University of Southern California Dissertations and Theses

Hypoxia-induced adaptive responses

(USC Thesis Other)

Hypoxia-induced adaptive responses

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

HYPOXIA-INDUCED ADAPTIVE RESPONSES

by

Jo-Lin Chen

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
(PHARMACOLOGY AND PHARMACEUTICAL SCIENCES)

August 2008

Copyright 2008 Jo-Lin Chen

ii
DEDICATION

To my mother and father,
my brother and sister-in-law
for their unconditional love and support.

iii
ACKNOWLEDGEMENTS

I would like to thank my advisor, Dr. David Ann, for his precious support and
guidance for these years of work and I benefit a lot in learning from him. I would like to
thank Dr. Helen Lin for her helpful discussion and suggestion about my project. I would
like to thank my Chairman of committee, Dr. Wei-Chiang Shen for his generosity in
taking care of my situation after Dr. Ann relocates to City of Hope Cancer Center and his
insightful suggestion of my project. I would like to thank my guidance committee
members: Dr. Sarah Hamm-Alvarez for her generous allowance to use the equipment in
her lab and helpful suggestion to my work, Dr. Kwang-Jin Kim for his cheerful
encouragement and invaluable discussion and comments into all my projects, and Dr.
James Ou in helping me throughout my graduate student life in USC and insightful
suggestion to my work. I also would like to thank Drs. Mary E. Reyland, Reen Wu, Yun
Qiu, and Anning Lin for providing me key materials in my work.
In addition, I wish to thank my laboratory colleagues, Dr. Carlos Clavijo, Dr. Ha-Van
Nguyen, Dr. Lee Boo, Dr. Hugo Lee, Dr. Xue-Fei Cao, Angela Li, Xu Li and Dr. Xiu-
Zhu Sun for their unconditional support and invaluable discussion.
Finally, I would like to deeply thank my parents, my family, and my friends for their
unconditional love, encouragement and support.

iv
TABLE OF CONTENTS

DEDICATION ii

ACKNOWLEDGEMENTS iii

LIST OF FIGURES vii

LIST OF ABBREVIATIONS x

ABSTRACT xiii

CHAPTER 1 INTRODUCTION 1
Rationale 1
Hypothesis 2
Background 2
• Hypoxia and HIF-1 α 2
• Desferroxamine (DFO), a hypoxia-mimetic agent, and its effects 5
• Salivary adaptive responses under hypoxic stress 5
• Structure and biological functions of PKCδ 8

CHAPTER 2 MATERIALS AND METHODS 12
• Cell lines and maintenance 12
• Chemicals and antibodies 13
• Production of lentiviral EGFP-SUMO-1 and EGFP-SUMO-1aa 13
in 293T cells
• Lentiviral transduction of Pa-4 cells 14
• Exposure of hypoxic condition (1% O
2
) 15
• Transient transfection and NF-κB transactivation assays 15
• Measurement of TER 16
• Preparation of SUMO-binding motif (SBM) peptide 17
• Whole cell lysate preparation and Western blot analyses 17
• Sample preparation and immunoprecipitation 18
• Cell viability measurement by MTT assays 19
• Cell cycle analyses by flow cytometry 19
• Confocal microscopy 20
• GFP-LC3 puncta analyses 20
• Statistical analyses 21

v
CHAPTER 3 SUMOYLATION ATTENUATES SENSITIVITY 22
TOWARDS HYPOXIC INJURY BY MODULATING
ADAPTIVE RESPONSES TO HYPOXIA IN SALIVARY
EPITHELIAL CELLS
Introduction 22
Results 25
• Exposure to either 1% O
2
or hypoxia-mimetic DFO stimulated global 25
SUMOylation
• SUMO-1 strengthens passive barrier properties against acute hypoxia 28
• SUMOylation facilitates the reassembly of F-Actin and ZO-1 after 31
prolonged hypoxic exposure
• SUMO-1 potentiates cell survival against acute hypoxia 33
• Effect of SUMOylation on DFO-induced ATM, PKCδ, and Caspase-3 36
activation
• Effect of SUMOylation on hypoxia-/DFO-induced NF-κB activation 40
Discussion 47

CHAPTER 4 PROTEIN KINASE Cδ-DEPENDENT AND 53
-INDEPENDENT SIGNALING IN GENOTOXIC
RESPONSE TO TREATMENT OF DESFERROXAMINE
(DFO), A HYPOXIA-MIMETIC AGENT
Introduction 53
Results 56
• PKCδ is proteolytically activated upon DFO-treatment 56
• DFO stimulates nuclear translocation of PKCδ 57
• Dysregulation of DFO-induced DNA damage response signaling by 61
sh-PKCδ or PKCδKD-EGFP
• Activation of pro-apoptotic Caspase-3 and pro-survival Akt in 63
response to DFO-treatment
• The involvement of PKCδ in DFO-induced cellular responses in 68
Pa-4 cells
Discussion 70

CHAPTER 5 EMERGING ROLE FOR THE PROTEIN KINASE Cδ 75
-DEPENDENT SIGNALING PATHWAY IN AUTOPHAGY
INDUCED BY DESFERROXAMINE, A HYPOXIA
-MIMETIC AGENT
Introduction 75
Results 77
• Increased autophagosome accumulation during DFO stress 77
• PKCδ is required for DFO-, but not EBSS-, induced autophagy 83
• JNK1 signaling pathway is required for PKCδ-mediated activation 87
of autophagy in DFO-treated cells

vi
• PKCδ stimulates the release of Beclin-1 from the complex of Beclin-1 92
and bcl-2 upon DFO-exposure
• Protective effect of autophagy on cell death induced by DFO 97
Discussion 100

CHAPTER 6 SUMMARY AND PERSPECTIVES 107
Summary 107
Perspectives 110

BIBLIOGRAPHY 111

vii
LIST OF FIGURES
Figure 1.1 The hypoxia-inducible factor 1 (HIF-1) pathway under 4
normoxia and hypoxia and its possible interfering agents.

Figure 1.2 Structure of PKCδ. 9

Figure 3.1 Establishment of SUMO-1-transduced Pa-4 cells. 26

Figure 3.2 Hypoxia or DFO induces global SUMOylation. 27
Figure 3.3 DFO induces global SUMOylation in primary salivary cells. 28
Figure 3.4 Overexpression of SUMO-1 protects TER over time. 30
Figure 3.5 SUMO-1 restores disrupted F-actin and ZO-1 assembly after 32
prolonged hypoxia.

Figure 3.6 SUMOylation antagonizes DFO-induced cell death. 34
Figure 3.7 SUMO-dependent protein-protein interaction is involved in 35
hypoxic responses.

Figure 3.8 Hypoxia renders S phase arrest. 36
Figure 3.9 Induction of ATM S1981 and H2AX S139 phosphorylation by 37
DFO treatment.

Figure 3.10 ATM activation proceeds Caspase-3 activation on DFO treatment. 39
Figure 3.11 ATM-deficient cells are relative to resistant to DFO treatment. 40
Figure 3.12 SUMO-1 attenuates NF-κB-dependent transcription by 41
cytoplasmic pathway.

Figure 3.13 SUMO-1 antagonizes MEKK1-mediated NF-κB activation by 42
stabilizing the steady-state level of IκB α.

Figure 3.14 SUMO-1 enhances the hypoxia-stimulated NF-κB-dependent 44
transcription.

Figure 3.15 SUMOylation enhances NF-κB DNA-binding activity. 45

Figure 3.16 SUMOylation retards DFO-induced decrease on steady-state 46
level of ZO-1.
viii

Figure 3.17 Hypoxia and DFO induce SUMOylation-regulated signalings to 46
modulate salivary adaptive responses.

Figure 4.1 DFO induces nuclear translocation of PKCδ. 59

Figure 4.2 PKCδ kinase function is required for its DFO-induced nuclear 60
translocation.

Figure 4.3 γ-H2AX accumulates in nuclei of Pa-4 cells in response to 62
DFO-treatment.

Figure 4.4 DFO-treatment induces p53 and Chk1 activation. 63

Figure 4.5 DFO induces activation of p53 and Chk1 in a Caspase-3 65
-independent manner.

Figure 4.6 Attenuation of DFO-induced Akt phosphorylations at Ser-473 66
by PKCδ.

Figure 4.7 PKCδKD downregulates hypoxia-induced PKCδ and Caspase-3 67
activation.

Figure 4.8 PKCδ is essential for DFO-mediated cytotoxicity. 69

Figure 4.9 Proposed model for DFO-induced activation of PKCδ, DNA 70
damage responses, and Akt to mediate salivary adaptive responses.

Figure 5.1 Treatment with DFO, a hypoxia-mimetic agent, induces 79
autophagic process.

Figure 5.2 DFO induces autophagosome accumulation in MEF/GFP-LC3 80
cells.

Figure 5.3 Validation of DFO-induced autophagic process. 82

Figure 5.4 Hypoxia (1% O
2
)-induced degradation of LC3-II and p62 is 83
inhibited by a combined treatment of lysosomal protease
inhibitors, E64d and pepstatin A.

Figure 5.5 PKCδ is essential for DFO-induced autophagy in Pa-4 cells. 84

Figure 5.6 PKCδ is essential for DFO-induced autophagy in MEF cells. 85

ix
Figure 5.7 DFO induces Y-64 and Y-155 phosphorylation of PKCδ. 86

Figure 5.8 PKCδ is required for DFO-induced JNK and c-Jun 88
phosphorylation in Pa-4 cells.

Figure 5.9 PKCδ is required for DFO-induced JNK and c-Jun 89
phosphorylation in MEF cells.

Figure 5.10 PKCδ is involved in UV-induced JNK activation in MEF cells. 89

Figure 5.11 Western analyses of JNK expression in MEF/WT, MEF/JNK1
-/-
, 90
and MEF/JNK2
-/-
cells.

Figure 5.12 JNK1, but not JNK2, is involved in DFO-induced autophagy. 91

Figure 5.13 PKCδ is required for DFO-induced dissociation of Beclin-1 93
from bcl-2.

Figure 5.14 DFO-induced JNK and bcl-2 phosphorylation is more 94
pronounced in MCF-7/Caspase-3 than in MCF-7/Neo cells.

Figure 5.15 DFO-treatment induces GFP-LC3 puncta accumulation in 96
MCF-7/Caspase-3, but not in MCF-7/Neo, cells.

Figure 5.16 Cytoprotective role of autophagy against DFO-induced cell death 97
in Pa-4 cells.

Figure 5.17 Lack of cytoprotection by PKCδ and JNK1 against long term 98
DFO-treatment.

Figure 5.18 Long-term DFO-treatment induces JNK activation. 99

Figure 5.19 Proposed signaling pathway underlying DFO-induced autophagic 100
process.

Figure 5.20 3-MA-treatment elicits no effect on JNK activation in both Pa-4 104
and MEF cells.

x
LIST OF ABBREVIATIONS

3-MA 3-methyladenine
Ab Antibody
ANOVA Analyses of variances
Atg Autophagy-related gene
ATM Ataxia-telangiectasia-mutated protein
BH3 Bcl-2 homology region-3
bHLH-PAS Basic-Helix-Loop-Helix-PAS
BNIP3 Bcl-2/adenovirus E1B 19kDa-interacting protein 3
CQ Chloroquine
CO-IP Co-immunoprecipitation
DFO Desferroxamine
DMEM Dulbecco’s modified Eagle medium
DSBs DNA double-stranded breaks
EBSS Earle’s balanced salt solution
EGFP Enhanced green fluorescent protein
ER Endoplasmic reticulum
FACS Fluorescence activated cell sorting
FBS Fetal bovine serum
GFP Green fluorescent protein
GFP-LC3 Green fluorescent protein-LC3
HIF-1α Hypoxia inducible factor 1-α
xi
HRE Hypoxia-responsive element
IGF Insulin-like growth factor
IKK IκB Kinase
IκB Inhibitor of NF-κB
iNOS Inducible nitric oxide synthase

IP Immunoprecipitation
JNK c-Jun NH
2
-terminal kinase
LC3 Microtubule-associated protein 1 light chain 3
mTOR Mammalian target of rapamycin
MEF Mouse embryonic fibroblasts
MOI Multiplicity of infection
MTT 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide
NEM N-ethylmaleimide
NEMO NF-κB essential modulator
NF-κB Nuclear factor-κB
NOS2 Nitric oxide synthase 2
NP-40 Nonidet P-40
PAI-1 Plasminogen activator inhibitor-1

PBS Phosphate buffered saline
PE Phosphatidylethanolamine
PHD Prolyl hydroxylase
PI3K Phosphoinositide 3-kinase
xii
PKCδ Protein kinase Cδ
RR Ribonucleotide reductase
SAPK/JNK Stress-activated protein kinase/c-Jun N-terminal kinase
SBM SUMO-binding motif
shRNA Short hairpin RNA
SDS Sodium dodecylsulfate
SDS-PAGE SDS-polyacrylamide gel electrophoresis
SUMO Small ubiquitin-like modifier
TER Transepithelial electrical resistance
TNF- α Tumor

necrosis factor- α
VEGF Vascular endothelial growth factor
VHL von Hippel-Lindau
WT Wild type

xiii
ABSTRACT
Hypoxia is a physiological condition defined as the oxygen concentration is lower
than 5% in general while normoxia is around 20%. It has been reported that hypoxia is
related to numerous

human diseases, such as respiratory diseases, vascular diseases,
neurodegeneration and cancer, and hypoxic stress activates various signal transduction
pathways to regulate cellular adaptive responses. Desferroxamine (DFO), a hypoxia-
mimetic agent, functions as an iron chelator and has been utilized in hypoxic study.
However, the molecular mechanisms underlying hypoxia- and/or DFO-induced adaptive
responses are still unclear. In this report, I demonstrate that hypoxia and DFO induce a
transient global SUMOylation to augment cell survival against acute hypoxic stress by
strengthening passive barrier properties, facilitating the reassembly of F-actin/ZO-1,
attenuating pro-apoptotic PKCδ/Caspase-3 activation, inducing S1981 phosphorylation of
ATM, and activating pro-survival NF-κB signaling pathways. Under prolonged hypoxic
treatment, DFO induces nuclear translocation and proteolytic cleavage of PKCδ and
PKCδ-dependent Caspase-3 activation to render a sustained DFO-elicited γ-H2AX
activation leading to apoptotic cell death. Intriguingly, PKCδ plays a fine-tuning role in
modulating DFO-induced Akt phosphorylation. Moreover, DFO-exposure also induces a
PKCδ-independent signaling and both PKCδ-dependent and -independent pathways
functionally cooperate to integrate pro-apoptotic/Caspase-3, pro-survival/Akt, and DNA
damage-induced DNA repair/cell cycle regulation signalings. An accumulation of
autophagosomes was observed during 2 to 4 h post-hypoxic treatment. Autophagy is a
tightly orchestrated intracellular process and is essential for cell survival or death in
xiv
response to stress conditions. DFO-treatment renders a rapid and transient
phosphorylation at Y-64 and Y-155 of PKC δ and conveys JNK1 Thr183/Tyr185-
phosphorylation. Inhibition of PKC δ by PKC δKD or PKC δ-knockout reduces DFO-
induced changes in LC3-II levels. The requirement of PKCδ is apparent for DFO-, but
not starvation-, induced autophagy. Both JNK1 activation and release of Beclin-1, a key
molecule in autophagic process, from inhibitory bcl-2 are PKC δ-dependent.
Significantly, inhibition of autophagy by 3-MA or Atg5-knockout presents a more
prevalence in cell death while PKCδ- or JNK1-deficient cells exhibit resistance to long-
term DFO-treatment. In summary, acute hypoxia/DFO stress induces a transient
activation of SUMOylation and autophagy to protect cells against hypoxic injury, while
prolonged treatment induces PKC δ proteolytic activation leading to apoptotic cell death.

1
CHAPTER 1
INTRODUCTION
Rationale
Hypoxia generally occurs in normal tissue during development, for example, embryos
are in the hypoxic environment before a blood circulatory system is established, and also
occurs in disease states, such as tumor proliferation. Numerous studies have suggested
that hypoxia and hypoxia-mimetic agent, desferroxamine (DFO), could be a novel
therapeutic approach for treating diseases, such as myocardial and cerebral ischemia and
cancer (13, 17, 29, 39, 133). Therefore, understanding the molecular mechanisms and
physiological outcomes underlying hypoxic condition is important to further develop
treatments to cure diseases.
It has been reported that hypoxic injuries effectively disrupt the integrity of epithelial
barrier and increase the movement of ions, large solutes,

and inflammatory cells across
epithelial tight junction structures (54). To achieve the goal of study in hypoxic stress-
induced adaptive responses, the rodent salivary glands have provided an unique and
useful model to investigate salivary physiology and pathophysiology, such as
fluid/protein secretion, cell proliferation/differentiation, signaling network, and cell-
lineage tracing (129, 171). In this report, salivary gland epithelial cells, Pa-4, have been
utilized as a model system to dissect the molecular mechanisms governing adaptive
responses to hypoxic stress, regulating tissue/cell-specific gene expression and their
biological functional changes.

2
Hypothesis
Since hypoxia causes different levels of cellular injuries depending on the dosage and
time period of hypoxic condition to activate complicated gene regulation in adapt to
stress-induced responses, it is important to understand the molecular mechanisms
underlying hypoxic stress. The present work is designed to study the hypoxia/DFO-
induced post-translational modification of proteins and investigate the gene regulation to
modulate adaptive responses. In this dissertation, the central hypotheses are that: 1) acute
hypoxia/DFO-induced overexpression of SUMO-1 protects salivary epithelial cells
against hypoxic injury by modulating protein activation and physiological adaptation in
response to hypoxia; 2) prolonged DFO-treatment-elicited genotoxic stress activates both
PKCδ-dependent and -independent signaling pathways to regulate salivary adaptive
responses; and 3) acute DFO-exposure induces transient autophagy to protect cells
against hypoxic stress through the activation of PKCδ/JNK1 pathway.

Background
Hypoxia and HIF-1 α
Hypoxia is a biological condition which is defined as the oxygen concentration is
lower than 5% in general while normoxia is around 20%. Hypoxia takes place in
physiologically normal development. For example, mammalian embryos are under a
period of time in an almost entirely hypoxic environment before blood vessels are
established. Also hypoxia occurs in the pathophysiological condition during tumor
formation and metastasis. Hypoxia is toxic to both normal and cancer cells; however,
tumor cells develop molecular adaptation to survive in a harsh condition (60, 122, 137,
3
141). More and more genes have been found to be regulated by hypoxia, such as hypoxia
inducible factor-1 α (HIF-1 α). It has been demonstrated that hypoxia-induced genes, such
as HIF-1 α, p53 and NOS, are involved in many cellular functions, including angiogenesis,
glucose metabolism, erythropoiesis, growth factors signaling, nitric oxide production and
reactive oxygen species, transcription factors, stress responses, immune responses,
apoptosis, pH regulation, tumor invasion and metastasis and lead to cell survival or cell
death in response to hypoxic stress (16, 19, 28, 43, 60, 64, 78, 98, 107, 125, 135, 154).
HIF-1 is a heterodimeric transcription factor acting as an oxygen sensor. The N-
terminal of HIF-1 is basic-helix-loop-helix-PAS (bHLH-PAS) structure and these two
domains are required for dimerization and DNA binding. On the other side, the C-
terminal of HIF-1 is required for transactivation. HIF-1 is consisted by α and α subunits.
HIF-1 α is sensitive to oxygen concentration and regulated by growth factors whereas
HIF-1 β is constitutively expressed. As shown in Figure 1.1, under normal condition
(normoxia), HIF-1α is hydroxylated by prolyl hydroxylase enzymes (PHDs) and this
modification is required to be recognized and bound by von Hippel-Lindau (VHL)
protein for subsequent ubiquitination and degradation in the proteasome (122). On the
other hand, under hypoxic condition, HIF-1α can not be hydroxylated by PHDs because
of lack-of-oxygen and therefore HIF-1 α protein remains stable. Stabilized HIF-1 α
translocates into nucleus and interacts with constitutively active HIF-1β and other
coactivators, CBP/p300, to bind to hypoxia-responsive element (HRE) therefore activate
transcriptional activities of target genes (16, 87, 108, 122, 138-140, 174). There are
many genes are transcriptionally regulated by HIF-1, including vascular endothelial
4
growth factor (VEGF), erythropoietin, insulin-like growth factor (IGF), heme oxygenase-
1, p21, nitric oxide synthase 2 (NOS2) (60).

Figure 1.1 The hypoxia-inducible factor 1 (HIF-1) pathway under normoxia and hypoxia and its
possible interfering agents. Under normoxia HIF-1 α is subject to oxygen-dependent hydroxylation on
specific proline residues in the oxygen dependent degradation domain (ODDD) and following
ubiquitination and proteasomal degradation that is mediated by the von Hippel-Lindau protein. This
mechanism is furthermore triggered through posttranslational HIF-1 α acetylation of lysine residues in the
ODDD and hydroxylation of asparagine residue within the C-TAD. These events do not occur under
hypoxia and so HIF-1 α is stabilized, translocates to the nucleus, interacts with hypoxia responsive elements
(HRE), and promotes the activation of target genes. Abbreviations: PI3K, phosphatidylinositol 3-kinase;
AKT, protein kinase B; mTOR, mammalian target of rapamycin; HIF-1 α/ β, hypoxia inducible factor α/ β;
PHD, prolyl hydroxylase; FIH-1, factor inhibiting HIF-1; ARD-1, N-acetyltransferase; ODDD, oxygen
dependent degradation domain; C-TAD, C-terminal-transactivation domain; pVHL, protein Von Hippel-
Lindau; MAPK, mitogen-activated protein kinase; HRE, hypoxia-response element; VEGF, vascular
endothelial growth factor; ET-1, endothelin-1; Glut-1, glucose transporter 1.(122)

5
Desferroxamine (DFO), a hypoxia-mimetic agent, and its effects
Desferroxamine (DFO) is an iron chelator which functions to deplete intracellular Fe,
an essential molecule for PHDs-mediated HIF-1 hydroxylation. DFO-exposure has been
reported to induce cell cycle arrest and apoptosis and it has shown anti-proliferative
activity in leukemia and neuroblastoma cells in vivo and in vitro (39, 85, 132). DFO is
also known to inhibit the activity of ribonucleotide reductase (RR) which is responsible
to convert ribonucleotides into deoxyribonucleotides for de novo DNA synthesis (151,
152). Iron deprivation has been demonstrated to regulate many critical cellular genes
expression. For example, HIF-1α is stabilized by DFO and subsequently transactivates
VEGF-1 expression level which is related to angiogenesis (7, 18, 100). Therefore, DFO
is also considered function as a potential anti-cancer agent by causing hypoxic condition
and decreasing cell viability since tumor growth requires lots of nutrients and oxygen.
Many reports suggest that DFO causes cell cycle arrest at G
1
/S phases and results in cell
death with high dose treatment (85, 116). My preliminary data have shown that
prolonged DFO-treatment induces cell death in salivary gland epithelial Pa-4 cells by
MTT assay. However, the precise molecular mechanisms are not fully understood.

Salivary adaptive responses under hypoxic stress
• Hypoxia induces HIF-1 α transcriptional activity.
The mammalian cells response to hypoxia is complex and varies at different oxygen
tension (55). In general, cells may adapt to hypoxia in numerous ways, including a
transition from oxidative phosphorylation to glycolysis and neovascularization. Many of
these metabolic responses are mediated by the transcription factor HIF-1 (27, 60).
6
Hypoxia induces HIF-1 α transcriptional activity to regulate downstream gene expression,
such as VEGF, resulting in the increase of cell growth. Our previous studies have
showed that VEGF, PAI-1, and iNOS are activated by Etk and promote cell survival and
proliferation in response to hypoxic stress (20, 21).

• Hypoxia enhances the expression level of SUMO-1.
It has been reported that hypoxic stress induces various post-translational
modification, such as phosphorylation and SUMOylation, to regulate gene expression and
stability (11, 78, 87, 108, 135). Hypoxia up-regulates the steady-state level of SUMO-1
by as much as 100-fold suggesting the potential important role of SUMOylation in
governing cellular hypoxic responses (26, 142). Bossis and Melchior et al. have reported
that oxidative stress regulates SUMOylation process (11). It supports my hypothesis that
alteration of SUMOylation process is a means for cells to respond to stress condition.
Recently, Cheng J. et al. report that HIF-1 α is SUMOylated by SUMO-1 and
deSUMOylated by SENP1 under hypoxic condition. Hypoxia-induced HIF-1 α
SUMOylation enhances HIF-1 α binding to VHL and sequentially degraded in a
proteasome-dependent manner in the nucleus (24). Therefore, SUMOylation serves as
ubiquitin-like proteasomal degradation and regulates protein stability in response to
hypoxia.

• Hypoxia induces apoptotic cell death.
Importantly, cells respond to hypoxia by diminishing their proliferative rates;
however, the underlying mechanism still remains elusive. Results from both invasive and
7
noninvasive studies of a variety of normal tissues and tumors suggest that under a less
stringent hypoxic condition, cells are viable, but non-proliferating, while cells accumulate
DNA damage and undergo cell death in a more stringent or prolonged hypoxic condition
(4). It has been observed that prolonged hypoxia renders a decrease in TER by increasing
the depolymerization of Actin and results in cell growth inhibition in salivary Pa-4 cells
(54). A numerous studies have shown that chronic hypoxia increase apoptotic cell death.
Therefore, I propose to characterize the molecular mechanisms underlying the
antagonistic interplay between hypoxia-induced pro-survival and anti-survival signals in
salivary cells.

• Hypoxia induces autophagy.
Autophagy is a metabolic process to degrade cytoplasmic proteins or organelles for
the maintenance of cellular homeostasis. It has been shown that chronic hypoxia (1% O
2
)
induces autophagosome accumulation, a key marker of autophagic process, through the
induction of HIF-1α and Bcl-2/adenovirus E1B 19kDa-interacting protein 3 (BNIP3; pro-
cell death Bcl-2 family member) and leads to autophagic cell death (2, 106). Intriguingly,
from my preliminary data, it shown that acute hypoxia (1% O
2
or DFO-treatment) also
induces autophagy in a transient manner and results in protecting salivary cells against
hypoxic stress. Therefore, I am interesting in investigating the detail molecular
mechanisms underlying acute hypoxia-induced autophagy.

8
Structure and biological functions of PKCδ
• Structure of PKCδ.
Protein kinase Cδ (PKCδ), a Serine/Threonine kinase, is activated by diverse stimuli
and plays a crucial role in regulating cell growth, proliferation, differentiation, and
apoptosis in cell type-specific manner (89, 115, 149). According to the differences of
regulatory domains, PKCδ can be activated by diacylglycerols (DAGs) but in a calcium-
independent manner. As shown in Figure 1.2, the highly conserved C-terminal catalytic
domain of PKCδ consists of ATP-binding C3 and protein-substrate-binding C4
subdomains, there are three conserved phosphorylation sites, T505, S643, and S662,
locating in the activation loop, turn motif and hydrophobic motif, respectively (115, 149).
In the N-terminal regulatory domain, it contains an autoinhibitory pseudosubstrate
domain between 2 subdomains, C1A and C1B, which are responsible to bind to
DAG/PMA, and one C2-like domain, lack of Asp residues for Ca
2+
binding. Therefore,
PKCδ is maximally activated by DAG/PMA without the requirement of Ca
2+
(115, 149).
9

Figure 1.2 Structure of PKCδ. (A) Structural domains of PKCδ implicated in functional regulation and (B)
sequences of the twin C1A and C1B domains. (149)

PKCδ is activated by a variety of mechanisms including membrane translocation (72,
161), proteolytically cleaved by Caspase-3 (35, 77), protein-protein interactions (8), and
tyrosine phosphorylation (33, 77). Many studies have been shown that activated PKCδ
translocates to different subcellular compartments, such as nucleus and mitochondria, in
response to cellular stimuli by distinctive mechanisms (14, 37, 105). Although PKCδ has
both pro-apoptotic and anti-apoptotic functions, most of studies focus on that PKCδ is
involved in the induction of apoptosis (14).
10
The classic activation model of PKCδ is activated by DAG/PMA-stimulated
membrane translocalization and phosphorylated in the C-terminal Ser/Thr residues of
PKCδ (130, 149). Caspase-3-mediated cleavage of PKCδ also plays an important role in
response to certain types of genotoxic stresses, such as etoposide, and lead to apoptotic
cell death (167). In addition, PKCδ phosphorylated on tyrosine residues has been
considered as a regulatory mechanism since this Y-phosphorylation is stimuli-dependent
and cell type-specific (130, 149). It has been reported that inactive PKCδ stays in a
“primed”, closure conformation status, once PKCδ is induced phosphorylation at
different tyrosine residues by stimuli, the conformation will change to expose catalytic
domain for further Ser/Thr phosphorylation and Caspase-3-mediated cleavage (115, 149).
This Y-phosphorylation is not conserved thought all the mammals and it is a transient
event depending on the stimuli and microenvironments (115, 130, 149, 167). Recently, it
has been demonstrated that Y-phosphorylation regulates nuclear translocation of PKCδ
suggesting that phosphorylation/dephosphorylation of PKCδ in the regulatory domain
serves as a switch to determine cell fate toward cell survival or cell death (68).

• Biological functions of PKCδ.
PKCδ has been implicated to regulate cell cycle and programmed cell death in many
cell types. Unlike other PKCs, PKCδ generally inhibits proliferation, induces cell cycle
arrest, and/or enhances the differentiation of various undifferentiated cell lines (115, 130,
167). To date, it has been identified many substrates of PKCδ, such as p53, STATs, p300,
DNA-PK, depending on cell type and extracellular stimuli (46, 70, 167, 168, 173). PKCδ
11
has been emerged as a regulator to modulate cellular function in both pro-survival and
pro-apoptotic manner. However, most of studies have shown that PKCδ is a pro-
apoptotic molecule. The function of PKCδ is also regulated by its location in the cells.
For example, PMA or oxidative stress induces PKCδ to localize to mitochondria for
mediating apoptotic signaling resulting in the loss of mitochondrial membrane potential,
release of cytochrome c and activation of Caspase-3 (67, 110). In this dissertation, I
propose to study the role of PKCδ in response to hypoxia- and/or DFO-induced adaptive
responses.

12
CHAPTER 2
MATERIALS AND METHODS
Cell lines and maintenance
The rat parotid epithelial cell lines, Pa-4, and derived lentiviral transduced cell lines
including Pa-4/PKCδWT-EGFP, Pa-4/PKCδKD-EGFP, Pa-4/shPKCδ, Pa-4/EGFP, Pa-
4/EGFP-SUMO-1 and Pa-4/EGFP-SUMO-1aa were cultured in Dulbecco’s modified
Eagle’s/F-12 (DME/F-12, 1:1 mixture) medium supplemented with 2.4% fetal bovine
serum (FBS), insulin (5 μg/ml), L-glutamine (2.3 mM), transferrin (5 μg/ml), epidermal
growth factor (25 ng/ml), hydrocortisone (1.1 μM), glutamate (5 mM), T
3
(3,3',5-triiodo-
L-thyronine) (1.7 nM), kanamycin (94 μg/ml) and fungizone (47 μg/ml) and maintained
at 35ºC in a humidified atmosphere of 5% CO
2
and 95% air. HEK 293 and 293T cells
were grown in Dulbecco’s modified Eagle medium (DMEM) supplemented with 10%
FBS plus 1% penicillin/streptomycin and cultured at 37ºC. HeLa cells were grown in
Dulbecco’s Modification of Eagle’s Medium (4.5 g/L glucose) supplemented with 10%
FBS, 1 mM sodium pyruvate, and 1% penicillin/streptomycin and maintained at 37ºC.
YZ5 (ATM
+/+
) and pEBS7 (ATM
-/-
) cell lines were cultured in DMEM plus 10% FBS,
1% penicillin/streptomycin and 100 μg/ml hygromycin and grown at 37ºC. MCF-7/Neo
and MCF-7/Caspase-3 cells were maintained in DMEM containing 10% FBS, 0.5 mg/ml
geneticin (G418; Sigma) and 1% antibiotics and grown at 37ºC. MEF/WT, MEF/JNK1
-/-
,
MEF/JNK2
-/-
, MEF/Atg5
-/-
, MEF/GFP-LC3 and MEF/PKCδ
-/-
cells were grown in
DMEM supplemented with 10% FBS plus 1% antibiotics and maintained at 37ºC.

13
Chemicals and antibodies
Desferroxamine (DFO), N-ethylmaleimide (NEM), Na
3
VO
4
, TNF- α, propidium
iodide, 3-methylalanine (3-MA), chloroquine (CQ), E64d, pepstatin A, EBSS, and
rapamycin were purchased from Sigma. Complete Protease Inhibitor Cocktail and
PhosSTOP Phosphatase Inhibitor Cocktail were purchased from Roche.
Anti-PKCδ, anti-GFP, anti-ZO-1, anti-cyclin A, anti-IκB α, anti-ATM, anti-H2AX,
anti-p53, anti-bcl-2, anti-tubulin antibodies were purchased from Santa Cruz
Biotechnologies. Antibodies against Caspase-3, phospho-Ser15-p53, Akt, phospho-
Ser473-Akt, phospho-Thr308-Akt, Chk1, phospho-Ser345-Chk1, phospho-Ser70-bcl-2,
JNK, phospho-Thr183/Tyr185-JNK, c-Jun, phospho-Ser63-c-Jun were purchased from
Cell Signaling Technology. Anti-phospho-Ser1981-ATM, anti-phospho-Ser139-H2AX,
and anti-phosphotyrosine (4G10) were purchased from Upstate Biotechnology. An anti-
GMP-1 (SUMO-1) was purchased from Zymed Laboratories. An anti-LC3 antibody was
purchased from MBL International Co. An anti-p62 antibody was purchased from
American Research Products. Antibodies against Beclin-1 were purchased from BD
Biosciences and Abcam. Anti-actin antibody was purchased from Chemicon
International.

Production of lentiviral EGFP-SUMO-1 and EGFP-SUMO-1aa in 293T cells
The EGFP-SUMO-1 fragment obtained from digestion of pEGFP-C1-SUMO-1 was
cloned into the PinA I and Hinc II sites of the pRRLsin.hCMV156 vector to construct
pRRLsin.hCMV-EGFP-SUMO-1 (Figure 3.1). HEK 293T cells (70% confluence) in 175
mm flasks were co-transfected with 21 μg pRRLsin.hCMV-EGFP (a human
14
immunodeficiency virus (HIV)-based self-inactivating (SIN) replication-defective
lentivirus transfer vector expressing either an enhanced green fluorescent protein
(EGFP)), -EGFP-SUMO-1, or -EGFP-SUMO-1aa (unconjugatable SUMO-1), 14 μg of
pΔ8.7 (for viral packaging), and 7 μg pVSV-G (for VSV-G pseudotyping) by calcium
phosphate precipitation. Chloroquine was added to a final concentration of 25 μM, and
cells were incubated in a 5% CO
2
incubator at 37ºC for 16 h. Chloroquine-containing
medium was replaced with culture medium containing 10 mM sodium butyrate, and cells
were incubated for at least another 8 h prior to the addition of fresh culture medium and
then incubated for an additional 16 h. Viral supernatant was collected, centrifuged at
2500 rpm for 10 min, and stored immediately at 4ºC. Supernatants were pooled and
concentrated using a Macrosep Centrifuge Device with a 300-kDa molecular mass cut-off
(Millipore) and a 0.45 μm syringe filter. Aliquots of concentrated virus were stored at -
80ºC. Titers of concentrated (2-4x10
8
transducing units (TU)/ml) lentiviral stocks
(pRRLsin.hCMV-EGFP and -EGFP-SUMO-1) were determined by infecting HEK 293T
cells in the presence of 6 μg/ml polybrene with serial dilutions based on results of
fluorescence-activated cell sorting (FACS).

Lentiviral transduction of Pa-4 cells
1x10
5
Pa-4 cells were plated to reach 50% confluence prior to transducing
(multiplicity of infection (MOI) of 0.1 to 40) with lentivirus encoding SUMO-1 in the
presence of 6 μg/ml polybrene to establish the optimal transduction efficiency. Twenty-
four hours after transduction, the polybrene-containing medium was replaced with fresh
15
culture medium and incubated for an additional 16 h before analyses. Seventy-two hours
after transduction, cells were washed and trypsinized for further propagation. An aliquot
of these cells was analyzed by FACS for determination of transduction efficiency. For
FACS analyses, trypsinized cells were re-suspended in 0.5 ml phosphate buffered saline
(PBS) (4x10
5
cells/ml). For generation of Pa-4/SUMO-1aa (unconjugatable form of
SUMO-1) and Pa-4/EGFP, Pa-4 cells were transduced as described above at an MOI of
10.

Exposure of hypoxic condition (1% O
2
)
Parental Pa-4 and transduced cells were seeded onto tissue culture-treated, 12 mm
polycarbonate filters (Costar-Corning), cultured for 2 days at 35ºC and 5% CO
2
in air
before being transferred to an exposure chamber, which was flushed with 1% O
2

balanced with 5% CO
2
and 94% N
2
. The chamber was then sealed airtight and kept at
35ºC for the duration of hypoxia treatment. For NF-κB reporter assays, transiently
transfected cells were exposed to 1% O
2
for 24 h as described below.

Transient transfection and NF-κB transactivation assays
Pa-4 and Pa-4/EGFP-SUMO-1 cells were transiently transfected with 0.5 μg of
pGL2-NF-κB luciferase reporter construct harboring an IL-6 promoter and two NF-κB
binding sites (a generous gift of Dr. Yun Yen, City of Hope National Medical Center,
CA) using Lipofectamine 2000 (Invitrogen) by following manufacture’s instruction.
One-tenth μg of the Renilla luciferase pRL-TK plasmid was co-transfected for
16
normalization of transfection efficiency. Six-hours after the start of transfection, cells
were

recovered overnight in 0.05% stripped-serum medium and subsequently exposed to
normoxic or hypoxic conditions (1% O
2
- or DFO-treatment) for 24 h prior to luciferase
assays. Relative luciferase activity from the firefly luciferase reporter gene was
determined and normalized to Renilla luciferase activity using the Dual Luciferase
Reporter Assay System (Promega). Fold-induction by hypoxia was calculated after
normalizing the pGL2-NF-κB reporter activities with the activities from co-transfected
pRL-TK. The nuclear protein preparation and TransBinding NF-κB Assays (Panomics)
were performed according to manufacturer’s instruction.

Measurement of TER
Bioelectrical parameters of epithelial cell monolayers were monitored using a
MilliCell ERS screening device (Millipore, Bedford, MA). Spontaneous potential
difference (SPD; expressed in mV (apical side as a reference)) and transepithelial
electrical resistance (TER; expressed in kΩ·cm
2
) were measured with a set of chopstick-
style electrodes. Offset potential differences generated by the voltage-sensing electrodes
and background electrical resistance contributed by both the bathing medium and the
filter membrane were measured at the beginning and end of each set of bioelectric
measurements using blank filters and corrected for. SPD and TER were measured up to 6
days. Data are presented as mean ± S.D. from at least 3 separate experiments performed
in triplicate.

17
Preparation of SUMO-binding motif (SBM) peptide
The peptide that noncovalently binds to SUMO (PIAS
X
) as well as a control peptide
with scrambled amino acid sequence with identical composition (146) was fused with
HIV-1 TAT nuclear localization signal. Peptides were synthesized by the Peptide
Synthesis Core Facility at the City of Hope National Medical Center, purified by HPLC,
and verified by mass spectrometry. Pa-4/SUMO-1 cells were pre-incubated with SBM
(10 μM) for 24 h followed by treatment with varying concentrations of DFO in the
presence or absence of 10 μM SBM and incubated for 72 h before determining cell
viability by MTT assays.

Whole cell lysate preparation and Western blot analyses
Cells were washed by 1xPBS twice before harvesting cells. For SUMOylation assays,
25 mM N-ethylmaleimide was included in the SDS lysis buffer (62.5 mM Tris, pH 6.8,
1% SDS, 10% glycerol, 1% β-mercaptoethanol). For protein phosphorylation studies,
whole cell lysates were extracted by SDS lysis buffer containing both the Complete
Protease Inhibitor Cocktail (Roche) and phosphatase inhibitor Na
3
VO
4
(2 mM; Sigma).
Cell lysates were collected by centrifugation and protein concentrations were determined
using the Bradford protein assay. Twenty to 50 μg of protein lysates were subjected to
SDS-PAGE analyses, followed by immunoblotting with respective antibodies as
indicated. To detect endogenous LC3-I and LC3-II, whole cell lysates were prepared
using a Triton X-100-based lysis buffer (20 mM Tris-HCl, pH 7.4, 150 mM NaCl, 1%
Triton X-100) plus the Complete Protease Inhibitor Cocktail (Roche). Equal amounts of
proteins were aliquoted and added with SDS lysis buffer right before boiling, followed by
18
SDS-PAGE and immunoblotting with an anti-LC3 antibody. Anti-tubulin and anti-actin
antibodies were used for quantifying the proteins loaded into each lane for internal
controls. Immunoblots were imaged with an enhanced chemiluminescence detection kit
(ECL-Plus, Amersham Pharmacia Biotech) and the VersaDoc 5000 Imaging System
(Bio-Rad). Densitometric data were captured, quantified with Quantity One Software
(Bio-Rad) and normalized with internal control proteins, individually, in each experiment.
The relative level of a particular protein altered by various treatments was then calculated
by setting the normalized value in the control as 1, assuming equal variances.

Sample preparation and immunoprecipitation
Whole cell lysates were prepared using RIPA buffer (25 mM Tris, 125 mM NaCl, 1%
Nonidet P-40 (NP-40), 0.1% SDS, 0.5% sodium deoxycholate, 0.004% sodium azide, pH
8.0) containing both the Complete Protease Inhibitor Cocktail and PhosSTOP
Phosphatase Inhibitor Cocktail (Roche). Protein concentrations were determined by
BCA assay (Bio-Rad). Whole cell lysates (1 mg) were incubated with 1-5 μg/ml of a
specific antibody as indicated at 4ºC for 2 h and incubated further with 20 μl Protein A/G
PLUS-Agarose beads (Santa Cruz Biotechnologies) at 4ºC overnight to capture immune
complexes. Immune complexes were then washed three times with PBS containing 1%
NP-40, 0.5% sodium deoxycholate and the Complete Protease Inhibitor Cocktail,
dissolved by adding SDS lysis buffer, boiled for 6 min, resolved by SDS-PAGE, and
blotted with antibodies. The levels for proteins of interest were determined using an
ECL-Plus kit. The relative level of immunoprecipitated complex was normalized with
19
total input proteins in lysates individually and normalized by setting the non-treatment
sample value as 1.

Cell viability measurement by MTT assays
Cells were seeded into 24-well plates to obtain a confluence of 35% on the day of the
experiment. Cells were then treated with different concentrations of DFO and medium
changed daily for 2 days. Twenty-four to 48 h after the start of DFO-treatment
(depending on cell types), 0.2 ml of 0.1 mg/ml 3-[4,5-dimethylthiazol-2-yl]-2,5-
diphenyltetrazolium bromide (MTT) (Sigma, St. Louis, MO) in OptiMEM I (Invitrogen)
was added to each well and the plate was incubated at 37ºC for an additional 1.5 h. The
MTT solution was then aspirated and 0.2 ml isopropanol was added to each well to
dissolve the formazan crystals. Absorbance was read immediately at 540 nm in a
scanning multi-well spectrophotometer. The results were depicted as percentage of cell
viability and reported as the mean ± S.D. of three independent experiments performed in
triplicates.

Cell cycle analyses by flow cytometry
Cells were seeded at 50-80% confluence in 35 mm dishes and synchronized by serum
starvation overnight. After treatment with DFO for different time points as indicated,
cells were fixed in 70% ethanol overnight and stained with propidium iodide (PI)/ Triton
X-100/ RNase A in PBS for 30 min at room temperature. Flow cytometry was performed
by using FACSCaliber (Becton Dickinson) in Norris Cancer Center Flow Cytometry
Core Facility (University of Southern California, CA). Statistical analyses was
20
performed by two-way analyses of variances (ANOVA) with randomized sample blocks
and post-hoc tests using Fisher’s least squared difference method of protected t-tests. p
< 0.05 was considered statistically significant.

Confocal microscopy
For hypoxic exposure, cells were incubated in medium with reduced serum (0.05%
FBS) for 24 h before exposure to 1% O
2
for time-course studies. At appropriate time
points, cells maintained under normoxic and hypoxic conditions or treated with DFO
were washed with PBS, fixed with 4% paraformaldehyde in PBS for 30 min at room
temperature and quenched with 50 mM NH
4
Cl in PBS. Fixed cells were then
permeabilized with 0.5% Triton X-100 in PBS for 15 min, blocked with 1% BSA in PBS,
and incubated with rhodamine-phalloidin (Chemicon International; 1:100) for 1 h for
detection of F-actin and an anti-ZO-1 antibody (Zymed Laboratories; 1:200) to detect
ZO-1, respectively. Processed cells were mounted with Prolong Gold Antifade reagent
(Molecular Probe) and examined by Zeiss Inverted LSM 510 Meta 2 Photon Microscope.

GFP-LC3 puncta analyses
Pa-4, MCF-7/Caspase-3 or MCF-7/Neo cells were transient transfected with GFP-
LC3 or GFP-N1 expression constructs by using Lipofectamine 2000 (Invitrogen) per
manufacturer’s instructions. For GFP-LC3 puncta analyses, cells at 36 h post-
transfection or stably transfected MEF/GFP-LC3 cells were seeded on coverslips placed
in the 6-well plates and cultured overnight, followed by various treatments for different
time periods. Cells were then fixed with 4% paraformaldehyde in phosphate-buffered
21
saline (PBS) for 30 min, incubated with DAPI (4'-6-Diamidine-2-phenyl indole;
Molecular Probes) for 10 min and mounted with Prolong Gold Antifade reagent
(Molecular Probes). Images were acquired with Zeiss Inverted LSM 510 Meta 2 Photon
Microscope and analyzed using Zeiss LSM Image Examiner. The GFP-LC3 puncta and
GFP-LC3-positive cells were examined and quantified in more than five fields per slide
on five slides. The increase in GFP-LC3 puncta represents the relative accumulation of
autophagosomes.

Statistical analyses
Experiments were independently carried out in duplicate at least three times, unless
stated otherwise. One representative data set from these three independent experiments is
presented where appropriate. Reporter activity shown is the mean ± S.D. based on at
least three independent transfection experiments. Error bar represents the S.D. of the
mean. The results were evaluated for statistical significance by two-way analyses of
variances (ANOVA) with randomized sample blocks, followed by post-hoc comparisons
based on Fisher's least squared difference method of protected t-tests. p values less than
0.05 were regarded as significant.

22
CHAPTER 3
SUMOYLATION ATTENUATES SENSITIVITY TOWARDS HYPOXIC INJURY
BY MODULATING ADAPTIVE RESPONSES TO HYPOXIA
IN SALIVARY EPITHELIAL CELLS
Introduction
Small ubiquitin-like modifier (SUMO) is the best characterized

member of a growing
family of ubiquitin-related proteins. SUMO

is conjugated to target proteins using an
enzyme conjugation

system similar to but distinct from that of ubiquitin (38, 71).
Importantly,

SUMO-1, -2, -3, and -4 have emerged as important posttranslational

modifiers that regulate diverse cellular functions including

intracellular targeting, DNA
repair, cell cycle progression,

and responses to extracellular stimuli (23, 38, 71). A large
numbers of

proteins are modified by SUMO, and recent proteomic studies

have shown
that as many as 400 yeast proteins are modified by

yeast SUMO homolog and 2683
potential SUMO substrates are conserved

in both humans and mice (119, 165, 176). The
precise functional differences

of various SUMO paralogs remain to be established.
However,

given the importance of SUMOylation, it is not surprising that

SUMO plays
important roles in the development of various diseases (10, 94, 148).

Although the
significance of SUMOylation in modulating cellular

adaptive responses is well
established, it is still not clear

how SUMO modification regulates specific key cellular
functions.

Accumulating evidence suggests the potential importance of SUMOylation

in
governing cellular hypoxic responses in that hypoxia up-regulates

the steady-state level of
SUMO-1 by as much as 100-fold (26, 142). Hypoxia is a (patho)physiological condition
23
that arises when

cellular oxygen demand exceeds supply. Regions of hypoxia occur

not
only in disease states but also during normal development.

For example, mammalian
embryos are, for significant periods

of time, in an almost entirely hypoxic environment
before a

blood circulatory system is established (22). Hypoxia is also

a physiological
inducer of the p53 tumor suppressor and provides

selective pressure during tumor growth
for the elimination of

cells with wild-type p53 and the clonal expansion of cells with

mutated p53 (82). It has been established that hypoxic cells acquire

genetic and adaptive
changes to survive and proliferate in a

hypoxic microenvironment, enabling eventual
evasion from hypoxia-induced

cell death (113).
As an activator of pro-inflammatory and anti-apoptotic genes,

the transcription factor
nuclear factor-κB (NF-κB) is a key factor

in determining whether cells survive after
being subjected to

genotoxic stress. Hypoxia-elicited phenotypic manifestations

have
been reported to be a consequence of the induction of tumor necrosis factor- α (TNF- α), a
known activator of NF-κB signaling (102, 150). NF-κB transcriptional activity is
stimulated by hypoxia (26, 102, 150). Hypoxia also appears to be a key factor involved
in the development of genetic instability (9). Recent studies have suggested that

hypoxia
elicits increased DNA damage, enhanced mutagenesis,

and functional impairment in
DNA repair pathways (reviewed in 58, 60). Because SUMOylation has been
demonstrated to

be involved in governing DNA repair and genomic stability (38), we
postulated that SUMO-1 functions as a central player in the

generalized hypoxic response.
SUMOylation of Inhibitor of NF-κB

(IκB) has been previously demonstrated to act in an
anti-NF-κB

fashion by attenuating the activation of NF-κB by various cytokines (36). In
24
contrast, reports from a recent study by Huang et al., (65) suggest that both SUMOylation
and ataxia-telangiectasia-mutated protein

(ATM) activation enhance genotoxic stress-
mediated NF-κB activation.

Hence, SUMO-1 can function in both anti-NF-κB and pro-
NF-κB manners,

depending on the individual stimuli and specific pathway used

for NF-
κB activation. However, questions remain regarding the

exact role of SUMOylation in
modulating NF-κB transactivation

in response to stress by hypoxia (1% O
2
) or hypoxia-
mimetic

desferroxamine (DFO).

In this project, I demonstrate that treatment with 1% O
2
or

DFO induces global
SUMOylation, disrupts epithelial barrier

function and ZO-1 assembly, is cytotoxic, and
activates genotoxic

signaling cascade. I next examined the effect of the augmented

SUMOylation process on modulating the stress response in salivary

epithelial Pa-4 cells
by investigating the following biological

events. Is enhanced SUMOylation capacity
beneficial for cell

survival under DFO treatment? Will SUMOylation elicit an inhibitory

or stimulatory effect on 1% O
2
- or DFO-dependent NF-κB activation?

Does
SUMOylation protect cells against hypoxia-mediated decreases

in transepithelial
electrical resistance (TER) and alter tight-junctional

protein distribution? To achieve this
goal, I stably overexpressed

SUMO-1 in rat salivary epithelial Pa-4 cells by lentivirus-
mediated

transduction to examine the effect of SUMOylation on governing

epithelial
homeostatic control mechanisms against exposure to

1% O
2
or DFO. Together, I
conclude that SUMOylation attenuates

activation of pro-apoptotic protein kinase Cδ
(PKCδ) and Caspase-3 while

promoting genotoxicity-induced NF-κB transactivation and
25
facilitating

TER restoration and assembly of tight junction-associated proteins

in response
to exposure to reduced oxygen tension or DFO.

Results
Exposure to either 1% O
2
or hypoxia-mimetic DFO stimulated global SUMOylation
Because the exact role of hypoxia-induced SUMO-1 overexpression

in modulating
hypoxic response is still unclear, I sought to

develop a cell model system to examine the
effect of augmented

SUMO-1 expression in governing various hypoxia-elicited cellular

responses, such as cell survival, NF-κB activation, and barrier

integrity. Because there is
a very limited (almost none) choice

of pharmacological reagents that can be used to
modulate SUMOylation

capacity in cells, we used an EGFP-SUMO-1 (Figure 3.1)
lentivirus-mediated

transduction system to enhance SUMOylation. First, by using

FACS
analyses to monitor EGFP expression, a dose-dependent transduction

was established in
Pa-4 cells by increasing MOIs from 0.1 to

40 to yield 21 to 99% EGFP-SUMO-1-positive
cells (data not shown). To assess the stability of transgene expression, transduced

cells
were passaged, stored, and revived for FACS analyses,

which showed that Pa-4/SUMO-1
cells remained > 99% of EGFP-SUMO-1

positive after repeated passage at least eight
times (data not

shown). Western analyses were performed to confirm that the

transduced
EGFP-SUMO-1 was functionally conjugated to the target

proteins by using whole-cell
lysates prepared from Pa-4/SUMO-1

cells (data not shown).

26
RRL/LTR
Ψ
RRE cPPT EGFP CMV SIN LTR PRE
Ψ
RRE cPPT CMV EGFP
SIN LTR PRE
RRE
RRL/LTR
SUMO-1
RRL/LTR
Ψ
RRE cPPT EGFP CMV SIN LTR PRE
Ψ
RRE cPPT CMV EGFP
SIN LTR PRE
RRE
RRL/LTR
SUMO-1

Figure 3.1 Establishment of SUMO-1-transduced Pa-4 cells. Diagram of pRRLsin.hCMV-EGFP-
SUMO-1 and parental constructs.

Consistent with previous reports in

other cell types (26, 142), 1% O
2
treatment
stimulated overall SUMOylation

profile by about 1.5-fold in both Pa-4 and Pa-4/SUMO-
1 cells

at 6 h post-treatment (Figure 3.2A). I next examined whether

iron chelator DFO is
able to mimic low O
2
tension to induce

SUMOylation process by treating both Pa-4 and
Pa-4/SUMO-1 cells

with 50 and 100 µM DFO, respectively. As shown in

Figure 3.2B,
Pa-4 cells treated with 100 µM DFO displayed

an enhanced SUMOylation profile similar
to that detected in

Figure 3.2A, whereas both 50 and 100 µM DFO treatment

elicited a
more sustained increase of overall SUMOylation pattern

in Pa-4/SUMO-1 cells.
Moreover, I also observed an increasing SUMOylation trend in salivary gland primary
cells with 200 μM DFO-treatment (Figure 3.3).

27

12 3 4 5 6
1% O
2
Pa-4 Pa-4/SUMO-1
0 6 30 0 6 30 h
SUMOylated proteins
Tubulin
Western: anti-SUMO-1
12 3 4 5 6
1% O
2
Pa-4 Pa-4/SUMO-1
0 6 30 0 6 30 h
SUMOylated proteins
Tubulin
Western: anti-SUMO-1
0
0.5
1
1.5
2
1% O
2
Pa-4 Pa-4/SUMO-1
0 6 30 0 6 30 h
Relative Induction of SUMOylation
0
0.5
1
1.5
2
1% O
2
Pa-4 Pa-4/SUMO-1
0 6 30 0 6 30 h
Relative Induction of SUMOylation
A
B
7 8910 11 12 12 3 4 5 6
DFO
SUMOylated proteins
Pa-4 Pa-4/SUMO-1
06 30 0 6 30 h
Tubulin
Pa-4 Pa-4/SUMO-1
06 30 0 6 30
50 μM 100 μM
7 8910 11 12 12 3 4 5 6
DFO
SUMOylated proteins
Pa-4 Pa-4/SUMO-1
06 30 0 6 30 h
Tubulin
Pa-4 Pa-4/SUMO-1
06 30 0 6 30
50 μM 100 μM
0
0.5
1
1.5
2
Pa-4 Pa-4/SUMO-1
06 30 0 6 30 h 50 μM DFO
Relative Induction of SUMOylation
0
0.5
1
1.5
2
Pa-4 Pa-4/SUMO-1
06 30 0 6 30 h 50 μM DFO
Pa-4 Pa-4/SUMO-1
06 30 0 6 30 h 50 μM DFO
Relative Induction of SUMOylation
0
0.5
1
1.5
2
2.5
h
Pa-4 Pa-4/SUMO-1
06300 630 100 μM DFO
Relative Induction of SUMOylation
0
0.5
1
1.5
2
2.5
h
Pa-4 Pa-4/SUMO-1
06300 630 100 μM DFO
Pa-4 Pa-4/SUMO-1
06300 630 100 μM DFO
Relative Induction of SUMOylation

Figure 3.2 Hypoxia or DFO induces global SUMOylation. Equal amounts of whole-cell lysates were
subjected to Western analyses with an anti-GMP-1 (SUMO-1) antibody. Representative Western analysis
data demonstrate overall increases in SUMOylation process after treatment with 1% O
2
(A) or DFO (50 or
100 µM) (B) for different time periods, as indicated. A filled arrow indicates the SUMO-modified
RanGAP1, and an unfilled arrow depicts the unmodified RanGAP1. After normalizing with that of tubulin,
the respective steady-state level of SUMOylated proteins in Pa-4 and Pa-4/SUMO-1 cells before treatment
was designated as 1. The relative induction of SUMOylation by treatment with 1% O
2
(A) or DFO (50 or
100 µM) (B) is shown.

28

Figure 3.3 DFO induces global SUMOylation in primary salivary cells. (A) Morphology of dissected
rat salivary gland (left panel) and primary culture cells grown on tissue culture dish (right panel). (B) DFO
elicits an overall increase of SUMOylation pattern in primary salivary cells. Primary salivary cells were
treated with 200 μM DFO for different time periods as indicated. Equal amounts of whole cell lysates were
subjected to Western blotting analyses with anti-GMP-1 (SUMO-1), anti-PKCδ and anti-tubulin antibodies.
Relative induction of SUMOylation level was quantified by Quantity One software (Bio-Rad) and
normalized with that of tubulin while untreated cell lysate was designated as 1.

SUMO-1 strengthens passive barrier properties against acute hypoxia
After verifying that treatment of either 1% O
2
or DFO increases

global SUMOylation
in both Pa-4 and Pa-4/SUMO-1 cells (Figure

3.2), I next investigated whether augmented
A
B
Primary salivary cells grown
on Tissue culture dish
(morphologically epithelium-like)
Dissected Salivary Gland
200 μM DFO 02 6 24
Primary Salivary Cells
Tubulin
PKCδ
SUMOylated
proteins
12 3 4
0
0.5
1
1.5
2
200 μM DFO
02 6 24 (h)
Primary Salivary Cells
Relative Induction of SUMOylation
A
B
Primary salivary cells grown
on Tissue culture dish
(morphologically epithelium-like)
Dissected Salivary Gland Primary salivary cells grown
on Tissue culture dish
(morphologically epithelium-like)
Dissected Salivary Gland
200 μM DFO 02 6 24
Primary Salivary Cells
Tubulin
PKCδ
SUMOylated
proteins
12 3 4
0
0.5
1
1.5
2
200 μM DFO
02 6 24 (h)
Primary Salivary Cells
Relative Induction of SUMOylation
200 μM DFO 02 6 24
Primary Salivary Cells
Tubulin
PKCδ
SUMOylated
proteins
12 3 4
200 μM DFO 02 6 24
Primary Salivary Cells
Tubulin
PKCδ
SUMOylated
proteins
12 3 4
0
0.5
1
1.5
2
200 μM DFO
02 6 24 (h)
Primary Salivary Cells
Relative Induction of SUMOylation
0
0.5
1
1.5
2
200 μM DFO
02 6 24 (h)
Primary Salivary Cells
0
0.5
1
1.5
2
200 μM DFO
02 6 24 (h)
Primary Salivary Cells
Relative Induction of SUMOylation
29
SUMOylation

process is (patho)physiologically relevant. Previously, it has been

demonstrated that hypoxic treatment causes a weakening of epithelial

barrier properties in
Pa-4 cells (54). To search for functional

consequence of increased SUMOylation, TER
was measured in Pa-4

and Pa-4/SUMO-1 cell monolayers under both normoxic and
hypoxic

conditions. When TER changes over time in normoxic Pa-4 and

Pa-4/SUMO-1
cells were measured, a notably higher baseline TER

was observed in Pa-4/SUMO-1 cells
(Figure 3.4A).
Based on my previous experience that TER increases at day 2

after seeding in our
experimental system (54), studies on TER

time courses under parallel normoxic or
hypoxic treatments were

performed on cells starting from 48 to 96 h after seeding

to
investigate the effect of SUMOylation on TER maintenance. Under normoxic conditions,
TER of Pa-4/SUMO-1 monolayers at

96 h displayed a more robust increase (up to almost
five-fold)

than that (~two-fold) observed in Pa-4 monolayers (Figure 3.4B),

indicating
that augmented SUMOylation strengthens passive barrier

properties of Pa-4 cells under
baseline conditions. By contrast,

there was only a transient TER increase detected in Pa-
4/SUMO-1

but not Pa-4 cells during the first 9-hour interval of hypoxia

(48 to 57 h post-
hypoxia; Figure 3.4C) when both Pa-4 and

Pa-4/SUMO-1 monolayers were subjected to
hypoxic 1% O
2
treatment. However, TERs of both types of monolayers eventually
decreased

to less than 65% of the initial value at 24 h post-hypoxia

(i.e. at 72 h in Figure
3.4C). Together, these data indicate

that enhanced SUMOylation in Pa-4/SUMO-1 cells
is able to transiently

blunt hypoxia-induced deterioration of passive permeability

barrier
properties up to 9 h post-hypoxia, and the efficacy

of SUMO-1 overexpression in
30
attenuating hypoxia-induced TER

reduction was not sustained at 24 h after 1% O
2

exposure

as indicated at 72 h in Figure 3.4C.

B Normoxia C Hypoxia
48 54 60 66 72 78 84 90 96
0
100
200
300
400
500
600
Time (h)
Pa-4
Pa-4/SUMO-1
TER (% of the value at T= 48 h)
*
*
*
*
*
*
‡
48 54 60 66 72 78 84 90 96
0
100
200
300
400
500
600
Time (h)
Pa-4
Pa-4/SUMO-1
TER (% of the value at T= 48 h)
*
*
*
*
*
*
‡
0
100
200
300
400
Time (h)
Pa-4
Pa-4/SUMO-1
48 54 60 66 72 78 84 90 96
**
*
*
*
*
TER (% of the value at T= 48 h)
0
100
200
300
400
Time (h)
Pa-4
Pa-4/SUMO-1
48 54 60 66 72 78 84 90 96
**
*
*
*
*
TER (% of the value at T= 48 h)
A
TER (kΩ·cm
2
)
0
0.5
1
2.5
1.5
2
Pa-4
Pa-4/SUMO-1
*
B Normoxia C Hypoxia C Hypoxia
48 54 60 66 72 78 84 90 96
0
100
200
300
400
500
600
Time (h)
Pa-4
Pa-4/SUMO-1
TER (% of the value at T= 48 h)
*
*
*
*
*
*
‡
48 54 60 66 72 78 84 90 96
0
100
200
300
400
500
600
Time (h)
Pa-4
Pa-4/SUMO-1
TER (% of the value at T= 48 h)
*
*
*
*
*
*
‡
0
100
200
300
400
Time (h)
Pa-4
Pa-4/SUMO-1
48 54 60 66 72 78 84 90 96
**
*
*
*
*
TER (% of the value at T= 48 h)
0
100
200
300
400
Time (h)
Pa-4
Pa-4/SUMO-1
48 54 60 66 72 78 84 90 96
**
*
*
*
*
TER (% of the value at T= 48 h)
A
TER (kΩ·cm
2
)
0
0.5
1
2.5
1.5
2
Pa-4
Pa-4/SUMO-1
*

Figure 3.4 Overexpression of SUMO-1 protects TER over time. (A) Higher TER in Pa-4/SUMO-1 cells
than that in Pa-4 cells. (B) Effect of SUMO-1 on TER under normoxic condition. TER was measured in Pa-
4 and Pa-4/SUMO-1 cells that were grown on Transwell filters at indicated time points. Time courses of
TER for Pa-4 and Pa-4/SUMO-1 cells between 48 and 96 h after seeding under normoxic condition were as
shown. (C) Overexpression of SUMO-1 protects TER from being compromised by hypoxia. Cell culture
and TER measurements were performed as described in (B). Hypoxic exposure was initiated at t = 48 h.
Data are presented as mean ± SD from three independent experiments, each performed in triplicate. * and ‡
denote significant difference in TER percentages between Pa-4 and Pa-4/SUMO-1 cells at corresponding
time points (p < 0.001 and p < 0.01, respectively).

During 24 to 48 h of treatment with 1% O
2
(indicated as

72 to 96 h in Figure 3.4C),
we noted an increase in TER in

Pa-4/SUMO-1 monolayers, recovering beyond the pre-
hypoxia level

(Figure 3.4C). As shown in Figure 3.4C, the TER reached its highest

value
(25% greater than its pre-hypoxia TER at 48 h [100%])

at 33 h post-hypoxia onset (i.e. at
81 h in Figure 3.4C)

in Pa-4 cells, and the TER in Pa-4/SUMO-1 cells also exceeded

by
150% (compared with that at 48 h) at 33 h post-hypoxia

onset and sustained for another
15 h. The highest TER value

(at 48 h post-hypoxia) in Pa-4/SUMO-1 cells was 200%
greater

than that at the onset of hypoxia (Figure 3.4C). However, these

values were still
less than the control (Figure 3.4B; normoxia) counterparts

at each time point. Although
31
the exact effects of SUMOylation

on the restoration of TER after prolonged exposure to
hypoxia

in Pa-4/SUMO-1 cells are difficult to interpret at this point,

these observations
were reproducible. Collectively, these data

indicate that overexpression of SUMO-1 in
Pa-4/SUMO-1 cells

plays a transient and partially protective role against injury

to passive
barrier properties exerted by early stage of hypoxic

stress and renders a higher TER value
after prolonged hypoxic

exposure.

SUMOylation facilitates the reassembly of F-Actin and ZO-1 after prolonged
hypoxic exposure
Higher TER value is a hallmark of polarized monolayers with

a stabilized tight
junction complex, which involves a network

of occludin and claudins linked to actin via
ZO proteins (155, 156). Thus, a decrease in TER may reflect changes in epithelial
integrity

and junction formation. In this regard, I next examined whether

the chronic
hypoxia-induced decrease in TER was a consequence

of altered F-actin and ZO-1
assembly in both Pa-4 and Pa-4/SUMO-1

cell monolayers under normoxia and hypoxia.
Pa-4 and Pa-4/SUMO-1

monolayers were exposed to 1% O
2
as described for TER
measurements

and processed for F-actin and ZO-1 visualization by confocal

microscopy.
Exposure of confluent Pa-4 (Figure 3.5, left panels)

and Pa-4/SUMO-1 (Figure 3.5, right
panels) monolayers to hypoxia

for 24 h resulted in irregular, discontinuous F-actin
patterns

at the cellular border and punctuate, gap-like appearances at

points of cellular
contacts (Figure 3.5, panel d), as it has previously

reported (54). Similar discontinuous
appearances were also observed

for ZO-1 after 24 h of hypoxia exposure in both Pa-4 and

Pa-4/SUMO-1 cells (Figure 3.5, panel e) compared with a smooth pattern

of ZO-1 in
32
cells under normoxic conditions. However, the appearance

of discontinuity of F-actin
(Figure 3.5, panel g) and ZO-1 (Figure 3.5, panel h) was less pronounced at 30 h of
hypoxia. Notably, at 30 h

post-hypoxic treatment, there were few irregular structures at

cellular borders and gap-like appearances at cellular contacts

detectable for both F-actin
and ZO-1 in Pa-4/SUMO-1 cells, whereas

gap-like structures at cellular borders were still
apparent

in Pa-4 monolayers (Figure 3.5, panels g and h). I concluded that chronic

(or
prolonged) hypoxia-induced TER decrease could be a consequence

of altered F-actin and
ZO-1 appearance and assembly pattern

of tight junctions, correlating reasonably well
with the decrease

of TER at 24 h and recovery at 30 h post-hypoxia onset

(Figure 3.4C).
Importantly, SUMOylation facilitates the restoration

of F-actin and ZO-1 assembly after
prolonged hypoxic exposure

(30 h) in Pa-4/SUMO-1 cells.

Pa-4 Pa-4/SUMO-1
c
f
i
c
f
c
f
i
c
f
Merge
c
f
i
c
f
i
Merge
Normoxia Hypoxia (24 h) Hypoxia (30 h)
a
d
g
a
d
→
a
d
g
a
d
F-Actin
b
e
h
b
e
b
e
h
b
e
ZO-1
g
a
d
→
g
a
d
a
d
F-Actin
h
e
b
→
→
h
e
b
e
b
ZO-1
Pa-4 Pa-4/SUMO-1
c
f
i
c
f
c
f
i
c
f
Merge
c
f
i
c
f
c
f
i
c
f
c
f
i
c
f
c
f
i
c
f
Merge
c
f
i
c
f
i
Merge
c
f
i
c
f
i
Merge
Normoxia Hypoxia (24 h) Hypoxia (30 h)
a
d
g
a
d
→
a
d
g
a
d
F-Actin
a
d
g
a
d
→
a
d
g
a
d
F-Actin
b
e
h
b
e
b
e
h
b
e
ZO-1
b
e
h
b
e
b
e
h
b
e
ZO-1
g
a
d
→
g
a
d
a
d
F-Actin
g
a
d
→
a
d
→
g
a
d
a
d
F-Actin
h
e
b
→
→
h
e
b
e
b
ZO-1
h
e
b
→
→
e
b
→
→
h
e
b
e
b
ZO-1

Figure 3.5 SUMO-1 restores disrupted F-actin and ZO-1 assembly after prolonged hypoxia. The
effects of hypoxia on the localization of F-actin (red) and ZO-1 (green) in Pa-4 and Pa-4/SUMO-1 cell
monolayers were assessed by confocal microscopy at indicated time points. Arrows indicate gap-like
(zagged) structures.
33
SUMO-1 potentiates cell survival against acute hypoxia
Because hypoxia-mimetic DFO has been shown to induce global

SUMOylation
(Figure 3.2B) and is known to cause cell growth arrest

and cell death (55), I treated
parental Pa-4, Pa-4/EGFP, Pa-4/SUMO-1,

and Pa-4/SUMO-1aa (unconjugatable form of
SUMO-1) cells with

increasing concentrations of DFO. As demonstrated in Figure

3.6,
DFO rendered a dose-dependent growth inhibition in all

four cell types. Notably, Pa-
4/SUMO-1 cells exhibited a more

pronounced resistance to cytotoxicity elicited by 50
µM

DFO than the other three cell types examined (Figure 3.6, lane

8). In addition, Pa-
4/SUMO-1aa cells were somewhat more sensitive

to DFO (50 µM) treatment than
parental Pa-4 cells

or Pa-4/EGFP cells (Figure 3.6, lane 11 versus lanes 2 and 5),

suggesting that the protective effect observed in Pa-4/SUMO-1

cells is specific. Because
there were no noticeable differences

between Pa-4 and Pa-4/EGFP cells (Figure 3.6, lanes
2 and 3

versus lanes 5 and 6) and both served as a control, only Pa-4

cells were used in
the following studies.

34

0
25
50
75
100
% Cell Viability
125
9 1 2 3 4 56 78 10 11 12
0 μM DFO
50 μM DFO
100 μM DFO
∗
†
‡
∗
†
‡
Pa-4 WT EGFP SUMO-1 SUMO-1aa
0
25
50
75
100
% Cell Viability
125
9 1 2 3 4 56 78 10 11 12
0 μM DFO
50 μM DFO
100 μM DFO
∗
†
‡
∗
†
‡
Pa-4 WT EGFP SUMO-1 SUMO-1aa

Figure 3.6 SUMOylation antagonizes DFO-induced cell death. Pa-4, Pa-4/EGFP, Pa-4/SUMO-1, and
Pa-4/SUMO-1aa cells were treated with increasing concentrations of DFO, as indicated. Cell survival was
measured by MTT assays at 24 h post-treatment. ∗ denotes significant difference (p < 0.001) between lanes
2 and 8. ‡ denotes significant difference (p < 0.01) between lanes 3 and 9. † denotes significant difference
(p < 0.05) between lanes 2 and 11.

To further ascertain the role of enhanced SUMOylation in protecting

Pa-4/SUMO-1
cells against DFO-elicited cell growth inhibition,

10 µM cell permeant peptide containing
SBM (146, 147) was

delivered into cells and treated with an increasing concentration

of
DFO (Figure 3.7). Cells that took up scrambled peptide and

were subsequently treated
with the same concentration of DFO

served as a control. There was not a notable
decrease in cell

viability in cells treated with SBM compared with control in

the absence
of DFO. However, in 100 µM DFO-treated

Pa-4/SUMO-1 cells, cell viability decreased
by more than 40%

in cells with 10 µM SBM compared with control. This

supported my
postulation that SUMOylation and SUMO-dependent

protein-protein interactions are
essential for modulating the

cellular adaptive responses to hypoxia.

35

DFO (μM) 0 50 100
0
20
40
60
80
100
120
% Cell Viability
Pa-4/SUMO-1
+SBM
-SBM
DFO (μM) 0 50 100
0
20
40
60
80
100
120
% Cell Viability
Pa-4/SUMO-1
+SBM
-SBM

Figure 3.7 SUMO-dependent protein-protein interaction is involved in hypoxic responses. Cell
permeant, SBM peptide and scrambled peptide (control peptide) were introduced into Pa-4/SUMO-1 cells
and treated with an increasing concentration of DFO (0, 50, and 100 µM).

DFO treatment has been reported to delay cells exiting from

S phase of cell cycle
transition (57). To explore the mechanism

underlying SUMO-dependent enhancement of
cell survival against

DFO, I then assayed whether the effect of SUMOylation is cell

cycle-dependent. As shown in Figure 3.8A, DFO (24 h) treatment

rendered 3.4% of Pa-4
and 7.1% of Pa-4/SUMO-1 cells to be accumulated

in G
2
/M phase, decreasing from 9.9%
and 7.9% of untreated Pa-4

and Pa-4/SUMO-1 cells, respectively. Notably, a
substantially

higher portion of Pa-4 cells was in the S phase of cell cycle,

whereas
overexpression of SUMO-1 rendered more cells entering

cell cycle transition, in response
to DFO (Figure 3.8A). I also

quantified the apoptosis of Pa-4 and Pa-4/SUMO-1 cells
after

DFO exposure by assessing the sub-G
1
cell population, a hallmark

of cell apoptosis.
A significant decrease in sub-G
1
cell population

was observed in Pa-4/SUMO-1 cells
compared with that of Pa-4

cells (Figure 3.8B). Together, it is possible that enhanced
36
SUMOylation

allows Pa-4 cells to escape DFO-elicited cell cycle S phase

arrest and
apoptosis.

AB
DFO (50 μM)04 24
Channels (FL2-A)
0 30 60 90 120 150
Channels (FL2-A)
0 30 60 90 120 150
Channels (FL2-A)
0 30 60 90 120 150
Channels (FL2-A)
0 30 60 90 120 150
Channels (FL2-A)
0 30 60 90 120 150
Channels (FL2-A)
0 30 60 90 120 150
Pa-4
Pa-4/SUMO-1
h
G
1
G
2
/M
↓ ↓
G
1
G
2
/M
↓ ↓
G
1
G
2
/M
↓ ↓
G
1
G
2
/M
↓ ↓
G
1
G
2
/M
↓ ↓
G
1
G
2
/M
↓ ↓
7.1 4.7 7.9 3.4 6.2 9.9
G
2
/M (%)
45.4 42.0 36.4 51.3 40.9 34.3 S (%)
47.4 53.3 55.7 45.4 52.9 55.7
G
1
(%)
24 h 4 h 0 h 24 h 4 h 0 h DFO (50 μM)
Pa-4/SUMO-1 Pa-4
DFO (50 μM)
∗
0 4 24 0 4 24 h
0
1
2
3
4
5
6
0
1
2
3
4
5
6
∗
Fold Induction in Sub-G
1
Cells
Pa-4 Pa-4/SUMO-1
AB
DFO (50 μM)04 24
Channels (FL2-A)
0 30 60 90 120 150
Channels (FL2-A)
0 30 60 90 120 150
Channels (FL2-A)
0 30 60 90 120 150
Channels (FL2-A)
0 30 60 90 120 150
Channels (FL2-A)
0 30 60 90 120 150
Channels (FL2-A)
0 30 60 90 120 150
Pa-4
Pa-4/SUMO-1
h
G
1
G
2
/M
↓ ↓
G
1
G
2
/M
↓ ↓
G
1
G
2
/M
↓ ↓
G
1
G
2
/M
↓ ↓
G
1
G
2
/M
↓ ↓
G
1
G
2
/M
↓ ↓
7.1 4.7 7.9 3.4 6.2 9.9
G
2
/M (%)
45.4 42.0 36.4 51.3 40.9 34.3 S (%)
47.4 53.3 55.7 45.4 52.9 55.7
G
1
(%)
24 h 4 h 0 h 24 h 4 h 0 h DFO (50 μM)
Pa-4/SUMO-1 Pa-4
DFO (50 μM)04 24
Channels (FL2-A)
0 30 60 90 120 150
Channels (FL2-A)
0 30 60 90 120 150
Channels (FL2-A)
0 30 60 90 120 150
Channels (FL2-A)
0 30 60 90 120 150
Channels (FL2-A)
0 30 60 90 120 150
Channels (FL2-A)
0 30 60 90 120 150
Pa-4
Pa-4/SUMO-1
h
G
1
G
2
/M
↓ ↓
G
1
G
2
/M
↓
G
2
/M
↓ ↓
G
1
G
2
/M
↓ ↓
G
1
G
2
/M
↓
G
2
/M
↓ ↓
G
1
G
2
/M
↓
G
2
/M
↓ ↓
G
1
G
2
/M
↓
G
2
/M
↓ ↓
7.1 4.7 7.9 3.4 6.2 9.9
G
2
/M (%)
45.4 42.0 36.4 51.3 40.9 34.3 S (%)
47.4 53.3 55.7 45.4 52.9 55.7
G
1
(%)
24 h 4 h 0 h 24 h 4 h 0 h DFO (50 μM)
Pa-4/SUMO-1 Pa-4
7.1 4.7 7.9 3.4 6.2 9.9
G
2
/M (%)
45.4 42.0 36.4 51.3 40.9 34.3 S (%)
47.4 53.3 55.7 45.4 52.9 55.7
G
1
(%)
24 h 4 h 0 h 24 h 4 h 0 h DFO (50 μM)
Pa-4/SUMO-1 Pa-4
DFO (50 μM)
∗
0 4 24 0 4 24 h
0
1
2
3
4
5
6
0
1
2
3
4
5
6
∗
Fold Induction in Sub-G
1
Cells
Pa-4 Pa-4/SUMO-1
DFO (50 μM)
∗
0 4 24 0 4 24 h
0
1
2
3
4
5
6
0
1
2
3
4
5
6
∗
Fold Induction in Sub-G
1
Cells
Pa-4 Pa-4/SUMO-1

Figure 3.8 Hypoxia renders S phase arrest. (A) Synchronized Pa-4 and Pa-4/SUMO-1 cells were treated
with either vehicle or 50 µM DFO for 4 and 24 h, respectively, and subjected to FACS analyses. The
populations of G
1
and G
2
/M cells were indicated by arrows. The percentage of cells in each phase of cell
cycle was summarized in the table. One representative FACS analysis from three independent analyses is
shown. (B) SUMOylation protects Pa-4 cells against DFO-elicited apoptosis as reflected by a decrease in
sub-G
1
population of Pa-4/SUMO-1 cells. The populations of sub-G
1
cells in (A) were normalized and
expressed as fold induction compared with that of vehicle-treated Pa-4 and Pa-4/SUMO-1 cells. * denotes
significant difference (p < 0.001) between lanes 3 and 6.

Effect of SUMOylation on DFO-induced ATM, PKCδ, and Caspase-3 activation
I further investigated whether DFO treatment induces DNA damage

by examining the
phosphorylation profiles of both ATM S1981

and H2AX S139 in DFO-treated Pa-4 cells
(Figure 3.9). The ATM

S1981 phosphorylation was detected at 4 h after treatment

and
was robustly stimulated at 24 h post-treatment in both

Pa-4 and Pa-4/SUMO-1 cells
(Figure 3.9A, lanes 5 and 6 versus

lane 4, top panel). Consistent with the DFO-mediated
37
ATM activation,

a marked induction of H2AX S139 phosphorylation was also noticed

in
both Pa-4 and Pa-4/SUMO-1 cells (Figures 3.9A and C).

AB
12 3 4 5 6
h 0 4 24 0 4 24 DFO (50 μM)
Cyclin A
Tubulin
Actin
SUMO-1
p-H2AX
p-ATM
Pa-4 Pa-4/SUMO-1
0
1
2
3
4
5
h 0 4 24 0 4 24 DFO (50 μM)
Pa-4 Pa-4/SUMO-1
Relative Induction
of p-ATM
5
0
1
2
3
4
Relative Induction
of p-H2AX
h 04 24 0 4 24 DFO (50 μM)
Pa-4 Pa-4/SUMO-1
h 04 24 0 4 24 DFO (50 μM)
Pa-4 Pa-4/SUMO-1
0
1
2
3
Relative Induction
of Cyclin A
12 3 4 5 6
h 0 4 24 0 4 24 DFO (50 μM)
Cyclin A
Tubulin
Actin
SUMO-1
p-H2AX
p-ATM
Pa-4 Pa-4/SUMO-1
12 3 4 5 6 12 3 4 5 6
h 0 4 24 0 4 24 DFO (50 μM)
Cyclin A
Tubulin
Actin
SUMO-1
p-H2AX
p-ATM
Pa-4 Pa-4/SUMO-1
0
1
2
3
4
5
h 0 4 24 0 4 24 DFO (50 μM)
Pa-4 Pa-4/SUMO-1
Relative Induction
of p-ATM
0
1
2
3
4
5
h 0 4 24 0 4 24 DFO (50 μM)
Pa-4 Pa-4/SUMO-1
Relative Induction
of p-ATM
5
0
1
2
3
4
Relative Induction
of p-H2AX
h 04 24 0 4 24 DFO (50 μM)
Pa-4 Pa-4/SUMO-1
5
0
1
2
3
4
Relative Induction
of p-H2AX
h 04 24 0 4 24 DFO (50 μM)
Pa-4 Pa-4/SUMO-1
h 04 24 0 4 24 DFO (50 μM)
Pa-4 Pa-4/SUMO-1
0
1
2
3
Relative Induction
of Cyclin A
h 04 24 0 4 24 DFO (50 μM)
Pa-4 Pa-4/SUMO-1
0
1
2
3
Relative Induction
of Cyclin A
CD

Figure 3.9 Induction of ATM S1981 and H2AX S139 phosphorylation by DFO treatment. Equal
amounts of protein lysates prepared from cells treated with DFO (50 µM) for different time periods as
indicated were subjected to Western analyses (A). After normalizing with that of tubulin, the respective
steady-state levels of S1981-phosphorylated ATM (p-ATM) and S-139-phosphorylated H2AX (p-H2AX)
in Pa-4 and Pa-4/SUMO-1 cells before treatment were designated as 1. The relative induction of p-ATM
(B), p-H2AX (C), and Cyclin A (D) by DFO is shown.

Because

PKCδ is one of the abundant PKC isoforms expressed in salivary

cells and is
proteolytically activated by Caspase-3 to induce

mitochondria-dependent apoptosis (131),
I next performed additional

experiments to address whether ATM S1981 phosphorylation
and

PKCδ/Caspase-3 activation are a sequential event in cells being

exposed to either 50
38
or 100 µM DFO. DFO-treatment

led to the generation of a 40-kDa fragment, a hallmark
of PKCδ

activation, at 48 h after treatment, and augmented SUMOylation

modestly
attenuated the induced PKCδ cleavage in Pa-4/SUMO-1

cells (Figures 3.10A and D).
Consistent with PKCδ cleavage pattern,

DFO treatment (24 h) elicited enhanced Caspase-
3 activation

in Pa-4 cells compared with Pa-4/SUMO-1 cells (Figures 3.10A and C). I
conclude that DFO treatment alone is sufficient to elicit

DNA damage and Caspase-
3/PKCδ activation in both Pa-4 cells and

Pa-4/SUMO-1 cells. Intriguingly, S1981-
phosphorylated ATM was

detected at 24 h after DFO treatment and before activation

of
the PKCδ/Caspase-3, which appeared at 48 h after treatment

in both Pa-4 and Pa-
4/SUMO-1 cells (Figure 3.10B). Together

with results shown in Figures 3.6 and 3.9,
SUMOylation conceivably

protected cells from DFO-elicited cell death by attenuating

the
activation of PKCδ and Caspase-3 on exposure to DFO.

39
AB
CD
Pa-4 Pa-4/SUMO-1
Time
DFO
0
0
24 48 24 48
50 100 50 100
0
0
24 48 24 48
50100 50 100
PKCδ
PKCδ activation
p-ATM
Pro-caspase-3
Tubulin
h
μM
Caspase-3 cleavage
Pa-4 Pa-4/SUMO-1
Time
DFO
0
0
24 48 24 48
50 100 50 100
0
0
24 48 24 48
50100 50 100
PKCδ
PKCδ activation
p-ATM
Pro-caspase-3
Tubulin
h
μM
Caspase-3 cleavage
Pa-4 Pa-4/SUMO-1
Time
DFO 100
h
μM
Relative Induction of Caspase-3 Cleavage
24
50
24 48
100
24
50
24
100
48
50
48
100
0
5
10
15
20
25
30
35
40
45
48
50
Pa-4 Pa-4/SUMO-1
Time
DFO 100
h
μM
Relative Induction of Caspase-3 Cleavage
24
50
24 48
100
24
50
24
100
48
50
48
100
0
5
10
15
20
25
30
35
40
45
0
5
10
15
20
25
30
35
40
45
0
5
10
15
20
25
30
35
40
45
48
50
Relative Induction of p-ATM
Pa-4 Pa-4/SUMO-1
Time
DFO
h
μM
24
50
48
100
24
50
24
100
48
50
48
100
48
50
0
5
10
15
20
25
30
35
40
24
100
Relative Induction of p-ATM
Pa-4 Pa-4/SUMO-1
Time
DFO
h
μM
24
50
48
100
24
50
24
100
48
50
48
100
48
50
0
5
10
15
20
25
30
35
40
0
5
10
15
20
25
30
35
40
24
100
Relative Induction of PKCδ Activation
Pa-4 Pa-4/SUMO-1
Time
DFO
24
50
24
100
48
100
24
50
24
100
48
50
48
100
h
μM
48
50
0
1
2
3
4
5
6
7
8
9
10
Relative Induction of PKCδ Activation
Pa-4 Pa-4/SUMO-1
Time
DFO
24
50
24
100
48
100
24
50
24
100
48
50
48
100
h
μM
48
50
Pa-4 Pa-4/SUMO-1
Time
DFO
24
50
24
100
48
100
24
50
24
100
48
50
48
100
h
μM
48
50
0
1
2
3
4
5
6
7
8
9
10
0
1
2
3
4
5
6
7
8
9
10

Figure 3.10 ATM activation proceeds Caspase-3 activation on DFO treatment. Pa-4 and Pa-4/SUMO-
1 cells were treated with an increasing concentration of DFO for different time periods as indicated and
subjected to Western analyses. The quantitative analyses on the relative induction of p-ATM (A), Caspase-
3 cleavage (B), and PKCδ (C) were performed as described in Figure 3.8.

It was previously reported that there is a decreased PKCδ expression

in ATM
-/-
cells
(91). To examine the role of PKCδ

activation in mediating DFO-elicited cell growth
inhibition,

I then treated pEBS7 (ATM
-/-
) and YZ5 (pEBS7+ATM)

cells with DFO
followed by MTT assays. As expected, the ATM-deficient

pEBS7 cells are relatively
40
resistant toward DFO treatment compared

with ATM-proficient YZ5 cells (Figure 3.11).
It was conceivable

that DFO-elicited PKCδ activation plays an essential role in

conveying
DFO-exerted cytotoxicity. Together, I postulate that

enhanced SUMOylation attenuates
PKCδ signaling cascade, resulting

in an increased cell survival in Pa-4/SUMO-1 cells
(Figure 3.6).

0
20
40
60
80
100
120
% Cell Viability
YZ5 (ATM+/+)
pEBS7 (ATM-/-)
DFO (μM) 0 50 100
0
20
40
60
80
100
120
% Cell Viability
YZ5 (ATM+/+)
pEBS7 (ATM-/-)
DFO (μM) 0 50 100

Figure 3.11 ATM-deficient cells are relative to resistant to DFO treatment. ATM-deficient pEBS7 and
ATM-proficient YZ5 cells were exposed to different concentrations of DFO, followed by MTT assays as
described in Figure 3.5.

Effect of SUMOylation on hypoxia-/DFO-induced NF-κB activation
To determine whether SUMO-1 has a distinct biological effect

on NF-κB activation
in response to either 1% O
2
or DFO exposure,

a luciferase reporter construct containing
NF-κB binding elements

was used in transfection assays. I first tested whether SUMO-1

attenuates NF-κB activation by cytoplasmic pathways. MEKK1, which

acts as an
upstream activator of IκB Kinase (IKK) (86, 112), was

chosen to activate NF-κB. It was
clear that wild-type SUMO-1

repressed MEKK1-mediated NF-κB reporter activation
41
(Figure 3.12,

lane 2 versus lane 1). Cotransfection with SUMO-1(M59V), harboring

an
M59V mutation analogous to the reported SUMO-4 M55V loss-of-function

mutation (10,
52), partially restored the transactivation of NF-κB

induced by MEKK1 (Figure 3.12,
lane 3 versus lane 2).

SUMO-1
SUMO-1(M59V)
MEKK1
-+-
-+ -
++ +
12000
10000
8000
6000
4000
2000
0
NF-κB Luciferase Activity
SUMO-1
SUMO-1(M59V)
MEKK1
-+-
-+ -
++ +
12000
10000
8000
6000
4000
2000
0
NF-κB Luciferase Activity

Figure 3.12 SUMO-1 attenuates NF-κB-dependent transcription by cytoplasmic pathway. HEK 293T
cells were cotransfected with NF-κB reporter, pRL-TK indicator, and a combination of pEGFP-C1 (vector
control), SUMO-1 (wild-type), SUMO-1(M59V), and MEKK1 expression constructs. The relative
luciferase activities from the firefly luciferase reporter gene were determined and normalized with Renilla
indicator luciferase activity, and the normalized NF-κB activity (x1000) is shown. SUMO-1 expression led
to a 10-fold repression of the MEKK1-activated NF-κB transactivation, whereas SUMO-1(M59V)
conveyed a 3.6-fold higher MEKK1-dependent NF-κB transactivation than did SUMO-1.

Immunoprecipitation

followed by Western analyses were performed to confirm that

IκB α was SUMOylated (Figure 3.13, top panel). The decreased

MEKK1-mediated NF-
κB activation in SUMO-1-transfected cells (Figure

3.12, lane 2 versus lane 1) resulted, at
least in part, from

the higher IκB α in MEKK1-transfected cells (Figure 3.13, bottom

panel, lane 4 versus lane 2). These data confirm that SUMO-1

is involved in attenuating
the MEKK1-mediated NF-κB transactivation.
42

4 3 2 1
IP: anti-SUMO-1
IB: anti-IκBα
SUMO-1 - + - +
IP
SUMO-IκBα
Input
IκBα
3.5
0
0.5
1
1.5
2
2.5
3
SUMO-1
MEKK1
Control
12 3 4
-+ -+
Relative IκBα Level
4 3 2 1
IP: anti-SUMO-1
IB: anti-IκBα
SUMO-1 - + - +
IP
SUMO-IκBα
Input
IκBα
4 3 2 1
IP: anti-SUMO-1
IB: anti-IκBα
SUMO-1 - + - +
IP
SUMO-IκBα
Input
IκBα
3.5
0
0.5
1
1.5
2
2.5
3
SUMO-1
MEKK1
Control
12 3 4
-+ -+
Relative IκBα Level
3.5
0
0.5
1
1.5
2
2.5
3
3.5
0
0.5
1
1.5
2
2.5
3
SUMO-1
MEKK1
Control
12 3 4
-+ -+
Relative IκBα Level

Figure 3.13 SUMO-1 antagonizes MEKK1-mediated NF-κB activation by stabilizing the steady-state
level of IκB α. HeLa cells (top panel) were transfected with or without pEGFP-SUMO-1. Equal amounts of
cell lysates were subjected to immunoprecipitation with an anti-SUMO-1 antibody, followed by Western
analyses with an anti-IκB α antibody (lanes 3 and 4). The inputs (5%; lanes 1 and 2) were also analyzed by
an anti-IκB α antibody. HEK 293T cells (bottom panel) were transfected with a vector (pEGFP-C1) alone
(lanes 1 and 2) or pEGFP-SUMO-1 (lanes 3 and 4) in the presence and absence of MEKK1. The effect of
SUMO-1 on MEKK1-mediated IκB α degradation was determined by densitometry of signals obtained
from Western analyses using an anti- IκB α antibody and anti-actin antibody (data not shown). After
normalizing with that of actin, the lowest steady-state level of IκB α is designated as 1 (lane 2). The relative
IκB α level, which represents one of three independent experiments, is shown.

I next performed the same NF-κB reporter assays in Pa-4 and

Pa-4/SUMO-1 cells
under a combination of hypoxia and TNF- α treatments. TNF- α (25 ng/ml) alone elicited
a very modest stimulating effect

on NF-κB activation in Pa-4 cells (Figure 3.14, lane 2
versus

lane 1), as previously reported in endothelial cells (21). Although I observed an
enhancement effect of 1% O
2
or 25 µM

DFO on NF-κB transactivation in Pa-4 cells
(Figure 3.14, lanes

5 and 9 versus lane 1), I also noticed that the basal NF-κB

activities in
43
Pa-4/SUMO-1 cells were higher than those in Pa-4

cells (Figure 3.14, lane 3 versus lane 1)
and that TNF- α (25 ng/ml)

failed to induce NF-κB activation in Pa-4/SUMO-1 cells
under

normoxic condition (Figure 3.14, lane 4 versus lane 3). In addition,

treatment of
Pa-4/SUMO-1 cells with either 1% O
2
or 25 µM

DFO was able to further enhance NF-κB
activation, despite their

high basal NF-κB level (Figure 3.14, lanes 7 and 11 versus lane

3). Addition of TNF- α (25 ng/ml) to either 1% O
2
- or DFO-treated

Pa-4 or Pa-4/SUMO-
1 cells had no stimulatory NF-κB activation

(Figure 3.14, lanes 6, 8, 10, and 12 versus
lanes 5, 7, 9, and

11). The lack of additional TNF- α-mediated NF-κB activation in

Pa-4
cells under hypoxic conditions supported our notion that

SUMOylation elicited
differential regulation on NF-κB-dependent

transactivation. Importantly, DFO/hypoxic
treatment in the context

of enhanced SUMOylation resulted in a markedly stronger NF-
κB

activation (Figure 3.14, lanes 7 and 11 versus lanes 5 and 9,

respectively).

44
0
2
4
6
8
10
12
14
Normoxia 1% O
2
Pa-4/SUMO-1
1 3 4 5 6 7 8 9 10 11 12
-+ - + - - + - + - +
2
∗
‡
∗
†
†
25 μM DFO
Pa-4
TNF-α +
‡
Fold Induction of NF-κB Activity
Pa-4/SUMO-1 Pa-4 Pa-4/SUMO-1 Pa-4
0
2
4
6
8
10
12
14
Normoxia 1% O
2
Pa-4/SUMO-1
1 3 4 5 6 7 8 9 10 11 12
-+ - + - - + - + - +
2
∗
‡
∗
†
†
25 μM DFO
Pa-4
TNF-α +
‡
Fold Induction of NF-κB Activity
Pa-4/SUMO-1 Pa-4 Pa-4/SUMO-1 Pa-4

Figure 3.14 SUMO-1 enhances the hypoxia-stimulated NF-κB-dependent transcription. Pa-4 and Pa-
4/SUMO-1 cells were transfected with an NF-κB reporter and pRL-TK indicator. After transfection, cells
were exposed to nonhypoxia, 1% O
2
, or 25 µM DFO for 24 h in the presence and absence of TNF- α (25
ng/ml). The normalized NF-κB activity under non-hypoxia in the absence of TNF- α was designated as 1
(lane 1). The fold induction was calculated by dividing the normalized luciferase activity under indicated
experimental conditions with the NF-κB activity observed under normoxic conditions in the absence of
TNF- α. * denotes significant difference (p < 0.001) between lanes 5 and 7. ‡ denotes significant difference
(p < 0.01) between lanes 9 and 11. † denotes significant difference (p < 0.05) between lanes 1 and 3.

By using TransBinding NF-κB Assays (Panomics),

I further confirmed that the
increased NF-κB-dependent reporter

transactivation resulted from enhanced NF-κB
activity (Figure 3.15).

45

0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
Pa-4 Pa-4/SUMO-1
-
+
-
+
Pa-4 Pa-4/SUMO-1
-
+
-
+
1% O
2
DFO 50 μM
Relative NF-κB DNA-binding
Activity
Treatment
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
Pa-4 Pa-4/SUMO-1
-
+
-
+
Pa-4 Pa-4/SUMO-1
-
+
-
+
1% O
2
DFO 50 μM
Relative NF-κB DNA-binding
Activity
Treatment

Figure 3.15 SUMOylation enhances NF-κB DNA-binding activity. Pa-4 and Pa-4/SUMO-1 cells were
treated with 1% O
2
or 50 µM DFO for 16 h and subjected to TransBinding NF-κB Assays (Panomics).

I next examined the steady-state level of ZO-1 protein to determine

the effect of
SUMOylation on governing endogenous ZO-1 expression

under DFO-stressed condition.
Consistent with results demonstrated

in Figures 3.4 and 3.5, there was a clear decrease of
the steady-state

level of ZO-1 in Pa-4 cells after DFO treatment, and enhanced

SUMOylation presumably retarded such a DFO-elicited down-regulation

of ZO-1 at 6 h
post-treatment (Figure 3.16). Together, I

postulate that SUMOylation probably protects
Pa-4 cells against

hypoxia- and DFO-elicited cytotoxicity by attenuating pro-apoptotic

signals, such as activation of PKCδ and Caspase-3, and by augmenting

pro-survival
signals, such as NF-κB activation and sustained

level of ZO-1, on hypoxic exposure
(Figure 3.17).

46
12 3 4 56
Tubulin
Actin
EGFP-SUMO-1
ZO-1
DFO (50 μM) 0 6 30 h
Pa-4 Pa-4/SUMO-1
06 30
12 3 4 56
Tubulin
Actin
EGFP-SUMO-1
ZO-1
DFO (50 μM) 0 6 30 h
Pa-4 Pa-4/SUMO-1
06 30
0
0.2
0.4
0.6
0.8
1
1.2
1
1.1
0.4
DFO (50 μM)
1
0.7
0.4
06 30 h 06 30
Pa-4 Pa-4/SUMO-1
Relative ZO-1 Expression
0
0.2
0.4
0.6
0.8
1
1.2
1
1.1
0.4
1
1.1
0.4
DFO (50 μM)
1
0.7
0.4
06 30
1
0.7
0.4
06 30 h 06 30
Pa-4 Pa-4/SUMO-1
Relative ZO-1 Expression

Figure 3.16 SUMOylation retards DFO-induced decrease on steady-state level of ZO-1. Pa-4 and Pa-
4/SUMO-1 cells were treated with 50 µM DFO for indicated time periods and subjected to Western
analyses with anti-ZO-1, anti-actin, and anti-tubulin antibodies, respectively. After normalizing with that of
tubulin, the respective steady-state level of ZO-1 in Pa-4 and Pa-4/SUMO-1 cells before treatment was
designated as 1. The relative ZO-1 level, which represents one of three independent experiments, is shown.

Hypoxia, DFO
DNA Double strand breaks
ATM activation
(S1981 phosphorylation)
γH2AX Phosphorylation
Cell cycle
arrest
Apoptosis DNA
repair
Cell death
Caspase-3 activation
PKCδ activation
Cell survival
NF-κB activation
Passive barrier
properties
TER F-actin/ZO-1
reassembly
Salivary adaptive responses
SUMOylation-regulated signalings
SUMOylation
SUMOylation
SUMOylation
SUMOylation
Hypoxia, DFO
DNA Double strand breaks
ATM activation
(S1981 phosphorylation)
γH2AX Phosphorylation
Cell cycle
arrest
Apoptosis DNA
repair
Cell death
Caspase-3 activation
PKCδ activation
Cell survival
NF-κB activation
Passive barrier
properties
TER F-actin/ZO-1
reassembly
Salivary adaptive responses
SUMOylation-regulated signalings
SUMOylation SUMOylation
SUMOylation
SUMOylation
SUMOylation

Figure 3.17 Hypoxia and DFO induce SUMOylation-regulated signalings to modulate salivary
adaptive responses. SUMOylation attenuates sensitivity toward hypoxia- or DFO-induced cell death by
down-regulating S139 phosphorylation of γH2AX, inhibiting proteolytic activation of PKCδ and Caspase-3,
inducing NF-κB transactivation, and strengthening passive barrier properties.

47
Discussion
I have investigated alterations in sensitivity toward hypoxia

(1% O
2
) or DFO as
reflected by enhanced cell survival and accelerated

reassembly of F-actin/ZO-1 and
restoration of transepithelial

electrical resistance on exposure of SUMO-1-overexpressing
salivary

Pa-4 cells to 1% O
2
or DFO. Together, I propose that 1% O
2
-

or DFO-induced
SUMOylation partly accounts for these manifested

phenotypes through reprogrammed
adaptive responses to hypoxia. For example, salivary epithelial Pa-4 cells possess the
ability

to restore their TER and F-actin/ZO-1 assembly after undergoing

prolonged
hypoxic exposure, and this aspect of the "recovery"

phase during the course of chronic
exposure to 1% O
2
is also

potentiated by increased SUMOylation.
Hypoxia-mediated signaling pathways play key roles in critical

developmental and
physiological processes including angiogenesis,

erythropoiesis, glucose transport,
glycolysis, iron transport,

cell survival, and proliferation (reviewed in 60 and 136). More
importantly, clinical and laboratory studies have

demonstrated that hypoxia-dependent
events are involved in numerous

human diseases, such as myocardial and cerebral
ischemia, pulmonary

hypertension, preeclampsia, intra-uterine growth retardation,

and
cancer (137). It gradually became appreciated that cells may

respond in a more
active/adaptive manner than a mere passive

fashion after prolonged hypoxic exposure,
correlating well with

the hypoxia-induced SUMOylation. Herein, I addressed the role

of
induced SUMOylation in modulating hypoxic responses by demonstrating

that
SUMOylation promotes transient TER maintenance, attenuates

Caspase-3/PKCδ
cleavage, and enhances genotoxicity-induced NF-κB

transactivation, maintaining
epithelial integrity of hypoxia

and DFO-treated salivary Pa-4 cells (Figure 3.17).
48
Conceivably, SUMOylation

could function as an important posttranslational modification

for protecting epithelial cells against DFO-elicited DNA damage

and also accelerate
"recovery" processes from chronic hypoxia-induced

tight junction disruption on
prolonged hypoxia.

What are the SUMOylation targets involved in modulating cellular

responses to
hypoxia or DFO? Bae et al., (3) and Shao et al., (142) have independently reported that
HIF-1 α can be modified by SUMOylation,

leading to an increase in HIF-1 α stability and
its transcriptional

activation during hypoxia. It was also indicated by Shao et

al., (142)
that the induction of SUMO-1 and SUMOylated HIF-1 α by hypoxia

is accompanied by
an increased production of vascular endothelial

growth factor, a known target of
hypoxia/HIF-1 α. The purposes

of this report are to demonstrate that DFO treatment also
induces

SUMOylation and to identify and characterize additional signaling

pathways that
participate in salivary adaptive responses to

stress by hypoxia or DFO. My results clearly
demonstrate that

SUMO-1-attenuated activation of PKCδ and Caspase-3 and SUMO-1-
enhanced

NF-κB activation in response to DFO are positively correlated

with enhanced
survival of DFO-treated Pa-4 cells.

Among various SUMOylation functions, the regulation of NF-κB

activation by
SUMOylation is perhaps one of the most intriguing

ones. Cytokine-mediated NF-κB
activation occurs through the signaling-induced

proteolytic degradation of IκB, via an
ubiquitination-mediated

system, rendering the nuclear translocation of NF-κB.
Polyubiquitination

of IκB α, mainly at lysine residues 21 and 22, targets it for proteasomal

degradation after extracellular stimuli-mediated IκB α phosphorylation. In contrast,
49
SUMOylation at lysine 21 of IκB α appears to protect

IκB α from proteasome-mediated
degradation, thus inhibiting NF-κB-dependent

transcription by cytokines (36). In
addition, it is also postulated

that SUMOylation of IκB α is a nuclear event to titrate out
nuclear

NF-κB (62). I first demonstrated that SUMO-1 antagonizes MEKK1-mediated

NF-κB activation (Figure 3.12), confirming the already published

role of SUMO-1 in
down-regulating cytokine-induced NF-κB activation.
However, a recent report by Huang et al., (65) suggested that SUMOylation

of NF-κB
essential modulator (NEMO), an essential NF-κB nuclear

modulator, allows NEMO to be
incorporated into ATM signaling

complex in response to genotoxic stress. Modified
NEMO then

exits the nucleus and associates with and activates the IKK

complex,
resulting in release of NF-κB from its IκB α inhibitor. Hence, the SUMO modification of
NEMO and the ATM activation

work in concert to activate NF-κB and its downstream
survival

pathway(s) against genotoxic stress. Consistent with previous

reports (56, 57), I
further demonstrated that hypoxia-mimetic

DFO exposure leads to ATM activation and
H2AX S139 phosphorylation

in salivary epithelial cells (Figure 3.9), hallmarks of DNA

damage, and causes Pa-4 cell cycle arrest in S phase (Figure

3.8). Importantly, by using
SBM to interfere with presumably

SUMO-dependent protein-protein interaction (146,
147), I was able

to partially reverse the observed resistance toward DFO-induced

cytotoxicity in Pa-4/SUMO-1 cells (Figure 3.7). Together with

Figure 3.9, I further
concluded that hypoxia-/DFO-induced NF-κB

activation (Figures 3.14 and 3.15) is
mediated by a genotoxic signaling

pathway, which is distinct from the aforementioned
SUMOylation-sensitive

NF-κB activation pathway used by cytokines. Moreover, I
50
demonstrated

that Pa-4/SUMO-1 cells exhibit a higher basal NF-κB activity

(Figures 3.14
and 3.15), supporting the notion that SUMO-1 also

acts in a pro-NF-κB manner to protect
cells against genotoxicity

generated endogenously, such as DNA replication. The higher

basal and hypoxia-/DFO-induced NF-κB activity may, at least in

part, result in the
development of resistance against injuries

caused by hypoxic exposure in Pa-4/SUMO-1
cells. The mechanism

underlying SUMOylation-mediated attenuation of DFO-triggered

PKCδ/Caspase-3 cleavage is still unclear and currently under

investigation.
The present study also provides evidence that overexpression

of SUMO-1 leads to a
transient protection of passive barrier

properties (e.g., TER) against acute hypoxic injury
(Figure 3.4C). The increase in TER under baseline conditions indicates that

SUMO-1
overexpression enhances epithelial integrity. However,

I postulate that this is probably
not by a direct effect by

enhanced SUMOylation on TER per se but may take place via a

regulatory mechanism used by SUMO-1 to enhance the expression

of tight junction-
associated proteins, subsequently leading

to the observed higher TER under acute
hypoxia (Figure 3.4C). For example, I presented evidence that there was a higher level

of
ZO-1 protein detected in Pa-4/SUMO-1 cells than that in Pa-4

cells at 6 h post- treatment
(Figure 3.16). Moreover, the

increase in NF-κB activity by hypoxia has been proposed to
protect

against hypoxia-induced increase in paracellular permeability (15), a hallmark of
deranged tight-junction properties, also lending

credence to our results. Therefore,
overexpression of SUMO-1

in Pa-4 cells may result in an enhancement of genotoxicity-
induced

NF-κB transactivation that subsequently induces transcription

and translation of
certain target genes, resulting in higher

TER under non-hypoxia and acute hypoxia, and
51
also leading to

accelerated reassembly of F-actin/ZO-1 under prolonged hypoxia.

There
are also studies that identified loss of organization

in the F-actin and ZO-1 organization
as being a critical event

leading to a lower TER and increased paracellular permeability

in
epithelial monolayers (15, 51, 54, 63, 117, 118). Results from my studies

supported this
notion by showing that hypoxic stress causes

a disruption of F-actin and ZO-1
organization and a decrease

of steady-state level of ZO-1 (Figures 3.5 and 3.16). In
addition,

TNF- α in hypoxia also functions in an autocrine manner, decreasing

barrier
function. Consequently, the failure of TNF- α to induce

NF-κB transactivation in
hypoxia-treated Pa-4/SUMO-1 cells (Figure

3.14) would protect barrier function.
Emerging evidence has suggested that the maintenance or breakdown

of epithelial
integrity has significant impact on proper tissue

and organ function (163). For example,
disruption of epithelial

or endothelial barrier formation may lead to altered pulmonary

permeability and airway fluid accumulation, unregulated cancer

cell growth and increased
invasiveness, and perturbation of

vascular permeability (45, 50, 92). Hence, it is not
surprising to

see that SUMOylation enhances both epithelial barrier function

and cell
survival in response to hypoxic stress. The exact nature

of cross-talk between signaling
pathways rendering the observed

protection of F-actin/ZO-1 assembly and the increased
resistance

against hypoxia remains enigmatic. Notwithstanding the uncertainty

concerning the detailed mechanism underlying these two processes,

my results
unequivocally demonstrate a crucial role of SUMOylation

in augmenting salivary
adaptive response(s) to exposure to either

hypoxia or DFO. Although it is possible that
these two events

are indirectly linked, I postulate that the reorganization

of tight junctions
and the genotoxic signalings may functionally

cooperate downstream from enhanced
52
SUMOylation process. One

plausible model is that genotoxicity-induced NF-κB
transactivation

is part of a central regulatory circuit in response to hypoxic

stress.

Comerford et al., (26) recently proposed that SUMOylation represents

a general "off
switch" to hypoxia-induced inflammatory phenotypes. I have provided evidence herein
to suggest that SUMO-1 overexpression

protects salivary Pa-4 cells against epithelial
injury by TER

maintenance after acute hypoxia, hence providing some protection

for
hypoxic cells. I propose a possible mechanism by which

enhanced SUMOylation might
be protective via a "preconditioning"

effect. Further studies will be needed to resolve
many remaining

questions about this system, including the identities of other

SUMOylation targets, the role of NF-κB activation in controlling

SUMO-dependent
resistance to hypoxia-mediated cell killing,

and the mechanisms underlying an
accelerated restoration of

passive barrier properties, e.g., TER in SUMO-1-expressing
cells,

after prolonged hypoxia. It is also known that tumor cells are

more resistant to
chemotherapy and radiation therapy under hypoxic

conditions (60, 61). Hence, the
activation of genotoxicity-induced

NF-κB pathway and the SUMOylation process are
likely to be important

targets for hypoxia-mimetic cytotoxins, because they represent

an
oxygen-sensitive activation/induction mechanism and contribute

to the altered signaling
events induced by hypoxia.

53
CHAPTER 4
PROTEIN KINASE Cδ-DEPENDENT AND -INDEPENDENT SIGNALING IN
GENOTOXIC RESPONSE TO TREATMENT OF DESFERROXAMINE (DFO),
A HYPOXIA-MIMETIC AGENT
Introduction
Desferroxamine (DFO), an iron chelator, is known to regulate the expression of
hypoxia-inducible genes, such as vascular endothelial growth factor (VEGF), by
stabilizing hypoxia inducible factor 1-alpha (HIF-1α) (18, 100). Iron chelation also
affects the activity of a number of enzymes, including those in the respiratory chain
complexes, which require the formation of heme groups or iron-sulfur clusters. As a
consequence, treatment with a sub-toxic concentration of DFO leads to a reduction of
total cellular ATP levels and a decrease in the proportion of cells with active
mitochondria (166). By taking advantage of this hypoxia-mimicking effect, DFO-
treatment has been widely used as a method to experimentally simulate hypoxic
conditions (69). Clinical trials have further confirmed a promising lead for DFO as a
therapeutic agent against certain types of cancer (18, 29, 100).
Hypoxia-reoxygenation has been shown to render DNA replication fork arrest,
followed by DNA damage and stabilization of p53, resulting in DNA damage response
(48, 56). The DNA damage response is usually initiated by the activation of the ataxia
telangiectasia mutated (ATM) and the homologous ATR proteins (143). ATM and ATR
share many common substrates and are usually activated with distinct kinetics or by
different genotoxins. Once activated, ATM phosphorylates various downstream
substrates, including p53, Chkl, Chk2 and histone H2AX. ATM activation exerts three
54
crucial functions: regulation and stimulation of DNA repair, signaling cell-cycle
checkpoints and signaling cell death (84). If un-repaired, this DNA damage, especially
DNA double-stranded breaks (DSBs) could trigger cell death. Caspases are intracellular
cysteine proteases that mediate cell death and inflammation. Caspase-3 is a major
mediator of apoptotic cell death (172). Conversely, the protein Akt is a serine-threonine
kinase that has a central role in provoking suppression of apoptosis. Inhibition of
apoptosis by Akt leads to increased cell survival (5). The substrates of Akt include pro-
apoptotic Bad, Caspase-9 and ASK1, leading to a direct suppression of mitochondrial
apoptotic pathway. As DFO mimics hypoxic condition in various cells and tissues, it will
be of great interests to evaluate the effect of DFO-treatment on modulating DNA damage
response and pro-/anti-apoptotic signaling and their crosstalk.
The protein kinase C (PKC) family of serine-threonine kinases is activated by diverse
stimuli and participates in a variety of cellular processes, such as growth, differentiation,
apoptosis and cellular senescence (6, 31, 53, 149, 164, 177). To date, the PKC family
comprises 11 isoforms that are subgrouped on the basis of structure and their activation
modes into three subfamilies: The classical PKCs (α, β and γ) that are activated by
diacylglycerol (DAG) and calcium; the novel PKCs (δ, ε, η and θ) that require DAG, but
not calcium, for activation; and the atypical PKCs (ζ and λ/τ) that can be activated in a
DAG- and calcium-independent manner (14, 114, 149). Studies with PKCδ
-/-
mice
suggest that PKCδ plays pivotal roles in the regulation of cell proliferation and apoptosis
(6, 66, 88, 149). Mechanistically, PKCδ is activated in response to numerous cellular
stimuli by various mechanisms, including membrane translocation (72, 161), protein-
55
protein interaction (8), tyrosine phosphorylation (33, 77) and proteolytic cleavage (33, 77,
131). The translocation of PKCδ to different subcellular compartments and/or proteolytic
cleavage can be induced by ceramide, TNF-α, UV irradiation, ionizing radiation,
testosterone, oxidative stress and etoposide (14, 37, 103, 105, 131).
Although PKCδ plays a central role in the regulation of responses incurred by a
variety of stimuli, a detailed characterization of the function of PKCδ and its underlying
mechanism under hypoxia- or DFO-induced stress condition is still elusive. Herein, I
report that PKCδ is a key regulator of the salivary adaptive signaling network in response
to the genotoxic stress elicited by DFO-treatment. I utilized lentiviral delivery of sh-
PKCδ, wild-type PKCδ-EGFP or Lys376Arg-mutated PKCδ-EGFP to investigate DFO-
mediated PKCδ activation and examine whether DNA damage and cell survival
pathways are differentially regulated in salivary epithelial Pa-4 cells with a number of
distinct PKCδ contexts upon the exposure of cells to DFO. Results from my studies
demonstrate a time-dependent proteolytic activation and nuclear translocation of PKCδ in
response to DFO-treatment. I further establish that the Lys376 residue of PKCδ is
essential for DFO-induced PKCδ activation and that Lys376Arg-mutated PKCδ-EGFP
functionally dysregulates PKCδ-dependent adaptive responses in a dominant-negative
manner. My results also suggest that DFO-treatment induces differential p53, Chk1 and
Akt activation profiles in Pa-4, Pa-4/sh-PKCδ and Pa-4/PKCδKD-EGFP cells. The latter
results provide a potential underlying mechanism by which fine-tuning of salivary
cellular responses to genotoxic responses upon the treatment of DFO may be
accomplished through PKCδ. A crosstalk among the activated PKCδ, Caspase-3, Akt
56
and ATM signaling pathways is proposed. To the best of our knowledge, this is the first
demonstration showing that PKCδ plays a unique role in mediating adaptive responses to
DFO-exerted genotoxicity.

Results
PKCδ is proteolytically activated upon DFO-treatment
In order to establish the role of PKCδ in DFO-induced cellular response(s), I
engineered a series of lentiviral constructs which express wild-type (WT) or kinase-dead
(KD; Lys376Arg) PKCδ fused to EGFP (PKCδWT-EGFP or PKCδKD-EGFP), or RNAi
(sh-PKCδ) for suppressing the expression of endogenous PKCδ, as described in Dr. C.
Clavijo’s dissertation. Since salivary Pa-4 cells express a high level of endogenous
PKCδ, they were chosen to be as an experimental model for our studies to achieve my
goal. Pa-4 cells transduced with individual lentivirus, harboring either PKCδWT-EGFP
or PKCδKD-EGFP, were then selected to obtain the highest protein expression clones of
Pa-4/PKCδWT-EGFP or Pa-4/PKCδKD-EGFP, respectively. To downregulate PKCδ
expression, four different sh-PKCδs were pooled in a simultaneous transduction to obtain
~40% PKCδ gene silencing effect. Two clones from pools showed ≥ 90%
downregulation of endogenous PKCδ expression were used for studies shown herein.
I next characterized whether DFO elicits PKCδ activation. One reported mechanism
underlying PKCδ activation is its proteolytic cleavage, which yields a 40-kDa catalytic
fragment (CF) with a constitutive kinase activity (34). The generation of PKCδ-CF has
been considered as a hallmark of PKCδ activation in apoptotic cells (37, 42, 47). As
57
demonstrated by Western blotting analyses, PKCδ was proteolytically activated in DFO-
treated Pa-4 cells, evident by the generation of PKCδ-CF. Since the treatment of 100 μM
or 200 μM DFO failed to yield additional accumulation of PKCδ-CF, 50 μM DFO-
treatment was utilized to treat Pa-4 cells in all remaining studies. It was also noted that
DFO (50 μM)-mediated generation of PKCδ-CF is not detectable until 24 h post-
treatment, but increases thereafter (shown in Dr. C. Clavijo’s dissertation). Pa-4/sh-
PKCδ cells served as a control to ascertain the validity of PKCδ-CF generation in Pa-4
cells in response to DFO-treatment. Intriguingly, PKCδKD-EGFP delayed the
accumulation of PKCδ-CF from endogenous PKCδ, indicating that Lys376Arg mutation
of PKCδKD-EGFP affects the generation of PKCδ-CF from endogenous PKCδ in DFO-
treated cells. Conceivably, the kinase activity is perhaps required for an earlier event
necessary for DFO-elicited PKCδ cleavage in Pa-4 cells.

DFO stimulates nuclear translocation of PKCδ
The activation of classical and novel PKC is often noted with their translocation from
the soluble to the particulate cellular fractions through the interaction between the C1
domain and membrane lipids (114). Since subcellular localization change is expected
upon PKCδ activation, I performed immunocytochemical analyses to investigate if DFO-
treatment induces PKCδ translocation to different cellular compartments. As I observed
consistently, endogenous PKCδ was seen throughout the cells prior to DFO-treatment.
Following treatment with DFO, PKCδ became enriched in the nucleus where the
intensity of signals detected from endogenous PKCδ reached maximum between 24 h and
58
32 h post-treatment, correlating with time-dependent generation of PKCδ-CF (Figure 4.1;
courtesy from Dr. C. Clavijo’s dissertation). I next tried to confirm this observation by
performing imaging analyses with Pa-4/PKCδWT-EGFP cells through monitoring EGFP
signals at different post-DFO-treatment time points. Notably, PKCδWT-EGFP exhibited
a restricted vesicle-like signal surrounding the nucleus in untreated cells, but translocated
to the nucleus in DFO-treated cells (Figure 4.2A; courtesy from Dr. C. Clavijo’s
dissertation). Although the gross intracellular localization of PKCδ and the trend of
nuclear translocation kinetics during the course of DFO-treatment were comparable
between the endogenous PKCδ and the stably expressed PKCδWT-EGFP, notable
differences existed. I cannot, however, rule out the possibility that EGFP-tagging
perhaps contributes, at least in part, to the localization of PKCδWT-EGFP fusion protein
and the kinetics of PKCδWT-EGFP nuclear translocation upon DFO-exposure, compared
to those seen with endogenous PKCδ.
On the other hand, these analyses also indicated that PKCδKD-EGFP failed to locate
in the nucleus of DFO-treated cells (Figure 4.2A; courtesy from Dr. C. Clavijo’s
dissertation). Instead, scanning confocal microscopy with analyses using reconstruction
of orthogonal image planes indicated that PKCδKD-EGFP was enriched at the plasma
membrane at 48 h post-treatment (Figure 4.2B; courtesy from Dr. C. Clavijo’s
dissertation), suggesting that the Lys-to-Arg mutation of PKCδ at amino acid residue 376
abolished its DFO-induced nuclear translocation capacity. Compared to PKCδWT-EGFP
signals, a substantial portion of PKCδKD-EGFP signals were localized in a peri-nuclear
compartment under control condition, while DFO-treatment enriched the peri-nuclear
59
localization of PKCδKD-EGFP. Although the exact mechanism underlying differential
localization between PKCδWT-EGFP and PKCδKD-EGFP remains elusive, I surmise that
transient recruitment of PKCδ to a peripheral cellular location, probably the plasma
membrane, is part of PKCδ sorting or activation process and that Lys376Arg mutation
renders PKCδKD-EGFP to be sequestered at the plasma membrane.

PKCδ
DAPI
Tubulin
Merged
DFO (50 μM): 0 h 24 h 32 h 48 h
ab c d
efg h
ij k l
mn o p
PKCδ
DAPI
Tubulin
Merged
DFO (50 μM): 0 h 24 h 32 h 48 h
ab c d
efg h
ij k l
mn o p

Figure 4.1 DFO induces nuclear translocation of PKCδ. Pa-4 cells were seeded on 16-well cell culture
chambers, treated with 50 μM DFO and harvested at the indicated time points. Cells were processed for
confocal microscopy as indicated in Materials and Methods. Immunostaining of endogenous PKCδ (green),
DAPI staining for identification of nuclei (blue) and immunostaining for tubulin (red) are shown. Images
were acquired using a Nikon epifluorescence inverted microscope and processed using the Metamorph
Imaging, Corel Photo-Paint and LSM 5 Image Browser Software.

60
B
0 h
12 h
48 h
24 h
36 h
PKCδWT-EGFP PKCδKD-EGFP
ij k l
mn o p
ab c d
ef g h
qr s t
DFO (50 μM):
-+ -+
A
a b
c d
Control
48 h
DFO 50 μM
48 h
PKCδWT-EGFP PKCδKD-EGFP B
0 h
12 h
48 h
24 h
36 h
PKCδWT-EGFP PKCδKD-EGFP
ij k l
mn o p
ab c d
ef g h
qr s t
DFO (50 μM):
-+ -+
0 h
12 h
48 h
24 h
36 h
PKCδWT-EGFP PKCδKD-EGFP
ij k l
mn o p
ab c d
ef g h
qr s t
DFO (50 μM):
-+ -+
A
a b
c d
Control
48 h
DFO 50 μM
48 h
PKCδWT-EGFP PKCδKD-EGFP
a b
c d
Control
48 h
DFO 50 μM
48 h
PKCδWT-EGFP PKCδKD-EGFP

Figure 4.2 PKCδ kinase function is required for its DFO-induced nuclear translocation. (A) Pa-
4/PKCδWT-EGFP or Pa-4/PKCδKD-EGFP cells were seeded in an 8-well cell culture chambers and either
treated with or without 50 μM DFO in full culture medium. Cell images were acquired at the indicated time
points using a Nikon epifluorescence inverted microscope and processed as in Figure 4.1. (B) Orthogonal
image planes were reconstructed using the C-Imaging, Corel Photo-Paint and LSM 5 Image Browser
Software. DFO-treatment induces a translocation of PKCδ-EGFP from cytoplasm to nucleus whereas
PKCδKD-EGFP accumulates at the plasma membrane.
61
Dysregulation of DFO-induced DNA damage response signaling by sh-PKCδ or
PKCδKD-EGFP
DFO-treatment was reported to affect genome integrity by us and others (32). I next
performed biochemical studies to investigate the crosstalk between ataxia telangiectasia
mutated (ATM) cascade and PKCδ signaling during DFO-induced adaptive responses. I
first evaluated the phosphorylation status of ATM and its substrate H2AX in Pa-4 cells
treated with DFO for different time periods. ATM activation was assessed by ATM
phosphorylation at Ser-1981 during the course of DFO-treatment (shown in Dr. C.
Clavijo’s dissertation). Moreover, it is likely that the manipulation of PKCδ level by use
of sh-PKCδ or PKCδ function by expression of PKCδKD-EGFP evoked a modest
modulation on the fold-of-activation or the onset of maximal induction, as noted by the
quantitative analysis of the relative pSer-1981-ATM level in each sample.
In addition, H2AX, a hallmark of genotoxicity, was phosphorylated at Ser-139 (γ-
H2AX) and, γ-H2AX signals accumulated with increasing periods of DFO-exposure,
which was further confirmed by confocal microscopy analyses (Figure 4.3; courtesy from
Dr. C. Carlos’s dissertation). There was an apparent localization of both γ-H2AX and
PKCδ in the nucleus of DFO-treated cells. I next examined whether PKCδ is involved in
DFO-elicited γ-H2AX activation. The sustained high level of γ-H2AX signals from 16 to
48 h post-DFO-treatment in both Pa-4/sh-PKCδ and Pa-4/PKCδKD-EGFP cells were
lacking, suggesting that PKCδ is not necessary for the initial γ-H2AX phosphorylation,
but may be essential for the maintenance of high level γ-H2AX. Taken together, I
62
surmised that PKCδ contributes, at least in part, to the pathways leading to the activation
of DNA damage response in DFO-treated cells.

γ-H2AX
DAPI
PKCδ
Merged
a b cd
e f gh
i j k l
m n op
DFO (50 μM): 0 h 24 h 32 h 48 h
γ-H2AX
DAPI
PKCδ
Merged
a b cd
e f gh
i j k l
m n op
DFO (50 μM): 0 h 24 h 32 h 48 h

Figure 4.3 γ-H2AX accumulates in nuclei of Pa-4 cells in response to DFO-treatment. Pa-4 cells were
seeded in 16-well cell culture chambers, treated with 50 μM DFO and harvested at the indicated time points.
Cells were processed for confocal microscopy analyses as described in Materials and Methods.
Immunostaining of endogenous γ-H2AX (green) and PKCδ (red), along with DAPI (blue) staining for
identification of nuclei are shown. Images were acquired using a Nikon epifluorescence inverted
microscope and processed using the Metamorph Imaging, Corel Photo-Paint and LSM 5 Image Browser
Software.

Phosphorylation of both p53 at Ser-15 and Chk1 at Ser-345 has also been shown to be
involved in DNA damage signaling (101, 111, 162). To determine if different PKCδ
contexts affect DFO-induced activation of molecule(s) other than ATM, I assessed
phosphorylation profile of p53 at Ser-15 (pSer15-p53) and Chk1 at Ser-345 (pSer345-
Chk1) upon exposure to DFO in parental and engineered Pa-4 cells. As shown in Figure
4.4, exposure to DFO caused a sustained induction of pSer15-p53 and a temporal
63
increase of pSer345-Chk1 signal in all cells tested. Both the durations and signals of
pSer345-Chk1 and pSer15-p53 were somehow altered in both DFO-treated Pa-4/sh-
PKCδ and Pa-4/PKCδKD-EGFP cells. Together, it is likely that DNA damage signaling
pathways other than ATM may also contribute to the DFO-mediated signaling events.

1 2 34 5 6 78 9 10 11 12
Actin
DFO (100 μM): 0 24 36 0 24 36 0 24 36 48 48 48
PKCδKD-EGFP sh-PKCδ Pa-4 WT
Chk1
PKCδ
PKCδKD-EGFP
p53
pSer15-p53
1 11.4 12.6 8.9 1 6.1 5.5 7 1 5.4 5.0 2.9
1 1.9 1.9 1.5 1 1.1 0.9 0.6 1 1.3 1 0.6
pSer345-Chk1
1 6.3 6.0 1.7 1 9.3 2.0 1.8 1 11.2 2.1 0.8
1 2 34 5 6 78 9 10 11 12
Actin
DFO (100 μM): 0 24 36 0 24 36 0 24 36 48 48 48
PKCδKD-EGFP sh-PKCδ Pa-4 WT
Chk1
PKCδ
PKCδKD-EGFP
p53
pSer15-p53
1 11.4 12.6 8.9 1 6.1 5.5 7 1 5.4 5.0 2.9
1 1.9 1.9 1.5 1 1.1 0.9 0.6 1 1.3 1 0.6
pSer345-Chk1
1 6.3 6.0 1.7 1 9.3 2.0 1.8 1 11.2 2.1 0.8

Figure 4.4 DFO-treatment induces p53 and Chk1 activation. Pa-4, Pa-4/sh-PKCδ and Pa-4/PKCδKD-
EGFP cells were treated with 100 μM DFO for the indicated time periods prior to harvesting cells. Equal
amounts of whole cell extracts were used for Western analyses with indicated antibodies, while an anti-
actin antibody was used to show loading control. Relative levels of pSer15-p53 and pSer345-Chk1 are
indicated as italic. One representative Western blot from three independent experiments is shown.

Activation of pro-apoptotic Caspase-3 and pro-survival Akt in response to DFO-
treatment
As Caspase-3 is a key effector component of several apoptotic signaling pathways (12,
124, 157), I next explored the possibility that Caspase-3 also participates in DFO/PKCδ-
mediated signaling. The characteristic cleavage of Caspase-3, a hallmark of Caspase-3
activation, appeared to occur around at 32 h post-DFO-treatment (shown in Dr. C.
64
Clavijo’s dissertation), correlating with PKCδ cleavage and γ-H2AX activation. Since
PKCδ has been implicated in a positive activation loop with Caspase-3 in other apoptotic
contexts (14), I also evaluated the effect of PKCδ expression on DFO-mediated Caspase-
3 activation. As expected, the steady-state levels of cleaved fragment of Caspase-3
(Caspase-3-CF) in Pa-4/sh-PKCδ and Pa-4/PKCδKD-EGFP cells were barely detectable
by Western analyses during the course of DFO-treatment, suggesting that PKCδ is
necessary for DFO-induced Caspase-3 activation.
As shown in Figure 4.5, I further evaluated the correlation between the Caspase-3 and
DFO-induced activation of PKCδ and DNA damage signaling in Caspase-3-deficienct
MCF7/Neo and MCF7/Caspase-3 cells (75). It is worth noting that DFO only induced
the appearance of PKCδ-CF in MCF7/Caspase-3, but not MCF7/Neo cells, suggesting
that Caspase-3 also participates in DFO-activated proteolytic cleavage of PKCδ. While
DFO-triggered phosphorylation of p53 at Ser-15 was not affected by the absence of
Caspase-3 in MCF7 cells, the induction of Chk1 phosphorylation at Ser-345 was notably
reduced in MCF7/Caspase-3 cells (Figure 4.5). Together with data shown in Figure 4.4,
these latter data led us to envisage the possibility that DFO-treatment activates p53 and
Chk1 utilizing PKCδ/Caspase3-independent pathways.

65

12 3 4 5 6 7 8
Actin
100 μM DFO 0 24 36 48 0 24 36 48 Time (h)
MCF7/Neo MCF7/Caspase-3
Caspase-3
PKCδ
pSer345-Chk1
Chk1
PKCδ-CF
1 1.3 1.1 1.8 1 1 4.7 4.1
Caspase-3-CF
11 1.7 1.4
pSer15-p53
p53
1 16.8 60.5 50.3 1 10.2 61.1 65.9
1 3.4 4.1 4.2 1 0.8 1.5 1.6
12 3 4 5 6 7 8
Actin
100 μM DFO 0 24 36 48 0 24 36 48 Time (h)
MCF7/Neo MCF7/Caspase-3
Caspase-3
PKCδ
pSer345-Chk1
Chk1
PKCδ-CF
1 1.3 1.1 1.8 1 1 4.7 4.1
Caspase-3-CF
11 1.7 1.4
pSer15-p53
p53
1 16.8 60.5 50.3 1 10.2 61.1 65.9
1 3.4 4.1 4.2 1 0.8 1.5 1.6

Figure 4.5 DFO induces activation of p53 and Chk1 in a Caspase-3-independent manner. MCF7/Neo
and MCF7/Caspase-3 cells were treated with 100 μM DFO for the indicated time periods prior to
harvesting. Equal amounts of whole cell extracts were used for Western analyses with indicated antibodies,
while an anti-actin antibody was used to show loading control. Relative levels of PKCδ-CF, Caspase-3-CF,
pSer15-p53 and pSer345-Chk1 are indicated as italic.

Next, I evaluated the effect of PKCδ on Akt activation profile during the DFO-
induced responses. As shown in Figure 4.6, the Ser-473-Akt was modestly
phosphorylated by DFO-treatment, while no marked difference in the profile of DFO-
induced Ser-473-Akt phosphorylation was noticed throughout the time course I examined
using the three cell lines. In some experiments, I observed pSer-473-Akt was induced at
around 8 h post-treatment with DFO, followed by a downregulation of pSer-473-Akt
signal until around 40 ~ 48 h post-treatment (Figure 4.6, upper panel) or with a higher
basal level of pSer-473-Akt signal in Pa-4 cells. In general, the DFO-stimulated Thr-
308-Akt phosphorylation signals were more noticeable, also a delayed event, in Pa-4, Pa-
4/sh-PKCδ and Pa-4/PKCδKD-EGFP cells, despite their distinct PKCδ context (Figure
66
4.6). By comparing their DFO-induced pSer-473-Akt and pThr-308-Akt profiles (Figure
4.6), PKCδ appears to play a fine-tuning role in modulation of DFO-triggered Akt
phosphorylation.

50 μM DFO:
Time (h): 0 8 16 24 32 40 48
Pa-4
Pa-4/sh-PKCδ
Pa-4/PKCδKD
pSer473-Akt
pThr308-Akt
Akt
Actin
1 2.5 1.3 0.6 0.9 1.6 2.4
1 1.2 1.3 1.6 2.1 7.8 5.5
Actin
pThr308-Akt
pSer473-Akt
Akt
1 3.4 2.9 1 1 1 1.3
1 0.8 1.1 1.2 3.8 10.1 5.4
Akt
Actin
1 1 1.3 1.9 1.7 3.1 2.7
1 2.2 1.8 1.6 1 0.8 0.7
pThr308-Akt
pSer473-Akt
50 μM DFO:
Time (h): 0 8 16 24 32 40 48
Pa-4
Pa-4/sh-PKCδ
Pa-4/PKCδKD
pSer473-Akt
pThr308-Akt
Akt
Actin
1 2.5 1.3 0.6 0.9 1.6 2.4
1 1.2 1.3 1.6 2.1 7.8 5.5
Actin
pThr308-Akt
pSer473-Akt
Akt
1 3.4 2.9 1 1 1 1.3
1 0.8 1.1 1.2 3.8 10.1 5.4
Akt
Actin
1 1 1.3 1.9 1.7 3.1 2.7
1 2.2 1.8 1.6 1 0.8 0.7
pThr308-Akt
pSer473-Akt

Figure 4.6 Attenuation of DFO-induced Akt phosphorylations at Ser-473 by PKCδ. Cells were treated
as indicated time periods prior to harvesting cells. Equal amounts of whole cell extracts were used for
Western analyses with anti-phospho-Ser-473-Akt, anti-phospho-Thr-308-Akt antibodies or an anti-actin
antibody to assess equal loading. Relative levels of pSer473-Akt and pThr308-Akt are indicated as italic.

Lastly, I examined whether I could extend my observations that PKCδKD-EGFP
attenuates the DFO-induced proteolytic activation of PKCδ and Caspase-3 to hypoxic
exposure. As shown in Figure 4.7, it appeared that PKCδKD-EGFP also modulated the
67
generation of PKCδ-CF and Caspase-3-CF upon prolonged exposure of 1% O
2
(Figure
4.7, lower panel), supporting the notion that PKCδ could play a vital role in modulating
the cellular response to adverse conditions, such as exposure to DFO or 1% O
2
. However,
1% O
2
-evoked PKCδ-CF began to appear in a significant level, 8 h earlier in Pa-
4/PKCδKD-EGFP cells than that observed in Pa-4 cells (Figure 4.7). The exact reason
underlying this phenomenon is still unclear.

1% O
2
Pa-4/PKCδKD
Time (h): 0 8 16 24 32 40 48
Pa-4
PKCδ-CF
PKCδ
Caspase-3
Caspase-3-CF
Tubulin
1 1 0.9 1.1 12.6 41.5 42.7
1 1.6 2 5.8 33.8 29.1 12.9
PKCδ-CF
Caspase-3
Caspase-3-CF
Tubulin
PKCδ
1 0.9 1.1 1.4 2.2 2.6 2.2
1 0.8 1.2 3.1 11.4 10.2 5.5
1% O
2
Pa-4/PKCδKD
Time (h): 0 8 16 24 32 40 48
Pa-4
PKCδ-CF
PKCδ
Caspase-3
Caspase-3-CF
Tubulin
1 1 0.9 1.1 12.6 41.5 42.7
1 1.6 2 5.8 33.8 29.1 12.9
PKCδ-CF
Caspase-3
Caspase-3-CF
Tubulin
PKCδ
1 0.9 1.1 1.4 2.2 2.6 2.2
1 0.8 1.2 3.1 11.4 10.2 5.5
Pa-4/PKCδKD
Time (h): 0 8 16 24 32 40 48
Pa-4
PKCδ-CF PKCδ-CF
PKCδ PKCδ
Caspase-3 Caspase-3
Caspase-3-CF Caspase-3-CF
Tubulin Tubulin
1 1 0.9 1.1 12.6 41.5 42.7 1 1 0.9 1.1 12.6 41.5 42.7
1 1.6 2 5.8 33.8 29.1 12.9 1 1.6 2 5.8 33.8 29.1 12.9
PKCδ-CF
Caspase-3
Caspase-3-CF
Tubulin
PKCδ
1 0.9 1.1 1.4 2.2 2.6 2.2
1 0.8 1.2 3.1 11.4 10.2 5.5

Figure 4.7 PKCδKD downregulates hypoxia-induced PKCδ and Caspase-3 activation. Pa-4 and Pa-
4/PKCδKD-EGFP cells were treated with 1% O
2
for the indicated time periods prior to harvesting cells.
Equal amounts of whole cell extracts were used for Western analyses with an anti-PKCδ or an anti-
Caspase-3 antibody, while an anti-tubulin antibody was utilized to assess equal loading. Relative levels of
PKCδ-CF and Caspase-3-CF in Pa-4 and Pa-4/PKCδKD-EGFP cells are indicated as italic. One
representative Western blot from three independent experiments is shown.

68
The involvement of PKCδ in DFO-induced cellular responses in Pa-4 cells
In order to further evaluate the contribution of PKCδ to DFO-mediated signaling
crosstalk and its biological consequence, these parental and engineered Pa-4 cells were
assayed for their sensitivity towards DFO-mediated cell growth inhibition by MTT assays.
As shown in Figure 4.8A, both Pa-4 cells with decreased expression of endogenous
PKCδ and Pa-4 cells expressing PKCδKD-EGFP were less sensitive to DFO-treatment
compared with that for the parental Pa-4 cells, the pool of transduced cells with
PKCδWT-EGFP, or cells transfected with the corresponding empty vector. These data
support our notion that PKCδ plays an antagonistic role by suppressing cell survival in
response to DFO. The lack of further enhancement in DFO-elicited cell growth
inhibition by exogenous PKCδWT-EGFP suggests that the endogenous PKCδ level in Pa-
4 cells may be sufficient for conveying the signaling elicited by DFO or PKCδ-
independent pathway may counter the effect of enhanced PKCδ/Caspase-3 pathway.
Since DFO-treatment is known to arrest Pa-4 cells in S-phase and induce apoptosis (116),
I also determined the role of PKCδ in DFO-elicited cell cycle dysregulation. Neither sh-
PKCδ nor exogenous PKCδKD-EGFP alone conveyed a marked effect on modulating cell
cycle progression prior to DFO-treatment. In Figure 4.8B, except for 24 h post-DFO-
exposure in Pa-4/PKCδKD-EGFP cells, significant increases in sub-G
1
cell population
were observed in DFO-treated cells compared to vehicle-treated cells in all three cell
lines, respectively. Moreover, I also observed significant decreases in sub-G
1
cell
population for both Pa-4/sh-PKCδ and Pa-4/PKCδKD-EGFP cells, as compared to that in
69
Pa-4 cells at both 24 h and 48 h post-DFO-treatment, respectively (Figure 4.8B).
Together, I concluded that PKCδ enhances cytotoxic effect of DFO in Pa-4 cells.

AB
0
20
40
60
80
100
120
% Cell Viability
DFO 0 100 (μM)
*
*
Pa-4
Pa-4/sh-PKCδ
Pa-4/PKCδWT
Pa-4/PKCδKD
0
20
40
60
80
100
120
% Cell Viability
DFO 0 100 (μM)
*
*
Pa-4
Pa-4/sh-PKCδ
Pa-4/PKCδWT
Pa-4/PKCδKD
100 μM DFO 02448
Pa-4 Pa-4/sh-PKCδ Pa-4/PKCδKD
(h)
0
10
20
30
40
50
60
70
80
Sub-G
1
cell population (%)
0 24 48 0 24 48
∗∗ ∗∗ ∗∗
∗∗ ∗∗ ∗∗ ∗∗ ∗∗
∗∗
∗∗ ∗
100 μM DFO 02448 02448
Pa-4 Pa-4/sh-PKCδ Pa-4/PKCδKD
(h)
0
10
20
30
40
50
60
70
80
Sub-G
1
cell population (%)
02448 0 24 48 0 24 48 02448
∗∗ ∗∗ ∗∗
∗∗ ∗∗ ∗∗ ∗∗ ∗∗ ∗∗ ∗∗ ∗∗ ∗∗ ∗∗
∗∗
∗∗ ∗
∗∗
∗∗ ∗

Figure 4.8 PKCδ is essential for DFO-mediated cytotoxicity. (A) sh-PKCδ or PKCδKD increases Pa-4
cell survival towards DFO-induced cell death. MTT assays were performed as described in Materials and
Methods. Results from three independent assays are shown. Error bars correspond to standard deviation;
*

denotes significant difference (p < 0.01) between corresponding cells. (B) Attenuation of PKCδ by sh-
PKCδ or PKCδKD protects Pa-4 cells against DFO-elicited apoptosis as reflected by a decrease in sub-G
1
population. The populations of sub-G
1
cells were normalized and expressed as percentage compared with
that of vehicle-treated Pa-4, Pa-4/sh-PKCδ and Pa-4/PKCδKD-EGFP cells, respectively. Data are presented
as mean ± S.D. from 3 separate experiments performed in triplicate.
**
denotes p < 0.01 and
*
denotes p <
0.05.

In summary, my data provide evidence for signaling crosstalk in DFO-treated Pa-4
cells and suggest that both PKCδ-dependent and -independent pathways functionally
cooperate to mediate salivary adaptive responses against DFO, as depicted in Figure 4.9.
I postulate that the DFO-mediated PKCδ activation is a part of a central regulatory circuit,
which directs an integrated balance between Caspase-3-/Akt-activated signaling and
DNA damage response.

70

DFO, Hypoxia
Caspase-3
?
Salivary Adaptive Responses
ATM/H2AX Chk1/p53
DNA Repair
Cell Cycle Regulation
Cell Death Cell Survival
PKCδ
Activation
DNA Damage
Response
Akt
Activation
?
DFO, Hypoxia
Caspase-3
?
Salivary Adaptive Responses
ATM/H2AX Chk1/p53
DNA Repair
Cell Cycle Regulation
Cell Death Cell Survival
PKCδ
Activation
DNA Damage
Response
Akt
Activation
?

Figure 4.9 Proposed model for DFO-induced activation of PKCδ, DNA damage responses, and Akt to
mediate salivary adaptive responses. First, DFO-treatment induces time-dependent nuclear translocation
and proteolytic cleavage of PKCδ and PKCδ-dependent Caspase-3 activation. Second, PKCδ activation
renders a sustained DFO-elicited γ-H2AX activation, a hallmark of DNA damage response. Third, DFO-
exposure also induces Chk1 and p53 phosphorylation in a PKCδ-independent manner. Furthermore, PKCδ
plays a fine-tuning role in modulation of DFO-induced Akt phosphorylation. Both PKCδ-dependent and -
independent pathways functionally cooperate to mediate salivary adaptive responses against DFO. DFO-
mediated PKCδ activation is a part of a central regulatory circuit, which directs an integrated balance
among pro-apoptotic/Caspase-3, pro-survival/Akt, and DNA damage-induced DNA repair/cell cycle
regulation signalings.

Discussion
Several lines of evidence provided herein support the notion that PKCδ activation is
involved in DFO-elicited signaling events. First, DFO-treatment induces time-dependent
nuclear translocation and proteolytic cleavage of PKCδ and PKCδ-dependent Caspase-3
activation (Figures 4.8A and B). Second, PKCδ partakes of the high level of sustained γ-
H2AX signal, a hallmark of DNA damage response, upon exposure to DFO (Figure 4.3).
71
Third, the kinase-dead PKCδ-EGFP is capable of attenuating the DFO-/hypoxia-induced
activation of endogenous PKCδ and Caspase-3 (Figure 4.7). In addition, PKCδKD-EGFP
fails to be translocated to the nucleus of DFO-treated cells (Figure 4.2) even though it
harbors PKCδ bipartite nuclear localization signal reported by DeVries et al., (37),
consistent with the proposed dominant-negative role of PKCδKD (93). Lastly, sh-PKCδ
or PKCδKD-EGFP conveys partial resistance to DFO-elicited cytotoxicity (Figures 4.8A
and B). Together, I conclude that PKCδ is an important component of the activated
signaling pathway, consequent to DFO-treatment. Furthermore, DFO-treatment also
induces Chk1 and p53 phosphorylation in a PKCδ-independent manner (Figures 4.4 and
4.5).
Although the hypoxia-mimetic effect of DFO has been well documented, no direct
link between DFO-mediated genotoxic stress and activation of a specific PKC isoform
has been described to date. My present work suggests that PKCδ is involved in
mediating DFO-induced genotoxic responses. First, I showed that DFO-induced
accumulation of DNA damage marker γ-H2AX is modulated by expression of sh-PKCδ
or Lys376Arg-mutated PKCδ-EGFP. Current thinking predicts that ATM is activated by
DNA double-stranded breaks (DSBs) and recruits more ATM molecules as well as other
repair proteins and sensors, such as MRN, 53BP1 and MDC1, to the DNA lesion (44).
Moreover, γ-H2AX can also be activated by DNA-PK and ATR, in addition to ATM.
While my present work suggests that PKCδ activation renders a sustained DFO-elicited
γ-H2AX activation (Figure 4.2), the exact molecular mechanism underlying this
observation is still elusive. It is possible that H2AX serves as a direct kinase substrate of
72
nucleus-translocated PKCδ, given that PKCδ is translocated to the nucleus during DFO-
treatment (Figure 4.2). Second, PKCδ is cleaved by a Caspase-3-dependent mechanism
to a 40-kDa PKCδ-CF, analogous to PKCδ activation in cells treated with genotoxic
DNA-damaging agents (37, 42), in DFO-treated cells (Figure 4.5). As a relatively higher
level of Caspase-3-CF was observed in Pa-4/PKCδWT-EGFP cells treated with DFO for
48 h, a feed forward pathway for PKCδ and Caspase-3 reciprocal activation (Figure 4.9)
is likely to exist in response to DFO-treatment, consistent with previous reports in other
apoptotic systems (79, 81, 89, 131). Moreover, overexpression of the PKCδ-CF has been
shown to induce chromatin condensation and DNA fragmentation, supporting a role for
PKCδ-CF in mediating cell response to genotoxic stress (47). Lastly, studies by others
have also shown that PKCδ interacts with p53 DINP1 and p53, leading to p53
phosphorylation upon exposure to genotoxicity (168) and that PKCδ mediates the
cytotoxic effect of arsenate-induced cell death (91). In addition, cells derived from
PKCδ
-/-
mice were shown to be defective in mitochondria-dependent apoptosis (66, 88).
These findings collectively support my proposition of pro-genotoxic role of PKCδ in
DFO-dependent signaling in salivary epithelial cells.
Previous studies have reported that treatment of cells with DNA damaging agents is
associated with translocation of PKCδ to the nucleus (37, 170). However, these studies
also showed that etoposide elicits PKCδ proteolytic cleavage and p53 phosphorylation
via a rapid induction in salivary cells in vitro and in vivo (37, 66). By contrast, my
present results suggest that the induction of PKCδ and Caspase-3 proteolytic cleavage
and p53 phosphorylation by DFO-treatment occurs in a delayed manner. One possible
73
explanation for this time-dependent action in DFO-treated cells is that DFO, a membrane
impermeant iron-chelator, is taken up primarily by fluid-phase endocytosis, travels
through various endosomes to late endosomes, and eventually accumulates in the
lysosomal compartment (40, 83). It is plausible to assume that the change in the extent of
DFO-chelatable irons in extracellular milieu, cytoplasmic and other subcellular
compartments and membrane destabilization of these organelle result in the activation of
various signaling pathways, including PKCδ, Caspase-3, ATM, p53 and Chk1, in a rather
delayed manner. In addition, I have recently reported that hypoxia- or DFO-treatment
leads to a transient activation of the SUMOylation process, and that overexpression of
SUMO-1 protects salivary Pa-4 cells against hypoxic injury (116). It is conceivable that
SUMOylation alters and/or delays the signaling event(s) activated by DFO-treatment.
While PKCδ plays a role in mediating cell response to DFO-treatment, as shown
herein, signaling pathway other than PKCδ also contributes in part to the effect by DFO
in salivary cells. It appears that PKCδ-independent signaling pathway is also activated
by DFO-treatment. For example, the activation of both Chk1 and p53 is also known to be
part of genotoxic responses. As shown in Figure 4.4, the DFO-exerted Chk1
phosphorylation at Ser-345 and p53 phosphorylation at Ser-15 were noted in cells lacking
functional PKCδ pathway, such as in Pa-4/sh-PKCδ and Pa-4/PKCδKD-EGFP cells.
DFO-treatment also induced Chk1 and p53 activation without generating PKCδ-CF in
MCF7/Neo cells (Figure 4.5). Consistent with our results is the report by Humphries et
al., who demonstrated the involvement of a signaling pathway other than PKCδ in
governing the steady-state level of p53 and its Ser-15 phosphorylation under genotoxic
74
conditions in cells prepared from PKCδ
-/-
and PKCδ
+/+
mice (66). Previously, Gibson et
al., have reported that hypoxia induces Chk2 activation in a ATM-dependent manner (48).
My results unequivocally showed that DFO not only activates ATM (and presumably its
downstream Chk2) but also induces PKCδ-independent Chk1 stimulation. Although
Chk1 functionally shares many substrates with Chk2, I suggest that Chk1 activation may
represent an additional pathway to ascertain a proper adaptive response to DFO-induced
genotoxic stress.
A complete picture of the signaling pathways and functions by PKCδ in DFO-treated
cells has yet to emerge. The presented studies expanded our understanding of PKCδ
biology, as well as perhaps the role of PKCδ in regulating DFO-, and perhaps hypoxia-,
induced adaptive response. Taken together, my data suggest the role for the crosstalk of
PKCδ and DNA damage response signaling pathways in DFO-mediated genotoxicity and
also demonstrate that both PKCδ-dependent and -independent pathways functionally
cooperate to mediate salivary adaptive responses against DFO (Figure 4.9).

75
CHAPTER 5
EMERGING ROLE FOR THE PROTEIN KINASE Cδ-DEPENDENT
SIGNALING PATHWAY IN AUTOPHAGY INDUCED BY DESFERROXAMINE,
A HYPOXIA-MIMETIC AGENT
Introduction
Autophagy, self-cannibalization to degrade cells’ own constituents including their
organelles, is induced by certain environmental cues, such as starvation, heat shock and
hypoxia (90) and references therein. In mammals, autophagy exhibits marked
associations with neurodegenerative diseases, cancer, cardiomyopathies, aging, type II
programmed cell death, bacterial invasion, MHC class II antigen presentation and cellular
maintenance (90). The Autophagy-related genes, Atg/Apg/Aut/Cvt genes, have been
isolated and characterized in yeast and mammals (90).
There are two ubiquitination-like conjugation systems required for autophagosome
formation (90). One system mediates the conjugation of Atg12-Atg5, and the other
system produces covalent linkage between Atg8 and phosphatidylethanolamine (PE).
Atg12 is first activated by Atg7, followed by transfer to Atg10, and finally covalently
attached to Atg5, a process requiring ATP (90). The Atg12-Atg5 conjugates localize to
autophagosome precursors, dissociate just before or after completion of autophagic
vacuole formation, and are essential for elongation of the isolation membrane (90).
Microtubule-associated protein 1 light chain 3 (Map1-LC3; LC3) is the mammalian
homologue of yeast Atg8 (73). The carboxyl terminal region of LC3 is cleaved by Atg4,
generating a soluble form known as LC3-I and exposing carboxyl terminal glycines
essential for further reactions (74). LC3-I, in turn, is modified to a membrane-bound
76
form of LC3-II (a LC3-phospholipid conjugate) by Atg7 and Atg3, El- and E2-like
enzymes (90), and localizes to autophagosomes and autolysosomes (73). Thus, the
relative amount of LC3-I-to-LC3-II conversion and the changes in LC3-II level via
degradation in those mammalian cells are a useful marker for the formation of
autophagosomes and autolysosomes, respectively. Importantly, LC3-II lipidation
depends on the Atg12-Atg5 conjugates, since LC3-II form is not observed at all in either
Atg5
-/-
cells or those cells engineered to express the conjugation-defective Atg5 mutant
(ATG5K130R) (59, 109).
Signaling pathways that regulate autophagy are extremely complex, as numerous
feedforward and feedback loops and cross-talks with many other signaling networks are
involved in coordinating cellular autophagy, proliferation and apoptosis. One of the key
regulators of autophagy is phosphoinositide 3-kinase (PI3K). Mammalian cells contain
three distinct types of PI3K, based on their substrate specificities and subunit
organizations (158). Class I PI3Ks produce PtdIns(3,4,5)P3 in vivo. Class II PI3Ks
appear to produce PtdIns(3,4)P2 and PtdIns3P. Finally, the class III PI3K, Vps34, only
produces PtdIns3P. While the class I PI3K activates the mammalian target of rapamycin
(mTOR), the interruption of mTOR-dependent signaling by rapamycin stimulates
autophagy in many cell types (134). By contrast, the pharmacological inhibitor, 3-
methyladenine (3-MA), which targets the class III PI3K/Vps34 inhibits the nucleation of
autophagosome vesicles (123).
Although autophagy has been characterized in many contexts, the signaling pathway
that activates autophagy in response to hypoxic stress has not yet been studied
extensively. Oxygen deprivation by chronic hypoxia is known to induce autophagy. For
77
example, Beclin-1 levels in the cortex and striatum are increased in cerebral ischemia (1).
The early changes of Beclin-1 levels in the penumbra occurred at 6 h, peaked at 24 h and
lasted for at least 2 days in the neuronal cells (128). Beclin-1 (also called Atg6) is a
phylogenetically conserved protein that is essential for the initiation of autophagy,
presumably via its interaction with the class III PI3K, Vps34 (90, 175). Originally,
human Beclin-1 is identified as an interactor of bcl-2 (95). Beclin-1 reportedly possesses
a so-called bcl-2 homology region-3 (BH3) domain (amino acid 114-123) that mediates
the interaction with bcl-2 and other bcl-2 homologues, such as bcl-X
L
and Mcl-1 (104).
It has recently become clear that overexpression of Beclin-1 or depletion of bcl-2
stimulates autophagy (121). Herein, I demonstrate a novel pathway by which a hypoxia-
mimetic, desferroxamine (DFO), utilizes a rapid activation of PKCδ signaling pathway to
release Beclin-1 from bcl-2 leading to autophagy induction.

Results
Increased autophagosome accumulation during DFO stress
I previously reported that treatment of cells with desferroxamine (DFO), a hypoxia-
mimetic agent, confers a crosstalk between the activation of protein kinase Cδ (PKC δ)
and apoptotic Caspase-3 pathways (25). As signaling pathways that regulate cell
proliferation and apoptosis are frequently associated with the regulation of autophagy, I
hypothesize that DFO-treatment induces autophagy. To test this hypothesis, GFP-LC3
fluorescence was used to monitor autophagosome accumulation in Pa-4 cells that had
been transiently transfected with an expressing vector for GFP-LC3 (73). The formation
78
of GFP-LC3-labeled structures, GFP-LC3 puncta, representing autophagosomes, was
induced in cells exposed to DFO (Figures 5.1A, upper panels a-f and 5.1B, indicated by
arrows). Morphometric analyses of the GFP fluorescence images that GFP-LC3 puncta
were transiently induced at 2 h post-treatment and tapered off after reaching maximal
induction at 4 h post-treatment (Figure 5.1). Since chloroquine (CQ) has been shown to
inhibit the fusion of lysosomes with autophagosomes and to amplify GFP-LC3 puncta
signal (80), treatment with CQ for 4 h was used to confirm the expected increase in GFP-
LC3 puncta accumulation (Figures 5.1A, middle panels, and 5.1C). In addition,
transfection with the parental vector GFP-N1 followed by DFO-exposure was used to
validate the specificity of our assays in response to DFO (Figure 5.1A, lower panels).

79
A
Pa-4
GFP-N1
GFP-LC3
(h) 50 μM DFO 24 6 12 24 0
10 μM CQ (h) 2 4 6 12 24 0
GFP-LC3
ad bf e c
Pa-4
GFP-N1
GFP-LC3
(h) 50 μM DFO 24 6 12 24 0 2 4 6 12 24 0
10 μM CQ (h) 2 4 6 12 24 0
GFP-LC3
ad bf e c
B
c
a
c
a
0
10
20
30
40
50
60
70
80
90
100
GFP-LC3+50 μM DFO
GFP-LC3+10 μM CQ
2 0 4 6 12 24 (h)
GFP-LC3 Puncta/GFP
Positive Cells (%)
0
10
20
30
40
50
60
70
80
90
100
0
10
20
30
40
50
60
70
80
90
100
GFP-LC3+50 μM DFO
GFP-LC3+10 μM CQ
GFP-LC3+50 μM DFO
GFP-LC3+10 μM CQ
2 0 4 6 12 24 (h)
GFP-LC3 Puncta/GFP
Positive Cells (%)
C

Figure 5.1 Treatment with DFO, a hypoxia-mimetic agent, induces autophagic process. (A) DFO-
treatment induces GFP-LC3 puncta accumulation in Pa-4 cells. Pa-4 cells were transfected with GFP-LC3
or GFP-N1 and treated with DFO (50 μM) or chloroquine (CQ, 10 μM) for indicated time periods. These
cells were then examined by confocal microscopy. Autophagosome formation was defined by the
accumulation of GFP-LC3 puncta (indicated by arrows). (B) Enlarged view of DFO-induced GFP-LC3
puncta in Pa-4 cells. Selected enlarged views of the boxed area in panels a and c of (A) are shown. (C)
Quantification of GFP-LC3 puncta in Pa-4 cells. The GFP-LC3 puncta were counted individually by three
investigators blind to experimental conditions. The ratio of GFP-LC3 puncta over GFP-positive cells was
shown as a percentage.

80
To avoid non-specific accumulation of GFP-LC3 in non-autophagic cells, as
described for some report on transient transfection (80), MEF/GFP-LC3 (MEF cells
stably transfected with GFP-LC3) were used to examine the ability of DFO to induce
GFP-LC3 puncta. As shown in Figure 5.2A (panel a), GFP-LC3 signals were equally
distributed throughout the cells prior to DFO-treatment. Upon exposure to DFO, green
signals redistributed from nuclei and aggregated in peri-nuclear areas where the intensity
of signals reached maximum at 4 h post-treatment (Figure 5.2B), followed by a gradual
decrease in signals for GFP-LC3 puncta at 6 h post-treatment (Figure 5.2A, panels a-e).

B A
DAPI
Merge
GFP
MEF/GFP-LC3
100 μM DFO
EBSS 4 h 1 h2 h 4 h6 h 0 h
ad cef b
DAPI
Merge
GFP
MEF/GFP-LC3
100 μM DFO
EBSS 4 h 1 h2 h 4 h6 h 0 h EBSS 4 h 1 h2 h 4 h6 h 0 h
ad cef b
a
d
a
d

Figure 5.2 DFO induces autophagosome accumulation in MEF/GFP-LC3 cells. (A) DFO-treatment
induces GFP-LC3 puncta accumulation in MEF/GFP-LC3 cells. MEF/GFP-LC3 cells were treated with
DFO (100 μM) for indicated time periods and subjected to confocal microscopy analyses. Green
fluorescence represents GFP-LC3 signals, while blue color represents nuclear staining (DAPI staining).
Images shown are representatives from one out of three independent experiments. Autophagosome
formation was indicated by accumulation of GFP-LC3 puncta (indicated by arrows). (B) Enlarged view of
DFO-induced GFP-LC3 puncta in MEF/GFP-LC3 cells. Selected enlarged views of the boxed area in
panels a and d of (A) are shown.

To further characterize DFO-induced autophagic process, the amount of membrane-
bound LC3-II (the phosphatidylethanolamine-conjugated form (74)) increased in MEF
81
cells exposed to DFO (Figure 5.3A, left panel), a hallmark of autophagy, when subjected
to Western analyses. Since endogenous LC3 in MEF cells is mainly present as the
membrane-bound LC3-II forms (Figure 5.3A, left panel), I hence monitored the time
course of the changes in LC3-II level as a means to assess autophagy in the remaining
studies. It has been shown by others that LC3-II level is increased and then decreased
inside autolysosomes, while LC3 on the outer membrane is deconjugated by Atg4 and
returns to cytosol (80). As Atg12-Atg5 conjugates have an E3-like activity to enhance
LC3 lipidation (59), no LC3 processing was observed in Atg5
-/-
cells during DFO-stress
(Figure 5.3A, right panel). I also observed a reproducible increase in LC3-I level (Figure
5.3) in response to DFO-treatment. Although it is possible that DFO induces LC3
expression, the exact nature underlying this phenomenon is still unclear. Next, I pre-
treated cells with lysosomal protease inhibitors, E64d and pepstatin A, for 2 h prior to
DFO-exposure to ascertain that DFO stimulates autophagic flux (80). As expected, the
pretreatment of cells with E64d and pepstatin A prevented the decrease in LC3-II level
and rendered a sustained level of LC3-II during DFO-treatment (Figure 5.3B, top panel,
lanes 2, 4, 6 and 8 versus lanes 1, 3, 5 and 7). Lastly, the p62/SQSTM1 serves as a link
between LC3 and ubiquitinated substances by incorporating itself into completed
autophagosome and then degraded by autolysosomes (80). The treatment of cells with
the lysosomal inhibitors also attenuated the rate of p62 degradation (Figure 5.3B, second
panel). Taken together, I conclude that autophagic process is stimulated in Pa-4 and
MEF cells upon the exposure to DFO.

82
B
A
1 1.2 1 0.8
Actin
MEF MEF/Atg5
-/-
02 4 60 2 4 6(h) 100 μM DFO
LC3-II
LC3-I
(Long Exp)
LC3-II (Short Exp)
1 1.2 1 0.8
Actin
MEF MEF/Atg5
-/-
02 4 60 2 4 6(h) 100 μM DFO
LC3-II
LC3-I
(Long Exp)
LC3-II (Short Exp)
MEF
200 μM DFO (h) 00 2 2 4 4 6 6
E64d+Pepstatin A (h) 02 0 2 0 2 0 2
Actin
LC3-II
12.1 1.6 2 1.3 2 1.3 1.7
p62
1 1 1.3 1.7 1.5 1.7 1.3 2
MEF
200 μM DFO (h) 00 2 2 4 4 6 6
E64d+Pepstatin A (h) 02 0 2 0 2 0 2
Actin
LC3-II
12.1 1.6 2 1.3 2 1.3 1.7
p62
1 1 1.3 1.7 1.5 1.7 1.3 2

Figure 5.3 Validation of DFO-induced autophagic process. (A) DFO induces the conversion of LC3-I to
LC3-II in wild-type MEF, but not in MEF/Atg5
-/-
, cells. Cells were treated with DFO (100 μM) for
indicated time periods before harvesting cells. Equal amounts of whole cell lysates were subjected to
Western analyses with anti-LC3 and anti-actin antibodies, respectively. Images were visualized with long
exposures (Long exp; upper panel) to show both LC3-I and LC3-II bands while the intensity of each LC3-
II band was also acquired with shorter exposures (Short Exp; middle panel) for quantification purpose.
Relative levels of LC3-II were quantified by Quantity One software (Bio-Rad) and shown as fold induction
(numbers in italic), while non-treatment (control) is designated as 1. An anti-actin antibody was used to
probe actin levels as an internal control for assessing equal loading. (B) DFO-enhanced degradation of
LC3-II and p62 is inhibited by a combined set of lysosomal protease inhibitors, E64d and pepstatin A.
Cells were pre-treated with a mixture of E64d and pepstatin A (10 μg/ml each), or vehicle for 2 h prior to
the treatment of DFO (200 μM) for indicated time periods. Whole cell lysates were harvested and subjected
to Western analyses with anti-LC3, anti-p62, and anti-actin antibodies, respectively. Relative levels of
specific proteins of interest were shown as fold increase (numbers in italic), while the level for non-
treatment (control) sample is set as 1. Long Exp: longer exposures; Short Exp: shorter exposures.

A similar observation was also made in Pa-4 cells exposed to 1% oxygen for 2 to 6 h
(Figure 5.4), confirming that hypoxic stress stimulates a rapid induction of autophagy.

83

Pa-4
1% O
2
(h)
E64d+Pepstatin A (h)
4
0
4
2
6
0
6
2
LC3-II
LC3-I
(Long Exp)
LC3-II (Short Exp)
Actin
p62
0
0
1
1
0
2
1.4
0.9
2
0
0.7
2.7
2
2
1.2
3
0.3
3.3
1.3
4.7
0.3
3.1
1.2
5.6
Pa-4
1% O
2
(h)
E64d+Pepstatin A (h)
4
0
4
0
4
2
4
2
6
0
6
0
6
2
6
2
LC3-II
LC3-I
(Long Exp)
LC3-II (Short Exp)
Actin
p62
0
0
0
0
1
1
1
1
0
2
0
2
1.4
0.9
1.4
0.9
2
0
2
0
0.7
2.7
0.7
2.7
2
2
2
2
1.2
3
1.2
3
0.3
3.3
0.3
3.3
1.3
4.7
1.3
4.7
0.3
3.1
0.3
3.1
1.2
5.6
1.2
5.6

Figure 5.4 Hypoxia (1% O
2
)-induced degradation of LC3-II and p62 is inhibited by a combined
treatment of lysosomal protease inhibitors, E64d and pepstatin A. Pa-4 cells were pre-treated with a
mixture of E64d and pepstatin A (10 μg/ml each) or vehicle for 2 h prior to exposure of 1% O
2
for
indicated time periods. Whole cell lysates were harvested and analyzed by Western analyses with anti-LC3,
anti-p62, and anti-actin antibodies, respectively. Relative levels of specific protein were shown as fold
increase (numbers in italic) and non-treatment (control) value is designated as 1. Long Exp: longer
exposures; Short Exp: shorter exposures.

PKCδ is required for DFO-, but not EBSS-, induced autophagy
To examine whether PKCδ is involved in DFO-induced autophagic process, I
demonstrated that the time-dependent changes in LC3-II level is almost completely
ameliorated in DFO-treated Pa-4/PKC δKD cells (Figure 5.5A, left panel). The lack of
LC3 processing in DFO-treated Pa-4/PKC δKD cells was in agreement with my previous
observations that K376R mutation of PKC δ, PKC δKD, is a dominant-negative form of
PKC δ (25). In contrast, amino acid deprivations (i.e., EBSS-treatment) induced the LC3
processing in both Pa-4 and Pa-4/PKC δKD cells (Figure 5.5A, right panel). These results
suggest that the autophagy pathway is functional in Pa-4/PKCδKD cells, whereas the
DFO-activated signaling pathway(s) to stimulate autophagy is disturbed by the dominant
negative PKCδKD. Moreover, the changes of LC3-II level were largely inhibited by pre-
84
treating cells with 3-methyladenine (3-MA) (Figure 5.5B), a nucleotide derivative that
inhibits the early stage of autophagosome formation (90).

(h) 01 2 6 0 1 2 6 EBSS
Pa-4 Pa-4/PKCδKD
LC3-II
LC3-I
Actin
1 0.8 0.9 0.5 1 1 0.9 0.6 LC3-II
1 0.7 0.5 0.4 1 0.9 0.6 0.6 LC3-I
(h) 01 2 6 0 1 2 6 EBSS
Pa-4 Pa-4/PKCδKD Pa-4 Pa-4/PKCδKD
LC3-II
LC3-I
LC3-II
LC3-I
Actin Actin
1 0.8 0.9 0.5 1 1 0.9 0.6 LC3-II
1 0.7 0.5 0.4 1 0.9 0.6 0.6 LC3-I
1 0.8 0.9 0.5 1 1 0.9 0.6 LC3-II
1 0.7 0.5 0.4 1 0.9 0.6 0.6 LC3-I
200 μM DFO
Pa-4/PKCδKD Pa-4
(h) 01 2 6 4012 6 4
1 1.4 1.2 1.2 1.1
LC3-II (Short Exp)
1 0.9 2.2 1.7 1.5
Actin
LC3-II
LC3-I
(Long Exp)
200 μM DFO
Pa-4/PKCδKD Pa-4 Pa-4/PKCδKD Pa-4
(h) 01 2 6 4012 6 4(h) 01 2 6 4 01 2 6 4012 6 4 01 2 6 4
1 1.4 1.2 1.2 1.1 1 1.4 1.2 1.2 1.1
LC3-II (Short Exp)
1 0.9 2.2 1.7 1.5 1 0.9 2.2 1.7 1.5
Actin
LC3-II
LC3-I
(Long Exp)
A

B
200 μM DFO (h)
10 mM 3-MA
024 6
222 2
02 4 6
22 2 2 (h)
Pa-4 Pa-4/PKCδKD
LC3-II
Actin
11 1 1 1 1.1 1 1
200 μM DFO (h)
10 mM 3-MA
024 6
222 2
02 4 6
22 2 2 (h)
200 μM DFO (h)
10 mM 3-MA
024 6
222 2
024 6
222 2
02 4 6
22 2 2
02 4 6
22 2 2 (h)
Pa-4 Pa-4/PKCδKD Pa-4 Pa-4/PKCδKD
LC3-II LC3-II
Actin Actin
11 1 1 1 1.1 1 1 11 1 1 1 1.1 1 1

Figure 5.5 PKCδ is essential for DFO-induced autophagy in Pa-4 cells. (A) DFO and EBSS use distinct
signaling mechanisms to induce autophagy. Both Pa-4 and Pa-4/PKCδKD cells were treated with DFO (200
μM, left panels) or EBSS (right panels) for indicated time periods. Whole cell lysates were subjected to
Western analyses with an anti-LC3 or anti-actin antibody. Both LC3-I and LC3-II images were shown with
longer exposures (Long Exp; upper left panel), while the intensity of LC3-II image was also acquired with
shorter exposures (Short Exp; middle left panel) for quantification purposes. Relative levels of LC3-II (left
panel) or LC3-I and LC3-II (right panel) were quantified by Quantity One software (Bio-Rad) and shown
as fold induction (numbers in italic) by setting non-treatment (control) value as 1 after normalizing with
individual actin levels. (B) 3-MA treatment inhibits the conversion of LC3-II in Pa-4 cells. Cells were pre-
treated with 3-MA (10 mM) for 2 h prior to DFO (200 μM)-treatment for indicated time periods. Whole
cell lysates were analyzed by Western analyses with anti-LC3 and anti-actin antibodies, respectively.
Relative levels of LC3-II were shown as numbers in italic.

Next, I sought to address the stimulatory role of PKC δ in DFO-induced autophagy
directly, rather than an off-target of PKC δKD, using both wild-type MEF and MEF/PKC δ
-
/-
cells to assess the LC3 processing in response to DFO stress. The expression profile of
85
PKC δ in wild-type MEF and MEF/PKC δ
-/-
cells was confirmed by Western analyses
(Figure 5.6A). As shown in Figure 5.6B, LC3-II levels peaked and then decreased, in
DFO-exposed wild-type MEFs (top panel, lanes 1 to 4), while LC3-II level remained
unchanged during the course of DFO-treatment in MEF/PKC δ
-/-
cells (top panel, lanes 5
to 8). Notably, more robust changes in LC3-II level (Figure 5.6B) were observed when
the concentration of DFO utilized to induce autophagy was raised from 100 μM (Figure
5.3A) to 200 μM (Figure 5.6B). Hence, a higher concentration of DFO was used in the
majority of experiments this point onward. Moreover, the observed LC3 processing in
wild-type MEFs was abolished by treatment with 3-MA (Figure 5.6B, 3
rd
panel, lanes 1
to 4). Taken together, I conclude that PKC δ is highly likely to be required for DFO to
activate the autophagosome formation/autophagy.

B A
MEF MEF/PKCδ
-/-
LC3-II
Actin
200 μM DFO (h) 0 2 4 6 0 246
1 2.1 3 1. 5 1 1 1 1
200 μM DFO (h) 0 2 4 6 0 246
10 mM 3-MA (h) 2 2 2 2 2 222
1 1 1 1 1 111
LC3-II
Actin
MEF MEF/PKCδ
-/-
LC3-II
Actin
200 μM DFO (h) 02 4 6 0 2 4 6 0 246 0 246
1 2.1 3 1. 5 1 1 1 1
200 μM DFO (h) 0 2 4 6 0 246 200 μM DFO (h) 0 2 4 6 0 246
10 mM 3-MA (h) 2 2 2 2 2 222 10 mM 3-MA (h) 2 2 2 2 2 222
1 1 1 1 1 111 1 1 1 1 1 111
LC3-II
Actin
MEF
PKCδ
-/-
WT
PKCδ
Actin
MEF
PKCδ
-/-
WT
PKCδ
-/-
WT
PKCδ
Actin

Figure 5.6 PKCδ is essential for DFO-induced autophagy in MEF cells. (A) Western blotting analyses
of PKCδ expression level in MEF and MEF/PKCδ
-/-
cell lines. (B) DFO-induced LC3-II conversion
requires PKCδ. Wild-type MEF and MEF/PKCδ
-/-
cells were treated with DFO (200 μM) in the absence
(upper two panels) or presence (lower two panels) of pre-treatment with 3-MA (10 mM, 2 h) for indicated
time periods prior to harvesting cells. Equal amounts of whole cell lysates were subjected to Western
analyses with anti-LC3 and anti-actin antibodies, respectively. Relative LC3-II level was quantitated, after
normalizing against individual actin levels by setting control value as 1.

86
Tyrosine phosphorylation of PKC δ at specific Tyr (Y)-residue(s) is one of the
activation mechanisms in response to different stimuli, such as UV, H
2
O
2
and etoposide
(68). As shown in Figure 5.7A, Y-phosphorylation of PKC δ was detected between 0.5 to
1 h post DFO-treatment. Moreover, when wild-type or EGFP-PKC δ mutated at either
Y64F, Y155F, or Y64/155F was transfected into MEF/PKC δ
-/-
cells followed by DFO-
treatment, both Y-64 and Y-155 appeared to be involved in DFO-induced PKC δ
activation, as seen by immunoprecipitation followed by Western analyses (Figure 5.7B).
The specific mechanisms underlying the observed inter-dependency between DFO-
induced Y-64 and Y-155 phosphorylation remain to be investigated.

EGFP-PKCδ WT Y64F Y155F Y64/155F
200 μM DFO (h) 0 2 0 2 0 2 02
MEF/PKCδ
-/-
1 1.9 0.6 0.6 0.7 0.7 0.6 0.7
IB: pTyr IP: GFP
IB: GFP
IB: Actin
5% input
024 1 0.5 (h)
IB: pTyr IP: PKCδ
200 μM DFO
5% input
1 1.2 1.5 2.3 1.2
IB: Actin
11 1 1 1.1
IB: PKCδ
AB
EGFP-PKCδ WT Y64F Y155F Y64/155F
200 μM DFO (h) 0 2 0 2 0 2 02
MEF/PKCδ
-/-
1 1.9 0.6 0.6 0.7 0.7 0.6 0.7
IB: pTyr IP: GFP
IB: GFP
IB: Actin
5% input
EGFP-PKCδ WT Y64F Y155F Y64/155F
200 μM DFO (h) 0 2 0 2 0 2 0 2 0 2 0 2 02 02
MEF/PKCδ
-/-
1 1.9 0.6 0.6 0.7 0.7 0.6 0.7
IB: pTyr IP: GFP
IB: GFP
IB: Actin
IB: GFP
IB: Actin
5% input
024 1 0.5 (h)
IB: pTyr IP: PKCδ
200 μM DFO
5% input
1 1.2 1.5 2.3 1.2
IB: Actin
11 1 1 1.1
IB: PKCδ
024 1 0.5 (h)
IB: pTyr IP: PKCδ
200 μM DFO
5% input
1 1.2 1.5 2.3 1.2
IB: Actin
11 1 1 1.1
IB: PKCδ
AB

Figure 5.7 DFO induces Y-64 and Y-155 phosphorylation of PKCδ. (A) DFO-induced tyrosine
phosphorylation of PKCδ. Wild-type MEF cells were treated with DFO (200 μM) for indicated time
periods prior to harvesting cells. Equal amounts of whole cell lysates were subjected to
immunoprecipitation with an anti-PKCδ antibody followed by immunoblotting analyses with an anti-
phosphotyrosine 4G10 antibody. Western analyses of 5% input lysates using an anti-PKCδ or anti-actin
antibody served as protein expression controls. The relative density of immunoprecipitated complex was
normalized with total input lysates individually and calculated by setting basal phosphorylation level prior
to DFO-treatment as 1. (B) Both Y-64 and Y-155 of PKCδ are involved in DFO-induced PKCδ
phosphorylation. MEF/PKCδ
-/-
cells were transiently transfected with wild-type (WT), Y64F-, Y155F-, or
Y64/155F-EGFP-PKCδ. Transfected cells were treated with DFO (200 μM) for indicated time periods.
Equal amounts of whole cell lysates were subjected to immunoprecipitation with an anti-GFP antibody,
followed by immunoblotting with an anti-phosphotyrosine 4G10 antibody. Western analyses of 5% input
lysates using an anti-GFP or anti-actin antibody served as protein expression controls. The relative level of
DFO-induced Y-phosphorylation was calculated after normalizing with total input proteins and by setting
intensity obtained in non-treated EGFP-PKCδ WT transfectant as 1. Long Exp: longer exposures; Short
Exp: shorter exposures.

87
JNK1 signaling pathway is required for PKCδ-mediated activation of autophagy in
DFO-treated cells
Since PKC δ is required for DFO-induced autophagy, I postulate that PKC δ-mediated
signaling activation is required for the formation of autophagosomes in response to DFO
stress. Previously, Liu et al., demonstrated that PKC δ augments UV-induced JNK
activation (97). I thus tested whether DFO activates JNK in Pa-4 and Pa-4/PKC δKD cells.
Since DFO-treatment conveyed a preferential activation on short forms of JNK1/2
(JNK1/2
S
) in both Pa-4 and MEF cells, only p-JNK1/2
S
images are shown in Figures 5.8
and 5.9. Notably, the exposure of cells to DFO results in a transient c-Jun
phosphorylation at Ser-63, in relation with JNK1/2
S
activation (phosphorylation of Thr-
183/Tyr-185) in Pa-4 (Figure 5.8A, lanes 1 to 4), but not in Pa-4/PKC δKD (Figure 5.8B,
lanes 1 to 4), cells. To my surprise, treatment with 3-MA almost abolished both JNK
activation and c-Jun phosphorylation (Figure 5.8A, lanes 5 to 8 versus lanes 1 to 4).

88
B A
Actin
JNK1/2
S
JNK1/2
L
p-JNK1/2
S
1 0.9 0.7 0.9 1 0.8 1.6 1.8
c-Jun
p-S63-c-Jun
1 1.2 1.2 1.4 1 0.8 0.8 1.2
(h) 02 4 6
Pa-4/PKCδKD
10 mM 3-MA
02 4 6
200 μM DFO 200 μM DFO
Actin
JNK1/2
S
JNK1/2
L
p-JNK1/2
S
1 0.9 0.7 0.9 1 0.8 1.6 1.8 1 0.9 0.7 0.9 1 0.8 1.6 1.8
c-Jun
p-S63-c-Jun
1 1.2 1.2 1.4 1 0.8 0.8 1.2 1 1.2 1.2 1.4 1 0.8 0.8 1.2
(h) 02 4 6 02 4 6
Pa-4/PKCδKD
10 mM 3-MA
02 4 6 02 4 6
200 μM DFO 200 μM DFO
(h) 0 246 0 2 4 6
Pa-4
10 mM 3-MA
200 μM DFO
JNK1/2
L
JNK1/2
S
1 1.3 1.9 2.1 1 1 1 1.3
Actin
c-Jun
1 1.1 2.1 1.4 1 1 1 1.1
p-JNK1/2
S
p-S63-c-Jun
200 μM DFO
(h) 0 246 0 2 4 6(h) 0 246 0 246 0 2 4 6 02 4 6
Pa-4
10 mM 3-MA
200 μM DFO
JNK1/2
L
JNK1/2
S
1 1.3 1.9 2.1 1 1 1 1.3 1 1.3 1.9 2.1 1 1 1 1.3
Actin Actin
c-Jun c-Jun
1 1.1 2.1 1.4 1 1 1 1.1 1 1.1 2.1 1.4 1 1 1 1.1
p-JNK1/2
S
p-JNK1/2
S
p-S63-c-Jun p-S63-c-Jun
200 μM DFO

Figure 5.8 PKCδ is required for DFO-induced JNK and c-Jun phosphorylation in Pa-4 cells. DFO-
induced JNK and c-Jun phosphorylation in PKCδ-competent cells are inhibited by 3-MA treatment. Pa-4
and Pa-4/PKCδKD (A and B) cells were treated with DFO (200 μM) in the absence or presence of pre-
treatment with 3-MA (10 mM, 2 h) for indicated time periods before harvesting cells. Equal amounts of
whole cell lysates from different groups of cells were subjected to immunoblotting analyses with indicated
antibodies. Individual levels of specific proteins were visualized and quantified by Quantity One software
(Bio-Rad). After normalizing with actin levels, the relative levels of each protein were shown as numbers
in italic where non-treatment control level is designated as 1.

An identical observation was made in the pair of MEF and MEF/PKCδ
-/-
cells
(Figures 5.9A and B). The enhancement afforded by PKCδ on UV-induced activation of
both long and short forms of JNK1/2 was demonstrated (Figure 5.10) in support of
validating our observations.

89
B A
JNK1/2
L
JNK1/2
S
1 2.5 2.8 2.9 1 1 1 1
1 1.6 2.4 2.6 1 1 1.1 1.3
p-S63-c-Jun
c-Jun
Actin
PKCδ
p-JNK1/2
S
(h) 02 4 6 02 4 6
MEF
10 mM 3-MA
200 μM DFO 200 μM DFO
JNK1/2
L
JNK1/2
S
1 2.5 2.8 2.9 1 1 1 1 1 2.5 2.8 2.9 1 1 1 1
1 1.6 2.4 2.6 1 1 1.1 1.3 1 1.6 2.4 2.6 1 1 1.1 1.3
p-S63-c-Jun p-S63-c-Jun
c-Jun c-Jun
Actin Actin
PKCδ PKCδ
p-JNK1/2
S
(h) 02 4 6 02 4 6 02 4 6 02 4 6
MEF
10 mM 3-MA
200 μM DFO 200 μM DFO
JNK1/2
L
JNK1/2
S
11.6 1.4 2 1 1 1 1
11.2 1.2 2 11 1.6 1.2
Actin
p-S63-c-Jun
PKCδ
c-Jun
p-JNK1/2
S
(h) 02 4 6 0 2 4 6
MEF/PKCδ
-/-
10 mM 3-MA
200 μM DFO 200 μM DFO
JNK1/2
L
JNK1/2
S
11.6 1.4 2 1 1 1 1 11.6 1.4 2 1 1 1 1
11.2 1.2 2 11 1.6 1.2 11.2 1.2 2 11 1.6 1.2
Actin
p-S63-c-Jun
PKCδ
c-Jun
p-JNK1/2
S
(h) 02 4 6 02 4 6 0 2 4 6 02 4 6
MEF/PKCδ
-/-
10 mM 3-MA
200 μM DFO 200 μM DFO

Figure 5.9 PKCδ is required for DFO-induced JNK and c-Jun phosphorylation in MEF cells. DFO-
induced JNK and c-Jun phosphorylation in PKCδ-competent cells are inhibited by 3-MA treatment. MEF
and MEF/PKCδ
-/-
(A and B) cells were treated with DFO (200 μM) in the absence or presence of pre-
treatment with 3-MA (10 mM, 2 h) for indicated time periods before harvesting cells. Equal amounts of
whole cell lysates from different groups of cells were subjected to immunoblotting analyses with indicated
antibodies. Individual levels of specific proteins were visualized and quantified by Quantity One software
(Bio-Rad). After normalizing with actin levels, the relative levels of each protein were shown as numbers
in italic where non-treatment control level is designated as 1.

MEF/WT MEF/PKCδ
-/-
40 J/cm
2
UV (h)
p-JNK1/2
S
Actin
0 0.5 1 2 0 0.5 1 2
1 35.2 27.5 12.9 117 11 4.1
p-JNK1/2
L
MEF/WT MEF/PKCδ
-/-
40 J/cm
2
UV (h)
p-JNK1/2
S
Actin
00.5 1 2 0 0.5 1 2 0 0.5 1 2 00.5 1 2
1 35.2 27.5 12.9 117 11 4.1 1 35.2 27.5 12.9 117 11 4.1
p-JNK1/2
L

Figure 5.10 PKCδ is involved in UV-induced JNK activation in MEF cells. MEF/WT and MEF/PKCδ
-/-

cells were exposed to UV (40 J/cm
2
) for the indicated time periods, followed by preparation of whole cell
lysates and subsequent Western analyses. Relative levels of p-JNK were normalized against corresponding
actin levels, where the fold inductions are shown with the numbers in italic by setting the control (non-
treatment) level as 1.

90
Two closely related JNK isoforms, JNK1 and JNK2, exhibit shared functions;
however, they also have distinct biological activities (96). To ascertain whether JNK1 or
JNK2 is required for DFO-induced autophagy, I examined LC3 processing and JNK
activation profile in response to DFO-treatment in MEFs that are deficient in either JNK1
or JNK2 allele. Immunoblotting analyses using an anti-JNK antibody, which can detect
both long forms (p54) and short forms (p46) of JNK1 or JNK2 confirmed the expression
of JNK1 and JNK2 in respective JNK2
-/-
and JNK1
-/-
MEF

cells (Figure 5.11).

MEF
JNK1
-/-
JNK2
-/-
WT
p46-JNK
p54-JNK
c-Jun
Actin
MEF
JNK1
-/-
JNK2
-/-
WT
p46-JNK
p54-JNK
c-Jun c-Jun
Actin Actin

Figure 5.11 Western analyses of JNK expression in MEF/WT, MEF/JNK1
-/-
, and MEF/JNK2
-/-
cells.
Equal amounts of whole cell lysates prepared from MEF, MEF/JNK1
-/-
, and MEF/JNK2
-/-
cells were
subjected to immunoblotting analyses with anti-JNK, anti-c-Jun and anti-actin antibodies, respectively, to
assess total proteins of interest.

As shown in Figure 5.12A, I only detected phospho-JNK1/2
S
signals in wild-type and
JNK2
-/-
, but not in JNK1
-/-
, MEF cells (top panels). Immunoblotting analyses using an
anti-phospho-c-Jun (Ser-63) antibody revealed that the c-Jun phosphorylation induced by
DFO was much weaker in JNK1
-/-
cells compared to those detected in wild-type and
JNK2
-/-
cells (Figure 5.12A, second panels). Endogenous LC3 processing was induced
by DFO in JNK2
-/-
, but not in JNK1
-/-
, cells in a manner similar to that in wild-type cells
(Figure 5.12B, top panels).
91

B
A
(h) 200 μM DFO
p-S63-c-Jun
Actin
p-JNK1/2
S
c-Jun
02 4 6
MEF/JNK2
-/-
2.4 2.1 1.6 1
2.8 3.4 2.3 1
02 4 6
MEF/JNK1
-/-
11.3 1.2 1
02 4 6
MEF
12 2.1 1.9
12 2.4 2.5
(h) 200 μM DFO
p-S63-c-Jun
Actin
p-JNK1/2
S
c-Jun
02 4 6 02 4 6
MEF/JNK2
-/-
2.4 2.1 1.6 1
2.8 3.4 2.3 1
02 4 6
MEF/JNK1
-/-
11.3 1.2 1
02 4 6 02 4 6
MEF
12 2.1 1.9
12 2.4 2.5
LC3-II
Actin
200 μM DFO
10 mM 3-MA
(h)
(h)
11 1 1 1 1 1 1 11 1 1
200 μM DFO (h)
1 2.1 3 1. 5 1 1.8 3.4 1. 6 11.1 1 1
Actin
0
2
2
2
4
2
6
2
02 4 6
MEF
0
2
2
2
4
2
6
2
02 4 6
MEF/JNK2
-/-
LC3-II
0
2
2
2
4
2
6
2
02 4 6
MEF/JNK1
-/-
LC3-II
Actin
200 μM DFO
10 mM 3-MA
(h)
(h)
11 1 1 1 1 1 1 11 1 1
200 μM DFO (h)
1 2.1 3 1. 5 1 1.8 3.4 1. 6 11.1 1 1
Actin
0
2
2
2
4
2
6
2
02 4 6 02 4 6
MEF
0
2
2
2
4
2
6
2
02 4 6 02 4 6
MEF/JNK2
-/-
LC3-II
0
2
2
2
4
2
6
2
02 4 6 02 4 6
MEF/JNK1
-/-

Figure 5.12 JNK1, but not JNK2, is involved in DFO-induced autophagy. (A) DFO induces JNK1
phosphorylation. MEF, MEF/JNK1
-/-
, and MEF/JNK2
-/-
cells were treated with DFO (200 μM) for
indicated time periods. Equal amounts of whole cell lysates were subjected to Western analyses with
specific antibodies as shown. The relative levels (numbers in italic) of specific proteins were calculated as
described above. (B) JNK1, but not JNK2, is responsible for DFO-induced autophagy. Cells were treated
with DFO (200 μM) in the absence (upper panels) or presence (lower panels) of pre-treatment with 3-MA
(10 mM, 2 h) for indicated time periods before harvesting cells. Equal amounts of lysates were subjected to
Western analyses with anti-LC3 and anti-actin antibodies, respectively. The relative levels (numbers in
italic) of LC3-II are normalized against their corresponding internal loading controls of actin, where non-
treatment value is set as 1.

These results shown in Figures 5.8, 5.9 and 5.12 together suggest that, while PKC δ is
involved in DFO-induced autophagy, JNK1 activation is necessary for DFO-induced LC3
processing. My observation that JNK activation is attenuated in Pa-4/PKC δKD and
92
PKC δ
-/-
cells (Figures 5.8B and 5.9B) also places JNK1 to the downstream of PKC δ in
the context of DFO stress.

PKCδ stimulates the release of Beclin-1 from the complex of Beclin-1 and bcl-2
upon DFO-exposure
The cross-talk between autophagy and apoptosis is further exemplified by a recent
finding that bcl-2, the prototypic apoptosis inhibitor, inhibits autophagy by binding to
Beclin-1 (121). To address whether DFO induces autophagy by dissociating Beclin-1
from the inhibitory bcl-2, immunoprecipitation by an anti-Beclin-1 antibody were
performed in MEF/WT and MEF/PKC δ
-/-
cells treated with DFO for different time
periods. It has been established that minimal levels of bcl-2 are co-immunoprecipitated
with Beclin-1 during autophagy-inducing condition, whereas high levels of bcl-2 co-
immunoprecipitated with Beclin-1 under autophagy-inhibitory condition (104, 121). As
shown in Figure 5.13, DFO-treatment resulted in a dissociation of Beclin-1 from bcl-2
exclusively in MEF/WT, but not in MEF/PKC δ
-/-
, cells, as evaluated by the extent of bcl-
2 remaining as a form complexed with Beclin-1. Taken together with result shown in
Figure 5.6B, the observation that decreased association of Beclin-1 with bcl-2 in response
to DFO-stress is PKCδ-dependent supports our contention that DFO induces autophagy
via PKC δ activation.

93

MEF/PKCδ
-/-
IB: Beclin-1
IB: bcl-2
11 11 1 1 1 1
MEF/WT 1 0.8 0.5 0.5 1 1 1 1
IB: bcl-2
IB: Beclin-1
200 μM DFO (h) 04 210 4 2 1
IP: Beclin-1 5% input
MEF/PKCδ
-/-
IB: Beclin-1
IB: bcl-2
11 11 1 1 1 1
IB: Beclin-1 IB: Beclin-1
IB: bcl-2 IB: bcl-2
11 11 1 1 1 1 11 11 1 1 1 1
MEF/WT 1 0.8 0.5 0.5 1 1 1 1
IB: bcl-2
IB: Beclin-1
1 0.8 0.5 0.5 1 1 1 1 1 0.8 0.5 0.5 1 1 1 1
IB: bcl-2 IB: bcl-2
IB: Beclin-1 IB: Beclin-1
200 μM DFO (h) 04 210 4 2 1 200 μM DFO (h) 04 210 4 2 1
IP: Beclin-1 5% input IP: Beclin-1 5% input

Figure 5.13 PKCδ is required for DFO-induced dissociation of Beclin-1 from bcl-2. MEF and
MEF/PKCδ
-/-
cells were treated with DFO (200 μM) for indicated time periods. Equal amounts of cell
lysates were subjected to immunoprecipitation with an anti-Beclin-1 antibody and followed by Western
analyses with anti-bcl-2 and anti-Beclin-1 antibodies, respectively. Five percent of total cell lysate inputs
analyzed by Western analyses as indicated served as protein expression controls. DFO-induced dissociation
rates (numbers in italic) of Beclin-1 from bcl-2 are assessed by the quantitative analyses of
immunoprecipitated bcl-2, as described previously.

It has been suggested that JNK1-mediated bcl-2 phosphorylation is essential to
relieve the inhibitory effect of bcl-2 on autophagy (120). Due to the commercial
antibody against p-Ser-70-bcl-2 only recognizes human phosphorylated bcl-2, we
examined the bcl-2 Ser-70 phosphorylation profiles during the course of DFO-treatment
in human breast cancer MCF-7/Caspase-3 and MCF-7/Neo cells. As shown in Figure
5.14, it appears that DFO-treatment resulted in a relatively rapid induction of JNK1 and
bcl-2 phosphorylation in MCF-7/Caspase-3 cells compared to that in MCF-7/Neo cells at
1 h post DFO-treatment. Thus, it is very likely that DFO induces the JNK1-mediated
phosphorylation of bcl-2 and as a result, inhibits interaction with Beclin-1 and stimulates
autophagy.

94

1 1.1 1.5 1.1 1 1.3 1.5 0.9
Actin
bcl-2
200 μM DFO
(h) 01 2 4 01 2 4
MCF7 Neo Caspase-3
Beclin-1
JNK1/2
L
JNK1/2
S
1 1.4 1.6 1.2 1 1.9 1.6 1.4
p-JNK1/2
L
p-S70-Bcl-2
1 1.1 1.5 1.1 1 1.3 1.5 0.9
Actin Actin
bcl-2 bcl-2
200 μM DFO
(h) 01 2 4 01 2 4 (h) 01 2 4 01 2 4 01 2 4 01 2 4
MCF7 Neo Caspase-3 MCF7 Neo Caspase-3
Beclin-1 Beclin-1
JNK1/2
L
JNK1/2
S
1 1.4 1.6 1.2 1 1.9 1.6 1.4
p-JNK1/2
L
p-JNK1/2
L
p-S70-Bcl-2 p-S70-Bcl-2

Figure 5.14 DFO-induced JNK and bcl-2 phosphorylation is more pronounced in MCF-7/Caspase-3
than in MCF-7/Neo cells. MCF-7/Caspase-3 and MCF-7/Neo cells were treated with DFO (200 μM) for
indicated time periods. Relative levels (numbers in italic) of phospho-Ser-70-bcl-2 and phospho-JNK, after
normalization with their corresponding actin levels, are shown.

Given the difference in terms of DFO-induced JNK1 and bcl-2 phosphorylation
between MCF-7/Caspase-3 and MCF-7/Neo cells, I hypothesize that the DFO-induced
autophagosome accumulation is more readily detectable in MCF-7/Caspase-3 cells. To
test this possibility, I evaluated the DFO-induced autophagosome formation in both
MCF-7/Neo and MCF-7/Caspase-3 cells. As shown in Figure 5.15A, treatment with
DFO for 4 h induced the accumulation of autophagosomes, which were recognized as the
GFP-LC3 puncta in MCF-7/Caspase-3 cells (left 3
rd
panels). In contrast, I rarely
observed GFP-LC3 puncta following the DFO-treatment in MCF-7/Neo cells (Figure
5.15A, right 6 top panels). To assess whether MCF-7/Neo cells are capable of forming
autophagosomes, I examined the autophagosome formation by treating MCF-7/Caspase-3
and MCF-7/Neo cells with CQ (10 μM) for 4 h. These CQ-treated MCF-7/Neo cells
95
exhibited a GFP-LC3 puncta profile comparable to that in MCF-7/Caspase-3 cells at 4 h
post-treatment, when autophagosomes accumulated (Figure 5.15A, bottom panels).
Taken together with results shown in Figures 5.13 and 5.14, I conclude that the increased
JNK1 activation and phosphorylation at Ser-70 of bcl-2 favor the autophagosome
formation upon exposure to DFO.

96
A
100 μM DFO
0 h
2 h
4 h
6 h
12 h
24 h
CQ 4 h
MCF-7/Caspase-3
GFP-LC3
MCF-7/Neo
Merge GFP-LC3 DAPI DAPI Merge
100 μM DFO
0 h
2 h
4 h
6 h
12 h
24 h
CQ 4 h
MCF-7/Caspase-3
GFP-LC3
MCF-7/Neo
Merge GFP-LC3 DAPI DAPI Merge

B
0
10
20
30
40
50
60
70
80
90
100
GFP-LC3 Puncta/GFP
Positive Cells (%)
2 0461224
MCF7/C3
MCF7/Neo
(h)
100 μM DFO
0
10
20
30
40
50
60
70
80
90
100
0
10
20
30
40
50
60
70
80
90
100
GFP-LC3 Puncta/GFP
Positive Cells (%)
2 0461224 2 0461224
MCF7/C3
MCF7/Neo
(h)
100 μM DFO

Figure 5.15 DFO-treatment induces GFP-LC3 puncta accumulation in MCF-7/Caspase-3, but not in
MCF-7/Neo, cells. (A) MCF-7/Caspase-3 and MCF-7/Neo cells were transfected with GFP-LC3 and
treated with DFO (100 μM) for indicated time periods or CQ (10 μM) for 4 h. Green color indicates GFP-
LC3 signals, whereas blue color represented represents nuclear staining (DAPI staining). Images shown are
representatives from one out of three independent experiments. Autophagosome formation was indicated
by accumulation of GFP-LC3 puncta (arrows). (B) Quantification of GFP-LC3 puncta in MCF-7 cells.
Cells undergoing autophagy were determined and enumerated as described in Figure 5.1C.
97
Protective effect of autophagy on cell death induced by DFO
Paradoxically, autophagy can serve to protect cells, but may also contribute to cell
damage (90). To examine whether the autophagy induced by DFO stress plays a role in
cell survival or cell death, Pa-4 cells in which autophagy was blocked using 3-MA were
exposed to DFO. The 3-MA-pretreated cells showed more prevalence in cell death at 24
h and 48 h after DFO-treatment compared to vehicle-pretreated cells (Figure 5.16).

200 μM DFO
2 mM 3-MA+200 μM DFO
10 μM Rapamycin+200 μM DFO
Non-treatment
∗∗ p < 0.01
0
20
40
60
80
100
120
(h) 0 24 48 24 48 24 48
Cell Viability (%)
∗∗
∗∗
∗∗
∗∗ ∗∗
∗∗ ∗∗
∗∗
∗∗
Pa-4
200 μM DFO
2 mM 3-MA+200 μM DFO
10 μM Rapamycin+200 μM DFO
Non-treatment
∗∗ p < 0.01
200 μM DFO
2 mM 3-MA+200 μM DFO
10 μM Rapamycin+200 μM DFO
Non-treatment
∗∗ p < 0.01
0
20
40
60
80
100
120
(h) 0 24 48 24 48 24 48
Cell Viability (%)
∗∗
∗∗
∗∗
∗∗ ∗∗
∗∗ ∗∗
∗∗
∗∗
Pa-4

Figure 5.16 Cytoprotective role of autophagy against DFO-induced cell death in Pa-4 cells. Cells were
subjected to pre-treatment with either vehicle, 3-MA (2 mM, 2 h), or rapamycin (10 μM, 2 h), following
combined treatment with DFO (200 μM) for indicated time periods. MTT assays were performed at the end
of indicated time periods to determine cell viability. Results from 6 independent experiments are shown as
percentage of cell viability, where the value corresponding to cells treated with vehicle alone is set as 100%.
Solid lines broken with ** denote p < 0.01.

In addition, cells pretreated with rapamycin, which was previously shown to
effectively inhibit mTOR and induce autophagy (134), were relatively resistant to DFO
only at 24 h following treatment. I further analyzed the sensitivity of MEF wild-type,
98
JNK1
-/-
, JNK2
-/-
, Atg5
-/-
and PKC δ
-/-
cells to DFO at 24 h and 48 h post-treatment. Atg5-
deficient cells showed significantly increased vulnerability to DFO, compared with wild-
type cells (Figure 5.17, lanes 2 and 3 versus lanes 11 and 12). However, either JNK1-
deficient or PKC δ-deficient cells exhibited a partial resistance to DFO-induced cell death
at 48 h post-treatment (Figure 5.17, lane 3 versus lanes 6 and 15).

∗ p < 0.05 ∗∗ p < 0.01
100 μM DFO
MEF JNK2
-/-
WT PKCδ
-/-
Atg5
-/-
JNK1
-/-
∗∗ ∗∗
∗∗ ∗∗
∗∗
∗∗
0 24 48 0 24 48
∗∗
02448 02448
∗∗
02448
(h)
0
20
40
60
80
100
120
Cell Viability (%)
∗∗
∗∗
∗∗ ∗∗ ∗∗ ∗∗ ∗∗ ∗∗ ∗∗ ∗∗ ∗∗ ∗
∗∗ ∗∗ ∗∗ ∗∗ ∗∗
∗
∗∗
∗ p < 0.05 ∗∗ p < 0.01
100 μM DFO
MEF JNK2
-/-
WT PKCδ
-/-
Atg5
-/-
JNK1
-/-
∗∗ ∗∗
∗∗ ∗∗
∗∗
∗∗
0 24 48 0 24 48
∗∗
02448 02448
∗∗
02448
(h)
0
20
40
60
80
100
120
Cell Viability (%)
∗∗
∗∗
∗∗ ∗∗ ∗∗ ∗∗ ∗∗ ∗∗ ∗∗ ∗∗ ∗∗ ∗
∗∗ ∗∗ ∗∗ ∗∗ ∗∗
∗
∗∗

Figure 5.17 Lack of cytoprotection by PKCδ and JNK1 against long term DFO-treatment. MEF WT,
JNK1
-/-
. JNK2
-/-
, Atg5
-/-
and PKCδ
-/-
cells were treated with DFO (100 μM) for 0, 24 or 48 h and subjected
to MTT analyses. Results from 9 independent experiments are shown as percentage of cell viability as
described in Figure 5.16. Solid lines broken with ** represent statistical significance at each time point
between (but within the same cell line) DFO-treated versus control cells. Dotted or dashed lines broken
with either * or ** represent statistical significance at 24 or 48 h post DFO-treatment between different cell
lines, respectively. * denotes p < 0.05, ** denotes p < 0.01.

99
Lastly, I evaluated JNK activation profile at 24 h and 48 h post-treatment with DFO
in Pa-4 and Pa-4/PKC δKD cells. In contrast to those shown in Figure 5.12A, the delayed
JNK activation at both short and long forms of JNK1/2 was observed at 48 h post DFO-
treatment in both Pa-4 and Pa-4/PKC δKD cells, albeit at different extents (Figure 5.18).

50 μM DFO (h) 0 24 48 0 24 48
Pa-4 Pa-4/PKCδKD
Tubulin
p54-JNK
p46-JNK
11.1 2.4 1 p-JNK2
L
1.4 1.6
11.1 9 1 1 4.3 p-JNK1
L
p-JNK1/2
L
p-JNK1/2
S
11.2 5.3 1 p-JNK2
S
35.4
10.8 2.1 1 1 1.8 p-JNK1
S
50 μM DFO (h) 0 24 48 0 24 48
Pa-4 Pa-4/PKCδKD
Tubulin
p54-JNK
p46-JNK
11.1 2.4 1 p-JNK2
L
1.4 1.6
11.1 9 1 1 4.3 p-JNK1
L
p-JNK1/2
L
p-JNK1/2
S
11.2 5.3 1 p-JNK2
S
35.4
10.8 2.1 1 1 1.8 p-JNK1
S

Figure 5.18 Long-term DFO-treatment induces JNK activation. Pa-4 and Pa-4/PKCδKD cells were
treated with DFO (50 μM) for 24 and 48 h. Equal amounts of cell lysates were subjected to Western
analyses with specific antibodies as indicated. Relative levels of phospho-JNK1/2 long and short forms,
after normalizing with its corresponding tubulin levels, are shown where the respective level in vehicle-
treated cells is set as 1.

In summary, these results indicate that PKCδ and JNK1 (short form) activation in the
early phase of DFO stress is required to activate autophagy, which plays a transient role
in protecting against cell death triggered by DFO stress (Figure 5.19, solid lines; please
see Discussion for details). In contrast, a sustained activation of both short and long
forms of JNK1/2 via PKC δ-dependent and -independent pathways by DFO could lead to
cell death (Figure 5.19, dotted lines).

100
Cell Death
3-MA
JNK1
Beclin-1
bcl-2
p
Autophagy Rapamycin
DFO
3-MA class III PI3K (?)
Tyr-phosphorylation ?
PKCδ PKCδKD
class III PI3K
Beclin-1
bcl-2
Cell Survival
?
Cell Death
3-MA
JNK1
Beclin-1 Beclin-1
bcl-2
p
bcl-2 bcl-2
p
Autophagy Rapamycin
DFO
3-MA class III PI3K (?)
Tyr-phosphorylation ?
PKCδ PKCδKD PKCδ PKCδKD
class III PI3K
Beclin-1 Beclin-1
bcl-2 bcl-2
Cell Survival
?

Figure 5.19 Proposed signaling pathway underlying DFO-induced autophagic process. DFO induces
PKCδ activation through tyrosine phosphorylation at Y64 and Y155 of PKCδ. DFO induces JNK1
activation through a PKCδ-dependent pathway, which is inhibited by 3-MA treatment. Activated JNK1 is
postulated to phosphorylate bcl-2, hence releasing Beclin-1 to induce autophagy. Rapamycin induces
autophagy through inhibition of mTOR signaling. Solid line signifies DFO-induced PKCδ-dependent
autophagic pathway while dotted line depicts DFO-induced apoptotic signalings leading to cell death. See
text for details.

Discussion
I show in this chapter that a hypoxia-mimetic, DFO, induces autophagic response and
that PKCδ/JNK1 activation in the early phase ensuing DFO-treatment is required for this
process. The following observations support this conclusion. First, DFO-treatment
stimulates the accumulation of GFP-LC3 puncta and LC3-II formation/degradation.
Second, DFO-induced autophagic process is attenuated in cells stably transfected with
101
PKCδKD, or in cells of PKCδ
-/-
, Atg5
-/-
and JNK1
-/-
contexts or in cells treated with a
class III PI3K inhibitor, 3-MA. Third, DFO-treatment leads to bcl-2 phosphorylation and
PKCδ-dependent dissociation of Beclin-1 from bcl-2. By using cells derived from two
different genetic approaches, dominant-negative PKCδKD and PKCδ-knockout, I
conclude that the DFO-mediated early activation of JNK1 is PKCδ-dependent. My
present study also suggests the possibility that autophagy could play a transient pro-
survival role against DFO-induced cell death.
Protein kinase C (PKC) comprises a family of 11 structurally related serine/threonine
protein kinases that regulate diverse biological functions (115). In the current study, I
provide evidence that the early activation of JNK1, downstream of PKC δ, is required for
DFO-induced autophagic response, providing a pro-survival signal. However, both
PKCδ- and JNK1-null cells exhibited resistance to DFO-elicited cell death (Figure 5.17).
This apparent discrepancy might be explained by the fact that the sustained activation of
PKCδ signaling is pro-apoptotic (25, 115). In accordance with these findings, PKC δ-
deficient mice exhibit resistance to irradiation and are defective in mitochondria-
dependent apoptosis (67, 88). PKC δ is also a Caspase substrate and Caspase-3-mediated
cleavage of PKCδ generates a constitutively active catalytic fragment (PKC δ-CF), which
is a potent inducer of apoptosis (25). To date, the mechanisms underlying apoptosis,
compared to autophagy, are much better elucidated. For example, PKC δ has been shown
to phosphorylate and activate Caspase-3 (160), to phosphorylate and target the anti-
apoptotic protein Mcl-1 for degradation (144), and to suppress phosphorylation of Akt
(25, 110). I have previously reported that the delayed activation of PKC δ and resultant
102
proteolytic cleavage followed by the subsequent nuclear translocation of PKC δ-CF favor
DFO-induced cell death (25). My current findings significantly advance the
understanding of the role of PKC δ in autophagy. The importance of time-dependent
PKC δ activation in differential cell fate determination is further supported by the
observation that PKC δ activation also plays a critical role in ER stress-mediated response
and cell death during ischemia-reperfusion injury (127). Using various genetically
modified knockout cells and pharmacological inhibitors (Figures 5.16 and 5.17), I herein
unequivocally demonstrated that PKC δ conveys dual roles, a transient pro-survival via
autophagy (this report) and a long term pro-death via Caspase-3 activation signals (25), to
regulate cell fate decision via post-translational modification in a time-dependent manner.
In support of the notion that the early activation of PKCδ is likely involved in DFO-
stimulated autophagy, PKC δ is also reportedly to be localized in the early endosomes (99,
153) in addition to its known role as a mediator for promoting cell apoptosis in the
nucleus (25). DFO, a membrane-impermeant iron chelator, is taken up primarily by fluid
phase endocytosis and travels through various endosomes (40). It is plausible that the
changes in membrane stability or microenvironment of these intracellular organelles upon
DFO-treatment lead to the activation of PKC δ through Y-phosphorylation. It has been
reported that various non-receptor tyrosine kinases, such as c-Abl and Lyn, phosphorylate
PKCδ in different tyrosine residues in response to specific stimuli leading to cell pro-
apoptosis (130, 167). Furthermore, Yoshidat K et al., have demonstrated that PKC δ
plays a critical role in the activation of SAPK/JNK through Lyn PKC δ MEKK1
MKK7 SAPK/JNK signaling in response to genotoxic stress (169). In my current
studies, I showed that DFO induces a rapid and transient phosphorylation of Y-64 and Y-
103
155 at the N-terminal regulatory domain of PKC δ (Figure 5.7). It is postulated widely
that PKC δ is in a “primed” closed conformation via domain-domain interaction, awaiting
stimulus for its activation (145). The phosphorylation and dephosphorylation of PKC δ at
Y-64 and Y-155 have been suggested to function as a regulatory switch to promote
nuclear translocation of PKC δ in etoposide-treated cells (68). However, which specific
tyrosine kinase is required on DFO-induced PKC δ Y-phosphorylation and the exact role
of phosphorylated PKC δ at both Y-64 and Y-155 in the scheme of DFO-induced
autophagy and PKC δ-dependent JNK1 activation are still under investigation.
My proposed model, based on the data shown herein, is depicted in Figure 5.19. In
brief, DFO-treatment leads to a rapid PKCδ and JNK1 (short form) activation. The
subsequent dissociation of Beclin-1 from protein complexes of bcl-2, presumably
phosphorylated by JNK1 (Figure 5.14; (120)), stimulates the autophagic process.
Autophagy could promote survival of DFO-treated cells by removing DFO-damaged
organelles; however, the effect is only transient. In agreement with my proposed model,
I demonstrate that 3-MA downregulates DFO-induced changes in LC3-II level (Figures
5.6B and 5.12B). Intriguingly, 3-MA also attenuates JNK1 activation in DFO-treated
MEF/WT and Pa-4 cells, reminiscent of that observed in MEF/PKC δ
-/-
and Pa-4/PKC δKD
cells (Figures 5.8 and 5.9), respectively, suggesting that a putative class III PI3K,
potentially Vps34, bridges the activation of signaling from PKC δ to JNK1. There was a
precedent that JNK activity is increased in PTEN knockout MEF cells (159). However, it
remains unclear how Vps34 or its product PtdIns3P induces JNK activation. Whether or
not 3-MA inhibits enzymes other than class III PI3K also remains to be elucidated.
Based on literature data, 3-MA is likely the preferential class III PI3K inhibitor, showing
104
no inhibition of class I PI3K (41). I further demonstrated that 3-MA by itself has no
observable effect on JNK1 activation (Figure 5.20). Based on my data, I thus propose
that a 3-MA-sensitive kinase complex, possibly containing class III PI3K, is involved in
transmitting DFO-activated PKC δ signals to JNK1 activation. In addition, the Beclin-1
binding partner, Vps34/class III PI3K, is predominantly found on endosomes and trans-
Golgi network (49). The close proximity of DFO, PKCδ and Vps34 located at the
endosomes further underscores the validity of our proposed model. Together, PKC δ
plays a critical role in the mechanism underlying the activation of autophagy pathway in
cells exposed to DFO or hypoxia. In addition, I have provided evidence that the class III
PI3K can be considered as part of the molecular machinery of PKC δ-JNK1 pathway for
DFO to induce autophagy.

MEF Pa-4
0 246 (h) 10 mM 3-MA 0 246
p-JNK1/2
S
111 1.4 11.1 1.4 1
JNK1/2
L
JNK1/2
S
Actin
p-S63-c-Jun
c-Jun
1 0.9 0.9 5.2 1 1.2 1.4 4.7
p-JNK1/2
L
MEF Pa-4
0 246 (h) 10 mM 3-MA 0 246
p-JNK1/2
S
111 1.4 11.1 1.4 1
JNK1/2
L
JNK1/2
S
Actin
p-S63-c-Jun
c-Jun
1 0.9 0.9 5.2 1 1.2 1.4 4.7
p-JNK1/2
L

Figure 5.20 3-MA-treatment elicits no effect on JNK activation in both Pa-4 and MEF cells. Cells
were treated with 10 mM 3-MA for the indicated time periods prior to harvesting for preparation of whole
cell lysates. Equal amounts of whole cell lysates were then subjected to Western analyses using specific
antibodies. Relative levels of phosphorylated proteins were normalized against their corresponding actin
levels, where the fold inductions are shown in italic by setting the control (non-treatment) value as 1.

105
Hypoxic stress is thought to be encountered during various pathological situations,
including cancer, myocardial infarction and stroke (126). Cells deprived of oxygen will
initially employ adaptive and survival strategies, but if hypoxia is sustained, cell death
will eventually ensue. Hypoxia-induced autophagy has been reported under a sustained
hypoxic condition through HIF-1α (Hypoxia-inducible factor-1 α), BNIP3 (Bcl-
2/adenovirus E1B 19kDa interacting protein 3), Atg5, and Beclin-1 (1, 90, 128 and
references therein). I have demonstrated herein that DFO-treatment utilizes PKCδ/JNK1
axis to induce a rapid and transient autophagic process (summarized in Figure 5.19) and
further shown that a similar rapid change in LC3 lipidation is observed in Pa-4 cells
exposed to 1% O
2
(Figure 5.4). I speculate that the identified PKCδ/JNK1 axis is also
utilized by 1% O
2
hypoxic exposure to convey a rapid and transient induction of
autophagy, which is currently under investigation. Chronic hypoxia is a typical
microenvironment occurring during tumor development, as rapid proliferation causes the
tumor to outgrow its available oxygen supply and triggers a variety of adverse effects
arising from metabolic stress. The precise mechanism of hypoxia-induced cell death
remains unclear as apoptosis, necrosis and autophagy have all been reported to occur in
response to hypoxic stress (30). Hypoxia-induced autophagy has been proposed to
mitigate genome damage caused by metabolic stress (76). Obviously, hypoxia/metabolic
stress-induced autophagy might also play “dual roles” in cell survival and cell death:
early induction of autophagy, as I demonstrated in this chapter, may contribute to a
protective response, whereas prolonged autophagy could lead to cell death. In line with
this scenario, I have shown that autophagy occurs following acute DFO-treatment, which
is likely to be an initial survival strategy (Figures 5.16 and 5.17). For example, cells are
106
shown to lead to decreased survival when autophagy is blocked by 3-MA, while cells
temporally exhibit increased survival when autophagy is stimulated by rapamycin.
A complete picture of the signaling pathways and associated functions of PKCδ
and/or JNK1 in DFO- (and hypoxia-) treated cells has yet to emerge. The data presented
herein expand the understanding of PKCδ/JNK1 biology, as well as the hitherto unknown
role of PKCδ/JNK1 in regulating DFO (and perhaps hypoxia)-induced autophagy. In
summary, my data suggest that the activated PKCδ/JNK1 pathway in response to DFO-
treatment is essential for the release of Beclin-1 from its inhibitor bcl-2 to stimulate
autophagy. The proposed signaling pathway will help us to understand the regulation of
autophagy, especially those activated independent of nutrient starvations.

107
CHAPTER 6
SUMMARY AND PERSPECTIVES
Summary
Hypoxia is a (patho)physiological condition. Hypoxic stress activates various signal
transduction pathways, including post-translational modification with the SUMO-1
protein (SUMOylation). It has been reported that SUMOylation regulates diverse cellular
functions including intracellular targeting, DNA repair, cell cycle progression, and
responses to extracellular stimuli. Accumulating evidence suggest the potential critical
role of SUMOylation in governing cellular hypoxic responses. In this dissertation, I
investigate the relationship between the 1% O
2
- and/or hypoxia-mimetic desferroxamine
(DFO)-stimulated SUMOylation process and the ability of cells to resist cell injury
elicited by these treatments, respectively, in the rat salivary Pa-4 epithelial cell line. I
have shown that exposure to 1% O
2
and/or DFO-treatment stimulates global
SUMOylation, resulting in a disrupted epithelial barrier function, F-actin and ZO-1
assembly, and activates DNA damage response signalings. By using salivary Pa-4 cells
stably transduced with lenti-SUMO-1 and a cell permeant peptide harboring SUMO-
binding motif to interfere with SUMO-dependent protein-protein interactions, we
demonstrate that SUMOylation augments cell survival against acute hypoxic stress-
induced cell growth inhibition by down-regulation of S139 H2AX phosphorylation and
attenuation of PKCδ activation and Caspase-3 cleavage, hallmarks of pro-apoptotic
signaling. Moreover, DFO-induced S1981 phosphorylation of ATM, a DNA damage
marker, precedes PKCδ and Caspase-3 activation. I further show that constitutive
SUMOylation facilitates 1% O
2
- or DFO-induced NF-κB transactivation, possibly via
108
activation of genotoxic signaling cascade. In addition, I report a transient preservation of
transepithelial electrical resistance (TER) during the early stage of hypoxia (1% O
2
) as
well as enhanced TER recovery following prolonged hypoxia in SUMO-1-expressing cell
monolayers. Taken together, my results unveil a previously unrecognized mechanism by
which SUMOylation and an intricate balance among activation of ATM, PKCδ, Caspase-
3 and NF-κB signaling pathways play critical roles in modulating salivary adaptive
responses to either 1% O
2
or DFO, governing salivary epithelial homeostasis.
Since I have observed PKCδ activation in response to hypoxic stress and DFO-
treatment and PKCδ has been reported to participate in a variety of signal transduction
pathways related apoptosis, cell proliferation and tumor suppression, I next investigate
the molecular mechanisms and cross-talks underlying DFO-induced PKCδ activation.
Herein, I demonstrate that PKCδ is proteolytically cleaved and translocated to the
nucleus in a time-dependent manner upon prolonged treatment of DFO. Specific
knockdown of the endogenous PKCδ by RNAi (shPKCδ) or expression of the kinase-
dead (Lys376Arg) mutant of PKCδ (PKCδKD) conferred a modulation on the cellular
adaptive responses to DFO-treatment. Notably, the time-dependent accumulation of
DFO-induced phosphorylation of Ser139-H2AX (γ-H2AX), hallmark for DNA damage,
was altered by knockdown of PKCδ and, shPKCδ completely abrogated the activation of
Caspase-3 in DFO-treated cells. Expression of PKCδKD-EGFP appears to abrogate
DFO-/hypoxia-induced activation of endogenous PKCδ and Caspase-3, suggesting that
PKCδKD-EGFP serves a dominant-negative function. Additionally, DFO-treatment also
led to the activation of Chk1, p53 and Akt, where DFO-induced activation of p53, Chk1
109
and Akt occurred in both PKCδ-dependent and -independent manners. In summary,
these findings suggest that the activation of PKCδ-mediated signaling network is one of
the critical contributing factors involved in fine-tuning of the DNA damage response to
prolonged DFO-treatment.
Recently, autophagy is found to be highly correlated to apoptotic cell death in terms
of the stress-induced regulatory responses. Autophagy is a tightly regulated intracellular
process for bulk degradation of cytoplasmic proteins or organelles that seems to be
essential for cell survival or death in response to stress conditions. Here I show that
treatment with DFO renders a rapid induction of autophagosome accumulation in Pa-4,
MEF/GFP-LC3 and MCF-7/Caspase-3 cells. PKC δ is rapidly and transient
phosphorylated at Y-64 and Y-155 in DFO-treated cells. Inhibition of PKC δ reduces the
changes in LC3-II levels, a hallmark of autophagic processing, in either Pa-4 cells stably
integrated with dominant-negative PKCδKD or MEF cells with PKCδ-null. Intriguingly,
the requirement of PKCδ is apparent for DFO-, but not starvation-, induced autophagy.
The importance of PKC δ in DFO-induced autophagy is further supported by my findings,
in that both JNK1 activation and release of Beclin-1 from inhibitory bcl-2 at 2 to 4 h post
DFO-treatment are PKC δ-dependent. Significantly, while inhibition of PKC δ-facilitated
autophagy by 3-MA or Atg5 knockout renders a more prevalence in cell death following
DFO-treatment, PKCδ- or JNK1-deficient cells exhibit resistance to long term exposure
of DFO. These results uncover a critical role for the early phase of PKC δ activation in
hypoxic stress-mediated autophagy induction and demonstrate the time-dependent dual
roles of PKCδ-dependent signaling in the cell fate determination.
110
In the overall scheme, acute hypoxia/DFO-treatment induces transient SUMOylation
and autophagy to protect cells against short-term hypoxic injury. However, if cells could
not overcome the damage under prolonged hypoxic condition, activation of
PKC δ/Caspase-3 signaling leads to apoptotic cell death.

Perspectives
In this dissertation, I have demonstrated that PKC δ plays a central role in governing
hypoxia- and/or DFO-induced salivary adaptive responses. This study helps us to further
understand PKC δ-regulated molecular and (patho)physiological mechanisms underlying
hypoxic conditions. For future perspectives, it is interesting to study that which upstream
tyrosine kinase is responsible to rapidly activate PKC δ and whether other tyrosine
residues of PKC δ are able to be phosphorylated in response to short-term 1% O
2
and/or
DFO-treatment. Furthermore, I have first uncovered that acute hypoxic treatment
induces autophagy in Pa-4 cells. However, the detail mechanism behind this finding is
still unclear. It is possible that PKC δ/JNK and some unknown molecules which are
pooled in the early endosomes play the essential role in the physiological scenarios which
is different from the chronic/prolonged hypoxia-induced activation pathway mediated by
HIF-1 α and BNIP3.

111
BIBLIOGRAPHY
1. Adhami, F., G. Liao, Y. M. Morozov, A. Schloemer, V. J. Schmithorst, J. N.
Lorenz, R. S. Dunn, C. V. Vorhees, M. Wills-Karp, J. L. Degen, R. J. Davis, N.
Mizushima, P. Rakic, B. J. Dardzinski, S. K. Holland, F. R. Sharp, and C. Y.
Kuan. 2006. Cerebral ischemia-hypoxia induces intravascular coagulation and
autophagy. Am J Pathol 169:566-83.
2. Azad, M. B., Y. Chen, E. S. Henson, J. Cizeau, E. McMillan-Ward, S. J. Israels,
and S. B. Gibson. 2008. Hypoxia induces autophagic cell death in apoptosis-
competent cells through a mechanism involving BNIP3. Autophagy 4:195-204.
3. Bae, S. H., J. W. Jeong, J. A. Park, S. H. Kim, M. K. Bae, S. J. Choi, and K. W.
Kim. 2004. Sumoylation increases HIF-1alpha stability and its transcriptional
activity. Biochem Biophys Res Commun 324:394-400.
4. Banan, A., J. Z. Fields, A. Farhadi, D. A. Talmage, L. Zhang, and A.
Keshavarzian. 2002. Activation of delta-isoform of protein kinase C is required for
oxidant-induced disruption of both the microtubule cytoskeleton and permeability
barrier of intestinal epithelia. J Pharmacol Exp Ther 303:17-28.
5. Bao, S., G. Ouyang, X. Bai, Z. Huang, C. Ma, M. Liu, R. Shao, R. M. Anderson,
J. N. Rich, and X. F. Wang. 2004. Periostin potently promotes metastatic growth
of colon cancer by augmenting cell survival via the Akt/PKB pathway. Cancer Cell
5:329-39.
6. Basu, A. 2003. Involvement of protein kinase C-delta in DNA damage-induced
apoptosis. J Cell Mol Med 7:341-50.
7. Beerepoot, L. V., D. T. Shima, M. Kuroki, K. T. Yeo, and E. E. Voest. 1996.
Up-regulation of vascular endothelial growth factor production by iron chelators.
Cancer Res 56:3747-51.
8. Benes, C. H., N. Wu, A. E. Elia, T. Dharia, L. C. Cantley, and S. P. Soltoff.
2005. The C2 domain of PKCdelta is a phosphotyrosine binding domain. Cell
121:271-80.
9. Bindra, R. S., P. J. Schaffer, A. Meng, J. Woo, K. Maseide, M. E. Roth, P.
Lizardi, D. W. Hedley, R. G. Bristow, and P. M. Glazer. 2004. Down-regulation
of Rad51 and decreased homologous recombination in hypoxic cancer cells. Mol
Cell Biol 24:8504-18.
10. Bohren, K. M., V. Nadkarni, J. H. Song, K. H. Gabbay, and D. Owerbach.
2004. A M55V polymorphism in a novel SUMO gene (SUMO-4) differentially
activates heat shock transcription factors and is associated with susceptibility to
type I diabetes mellitus. J Biol Chem 279:27233-8.
112
11. Bossis, G., and F. Melchior. 2006. Regulation of SUMOylation by reversible
oxidation of SUMO conjugating enzymes. Mol Cell 21:349-57.
12. Boyce, M., A. Degterev, and J. Yuan. 2004. Caspases: an ancient cellular sword
of Damocles. Cell Death Differ 11:29-37.
13. Brahimi-Horn, M. C., J. Chiche, and J. Pouyssegur. 2007. Hypoxia and cancer. J
Mol Med 85:1301-7.
14. Brodie, C., and P. M. Blumberg. 2003. Regulation of cell apoptosis by protein
kinase c delta. Apoptosis 8:19-27.
15. Brown, R. C., K. S. Mark, R. D. Egleton, J. D. Huber, A. R. Burroughs, and T.
P. Davis. 2003. Protection against hypoxia-induced increase in blood-brain barrier
permeability: role of tight junction proteins and NFkappaB. J Cell Sci 116:693-700.
16. Bunn, H. F., J. Gu, L. E. Huang, J. W. Park, and H. Zhu. 1998. Erythropoietin:
a model system for studying oxygen-dependent gene regulation. J Exp Biol
201:1197-201.
17. Buss, J. L., B. T. Greene, J. Turner, F. M. Torti, and S. V. Torti. 2004. Iron
chelators in cancer chemotherapy. Curr Top Med Chem 4:1623-35.
18. Buss, J. L., F. M. Torti, and S. V. Torti. 2003. The role of iron chelation in cancer
therapy. Curr Med Chem 10:1021-34.
19. Callapina, M., J. Zhou, T. Schmid, R. Kohl, and B. Brune. 2005. NO restores
HIF-1alpha hydroxylation during hypoxia: role of reactive oxygen species. Free
Radic Biol Med 39:925-36.
20. Chau, C. H., K. Y. Chen, H. T. Deng, K. J. Kim, K. Hosoya, T. Terasaki, H. M.
Shih, and D. K. Ann. 2002. Coordinating Etk/Bmx activation and VEGF
upregulation to promote cell survival and proliferation. Oncogene 21:8817-29.
21. Chau, C. H., C. A. Clavijo, H. T. Deng, Q. Zhang, K. J. Kim, Y. Qiu, A. D. Le,
and D. K. Ann. 2005. Etk/Bmx mediates expression of stress-induced adaptive
genes VEGF, PAI-1, and iNOS via multiple signaling cascades in different cell
systems. Am J Physiol Cell Physiol 289:C444-54.
22. Chen, E. Y., M. Fujinaga, and A. J. Giaccia. 1999. Hypoxic microenvironment
within an embryo induces apoptosis and is essential for proper morphological
development. Teratology 60:215-25.
23. Chen, X. L., A. Reindle, and E. S. Johnson. 2005. Misregulation of 2 microm
circle copy number in a SUMO pathway mutant. Mol Cell Biol 25:4311-20.
113
24. Cheng, J., X. Kang, S. Zhang, and E. T. Yeh. 2007. SUMO-specific protease 1 is
essential for stabilization of HIF1alpha during hypoxia. Cell 131:584-95.
25. Clavijo, C., J. L. Chen, K. J. Kim, M. E. Reyland, and D. K. Ann. 2007. Protein
kinase Cdelta-dependent and -independent signaling in genotoxic response to
treatment of desferroxamine, a hypoxia-mimetic agent. Am J Physiol Cell Physiol
292:C2150-60.
26. Comerford, K. M., M. O. Leonard, J. Karhausen, R. Carey, S. P. Colgan, and
C. T. Taylor. 2003. Small ubiquitin-related modifier-1 modification mediates
resolution of CREB-dependent responses to hypoxia. Proc Natl Acad Sci U S A
100:986-91.
27. Dang, C. V., and G. L. Semenza. 1999. Oncogenic alterations of metabolism.
Trends Biochem Sci 24:68-72.
28. Dawn, B., and R. Bolli. 2005. HO-1 induction by HIF-1: a new mechanism for
delayed cardioprotection? Am J Physiol Heart Circ Physiol 289:H522-4.
29. Dayani, P. N., M. C. Bishop, K. Black, and P. M. Zeltzer. 2004. Desferoxamine
(DFO)--mediated iron chelation: rationale for a novel approach to therapy for brain
cancer. J Neurooncol 67:367-77.
30. Degenhardt, K., R. Mathew, B. Beaudoin, K. Bray, D. Anderson, G. Chen, C.
Mukherjee, Y. Shi, C. Gelinas, Y. Fan, D. A. Nelson, S. Jin, and E. White. 2006.
Autophagy promotes tumor cell survival and restricts necrosis, inflammation, and
tumorigenesis. Cancer Cell 10:51-64.
31. Dempsey, E. C., A. C. Newton, D. Mochly-Rosen, A. P. Fields, M. E. Reyland,
P. A. Insel, and R. O. Messing. 2000. Protein kinase C isozymes and the
regulation of diverse cell responses. Am J Physiol Lung Cell Mol Physiol
279:L429-38.
32. Dendorfer, A., M. Heidbreder, T. Hellwig-Burgel, O. Johren, F. Qadri, and P.
Dominiak. 2005. Deferoxamine induces prolonged cardiac preconditioning via
accumulation of oxygen radicals. Free Radic Biol Med 38:117-24.
33. Denning, M. F., A. A. Dlugosz, D. W. Threadgill, T. Magnuson, and S. H.
Yuspa. 1996. Activation of the epidermal growth factor receptor signal
transduction pathway stimulates tyrosine phosphorylation of protein kinase C delta.
J Biol Chem 271:5325-31.
34. Denning, M. F., Y. Wang, B. J. Nickoloff, and T. Wrone-Smith. 1998. Protein
kinase Cdelta is activated by caspase-dependent proteolysis during ultraviolet
radiation-induced apoptosis of human keratinocytes. J Biol Chem 273:29995-30002.
114
35. Denning, M. F., Y. Wang, S. Tibudan, S. Alkan, B. J. Nickoloff, and J. Z. Qin.
2002. Caspase activation and disruption of mitochondrial membrane potential
during UV radiation-induced apoptosis of human keratinocytes requires activation
of protein kinase C. Cell Death Differ 9:40-52.
36. Desterro, J. M., M. S. Rodriguez, and R. T. Hay. 1998. SUMO-1 modification of
IkappaBalpha inhibits NF-kappaB activation. Mol Cell 2:233-9.
37. DeVries, T. A., M. C. Neville, and M. E. Reyland. 2002. Nuclear import of
PKCdelta is required for apoptosis: identification of a novel nuclear import
sequence. Embo J 21:6050-60.
38. Dohmen, R. J. 2004. SUMO protein modification. Biochim Biophys Acta
1695:113-31.
39. Donfrancesco, A., G. Deb, L. De Sio, R. Cozza, and A. Castellano. 1996. Role of
deferoxamine in tumor therapy. Acta Haematol 95:66-9.
40. Doulias, P. T., S. Christoforidis, U. T. Brunk, and D. Galaris. 2003. Endosomal
and lysosomal effects of desferrioxamine: protection of HeLa cells from hydrogen
peroxide-induced DNA damage and induction of cell-cycle arrest. Free Radic Biol
Med 35:719-28.
41. Egami, Y., and N. Araki. 2008. Characterization of Rab21-positive tubular
endosomes induced by PI3K inhibitors. Exp Cell Res 314:729-37.
42. Emoto, Y., Y. Manome, G. Meinhardt, H. Kisaki, S. Kharbanda, M. Robertson,
T. Ghayur, W. W. Wong, R. Kamen, R. Weichselbaum, and et al. 1995.
Proteolytic activation of protein kinase C delta by an ICE-like protease in apoptotic
cells. Embo J 14:6148-56.
43. Fandrey, J. 2004. Oxygen-dependent and tissue-specific regulation of
erythropoietin gene expression. Am J Physiol Regul Integr Comp Physiol
286:R977-88.
44. Fernandez-Capetillo, O., A. Lee, M. Nussenzweig, and A. Nussenzweig. 2004.
H2AX: the histone guardian of the genome. DNA Repair (Amst) 3:959-67.
45. Finigan, J. H., S. M. Dudek, P. A. Singleton, E. T. Chiang, J. R. Jacobson, S. M.
Camp, S. Q. Ye, and J. G. Garcia. 2005. Activated protein C mediates novel lung
endothelial barrier enhancement: role of sphingosine 1-phosphate receptor
transactivation. J Biol Chem 280:17286-93.

115
46. Gartsbein, M., A. Alt, K. Hashimoto, K. Nakajima, T. Kuroki, and T.
Tennenbaum. 2006. The role of protein kinase C delta activation and STAT3
Ser727 phosphorylation in insulin-induced keratinocyte proliferation. J Cell Sci
119:470-81.
47. Ghayur, T., M. Hugunin, R. V. Talanian, S. Ratnofsky, C. Quinlan, Y. Emoto,
P. Pandey, R. Datta, Y. Huang, S. Kharbanda, H. Allen, R. Kamen, W. Wong,
and D. Kufe. 1996. Proteolytic activation of protein kinase C delta by an ICE/CED
3-like protease induces characteristics of apoptosis. J Exp Med 184:2399-404.
48. Gibson, S. L., R. S. Bindra, and P. M. Glazer. 2005. Hypoxia-induced
phosphorylation of Chk2 in an ataxia telangiectasia mutated-dependent manner.
Cancer Res 65:10734-41.
49. Gillooly, D. J., I. C. Morrow, M. Lindsay, R. Gould, N. J. Bryant, J. M.
Gaullier, R. G. Parton, and H. Stenmark. 2000. Localization of
phosphatidylinositol 3-phosphate in yeast and mammalian cells. Embo J 19:4577-
88.
50. Gon, Y., M. R. Wood, W. B. Kiosses, E. Jo, M. G. Sanna, J. Chun, and H.
Rosen. 2005. S1P3 receptor-induced reorganization of epithelial tight junctions
compromises lung barrier integrity and is potentiated by TNF. Proc Natl Acad Sci
U S A 102:9270-5.
51. Gumbiner, B., T. Lowenkopf, and D. Apatira. 1991. Identification of a 160-kDa
polypeptide that binds to the tight junction protein ZO-1. Proc Natl Acad Sci U S A
88:3460-4.
52. Guo, D., M. Li, Y. Zhang, P. Yang, S. Eckenrode, D. Hopkins, W. Zheng, S.
Purohit, R. H. Podolsky, A. Muir, J. Wang, Z. Dong, T. Brusko, M. Atkinson,
P. Pozzilli, A. Zeidler, L. J. Raffel, C. O. Jacob, Y. Park, M. Serrano-Rios, M.
T. Larrad, Z. Zhang, H. J. Garchon, J. F. Bach, J. I. Rotter, J. X. She, and C.
Y. Wang. 2004. A functional variant of SUMO4, a new I kappa B alpha modifier,
is associated with type 1 diabetes. Nat Genet 36:837-41.
53. Gutcher, I., P. R. Webb, and N. G. Anderson. 2003. The isoform-specific
regulation of apoptosis by protein kinase C. Cell Mol Life Sci 60:1061-70.
54. Hamm-Alvarez, S. F., A. Chang, Y. Wang, G. Jerdeva, H. H. Lin, K. J. Kim,
and D. K. Ann. 2001. Etk/Bmx activation modulates barrier function in epithelial
cells. Am J Physiol Cell Physiol 280:C1657-68.
55. Hammond, E. M., N. C. Denko, M. J. Dorie, R. T. Abraham, and A. J. Giaccia.
2002. Hypoxia links ATR and p53 through replication arrest. Mol Cell Biol
22:1834-43.
116
56. Hammond, E. M., M. J. Dorie, and A. J. Giaccia. 2003. ATR/ATM targets are
phosphorylated by ATR in response to hypoxia and ATM in response to
reoxygenation. J Biol Chem 278:12207-13.
57. Hammond, E. M., M. J. Dorie, and A. J. Giaccia. 2004. Inhibition of ATR leads
to increased sensitivity to hypoxia/reoxygenation. Cancer Res 64:6556-62.
58. Hammond, E. M., and A. J. Giaccia. 2004. The role of ATM and ATR in the
cellular response to hypoxia and re-oxygenation. DNA Repair (Amst) 3:1117-22.
59. Hanada, T., N. N. Noda, Y. Satomi, Y. Ichimura, Y. Fujioka, T. Takao, F.
Inagaki, and Y. Ohsumi. 2007. The Atg12-Atg5 conjugate has a novel E3-like
activity for protein lipidation in autophagy. J Biol Chem 282:37298-302.
60. Harris, A. L. 2002. Hypoxia--a key regulatory factor in tumour growth. Nat Rev
Cancer 2:38-47.
61. Harrison, L., and K. Blackwell. 2004. Hypoxia and anemia: factors in decreased
sensitivity to radiation therapy and chemotherapy? Oncologist 9 Suppl 5:31-40.
62. Hay, R. T., L. Vuillard, J. M. Desterro, and M. S. Rodriguez. 1999. Control of
NF-kappa B transcriptional activation by signal induced proteolysis of I kappa B
alpha. Philos Trans R Soc Lond B Biol Sci 354:1601-9.
63. Hecht, G., C. Pothoulakis, J. T. LaMont, and J. L. Madara. 1988. Clostridium
difficile toxin A perturbs cytoskeletal structure and tight junction permeability of
cultured human intestinal epithelial monolayers. J Clin Invest 82:1516-24.
64. Hellwig-Burgel, T., D. P. Stiehl, A. E. Wagner, E. Metzen, and W. Jelkmann.
2005. Review: hypoxia-inducible factor-1 (HIF-1): a novel transcription factor in
immune reactions. J Interferon Cytokine Res 25:297-310.
65. Huang, T. T., S. M. Wuerzberger-Davis, Z. H. Wu, and S. Miyamoto. 2003.
Sequential modification of NEMO/IKKgamma by SUMO-1 and ubiquitin mediates
NF-kappaB activation by genotoxic stress. Cell 115:565-76.
66. Humphries, M. J., K. H. Limesand, J. C. Schneider, K. I. Nakayama, S. M.
Anderson, and M. E. Reyland. 2006. Suppression of Apoptosis in the Protein
Kinase C{delta} Null Mouse in Vivo. J Biol Chem 281:9728-37.
67. Humphries, M. J., K. H. Limesand, J. C. Schneider, K. I. Nakayama, S. M.
Anderson, and M. E. Reyland. 2006. Suppression of apoptosis in the protein
kinase Cdelta null mouse in vivo. J Biol Chem 281:9728-37.
68. Humphries, M. J., A. M. Ohm, J. Schaack, T. S. Adwan, and M. E. Reyland.
2007. Tyrosine phosphorylation regulates nuclear translocation of PKCdelta.
Oncogene 27:3045-53.
117
69. Ijichi, A., S. Sakuma, and P. J. Tofilon. 1995. Hypoxia-induced vascular
endothelial growth factor expression in normal rat astrocyte cultures. Glia 14:87-93.
70. Johnson, C. L., D. Lu, J. Huang, and A. Basu. 2002. Regulation of p53
stabilization by DNA damage and protein kinase C. Mol Cancer Ther 1:861-7.
71. Johnson, E. S. 2004. Protein modification by SUMO. Annu Rev Biochem 73:355-
82.
72. Joseloff, E., C. Cataisson, H. Aamodt, H. Ocheni, P. Blumberg, A. J. Kraker,
and S. H. Yuspa. 2002. Src family kinases phosphorylate protein kinase C delta on
tyrosine residues and modify the neoplastic phenotype of skin keratinocytes. J Biol
Chem 277:12318-23.
73. Kabeya, Y., N. Mizushima, T. Ueno, A. Yamamoto, T. Kirisako, T. Noda, E.
Kominami, Y. Ohsumi, and T. Yoshimori. 2000. LC3, a mammalian homologue
of yeast Apg8p, is localized in autophagosome membranes after processing. Embo J
19:5720-8.
74. Kabeya, Y., N. Mizushima, A. Yamamoto, S. Oshitani-Okamoto, Y. Ohsumi,
and T. Yoshimori. 2004. LC3, GABARAP and GATE16 localize to
autophagosomal membrane depending on form-II formation. J Cell Sci 117:2805-
12.
75. Kagawa, S., J. Gu, T. Honda, T. J. McDonnell, S. G. Swisher, J. A. Roth, and B.
Fang. 2001. Deficiency of caspase-3 in MCF7 cells blocks Bax-mediated nuclear
fragmentation but not cell death. Clin Cancer Res 7:1474-80.
76. Karantza-Wadsworth, V., S. Patel, O. Kravchuk, G. Chen, R. Mathew, S. Jin,
and E. White. 2007. Autophagy mitigates metabolic stress and genome damage in
mammary tumorigenesis. Genes Dev 21:1621-35.
77. Kaul, S., V. Anantharam, Y. Yang, C. J. Choi, A. Kanthasamy, and A. G.
Kanthasamy. 2005. Tyrosine phosphorylation regulates the proteolytic activation
of protein kinase Cdelta in dopaminergic neuronal cells. J Biol Chem 280:28721-30.
78. Kietzmann, T., A. Krones-Herzig, and K. Jungermann. 2002. Signaling cross-
talk between hypoxia and glucose via hypoxia-inducible factor 1 and glucose
response elements. Biochem Pharmacol 64:903-11.
79. Kitazawa, M., V. Anantharam, and A. G. Kanthasamy. 2003. Dieldrin induces
apoptosis by promoting caspase-3-dependent proteolytic cleavage of protein kinase
Cdelta in dopaminergic cells: relevance to oxidative stress and dopaminergic
degeneration. Neuroscience 119:945-64.
118
80. Klionsky, D. J., H. Abeliovich, P. Agostinis, D. K. Agrawal, G. Aliev, D. S.
Askew, M. Baba, E. H. Baehrecke, B. A. Bahr, A. Ballabio, B. A. Bamber, D. C.
Bassham, E. Bergamini, X. Bi, M. Biard-Piechaczyk, J. S. Blum, D. E.
Bredesen, J. L. Brodsky, J. H. Brumell, U. T. Brunk, W. Bursch, N.
Camougrand, E. Cebollero, F. Cecconi, Y. Chen, L. S. Chin, A. Choi, C. T.
Chu, J. Chung, P. G. Clarke, R. S. Clark, S. G. Clarke, C. Clave, J. L.
Cleveland, P. Codogno, M. I. Colombo, A. Coto-Montes, J. M. Cregg, A. M.
Cuervo, J. Debnath, F. Demarchi, P. B. Dennis, P. A. Dennis, V. Deretic, R. J.
Devenish, F. Di Sano, J. F. Dice, M. Difiglia, S. Dinesh-Kumar, C. W.
Distelhorst, M. Djavaheri-Mergny, F. C. Dorsey, W. Droge, M. Dron, W. A.
Dunn, Jr., M. Duszenko, N. T. Eissa, Z. Elazar, A. Esclatine, E. L. Eskelinen, L.
Fesus, K. D. Finley, J. M. Fuentes, J. Fueyo, K. Fujisaki, B. Galliot, F. B. Gao,
D. A. Gewirtz, S. B. Gibson, A. Gohla, A. L. Goldberg, R. Gonzalez, C.
Gonzalez-Estevez, S. Gorski, R. A. Gottlieb, D. Haussinger, Y. W. He, K.
Heidenreich, J. A. Hill, M. Hoyer-Hansen, X. Hu, W. P. Huang, A. Iwasaki, M.
Jaattela, W. T. Jackson, X. Jiang, S. Jin, T. Johansen, J. U. Jung, M.
Kadowaki, C. Kang, A. Kelekar, D. H. Kessel, J. A. Kiel, H. P. Kim, A. Kimchi,
T. J. Kinsella, K. Kiselyov, K. Kitamoto, E. Knecht, et al. 2008. Guidelines for
the use and interpretation of assays for monitoring autophagy in higher eukaryotes.
Autophagy 4:151-75.
81. Koriyama, H., Z. Kouchi, T. Umeda, T. C. Saido, T. Momoi, S. Ishiura, and K.
Suzuki. 1999. Proteolytic activation of protein kinase C delta and epsilon by
caspase-3 in U937 cells during chemotherapeutic agent-induced apoptosis. Cell
Signal 11:831-8.
82. Koumenis, C., R. Alarcon, E. Hammond, P. Sutphin, W. Hoffman, M. Murphy,
J. Derr, Y. Taya, S. W. Lowe, M. Kastan, and A. Giaccia. 2001. Regulation of
p53 by hypoxia: dissociation of transcriptional repression and apoptosis from p53-
dependent transactivation. Mol Cell Biol 21:1297-310.
83. Kurz, T., A. Leake, T. Von Zglinicki, and U. T. Brunk. 2004. Relocalized redox-
active lysosomal iron is an important mediator of oxidative-stress-induced DNA
damage. Biochem J 378:1039-45.
84. Lavin, M. F., G. Birrell, P. Chen, S. Kozlov, S. Scott, and N. Gueven. 2005.
ATM signaling and genomic stability in response to DNA damage. Mutat Res
569:123-32.
85. Le, N. T., and D. R. Richardson. 2002. The role of iron in cell cycle progression
and the proliferation of neoplastic cells. Biochim Biophys Acta 1603:31-46.
86. Lee, F. S., J. Hagler, Z. J. Chen, and T. Maniatis. 1997. Activation of the
IkappaB alpha kinase complex by MEKK1, a kinase of the JNK pathway. Cell
88:213-22.
119
87. Lee, J. W., S. H. Bae, J. W. Jeong, S. H. Kim, and K. W. Kim. 2004. Hypoxia-
inducible factor (HIF-1)alpha: its protein stability and biological functions. Exp
Mol Med 36:1-12.
88. Leitges, M., M. Mayr, U. Braun, U. Mayr, C. Li, G. Pfister, N. Ghaffari-
Tabrizi, G. Baier, Y. Hu, and Q. Xu. 2001. Exacerbated vein graft arteriosclerosis
in protein kinase Cdelta-null mice. J Clin Invest 108:1505-12.
89. Leverrier, S., A. Vallentin, and D. Joubert. 2002. Positive feedback of protein
kinase C proteolytic activation during apoptosis. Biochem J 368:905-13.
90. Levine, B., and G. Kroemer. 2008. Autophagy in the pathogenesis of disease. Cell
132:27-42.
91. Li, B., X. Wang, N. Rasheed, Y. Hu, S. Boast, T. Ishii, K. Nakayama, K. I.
Nakayama, and S. P. Goff. 2004. Distinct roles of c-Abl and Atm in oxidative
stress response are mediated by protein kinase C delta. Genes Dev 18:1824-37.
92. Li, L., J. Backer, A. S. Wong, E. L. Schwanke, B. G. Stewart, and M. Pasdar.
2003. Bcl-2 expression decreases cadherin-mediated cell-cell adhesion. J Cell Sci
116:3687-700.
93. Li, W., J. C. Yu, D. Y. Shin, and J. H. Pierce. 1995. Characterization of a protein
kinase C-delta (PKC-delta) ATP binding mutant. An inactive enzyme that
competitively inhibits wild type PKC-delta enzymatic activity. J Biol Chem
270:8311-8.
94. Li, Y., H. Wang, S. Wang, D. Quon, Y. W. Liu, and B. Cordell. 2003. Positive
and negative regulation of APP amyloidogenesis by sumoylation. Proc Natl Acad
Sci U S A 100:259-64.
95. Liang, X. H., S. Jackson, M. Seaman, K. Brown, B. Kempkes, H. Hibshoosh,
and B. Levine. 1999. Induction of autophagy and inhibition of tumorigenesis by
beclin 1. Nature 402:672-6.
96. Liu, J., Y. Minemoto, and A. Lin. 2004. c-Jun N-terminal protein kinase 1 (JNK1),
but not JNK2, is essential for tumor necrosis factor alpha-induced c-Jun kinase
activation and apoptosis. Mol Cell Biol 24:10844-56.
97. Liu, J., D. Yang, Y. Minemoto, M. Leitges, M. R. Rosner, and A. Lin. 2006.
NF-kappaB is required for UV-induced JNK activation via induction of PKCdelta.
Mol Cell 21:467-80.
98. Liu, Q., U. Moller, D. Flugel, and T. Kietzmann. 2004. Induction of plasminogen
activator inhibitor I gene expression by intracellular calcium via hypoxia-inducible
factor-1. Blood 104:3993-4001.
120
99. Llado, A., F. Tebar, M. Calvo, J. Moreto, A. Sorkin, and C. Enrich. 2004.
Protein kinaseCdelta-calmodulin crosstalk regulates epidermal growth factor
receptor exit from early endosomes. Mol Biol Cell 15:4877-91.
100. Lovejoy, D. B., and D. R. Richardson. 2003. Iron chelators as anti-neoplastic
agents: current developments and promise of the PIH class of chelators. Curr Med
Chem 10:1035-49.
101. Lukas, C., J. Falck, J. Bartkova, J. Bartek, and J. Lukas. 2003. Distinct
spatiotemporal dynamics of mammalian checkpoint regulators induced by DNA
damage. Nat Cell Biol 5:255-260.
102. Ma, T. Y., G. K. Iwamoto, N. T. Hoa, V. Akotia, A. Pedram, M. A. Boivin, and
H. M. Said. 2004. TNF-alpha-induced increase in intestinal epithelial tight junction
permeability requires NF-kappa B activation. Am J Physiol Gastrointest Liver
Physiol 286:G367-76.
103. Maddali, K. K., D. H. Korzick, D. L. Tharp, and D. K. Bowles. 2005. PKCdelta
mediates testosterone-induced increases in coronary smooth muscle Cav1.2. J Biol
Chem 280:43024-9.
104. Maiuri, M. C., G. Le Toumelin, A. Criollo, J. C. Rain, F. Gautier, P. Juin, E.
Tasdemir, G. Pierron, K. Troulinaki, N. Tavernarakis, J. A. Hickman, O.
Geneste, and G. Kroemer. 2007. Functional and physical interaction between Bcl-
X(L) and a BH3-like domain in Beclin-1. Embo J 26:2527-39.
105. Matassa, A. A., L. Carpenter, T. J. Biden, M. J. Humphries, and M. E.
Reyland. 2001. PKCdelta is required for mitochondrial-dependent apoptosis in
salivary epithelial cells. J Biol Chem 276:29719-28.
106. Matsui, Y., S. Kyoi, H. Takagi, C. P. Hsu, N. Hariharan, T. Ago, S. F. Vatner,
and J. Sadoshima. 2008. Molecular Mechanisms and Physiological Significance of
Autophagy during Myocardial Ischemia and Reperfusion. Autophagy 4:409-15.
107. Metzen, E., J. Zhou, W. Jelkmann, J. Fandrey, and B. Brune. 2003. Nitric oxide
impairs normoxic degradation of HIF-1alpha by inhibition of prolyl hydroxylases.
Mol Biol Cell 14:3470-81.
108. Minet, E., G. Michel, D. Mottet, M. Raes, and C. Michiels. 2001. Transduction
pathways involved in Hypoxia-Inducible Factor-1 phosphorylation and activation.
Free Radic Biol Med 31:847-55.
109. Mizushima, N., A. Yamamoto, M. Hatano, Y. Kobayashi, Y. Kabeya, K.
Suzuki, T. Tokuhisa, Y. Ohsumi, and T. Yoshimori. 2001. Dissection of
autophagosome formation using Apg5-deficient mouse embryonic stem cells. J Cell
Biol 152:657-68.
121
110. Murriel, C. L., E. Churchill, K. Inagaki, L. I. Szweda, and D. Mochly-Rosen.
2004. Protein kinase Cdelta activation induces apoptosis in response to cardiac
ischemia and reperfusion damage: a mechanism involving BAD and the
mitochondria. J Biol Chem 279:47985-91.
111. Myers, J. S., and D. Cortez. 2006. Rapid Activation of ATR by Ionizing Radiation
Requires ATM and Mre11. J Biol Chem 281:9346-50.
112. Nakano, H., M. Shindo, S. Sakon, S. Nishinaka, M. Mihara, H. Yagita, and K.
Okumura. 1998. Differential regulation of IkappaB kinase alpha and beta by two
upstream kinases, NF-kappaB-inducing kinase and mitogen-activated protein
kinase/ERK kinase kinase-1. Proc Natl Acad Sci U S A 95:3537-42.
113. Nelson, D. A., T. T. Tan, A. B. Rabson, D. Anderson, K. Degenhardt, and E.
White. 2004. Hypoxia and defective apoptosis drive genomic instability and
tumorigenesis. Genes Dev 18:2095-107.
114. Newton, A. C. 2001. Protein kinase C: structural and spatial regulation by
phosphorylation, cofactors, and macromolecular interactions. Chem Rev 101:2353-
64.
115. Newton, A. C. 2003. Regulation of the ABC kinases by phosphorylation: protein
kinase C as a paradigm. Biochem J 370:361-71.
116. Nguyen, H. V., J. L. Chen, J. Zhong, K. J. Kim, E. D. Crandall, Z. Borok, Y.
Chen, and D. K. Ann. 2006. SUMOylation Attenuates Sensitivity toward
Hypoxia- or Desferroxamine-Induced Injury by Modulating Adaptive Responses in
Salivary Epithelial Cells. Am J Pathol 168:1452-63.
117. Nusrat, A., M. Giry, J. R. Turner, S. P. Colgan, C. A. Parkos, D. Carnes, E.
Lemichez, P. Boquet, and J. L. Madara. 1995. Rho protein regulates tight
junctions and perijunctional actin organization in polarized epithelia. Proc Natl
Acad Sci U S A 92:10629-33.
118. Nusrat, A., C. von Eichel-Streiber, J. R. Turner, P. Verkade, J. L. Madara,
and C. A. Parkos. 2001. Clostridium difficile toxins disrupt epithelial barrier
function by altering membrane microdomain localization of tight junction proteins.
Infect Immun 69:1329-36.
119. Panse, V. G., U. Hardeland, T. Werner, B. Kuster, and E. Hurt. 2004. A
proteome-wide approach identifies sumoylated substrate proteins in yeast. J Biol
Chem 279:41346-51.
120. Pattingre, S., L. Espert, M. Biard-Piechaczyk, and P. Codogno. 2008.
Regulation of macroautophagy by mTOR and Beclin 1 complexes. Biochimie
90:313-23.
122
121. Pattingre, S., A. Tassa, X. Qu, R. Garuti, X. H. Liang, N. Mizushima, M.
Packer, M. D. Schneider, and B. Levine. 2005. Bcl-2 antiapoptotic proteins
inhibit Beclin 1-dependent autophagy. Cell 122:927-39.
122. Paul, S. A., J. W. Simons, and N. J. Mabjeesh. 2004. HIF at the crossroads
between ischemia and carcinogenesis. J Cell Physiol 200:20-30.
123. Petiot, A., E. Ogier-Denis, E. F. Blommaart, A. J. Meijer, and P. Codogno.
2000. Distinct classes of phosphatidylinositol 3'-kinases are involved in signaling
pathways that control macroautophagy in HT-29 cells. J Biol Chem 275:992-8.
124. Philchenkov, A. 2004. Caspases: potential targets for regulating cell death. J Cell
Mol Med 8:432-44.
125. Postovit, L. M., R. Sullivan, M. A. Adams, and C. H. Graham. 2005. Nitric
oxide signalling and cellular adaptations to changes in oxygenation. Toxicology
208:235-48.
126. Pouyssegur, J., F. Dayan, and N. M. Mazure. 2006. Hypoxia signalling in cancer
and approaches to enforce tumour regression. Nature 441:437-43.
127. Qi, X., A. Vallentin, E. Churchill, and D. Mochly-Rosen. 2007. deltaPKC
participates in the endoplasmic reticulum stress-induced response in cultured
cardiac myocytes and ischemic heart. J Mol Cell Cardiol 43:420-8.
128. Rami, A. 2008. Upregulation of Beclin 1 in the ischemic penumbra. Autophagy
4:227-9.
129. Redman, R. S. 1987. The salivary system, p. 1-20. In L. M. Sreebny (ed.),
Development of salivary glands. CRC press, Inc., Boca Raton, FL.
130. Reyland, M. E. 2007. Protein kinase Cdelta and apoptosis. Biochem Soc Trans
35:1001-4.
131. Reyland, M. E., S. M. Anderson, A. A. Matassa, K. A. Barzen, and D. O.
Quissell. 1999. Protein kinase C delta is essential for etoposide-induced apoptosis
in salivary gland acinar cells. J Biol Chem 274:19115-23.
132. Richardson, D. R. 1997. Potential of iron chelators as effective antiproliferative
agents. Can J Physiol Pharmacol 75:1164-80.
133. Richardson, D. R. 2002. Therapeutic potential of iron chelators in cancer therapy.
Adv Exp Med Biol 509:231-49.
134. Rubinsztein, D. C., J. E. Gestwicki, L. O. Murphy, and D. J. Klionsky. 2007.
Potential therapeutic applications of autophagy. Nat Rev Drug Discov 6:304-12.
123
135. Schmid, T., J. Zhou, and B. Brune. 2004. HIF-1 and p53: communication of
transcription factors under hypoxia. J Cell Mol Med 8:423-31.
136. Semenza, G. 2002. Signal transduction to hypoxia-inducible factor 1. Biochem
Pharmacol 64:993-8.
137. Semenza, G. L. 2000. HIF-1 and human disease: one highly involved factor. Genes
Dev 14:1983-91.
138. Semenza, G. L. 2001. HIF-1 and mechanisms of hypoxia sensing. Curr Opin Cell
Biol 13:167-71.
139. Semenza, G. L. 2001. HIF-1, O(2), and the 3 PHDs: how animal cells signal
hypoxia to the nucleus. Cell 107:1-3.
140. Semenza, G. L. 2004. Hydroxylation of HIF-1: oxygen sensing at the molecular
level. Physiology (Bethesda) 19:176-82.
141. Semenza, G. L. 2001. Hypoxia-inducible factor 1: oxygen homeostasis and disease
pathophysiology. Trends Mol Med 7:345-50.
142. Shao, R., F. P. Zhang, F. Tian, P. Anders Friberg, X. Wang, H. Sjoland, and H.
Billig. 2004. Increase of SUMO-1 expression in response to hypoxia: direct
interaction with HIF-1alpha in adult mouse brain and heart in vivo. FEBS Lett
569:293-300.
143. Shiloh, Y., Y. Andegeko, and I. Tsarfaty. 2004. In search of drug treatment for
genetic defects in the DNA damage response: the example of ataxia-telangiectasia.
Semin Cancer Biol 14:295-305.
144. Sitailo, L. A., S. S. Tibudan, and M. F. Denning. 2006. The protein kinase C delta
catalytic fragment targets Mcl-1 for degradation to trigger apoptosis. J Biol Chem
281:29703-10.
145. Solodukhin, A. S., R. H. Kretsinger, and J. J. Sando. 2007. Initial three-
dimensional reconstructions of protein kinase C delta from two-dimensional
crystals on lipid monolayers. Cell Signal 19:2035-45.
146. Song, J., L. K. Durrin, T. A. Wilkinson, T. G. Krontiris, and Y. Chen. 2004.
Identification of a SUMO-binding motif that recognizes SUMO-modified proteins.
Proc Natl Acad Sci U S A 101:14373-8.
147. Song, J., Z. Zhang, W. Hu, and Y. Chen. 2005. Small ubiquitin-like modifier
(SUMO) recognition of a SUMO binding motif: a reversal of the bound orientation.
J Biol Chem 280:40122-9.
124
148. Steffan, J. S., N. Agrawal, J. Pallos, E. Rockabrand, L. C. Trotman, N. Slepko,
K. Illes, T. Lukacsovich, Y. Z. Zhu, E. Cattaneo, P. P. Pandolfi, L. M.
Thompson, and J. L. Marsh. 2004. SUMO modification of Huntingtin and
Huntington's disease pathology. Science 304:100-4.
149. Steinberg, S. F. 2004. Distinctive activation mechanisms and functions for protein
kinase Cdelta. Biochem J 384:449-59.
150. Taylor, C. T., A. L. Dzus, and S. P. Colgan. 1998. Autocrine regulation of
epithelial permeability by hypoxia: role for polarized release of tumor necrosis
factor alpha. Gastroenterology 114:657-68.
151. Thelander, L., A. Graslund, and M. Thelander. 1983. Continual presence of
oxygen and iron required for mammalian ribonucleotide reduction: possible
regulation mechanism. Biochem Biophys Res Commun 110:859-65.
152. Thelander, L., and P. Reichard. 1979. Reduction of ribonucleotides. Annu Rev
Biochem 48:133-58.
153. Torgersen, M. L., S. Walchli, S. Grimmer, S. S. Skanland, and K. Sandvig.
2007. Protein kinase Cdelta is activated by Shiga toxin and regulates its transport. J
Biol Chem 282:16317-28.
154. Treins, C., S. Giorgetti-Peraldi, J. Murdaca, G. L. Semenza, and E. Van
Obberghen. 2002. Insulin stimulates hypoxia-inducible factor 1 through a
phosphatidylinositol 3-kinase/target of rapamycin-dependent signaling pathway. J
Biol Chem 277:27975-81.
155. Tsukita, S., and M. Furuse. 1999. Occludin and claudins in tight-junction strands:
leading or supporting players? Trends Cell Biol 9:268-73.
156. Tsukita, S., M. Furuse, and M. Itoh. 2001. Multifunctional strands in tight
junctions. Nat Rev Mol Cell Biol 2:285-93.
157. Twomey, C., and J. V. McCarthy. 2005. Pathways of apoptosis and importance in
development. J Cell Mol Med 9:345-59.
158. Vanhaesebroeck, B., S. J. Leevers, K. Ahmadi, J. Timms, R. Katso, P. C.
Driscoll, R. Woscholski, P. J. Parker, and M. D. Waterfield. 2001. Synthesis and
function of 3-phosphorylated inositol lipids. Annu Rev Biochem 70:535-602.
159. Vivanco, I., N. Palaskas, C. Tran, S. P. Finn, G. Getz, N. J. Kennedy, J. Jiao, J.
Rose, W. Xie, M. Loda, T. Golub, I. K. Mellinghoff, R. J. Davis, H. Wu, and C.
L. Sawyers. 2007. Identification of the JNK signaling pathway as a functional
target of the tumor suppressor PTEN. Cancer Cell 11:555-69.
125
160. Voss, O. H., S. Kim, M. D. Wewers, and A. I. Doseff. 2005. Regulation of
monocyte apoptosis by the protein kinase Cdelta-dependent phosphorylation of
caspase-3. J Biol Chem 280:17371-9.
161. Wang, Q. J., D. Bhattacharyya, S. Garfield, K. Nacro, V. E. Marquez, and P.
M. Blumberg. 1999. Differential localization of protein kinase C delta by phorbol
esters and related compounds using a fusion protein with green fluorescent protein.
J Biol Chem 274:37233-9.
162. Wang, S., M. Guo, H. Ouyang, X. Li, C. Cordon-Cardo, A. Kurimasa, D. J.
Chen, Z. Fuks, C. C. Ling, and G. C. Li. 2000. The catalytic subunit of DNA-
dependent protein kinase selectively regulates p53-dependent apoptosis but not cell-
cycle arrest. Proc Natl Acad Sci U S A 97:1584-8.
163. Weinacker, A. B., and L. T. Vaszar. 2001. Acute respiratory distress syndrome:
physiology and new management strategies. Annu Rev Med 52:221-37.
164. Wheaton, K., and K. Riabowol. 2004. Protein kinase C delta blocks immediate-
early gene expression in senescent cells by inactivating serum response factor. Mol
Cell Biol 24:7298-311.
165. Wohlschlegel, J. A., E. S. Johnson, S. I. Reed, and J. R. Yates, 3rd. 2004. Global
analysis of protein sumoylation in Saccharomyces cerevisiae. J Biol Chem
279:45662-8.
166. Yoon, Y. S., H. Cho, J. H. Lee, and G. Yoon. 2004. Mitochondrial dysfunction
via disruption of complex II activity during iron chelation-induced senescence-like
growth arrest of Chang cells. Ann N Y Acad Sci 1011:123-32.
167. Yoshida, K. 2007. PKCdelta signaling: mechanisms of DNA damage response and
apoptosis. Cell Signal 19:892-901.
168. Yoshida, K., H. Liu, and Y. Miki. 2006. Protein kinase C delta regulates Ser46
phosphorylation of p53 tumor suppressor in the apoptotic response to DNA damage.
J Biol Chem 281:5734-40.
169. Yoshida, K., Y. Miki, and D. Kufe. 2002. Activation of SAPK/JNK signaling by
protein kinase Cdelta in response to DNA damage. J Biol Chem 277:48372-8.
170. Yoshida, K., H. G. Wang, Y. Miki, and D. Kufe. 2003. Protein kinase Cdelta is
responsible for constitutive and DNA damage-induced phosphorylation of Rad9.
Embo J 22:1431-41.
171. Yu, J. H., M. R. Mark, and R. S. Redman. 1987. Effect of clonidine on secretion
of fluid and ions by the parotid and submandibular glands of the rat. Arch Oral Biol
32:27-33.
126
172. Yuan, J. 2006. Divergence from a dedicated cellular suicide mechanism: exploring
the evolution of cell death. Mol Cell 23:1-12.
173. Yuan, L. W., J. W. Soh, and I. B. Weinstein. 2002. Inhibition of histone
acetyltransferase function of p300 by PKCdelta. Biochim Biophys Acta 1592:205-
11.
174. Zagorska, A., and J. Dulak. 2004. HIF-1: the knowns and unknowns of hypoxia
sensing. Acta Biochim Pol 51:563-85.
175. Zeng, X., J. H. Overmeyer, and W. A. Maltese. 2006. Functional specificity of
the mammalian Beclin-Vps34 PI 3-kinase complex in macroautophagy versus
endocytosis and lysosomal enzyme trafficking. J Cell Sci 119:259-70.
176. Zhou, F., Y. Xue, H. Lu, G. Chen, and X. Yao. 2005. A genome-wide analysis of
sumoylation-related biological processes and functions in human nucleus. FEBS
Lett 579:3369-75.
177. Zhou, L., and M. B. Hershenson. 2003. Mitogenic signaling pathways in airway
smooth muscle. Respir Physiol Neurobiol 137:295-308.

Abstract (if available)

Abstract Hypoxia is a physiological condition defined as the oxygen concentration is lower than 5% in general while normoxia is around 20%. It has been reported that hypoxia is related to numerous human diseases, such as respiratory diseases, vascular diseases, neurodegeneration and cancer, and hypoxic stress activates various signal transduction pathways to regulate cellular adaptive responses. Desferroxamine (DFO), a hypoxia-mimetic agent, functions as an iron chelator and has been utilized in hypoxic study. However, the molecular mechanisms underlying hypoxia- and/or DFO-induced adaptive responses are still unclear. In this report, I demonstrate that hypoxia and DFO induce a transient global SUMOylation to augment cell survival against acute hypoxic stress by strengthening passive barrier properties, facilitating the reassembly of F-actin/ZO-1, attenuating pro-apoptotic PKCdelta/Caspase-3 activation, inducing S1981 phosphorylation of ATM, and activating pro-survival NF-kappaB signaling pathways. Under prolonged hypoxic treatment, DFO induces nuclear translocation and proteolytic cleavage of PKCdelta and PKCdelta-dependent Caspase-3 activation to render a sustained DFO-elicited gamma-H2AX activation leading to apoptotic cell death. Intriguingly, PKCdelta plays a fine-tuning role in modulating DFO-induced Akt phosphorylation. Moreover, DFO-exposure also induces a PKCdelta-independent signaling and both PKCdelta-dependent and -independent pathways functionally cooperate to integrate pro-apoptotic/Caspase-3, pro-survival/Akt, and DNA damage-induced DNA repair/cell cycle regulation signalings.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Bimodal effects of bone morphogenetic proteins in prostate cancer

PDF

Targeting cellular redox modulations for pancreatic cancer treatment

PDF

Translational regulation and endosomal trafficking in synaptic adaptation to stress

PDF

PTEN loss antagonizes aging through promoting regeneration and prevents oxidative stress induced cell death

PDF

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

PDF

To adapt or not to adapt: the age-specific and sex-dependent differences in the adaptive stress response

PDF

Differential effects of starvation in normal and cancer cells: from EGR1-dependent protection to p53-mediated apoptosis

PDF

Modulation of transcription and receptor function with synthetic small molecules and multi-finctional integrin ligands

PDF

Development of novel small molecules targeting mitochondrial and oxidative stress signaling pathways for pancreatic cancer therapy

Asset Metadata

Creator Chen, Jo-Lin (author)

Core Title Hypoxia-induced adaptive responses

School School of Pharmacy

Degree Doctor of Philosophy

Degree Program Molecular Pharmacology

Publication Date 07/21/2008

Defense Date 04/17/2008

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

Tag apoptosis,autophagy,desferroxamine,DNA damage,hypoxia,OAI-PMH Harvest,protein kinase C delta,signal transduction,SUMOylation

Language English

Advisor Shen, Wei-Chiang (committee chair), Ann, David K. (committee member), Hamm-Alvarez, Sarah F. (committee member), Kim, Kwang-Jin (committee member), Ou, Jing-Hsiung James (committee member)

Creator Email jolinchen118@yahoo.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m1366

Unique identifier UC1120018

Identifier etd-Chen-20080721 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-194733 (legacy record id),usctheses-m1366 (legacy record id)

Legacy Identifier etd-Chen-20080721.pdf

Dmrecord 194733

Document Type Dissertation

Rights Chen, Jo-Lin

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