Embed Size (px)
Citation preview
Role of Receptor Tyrosine Kinase Regulator Sprouty
in Ovarian Cancer Cells
by
Wai Kin So
BSc, The Chinese University of Hong Kong, 2002
MPhil, The Chinese University of Hong Kong, 2004
A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF
THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
in
The Faculty of Graduate Studies
(Reproductive and Developmental Sciences)
THE UNIVERSITY OF BRITISH COLUMBIA
(Vancouver)
April, 2012
© Wai Kin So 2012
ii
Abstract
Aberrant epidermal growth factor receptor (EGFR) activity contributes to the
development of epithelial ovarian cancer (EOC), a common and lethal female
malignancy. Elucidating the regulation of EGFR function will improve treatments for
EOC and the survival of patients.
This study aims to elucidate the role of Sprouty (SPRY) proteins, which are EGFR
regulators, in EOC. The investigation began with demonstrating the downregulation of
mRNA levels of two SPRY members, SPRY2 and SPRY4, in EOC tissues and/or cell lines.
Deletion of the SPRY2 gene was found to cause reduced SPRY2 mRNA. Loss of the
SPRY2 gene and thus its expression are particularly common in high-grade serous tumors,
suggesting that SPRY2 deficiency may be involved in the pathogenesis of this prevailing
subtype of EOC.
The regulatory mechanisms of SPRY level are incompletely understood. The
EGFR ligand EGF strongly upregulates SPRY4 protein level primarily through the ERK
pathway. In addition, the PI3K/AKT pathway and hypoxia-inducible factor-1 (HIF-1α)
have been shown to be involved in SPRY4 regulation, allowing the possibility that
SPRY4 is regulated by micro-environmental (hypoxia) and genetic (PI3K mutation)
abnormalities.
Functionally, SPRY2 and SPRY4 counteract various aspects of EGFR activity and
generally have tumor suppressor functions. First, in contrast to the EGFR, SPRY2 and
SPRY4 prevent loss of cell adhesion by E-cadherin and therefore suppress cancer cell
invasion. Second, SPRY4 inhibits PI3K/AKT signalling activated by EGF, as AKT
iii
activation is enhanced in the absence of SPRY4. Finally, the HIF-1α oncogene has been
identified as a novel SPRY4 target. In ovarian cancer cell lines, SPRY4 suppresses the
basal and EGF-stimulated expression of HIF-1α. The negative effects of SPRY4 on HIF-
1α are also reflected by modulation of HIF-1 activity and target gene expression. SPRY4
has also been shown to destabilise HIF-1α protein, independent of the classic HIF-1α
degradation pathway.
The current study investigated the expression, regulation and function of SPRY in
ovarian cancer. Understanding the tumor suppressor role of SPRY will not only enhance
our knowledge about the pathophysiology of ovarian cancer but also identifies a possible
therapeutic intervention against this lethal malignancy.
iv
Preface
1. A version of Chapter 2 has been submitted and is under revision. Wai-Kin So, Alice S.
T. Wong, David G. Huntsman, C. Blake Gilks, and Peter C. K. Leung. Genetic
Inactivation of Sprouty2 Promotes Epidermal Growth Factor-induced E-cadherin
Downregulation and Invasion in Ovarian Cancer Cells.
Contributions
David G. Huntsman, C. Blake Gilks and I participated in design and performance of the
study. I drafted the manuscript. Peter C. K. Leung, Alice S. T. Wong, David G.
Huntsman and C. Blake Gilks read, critically revised and approved the manuscript.
Ethics Approval
Approvals for the study were obtained from the University of British Columbia Research
Ethics Board (#H04-60102, #H02-61375 and #H03-70606) and written informed
consents from all participants involved in the study were obtained.
2. A manuscript based on a version of Chapter 3 has been under preparation. Wai-Kin So,
Man-Tat Lau, and Peter C. K. Leung. An Amphiregulin and Sprouty4 Loop Regulates
Ovarian Cancer Cell Invasiveness Via and E-cadherin-dependent Mechanism.
Contributions
I participated in design of the study, Man-Tat Lau and I performed experiments. I drafted
the manuscript.
v
Presentation
5th International Epithelial-Mesenchymal Transition Meeting, Singapore, Oct. 2011
Poster presentation on An Amphiregulin and Sprouty4 Loop Regulates Ovarian Cancer
Cell Invasiveness Via and E-cadherin-dependent Mechanism. Wai-Kin So and Peter C.
K. Leung.
vi
Table of Contents
Abstract ............................................................................................................................... ii!Preface................................................................................................................................ iv!Table of Contents............................................................................................................... vi!List of Tables ..................................................................................................................... ix!List of Figures ..................................................................................................................... x!List of Abbreviations ........................................................................................................ xii!Acknowledgements.......................................................................................................... xvi!Chapter 1 Introduction ........................................................................................................ 1!
1.1 Ovarian cancer .......................................................................................................... 1!1.2 Tumorigenesis of EOCs............................................................................................ 1!1.3 Classification of EOCs.............................................................................................. 3!
1.3.1 High-grade serous carcinoma (HGSC) .............................................................. 3!1.3.2 Low-grade serous carcinoma (LGSC) ............................................................... 5!1.3.3 Endometrioid carcinoma (EC) ........................................................................... 6!1.3.4 Clear cell carcinoma (CCC)............................................................................... 7!1.3.5 Mucinous tumors ............................................................................................... 7!
1.4 Epidermal growth factor (EGF) and the EGF receptor (EGFR)............................... 8!1.4.1 EGFR structure and signaling............................................................................ 8!1.4.2 EGFR expression and mutations in ovarian cancer ........................................... 9!1.4.3 EGFR ligands................................................................................................... 11!1.4.4 Functions of the EGFR and it ligands in ovarian cancer ................................. 13!1.4.5 Clinical activities of EGFR-targeted therapies. ............................................... 15!
1.5 Sprouty (SPRY) ...................................................................................................... 16!1.5.1 SPRY structure................................................................................................. 16!1.5.2 SPRY functions and mechanisms .................................................................... 17!1.5.3 Regulation of SPRY activity............................................................................ 18!1.5.4 SPRY expressions in cancer ............................................................................ 19!1.5.5 SPRY as tumor suppressor genes .................................................................... 19!
1.6 Hypoxia-inducible factor-1 alpha (HIF-1α)............................................................ 19!1.6.1 HIF-1α structure............................................................................................... 20!1.6.2 HIF-1α regulation ............................................................................................ 21!
1.6.2.1 Hypoxia..................................................................................................... 21!1.6.2.2 Regulation of HIF-1α in normoxia........................................................... 22!
1.6.2.2.1 VHL mutations ................................................................................... 22!1.6.2.2.2 AKT and PI3K ................................................................................... 23!1.6.2.2.3 Glycogen synthase kinase 3β (GSK3β) ............................................. 25!1.6.2.2.4 Heat shock protein 90 (HSP90) ......................................................... 26!1.6.2.2.5 Hormonal regulation .......................................................................... 27!
1.6.3 HIF-1α expressions in cancer .......................................................................... 27!1.6.4 HIF-1α functions in cancer.............................................................................. 28!
vii
1.6.4.1 Angiogenesis............................................................................................. 28!1.6.4.2 Metastasis.................................................................................................. 29!
1.7 Hypothesis and objectives....................................................................................... 30!2.1 Introduction............................................................................................................. 37!2.2 Materials and methods ............................................................................................ 39!
2.2.1 The Human Exonic Evidence-Based Oligonucleotide microarray (HEEBO). 39!2.2.2 Molecular inversion probe (MIP) copy number analysis ................................ 39!2.2.3 The Cancer Genome Atlas (TCGA) ................................................................ 40!2.2.5 Real-time PCR ................................................................................................. 41!2.2.6 Antibodies ........................................................................................................ 41!2.2.7 Invasion assay .................................................................................................. 42!2.2.8 Statistical analysis............................................................................................ 42!
2.3 Results..................................................................................................................... 43!2.3.1 Levels of SPRY mRNA in ovarian tumors of different pathological subtypes 43!2.3.2 Levels of SPRY mRNA in immortalised ovarian surface epithelium (IOSE) and EOC-derived cell lines.............................................................................................. 44!2.3.3 A deletion event in the proximity of the SPRY2 locus..................................... 44!2.3.4 SPRY2 deletion may lead to reduced SPRY2 mRNA level............................. 45!2.3.5 SPRY2 reversed EGF-suppressed E-cadherin protein expression and antagonized EGF-induced cell invasion ................................................................... 46!2.3.6 SPRY2 and E-cadherin proteins displayed a positive correlation in human ovarian cancer cell lines and tumors......................................................................... 47!
2.4 Discussion ............................................................................................................... 47!Chapter 3 An amphiregulin and Sprouty4 loop regulates ovarian cancer cell invasiveness via an E-cadherin-dependent mechanism ......................................................................... 60!
3.1 Introduction............................................................................................................. 60!3.2 Materials and methods ............................................................................................ 62!
3.2.1 Cell culture and reagents.................................................................................. 62!3.2.2 Transfection ..................................................................................................... 63!3.2.3 Real-time PCR ................................................................................................. 63!3.2.4 Western blot analysis ....................................................................................... 64!3.2.5 Invasion assay .................................................................................................. 64!3.2.6 Statistical analysis............................................................................................ 65!
3.3 Results..................................................................................................................... 65!3.3.1 AREG promoted invasion of ovarian cancer cells........................................... 65!3.3.2 AREG reduced E-cadherin levels, and E-cadherin overexpression blocks AREG-induced invasion ........................................................................................... 66!3.3.3 AREG suppressed E-cadherin level and promotes cell invasion via the EGFR................................................................................................................................... 66!3.3.4 AREG induced Slug expression....................................................................... 67!3.3.5 The MAPK/ERK and PI3K/AKT pathways mediated the effects of AREG on SLUG mRNA and E-cadherin levels and cell invasion ............................................ 67!3.3.6 AREG induced SPRY4 expression.................................................................. 67!3.3.7 SPRY4 knockdown enhanced AREG-induced E-cadherin suppression and invasion ..................................................................................................................... 68!
3.4 Discussion ............................................................................................................... 68!
viii
Chapter 4 Sprouty4 feedback regulates epidermal growth factor/AKT/hypoxia-inducible factor-1 alpha axis in ovarian cancer cells ........................................................................ 82!
4.1 Introduction............................................................................................................. 82!4.2 Materials and methods ............................................................................................ 84!
4.2.1 Cell culture and reagents.................................................................................. 84!4.2.2 Transfection ..................................................................................................... 84!4.2.3 Real-time PCR ................................................................................................. 85!4.2.4 Western blot analysis ....................................................................................... 85!4.2.5 Luciferase assay ............................................................................................... 86!4.2.6 Statistical analyses ........................................................................................... 86!
4.3 Results..................................................................................................................... 87!4.3.1 EGF increased SPRY4 in ovarian cancer cells ................................................ 87!4.3.2 The MEK/ERK and PI3K/AKT pathways mediated the effects of EGF on SPRY4 levels ............................................................................................................ 87!4.3.3 EGF induced HIF-1α via the PI3K/AKT pathway .......................................... 88!4.3.4 HIF-1α plays a minor role in EGF-induced SPRY4 level ............................... 88!4.3.5 SPRY4 overexpression reversed EGF-induced HIF-1α levels and HIF-1 activity....................................................................................................................... 89!4.3.6 SPRY4 knockdown enhanced the effect of EGF on HIF-1α ........................... 89!4.3.7 AKT pathway mediated HIF-1α regulation by EGF and SPRY4.................... 90!
4.4 Discussion ............................................................................................................... 90!5.1 Introduction........................................................................................................... 103!5.2 Materials and methods .......................................................................................... 105!
5.2.1 Cell culture and reagents................................................................................ 105!5.2.2 Transfection ................................................................................................... 105!5.2.3 Real-time PCR ............................................................................................... 106!5.2.4 Western blot analysis ..................................................................................... 106!5.2.5 Statistical analysis.......................................................................................... 107!
5.3 Results................................................................................................................... 107!5.3.1 SPRY4 negatively regulated HIF-1α expression levels in ovarian cancer cells................................................................................................................................. 107!5.3.2 SPRY4 negatively regulated HIF-1 activity .................................................. 108!5.3.3 SPRY4 regulated HIF-1α protein half-life without affecting Hif-1α mRNA levels ....................................................................................................................... 109!5.3.4 HIF-1α modulation by SPRY4 was independent of PHD activity ................ 109!
5.4 Discussion ............................................................................................................. 110!References....................................................................................................................... 129!
ix
List of Tables
Table 2.1 Comparison between mean SPRY2 mRNA levels of various histopathological types using Student’s t test..……….…………………………….…………………...…..51
Table 2.2 MIP analysis of loss of the markers flanking the SPRY2 and SPRY4 loci in ovarian tumors………………………………………………………………………..….52
x
List of Figures
Figure 1.1 The chart illustrates the pathogenesis of epithelial ovarian cancers…..……..33
Figure 1.2 The structure and signaling of EGFR.…………..……...…………………….34
Figure 1.3 A schematic diagram of human SPRY depicting its structure, its interacting partners and the corresponding functional consequences. ………………………………35
Figure 1.4 The diagram illustrates the regulation of HIF-1α. ……………………..……36
Figure 2.1 The box plot displays the mean SPRY2 mRNA levels in serous (high-grade, low-grade or borderline), endometrioid (carcinoma or borderline tumor), clear cell ovarian tumors and normal Fallopian tube. (HEEBO array data)……………………….54
Figure 2.2 Comparison of SPRY2 and SPRY4 mRNA levels in immortalised OSE (IOSE) and ovarian cancer cell lines by real-time PCR...……………………………………..…55
Figure 2.3 A schematic representation of the MIP copy number assay results of 28 high-grade serous carcinomas...………………………………………………………...….….56
Figure 2.4 The effect of SPRY2 on EGF-induced E-cadherin suppression and cell invasion.…………………...……………………………………………………………..57
Figure 2.5 The correlation between SPRY2 and E-cadherin protein expression in ovarian cancers..…………………………………………………..…………………………..….59
Figure 3.1 AREG promoted ovarian cancer cells invasion.….……………………..…...73
Figure 3.2 AREG reduced E-cadherin levels and E-cadherin overexpression blocked AREG-induced invasion. ………………..………………………………………………74
Figure 3.3 AREG suppressed E-cadherin level and promoted cell invasion via the EGFR. ………………………………………………………………………………….………..75
Figure 3.4 AREG induced SLUG mRNA ……………………………….……………....76
Figure 3.5 The MAPK/ERK and PI3K/AKT pathways mediated the effects of AREG on Slug and E-cadherin level and cell invasion. …………………………………………....77
Figure 3.6 AREG induced SPRY4…………………………………………………..…..79
Figure 3.7 SPRY4 knockdown enhanced AREG-induced E-cadherin suppression and invasion..……………………………………………...…………..……………..……….80
Figure 4.1 EGF induced SPRY4 level in ovarian cancer cells...…………………..…….94
Figure 4.2 The MEK/ERK and PI3K/AKT pathways mediated the effects of EGF on SPRY4 levels..………………………………………………………...…………………95
xi
Figure 4.3 EGF induced HIF-1α and HIF-1 activity via the PI3K/AKT pathway.………………………………………………………………………………….96
Figure 4.4 HIF-1α plays a minor role in EGF-induced SPRY4 level..…...……………..97
Figure 4.5 SPRY4 overexpression reversed EGF-induced HIF-1α expression and HIF-1 activity.……………………………………………………………………….………..…99
Figure 4.6 SPRY4 knockdown enhanced EGF effect on HIF-1α...…………..…..........101
Figure 4.7 The PI3K/AKT pathway mediated HIF-1α regulation by EGF and SPRY4……………………………………………………………………………….....102
Figure 5.1 SPRY4 negatively regulated HIF-1α levels in ovarian cancer cells. …………………………………………………………………………………...….…..113
Figure 5.2 SPRY4 negatively regulated HIF-1 activity. ……………………….………114
Figure 5.3 SPRY4 regulated HIF-1α protein half-life without affecting Hif-1α mRNA levels. ……...………………………………...…….……………………….…..............115
Figure 5.4 HIF-1α modulation by SPRY4 acts independently of PHD activity. …………………………………………………………………………………………..117
Figure 6.1 The diagram summarizes the findings. ……………………………………..128
xii
List of Abbreviations
AREG Amphiregulin
BRCA1 Breast cancer 1, early onset
BTC Betacellulin
CCC Clear cell carcinoma
CTNNB Cadherin-associated protein beta
DM Double mutant
DMSO Dimethyl sulfoxide
EC Endometrioid carcinoma
ECD Extracellular domain
ECM Extracellular matrix
EGF Epidermal growth factor
EMT Epithelial-mesenchymal transition
EOC Epithelial ovarian cancer
EPI Epiregulin
ERBB Erythroblastic leukemia viral oncogene homolog
ERK Extracellular signal regulated protein kinase
FBS Fetal bovine serum
FGF Fibroblast growth factor
GAPDH Gylceraldehyde-3-phosphate dehydrogenase
GIST Gastrointestinal stromal tumor
GSK-3 Glycogen synthase kinase-3
xiii
HB-EGF Heparin-binding epidermal growth factor
HEEBO Human Exonic Evidence-Based Oligonucleotide
HER2 Human epidermal growth factor receptor 2
HGF Hepatocyte growth factor
HGSC High-grade serous carcinoma
HIF-1α Hypoxia-inducible factor-1 alpha
HIF-1β Hypoxia-inducible factor-1 beta
HRE Hypoxia responsive element
HRG Heregulin
HSP Heat shock protein
ICM Intracellular domain
IGF-I Insulin-like growth factor-I
IOSE Immortalized ovarian surface epithelium
JM Juxtamembrane domain
JNK Jun N-terminal protein kinase
LGSC Low-grade serous carcinoma
LH Luteinizing hormone
LOH Loss of heterozygosity
mAb Monoclonal antibody
MAPK Mitogen-activated protein kinase
MDCK Madin-darby canine kidney
MDM2 Murine double minute 2
xiv
MET Mesenchymal epithelial transition factor
MIP Molecular inversion probe
MMP Matrix metalloproteinase
mTOR Mammalian target of rapamycin
NSCLC Non-small-cell lung cancer
ODD Oxygen-dependent degradation
OSE Ovarian surface epithelium
PAI-1 Plasminogen activator inhibitor-1
PDGF Platelet-derived growth factor
PI3K Phosphatidylinositol 3-kinase
PKA Protein kinase A
PHD Prolyl hydroxylase
PTEN Phosphatase and tensin homolog deleted on chromosome 10
PVDF Polyvinylidene fluoride
RAS Rat sarcoma
RBD RAF1-binding domain
RCC Renal clear cell carcinoma
RD Regulatory domain
RING Really interesting new gene
ROS Reactive oxygen species
RTK Receptor tyrosine kinase
SD Standard deviation
SDS Sodium dodecyl sulphate
xv
siRNA Small interference RNA
SIAH Seven in absentia homolog
SNP Sodium nitroprusside
SOS Son of Sevenless
STAT Signal transducer and activation of transcription
STIC Serous tubal intraepithelial carcinoma
Sp1 Stimulating protein 1
SPRY Sprouty
TCGA The Cancer Genome Atlas
TESK1 Testicular protein kinase 1
TGFα Transforming growth factor alpha
TGFβ Transforming growth factor beta
TKD Tyrosine kinase domain
TMD Transmembrane domain
TKI Tyrosine kinase inhibitor
uPA Urokinase-type plasminogen activator
VEGF Vascular endothelial growth factor
VHL Von Hippel-Lindau
WT-1 Wilms tumor 1
WT Wild-type
ZEB1 Zinc finger E-box-binding homeobox 1
xvi
Acknowledgements
First of all, I would like to express my greatest gratitude to my supervisor, Dr. Peter C.K.
Leung for his invaluable support, patient guidance and warm encouragement during my
study. In addition, I would like to express my sincere gratitude to my supervisory
committee members, Drs. Geoffrey Hammond, Blake Gilks, Y.Z. Wang and Mark Carey
for their scientific criticisms and advice and my experiments and thesis. I would also
want to thanks Drs. Blake Gilks, David Huntsman and Alice S.T. Wong from the
University of Hong Kong for their help and opinions of my experiments.
I also would like to thank Dr. Christian Klausen and Ms. Roshni Nair for their help over
the years.
Here I also thank the Interdisciplinary Women’s Reproductive Health Research Training
Program providing me scholarship.
Lastly, I wish to thank my parents and family members for their love, understanding and
endless support throughout my study.
1
Chapter 1 Introduction
1.1 Ovarian cancer
Ovarian cancer is the second most prevalent gynaecological malignancy (after
endometrial cancer) among women in the United States (1). A woman’s lifetime risk of
developing ovarian cancer is 1 in 70 (2). Ovarian cancer is the most lethal of all
gynaecologic malignancies with a 5-year survival rate of 30% – 40% (3), and overall
survival has not changed in decades (4). The poor patient outcome is mainly due to the
lack of a reliable screening test for early disease detection. Most ovarian carcinomas are
diagnosed at a late stage, after invasion to the intra-peritoneal cavity, and are thus
inoperable (3). In addition, many ovarian carcinomas respond poorly to therapy (1, 5) or
recur with the development of chemoresistance (1).
Based on their origins, approximately 90% of all human ovarian cancers are
categorised as epithelial ovarian carcinomas (EOCs), which originate in the ovarian
surface epithelium (OSE), and the rest are derived from granulosa, stromal or germ cells
(3). However, evidence supporting an origin of these tumors outside the ovary is
accumulating. Therefore, the paradigm of the OSE as the precursor of EOCs is being
challenged, and the cellular origin of EOC remains controversial.
1.2 Tumorigenesis of EOCs
Fathalla proposed the ‘incessant ovulation theory’ in 1971, suggesting that repeat
ovulation, surface rupture and subsequent repair lead to the trapping of the OSE in the
ovarian stroma and formation of inclusion cysts. Inclusion cysts secrete somatic growth
factors that lead to cell proliferation, genetic aberrations and finally malignant
2
transformation (6). This hypothesis is supported by substantial epidemiological data. One
case–control study of 150 ovarian cancer patients under the age of 50 years demonstrated
that the risk of ovarian cancer decreased with increasing numbers of live births,
increasing numbers of incomplete pregnancies and the use of oral contraceptives (7).
Another prevailing hypothesis addressing the development of ovarian cancer was
proposed by Cramer and Welch in 1983. Their ‘gonadotropin theory’ proposed that
excessive gonadotropin stimulation contributes to ovarian carcinogenesis (8). The risk of
ovarian cancer increases during the perimenopausal period, when serum gonadotropin
levels peak and thereafter remain elevated (9, 10). Moreover, only 10% – 15% of tumors
appear in premenopausal women (11). Likewise, polycystic ovary syndrome patients
(with high luteinising hormone levels) are more prone to ovarian cancer (12).
Epidemiologic evidence supports the idea that pregnancies, breast feeding, and oral
contraceptive use, which suppress pituitary gonadotropin secretion, reduce the risk of
ovarian cancer (13-16). Mesothelial OSE has been proposed to undergo Müllerian
differentiation during ovarian carcinogenesis. As a result, the OSE loses its mesothelial
characteristics and acquires the properties of the Müllerian system (3). This characteristic
readily explains why histological and immunocytochemical analyses of EOCs including
serous (Fallopian tube), endometrioid (endometrium), clear cell (vaginal) and mucinous
(endocervix) reveal characteristics of Müllerian epithelia rather than mesothelial tumors
(3). For instance, serous carcinomas express PAX2 or PAX8, which are normally
expressed in Fallopian tube epithelia (17, 18). However, the similarities of EOCs to
Müllerian tumors support the possibility of a Müllerian lineage and thus provide an
alternative theory of EOC histogenesis (4, 19). For instance, instead of OSE, serous
3
carcinomas in Fallopian tube are proposed to be precursors of high-grade serous tumors
(20). On the other hand, endometrioid and clear cell carcinomas are suggested to arise
from endometriosis (21).
1.3 Classification of EOCs
Rather than a single entity, EOCs are heterogeneous diseases comprising tumors
of different subtypes. Serous (high-grade and low-grade), endometrioid, clear cell and
mucinous are the major subtypes of EOCs. Their distinct genetic abnormalities and
oncogenic pathways and differential responses to chemotherapy (1, 22) emphasise the
importance of subtype diagnosis and subtype-specific therapies (23).
1.3.1 High-grade serous carcinoma (HGSC)
HGSC is the most common type of EOC and accounts for 60% of EOC cases and
90% of all serous carcinomas (1). HGSCs are usually diagnosed at an advanced stage
(22) 85% of patients present with widespread peritoneal metastases (1). Accurate
diagnosis of HGSC depends on squamous differentiation and areas with solid growth
(22). For problematic cases, Wilms tumor 1 (WT-1) is an useful immunomarkers due to
its specific expression in HGSCbut not other EOCs (24).
Almost all HGSCs harbour TP53 mutations (up to 97.6%) (25) (Fig. 1.1).
Additional mechanisms including TP53 loss of heterozygosity (LOH) (26) and MDM2 or
MDM4 (specific p53 inhibitors) amplification (25) contribute to aberrant p53 expression.
Although p53 dysfunction is a ubiquitous feature of HGSCs, the data regarding the
prognostic value of p53 are conflicting (25, 27).
4
Other characteristic genetic abnormalities of HGSCs include loss of BRCA1 and
BRCA2 (Fig. 1.1). Both BRCA genes can be inactivated through germline or somatic
mutations, and BRCA1 is also silenced epigenetically through promoter hypermethylation
(28), leading to familial and sporadic HGSC. Among BRCA-related hereditary ovarian
cancers, up to 57% -100% are HGSCs (22, 29), highlighting the importance of BRCA in
HGSC tumorigenesis.
Owing to their role in DNA repair, BRCA and p53 dysfunction lead to genomic
instability and allow accumulation of further genetic alterations along the path of tumor
progression. HGSCs normally display aneuploidy and intratumoral genetic heterogeneity
(30).
In addition to the conventional theory of development of EOCs from the OSE or
cortical inclusion cysts (3), there is an emerging view that EOCs may derive from cells
outside the ovary, which subsequently implant and expand within the ovary and present as
tumors originating in ovary (4). This mechanism may explain why the OSE has a
mesothelial phenotype whereas HGSCs have Müllerian morphology and express a
Müllerian marker (PAX8) but not mesothelial markers (18). Serous tubal intraepithelial
carcinomas (STICs) in the fimbriated end of Fallopian tubes are proposed to be the
precursors of HGSCs of the ovary (20). In addition to the high incidence of dysplastic
changes in the Fallopian tubes of women with a genetic predisposition for familial
ovarian cancer (31), up to 50% - 60% of HGSC patients also have STICs (32, 33). A
clonal relationship between STICs and HGSCs is further supported by identical TP53
mutations in the tumors (34) (Fig. 1.1). A gene profiling study showed that HGSCs and
the Fallopian tube epithelium share similar gene expression profiles, and HGSCs were
5
found to be less closely related to the OSE (35), suggesting that HGSCs may originate
from the Fallopian tube epithelium rather than the OSE.
Surgery is the primary treatment for all ovarian cancers including HGSC.
Although the majority (70% - 80%) of HGSC patients show initial responses to
platinum/taxane-based chemotherapy (1, 22), 70% of them experience a recurrence, and
many recurrent tumors become resistant to etoposide and doxorubicin (1). All of these
factors together, combined with the late diagnosis, contribute to the poor prognosis of
HGSC patients, who have a 10% - 20% 5-year survival rate (1).
1.3.2 Low-grade serous carcinoma (LGSC)
Among all ovarian serous cancers, 10% are LGSCs. Differentiation diagnosis
between LGSC and HGSC is usually based in the uniformity of nuclei. In addition,
psammoma bodies, differentiated architecture with papillary growth are characteristic
feathers of LGSC (22). The prognosis of LGSC patients is better than that for HGSC
patients (1). Rarely, LGSCs are found concurrently with HGSCs, suggesting the
progression of LGSC to HGSG (36) (Fig. 1.1). However, it is well accepted that low-
grade and high-grade serous carcinomas are fundamentally distinct tumors (22). LGSCs
rarely harbour TP53 or germline BRCA mutations (1, 22). Instead, activating mutations of
KRAS, BRAF, (37) or ERBB2 (38) are common. Accordingly, 60-70% of LGSCs express
active MAPK (39), highlighting the role of the KRAS/BRAF/MEK/MAPK pathway in
LGSC pathogenesis. Mutations of KRAS and BRAF are also found in serous borderline
tumors, which are believed to be the immediate precursors of LGSCs (1) (Fig. 1.1).
6
1.3.3 Endometrioid carcinoma (EC)
EC is the second most common histologic subtype of EOC, accounting for 10% -
20% of cases (40). ECs are mostly diagnosed at stages I and II, which accounts for the
better prognosis than other subtypes (22). EC are commonly characterized with squamous
or mucinous differentiation (22). Some clinicians diagnosed high-grade carcinomas with
glandular differentiation as EC, however, this diagnostic is doubted as glandular
differentiation is also involved in serous neoplasia (22).
Similar to high-grade serous, high-grade EC may arise de novo or by rare stepwise
progression from borderline and low-grade tumors (40). High-grade de novo ECs
typically have mutations in TP53 (41) (Fig. 1.1). On the other hand, the genetic defects
found in borderline or low-grade endometrioid tumors are also commonly found in high-
grade tumors. For instance, CTNNB1 (38% - 50% of cases) and PTEN (phosphatase and
tensin homolog deleted on chromosome 10) (20%) mutations have been identified in
both, suggesting a precursor role of these lesions (22, 42) (Fig. 1.1). The CTNNB1 gene
mutation encodes a β-catenin protein that is resistant to degradation, which allows
stabilization and nuclear accumulation of β-catenin (43) and can cause endometrioid
carcinogenesis. There is also evidence suggesting that endometriosis gives rise to EC.
ECs share similar molecular signatures with adjacent endometriosis, including LOH of
chromosome 12 (44) and 10q23 (45) mutations in ARID1A tumor suppressor gene, which
encodes adenine-thymine (AT)-rich interactive domain-containing protein 1A (21) and
PTEN (45). Furthermore, up to 42% of EC cases are associated with endometriosis (46),
and 67% - 100% of endometriosis cases are associated with coexisting EC (19) (Fig. 1.1).
7
1.3.4 Clear cell carcinoma (CCC)
CCCs account for approximately 10% of all EOC cases. Though CCC patients are
usually diagnosed at stages I or II, chemo-resistance, relapse after surgery and the
aggressiveness of CCCs make the prognosis of CCC patients worse than that of other
ovarian carcinomas at early stages (1, 22). The presence of clear cells is not sufficient for
a diagnosis of CCC. Additional criteria for CCC include papillary/tubulocystic
architecture and low mitotic rate (22).
In sharp contrast to HGSCs, most CCCs contain wild-type TP53 (47) and
BRCA1/2 (48). CCCs are best characterised by aberrant activation of PI3K signalling,
including activating mutations of PI3KCA (42, 47), loss of PTEN expression (49) and
activated mTOR (mammalian target of rapamycin) (50) (Fig. 1.1). Recently, two groups
demonstrated frequent somatic mutation of the ARID1A tumor suppressor gene in CCCs
(21, 51). Furthermore, ARID1A mutations are found in CCCs as well as in adjacent
atypical endometriosis, but not in distant endometriosis (21). This difference implicates
ARID1A mutations as early events in the transformation of endometriosis into cancer and
strengthens the association between endometriosis and clear cell carcinogenesis.
1.3.5 Mucinous tumors
Although mucinous tumors constitute 10% - 15% of all primary ovarian tumors,
most of them are benign and borderline tumors. Only 3% - 4% are mucinous carcinomas
(22), making them one of the rarest types of EOCs (4). Most of these tumors show
gastrointestinal differentiation and contain goblet cells, which are key features for their
8
diagnosis. However, mucinous tumors are often heterogeneous, making adequate
sampling for histological examination critical (22).
The origin of mucinous tumors are unclear, and based on their non-Müllerian
phenotype, development from cortical inclusion cysts is unlikely (4). KRAS mutations are
the most frequent genetic defects in mucinous carcinomas (85% of cases) and represent
an early event in mucinous tumorigenesis. In addition, the ERBB2 gene is amplified in
15% to 20% of mucinous carcinomas (22) (Fig. 1.1).
1.4 Epidermal growth factor (EGF) and the EGF receptor (EGFR)
Numerous steroids, growth factors and cytokines secreted by ovary stromal or
cancer cells promote neoplastic transformation and progression of ovarian cancer (52). Of
these, EGF-like growth factors and their cognate receptors have been the most
extensively studied.
1.4.1 EGFR structure and signaling
The EGFR, also known as ERBB1/HER1, together with HER2/neu,
ERBB3/HER3 and ERBB4/HER4 comprise the ERBB (or HER) family (53). All four
members are Type I transmembrane tyrosine kinase receptors that consist of an
extracellular domain (ECD), a transmembrane domain (TMD) and an intracellular domain
(ICD) (Fig. 1.2). The ICD can be functionally further divided into i) the juxtamembrane
domain (JD), which mediates EGFR internalisation and signalling specificity (54, 55); ii)
the tyrosine kinase domain (TKD), which possesses intrinsic kinase activity (56); and iii)
the carboxy-terminal regulatory domain (RD). Upon ligand binding, the ECD undergoes a
conformational change that allows dimerization of ERBBs (Fig. 1.2). Dimerization can
9
be homodimerization or heterodimerization between various ERBBs (57) and HER2 is
the most common binding partner for ERBBs (58). In the dimer complexes, the TKD
undergoes tyrosine phosphorylation. Subsequently, tyrosine residues in the RD are
transphosphorylated, in turn recruiting diverse cytoplasmic adaptor proteins and
activating a variety of signalling pathways (53). In ERBB3, a single amino substitution in
the TKD abrogates the intrinsic kinase activity, making ERBB3 homodimers inactive and
the signal transduction of ERBB3 relies on heterodimerization with other ERBB members
(56).
The ERBB family members invariably activate the ERK cascade through
recruiting the adaptor proteins Grb2 or Shc (59) (Fig. 1.2). The PI3K/AKT pathway is
triggered by most activated ERBB dimers, which have different kinetics and potency.
Such differences are probably due to the fact that ERBB3 and ERBB4 are capable of
direct binding to the PI3K p85 regulatory subunit, whereas the EGFR and HER2 couple
indirectly with p85 through adaptor proteins or ERBB3/ERBB4 (60). Phosphorylated-
AKT was detected in 68% of ovarian carcinomas on a tissue microarray (61). Activation
of these pathways in cancer has been associated with increased cell survival, growth,
angiogenesis and metastasis (57, 62). Furthermore, aberrant levels of the EGFR and
HER2, as well as their downstream ERK and AKT, lead to resistance to platinum-based
chemotherapy (61, 63, 64).
1.4.2 EGFR expression and mutations in ovarian cancer
Aberrant EGFR activity is achieved through a variety of mechanisms, including
overproduction of ligands, overexpression of receptors and constitutive activation of
10
receptors through mutation or loss of regulators (57). EGFR overexpression and mutation
are frequent events in human malignancies; for example, the EGFR gene is amplified in
up to 40% of gliomas (65). There are many published reports regarding EGFR expression
and its prognostic significance in ovarian cancer. In general, increased EGFR is
correlated with more aggressive diseases and poorer patient outcomes (53). In various
studies, the frequency of immunohistochemical detection of EGFR expression in ovarian
cancer tissues ranges from 4% - 100% (53). EGFR gene amplification is seen in 43% of
serous carcinomas (66), and EGFR overexpression is more common in serous (66, 67)
and endometrioid (68) carcinoma subtypes. Both a higher gene copy number and protein
expression correlate with the serous subtype (66). Higher levels of the EGFR are found in
omental metastases (69) and carcinomas of advanced stages (68, 70), implicating a role
for the EGFR in tumor progression and metastasis. This role may explain the fact that
high EGFR expression correlates with unfavourable outcomes in terms of overall survival
(66-68, 71) and disease-free survival (67). Increased EGFR expression is also associated
with carcinomas (compared with LMP or benign tumors) (70, 72), higher tumor grade
(66, 73), higher proliferative index (66, 73), larger residual tumor size (66, 68), and
advanced age of patients (66, 74). However, data from numerous studies are highly
variable and contradictory; many investigators have found no relationship between the
EGFR and the clinical features and prognostic factors mentioned above (53).
In addition to gene amplification and protein overexpression, EGFR hyperactivity
may arise from EGFR gene mutations (75). An 801-base pair in-frame deletion in the
extracellular domain of the EGFR results in a constitutively active EGFR variant III
(EGFRvIII) mutant. Independent of ligands, this mutant is constitutively active in terms
11
of receptor dimerization, autophosphorylation and downstream signalling activation (75,
76). EGFRvIII is expressed in a large proportion of gliomas and breast carcinomas and
75% of ovarian carcinomas (77). However, other reports showed that EGFRvIII is rare in
EOC (66, 78) Expression of this mutant in an EOC cell line causes epithelial-
mesenchymal (EMT) transition, cell migration and metastasis (79). Mutations in the
catalytic domain (exon 18, 19, 20 or 21) of EGFR are less frequent in ovarian carcinomas
(66, 74, 80). Activating mutations of the EGFR enhance the sensitivity of lung cancer
patients to EGFR inhibitors (81). In ovarian cancers patients, responses to gefitinib (a
small molecule tyrosine kinase inhibitor) have been reported to be independent of EGFR
mutational status in one study (82), but another reported that EGFR-activating mutations
have also been reported to confer patient responses to gefitinib and lead to longer
progression-free survival (67).
In contrast to numerous investigations of EGFR expression and mutations, only a
few studies regarding EGFR activation status in ovarian carcinomas have been published.
These studies reported values for EGFR activation, as detected by immunohistochemistry,
of 11.8%, 35% and 100% (62, 83, 84). EGFR phosphorylation was also found to be
positively correlated with a metastasis-promoting protein (matrix metalloproteinase 9)
and negatively correlated with E-cadherin, which inhibits metastasis (84).
1.4.3 EGFR ligands
The activation of ERBBs is mediated through ligand-dependent mechanisms.
Ligands of ERBBs contain an EGF-like domain and three intramolecular disulphide
bonds and are synthesised as plasma membrane-bound precursors (85). Based on their
12
binding specificity, they are classified into four groups: EGF, transforming growth factor-
α (TGF-α) and amphiregulin (AREG) bind the EGFR exclusively; epiregulin (EPI),
betacellulin (BTC) and heparin-binding-EGF (HB-EGF) have dual specificity and bind
the EGFR and ERBB4; and the heregulin (HRG) family constitute the remaining two
groups, which bind both ERBB3 and ERBB4 (HRG-1 and HRG-2) or exclusively to
ERBB4 (HRG-3 and HRG-4) (53). No HER2-specific ligand has been identified yet (57).
Most of these ligands are overexpressed in and correlate with the clinical features
of different types of cancers, including ovarian cancer (57). EGF has been detected in
borderline, low-grade tumors, carcinomas and EOC cell lines (53, 86). Furthermore, up to
72% of epithelial ovarian tumors that are examined are immunopositive for EGF, and a
higher EGF expression level has been found in mucinous cystadenocarcinomas compared
with that in LMP mucinous tumors (87). The EGFR-exclusive ligand TGF-α is frequently
detected in EOCs, with one study showing expression in 77% of tissues examined (53).
The expression of TGF-α correlates with disease grade and stage (86). Concomitant high
TGF-α levels and peak incidence of the disease in post-menopausal women suggest the
importance of this growth factor in ovarian cancer pathology (88). HB-EGF expression is
significantly higher in ovarian cancer than in normal ovaries (89, 90). Furthermore,
elevated HB-EGF protein levels were detected in ascites fluid from ovarian cancer
patients compared with that in the peritoneal fluid of normal women (90). A higher HB-
EGF expression level is significantly associated with shorter progression-free survival of
patients (89). Further, the levels of AREG found in ovarian cancer cell lines, tissues and
the peritoneal fluid of ovarian cancer patients are higher than those of TGF-α and EGF
13
(89, 91, 92). To date, however, no studies have reported the expression of the other two
EGFR-binding ligands, EPI and BTC, in ovarian carcinomas tissues (53).
It is important to note that the ERBB family members can also be transactivated
by heterologous signals including those from G protein coupled receptors, independent of
their cognate ligands (93-96). Cross activation of ERBBs can be achieved through direct
phosphorylation or induction of ectodomain shedding of ligands (90). For example, the
chemokine CXCL1 transactivates the EGFR through proteolytic cleavage of HB-EGF,
leading to MAPK activation and the proliferation of ovarian cancer cells (97).
1.4.4 Functions of the EGFR and it ligands in ovarian cancer
EGF and TGF-α are the most well-studied ligands of the EGFR. They form
autocrine loops with the EGFR and have numerous regulatory functions in OSE and
ovarian cancer cells. Gene profiling studies of rat OSE have demonstrated that genes
involved in the cell cycle, proliferation, and apoptosis are targets of EGF (98). EGF
treatment increases the proliferation of human primary OSE cells and ovarian cancer cells
in culture (99, 100). In vitro treatment with TGF-α increased cell proliferation in eight
ovarian cancer cell lines in one study (101). TGF-α is also synthesised by cancer cell
lines (102), and disruption of the TGF-α/EGFR autocrine loop by a TGF-α neutralising
antibody or antisense oligodeoxynucleotides blocks proliferation (102, 103). Accordingly,
the in vivo growth of inoculated tumors is also blocked by TGF-α antibody treatment
(104). Interestingly, growth regulation by AREG is unique, as AREG has a biphasic effect
on OSE and ovarian cancer cells (105).!Ovarian cancer cell lines stably transfected with
an antisense EGFR also show decreased proliferation (106). With its potent mitogenic
14
potential, the EGFR and ligand autocrine loops confer a selective advantage for cells and
are implicated in clonal expansion and in the accumulation of mutations within tumors.
In addition to cell growth, the EGFR and its ligands are implicated in metastasis.
In response to EGF treatment, OSE cells (107) and ovarian cancer cells undergo EMT
(108, 109) and EMT-associated N-cadherin and vimentin expression changes (108). It is
important to note that during EMT, cells adopt a fibroblastic phenotype with reduced
intercellular contact and increased cell motility. Consistent with these observations, tumor
cell migration and invasion have been repeatedly demonstrated to be stimulated by EGF
(84, 99, 108, 110, 111). Motility may be modulated through regulating the integrin
system, which comprises transmembrane receptors forming bridges between the
extracellular matrix (ECM) and the cytoskeleton and plays roles in adhesion and
migration. EGF activates STAT3 signalling and regulates the levels of integrin α2, α6
and β1 (108). Alternatively, invasiveness may be acquired through promoting proteolytic
degradation of the ECM; indeed, EGF increases the expression of uPAR (urokinase-type
plasminogen activator receptor) (110) and MMP9 (84, 111). Furthermore, MMP9 is
responsible for the EGF-induced loss of E-cadherin and disruption of adhesion junctions,
leading to a migratory and invasive phenotype (84). In contrast, silencing of the EGFR
suppresses integrin expression, MMP9 activity, cell adhesion and the invasion of ovarian
cancer cells (106, 112). However, no studies on the role of AREG, EPI or BTC in ovarian
cancer cell invasion have been published.
Furthermore, blocking the EGFR suppresses in vivo ovarian cancer cell
tumorigenicity in nude mice (106, 113). EGF also plays a role in regulating the sensitivity
15
of ovarian cancer cells to cisplatin in vitro (114), and dominant-negative EGFR can
partially restore cisplatin sensitivity in drug-resistant ovarian cancer cells (113).
1.4.5 Clinical activities of EGFR-targeted therapies.
Agents disrupting EGFR activities have been evaluated for their feasibility as
ovarian cancer therapies, for examples, small molecule tyrosine kinase inhibitors (TKIs)
targeting EGFR kinase domain and monoclonal antibodies (mAb) binding EGFR ECD.
Generally, these therapeutics have very limited clinical activities in ovarian cancers (59).
Gefitinib and erlotinib are EGFR TKIs which inhibit ovarian cancer cells and xenografts
growth. No patients showed complete response or partial response in a phase II clinical
trial of gefitinib (83) and only one objective response (4 % of cases) reported in another
trial (74). In a phase II trial of erlotinib, 6% of patients demonstrated partial response and
no complete response has been observed (59). Furthermore, trials have been performed
using TKIs in combinations with cytotoxic chemotherapies like docetaxel and carboplatin
(115, 116). However, none of these treatment regimes showed improvement in clinical
activities (115, 116). On the contrary, significant clinical response to gefitinib and
erlotinib have been observed in 10 % - 30 % NSCLC patients (117). In addition to TKIs,
results from clinical trials of EGFR mAbs are also disappointing. These antibodies block
ligand binding and receptor dimerization, they also reduce EGFR levels through
promoting EGFR internalization and degradation (116). The first clinically tested EGFR
mAb, cetuximab, showed partial response in 4 % of patients (118). In the phase II trial of
another EGFR antibody, matuzumab, neither partial response nor complete response has
been demonstrated (119).
16
1.5 Sprouty (SPRY)
In the past decade, a novel family of cytoplasmic proteins called SPRY have been
shown to play potent regulatory roles during Drosophila and mouse development (120-
122). Among these studies, the overexpression of SPRY has been shown to mimic the
effects of EGFR signalling loss and prevent EGFR-induced phenotypic changes (121).
Many additional studies have demonstrated that SPRY regulates the RAS/RAF/ERK
pathway that is activated by various RTKs (122-125).
1.5.1 SPRY structure
One Drosophila and four mammalian SPRY homologs have been identified with
conserved structural characteristics, namely SPRY1, SPRY2, SPRY3 and SPRY4 (121,
122, 126). All SPRY members share a conserved C-terminal cysteine-rich region, with
44% – 52% identity between Drosophila spry and mouse Spry1, 2 and 4 proteins (122).
They also contain a conserved SPRY domain that mediates binding to RAF1 (RAF1-
binding domain, RBD) (127) (Fig. 1.3). The C-terminal domain is responsible for plasma
membrane localisation (121, 128, 129). Deletion of the C-terminal domain abolishes
membrane localisation and many of the biological functions of SPRY (121, 125, 128). In
contrast, the N-terminal domains of the SPRY proteins are more divergent and are
proposed to account for the substrate diversity and specificity of the individual SPRY
members (122). However, the N-terminal domains of all SPRY proteins invariably
contain a conserved tyrosine residue (Fig. 1.3). The tyrosine is phosphorylated upon
growth factor stimulation and mediates membrane translocation. Most of the functions of
17
SPRY1 and SPRY2 (126, 130, 131) but not SPRY4 (126) are dependent on the
phosphorylation of this tyrosine.
1.5.2 SPRY functions and mechanisms
It is well known that downstream of various RTKs, SPRY proteins specifically
inhibit the RAS/RAF/ERK pathway without affecting the p38, JNK or PI3K/AKT
pathways (124, 125). However, there is still no consensus on how SPRY blocks ERK
activation. Drosophila spry regulates ERK signalling at a point downstream of EGFR and
upstream of RAS and RAF (121). In mammalian cells, by direct interaction with Grb2,
SPRY1 and SPRY2 decrease RAS activation, RAS-RAF binding and RAF activation
(124) (Fig. 1.3). Alternatively, downstream of RAS and independent of RAS activation,
SPRY2 inhibits FGF-induced RAF activation (125). Interestingly, SPRY is only capable
of binding to wild-type RAF, but oncogenic RAF is refractory to SPRY inhibition,
leading to unchecked ERK activation in cell harbouring oncogenic RAF mutations (132).
The evidence to date suggests the existence of multiple mechanisms that depend on the
cellular context and/or the identity of the RTK (127).
In contrast to initial observations (124, 125), accumulating evidence supports the
mechanism that the PI3K/AKT cascade is also a target of SPRY (133, 134). To date,
there has only been one study on the mechanism of AKT repression by SPRY, which
demonstrated that SPRY2 increases the amount of PTEN and decreases its
phosphorylation, leading to a decreased activation of AKT by EGF.
In contrast to SPRY2, the understanding of the regulatory mechanisms of SPRY 4
is extremely limited and conflicting. SPRY4 was first reported to represses RTK-induced
18
ERK phosphorylation upstream of, or parallel to, RAS (135, 136). However, later studies
found that mouse SPRY4 interferes with the VEGF-induced RAS-independent activation
of RAF1 through a direct association with RAF1 (137). SPRY4 can also interact with
TESK1 (testicular protein kinase 1) to inhibit growth factor-induced RAS/MAPK
signalling (136, 138) (Fig. 1.3).
Although SPRY proteins are generally considered to be inhibitory, SPRY2 may
enhance EGF signalling in a cell type-specific manner (127). Cbl proteins are E3
ubiquitin ligases that recognise, ubiquitinate and target RTKs for degradation (139-141).
Through direct interaction with the Cbl RING finger domain (142, 143), SPRY2 is
capable of sequestering Cbl and thereby protects EGFR from degradation (144), resulting
in higher EGFR levels and sustained activation (127) (Fig. 1.3).
1.5.3 Regulation of SPRY activity
The feedback nature of SPRY arises from the prompt activation of SPRY by RTK
activity. In addition to phosphorylation and membrane translocation as described above,
the transcriptional regulation of SPRY is also important. During Drosophila embryonic
development, SPRY level is higher in cells that are responsive to FGF and EGF (121,
123, 145, 146). Levels of SPRY2 and SPRY4 have been shown to be upregulated by
activation of RAS/ERK signalling upon EGF and FGF stimulation in vitro (147).
Similarly, in cancer cells, SPRY level is induced by growth factors (148) and oncogenic
mutations downstream of RTK (149).
19
1.5.4 SPRY expressions in cancer
Aside from a few exceptions (132, 150), the expression of SPRY has been found
to be downregulated in cancer cells in most reports. SPRY2 and/or SPRY1 are
significantly down-regulated in breast (151), prostate (152, 153), endometrial (154), and
liver (155) cancers. The effects of the altered expression of different SPRY isoforms is
likely cancer type-specific. The same group recently demonstrated a lower level of
SPRY2, but not SPRY1, in hepatocellular carcinoma (155). Furthermore, the mechanisms
of downregulation are also variable among different types of cancers. For example, in
prostate cancer, SPRY2 is silenced through both epigenetic promoter hypermethylation
and LOH (156). In contrast, SPRY2 loss in breast and liver cancer is independent of
promoter methylation (151, 155). The prognostic value of SPRY has been demonstrated
in renal (157), liver (158) and prostate (153) cancer patients. Moreover, SPRY4 mRNA
level is as a reliable marker of the response of gastrointestinal stromal tumors to Gleevec
treatment (150).
1.5.5 SPRY as tumor suppressor genes
The tumor suppressor role of SPRY is supported by its negative effects on
tumorigenesis. SPRY have been reported to inhibit proliferation (151, 155, 159, 160),
migration and invasion (148, 159-161), cell cycle progression (148), in vitro and in vivo
tumorigenesis (148, 151) and metastasis (162).
1.6 Hypoxia-inducible factor-1 alpha (HIF-1α)
Hypoxia, a low oxygen tension condition, is a common phenomenon in solid
tumors and is a driving force of tumor progression. Although severe or prolonged hypoxia
20
is toxic to cancer cells, it selects cancer cells that with adaptive and genetic changes that
allow them to survive and proliferate (163). Most biological changes in response to
hypoxia are mediated through the induction of the critical molecular mediators of
hypoxia, the hypoxia-inducible factors (HIFs). Following stabilization and nuclear
translocation in hypoxic conditions, HIF binds to hypoxia-response elements (HREs) to
induce expression of numerous hypoxia-response genes (Fig. 1.4), thereby regulating
multiple steps of tumorigenesis, including angiogenesis, metastasis and resistance to
therapy (163). These processes contribute to the malignant phenotype, increase the rate of
mutation and decrease overall patient survival (164).
1.6.1 HIF-1α structure
The HIF family, including HIF-1, HIF-2 and HIF-3, belongs to the PAS (PER-
ARNT (arylhydrocarbon receptor nuclear translocator)-SIM) family of basic helix-loop-
helix (bHLH) transcription factors. HIFs bind to DNA as heterodimers of a constitutively
expressed HIF-1β subunit, also known as ARNT, and an oxygen-dependent α subunit
(HIF-1α, HIF-2α and HIF-3α). Structural analyses revealed that HIF-1α contains a bHLH
domain (for dimerization and DNA binding), a PAS domain (for dimerization and target
gene specificity), two transactivation domains (N-terminal, NTAD and C-terminal,
CTAD) and an oxygen-dependent degradation (ODD) domain (Fig. 1.4). The ODD
domain is required for the ubiquitin–proteasome degradation pathway under aerobic
conditions, and due to the absence of the ODD domain, HIF-1β is constitutively
expressed in normoxia.
21
1.6.2 HIF-1α regulation
The abilities of HIFs to act as the master regulator of cellular responses to
hypoxia and in O2-homeostasis arises from the tight regulation of its activities by O2
availability. The activities of HIFs are largely controlled through the availability of the α
subunit, whose half-life is extremely short under normal oxygen tension (165). Therefore,
the majority of investigations focus on the alpha subunits, and HIF-1α is the most
extensively studied. In addition to tissue hypoxia, HIF-1α stabilization can be regulated
by hormonal stimulation and genetic alterations under normoxic conditions.
1.6.2.1 Hypoxia
Under well-oxygenated conditions, conserved proline residues (Pro402 and
Pro564) within the ODD domain of HIF-1α are hydroxylated by prolyl hydroxylases
(PHDs) (Fig. 1.4). Hydroxylated HIF-1α is recognised and ubiquitinated by Von Hippel-
Lindau (VHL), the substrate recognition component of an E3 ligase complex.
Ubiquitinated-HIF-1α is then targeted to the proteasomal degradation pathway (166). As
the enzymatic activities of PHDs require O2 as a cosubstrate, under hypoxia, PHDs are
inactive, hydroxylation of HIF-1α is suppressed, and HIF-1α is stabilized (Fig. 1.4).
PHDs are transcriptionally regulated by hypoxia in both HIF-dependent and HIF-
independent pathways, which creates a negative feedback loop (165). Hypoxic induction
of PHD2 and PHD3 is found in a wide range of cell types; however, induction is not
observed in cells lacking HIF-1α or HIF-1β (167). Silencing HIF-1α reduces the
induction of PHD2 and PHD3 by hypoxia (168). The HIF-independent pathway is
mediated by SIAH1 and SIAH2, specific E3 ligases of PHD1 and PHD3 (169, 170),
22
whose expression is transcriptionally upregulated by hypoxia in a HIF-independent
manner (171). Hypoxia-independent regulation of PHDs by cellular and hormonal factors
such as oncogenic RAS and Src (see below), reactive oxygen species (ROS) (172), TGF-
β (173) and endothelin (174) serves as one of the mechanisms of HIF-1α regulation in
normoxia.
Stabilized HIF-1α translocates into the nucleus and then dimerizes with HIF-
1β to constitute HIF-1. HIF-1 heterodimers activate transcription by recruiting the
transcriptional coactivators p300 and CREB binding protein (CBP). Interaction between
HIF-1α and p300/CBP is regulated by factor inhibiting HIF-1 (FIH-1) in an oxygen-
dependent manner. FIH, which belongs to the 2-oxoglutarate and Fe(II)-dependent
dioxygenase superfamily, hydroxylates the conserved asparagine (Asn803) residue within
the HIF-1α CTAD and prevents HIF-1α/p300 interaction (175). When hypoxia prevents
asparagine hydroxylation by FIH, the CTAD is activated, interacts with p300/CBP and
binds to HREs to regulate hypoxia-regulated gene expression. Therefore, in addition to
HIF-1α stabilization, the complete activation of HIF transcriptional activity also requires
CTAD activation.! Kung et al. showed that blocking the interaction of HIF-1α with
p300/CBP attenuates hypoxia-inducible gene expression, reduces capillary formation and
inhibits tumor growth in nude mouse xenografts (176).
1.6.2.2 Regulation of HIF-1α in normoxia
1.6.2.2.1 VHL mutations
HIF-1α can also be activated in tumors under normoxic conditions through
genetic alterations in the oxygen-signalling pathway. As mentioned earlier, the VHL
23
tumor suppressor gene product, pVHL, functions as the substrate recognition subunit of
the ubiquitin E3 ligase complex that targets HIF-1α for degradation. This interaction is
dependent on the hydroxylation of one or both of the two conserved prolines (Pro402 and
Pro564) of HIF-1α by PHDs. Due to the O2-sensing function of PHDs, VHL loss-of-
function results in HIF stabilization regardless of oxygen tension (177, 178).
Inactivating germline VHL mutations predispose individuals to VHL diseases
including renal clear cell carcinomas (RCCs) and haemangioblastomas (179, 180). Using
immunohistochemistry, HIF-1α and HIF-2α proteins are found to be overexpressed in all
tested RCCs and haemangioblastomas (181). Similarly, cells lacking the VHL wild-type
protein constitutively express HIF-1α protein in normoxia (181, 182), which can be
rescued by the reintroduction of a functional VHL (181). Accordingly, the mRNA of the
hypoxia-regulated genes VEGF (181, 183) and GLUT1 (181, 184) are found to be higher
in VHL-defective cells. Finally, expression of wild-type VHL in VHL-defective RCC cell
lines prevents tumor formation in nude mice (181, 185).
1.6.2.2.2 AKT and PI3K
Ample evidence suggests that PI3K/AKT signalling can also stabilize HIF-1α and
induce HIF-1 activity (186-192). In hypoxic prostate cancer cells, inhibition of PI3K
using pharmacologic inhibitors or siRNA knockdown decreases HIF-1α levels (188, 189)
and VEGF promoter activity (190). The expression of HIF-1α and VEGF and the
angiogenesis of prostate tumor xenografts are inhibited by dominant negative AKT
mutants (189). HIF-1α stabilization by AKT does not require functional VHL (193).
24
Activation of the PI3K/AKT pathway can arise from amplification of the
phosphatidylinositol 3-kinase catalytic subunit α (PIK3CA) oncogene in human
malignancies, including ovarian cancer (194). PIK3CA level is positively correlated with
VEGF expression in ovarian cancer cell lines. Treatment with a PI3K inhibitor abrogates
HIF-1α and VEGF expression (195). PIK3CA level has also been detected in most
advanced-stage ovarian cancer biopsies and positively correlates with the VEGF level
and the extent of microvascular development (195).
Moreover, components of the PI3K/AKT signalling pathway, both upstream and
downstream, have been shown to be involved in HIF-1α regulation, including PTEN, a
molecular inhibitor of AKT. Glioblastoma cell lines lacking functional PTEN express
high levels of VEGF even in normoxia, which is suppressed by restoration of wild-type
PTEN (187). Overexpression of PTEN in prostate tumor cells reduces HIF-1α levels
(190), VEGF expression, angiogenesis and tumor growth (189). Downstream of AKT,
inhibition of mTOR (mammalian target of rapamycin) reverses the effect of AKT on HIF-
1α and target gene expressions (191, 196).
AKT has been shown to stabilize HIF-1α without impairing prolyl hydroxylation
and HIF-1α degradation, as revealed by antibodies specific for hydroxylated proline
(197). However, alternative mechanisms for normoxic HIF-1α stabilization exist. For
example, AKT positively regulates HIF-1α levels in glioblastoma cells through increased
protein translation (190).
In addition to PI3K and AKT, oncogenes such as RAS (198, 199) and Src (200)
have been implicated in the normoxic activation of HIF-1α. Introduction of the
25
oncogenic mutant RASV12 and v-Src into cells increases the HIF-1α level and HIF-1
activity. Remarkably, RASV12 and v-Src abolish the hydroxylation of Pro564 in the
ODD domain and hence inhibit HIF-1α degradation. In sharp contrast, cells transfected
with constitutively active AKT upregulate HIF-1α without affecting hydroxylation. This
finding indicates that these oncogenes may stabilize HIF-1α through hydroxylation-
dependent or hydroxylation-independent pathways (197).
1.6.2.2.3 Glycogen synthase kinase 3β (GSK3β)
The AKT target glycogen synthase kinase 3 (GSK3, exists in GSK3α or
GSK3β isoforms) is a negative HIF-1α regulator. Inhibition of GSK3β using LiCl or
GSK3β knockdown increases HIF-1α accumulation and the level of the HIF-1 target PAI-
1 in normoxia, whereas GSK3β overexpression reduces HIF-1α (201). Notably, GSK3β
mediates the balance between HIF-1α stabilization/degradation as a function of the
duration of hypoxia. Mottet et al. showed that short (5 hrs) incubations of prostate tumor
cells in hypoxia lead to AKT activation and the inhibition of phosphorylation of GSK3β
on Ser9, therefore reducing GSK3β activity and resulting in HIF-1α accumulation as well
as increased HIF-1 transcriptional activity. In contrast, prolonged (16 hr) hypoxia
inactivates AKT and hence activates GSK3β, resulting in decreased HIF-1α protein levels
(192). HIF-1α destabilization by GSK3β is via the NTAD is independent of HIF-1α
hydroxylation and pVHL activity (201).
26
1.6.2.2.4 Heat shock protein 90 (HSP90)
HSP90 interacts with HIF-1α (193, 202, 203) as well as HIF-2α and HIF-3α
(204). These associations are mediated through the HIF-1α PAS domain (203, 205) and
occur in both normoxia and hypoxia (193, 203). Although HSP90 is associated with HIF-
1α in the cytoplasm, HSP90 does not cotranslocate into the nucleus with HIF-1α (206).
The mechanism underlying HIF-1α protection by HSP90 is not yet fully understood and
is thought to occur via VHL-independent degradation (193, 204).
In hypoxia, HSP90 is induced by (206), binds to and protects HIF-1α. Inhibition
of HSP90 by geldanamycin results in loss of HIF-1α accumulation and HIF-1 activation
in hypoxic VHL-defective cells (202, 206).
The level of HSP90 is enhanced by AKT in normoxia. Inhibitors of AKT, but not
MAPK inhibitors, reduce HSP90 expression (193). In VHL-defective RCC cells,
inhibition of the AKT pathway promotes HIF-1α degradation as efficiently as the HSP90
inhibitor geldanamycin (193). As the PI3K/AKT pathway has no effect on HIF-1α
hydroxylation, protection of HIF-1α by HSP90 increases HIF-1α stabilization by AKT,
independent of proline hydroxylation, subsequent VHL ubiquitination and proteasomal
degradation. �
In addition to AKT, HSP90 mediates the HIF-1α regulation of the kinase RACK1
(receptor of activated protein kinase C) (207). Moreover, carbon monoxide, an
environmental factor implicated in angiogenesis (208), stabilizes HIF-1α and promotes
VEGF expression through enhancing HIF-1α/HSP90 interaction (209).
27
1.6.2.2.5 Hormonal regulation
Another common mechanism for the induction of HIF-1α in normoxia involves a
large variety of growth factors and cytokines. EGF (188) and lysophosphatidic acid (210)
induce HIF-1α through AKT activation to stimulate VEGF expression. IGF-II regulates
cell survival through AKT and HIF-1α induction (211). TGF-β regulates the expression
of HIF-1α and the HIF-1 targets PHD2 and PAI (173). Other hormones including IGF-I
(187, 191, 212, 213), PGDF (186) and follicle-stimulating hormone (FSH) (214) have
been reported to regulate the expression of HIF-1α.
1.6.3 HIF-1α expressions in cancer
The expression of HIF-1α has been comprehensively investigated in panels of
human normal and cancerous biopsies by immunohistochemistry (215, 216). Most normal
tissues show no HIF-1α immunoreactivity. In contrast, overexpression of HIF-1α was
detected in 50% – 60% of malignant samples in one study, which represented most types
of tumors examined, including breast, liver, lung, colon, prostate, ovarian, renal and
pancreatic cancer. Furthermore, HIF-1α overexpression was found to be associated with
metastasis and high-grade tumors. Two-thirds of the regional lymph node and bone
metastases were HIF-1α positive. In addition, a high level of HIF-1α was more
commonly found in breast metastases than in primary tumors. Among brain tumors of
different grades, the strongest HIF-1α expression level was detected in the most
malignant and vascularised tumors (215). Both borderline ovarian tumors and epithelial
ovarian cancers were positive for HIF-1α expression (217, 218). HIF-1α has been shown
to be a negative prognostic factor in most cancers (215, 219, 220)
28
1.6.4 HIF-1α functions in cancer
1.6.4.1 Angiogenesis
When solid tumors expand, tumor hypoxia increases due to high metabolism and
excessive oxygen consumption as well as the increased distance between tumor cells and
local capillaries (221). To promote blood vessel formation and sustain growth in hypoxic
microenvironments, tumors transition from a nonangiogenic to angiogenic phenotype,
termed the “angiogenic switch” (222). As a prime physiological regulator of the
angiogenic switch (223), a positive correlation has been found between HIF-1α
overexpression and tumor vascular density (215) in brain and ovarian (217) tumor
biopsies.
During the angiogenic switch, hypoxia shifts the balance toward proangiogenic
factors and reduces the expression of antiangiogenic factors such as thrombospondin-1
and thrombospondin-2 (224). To actively promote angiogenesis, HIF activates the
expression of various angiogenic proteins including growth factors, cytokines, and a
number of small molecules, including the key angiogenic factor VEGF. VEGFR1 and
VEGFR2 on endothelial cells mediate the mitogenic effects of VEGF on endothelial cells.
Therefore, VEGF has strong angiogenic activity, and treatment with a VEGFR inhibitor
results in a 60% reduction in tumor vasculature in mouse models (225).
The VEGF promoter contains HRE binding sites and is a direct transcriptional
target of HIF-1 (226). Accordingly, xenografts of HIF-1β-deficient hepatoma cells exhibit
reduced levels of hypoxia-induced VEGF mRNA and reduced tumor vascularisation and
proliferation (227). HIF-1α is also expressed in endothelial cells and mediates autocrine
signalling of VEGF/VEGFR2 in endothelial cell proliferation and blood vessel formation
29
(228). More importantly, human tumors with high HIF-1α and VEGF expression levels
are correlated with more aggressive and malignant phenotypes (229).
In addition to the hypoxia-mediated activation of HIF, the AKT/HIF-1α axis also
mediates VEGF level and angiogenesis stimulated by growth factors (188, 191) and other
hormones (210, 214) under normoxia.
1.6.4.2 Metastasis
Metastasis is the primary cause of cancer-related deaths. It is a multistep process:
from the initial epithelial-mesenchymal transition (EMT), dissemination, and homing to
the final organotropic colonisation. The role of hypoxia and HIF in tumor metastasis is
supported by the observation that HIF-1α expression is upregulated or more commonly
found in metastases and high-grade tumors (215). Mechanistically, hypoxia and HIF are
potent regulators of many key factors that determine tumor cell invasiveness and
metastatic potential, namely E-cadherin, lysyl oxidase (LOX), chemokine receptor
CXCR4 (CXCR4), stromal-derived factor 1 (SDF-1) and plasminogen activator inhibitor-
1 (PAI-1).
HIF-1 promotes metastasis through repressing E-cadherin, a major cell adhesion
molecule that maintains epithelial integrity. Loss of E-cadherin is a hallmark of and
prerequisite for metastasis. Reduced E-cadherin expression is associated with increased
metastasis in patients with ovarian (230), breast (231), prostate (232) and lung cancers
(233). The antimetastatic role of E-cadherin is supported by the finding that restoration of
E-cadherin in cancer cells inhibits metastasis (234). Repressed E-cadherin levels have
been found to be correlated with HIF-1α expression in ovarian (235) and RCC carcinoma
biopsies (236). In hypoxic ovarian cancer cells and VHL-defective RCC cells, HIF-1α is
30
overexpressed and EMT and cell invasion are increased in vitro (235, 236). E-cadherin in
these cells is repressed by HIF-1α through inducing the E-cadherin-specific repressors
Snail or ZEB1 (235, 236).
HIF also promotes metastasis through modulating the extracellular matrix. LOX is
an enzyme that is involved in extracellular matrix formation. HIF-1 is a potent inducer of
LOX, and secreted LOX contributes to the invasive properties of hypoxic human cancer
cells. The importance of LOX in hypoxia-induced metastasis is directly underscored by
the finding that inhibition of LOX is sufficient to attenuate hypoxia-induced metastasis in
mice. In patients with breast and head and neck tumors, a high LOX expression level is
correlated with hypoxia and poor distant metastasis-free and overall survival (237).
CXCR4 is the most common chemokine receptor that is expressed in tumor cells.
It is a direct HIF target in hypoxic lung, ovarian, breast and renal cell carcinoma cells
(238). Its ligand, SDF1, is highly expressed in common sites of metastasis, including the
lung, bone marrow, and liver (220). Interactions between CXCR4 and its ligand SDF-1
mediate the metastatic homing of tumor cells in response to hypoxia (220, 238)
Another invasion/metastasis related gene PAI-1 is a direct target of HIF-1α. PAI-
1 levels are enhanced through PI3K/AKT and ERK1/2 via HIF-1α in prostate and gastric
cancer cells (212, 239).
1.7 Hypothesis and objectives
Loss of expression or activity of SPRY has been demonstrated in various human
malignancies. Nevertheless, no studies on SPRY in ovarian cancer have been reported.
Given the implications of various RTKs in ovarian tumorigenesis, I tested the hypothesis
that SPRY acts as a tumor suppressor in ovarian cancer. The main objectives of this thesis
31
are to investigate the expression and functions of SPRY in ovarian cancer, specifically in
the context of EGFR and HIF-1α signalling.
Objective 1. In Chapter 2, I investigated whether the SPRY mRNA is deregulated and
examined the function of SPRY2 in ovarian cancer.
1) The mRNA levels of SPRY members in ovarian cancer tissues and cell lines was
investigated.
2) The incidence of SPRY2 gene deletion was evaluated.
3) The role of SPRY2 in EGF-induced cell invasion was tested.
Objective 2. In Chapter 3, I tested the roles of an alternative EGFR ligand (AREG) and
regulator (SPRY4) in ovarian cancer invasion.
1) The effect of AREG on invasion was tested.
2) The underlying molecular pathways were elucidated.
3) The effect of AREG on SPRY4 level was examined.
4) The effect of SPRY4 on AREG-induced invasion was investigated.
Objective 3. In Chapter 4, I explored the regulation of SPRY4 by EGF and the effects of
SPRY4 on EGF signalling.
1) The role of AKT/HIF-1α in EGF-induced SPRY4 level was tested.
2) The feedback of SPRY4 on the EGF-induced level and activity of HIF-1α was
evaluated.
32
3) The role of AKT in SPRY4-mediated HIF-1α regulation was investigated.
Objective 4. In Chapter 4, I extended the investigation to the underlying mechanisms of
SPRY4 inhibition of HIF-1α level.
1) The inhibition of HIF-1α level by SPRY4 was confirmed.
2) The level of regulation of HIF-1α by SPRY4 was determined.
3) The roles of PHD in the SPRY4 regulation of HIF-1α level was investigated.
33
Figure 1.1 The chart illustrates the pathogenesis of epithelial ovarian cancers. Low-grade
serous carcinomas arise from serous borderline tumors containing activating KRAS,
BRAF or ERBB2 mutations. Rarely, low-grade serous tumors progress to high-grade
tumors. High-grade serous carcinomas develop from the Fallopian tube or ovarian surface
epithelium through TP53 and BRCA mutations. KRAS mutations and ERBB2 gene
amplification are frequently found in mucinous tumors. Clear cell carcinomas contain
PI3KCA mutations or loss of PTEN. PTEN and CTNNB1 mutations may lead to the
formation of endometrioid carcinomas. Alternatively, endometrioid carcinomas may
contain TP53 mutations. Endometrioid carcinomas and clear cell carcinomas may also
arise from endometriosis with ARID1A mutations. Modified from Lalwani, N et al, 2011
(1).
34
Figure 1.2 The structure and signaling of EGFR. EGFR consists of an extracellular
domain (ECD), a transmembrane domain (TMD) and an intracellular domain (ICD). The
ICD can be further divided into a juxtamembrane domain (JD), a tyrosine kinase domain
(TKD) and a regulatory domain (RD). Upon the binding of EGFR ligands, such as EGF,
transforming growth factor α (TGFα) or amphiregulin (AREG), the TKD is
phosphorylated and thereby activated; the RD is then transactivated. The phosphorylated
RD recruits adaptor proteins, including Shc, Grb2 and SOS, and triggers the
RAS/RAF/MEK/ERK cascade. Activated EGFR also stimulates other MAPK pathways
(p38 and JNK) and the PI3K/AKT pathway.
35
Figure 1.3 A schematic diagram of human SPRY depicting its structure, its interacting
partners and the corresponding functional consequences. The C-termini of SPRY proteins
are more conserved and contain a RAF1-binding domain (RBD), which mediates the
interaction with RAF and leads to the inhibition of ERK activation. ERK inhibition is
also achieved through SPRY interaction with Testicular protein kinase 1 (TESK1). SPRY
also interfere with the RAS/RAF/MEK/ERK pathway through interaction with Grb2. The
N-termini of SPRY are more divergent but invariably contain a conserved tyrosine
residue. Tyrosine-phosphorylated SPRY binds with Cbl, which inhibits EGFR
degradation and increases EGFR signaling. Interaction between SPRY and Seven in
absentia homolog 2 (SIAH2) promotes the ubiquitination of SPRY and target SPRY for
proteasomal degradation. Modified from Edwin, F et al, 2009 (240).
36
Figure 1.4 The diagram illustrates the regulation of HIF-1α. HIF-1α is mainly regulated
post-transcriptionally. Under normoxia, PHD proteins hydroxylate the two proline
residues within the HIF-1α oxygen-dependent domain (ODD) domain. Hydroxylated
HIF-1α is prone to Von Hippel-Lindau (VHL)-mediated degradation. When PHDs are
degraded by Seven in absentia homolog 2 (SIAH2) or inactivated by hypoxia, growth
factors (GFs), reactive oxygen species (ROS) or oncogenic RAS, HIF-1α degradation is
prevented. In contrast, the PI3K/AKT/glycogen synthase kinase-3β (GSK3β) pathway
regulates HIF-1α independently of PHD activity. When HIF-1α is stabilized, it would
translocate into the nucleus and dimerize with HIF-1β. With the recruitment of the p300
co-factor, the complex recognizes hypoxia-responsive elements (HRE) in the promoters
of target genes, which are implicated in processes such as angiogenesis, invasion,
proliferation and apoptosis.
37
Chapter 2 Genetic inactivation of Sprouty2 promotes epidermal growth factor-induced E-
cadherin downregulation and invasion in ovarian cancer cells
2.1 Introduction
Ovarian cancer is the fifth leading cause of all female cancer-related deaths and the
most lethal gynaecologic malignancy in North America. Approximately 60% of women
who develop ovarian cancer will die from the disease (2, 241). Although epithelial
ovarian cancer (EOC) includes a majority of all ovarian carcinomas, its origin and
aetiology have not yet been completely elucidated. Current evidence suggests that
malfunction of receptor tyrosine kinases (RTKs), including epidermal growth factor
receptor (EGFR), contributes to the development of EOC (52). EGFR and its ligands play
a critical role in cell proliferation, survival, and tumor metastasis (53, 242). Moreover,
increased expression levels of EGFR-specific ligands have been identified in ovarian
cancer cells (89) and ascitic fluid (90), and EGFR expression levels were found to be
elevated in advanced stage disease and metastases (53).
In addition to RTK abnormalities, the loss of endogenous regulators represents an
alternative mechanism that leads to aberrant RTK activity. During the past decade, a
novel family of cytoplasmic proteins, Sprouty (SPRY), has been identified as feedback
inhibitors of the RAS/MAPK/ERK signalling cascade triggered by RTKs. Consistent
with their inhibitory involvement in the RTK-stimulated ERK pathway, SPRY is a
putative tumor suppressor, and cells lacking either SPRY expression or function may be
hypersensitive to mitogenic and metastatic signals. SPRY isoforms have been found to be
downregulated in most malignancies studied, including breast (151), prostate (152, 153,
38
156, 161), liver (155) and lung (160) cancers. Functionally, SPRY has been reported to
interfere in steps of tumorigenesis, including proliferation (151, 155, 160), migration
(148, 160, 161), invasion (148), and cell cycle (148) as well as in vitro (148) and in vivo
(151) tumorigenic potential and formation of metastases (162). The prognostic value
associated with SPRY levels have been established in renal (157), liver (158) and prostate
(153) cancer patients. Moreover, SPRY4 mRNA level serves as a reliable response
marker to Gleevec treatment for patients with gastrointestinal stromal tumors (150).
However, the mechanism by which SPRY is silenced appears to vary depending on the
cancer type (151-153, 155, 156, 160, 161), and contradicting data exist for similar
malignancies (153, 155, 156, 158).
The present study aimed to investigate, in addition to any SPRY downregulation, the
underlying mechanisms and functions of SPRY in ovarian cancer. We demonstrate that
SPRY2 mRNA is downregulated in both biopsies and permanent cell lines derived from
EOC. Furthermore SPRY2 gene is deleted in some of the ovarian tumors and is possibly a
cause of the reduced SPRY2 mRNA. We further demonstrate a positive correlation
between SPRY2 and E-cadherin. Furthermore, the re-introduction of SPRY2 in ovarian
cancer cells partially restored E-cadherin expression levels suppressed by EGF and
antagonized the effect of EGF-induced invasion.
39
2.2 Materials and methods
2.2.1 The Human Exonic Evidence-Based Oligonucleotide microarray (HEEBO)
The HEEBO microarray (Stanford, CA, USA) employed for the study included
44,544 70-mer probes that were designed using a transcriptome-based annotation of
exonic structure for genomic loci. Pooled RNA from 10 human cancer cell lines of
different origins (Stratagene, Universal Human Reference RNA, Cat 740000) for broad
gene coverage on the array was included as a reference. We examined the mRNA
expression profiles of a series of ovarian tumors from the Vancouver General Hospital
tumor bank obtained from patients who were undergoing surgery during 2004 and 2005.
These cases included the following subtypes: high-grade serous (n = 35), low-grade
serous (n = 2), endometrioid (n = 7), clear cell (n = 3), serous borderline (n = 1),
endometrioid borderline tumor (n = 1) and normal Fallopian tube (n = 1). Approval for
the study was obtained from the University of British Columbia Research Ethics Board
(#H04-60102), and written informed consent was obtained from all participants involved
in the study. HEEBO was performed as described previously (243).
2.2.2 Molecular inversion probe (MIP) copy number analysis
To determine whether deletion of the SPRY2 and SPRY4 genes occurs in human
ovarian tumors, samples from another cohort of patients were utilised to perform MIP
copy number analysis. All women undergoing primary debulking surgery at the
Vancouver General Hospital and British Columbia Cancer Agency in Vancouver,
Canada, between January 2004 and September 2005 were invited, except those with
mucinous and borderline tumors or who had received pre-operative chemotherapy. The
40
pathology data were reviewed by a pathologist (CBG). The classification and grading of
tumors were performed as described previously (244). We included 28 high-grade serous,
5 high-grade serous/undifferentiated, 3 high-grade undifferentiated, 5 endometrioid, 4
clear cell and 1 low-grade serous tumors. Ethical approval was obtained from the
University of British Columbia Research Ethics Board (#H02-61375 and #H03-70606),
and written informed consent was obtained from all participants involved in the study.
The MIP copy number assay and copy number estimation were performed as described
previously (245). Copy numbers over 3.0 were considered amplification events and copy
numbers below 1.5 were considered deletion events.
2.2.3 The Cancer Genome Atlas (TCGA)
To obtain direct evidence to support a dependency of reduced SPRY2 mRNA
level on gene deletion, we retrieved data from the TCGA database portal
(http://cancergenome.nih.gov/) in September 2011. Copy number data for 585 ovarian
serous cystadenocarcinomas and 587 normal samples (569 matched and 18 unmatched),
as well as data for the gene expression profile of 584 tumors and 18 unmatched normal
samples, were extracted.
2.2.4 Cell culture and reagents
Four non-tumorigenic SV40 Tag-immortalised OSE-derived lines, IOSE-29, IOSE-
80, IOSE-120, and IOSE-398, were generous gifts from Dr. Nelly Auersperg (University
of British Columbia) (246). The human ovarian adenocarcinoma cell line, BG-1, was
kindly provided by Dr. K.S. Korach (National Institute of Environmental Health Sciences,
NIH, Research Triangle Park, NC) (247). CaOV3, OVCAR3 and SKOV3 ovarian cancer
41
cell lines were obtained from American Type Culture Collection (Manassas, VA). The
cell lines were cultured in MCDB 105 / M199 (1:1), supplemented with 10% heat-
inactivated fetal bovine serum (FBS), 100 IU/ml penicillin and 100 g/ml streptomycin.
The cells were cultured at 37°C and 5% CO2. Fetal bovine serum was purchased from
Hyclone Laboratories (Logan, UT). Human recombinant EGF and other tissue culture
materials were obtained from Sigma Chemical Co. (St. Louis, MO) unless otherwise
stated. Human SPRY2 overexpression constructs (FLAG-SPRY2) and the empty pXJ40-
FLAG vector, which were transfected using Lipofectamine™ 2000 (Invitrogen, Carlsbad,
CA), were gifts from Dr. Graeme R. Guy (the Institute of Molecular and Cell Biology,
Singapore) (143).
2.2.5 Real-time PCR
Real-time PCR was performed using an ABI 7300 real-time thermal cycler (ABI,
Hercules, CA). SPRY2, SPRY4 and the internal control, GAPDH, were amplified in
duplicate with the following PCR primers: SPRY2, forward 5’-
CCCCTCTGTCCAGATCCATA-3’ and reverse 5’-CCCAAATCTTCCTTGCTCAG-3’;
SPRY4, forward 5’-AGCCTGTATTGAGCGGTTTG-3’ and reverse 5’-
GGTCAATGGGTAGGATGGTG-3’, and GAPDH, forward 5’-
GAGTCAACGGATTTGGTCGT-3’ and reverse 5’-GACAAGCTTCCCGTTCTCAG-3’.
2.2.6 Antibodies
Specific antibodies were used to detect proteins via Western blot analysis: the
anti-E-cadherin antibody was obtained from BD Transduction Laboratories (Lexington,
42
KY), the anti-SPRY2 antibody was purchased from Sigma and the anti-β-actin antibody
was purchased from Santa Cruz Biotechnology (Santa Cruz, CA).
2.2.7 Invasion assay
Twenty-four-well transwell filters with an 8-µm pore coated with 1 mg/ml Matrigel
(50 µl/well; BD Sciences, Mississauga, ON, Canada) were used to assess cell invasion.
SKOV3 cells transfected with either control or SPRY2-overexpression constructs were
trypsinized and resuspended in 0.1% FBS medium, with or without EGF, and then seeded
in triplicate in the upper chamber. Next, 1% FBS medium was added to the lower wells.
The chambers were incubated for 24 h at 37°C in a 5% CO2 atmosphere. Cells that did
not penetrate the filter were removed, and invaded cells on the lower surface of the filter
were fixed with ice-cold methanol, stained with Hoechst 33258, and counted by
epifluorescence microscopy using Northern Eclipse 6.0 software (Empix Imaging,
Mississauga, ON, Canada). Triplicate inserts were used for each individual experiment,
and the results are presented as the mean values.
2.2.8 Statistical analysis
Statistical analysis was performed using Prism Graphing software. Differential
variations in SPRY mRNA levels among ovarian tumor subtypes were assessed using the
Kruskal-Wallis rank test followed by Student’s t test comparing each tumor subtype. The
relative quantification of mRNA expression levels, as assessed by real-time PCR, was
calculated using the 2–ΔΔ Ct method. For the invasion assay and the comparison of E-
cadherin expression levels with the control (using SPRY2-overexpressing cells), a one-
way ANOVA and nonparametric Column Analysis was performed followed by Tukey’s
43
Multiple Comparison Test to compare all pairs of columns. Columns are not denoted by
the same letter are statistically different (P < 0.05). Data are presented as the mean ± SD
of three or four independent experiments. The Pearson correlation coefficient (r) and
associated probability (P) were calculated when comparing E-cadherin and SPRY2
protein levels using the Spearman nonparametric correlation.
2.3 Results
2.3.1 Levels of SPRY mRNA in ovarian tumors of different pathological subtypes
To examine whether SPRY mRNA were downregulated in ovarian tumors, as is
reported for other malignancies, we compared EOC samples of various histopathological
types, including serous (high-grade, low-grade or borderline), endometrioid (carcinoma
or borderline tumor), clear cell and normal Fallopian tube. SPRY1, SPRY2 and SPRY4
mRNA were included in our array. Significant differences in the SPRY2 mRNA levels
were observed in various tumors types (Kruskal-Wallis test, P = 0.0091, Fig. 2.1). The
lowest mean level of SPRY2 mRNA was observed in the clear cell sample followed by
the high-grade serous carcinomas. All other samples displayed higher mean SPRY2
mRNA levels compared with the reference level. To further assess the variations in
SPRY2 mRNA levels among the categories, we extended our study to perform
comparisons between each subtype using Student's t test. The difference in SPRY2
mRNA levels between high-grade and low-grade serous was statistically significant (P =
0.022, Table 2.1). When comparing pathological subtypes, the mean SPRY2 mRNA
levels in serous and clear cell carcinomas were statistically lower than that of
44
endometrioid carcinomas (serous vs. endometrioid, P = 0.0007; clear cell vs.
endometrioid, P = 0.022). SPRY4 mRNA levels were not significantly different between
the various subtypes (P = 0.33, data not shown); however, the mean expression level in
high-grade serous was lower than that in the reference and was the lowest among all the
specimens. We utilised three oligonucleotides in the array that corresponded to different
SPRY1 transcripts, for which the expression levels were similar for each subtype and
therefore did not display significant differences (P = 0.396, 0.706 and 0.813, data not
shown).
2.3.2 Levels of SPRY mRNA in immortalised ovarian surface epithelium (IOSE) and
EOC-derived cell lines
Next, we extended our expression analysis to include ovarian cancer cells cultured in
vitro. To enhance data reliability, 4 IOSE cell lines established from individual patients
were included as references for the comparison of SPRY mRNA levels in ovarian cancer
cell lines. Using real-time PCR, we confirmed the observation of reduced SPRY2 mRNA
(Fig. 2.2A) expression in most (3 out of 4) ovarian cancer cell lines tested (BG-1,
OVCAR3 and SKOV3). In addition, the SPRY4 mRNA levels observed in all 4 cancer
cell lines were consistently lower than those of the IOSE cell lines (Fig. 2.2B).
2.3.3 A deletion event in the proximity of the SPRY2 locus
MIP copy number analysis was performed to examine whether the reduced SPRY
levels observed were due to chromosomal changes of the SPRY genes. As there are no
specific intragenic probes available, we used probes closely flanking the SPRY2 and
SPRY4 loci as surrogate markers. For SPRY2, two markers tightly flanking the locus (0.2
45
Mb proximal and 0.5 Mb distal) exhibited a 23.4% (11/46) and 19.6% (9/46) loss,
respectively, in 46 tumors. Notably, with one exception in undifferentiated high-grade
tumor, nearly all of the deletion events observed were identified in high-grade serous
carcinomas (Table 2.2); none of the endometrioid, clear cell, or low-grade serous tumors
exhibited loss of the SPRY2 gene. In high-grade serous samples, the frequencies of
SPRY2 loss were found to be 32.1% (9/28) to 35.7% (10/28). Two additional markers
(0.9 Mb proximal and 1.5 Mb distal) showed a similar frequency of loss (25% and 35.7%,
respectively) (Fig. 2.3). In contrast, deletion of the markers (0.4 Mb proximal and 90 kb
distal) closest to the SPRY4 locus was rarely found, with only 3.6% (1/28) and 7.1%
(2/28) of high-grade serous tumors, respectively. Two markers more distant from the
SPRY4 gene exhibited only two cases of loss (7.1%) (Fig. 2.3).
2.3.4 SPRY2 deletion may lead to reduced SPRY2 mRNA level
In a TCGA data set, at a percentage lower than our MIP study, 24% of ovarian
serous cystadenocarcinoma samples displayed a decrease in gene copy number. A
majority (67%) of the samples showed decreased SPRY2 mRNA level (log2
tumor/normal ratio < -0.5), therefore, suggesting that SPRY2 gene deletion was
responsible for the reduced SPRY2 mRNA level in the samples. According to the gene
expression analysis, decreased SPRY2 mRNA level was detected in 54� of the tumors.
However, when we analysed gene copy number in these tumors, only 32% showed a
decrease in copy number, thereby indicating that additional mechanisms may contribute
to the reduced SPRY2 mRNA level.
46
2.3.5 SPRY2 reversed EGF-suppressed E-cadherin protein expression and
antagonized EGF-induced cell invasion
In our previous report (248), the SKOV3 cell line was shown to respond
significantly to EGF treatment and showed a significant decrease in E-cadherin
expression levels. We used this cell model to investigate the effect of SPRY on EGF-
induced reduction in E-cadherin levels. We transiently overexpressed SPRY2 in SKOV3
cells, and gene overexpression was subsequently confirmed using antibodies specifically
recognising SPRY2 protein (Fig. 2.4A). SPRY2 overxpression has no effect on cell
morphology (data not shown) and basal E-cadherin (Fig. 2.4B). After EGF treatment E-
cadherin levels were significantly suppressed and the decrease was partially reversed by
SPRY2, which results in higher E-cadherin levels in SPRY2-overexpressing cells
compared with cells transfected with empty vector (Fig. 2.4B). Together, these results
support an antagonizing effect of SPRY2 on EGF activity of repressing E-cadherin
protein. The negative effect of SPRY2 on EGF regulation of E-cadherin prompted us to
evaluate the effect of SPRY2 on EGF-stimulated cell invasion using the transwell
invasion assay. In correlation with the effect on E-cadherin protein, SPRY2 counteracted
the effect of EGF on invasion and SPRY2-overexpressing cells showed reduced
invasiveness under EGF stimulation, whereas SPRY2 had no effect on basal invasion
(Fig 2.4C, D).
47
2.3.6 SPRY2 and E-cadherin proteins displayed a positive correlation in human
ovarian cancer cell lines and tumors
To test whether the positive effect of SPRY2 on E-cadherin level is reflected in
the endogenous expression levels, we analysed the SPRY2 and E-cadherin expression
levels in three ovarian cancer cell lines and a panel of eleven high-grade serous ovarian
tumors isolated from patients. In most of the samples, we found parallel SPRY2 and E-
cadherin expression levels (Fig. 2.5A). This finding indicated a positive correlation
between the expression levels of the two proteins. Moreover, the correlation was
statistically significant (Pearson correlation coefficient, r = 0.5825 and P = 0.0288) in a
Spearman nonparametric correlation analysis (Fig. 2.5B).
2.4 Discussion
SPRY proteins have been identified as endogenous inhibitors of the
RAS/MAPK/ERK pathway downstream from RTKs. Aberrant activity of various RTKs,
especially EGFR, plays an essential role in malignancy development, including ovarian
cancer. Therefore, we aimed to examine the expression profile and function of SPRY in
ovarian cancer, which remains largely unknown. First, we demonstrated a reduction in
SPRY2 mRNA level in ovarian cancer cell lines and clinical samples (Fig. 2.1, 2.2A). It is
important to note that both the significant difference observed between the SPRY2 mRNA
levels in high-grade serous and endometrioid samples in the HEEBO analysis and the
observation that most (12 out of 13) SPRY2 losses were found in high-grade serous
carcinomas suggests the involvement of SPRY2 in serous EOC pathogenesis. These
48
results also demonstrate that SPRY2 represents a potential molecular marker for the
identification of serous carcinomas, which includes a majority (approximately 80%) of all
EOCs (3).
We next evaluated the occurrence of the chromosomal deletion of the SPRY2 gene.
The SPRY2 locus has been mapped to 13q31.1 (156). Previous cytogenetic studies have
shown that chromosomal loss of either 13 or 13q is a frequent event in both hereditary
(249, 250) and sporadic ovarian carcinomas (251, 252). This frequency can be explained
by the existence of known tumor-suppressor genes on 13q including BRCA2 (13q12-13),
RB (13q14) and protocadherin 9 (13q21-2). The data presented here support these studies
by demonstrating the chromosomal deletion of the SPRY2 locus, which occurred in a
moderate proportion of cases (Fig. 2.3). Furthermore, a minor proportion (16%) of serous
cystadenocarcinomas from the TCGA database showed both gene deletion and reduced
expression, thereby suggesting that genetic aberration is an underlying cause of SPRY2
mRNA downregulation in ovarian cancer. In contrast to the SPRY2 results, deletion of the
markers flanking the SPRY4 gene (90 kb to 0.7 Mb) occurred in only 2 cases (Table 2.2).
The deletion of the SPRY4 gene is a rare event in ovarian cancer and unlikely an
explanation for the reduced SPRY4 mRNA levels observed in cell lines.
According to TCGA gene expression analysis, only 32% of tumors with reduced
SPRY2 mRNA level exhibited a gene deletion, thereby indicating the requirement of
additional silencing mechanisms. The presence of CpG islands in the SPRY2 5’
regulatory regions suggests the potential involvement of epigenetic mechanisms in the
regulation of SPRY2 expression. Reports on the methylation status of the SPRY2
promoter in liver and prostate cancer samples have been contradictory and inconclusive
49
(153, 155, 156, 158). Our preliminary experiment demonstrated a robust reactivation of
SPRY4 mRNA level in BG-1, CaOV3 and OVCAR3 cells treated with the demethylating
agent, 5-aza-2-deoxycytidine (data unpublished). However, only a mild induction of
SPRY2 mRNA level was observed in 2 out of 3 cell lines (data not shown), thereby
suggesting that promoter hypermethylation may play a minor role in the silencing of
SPRY2 expression in this type of malignancy.
Reduced E-cadherin expression has been found to be associated with advanced stage
disease, poor differentiation, metastasis (230) and serous subtype tumors (253). Many
reports, including our recent report on EGF-repressed E-cadherin expression via the
induction of Snail and Slug in SKOV3 cells (248), have shown that E-cadherin is silenced
at a transcriptional level, either through epigenetic promoter hypermethylation (230, 254)
or repression by transcriptional repressor induction (255, 256). The present study
provides evidence that SPRY2 positively regulates and correlates with E-cadherin
expression, thereby suggesting the possibility that loss of SPRY2 might represent an
additional mechanism whereby ovarian cancer cells lose E-cadherin and gain malignant
properties. Furthermore, among various ovarian tumor subtypes, high-grade serous
tumors exhibited SPRY2 gene deletion, reduced SPRY2 mRNA level and a correlation
between SPRY2 and E-cadherin expression levels, thereby implying that SPRY2 may
play a role in the tumorigenesis of high-grade serous tumors.
In summary, our findings demonstrate that the SPRY2 level is decreased in ovarian
cancers due to a chromosomal deletion, and the reintroduction of SPRY2 diminishes
EGF-induced cell invasiveness by restoring the EGF-repressed intercellular adhesion
molecule E-cadherin protein. Therefore, genetic downregulation of SPRY2 may lead to a
50
decrease in E-cadherin expression, which promotes ovarian cancer progression. Further
clarifying the mechanisms underlying SPRY function and regulation will not only
advance our understanding of ovarian cancer progression but also facilitate the
development of novel therapeutic strategies for the treatment of ovarian cancer.
51
Table 2.1 Comparison between mean SPRY2 mRNA levels of various histopathological types by Student’s t test
The HEEBO microarray was performed to examined the SPRY2 mRNA
expression profiles of a series of ovarian tumors: high-grade serous (n = 35), low-grade
serous (n = 2), endometrioid (n = 7), clear cell (n = 3), serous borderline (n = 1),
endometrioid borderline tumor (n = 1) and normal Fallopian tube (n = 1). Student's t test
was performed to assess the variations in SPRY2 mRNA levels between each pair of
subtypes. Means not denoted by the same letter are significantly different.
Mean
Serous borderline A B 1.83
Low-grade serous A 1.75
Endometrioid borderline A B 1.54
Endometrioid A 1.47
Normal Fallopian tube A B 1.46
High-grade serous B -0.50
Clear cell B -0.66
Means not denoted by the same letter are significantly different
52
Table 2.2 MIP analysis of loss of the markers flanking the SPRY2 and SPRY4 loci in ovarian tumors
SPRY2 SPRY4
number histopathology proximal 2 proximal 1 distal 1 distal 2 proximal distal
223 serous HG 1.40 1.40 1.58 1.58 2.09 1.92
329 serous HG 1.37 1.20 1.26 1.26 1.44 1.33
293 serous HG 1.94 1.88 1.88 1.88 2.10 2.22
283 serous HG 1.44 1.09 1.30 1.30 2.43 2.13
239 serous HG 2.41 2.24 2.28 2.28 2.42 2.31
327 serous HG 2.26 1.98 2.49 2.13 2.31 2.07
379 serous HG 1.58 1.58 1.41 1.36 2.09 2.09
163 serous HG 1.46 1.43 1.44 1.44 1.77 1.99
305 serous HG 2.09 2.09 1.94 1.94 2.61 2.32
212 serous HG 1.54 2.76 1.54 2.12 1.83 2.36
330 serous HG 2.34 2.17 2.57 2.56 1.56 1.41
332 serous HG 2.08 1.69 1.30 1.30 2.12 1.77
388 serous HG 2.97 2.97 3.01 2.56 2.05 1.94
363 serous HG 1.57 1.36 1.36 1.32 2.21 2.27
344 serous HG 2.06 1.57 2.26 1.74 2.73 2.44
345 serous HG 1.29 1.28 1.18 1.18 2.47 2.91
384 serous HG 1.93 1.82 1.97 1.74 2.71 2.36
178 serous HG 2.04 2.02 2.02 1.57 2.67 2.35
229 serous HG 1.35 1.35 1.53 1.46 2.26 2.10
309 serous HG 2.36 2.36 2.45 2.21 2.23 2.23
394 serous HG 1.96 1.83 1.83 1.82 2.22 2.09
195 serous HG 1.75 1.24 1.20 1.20 2.39 2.39
236 serous HG 1.42 1.42 1.53 1.53 1.59 1.90
172 serous HG 2.75 2.22 2.13 1.99 1.62 1.55
254 serous HG 1.68 1.68 1.75 1.84 2.05 2.05
53
SPRY2 SPRY4
number histopathology proximal 2 proximal 1 distal 1 distal 2 proximal distal
319 serous HG 2.51 2.16 2.30 2.30 2.15 2.21
372 serous HG 1.50 1.36 1.36 1.30 2.35 2.41
297 serous HG 2.03 2.00 2.09 1.86 2.27 2.19
208 undifferentiated HG 3.03 2.53 2.72 2.72 1.81 1.81
273 undifferentiated HG 1.67 1.48 1.54 1.54 1.94 1.66
240 undifferentiated HG 2.27 2.10 2.27 2.30 2.10 2.02
161 serous/undifferentiated HG 1.87 1.72 1.67 1.67 1.87 1.82
336 serous/undifferentiated HG 1.88 2.27 1.83 1.83 1.87 1.87
186 serous/undifferentiated HG 2.44 2.09 2.09 2.36 2.60 2.72
201 serous/undifferentiated HG 2.23 2.23 1.81 1.81 3.24 3.24
280 serous/undifferentiated HG 2.62 2.57 2.54 2.54 1.89 2.25
198 clear cell 2.18 2.01 2.23 1.88 3.04 2.42
213 clear cell 2.14 2.14 1.86 1.86 2.01 1.98
219 clear cell 2.48 2.50 2.61 2.61 2.26 2.07
392 clear cell 2.04 1.81 2.04 1.64 2.54 2.31
242 endometrioid – G2 2.37 2.02 2.37 1.91 2.58 2.14
281 endometrioid – G2 2.23 2.41 2.45 2.45 2.23 2.23
334 endometrioid – G1 2.02 1.94 2.32 2.23 2.35 1.92
156 endometrioid – G2 1.93 1.93 1.90 1.90 2.18 1.88
343 endometrioid – G2 2.46 2.03 2.03 2.03 2.26 2.33
324 serous LG 2.00 1.97 2.00 2.00 1.92 2.13
MIP analysis were performed on 28 high-grade serous, 5 high-grade serous/undifferentiated, 3 high-grade undifferentiated, 5 endometrioid, 4 clear cell and 1 low-grade serous ovarian tumors to detect presence of the markers flanking the SPRY2 and SPRY4 loci. The MIP copy numbers below 1.5 were considered deletion events (bolded). HG: high-grade; LG: low-grade; G1: grade 1; G2: grade 2; G3: grade 3.
54
Figure 2.1 The box plot displays the HEEBO microarray result of SPRY2 mRNA levels
in ovarian tumors of various subtypes: high-grade serous (n = 35), low-grade serous (n =
2), endometrioid (n = 7), clear cell (n = 3), serous borderline (n = 1), endometrioid
borderline tumor (n = 1) and normal Fallopian tube (n = 1). Pooled RNA from 10 human
cell lines for broad gene coverage on the array was included as a reference (0). Dots
represent individual samples.
55
A
B
Figure 2.2 Comparison of SPRY2 and SPRY4 mRNA levels in immortalised OSE (IOSE)
and ovarian cancer cell lines by real-time PCR. Real-time data of SPRY2 and SPRY4
mRNA were normalized against the internal control, GAPDH. The average of normalized
expression levels of the IOSE cell lines was calculated and set to 100% (dotted horizontal
line). The relative levels of SPRY2 and (A) SPRY4 (B) mRNA were expressed as a
percentage of the mean expression in IOSE cell lines. Columns: mean of three passages
of the cell lines; bars: SD.
56
Figure 2.3 A schematic representation of the MIP copy number assay results of 28 high-
grade serous carcinomas. The diagram shows the MIP ID, position and the corresponding
percentage of loss of markers flanking the SPRY2 locus (left) and SPRY4 locus (right).
57
A B
C
D
58
Figure 2.4 The effect of SPRY2 on EGF-induced E-cadherin suppression and cell
invasion. A. SKOV3 cells were transiently transfected with either the empty pXJ40-
FLAG vector or FLAG-SPRY2 constructs. Starved SKOV3 cells were treated with 100
ng/ml EGF. Total protein was collected after 24 hr and analysed by Western blotting
using anti-E-cadherin, anti-SPRY2 or anti-β-actin antibodies. B. The protein signal
intensities were quantified and normalized against the internal control. The data are
expressed as a percentage of control vector-transfected cells without EGF treatment and
represents the mean ± SD of three independent experiments. C. SKOV3 cells transfected
with either control or SPRY2-overexpression plasmid were seeded in Matrigel-coated
transwell filters and cultured with 100 ng/ml EGF for 24 h. Invaded cells were then
stained and quantified. The data are shown as the mean ± SD of four independent
experiments. D. Representative photos of the invasion assay. The mean values that are
not denoted by the same letter are significantly different. Open bars: control treatment;
Filled bars: EGF treatment.
59
A
B
Figure 2.5 The correlation between SPRY2 and E-cadherin protein expression in ovarian
cancers. A. Total protein from ovarian cancer cell lines and high-grade serous ovarian
tumors was extracted and analysed via Western blot. B. Protein signal intensities were
quantified and a correlation was assessed using the Spearman nonparametric correlation
method.
60
Chapter 3 An amphiregulin and Sprouty4 loop regulates ovarian cancer cell
invasiveness via an E-cadherin-dependent mechanism
3.1 Introduction
Ovarian cancer is a common and the most lethal gynecological cancer. The high
mortality rate is caused by a lack of reliable screening tests and obvious symptoms, which
frequently results in diagnosis at advanced disease stages when peritoneal metastasis is
already present (3). Impairment of the epidermal growth factor (EGF) system is known to
play a role in ovarian cancer by directly enhancing the invasiveness and metastatic
potential of cancer cells (69, 84, 99, 108-111). EGFR amplification, mutation and protein
overexpression have been reported in ovarian cancer (57, 66, 67, 75). Alternatively,
aberrant EGFR activity may be a result of overproduction of EGFR ligands.
Among the EGFR-binding ligands, amphiregulin (AREG) levels are higher than
TGF-α and EGF levels in ovarian cancer tissues and cell lines (89, 91). Similarly, the
concentration of AREG in the peritoneal fluid of ovarian cancer patients at different
stages of the disease ranges between 203 – 225 pg/ml and is significantly higher than the
concentration of TGF-α (2.01 – 8.33 pg/ml) (92). These data suggest that AREG is an
important ligand activating EGFR in cancer cells.
Higher local AREG concentration may arise from the juxtacrine action of AREG
(257). Furthermore, in regards to ovarian cancer, AREG has been shown to be
upregulated by luteinizing hormone (LH) during ovulation (258). This finding not only
suggests a high local concentration of AREG in the ovaries than the circulation but also
suggests a causal link between LH, AREG and ovarian cancer. Our laboratory has
61
previously demonstrated that LH stimulates ovarian cancer migration and invasion (259,
260). Considering that AREG is induced by and mediates the actions of LH (258),
exposure to AREG may in turn promote neoplastic transformation and progression. To
date, studies of the physiological role of AREG in ovarian cancer have been restricted to
its role in cell proliferation (91, 105). In ovarian cancer cells and normal ovarian surface
epithelial cells, AREG induces a biphasic regulation of proliferation (105).
Alternatively, EGFR hyperactivity may be the result of the deregulation of
downstream effectors and regulators (261, 262), for instance, Sprouty (SPRY). SPRY
members (SPRY1-4) have been proposed as general inhibitors of signalling downstream
of EGFR and other receptor tyrosine kinases (RTKs) (127). In various normal cell
systems, SPRY levels and activities are regulated by growth factors through activating
RTKs (121, 123, 145-147). Similarly, SPRY levels have been found to be induced by
growth factors (148) and oncogenic mutations downstream of RTKs in cancer cells (149).
SPRY members act as tumor suppressor genes and negate many aspects of tumorigenesis,
including cell invasion and cancer metastasis (148, 159-162). Our laboratory has recently
found that SPRY2 antagonises EGF-induced invasion in ovarian cancer cells (So et al
unpublished).
To evaluate the proinvasive potential of AREG, two invasive ovarian cancer cell
lines were treated with AREG, and invasiveness was assayed. We demonstrated a
significant effect of AREG in promoting cell invasion through activation of the
MAPK/ERK and PI3K/AKT pathways, induction of SLUG mRNA expression and
reduction in level of the adhesion molecule E-cadherin. Using the AREG-induced E-
cadherin downregulation and cell invasion as physiological endpoints, we tested the
62
hypothesis that AREG triggers a SPRY-mediated feedback response. We first showed that
SPRY4 was significantly induced by AREG. To assess the importance of SPRY4
upregulation, SPRY4 was silenced using siRNA. Compared with control cells, ovarian
cancer cells depleted of SPRY4 responded to AREG more significantly in terms of E-
cadherin downregulation and cell invasion. These data support the presence and
functionality of an AREG/SPRY4 loop in ovarian cancer invasion regulation.
3.2 Materials and methods
3.2.1 Cell culture and reagents
The SKOV3 ovarian cancer cell line was obtained from the American Type Culture
Collection (Manassas, VA). Our laboratory previously showed that SKOV3 is invasive
and its invasiveness is increased by EGF (248). The cells were cultured in MCDB
105/M199 (1:1) supplemented with 5% heat-inactivated fetal bovine serum, 100 IU/ml
penicillin and 100 g/ml streptomycin. The cells were cultured at 37°C and 5% CO2. Fetal
bovine serum was purchased from Hyclone Laboratories (Logan, UT). The MEK/ERK
inhibitor U0126 was purchased from Calbiochem (San Diego, CA). Human recombinant
AREG, the PI3K/AKT inhibitor LY294002, the EGFR inhibitor AG1478 and other tissue
culture materials were obtained from Sigma Chemical Co. (St. Louis, MO) unless
otherwise stated.
63
3.2.2 Transfection
The empty pcDNA3.1 vector was obtained from Invitrogen (Carlsbad, CA). A
human E-cadherin expression vector (plasmid 28009) was purchased from Addgene
(Cambridge MA). Plasmids were transfected using Lipofectamine™ 2000 (Invitrogen,
Carlsbad, CA). ON-TARGETplus SMARTpool SPRY4 siRNA and non-targeting siRNA
(Dharmacon, Lafayette, CO) were transfected using Lipofectamine RNAiMAX
(Invitrogen, Carlsbad, CA).
3.2.3 Real-time PCR
Total RNA was extracted from cells using TRIzoL Reagent (Invitrogen) and used
in first-strand DNA (cDNA) synthesis using the Invitrogen Super-ScriptTM first strand
synthesis system for real-time PCR according to the manufacturer’s protocol. Real-time
PCR was performed in an ABI 7300 real-time thermal cycler (ABI, Hercules, CA). The
amplifications of E-cadherin, SLUG, ZEB1, SPRY2, SPRY4 and the internal control
Gapdh were performed as follows: a 3 min hot start at 95ºC followed by 40 cycles of
denaturation at 95ºC for 15 sec and amplification at 60ºC for 1 min. PCR reactions were
performed in duplicate with the following PCR primers: E-cadherin, forward 5′-
GGGTGACTACAAAATCAATC-3′ and reverse 5′-AAAGAGCCCTTACTGCCCCC-3′;
SLUG, forward 5′-TTCGGACCCACACATTACCT-3′ and reverse 5′-
GCAGTGAGGGCAAGAAAAAG -3′; ZEB1 forward 5′-
GCACCTGAAGAGGACCAGAG-3′ and reverse 5′-TGCATCTGGTGTTCCATTTT-3′;
SPRY2, forward 5’-TTGCACATCGCAGAAAGAAG-3’ and reverse 5’-
GAAGTGTGGTCACTCCAGCA-3’; SPRY4, forward 5’-
64
AGCCTGTATTGAGCGGTTTG-3’ and reverse 5’-GGTCAATGGGTAGGATGGTG-
3’; GAPDH, forward 5′-GAGTCAACGGATTTGGTCGT-3′ and reverse 5′-
GACAAGCTTCCCGTTCTCAG-3′.
3.2.4 Western blot analysis
Equal amounts of total cell lysate were resolved on 7.5% SDS-PAGE gels and
electrotransferred to a PVDF membrane. After blocking for 1 hr with 5% nonfat dry milk
in TBS-T buffer (20 mM Tris-HCl, pH 7.4, 150 mM NaCl, 0.1% Tween-20), the blots
were probed for overnight at 4 ºC with the following primary antibodies: anti-E-cadherin
antibody, which was obtained from BD Transduction Laboratories (Lexington, KY); the
anti-SPRY4 antibody was obtained from Abcam (Cambridge, MA); anti-phospho-AKT
(Ser473), anti-total AKT, anti-phospho-ERK1/2 (Thr202/Tyr204) and anti-ERK1/2
antibody, which were purchased from Cell Signaling, Inc. (Austin, TX) and anti-β-actin
antibody, which was obtained from Santa Cruz Biotechnology (Santa Cruz, CA). The
blots were then incubated with a peroxidase-conjugated secondary antibody (Bio-Rad)
for 1 hr followed by detection with ECL chemiluminescence reagent (Amersham,
Arlington Heights, IL) and exposure on X-ray films.
3.2.5 Invasion assay
To assess invasion, 24-well transwell filters with an 8-µm pore coated with 1
mg/ml Matrigel (50 µl/well; BD Sciences, Mississauga, ON) were used. Ovarian cancer
cells were trypsinised, re-suspended in 0.1% FBS medium and seeded in triplicate in the
upper chamber. Medium containing 1% FBS was added to the lower wells. The chambers
were incubated for 24 hrs at 37°C in a 5% CO2 atmosphere. The cells that did not
65
penetrate the filter were wiped off. The invading cells on the lower surface of the filter
were fixed with ice-cold methanol, stained with Hoechst 33258 and counted through
epifluorescence microscopy with Northern Eclipse 6.0 software (Empix Imaging,
Mississauga, ON). Triplicate inserts were used for each individual experiment, and the
results are presented as the mean numbers.
3.2.6 Statistical analysis
Real-time PCR quantification of mRNA level was calculated using the 2–ΔΔ Ct
method. Data are presented as the mean ± SD of three independent experiments or
triplicates in a representative experiment and were analyzed by one-way ANOVA
followed by Tukey’s post-hoc test using GraphPad Prism 5 (GraphPad Software, San
Diego, CA) to compare all pairs of columns. Columns are not denoted by the same letter
are statistically different. Means not denoted by the same letter are significantly different
(P < 0.05).
3.3 Results
3.3.1 AREG promoted invasion of ovarian cancer cells
We tested the effect of AREG on the invasion of ovarian cancer cells using a
Matrigel-coated transwell assay. SKOV3 and OVCAR5 cells were incubated with
different concentrations of AREG and allowed to invade across the transwell for 24 hrs.
AREG promoted invasion of both cell lines (Fig. 3.1). At 1 ng/ml, AREG was ineffective
66
at stimulating invasion, whereas invasion was significantly increased by 10 and 100
ng/ml AREG.
3.3.2 AREG reduced E-cadherin levels, and E-cadherin overexpression blocks
AREG-induced invasion
To test the involvement of the adhesion molecule E-cadherin on the proinvasive
effect of AREG, we first examined the impact of AREG on E-cadherin levels. In
agreement with its stimulatory effect on cell invasion, AREG treatment resulted in
suppression of E-cadherin protein levels (Fig. 3.2A: SKOV3; Fig. 3.2B: OVCAR5). We
next tested whether this effect of AREG involved transcriptional regulation of E-
cadherin. Real-time PCR shown that AREG suppresses E-cadherin mRNA significantly
with a maximal effect after 24 hrs of treatment (Fig. 3.2C).
To establish a definite causal link between E-cadherin suppression and an increase
in cell invasiveness, we transfected SKOV3 cells with a human E-cadherin expression
plasmid. Treatment with AREG increased invasion of cells transfected with a control
vector; however, E-cadherin overexpression not only suppressed basal cell invasion but
also decreased AREG-induced invasion (Fig. 3.2D)
3.3.3 AREG suppressed E-cadherin level and promotes cell invasion via the EGFR
To confirm that AREG-induced cell invasion was mediated by the EGFR, we used
the pharmacological EGFR inhibitor, AG1478, to specifically block EGFR activity in
SKOV3 cells. As shown in Fig. 3.3A and B, AG1478 markedly diminished the AREG-
induced reduction in E-cadherin levels and cell invasion.
67
3.3.4 AREG induced Slug expression
AREG strongly induced the expression of Slug and, to a lesser extent, Zeb1 (Fig.
3.4A-D), which are two transcriptional repressors of E-cadherin. Importantly, the
maximal effects on SLUG and ZEB1 mRNA levels preceded changes in E-cadherin levels
(Fig. 3.2C), implying that these transcriptional repressors mediate the suppression of E-
cadherin by AREG.
3.3.5 The MAPK/ERK and PI3K/AKT pathways mediated the effects of AREG on
SLUG mRNA and E-cadherin levels and cell invasion
Treatment of SKOV3 cells with AREG induced rapid activation of ERK and AKT
(Fig. 3.5A). To elucidate whether these molecular pathways mediate these effects of
AREG, SKOV3 cells were co-treated with AREG in the presence of the MEK/ERK
inhibitor, U0126, or the PI3K/AKT inhibitor, LY294002. As shown in Fig. 3.5B and
3.5C, the inhibitors effectively blocked the effects of AREG on SLUG mRNA induction
and E-cadherin suppression. Furthermore, blocking these pathways totally abolished
AREG-stimulated SKOV3 cell invasion (Fig. 3.5D).
3.3.6 AREG induced SPRY4 expression
SPRY deficiencies are common in cancers and may result in excessive growth factor
activities (152). Therefore, it was important to investigate the integrity and functionality
of a SPRY feedback loop in ovarian cancer cells. When SKOV3 cells were treated with
increasing concentrations of AREG, SPRY2 and SPRY4 mRNA levels was elevated, but
SPRY4 was stimulated to a much greater extent (approximately 4 fold versus 10 fold)
(Fig. 3.6A). In a time course experiment, the stimulatory effect of AREG on SPRY4
68
mRNA peaked at 3 hrs (Fig. 3.6B). Induction of SPRY4 expression was also confirmed at
the protein level (Fig. 3.6C).
3.3.7 SPRY4 knockdown enhanced AREG-induced E-cadherin suppression and
invasion
To understand the importance of SPRY4 induction, we evaluated AREG function
in the absence of SPRY4. Using SPRY4-targeting siRNA (siSPRY4), both endogenous
and AREG-induced SPRY4 levels were completely depleted (Fig. 3.7A). The siSPRY4
reduced the basal E-cadherin levels compared to cells transfected with control siRNA
(siCtrl) (Fig. 3.7A). When the transfected cells were treated with AREG, E-cadherin
protein levels were reduced, and this suppressive effect was more obvious in siSPRY4-
transfected cells (Fig. 3.7A). Interestingly, siSPRY4 had no effect on E-cadherin mRNA
level (Fig. 3.7B). We next assayed the influence of SPRY4 on cell invasion. Similar to
the results of E-cadherin levels, siSPRY4 increased the basal level of invasion and
amplified the invasion caused by AREG (Fig. 3.7C, D).
3.4 Discussion
The current study was undertaken to investigate the functional relationships between
AREG and SPRY in ovarian cancer cell invasiveness. We demonstrate that AREG
promotes ovarian cancer cell invasion through activation of ERK and AKT, induction of
Slug and reduction of E-cadherin levels. Furthermore, we showed that AREG elicited a
regulatory feedback response through the induction of SPRY4, which, in turn, reversed
the E-cadherin suppression and antagonised the cell invasion mediated by AREG. We
69
confirmed that these AREG actions are mediated by EGFR (Fig. 3.3). Interestingly, the
EGFR inhibitor and inhibitors of the MEK/ERK and PI3K/AKT pathways also increased
the basal E-cadherin level and reduced basal invasion (Fig. 3.3). These data are explained
by the autocrine nature of AREG and other EGFR ligands; ample evidence have
demonstrated that ovarian cancer cells express these ligands, which constitute autocrine
loops with EGFR (53).
This is the first report to demonstrate an invasion-promoting role for AREG in
ovarian cancer cells (Fig. 3.1A). Although it is not a novel finding that AREG promotes
cancer cell invasion and metastasis, most of the reported proinvasive effects of AREG are
associated with protease-mediated extracellular matrix (ECM) degradation. For example,
urokinase and plasminogen activator inhibitor-1 are responsible for AREG-induced
invasion of breast cancer cells (263). Moreover, the levels and activities of matrix
metalloproteinases are upregulated by AREG to mediate invasion of breast cancer cells
(264, 265), head and neck squamous carcinoma cells (266) and mesothelioma cells (267).
AREG also regulates levels of extracellular matrix metalloproteinase inducers in
transformed breast epithelial cells (264). In addition, integrin, which couples ECM to the
intracellular cytoskeleton and whose level and activation are altered during colon cancer
cell invasion, is induced by AREG (268). The involvement of these molecules in AREG-
induced invasion is a possible explanation to the observation that MEK/ERK and
PI3K/AKT inhibitors have greater effects on invasion than on E-cadherin level (Fig. 3.5
C, D). Here we show that AREG induces SLUG mRNA expression (Fig. 3.4), suppresses
E-cadherin levels (Fig. 3.2) and stimulates ovarian cancer cell invasion (Fig. 3.1).
Similarly, AREG has been reported to reduce E-cadherin level and adherence in
70
keratinocytes (269). Moreover, AREG promotes a reduction in membrane-localized E-
cadherin and a motile morphology in MCDK cells (270), suggesting that E-cadherin is a
common mediator of AREG-stimulated cell motility.
Although SPRY are general inhibitors of signalling downstream of RTKs (127),
there is evidence that individual SPRY proteins are preferentially regulated by and
associate with different growth factors. For example, fibroblast growth factor (FGF)
induces the expression of SPRY2 and SPRY4 (271), whereas SPRY1 mRNA is
transiently downregulated by FGF (272). Furthermore, an insensitivity to the growth
factor-induced upregulation of SPRY has been observed in cancer cell lines. FGF-2
treatment causes a rapid and transient increase in normal prostate epithelial cell SPRY1
mRNA and protein levels; however, FGF-2 reduces SPRY1 mRNA levels in neoplastic
epithelia (152). Refractory to growth factor stimulation results in excessive growth factor
signals and would favour tumorigenesis. In ovarian cancer cells, both SPRY2 and SPRY4
mRNA levels were elevated by AREG, and the increase was substantially higher for
SPRY4 than SPRY2 mRNA (Fig. 3.1A). This result may suggest a stronger functional
relevance between AREG and SPRY4 than SPRY2. However, SPRY4 inhibition on
EGFR activity has been demonstrated in some but not all studies (136, 137, 273, 274).
Therefore it was of great interest to elucidate the regulatory role of SPRY4 on AREG
function. Consistent with the inhibitory nature of SPRY, SPRY4 siRNA amplified the
effect of AREG and led to a further decrease in E-cadherin levels. More importantly,
treating SPRY4-depleted cells with AREG resulted in a more significant increase in
invasion then AREG or siRNA treatment alone. These data indicate that SPRY4 levels
induced by AREG would feed back to counteract the effects of AREG.
71
In sharp contrast to SPRY2, the molecular targets of SPRY4 are poorly defined
(127). SPRY4 binds directly to RAF1 (137) and BRAF (132) to inhibit VEGF-induced
ERK activation (137). SPRY4 also binds constitutively to TESK1 (testicular protein
kinase 1). The SPRY4/TESK interaction is increased by EGF and suppresses integrin-
mediated cell spreading (136, 138). In addition, a recent report showed that SPRY4
overexpression increased levels of E-cadherin (159). In agreement, the current study
showed that the SPRY4 and E-cadherin protein levels were positively associated and that
SPRY4 depletion led to a decrease in E-cadherin level (Fig. 3.7A). All of these data
suggest that E-cadherin is a molecular target of SPRY4 and that it is positively regulated
by SPRY4. The loss of the epithelial marker E-cadherin is a hallmark of epithelial-to-
mesenchymal transition and leads to the loss of cell-cell contact and the adoption of a
motile phenotype, which leads to an increase in invasiveness (Fig. 3.7C, D). Together
with our previous demonstration of the E-cadherin inhibition of ovarian cancer
proliferation (275), SPRY4 may potentially regulate various aspects of ovarian cancer
tumorigenesis through the modulation of E-cadherin.
SPRY have been shown to execute their tumor suppressing roles in response to both
antitumorigenic and oncogenic stimulations. For example, SPRY4 is upregulated by
Wnt7A/Fzd9 and peroxisome proliferator-activated receptor gamma pathway to mediate
their effects on non-small-cell lung cancer cell migration and invasion inhibition (159,
276). On the contrary, in our study, SPRY4 was downstream to an invasion-promoting
growth factor and played a role as a mediator of the inhibitory feedback loop. Similarly,
significant Spry2 mRNA expression is induced in lung epithelium in mice carrying a
germline oncogenic K-rasG12D mutation. Such SPRY2 upregulation is antitumorigenic
72
in nature as number of tumors as well as overall tumor burden is significantly increased
after crossing the mice with a SPRY2-null background (149). These examples indicate
that the upregulation of the tumor suppressor SPRY mediates antitumorigenic functions
during tumorigenesis.
73
A B
C
D
Figure 3.1 AREG promoted ovarian cancer cells invasion. A and B, SKOV3 and
OVCAR5 cells were trypsinized and seeded onto Matrigel-coated transwell inserts with 0
- 100 ng/ml AREG. Cells were allowed to invade for 24 hrs. Invaded cells were stained
and quantified. Data are shown as the mean ± SD of three independent experiments.
Means not denoted by the same letter are significantly different. C and D show
representative photos of the invasion assay.
74
A B
C D
Figure 3.2 AREG reduced E-cadherin levels and E-cadherin overexpression blocked
AREG-induced invasion. SKOV3 (A) and OVACR5 (B) cells were treated with 0 - 100
ng/ml AREG for 24 hrs. E-cadherin and the internal control actin were analyzed using
Western blot. C, SKOV3 cells were treated with AREG over a 24-hr time course. The
effect of AREG on mRNA levels of E-cadherin and the internal control Gapdh were
assayed. D, SKOV3 cells were transiently transfected with control (empty vector) or
human E-cadherin expression (E-cadherin) vectors. After 48 hrs, transfected cell were
seeded onto Matrigel-coated transwell inserts for 24 hrs. Invaded cells were stained and
quantified. Open bars: control treatment; filled bars: AREG treatment (10 ng/ml).
Relative levels are expressed as a percentage of control. Data are shown as the mean ±
SD of three independent experiments (A, B) or triplicates of a representative experiment
(C, D). Means not denoted by the same letter are significantly different.
75
A
B
Figure 3.3 AREG suppressed E-cadherin level and promoted cell invasion via the EGFR.
A, SKOV3 cells were pretreated with AG1478 (10 µM) for 30 min before the addition of
AREG (10 ng/ml) for 24 hrs. Changes in E-cadherin levels were detected by Western
blot. Data are shown as the mean ± SD of three independent experiments. B, SKOV3
cells were trypsinized and incubated with AG1478 for 30 min. Cells were then co-treated
with AREG (10 ng/ml) and seeded onto Matrigel-coated transwells for 24 hrs. Invaded
cells were stained and quantified. Relative levels are expressed as a percentage of control.
Data are shown as the mean ± SD of triplicates of a representative experiment. Means not
denoted by the same letter are significantly different. Open bars: control treatment; filled
bars: AREG treatment.
76
A B
C D
Figure 3.4 AREG induced SLUG mRNA. SKOV3 cells were treated with 100 ng/ml
AREG for 0 to 24 hrs as indicated. SLUG (A) and ZEB1 (B) mRNA levels were then
analyzed by real-time PCR. SKOV3 cells were treated with different doses of AREG for
3 hrs before being collected for SLUG (C) or ZEB1 (D) real-time PCR. Relative mRNA
levels of SLUG and ZEB1 were expressed as a percentage of the control. Data are shown
as the mean ± SD of triplicates of a representative experiment. Means not denoted by the
same letter are significantly different.
77
A B
C D
78
Figure 3.5 The MAPK/ERK and PI3K/AKT pathways mediated the effects of AREG on
Slug and E-cadherin level and cell invasion. A, SKOV3 cells were treated with 10 ng/ml
AREG for 0.5 and 3 hrs. ERK and AKT activation were detected using Western blot
assays. B and C, SKOV3 cells were pre-incubated with U0126 (10 µM) and LY294002
(25 µM) for 30 min prior to co-treatment with AREG. Cells were harvested after 3 hrs
and SLUG mRNA level was analyzed by real-time PCR. E-cadherin protein level was
analyzed by Western blot after 24 hrs. D, trypsinized SKOV3 cells were pre-treated with
U0126 (10 µM) and LY294002 (25 µM) for 30 min and co-treated with AREG in
Matrigel-coated transwell inserts for 24 hrs. Invaded cells were stained and quantified.
Relative levels are expressed as a percentage of the control. Data are shown as the mean
± SD of triplicates of a representative experiment (B) or three independent experiments
(C, D). Means not denoted by the same letter are significantly different. Open bars:
control treatment; filled bars: AREG treatment.
79
A
C
B
Figure 3.6 AREG induced SPRY4. SKOV3 cells were treated with AREG at various
concentrations (A) or for various durations (B). The mRNA levels of SPRY2 and SPRY4
were detected with real-time PCR. C, SKOV3 cells were treated with various
concentrations of AREG, and the effect on the SPRY4 level was assayed using Western
blotting. The relative levels are expressed as a percentage of the control. The data are
shown as the mean ± SD of triplicates of a representative experiment (A, B) or three
independent experiments (C). Means not denoted by the same letter are significantly
different.
80
A B
C
D
81
Figure 3.7 SPRY4 knockdown enhanced AREG-induced E-cadherin suppression and
invasion. A, SKOV3 cells were transiently transfected with control siRNA (siCtrl) or
SPRY4 siRNA (siSPRY4) for 48 hrs. After transfection, the cells were treated with 10
ng/ml AREG for 24 hrs, and E-cadherin and SPRY4 levels were assayed through
immunoblotting. Data is shown as mean ± SD of three independent experiments. B, The
transfected cells were treated with 10 ng/ml AREG for 12 hrs and subjected to real-time
PCR to determine E-cadherin mRNA level. Data is shown as mean ± SD of triplicates in
a representative experiment. C, At 48 hrs after transfection, SKOV3 cells were
trypsinised, seeded in Matrigel-coated transwells and treated with 10 ng/ml AREG for 24
hrs. The invading cells were stained and quantified. The data are shown as the mean ± SD
of triplicates in a representative experiment. D, Representative images of the invasion
assay. Relative levels were expressed as a percentage of control. Means not denoted by
the same letter are significantly different. Open bars: control treatment; filled bars: AREG
treatment.
82
Chapter 4 Sprouty4 feedback regulates epidermal growth factor/AKT/hypoxia-
inducible factor-1 alpha axis in ovarian cancer cells
4.1 Introduction
Functioning as a negative feedback regulator of receptor tyrosine kinase (RTK)/ERK
signalling, Sprouty (SPRY) expression is primarily regulated by RTK activation. Growth
factor stimulation has been widely demonstrated to be essential and sufficient for SPRY
gene expression (122, 145, 277, 278). Various groups have shown that different growth
factors and phorbol 12,13-dibutyrate induce SPRY expression in an ERK pathway-
dependent manner, downstream from RTK activation (147, 271). In human tumor cells
with constitutive ERK activation, SPRY levels are elevated, which can be reduced by an
ERK pathway inhibitor (147).
To elucidate the regulatory mechanisms that control SPRY levels, the human SPRY2
and SPRY4 promoters have been recently cloned and multiple potential transcription
factor binding sites were identified on these promoters (279, 280), thereby suggesting the
regulation of SPRY by pathways other than ERK signalling. A putative hypoxia-
inducible factor-1 alpha (HIF-1α) binding site has been identified on the human SPRY4
promoter (280). Accordingly, SPRY4 level has been shown to be elevated by chemical-
mimics of hypoxia in both normal and malignant cells (281), suggesting a positive effect
of HIF-1α on SPRY4 levels.
In addition to tissue hypoxia, HIF-1α levels and HIF-1 activity have been shown to be
modulated by several environmental (211, 282) and hormonal factors (174, 283),
including growth factors (188, 191). Our laboratory has recently reported that epidermal
83
growth factor (EGF) could strongly elevate HIF-1α levels in ovarian cancer cells, which
mediated EGF-induced E-cadherin downregulation and cell invasion (unpublished data).
We sought to investigate whether HIF-1α plays a role in mediating the regulatory effect
of EGF on SPRY4 level.
Emerging experimental data shows that SPRY could be regulated by and modulate
additional signalling pathways. For example, fibroblast growth factor (FGF) induced
SPRY1 and SPRY2 expression via a Ca2+-dependent pathway (284), while Xenopus
SPRY can inhibit calcium signalling (285). In colon cancer cells, the SPRY2 levels were
suppressed by an active vitamin D metabolite (1,25(OH)2D3) through an E-cadherin-
dependent mechanism, and SPRY2, in turn, repressed 1,25(OH)2D3-induced E-cadherin
expression (286). Recently, an HIF-1α regulator, Seven-in-Absentia homologue-2
(SIAH2) (169), was found to interact with SPRY4 (287), which suggests that SPRY4
may play a role in HIF-1α regulation. However, direct information regarding the effect of
SPRY4 on HIF-1α levels and a functional linkage between SPRY4 and HIF-1α are still
lacking.
We demonstrated that EGF has a strong inducing effect on the SPRY4 level in
ovarian cancer cells via an AKT- and HIF-1α- dependent mechanism. Functionally,
SPRY4 feedback inhibits AKT activation and antagonizes EGF-induced HIF-1α levels
and HIF-1 activity. Together, our data demonstrate the presence and functionality of a
novel EGFR/AKT/HIF-1α and SPRY4 feedback loop in ovarian cancer cells.
84
4.2 Materials and methods
4.2.1 Cell culture and reagents
SKOV3, OVCAR3 and OVCAR4 ovarian cancer cell lines were obtained from
American Type Culture Collection (Manassas, VA). The cell lines were cultured in
MCDB 105/M199 (1:1), supplemented with 5% heat-inactivated fetal bovine serum, 100
IU/ml penicillin and 100 g/ml streptomycin. The cells were cultured at 37°C and 5%
CO2. Fetal bovine serum was purchased from Hyclone Laboratories (Logan, UT). The
MEK/ERK inhibitor, U0126, was purchased from Calbiochem (San Diego, CA). Human
recombinant EGF, MEK/ERK inhibitor (PD98059), PI3K/AKT inhibitors (LY294002
and Wortmannin) and other tissue culture materials were obtained from Sigma Chemical
Co. (St. Louis, MO), unless otherwise stated.
4.2.2 Transfection
SPRY4 promoter-luciferase reporter constructs were kindly provided by Dr. D.
Warburton (Children Hospital Los Angeles, California). The FLAG-SPRY4 construct
and the empty pXJ40-FLAG vector were generous gifts from Dr. Graeme R. Guy
(Institute of Molecular and Cell Biology, Singapore). The hypoxia responsive element
(HRE)-luciferase reporter construct was purchased from Addgene (Cambridge, MA). The
plasmids were transfected using Lipofectamine™ 2000 (Invitrogen, Carlsbad, CA). ON-
TARGETplus SMARTpool non-targeting (siCtrl), HIF-1α siRNA (HIF-1α), SPRY4
siRNA (siSPRY4) and non-targeting siRNA (Dharmacon, Lafayette, CO) were
transfected using Lipofectamine RNAiMAX (Invitrogen).
85
4.2.3 Real-time PCR
Total RNA extracted from cells using TRIzoL Reagent (Invitrogen) was used in
first-strand DNA (cDNA) synthesis using Invitrogen Super-ScriptTM first-strand
synthesis system for real-time PCR according to the manufacturer’s protocol. Real-time
PCR was performed using an ABI 7300 real-time thermal cycler (ABI, Hercules, CA).
The detections of SPRY4, VEGF, and the internal control, GAPDH, mRNAs were
performed as follows: a 3-min hot start at 95ºC followed by 40 cycles of denaturation at
95ºC for 15 sec, and amplification at 60ºC for 1 min. PCR reactions were performed in
duplicates with the following PCR primers: SPRY4, forward 5’-
AGCCTGTATTGAGCGGTTTG-3’ and reverse 5’-GGTCAATGGGTAGGATGGTG-
3’; VEGF, forward 5’-GGCTCTAGATCGGGCCTCCGAAACCAT-3’ and reverse 5’-
GGCTCTAGAGCGCAGAGTCTCCTCTTC-3’; and GAPDH, forward 5’-
GAGTCAACGGATTTGGTCGT-3’ and reverse 5’-GACAAGCTTCCCGTTCTCAG-3’.
4.2.4 Western blot analysis
Equal amounts of total cell lysates were resolved in 7.5% SDS-PAGE and
electrotransferred to a PVDF membrane. After blocking for 1 hr with 5% non-fat dry
milk in TBS-T buffer (20 mM Tris-HCl, pH 7.4, 150 mM NaCl, 0.1% Tween-20), the
blots were probed for overnight at 4 ºC with the appropriate primary antibodies. The
antibody for HIF-1α was purchased from BD Transduction Laboratories (Lexington,
KY), the anti-SPRY4 antibody was obtained from Abcam (Cambridge, MA), the anti-β-
actin antibody was obtained from Santa Cruz Biotechnology (Santa Cruz, CA), the anti-
phospho-EGFR (Tyr992), anti-EGFR, anti-phospho-AKT (Ser473), anti-total AKT, anti-
86
phospho-ERK1/2 (Thr202/Tyr204), anti-ERK1/2 and PTEN antibodies were purchased
from Cell Signaling, Inc. (Austin, TX). The blots were then treated with a peroxidase-
conjugated secondary antibody (Bio-Rad) for 1 hr followed by a detection step with ECL
chemiluminescence reagent (Amersham, Arlington Heights, IL) and exposure to X-ray
films.
4.2.5 Luciferase assay
Cells transfected with luciferase reporter constructs were treated with EGF for 24
hours. After treatment, the cells were lysed, and total cell lysates were centrifuged at
13,000 rpm for 10 min to remove any cell debris. Luciferase activity was determined
according to the manufacturer’s instructions (Promega).
4.2.6 Statistical analyses
For real-time PCR data, the relative quantification of levels was calculated using
the 2–ΔΔ Ct method. Data are presented as the mean ± SD of three independent
experiments or triplicates in a representative experiment and were analyzed by one-way
ANOVA followed by Tukey’s post-hoc test using GraphPad Prism 5 (GraphPad
Software, San Diego, CA) to compare all pairs of columns. Columns are not denoted by
the same letter are statistically different. Means not denoted by the same letter are
significantly different (P < 0.05).
87
4.3 Results
4.3.1 EGF increased SPRY4 in ovarian cancer cells
To test the effect of EGF on the SPRY4 level, SKOV3 cells were challenged with
either 100 ng/ml EGF over a 24-hr time course or various EGF concentrations for 3 hrs.
SPRY4 mRNA levels were significantly elevated (Fig. 4.1A, B). An inductive effect of
EGF was also observed in two additional ovarian cancer cell lines, OVCAR3 and
OVACR4 (data not shown). A similar robust inducing effect was also observed at the
SPRY4 protein level (Fig. 4.1C). To evaluate the contribution of transcriptional induction
via promoter activation in SPRY4 mRNA induction, we transiently transfected reporter
gene constructs containing different lengths of the SPRY4 5’ flanking region into SKOV3
cells. Under EGF treatment, no induction was observed in the minimal promoter region (-
31/+56). The promoter region (-69/+56) contains the consensus sequence for the binding
of HIF-1 and signal transducer and activator of transcription 5. In addition, the promoter
region (-4446/+56) contains many putative binding sites for transcription factors,
including stimulating protein 1 (Sp1), activator protein 2, Elk-1, and WT-1 (280). Under
EGF treatment, luciferase activities driven by regions (-69/+56) and (-4446/+56) were
significantly induced when compared with the control treatment (Fig. 4.1D).
4.3.2 The MEK/ERK and PI3K/AKT pathways mediated the effects of EGF on
SPRY4 levels
To evaluate the role of the MEK/ERK and PI3K/AKT pathways in EGF-induced
SPRY4, SKOV3 cells were pre-incubated with inhibitors for the specific pathways for 30
mins. The MER/ERK inhibitor significantly abolished the effect of EGF on both SPRY4
88
mRNA and SPRY4 protein (Fig. 4.2A and 4.2B). Although PI3K inhibitor partially
blocked SPRY4 mRNA increased by EGF (Fig. 4.2A), it blocked completely blocked the
increase in SPRY4 levels (Fig. 4.2B).
4.3.3 EGF induced HIF-1α via the PI3K/AKT pathway
Our previous and present reports demonstrated a strong effect of EGF on HIF-1α
levels (Fig. 4.3A and Cheng et al., unpublished). When pre-treated with the PI3K/AKT
pathway inhibitor, LY294002, both basal and EGF-induced HIF-1α levels were reduced
significantly, whereas the MEK/ERK inhibitor slightly suppressed EGF-induced HIF-1α
levels (Fig. 4.3B). Moreover, the cells were transfected with an HRE-luciferase reporter
gene construct by which the activity was driven by HIF. EGF-induced HRE-luciferase
activity was completely blocked by the PI3K/AKT pathway inhibitors, Wortmannin and
LY294002, and not by the MER/ERK inhibitor (Fig. 4.3C).
4.3.4 HIF-1α plays a minor role in EGF-induced SPRY4 level
The presence of the HIF responsive element (HRE) in the human SPRY4 promoter
(280) prompted us to question whether HIF-1α mediates the effect of EGF on SPRY4
regulation. At the promoter level, we performed site-directed mutagenesis of the HRE
located within the SPRY4 promoter (Fig. 4.4A). EGF increased luciferase activity in cells
transfected with the wild-type promoter construct. In contrast, mutation of the HRE in the
SPRY4 promoter reduced both the basal and luciferase activities induced by EGF (Fig
4.4A). Next, we depleted endogenous HIF-1α expression using specific siRNA (Fig.
4.4B). However, we did not observe a significant reduction in EGF-induced SPRY4
mRNA levels (Fig. 4.4C) following HIF-1α knockdown, although the mRNA level of the
89
HIF-1α target, VEGF, increase by EGF being partially reduced (Fig. 4.4C). Accordingly,
HIF-1α knockdown did not alter the effect of EGF on SPRY4 levels (Fig. 4.4D).
4.3.5 SPRY4 overexpression reversed EGF-induced HIF-1α levels and HIF-1
activity
We next investigated the mechanism by which SPRY4 impacts HIF-1α levels and
HIF activity. We overexpressed human SPRY4 in SKOV3 cells, which express low
levels of SPRY4. Overexpression was confirmed by Western blot (Fig. 4.5A). After EGF
treatment, HIF-1α levels were elevated. However, SPRY4-overexpressing cells displayed
lower HIF-1α levels when compared with cells transfected with the control vector (Fig.
4.5A). To evaluate whether the altered HIF-1α levels were accompanied by a change in
HIF-1 activity, two experiments were conducted. First, an HRE-luciferase reporter gene
construct was co-transfected with SPRY4 overexpression vector into SKOV3 cells. EGF
strongly increased the HRE-mediated luciferase activity, which was markedly reduced by
SPRY4 overexpression (Fig. 4.5B). Similarly, SPRY4 also antagonized the inductive
effect of EGF on VEGF mRNA (Fig 4.5C).
4.3.6 SPRY4 knockdown enhanced the effect of EGF on HIF-1α
To eliminate the possibility of non-specific effects due to massive SPRY4
overexpression, we confirmed our observation via SPRY4 knockdown using specific
siRNA in cell lines expressing higher levels of endogenous SPRY4 (OVCAR3 and
OVACAR4). In siSPRY4-treated OVCAR3 cells, EGF elicited a higher HIF-1α level
than that of the cells treated with siCtrl (Fig. 4.6A, C).
90
4.3.7 AKT pathway mediated HIF-1α regulation by EGF and SPRY4
In the overexpression and knockdown experiments, we observed changes in the
phospho-AKT (pAKT) levels similar to those observed for HIF-1α (Fig. 4.5A, 4.6A). In
SPRY4-overexpressing SKOV3 cells, the pAKT level was lower than that of the control
cells (Fig. 4.5A). In contrast, AKT activation in SPRY4 siRNA-treated OVCAR3 cells
was increased (Fig. 4.6A, C). To further confirm the causal role of the elevated AKT
activation of HIF-1α levels, we co-incubated the cells with PI3K/AKT inhibitors. In
OVCAR3 cells, when AKT activation was inhibited, the SPRY4-augmented EGF-
induced HIF-1α expression was completely impaired (Fig. 4.7).
4.4 Discussion
Despite the importance of RTKs and its downstream ERK pathway in the regulation
of SPRY expression is well-known (127, 147, 271), the detailed molecular mechanisms
of SPRY regulation are not completely understood. In the present study, we confirmed
the importance of ERK in SPRY4 expressions as MEK/ERK inhibitor reduced the basal
levels of SPRY4 protein (Fig. 4.2), furthermore we detected a strong stimulation of the
SPRY4 level by EGF that was primarily mediated by the ERK pathway (Fig. 4.2). In
addition, the (-4446/+56) promoter construct responded to EGF more prominently than
the region (-69/+56), this is probably due to the region (-4446/+56) contains many
putative sites for transcription factors downstream to ERK, for example Sp1 and Elk-1
(280). In contrast, inhibition of the PI3K/AKT pathway exerted more prominent
inhibitory effect on and SPRY4 protein than SPRY4 mRNA levels (Fig. 4.2), thereby
91
suggesting differential regulatory mechanisms at the transcriptional and post-
transcriptional levels by the AKT pathway. AKT has been shown to bind to an E3 ligase,
SIAH2 (288), which can directly interact with SPRY4 (287). This finding suggests that
AKT regulates SPRY4 by modifying SPRY4 degradation and has more significant effects
on SPRY4 protein than SPRY4 mRNA.
When SKOV3 cells were treated with two different PI3K/AKT inhibitors, the basal
HRE-luciferase activities were reduced; this observation agrees with the previous finding
that basal HIF-1α levels in ovarian cancer cell lines were PI3K-dependent (289). We also
showed that EGF induced HIF-1α accumulation and activity via the PI3K/AKT pathway
(Fig. 4.3). This A recent article showing a positive effect of hypoxia on SPRY4 level
(281), therefore we then tested the cross-talk of EGF and HIF-1α on SPRY4 regulation.
Mutation of the HRE within the human SPRY4 promoter, which impaired both basal and
EGF-induced SPRY4 promoter activity, highlights the importance of HIF-1α in SPRY4
transcriptional regulation. However, HIF-1α knockdown failed to decrease EGF-induced
SPRY4 mRNA and SPRY4 protein levels (Fig. 4.4C, D). The ineffectiveness of HIF-1α
knockdown could be attributed to the dominance of the ERK signal over the HIF-1α
pathway, thereby rendering the effects insignificant. In addition, the human SPRY4
promoter HIF-1α overlaps with CpG islands, which has been found to be
hypermethylated in prostate cancer (161). The SPRY4 promoter is likely hypermethylated
in ovarian cancer cell lines, as SPRY4 expression could be reactivated using a
demethylating agent treatment (unpublished data). Methylation may hinder HIF-1α
accessibility to HRE, thus rendering the promoter refractory to the HIF-1α level
dynamics. The same concept has been applied to a well-known HIF-1α target, HIF
92
prolyl-hydroxylase 3 (PHD3). The PHD3 promoters in various human carcinoma cell
lines have been silenced by aberrant methylation, causing these cells to express reduced
basal and hypoxia-upregulated PHD3 mRNA (290). This notion also provides an
explanation by which the promoter construct can be activated by EGF via the HRE,
regardless of the endogenous SPRY4 promoter methylation status.
Hypoxia stabilizes and causes the accumulation of HIF-1α, which acts as the
regulator of a repertoire of hypoxic responses, including angiogenesis (163). To date, a
direct functional connection between SPRY4 and HIF-1α has not been described,
although there are lines of experimental evidence suggesting that SPRY4 and HIF-1α are
functionally linked. As demonstrated with a Spry4-knockout mouse model and ex vivo
assays, SPRY4 functions as a negative angiogenic regulator, for example, SPRY4-
negative tumors have enhanced vascularisation (274). Strikingly, in SPRY4-knockout
cells, growth factors such as bFGF and VEGFA induced a greater AKT activation (274).
However, not addressed in that study was the negative effect of SPRY4 on HIF-1α levels
and activity, possibly an additional cause of the SPRY4 antiangiogenic effect. Our
observation supports a novel regulatory involvement of SPRY4 on HIF-1α.
Next, we investigated the mechanism underlying the effect of SPRY4 on EGF-
induced HIF-1α expression. We first delineated the signalling pathways mediating the
effect of EGF. Pre-incubation with PI3K/AKT inhibitors but not MEK inhibitor
effectively blocked the basal and EGF-induced HIF-1α and HIF-1 activity (Fig. 4.3B, C).
The importance of the PI3K/AKT pathway in EGF-induced HIF-1α accumulation has
been previously reported in prostate cancer cells (188). We observed concurrent increases
93
in EGF-induced HIF-1α level and AKT activation in SPRY4 siRNA-treated cells,
whereas the levels of both HIF-1α and phospho-AKT were reduced in SPRY4
overexpressing cells (Fig. 4.5, 4.6). The inhibition of AKT activation abrogated the effect
of SPRY4 siRNA on HIF-1α expression (Fig. 4.7). These data suggest that SPRY4
regulates HIF-1α via EGF-induced AKT signalling. Although inhibition of the
RTK/RAS/MAPK/ERK pathway is the principal action of SPRY (127), emerging data
have suggested that AKT is also a target of SPRY activity (134, 274). Another SPRY
member, SPRY2, has been shown to regulate AKT activity by modulating the content
and activity of phosphatase and tensin homologue deleted on chromosome 10 (PTEN)
(134). However, PTEN levels were not affected by both SPRY4 knockdown and
overexpression (Fig. 4.5A, 4.6A), therefore the mechanism by which SPRY4 regulates
AKT and HIF-1α remains unknown.
94
A B
C D E
Figure 4.1 EGF induced SPRY4 expression in ovarian cancer cells. SKOV3 cells were
treated with EGF time (A) or dose (B) courses and assayed for SPRY4 mRNA level using
real-time PCR. The relative levels of SPRY4 mRNA are expressed as a percentage of the
control. C. SKOV3 cells were treated with 10 ng/ml EGF for various durations and
assayed for SPRY4 protein level by immunoblotting. D. SKOV3 cells were transfected
with SPRY4 promoter constructs and treated with EGF for 24 hrs. Cell lysate was
collected, and luciferase activity was assessed. The relative levels of luciferase activity
are expressed as a percentage of the control treatment of (-31/+56). E. OVCAR3 and
OVCAR4 cells were treated with 100 ng/ml EGF for 3 hrs and SPRY4 mRNA were
assayed using real-time PCR. The data are presented as the mean ± SD of three
independent experiments (A, C) or triplicates in a representative experiment (B, D, E).
The mean values that are not denoted by the same letter are significantly different.
95
A
B
Figure 4.2 The MEK/ERK and PI3K/AKT pathways mediated the effects of EGF on
SPRY4 mRNA (A) and SPRY4 protein (B) levels. SKOV3 cells were pre-incubated with
either MEK/ERK inhibitor (U0126, 10 µM) or PI3K/AKT inhibitor (LY294002, 25 µM)
for 30 min prior to EGF (10 ng/ml) treatment. The cells were harvested for real-time PCR
(A) and Western blot (B) analysis after 3 hrs and 24 hrs, respectively. Relative levels of
SPRY4 mRNA and SPRY4 are expressed as a percentage of the control. The data are
presented as the mean ± SD of three independent experiments. Mean values that are not
denoted by the same letter are significantly different.
96
A
B C
Figure 4.3 EGF induced HIF-1α and HIF-1 activity via the PI3K/AKT pathway. A.
SKOV3 cells were treated with EGF for different durations and HIF-1α protein levels
were assessed. B. SKOV3 cells were pre-incubated with PI3K/AKT inhibitor
(LY294002, 25 µM) and MEK/ERK inhibitor (PD98059, 10 µM) for 30 min prior to
EGF treatment. Their effects on HIF-1α protein were assayed after 30 min. C. SKOV3
cells were transfected with HRE-luciferase construct, pre-treated with PI3K/AKT
inhibitors (LY294002, 25 µM; Wortmannin 1 µM) and MEK/ERK inhibitor (U0126, 10
µM) and treated with EGF for 24 hrs. Cell lysates were collected, and luciferase activity
was assessed. The relative levels of luciferase activity are expressed as a percentage of
the control. The data are shown as the mean ± SD of triplicates in a representative
experiment. Mean values that are not denoted by the same letter are significantly
different.
97
A B
C D
E
98
Figure 4.4 HIF-1α plays a minor role in EGF-induced SPRY4 expression. A. A diagram
displaying the site-directed mutagenesis of the HIF-1α binding site of the SPRY4 (-
69/+56) promoter region (upper panel). SKOV3 cells were transfected with SPRY4 (-
69/+56) promoter-luciferase constructs containing either wild-type or mutated HRE.
After treatment with 10 ng/ml EGF for 24 hrs, cell lysates were collected for the
luciferase activity assay. The relative levels of luciferase activity are expressed as a
percentage of the control (lower panel). B. SKOV3 cells were transfected with either 50
nM non-targeting siRNA (siCtrl) or HIF-1α siRNA (siHIF-1α) and treated with 10 ng/ml
EGF for 3 hrs. Cells were collected for Western blot analysis of HIF-1α protein (B) as
well as real-time PCR for SPRY4 (C) and VEGF (D) mRNAs. E. Non-targeting siRNA or
HIF-1α siRNA-transfected cells treated with EGF for 24 hrs were harvested, and SPRY4
level was assayed by Western blot. The relative levels are expressed as a percentage of
the control. The data are shown as the mean ± SD of triplicates in a representative
experiment (A) or three independent experiments (C, D, E). Mean values that are not
denoted by the same letter are significantly different.
99
A
B C
100
Figure 4.5 SPRY4 overexpression reversed EGF-induced HIF-1α expression and
HIF-1 activity. A. SKOV3 cells were transfected with either SPRY4
overexpression or empty vectors. Transfected cells were treated with EGF for 3
hrs, and cell lysates were collected for HIF-1α protein analysis. SKOV3 cells
were co-transfected with HRE-luciferase construct and either SPRY4
overexpression vector or control vector. After treatment with EGF for 24 hrs, cell
lysates were collected for luciferase assay. B. SKOV3 cells were transfected with
the SPRY4 overexpression vector and treated with EGF. After a 3-hr treatment,
RNA was collected and VEGF mRNA level was assayed by real-time PCR.
Relative activity levels are expressed as a percentage of the control. The data are
shown as the mean ± SD of three independent experiments. Mean values that are
not denoted by the same letter are significantly different.
101
A
B C
Figure 4.6 SPRY4 knockdown enhanced EGF effect on HIF-1α. A, OVCAR3 cells were
transfected with 50 nM siCtrl or siSPRY4 for 48 hrs. Transfected cells were treated with
10 ng/ml EGF for 30 min. Levels of HIF-1α, PTEN, and pAKT were then detected by
Western blot. Quantification of pAKT (B) and HIF-1α (C) levels. The relative levels are
expressed as a percentage of the EGF-treated siCtrl cells. The data are shown as the mean
± SD of three independent experiments. Means not denoted by the same letter are
significantly different. Open bars: siCtrl transfection; filled bars: siSPRY4 transfection.
102
A
B
Figure 4.7 The PI3K/AKT pathway mediated HIF-1α regulation by EGF and SPRY4. A,
OVCAR3 cells were transfected with 50 nM siCtrl or siSPRY4 for 48 hrs. Transfected
cells pre-treated with PI3K/AKT inhibitors (Wortmannin, 1 µM and LY294002, 25 µM)
for 30 min and then treated with 10 ng/ml EGF for 30 min. HIF-1α and pAKT were
detected by Western blot. B, Quantification of HIF-1α levels. The relative levels are
expressed as a percentage of the EGF-treated siCtrl cells. The data are shown as the mean
± SD of three independent experiments. Open bars: siCtrl transfection; filled bars:
siSPRY4 transfection.
103
Chapter 5 Hypoxia-inducible factor-1 alpha destabilization by Sprouty4, independent of
prolyl hydroxylases activity
5.1 Introduction
Hypoxia-inducible factors (HIFs) regulate O2-homeostasis during both physiological
and pathological processes (163). HIFs are heterodimeric transcription factors composed
of a constitutively expressed nuclear protein HIF-1β and one of the three closely related
forms of the α subunit (HIF-1α, HIF-2α and HIF-3α). The activity of HIF-1 is tightly
controlled through the availability of the α subunit, as the half-life of HIF-1α is
extremely short in normoxia (165). Under well-oxygenated conditions, the oxygen-
dependent degradation (ODD) domain of HIF-1α is hydroxylated by prolyl hydroxylases
(PHDs) and allows binding of the Von Hippel-Lindau (VHL) E3 ligase, which
ubiquitinates HIF-1α and targets HIF-1α for proteasomal degradation (166). There are
three PHDs: PHD1 and PHD3 hydroxylate HIF-1α in hypoxia, whereas PHD2 is the
main PHD responsible for normoxic hydroxylation (291). As the enzymatic activities of
PHDs require O2 as a co-substrate, under hypoxia, PHDs are inactive and hydroxylation
of HIF-1α is suppressed; HIF-1α is thereby stabilized, which then translocates into the
nucleus and dimerizes with HIF-1β to constitue HIF-1. The HIF-1 transcription factor
binds to hypoxia-responsive element (HRE) and induces gene transcription. Moreover,
HIF-1α regulation is further complicated by the SIAH E3 ubiquitin ligases. SIAH ligases
are responsible for the degradations of several proteins, including PHDs. Overexpression
of SIAH2 decreased cellular PHD levels, thereby allowing the stabilization and
accumulation of HIF-1α (169).
104
In addition to tissue hypoxia, HIF-1α levels and activity have been shown to be
modulated by several genetic (197), environmental (211, 282) and hormonal factors (174,
283), including growth factors (188, 191). Recently, our laboratory has reported that HIF-
1α is regulated by epidermal growth factor (EGF) and the EGFR regulator, Sprouty4
(SPRY4) (So et al. unpublished).
As regulators of EGFR and other receptor tyrosine kinases (RTKs), SPRY proteins
function as tumor-suppressor genes and have been implicated in many tumorigenic
processes, including angiogenesis (292). Direct evidence of SPRY involvement in
angiogenesis comes from a study in which tumor cells were transplanted into Spry4-
knockout mice and found to grow much faster and be associated with enhanced
neovascularisation compared to those in wild-type mice (274). Moreover, SPRY4
overexpression in embryonic endothelial cells was shown to minimise branching and
sprouting of small vessels (135).
Although SPRY4 and HIF-1α may be functionally linked, the impact of SPRY4 on
the HIF-1α pathway has not been clearly elucidated. Moreover, SPRY4 has been recently
demonstrated to interact with SIAH2 (287), which opens the possibility that, through
interaction with SIAH, SPRY4 regulates HIF-1α hydroxylation and, therefore, HIF-1α
degradation.
We have demonstrated in this report that SPRY4 negatively regulates HIF-1α levels,
as HIF-1α levels were increased in SPRY4 siRNA-treated cells. Accordingly, HIF-1
activity and levels of HIF-1 target genes were modulated by SPRY4. Without altering
105
HIF-1α mRNA levels, SPRY4 was found to increase the half-life of HIF-1α protein, in a
mechanism independent of PHD activity and HIF-1α hydroxylation.
5.2 Materials and methods
5.2.1 Cell culture and reagents
SKOV3, OVCAR3 and OVCAR4 ovarian cancer cell lines were obtained from
American Type Culture Collection (Manassas, VA). The cell lines were cultured in
MCDB 105/M199 (1:1), supplemented with 5% heat-inactivated fetal bovine serum, 100
IU/ml penicillin and 100 g/ml streptomycin. The cells were cultured at 37°C and 5%
CO2. Fetal bovine serum was purchased from Hyclone Laboratories (Logan, UT).
Human recombinant EGF, cyclohexamide (CHX), MG132 and other tissue culture
materials were obtained from Sigma Chemical Co. (St. Louis, MO) unless otherwise
stated.
5.2.2 Transfection
The FLAG-SPRY4 construct and the empty pXJ40-FLAG vector were generous
gifts from Dr. Graeme R. Guy (Institute of Molecular and Cell Biology, Singapore). The
hypoxia responsive element-luciferase reporter construct, ODD-domain luciferase fusion
construct, PHD2 expression vector, HIF-1α expression vector (WT HIF-1α) and double-
mutant HIF-1α (Pro402Ala and Pro564Ala) (DM HIF-1α) expression vector were
purchased from Addgene (Cambridge, MA). Plasmids were transfected using
Lipofectamine™ 2000 (Invitrogen, Carlsbad, CA). ON-TARGETplus SMARTpool
106
SPRY4 siRNA and non-targeting siRNA (Dharmacon, Lafayette, CO) were transfected
using Lipofectamine RNAiMAX (Invitrogen).
5.2.3 Real-time PCR
Total RNA extracted from cells using TRIzoL Reagent (Invitrogen) was used in
first-strand DNA (cDNA) synthesis using Invitrogen Super-ScriptTM first-strand
synthesis system for real-time PCR according to the manufacturer’s protocol. Real-time
PCR was performed using an ABI 7300 real-time thermal cycler (ABI, Hercules, CA).
The detections of HIF-1α, PAI-1, PHD3 and the internal control, GAPDH, were
performed as follows: a 3-min hot start at 95ºC followed by 40 cycles of denaturation at
95ºC for 15 sec, and amplification at 60ºC for 1 min. PCR reactions were performed in
duplicate with the following PCR primers: HIF-1α, 5’-
TCATCCAAGAAGCCCTAACG-3’ and reverse 5’-TCGCTTTCTCTGAGCATTCTGC-
3’; PAI-1 forward 5’-GGACAGACCCTTCCTCTTTGT-3’ and reverse 5’-
TCCATCACTTGGCCCATGAA-3; PHD3, forward 5’-ATCAGCTTCCTCCTGTCCC-
3’ and reverse 5’-CAGCGACCATCACCGTTG-3’; and GAPDH, forward 5’-
GAGTCAACGGATTTGGTCGT-3’ and reverse 5’- GACAAGCTTCCCGTTCTCAG-
3’.
5.2.4 Western blot analysis
Equal amounts of total cell lysates were resolved on 7.5% SDS-PAGE gels and
electrotransferred to a PVDF membrane. After blocking for 1 hr with 5% non-fat dry
milk in TBS-T buffer (20 mM Tris-HCl, pH 7.4, 150 mM NaCl, 0.1% Tween-20), the
blots were probed for overnight at 4 ºC with primary antibodies. The HIF-1α antibody
107
was purchased from BD Transduction Laboratories (Lexington, KY), the anti-SPRY4
antibody was obtained from Abcam (Cambridge, MA), and the anti-β-actin antibody was
obtained from Santa Cruz Biotechnology (Santa Cruz, CA). The anti-hydroxylated-HIF-
1α (Pro402) antibody was obtained from Bethyl Laboratories, Inc. (Montgomery, TX).
The anti-hydroxylated-HIF-1α (Pro564) was purchased from Cell Signaling, Inc. (Austin,
TX). The blots were then incubated with a peroxidase-conjugated secondary antibody
(Bio-Rad) for 1 hr, followed by detection with ECL chemiluminescence reagent
(Amersham, Arlington Heights, IL) and exposure to X-ray films.
5.2.5 Statistical analysis
For real-time PCR data, the relative quantification of levels was calculated using
the 2–ΔΔ Ct method. Data are presented as the mean ± SD of three independent
experiments or triplicates in a representative experiment and were analyzed by one-way
ANOVA followed by Tukey’s post-hoc test using GraphPad Prism 5 (GraphPad
Software, San Diego, CA) to compare all pairs of columns. Columns are not denoted by
the same letter are statistically different. Means not denoted by the same letter are
significantly different (P < 0.05).
5.3 Results
5.3.1 SPRY4 negatively regulated HIF-1α expression levels in ovarian cancer cells
To test the effect of SPRY4 on HIF-1α levels, we used either siRNA knockdown or
overexpression to manipulate SPRY4 levels in the cells. SPRY4 siRNA effectively
108
depleted endogenous SPRY4 protein and caused an increase in HIF-1α levels in the three
cell lines tested, with OVCAR4 displaying the most significant response (Fig. 5.1A).
Consistent with the siRNA experiment, SPRY4 overexpression decreased the HIF-1α
content in the cell lines (Fig. 5.1B). Furthermore, we found that the effect was specific for
SPRY4, as HIF-1α expression levels were not increased in cells treated with SPRY2
siRNA (Fig. 5.1C). Among the three cell lines, OVCAR4 displays the highest levels of
endogenous SPRY4 expression and clearly responds to SPRY4 siRNA; therefore,
OVACR4 was used for most of the subsequent experiments.
5.3.2 SPRY4 negatively regulated HIF-1 activity
We next investigated whether the increase in HIF-1α expression levels caused by
SPRY4 siRNA would lead to a parallel change in HIF-1 activity. When co-transfected
with HRE-luciferase reporter construct, SPRY4 siRNA caused an increase in HRE-driven
luciferase activity (OVCAR4: Fig. 5.2A; SKOV3 and OVCAR3: data not shown).
Accordingly, endogenous mRNA expression of two HIF transcriptional targets,
plasminogen activator inhibitor-1 (PAI-1) (293) and prolyl hydroxylase-3 (PHD3) (294),
were increased by SPRY4 siRNA when compared with the control (Fig. 5.2B). To
confirm that these increases were mediated by elevated HIF-1 activity, we reduced HIF-
1α expression with sodium nitroprusside (SNP) ((295) and Fig. 5.2C) and found that SNP
reversed the positive impacts of SPRY4 siRNA on PAI-1 and PHD3 mRNA levels (Fig.
5.2C, D).
109
5.3.3 SPRY4 regulated HIF-1α protein half-life without affecting Hif-1α mRNA
levels
HIF-1α is regulated at the transcriptional, translational and post-translational
levels. We measured HIF-1α mRNA levels in OVCAR4 cells transfected with SPRY4
siRNA for 24 or 48 hrs and found that depleting SPRY4 had no effect on HIF-1α mRNA
levels (Fig. 5.3A). Next, we tested the possibility that SPRY4 siRNA increased HIF-1α
levels by inhibiting its degradation. After treating cells with the proteasomal inhibitor,
MG132, HIF-1α accumulated in the cells, and the change in HIF-1α levels induced by
SPRY4 siRNA was completely blocked (Fig. 5.3B). To show that SPRY4 exerts its effect
via altering HIF-1α protein levels, HIF-1α was overexpressed in OVCAR4 cells, and the
levels were assayed by Western blot. We found that SPRY4 siRNA causes an
accumulation of overexpressed HIF-1α as effectively as the endogenous HIF-1α (Fig.
5.3C). To test the effect of SPRY4 on the half-life of HIF-1α protein, the cells were
treated with protein synthesis inhibitor cycloheximide (CHX). CHX treatment caused a
rapid and significant reduction in HIF-1α levels, but the rate of reduction in SPRY4
siRNA-treated cells was more moderate than that in the control (Fig. 5.3D)
5.3.4 HIF-1α modulation by SPRY4 was independent of PHD activity
We also tested the hypothesis that SPRY4 regulates HIF-1α by modulating PHD-
mediated hydroxylation activity. We monitored the changes in PHD levels in response to
SPRY4. Contrasting its HIF-1α−stabilizing effect, PHD levels were not reduced; instead,
SPRY4 siRNA increased PHD2 and PHD3 levels slightly. HIF-1α siRNA reversed the
effect of SPRY4 siRNA (Fig. 5.4A). PHD2 hydroxylates HIF-1α in normoxia (291), but
110
PHD2 overexpression was unable to reverse SPRY4 siRNA-induced HIF-1α levels (Fig.
5.4B). Two known HIF-1α activators, CoCl2 and DMOG, completely abrogated
hydroxylation of Pro402 and Pro564 (Fig. 5.4C). In contrast, SPRY4 knockdown had no
effect on the levels of hydroxylated HIF-1α protein (Fig. 5.4C). Plasmids for wild-type
HIF-1α (WT HIF-1α) or a double mutant with Pro402Ala and Pro564Ala substitutions,
which is resistant to PHD-mediated degradation (DM HIF-1α), were co-transfected with
SPRY4 siRNA in OVCAR4 cells. Treatment with SPRY4 siRNA increased the levels of
both wild-type and mutant HIF-1α (Fig. 5.4D). Finally, CoCl2 was capable of stabilizing
the fusion of the HIF-1α ODD domain and a luciferase reporter gene, leading to an
increase in luciferase activity (Fig. 5.4E). SPRY4 knockdown failed to act via the HIF-1α
ODD domain-induced increase in luciferase levels (Fig. 5.4E).
5.4 Discussion
Previously, we demonstrated that SPRY4 negatively regulates HIF-1α (So et al.,
unpublished). As the interaction between SPRY4 and a PHD-degrading E3-ligase, SIAH,
was recently reported (287), we tested the hypothesis that SPRY4 regulates HIF-1α by
modulating PHD levels and/or activity. We demonstrated that SPRY4 plays a negative
role in HIF-1α expression and HIF activity (Fig. 5.1, 5.2). Next, we demonstrated that
SPRY4 also functions on overexpressed HIF-1α (Fig. 5.3C), while HIF-1α mRNA levels
were not affected (Fig. 5.3A), indicating that SPRY4 directly acts at the HIF-1α protein
and independent of HIF-1α mRNA. Furthermore, the effect of SPRY4 is suggested to
involve protein degradation as the effect of SPRY4 was less obvious when protein
111
degradation is blocked (Fig. 5.3B). Finally, we demonstrated a direct effect of SPRY4
siRNA on prolonging HIF-1α protein half-life (Fig. 5.3D).
HIF-1α stabilization by SPRY4 knockdown is an additional example of a protein
whose stability is influenced by SPRY. A well-elucidated example of SPRY-regulated
protein degradation is its involvement in abrogating Cbl-mediated EGFR degradation.
Cbl proteins are E3 ubiquitin ligases, which recognise ubiquitinate and target RTKs to
degradation (139-141). Through direct interaction with the Cbl RING finger domain
(142, 143), SPRY2 is capable of sequestering Cbl and thus protects EGFR from
degradation (144). Similarly, SPRY1, 2 and 4 were found to interact with the RING
finger domain of another E3 ligase, SIAH2 (287), thereby suggesting that SPRY may
interfere with the degradation of SIAH targets, including PHDs (169). This observation
has prompted us to test this possibility.
However, our results clearly excluded the involvement of PHD activity in SPRY4-
regulated HIF-1α stability. First, reduced PHD levels and activity are anticipated to
stabilize HIF-1α; but, SPRY4 siRNA increased PHD2 and PHD3 levels (Fig. 5.4A), and
PHD2 overexpression failed to abolish SPRY4 activity (Fig. 5.4B). In parallel, the
hydroxylation status of Pro402 and Pro564 were not altered by SPRY4 siRNA treatment
(Fig. 5.4C). SPRY4 had similar effects on wild-type and mutant HIF-1α with proline
substitutions (Fig. 5.4D). SPRY4 knockdown was incapable of stabilizing the ODD-
fusion protein (Fig. 5.4E). This finding confirms that the ODD domain is not a SPRY4
target. These results suggest that SPRY4 regulates HIF via a yet-to-be determined
mechanism that is independent of the hydroxylation of the ODD domain. Moreover, the
increase in PHD2 and PHD3 levels induced by SPRY4 knockdown provides additional
112
evidence of elevated HIF-1 activity, as both are known HIF targets (169). PHD2 and
PHD3 levels were reduced by HIF-1α siRNA (Fig. 5.4A).
The negative impact on HIF-1α exerted by SPRY4 together with the known SPRY4
induction by hypoxia and hypoxia mimetic (281) suggests reciprocal regulation between
SPRY4 and HIF-1α and constitutes a novel negative feedback loop operating in cancer
cells. Moreover, SPRY2 expression in colon cancer cells was repressed by 1,25(OH)2D3
(an active vitamin D metabolite) via an E-cadherin-dependent pathway, and SPRY2 in
turn repressed 1,25(OH)2D3-induced E-cadherin expression (286). These examples
broaden SPRY activity to signalling pathways other than the RTK/ERK pathway.
Therefore, the factors or dynamics of the microenvironment that induce SPRY would
trigger a counter regulatory loop by SPRY.
Stabilization and nuclear localisation of HIF-1α are common in most cancers,
including ovarian cancer (215, 216). Furthermore, HIF-1α overexpression is significantly
correlated with enhanced microvessel density in ovarian tumors (217). The negative
involvement of SPRY4 on HIF-1α levels demonstrated here may provide explanations
for the accumulation of HIF-1α protein and the antiangiogenic role of SPRY4 reported
previously (274). Our laboratory has previously shown decreased SPRY4 expression in
ovarian cancer cell lines (So et al., unpublished), which possibly contributed to HIF-1α
accumulation and an angiogenic switch in tumors and loss of SPRY4 may play a crucial
role during ovarian cancer progression.
113
A
B
C
Figure 5.1 SPRY4 negatively regulates HIF-1α levels in ovarian cancer cells. OVCAR4,
OVCAR3 and SKOV3 cells were transfected with 50 nM of control siCtrl or siSPRY4
(A) or with 50 nM of the SPRY4 overexpression vector (B). Transfected cells were
cultured for 48 hrs. An equal amount of total protein was loaded in each gel lane, and
HIF-1α was detected using Western blot analysis. C. Prior to HIF-1α detection using
Western blot analysis, OVCAR4 cells were transfected for 48 hrs with siCtrl, siSPRY2 or
siSPRY4.
114
A B
C D
Figure 5.2 SPRY4 negatively regulated HIF-1 activity. A. OVCAR4 cells were co-
transfected with an HRE-luciferase construct and 50 nM siCtrl or siSPRY4. After 48 hrs,
cell lysates were collected for luciferase assay. B. OVCAR4 cells were transfected with
SPRY4 siRNA for 48 hrs. RNA was collected and PAI-1 and PHD3 mRNA levels were
assayed using real-time PCR. C. OVCAR4 cells were transfected with siCtrl or siSPRY4
and incubated with the HIF-1α inhibitor SNP (500 µM) for 48 hrs, and HIF-1α (C) and
PAI-1 and PHD3 mRNA levels were then assayed. (D). Relative mRNA levels are
expressed as a percentage of the control. The data are shown as the mean ± SD of
triplicates in a representative experiment (A) or three independent experiments (B, D).
Mean values that are not denoted by the same letter are significantly different. **, P <
0.01; ***, P < 0.005.
115
A B
C
D
116
Figure 5.3 SPRY4 regulated HIF-1α protein half-life without affecting Hif-1α mRNA
levels. A. OVCAR4 cells were transfected with 50 nM siCtrl or siSPRY4 for 48 hrs.
RNA was collected and HIF-1α mRNA levels were assayed. B. OVCAR4 cells were
transfected with siCtrl or siSPRY4 for 48 hrs. MG132 (5 µM) was added 8 hrs before
harvesting protein for Western blot analysis. C. OVCAR4 cells were co-transfected with
an HIF-1α overexpression construct and siCtrl or siSPRY4. After 48 hrs, cell lysates
were collected for HIF-1α detection analysis. *, P < 0.05. D. OVCAR4 cells were
transfected with SPRY4 siRNA for 48 hrs. CHX (1 µg/ml) was added, and cell lysates
were then collected at various time points for HIF-1α protein level analysis. Relative
levels were expressed as a percentage of the control. The data are shown as the mean ±
SD of three independent experiments. Mean values that are not denoted by the same letter
are significantly different.
117
A B
C
D E
118
Figure 5.4 HIF-1α modulation by SPRY4 acts independently of PHD activity. A.
OVCAR4 cells were transfected with 50 nM siCtrl or siSPRY4 for 48 hrs and assayed for
PHD protein levels using Western blot. B. OVCAR4 cells were co-transfected with a
PHD2 overexpression construct and siSPRY4. After 48 hrs, cell lysates were collected
for detection of HIF-1α levels using Western blot. C. OVCAR4 cells were treated with
CoCl2 (100 µm) or DMOG (1 mM) in the presence of MG132 (5 µM) for 8 hrs (left
panel), or OVCAR4 cells were transfected with siCtrl or siSPRY4 for 48 hrs and treated
with MG132 (5 µM) for 8 hrs (right panel). Protein was collected and assessed for
hydroxylate HIF-1α at Pro402 or Pro564 using Western blot. D. OVCAR4 cells were co-
transfected with an empty vector (Ctrl), an overexpression construct for wild-type HIF-
1α (WT HIF-1α ) or a double-mutant HIF-1α (DM HIF-1α) and siCtrl or siSPRY4.
After 48 hrs, cell lysates were collected for the detection of HIF-1α using Western blot.
The data are shown as the mean ± SD of three independent experiments. *, P < 0.05. E.
OVCAR4 cells were co-transfected with the ODD-luciferase construct and siCtrl or
siSPRY4 for 48 hrs. Cell lysates were collected for the luciferase activity assay. Relative
activity levels are expressed as a percentage of the control. The data are shown as the
mean ± SD of triplicates in a representative experiment. Mean values that are not denoted
by the same letter are significantly different.
119
Chapter 6 Conclusions and future directions
Summary and significance of research findings
Although epithelial ovarian cancer (EOC) comprises the majority of ovarian
carcinomas, its origin and etiology have yet to be completely elucidated. The evidence to
date suggests that malfunctions of receptor tyrosine kinases (RTKs), including the
epidermal growth factor receptor (EGFR), contribute to the development of EOC (52).
Indeed, the EGFR and its ligands play critical roles in cell proliferation and survival as
well as tumor metastasis (53, 242). This study focuses on the intracellular EGFR regulator
Sprouty (SPRY) as well as the EGFR ligand amphiregulin (AREG) in EOC.
Aberrant EGFR activity is common in ovarian cancer and can arise from EGFR
amplification or activating mutations, or overexpression of EGFR or its ligands (65, 66,
75, 77, 88, 89, 91, 92). In addition to these abnormalities, loss of endogenous regulators is
an alternative mechanism that leads to aberrant EGFR activity. SPRY proteins have been
found to be deregulated in most malignancies tested (151-153, 155, 156, 160, 161), and
SPRY-defective cells are hypersensitive to mitogenic and metastatic signals. Furthermore,
SPRY proteins have been reported to regulate various aspects of tumorigenesis (148, 151,
155, 160-162).
When I examined SPRY level abnormalities in ovarian cancer, SPRY2
downregulation and SPRY2 deletion were detected in high-grade serous tumors. This
observation has been confirmed with the TCGA database, which contains data from more
than 500 serous cystadenocarcinomas.
120
Next, I extended my research to investigate the antitumoral functions of SPRY2.
Based on the positive correlation between endogenous SPRY2 and E-cadherin levels in
cell lines and high-grade serous tumors, I hypothesised that SPRY2 may regulate
invasiveness through modulating E-cadherin. My data showed that expression of SPRY2
antagonized the effects of EGF on E-cadherin protein repression and cell invasion (Fig.
6.1). The following studies demonstrated that depleting SPRY4 enhances E-cadherin
downregulation and invasion induced by another EGFR ligand, AREG (Fig. 6.1). These
findings are in agreement with previous reports on the antiinvasive role of SPRY (148,
159-161) and add ovarian cancer, an important and lethal gynaecologic cancer, to the list
of malignancies with SPRY deficiency. Any mechanism that causes the downregulation
of the SPRY level would result in increased stimulation from growth factors, which
favour transformation and the progression of cancer, even in circumstances without ligand
overproduction or receptor dysfunction.
Though there are numerous reports supporting a stimulatory role of EGFR ligands
in ovarian cancer cell invasion and metastasis, most of those studies have focused on
EGF, transforming growth factor-α and heparin binding-EGF (84, 99, 108, 110, 111,
296). Few reports on the effects of AREG have been published. Thus, the effects of
AREG on invasion and the mechanisms of action were investigated. AREG
downregulates E-cadherin and promotes invasion (Fig. 6.1). In addition, AREG induces
SPRY4 expression and activates the negative feedback action of SPRY4 (Fig. 6.1).
Although it is well known that expression of SPRY is induced by RTK/RAS/ERK
and that SPRY feedbacks to antagonize the pathway, many aspects regarding the
functions of SPRY remain unclear. Specifically, it remains to be determined how SPRY
121
regulates ERK activation (123-125, 145). Second, the paradox of both the positive and
negative effects of SPRY2 on EGFR signalling needs to be clarified (121, 297). Third,
the initial paradigm that SPRY is specific for ERK has recently been challenged.
Emerging experimental data show that SPRY is regulated by and modulates more diverse
signalling pathways (133, 134, 298). For example, Wnt signalling induces SPRY4
expression (159), FGF induces SPRY1 and SPRY2 expression through a Ca2+-dependent
pathway (284), and Xenopus SPRY inhibits calcium signalling (285). Clearly, more
studies must be completed to obtain a better understanding of SPRY function.
As SPRY4 is regulated by HIF-1α and SPRY4 regulates angiogenesis (274),
SPRY4 and HIF-1α may also constitute a feedback loop that does not involve the ERK
pathway. This hypothesis is strengthened by the existence of a feedback loop comprising
an active vitamin D metabolite (1,25(OH)2D3) and SPRY2 in colon cancer cells (286).
Together, our data demonstrate the presence and functionality of a novel
EGFR/AKT/HIF-1α and SPRY4 feedback loop in ovarian cancer cells, in which EGF
induces SPRY4 expression through an AKT- and HIF-1α-dependent mechanism (Fig.
6.1). In turn, SPRY4 decreases AKT activation to antagonize EGF-induced expression of
HIF-1α (Fig. 6.1). SPRY2 regulates AKT activity through modulating the expression and
activity of phosphatase and tensin homolog deleted on chromosome 10 (PTEN) (134).
However, the mechanism by which SPRY4 antagonizes AKT and HIF-1α remains
unknown. These findings not only identify a SPRY target outside of the ERK pathways
but also demonstrate the negative effects of SPRY on AKT and HIF-1α. Furthermore, as
AKT and HIF-1α are oncogenic in ovarian cancer (215-217), the negative nature of
SPRY4 on AKT and HIF-1α suggests that SPRY4 is a tumor suppressor.
122
Lastly, I tested whether SPRY4 regulates HIF-1α through the PHD-mediated
HIF-1α degradation pathway. SPRY4 physically interacts with SIAH2 (287), and SIAH
proteins are responsible for PHD degradation and thus prevent HIF-1α hydroxylation by
PHD. As this mechanism causes HIF-1α stabilization and accumulation (169), I
hypothesised that SPRY4 may sequester SIAH and prevent PHD degradation, which
would allow PHDs to hydroxylate HIF-1α and lead to HIF-1α degradation. However, my
data disproved this hypothesis and showed the following: 1) SPRY4 knockdown
stabilizes HIF-1α without reducing PHD levels, 2) PHD2 overexpression fails to reverse
HIF-1α accumulation induced by SPRY4 knockdown, 3) SPRY4 knockdown does not
alter the level of hydroxylated HIF-1α, 4) SPRY4 knockdown stabilizes a HIF-1α mutant
that is refractory to PHD-mediated degradation and 5) SPRY4 knockdown is unable to
stabilize a fusion protein containing the PHD-targeted domain.
Potential applications of the research findings
The current project aims to investigate the expression, functions and mechanisms
of the action of SPRY isoforms in ovarian cancer. These studies enhance our
understanding of the pathogenesis of ovarian cancer. Furthermore, due to their critical
roles in regulating RTK signallings and the downstream processes, SPRY isoforms may
be attractive targets for drug intervention or gene therapy in the treatment of ovarian
cancer.
Ovarian cancer is the most lethal of all the gynaecologic malignancies, with a 5-
year survival rate of 30% – 40% (3), and overall survival has not improved for decades
(4). One of the pitfalls in the current management of ovarian cancer is that although
123
ovarian carcinomas of different subtypes are distinct diseases caused by different
pathways that respond differently to therapies, the current treatment protocols are not
subtype specific (22). Ovarian cancer treatment should be subtype specific or
individualized, which has been proposed for other cancers (299, 300). To achieve
successful subtype-specific treatment, accurate, consistent and reproducible classification
is critical. Pathologists encounter challenges in the diagnosis of the tumors, especially
between high-grade serous carcinoma (HGSC) and endometrioid carcinoma (EC), clear
cell carcinoma (CCC) or low-grade serous carcinoma (LGSC) (22). Therefore, the
discovery of reliable biomarkers for the differential diagnosis is necessary.
The distinct SPRY2 mRNA level and SPRY2 deletion patterns across various
subtypes suggest the potential of SPRY2 as a diagnostic marker for identifying HGSC.
The majority of HGSC samples show reduced SPRY2 mRNA level, and the mean SPRY2
mRNA level of HGSCs is statistically significantly lower than those of ECs and LGSCs.
Furthermore, deletions are almost exclusively found in HGSC, and neither EC, CCC nor
LGSC exhibit loss of the SPRY2 gene. These observations suggest that SPRY2 loss is
important and likely specific to HGSC tumorigenesis. The potential of SPRY2 as a
marker is further supported by the high incidence (more than 1/3 of cases) of SPRY2
mRNA level reduction among HGSCs. These studies need to be performed on a larger
number of ovarian cancer cases or tested in combination with other known HGSC-
specific biomarkers such as WT-1, which is detected in 75% of HGSC (24), to increase
the reliability of diagnosis. In breast cancer patients, the association between low SPRY2
level and higher pathological grade suggests that SPRY2 is a significant independent
prognostic factor (301). SPRY4 mRNA level has been identified as a reliable response
124
marker to Imatinib (c-KIT inhibitor) treatment in patients with gastrointestinal stromal
tumors (GIST) (150).
In addition to potential diagnostic and prognostic applications, SPRY proteins
may be good molecular targets for therapeutic intervention in cancers (286, 292). In
accordance with many other reports, this study demonstrates the altered expression and
tumor suppressing effects of SPRY in ovarian cancer, making SPRY candidate targets for
ovarian cancer treatment. It is important to note that in addition to the antagonistic effects
on the EGFR (activated by EGF and AREG), the regulation of SPRY by epiregulin (binds
EGFR and ERBB4), hepatocyte growth factor and fibroblast growth factor-2 (data not
shown) suggest the functional interaction of SPRY with these RTKs. Furthermore, SPRY
proteins have been reported to antagonize vascular endothelial growth factor (137, 274,
302) and glial cell line-derived neurotrophic factor (303). As most of these RTKs are
implicated in ovarian cancer development (52) and are targets for therapy (59, 304),
SPRY-based therapy may be superior to treatments that target a single RTK.
Downstream of RTKs, SPRY may act by sequestering RAS, RAF and Src (305)
(123-125). The PI3K/AKT pathway has been shown to be regulated by SPRY (133, 134),
and the current study confirms this observation. This study is also the first report on
SPRY regulation of HIF-1α. All of the SPRY target molecules and related pathways
identified thus far are implicated in oncogenic processes. Therefore, it is logical to
hypothesize that SPRY-targeted therapies could have potent antitumoral activity against
many processes simultaneously. SPRY members have been shown to negatively regulate
tumorigenic processes including proliferation (134, 151, 152, 155, 160), migration (148,
160, 161), invasion (148), cell cycle (148), apoptosis (306), angiogenesis (135, 272), in
125
vitro (148) and in vivo (151) tumorigenesis and metastasis (162). Furthermore, although
the roles played by SPRY remain unclear, SPRY2 and SPRY4 have been shown to reflect
breast and GIST cancer patient responses to therapies targeting HER2 and c-KIT,
respectively (150, 301). This finding opens the possibility of enhancing patient sensitivity
to chemotherapy through modulating SPRY.
Future work: regulation of E-cadherin by SPRY
E-cadherin serves to maintain intercellular contact, and loss of E-cadherin in
adhesion junctions is a prerequisite for cancer metastasis. E-cadherin can be silenced in
ovarian cancer through epigenetic silencing (promoter hypermethylation) (230, 254) and
direct transcriptional repression (by SNAIL and SLUG) (235, 255, 256, 296). It is still
unclear how SPRY2 regulates E-cadherin in ovarian cancer cells. Previously, SPRY2 was
shown to regulate colon cancer cell expression of E-cadherin through modulating levels
of ZEB1, a transcriptional repressor of E-cadherin (286). In contrast, I showed that
SPRY4 knockdown decreases E-cadherin protein levels independent of E-cadherin
mRNA level (Fig. 3.7B). Additionally, my preliminary data shows that SPRY2 increases
E-cadherin protein levels but is ineffective in regulating E-cadherin mRNA. These data
suggest that a SPRY mediates E-cadherin through a post-transcriptional mechanism.
Regulation of protein degradation is an important mechanism of post-
transcriptional regulation. It has been demonstrated that SPRY abrogates Cbl-mediated
EGFR degradation. Cbl proteins are E3 ubiquitin ligases that recognise, ubiquitinate and
target EGFR to endocytosis pathways and degradation (139-141). Cbl binds to the well-
conserved Cbl-TKB binding site through its RING finger domain at the N-terminus of
126
SPRY2 (142, 143). Thus, SPRY2 effectively sequesters Cbl and protects the EGFR from
degradation (144). Similar interactions between the RING finger domain of another E3
ligase, SIAH2, and the N-terminus of SPRY2 (287) suggest that this is a common
function for SPRY2. For instance, Hakai, a ubiquitin ligase that structurally resembles
Cbl (Hakai is the Japanese word for 'destruction') has been shown to interact with E-
cadherin and, through ubiquitination, initiate endocytosis and the subsequent degradation
of E-cadherin. Expression of Hakai in MDCK epithelial cells promotes endocytosis of E-
cadherin, disrupts cell-cell contacts and enhances cell motility (307). In ovarian cancer
cells, both Hakai mRNA and protein are expressed (preliminary data). One may speculate
that SPRY interacts with Hakai via the RING finger domain, thereby sequestering Hakai
and blocking E-cadherin degradation. As such, enhanced degradation from loss of SPRY
might represent an additional mechanism whereby ovarian cancer cells lose E-cadherin,
increasing the invasive potential of cancer cells. The interaction between SPRY and
Hakai and their cooperation in E-cadherin degradation certainly warrant further detailed
investigation.
Although SPRY has been shown to interact with E3 ligases (142, 143, 287) and to
regulate protein degradation (144), the roles of SPRY in the regulation of protein
ubiquitination and degradation have not been examined in detail. The proper balance
between ubiquitination and deubiquitination of cellular proteins is crucial for normal cell
cycling and function. Indeed, many oncoproteins and tumor suppressors are involved in
ubiquitination (usually as E3 ligases) or are deubiquitinating enzymes. This fact is best
exemplified by the inactivating germline VHL mutation in renal clear cell carcinomas
(RCCs) (179, 180), which results in the stabilization of HIF-1α (181) and a subsequent
127
excess of angiogenic signals and vascularization (181, 183). In addition, Murine double
minute 2 acts as an E3 ligase of p53 and is overexpressed in many cancers, including
ovarian cancer, which contributes to the reduced p53 expression (25). Notably, both HIF-
1α (our study) and p53 (159) are targets of SPRY4. Perturbations of the proteasome
pathway result in pathogenic malignancies that affect tumor progression, drug resistance,
and altered immune surveillance; thus, the proteasome is an attractive target for cancer
therapies (308). Bortezomib, previously known as PS-341, is a specific inhibitor of the
proteasome pathway. Bortezomib induces apoptosis (309), inhibits angiogenesis (310,
311) and increases survival of the xenograft mice model (310, 312). Bortezomib is also
the first proteasome inhibitor to enter a clinical trial aimed at treating myeloma (311),
which highlights the significance of the ubiquitin-proteasome pathway in cancer. Taken
together, these data suggest the importance of evaluating the roles of SPRY in the
modulation of the protein degradation pathway.
128
Figure 6.1 The diagram summarizes the findings. AREG promotes ovarian cancer cell
invasion by reducing E-cadherin. The loss of SPRY2 and the induction of SPRY4 by
EGFR ligands (EGF and AREG) have been demonstrated. SPRY2 and SPRY4
antagonize the effects of EGF and AREG on the suppression of E-cadherin and invasion
stimulation. Furthermore, the AKT pathway mediates EGF induction of SPRY4. AKT
and HIF-1α are identified as novel targets of SPRY4, suggesting the possibility of
regulation of the downstream oncogenic pathways by SPRY4.
129
References
1. Lalwani N, Prasad SR, Vikram R, Shanbhogue AK, Huettner PC, Fasih N 2011 Histologic, molecular, and cytogenetic features of ovarian cancers: implications for diagnosis and treatment. Radiographics 31:625-646
2. Murdoch WJ 1996 Ovarian surface epithelium, ovulation and carcinogenesis. Biol Rev Camb Philos Soc 71:529-543
3. Auersperg N, Wong AS, Choi KC, Kang SK, Leung PC 2001 Ovarian surface epithelium: biology, endocrinology, and pathology. Endocr Rev 22:255-288
4. Kurman RJ, Shih Ie M 2011 Molecular pathogenesis and extraovarian origin of epithelial ovarian cancer--shifting the paradigm. Hum Pathol 42:918-931
5. Anglesio MS, Carey MS, Kobel M, Mackay H, Huntsman DG 2010 Clear cell carcinoma of the ovary: a report from the first Ovarian Clear Cell Symposium, June 24th, 2010. Gynecol Oncol 121:407-415
6. Fathalla MF 1971 Incessant ovulation--a factor in ovarian neoplasia? Lancet 2:163
7. Casagrande JT, Louie EW, Pike MC, Roy S, Ross RK, Henderson BE 1979 "Incessant ovulation" and ovarian cancer. Lancet 2:170-173
8. Cramer DW, Welch WR 1983 Determinants of ovarian cancer risk. II. Inferences regarding pathogenesis. J Natl Cancer Inst 71:717-721
9. Chakravarti S, Collins WP, Forecast JD, Newton JR, Oram DH, Studd JW 1976 Hormonal profiles after the menopause. Br Med J 2:784-787
10. Scaglia H, Medina M, Pinto-Ferreira AL, Vazques G, Gual C, Perez-Palacios G 1976 Pituitary LH and FSH secretion and responsiveness in women of old age. Acta Endocrinol (Copenh) 81:673-679
11. Sell A, Bertelsen K, Andersen JE, Stroyer I, Panduro J 1990 Randomized study of whole-abdomen irradiation versus pelvic irradiation plus cyclophosphamide in treatment of early ovarian cancer. Gynecol Oncol 37:367-373
12. Schildkraut JM, Schwingl PJ, Bastos E, Evanoff A, Hughes C 1996 Epithelial ovarian cancer risk among women with polycystic ovary syndrome. Obstet Gynecol 88:554-559
13. Whittemore AS, Harris R, Itnyre J 1992 Characteristics relating to ovarian cancer risk: collaborative analysis of 12 US case-control studies. IV. The pathogenesis of epithelial ovarian cancer. Collaborative Ovarian Cancer Group. Am J Epidemiol 136:1212-1220
14. La Vecchia C, Franceschi S 1999 Oral contraceptives and ovarian cancer. Eur J Cancer Prev 8:297-304
15. Shoham Z 1994 Epidemiology, etiology, and fertility drugs in ovarian epithelial carcinoma: where are we today? Fertil Steril 62:433-448
16. Rao BR, Slotman BJ 1991 Endocrine factors in common epithelial ovarian cancer. Endocr Rev 12:14-26
17. Tung CS, Mok SC, Tsang YT, Zu Z, Song H, Liu J, Deavers MT, Malpica A, Wolf JK, Lu KH, Gershenson DM, Wong KK 2009 PAX2 expression in low malignant potential ovarian tumors and low-grade ovarian serous carcinomas. Mod Pathol 22:1243-1250
130
18. Kurman RJ, Shih Ie M 2010 The origin and pathogenesis of epithelial ovarian cancer: a proposed unifying theory. Am J Surg Pathol 34:433-443
19. Bell DA 2005 Origins and molecular pathology of ovarian cancer. Mod Pathol 18 Suppl 2:S19-32
20. Piek JM, van Diest PJ, Zweemer RP, Kenemans P, Verheijen RH 2001 Tubal ligation and risk of ovarian cancer. Lancet 358:844
21. Wiegand KC, Shah SP, Al-Agha OM, Zhao Y, Tse K, Zeng T, Senz J, McConechy MK, Anglesio MS, Kalloger SE, Yang W, Heravi-Moussavi A, Giuliany R, Chow C, Fee J, Zayed A, Prentice L, Melnyk N, Turashvili G, Delaney AD, Madore J, Yip S, McPherson AW, Ha G, Bell L, Fereday S, Tam A, Galletta L, Tonin PN, Provencher D, Miller D, Jones SJ, Moore RA, Morin GB, Oloumi A, Boyd N, Aparicio SA, Shih Ie M, Mes-Masson AM, Bowtell DD, Hirst M, Gilks B, Marra MA, Huntsman DG 2010 ARID1A mutations in endometriosis-associated ovarian carcinomas. N Engl J Med 363:1532-1543
22. Gilks CB, Prat J 2009 Ovarian carcinoma pathology and genetics: recent advances. Hum Pathol 40:1213-1223
23. Yap TA, Carden CP, Kaye SB 2009 Beyond chemotherapy: targeted therapies in ovarian cancer. Nat Rev Cancer 9:167-181
24. Kobel M, Kalloger SE, Boyd N, McKinney S, Mehl E, Palmer C, Leung S, Bowen NJ, Ionescu DN, Rajput A, Prentice LM, Miller D, Santos J, Swenerton K, Gilks CB, Huntsman D 2008 Ovarian carcinoma subtypes are different diseases: implications for biomarker studies. PLoS Med 5:e232
25. Ahmed AA, Etemadmoghadam D, Temple J, Lynch AG, Riad M, Sharma R, Stewart C, Fereday S, Caldas C, Defazio A, Bowtell D, Brenton JD 2010 Driver mutations in TP53 are ubiquitous in high grade serous carcinoma of the ovary. J Pathol 221:49-56
26. Werness BA, Parvatiyar P, Ramus SJ, Whittemore AS, Garlinghouse-Jones K, Oakley-Girvan I, DiCioccio RA, Wiest J, Tsukada Y, Ponder BA, Piver MS 2000 Ovarian carcinoma in situ with germline BRCA1 mutation and loss of heterozygosity at BRCA1 and TP53. J Natl Cancer Inst 92:1088-1091
27. Kobel M, Reuss A, Bois A, Kommoss S, Kommoss F, Gao D, Kalloger SE, Huntsman DG, Gilks CB 2010 The biological and clinical value of p53 expression in pelvic high-grade serous carcinomas. J Pathol 222:191-198
28. Senturk E, Cohen S, Dottino PR, Martignetti JA 2010 A critical re-appraisal of BRCA1 methylation studies in ovarian cancer. Gynecol Oncol 119:376-383
29. Callahan MJ, Crum CP, Medeiros F, Kindelberger DW, Elvin JA, Garber JE, Feltmate CM, Berkowitz RS, Muto MG 2007 Primary fallopian tube malignancies in BRCA-positive women undergoing surgery for ovarian cancer risk reduction. J Clin Oncol 25:3985-3990
30. Khalique L, Ayhan A, Weale ME, Jacobs IJ, Ramus SJ, Gayther SA 2007 Genetic intra-tumour heterogeneity in epithelial ovarian cancer and its implications for molecular diagnosis of tumours. J Pathol 211:286-295
31. Piek JM, van Diest PJ, Zweemer RP, Jansen JW, Poort-Keesom RJ, Menko FH, Gille JJ, Jongsma AP, Pals G, Kenemans P, Verheijen RH 2001 Dysplastic changes in prophylactically removed Fallopian tubes of women predisposed to developing ovarian cancer. J Pathol 195:451-456
131
32. Kindelberger DW, Lee Y, Miron A, Hirsch MS, Feltmate C, Medeiros F, Callahan MJ, Garner EO, Gordon RW, Birch C, Berkowitz RS, Muto MG, Crum CP 2007 Intraepithelial carcinoma of the fimbria and pelvic serous carcinoma: Evidence for a causal relationship. Am J Surg Pathol 31:161-169
33. Przybycin CG, Kurman RJ, Ronnett BM, Shih Ie M, Vang R 2010 Are all pelvic (nonuterine) serous carcinomas of tubal origin? Am J Surg Pathol 34:1407-1416
34. Lee Y, Miron A, Drapkin R, Nucci MR, Medeiros F, Saleemuddin A, Garber J, Birch C, Mou H, Gordon RW, Cramer DW, McKeon FD, Crum CP 2007 A candidate precursor to serous carcinoma that originates in the distal fallopian tube. J Pathol 211:26-35
35. Marquez RT, Baggerly KA, Patterson AP, Liu J, Broaddus R, Frumovitz M, Atkinson EN, Smith DI, Hartmann L, Fishman D, Berchuck A, Whitaker R, Gershenson DM, Mills GB, Bast RC, Jr., Lu KH 2005 Patterns of gene expression in different histotypes of epithelial ovarian cancer correlate with those in normal fallopian tube, endometrium, and colon. Clin Cancer Res 11:6116-6126
36. Dehari R, Kurman RJ, Logani S, Shih Ie M 2007 The development of high-grade serous carcinoma from atypical proliferative (borderline) serous tumors and low-grade micropapillary serous carcinoma: a morphologic and molecular genetic analysis. Am J Surg Pathol 31:1007-1012
37. Singer G, Oldt R, 3rd, Cohen Y, Wang BG, Sidransky D, Kurman RJ, Shih Ie M 2003 Mutations in BRAF and KRAS characterize the development of low-grade ovarian serous carcinoma. J Natl Cancer Inst 95:484-486
38. Anglesio MS, Arnold JM, George J, Tinker AV, Tothill R, Waddell N, Simms L, Locandro B, Fereday S, Traficante N, Russell P, Sharma R, Birrer MJ, deFazio A, Chenevix-Trench G, Bowtell DD 2008 Mutation of ERBB2 provides a novel alternative mechanism for the ubiquitous activation of RAS-MAPK in ovarian serous low malignant potential tumors. Mol Cancer Res 6:1678-1690
39. Hsu CY, Bristow R, Cha MS, Wang BG, Ho CL, Kurman RJ, Wang TL, Shih Ie M 2004 Characterization of active mitogen-activated protein kinase in ovarian serous carcinomas. Clin Cancer Res 10:6432-6436
40. Oniciu DC, Dasseux JL, Yang J, Mueller R, Pop E, Denysenko A, Duan C, Huang TB, Zhang L, Krause BR, Drake SL, Lalwani N, Cramer CT, Goetz B, Pape ME, McKee A, Fici GJ, Lutostanski JM, Brown SC, Bisgaier CL 2006 Influence of various central moieties on the hypolipidemic properties of long hydrocarbon chain diols and diacids. J Med Chem 49:334-348
41. Kobayashi H 2009 Ovarian cancer in endometriosis: epidemiology, natural history, and clinical diagnosis. Int J Clin Oncol 14:378-382
42. Campbell IG, Russell SE, Choong DY, Montgomery KG, Ciavarella ML, Hooi CS, Cristiano BE, Pearson RB, Phillips WA 2004 Mutation of the PIK3CA gene in ovarian and breast cancer. Cancer Res 64:7678-7681
43. Polakis P 2007 The many ways of Wnt in cancer. Curr Opin Genet Dev 17:45-51 44. Jiang X, Morland SJ, Hitchcock A, Thomas EJ, Campbell IG 1998 Allelotyping
of endometriosis with adjacent ovarian carcinoma reveals evidence of a common lineage. Cancer Res 58:1707-1712
45. Sato N, Tsunoda H, Nishida M, Morishita Y, Takimoto Y, Kubo T, Noguchi M 2000 Loss of heterozygosity on 10q23.3 and mutation of the tumor suppressor
132
gene PTEN in benign endometrial cyst of the ovary: possible sequence progression from benign endometrial cyst to endometrioid carcinoma and clear cell carcinoma of the ovary. Cancer Res 60:7052-7056
46. McMeekin DS, Burger RA, Manetta A, DiSaia P, Berman ML 1995 Endometrioid adenocarcinoma of the ovary and its relationship to endometriosis. Gynecol Oncol 59:81-86
47. Kuo KT, Mao TL, Jones S, Veras E, Ayhan A, Wang TL, Glas R, Slamon D, Velculescu VE, Kuman RJ, Shih Ie M 2009 Frequent activating mutations of PIK3CA in ovarian clear cell carcinoma. Am J Pathol 174:1597-1601
48. Risch HA, McLaughlin JR, Cole DE, Rosen B, Bradley L, Fan I, Tang J, Li S, Zhang S, Shaw PA, Narod SA 2006 Population BRCA1 and BRCA2 mutation frequencies and cancer penetrances: a kin-cohort study in Ontario, Canada. J Natl Cancer Inst 98:1694-1706
49. Hashiguchi Y, Tsuda H, Inoue T, Berkowitz RS, Mok SC 2006 PTEN expression in clear cell adenocarcinoma of the ovary. Gynecol Oncol 101:71-75
50. Miyazawa M, Yasuda M, Fujita M, Kajiwara H, Hirabayashi K, Takekoshi S, Hirasawa T, Murakami M, Ogane N, Kiguchi K, Ishiwata I, Mikami M, Osamura RY 2009 Therapeutic strategy targeting the mTOR-HIF-1alpha-VEGF pathway in ovarian clear cell adenocarcinoma. Pathol Int 59:19-27
51. Jones S, Wang TL, Shih Ie M, Mao TL, Nakayama K, Roden R, Glas R, Slamon D, Diaz LA, Jr., Vogelstein B, Kinzler KW, Velculescu VE, Papadopoulos N 2010 Frequent mutations of chromatin remodeling gene ARID1A in ovarian clear cell carcinoma. Science 330:228-231
52. Wong AS, Leung PC 2007 Role of endocrine and growth factors on the ovarian surface epithelium. J Obstet Gynaecol Res 33:3-16
53. Lafky JM, Wilken JA, Baron AT, Maihle NJ 2008 Clinical implications of the ErbB/epidermal growth factor (EGF) receptor family and its ligands in ovarian cancer. Biochim Biophys Acta 1785:232-265
54. Lin CR, Chen WS, Lazar CS, Carpenter CD, Gill GN, Evans RM, Rosenfeld MG 1986 Protein kinase C phosphorylation at Thr 654 of the unoccupied EGF receptor and EGF binding regulate functional receptor loss by independent mechanisms. Cell 44:839-848
55. Segatto O, Lonardo F, Wexler D, Fazioli F, Pierce JH, Bottaro DP, White MF, Di Fiore PP 1991 The juxtamembrane regions of the epidermal growth factor receptor and gp185erbB-2 determine the specificity of signal transduction. Mol Cell Biol 11:3191-3202
56. Guy PM, Platko JV, Cantley LC, Cerione RA, Carraway KL, 3rd 1994 Insect cell-expressed p180erbB3 possesses an impaired tyrosine kinase activity. Proc Natl Acad Sci U S A 91:8132-8136
57. Yarden Y, Sliwkowski MX 2001 Untangling the ErbB signalling network. Nat Rev Mol Cell Biol 2:127-137
58. Tzahar E, Waterman H, Chen X, Levkowitz G, Karunagaran D, Lavi S, Ratzkin BJ, Yarden Y 1996 A hierarchical network of interreceptor interactions determines signal transduction by Neu differentiation factor/neuregulin and epidermal growth factor. Mol Cell Biol 16:5276-5287
133
59. Sheng Q, Liu J 2011 The therapeutic potential of targeting the EGFR family in epithelial ovarian cancer. Br J Cancer 104:1241-1245
60. Soltoff SP, Cantley LC 1996 p120cbl is a cytosolic adapter protein that associates with phosphoinositide 3-kinase in response to epidermal growth factor in PC12 and other cells. J Biol Chem 271:563-567
61. Altomare DA, Wang HQ, Skele KL, De Rienzo A, Klein-Szanto AJ, Godwin AK, Testa JR 2004 AKT and mTOR phosphorylation is frequently detected in ovarian cancer and can be targeted to disrupt ovarian tumor cell growth. Oncogene 23:5853-5857
62. de Graeff P, Crijns AP, Ten Hoor KA, Klip HG, Hollema H, Oien K, Bartlett JM, Wisman GB, de Bock GH, de Vries EG, de Jong S, van der Zee AG 2008 The ErbB signalling pathway: protein expression and prognostic value in epithelial ovarian cancer. Br J Cancer 99:341-349
63. Lee S, Choi EJ, Jin C, Kim DH 2005 Activation of PI3K/Akt pathway by PTEN reduction and PIK3CA mRNA amplification contributes to cisplatin resistance in an ovarian cancer cell line. Gynecol Oncol 97:26-34
64. Qiu L, Di W, Jiang Q, Scheffler E, Derby S, Yang J, Kouttab N, Wanebo H, Yan B, Wan Y 2005 Targeted inhibition of transient activation of the EGFR-mediated cell survival pathway enhances paclitaxel-induced ovarian cancer cell death. Int J Oncol 27:1441-1448
65. Wikstrand CJ, Reist CJ, Archer GE, Zalutsky MR, Bigner DD 1998 The class III variant of the epidermal growth factor receptor (EGFRvIII): characterization and utilization as an immunotherapeutic target. J Neurovirol 4:148-158
66. Lassus H, Sihto H, Leminen A, Joensuu H, Isola J, Nupponen NN, Butzow R 2006 Gene amplification, mutation, and protein expression of EGFR and mutations of ERBB2 in serous ovarian carcinoma. J Mol Med 84:671-681
67. Skirnisdottir I, Seidal T, Karlsson MG, Sorbe B 2005 Clinical and biological characteristics of clear cell carcinomas of the ovary in FIGO stages I-II. Int J Oncol 26:177-183
68. Leng J, Lang J, Shen K, Guo L 1997 Overexpression of p53, EGFR, c-erbB2 and c-erbB3 in endometrioid carcinoma of the ovary. Chin Med Sci J 12:67-70
69. Scambia G, Benedetti Panici P, Battaglia F, Ferrandina G, Baiocchi G, Greggi S, De Vincenzo R, Mancuso S 1992 Significance of epidermal growth factor receptor in advanced ovarian cancer. J Clin Oncol 10:529-535
70. Harlozinska A, Bar JK, Sobanska E, Goluda M 1998 Epidermal growth factor receptor and c-erbB-2 oncoproteins in tissue and tumor effusion cells of histopathologically different ovarian neoplasms. Tumour Biol 19:364-373
71. Psyrri A, Kassar M, Yu Z, Bamias A, Weinberger PM, Markakis S, Kowalski D, Camp RL, Rimm DL, Dimopoulos MA 2005 Effect of epidermal growth factor receptor expression level on survival in patients with epithelial ovarian cancer. Clin Cancer Res 11:8637-8643
72. Khalifa MA, Lacher DA, Lage JM, Mannel RS, Walker JL, Angros LH, Min KW 1997 Immunohistochemical assessment of proliferation markers and altered gene expression in archival specimens of ovarian epithelial tumors. Cancer Detect Prev 21:532-539
134
73. Goff BA, Shy K, Greer BE, Muntz HG, Skelly M, Gown AM 1996 Overexpression and relationships of HER-2/neu, epidermal growth factor receptor, p53, Ki-67, and tumor necrosis factor alpha in epithelial ovarian cancer. Eur J Gynaecol Oncol 17:487-492
74. Schilder RJ, Sill MW, Chen X, Darcy KM, Decesare SL, Lewandowski G, Lee RB, Arciero CA, Wu H, Godwin AK 2005 Phase II study of gefitinib in patients with relapsed or persistent ovarian or primary peritoneal carcinoma and evaluation of epidermal growth factor receptor mutations and immunohistochemical expression: a Gynecologic Oncology Group Study. Clin Cancer Res 11:5539-5548
75. Kuan CT, Wikstrand CJ, Bigner DD 2001 EGF mutant receptor vIII as a molecular target in cancer therapy. Endocr Relat Cancer 8:83-96
76. Pedersen MW, Meltorn M, Damstrup L, Poulsen HS 2001 The type III epidermal growth factor receptor mutation. Biological significance and potential target for anti-cancer therapy. Ann Oncol 12:745-760
77. Moscatello DK, Holgado-Madruga M, Godwin AK, Ramirez G, Gunn G, Zoltick PW, Biegel JA, Hayes RL, Wong AJ 1995 Frequent expression of a mutant epidermal growth factor receptor in multiple human tumors. Cancer Res 55:5536-5539
78. Steffensen KD, Waldstrom M, Olsen DA, Corydon T, Lorentzen KA, Knudsen HJ, Jeppesen U, Brandslund I, Jakobsen A 2008 Mutant epidermal growth factor receptor in benign, borderline, and malignant ovarian tumors. Clin Cancer Res 14:3278-3282
79. Zeineldin R, Rosenberg M, Ortega D, Buhr C, Chavez MG, Stack MS, Kusewitt DF, Hudson LG 2006 Mesenchymal transformation in epithelial ovarian tumor cells expressing epidermal growth factor receptor variant III. Mol Carcinog 45:851-860
80. Tanaka Y, Terai Y, Tanabe A, Sasaki H, Sekijima T, Fujiwara S, Yamashita Y, Kanemura M, Ueda M, Sugita M, Franklin WA, Ohmichi M 2011 Prognostic effect of epidermal growth factor receptor gene mutations and the aberrant phosphorylation of Akt and ERK in ovarian cancer. Cancer Biol Ther 11:50-57
81. Paez JG, Janne PA, Lee JC, Tracy S, Greulich H, Gabriel S, Herman P, Kaye FJ, Lindeman N, Boggon TJ, Naoki K, Sasaki H, Fujii Y, Eck MJ, Sellers WR, Johnson BE, Meyerson M 2004 EGFR mutations in lung cancer: correlation with clinical response to gefitinib therapy. Science 304:1497-1500
82. Lacroix L, Pautier P, Duvillard P, Motte N, Saulnier P, Bidart JM, Soria JC 2006 Response of ovarian carcinomas to gefitinib-carboplatin-paclitaxel combination is not associated with EGFR kinase domain somatic mutations. Int J Cancer 118:1068-1069
83. Posadas EM, Liel MS, Kwitkowski V, Minasian L, Godwin AK, Hussain MM, Espina V, Wood BJ, Steinberg SM, Kohn EC 2007 A phase II and pharmacodynamic study of gefitinib in patients with refractory or recurrent epithelial ovarian cancer. Cancer 109:1323-1330
84. Cowden Dahl KD, Symowicz J, Ning Y, Gutierrez E, Fishman DA, Adley BP, Stack MS, Hudson LG 2008 Matrix metalloproteinase 9 is a mediator of
135
epidermal growth factor-dependent e-cadherin loss in ovarian carcinoma cells. Cancer Res 68:4606-4613
85. Harris RC, Chung E, Coffey RJ 2003 EGF receptor ligands. Exp Cell Res 284:2-13
86. Stromberg K, Johnson GR, O'Connor DM, Sorensen CM, Gullick WJ, Kannan B 1994 Frequent immunohistochemical detection of EGF supergene family members in ovarian carcinogenesis. Int J Gynecol Pathol 13:342-347
87. Niikura H, Sasano H, Sato S, Yajima A 1997 Expression of epidermal growth factor-related proteins and epidermal growth factor receptor in common epithelial ovarian tumors. Int J Gynecol Pathol 16:60-68
88. Owens OJ, Leake RE 1992 Growth factor content in normal and benign ovarian tumours. Eur J Obstet Gynecol Reprod Biol 47:223-228
89. Tanaka Y, Miyamoto S, Suzuki SO, Oki E, Yagi H, Sonoda K, Yamazaki A, Mizushima H, Maehara Y, Mekada E, Nakano H 2005 Clinical significance of heparin-binding epidermal growth factor-like growth factor and a disintegrin and metalloprotease 17 expression in human ovarian cancer. Clin Cancer Res 11:4783-4792
90. Miyamoto S, Hirata M, Yamazaki A, Kageyama T, Hasuwa H, Mizushima H, Tanaka Y, Yagi H, Sonoda K, Kai M, Kanoh H, Nakano H, Mekada E 2004 Heparin-binding EGF-like growth factor is a promising target for ovarian cancer therapy. Cancer Res 64:5720-5727
91. Yotsumoto F, Yagi H, Suzuki SO, Oki E, Tsujioka H, Hachisuga T, Sonoda K, Kawarabayashi T, Mekada E, Miyamoto S 2008 Validation of HB-EGF and amphiregulin as targets for human cancer therapy. Biochem Biophys Res Commun 365:555-561
92. Yagi H, Miyamoto S, Tanaka Y, Sonoda K, Kobayashi H, Kishikawa T, Iwamoto R, Mekada E, Nakano H 2005 Clinical significance of heparin-binding epidermal growth factor-like growth factor in peritoneal fluid of ovarian cancer. Br J Cancer 92:1737-1745
93. Lin CI, Chen CN, Huang MT, Lee SJ, Lin CH, Chang CC, Lee H 2008 Lysophosphatidic acid upregulates vascular endothelial growth factor-C and tube formation in human endothelial cells through LPA(1/3), COX-2, and NF-kappaB activation- and EGFR transactivation-dependent mechanisms. Cell Signal 20:1804-1814
94. Keely SJ, Uribe JM, Barrett KE 1998 Carbachol stimulates transactivation of epidermal growth factor receptor and mitogen-activated protein kinase in T84 cells. Implications for carbachol-stimulated chloride secretion. J Biol Chem 273:27111-27117
95. Ushio-Fukai M, Griendling KK, Becker PL, Hilenski L, Halleran S, Alexander RW 2001 Epidermal growth factor receptor transactivation by angiotensin II requires reactive oxygen species in vascular smooth muscle cells. Arterioscler Thromb Vasc Biol 21:489-495
96. Mori K, Kitayama J, Shida D, Yamashita H, Watanabe T, Nagawa H 2006 Lysophosphatidic acid-induced effects in human colon carcinoma DLD1 cells are partially dependent on transactivation of epidermal growth factor receptor. J Surg Res 132:56-61
136
97. Bolitho C, Hahn MA, Baxter RC, Marsh DJ 2010 The chemokine CXCL1 induces proliferation in epithelial ovarian cancer cells by transactivation of the epidermal growth factor receptor. Endocr Relat Cancer 17:929-940
98. Abdollahi A, Gruver BN, Patriotis C, Hamilton TC 2003 Identification of epidermal growth factor-responsive genes in normal rat ovarian surface epithelial cells. Biochem Biophys Res Commun 307:188-197
99. Chen Z, Fadiel A, Feng Y, Ohtani K, Rutherford T, Naftolin F 2001 Ovarian epithelial carcinoma tyrosine phosphorylation, cell proliferation, and ezrin translocation are stimulated by interleukin 1alpha and epidermal growth factor. Cancer 92:3068-3075
100. Siemens CH, Auersperg N 1988 Serial propagation of human ovarian surface epithelium in tissue culture. J Cell Physiol 134:347-356
101. Stromberg K, Collins TJt, Gordon AW, Jackson CL, Johnson GR 1992 Transforming growth factor-alpha acts as an autocrine growth factor in ovarian carcinoma cell lines. Cancer Res 52:341-347
102. Casamassimi A, De Luca A, Agrawal S, Stromberg K, Salomon DS, Normanno N 2000 EGF-related antisense oligonucleotides inhibit the proliferation of human ovarian carcinoma cells. Ann Oncol 11:319-325
103. Morishige K, Kurachi H, Amemiya K, Fujita Y, Yamamoto T, Miyake A, Tanizawa O 1991 Evidence for the involvement of transforming growth factor alpha and epidermal growth factor receptor autocrine growth mechanism in primary human ovarian cancers in vitro. Cancer Res 51:5322-5328
104. Kurachi H, Morishige K, Amemiya K, Adachi H, Hirota K, Miyake A, Tanizawa O 1991 Importance of transforming growth factor alpha/epidermal growth factor receptor autocrine growth mechanism in an ovarian cancer cell line in vivo. Cancer Res 51:5956-5959
105. Johnson GR, Saeki T, Auersperg N, Gordon AW, Shoyab M, Salomon DS, Stromberg K 1991 Response to and expression of amphiregulin by ovarian carcinoma and normal ovarian surface epithelial cells: nuclear localization of endogenous amphiregulin. Biochem Biophys Res Commun 180:481-488
106. Alper O, De Santis ML, Stromberg K, Hacker NF, Cho-Chung YS, Salomon DS 2000 Anti-sense suppression of epidermal growth factor receptor expression alters cellular proliferation, cell-adhesion and tumorigenicity in ovarian cancer cells. Int J Cancer 88:566-574
107. Ahmed N, Maines-Bandiera S, Quinn MA, Unger WG, Dedhar S, Auersperg N 2006 Molecular pathways regulating EGF-induced epithelio-mesenchymal transition in human ovarian surface epithelium. Am J Physiol Cell Physiol 290:C1532-1542
108. Colomiere M, Findlay J, Ackland L, Ahmed N 2009 Epidermal growth factor-induced ovarian carcinoma cell migration is associated with JAK2/STAT3 signals and changes in the abundance and localization of alpha6beta1 integrin. Int J Biochem Cell Biol 41:1034-1045
109. Lim R, Ahmed N, Borregaard N, Riley C, Wafai R, Thompson EW, Quinn MA, Rice GE 2007 Neutrophil gelatinase-associated lipocalin (NGAL) an early-screening biomarker for ovarian cancer: NGAL is associated with epidermal
137
growth factor-induced epithelio-mesenchymal transition. Int J Cancer 120:2426-2434
110. Henic E, Noskova V, Hoyer-Hansen G, Hansson S, Casslen B 2009 Estradiol attenuates EGF-induced rapid uPAR mobilization and cell migration via the G-protein-coupled receptor 30 in ovarian cancer cells. Int J Gynecol Cancer 19:214-222
111. Ellerbroek SM, Halbleib JM, Benavidez M, Warmka JK, Wattenberg EV, Stack MS, Hudson LG 2001 Phosphatidylinositol 3-kinase activity in epidermal growth factor-stimulated matrix metalloproteinase-9 production and cell surface association. Cancer Res 61:1855-1861
112. Alper O, Bergmann-Leitner ES, Bennett TA, Hacker NF, Stromberg K, Stetler-Stevenson WG 2001 Epidermal growth factor receptor signaling and the invasive phenotype of ovarian carcinoma cells. J Natl Cancer Inst 93:1375-1384
113. Chan JK, Pham H, You XJ, Cloven NG, Burger RA, Rose GS, Van Nostrand K, Korc M, Disaia PJ, Fan H 2005 Suppression of ovarian cancer cell tumorigenicity and evasion of Cisplatin resistance using a truncated epidermal growth factor receptor in a rat model. Cancer Res 65:3243-3248
114. Christen RD, Hom DK, Porter DC, Andrews PA, MacLeod CL, Hafstrom L, Howell SB 1990 Epidermal growth factor regulates the in vitro sensitivity of human ovarian carcinoma cells to cisplatin. J Clin Invest 86:1632-1640
115. Blank SV, Christos P, Curtin JP, Goldman N, Runowicz CD, Sparano JA, Liebes L, Chen HX, Muggia FM 2010 Erlotinib added to carboplatin and paclitaxel as first-line treatment of ovarian cancer: a phase II study based on surgical reassessment. Gynecol Oncol 119:451-456
116. Siwak DR, Carey M, Hennessy BT, Nguyen CT, McGahren Murray MJ, Nolden L, Mills GB 2010 Targeting the epidermal growth factor receptor in epithelial ovarian cancer: current knowledge and future challenges. J Oncol 2010:568938
117. Kumar A, Petri ET, Halmos B, Boggon TJ 2008 Structure and clinical relevance of the epidermal growth factor receptor in human cancer. J Clin Oncol 26:1742-1751
118. Schilder RJ, Pathak HB, Lokshin AE, Holloway RW, Alvarez RD, Aghajanian C, Min H, Devarajan K, Ross E, Drescher CW, Godwin AK 2009 Phase II trial of single agent cetuximab in patients with persistent or recurrent epithelial ovarian or primary peritoneal carcinoma with the potential for dose escalation to rash. Gynecol Oncol 113:21-27
119. Seiden MV, Burris HA, Matulonis U, Hall JB, Armstrong DK, Speyer J, Weber JD, Muggia F 2007 A phase II trial of EMD72000 (matuzumab), a humanized anti-EGFR monoclonal antibody, in patients with platinum-resistant ovarian and primary peritoneal malignancies. Gynecol Oncol 104:727-731
120. Hacohen N, Kramer S, Sutherland D, Hiromi Y, Krasnow MA 1998 Expression of Sprouty genes 1, 2 and 4 during mouse organogenesis. Cell 92:253-263
121. Casci T, Vinos J, Freeman M 1999 Sprouty, an intracellular inhibitor of Ras signaling. Cell 96:655-665
122. Minowada G, Jarvis LA, Chi CL, Neubuser A, Sun X, Hacohen N, Krasnow MA, Martin GR 1999 Vertebrate Sprouty genes are induced by FGF signaling and can cause chondrodysplasia when overexpressed. Development 126:4465-4475
138
123. Reich A, Sapir A, Shilo B 1999 Sprouty is a general inhibitor of receptor tyrosine kinase signaling. Development 126:4139-4147
124. Gross I, Bassit B, Benezra M, Licht JD 2001 Mammalian sprouty proteins inhibit cell growth and differentiation by preventing ras activation. J Biol Chem 276:46460-46468
125. Yusoff P, Lao DH, Ong SH, Wong ES, Lim J, Lo TL, Leong HF, Fong CW, Guy GR 2002 Sprouty2 inhibits the Ras/MAP kinase pathway by inhibiting the activation of Raf. J Biol Chem 277:3195-3201
126. Mason JM, Morrison DJ, Bassit B, Dimri M, Band H, Licht JD, Gross I 2004 Tyrosine phosphorylation of Sprouty proteins regulates their ability to inhibit growth factor signaling: a dual feedback loop. Mol Biol Cell 15:2176-2188
127. Mason JM, Morrison DJ, Basson MA, Licht JD 2006 Sprouty proteins: multifaceted negative-feedback regulators of receptor tyrosine kinase signaling. Trends Cell Biol 16:45-54
128. Yigzaw Y, Cartin L, Pierre S, Scholich K, Patel TB 2001 The C terminus of sprouty is important for modulation of cellular migration and proliferation. J Biol Chem 276:22742-22747
129. Lim J, Wong ES, Ong SH, Yusoff P, Low BC, Guy GR 2000 Sprouty proteins are targeted to membrane ruffles upon growth factor receptor tyrosine kinase activation. Identification of a novel translocation domain. J Biol Chem 275:32837-32845
130. Hanafusa H, Torii S, Yasunaga T, Nishida E 2002 Sprouty1 and Sprouty2 provide a control mechanism for the Ras/MAPK signalling pathway. Nat Cell Biol 4:850-858
131. Hall AB, Jura N, DaSilva J, Jang YJ, Gong D, Bar-Sagi D 2003 hSpry2 is targeted to the ubiquitin-dependent proteasome pathway by c-Cbl. Curr Biol 13:308-314
132. Tsavachidou D, Coleman ML, Athanasiadis G, Li S, Licht JD, Olson MF, Weber BL 2004 SPRY2 is an inhibitor of the ras/extracellular signal-regulated kinase pathway in melanocytes and melanoma cells with wild-type BRAF but not with the V599E mutant. Cancer Res 64:5556-5559
133. de Alvaro C, Martinez N, Rojas JM, Lorenzo M 2005 Sprouty-2 overexpression in C2C12 cells confers myogenic differentiation properties in the presence of FGF2. Mol Biol Cell 16:4454-4461
134. Edwin F, Singh R, Endersby R, Baker SJ, Patel TB 2006 The tumor suppressor PTEN is necessary for human Sprouty 2-mediated inhibition of cell proliferation. J Biol Chem 281:4816-4822
135. Lee SH, Schloss DJ, Jarvis L, Krasnow MA, Swain JL 2001 Inhibition of angiogenesis by a mouse sprouty protein. J Biol Chem 276:4128-4133
136. Leeksma OC, Van Achterberg TA, Tsumura Y, Toshima J, Eldering E, Kroes WG, Mellink C, Spaargaren M, Mizuno K, Pannekoek H, de Vries CJ 2002 Human sprouty 4, a new ras antagonist on 5q31, interacts with the dual specificity kinase TESK1. Eur J Biochem 269:2546-2556
137. Sasaki A, Taketomi T, Kato R, Saeki K, Nonami A, Sasaki M, Kuriyama M, Saito N, Shibuya M, Yoshimura A 2003 Mammalian Sprouty4 suppresses Ras-independent ERK activation by binding to Raf1. Nat Cell Biol 5:427-432
139
138. Tsumura Y, Toshima J, Leeksma OC, Ohashi K, Mizuno K 2005 Sprouty-4 negatively regulates cell spreading by inhibiting the kinase activity of testicular protein kinase. Biochem J 387:627-637
139. Miyake S, Lupher ML, Jr., Druker B, Band H 1998 The tyrosine kinase regulator Cbl enhances the ubiquitination and degradation of the platelet-derived growth factor receptor alpha. Proc Natl Acad Sci U S A 95:7927-7932
140. Levkowitz G, Waterman H, Zamir E, Kam Z, Oved S, Langdon WY, Beguinot L, Geiger B, Yarden Y 1998 c-Cbl/Sli-1 regulates endocytic sorting and ubiquitination of the epidermal growth factor receptor. Genes Dev 12:3663-3674
141. Ettenberg SA, Magnifico A, Cuello M, Nau MM, Rubinstein YR, Yarden Y, Weissman AM, Lipkowitz S 2001 Cbl-b-dependent coordinated degradation of the epidermal growth factor receptor signaling complex. J Biol Chem 276:27677-27684
142. Guy GR, Jackson RA, Yusoff P, Chow SY 2009 Sprouty proteins: modified modulators, matchmakers or missing links? J Endocrinol 203:191-202
143. Wong ES, Lim J, Low BC, Chen Q, Guy GR 2001 Evidence for direct interaction between Sprouty and Cbl. J Biol Chem 276:5866-5875
144. Wong ES, Fong CW, Lim J, Yusoff P, Low BC, Langdon WY, Guy GR 2002 Sprouty2 attenuates epidermal growth factor receptor ubiquitylation and endocytosis, and consequently enhances Ras/ERK signalling. Embo J 21:4796-4808
145. Hacohen N, Kramer S, Sutherland D, Hiromi Y, Krasnow MA 1998 sprouty encodes a novel antagonist of FGF signaling that patterns apical branching of the Drosophila airways. Cell 92:253-263
146. Kramer S, Okabe M, Hacohen N, Krasnow MA, Hiromi Y 1999 Sprouty: a common antagonist of FGF and EGF signaling pathways in Drosophila. Development 126:2515-2525
147. Ozaki K, Kadomoto R, Asato K, Tanimura S, Itoh N, Kohno M 2001 ERK pathway positively regulates the expression of Sprouty genes. Biochem Biophys Res Commun 285:1084-1088
148. Lee CC, Putnam AJ, Miranti CK, Gustafson M, Wang LM, Vande Woude GF, Gao CF 2004 Overexpression of sprouty 2 inhibits HGF/SF-mediated cell growth, invasion, migration, and cytokinesis. Oncogene 23:5193-5202
149. Shaw AT, Meissner A, Dowdle JA, Crowley D, Magendantz M, Ouyang C, Parisi T, Rajagopal J, Blank LJ, Bronson RT, Stone JR, Tuveson DA, Jaenisch R, Jacks T 2007 Sprouty-2 regulates oncogenic K-ras in lung development and tumorigenesis. Genes Dev 21:694-707
150. Frolov A, Chahwan S, Ochs M, Arnoletti JP, Pan ZZ, Favorova O, Fletcher J, von Mehren M, Eisenberg B, Godwin AK 2003 Response markers and the molecular mechanisms of action of Gleevec in gastrointestinal stromal tumors. Mol Cancer Ther 2:699-709
151. Lo TL, Yusoff P, Fong CW, Guo K, McCaw BJ, Phillips WA, Yang H, Wong ES, Leong HF, Zeng Q, Putti TC, Guy GR 2004 The ras/mitogen-activated protein kinase pathway inhibitor and likely tumor suppressor proteins, sprouty 1 and sprouty 2 are deregulated in breast cancer. Cancer Res 64:6127-6136
140
152. Kwabi-Addo B, Wang J, Erdem H, Vaid A, Castro P, Ayala G, Ittmann M 2004 The expression of Sprouty1, an inhibitor of fibroblast growth factor signal transduction, is decreased in human prostate cancer. Cancer Res 64:4728-4735
153. Fritzsche S, Kenzelmann M, Hoffmann MJ, Muller M, Engers R, Grone HJ, Schulz WA 2006 Concomitant down-regulation of SPRY1 and SPRY2 in prostate carcinoma. Endocr Relat Cancer 13:839-849
154. Velasco A, Pallares J, Santacana M, Gatius S, Fernandez M, Domingo M, Valls J, Yeramian A, Encinas M, Dolcet X, Matias-Guiu X Promoter hypermethylation and expression of sprouty 2 in endometrial carcinoma. Hum Pathol 42:185-193
155. Fong CW, Chua MS, McKie AB, Ling SH, Mason V, Li R, Yusoff P, Lo TL, Leung HY, So SK, Guy GR 2006 Sprouty 2, an inhibitor of mitogen-activated protein kinase signaling, is down-regulated in hepatocellular carcinoma. Cancer Res 66:2048-2058
156. McKie AB, Douglas DA, Olijslagers S, Graham J, Omar MM, Heer R, Gnanapragasam VJ, Robson CN, Leung HY 2005 Epigenetic inactivation of the human sprouty2 (hSPRY2) homologue in prostate cancer. Oncogene 24:2166-2174
157. Takahashi M, Rhodes DR, Furge KA, Kanayama H, Kagawa S, Haab BB, Teh BT 2001 Gene expression profiling of clear cell renal cell carcinoma: gene identification and prognostic classification. Proc Natl Acad Sci U S A 98:9754-9759
158. Calvisi DF, Ladu S, Gorden A, Farina M, Lee JS, Conner EA, Schroeder I, Factor VM, Thorgeirsson SS 2007 Mechanistic and prognostic significance of aberrant methylation in the molecular pathogenesis of human hepatocellular carcinoma. J Clin Invest 117:2713-2722
159. Tennis MA, Van Scoyk MM, Freeman SV, Vandervest KM, Nemenoff RA, Winn RA 2010 Sprouty-4 inhibits transformed cell growth, migration and invasion, and epithelial-mesenchymal transition, and is regulated by Wnt7A through PPARgamma in non-small cell lung cancer. Mol Cancer Res 8:833-843
160. Sutterluty H, Mayer CE, Setinek U, Attems J, Ovtcharov S, Mikula M, Mikulits W, Micksche M, Berger W 2007 Down-regulation of Sprouty2 in non-small cell lung cancer contributes to tumor malignancy via extracellular signal-regulated kinase pathway-dependent and -independent mechanisms. Mol Cancer Res 5:509-520
161. Wang J, Thompson B, Ren C, Ittmann M, Kwabi-Addo B 2006 Sprouty4, a suppressor of tumor cell motility, is down regulated by DNA methylation in human prostate cancer. Prostate 66:613-624
162. Miyoshi K, Wakioka T, Nishinakamura H, Kamio M, Yang L, Inoue M, Hasegawa M, Yonemitsu Y, Komiya S, Yoshimura A 2004 The Sprouty-related protein, Spred, inhibits cell motility, metastasis, and Rho-mediated actin reorganization. Oncogene 23:5567-5576
163. Harris AL 2002 Hypoxia--a key regulatory factor in tumour growth. Nat Rev Cancer 2:38-47
164. Zhang Y, Li M, Yao Q, Chen C 2007 Recent advances in tumor hypoxia: tumor progression, molecular mechanisms, and therapeutic implications. Med Sci Monit 13:RA175-180
141
165. Berra E, Ginouves A, Pouyssegur J 2006 The hypoxia-inducible-factor hydroxylases bring fresh air into hypoxia signalling. EMBO Rep 7:41-45
166. Kim W, Kaelin WG, Jr. 2003 The von Hippel-Lindau tumor suppressor protein: new insights into oxygen sensing and cancer. Curr Opin Genet Dev 13:55-60
167. del Peso L, Castellanos MC, Temes E, Martin-Puig S, Cuevas Y, Olmos G, Landazuri MO 2003 The von Hippel Lindau/hypoxia-inducible factor (HIF) pathway regulates the transcription of the HIF-proline hydroxylase genes in response to low oxygen. J Biol Chem 278:48690-48695
168. Aprelikova O, Chandramouli GV, Wood M, Vasselli JR, Riss J, Maranchie JK, Linehan WM, Barrett JC 2004 Regulation of HIF prolyl hydroxylases by hypoxia-inducible factors. J Cell Biochem 92:491-501
169. Nakayama K, Frew IJ, Hagensen M, Skals M, Habelhah H, Bhoumik A, Kadoya T, Erdjument-Bromage H, Tempst P, Frappell PB, Bowtell DD, Ronai Z 2004 Siah2 regulates stability of prolyl-hydroxylases, controls HIF1alpha abundance, and modulates physiological responses to hypoxia. Cell 117:941-952
170. Nakayama K, Gazdoiu S, Abraham R, Pan ZQ, Ronai Z 2007 Hypoxia-induced assembly of prolyl hydroxylase PHD3 into complexes: implications for its activity and susceptibility for degradation by the E3 ligase Siah2. Biochem J 401:217-226
171. Nakayama K, Ronai Z 2004 Siah: new players in the cellular response to hypoxia. Cell Cycle 3:1345-1347
172. Callapina M, Zhou J, Schmid T, Kohl R, Brune B 2005 NO restores HIF-1alpha hydroxylation during hypoxia: role of reactive oxygen species. Free Radic Biol Med 39:925-936
173. Ueno M, Maeno T, Nomura M, Aoyagi-Ikeda K, Matsui H, Hara K, Tanaka T, Iso T, Suga T, Kurabayashi M 2011 Hypoxia-inducible factor-1alpha mediates TGF-beta-induced PAI-1 production in alveolar macrophages in pulmonary fibrosis. Am J Physiol Lung Cell Mol Physiol 300:L740-752
174. Spinella F, Rosano L, Del Duca M, Di Castro V, Nicotra MR, Natali PG, Bagnato A 2010 Endothelin-1 inhibits prolyl hydroxylase domain 2 to activate hypoxia-inducible factor-1alpha in melanoma cells. PLoS One 5:e11241
175. Lando D, Peet DJ, Whelan DA, Gorman JJ, Whitelaw ML 2002 Asparagine hydroxylation of the HIF transactivation domain a hypoxic switch. Science 295:858-861
176. Kung AL, Wang S, Klco JM, Kaelin WG, Livingston DM 2000 Suppression of tumor growth through disruption of hypoxia-inducible transcription. Nat Med 6:1335-1340
177. Ivan M, Kondo K, Yang H, Kim W, Valiando J, Ohh M, Salic A, Asara JM, Lane WS, Kaelin WG, Jr. 2001 HIFalpha targeted for VHL-mediated destruction by proline hydroxylation: implications for O2 sensing. Science 292:464-468
178. Jaakkola P, Mole DR, Tian YM, Wilson MI, Gielbert J, Gaskell SJ, Kriegsheim A, Hebestreit HF, Mukherji M, Schofield CJ, Maxwell PH, Pugh CW, Ratcliffe PJ 2001 Targeting of HIF-alpha to the von Hippel-Lindau ubiquitylation complex by O2-regulated prolyl hydroxylation. Science 292:468-472
179. Neumann HP, Lips CJ, Hsia YE, Zbar B 1995 Von Hippel-Lindau syndrome. Brain Pathol 5:181-193
142
180. Zhuang Z, Gnarra JR, Dudley CF, Zbar B, Linehan WM, Lubensky IA 1996 Detection of von Hippel-Lindau disease gene mutations in paraffin-embedded sporadic renal cell carcinoma specimens. Mod Pathol 9:838-842
181. Krieg M, Haas R, Brauch H, Acker T, Flamme I, Plate KH 2000 Up-regulation of hypoxia-inducible factors HIF-1alpha and HIF-2alpha under normoxic conditions in renal carcinoma cells by von Hippel-Lindau tumor suppressor gene loss of function. Oncogene 19:5435-5443
182. Maxwell PH, Wiesener MS, Chang GW, Clifford SC, Vaux EC, Cockman ME, Wykoff CC, Pugh CW, Maher ER, Ratcliffe PJ 1999 The tumour suppressor protein VHL targets hypoxia-inducible factors for oxygen-dependent proteolysis. Nature 399:271-275
183. Gnarra JR, Zhou S, Merrill MJ, Wagner JR, Krumm A, Papavassiliou E, Oldfield EH, Klausner RD, Linehan WM 1996 Post-transcriptional regulation of vascular endothelial growth factor mRNA by the product of the VHL tumor suppressor gene. Proc Natl Acad Sci U S A 93:10589-10594
184. Iliopoulos O, Levy AP, Jiang C, Kaelin WG, Jr., Goldberg MA 1996 Negative regulation of hypoxia-inducible genes by the von Hippel-Lindau protein. Proc Natl Acad Sci U S A 93:10595-10599
185. Iliopoulos O, Kibel A, Gray S, Kaelin WG, Jr. 1995 Tumour suppression by the human von Hippel-Lindau gene product. Nat Med 1:822-826
186. Chen EY, Mazure NM, Cooper JA, Giaccia AJ 2001 Hypoxia activates a platelet-derived growth factor receptor/phosphatidylinositol 3-kinase/Akt pathway that results in glycogen synthase kinase-3 inactivation. Cancer Res 61:2429-2433
187. Zundel W, Schindler C, Haas-Kogan D, Koong A, Kaper F, Chen E, Gottschalk AR, Ryan HE, Johnson RS, Jefferson AB, Stokoe D, Giaccia AJ 2000 Loss of PTEN facilitates HIF-1-mediated gene expression. Genes Dev 14:391-396
188. Zhong H, Chiles K, Feldser D, Laughner E, Hanrahan C, Georgescu MM, Simons JW, Semenza GL 2000 Modulation of hypoxia-inducible factor 1alpha expression by the epidermal growth factor/phosphatidylinositol 3-kinase/PTEN/AKT/FRAP pathway in human prostate cancer cells: implications for tumor angiogenesis and therapeutics. Cancer Res 60:1541-1545
189. Fang J, Ding M, Yang L, Liu LZ, Jiang BH 2007 PI3K/PTEN/AKT signaling regulates prostate tumor angiogenesis. Cell Signal 19:2487-2497
190. Pore N, Jiang Z, Shu HK, Bernhard E, Kao GD, Maity A 2006 Akt1 activation can augment hypoxia-inducible factor-1alpha expression by increasing protein translation through a mammalian target of rapamycin-independent pathway. Mol Cancer Res 4:471-479
191. Treins C, Giorgetti-Peraldi S, Murdaca J, Monthouel-Kartmann MN, Van Obberghen E 2005 Regulation of hypoxia-inducible factor (HIF)-1 activity and expression of HIF hydroxylases in response to insulin-like growth factor I. Mol Endocrinol 19:1304-1317
192. Mottet D, Dumont V, Deccache Y, Demazy C, Ninane N, Raes M, Michiels C 2003 Regulation of hypoxia-inducible factor-1alpha protein level during hypoxic conditions by the phosphatidylinositol 3-kinase/Akt/glycogen synthase kinase 3beta pathway in HepG2 cells. J Biol Chem 278:31277-31285
143
193. Zhou J, Schmid T, Frank R, Brune B 2004 PI3K/Akt is required for heat shock proteins to protect hypoxia-inducible factor 1alpha from pVHL-independent degradation. J Biol Chem 279:13506-13513
194. Shayesteh L, Lu Y, Kuo WL, Baldocchi R, Godfrey T, Collins C, Pinkel D, Powell B, Mills GB, Gray JW 1999 PIK3CA is implicated as an oncogene in ovarian cancer. Nat Genet 21:99-102
195. Zhang L, Yang N, Katsaros D, Huang W, Park JW, Fracchioli S, Vezzani C, Rigault de la Longrais IA, Yao W, Rubin SC, Coukos G 2003 The oncogene phosphatidylinositol 3'-kinase catalytic subunit alpha promotes angiogenesis via vascular endothelial growth factor in ovarian carcinoma. Cancer Res 63:4225-4231
196. Majumder PK, Febbo PG, Bikoff R, Berger R, Xue Q, McMahon LM, Manola J, Brugarolas J, McDonnell TJ, Golub TR, Loda M, Lane HA, Sellers WR 2004 mTOR inhibition reverses Akt-dependent prostate intraepithelial neoplasia through regulation of apoptotic and HIF-1-dependent pathways. Nat Med 10:594-601
197. Chan DA, Sutphin PD, Denko NC, Giaccia AJ 2002 Role of prolyl hydroxylation in oncogenically stabilized hypoxia-inducible factor-1alpha. J Biol Chem 277:40112-40117
198. Lim JH, Lee ES, You HJ, Lee JW, Park JW, Chun YS 2004 Ras-dependent induction of HIF-1alpha785 via the Raf/MEK/ERK pathway: a novel mechanism of Ras-mediated tumor promotion. Oncogene 23:9427-9431
199. Blancher C, Moore JW, Robertson N, Harris AL 2001 Effects of ras and von Hippel-Lindau (VHL) gene mutations on hypoxia-inducible factor (HIF)-1alpha, HIF-2alpha, and vascular endothelial growth factor expression and their regulation by the phosphatidylinositol 3'-kinase/Akt signaling pathway. Cancer Res 61:7349-7355
200. Jiang BH, Agani F, Passaniti A, Semenza GL 1997 V-SRC induces expression of hypoxia-inducible factor 1 (HIF-1) and transcription of genes encoding vascular endothelial growth factor and enolase 1: involvement of HIF-1 in tumor progression. Cancer Res 57:5328-5335
201. Flugel D, Gorlach A, Michiels C, Kietzmann T 2007 Glycogen synthase kinase 3 phosphorylates hypoxia-inducible factor 1alpha and mediates its destabilization in a VHL-independent manner. Mol Cell Biol 27:3253-3265
202. Gradin K, McGuire J, Wenger RH, Kvietikova I, fhitelaw ML, Toftgard R, Tora L, Gassmann M, Poellinger L 1996 Functional interference between hypoxia and dioxin signal transduction pathways: competition for recruitment of the Arnt transcription factor. Mol Cell Biol 16:5221-5231
203. Minet E, Mottet D, Michel G, Roland I, Raes M, Remacle J, Michiels C 1999 Hypoxia-induced activation of HIF-1: role of HIF-1alpha-Hsp90 interaction. FEBS Lett 460:251-256
204. Katschinski DM, Le L, Schindler SG, Thomas T, Voss AK, Wenger RH 2004 Interaction of the PAS B domain with HSP90 accelerates hypoxia-inducible factor-1alpha stabilization. Cell Physiol Biochem 14:351-360
205. Isaacs JS, Jung YJ, Neckers L 2004 Aryl hydrocarbon nuclear translocator (ARNT) promotes oxygen-independent stabilization of hypoxia-inducible factor-
144
1alpha by modulating an Hsp90-dependent regulatory pathway. J Biol Chem 279:16128-16135
206. Minet E, Ernest I, Michel G, Roland I, Remacle J, Raes M, Michiels C 1999 HIF1A gene transcription is dependent on a core promoter sequence encompassing activating and inhibiting sequences located upstream from the transcription initiation site and cis elements located within the 5'UTR. Biochem Biophys Res Commun 261:534-540
207. Liu YV, Baek JH, Zhang H, Diez R, Cole RN, Semenza GL 2007 RACK1 competes with HSP90 for binding to HIF-1alpha and is required for O(2)-independent and HSP90 inhibitor-induced degradation of HIF-1alpha. Mol Cell 25:207-217
208. Dulak J, Deshane J, Jozkowicz A, Agarwal A 2008 Heme oxygenase-1 and carbon monoxide in vascular pathobiology: focus on angiogenesis. Circulation 117:231-241
209. Choi YK, Kim CK, Lee H, Jeoung D, Ha KS, Kwon YG, Kim KW, Kim YM 2010 Carbon monoxide promotes VEGF expression by increasing HIF-1alpha protein level via two distinct mechanisms, translational activation and stabilization of HIF-1alpha protein. J Biol Chem 285:32116-32125
210. Lee J, Park SY, Lee EK, Park CG, Chung HC, Rha SY, Kim YK, Bae GU, Kim BK, Han JW, Lee HY 2006 Activation of hypoxia-inducible factor-1alpha is necessary for lysophosphatidic acid-induced vascular endothelial growth factor expression. Clin Cancer Res 12:6351-6358
211. Thomas R, Kim MH 2008 HIF-1 alpha: a key survival factor for serum-deprived prostate cancer cells. Prostate 68:1405-1415
212. Dimova EY, Moller U, Herzig S, Fink T, Zachar V, Ebbesen P, Kietzmann T 2005 Transcriptional regulation of plasminogen activator inhibitor-1 expression by insulin-like growth factor-1 via MAP kinases and hypoxia-inducible factor-1 in HepG2 cells. Thromb Haemost 93:1176-1184
213. Bardos JI, Chau NM, Ashcroft M 2004 Growth factor-mediated induction of HDM2 positively regulates hypoxia-inducible factor 1alpha expression. Mol Cell Biol 24:2905-2914
214. Huang Y, Hua K, Zhou X, Jin H, Chen X, Lu X, Yu Y, Zha X, Feng Y 2008 Activation of the PI3K/AKT pathway mediates FSH-stimulated VEGF expression in ovarian serous cystadenocarcinoma. Cell Res 18:780-791
215. Zhong H, De Marzo AM, Laughner E, Lim M, Hilton DA, Zagzag D, Buechler P, Isaacs WB, Semenza GL, Simons JW 1999 Overexpression of hypoxia-inducible factor 1alpha in common human cancers and their metastases. Cancer Res 59:5830-5835
216. Talks KL, Turley H, Gatter KC, Maxwell PH, Pugh CW, Ratcliffe PJ, Harris AL 2000 The expression and distribution of the hypoxia-inducible factors HIF-1alpha and HIF-2alpha in normal human tissues, cancers, and tumor-associated macrophages. Am J Pathol 157:411-421
217. Birner P, Schindl M, Obermair A, Breitenecker G, Oberhuber G 2001 Expression of hypoxia-inducible factor 1alpha in epithelial ovarian tumors: its impact on prognosis and on response to chemotherapy. Clin Cancer Res 7:1661-1668
145
218. Daponte A, Ioannou M, Mylonis I, Simos G, Minas M, Messinis IE, Koukoulis G 2008 Prognostic significance of Hypoxia-Inducible Factor 1 alpha(HIF-1 alpha) expression in serous ovarian cancer: an immunohistochemical study. BMC Cancer 8:335
219. Semenza GL 2003 Targeting HIF-1 for cancer therapy. Nat Rev Cancer 3:721-732
220. Rankin EB, Giaccia AJ 2008 The role of hypoxia-inducible factors in tumorigenesis. Cell Death Differ 15:678-685
221. North S, Moenner M, Bikfalvi A 2005 Recent developments in the regulation of the angiogenic switch by cellular stress factors in tumors. Cancer Lett 218:1-14
222. Folkman J, Watson K, Ingber D, Hanahan D 1989 Induction of angiogenesis during the transition from hyperplasia to neoplasia. Nature 339:58-61
223. Giordano FJ, Johnson RS 2001 Angiogenesis: the role of the microenvironment in flipping the switch. Curr Opin Genet Dev 11:35-40
224. de Fraipont F, Nicholson AC, Feige JJ, Van Meir EG 2001 Thrombospondins and tumor angiogenesis. Trends Mol Med 7:401-407
225. Mancuso MR, Davis R, Norberg SM, O'Brien S, Sennino B, Nakahara T, Yao VJ, Inai T, Brooks P, Freimark B, Shalinsky DR, Hu-Lowe DD, McDonald DM 2006 Rapid vascular regrowth in tumors after reversal of VEGF inhibition. J Clin Invest 116:2610-2621
226. Forsythe JA, Jiang BH, Iyer NV, Agani F, Leung SW, Koos RD, Semenza GL 1996 Activation of vascular endothelial growth factor gene transcription by hypoxia-inducible factor 1. Mol Cell Biol 16:4604-4613
227. Maxwell PH, Dachs GU, Gleadle JM, Nicholls LG, Harris AL, Stratford IJ, Hankinson O, Pugh CW, Ratcliffe PJ 1997 Hypoxia-inducible factor-1 modulates gene expression in solid tumors and influences both angiogenesis and tumor growth. Proc Natl Acad Sci U S A 94:8104-8109
228. Tang N, Wang L, Esko J, Giordano FJ, Huang Y, Gerber HP, Ferrara N, Johnson RS 2004 Loss of HIF-1alpha in endothelial cells disrupts a hypoxia-driven VEGF autocrine loop necessary for tumorigenesis. Cancer Cell 6:485-495
229. Jensen RL, Ragel BT, Whang K, Gillespie D 2006 Inhibition of hypoxia inducible factor-1alpha (HIF-1alpha) decreases vascular endothelial growth factor (VEGF) secretion and tumor growth in malignant gliomas. J Neurooncol 78:233-247
230. Yuecheng Y, Hongmei L, Xiaoyan X 2006 Clinical evaluation of E-cadherin expression and its regulation mechanism in epithelial ovarian cancer. Clin Exp Metastasis 23:65-74
231. Brinck U, Jacobs S, Neuss M, Tory K, Rath W, Kulle B, Fuzesi L 2004 Diffuse growth pattern affects E-cadherin expression in invasive breast cancer. Anticancer Res 24:2237-2242
232. Kallakury BV, Sheehan CE, Winn-Deen E, Oliver J, Fisher HA, Kaufman RP, Jr., Ross JS 2001 Decreased expression of catenins (alpha and beta), p120 CTN, and E-cadherin cell adhesion proteins and E-cadherin gene promoter methylation in prostatic adenocarcinomas. Cancer 92:2786-2795
233. Sulzer MA, Leers MP, van Noord JA, Bollen EC, Theunissen PH 1998 Reduced E-cadherin expression is associated with increased lymph node metastasis and
146
unfavorable prognosis in non-small cell lung cancer. Am J Respir Crit Care Med 157:1319-1323
234. Hanahan D, Weinberg RA 2000 The hallmarks of cancer. Cell 100:57-70 235. Imai T, Horiuchi A, Wang C, Oka K, Ohira S, Nikaido T, Konishi I 2003 Hypoxia
attenuates the expression of E-cadherin via up-regulation of SNAIL in ovarian carcinoma cells. Am J Pathol 163:1437-1447
236. Krishnamachary B, Zagzag D, Nagasawa H, Rainey K, Okuyama H, Baek JH, Semenza GL 2006 Hypoxia-inducible factor-1-dependent repression of E-cadherin in von Hippel-Lindau tumor suppressor-null renal cell carcinoma mediated by TCF3, ZFHX1A, and ZFHX1B. Cancer Res 66:2725-2731
237. Erler JT, Bennewith KL, Nicolau M, Dornhofer N, Kong C, Le QT, Chi JT, Jeffrey SS, Giaccia AJ 2006 Lysyl oxidase is essential for hypoxia-induced metastasis. Nature 440:1222-1226
238. Lu X, Kang Y 2010 Hypoxia and hypoxia-inducible factors: master regulators of metastasis. Clin Cancer Res 16:5928-5935
239. Lin MT, Kuo IH, Chang CC, Chu CY, Chen HY, Lin BR, Sureshbabu M, Shih HJ, Kuo ML 2008 Involvement of hypoxia-inducing factor-1alpha-dependent plasminogen activator inhibitor-1 up-regulation in Cyr61/CCN1-induced gastric cancer cell invasion. J Biol Chem 283:15807-15815
240. Edwin F, Anderson K, Ying C, Patel TB 2009 Intermolecular interactions of Sprouty proteins and their implications in development and disease. Mol Pharmacol 76:679-691
241. Cannistra SA 1993 Cancer of the ovary. N Engl J Med 329:1550-1559 242. Zeineldin R, Muller CY, Stack MS, Hudson LG 2010 Targeting the EGF receptor
for ovarian cancer therapy. J Oncol 2010:414676 243. Pradhan M, Risberg BA, Trope CG, van de Rijn M, Gilks CB, Lee CH 2010
Gross genomic alterations and gene expression profiles of high- grade serous carcinoma of the ovary with and without BRCA1 inactivation. BMC Cancer 10:493
244. Press JZ, De Luca A, Boyd N, Young S, Troussard A, Ridge Y, Kaurah P, Kalloger SE, Blood KA, Smith M, Spellman PT, Wang Y, Miller DM, Horsman D, Faham M, Gilks CB, Gray J, Huntsman DG 2008 Ovarian carcinomas with genetic and epigenetic BRCA1 loss have distinct molecular abnormalities. BMC Cancer 8:17
245. Brown LA, Kalloger SE, Miller MA, Shih Ie M, McKinney SE, Santos JL, Swenerton K, Spellman PT, Gray J, Gilks CB, Huntsman DG 2008 Amplification of 11q13 in ovarian carcinoma. Genes Chromosomes Cancer 47:481-489
246. Maines-Bandiera SL, Kruk PA, Auersperg N 1992 Simian virus 40-transformed human ovarian surface epithelial cells escape normal growth controls but retain morphogenetic responses to extracellular matrix. Am J Obstet Gynecol 167:729-735
247. Geisinger KR, Kute TE, Pettenati MJ, Welander CE, Dennard Y, Collins LA, Berens ME 1989 Characterization of a human ovarian carcinoma cell line with estrogen and progesterone receptors. Cancer 63:280-288
147
248. Cheng JC, Klausen C, Leung PC 2010 Hydrogen peroxide mediates EGF-induced down-regulation of E-cadherin expression via p38 MAPK and snail in human ovarian cancer cells. Molecular endocrinology (Baltimore, Md 24:1569-1580
249. Zweemer RP, Ryan A, Snijders AM, Hermsen MA, Meijer GA, Beller U, Menko FH, Jacobs IJ, Baak JP, Verheijen RH, Kenemans P, van Diest PJ 2001 Comparative genomic hybridization of microdissected familial ovarian carcinoma: two deleted regions on chromosome 15q not previously identified in sporadic ovarian carcinoma. Lab Invest 81:1363-1370
250. Jongsma AP, Piek JM, Zweemer RP, Verheijen RH, Klein Gebbinck JW, van Kamp GJ, Jacobs IJ, Shaw P, van Diest PJ, Kenemans P 2002 Molecular evidence for putative tumour suppressor genes on chromosome 13q specific to BRCA1 related ovarian and fallopian tube cancer. Mol Pathol 55:305-309
251. Kim TM, Benedict WF, Xu HJ, Hu SX, Gosewehr J, Velicescu M, Yin E, Zheng J, D'Ablaing G, Dubeau L 1994 Loss of heterozygosity on chromosome 13 is common only in the biologically more aggressive subtypes of ovarian epithelial tumors and is associated with normal retinoblastoma gene expression. Cancer Res 54:605-609
252. Yang-Feng TL, Li S, Han H, Schwartz PE 1992 Frequent loss of heterozygosity on chromosomes Xp and 13q in human ovarian cancer. Int J Cancer 52:575-580
253. Voutilainen KA, Anttila MA, Sillanpaa SM, Ropponen KM, Saarikoski SV, Juhola MT, Kosma VM 2006 Prognostic significance of E-cadherin-catenin complex in epithelial ovarian cancer. J Clin Pathol 59:460-467
254. Rathi A, Virmani AK, Schorge JO, Elias KJ, Maruyama R, Minna JD, Mok SC, Girard L, Fishman DA, Gazdar AF 2002 Methylation profiles of sporadic ovarian tumors and nonmalignant ovaries from high-risk women. Clin Cancer Res 8:3324-3331
255. Pon YL, Zhou HY, Cheung AN, Ngan HY, Wong AS 2008 p70 S6 kinase promotes epithelial to mesenchymal transition through snail induction in ovarian cancer cells. Cancer Res 68:6524-6532
256. Park SH, Cheung LW, Wong AS, Leung PC 2008 Estrogen regulates Snail and Slug in the down-regulation of E-cadherin and induces metastatic potential of ovarian cancer cells through estrogen receptor alpha. Mol Endocrinol 22:2085-2098
257. Willmarth NE, Ethier SP 2006 Autocrine and juxtacrine effects of amphiregulin on the proliferative, invasive, and migratory properties of normal and neoplastic human mammary epithelial cells. J Biol Chem 281:37728-37737
258. Park JY, Su YQ, Ariga M, Law E, Jin SL, Conti M 2004 EGF-like growth factors as mediators of LH action in the ovulatory follicle. Science 303:682-684
259. Choi JH, Choi KC, Auersperg N, Leung PC 2006 Gonadotropins activate proteolysis and increase invasion through protein kinase A and phosphatidylinositol 3-kinase pathways in human epithelial ovarian cancer cells. Cancer Res 66:3912-3920
260. Lau MT, Wong AS, Leung PC 2010 Gonadotropins induce tumor cell migration and invasion by increasing cyclooxygenases expression and prostaglandin E(2) production in human ovarian cancer cells. Endocrinology 151:2985-2993
148
261. Zhang YW, Vande Woude GF 2007 Mig-6, signal transduction, stress response and cancer. Cell Cycle 6:507-513
262. Wakioka T, Sasaki A, Kato R, Shouda T, Matsumoto A, Miyoshi K, Tsuneoka M, Komiya S, Baron R, Yoshimura A 2001 Spred is a Sprouty-related suppressor of Ras signalling. Nature 412:647-651
263. Willmarth NE, Ethier SP 2008 Amphiregulin as a novel target for breast cancer therapy. J Mammary Gland Biol Neoplasia 13:171-179
264. Menashi S, Serova M, Ma L, Vignot S, Mourah S, Calvo F 2003 Regulation of extracellular matrix metalloproteinase inducer and matrix metalloproteinase expression by amphiregulin in transformed human breast epithelial cells. Cancer Res 63:7575-7580
265. Kondapaka SB, Fridman R, Reddy KB 1997 Epidermal growth factor and amphiregulin up-regulate matrix metalloproteinase-9 (MMP-9) in human breast cancer cells. Int J Cancer 70:722-726
266. P Oc, Modjtahedi H, Rhys-Evans P, Court WJ, Box GM, Eccles SA 2000 Epidermal growth factor-like ligands differentially up-regulate matrix metalloproteinase 9 in head and neck squamous carcinoma cells. Cancer Res 60:1121-1128
267. Liu Z, Klominek J 2003 Regulation of matrix metalloprotease activity in malignant mesothelioma cell lines by growth factors. Thorax 58:198-203
268. Picihard V, Berthois Y, Roccabianca M, Prevot C, Sarrazin M, Portugal H, Kumar S, Kumar P, Rognoni JB 2006 Concomitant cell growth and differentiation are dependent on erbB1 and integrin activation in an autonomously surviving colon adenocarcinoma: involvement of autocrine amphiregulin secretion. Anticancer Res 26:2769-2783
269. Chung E, Cook PW, Parkos CA, Park YK, Pittelkow MR, Coffey RJ 2005 Amphiregulin causes functional downregulation of adherens junctions in psoriasis. J Invest Dermatol 124:1134-1140
270. Chung E, Graves-Deal R, Franklin JL, Coffey RJ 2005 Differential effects of amphiregulin and TGF-alpha on the morphology of MDCK cells. Exp Cell Res 309:149-160
271. Sasaki A, Taketomi T, Wakioka T, Kato R, Yoshimura A 2001 Identification of a dominant negative mutant of Sprouty that potentiates fibroblast growth factor- but not epidermal growth factor-induced ERK activation. J Biol Chem 276:36804-36808
272. Impagnatiello MA, Weitzer S, Gannon G, Compagni A, Cotten M, Christofori G 2001 Mammalian sprouty-1 and -2 are membrane-anchored phosphoprotein inhibitors of growth factor signaling in endothelial cells. J Cell Biol 152:1087-1098
273. Ozaki K, Miyazaki S, Tanimura S, Kohno M 2005 Efficient suppression of FGF-2-induced ERK activation by the cooperative interaction among mammalian Sprouty isoforms. J Cell Sci 118:5861-5871
274. Taniguchi K, Ishizaki T, Ayada T, Sugiyama Y, Wakabayashi Y, Sekiya T, Nakagawa R, Yoshimura A 2009 Sprouty4 deficiency potentiates Ras-independent angiogenic signals and tumor growth. Cancer Sci 100:1648-1654
149
275. Lau MT, Klausen C, Leung PC 2011 E-cadherin inhibits tumor cell growth by suppressing PI3K/Akt signaling via beta-catenin-Egr1-mediated PTEN expression. Oncogene 30:2753-2766
276. Winn RA, Van Scoyk M, Hammond M, Rodriguez K, Crossno JT, Jr., Heasley LE, Nemenoff RA 2006 Antitumorigenic effect of Wnt 7a and Fzd 9 in non-small cell lung cancer cells is mediated through ERK-5-dependent activation of peroxisome proliferator-activated receptor gamma. J Biol Chem 281:26943-26950
277. Chambers D, Mason I 2000 Expression of sprouty2 during early development of the chick embryo is coincident with known sites of FGF signalling. Mech Dev 91:361-364
278. Chi L, Zhang S, Lin Y, Prunskaite-Hyyrylainen R, Vuolteenaho R, Itaranta P, Vainio S 2004 Sprouty proteins regulate ureteric branching by coordinating reciprocal epithelial Wnt11, mesenchymal Gdnf and stromal Fgf7 signalling during kidney development. Development 131:3345-3356
279. Ding W, Bellusci S, Shi W, Warburton D 2003 Functional analysis of the human Sprouty2 gene promoter. Gene 322:175-185
280. Ding W, Bellusci S, Shi W, Warburton D 2004 Genomic structure and promoter characterization of the human Sprouty4 gene, a novel regulator of lung morphogenesis. Am J Physiol Lung Cell Mol Physiol 287:L52-59
281. Haigl B, Mayer CE, Siegwart G, Sutterluty H 2010 Sprouty4 levels are increased under hypoxic conditions by enhanced mRNA stability and transcription. Biol Chem 391:813-821
282. Dayan F, Bilton RL, Laferriere J, Trottier E, Roux D, Pouyssegur J, Mazure NM 2009 Activation of HIF-1alpha in exponentially growing cells via hypoxic stimulation is independent of the Akt/mTOR pathway. J Cell Physiol 218:167-174
283. Park SY, Jeong KJ, Lee J, Yoon DS, Choi WS, Kim YK, Han JW, Kim YM, Kim BK, Lee HY 2007 Hypoxia enhances LPA-induced HIF-1alpha and VEGF expression: their inhibition by resveratrol. Cancer Lett 258:63-69
284. Abe M, Naski MC 2004 Regulation of sprouty expression by PLCgamma and calcium-dependent signals. Biochem Biophys Res Commun 323:1040-1047
285. Nutt SL, Dingwell KS, Holt CE, Amaya E 2001 Xenopus Sprouty2 inhibits FGF-mediated gastrulation movements but does not affect mesoderm induction and patterning. Genes Dev 15:1152-1166
286. Barbachano A, Ordonez-Moran P, Garcia JM, Sanchez A, Pereira F, Larriba MJ, Martinez N, Hernandez J, Landolfi S, Bonilla F, Palmer HG, Rojas JM, Munoz A 2010 SPROUTY-2 and E-cadherin regulate reciprocally and dictate colon cancer cell tumourigenicity. Oncogene 29:4800–4813
287. Nadeau RJ, Toher JL, Yang X, Kovalenko D, Friesel R 2007 Regulation of Sprouty2 stability by mammalian Seven-in-Absentia homolog 2. J Cell Biochem 100:151-160
288. Nakayama K, Qi J, Ronai Z 2009 The ubiquitin ligase Siah2 and the hypoxia response. Mol Cancer Res 7:443-451
289. Skinner HD, Zheng JZ, Fang J, Agani F, Jiang BH 2004 Vascular endothelial growth factor transcriptional activation is mediated by hypoxia-inducible factor 1alpha, HDM2, and p70S6K1 in response to phosphatidylinositol 3-kinase/AKT signaling. J Biol Chem 279:45643-45651
150
290. Place TL, Fitzgerald MP, Venkataraman S, Vorrink SU, Case AJ, Teoh ML, Domann FE 2011 Aberrant promoter CpG methylation is a mechanism for impaired PHD3 expression in a diverse set of malignant cells. PLoS One 6:e14617
291. Berra E, Benizri E, Ginouves A, Volmat V, Roux D, Pouyssegur J 2003 HIF prolyl-hydroxylase 2 is the key oxygen sensor setting low steady-state levels of HIF-1alpha in normoxia. Embo J 22:4082-4090
292. Lo TL, Fong CW, Yusoff P, McKie AB, Chua MS, Leung HY, Guy GR 2006 Sprouty and cancer: The first terms report. Cancer Lett
293. Liao H, Hyman MC, Lawrence DA, Pinsky DJ 2007 Molecular regulation of the PAI-1 gene by hypoxia: contributions of Egr-1, HIF-1alpha, and C/EBPalpha. FASEB J 21:935-949
294. Wenger RH, Stiehl DP, Camenisch G 2005 Integration of oxygen signaling at the consensus HRE. Sci STKE 2005:re12
295. Takabuchi S, Hirota K, Nishi K, Oda S, Oda T, Shingu K, Takabayashi A, Adachi T, Semenza GL, Fukuda K 2004 The inhibitory effect of sodium nitroprusside on HIF-1 activation is not dependent on nitric oxide-soluble guanylyl cyclase pathway. Biochem Biophys Res Commun 324:417-423
296. Yagi H, Yotsumoto F, Miyamoto S 2008 Heparin-binding epidermal growth factor-like growth factor promotes transcoelomic metastasis in ovarian cancer through epithelial-mesenchymal transition. Mol Cancer Ther 7:3441-3451
297. Waterman H, Katz M, Rubin C, Shtiegman K, Lavi S, Elson A, Jovin T, Yarden Y 2002 A mutant EGF-receptor defective in ubiquitylation and endocytosis unveils a role for Grb2 in negative signaling. Embo J 21:303-313
298. Akbulut S, Reddi AL, Aggarwal P, Ambardekar C, Canciani B, Kim MK, Hix L, Vilimas T, Mason J, Basson MA, Lovatt M, Powell J, Collins S, Quatela S, Phillips M, Licht JD 2010 Sprouty proteins inhibit receptor-mediated activation of phosphatidylinositol-specific phospholipase C. Mol Biol Cell 21:3487-3496
299. Bertheau P, Turpin E, Rickman DS, Espie M, de Reynies A, Feugeas JP, Plassa LF, Soliman H, Varna M, de Roquancourt A, Lehmann-Che J, Beuzard Y, Marty M, Misset JL, Janin A, de The H 2007 Exquisite sensitivity of TP53 mutant and basal breast cancers to a dose-dense epirubicin-cyclophosphamide regimen. PLoS Med 4:e90
300. Wright G, Tan B, Rosenwald A, Hurt EH, Wiestner A, Staudt LM 2003 A gene expression-based method to diagnose clinically distinct subgroups of diffuse large B cell lymphoma. Proc Natl Acad Sci U S A 100:9991-9996
301. Faratian D, Sims AH, Mullen P, Kay C, Um I, Langdon SP, Harrison DJ 2011 Sprouty 2 is an independent prognostic factor in breast cancer and may be useful in stratifying patients for trastuzumab therapy. PLoS One 6:e23772
302. Huebert RC, Li Q, Adhikari N, Charles NJ, Han X, Ezzat MK, Grindle S, Park S, Ormaza S, Fermin D, Miller LW, Hall JL 2004 Identification and regulation of Sprouty1, a negative inhibitor of the ERK cascade, in the human heart. Physiol Genomics 18:284-289
303. Basson MA, Akbulut S, Watson-Johnson J, Simon R, Carroll TJ, Shakya R, Gross I, Martin GR, Lufkin T, McMahon AP, Wilson PD, Costantini FD, Mason IJ,
151
Licht JD 2005 Sprouty1 is a critical regulator of GDNF/RET-mediated kidney induction. Dev Cell 8:229-239
304. Zillhardt M, Park SM, Romero IL, Sawada K, Montag A, Krausz T, Yamada SD, Peter ME, Lengyel E 2011 Foretinib (GSK1363089), an orally available multikinase inhibitor of c-Met and VEGFR-2, blocks proliferation, induces anoikis, and impairs ovarian cancer metastasis. Clin Cancer Res 17:4042-4051
305. Li X, Brunton VG, Burgar HR, Wheldon LM, Heath JK 2004 FRS2-dependent SRC activation is required for fibroblast growth factor receptor-induced phosphorylation of Sprouty and suppression of ERK activity. J Cell Sci 117:6007-6017
306. Edwin F, Patel TB 2008 A novel role of Sprouty 2 in regulating cellular apoptosis. J Biol Chem 283:3181-3190
307. Fujita Y, Krause G, Scheffner M, Zechner D, Leddy HE, Behrens J, Sommer T, Birchmeier W 2002 Hakai, a c-Cbl-like protein, ubiquitinates and induces endocytosis of the E-cadherin complex. Nat Cell Biol 4:222-231
308. Ciechanover A, Iwai K 2004 The ubiquitin system: from basic mechanisms to the patient bed. IUBMB Life 56:193-201
309. Frankel A, Man S, Elliott P, Adams J, Kerbel RS 2000 Lack of multicellular drug resistance observed in human ovarian and prostate carcinoma treated with the proteasome inhibitor PS-341. Clin Cancer Res 6:3719-3728
310. LeBlanc R, Catley LP, Hideshima T, Lentzsch S, Mitsiades CS, Mitsiades N, Neuberg D, Goloubeva O, Pien CS, Adams J, Gupta D, Richardson PG, Munshi NC, Anderson KC 2002 Proteasome inhibitor PS-341 inhibits human myeloma cell growth in vivo and prolongs survival in a murine model. Cancer Res 62:4996-5000
311. Rajkumar SV, Richardson PG, Hideshima T, Anderson KC 2005 Proteasome inhibition as a novel therapeutic target in human cancer. J Clin Oncol 23:630-639
312. Adams J, Palombella VJ, Sausville EA, Johnson J, Destree A, Lazarus DD, Maas J, Pien CS, Prakash S, Elliott PJ 1999 Proteasome inhibitors: a novel class of potent and effective antitumor agents. Cancer Res 59:2615-2622
