX614 had no effect on BDNF expression levels, and Nife- 111
dipine, calcium-free incubation solution, or ryanodine blocked
CX614-induced increase in TrkB phosphorylation in acute
hippocampal slices
4.3.5. Actinomycin D had no effect on CX614-induced increase in 115
TrkB phosphorylation, but blocked CX614-induced Arc
synthesis in primary cortical neuron cultures
4.3.6. BDNF and CX614 stimulate dendritic protein translation 116
4.4. Discussion 116

5. Chapter 5: Conclusions 124

References 133

Appendix: Publications 147

vii
List of Figures

Figure 1. Structure of EUK-207 and EUK-189 14

Figure 2. Effects of EUK-189 and EUK-207 on OGD-induced LDH 29
release in acute hippocampal slices from 2-month-old rats

Figure 3. Effects of EUK-189 and EUK-207 on OGD-induced increase 30
in PI uptake in different regions of acute hippocampal slices from
2-month-old rats

Figure 4. Effects of EUK-189 and EUK-207 on OGD-induced ATP 32
depletion in acute hippocampal slices from 2-month-old rats

Figure 5. Effects of EUK-189 and EUK-207 on OGD-induced ROS 34
generation in acute hippocampal slices from 2-month-old rats

Figure 6. Effects of U0126, EUK-189, or EUK-207 on OGD-induced 35
changes in ERK1/2 phosphorylation in acute hippocampal slices from
2-month-old rats

Figure 7. Effects of U0126, EUK-189, or EUK-207 on OGD-induced 37
LDH release in acute hippocampal slices from 2-month-old rats

Figure 8. Effects of U0126 or EUK-207 on ERK1/2 phosphorylation, 38
and on OGD-induced LDH release in acute hippocampal slices from
postnatal day 10 rats

Figure 9. Effects of EUK-207 on OGD-induced changes of p38 and 40
JNK phosphorylation in acute hippocampal slices from 2-month-old
and p10 rats

Figure 10. Effects of EUK-207, PD98059, and SB203580 on OGD- 41
induced LDH release in acute hippocampal slices from 2-month-old rats

Figure 11. Effects of EUK-207, PD98059, and SB203580 on OGD- 42
induced LDH release in acute hippocampal slices from postnatal day
10 rats

Figure 12. Effects of EUK-207 on LDH release elicited by different 50
periods of OGD in organotypic hippocampal slice cultures

Figure 13. Effects of different concentrations of EUK-207 on LDH 51
release elicited by 1 h OGD in organotypic hippocampal slice cultures

viii
Figure 14. Effects of pre- or post-treatment with EUK-207 on 1 h 52
OGD-induced LDH release and rate of LDH release at various times
after OGD in organotypic hippocampal slice cultures

Figure 15. Effects of EUK-207 on OGD-induced increase in PI uptake 54
in organotypic hippocampal slice cultures

Figure 16. ROS levels, measured with DCF fluorescence during and 56
after OGD

Figure 17. Effects of OGD and EUK-207 on lipid peroxidation in 57
cultured hippocampal slices

Figure 18. Effects of OGD and EUK-207 on apoptosis-inducing factor 58
(AIF) and cytochrome c (cyto c) levels in different subcellular fractions
in cultured hippocampal slices

Figure 19. Developmental changes of NR1, NR2A or NR2B subunits in 66
rat hippocampus

Figure 20. Effects of NMDA treatment on LDH release in acute hippo- 73
campal slices from rats of different ages

Figure 21. Effects of NR2A, NR2B antagonists on NMDA-induced 74
LDH release in acute hippocampal slices from rats of different ages

Figure 22. Effects of NMDA treatment on Propidium Iodide (PI) 75
staining in acute hippocampal slices from rats of different ages

Figure 23. Effects of NMDA treatment on Nissl staining in acute slices 77
from 1-week-old and 3-month-old rats

Figure 24. Effects of NMDA on calpain-mediated spectrin degradation 78
in acute hippocampal slices from rats of different postnatal ages

Figure 25. Effects of calpain inhibitor III on NMDA-induced calpain 80
activation and toxicity in acute hippocampal slices from rats of different
postnatal ages

Figure 26. Effects of pretreatment with APV or calpain inhibitor III on 81
NMDA-induced calpain activation in acute hippocampal slices of 3-
month-old rats

ix
Figure 27. Effects of pretreatment with APV or calpain inhibitor III on 82
NMDA-induced LDH release in acute hippocampal slices of 3-month-
old rats

Figure 28. Effects of oxygen/glucose deprivation (OGD) on LDH release 85
in acute hippocampal slices from 1-week-old and 3-month-old rats

Figure 29. Effects of oxygen/glucose deprivation (OGD) on calpain- 86
mediated spectrin degradation in acute hippocampal slices from 1-week-
old and 3-month-old rats

Figure 30. Effects of NMDA treatment on PSD95 levels in acute 87
hippocampal slices from 1-week-old and 3-month-old rats

Figure 31. Scheme of BDNF activated protein translation 98

Figure 32. Effects of CNQX on CX614-induced increase in TrkB 103
phosphorylation in acute hippocampal slices

Figure 33. Effects of CNQX on CX614-induced increase in mTOR 104
phosphorylation in acute hippocampal slices

Figure 34. Effects of CNQX on CX614-induced increase in 4EBP1 105
phosphorylation in acute hippocampal slices

Figure 35. Effects of K252a on BDNF or CX614-induced increase in 107
TrkB phosphorylation in acute hippocampal slices

Figure 36. Effects of K252a on BDNF or CX614-induced increase in 108
mTOR phosphorylation in acute hippocampal slices

Figure 37. Effects of K252a on BDNF or CX614-induced increase in 109
4EBP1 phosphorylation in acute hippocampal slices

Figure 38. Effects of rapamycin on BDNF or CX614-induced increase 110
in 4EBP1 phosphorylation in acute hippocampal slices

Figure 39. Effects of CX614 on BDNF expression levels in acute hippo- 112
campal slices

Figure 40. Effects of nifedipine, or calcium-free incubation solution on 113
CX614-induced increase in TrkB phosphorylation in acute hippocampal
Slices

x
Figure 41. Effects of ryanodine on CX614-induced increase in TrkB 114
phosphorylation in acute hippocampal slices

Figure 42. Effects of actinomycin D on CX614-induced increase in 117
TrkB phosphorylation and Arc protein levels in primary cortical
neuronal cultures

Figure 43. Effects of BDNF and CX614 on dendritic local protein 118
synthesis in primary cultured cortical neurons

Figure 44. Schematic representation of the mechanisms underlying 119
CX614-activated dendritic protein translation

xi
Abstract
Oxygen/glucose deprivation (OGD) is widely used as an in vitro model of
ischemia, and mechanisms underlying OGD-induced neuronal death are not
completely understood. I tested the participation of reactive oxygen species (ROS) in
OGD by using EUK-207, a synthetic superoxide dismutase/catalase mimetic. EUK-
207 provides neuroprotection against OGD-induced cell death both in acute and
cultured hippocampal slices. In cultured slices, this effect is related to decreased free
radical accumulation, reduced release of apoptosis-inducing factor and reduced lipid
peroxidation. In acute slices, protective effects of EUK-207 are also related to
elimination of free radical accumulation and partial reversal of ATP depletion.
Excitotoxicity induced by overactivation of NMDA receptors, a subtype of
ionotropic glutamate receptors, is involved in OGD-induced cell death in slices from
young but not adult rats. To better understand the mechanisms of excitotoxic cell
death, I studied the effects of NMDA treatment on acute hippocampal slices from
both neonatal and mature rats, and in particular, the role of calpain-mediated spectrin
degradation. NMDA treatment results in cell death and spectrin degradation in slices
from young but not adult rats, effects that are partly abolished by NMDA receptor
subunit NR2B antagonists and by calpain inhibitor III, but not affected by a NR2A
specific antagonist, suggesting that developmental changes in NMDA receptor
subunit composition contribute to developmental changes in NMDA toxicity and
possibly OGD-induced cell death.
While overactivation of NMDA receptors is involved in neurodegeneration,
modulation of AMPA receptors, another ionotropic glutamate receptor subtype, is
xii
involved in synaptic plasticity. Positive AMPA receptor modulators (such as CX614)
facilitate LTP induction and improve performance in several learning tasks. We
demonstrated that CX614 rapidly activates the dendritic translation machinery and
increases the dendritic expression of Arc, a translation reporter, in a brain-derived
neurotrophic factor- (BDNF) dependent manner. AMPA receptor activation,
extracellular calcium influx, intracellular calcium release, and activation of BDNF
receptor TrkB, are all required for CX614-induced effects on protein translation,
indicating that BDNF secretion is involved in these effects. Our results demonstrate
that positive modulation of AMPA receptors stimulates AMPA receptor- and TrkB
receptor- dependent dendritic protein translation, an effect mediated by BDNF
secretion.
1
1. Chapter 1: Introduction
Hippocampus plays an essential role in learning and memory and numerous
studies have been directed at understanding the mechanisms underlying these
processes. In particular, the role of protein synthesis and more specifically of
dendritic protein synthesis has been extensively debated, although it remains unclear
how it is triggered and how it contributes to synaptic plasticity. Some of the studies
in this dissertation have been directed at this question (Part III). Hippocampus is also
extremely vulnerable to numerous stressors including excitotoxicity and stroke, and
the mechanisms underlying vulnerability of hippocampal neurons to stress are still
not completely understood. Other studies in this dissertation are concerned with this
problem (Part I & II).
Stroke is the rapid interruption of blood supply to the brain. There are two
types of stroke, ischemic and hemorrhagic strokes, with the former accounting for
70% to 80% of stroke cases (Tabakman et al., 2005). With the model of
oxygen/glucose deprivation (OGD) on hippocampal slices, my study mainly focused
on mechanisms involved in ischemic cell death. Although there has been numerous
studies on ischemic therapy, few drugs have shown neuroprotective effects due to a
narrow window of effective drug application after the onset of symptons. This failure
of finding effective drugs is in part due to the rapid, severe, and irreversible damage
ischemia elicits in the cerebral regions that are affected by the shortage of blood
supply and thus experience low levels of oxygen and energy source.
The process of cell death after ischemia and reperfusion is very complicated,
because numerous cellular events such as decreased ATP levels, generation of ROS,
2
synaptic release of glutamate, activation of calcium-dependent protease, activation of
apoptotic pathways and others are engaged by the initial decrease in oxygen and
energy supplies (Koistinaho and Koistinaho, 2005;Zipp and Aktas, 2006;Won et al.,
2002). In addition to the complexity of the various events involved in ischemic cell
death, there are many crosstalks among those events, which make it more difficult to
understand the process of ischemic cell death. To alleviate ischemia-induced damage,
it is important to find key factors that are the main cause of ischemic cell death and
to discover drugs which could stop the activation of these key factors, increase the
tolerance of cerebral neurons to low oxygen and energy, and extend the window for
clinical therapy.
It has been proposed that among all events that are involved in ischemic cell
death, ROS generation is the key factor contributing to necrotic cell death and other
types of ischemic cell death (Lipton, 1999). Whether excess ROS are produced
mainly during or after ischemia is not clear. While ROS levels are elevated during
ischemia (Piantadosi and Zhang, 1996), some studies have suggested that more free
radicals are generated during the early period of reperfusion (Piantadosi and Zhang,
1996;Dirnagl et al., 1995;Oliver et al., 1990). Once generated, ROS cause lipid
peroxidation, protein oxidation and DNA mutations (Chen and Yu, 1994;Ikeda and
Long, 1990;Kehrer, 1993), and ROS also induce increased glutamate release by
inhibiting Na
+
-K
+
-APTase.
In my work, I used a synthetic superoxide dismutase/catalase mimetic EUK-
207, which eliminates both superoxide and hydrogen peroxide, to answer several
questions: i) Do ROS scavengers provide protection in hippocampal slices subjected
3
to OGD and recovery? ii) What is the optimal time for EUK-207 treatment to
achieve maximal protection and when are ROS generated during OGD and recovery?
iii) What are the contributions to neuronal death of other events such as ATP
depletion, activation or inactivation of mitogen-activated protein kinases (MAPKs),
and overactivation of NMDA receptors? iv) Do those events play different roles in
OGD-induced cell death in acute hippocampal slices prepared from newborn and
adult rats? My results show that EUK-207 significantly reduces OGD-induced cell
death by inhibiting free radical accumulation both in acute and cultured hippocampal
slices. Although it has been generally assumed that more free radicals are generated
during reperfusion, my results indicate that increased formation of reactive oxygen
species takes place relatively early during OGD, and EUK-207 produces better
neuroprotection when it is present in slices at the beginning of the OGD period than
only during the recovery period.
The direct consequence of lack of oxygen and energy during ischemia is the
decrease of ATP levels in living cells, which takes place within seconds to minutes
after the onset of ischemia. It has been reported that neonatal animals show
anaerobic ATP production and conserved energy during ischemia (Bickler et al.,
1993;Folbergrova, 1993). It was therefore interesting to test whether ATP levels
exhibited a different pattern of changes with OGD treatment in acute hippocampal
slices from newborn rats compared to those from adult rats, and whether ATP
depletion was exacerbated by ROS generation. My results indicate that in acute
hippocampal slices from 2-month-old rats, ATP levels decrease rapidly with OGD
treatment, and significant ATP recovery is observed when EUK-207 is applied both
4
during OGD and recovery. In acute hippocampal slices from postnatal day 10 (p10)
rats, the decrease of ATP levels during OGD is relatively slower than slices from
adult rats, thus supporting the notion that neonatal animals are better able to maintain
ATP levels under OGD conditions.
The role of MAPKs in ischemic cell death is still ambiguous. Depending on
the ischemic models, MAPKs have been reported to be either activated or inactivated
following ischemia and reperfusion, and activation of this pathway has been reported
to promote neuronal survival as well as cell death (Namura et al., 2001;Zhu et al.,
2005;Murray et al., 1998;Fahlman et al., 2002). My results indicate that the
extracellular signal-regulated kinases 1 and 2 (ERK1/2, one member of the MAPKs)
is inactivated after OGD and recovery in slices from 2-month-old rats, but remains
unchanged in hippocampal slices prepared from p10 rats. This effect can be related
to a slower decrease of ATP levels in slices from neonatal rats compared to those
from adult rats after OGD treatment.
Glutamate is the major excitatory neurotransmitter in the central nervous
system. Ionotropic glutamate

receptors are classified into three subtypes, the N-
methyl-D-aspartate (NMDA), alpha-amino-3-hydroxy-5-methylisoxazole-4-
propionate (AMPA), and kainate (KA) receptors (Collingridge and Lester, 1989).
Although NMDA receptors (NMDARs) play a crucial role in neuronal development
and plasticity, overactivation of NMDARs causes excitotoxic cell death (Dingledine
et al., 1999). During ischemia, the rapid decrease in ATP levels in cells induces the
collapse of the mitochondrial membrane potential, malfunction of the Na
+
-K
+
-
APTase, reverse functioning of glutamate transporters, and overactivation of NMDA
5
receptors. It was therefore interesting as well to study whether similar processes were
taking place in hippocampal slices from neonatal and adult rats by OGD. My results
show that 1 h OGD treatment induces equivalent increase in cell death in acute
hippocampal slices from 1-week-old and 3-month-old rats. However, OGD-induced
toxicity in slices of newborn or adult rats involves different mechanisms, since
MK801, a NMDA receptor antagonist, partly blocks OGD-induced cell death in
slices from newborn rats, but had no effect on slices from adult rats, suggesting that
activation of NMDA receptors is involved in OGD-induced cell death in acute
hippocampal slices from newborn, but not adult rats.
As my results suggested that activation of NMDA receptors differentially
contributed to OGD-induced cell death in slices from newborn and adult rats, I
studied developmental changes in NMDA-induced excitotoxicity in acute slices from
rats of different ages. NMDA treatment induces a time-dependent increase in cell
death in slices from 1-week-old rats, and NMDA-induced toxicity decreases with age.
It has been reported that the expression of NMDA receptors undergoes subunit- and
region-related changes during postnatal development, with high NR2B (NMDA
receptor subunit 2B) and low NR2A (NMDA receptor subunit 2A) expression after
birth, and increased expression of NR2A and decreased NR2B during postnatal
development (Liu et al., 2004;Sans et al., 2000). Although NR2A- and NR2B-
containing NMDA receptors are involved in different functions in synaptic plasticity
(Kim et al., 2005;Krapivinsky et al., 2003;Liu et al., 2004), the roles of NR2A and
NR2B subunits in NMDA-induced excitotoxicity are not clear. In my experiments,
with specific NR2A and NA2B antagonists, I demonstrated that it is NR2B but not
6
NR2A subunits that play an important role in NMDA-induced excitotoxicity and
possibly in OGD-induced cell death in acute slices from newborn rats, and my work
is the first to use the model of acute hippocampal slices to relate developmental
changes of NMDA-induced toxicity to developmental changes of NMDA receptor
composition.
One of the consequences of NMDA receptor activation is an increase in
intracellular calcium levels and activation of calcium-dependent proteases such as
calpains. Although calpains have been implicated under physiological conditions in
synaptic modifications and plasticity, they are also involved in oxidative stress and
neuronal death (Ray et al., 2000;Kelly and Ferreira, 2006;Lynch and Baudry, 1987);
however, the role of calpain in excitotoxicity has been controversial, with studies
reporting clear evidence for the involvement of this protease, while others reporting
the opposite (Bizat et al., 2003; Korhonen et al., 2005). It has been shown that from
newborn to adult animals, there is a switch from anaerobic to aerobic metabolism
during development, and this switch could contribute to differences between young
and mature cells in balancing intracellular calcium levels after the breakdown of
calcium homeostasis (Baudry and Lynch, 1985). It was therefore possible that
newborn and adult rats exhibit different pattern of calpain activation with NMDA
treatment. My results indicate that NMDA rapidly activates calpain in slices from
neonatal but not adult rats, therefore further supporting the notion that mechanisms
of NMDA-induced toxicity are developmentally regulated. Inhibiting calpain activity
significantly reduces NMDA-mediated neurotoxicity in slices from neonatal rats,
7
establishing that calpain activation plays a significant role in neurotoxicity in
neonatal rat brain.
While overactivation of NMDA receptors is involved in neurodegeneration,
overactivation of AMPA receptors, another subtype of ionotropic glutamate receptors,
has also been reported to induce an increase in intracellular calcium levels and
calpain-mediated programmed cell death and edematous necrosis in cerebellar
neurons (Chen et al., 2007;Mansouri et al., 2007). However, a recent study indicated
that positive modulation of AMPA receptors with aniracetam together with a high
concentration of AMPA (500 µM) protected neurons against glutamate excitotoxicity,
an effect mediated by the upregulation of brain-derived neurotropic factor (BDNF)
release (Wu et al., 2004). BDNF is a neurotrophic factor that promotes neurite
outgrowth and differentiation during development, and stimulates neurogenesis and
mediates neuronal survival in mature nervous system (Chao, 2003). BDNF has also
been reported to stimulate dendritic protein translation through the activation of the
mammalian target of rapamycin (mTOR) pathway in primary cortical neurons (Takei
et al., 2004). Regulation of dendritic translation plays a critical role in synaptic
plasticity, in particular in long-term potentiation (LTP) and long-term depression
(LTD), and inhibition of mRNA translation prevents memory consolidation (Huber et
al., 2000;Pfeiffer and Huber, 2006;Wells and Fallon, 2000). BNDF has been shown
to play a key role in synaptic plasticity and learning (Jourdi et al., 2003;Rex et al.,
2007), possibly through the regulation of dendritic protein translation. Thus, we were
interested to determine whether positive modulation of AMPA receptors could be
involved not only in excitotoxicity and neurodegeneration, but also in modulation of
8
protein translation and synaptic plasticity. In addition, we tested the hypothesis that
AMPA receptor-mediated upregulation of dendritic protein translation could be
mediated by BDNF release.
In our experiments, CX614, one of the positive AMPA receptor modulators
(A.K.A ampakines), was used to test the effect of positive modulation of AMPA
receptors on protein translation. Ampakines are small molecules that allosterically
modulate the properties of AMPA receptors (rates of channel opening/closing, and
desensitization) and potentiate AMPA receptor-mediated current (Arai et al., 1996).
Our results indicate that in acute hippocampal slices or primary neuronal cultures,
acute CX614 treatment activates mTOR and targets downstream of mTOR, such as
4EBP1 and p70S6K, which are involved in protein translation. CX614 also rapidly
increases the dendritic synthesis of myristoylated GFP, a translation reporter,
confirming that positive modulation of AMPA receptors increases dendritic protein
translation, mainly through increased BDNF release but not total BDNF protein
levels, which is in good agreement with the report of Kolarow (Kolarow et al., 2007).
Ampakines have been reported to increase learning and memories in rats
tested in a radial maze task and an odor-matching task (Staubli et al., 1994), and also
be effective in monkeys and humans for the treatment of schizophrenia and for
alleviating the impairment of performance due to sleep deprivation (Lynch,
2006;Goff et al., 2001;Porrino et al., 2005). The exact mechanisms responsible for
ampakine-mediated learning enhancement are not completely clear. Our results show
for the first time that positive modulation of AMPA receptors activates the protein
translation machinery and induces dendritic translation of proteins such as Arc and
9
CamKII, which are involved in synaptic plasticity. Therefore, our results provide a
new explanation for the effects of positive AMPA receptor modulators on facilitation
of LTP and improved learning and memory. In particular, unlike the upregulation of
BDNF by aniracetam, which requires the presence of high AMPA concentrations,
CX614 increases dendritic protein translation without the need for exogenous AMPA
receptor agonists, indicating that stimulation of protein translation with CX614 can
occur at basal/physiological levels of glutamate release. Unlike BDNF, ampakines
are orally bioactive and able to cross the blood-brain barrier (Lynch, 2002). Their
abilities to increase the levels of neurotrophins such as BDNF make them an
important candidate to improve cognitive functions in humans and provide a
potential therapeutic method to treat neurodegenerative diseases such as mild
cognitive impairment and Alzheimer’s disease.

10
2. Chapter 2
Protective effects of superoxide dismutase/catalase mimetics against
oxygen/glucose deprivation-induced cell death in hippocampus

2.1. Introduction
2.1.1 Cerebral ischemia
Stroke is a rapid loss of brain function due to the disturbance of blood supply
to the brain, and it is the third leading cause of death in the United States. There are
two types of stroke, ischemia (failure of blood supply) and hemorrhage (bleeding
into the subarachnoid space), and ischemia accounts for 70% to 80% of stroke cases
(Tabakman et al., 2005). Cerebral ischemia is the block of oxygenated blood supply
to the brain usually due to blood clots, and it triggers a variety of pathological events,
including excitotoxicity, inflammation, delayed neuronal dysfunction and cell death,
and has also been implicated in Alzheimer’s and other neurodegenerative diseases
(Koistinaho and Koistinaho, 2005;Zipp and Aktas, 2006).
The mechanisms underlying ischemic neuronal death are not completely
understood. Ischemia and reperfusion can induce a lot of structural and biochemical
changes that are complicated due to many interactions between these changes.
Cerebral ischemia has been shown to elicit inhibition of oxidative phosphorylation
with rapid decrease of ATP levels, excessive generation of reactive oxygen species
by the mitochondrial respiratory chain, membrane depolarization as lack of ATP
substrate for the Na
+
-K
+
pump, release of glutamate followed by overactivation of
11
glutamate receptors, increased intracellular calcium concentration and breakdown of
intracellular calcium homeostasis, activation of calcium dependent proteases, and
ultimately cell death by necrosis or apoptosis (Koistinaho and Koistinaho, 2005;Zipp
and Aktas, 2006;Won et al., 2002). Depending on different ischemia, mild or severe,
it is uncertain that whether all these events happen in ischemia-induced cell death, or
just parts of these various changes are involved. The exact temporal sequence and the
relative importance of these different events remain far from being understood. The
involvement of all those events in acute and cultured hippocampal slices during and
after in vitro ischemia is investigated in my experiments, with a specific focus on
ischemia-induced free radical generation.

2.1.2. Different ischemic models
Different models have been used to study ischemia-induced damage. In in
vitro model, primary neuronal or glial cultures from cortex, hippocampus, or
cerebellum are exposed to N
2
/CO
2
equilibrated bath solution with (anoxia or hypoxia)
or without glucose (oxygen/glucose deprivation). Since 1995, organotypic
hippocampal slice cultures (OHSCs) have been used to study ischemic damages.
Compared to dissociated neuronal cultures, OHSCs have become a more valuable
model because they are believed to better mimic in vivo conditions (Strasser and
Fischer, 1995). Although not used so popularly, acute hippocampal slices are also a
good model to study ischemic damages because acute slices can be obtained from
both neonatal and mature animals, and it is a good model to study developmental
changes of ischemic damage from young to adult animals. However, in acute slices,
12
studies of the long effects (hours to days) of the delayed ischemic damages are
limited. On the contrary, tissue from prenatal or juvenile rats or mice is required for
organotypic slice cultures, but they are a good model to study ischemia-induced
cellular changes in a relatively long time. In my experiments, both acute and cultured
hippocampal slices are used to study acute and delayed effects of ischemic cell death
and also to test some developmental changes of ischemic cell death between
newborn and adult rats.
There are three main in vivo ischemic models, global ischemia, focal
ischemia and hypoxia/ischemia. In global ischemic model, normally four-vessel
occlusion or two-vessel occlusion combined with hypotension is applied in rats,
while two-vessel occlusion is performed in gerbils. In focal ischemic models, one
middle cerebral artery is occluded, and sometimes combined with carotid artery
occlusion. Compared with global ischemia, in focal ischemia there are graduated
damages from the core of the lesion to its surrounded region, and the blood flow and
tissue vulnerability to ischemia are different from core to its penumbra, so focal
ischemic insult is more complex than global ischemia. In hypoxia/ischemia model,
unilateral carotid artery occlusion is combined with hypoxia, and this model can
produce similar effect as focal ischemia when expose rats to 60 min 8% hypoxia or
as global ischemia when expose rats to 15 min 3% hypoxia (Williams et al., 1994).

2.1.3. Reactive oxygen species (ROS) and superoxide dismutase/catalase
mimetics

13
Superoxide, and hydrogen peroxide are two important ROS produced mainly
within the mitochondria from both Complexes I and III of the electron transport
chain (Cadenas and Sies, 1998). At moderate concentrations, ROS are involved in
defense against infectious bacteria and in a number of cellular signaling pathways
(Valko et al., 2004). However, overproduction of ROS results in damage to cellular
lipids, proteins, or DNA, and it has been reported to induce both necrosis and
apoptosis (Henderson et al., 2006;Noh et al., 2006). Antioxidants that can scavenge
ROS have been tested in cancer, aging, and different diseases induced by ROS.
Free radicals, such as superoxide radicals and hydroxyl radicals, are
important ROS involved in ischemic cell death. Under normal conditions, the rate of
free radicals formation is equal to that of their elimination. During ischemia and
reperfusion, the balance is broken either by increased free radical production or
decreased activity of cellular defense system. Whether excess ROS are produced
mainly during or after ischemia is not clear. It has been reported that increased free
radical formation happens both during 15 min global ischemia (Piantadosi and Zhang,
1996) and during 30 min focal ischemia (Kinuta et al., 1989), however, it has been
suggested that more free radicals are generated during the early period of reperfusion
(Piantadosi and Zhang, 1996;Dirnagl et al., 1995;Oliver et al., 1990). In my
experiments, the free radical scavenger is applied before, during, and after OGD
treatment to find when is the best time to eliminate excess generation of free radicals
and thus to provide maximal protection.
Although there are several mechanisms about the source of free radical
production during ischemia, including accumulation of xanthine/hypoxanthine
14

Fig 1. Structure of EUK-207 and EUK-189 (from Liu, 2003).
15
during global ischemia, accumulation of arachidonic acid via the cyclooxygenase or
lipoxygenase pathway during global and focal ischemia (Tegtmeier et al., 1990),
excess free radicals are believed to be mainly generated in mitochondria with
depletion of ATP, rise of Ca
2+
, and the opening of mitochondria transition pore
(MTP). The very reactive peroxynitrite is generated by the reaction of NO and
superoxide, and is also increased 4 h after 30 min global ischemia in rats (Forman et
al., 1998). Once the levels of free radicals increase during ischemia and reperfusion,
this status can persist for many hours and even days. The exact relations between
free radical generation and cell death are not clear. Free radicals react strongly with
unsaturated lipids with the formation of peroxides and aldehydes, which decrease
membrane fluidity (Chen and Yu, 1994;Ikeda and Long, 1990;Kehrer, 1993). Free
radicals also oxidize protein side chains with the formation of carbonyl groups, and
interact with DNA to produce single-strand breaks in DNA, which activate poly
(ADP-ribose) polymerase (PARP), deplete NAD levels and cause the inhibition of
mitochondria function (Szabo, 1996). During ischemia, free radicals might also
oxidize Na
+
-K
+
-APTase and make it ready to be degraded by calpain (Zolotarjova et
al., 1994). The dysfunction of Na
+
-K
+
-APTase leads to cell depolarization, an
increase in glutamate release, and the breakdown of calcium homeostasis, with a
resultant cell death by necrosis or apoptosis.
In my experiments, the participation of reactive oxygen species in ischemia
was tested by using two salen-manganese (Mn) complexes, EUK-189 and EUK-207.
EUK-189 and EUK-207 have been shown to act as superoxide dismutase/catalase
mimetics, thus eliminating both superoxide and hydrogen peroxide. The protective
16
effects of exogenously administrated superoxide dismutase (SOD) or catalase have
been tested in several studies (Liu et al., 1989;Armstead et al., 1992), and mild
protective effects were observed when these enzymes were applied before ischemic
damage, and there was no protection when enzymes were applied after ischemia. The
limited protective effect might be related to the low plasma stability of these
enzymes or to the difficulty to deliver these enzymes to the injury site.
To mimic the antioxidant properties of endogenous enzymes, three major
types of SOD mimetics have been developed. They are macrocyclic Mn complexes,
salen Mn complexes, and Mn porphyrin complexes (Pong et al., 2001). The
macrocyclic Mn complexes only show specificity for superoxide, while the salen Mn
complexes, and Mn porphyrins have been reported to show both superoxide and
hydrogen peroxide scavenging activities (Doctrow et al., 2002). The salen-
manganese complexes are the main focus of my study, including EUK-8, the
prototype molecule and one of the most water soluble compound of the EUK series;
EUK-134, with significantly higher catalytic activities than EUK-8; EUK-189 and
EUK-207, with similar catalase activity and higher lipophilicity and neuroprotective
activity when compared to EUK-134 (Liu et al., 2003;Baker et al., 1998).
Compared to endogenous free radical eliminating enzymes, EUK compounds
have lower molecular weight and are easier to be delivered to the site that is under
stress of excess free radical generation. The EUK compounds have shown efficacy in
different disease models associated with ROS formation. For example, EUK-134
reduced 1-methyl-4-phenylpyridinium (MPP
+
) and 6-hydroxydopamine (6-OHDA)-
induced toxicity in primary dopaminergic neuronal cultures, suggesting a potential
17
therapeutic role for EUK compounds in the treatment of Parkinson’s disease (Pong et
al., 2000). EUK-8 and EUK-134 reduced levels of oxidative stress and prolonged
survival chance in a mouse amyotrophic lateral sclerosis model (Jung et al., 2001).
EUK-134 and EUK-189 protected cultured cortical neurons from staurosporine-
induced oxidative stress, mitochondrial dysfunction, and neuronal apoptosis (Pong et
al., 2001); EUK-134 also completely blocked excessive generation of reactive
oxygen species induced by zinc and reduced subsequent oxidative damage in
cultured cortical neurons (Pong et al., 2002). EUK-8 and EUK-134 have been
reported to attenuate hemorrhage- and resuscitation-induced liver and pancreatic
injury and renal dysfunction, suggesting EUK compounds might be one therapeutic
approach for multiple organ failure in hemorrhagic shock (Izumi et al., 2002). In an
Alzheimer’s disease (AD) model with the activation of microglia by 42-amino-acid
form of the beta-amyloid peptide, EUK-8, EUK-134 and EUK-189 all provided
protection (Anderson et al., 2001). Reactive oxygen species have also been
suggested to be involved in aging, and it has been proved that when ROS were
eliminated with EUK-134 or EUK-189, the life-span of Caenorhabditis elegans was
increased by a mean of 44 percent, and the life-span of superoxide dismutase 2 (sod2)
nullizygous mice, which exhibited mitochondrial defects and severe tissue
pathologies, was extended by threefold (Melov et al., 2000;Melov et al., 2001).
Oxidative stress is involved in cognitive impairment and the loss of learning and
memory function in aged people. It has been reported that with chronic application
of EUK-189 or EUK-207, the cognitive deficits and protein oxidation were almost
completely reversed, and lipid peroxidation was also significantly reduced, in 11-
18
month-old mice (Liu et al., 2003) or in 23-month-old mice (Clausen et al., 2008).
Different EUK compounds differ in their SOD activity, catalase activity, lipophilicity
and stability, and all these properties determine their neuroprotective efficiency.

2.1.4. ATP depletion, necrosis and apoptosis in ischemia
Ischemia induces rapid ATP depletion, and ATP depletion further contributes
to the collapse of the mitochondrial membrane potential, the opening of
mitochondrial permeability transition pore (MPTP), the dissipation of
transmembrane electro-chemical mitochondrial gradient, and the release of small
proteins such as cytochrome c from mitochondria (Fiskum et al., 1999;Murphy et al.,
1999). It has been suggested that the degree of ATP depletion determines the form of
death cell exhibits, with gradual ATP depletion after mild ischemia inducing caspase-
3 release and apoptosis, and rapid ATP depletion with severe ischemia eliciting
cytochrome c release and necrosis (Li et al., 1997;Saikumar et al., 1998). Drugs that
prevent ATP depletion or promote ATP recovery after ischemia have been reported to
increase cell survival (Riepe et al., 1997;Galeffi et al., 2000).
Ischemia may induce cell death either through necrosis or apoptosis. Necrosis
is a rapid form of cell death that is due to the breakdown of ionic homeostasis and
apoptosis is a form of relatively delayed cell death involving genetic programs,
although there is no distinct time point during reperfusion to separate necrotic from
apoptotic cell death (Banasiak et al., 2000). Normally necrosis takes place when cells
suffer from extreme stress and it is characterized by cytoplasmic swelling, cell
membrane rupture, and the release of intracellular contents to neighboring cells
19
(Majno and Joris, 1995). Compared to necrosis, the morphological changes of
apoptosis are easier to recognize and are characterized by chromatin condensation,
DNA fragmentation, plasma membrane blebbing, formation of apoptotic bodies,
caspase activation, and the release of mitochondrial intermediate proteins (Van Loo
et al., 2002;Krysko et al., 2008).
Generally the apoptosis-inducing factor (AIF) is localized in mitochondria,
and in OGD-treated neuronal cultures and in focal and global ischemia AIF has been
reported to be translocated from mitochondria to the nucleus (Cao et al.,
2003;Komjati et al., 2004), which induces chromatin condensation, DNA
fragmentation and apoptosis that is independent of caspase activity (Daugas et al.,
2000). In transient and permanent cerebral ischemia, cytochrome c is also released
from mitochondria to cytosol (Perez-Pinzon et al., 1999;Sugawara et al., 1999),
where cytochrome c binds to Apaf-1, recruits procaspase-9 to form the apoptosome,
and initiates further apoptotic cascades (Li et al., 1997). In my experiments, the
release of AIF and cytochrome c from mitochondria in the presence or absence of
EUK-207 during early recovery is tested to see if OGD-induced activation of pro-
apoptotic factors is related to ROS generation.

2.1.5. MAPK
Mitogen-activated protein kinases (MAPKs) are another type of intracellular
pathway frequently implicated in mechanisms of cell death and survival. MAPKs
consist of the extracellular signal-regulated kinases 1 and 2 (ERK1/2), p38, and
stress-activated protein kinases (SAPKs)/c-Jun N-terminal kinase (JNK). In
20
particular, the role of ERK1/2 in ischemia remains ambiguous, as ERK1/2 has been
shown to be either activated or inactivated following ischemia and reperfusion
depending on the different ischemic models, and activation of this pathway has been
reported to promote neuronal survival as well as cell death (Namura et al., 2001;Zhu
et al., 2005;Murray et al., 1998;Fahlman et al., 2002). MEK1/2 is a serine/threonine
protein kinase that activates ERK1/2, and MEK1/2 inhibitors such as U0126 and
PD98059, which are used in my experiments, are widely used to study the role of
MEK/ERK in different animal ischemic models (Wang et al., 2003;Namura et al.,
2001).
P38 and JNK are also involved in cellular responses to stress such as cerebral
ischemia, and p38 inhibition has been shown to provide neuronal protection in
cerebral ischemia (Sugino et al., 2000;Barone et al., 2001), although p38 activation is
also involved in neuronal protection against some insults (Claytor et al., 2007;Lin et
al., 2006). JNK is composed of JNK1, JNK2, and JNK3, with JNK1 and JNK2 found
in most cells and tissues, while JNK3 found mainly in the brain, heart and testis
(Bode and Dong, 2007). In a middle cerebral artery occlusion model, intraventricular
administration of a cell-penetrating JNK inhibitory peptide (D-JNK1), at 6 h after
occlusion significantly blocked lesion volume by 90% (Borsello et al., 2003). JNK3
has been reported to be all activated during cerebral ischemia, and when it was
inhibited by K252a without any effects on protein expression levels, ischemic cell
death in CA1 pyramidal cells was greatly reduced (Pan et al., 2005).

2.2. Materials and Methods
21
2.2.1. Acute hippocampal slice preparation
Hippocampi were rapidly dissected from postnatal 10-day or 2-month old
Sprague–Dawley rats, submerged in chilled cutting medium containing (in mM): 220
sucrose, 20 NaCl, 2.5 KCl, 1.25 NaH
2
PO
4
, 26 NaHCO
3
, 10 glucose, 2 ascorbic acid,
2 MgSO
4,
bubbled with 95% O
2
–5% CO
2
, and then cut into transverse slices (400
µm thick) using a McIlwain tissue chopper. After isolation, hippocampal slices were
placed in incubation baskets in an artificial cerebral-spinal fluid (aCSF) containing
(in mM): 124 NaCl, 5 KCl, 1.25 NaH
2
PO
4
, 26 NaHCO
3
, 2 ascorbic acid, 10 glucose,
1.5 MgSO
4
, 2.5 CaCl
2
, saturated with 95% O
2
–5% CO
2
and incubated for a 1 h–
recovery period at 37 °C.

2.2.2. OGD treatment for acute hippocampal slices
After 1 h incubation, hippocampal slices were washed twice with OGD
solution containing (in mM): 124 NaCl, 5 KCl, 1.25 NaH
2
PO
4
, 26 NaHCO
3
, 1.5
MgSO
4
, 2.5 CaCl
2
(pH 7.5), and then transferred into individual vials with 2 slices
per vial in 1.5 ml OGD solution, previously bubbled with 95% nitrogen–5% CO
2
for
30 min. Hippocampal slices were incubated in OGD solution at 35
o
C for different
time in anaerobic vials saturated with 95% nitrogen–5% CO
2
; in some cases, slices
were then collected and processed for western blots. For LDH assay, incubation
medium was collected after the 2 h incubation period, and slices were washed with
aCSF and further incubated for 2.5 h in 1.5 ml fresh aCSF solution saturated with
95% O
2
–5% CO
2
. When hippocampal slices were treated with drugs dissolved in
22
DMSO, DMSO was also added to the control group at the same final concentration
as used for the drugs (0.05%–0.1%).

2.2.3. Organotypic hippocampal slice culture
Organotypic hippocampal slice cultures (OHSC) were prepared from
postnatal 7-9 day old Sprague-Dawley rats according to the method described before
(Stoppini et al., 1991). Hippocampi were rapidly dissected and transverse slices (400
µm thick) were prepared with a McIlwain tissue chopper and placed on porous
Millipore membrane inserts in a 6-well plate with each well containing 1ml culture
medium (50% basal medium eagle (BME), 25% horse serum (HS), 25% Earle’s
Balanced Salts (EBSS), 1mM glutamine, 25 mM HEPES, 15 mM glucose, 2 mM
NaHCO
3
, 100 U/ml penicillin and 100 µg/ml streptomycin, pH 7.2). Culture medium
was changed three times a week and slices were cultured for 2 weeks at 35 °C
saturated with 5% CO
2
.

2.2.4. OGD treatment for cultured hippocampal slices
Cultured hippocampal slices were washed twice with sterile PBS, and then
transferred into new six-well plates with each well containing 1ml of glucose-free
and serum-free DMEM which had been pre-bubbled with 95% N
2
–5% CO
2
for 30
minutes. Hippocampal slices were incubated at 36 °C for 1 h in an anaerobic
chamber saturated with 95% N
2
–5% CO
2
.

At the end of the OGD period,
hippocampal slices were washed twice and kept in serum-free culture medium for 3
h, 6 h, or 24 h before cell viability assay.
23

2.2.5. Cytosolic and nuclear proteins preparation
Hippocampal slices were homogenized gently in a buffer containing: 20 mM
HEPES (pH 7.5), 250 mM sucrose, 10 mM KCl, 1.5 mM MgCl
2
, 1 mM EDTA, 1
mM EGTA, 0.5 mM dithiothreitol (DTT), 0.5 mM phenylmethylsulphonyl fluoride
(PMSF), 2 mg/mL leupeptin, and 1:200 protease inhibitor cocktail (Sigma).
Homogenates were centrifuged at 500 g for 5 min at 4 °C and the pellet was used for
western blots to detect nuclear AIF. The supernatants were centrifuged at 1000 g for
10 min at 4 °C, and then the supernatants after centrifuge (1000 g) were centrifuged
at 8000 g for 30 min at 4 °C. The supernatants after centrifuge (8000 g) were further
centrifuged at 100000 g for 60 min at 4 °C and the resulting supernatants were used
as the cytosolic fraction for analysis of cytochrome c release.

2.2.6. Western blots
Hippocampal slices were homogenized and then sonicated with a tip
sonicator on ice for 6 seconds twice in a lysis buffer containing (in mM): 150 NaCl,
5 EDTA, 1% Triton X-100, 10 mM Tris-HCl (pH 7.4), 0.5 mM
phenylmethylsulphonyl fluoride (PMSF), 2 mg/mL leupeptin, and 1:100 protease
inhibitor cocktail. After sample preparation, proteins were loaded to each lane of 6%,
10% or 12% SDS-PAGE gels and, after separation, proteins were transferred onto
PVDF membranes. The PVDF membranes were blocked with 5% non-fat milk at
room temperature for 1 h and probed with different primary antibodies (actin
1:10,000 dilution; pJNK and JNK antibodies, 1:2,000 dilution; other primary
24
antibodies, 1:1,000 dilution) at 4
o
C overnight. Membranes were then incubated with
secondary antibodies for 1 h and developed with ECL solutions. Western blots were
scanned and analyzed quantitatively by densitometry with ImageJ software. Data
were generally calculated as fold of control and expressed as means ± SEM from at
least four independent experiments. Student’s t-test was used for statistical analyses
and p values < 0.05 were considered as statistically significant.

2.2.7. Cell viability assays - LDH assay
Neuronal damage was assessed by measurement of lactate dehydrogenase
(LDH) released into the incubation solution (Koh and Choi, 1987;Bruce et al., 1995).
At the end of the various treatments, 0.3 ml of medium solution was mixed with 0.7
ml potassium phosphate buffer (100 mM K
2
HPO
4
, adjusted to pH 7.5 with KH
2
PO
4
).
After 20 minutes, 0.5 ml freshly made solution with sodium pyruvate and NADH
was added to the mixed solution immediately followed by measuring absorbance at
340 nm at 1 min interval. LDH release was normalized to protein concentration and
results are shown as fold of controls.

2.2.8. Cell viability assays - PI uptake assay
Neuronal damage was also assessed by propidium iodide (PI) uptake as
previously described (Laake et al., 1999). For cultured organotypic slices treated
with OGD and recovery, PI (4.6 µg/ml) was added to culture medium in the
beginning of different treatments. PI uptake was visualized using a 5X objective with
a microscope fitted with fluorescence detection, and images of PI-labeled slices were
25
captured with a CCD camera; at this magnification, one image was sufficient to
analyze an entire hippocampal slice. To obtain the best intensity of images and to
avoid saturation, all cultured hippocampal slices were exposed for 100 ms, and the
camera gain was kept constant throughout each experiment. Fluorescence intensity
was estimated by the following method: first, images were adjusted to gray levels
and captured with Adobe Photoshop, with the background of images in white and PI-
stained structures in black; second, modified images were analyzed quantitatively by
densitometry with ImageJ software. Data are generally shown as means ± SEM from
the indicated number of independent experiments.
For acute slices treated with OGD and recovery, PI (4.6 µg/mL) was added to
the incubation medium at the beginning of treatment, and PI uptake at the indicated
times after treatment in hippocampal subfields CA1, CA3, and DG was visualized
using a 10X objective with a LSM 510 META inverted microscope fitted with a laser
confocal system (Zeiss, Germany). Excitation was set at 543 nm and emission was
detected with a 560 nm long-pass filter. For each slice, the exact location of the slice
surface was determined with the addition of microbeads (Invitrogen) and by moving
the stage up and down until the layer with the brightest fluorescent intensity was
found. Images of CA1, CA3 or DG 60 µm below surface were then acquired and
analyzed. Fluorescence intensity of confocal images was estimated by the following
method: first, images were adjusted to gray levels and captured with Adobe
Photoshop, with the background of images in white and PI-stained structures in black;
second, modified images were analyzed quantitatively by densitometry with ImageJ
26
software. Finally, data were normalized to the average value found in control slices
for each experiment.

2.2.9. ATP measurement
ATP levels were determined by the luciferin-luciferase luminescent reaction
using the method described by Rieger (Rieger, 1997). The ATP assay kit was
purchased from Sigma (FL-ASC). After treatment, hippocampal slices were
homogenized and then centrifuged at 12,000 g for 5 min. The supernatant was
collected and processed following the kit instructions. Luminescence produced by
the luciferin-luciferase reaction was recorded with a scintillation counter. ATP levels
were normalized to protein concentration and expressed as nanomole ATP per
milligram protein.

2.2.10. Evaluation of ROS production in slices
Levels of ROS production in slices were determined with the fluorescent
probe 2’,7’-dichlorofluorescin diacetate (DCFH-DA). A 20 mM DCFH-DA stock
solution was prepared and stored at -20
o
C. DCFH-DA was added to slices bath
solution or media 1 h before the end of treatment, with a final 10 µM concentration.
For OGD time course experiment (10, 30, 60 min), DCFH-DA was added to slices
before OGD treatment to make sure all slices were incubated with DCFH-DA for 1 h.
Fluorescent images were obtained with a 5X objective from a microscope fitted with
fluorescence detection, and captured with a CCD camera; at this magnification, one
27
image was sufficient to analyze an entire hippocampal slice. Fluorescence intensity
was estimated by the same method as for PI uptake described above.

2.2.11. TBARS assay
The lipid peroxidation was measured with TBARS assay. After different
treatment, hippocampal slices were homogenized in 250 µl of TBARS
Homogenization Buffer (THB) containing: 2.5% SDS, 6.25 µM deferoxamine, and
2.5 µM probucol (for blank, no slices were added to 250 µl of TBARS
Homogenization Buffer). Then 300 µl of TBA reagent (dissolve 133 mg
thiobarbituric acid in 9.867 ml of warm double distilled water, heat in 65.0
o
C until
thiobarbituric acid dissolves completely. Then allow thiobarbituric acid solution cool
at room temperature for 20 min, and mix 3.75 ml of prepared thiobarbituric acid
solution with 6.25 ml of 20% acetic acid, and adjust pH to 3.5 before the
thiobarbituric acid precipitates. Made fresh on the day of assay) was added to 200 µl
of homogenate (the remaining 50 µl of homogenate was used for protein assay) and
the mix was vortexed for 15 s. The mixed samples were incubated at 95
o
C for 1 h,
cooled at room temperature for 20 min, and 300 µl 15:1 n-butanol / pyridine solution
was added to each sample and inverted 10 times followd by vortexing for 30 s. The
mixed samples were centrifuged at 4000 g for 10 min at room temperature. The
organic layer (top layer) was collected and 200µl of each sample was pipette into a
96-well plate and measured at 532 nm.

2.2.12. Statistical analysis
28
Data were generally calculated as fold of control and expressed as means ±
SEM from the indicated number of independent experiments. One-way ANOV A with
Tukey’s post hoc test was used to analyze the effects of EUK-189 and EUK-207 on
OGD-induced LDH release and PI uptake in acute hippocampal slices, and to
analyze the effects of different concentration of EUK-207 on OGD-induced LDH
release in culture hippocampal slices. Student’s t-test was used for statistical analyses
of other results. P values < 0.05 were considered as statistically significant.

2.3. Results and Discussions
2.3.1. Acute hippocampal slices
2.3.1.1. EUK-189 and EUK-207 attenuate OGD-induced cell death in acute
hippocampal slices from 2-month-old rats
Acute hippocampal slices from 2-month-old rats were subjected to OGD for
2 h, in the absence or presence of EUK-189 (50 µM) or EUK-207 (50 µM), and LDH
release was measured at the end of treatment. OGD treatment induced a 4.13 ± 0.26
fold increase in LDH release, which was partly blocked by EUK-189 (3.21 ± 0.20)
and EUK-207 (2.93 ± 0.21, Fig. 2A). At the end of 2 h OGD treatment, hippocampal
slices were transferred to fresh aCSF solution and further incubated for 2.5 h in the
absence or presence of EUK-189 or EUK-207, saturated with 95% O
2
–5% CO
2
.
LDH release was measured at the end of this recovery period. Compared to control,
OGD/recovery treatment induced a 7.33 ± 0.46 fold increase in LDH release, which
was again significantly blocked by EUK-189 (5.80 ± 0.32) and EUK-207 (3.85 ±
0.24, Fig. 2B). When hippocampal slices were first subjected to OGD for 2 h, and
29

Figure 2. Effects of EUK-189 and EUK-207 on OGD-induced LDH release in acute
hippocampal slices from 2-month-old rats. (A) Hippocampal slices were subjected to
OGD for 2 h in the absence or presence of EUK-189 (50 µM) or EUK-207 (50 µM).
LDH release was measured at the end of 2 h OGD treatment. Results are expressed
as fold increase over control values and are means ± S.E.M. of 25 experiments. (B)
Hippocampal slices were subjected to OGD for 2 h followed by 2.5 h recovery (in
fresh aCSF in the presence of normal O
2
/CO
2
conditions), in the absence or presence
of EUK-189 or EUK-207 throughout the treatment. LDH release was measured at
the end of the 2.5 h recovery treatment. Results are means ± S.E.M. of 30
experiments. (C) All hippocampal slices were subjected to OGD for 2 h, and
transferred to fresh aCSF and treated for 2.5 h in the absence or presence of EUK-
189 or EUK-207 (with EUK-189 or EUK-207 applied only during the 2.5 h
recovery). LDH release was measured at the end of the 2.5 h treatment. Results are
means ± S.E.M. of 8 experiments.
30

Figure 3. Effects of EUK-189 and EUK-207 on OGD-induced increase in PI uptake
in different regions of acute hippocampal slices from 2-month-old rats. Hippocampal
slices were subjected to OGD for 2 h followed by 2.5 h recovery in the absence or
presence of EUK-189 (50 µM) or EUK-207 (50 µM). Representative images of PI
staining in CA1 region (A), CA3 region (B), or DG region (C) 60 µm below the
surface of acute hippocampal slices are displayed. D, E, and F. Quantification of PI
staining in different hippocampal subfields (D: CA1; E: CA3; F: DG). PI staining
intensity was analyzed as described under Materials and Methods. Results are
expressed as percent of control values and are means ± S.E.M. of 11 experiments.
Statistical analysis was done by one-way ANOVA followed by Tukey’s post hoc test
(* P < 0.05, ** P < 0.01).
31
then transferred to fresh aCSF and incubated for 2.5 h in the absence or presence of
EUK-189 or EUK-207 (EUK-189 or EUK-207 applied only during the 2.5 h
recovery period), only EUK-207 significantly reduced OGD-induced LDH release
(4.90 ± 0.41 vs 7.33 ± 0.46, Fig. 2C).
The neuroprotective effects of EUK-189 and EUK-207 against OGD-induced
cell death were also determined by analyzing propidium iodide (PI) uptake in CA1,
CA3 and DG regions 60 µm below surface (selection of this level was based on
analysis of PI uptake in control slices; the upper layers exhibited intense PI staining
due to neuronal damage produced during slice preparation) (Fig. 3). Hippocampal
slices were treated with OGD for 2 h followed by 2.5 h recovery, in the absence or
presence of EUK-189 (50 µM) or EUK-207 (50 µM) throughout the incubation.
OGD treatment induced a significant increase in PI staining in the pyramidal layers
of CA1 (1.81 ± 0.19 fold increase over control values) and CA3 (1.97 ± 0.17 fold
increase over control values), and in the granular layer of DG (1.56 ± 0.14 fold
increase over control values). EUK-189 partly decreased PI uptake (1.35 ± 0.21 in
CA1, 1.51 ± 0.11 in CA3, and 1.26 ± 0.17 in DG , when compared to control values),
while EUK-207 completely reversed OGD-induced PI uptake (0.97 ± 0.11 in CA1,
1.13 ± 0.16 in CA3, and 0.92 ± 0.14 in DG , when compared to control values). One-
way ANOV A and Tukey’s post hoc test indicated that, when compared to EUK-189,
EUK-207 provided more protection against OGD-induced neuronal death.

2.3.1.2 EUK-189 and EUK-207 partly blocked OGD-induced ATP
depletion in acute hippocampal slices
32

Figure 4. Effects of EUK-189 and EUK-207 on OGD-induced ATP depletion in
acute hippocampal slices from 2-month-old rats. (A) Hippocampal slices were
subjected to OGD for 2 h in the absence or presence of EUK-189 (50 µM) or EUK-
207 (50 µM). ATP levels were measured at the end of the 2 h OGD treatment. (B)
Hippocampal slices were subjected to OGD for 2 h followed by 2.5 h recovery in the
absence or presence of EUK-189 or EUK-207. ATP levels were measured at the end
of the 2.5 h recovery treatment. (C) All hippocampal slices were subjected to OGD
for 2 h. They were then transferred to fresh aCSF and treated for 2.5 h in the absence
or presence of EUK-189 or EUK-207 (with EUK-189 or EUK-207 applied only
during the 2.5 h recovery period). ATP levels were measured at the end of treatment.
Results are expressed as percent decrease over the respective control values and are
means ± S.E.M. of 5 experiments. Statistical analysis was done by student’s t-test (*
p < 0.05 as compared to OGD-treated slices).
33
ATP depletion is one of the initial events triggered by ischemia and has been
proposed to play a crucial role in cell death. To examine whether treatment with
EUK-189 or EUK-207 had any effect on OGD-induced ATP depletion, ATP levels
were measured following various experimental treatments. Compared to control,
hippocampal slices subjected to OGD for 2 h exhibited ATP levels representing about
25% of control values, an effect that was only slightly changed by treatment with
EUK-189 or EUK-207 (Fig. 4A). ATP levels were also measured at the end of the 2.5
h recovery period following the 2 h OGD treatment in the absence or presence of
EUK-189 or EUK-207 throughout the incubation. Partial recovery of ATP levels
(55% of control values) was observed after 2.5 h recovery under control conditions,
and this recovery of ATP levels was further enhanced by treatment with EUK-189 or
EUK-207 during the OGD and recovery periods (up to 73% or 76% of control values
respectively, an effect which was significantly different from OGD treatment alone
(Fig. 4B) On the other hand, when EUK-189 or EUK-207 was applied only during
the 2.5 h recovery period, only small and no significant increases in ATP levels were
observed (Fig. 4C).

2.3.1.3 EUK-189 and EUK-207 completely blocked OGD-induced ROS
generation in acute hippocampal slices
To determine whether EUK-189 or EUK-207 protected hippocampal slices
from OGD-induced cell death by eliminating free radical formation, ROS
accumulation was evaluated with the fluorescent probe DCF. After 2 h OGD
treatment followed by 2.5 h recovery in the absence or presence of EUK-189 or
34

Figure 5. Effects of EUK-189 and EUK-207 on OGD-induced ROS generation in
acute hippocampal slices from 2-month-old rats. Hippocampal slices were subjected
to OGD for 2 h followed by 1.5 h recovery in the absence or presence of EUK-189
(50 µM) or EUK-207 (50 µM), and then incubated with fresh aCSF with DCFH-DA
(10 µM) for another 1 h-period of recovery. (A) Representative images of DCF
fluorescence in whole hippocampal slices subjected to OGD for 2 h followed by 2.5
h recovery in the absence or presence of EUK-189 or EUK-207. (B) Quantification
of DCF fluorescence intensity. Results are expressed as percent of control values and
are means ± S.E.M. of 15 experiments. (* p < 0.05 as compared to OGD-treated
slices, student’s t-test).
35

Figure 6. Effects of U0126, EUK-189, or EUK-207 on OGD-induced changes in
ERK1/2 phosphorylation in acute hippocampal slices from 2-month-old rats. Acute
hippocampal slices were subjected to OGD for 2 h followed by 2.5 h recovery. At
the end of incubation, slices were homogenized and aliquots were processed for
immunoblotting with antibodies against double-phosphorylated ERK (pERK) and
total ERK. (A) Representative images of western blots indicating the pERK and
ERK levels at 42 and 44 kDa in the absence or presence of EUK-189 (50 µM), EUK-
207 (50 µM), EUK-189 plus U0126, EUK-207 plus U0126, or U0126 (10 µM). (B)
Quantitative analysis of blots similar to those shown in A. Blots were scanned and
the intensities of pERK bands were quantified and normalized to the intensities of
ERK bands. Results were expressed as percent of control values and the data
represent means ± S.E.M. of 5 experiments.
36
EUK-207, DCF fluorescence over the whole hippocampal slices was analyzed with
fluorescent microscopy. Compared to untreated control slices, hippocampal slices
subjected to OGD exhibited a significant increase in fluorescence intensity (about
156% of control values, Fig. 5); in contrast, slices subjected to OGD in the presence
of EUK-189 or EUK-207 exhibited a decrease in DCF fluorescence intensity (86%
and 80% of control values, respectively).

2.3.1.4 The protective effect of EUK-207 against OGD-induced cell
death did not involve the ERK or the p38 pathway
To determine the role of the ERK pathway in my ischemia model, I first
determined whether ERK was activated as assessed with measurement of levels of
double-phosphorylated ERK1/2 in hippocampal slices subjected to OGD for 2 h
followed by 2.5 h recovery (Fig. 6A). Under these conditions, p-ERK1/2 levels were
decreased to about 20% of control values. Addition of EUK-189 or EUK-207 in the
incubation medium throughout the OGD and recovery periods resulted in a
significant recovery of p-ERK1/2 levels to 80% and 50% of control values,
respectively (Fig. 6B). When the MEK inhibitor, U0126 (10 µM), was applied alone
or together with EUK-189 or EUK-207, it completely blocked ERK1/2
phosphorylation, as expected. However, treatment with U0126 had no effect on
OGD-induced LDH release or on the protective effects of EUK-189 and EUK-207
against OGD-induced LDH release (Fig. 7A, B).
To see if ERK phosphorylation levels show different changes after OGD
treatment during development, I also tested the roles of ROS and of ERK on OGD-
37

Figure 7. Effects of U0126, EUK-189, or EUK-207 on OGD-induced LDH release
in acute hippocampal slices from 2-month-old rats. (A) Hippocampal slices were
subjected to OGD for 2 h and LDH release was measured at the end of the 2 h OGD
treatment. Results are expressed as fold increase over control values and are means ±
S.E.M. of 10 experiments. (B) Hippocampal slices were subjected to OGD for 2 h
followed by 2.5 h recovery. LDH release was measured at the end of the 2.5 h
recovery treatment. Results are means ± S.E.M. of 10 experiments. (* p < 0.05 as
compared to OGD-treated slices, student’s t-test).
38

Figure 8. Effects of U0126 or EUK-207 on ERK1/2 phosphorylation, and on OGD-
induced LDH release in acute hippocampal slices from postnatal day 10 rats. Acute
hippocampal slices from postnatal day 10 rats were subjected to OGD for 2 h
followed by 2.5 h recovery in the absence or presence of U0126 (10 µM) or EUK-
207 (50 µM). At the end of treatment, slices were homogenized and aliquots were
processed for immunoblotting with pERK and ERK antibodies. LDH release in
medium was measured as an index of toxicity. (A) Representative images of western
blots indicating the pERK and ERK levels at 42 and 44 kDa in the absence or
presence of EUK-207, U0126, or EUK-207 plus U0126. (B) Quantitative analysis of
blots similar to those shown in (A). The results are expressed as percent of control
values and the data represent means ± S.E.M. of 4 experiments. (C) Hippocampal
slices were subjected to OGD for 2 h followed by 2.5 h recovery, in the absence or
presence of EUK-207, U0126, or EUK-207 plus U0126. LDH release was measured
at the end of the 2.5 h recovery treatment. Results are expressed as fold increase over
control values and are means ± S.E.M. of 10 experiments. (* p < 0.05 as compared to
OGD-treated slices, student’s t-test).
39
induced cell death in slices from neonatal rats. In contrast to the striking OGD-
induced decrease in ERK phosphorylation in slices from adult rats, slices prepared
from postnatal day 10 rats and subjected to OGD for 2 h followed by 2.5 h recovery
showed a slight increase in ERK1/2 phosphorylation (Fig. 8A,B). OGD-induced
LDH release was also significantly reduced by EUK-207 (Fig. 8C). Treatment with
EUK-207 did not modify ERK1/2 phosphorylation levels when compared to either
control or OGD. As observed in slices from adult rats, U0126 completely blocked
ERK1/2 activation when applied alone or together with EUK-207, and also had no
effect on OGD-induced LDH release either by itself or in the presence of EUK-207
(Fig. 8).
To determine the roles of p38 and JNK pathways in the neuroprotective
effects of EUK-207, I tested the levels of phosphorylated p38 and JNK. Acute
hippocampal slices from 2-month-old and postnatal day 10 rats were subjected to
OGD for 2 h followed by 2.5 h recovery. Under these conditions, p-p38 levels were
slightly decreased to 81% of control values (p >0.05, t-test) in slices from postnatal
day 10 rats and decreased to 60% of control values (p <0.05, t-test) in slices from 2-
month-old rats (Fig. 9A,B). ANOV A with Tukey’s post hoc test indicated that in
slices from either 2-month-old or postnatal day 10 rats, there were no significant
differences between any group of treatment when slices were subjected to OGD in
the absence or presence of EUK-207 (50 µM), EUK-207 plus PD98059 (MEK1/2
inhibitor, 10 µM), or EUK-207 plus SB203580 (p38 inhibitor, 1 µM). When slices
were subjected to OGD for 2 h followed by 2.5 h recovery, p-JNK levels were
significantly decreased to below 20% of control values (p <0.05, t-test) in slices from
40

Figure 9. Effects of EUK-207 on OGD-induced changes of p38 and JNK
phosphorylation in acute hippocampal slices from 2-month-old and p10 rats. Acute
hippocampal slices were subjected to OGD for 2 h followed by 2.5 h recovery. At
the end of incubation, slices were homogenized and aliquots were processed for
immunoblotting with antibodies against phosphorylated p38 or JNK and total p38 or
JNK. (A) Representative images of western blots indicating the p-p38 and p38 levels
at 43 kDa in the absence or presence of EUK-207 (50 µM), EUK-207 plus MEK
inhibitor PD98059 (10 µM), or EUK-207 plus p38 inhibitor SB203580 (1 µM). (B)
Quantitative analysis of blots similar to those shown in (A). Results are expressed as
percent of control values and the data represent means ± S.E.M. of 4 experiments. (C)
Representative images of western blots indicating the p-JNK and JNK levels at 46
and 54 kDa. (D) Quantitative analysis of blots similar to those shown in C. Results
represent means ± S.E.M. of 5 experiments. (PD: PD98059; SB: SB203580)
41

Figure 10. Effects of EUK-207, PD98059, and SB203580 on OGD-induced LDH
release in acute hippocampal slices from 2-month-old rats. Hippocampal slices from
2-month-old rats were subjected to (A) 2 h OGD or (B) 2 h OGD followed by 2.5 h
recovery, in the absence or presence of EUK-207 (50 µM), EUK-207 plus PD
(PD98059; 10 µM), or EUK-207 plus SB (SB203580; 1 µM). Results are expressed
as fold increase over control values and are means ± S.E.M. of 10 experiments.
42

Figure 11. Effects of EUK-207, PD98059, and SB203580 on OGD-induced LDH
release in acute hippocampal slices from postnatal day 10 rats. Hippocampal slices
from postnatal day 10 rats were subjected to (A) 2 h OGD or (B) 2 h OGD followed
by 2.5 h recovery, in the absence or presence of EUK-207, EUK-207 plus PD
(PD98059; 10 µM), or EUK-207 plus SB (SB203580; 1 µM). Results are expressed
as fold increase over control values and are means ± S.E.M. of 10 experiments. (* p
< 0.05 as compared to OGD-treated slices, student’s t-test).
43
both 2-month-old and postnatal day 10 rats (Fig. 9C,D). ANOV A with Tukey’s post
hoc test indicated that there were no significant differences between any treatments
in slices from both ages either. When hippocampal slices from 2-month-old (Fig. 10)
and postnatal day 10 (Fig. 11) rats were subjected to 2 h OGD w/o recovery, or 2 h
OGD followed by 2.5 h recovery, PD98059 or SB203580 did not modify OGD-
induced LDH release or the protective effects of EUK-207 against OGD-induced
LDH release when applied alone or together with EUK-207.

2.3.1.5. Discussion
My results indicate that, among the multiple pathways activated as a result of
ischemia/reperfusion, excessive production of reactive oxygen species represents a
critical step in the rapid phase of neurodegeneration observed in both neonatal and
adult hippocampal slices. On the other hand, ERK, p38, and JNK do not seem to play
a major role in this initial phase of neuronal death. OGD has been shown to elicit
rapid decrease in ATP levels, increased intracellular calcium concentration, release of
glutamate followed by overactivation of glutamate receptors, excessive generation of
reactive-oxygen species (ROS), mitochondria dysfunction, and ultimately cell death
by both necrosis- and apoptosis-mediated cell death (Koistinaho and Koistinaho,
2005;Zipp and Aktas, 2006;Won et al., 2002). The exact temporal sequence of events
and the relative importance of these different events remain far from being
understood. My results indicate that increased formation of reactive oxygen species
might take place relatively early in this sequence, possibly even before recovery
occurs, as the two synthetic superoxide dismutase/catalase mimetics, EUK-189 and
44
EUK-207, produced better neuroprotection in acute hippocampal slices when applied
at the beginning of the OGD period than only during the recovery period.
In his comprehensive 1999 review, P. Lipton proposed a sequence of events
comprising initiators and activators followed by perpetrators leading from the initial
ischemic insults to cell death (Lipton, 1999). Accordingly, the loss of ATP initiates
membrane depolarization, glutamate release, and increased intracellular calcium.
Ischemia also induced gene activation possibly mediated through MAP kinase
activation, as well as increased production of oxygen free radicals and peroxynitrite.
These events are followed by activation of proteases such as calpain, mitochondrial
dysfunction, prolonged changes in kinases and phosphatases, cytoskeletal damage
and cell death. It has also been suggested that the degree of ATP depletion
determines the form of death cells will exhibit, with gradual ATP depletion after mild
ischemia inducing caspase-3 release and apoptosis, rapid ATP depletion with severe
ischemia eliciting extreme cytochrome c release and necrosis (Li et al.,
1997;Saikumar et al., 1998). Drugs that prevent ATP depletion or promote ATP
recovery after ischemia have been reported to increase cell survival when applied
before, during or after the ischemic insult (Riepe et al., 1997;Galeffi et al., 2000).
In my model, 2 h OGD treatment decreased ATP levels to about 25% of
control, an effect that is not modified by the presence of EUK-189 or EUK-207.
Thus, as expected, the neuroprotective effects of these compounds are not due to a
prevention of the ATP loss produced by ischemia. The significant degree of ATP
recovery observed when EUK-189 or EUK-207 was applied during and after the
OGD periods, as compared to control, is probably due to the decrease in cell damage
45
produced by these compounds. ATP depletion can further contribute to the collapse
of the mitochondrial membrane potential, the opening of mitochondrial permeability
transition pore (MPTP), the release of small proteins such as apoptosis-inducing
factor (AIF) and cytochrome c, and accumulation of free radicals (Fiskum et al.,
1999;Murphy et al., 1999). ATP depletion initiates MPTP opening and increases
ROS levels, which then in turn potentiate ATP depletion. It is thus expected that the
EUK compounds will provide some degree of protection for mitochondria as
previously described in the SOD1 null mice (Melov et al., 2001).
As synthetic superoxide dismutase/catalase mimetics, EUK-189 and EUK-
207 have been reported to provide protection in various animal models of human
diseases (Melov et al., 2000;Melov et al., 2001;Liu et al., 2003). My DCF
fluorescence results indicated that EUK-189 or EUK-207 blocked the formation of
ROS elicited by OGD and recovery. These results strongly suggest that the rapid
OGD and recovery-induced cell death in acute hippocampal slices is, at least in part,
due to increased free radical production, and that the neuroprotective effects of EUK-
207 and EUK-189 are linked to their ability to eliminate ROS. Although EUK-189
and EUK-207 are both synthetic superoxide dismutase/catalase mimetics with
relatively similar structures (Liu et al., 2003), my results with LDH release and PI
uptake indicated that EUK-207 exhibited better neuroprotective effects than EUK-
189. The PI staining results in different hippocampal subfields indicated that PI
uptake after OGD/recovery was higher in CA1 and CA3 than in DG, and that only
EUK-207 exhibited significant neuroprotection in all three subfields. The difference
in neuroprotective effects between EUK-189 and EUK-207 might be due to their
46
structural differences as EUK-207 has similar catalytic activities but greater
biological stability than EUK-189 (Liu et al., 2003), which might be an important
factor in my OGD model in acute hippocampal slices.
MAP Kinase pathways have often been implicated in ischemia-induced cell
death, although the role of MEK/ERK pathway remains controversial. It has
generally been proposed that MEK/ERK activation is involved both in cell death and
neuroprotective/survival effects during ischemia and after reperfusion (Alessandrini
et al., 1999;Fahlman et al., 2002). Some studies have shown that MEK inhibitors
such as U0126 or PD-98059 can reduce cell death induced by seizures, glutamate
excitotoxicity, and OGD (Namura et al., 2001;Zhu et al., 2005;Murray et al., 1998);
however, recent studies have also shown that inhibition of the MEK/ERK pathway
blocked the neuroprotective effects of N-acetyl-O-methyldopamine (NAMDA), or
fructose-1,6-bisphosphate (FBP) against ischemia/hypoxia (Fahlman et al.,
2002;Park et al., 2004), or had no effect in ischemia-induced cell death (Sugino et al.,
2000;Abe and Saito, 2000). The dual role of the MEK/ERK pathway may depend on
the animal model, or the duration of ERK activation (Park et al., 2004).
In my experiments, ERK1/2 was inactivated after OGD and recovery in slices
from 2-month-old rats, an effect that was partially restored by EUK-189 or EUK-207.
The MEK inhibitor U0126 completely blocked ERK1/2 activation whether slices
were treated with or without EUK-189 or EUK-207; however, U0126 did not modify
OGD-induced cell death, nor did it modify EUK-189- or EUK-207-mediated
protection against OGD-induced cell death. Moreover, while OGD and recovery
treatment elicited high LDH release in hippocampal slices prepared from postnatal
47
10 rats, ERK1/2 activation levels were not affected and U0126 did not modify EUK-
207 mediated neuroprotection against OGD-induced cell death under these
conditions. Thus, my results clearly indicate that ERK1/2 is not involved in early
ischemia-induced cell death. It remains possible though that this pathway participates
in more delayed forms of ischemia-induced cell death.
A similar conclusion can be drawn for the other members of the MAP kinase
pathways, p30 and JNK. EUK-207 had no effect on OGD-induced slight p38
dephosphorylation and significant JNK dephosphorylation in slices from either 2-
month-old or postnatal day 10 rats. Although p38 inhibition has been shown to
provide neuronal protection in cerebral ischemia (Sugino et al., 2000;Barone et al.,
2001), in acute hippocampal slices the p38 inhibitors, SB203580 or SB202190 (data
not show), did not modify OGD-induced cell death or EUK-207-mediated
neuroprotection against OGD-induced cell death. All these results suggest that the
MAP kinase pathways are not involved in OGD-induced cell death and are not
involved in the neuroprotective effects of EUK-189 or EUK-207 against the rapid
phase of OGD-induced cell death in acute hippocampal slices.
We previously reported data indicating that the mechanisms underlying
OGD-induced cell death were different in slices from neonatal and adult rats (Zhou
and Baudry, 2006), and my present results further confirm this interpretation. In the
present experiments, OGD/recovery in neonatal slices was not associated with
decreased levels in ERK1/2, in contrast to what was observed in slices prepared from
adult rats. This effect could be related to the much lower decrease in ATP levels
resulting from OGD in slices from neonatal compared to adult slices (data not
48
shown), an effect possibly correlated to the anaerobic nature of metabolism in
neonatal animals (Bickler et al., 1993;Folbergrova, 1993).
In conclusion, my data indicate that like many salen-manganese complexes
that are neuroprotective in various models of neurological diseases, EUK-189 and
EUK-207 provide significant protection against rapid OGD-induced cell death in
acute hippocampal slices, by eliminating free radicals generation, but not through the
MAP kinase pathways. While my results also document the critical roles of oxygen
free radicals and rapid ischemia-induced neuronal death, further studies are required
to better understand the links between early events set up by ischemic insults and
delayed forms of ischemia-induced cell death. To better understand the mechanisms
of relatively delayed ischemic cell death, organotypic hippocampal slice cultures
were treated with OGD and protective effects of EUK-207 was studied in the second
part of this chapter.
49
2.3.2. Organotypic hippocampal slice cultures
2.3.2.1. EUK-207 attenuated OGD-induced cell death in organotypic
hippocampal slice cultures
To test the effects of EUK-207 on OGD-induced cell death, cultured
hippocampal slices were subjected to 30, 40, or 60 min of OGD followed by 3 or 24
h recovery in regular medium with glucose and oxygen, in the absence or presence of
EUK-207 (40 µM, applied 1 h before OGD and during OGD), and LDH release was
measured at the end of the recovery periods. Compared to control, 30, 40, or 60 min
of OGD followed by 3 h recovery induced a 2.50 ± 0.32, 4.25 ± 0.46, or 12.0 ± 1.2
fold increase in LDH release respectively; 60 min OGD-induced LDH release was
significantly reduced by EUK-207 (7.12 ± 0.74, p <0.01, Fig. 12A, n=6). Compared
to control, 30, 40, or 60 min of OGD followed by 24 h recovery induced a 3.09 ±
0.24, 8.64 ± 0.71, or 20.0 ± 1.6 fold increase in LDH release respectively; 60 min
OGD-induced LDH release was also attenuated when hippocampal slices were pre-
treated with EUK-207 (15.0 ± 1.0, p <0.05, Fig. 12B, n=6). I also tested the effects
of different EUK-207 concentrations on OGD-induced LDH release. Hippocampal
slices were subjected to 1 h OGD followed by 3 h or 24 h recovery in the absence or
presence of EUK-207 (1 µM, 10 µM, 40 µM, or 100 µM, applied 1 h before OGD
and during OGD), and LDH release was measured after recovery. Compared to
control, 1 h OGD followed by 3 h or 24 h recovery induced a 12.7 ± 1.4 or 23.1 ± 1.8
fold increase in LDH release, which was reduced in a concentration-dependent
manner by EUK-207, although only the effect of 40 µM EUK-207 was significant
(62% or 73% of 3 h or 24 h OGD alone, respectively, Fig. 13, n=5).
50

Figure 12. Effects of EUK-207 on LDH release elicited by different periods of OGD
in organotypic hippocampal slice cultures. (A) Hippocampal slices were subjected to
30, 40, or 60 min of OGD followed by 3 h recovery in regular medium with glucose
and oxygen, in the absence or presence of EUK-207 (40 µM, applied 1 h before and
during OGD). LDH release was measured at the end of 3 h recovery. (B)
Hippocampal slices were subjected to 30, 40, or 60 min of OGD followed by 24 h
recovery in regular medium, in the absence or presence of EUK-207 (40 µM, applied
1 h before and during OGD). LDH release was measured at the end of 24 h recovery.
Statistical analysis was done by student’s t-test (* p < 0.05, ** P < 0.01).
51

Figure 13. Effects of different concentrations of EUK-207 on LDH release elicited
by 1 h OGD in organotypic hippocampal slice cultures. Hippocampal slices were
subjected to 1 h OGD followed by 3 h or 24 h recovery, in the absence or presence of
EUK-207 (1 µM, 10 µM, 40 µM, or 100 µM, applied 1 h before and during OGD).
LDH release was measured at the end of 3 h or 24 h recovery. Statistical analysis
was done by one-way ANOVA followed by Tukey’s post hoc test (* P < 0.05 as
compared to slices treated with OGD and 0 µM EUK-207).
52

Figure 14. Effects of pre- or post-treatment with EUK-207 on 1 h OGD-induced
LDH release and rate of LDH release at various times after OGD in organotypic
hippocampal slice cultures. (A) Hippocampal slices were subjected to 1 h OGD
followed by 24 h recovery in regular medium, in the absence or presence of EUK-
207. EUK-207 (40 µM) was applied in following conditions: (1) 2 h before OGD, (2)
1 h before OGD, (3) during 1 h OGD, or (4) during 24 h recovery (EUK-207, 20
µM). LDH release was measured at the end of 24 h recovery. (B) Rate of LDH
release at various times after 1 h OGD treatment. Statistical analysis was done by
student’s t-test (* p < 0.05, ** p < 0.01, as compared to OGD group).
53
To determine the best time for EUK-207 addition, cultured hippocampal
slices were treated with 40 µM EUK-207 2 h or 1 h before OGD, during the 1 h
OGD period, or during the 24 h recovery period (20 µM EUK-207). Although
application of EUK-207 during the 24 h recovery period slightly decreased OGD-
induced LDH release (86% of OGD alone), only the addition of EUK-207 2 h or 1 h
before OGD significantly reduced OGD-induced LDH release to 70% or 75% of
OGD alone (Fig. 14A, n =6). It was of interest to determine the pattern of LDH
release during and after OGD. Therefore, we calculated the rate of LDH release
(expressed in LDH release/h) during various periods starting at the end of the OGD
period and until 24 h of recovery. The highest rate of release, and presumably the
highest rate of neuronal damage, was observed between 0 and 3 h after OGD. By 6
and 24 h after OGD, the rate of LDH release decreased, but was still significantly
higher than in control slices (Fig. 14B).
The protective effects of EUK-207 against OGD-induced cell death were also
determined by analyzing propidium iodide (PI) uptake. Cultured hippocampal slices
were subjected to 1 h OGD followed by 24 h recovery in the absence or presence of
EUK-207 (40 µM EUK-207 applied 1 h before OGD, and 20 µM EUK-207 applied
during 24 h recovery). Compared to control, 1 h OGD followed with 24 h recovery
significantly increased propidium iodide uptake throughout the hippocampus (3.20 ±
0.16), and this increase was significantly reduced by EUK-207 (2.34 ± 0.15, Fig. 15,
n=12). Treatment with EUK-207 alone did not modify PI uptake.

2.3.2.2 EUK-207 protects organotypic hippocampal slice cultures from
54

Figure 15. Effects of EUK-207 on OGD-induced increase in PI uptake in
organotypic hippocampal slice cultures. (A) Cultured hippocampal slices were
subjected to OGD for 1 h followed by 24 h recovery in the absence or presence of
EUK-207 (40 µM EUK-207 applied 1 h before OGD, and 20 µM EUK-207 applied
during 24 h recovery). (B) Quantification of PI staining (** p < 0.01, student’s t-test).
55
OGD-induced cell death by attenuation of OGD-induced ROS
generation, lipid peroxidation and AIF translocation
To determine whether EUK-207 protected cultured hippocampal slices from
OGD-induced cell death by eliminating free radical formation, ROS accumulation
was evaluated with the fluorescent probe DCF. The amount of free radicals generated
during OGD and recovery was studied by measuring DCF fluorescence in
hippocampal slices subjected to 0 (control), 10, 30, or 60 min OGD, as well as in
slices observed 5, 10, 30 min, 1 h, 3 h, 6 h, or 24 h after 1 h OGD (Fig. 16A,B).
Compared to control, DCF fluorescence increased steadily with time of OGD
treatment; interestingly DCF fluorescence decreased sharply at the beginning of the
recovery period, before increasing slowly with further recovery. When hippocampal
slices were subjected to 1 h OGD followed by 3 h recovery in the absence or
presence of EUK-207 (40 µM, 1 h pretreatment before OGD), EUK-207
significantly reduced OGD and recovery-induced increase in DCF fluorescence
intensity from 1.48 ± 0.09 to 1.14 ± 0.08 (Fig. 16C, n=12).
When excess ROS is generated during ischemia or OGD, free radicals react
strongly with unsaturated lipids and cause lipid peroxidation, which decreases cell
membrane fluidity and induces cell death. To test the effects of OGD and EUK-207
on lipid peroxidation, cultured hippocampal slices were subjected to 1 h OGD
followed by different times of recovery, and lipid peroxidation was measured with
the TBARS assay at 0, 3, 6, or 24 h following 1 h OGD. Lipid peroxidation results
were normalized to protein levels in slices to minimize differences between slices.
Compared to control, lipid peroxidation increased after 1 h OGD treatment (120% of
56

Figure 16. ROS levels, measured with DCF fluorescence during and after OGD.
Cultured hippocampal slices were subjected to OGD for different times, or to 1 h
OGD followed by different times of recovery. (A) Representative images of DCF
fluorescence in cultured hippocampal slices subjected to OGD for 0 (control), 10, 30,
or 60 min, or 1 h OGD followed by 5, 10, 30, or 60 min recovery. (B) Quantification
of DCF fluorescence intensity (including 3 h, 6 h and 24 h recovery after 1 h OGD).
(C) Hippocampal slices were subjected to 1 h OGD followed by 3 h recovery in the
absence or presence of EUK-207 (40 µM, applied 1 h before OGD. * p < 0.05 as
compared to OGD group, student’s t-test).
57

Figure 17. Effects of OGD and EUK-207 on lipid peroxidation in cultured
hippocampal slices. Cultured hippocampal slices were subjected to 1 h OGD
followed by different times of recovery. Lipid peroxidation was measured with the
TBARS assay and normalized to protein levels. (A) Lipid peroxidation levels in
hippocampal slices treated with 1 h OGD followed by 0, 3, 6, or 24 h recovery. (B)
Hippocampal slices were subjected to 1 h OGD followed by 6 h recovery in the
absence or presence of EUK-207 (40 µM, applied 1 h before OGD and during OGD.
* p < 0.05 as compared to OGD group, student’s t-test).
58

Figure 18. Effects of OGD and EUK-207 on apoptosis-inducing factor (AIF) and
cytochrome c (cyto c) levels in different subcellular fractions in cultured
hippocampal slices. Cultured hippocampal slices were subjected to 1 h OGD
followed by 3 h recovery in the absence or presence of EUK-207 (40 µM, applied 1
h before OGD and during OGD). At the end of recovery, slices were homogenized,
and different subcellular fractions were prepared by differential centrifugation. The
nuclear fraction was processed for immunoblotting with an antibody against AIF,
and the cytosolic fraction was probed with an antibody against cytochrome c. (A)
Representative images and quantitative analysis of western blots probed with an AIF
antibody. (B) Representative images and quantitative analysis of western blots
probed with a cytochrome c antibody. (* p < 0.05 as compared to OGD group,
student’s t-test).
59
control level), peaked at 6 h of recovery (147% of control level), and remained
slightly higher than control at 24 h of recovery (112% of control level, Fig. 17A).
EUK-207 (40 µM, applied 1 h before OGD and during OGD) significantly reduced
lipid peroxidation measured at 6 h recovery from 1.71 ± 0.15 to 1.28 ± 0.11 (Fig.
17B, n=6).
Hippocampal slices subjected to OGD treatment might undergo cell death
either by necrosis or by apoptosis. To determine whether apoptosis contributes to
OGD-induced cell death, I evaluated AIF translocation from mitochondria into
nucleus, and cytochrome c release from mitochondria into cytosol after OGD.
Cultured hippocampal slices were subjected to 1 h OGD followed by 3 h recovery in
the absence or presence of EUK-207 (40 µM, applied 1 h before OGD and during
OGD). At the end of recovery, slices were homogenized and centrifuged, and the
nuclear fraction was processed for immunoblotting with an antibody against AIF,
while the cytosolic fraction was probed with an antibody against cytochrome c.
Compared to control, 1 h OGD and 3 h recovery induced an increase in nuclear AIF
levels (1.52 ± 0.07), an effect which was significantly blocked by EUK-207 (p <0.05,
Fig. 18A, n =3). Levels of cytosolic cytochrome c were also increased (2.26 ± 0.73)
with 1 h OGD and 3 h recovery, and although this effect was reduced by EUK-207,
the decrease did not reach statistical significance (p >0.05, Fig. 18B, n=4).

2.3.2.3. Discussion
Both acute and cultured hippocampal slices have been used as in vitro
ischemic models because they provide better approximations of in vivo conditions as
60
compared to dissociated neuronal cultures (Strasser and Fischer, 1995). We have
reported that EUK-207 and EUK-189 protected acute hippocampal slices from both
2-month-old and postnatal-day-ten rats from OGD and recovery-induced cell death
(Zhou et al., 2007). I proposed that the protective effects of these synthetic
SOD/catalase mimetics involved the elimination of free radicals and the partial
reversal of ATP depletion. EUK-207 provided better protection than EUK-189, an
effect which could be due to the structure differences between the compounds, as
EUK-207 has similar catalytic activities but greater biological stability than EUK-
189 (Liu et al., 2003). Although not as widely used as cultured slices, acute
hippocampal slices represent a good model to study ischemic damage because they
can be prepared from both neonatal and mature rats, thus providing a model to study
developmental as well as age-related changes in ischemic damage. On the other hand,
the use of acute slices prevent the possibility of studying the mechanisms involved in
delayed ischemic damage taking place in hours to days following the ischemic attack.
Therefore, cultured slices represent a better model to study the cellular and
functional changes taking place over prolonged periods of recovery that would better
reflect events taking place during reperfusion in in vivo models. Moreover, cultured
slices exhibit many features of in vivo tissues including morphological organization,
receptor expression, and synaptic function (Frotscher et al., 1995;Gahwiler et al.,
1997).
Under normal conditions, the rate of ROS formation is equal to that of their
elimination. However, during ischemia and reperfusion, this balance is perturbed
either due to increased ROS production or decreased activity of cellular defense
61
systems (Valko et al., 2007). In the present study, my results indicate that excessive
ROS production plays an important role in OGD- and recovery-induced cell death in
cultured hippocampal slices, as incubation of cultured slices with EUK-207 for 1 h
or 2 h before OGD significantly reduced cell death when measured with either LDH
release or PI uptake. When EUK-207 was applied only during the recovery period,
LDH release was slightly but not significantly decreased to 86% of OGD treatment
alone. When EUK-207 was applied only during 1 h OGD, little protection was
observed. The optimal degree of protection was obtained when EUK-207 was
applied before the OGD period, suggesting that the compound needs to be present in
the tissues before OGD to exhibit neuroprotection. It is also important to stress that
the maximal degree of protection, determined by either LDH release or PI uptake,
represented only 30-40%. These results suggest that ROS accumulation is likely to
represent one of the critical factors in ischemic cell death, and that although ROS are
generated throughout the OGD and the recovery period, there is a critical period at
the beginning of the OGD period that triggers events leading to irreversible neuronal
damage. This result has obvious consequences regarding the potential use of ROS
scavengers for the treatment of stroke in humans, and could account for the recent
failure of NXY-059 in clinical trials (Savitz and Fisher, 2007).
While increased free radical formation was found during both 15 min global
ischemia (Piantadosi and Zhang, 1996) and 30 min focal ischemia (Kinuta et al.,
1989), it has been generally assumed that more free radicals are generated during
reperfusion (Piantadosi and Zhang, 1996;Dirnagl et al., 1995;Oliver et al., 1990). In
my experiments, ROS levels measured with DCF fluorescence were increased within
62
5 min of OGD (data not shown) and increased steadily with OGD duration. These
results are in good agreement with several recent reports indicating that ROS are
generated during period of hypoxia and hypoglycemia (Abramov et al., 2007;Yu et
al., 2008;McGowan et al., 2006;Moro et al., 2005). My results are also consistent
with the fact that it is necessary to apply EUK-207 before OGD treatment to achieve
maximal protection. Interestingly, enhanced DCF fluorescence dropped sharply at
the beginning of recovery and increased again slowly during recovery. Although
DCF fluorescence is widely used as an index of oxidative stress due to its high
sensitivity, it has been reported that under aerobic conditions, the production of
fluorescent DCF from DCFH generates the DCF semiquinone free radical, which
could produce superoxide and hydrogen peroxide, thus further increasing DCFH
oxidation and leading to more DCF fluorescence (Marchesi et al., 1999;Bonini et al.,
2006). In my experiment, DCF fluorescence was generated with OGD treatment in
an anaerobic environment, and thus superoxide and hydrogen peroxide formation as
by-products should be much smaller than under aerobic conditions. However, due to
its superoxide dismutase/catalase activity, EUK-207 appears to interfere with the
DCF assay, possibly by generating hydrogen peroxide as an interim substrate. To
avoid this interference, cultured slices were treated with EUK-207 1 h before OGD,
and DCF fluorescence was measured after 3 h recovery; under these conditions, DCF
fluorescence was significantly lower than in untreated slices. This result strongly
suggests that EUK-207 protects cultures hippocampal slices from OGD-induced cell
death by elimination of ROS accumulation.
The protective effects of EUK-207 against OGD were measured with both
63
LDH release and PI uptake, and these two assays are generally considered as
markers of necrotic cell death due to lethal membrane injury (Dursun et al., 2006).
After cultured slices were subjected to OGD, LDH release reached its highest level
in the first 3 h of recovery, before gradually decreasing at 6 h and 24 h of recovery,
suggesting that after 1 h OGD treatment in cultured slices, more necrotic cell death
took place in the early period of recovery. In contrast to early cell death through
necrosis, apoptosis has generally been considered to be involved in delayed ischemic
cell death (Banasiak et al., 2000). In my experiments, activation of pro-apoptotic
factors was detected relatively early after OGD treatment. In particular, AIF
translocation to the nucleus as well as cytochrome c release from mitochondria was
observed at 3 h of recovery. EUK-207 partly reduced OGD-induced cytochrome c
release and significantly decreased AIF translocation into nucleus. By inhibition of
early pro-apoptotic factor activation, further activation of downstream apoptotic
pathways might be attenuated by EUK-207, suggesting that EUK-207 might also
reduce apoptosis-induced cell death.
In transient forebrain ischemia and permanent focal ischemia, lipid
peroxidation levels have been reported to increase in ischemia-sensitive regions such
as hippocampus and striatum (Bromont et al., 1989;Sharma and Kaundal, 2007).
Both the highly reactive hydroxyl and peroxynitrite radicals can initiate lipid
peroxidation. Once generated, lipid peroxidation can break membrane integrity,
change membrane permeability, fluidity and ion transport, and induce further ROS
production with mitochondrial damage (Green and Reed, 1998;Nigam and Schewe,
2000). In cultured hippocampal slices, when assessed with the thiobarbituric acid
64
assay, lipid peroxidation levels increased after 1 h OGD treatment and peaked at 6 h
recovery. EUK-207 significantly reduced OGD-induced increase in lipid
peroxidation, a result consistent with the notion that lipid peroxidation is
downstream from ROS generation.
In summary, by inhibition of ROS generation during OGD and recovery,
EUK-207 attenuated OGD-induced lipid peroxidation and pro-apoptotic factor
activation, and protected cultured hippocampal slices from ischemic cell death. The
partial protection provided by EUK-207 against OGD-induced cell death also
indicates that factors other than increased ROS formation are involved in cell death.
Importantly, my results indicate that the critical source of ROS formation in ischemia
and reperfusion occurs very early during the ischemic period, and that the use of
ROS scavengers in clinical application might be much more restricted than
previously thought.
65
3. Chapter 3
Developmental changes in NMDA neurotoxicity reflect developmental
changes in subunit composition of NMDA receptors

3.1. Introduction
3.1.1. NMDA receptors
Glutamate is the principal excitatory neurotransmitter in the central nervous
system (CNS) (Collingridge and Lester, 1989). There are two basic types of
glutamate receptors, the ionotropic glutamate receptors and the metabotropic
glutamate receptors, with the former the focus of my study. Ionotropic glutamate

receptors consist of three subtypes, the N-methyl-D-aspartate (NMDA), alpha-
amino-3-hydroxy-5-methylisoxazole-4-propionate (AMPA), and kainate (KA)
receptors. NMDA receptors (NMDARs) play a crucial role in neuronal development,

plasticity and survival; in addition, overactivation of NMDARs is involved in the
pathophysiology of epileptic seizures, hypoxic-ischemic neuronal damage and
excitotoxic cell death (Dingledine et al., 1999). The intracellular signaling pathways
mediating NMDA excitotoxicity are still debated (Waxman and Lynch, 2005). Most
researchers agree that NMDARs produce neurotoxicity via rapid Ca
2+
influx leading
to cellular Ca
2+
overload. Several studies have revealed developmental changes in
brain susceptibility to excitotoxicity (Kuroiwa and Okeda, 1994), although the
mechanisms responsible for these changes have not yet been elucidated. During
ischemia, the decrease of ATP levels in living cells induces the collapse of the
66

Figure 19. Developmental changes of NR1, NR2A or NR2B subunits in rat
hippocampus (from Sans, 2000).
67
mitochondrial membrane potential, malfunction of Na
+
-K
+
-APTase, the reverse of
glutamate transporters, and overactivation of NMDA receptors. The contribution of
NMDA-induced excitotoxicity in ischemic cell death during different developmental
stages is not clear either.
Several NMDAR subtypes have been identified, differing in kinetic
properties, sensitivity to various ligands, permeability to divalent ions, and
interactions with intracellular proteins(Cull-Candy et al., 2001). NMDARs are
composed of two NR1 subunits and at least one type of NR2 subunits with
predominantly NR2A or NR2B subunits in adult rat hippocampus (Wenzel et al.,
1997). As figure 19 shows, during the postnatal period, cortical

neurons exhibit
changes in kinetics of NMDAR-mediated

EPSCs (Barth and Malenka, 2001;Lu et al.,
2001), corresponding to a developmental change

in the composition of the NMDARs
from predominantly NR1/NR2B

to NR1/NR2A oligomers (Liu et al., 2004). It has
been recently proposed that NR2A- and NR2B-containing NMDA receptors are
linked to different intracellular cascades and participate in different functions in
synaptic plasticity and pathological conditions (Kim et al., 2005;Krapivinsky et al.,
2003;Liu et al., 2004), however, the role of NR2A or NR2B in NMDA-induced
excitotoxicity requires to be elucidated.

3.1.2. Calpain
One of the consequences of NMDA receptor activation is an increase in
intracellular calcium levels and activation of calcium-dependent neutral proteases
such as calpains which are ubiquitously distributed in the nervous system (Carafoli
68
and Molinari, 1998). There are 13 different calpain gene products, differing in their
N-terminal sequence, structures of regulatory domain, and calcium binding sites.
Two of the best characterized calpains are μ-calpain and M-calpain, with μ-calpain
requiring micromolar concentration of calcium while M-calpain requiring milimolar
of calcium to be activated in vitro (Glading et al., 2002). Although calpains are
implicated under physiological conditions in synaptic modifications and plasticity,
they are also involved in oxidative stress and neuronal death s (Ray et al., 2000;Kelly
and Ferreira, 2006;Lynch and Baudry, 1987), although the role of calpain in
excitotoxicity has been controversial, with studies reporting clear evidence for the
involvement of this protease, while others reporting the opposite (Bizat et al., 2003;
Korhonen et al., 2005). It has been shown that from newborn to adult animals, there
is a switch from anaerobic to aerobic metabolism during development, and this
switch could contribute to differences between young and mature cells in balancing
intracellular calcium levels after the breakdown of calcium homeostasis (Baudry and
Lynch, 1985), so it is possible that newborn and adult rats might show different
pattern of calpain activation with NMDA treatment and subsequent increase in
intracellular calcium levels. Calpain is also activated in brain ischemia and
reperfusion (Yamashima et al., 2003), and the calpain inhibitor MDL 28170 has been
reported to protect newborn rat brain from hypoxic ischemia (HI) by decreasing both
necrosis and apoptosis (Kawamura et al., 2005).

3.1.3. PSD-95 and NMDA receptors
69
PSD-95 is also known as SAP90 and is one of the most abundant proteins in
the postsynaptic density (PSD) which locates in postsynapse and contains multiple
macromolecular protein complexes (Kim et al., 2005;Sheng and Hoogenraad, 2007).
From the N-terminus, the protein of PSD-95 contains three PDZ domains, an SH3
domain, and a GK domain. Each PDZ is a module of ninety residues and the
members of the PSD-95 family normally show tandem arrangements of several PDZ
sequences. The SH3 domain may interact with the GK domain and stabilize the
structure of PSD-95 (McGee et al., 2001). PSD-95 contributes to the stabilization of
membrane proteins at synapses by interacting with those proteins and plays an
important role in synaptic plasticity. PSD-95 has been reported to be involved in
synaptic plasticity, and in cultured hippocampal neurons, overexpression of PSD-95
increases the number and size of dendritic spines and glutamate receptor activities
(El Husseini et al., 2000).
A recent study shows that during water maze training, NMDAR subunits
(NR1, NR2A, NR2B) and PSD-95 are rapidly recruited to lipid rafts with an increase
in the synaptic transmission efficiency (Delint-Ramirez et al., 2008). The PSD-95
PDZ domain interacts with the C-terminal of NR2 subunits of NMDA receptors and
links NMDA receptors to downstream neurotoxic signaling pathways. In cultured
cortical neurons, disruption of the interaction between NMDA receptors and PSD-95
by suppressing PSD-95 expression has been reported to uncouple NMDA receptor
activity from nitric oxide production and attenuate overactivation of NMDA
receptors induced excitotoxicity (Sattler et al., 1999). When the interactions between
NMDA receptors and PSD-95 are perturbed with a Tat peptide containing the C-
70
terminal residues of the NR2B subunit, neuronal excitotoxicity and ischemic cell
death are also significantly reduced (Cui et al., 2007). NMDA-induced, calpain-
mediated PSD-95 has been reported in organotypic hippocampal slice cultures (Lu et
al., 2000). In the present experiment, developmental changes of NMDA-induced
excitotoxicity, calpain activation, and PSD-95 degradation are studied in a model of
acute hippocampal slices from rats of different ages, and possible mechanisms
involved in these processes are also discussed.

3.2. Materials and Methods
3.2.1. NMDA treatment
After 1 h recovery, hippocampal slices were washed twice with fresh aCSF
and then gently transferred into individual vials, with 2 hippocampal slices per vial,
in 2 ml aCSF containing (in mM): 124 NaCl, 5 KCl, 1.25 NaH
2
PO
4
, 26 NaHCO
3
, 10
glucose, 1.5 MgSO
4
, 2.5 CaCl
2
, saturated with 95% O
2
–5% CO
2
, and further
incubated in the absence or presence of NMDA (100 µM) for 1 h or 3 h at 34 °C. For
NMDA receptor subunit antagonist treatment, slices were incubated with ifenprodil,
Ro25-6981, or NVP-AAM077 for 20 min before adding NMDA.

3.2.2. OGD treatment and LDH assay
After 1 h recovery, hippocampal slices were washed twice with OGD
solution containing (in mM): 124 NaCl, 5 KCl, 1.25 NaH
2
PO
4
, 26 NaHCO
3
, 1.5
MgSO
4
, 2.5 CaCl
2
(pH 7.5), and then transferred into individual vials with 2 slices
per vial in 2 ml OGD solution, previously bubbled with nitrogen for 20 min.
71
Hippocampal slices were incubated in OGD solution at 34 °C for 1 h in anaerobic
vials saturated with nitrogen; in some cases, slices were then collected and processed
for western blots. For LDH assay, incubation medium was collected after the 1 h
incubation period, and slices were washed with aCSF and further incubated for 2 h in
2 ml fresh aCSF solution saturated with 95% O
2
–5% CO
2
. The final LDH release
was the combination of LDH release during the 1 h and 2 h incubation. To measure
LDH release, 0.3 ml of medium solution was mixed with 0.7 ml potassium
phosphate buffer, and after 20 minutes was further mixed with 0.5 ml freshly made
solution containing sodium pyruvate and NADH and immediately followed by
measuring absorbance at 340 nM at 1 min interval.

3.2.3. PI uptake assay
Neuronal damage was also assessed by propidium iodide (PI) uptake as
previously described (Laake et al., 1999). For cell death assay for slices treated with
NMDA or NMDA receptor antagonists, PI (4.6 µg/ml) was added to bath solution
together with different treatment, and after 3 h incubation, slices were fixed
overnight in 4%

paraformaldehyde in 0.1M phosphate buffer at pH 7.4. Then the
slices were transferred to 0.1M phosphate buffer with 20% sucrose for 24 h and
sectioned at 20 µm using a freezing microtome. PI uptake was visualized using a 5X
objective with a microscope fitted with fluorescence detection, and images of PI-
labeled slices were captured with a CCD camera; at this magnification, one image
was sufficient to analyze an entire hippocampal slice. To obtain the best intensity of
images and to avoid saturation, all acute hippocampal slices were exposed for 50 ms,
72
and the camera gain was kept constant throughout each experiment. Fluorescence
intensity was estimated by the following method: first, images were adjusted to gray
levels and captured with Adobe Photoshop, with the background of images in white
and PI-stained structures in black; second, modified images were analyzed
quantitatively by densitometry with ImageJ software. Data are generally shown as
means ± SEM from the indicated number of independent experiments.

3.2.4. Nissl staining
For nissl staining, acute hippocampal slices were also fixed overnight in 4%

paraformaldehyde and transferred to 0.1M phosphate buffer with 20% sucrose for 24
h and sectioned at 20 µm using a freezing microtome. After sections (20 µm) had
been cut, they were mounted and dried onto slides which were passed through
different baths in the following order: 100% EtOH (ethyl alcohol) 2 minutes, xylene
2 minutes, 100% EtOH 2 minutes, 70% EtOH 2 minutes, distilled water 5 minutes,
cresyl violet 3 minutes, distilled water 2 dips, 70% EtOH 5 minutes, 80% EtOH 2
minutes, 90% EtOH 2 minutes, 95% EtOH 2 minutes, 100% EtOH 5 minutes, xylene
5 minutes, and then mounted with permount and waited for 24 h before observed
under microscope.

3.3. Results
3.3.1. NMDA treatment elicits rapid excitotoxicity in acute hippocampal
slices of young but not adult rats

73

Figure 20. Effects of NMDA treatment on LDH release in acute hippocampal slices
from rats of different ages. Hippocampal slices were prepared from rats of the
indicated ages and were recovered for 1 h before adding NMDA. (A) & (B) NMDA-
induced LDH release in acute hippocampal slices from rats of various postnatal ages.
Slices were incubated with NMDA (100 µM) in the absence or presence of APV (50
µM) or MK801 (10 µM) for 1 h (A) or 3 h (B). Results are expressed as fold of
increase over the respective control values and are means ± S.E.M. of 12
experiments. (C) LDH release expressed as fold of values measured in control slices
from 1-week-old rats. Means ± S.E.M. of 10 experiments.
74

Figure 21. Effects of NR2A, NR2B antagonists on NMDA-induced LDH release in
acute hippocampal slices from rats of different ages. Hippocampal slices were
prepared from 1-week-old, 3-week-old and 3-month-old rats, and were incubated
with or without NMDA in the absence or presence of ifenprodil (5 µM), Ro25
(Ro25-6981; 0.5 µM) or NVP (NVP-AAM077; 0.5 µM) for 3 h. Results are
expressed as fold of increase over the respective control values and are means ±
S.E.M. of 8 experiments. * p < 0.05 as compared to control; † p < 0.05 as compared
to NMDA treated slices.
75

Figure 22. Effects of NMDA treatment on Propidium Iodide (PI) staining in acute
hippocampal slices from rats of different ages. (A) Representative images of PI
staining in acute slices of 1-week-old and 3-month-old rats treated in the absence
(control) or presence of NMDA, NMDA plus ifenprodil (5 µM) or NMDA plus NVP
(NVP-AAM077, 0.5 µM). (B) Quantification of PI staining. Results are expressed as
fold of increase over the respective control values and are means ± S.E.M. of 5
experiments. * p < 0.05 as compared to control; † p < 0.05 as compared to NMDA
treated slices.
76
Acute hippocampal slices from rats of various postnatal ages were treated
with NMDA (100 µM) for 1 h or 3 h. Because LDH release was relatively low in the
first hour, 1 h NMDA treatment was followed by a 3 h recovery period in fresh aCSF
to allow enough LDH release into the incubation solution. A 1-hour NMDA
treatment induced a significant release of LDH in slices from 1-3-week-old rats but
not in slices from 2-month- or 3-month-old rats (Fig. 20A). A similar pattern was
observed following a 3 h treatment with NMDA (Fig. 20B). In both cases (1 or 3 h
NMDA treatment), NMDA-induced LDH release was partly blocked by APV (50
µM), and completely blocked by MK801 (10 µM) (Fig. 20A,B). Note that basal
LDH release was higher in slices from 3 month-old rats as compared to that in slices
from 1 week-old rats; however, NMDA treatment did not result in higher LDH
release than in control slices (Fig. 20C).

3.3.2. NMDA-induced excitotoxicity in acute hippocampal slices from
1-week-old rats is mediated through NR2B-containing NMDA receptors
NMDA-induced LDH release in slices from 1-week-old rats was almost
completely blocked by ifenprodil (5 µM) or Ro25-6981 (0.5 µM), antagonists with a
higher affinity for NR2B-containing NMDA receptors. In contrast, application of
NVP-AAM077 (0.5 µM), an antagonist selective for NR2A-containing NMDA
receptors, exhibited no significant protection. In slices from 3-week-old rats,
although both ifenprodil and NVP-AAM077 attenuated NMDA-induced LDH
release, only the effect of ifenprodil reached statistical significance (Fig. 21).

77

Figure 23. Effects of NMDA treatment on Nissl staining in acute slices from 1-
week-old and 3-month-old rats. Acute hippocampal slices from 1-week-old or 3-
month-old rats were incubated with 100 µM NMDA for 3 h in the absence or
presence of ifenprodil (Ifen; 5 µM), NVP-AAM077 (NVP; 0.5 µM), or MK-801
(MK; 10 µM). At the end of incubation, sections were fixed and processed for Nissl
staining.
78

Figure 24. Effects of NMDA on calpain-mediated spectrin degradation in acute
hippocampal slices from rats of different postnatal ages. Acute hippocampal slices
were prepared from 1-week-old, 3-week-old and 3-month-old rats and were
incubated under various experimental conditions. At the end of incubation, slices
were sonicated and aliquots were processed for immunoblotting with spectrin
antibodies. (A) Representative images of western blots indicating the levels of the
calpain-mediated spectrin breakdown products at 150 kDa and 145 kDa (SBDP,
arrows) in slices from rats of the indicated ages incubated in the absence (control) or
presence of NMDA, NMDA plus ifenprodil, NMDA plus NVP (NVP-AAM077), or
NMDA plus MK801. (B) Quantitative analysis of blots similar to those shown in (A).
Blots were scanned and the intensities of SBDP bands were quantified and expressed
as fold increase over the respective control values and the data represent means ±
S.E.M. of 4 experiments. * p < 0.05 as compared to control; † p < 0.05 as compared
to NMDA-treated slices.
79
Propidium iodide (PI) staining also indicated that NMDA treatment induced
higher cell death in slices from 1-week-old rats than in slices from adults (Fig. 22).
Quantitative analysis of staining intensity confirmed the visual impression and
indicated that, as compared to control, NMDA treatment induced a 2.75 ± 0.23 fold
increase in PI staining; this effect could also be completely blocked by MK801 (0.89
± 0.10) and ifenprodil (1.23 ± 0.21), but not by NVP-AAM077. In slices from adult
rats, NMDA treatment (1.12 ± 0.08) had no significant effect as compared to control.
The effect of NMDA treatment was not modified in the presence of MK801,
ifenprodil or NVP-AAM077.
With Nissl staining, NMDA treatment also resulted in a different pattern of
staining in slices from 1-week-old rats than in slices from adults (Fig. 23). In the
former, Nissl staining was increased in CA1 pyramidal cells and the cells appeared
more shrunk. In slices from adult rats, NMDA treatment induced a loss of staining.
In both cases, NMDA-induced changes in Nissl staining were blocked by MK-801,
and the effects of NMDA treatment in slices from both 1-week-old and adult rats
were also blocked by ifenprodil but not by NVP-AAM077 (Fig. 23).

3.3.3. Calpain activation is necessary for NMDA-induced excitotoxicity
in acute hippocampal slices of young rats
Spectrin is one of the preferred substrates of calpain and calpain-mediated
spectrin degradation is now widely used as a marker for calpain activation. NMDA
treatment induced a 3.87 ± 0.46 fold increase in spectrin degradation compared to
control in slices of 1-week-old rats. The enhanced spectrin degradation was blocked
by MK801 (0.41 ± 0.06) and markedly decreased by ifenprodil (1.56 ± 0.30), but
80

Figure 25. Effects of calpain inhibitor III on NMDA-induced calpain activation and
toxicity in acute hippocampal slices from rats of different postnatal ages. (A)
Representative images of western blots showing calpain-mediated spectrin
breakdown products at 150 kDa and 145 kDa (arrows) in slices treated in the absence
(control) or presence of NMDA, NMDA plus calpain inhibitor III (10 µM) or
NMDA plus EGTA (2 mM) w/o calcium. (B) Quantitative analysis of blots similar
to those shown in A. Blots were scanned and the intensities of SBDP bands were
quantified and expressed as fold increase over the respective control values and the
data represent means ± S.E.M. of 5 experiments. (C) LDH release in the medium of
slices treated in the absence (control) or presence of NMDA or NMDA plus calpain
inhibitor III (10 µM) in slices from rats of different ages. Results are expressed as
fold of increase over the respective control values and are means ± S.E.M. of 6
experiments. * p < 0.05 as compared to control; † p < 0.05 as compared to NMDA-
treated slices.
81

Figure 26. Effects of pretreatment with APV or calpain inhibitor III on NMDA-
induced calpain activation in acute hippocampal slices of 3-month-old rats. (A)
Representative images of western blots showing calpain-mediated spectrin
breakdown products at 150 kDa and 145 kDa (arrows) in slices from 3-month-old
rats pre-incubated with APV (50 µM) or calpain inhibitor III (10 µM) during slice
preparation and recovery, and then treated with NMDA (100 µM) for different
periods of time. (B) Quantitative analysis of blots similar to those shown in (A).
Blots were scanned and the intensities of bands were quantified and expressed as
percentage of the respective control values and the data represent means ± S.E.M. of
4 experiments. * p < 0.05 as compared to control.
82

Figure 27. Effects of pretreatment with APV or calpain inhibitor III on NMDA-
induced LDH release in acute hippocampal slices of 3-month-old rats. Acute slices
were treated in the absence (control) or presence of NMDA and NMDA plus MK801,
following pretreatment with APV or calpain inhibitor III during slice preparation and
1 h recovery, and LDH was measured at the end of treatment. Results are expressed
as percentage of the respective control values and are means ± S.E.M. of 6
experiments. * p < 0.05 as compared to control.
83
was not significantly affected by NVP-AAM077 (3.64 ± 0.42) (Fig. 24). Note that
MK801 resulted in a decrease in spectrin degradation below control values,
suggesting that some NMDA receptor activation takes place during the incubation of
slices even in the absence of exogenous NMDA. In slices from 3-week-old rats,
NMDA treatment induced a 1.92 ± 0.13 fold increase in spectrin degradation, which
was partly blocked by ifenprodil (1.55 ± 0.09) and by NVP-AAM077 (1.47 ± 0.07),
and completely blocked by MK801 (1.21 ± 0.17). In contrast, NMDA treatment had
no effect on spectrin degradation in slices from 3-month-old rats (Fig. 24A). When
NMDA was applied together with the membrane permeable calpain inhibitor, calpain
inhibitor III, or with EGTA in the absence of calcium, spectrin degradation was
decreased below control levels in slices from rats of all ages (Fig. 25A). In slices
from 1-week-old rats, calpain inhibitor III also significantly reduced NMDA-induced
LDH release from 4.03 ± 0.22 to 2.62 ± 0.19 (Fig. 25C).

3.3.4. The lack of NMDA-induced toxicity in slices from adult rats is not
due to a priming effect
It has previously been reported that low levels of NMDA receptor activation
could produce a refractory period to subsequent NMDA receptor activation
(Vyklicky, 1993). To test the hypothesis that the lack of NMDA toxicity observed in
slices from adult rats could be due to the release of glutamate and the activation of
NMDA receptors during the preparation of slices from adult rats and/or the
preincubation period, APV (50 µM) or calpain inhibitor III (10 µM) was added to
both cutting medium and recovery solution before adding NMDA (100 µM). When
84
APV was present in the cutting medium and during the preincubation period, neither
LDH release nor spectrin degradation were increased following NMDA treatment for
periods ranging from 10 min to 3 h when compared to control (Fig. 26). With calpain
inhibitor III present in the cutting medium and during the preincubation period, there
were similarly no obvious changes in NMDA-induced LDH release as compared to
control (Fig. 27). However, under these conditions, spectrin degradation was greatly
reduced (by about 60%) in the absence or presence of NMDA for 10 min, or 1 h, and
spectrin degradation levels increased to control levels after 3 h incubation in the
absence of calpain inhibitor III.

3.3.5. Oxygen-glucose deprivation (OGD) induces excitotoxicity in
acute slices from both young and adult rats
To determine whether acute slices from adult rats were refractory to toxicity,
I compared the toxic effects of another insult, oxygen-glucose deprivation (OGD), in
hippocampal slices from young and adult rats. In contrast to the lack of effect of
NMDA treatment on LDH release in slices from adult rats, OGD treatment induced
significant LDH release in slices from adult rats. Increasing the duration of OGD
treatment from 20 to 60 min induced increased LDH release as compared to control
conditions (Fig. 28A). In addition, in slices from 1-week-old rats, MK801 (10 µM)
or calpain inhibitor III (10 µM) significantly reduced OGD-induced LDH release by
37% or 52%; however, this effect was not observed in slices from 3-month-old rats
(Fig. 28B).

85

Figure 28. Effects of oxygen/glucose deprivation (OGD) on LDH release in acute
hippocampal slices from 1-week-old and 3-month-old rats. (A) Acute slices from 3-
month-old rats were incubated for different periods of time (20 min, 40 min and 60
min) in the absence of OGD without or with MK801 (10 µM). Results are expressed
as fold of increase over the respective control values and are means ± S.E.M. of 5
experiments. (B) Acute slices from 1-week-old and 3-month-old rats were incubated
for 1 h in the absence of oxygen and glucose without or with MK801 (10 µM) or
calpain inhibitor III (10 µM). Results are expressed as fold of increase over the
respective control values and are means ± S.E.M. of 7 experiments. * p < 0.05 as
compared to controls; † p < 0.05 as compared to OGD-treated slices.
86

Figure 29. Effects of oxygen/glucose deprivation (OGD) on calpain-mediated
spectrin degradation in acute hippocampal slices from 1-week-old and 3-month-old
rats. (A) & (B) Representative images of western blots showing calpain-mediated
spectrin breakdown products at 150 kDa and 145 kDa (arrows) in slices from 1-
week-old and 3-month-old rats subjected to OGD (1 h), OGD plus ifenprodil (5 µM),
OGD plus NVP (NVP-AAM077, 0.5 µM), OGD plus MK801 (10 µM) or OGD plus
calpain inhibitor III (10 µM). (C) & (D) Quantitative analysis of blots similar to
those shown in (A) & (B). Blots were scanned and the intensities of bands were
quantified and expressed as fold increase over the respective control values and the
data represent means ± S.E.M. of 5 experiments. * p < 0.05 as compared to control;
† p < 0.05 as compared to OGD-treated slices.
87

Figure 30. Effects of NMDA treatment on PSD95 levels in acute hippocampal slices
from 1-week-old and 3-month-old rats. (A) Representative western blots of PSD95
(arrow) in acute slices from 1-week-old and 3-month-old rats incubated in the
absence or presence of NMDA for 1 h. (B) Representative western blots of PSD95 in
acute slices from 1-week-old and 3-month-old rats incubated in the absence or
presence of NMDA for 1 h, NMDA plus ifenprodil (5 µM), NMDA plus NVP
(NVP-AAM077, 0.5 µM) or NMDA plus calpain inhibitor III (10 µM). (C) & (D)
Quantitative analysis of blots similar to those shown in A & B. Blots were scanned
and the intensities of bands were quantified and expressed as percentage of the
respective control values and the data represent means ± S.E.M. of 4 experiments. *
p < 0.05 as compared to controls; † p < 0.05 as compared to NMDA-treated slices.

88
One hour OGD treatment induced a 2.73 ± 0.19 fold increase in spectrin
degradation in slices from 1-week-old rats, an effect that was markedly attenuated by
MK801 (1.31 ± 0.20) or completely blocked by calpain inhibitor III (0.75 ± 0.08)
(Fig. 29). The same OGD treatment induced a small increase in spectrin degradation
in slices from 3-month-old rats (1.34 ± 0.14) and this effect was also blocked by
calpain inhibitor III (0.51 ± 0.07) but not by MK801 (Fig. 29B). NVP-AAM077 (0.5
µM) had no effect on OGD-induced spectrin degradation in slices from either young
or adult rats, while ifenprodil (5 µM) slightly reduced OGD-induced spectrin
degradation in slices from 1-week-old rats (Fig. 29A).

3.3.6. NMDA treatment induces calpain-mediated PSD95 truncation
PSD95 is a major postsynaptic density protein and plays an important role in
the assembly and organization of postsynaptic components of excitatory synapses.
PSD95 levels were much higher in hippocampal slices from 3-month-old than 1-
week-old rats. NMDA treatment induced significant PSD95 degradation in slices
from 1-week-old rats, but had little effect on PSD95 levels in slices from 3-month-
old rats (Fig. 30). Surprisingly, in slices from 1-week-old rats, neither NVP-
AAM077 nor ifenprodil modified NMDA-induced PSD95 degradation, although
calpain inhibitor III totally blocked NMDA-induced PSD95 degradation, and in fact
slightly increased PSD95 levels, suggesting that some degree of PSD95 degradation
might take place during the incubation period (Fig. 30).

3.4. Discussion
89
My results indicate that several changes in the mechanisms of NMDA-
mediated neurotoxicity take place during the postnatal period. First, NMDA elicits a
rapid neurotoxicity in acute slices from young but not adult rat hippocampus. This
effect is observed whether I used LDH release in the medium or PI uptake in
damaged cells to assess cell damage. Few studies have investigated developmental
changes in NMDA toxicity. Using cultured hippocampal slices, no significant
changes in NMDA-mediated toxicity were found with increasing periods of cultures
ranging from 1 week to 4 weeks (Bruce et al., 1995). In contrast, McDonald and coll.
(1988) reported a large decrease in NMDA toxicity during the developmental period
using direct in vivo injection of NMDA in striatum and hippocampus. My current
results therefore indicate that acute hippocampal slices provide a useful model to
study mechanisms of NMDA toxicity as it reproduces these in vivo changes.
Second, NMDA treatment rapidly activates the calcium-dependent protease
calpain in slices from neonatal but not adult rats. We had previously shown that
calpain levels are highest in neonatal rats and decline during the postnatal period
(Baudry et al., 1986), and this effect could account for the observed decrease in
NMDA-mediated calpain activation. In addition, it is possible that adult neurons
have a higher capacity to regulate intracellular calcium levels than neonatal ones, and
that this limits the ability of NMDA treatment to activate calpain. We previously
discussed the possibility that the switch from anaerobic to aerobic metabolism during
the postnatal period plays an important role in developmental changes in synaptic
plasticity and responses to injury, as it provides for increased mitochondrial capacity
to synthesize ATP and to absorb large calcium loads (Baudry and Lynch, 1985). In
90
any event, as blockade of calpain activity significantly reduced NMDA-mediated
neurotoxicity in slices from neonatal rats, my results clearly establish that calpain
activation plays a significant role in neurotoxicity in neonatal rat brain. The role of
calpain in excitotoxicity has been controversial, with studies reporting clear evidence
for the involvement of this protease, while others reporting the opposite (Bizat et al.,
2003; Korhonen et al., 2005). Recent studies using transgenic mice overexpressing
human calpastatin, the endogenous inhibitor of calpain, provide clear evidence for a
critical role of calpain in excitotoxicity (Higuchi et al., 2005).
Several mechanisms could account for the observed developmental changes
in NMDA toxicity. First, the expression of NMDA receptors undergoes subunit- and
region-related changes during postnatal development, with high NR2B and low
NR2A expression at postnatal day 2, and increased expression of NR1 and NR2A
during postnatal development (Liu et al., 2004;Sans et al., 2000). This switch in
NMDA receptors from predominantly NR2B-containing receptors to predominantly
synaptic NR2A-containing receptors matches well with my results, as NMDA-
induced toxicity in slices from 1-week-old rats was completely abolished by
ifenprodil, an antagonist of NR2B-containing receptors and not affected by NVP-
AAM077, an NR2A-containing receptor antagonist. Moreover, the developmental
profile of NMDA-mediated toxicity also matches well the developmental changes in
the ratio NR2B/NR2A (Liu et al., 2004). Furthermore, NR2A- and NR2B-containing
NMDA receptors have been shown to activate different intracellular cascades,
although the results have not been consistent between in vitro and in vivo
experiments (Krapivinsky, 2003; Liu et al., 2004). Such differences may be related to
91
the previously reported differences in the stimulation of synaptic (mostly NR2A-
containing receptors) and extrasynaptic (mostly NR2B-containing receptors) NMDA
receptors, with the former leading to CREB activation and increased BDNF
expression and neuronal survival and the latter leading to neuronal death
(Hardingham and Bading, 2002).
Interestingly, blockade of NR2B-containing receptors almost completely
blocked NMDA-mediated spectrin degradation, indicating that activation of NR2B-
containing receptors leads to calpain activation. Furthermore, results obtained in
slices from 3-week-old rats indicated that blockade of either NR2A-containing or
NR2B-containing receptors elicited a significant decrease in NMDA-mediated
spectrin degradation, suggesting that activation of either type of receptors leads to
calpain stimulation. Nevertheless, blockade of NR2B-containing but not NR2A-
containing receptors was neuroprotective in slices from 3-week-old rats.
These results strongly support the idea that the activation of NR2B-
containing NMDA receptors is more critical for NMDA-mediated neurotoxicity, and
that the postnatal decrease in these receptors is responsible for the developmental
changes in NMDA toxicity. Surprisingly, blocking either receptor did not prevent
calpain-mediated NMDA-mediated degradation of PSD-95 in slices from 1-week-old
rats. This result suggests that there might exist additional subtypes of NMDA
receptors that are linked to calpain activation at this developmental stage and closely
associated with PSD-95. Finally, the lack of NMDA toxicity in slices from adult rats
could be due to high platelet-derived growth factor B-chain (PDGF-B) expression
92
levels as PDGF-B has been reported to protect neurons from NMDA induced
excitotoxicity (Egawa-Tsuzuki et al., 2004).
Several experiments were performed to eliminate alternative explanations for
the lack of NMDA toxicity in slices from adult rats. I first showed that the lack of
NMDA toxicity was not due to the priming of the receptors during slice preparation
and pre-incubation, as previous studies have shown that activation of NMDA
receptors is followed by a refractory period (Izumi et al., 1992). I also compared the
effects of another type of insults, oxygen/glucose deprivation (OGD) in slices from
neonatal and adult rats (Taylor et al., 1999). In my experiment 1 h OGD treatment
produced equivalent and high levels of LDH release in acute hippocampal slices
from 1-week-old and 3-month-old rats. However, OGD-induced toxicity in slices of
young or adult rats also involved different mechanisms, since MK801 and calpain
inhibitor III could partly block OGD-induced LDH release in slices from 1-week-old
rats, but had no effect on slices from adult rats. These results indicate that activation
of NMDA receptors and calpain is involved in OGD-induced cell death in slices
from young, but not adult rats. Although synaptic glutamate release is increased by
OGD treatment, and OGD-induced cell death could be reduced by NMDA receptor
inhibitors (Beck et al., ;Fujimoto et al., ), recent studies have indicated that in
ischemic brain, acidosis might play a key role. It appears more likely that ischemia-
induced neuronal injury is mediated by acidosis and activation of Ca
2+
-permeable
acid-sensing ion channels (ASICs) (Xiong et al., ). This might explain why glutamate
antagonists have failed to show effective neuroprotection in stroke in multiple human
trials.
93
In conclusion, my results showed that NMDA treatment exhibited
developmental decrease in excitotoxicity in acute hippocampal slices. In slices from
neonatal rats, NMDA-induced excitotoxicity was mainly due to the activation of
NR2B-containing receptors and calpain. In contrast, a 1-hour OGD treatment elicited
a similar degree of neurotoxicity in acute hippocampal slices from rats of different
ages, although the mechanisms by which OGD induced neurotoxicity were also
different between young and old rats. Finally, our results indicate that the use of
acute hippocampal slices provide an interesting model to study mechanisms of
NMDA toxicity that can be useful to better understand mechanisms of
neurodegeneration.
94
4. Chapter 4
Positive modulation of AMPA receptors stimulates BDNF-mediated
dendritic protein translation

4.1. Introduction
While overactivation of NMDA receptors is involved in neurodegeneration,
the present experiments (part III) investigate whether positive modulation of AMPA
receptors, another subtype of ionotropic glutamate receptors, is involved in dendritic
protein translation and synaptic plasticity. A recent study shows that the positive
modulation of AMPA receptors with aniracetam plus high concentration of AMPA
(500 µM) protects neurons against glutamate-induced excitotoxicity through the
upregulation of brain-derived neurotropic factor (BDNF) release (Wu et al., 2004).
Dendritic translation plays a critical role in synaptic plasticity, in particular in long-
term potentiation (LTP) and long-term depression (LTD) (Huber et al., 2000;Pfeiffer
and Huber, 2006;Wells and Fallon, 2000), and possibly partly through the regulation
of dendritic protein translation (Takei et al., 2004), BNDF has been shown to play a
key role in synaptic plasticity and learning (Jourdi et al., 2003;Rex et al., 2007). Our
results indicate that, in the absence of exogenous AMPA receptor agonists and with
the upregulation of BDNF release, positive modulation of AMPA receptors increases
protein translation and synaptic plasticity.
4.1.1 AMPA receptor positive modulators
95
AMPA receptor positive modulators (also know as ampakines, such as
CX614) are small molecules derived from the structure of aniracetam, a drug that
potentiates AMPA receptor mediated current. Ampakines are not agonists of AMPA
receptors. Instead, they modulate the rate constant of AMPA receptor for transmitter
binding, channel opening, and desensitization (Arai et al., 1996).
Ampakines can be classified into two distinct subfamilies based on their
physiological properties, with type one (like CX546) effective in prolonging synaptic
responses while type two (like CX516) effective in increasing amplitude (Arai and
Kessler, 2007). These two types of ampakines might bind to different sites of AMPA
receptors because there is no competition between these two types in binding assays.
Ampakines are able to cross the blood-brain barrier, and their abilities to increase the
levels of neurotrophin make them an important candidate to enhance learning and
memory. Ampakines have been reported to increase the encoding of several
memories in rats, including retention of memory in a radial maze task and in an odor-
matching task (Staubli et al., 1994), and also be effective in monkeys and humans for
the treatment of schizophrenia and for alleviating the impairment of performance due
to sleep deprivation (Lynch, 2006;Goff et al., 2001;Porrino et al., 2005).. Because
ampakines are orally bioactive and are able to improve cognitive functions in
humans in a broad range of tests without any detective side effects (Lynch, 2002),
ampakines provide a potential therapeutic approach to treat neurological as well as
neurodegenerative diseases such as mild cognitive impairment and Alzheimer’s
disease.
96
In cultured rat entorhinal/hippocampal slices, the mRNA and protein levels of
BDNF are reversibly increased by CX614 and CX546, which are blocked by AMPA
but not NMDA receptor antagonists (Lauterborn et al., 2000). BDNF mRNA levels
increase with 3 h CX614 treatment, and BDNF protein levels increase with 24 h or
48 h CX614 treatment, and then begin to decrease with longer CX614 treatment,
with a down-regulation of the AMPA receptors levels (Lauterborn et al., 2003). A “24
h on/24 h off” CX614 treatment protocol consistently maintains the increased BDNF
protein levels for up to five days, proving that a sustained elevation of neurotrophin
levels can be achieved with chronic ampakine treatment.

4.1.2 Brain-derived neurotrophic factor (BDNF)
4.1.2.1. Neurotrophins
BDNF is a neurotrophic factor belonging to the neurotrophin family that
promotes neurite outgrowth and differentiation during development, regulates
neuronal survival, and stimulates and controls neurogenesis in mature nervous
system (Chao, 2003). In the neurotrophin family, nerve growth factor (NGF) and
BDNF were the first and the second neurotrophic factors to be characterized, and
neurotrophin-3 (NT-3) and neurotrophin-4/5 (NT-4/5) were discovered thereafter
(Kalb, 2005).
The effects of neurotrophins are mainly through two different receptors. The
first is Trk (tropomyosin related kinase) receptor, which is composed of three
receptors: TrkA receptor is preferentially activated by NGF and NT-3, TrkB receptor
by BDNF and NT-4, and TrkC by NT-3. The activation of Trk receptor leads to
97
dimerization and recruitment of intracellular signaling pathways by phosphorylaiton
of tyrosine residues in its intracellular autoregulatory loop (Huang and Reichardt,
2003). The second receptor is the p75 neurotrophin receptor which forms multimeric
receptor complexes with other receptors after binding with neurotrophins or
proneurotrophins, and mediates neurite outgrowth, myelination and also apoptosis
after activation (Barker, 2004). BDNF and the activation of its receptor have been
suggested to have therapeutic potential in some neurological diseases, including
Parkinson’s disease (PD), Alzheimer’s disease (AD), stroke, schizophrenia, and
amyotrophic lateral sclerosis (ALS) (Price et al., 2007).

4.1.2.2. BDNF, protein translation and synaptic plasticity
The mRNA and protein of TrkB are widely distributed in the brain, and in adult rat
hippocampus, TrkB receptors are mainly distributed in the axons and dendrites of
pyramidal and granule cells, with a small portion of receptors in cell bodies (Drake et
al., 1999). In normal conditions, TrkB receptors are located intracellularly and show
no response to extracellular BDNF, except in dendritic spines (Drake et al., 1999). In
retinal ganglion cells or spinal motor neurons, when cell membrane is depolarized,
both intracellular cAMP levels and cell plasma membrane associated TrkB receptors
may rapidly increase (Meyer-Franke et al., 1998). TrkB receptors are found in
neuronal structures involved in synapse formation, such as axonal growth cone
(Gomes et al., 2003). In TrkB-knockout mices, reduced synapse numbers and
important structural changes of presynaptic boutons are both observed (Martinez et
al., 1998). In organotypic hippocampal slice cultures, BDNF treatment increases
98

Figure 31. Scheme of BDNF activated protein translation (from Takei, 2004).
99
spine density in apical dendrites of CA1 pyramidal neurons. This effect is
independent of action potentials and spontaneous synaptic transmission is sufficient
for BDNF to induce new spine formation (Tyler and Pozzo-Miller, 2003). In
hippocampus, LTP is attenuated when hippocampal slices are treated with TrkB-
immunoglobulin G fusion protein, a BDNF scavenger, and LTP is also decreased in
BDNF-knockout mice in schaffer collateral CA1 synapse, which can be rescued by
exogenous BDNF application (Patterson et al., 1996). All studies above indicate that
in response to synaptic activity, BDNF may contribute to synapse formation,
structural alteration and synaptic plasticity.
In primary cultured cortical neurons, BDNF activates mammalian target of
rapamycin- (mTOR) dependent translation machinery and local protein translation, a
process involving both eIF4E/4E-binding protein (4EBP) and p70S6
kinase/ribosomal S6 protein. BDNF activated mTOR signaling pathway is elucidated
in Fig. 31 (Takei et al., 2004). The process of mRNA translation normally is divided
into three different stages, including initiation, elongation and termination, and in
mammalian system, each stage requires different translation factors such as
eukaryotic initiation factors (eIFs), eukaryotic elongation factors (eEFs), and
eukaryotic termination (release) factors (eRFs) (Proud, 2007). In serum, insulin or
growth factors stimulated protein translation, mTOR plays a key role in the
phosphorylation and regulation of these translation factors. Two important factors
downstream of mTOR are 4EBP1 and p70S6 kinase. 4EBP1 binds to eIF4E and
prevents it from interacting with the scaffold protein eIF4G to form an eIF4F
complex that is important for the initiation of protein translation. Phosphorylation of
100
4EBP1 by mTOR releases it from eIF4E and results in an increase in cap-dependent
translation. Activation of PI3K/mTOR pathway may regulate P70S6 phosphorylation,
which further phosphorylates the ribosome protein S6 kinase (S6K) of the 40S
ribosomal subunit. Activation of S6k may stimulate protein synthesis when growth
factors are available(Hannan et al., 2003). Protein synthesis is required for new
synapse formation and synaptic plasticity including LTP. LTP can be classified as the
early LTP that reflects posttranslational processes and late LTP that requires novel
protein synthesis (Huang and Kandel, 1994). LTP can be abolished with protein
synthesis inhibitor that is applied 15-20 min after LTP induction, and expression of
elongation factor 1A (eEF1A) is increased 5 min after stimulation in dendrites that
have been severed from cell bodies, suggesting that plasticity-related protein is
produced rapidly and locally in dendrites after stimulation (Tsokas et al., 2005). With
the upregulation of BDNF release, our results show that positive modulation of
AMPA receptors activates protein translation machinery, increases dendritic protein
translation, and provides an explanation for the effects of positive AMPA receptor
modulators on facilitation of LTP and improved learning and memory.

4.2. Methods
4.2.1. Acute hippocampal slice preparation
Hippocampi were rapidly dissected from postnatal-day-20 Sprague–Dawley
rats, submerged in chilled cutting medium containing (in mM): 220 sucrose, 20 NaCl,
2.5 KCl, 1.25 NaH
2
PO
4
, 26 NaHCO
3
, 10 glucose, 2 ascorbic acid, 2 MgSO
4
,
bubbled
with 95% O
2
–5% CO
2
, and then cut into transverse slices (400 µm thick) using a
101
McIlwain tissue chopper. After isolation, hippocampal slices were placed in
incubation baskets in an artificial cerebro-spinal fluid (aCSF) containing (in mM):
124 NaCl, 5 KCl, 1.25 NaH
2
PO
4
, 26 NaHCO
3
, 2 ascorbic acid, 10 glucose, 1.5
MgSO
4
, 2.5 CaCl
2
, saturated with 95% O
2
–5% CO
2
and incubated for a 1 h-recovery
period at 37 °C. After 1 h recovery, slices were treated with CX614 or BNDF for 1 h
and were collected at the end of treatment for western blot assay.

4.2.2. Primary cortical neuronal cultures
Primary neuronal cultures were prepared from E18 Sprague-Dawley embryos.
The cortical tissues were dissociated and incubated at 37
o
C for 15 min, in Hank’s
solution with 0.25% trypsin/EDTA and DNAse. The tissues were then rinsed in 10
ml Hank’s solution and further dissociated using a needle and a 10 ml serological
syringe. After dissociation, tissues were centrifuged twice and the pellet was filtered
with a 100 µm filter and after cell numbers were counted with Trypan Blue, neurons
were plated in six-well plates pre-coated with poly D-lysine (50 µg/ml), in the
neurobasal medium supplemented with B27, 0.5 mM glutamine, and 12.5 mM
glutamate. Neurons were cultured at 37
o
C in a humidified incubator containing 95%
air/5%CO
2
for a day, and switched to culture maintenance medium with neurobasal
medium supplemented with B27 and 0.5 mM glutamine. Culture medium was
switched twice a week until neurons were ready for experiments at 10-14 DIV .

4.2.3. Western blots
102
Hippocampal slices were homogenized and cortical neurons were collected in
a lysis buffer containing (in mM): 150 NaCl, 5 EDTA, 1% SDS, 10 mM Tris-HCl
(pH 7.4), 0.5 mM phenylmethylsulphonyl fluoride (PMSF), 2 mg/mL leupeptin, and
1:100 protease inhibitor cocktail. After sample preparation, proteins were loaded to
each lane of 6%, 8% or 10% SDS-PAGE gels and, after separation, proteins were
transferred onto PVDF membranes. The PVDF membranes were blocked with 5%
non-fat milk at room temperature for 1 h and probed with different primary
antibodies (actin, 1:10,000 dilution; Arc, 1:5000 dilution; other primary antibodies,
1:1,000 dilution) at 4
o
C overnight. Membranes were then incubated with secondary
antibodies for 1 h and developed with ECL solutions. Western blots were scanned
and analyzed quantitatively by densitometry with ImageJ software. Data were
generally calculated as fold of control and expressed as means ± S.E.M. from at least
three independent experiments.

4.3. Results and discussion
4.3.1. CNQX blocked CX614-induced increase in TrkB, mTOR or
4EBP1 phosphorylation in acute hippocampal slices
Acute hippocampal slices were prepared from postnatal-day-20 rats. After
preparation and recovery in 37
o
C for 1 h, slices were incubated in aCSF solution in
the absence or presence of CX614 (10 µM for 1 h), CNQX (50 µM for 90 min), or
CX614 plus CNQX (50 µM CNQX pretreated for 30 min, and then co-treated with
10 µM CX614 for 1 h). At the end of incubation, slices were homogenized and
aliquots were processed for immunoblotting with antibodies against phosphorylated
103

Figure 32. Effects of CNQX on CX614-induced increase in TrkB phosphorylation in
acute hippocampal slices. Acute hippocampal slices were prepared from postnatal-
day-20 rats and incubated in aCSF solution in the absence or presence of CX614 (10
µM for 1 h), CNQX (50 µM for 90 min), or CX614 plus CNQX (50 µM CNQX
pretreated for 30 min, and then co-treated with 10 µM CX614 for 1 h). At the end of
incubation, slices were homogenized and aliquots were processed for
immunoblotting with antibodies against phosphorylated TrkB, TrkB and actin.
Quantitative analysis of blots is expressed as fold of control values and the data
represent means ± S.E.M. of 4 experiments. * p < 0.05 as compared with control
group, student’s t-test.
104

Figure 33. Effects of CNQX on CX614-induced increase in mTOR phosphorylation
in acute hippocampal slices. Acute hippocampal slices were incubated in aCSF
solution in the absence or presence of CX614 (10 µM for 1 h), CNQX (50 µM for 90
min), or CX614 plus CNQX (50 µM CNQX pretreated for 30 min, and then co-
treated with 10 µM CX614 for 1 h). At the end of incubation, slices were
homogenized and aliquots were processed for immunoblotting with antibodies
against phosphorylated mTOR, mTOR and actin. Quantitative analysis of blots is
expressed as fold of control values and the data represent means ± S.E.M. of 6
experiments. * p < 0.05 as compared with control group, student’s t-test.
105

Figure 34. Effects of CNQX on CX614-induced increase in 4EBP1 phosphorylation
in acute hippocampal slices. Acute hippocampal slices were incubated in aCSF
solution in the absence or presence of CX614 (10 µM for 1 h), CNQX (50 µM for 90
min), or CX614 plus CNQX (50 µM CNQX pretreated for 30 min, and then co-
treated with 10 µM CX614 for 1 h). At the end of incubation, slices were
homogenized and aliquots were processed for immunoblotting with antibodies
against phosphorylated 4EBP1, 4EBP1 and actin. Quantitative analysis of blots is
expressed as fold of control values and the data represent means ± S.E.M. of 4
experiments. * p < 0.05 as compared with control group, student’s t-test.
106
TrkB, phosphorylated mTOR, or phosphorylated 4EBP1. Protein aliquots were also
probed with antibodies against TrkB, mTOR, or 4EBP1 to test if the total levels of
TrkB, mTOR, or 4EBP1 change after CX614 or CNQX treatment, and all samples
were probed with actin as western blot protein loading control. Compared to control
slices without CX614 or CNQX treatment, acute slices treated with CX614 for 1 h
had a significant increase in TrkB, mTOR or 4EBP1 phosphorylation (Fig. 32, Fig.
33, and Fig. 34, * p < 0.05 as compared to control values, student’s t-test), which was
completely blocked by CNQX, an AMPA receptor antagonist. Slices treated with
CNQX alone showed no changes of TrkB, mTOR or 4EBP1 phosphorylation.

4.3.2. K252a blocked BDNF- or CX614- induced increase in TrkB,
mTOR or 4EBP1 phosphorylation in acute hippocampal slices
Acute hippocampal slices were incubated in aCSF solution in the absence or
presence of BDNF (100 ng/ml for 1 h), BDNF plus K252a, CX614 (10 µM for 1 h),
CX614 plus K252a, and K252a (1 µM for 90 min). For BDNF plus K252a or CX614
plus K252a, slices were pretreated with 1 µM of K252a for 30 min, and then co-
treated with BDNF or CX614 for 60 min. At the end of incubation, slices were
homogenized and aliquots were processed for immunoblotting with antibodies
against phosphorylated TrkB, phosphorylated mTOR, or phosphorylated 4EBP1.
Protein aliquots were also probed with antibodies against TrkB, mTOR, or 4EBP1 to
test if the total levels of TrkB, mTOR, or 4EBP1 change after BDNF, CX614 or
K252a treatment. Protein aliquots were also probed with actin as protein loading
control. Compared to control slices, slices treated with BDNF or CX614 for 1 h had
107

Figure 35. Effects of K252a on BDNF or CX614 -induced increase in TrkB
phosphorylation in acute hippocampal slices. Acute hippocampal slices were
incubated in aCSF solution in the absence or presence of BDNF (100 ng/ml for 1 h),
BDNF plus K252a, CX614 (10 µM for 1 h), CX614 plus K252a, and K252a (1 µM
for 90 min). For BDNF plus K252a or CX614 plus K252a, slices were pretreated
with 1 µM of K252a for 30 min, and then co-treated with BDNF or CX614 for 60
min. At the end of incubation, slices were homogenized and aliquots were processed
for immunoblotting with antibodies against phosphorylated TrkB, TrkB and actin.
Quantitative analysis of blots is expressed as fold of control values and the data
represent means ± S.E.M. of 4 experiments. * p < 0.05 as compared with control
group, student’s t-test.
108

Figure 36. Effects of K252a on BDNF or CX614- induced increase in mTOR
phosphorylation in acute hippocampal slices. Acute hippocampal slices were
incubated in aCSF solution in the absence or presence of BDNF (100 ng/ml for 1 h),
BDNF plus K252a, CX614 (10 µM for 1 h), CX614 plus K252a, and K252a (1 µM
for 90 min). For BDNF plus K252a or CX614 plus K252a, slices were pretreated
with 1 µM of K252a for 30 min, and then co-treated with BDNF or CX614 for 60
min. At the end of incubation, slices were homogenized and aliquots were processed
for immunoblotting with antibodies against phosphorylated mTOR, mTOR and actin.
Quantitative analysis of blots is expressed as fold of control values and the data
represent means ± S.E.M. of 5 experiments. * p < 0.05 as compared with control
group, student’s t-test.
109

Figure 37. Effects of K252a on BDNF or CX614- induced increase in 4EBP1
phosphorylation in acute hippocampal slices. Acute hippocampal slices were
incubated in aCSF solution in the absence or presence of BDNF (100 ng/ml for 1 h),
BDNF plus K252a, CX614 (10 µM for 1 h), CX614 plus K252a, and K252a (1 µM
for 90 min). For BDNF plus K252a or CX614 plus K252a, slices were pretreated
with 1 µM of K252a for 30 min, and then co-treated with BDNF or CX614 for 60
min. At the end of incubation, slices were homogenized and aliquots were processed
for immunoblotting with antibodies against phosphorylated 4EBP1, 4EBP1 and actin.
Quantitative analysis of blots is expressed as fold of control values and the data
represent means ± S.E.M. of 4 experiments. * p < 0.05 as compared with control
group, student’s t-test.
110

Figure 38. Effects of rapamycin on BDNF or CX614-induced increase in 4EBP1
phosphorylation in acute hippocampal slices. Acute hippocampal slices were treated
with or without rapamycin (200 nM for 1 h), BDNF (100 ng/ml for 1 h), BDNF plus
rapamycin, CX614 (10 µM for 1 h), and CX614 plus rapamycin. For BDNF plus
rapamycin or CX614 plus rapamycin, slices were pretreated with 200 nM of
rapamycin for 30min, and then co-treated with BDNF or CX614 for 60 min. At the
end of treatment, slices were homogenized and aliquots were processed for
immunoblotting with antibodies against phosphorylated 4EBP1. Quantitative
analysis of blots is expressed as fold of control values and the data represent means ±
S.E.M. of 3 experiments. * p < 0.05 as compared with BDNF-treated group, ** p <
0.05 as compared with control group, † p < 0.05 as compared with CX614-treated
group, student’s t-test.

111
a significant increase in TrkB, mTOR or 4EBP1 phosphorylation (Fig. 35, Fig. 36,
and Fig. 37, * p < 0.05 as compared to control values, student’s t-test). K252a, a
TrkB receptor inhibitor, significantly reduced TrkB or 4EBP1 phosphorylation to
lower than control levels when applied alone or together with CX614. K252a also
completely blocked BDNF-induced increase in TrkB phosphorylation.

4.3.3. Rapamycin blocked BDNF- or CX614- induced increase in
4EBP1 phosphorylation in acute hippocampal slices
Acute hippocampal slices were incubated in aCSF solution in the absence or
presence of BDNF (100 ng/ml for 1 h), BDNF plus rapamycin, CX614 (10 µM for 1
h), CX614 plus rapamycin, and rapamycin (200 nM for 1 h). For BDNF plus
rapamycin or CX614 plus rapamycin, slices were pretreated with 200 nM of
rapamycin for 30 min, and then co-treated with BDNF or CX614 for 60 min. At the
end of incubation, slices were homogenized and aliquots were processed for
immunoblotting with antibodies against phosphorylated 4EBP1. Compared to control
slices, slices treated with BDNF or CX614 for 1 h had a significant increase in
4EBP1 phosphorylation, which was significantly reduced by rapamycin (Fig. 38).

4.3.4. CX614 had no effect on BDNF expression levels, and Nifedipine,
calcium-free incubation solution, or ryanodine blocked CX614-induced
increase in TrkB phosphorylation in acute hippocampal slices

112

Figure 39. Effects of CX614 on BDNF expression levels in acute hippocampal slices.
Acute hippocampal slices were treated with CX614 (10 µM) for 1 h and at the end of
treatment, slices were homogenized and aliquots were processed for immunoblotting
with antibodies against BDNF. Quantitative analysis of blots is expressed as fold of
control values and the data represent means ± S.E.M. of 4 experiments.
113

Figure 40. Effects of nifedipine, or calcium-free incubation solution on CX614-
induced increase in TrkB phosphorylation in acute hippocampal slices. Acute
hippocampal slices were incubated in aCSF solution in the absence or presence of
CX614 (10 µM for 1 h), CX614 with calcium-free aCSF solution, CX614 plus
nifedipine (10 µM nifedipine pretreated for 10 min, and then co-treated with 10 µM
CX614 for 1 h), calcium-free solution, and nifedipine (10 µM for 70 min). At the end
of incubation, slices were homogenized and aliquots were processed for
immunoblotting with antibodies against phosphorylated TrkB, and TrkB.
Quantitative analysis of blots is expressed as fold of control values and the data
represent means ± S.E.M. of 4 experiments. * p < 0.05 as compared with control
group, student’s t-test.
114

Figure 41. Effects of ryanodine on CX614-induced increase in TrkB phosphorylation
in acute hippocampal slices. Acute hippocampal slices were incubated in aCSF
solution in the absence or presence of CX614 (10 µM for 1 h), CX614 plus
ryanodine (100 µM ryanodine pretreated for 10 min, and then co-treated with 10 µM
CX614 for 1 h), and ryanodine (10 µM for 70 min). At the end of incubation, slices
were homogenized and aliquots were processed for immunoblotting with antibodies
against phosphorylated TrkB, and TrkB. Quantitative analysis of blots is expressed
as fold of control values and the data represent means ± S.E.M. of 4 experiments. * p
< 0.05 as compared with control group, student’s t-test.
115
After preparation and 1 h recovery, acute hippocampal slices were treated with
CX614 (10 µM) for 1 h and at the end of treatment, slices were homogenized and
aliquots were processed for immunoblotting with antibodies against BDNF.
Compared to control slices, slices treated with CX614 for 1 h had no changes on
total BDNF protein levels (Fig. 39).
After 1 h recovery, acute hippocampal slices were incubated in aCSF solution
in the absence or presence of CX614 (10 µM for 1 h), CX614 with calcium-free
aCSF solution, CX614 plus nifedipine or ryanodine (10 µM nifedipine or 100 µM
ryanodine pretreated for 10 min, and then co-treated with 10 µM CX614 for 1 h),
calcium free aCSF solution, nifedipine (10 µM for 70 min), and ryanodine (100 µM).
At the end of incubation, slices were homogenized and aliquots were processed for
immunoblotting with antibodies against phosphorylated TrkB. Protein aliquots were
probed with antibodies against TrkB to test if there were changes of total TrkB levels
after CX614, nifedipine, calcium-free, or ryanodine treatment. Compared to control,
acute slices treated with CX614 for 1 h had a significant increase in TrkB
phosphorylation, which was almost completely blocked by nifedipine, the blocker of
voltage-dependent L-type calcium channels, by deprivation of calcium in incubation
solution, or by ryanodine, a blocker of calcium influx-induced internal calcium
release (Fig. 40 and Fig. 41, * p < 0.05 as compared to control values, student’s t-
test).

4.3.5. Actinomycin D had no effect on CX614-induced increase in TrkB
phosphorylation, but blocked CX614-induced Arc synthesis in primary
116
cortical neuronal cultures
Cultured cortical neurons were treated with or without CX614 (10 µM for 1
h), CX614 plus actinomycin D (10 µM actinomycin D pretreated for 30 min, and
then co-treated with 10 µM CX614 for 1 h), and actinomycin D (10 µM for 90 min).
At the end of treatment, neurons were collected in lysis buffer and aliquots were
processed for immunoblotting with antibodies against phosphorylated TrkB, Arc and
actin. Actinomycin D had no effect on CX614-induced increase in TrkB
phosphorylation, but completely blocked CX614-mediated up-regulation of Arc
protein levels (Fig. 42), suggesting that although transcription is required for
expression of Arc levels, it is not involved in the activation of TrkB receptors.

4.3.6. BDNF and CX614 stimulate dendritic protein translation
After cultured for 14 days, cortical neuronal cultures were transfected with
myristoylated GFP-CaMKII construct, a protein translation reporter. One day after
transfection, selected dendrites were bleached with UV light until fluorescence
became invisible and then treated with BDNF (50 ng/ml) or CX614 (10 µM)
immediately. The time of the start of treatment was designated as time zero, and
fluorescent images were taken every 30 seconds and the fluorescent intencity was
expressed as fold of increase compared to each time zero values (Fig. 43). Both
BDNF and CX614 rapidly (within minutes) stimulated GFP protein synthesis in
dendrites of cultured neurons, and the rate of fluorescence recovery was faster when
compared to vehicle-treated controls.

117

Figure 42. Effects of actinomycin D on CX614-induced increase in TrkB
phosphorylation and Arc protein levels in primary cortical neuron cultures. Cultured
cortical neurons were treated with or without CX614 (10 µM for 1 h), CX614 plus
actinomycin D (1 µg/ml actinomycin D pretreated for 30 min, and then co-treated
with 10 µM CX614 for 1 h), and actinomycin D (1 µg/ml for 90 min). At the end of
treatment, neurons were collected in lysis buffer and aliquots were processed for
immunoblotting with antibodies against phosphorylated TrkB, Arc and actin.
Quantitative analysis of blots is expressed as fold of control values and the data
represent means ± S.E.M. of 3 experiments. * p < 0.05 as compared with control
group, ** p < 0.01 as compared with control group, † p < 0.05 as compared with
CX614-treated group, student’s t-test.

118

Figure 43. Effects of BDNF and CX614 on dendritic local protein synthesis in
primary cultured cortical neurons. Cultured cortical neurons were transfected with
myristoylated GFP-CaMKII construct. One day after transfection, selected dendrites
were bleached with UV light until fluorescence became invisible and then treated
with BDNF (50 ng/ml) or CX614 (10 µM) immediately. The time of the start of
treatment was designated as time zero, and fluorescent images were taken every 30
seconds (from Jourdi, Hsu, and Zhou, 2008).
119

Figure 44. Schematic representation of the mechanisms underlying CX614-activated
dendritic protein translation.
120
4.4. Discussion
The chronic effects of ampakines on BDNF mRNA and protein levels have
been reported before (Lauterborn et al., 2000). Our results show that 1 h CX614
treatment activates TrkB receptor as well as mTOR pathway, and up-regulates
dendritic protein translation. The possible mechanisms of CX-614 activated protein
translation machinery is illustrated in figure 44. Although translation pathways can
be activated by stimulation of NMDA receptors (Huber et al., 2000;Gong et al.,
2006), our results indicated that the effects of ampakine on protein translation is
mainly through the activation of AMPA receptors, because only AMPA receptor
antagonist CNQX, but not NMDA receptor antagonist APV (data not shown), can
block CX614-induced TrkB, mTOR, and 4EBP1 activation.
Tetrodotoxin (TTX) also blocked the activation of protein translation
pathways (data not show), suggesting that activation of synaptic voltage-gated
sodium channels is required in ampakine-activated translation. BDNF plays an
important role in synaptic activity (Patterson et al., 1996). In newborn rat
hippocampal neurons, TTX-dependent BDNF release was observed in electrical
stimulaton, without activation of glutamate receptors. Calcium influx through N-type
calcium channels and intracellular calcium release is necessary for BDNF release
(Balkowiec and Katz, 2002). Calcium influx- and internal calcium release-
dependent mammalian neurotrophins (NTs) secretion was also reported by Kolarow
(Kolarow et al., 2007). Our results indicated that CX614-induced TrkB receptor
phosphorylation was completely blocked by voltage-dependent calcium channel
blocker nifedipine, or when hippocampal slices were incubated in a calcium-free
121
solution. Although AMPA receptors are permeable to sodium ions and normally
impermeable to calcium ions in developed central nervous system (Iizuka et al.,
2000;Kumar et al., 2002), activation of AMPA receptors can also lead to the increase
in intracellular calcium, because influx of sodium ions may cause depolarization
through activation of NMDA receptors or voltage-dependent calcium channels and
result in the increased levels of intracellular calcium (Hoyt et al., 1998).
Our results also show that CX614-induced TrkB receptor activation is
blocked by ryanodine, an inhibitor of calcium influx-induced calcium release from
internal calcium stores. It has been reported that when ryanodine receptors are
deleted, AMPA receptor-mediated LTP in CA1 region of hippocampus becomes
smaller than those of wild-type mice, suggesting that ryanodine receptors are
involved in AMPA receptor mediated synaptic plasticity (Shimuta et al., 2001). My
results suggest that acute application of ampakine activates AMPA receptors and
induces increase in intracellular calcium levels through calcium influx and ryanodine
receptor-regulated internal calcium release, which might increase synaptic BDNF
levels and TrkB receptor activation.
Calcium may influx either through calcium permeable NMDA receptors or
through voltage dependent L-type calcium channels, and both pathways may
contribute to increased internal calcium levels and activity-dependent postsynaptic
NT secretion (Kolarow et al., 2007). However, in my experiments, BDNF release
induced by calcium influx through NMDA receptors is ruled out, because NMDA
receptor antagonist APV fails to block the effect of ampakine. NMDA receptors
mediated BDNF expression is complicated, with NMDA receptor antagonist MK801
122
either increasing BDNF expression in the retrosplenial and entorhinal cortex, or
decreasing BDNF levels in the granule cell layer of hippocampus (Jeon et al., 2006).
Because APV fails to block CX614-induced activation of mTOR translation pathway,
it is possible that the calcium influx through NMDA receptors is below the threshold
to induce BDNF release, or the calcium influx through NMDA receptors or voltage-
dependent calcium channels contributes differently to BNDF release.
Acute ampakine treatment may activate TrkB receptors and downstream
protein translation machinery through the increase in total BDNF expression levels,
or the increase in BDNF release from existing synaptic BDNF stores, or by
transactivation of BDNF receptors through intracellular signal pathways. Our results
indicate that 1 h CX614 has no effect on total BDNF levels in acute hippocampal
slices, suggesting that the upregulation of CX614-dependent translation is not due to
enhanced BDNF expression, but rather due to increased BDNF secretion or TrkB
receptor transactivation. When TrkB-Fc, an extracellular BDNF scavenger (TrkB-Fc
homodimer is about 120 to 130 kDa and not able to cross cell membrane when
applied in incubation medium), is applied together with CX614, CX614-induced
increase in Arc expression and mTOR phosphorylation is significantly blocked (data
not show), suggesting that BDNF release may at least partly contribute to CX614-
induced protein translation. It has been reported that stimulation of AMPA receptors
rapidly activates Src-family protein tyrosine kinase Lyn, which activates MAPK
signaling pathway and the protein translation pathway independent of calcium or
sodium influx through AMPA receptors (Hayashi et al., 1999). It is possible that the
kinase Lyn may indirectly regulate the effect of ampakine on protein translation.
123
Activation of TrkB receptors increases the phosphorylation of mTOR, which
activates downstream eIF4E and S6 kinase and initiates the process of protein
translation (Takei et al., 2004). My results indicated that rapamycin, a blocker of
mTOR, inhibits BDNF and CX614-induced phosphorylation of 4EBP1, proving that
mTOR plays a key role in ampakine-induced protein translation. In primary cortical
neurons, CX614 increased the total Arc protein levels and dendritic synthesis of
myristoylated GFP (myr-GFP). This effect of CX614 might help to explain the
effects of ampakines on the enhancement of synaptic plasticity and learning.
It is interesting that both mTOR blocker rapamycin and transcription inhibitor
actinomycin D blocks CX614-mediated upregulation of total Arc levels, however,
actinomycin D has no effect on CX614-induced TrkB phosphorylation. My results
indicate that activation of TrkB receptor by increased BDNF release or
transactivation is independent of transcription, but the downstream protein synthesis
requires the activation of both transcription and translation, and the lack of either
process may lead to decreased rate of protein synthesis, such as the immediate early
genes Arc.
In summary, our results demonstrate that acute application of ampakines
activates BDNF receptor and dendritic local protein translation, and this process
requires extracellular calcium influx, internal calcium release, and activation of
mTOR pathway. Our results are the first to show that positive modulation of AMPA
receptors activates translation machinery and increases local protein synthesis which
are involved in synaptic plasticity, and suggest that ampakines may contribute to
consolidation of memory and to the therapy of cognitive impairment in elder patients.
124
5. Chapter 5: Conclusions

Stroke is the rapid interruption of blood supply to the brain, and it is the third
leading cause of death in the United States. Although there have been numerous
studies related to developing new ischemic therapies, few drugs have shown
protective effects. Many mechanisms have been proposed to account for ischemic
cell death and their studies are complicated due to the various events involved in
ischemia and reperfusion and the many crosstalks among those events. To alleviate
ischemia-induced neuronal damage, it is important to find key factors that are the
main cause of ischemic cell death and to find drugs which can stop the activation of
these key factors during ischemia and reperfusion (recovery). In the first part of my
dissertation, I tested the hypothesis that ROS accumulation represents a key factor to
account for both acute (2 to 4 h) and relatively delayed (24 h) ischemic cell death.
In my experiments, I used OGD in acute and cultured hippocampal slices as
well as two synthetic superoxide dismutase/catalase mimetics, EUK-207 and EUK-
189, to investigate the roles of reactive oxygen species and other events such as the
depletion of ATP levels, MAPKs, lipid peroxidation, and the activation of pro-
apoptotic factors in ischemic death. Application of EUK-189 or EUK-207 during
OGD and recovery decreased LDH release and PI uptake in acute slices from 2-
month-old rats. In acute slices, EUK-189 or EUK-207 also partly blocked OGD-
induced ATP depletion and ROS generation. Although it has been generally assumed
that more free radicals are generated during reperfusion, our results indicate that
increased formation of reactive oxygen species might take place relatively early
during OGD, because EUK-189 and EUK-207 produced better neuroprotection when
125
applied at the beginning of the OGD period than during the recovery period. In acute
hippocampal slices from 2-month-old rats, ATP levels decreased rapidly with OGD
treatment, and significant ATP recovery was observed only when EUK-189 or EUK-
207 was applied at the beginning of OGD, suggesting that inhibition of early ROS
accumulation during OGD can partly prevent ATP depletion, which is the initial part
of the cascade activated in ischemic cell death
Results in cultured hippocampal slices also indicated that excessive ROS
accumulation takes place very early at the beginning of OGD. Our results show that
EUK-207 has to be present in cultured slices before OGD treatment, instead of being
present during OGD and recovery, to obtain maximal neuroprotection. When OGD-
induced accumulation of ROS was evaluated with the fluorescent probe DCF, DCF
fluorescence had a much larger increase during OGD treatment than during recovery.
Combining the results from both acute and cultured slices, we conclude that ROS are
generated throughout the OGD and the recovery period, but that there is a critical
period at the beginning of the OGD treatment; we propose that ROS generated
during this period trigger events that lead to irreversible ischemic neuronal damage.
Our results have obvious consequences regarding the potential use of ROS
scavengers for the treatment of stroke in humans, and could account for the recent
failure of the free radical scavenger NXY-059 in clinical trials.
In cultured hippocampal slices, AIF translocation to the nucleus as well as
cytochrome c release from mitochondria were observed early after 1 h OGD, and
EUK-207 significantly decreased AIF translocation into nucleus. Lipid peroxidation
levels peaked at 6 h of recovery and were also significantly reduced by EUK-207.
126
Therefore, it is clear that by inhibiting ROS accumulation during OGD and recovery,
EUK-207 attenuates OGD-induced lipid peroxidation and pro-apoptotic factor
activation in cultured slices, partly restores ATP depletion in acute slices, and
protects both cultured and acute hippocampal slices from ischemic cell death. My
results indicate that among all the events that are involved in ischemic cell death,
ROS generation/accumulation plays a key role in necrotic cell death, supporting the
further evaluation of this class of free radical scavengers for the treatment of
ischemic cell damage.
In my work, I compared changes in different factors that have been proposed
to contribute to ischemic cell death using acute hippocampal slices prepared from
newborn and adult rats. My results clearly indicated that the rate of ATP depletion,
activation/inactivation of ERK1/2, and NMDA-induced excitotoxicity exhibited
different patterns in young and adult rats. Thus, while ATP levels decreased rapidly
with OGD treatment in acute hippocampal slices from 2-month-old rats, the decrease
was much slower in acute hippocampal slices from postnatal day 10 (p10) rats.
Similarly, extracellular signal-regulated kinases 1 and 2 (ERK1/2) was inactivated
after OGD in slices from 2-month-old rats, but remained unchanged or even slightly
increased in hippocampal slices prepared from p10 rats. The role of ERK1/2 in
ischemic cell death is ambiguous, as ERK1/2 has been reported to be either activated
or inactivated following ischemia and reperfusion, and ERK1/2 activation has been
reported to promote neuronal survival as well as cell death (Namura et al., 2001;Zhu
et al., 2005;Murray et al., 1998;Fahlman et al., 2002). My results showed that
activation or inactivation of ERK1/2 during ischemia not only depends on different
127
ischemic models, but also on different developmental stages. Furthermore, my
results indicate that changes in ERK1/2 following OGD are not related to OGD-
induced cell death. Nevertheless, there remain some questions that need to be
answered, such as what are the mechanisms causing developmental changes in
ERK1/2 activation during OGD treatment, and whether these changes are related to
the slower decrease in ATP levels in slices prepared from younger rats?
In addition, a developmental change in the contribution of NMDA receptors
in excitotoxicity was also observed in OGD-induced cell death in slices from
neonatal and adult rats. My results indicated that 1 h OGD treatment induced similar
increase in cell death in acute hippocampal slices from 1-week-old and 3-month-old
rats; however, the NMDA receptor antagonist MK801 partly blocked OGD-induced
cell death in slices from newborn rats, but had no effect on slices from adult rats,
suggesting that activation of NMDA receptors is involved in OGD-induced cell death
in acute hippocampal slices from newborn, but not adult rats. Excitotoxicity is
generally studied in dissociated neurons, cultured hippocampal slices or intact
animals; however, the requirements of dissociated neurons or cultured slices to use
prenatal or juvenile rats seriously limit the advantages of these in vitro approaches.
My experiments are the first to use acute hippocampal slices to study the role of
excitotoxicity in OGD-induced cell death during development, and to investigate
developmental changes of overactivation of NMDA receptors induced excitotoxicity
and to relate these changes to developmental changes in NMDA receptor subunit
composition.
128
In acute hippocampal slices, NMDA elicited a rapid neurotoxicity from
young but not adult rats. NMDA treatment also activated the calcium-dependent
protease calpain in slices from neonatal but not adult rats; one possible explanation
for this difference in calpain activation could be due to the fact that adult neurons
have a higher capacity to regulate intracellular calcium levels than neonatal ones.
The role of calpain in excitotoxicity has been controversial, with studies reporting
clear evidence for the involvement of this protease, while others reporting the
opposite (Bizat et al., 2003; Korhonen et al., 2005). My results indicate that blockade
of calpain activity significantly reduces NMDA-mediated neurotoxicity in slices
from neonatal rats but not from adult rats, suggesting that calpain activation plays an
important role in neurotoxicity in neonatal but not in adult rat brain.
Although NR2A- and NR2B-containing NMDA receptors are involved in
different functions in synaptic plasticity (Kim et al., 2005;Krapivinsky et al.,
2003;Liu et al., 2004), the role of NR2A and NR2B subunits in NMDA-induced
excitotoxicity is not clear. The developmental switch in NMDA receptors from
predominantly NR2B-containing receptors to predominantly NR2A-containing
receptors matches well with my results, as NMDA-induced toxicity in slices from
neonatal rats was completely abolished by NR2B but not NR2A antagonists.
Blockade of NR2B-containing receptors in slices from neonatal rats also almost
completely blocked NMDA-mediated spectrin degradation, indicating that activation
of NR2B-containing receptors led to calpain activation. My results strongly support
the notion that activation of NR2B-containing NMDA receptors is more critical for
NMDA-mediated neurotoxicity, and that the postnatal decrease in these receptors is
129
responsible for the developmental changes in NMDA toxicity. Based on my results, I
hypothesize that it is the ratio of NR2B to NR2A in synapse that determines whether
overactivation of NMDA receptors would induce exitotoxicity. Increased NR2A
expression levels during postnatal development might reduce NR2B subunit-induced
toxicity and/or even provide protection against excitotoxicity or ischemia-induced
cell death. My hypothesis is consistent with the report that in mature synapses, the
stimulation of synaptic NMDA receptors (mostly containing NR2A subunits) leads to
CREB activation and increased BDNF expression and neuronal survival, while the
stimulation of extrasynaptic NMDA receptors (mostly containing NR2B subunits)
leads to toxicity and neuronal death (Hardingham and Bading, 2002), although
results from this study have been challenged (ref). My results clearly confirmed that
NMDA receptor-induced toxicity is involved in OGD-induced cell death, especially
in neonatal rats. Antagonists of NMDA receptors have been tested in animal models
of excitotoxicity including stroke and have shown great therapeutic potential in a
variety of neurodegenerative disorders such as acute cerebral ischemic conditions.
However, NMDA antagonists have failed in human clinical trials, possibly because
NMDA receptor site antagonists or channel blockers also inhibit normal NMDA
receptor functions and produce a variety of unwanted side effects such as
psychotomimetic effects and memory impairment. My studies, by elucidating the
roles of NR2A and NR2B in NMDA receptor-induced toxicity, support the idea that
instead of using NMDA receptor antagonists, specific antagonists of NMDA receptor
subunits should be tested in order to minimize the side effects and to optimize
neuroprotection.
130
While overactivation of NMDA receptors is involved in neurodegeneration,
our results show that positive modulation of AMPA receptors, another subtype of
ionotropic glutamate receptors, is involved in dendritic protein translation and
synaptic plasticity. In contrast to the study reporting that positive modulation of
AMPA receptors with aniracetam in association with a high concentration of AMPA
(500 µM) could protect neurons against glutamate-induced excitotoxicity through
BDNF release (Wu et al., 2004), our results indicate that AMPA receptor positive
modulators (ampakines), by prolonging the effects of endogenous glutamate at
AMPA receptors, elicit calcium influx through voltage-gated calcium channels and
intracellular calcium release from internal calcium stores; these events are both
necessary to produce BDNF release presumably from postsynaptic sites (although it
is not possible to eliminate the possibility that similar events take place
presynaptically), leading to TrkB receptor activation. As a result of TrkB receptor
activation, ampakines increase mTOR phosphorylation, activate the downstream
effectors eIF4E and p70S6 kinase, and increase total Arc levels and dendritic
synthesis of myristoylated GFP, a translation reporter. Our results show for the first
time that positive modulation of AMPA receptors increases dendritic protein
translation, mainly through increased BDNF release but not total BDNF protein
levels; in addition, this process is independent of transcription because actinomycin
D, a transcription inhibitor, fails to block ampakine-induced TrkB activation. The
effects of ampakines on dendritic protein translation could well be responsible for the
learning and memory-enhancing properties of this family of molecules, as well as for
their neuroprotective properties that have been shown in various models of
131
neurodegeneration. Our results support the potential usage of ampakines for the
improvement of cognitive functions in patients suffering from neurodegenerative
diseases such as mild cognitive impairment and Alzheimer’s disease.
In conclusion, my results show that glutamate by activating NMDA receptors
participates in ischemia, excitotoxicity and neurodegeneration, and by activating
AMPA receptors in the presence of AMPA receptor positive modulators stimulates
dendritic protein synthesis, thereby playing a critical role in synaptic plasticity and
learning. Furthermore, activity-dependent activation of excitatory synapses results in
cell membrane depolarization, calcium influx, brief activation of calcium-dependent
signaling cascades (such as activation of calpain or CaMkIIα), leading to changes in
dendritic cytoskeleton and protein synthesis; on the other hand, excessive activation
of excitatory synapses possibly coupled with malfunction of glutamate transporters
elicits excessive calcium influx, prolonged activation of calcium-dependent proteases
such as calpain, breakdown of calcium homeostasis and neuronal damage. It is clear
that calcium influx and calpain activation may be the common factors involved in
both cases. Our laboratory previously showed that calpain-mediated mGluR1α
truncation enhances NMDA-induced excitotoxicity (Xu et al., 2007), while it has
also been reported that cleavage of β-catenins by NMDA receptor-activated calpain
produces stable fragments that link calpain to genetic control of neuronal plasticity
(Abe and Takeichi, 2007). In my experiments, the roles of calpain in NMDA-
mediated excitotoxicity and OGD-induced cell death have been evaluated. As
calcium influx and subsequent increase in intracellular calcium levels are essential in
dendritic protein synthesis regulated by positive modulation of AMPA receptors, the
132
role of calpain in this process is also worthy to be further investigated. While my
studies cover both neurodegeneration and synaptic plasticity, the common pathways
they share suggest that depending on activation conditions, the same pathway might
act on distinct substrates and produce quite different results.

133
References
Abe K, Saito H (2000) Amyloid beta neurotoxicity not mediated by the mitogen-
activated protein kinase cascade in cultured rat hippocampal and cortical neurons.
Neurosci Lett 292: 1-4.
Abe K, Takeichi M (2007) NMDA-receptor activation induces calpain-mediated
beta-catenin cleavages for triggering gene expression. Neuron 53: 387-397.
Abramov AY, Scorziello A, Duchen MR (2007) Three distinct mechanisms generate
oxygen free radicals in neurons and contribute to cell death during anoxia and
reoxygenation. J Neurosci 27: 1129-1138.
Alessandrini A, Namura S, Moskowitz MA, Bonventre JV (1999) MEK1 protein
kinase inhibition protects against damage resulting from focal cerebral ischemia.
Proc Natl Acad Sci U S A 96: 12866-12869.
Anderson I, Adinolfi C, Doctrow S, Huffman K, Joy KA, Malfroy B, Soden P,
Rupniak HT, Barnes JC (2001) Oxidative signalling and inflammatory pathways in
Alzheimer's disease. Biochem Soc Symp 141-149.
Arai A, Kessler M, Rogers G, Lynch G (1996) Effects of a memory-enhancing drug
on DL-alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor currents
and synaptic transmission in hippocampus. J Pharmacol Exp Ther 278: 627-638.
Arai AC, Kessler M (2007) Pharmacology of ampakine modulators: from AMPA
receptors to synapses and behavior. Curr Drug Targets 8: 583-602.
Armstead WM, Mirro R, Thelin OP, Shibata M, Zuckerman SL, Shanklin DR,
Busija DW, Leffler CW (1992) Polyethylene glycol superoxide dismutase and
catalase attenuate increased blood-brain barrier permeability after ischemia in piglets.
Stroke 23: 755-762.
Baker K, Marcus CB, Huffman K, Kruk H, Malfroy B, Doctrow SR (1998) Synthetic
combined superoxide dismutase/catalase mimetics are protective as a delayed
treatment in a rat stroke model: a key role for reactive oxygen species in ischemic
brain injury. J Pharmacol Exp Ther 284: 215-221.
Balkowiec A, Katz DM (2002) Cellular mechanisms regulating activity-dependent
release of native brain-derived neurotrophic factor from hippocampal neurons. J
Neurosci 22: 10399-10407.
Banasiak KJ, Xia Y, Haddad GG (2000) Mechanisms underlying hypoxia-induced
neuronal apoptosis. Prog Neurobiol 62: 215-249.
Barker PA (2004) p75NTR is positively promiscuous: novel partners and new
insights. Neuron 42: 529-533.
134
Barone FC, Irving EA, Ray AM, Lee JC, Kassis S, Kumar S, Badger AM, Legos JJ,
Erhardt JA, Ohlstein EH, Hunter AJ, Harrison DC, Philpott K, Smith BR, Adams JL,
Parsons AA (2001) Inhibition of p38 mitogen-activated protein kinase provides
neuroprotection in cerebral focal ischemia. Med Res Rev 21: 129-145.
Barth AL, Malenka RC (2001) NMDAR EPSC kinetics do not regulate the critical
period for LTP at thalamocortical synapses. Nat Neurosci 4: 235-236.
Beck J, Lenart B, Kintner DB, Sun D Na-K-Cl cotransporter contributes to
glutamate-mediated excitotoxicity. J Neurosci 2003 Jun;15;23: 5061-5068.
Bickler PE, Gallego SM, Hansen BM (1993) Developmental changes in intracellular
calcium regulation in rat cerebral cortex during hypoxia. J Cereb Blood Flow Metab
13: 811-819.
Bode AM, Dong Z (2007) The functional contrariety of JNK. Mol Carcinog 46: 591-
598.
Bonini MG, Rota C, Tomasi A, Mason RP (2006) The oxidation of 2',7'-
dichlorofluorescin to reactive oxygen species: a self-fulfilling prophesy? Free Radic
Biol Med 40: 968-975.
Borsello T, Clarke PG, Hirt L, Vercelli A, Repici M, Schorderet DF, Bogousslavsky
J, Bonny C (2003) A peptide inhibitor of c-Jun N-terminal kinase protects against
excitotoxicity and cerebral ischemia. Nat Med 9: 1180-1186.
Bromont C, Marie C, Bralet J (1989) Increased lipid peroxidation in vulnerable brain
regions after transient forebrain ischemia in rats. Stroke 20: 918-924.
Bruce AJ, Sakhi S, Schreiber SS, Baudry M (1995) Development of kainic acid and
N-methyl-D-aspartic acid toxicity in organotypic hippocampal cultures. Exp Neurol
132: 209-219.
Cadenas E, Sies H (1998) The lag phase. Free Radic Res 28: 601-609.
Cao G, Clark RS, Pei W, Yin W, Zhang F, Sun FY, Graham SH, Chen J (2003)
Translocation of apoptosis-inducing factor in vulnerable neurons after transient
cerebral ischemia and in neuronal cultures after oxygen-glucose deprivation. J Cereb
Blood Flow Metab 23: 1137-1150.
Carafoli E, Molinari M (1998) Calpain: a protease in search of a function? Biochem
Biophys Res Commun 247: 193-203.
Chao MV (2003) Neurotrophins and their receptors: a convergence point for many
signalling pathways. Nat Rev Neurosci 4: 299-309.
135
Chen H, Kintner DB, Jones M, Matsuda T, Baba A, Kiedrowski L, Sun D (2007)
AMPA-mediated excitotoxicity in oligodendrocytes: role for Na(+)-K(+)-Cl(-) co-
transport and reversal of Na(+)/Ca(2+) exchanger. J Neurochem 102: 1783-1795.
Chen JJ, Yu BP (1994) Alterations in mitochondrial membrane fluidity by lipid
peroxidation products. Free Radic Biol Med 17: 411-418.
Clausen A, Doctrow S, Baudry M (2008) Prevention of cognitive deficits and brain
oxidative stress with superoxide dismutase/catalase mimetics in aged mice.
Neurobiol Aging.
Claytor RB, Aranson NJ, Ignotz RA, Lalikos JF, Dunn RM (2007) Remote ischemic
preconditioning modulates p38 MAP kinase in rat adipocutaneous flaps. J Reconstr
Microsurg 23: 93-98.
Collingridge GL, Lester RA (1989) Excitatory amino acid receptors in the vertebrate
central nervous system. Pharmacol Rev 41: 143-210.
Cui H, Hayashi A, Sun HS, Belmares MP, Cobey C, Phan T, Schweizer J, Salter
MW, Wang YT, Tasker RA, Garman D, Rabinowitz J, Lu PS, Tymianski M (2007)
PDZ protein interactions underlying NMDA receptor-mediated excitotoxicity and
neuroprotection by PSD-95 inhibitors. J Neurosci 27: 9901-9915.
Cull-Candy S, Brickley S, Farrant M (2001) NMDA receptor subunits: diversity,
development and disease. Curr Opin Neurobiol 11: 327-335.
Daugas E, Susin SA, Zamzami N, Ferri KF, Irinopoulou T, Larochette N, Prevost
MC, Leber B, Andrews D, Penninger J, Kroemer G (2000) Mitochondrio-nuclear
translocation of AIF in apoptosis and necrosis. FASEB J 14: 729-739.
Delint-Ramirez I, Salcedo-Tello P, Bermudez-Rattoni F (2008) Spatial memory
formation induces recruitment of NMDA receptor and PSD-95 to synaptic lipid rafts.
J Neurochem 106: 1658-1668.
Dingledine R, Borges K, Bowie D, Traynelis SF (1999) The glutamate receptor ion
channels. Pharmacol Rev 51: 7-61.
Dirnagl U, Lindauer U, Them A, Schreiber S, Pfister HW, Koedel U, Reszka R,
Freyer D, Villringer A (1995) Global cerebral ischemia in the rat: online monitoring
of oxygen free radical production using chemiluminescence in vivo. J Cereb Blood
Flow Metab 15: 929-940.
Doctrow SR, Huffman K, Marcus CB, Tocco G, Malfroy E, Adinolfi CA, Kruk H,
Baker K, Lazarowych N, Mascarenhas J, Malfroy B (2002) Salen-manganese
complexes as catalytic scavengers of hydrogen peroxide and cytoprotective agents:
structure-activity relationship studies. J Med Chem 45: 4549-4558.
136
Drake CT, Milner TA, Patterson SL (1999) Ultrastructural localization of full-length
trkB immunoreactivity in rat hippocampus suggests multiple roles in modulating
activity-dependent synaptic plasticity. J Neurosci 19: 8009-8026.
Dursun B, He Z, Somerset H, Oh DJ, Faubel S, Edelstein CL (2006) Caspases and
calpain are independent mediators of cisplatin-induced endothelial cell necrosis. Am
J Physiol Renal Physiol 291: F578-F587.
Egawa-Tsuzuki T, Ohno M, Tanaka N, Takeuchi Y, Uramoto H, Faigle R, Funa K,
Ishii Y, Sasahara M (2004) The PDGF B-chain is involved in the ontogenic
susceptibility of the developing rat brain to NMDA toxicity. Exp Neurol 186: 89-98.
El Husseini AE, Schnell E, Chetkovich DM, Nicoll RA, Bredt DS (2000) PSD-95
involvement in maturation of excitatory synapses. Science 290: 1364-1368.
Fahlman CS, Bickler PE, Sullivan B, Gregory GA (2002) Activation of the
neuroprotective ERK signaling pathway by fructose-1,6-bisphosphate during
hypoxia involves intracellular Ca2+ and phospholipase C. Brain Res 958: 43-51.
Fiskum G, Murphy AN, Beal MF (1999) Mitochondria in neurodegeneration: acute
ischemia and chronic neurodegenerative diseases. J Cereb Blood Flow Metab 19:
351-369.
Folbergrova J (1993) Cerebral energy state of neonatal rats during seizures induced
by homocysteine. Physiol Res 42: 155-160.
Forman LJ, Liu P, Nagele RG, Yin K, Wong PY (1998) Augmentation of nitric
oxide, superoxide, and peroxynitrite production during cerebral ischemia and
reperfusion in the rat. Neurochem Res 23: 141-148.
Frotscher M, Zafirov S, Heimrich B (1995) Development of identified neuronal
types and of specific synaptic connections in slice cultures of rat hippocampus. Prog
Neurobiol 45: 143-164.
Fujimoto S, Katsuki H, Kume T, Kaneko S, Akaike A Mechanisms of oxygen
glucose deprivation-induced glutamate release from cerebrocortical slice cultures.
Neurosci Res 2004 Oct;50: 179-187.
Gahwiler BH, Capogna M, Debanne D, McKinney RA, Thompson SM (1997)
Organotypic slice cultures: a technique has come of age. Trends Neurosci 20: 471-
477.
Galeffi F, Sinnar S, Schwartz-Bloom RD (2000) Diazepam promotes ATP recovery
and prevents cytochrome c release in hippocampal slices after in vitro ischemia. J
Neurochem 75: 1242-1249.
Glading A, Lauffenburger DA, Wells A (2002) Cutting to the chase: calpain
proteases in cell motility. Trends Cell Biol 12: 46-54.
137
Goff DC, Leahy L, Berman I, Posever T, Herz L, Leon AC, Johnson SA, Lynch G
(2001) A placebo-controlled pilot study of the ampakine CX516 added to clozapine
in schizophrenia. J Clin Psychopharmacol 21: 484-487.
Gomes AR, Correia SS, Carvalho AL, Duarte CB (2003) Regulation of AMPA
receptor activity, synaptic targeting and recycling: role in synaptic plasticity.
Neurochem Res 28: 1459-1473.
Gong R, Park CS, Abbassi NR, Tang SJ (2006) Roles of glutamate receptors and the
mammalian target of rapamycin (mTOR) signaling pathway in activity-dependent
dendritic protein synthesis in hippocampal neurons. J Biol Chem 281: 18802-18815.
Green DR, Reed JC (1998) Mitochondria and apoptosis. Science 281: 1309-1312.
Hannan KM, Brandenburger Y, Jenkins A, Sharkey K, Cavanaugh A, Rothblum L,
Moss T, Poortinga G, McArthur GA, Pearson RB, Hannan RD (2003) mTOR-
dependent regulation of ribosomal gene transcription requires S6K1 and is mediated
by phosphorylation of the carboxy-terminal activation domain of the nucleolar
transcription factor UBF. Mol Cell Biol 23: 8862-8877.
Hayashi T, Umemori H, Mishina M, Yamamoto T (1999) The AMPA receptor
interacts with and signals through the protein tyrosine kinase Lyn. Nature 397: 72-76.
Henderson D, Bielefeld EC, Harris KC, Hu BH (2006) The role of oxidative stress in
noise-induced hearing loss. Ear Hear 27: 1-19.
Hoyt KR, Arden SR, Aizenman E, Reynolds IJ (1998) Reverse Na+/Ca2+ exchange
contributes to glutamate-induced intracellular Ca2+ concentration increases in
cultured rat forebrain neurons. Mol Pharmacol 53: 742-749.
Huang EJ, Reichardt LF (2003) Trk receptors: roles in neuronal signal transduction.
Annu Rev Biochem 72: 609-642.
Huang YY, Kandel ER (1994) Recruitment of long-lasting and protein kinase A-
dependent long-term potentiation in the CA1 region of hippocampus requires
repeated tetanization. Learn Mem 1: 74-82.
Huber KM, Kayser MS, Bear MF (2000) Role for rapid dendritic protein synthesis in
hippocampal mGluR-dependent long-term depression. Science 288: 1254-1257.
Iizuka M, Nishimura S, Wakamori M, Akiba I, Imoto K, Barsoumian EL (2000) The
lethal expression of the GluR2flip/GluR4flip AMPA receptor in HEK293 cells. Eur J
Neurosci 12: 3900-3908.
Ikeda Y, Long DM (1990) The molecular basis of brain injury and brain edema: the
role of oxygen free radicals. Neurosurgery 27: 1-11.
138
Izumi M, McDonald MC, Sharpe MA, Chatterjee PK, Thiemermann C (2002)
Superoxide dismutase mimetics with catalase activity reduce the organ injury in
hemorrhagic shock. Shock 18: 230-235.
Jeon GS, Park SH, Lee KJ, Lee MS, Chun BG, Shin KH (2006) Valproate prevents
MK801-induced changes in brain-derived neurotrophic factor mRNA in the rat brain.
Eur J Pharmacol 545: 142-146.
Jourdi H, Iwakura Y, Narisawa-Saito M, Ibaraki K, Xiong H, Watanabe M, Hayashi
Y, Takei N, Nawa H (2003) Brain-derived neurotrophic factor signal enhances and
maintains the expression of AMPA receptor-associated PDZ proteins in developing
cortical neurons. Dev Biol 263: 216-230.
Jourdi H, Hsu Y, Zhou M, Qin Q, Bi X, Baudry M, Positive AMPA receptor
modulation rapidly stimulates dendritidc mRNA translation through BDNF release.
Submitted.

Jung C, Rong Y, Doctrow S, Baudry M, Malfroy B, Xu Z (2001) Synthetic
superoxide dismutase/catalase mimetics reduce oxidative stress and prolong survival
in a mouse amyotrophic lateral sclerosis model. Neurosci Lett 304: 157-160.
Kalb R (2005) The protean actions of neurotrophins and their receptors on the life
and death of neurons. Trends Neurosci 28: 5-11.
Kawamura M, Nakajima W, Ishida A, Ohmura A, Miura S, Takada G (2005)
Calpain inhibitor MDL 28170 protects hypoxic-ischemic brain injury in neonatal rats
by inhibition of both apoptosis and necrosis. Brain Res 1037: 59-69.
Kehrer JP (1993) Free radicals as mediators of tissue injury and disease. Crit Rev
Toxicol 23: 21-48.
Kelly BL, Ferreira A (2006) Beta-amyloid-induced dynamin 1 degradation is
mediated by NMDA receptors in hippocampal neurons. J Biol Chem.
Kim MJ, Dunah AW, Wang YT, Sheng M (2005) Differential roles of NR2A- and
NR2B-containing NMDA receptors in Ras-ERK signaling and AMPA receptor
trafficking. Neuron 46: 745-760.
Kinuta Y, Kikuchi H, Ishikawa M, Kimura M, Itokawa Y (1989) Lipid peroxidation
in focal cerebral ischemia. J Neurosurg 71: 421-429.
Koh JY, Choi DW (1987) Quantitative determination of glutamate mediated cortical
neuronal injury in cell culture by lactate dehydrogenase efflux assay. J Neurosci
Methods 20: 83-90.
Koistinaho M, Koistinaho J (2005) Interactions between Alzheimer's disease and
cerebral ischemia--focus on inflammation. Brain Res Brain Res Rev 48: 240-250.
139
Kolarow R, Brigadski T, Lessmann V (2007) Postsynaptic secretion of BDNF and
NT-3 from hippocampal neurons depends on calcium calmodulin kinase II signaling
and proceeds via delayed fusion pore opening. J Neurosci 27: 10350-10364.
Komjati K, Mabley JG, Virag L, Southan GJ, Salzman AL, Szabo C (2004)
Poly(ADP-ribose) polymerase inhibition protect neurons and the white matter and
regulates the translocation of apoptosis-inducing factor in stroke. Int J Mol Med 13:
373-382.
Krapivinsky G, Krapivinsky L, Manasian Y, Ivanov A, Tyzio R, Pellegrino C, Ben
Ari Y, Clapham DE, Medina I (2003) The NMDA receptor is coupled to the ERK
pathway by a direct interaction between NR2B and RasGRF1. Neuron 40: 775-784.
Krysko DV, Vanden Berghe T, D'Herde K, Vandenabeele P (2008) Apoptosis and
necrosis: detection, discrimination and phagocytosis. Methods 44: 205-221.
Kumar SS, Bacci A, Kharazia V, Huguenard JR (2002) A developmental switch of
AMPA receptor subunits in neocortical pyramidal neurons. J Neurosci 22: 3005-
3015.
Kuroiwa T, Okeda R (1994) Neuropathology of cerebral ischemia and hypoxia:
recent advances in experimental studies on its pathogenesis. Pathol Int 44: 171-181.
Laake JH, Haug FM, Wieloch T, Ottersen OP (1999) A simple in vitro model of
ischemia based on hippocampal slice cultures and propidium iodide fluorescence.
Brain Res Brain Res Protoc 4: 173-184.
Lauterborn JC, Lynch G, Vanderklish P, Arai A, Gall CM (2000) Positive
modulation of AMPA receptors increases neurotrophin expression by hippocampal
and cortical neurons. J Neurosci 20: 8-21.
Lauterborn JC, Truong GS, Baudry M, Bi X, Lynch G, Gall CM (2003) Chronic
elevation of brain-derived neurotrophic factor by ampakines. J Pharmacol Exp Ther
307: 297-305.
Li P, Nijhawan D, Budihardjo I, Srinivasula SM, Ahmad M, Alnemri ES, Wang X
(1997) Cytochrome c and dATP-dependent formation of Apaf-1/caspase-9 complex
initiates an apoptotic protease cascade. Cell 91: 479-489.
Lin CH, Kuo SC, Huang LJ, Gean PW (2006) Neuroprotective effect of N-
acetylcysteine on neuronal apoptosis induced by a synthetic gingerdione compound:
involvement of ERK and p38 phosphorylation. J Neurosci Res 84: 1485-1494.
Lipton P (1999) Ischemic cell death in brain neurons. Physiol Rev 79: 1431-1568.
Liu R, Liu IY, Bi X, Thompson RF, Doctrow SR, Malfroy B, Baudry M (2003)
Reversal of age-related learning deficits and brain oxidative stress in mice with
superoxide dismutase/catalase mimetics. Proc Natl Acad Sci U S A 100: 8526-8531.
140
Liu TH, Beckman JS, Freeman BA, Hogan EL, Hsu CY (1989) Polyethylene glycol-
conjugated superoxide dismutase and catalase reduce ischemic brain injury. Am J
Physiol 256: H589-H593.
Liu XB, Murray KD, Jones EG (2004) Switching of NMDA receptor 2A and 2B
subunits at thalamic and cortical synapses during early postnatal development. J
Neurosci 24: 8885-8895.
Lu HC, Gonzalez E, Crair MC (2001) Barrel cortex critical period plasticity is
independent of changes in NMDA receptor subunit composition. Neuron 32: 619-
634.
Lu X, Rong Y, Baudry M (2000) Calpain-mediated degradation of PSD-95 in
developing and adult rat brain. Neurosci Lett 286: 149-153.
Lynch G (2002) Memory enhancement: the search for mechanism-based drugs. Nat
Neurosci 5 Suppl: 1035-1038.
Lynch G (2006) Glutamate-based therapeutic approaches: ampakines. Curr Opin
Pharmacol 6: 82-88.
Lynch G, Baudry M (1987) Brain spectrin, calpain and long-term changes in
synaptic efficacy. Brain Res Bull 18: 809-815.
Majno G, Joris I (1995) Apoptosis, oncosis, and necrosis. An overview of cell death.
Am J Pathol 146: 3-15.
Mansouri B, Henne WM, Oomman SK, Bliss R, Attridge J, Finckbone V, Zeitouni T,
Hoffman T, Bahr BA, Strahlendorf HK, Strahlendorf JC (2007) Involvement of
calpain in AMPA-induced toxicity to rat cerebellar Purkinje neurons. Eur J
Pharmacol 557: 106-114.
Marchesi E, Rota C, Fann YC, Chignell CF, Mason RP (1999) Photoreduction of the
fluorescent dye 2'-7'-dichlorofluorescein: a spin trapping and direct electron spin
resonance study with implications for oxidative stress measurements. Free Radic
Biol Med 26: 148-161.
Martinez A, Alcantara S, Borrell V, Del Rio JA, Blasi J, Otal R, Campos N, Boronat
A, Barbacid M, Silos-Santiago I, Soriano E (1998) TrkB and TrkC signaling are
required for maturation and synaptogenesis of hippocampal connections. J Neurosci
18: 7336-7350.
McGee AW, Dakoji SR, Olsen O, Bredt DS, Lim WA, Prehoda KE (2001) Structure
of the SH3-guanylate kinase module from PSD-95 suggests a mechanism for
regulated assembly of MAGUK scaffolding proteins. Mol Cell 8: 1291-1301.
141
McGowan JE, Chen L, Gao D, Trush M, Wei C (2006) Increased mitochondrial
reactive oxygen species production in newborn brain during hypoglycemia. Neurosci
Lett 399: 111-114.
Melov S, Doctrow SR, Schneider JA, Haberson J, Patel M, Coskun PE, Huffman K,
Wallace DC, Malfroy B (2001) Lifespan extension and rescue of spongiform
encephalopathy in superoxide dismutase 2 nullizygous mice treated with superoxide
dismutase-catalase mimetics. J Neurosci 21: 8348-8353.
Melov S, Ravenscroft J, Malik S, Gill MS, Walker DW, Clayton PE, Wallace DC,
Malfroy B, Doctrow SR, Lithgow GJ (2000) Extension of life-span with superoxide
dismutase/catalase mimetics. Science 289: 1567-1569.
Meyer-Franke A, Wilkinson GA, Kruttgen A, Hu M, Munro E, Hanson MG, Jr.,
Reichardt LF, Barres BA (1998) Depolarization and cAMP elevation rapidly recruit
TrkB to the plasma membrane of CNS neurons. Neuron 21: 681-693.
Moro MA, Almeida A, Bolanos JP, Lizasoain I (2005) Mitochondrial respiratory
chain and free radical generation in stroke. Free Radic Biol Med 39: 1291-1304.
Murphy AN, Fiskum G, Beal MF (1999) Mitochondria in neurodegeneration:
bioenergetic function in cell life and death. J Cereb Blood Flow Metab 19: 231-245.
Murray B, Alessandrini A, Cole AJ, Yee AG, Furshpan EJ (1998) Inhibition of the
p44/42 MAP kinase pathway protects hippocampal neurons in a cell-culture model
of seizure activity. Proc Natl Acad Sci U S A 95: 11975-11980.
Namura S, Iihara K, Takami S, Nagata I, Kikuchi H, Matsushita K, Moskowitz MA,
Bonventre JV, Alessandrini A (2001) Intravenous administration of MEK inhibitor
U0126 affords brain protection against forebrain ischemia and focal cerebral
ischemia. Proc Natl Acad Sci U S A 98: 11569-11574.
Nigam S, Schewe T (2000) Phospholipase A(2)s and lipid peroxidation. Biochim
Biophys Acta 1488: 167-181.
Noh HS, Hah YS, Nilufar R, Han J, Bong JH, Kang SS, Cho GJ, Choi WS (2006)
Acetoacetate protects neuronal cells from oxidative glutamate toxicity. J Neurosci
Res 83: 702-709.
Oliver CN, Starke-Reed PE, Stadtman ER, Liu GJ, Carney JM, Floyd RA (1990)
Oxidative damage to brain proteins, loss of glutamine synthetase activity, and
production of free radicals during ischemia/reperfusion-induced injury to gerbil brain.
Proc Natl Acad Sci U S A 87: 5144-5147.
Pan J, Zhang QG, Zhang GY (2005) The neuroprotective effects of K252a through
inhibiting MLK3/MKK7/JNK3 signaling pathway on ischemic brain injury in rat
hippocampal CA1 region. Neuroscience 131: 147-159.
142
Park EM, Joh TH, Volpe BT, Chu CK, Song G, Cho S (2004) A neuroprotective role
of extracellular signal-regulated kinase in N-acetyl-O-methyldopamine-treated
hippocampal neurons after exposure to in vitro and in vivo ischemia. Neuroscience
123: 147-154.
Patterson SL, Abel T, Deuel TA, Martin KC, Rose JC, Kandel ER (1996)
Recombinant BDNF rescues deficits in basal synaptic transmission and hippocampal
LTP in BDNF knockout mice. Neuron 16: 1137-1145.
Perez-Pinzon MA, Xu GP, Born J, Lorenzo J, Busto R, Rosenthal M, Sick TJ (1999)
Cytochrome C is released from mitochondria into the cytosol after cerebral anoxia or
ischemia. J Cereb Blood Flow Metab 19: 39-43.
Pfeiffer BE, Huber KM (2006) Current advances in local protein synthesis and
synaptic plasticity. J Neurosci 26: 7147-7150.
Piantadosi CA, Zhang J (1996) Mitochondrial generation of reactive oxygen species
after brain ischemia in the rat. Stroke 27: 327-331.
Pong K, Doctrow SR, Baudry M (2000) Prevention of 1-methyl-4-phenylpyridinium-
and 6-hydroxydopamine-induced nitration of tyrosine hydroxylase and neurotoxicity
by EUK-134, a superoxide dismutase and catalase mimetic, in cultured dopaminergic
neurons. Brain Res 881: 182-189.
Pong K, Doctrow SR, Huffman K, Adinolfi CA, Baudry M (2001) Attenuation of
staurosporine-induced apoptosis, oxidative stress, and mitochondrial dysfunction by
synthetic superoxide dismutase and catalase mimetics, in cultured cortical neurons.
Exp Neurol 171: 84-97.
Pong K, Rong Y, Doctrow SR, Baudry M (2002) Attenuation of zinc-induced
intracellular dysfunction and neurotoxicity by a synthetic superoxide
dismutase/catalase mimetic, in cultured cortical neurons. Brain Res 950: 218-230.
Porrino LJ, Daunais JB, Rogers GA, Hampson RE, Deadwyler SA (2005)
Facilitation of task performance and removal of the effects of sleep deprivation by an
ampakine (CX717) in nonhuman primates. PLoS Biol 3: e299.
Price RD, Milne SA, Sharkey J, Matsuoka N (2007) Advances in small molecules
promoting neurotrophic function. Pharmacol Ther 115: 292-306.
Proud CG (2007) Signalling to translation: how signal transduction pathways control
the protein synthetic machinery. Biochem J 403: 217-234.
Ray SK, Fidan M, Nowak MW, Wilford GG, Hogan EL, Banik NL (2000) Oxidative
stress and Ca2+ influx upregulate calpain and induce apoptosis in PC12 cells. Brain
Res 852: 326-334.
143
Rex CS, Lin CY, Kramar EA, Chen LY, Gall CM, Lynch G (2007) Brain-derived
neurotrophic factor promotes long-term potentiation-related cytoskeletal changes in
adult hippocampus. J Neurosci 27: 3017-3029.
Rieger D (1997) Batch analysis of the ATP content of bovine sperm, oocytes, and
early embryos using a scintillation counter to measure the chemiluminescence
produced by the luciferin-luciferase reaction. Anal Biochem 246: 67-70.
Riepe MW, Kasischke K, Raupach A (1997) Acetylsalicylic acid increases tolerance
against hypoxic and chemical hypoxia. Stroke 28: 2006-2011.
Saikumar P, Dong Z, Weinberg JM, Venkatachalam MA (1998) Mechanisms of cell
death in hypoxia/reoxygenation injury. Oncogene 17: 3341-3349.
Sans N, Petralia RS, Wang YX, Blahos J, Hell JW, Wenthold RJ (2000) A
developmental change in NMDA receptor-associated proteins at hippocampal
synapses. J Neurosci 20: 1260-1271.
Sattler R, Xiong Z, Lu WY, Hafner M, MacDonald JF, Tymianski M (1999) Specific
coupling of NMDA receptor activation to nitric oxide neurotoxicity by PSD-95
protein. Science 284: 1845-1848.
Savitz SI, Fisher M (2007) Future of neuroprotection for acute stroke: in the
aftermath of the SAINT trials. Ann Neurol 61: 396-402.
Sharma SS, Kaundal RK (2007) Neuroprotective effects of 6-hydroxy-2,5,7,8-
tetramethylchroman-2-carboxylic acid (Trolox), an antioxidant in middle cerebral
artery occlusion induced focal cerebral ischemia in rats. Neurol Res 29: 304-309.
Sheng M, Hoogenraad CC (2007) The postsynaptic architecture of excitatory
synapses: a more quantitative view. Annu Rev Biochem 76: 823-847.
Shimuta M, Yoshikawa M, Fukaya M, Watanabe M, Takeshima H, Manabe T (2001)
Postsynaptic modulation of AMPA receptor-mediated synaptic responses and LTP
by the type 3 ryanodine receptor. Mol Cell Neurosci 17: 921-930.
Staubli U, Perez Y, Xu FB, Rogers G, Ingvar M, Stone-Elander S, Lynch G (1994)
Centrally active modulators of glutamate receptors facilitate the induction of long-
term potentiation in vivo. Proc Natl Acad Sci U S A 91: 11158-11162.
Stoppini L, Buchs PA, Muller D (1991) A simple method for organotypic cultures of
nervous tissue. J Neurosci Methods 37: 173-182.
Strasser U, Fischer G (1995) Quantitative measurement of neuronal degeneration in
organotypic hippocampal cultures after combined oxygen/glucose deprivation. J
Neurosci Methods 57: 177-186.
144
Sugawara T, Fujimura M, Morita-Fujimura Y, Kawase M, Chan PH (1999)
Mitochondrial release of cytochrome c corresponds to the selective vulnerability of
hippocampal CA1 neurons in rats after transient global cerebral ischemia. J Neurosci
19: RC39.
Sugino T, Nozaki K, Takagi Y, Hattori I, Hashimoto N, Moriguchi T, Nishida E
(2000) Activation of mitogen-activated protein kinases after transient forebrain
ischemia in gerbil hippocampus. J Neurosci 20: 4506-4514.
Szabo C (1996) DNA strand breakage and activation of poly-ADP ribosyltransferase:
a cytotoxic pathway triggered by peroxynitrite. Free Radic Biol Med 21: 855-869.
Tabakman R, Jiang H, Shahar I, Arien-Zakay H, Levine RA, Lazarovici P (2005)
Neuroprotection by NGF in the PC12 in vitro OGD model: involvement of mitogen-
activated protein kinases and gene expression. Ann N Y Acad Sci 1053: 84-96.
Takei N, Inamura N, Kawamura M, Namba H, Hara K, Yonezawa K, Nawa H (2004)
Brain-derived neurotrophic factor induces mammalian target of rapamycin-
dependent local activation of translation machinery and protein synthesis in neuronal
dendrites. J Neurosci 24: 9760-9769.
Tegtmeier F, Weber C, Heister U, Haker I, Scheller D, Nikolov R, Holler M (1990)
Eicosanoids in rat brain during ischemia and reperfusion--correlation to DC
depolarization. J Cereb Blood Flow Metab 10: 358-364.
Tsokas P, Grace EA, Chan P, Ma T, Sealfon SC, Iyengar R, Landau EM, Blitzer RD
(2005) Local protein synthesis mediates a rapid increase in dendritic elongation
factor 1A after induction of late long-term potentiation. J Neurosci 25: 5833-5843.
Tyler WJ, Pozzo-Miller L (2003) Miniature synaptic transmission and BDNF
modulate dendritic spine growth and form in rat CA1 neurones. J Physiol 553: 497-
509.
Valko M, Izakovic M, Mazur M, Rhodes CJ, Telser J (2004) Role of oxygen radicals
in DNA damage and cancer incidence. Mol Cell Biochem 266: 37-56.
Valko M, Leibfritz D, Moncol J, Cronin MT, Mazur M, Telser J (2007) Free radicals
and antioxidants in normal physiological functions and human disease. Int J
Biochem Cell Biol 39: 44-84.
Van Loo G, Demol H, van Gurp M, Hoorelbeke B, Schotte P, Beyaert R,
Zhivotovsky B, Gevaert K, Declercq W, Vandekerckhove J, Vandenabeele P (2002)
A matrix-assisted laser desorption ionization post-source decay (MALDI-PSD)
analysis of proteins released from isolated liver mitochondria treated with
recombinant truncated Bid. Cell Death Differ 9: 301-308.

145
Wang X, Wang H, Xu L, Rozanski DJ, Sugawara T, Chan PH, Trzaskos JM,
Feuerstein GZ (2003) Significant neuroprotection against ischemic brain injury by
inhibition of the MEK1 protein kinase in mice: exploration of potential mechanism
associated with apoptosis. J Pharmacol Exp Ther 304: 172-178.
Waxman EA, Lynch DR (2005) N-methyl-D-aspartate receptor subtypes: multiple
roles in excitotoxicity and neurological disease. Neuroscientist 11: 37-49.
Wells DG, Fallon JR (2000) In search of molecular memory: experience-driven
protein synthesis. Cell Mol Life Sci 57: 1335-1339.
Wenzel A, Fritschy JM, Mohler H, Benke D (1997) NMDA receptor heterogeneity
during postnatal development of the rat brain: differential expression of the NR2A,
NR2B, and NR2C subunit proteins. J Neurochem 68: 469-478.
Williams GD, Towfighi J, Smith MB (1994) Cerebral energy metabolism during
hypoxia-ischemia correlates with brain damage: a 31P NMR study in unanesthetized
immature rats. Neurosci Lett 170: 31-34.
Won SJ, Kim DY, Gwag BJ (2002) Cellular and molecular pathways of ischemic
neuronal death. J Biochem Mol Biol 35: 67-86.
Wu X, Zhu D, Jiang X, Okagaki P, Mearow K, Zhu G, McCall S, Banaudha K,
Lipsky RH, Marini AM (2004) AMPA protects cultured neurons against glutamate
excitotoxicity through a phosphatidylinositol 3-kinase-dependent activation in
extracellular signal-regulated kinase to upregulate BDNF gene expression. J
Neurochem 90: 807-818.
Xiong ZG, Zhu XM, Chu XP, Minami M, Hey J, Wei WL, MacDonald JF, Wemmie
JA, Price MP, Welsh MJ, Simon RP Neuroprotection in ischemia: blocking calcium-
permeable acid-sensing ion channels. Cell 2004 Sep;17;118: 687-698.
Xu W, Wong TP, Chery N, Gaertner T, Wang YT, Baudry M (2007) Calpain-
mediated mGluR1alpha truncation: a key step in excitotoxicity. Neuron 53: 399-412.
Yamashima T, Tonchev AB, Tsukada T, Saido TC, Imajoh-Ohmi S, Momoi T,
Kominami E (2003) Sustained calpain activation associated with lysosomal rupture
executes necrosis of the postischemic CA1 neurons in primates. Hippocampus 13:
791-800.
Yu S, Liu M, Gu X, Ding F (2008) Neuroprotective Effects of Salidroside in the
PC12 Cell Model Exposed to Hypoglycemia and Serum Limitation. Cell Mol
Neurobiol.
Zhou M, Baudry M (2006) Developmental changes in NMDA neurotoxicity reflect
developmental changes in subunit composition of NMDA receptors. J Neurosci 26:
2956-2963.
146
Zhou M, Dominguez R, Baudry M (2007) Superoxide dismutase/catalase mimetics
but not MAP kinase inhibitors are neuroprotective against oxygen/glucose
deprivation-induced neuronal death in hippocampus. J Neurochem 103: 2212-2223.
Zhu D, Wu X, Strauss KI, Lipsky RH, Qureshi Z, Terhakopian A, Novelli A,
Banaudha K, Marini AM (2005) N-methyl-D-aspartate and TrkB receptors protect
neurons against glutamate excitotoxicity through an extracellular signal-regulated
kinase pathway. J Neurosci Res 80: 104-113.
Zipp F, Aktas O (2006) The brain as a target of inflammation: common pathways
link inflammatory and neurodegenerative diseases. Trends Neurosci 29: 518-527.
Zolotarjova N, Ho C, Mellgren RL, Askari A, Huang WH (1994) Different
sensitivities of native and oxidized forms of Na+/K(+)-ATPase to intracellular
proteinases. Biochim Biophys Acta 1192: 125-131.

147
Appendix: Publications

1. Zhou M, Baudry M. Developmental changes in NMDA neurotoxicity reflect
developmental changes in subunit composition of NMDA receptors. J Neurosci.
2006 Mar 15;26(11):2956-63

2. Xu W, Zhou MO, Baudry M, Neuroprotection by cell permeable TAT-mGluR1
peptide in ischemia: synergy between carrier and cargo sequences. Neuroscientist.
2007 Nov 13.

3. Zhou M, Dominguez R, Baudry M, Superoxide dismutase/catalase mimetics but
not MAP kinase inhibitors are neuroprotective against oxygen/glucose deprivation-
induced neuronal death in hippocampus. J Neurochem. 2007 Dec;103(6):2212-23.

Abstract (if available)

Abstract Oxygen/glucose deprivation (OGD) is widely used as an in vitro model of ischemia, and mechanisms underlying OGD-induced neuronal death are not completely understood. I tested the participation of reactive oxygen species (ROS) in OGD by using EUK-207, a synthetic superoxide dismutase/catalase mimetic. EUK-207 provides neuroprotection against OGD-induced cell death both in acute and cultured hippocampal slices. In cultured slices, this effect is related to decreased free radical accumulation, reduced release of apoptosis-inducing factor and reduced lipid peroxidation. In acute slices, protective effects of EUK-207 are also related to elimination of free radical accumulation and partial reversal of ATP depletion. Excitotoxicity induced by overactivation of NMDA receptors, a subtype of ionotropic glutamate receptors, is involved in OGD-induced cell death in slices from young but not adult rats. To better understand the mechanisms of excitotoxic cell death, I studied the effects of NMDA treatment on acute hippocampal slices from both neonatal and mature rats, and in particular, the role of calpain-mediated spectrin degradation. NMDA treatment results in cell death and spectrin degradation in slices from young but not adult rats, effects that are partly abolished by NMDA receptor subunit NR2B antagonists and by calpain inhibitor III, but not affected by a NR2A specific antagonist, suggesting that developmental changes in NMDA receptor subunit composition contribute to developmental changes in NMDA toxicity and possibly OGD-induced cell death.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

An orchestra of glutamate receptors and transporters in synaptic transmission, plasticity and excitotoxicity

PDF

Estrogen, progesterone and BDNF interactions: roles in neuroprotection

PDF

Estrogen and progesterone interaction on synaptic transmission and LTP in rodent hippocampus

PDF

Non-apoptotic involvement of caspases in astrogliosis

PDF

C-jun N-terminal Kinase (JNK) mediated inhibition of Pyruvate Dehydrogenase (PDH) activity and its effect on mitochondrial metabolism during brain aging

PDF

Role of neuronal nitric oxide synthase in aging and neurodegeneration

Asset Metadata

Creator Zhou, Miou (author)

Core Title Studies on plasticity and neurodegeneration in rat hippocampus

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Neuroscience

Publication Date 09/16/2008

Defense Date 08/07/2008

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

Tag ampakine,excitotoxicity,hippocampus,ischemia,neurodegeneration,OAI-PMH Harvest,plasticity

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Baudry, Michel (committee chair), Ko, Chien-Ping (committee member), Qin, Peter Z. (committee member), Thompson, Richard (committee member)

Creator Email miouzhou@usc.edu,zhoumiou@hotmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m1600

Unique identifier UC1129665

Identifier etd-Zhou-2374 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-94181 (legacy record id),usctheses-m1600 (legacy record id)

Legacy Identifier etd-Zhou-2374.pdf

Dmrecord 94181

Document Type Dissertation

Rights Zhou, Miou

Type texts

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

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu