The Impact of Chromosomal Aberrations on the
Regulation of Kallikrein 6 Expression in Serous Ovarian
Carcinoma
by
Jane Marie Bayani
A thesis submitted in conformity with the requirements for the degree of Doctorate of Philosophy
Department of Laboratory Medicine and Pathobiology University of Toronto
© Copyright by Jane Marie Bayani 2013
ii
The Impact of Chromosomal Aberrations on the Regulation of
Kallikrein 6 Expression in Serous Ovarian Carcinoma
Jane Marie Bayani
Doctorate of Philosophy
Department of Laboratory Medicine and Pathobiology University of Toronto
2013
Abstract
Ovarian cancer (OCa) remains the leading cause of death due to a gynecologic malignancy in
North American women, and the pathogenesis of this disease is a consequence of the interplay
between DNA, RNA and proteins. The genomes of these cancers are characterized by numerical
and structural aberrations, resulting in copy-number changes of the affected regions. The serine
protease, Kallikrein 6 (KLK6), is a promising biomarker and is over-expressed in OCa.
However, the mechanisms leading to the observed KLK6 overexpression are poorly understood;
and to date, no study examining the chromosomal contributions to the overexpression have been
conducted. Utilization of multi-colour Fluorescence in situ Hybridization (FISH)-based
technologies to untreated primary serous OCa samples and cancer cell lines, showed that the
KLK locus, on 19q13.3/4, is involved in both numerical and structural aberrations; was subject to
high-level copy-number heterogeneity (p<0.001); and structural rearrangements of 19q were
significantly co-related to grade (p<0.001). Patients with a loss of the KLK locus, or no
structural rearrangement on 19q, experienced a trend towards longer disease free survival (DFS)
and better overall survival (OS), over those with a gain or amplification, or with breakage events
on 19q. KLK6-specific immunohistochemistry (IHC) showed weak correlation with KLK6 copy-
iii
number, suggesting other mechanisms together with copy-number, drives its over-expression.
Among these mechanisms are microRNAs (miRNAs), also shown to be affected by the copy-
number changes in OCas. Therefore, we investigated the role of miRNAs in OCa and their role
in KLK6 regulation. Specifically, we examined the copy-number status and miRNA expression
in a representative OCa cell line, OVCAR-3. miRNA expression profiling of OCa cell lines and
primary tumours showed their differential expression, including the decrease in expression of the
let-7 family members, which are predicted to target KLK6. Indeed, when hsa-let-7a was
transiently transfected into OVCAR-3, a reduction of secreted KLK6 protein was detected. Thus,
the contribution of numerical and structural aberrations of the OCa genome can directly affect
the expression KLK6 through copy number, but is also aided post-transcriptionally by miRNAs.
iv
Dedication
This body of work is dedicated with love,
to my parents, Jean and Earl Bayani;
to my husband, Ash;
and our children, Leila and Ryan.
I am everything because of you, and nothing without you.
v
Acknowledgments
The completion of this work not only symbolizes the period of time spent in the PhD program, but a long journey that has taken me from being a disillusioned university graduate, through to a career in molecular cytogenetics. Therefore, this work not only represents the contributions of many to my scientific life, but also to my personal life. Thank you to Dr. Eleftherios Diamandis for taking the chance on a staunch molecular cytogeneticist and introducing her to the world of proteomics. Through you, I have seen the rewards that come with taking chances; and the lessons to be learned from the failures. Because I believe everything happens for a reason and in its own time, it is perhaps no accident that an innocuous collaboration would become the basis of my doctoral work. You once said that everyone comes to the lab with a story – indeed we do. To Dr. Jeremy Squire and Dr. Maria Zielenska, who started me on this journey, I am grateful for your unending support and friendship; and to have been part of your research family. Thank you for nurturing my desire for discovery and helping me find my way. This work could not have been accomplished without the expertise of my research families. Thank you to all the members of the Advanced Centre for Detection of Cancer (ACDC) laboratory for sharing with me your knowledge and your friendship. Moreover, I am grateful to all the members of the Squire and Zielenska laboratories for your many years of continued support and friendship. To my thesis committee members, Dr. Ivan Blasutig and Dr. David Irwin, I appreciate your contributions; and to Dr. Harry Elsholtz for your guidance. My special thanks to Dr. Sabine Mai for acting as my external examiner. I also thank my dear friends Ajay Pandita and Paula Marrano, who have been with me since the beginning. This journey could not have been completed without the support of my family. My love and thanks to my parents, who taught us that we should never stop learning; and to my sisters, Jen and Jamie, and their families. I also give my sincere thanks to my mother- and father-in-law and extended family. Finally, to my husband and my children, thank you for always being there for me – I love you with all my heart.
vi
Table of Contents
Dedication ...................................................................................................................................... iv
Acknowledgments ........................................................................................................................... v
Table of Contents ........................................................................................................................... vi
List of Abbreviations ..................................................................................................................... xi
List of Tables ............................................................................................................................. xviii
List of Figures .............................................................................................................................. xix
Chapter 1 ......................................................................................................................................... 1
1 Ovarian Carcinoma .................................................................................................................... 2
1.1 Canadian Statistics .............................................................................................................. 2
1.2 Etiology ............................................................................................................................... 2
1.3 Diagnosis and Treatment .................................................................................................... 3
1.4 Pathology ............................................................................................................................ 5
1.5 Molecular Pathology ........................................................................................................... 6
1.5.1 Hereditary and Sporadic Ovarian Cancers .............................................................. 6
1.5.2 Type I and Type II Ovarian Cancers ....................................................................... 7
1.5.3 Ovarian Cancer Genomics and Transcriptomics .................................................. 10
1.6 Chromosomal Instability ................................................................................................... 11
1.7 Defining Chromosomal Instability in Ovarian Carcinoma ............................................... 13
1.8 The Current Status of Ovarian Cancer Biomarkers: Kallikrein 6 (KLK6) a Promising Ovarian Cancer Biomarker ............................................................................................... 16
1.8.1 The Kallikreins (KLK) and Kallkrein 6 (KLK6): A Promising Biomarker for Ovarian Cancer ..................................................................................................... 18
1.8.2 Kallikrein 6: Historical Perspectives and Genetics ............................................... 19
1.8.3 KLK6 mRNA ........................................................................................................ 20
1.8.4 KLK6 Protein ........................................................................................................ 22
vii
1.8.5 Role of KLK6 in Disease ...................................................................................... 26
1.8.5.1 Inflammation, Immunity and Skin Pathophysiology .............................. 27
1.8.5.2 Neurodegeneration.................................................................................. 27
1.8.5.3 Tumourigenesis ...................................................................................... 28
1.8.6 Clinical Utility of KLK6 in Ovarian Cancer ......................................................... 30
1.8.7 Regulation of KLK6 Expression and Activity ...................................................... 32
1.9 Rationale and Hypothesis ................................................................................................. 36
1.10 Objectives ......................................................................................................................... 36
Chapter 2 ....................................................................................................................................... 38
2 Distinct Patterns of Structural and Numerical Chromosomal Instability Characterize Sporadic Ovarian Cancer ......................................................................................................... 39
2.1 Introduction ....................................................................................................................... 39
2.2 Materials and Methods ...................................................................................................... 40
2.2.1 Patient Specimens and Cell Lines ......................................................................... 40
2.2.2 Peptide Nucleic Acid (PNA) Fluorescence in situ Hybridization (FISH) and Spectral Karyotyping Analysis ............................................................................. 41
2.2.3 Centrosome Immunostaining ................................................................................ 42
2.3 Results ............................................................................................................................... 43
2.3.1 Numerical Chromosomal Instability: N-CIN ........................................................ 43
2.3.2 Structural Chromosomal Instability: S-CIN ......................................................... 44
2.3.3 Multicentric Chromosomes ................................................................................... 45
2.3.4 Centrosome Aberrations and Association with Increasing N-CIN ....................... 47
2.4 Discussion ......................................................................................................................... 48
Chapter 3 ....................................................................................................................................... 55
3 Impact of Cytogenetic and Genomic Aberrations of the Kallikrein Locus in Ovarian Cancer ...................................................................................................................................... 56
3.1 Introduction ....................................................................................................................... 56
viii
3.2 Materials and Methods ...................................................................................................... 57
3.2.1 Patient Specimens and Cell Lines ......................................................................... 57
3.2.2 Fluorescence in situ Hybridization (FISH) ........................................................... 58
3.2.3 Spectral Karyotyping ............................................................................................ 59
3.2.4 Array Comparative Genomic Hybridization (aCGH) ........................................... 59
3.2.5 Measurement of KLKs by ELISA ........................................................................ 60
3.3 Results ............................................................................................................................... 63
3.3.1 Fine Structural Analysis of Chromosomal Rearrangements Affecting 19q13 in Cancer Cell Lines .................................................................................................. 63
3.3.2 Chromosomal Rearrangements of 19q13 in Ovarian Cancer Patient Samples ..... 66
3.3.3 High-Resolution aCGH and KLK-specific FISH Analysis Confirms Copy-Number Imbalances .............................................................................................. 69
3.3.4 Protein Expression and Relationship to Copy-Number and Structural Rearrangements in Cancer Cell Lines and Ovarian Specimens ............................ 70
3.4 Discussion ......................................................................................................................... 73
Chapter 4 ....................................................................................................................................... 76
4 Genomic Instability and Copy-Number Heterogeneity of Chromosome 19q, Including the Kallikrein Locus, In Ovarian Carcinomas ............................................................................... 77
4.1 Introduction ....................................................................................................................... 77
4.2 Materials and Methods ...................................................................................................... 79
4.2.1 Patient Specimens ................................................................................................. 79
4.2.2 Fluorescence in situ Hybridization (FISH) ........................................................... 79
4.2.3 Measurement of Chromosomal Instability ............................................................ 80
4.2.4 Immunohistochemistry (IHC) ............................................................................... 81
4.2.5 Statistical Methods ................................................................................................ 82
4.3 Results ............................................................................................................................... 82
4.3.1 Patient Cohort ....................................................................................................... 82
ix
4.3.2 The KLK Locus (19q13.3/13.4) is Subject to Frequent Copy-Number Alterations and Chromosomal Instability ............................................................. 82
4.3.3 Structural Alterations of Chromosome 19q are Associated with Age and Grade ..................................................................................................................... 86
4.3.4 Serous Ovarian Carcinomas Show a Range of KLK6 and p53 Expression ......... 87
4.3.5 Univariate and Multivariate Analyses .................................................................. 87
4.4 Discussion ......................................................................................................................... 91
Chapter 5 ....................................................................................................................................... 98
5 Modulation of KLK6 Protein Expression in the Ovarian Cancer Cell Line OVCAR-3 by miRNA Let-7a .......................................................................................................................... 99
5.1 Introduction ....................................................................................................................... 99
5.2 Materials and Methods .................................................................................................... 100
5.2.1 Fluorescence in situ Hybridization (FISH), Spectral Karyotyping (SKY) and Array Comparative Genomic Hybridization (aCGH) ......................................... 100
5.2.2 microRNA Profiling ............................................................................................ 101
5.2.3 Identification of miRNAs Predicted to Target KLK6 ......................................... 102
5.2.4 Transient Transfection of miRNAs to OVCAR-3 .............................................. 102
5.2.5 Quantitative Reverse Transcription Polymerase Chain Reaction (qRT-PCR) ... 103
5.2.6 Measurement of KLK6 and KLK10 by ELISA .................................................. 103
5.3 Results ............................................................................................................................. 104
5.3.1 Identification of Predicted miRNAs Regulating KLK6 and miRNA Profiling .. 104
5.3.2 OVCAR-3 Cytogenomics ................................................................................... 109
5.4 Discussion ....................................................................................................................... 115
Chapter 6 ..................................................................................................................................... 120
6 Summary and Future Directions ............................................................................................ 121
6.1 Summary ......................................................................................................................... 121
6.2 Future Directions ............................................................................................................ 126
References ................................................................................................................................... 129
x
Summary of Publications ............................................................................................................ 156
xi
List of Abbreviations
5-aza 5-Aza-2′-deoxycytidine
aCGH Array Comparative Genomic Hybridization
ACTION Adjuvant Chemotherapy Trial in Ovarian Neoplasia
ACTN4 Alpha-actinin-4
AD Alzheimer's Disease
AKT v-akt murine thymoma viral oncogene homolog 1
AKT2 v-akt murine thymoma viral oncogene homolog 2
Ala Alanine
APC Adenomatous polyposis coli
Apo A-1 Apolipoprotein A-I
APP Amyloid precursor protein
ARID1A AT-rich interactive domain-containing protein 1A
Asp Aspartic acid
ATCC American Type Culture Collection
AUC Area Under the Curve
β2M β2 microglobulin
BAC Bacterial Artificial Chromosome
Bcl-2 B-cell lymphoma 2
Bcl-XL B-cell lymphoma-extra large
BFB Breakage Fusion Bridge
Bim BCL2-like 11
bp base pair
BRAF v-Raf murine sarcoma viral oncogene homolog B1
xii
BRCA1 Breast Cancer 1, early onset
BRCA2 Breast Cancer 2, early onset
CA-125 Carbohydrate Antigen 125
CCND2 Cyclin D2
CCNE1 Cyclin E1
CGH Comparative Genomic Hybridization
CIN Chromosomal Instability
CNS Central Nervous System
COSMIC Catalogue Of Somatic Mutations In Cancer
CpG Cytosine phosphate Guanine
CT Computed Tomography
CTNNB1 Beta-catenin
DAPI 4',6-diamidino-2-phenylindole
dATP 2'-deoxyadenosine triphosphate
dCTP 2'-deoxycytidine triphosphate
DEAC 7- diethylaminocoumarin-3-carboxylic acid
DFS Disease Free Survival
dGTP 2'-deoxyguanosine triphosphate
DNA Deoxyribonucleic Acid
DSB Double Strand Break
dTTP 2'-deoxythymidine triphosphate
EDRN Early Detection Research Network
EGFR Epidermal Growth Factor Receptor
ELF-2 E74-like factor 2 (ets domain transcription factor)
ELISA Enzyme-linked immunosorbent assay
xiii
ELK-1 E twenty-six (ETS)-like transcription factor 1
EMT Epithelial to Mesenchymal Transition
FDA Food and Drug Administration
FFPE Formalin Fixed Paraffin Embedded
FIGO International Federation of Gynecology and Obstetrics
FISH Fluorescence in situ Hybridization
FITC Fluorescein isothiocyanate
Glu Glutamic acid
hsa-let-7 human-lethal-7
HE4 Human Epididymis Protein 4
HER2/neu Human Epidermal Growth Factor Receptor
HGSC High-Grade Serous Carcinoma
His Histidine
HMGA2 High-mobility group AT-hook 2
HNPCC Hereditary nonpolyposis colorectal cancer
HR Hazard Ratio
HR Homologous Recombination
HRE Hormone Response Element
HSR Homogeneous Staining Region
ICGC International Cancer Genome Consortium
ICON1 International Collaborative Ovarian Neoplasm Trial 1
ID4 Inhibitor of DNA binding 4
IGF2 Insulin-like Growth Factor 2
IHC Immunohistochemistry
IRF2BP2 Interferon Regulatory Factor 2 Binding Protein 2
xiv
ISCN International System for Human Cytogenetic Nomenclature
KLK Kallikrein
KLK1 Kallikrein 1
KLK10 Kallikrein-related peptidase 10
KLK11 Kallikrein-related peptidase 11
KLK12 Kallikrein-related peptidase 12
KLK13 Kallikrein-related peptidase 13
KLK14 Kallikrein-related peptidase 14
KLK15 Kallikrein-related peptidase 15
KLK2 Kallikrein-related peptidase 2
KLK3 Kallikrein-related peptidase 3
KLK4 Kallikrein-related peptidase 4
KLK5 Kallikrein-related peptidase 5
KLK6 Kallikrein-related peptidase 6
KLK7 Kallikrein-related peptidase 7
KLK8 Kallikrein-related peptidase 8
KLK9 Kallikrein-related peptidase 9
KRAS v-Ki-ras2 Kirsten rat sarcoma viral oncogene homolog
Leu Leucine
LGSC Low-Grade Serous Cancer
Lys Lysine
mCGH metaphase Comparative Genomic Hybridization
MECOM MDS1 and EVI1 complex locus
MFISH Multi-Colour Fluorescence in situ Hybridization
MIN Microsatellite Instability
xv
miRNA micro Ribonucleic Acid
MLH1 mutL homolog 1, colon cancer, nonpolyposis type 2 (E. coli)
MMR Mismatch Repair
mRNA messenger Ribonucleic Acid
MS Multiple Sclerosis
MS Mass Spectrometry
MSH2 mutS homolog 2, colon cancer, nonpolyposis type 1 (E. coli)
MSH6 mutS homolog 6 (E. coli)
MSI Microsatellite Instability
MUC16 Mucin 16
MYC/MYCC v-myc myelocytomatosis viral oncogene homolog (avian)
NCBI National Center for Biotechnology Information
N-CIN Numerical Chromosomal Instability
NF1 Neurofibromin 1
NHL Normal Human Lymphocyte
NHR Non-Homologous Repair
OCa Ovarian Carcinoma
OS Overall Survival
OSE Ovarian Surface Epithelium
P1 Promoter 1
P2 Promoter 2
p21 cip Cyclin-dependent kinase inhibitor 1A (p21, Cip1)
PAX8 Paired box 8
PBS Phosphate Buffered Saline
PCR Polymerase Chain Reaction
xvi
PD Parkinson's Disease
PFS Progression Free Survival
PIK3CA Phosphatidylinositol-4,5-bisphosphate 3-kinase, catalytic subunit alpha
PMS2 PMS2 postmeiotic segregation increased 2 (S. cerevisiae)
PNA Peptide Nucleic Acid
PPP2R1A Protein Phosphatase 2, Regulatory subunit A, alpha
pre-miRNA Precursor micro Ribonucleic Acid
pri-miRNA Primary micro Ribonucleic Acid
PSA Prostate Specific Antigen/ Kallikrein-related peptidase 3
PTEN Phosphatase and Tensin homolog
qRT-PCR quantitative Real Time-Polymerase Chain Reaction
RAB40C RAB40C, member RAS oncogene family
RB1 Retinoblastoma 1
RGB Red Green Blue
RMI Risk of Malignancy Index
ROMA Risk of Malignancy Algorithm
RT Room Temperature
S-CIN Structural Chromosomal Instability
Ser Serine
SILAC Stable Isotope Labeling by Amino acids in Cell culture
siRNA small interfering Ribonucleic Acid
SKY Spectral Karyotyping
SNP Single Nucleotide Polymorphism
SPORE Specialized Program of Research Excellence
SSC Saline Sodium Citrate
xvii
STAT Signal Transducer and Activator of Transcription
STIC Serous Tubal Intraepithelial Carcinoma
TCGA The Cancer Genome Atlas
TERT telomerase reverse transcriptase
Tfr transferrin receptor 2
TIC Tubal Intraepithelial Cancer
TMA Tissue Microarray
TP53 Tumor Protein p53
TT transthyretin
UTR Untranslated Region
v-MYB v-Myb myeloblastosis viral oncogene homolog (avian)
ZMYND8 Zinc finger, MYND-type containing 8
xviii
List of Tables
Chapter 1
Chapter 2
Table 2.1 Summary of N-CIN Indices, S-CIN Events and Centrosomal Aberrations in Primary Ovarian Carcinomas
Table 2.2 Raw FISH Scoring Values
Table 2.3 Composite SKY Karyotypes of OCA27A/B and OCA714
Chapter 3
Table 3.1 KLK-Specific FISH: Analysis of Cell Lines
Table 3.2 ELISA findings for cancer cell lines and patient specimens
Chapter 4
Table 4.1 Clinical characteristics of patients stratified by KLK copy number:
Table 4.2 Clinical characteristics of patients stratified by 19q-rearrangement
Table 4.3 Hazard ratio (HR) estimated from the Cox regression model for disease-free survival (DFS).
Table 4.4 Hazard ratio (HR) estimated from the Cox regression model for overall survival (OS).
Chapter 5
Table 5.1 Expression values of miRNAs predicted to target KLK6 in OVCAR-3
Table 5.2 Expression values of miRNAs predicted to target KLK10 in OVCAR-3
Table 5.3 Summary of miRNA Expression and Copy-Number Changes of miRNAs Predicted to Target KLK6
Table 5.4 Summary of miRNA Expression and Copy-Number Changes of miRNAs Predicted to Target KLK10
xix
List of Figures
Chapter 1
Figure 1.1 Pathology of Ovarian Carcinoma
Figure 1.2 Numerical and Structural Chromosomal Aberrations in Ovarian Carcinoma
Figure 1.3 The Human Kallikrein Locus at Chromosome 19q13.3/q13.4.
Figure 1.4 Summary of KLK6 Properties
Figure 1.5 Consequences of Dysregulated KLK 6 Expression in Cancer and Neurodegenerative Diseases
Figure 1.6 Mechanisms of Regulation of KLK6 Expression
Chapter 2
Figure 2.1 Genomic Instability Findings in Primary Ovarian Cancers by Interphase FISH, SKY and Centrosomal Analysis.
Figure 2.2 Increasing Copy-Number Instability is Associated with Changes in Ploidy.
Figure 2.3 Interplay between N-CIN, S-CIN, Centrosomal Aberrations and Ploidy.
Chapter 3
Figure 3.1 Summary of FISH findings of the KLK locus in cancer cell lines and ovarian cancer patients
Figure 3.2. KLK status in cancer cell lines by sequential SKY and FISH.
Figure 3.4 aCGH and FISH validation of copy number imbalances of the KLK locus in BT-474.
Chapter 4
Figure 4.1 KLK Copy-Number Analysis of Serous Ovarian Carcinomas by FISH and Immunohistochemistry.
Figure 4.2 Kaplan Meier Curves for Disease Free Survival (DFS) and Overall Survival (OS
xx
Chapter 5
Figure 5.1 miRNA Profiling of OVCAR-3 and miRNAs Predicted to Target KLK6
Figure 5.2 Cytogenomic Analysis of OVCAR-3
Figure 5.3 Cytogenomic Analysis of hsa-let-7a Family Members in OVCAR-3
1
Chapter 1
Introduction
Chapter 1 includes an abridged version of the published manuscript “Review: The Physiology and Pathobiology of Human Kallikrein-Related Peptidase 6 (KLK6)” by Bayani J and Diamandis EP. Clin Chem Lab Med. 2011. Nov 3;50(2):211-33.
2
1 Ovarian Carcinoma
1.1 Canadian Statistics
Ovarian cancer (OCa) is the leading cause of death from a gynecological malignancy and the
fourth leading cause of cancer death among Canadian women. Based on 2011 statistics, there
will be an estimated 2,600 new cases and 1,750 estimated deaths in Canada due to OCa
(http://www.cancer.ca). Though comprising only 3% of new cancer cases and 4.9% of cancer
deaths, it remains a deadly disease with an estimated five year survival of only 42%. The
lifetime probability of Canadian women developing OCa is 1.5%, while the probability of dying
from OCa is 1.1%.
1.2 Etiology
Like many cancers, the exact causes are unclear; however it is widely accepted that a number of
factors contribute to the etiology of OCa. The majority of OCa seen in the clinic arises
sporadically, with approximately 5-10% thought to be hereditary. BRCA1 and BRCA2 gene
mutations are most commonly associated with familial ovarian and breast cancer, comprising the
majority of hereditary OCa (Sogaard et al., 2006). Women with germline mutations of BRCA1
or BRCA2 have a 30%-70% chance of developing OCa by age 70 (Risch et al., 2006). BRCA1
and BRCA2 are critical components of the homologous recombination DNA repair pathway
involved in double-stranded breaks (DSB) (Deng, 2006). However, OCa has also been linked to
the spectrum of cancers associated with hereditary nonpolyposis colorectal cancer syndrome
(HNPCC) (Lynch et al., 2009), and familial cancer syndromes such as Peutz-Jegher and Multiple
Endocrine Neoplasia (MEN) type I (Lynch et al., 2009; Papageorgiou and Stratakis, 2002).
Normal reproductive and hormonal factors also play a role in the etiology of the disease.
Briefly, early-aged menstruation and late-aged menopause have been linked with increase OCa
cancer risk, which is in keeping with the incessant ovulation hypothesis (Fathalla, 1971). It
postulates the extended wound-healing events at the ovarian surface epithelium may lead to
greater cancer risk (Cramer et al., 1983). Thus factors that reduce ovulatory events, such as
pregnancy, appears to confer a protective role, with reports suggesting the risk of OCa among
parous women could be decreased as much as 50% as compared to nulliparous women (Salvador
3
et al., 2009). Lactation and use of oral contraceptives may also provide some decreased risk,
though the benefit is generally modest (reviewed by Salvador et al., 2009). The role for
hormone replacement remains less consistent and more controversial (Hinds and Price, 2010),
with the hypothesis that replacement of estrogen could stimulate cancer cells. Indeed clinicians
and the patients must weigh the potential risks and benefits of hormone replacement therapies.
Due to the link between BRCA1/2 carriers and OCa, prophylactic oophorectomy among these
higher risk women greatly reduces the chances of developing the cancer and may include the
prophylactic removal of the fallopian tubes (Miller et al., 2010). Moreover, tubal ligation has
also been shown to reduce cancer risk (Cibula et al., 2011). The link between the fallopian tubes
and OCa has been further strengthened by the revelation of tubal intraepithelial carcinomas in
prophylactically removed fallopian tubes (discussed below). Finally environmental and dietary
factors continue to contribute to the etiology of the disease, with the observation of increased
incidences of OCa among industrialize countries. These observations suggest the exposure to
carcinogens and perhaps intake of more processed foods, animal products, coffee, alcohol and
tobacco among industrialized nations increases OCa risk (Benedet et al., 2000).
1.3 Diagnosis and Treatment
Unfortunately, there are no overt or specific symptoms associated with early OCa. However,
symptoms may include vague abdominal pain or discomfort, menstrual irregularities, dyspepsia
and other digestive problems. Later-staged disease may present with the accumulation of ascites
fluid leading to the distention of the abdomen. For sporadic cancers, early diagnosis continues to
be problematic; however hereditary cancers have the (ironic) benefit of family history that offers
prophylactic options. Indeed, prophylactic bilateral salpingo-oophorectomy among these higher
risk women greatly reduces the chances of developing the cancer (reviewed by Miller et al.,
2010; and Lynch et al., 2009). However, such an invasive procedure has profound physical and
emotional implications. Patient age, particularly between 40 and 69, together with any
suspicious symptoms, are cause of further investigation. The International Federation of
Gynaecologists and Obstetricians (FIGO) recommendations for the treatment for suspicious OCa
are described briefly here (Benedet et al., 2000). A detailed medical history, complete pelvic
exam, together with testing for tumour markers, mark the exploratory investigation, to be
followed by chest radiograms, pelvic ultrasound and computerized tomography (CT).
4
Laparoscopic surgery serves to stage the cancer and identify the histological subtype. Based on
the extent of disease dissemination, total hysterectomy and bilateral salpingo-oophorectomy can
be expected with the goal of maximal cytoreduction.
The inclusion of adjuvant treatment for early staged disease appears to be most beneficial for
patients with non-optimally staged early cancers based on the International Collaborative
Ovarian Neoplasm Trial 1 (ICON1) and the Adjuvant Chemotherapy Trial in Ovarian Neoplasia
(ACTION) (Trimbos et al., 2003). Typically such first-line chemotherapeutic agents for both
early and advanced staged disease include Carboplatin alone or in combination with Paxitaxel.
The anticancer activity of platinum-based therapies such as Carboplatin or Cisplatin lies in the
ability of these drugs to create links within DNA strands and inhibiting replication and inducing
apoptosis (Siddik, 2003). Paxitaxel, belonging to a family of taxanes, also serves to inhibit cell
proliferation, but achieves this goal by binding tubulin to prevent microtubule disassembly. This
prevents the completion of mitosis during M-phase (Orr et al., 2003). Recurrent diseases are
more aggressively treated with platinum drugs (Carboplatin or Cisplatin) and Paxitaxel, as well
as the addition of other DNA intercalating drugs such as liposomal Doxorubicin (Caelyx or
Doxil); the topoisomerase I inhibitors (Rasheed and Rubin, 2003) Topotecan (Hycamtin) and
Etoposide; or the nucleoside analog (Ewald et al., 2008) Gemcitabine. For patients that have no
clinical evidence of disease, a second-look laparotomy may be ordered to determine the response
to treatment. Moreover, patients may also undergo a secondary cytoreduction procedure
following the completion of first-line chemotherapy. Management of patients during follow-up
is typically every three months in the first few years after treatment, and three to two additional
times a year after the fifth year; and involves physical examinations, blood work and radiological
tests. The vast majority of patients will relapse, however the success of further treatment of
these patients appears to depend on the disease-free interval of the first line therapy. Patients
with a disease-free interval of six months or greater are considered platinum-sensitive, while
those with a shorter disease-free interval are considered platinum-refractory. Patients deemed
platinum-sensitive may benefit from further combined platinum treatment, whereas refractory
patients may enroll in clinical trials or consider non-platinum therapeutics (Ledermann et al.,
2011; Pfisterer and Ledermann, 2006; Pfisterer et al., 2006).
5
1.4 Pathology
In recent years, compelling evidence suggests that serous “ovarian” cancers may in fact originate
from the distal fallopian tubes (reviewed by Salvador et al., 2009). These claims have been
largely supported by the fact that the native histology of the fallopian tube is Müllerian in nature,
consistent with observation that OCas exhibit a Müllerian phenotype. This is in contrast to the
characteristics of the normal ovarian surface epithelium (OSE), which is continuous with the
mesothelium of the pelvic organs, derived from the coelomic epithelium, and are generally
poorly differentiated (Auersperg et al., 2001). Thus, there is debate as to the origins of epithelial
OCa (Salvador et al., 2009). The argument for the malignant transformation of the OSE, and
those lining inclusion cysts within the ovary, is consistent with reports identifying lesions only
confined to the ovary. Moreover, unlike other normal adult epithelia, OSEs can convert to either
mesenchymal or epithelial phenotypes in response to different stimuli (Auersperg et al., 2001).
However, until recently, investigators did not systematically examine the fallopian tubes; simply
assuming that any fallopian tube lesions constituted disseminated disease from the ovary. The
identification of tubal intraepithelial carcinomas (TIC) and serous tubal intraepithelial
carcinomas (STIC) (reviewed by Kurman and Shih Ie, 2011) in the fimbria of the fallopian tubes
from prophylactic oophorectomies of BRCA-positive women (Crum et al., 2007; Lee et al.,
2007), offers the possibility that exfoliated TIC/STIC cells implant themselves onto the ovarian
surface epithelium; or to the ovulation site which are then incorporated into cortical inclusion
cysts (Kurman and Shih Ie, 2011) (Figure 1.1A). Indeed, there is speculation that both
hypotheses are correct and may explain the differences between the molecular subtypes
(reviewed by Karst and Drapkin, 2010).
There are now five major histological subtypes to what is classified as epithelial OCa (Figure
1.1B): high-grade serous carcinomas, low-grade serous carcinomas, endometrioid, clear cell and
mucinous tumours (Gilks and Prat, 2009). High-grade serous tumours comprise the majority of
cancers at a frequency of approximately 45% and are described as ciliated columnar cells of
tubal phenotype. Low-grade serous tumours comprise the remaining 15% of the serous
histological subtype, and are molecularly distinct from high-grade serous cancers. Endometrioid
cancers comprise approximately 20% of cancers, are often highly malignant and characterized by
pseduostratified columnar cells with an endometrioid phenotype. These cellular features suggest
that these cancers may represent the malignant transformation of pelvic and ovarian
6
endometriotic lesions. Clear cell tumours, which are cytoplasmically vacuolated, are also
believed to originate from transformed endometriosis cells, and account for approximately 10%
of cases seen. Finally mucinous tumours, which are the rarest subtype seen at 3-5%, do not have
Müllerian features like the other subtypes; but may originate from transitional-type epithelium
located at the tubal-peritoneal junction (Kurman and Shih Ie, 2011).
As with other cancers, histological grade generally refers to the overall microscopic appearance
of the cells into well differentiated (G1), moderate (G2) and poorly differentiated (G3) (Benedet
et al., 2000). OCa staging reflects the surgical observations at the time of diagnosis, and for
OCa, the FIGO criteria has been adopted (Benedet et al., 2000). Generally, stage I disease is
limited to the ovaries, stage II cancers are limited to the pelvis, stage III cancers spread beyond
the pelvis, and stage IV disease represents distant metastasis to the chest, liver and other organs
1.5 Molecular Pathology
1.5.1 Hereditary and Sporadic Ovarian Cancers
Approximately 10% of all epithelial OCas are hereditary, with the vast majority of these due to
germline mutations in BRCA1 or BRCA2 (Lynch et al., 2009; Risch et al., 2006; Sogaard et al.,
2006). However, mutations in at least four mismatch repair (MMR) genes (MLH1, MSH2, MSH6
and PMS2), associated with Lynch syndrome account, for up to 15% of hereditary cancers
(reviewed by Lynch et al. 2009). Lynch et al., 2009, provides a comprehensive review regarding
hereditary OCa, and the salient points will be briefly summarized as follows. Numerous studies
have confirmed the increase risk of OCa among specific BRCA1 and BRCA2 mutation carriers,
not only among the general population (Lynch et al., 2009), but also among those in closed
populations such as the Ashkenazi Jews (King et al., 2003; Tobias et al., 2000), and Icelanders
(Gudmundsson et al., 1996; Mikaelsdottir et al., 2004; Rafnar et al., 2004). Lynch syndrome,
known more commonly as HNPCC, predisposes affected individuals to colon cancer and other
malignancies including OCa. The consensus of studies suggests that MMR genes play a role in
the pathogenesis of the specific histological subtypes, particularly non-serous subtypes. Though
genetic predisposition confers greater risk, several studies have noted the relatively improved
prognosis and response to standard chemotherapy among the hereditary OCas (Cass et al., 2003;
Lynch et al., 2009; Pal et al., 2007). Serous carcinomas comprise the majority of OCas seen
7
among BRCA1/2 patients (~80%), while Lynch syndrome patients predominantly present with
endometrioid, mixed-endometrioid, mixed-mucinous and mixed- clear cell phenotypes, with
serous tumour comprising 25% (Lynch et al., 2009). In addition to mutations of the BRCA1/2
and MMR genes, a number of other genes are associated with hereditary OCa, and include the
oncogenes EGFR, HER2-neu, MYCC, KRAS, and β-catenin (CTNNB1) and tumour suppressor
genes TP53 and PTEN. In the case of oncogenes, the observed overexpression of these genes
appears to be correlated to activating mutations, or gene amplification. Similarly, mutations,
deletions or epigenetic silencing of TP53 and PTEN results in loss of function.
Sporadic OCa, those cancers not associated with hereditary factors, account for the majority of
cases. The median age at diagnosis of sporadic cancers is 63, in contrast to inherited cancers at
age 54, however there is no difference between the average age of diagnosis between BRCA2
carriers and sporadics (Boyd et al., 2000). Unlike hereditary cancers, sporadic cancers are the
most frequently seen subtype in the clinic and, frequently present at late stage and high grade.
Mutations in several oncogenes include KRAS, BRAF, PIK3CA and CTNNB1 (Landen et al.,
2008). Mutations in TP53 are frequently seen in the high-grade serous cancers and endometrioid
cancers, though less in the other subtypes.
1.5.2 Type I and Type II Ovarian Cancers
Ovarian cancer is increasingly recognized as a heterogeneous disease, not only by histological
subtype and tissues of origin, but in the recognition that they possess distinctive
clinicopathological and molecular features (reviewed by Kurman and Shih Ie, 2011). Kurman
and Shih have proposed two molecular subtypes, irrespective of histological classification: Type
I and Type II OCas. Type I tumours are defined broadly as being low-grade and include low-
grade serous carcinomas (LGSC), low-grade endometrioid, clear cell and mucinous cancers.
They develop from their putative precursor lesions and typically are large masses that are
confined to the ovary (stage I). Moreover, these tumours are relatively indolent with a good
prognosis. Genomically, these tumours have greater stability, possessing somatic sequence
mutations in KRAS, BRAF, PTEN, PIK3CA, CTNNB1, ARID1A and PPP2R1A. Type I tumours
have been hypothesized to arise from the malignant transformation of the OSE (Karst and
Drapkin, 2010). Type II tumours, however, comprise the majority of cancers which are high-
grade; including high-grade serous carcinomas (HGSC), high-grade endometrioid carcinomas,
8
carcinosarcomas and undifferentiated sarcomas. Type II cancers are often present in the
advanced stages (II-IV), and are highly proliferative and aggressive and recently hypothesized to
arise from TIC/STIC lesions in the fallopian tube (Karst and Drapkin, 2010). The majority of
studies focus on HGSC; which comprise over 75% of the cases, are prototypical Type II
tumours, exhibit unstable genomes and frequently possess TP53 mutations. Moreover,
inactivation of the BRCA pathways, either by mutation or methylation occurs in up to 50% of
HGSC. This is in contrast to Type I cancers, which do not possess mutations in TP53, or have
perturbations of the BRCA pathway (Kurman and Shih Ie, 2011) (Figure 1.1C).
9
Figure 1.1. Pathology of Ovarian Carcinoma. A) Aside from the malignant transformation of the ovarian surface epithelium, OCa is hypothesized to originate from lesions in the distal fallopian tube and exfoliating onto the surface of the ovary or cortical inclusion cysts. Shown is a histological section showing STIC lesions within the fallopian tube positive for mutant p53, with regions of normal p53 expression. B) Five histological subtypes of ovarian carcinoma. Shown are representative histological sections illustrating the morphologic characteristics of the major histological subtypes of epithelial OCa. C) Molecular classification of Type I and Type II ovarian carcinomas. Molecular classification based on grade, stage, immunohistological testing, genomics and response the chemotherapeutics.
10
1.5.3 Ovarian Cancer Genomics and Transcriptomics
The technical advances of the past decade has resulted in the (continued) amassing of high-
throughput data; including, and not limited to, the sequencing of whole OCa genomes, the
expression profiling of these genomes and, increasingly, the proteomes of these genomes
(Cancer Genome Atlas Research Network, 2011; Gorringe and Campbell, 2009). A common
feature of OCas, are their abnormal karyotypes (Bayani et al., 2002; Shridhar et al., 2001; Taetle
et al., 1999a; Taetle et al., 1999b). Indeed, traditional cytogenetics (Taetle et al., 1999a; Taetle et
al., 1999b) have revealed the karyotypes of OCas to possess both numerical and structural
rearrangements, which were further elucidated with the advent of simultaneous whole
chromosomal Fluorescence in situ Hybridization (FISH)-painting methods such as Spectral
Karyotyping (SKY) (Macville et al., 1997; Schrock et al., 1997) or Multi-colour FISH (MFISH)
(Liehr et al., 2004). Moreover, Comparative Genomic Hybridization (CGH) to both metaphase
and to microarrays delineated a recurrent pattern of chromosomal gains, losses and
amplifications, which permitted the identification of putative oncogenes and tumour suppressor
genes (reviewed by Gorringe and Campbell, 2009). With the advent of higher-resolution Single
Nucleotide Polymorphic (SNP) arrays as well as lower-resolution array platforms, we now have
the ability to distinguish regions of copy-number alteration and allelic imbalances. This has
culminated in the recent release by The Cancer Genome Atlas Research Network (TCGA)
(Cancer Genome Atlas Research Network, 2011) of a catalog of molecular aberrations relevant
to OCa. In this ambitious endeavor, 489 clinically annotated stage II to stage IV high-grade
serous ovarian cancers and matched normal DNAs were analysed for DNA copy-number, DNA
promoter methylation, mRNA expression, microRNA (miRNA) expression, and whole genome
and exome DNA sequencing (in a subset of tumours). Not surprisingly, mutational analysis
confirmed previous findings that mutations in TP53 were frequent events (303 of 316), and that
BRCA1 and BRCA2 germline mutations occurred in 9% and 8% of cases respectively. Other
statistically significant mutated genes included RB1, NF1, FAT3, CSMD3, GABRA6 and CDK12;
while copy-number analysis largely confirmed the findings from previous CGH studies. Perhaps
the most interesting findings of this study were the results of the mRNA, miRNA and DNA
methylation analyses, with the revelation of at least four expression subtypes: immunoreactive,
differentiated, proliferative and mesenchymal. These transcriptional subtypes were largely
correlative to the hierarchical clustering described by Tothill and colleagues (Tothill et al., 2008),
11
who described a six-member classification based on the profiling of endometrioid and serous
cancers. A general concordance between DNA methylation and reduced expression was
observed, and these studies confirmed the previous observations that the BRCA1 promoter is
hypermethylated and silenced in approximately 11% of cancers. Correlation to the clinical data
showed no significant differences between the transcriptional subtypes and survival duration;
however a 193-gene transcriptional signature predictive for overall survival was defined. One
hundred and eight genes were correlated with poor survival and 85 were correlated with good
survival, which were then validated on an independent set of 255 samples.
Prior to the TCGA findings, interesting results from the transcriptional profiling a decade ago, by
Jazaeri et al. 2002, found the greatest difference between BRCA1 and BRCA2 cancers. Sixty one
clinically confirmed OCas, comprising sporadic and hereditary cancers, were analysed using
cDNA microarrays. Among the hereditary cancers, tumours segregated according to their BRCA
status. Interestingly, the sporadic cancers segregated as being either BRCA1-like or BRCA2-like,
suggesting that dysregulated BRCA pathways are critical to the pathogenesis of both hereditary
and sporadic OCa. These early observations, were largely confirmed by the TCGA (Cancer
Genome Atlas Research Network, 2011) and more recently by Yang et al. (Yang et al., 2011a),
showing the greatest transcriptional differences occur between BRCA1- and BRCA2-mutated
tumours.
1.6 Chromosomal Instability
The estimated rate of somatic cell mutation in humans is approximately 1.1×10−8 per base pair
per generation (Loeb and Christians, 1996; Roach et al., 2010). Therefore, genetic or genomic
instability refers to the overall process that increases this rate of mutation and more importantly,
results in a phenotypic change. Genomic instability is classically categorized into two main
forms: microsatellite instability (MIN) and chromosomal instability (CIN); with the CIN
phenotype predominantly seen in OCa. The prototypic example for defining genomic instability
comes from studies in colorectal cancer, where distinct genomic phenotypes are seen between
hereditary and sporadic cases (Vilar and Gruber, 2010). The MIN phenotype causes the
shortening or expansion of short tandem repeat units that make up microsatellite sequences
scattered throughout the genome. Such events are indicative of underlying MMR deficiencies
associated with hereditary colorectal cancer (Lynch and de la Chapelle, 2003). Patients affected
12
by HNPCC with germline mutations of a variety of MMR genes, previously discussed, possess
cancer genomes that are predominantly near-diploid. In contrast, sporadic cancers comprise the
majority of colorectal cancers which exhibit the CIN phenotype (Lengauer et al., 1997); that is,
genomes with gross chromosomal changes. Among these patients, adenomatous polyps arise
through the bi-allelic loss in function of the adenomatous polyposis coli (APC) gene. The
progression of polyps to greater dysplasia coincides with the acquisition of mutations to activate
oncogenes; or results in the loss of tumor suppressor genes (Fearon and Vogelstein, 1990). The
consequences of these acquired mutations are the perturbations in cell cycle checkpoints and cell
division, leading to the observed chromosomal changes. Just as genomic instability refers to a
rate of genomic change, CIN refers to the excessive rate of chromosomal change (Lengauer et
al., 1997). The term aneuploidy is often interchangeably used with CIN; however it is important
to mention that the presence of aneuploidy alone does not constitute CIN, since a genome may
be stably aneuploid as in the case individuals affected by Downs Syndrome. Therefore, the
extent of clonality; or lack thereof, defines the level of CIN. As such, a tumor genome
composed of many different DNA populations can be characterized as having greater CIN, than
one where the genomes of individual tumor cells are more similar.
Many studies that investigate CIN, focus primarily on whole chromosomal gains or losses, and
fail to recognize the importance of structural rearrangements as an additional measure of
chromosomal instability. Changes at the chromosomal level encompass both numerical CIN (N-
CIN) as well as structural CIN (S-CIN) (Bayani et al., 2007). Generally, N-CIN describes the
on-going gains or losses of whole chromosomes, whole chromosome arms, or changes in DNA
ploidy (ie. multiples of a haploid number – 2n/3n/4n etc). The mechanisms mediating individual
chromosomal gains/losses and ploidy changes are influenced by errors in chromosomal
segregation and spindle attachment as well as supernumerary centrosomes (reviewed by Holland
and Cleveland, 2009; and Gisselsson, 2011). In contrast, S-CIN refers to on-going
chromosomal breakage events, which include chromosomal additions, deletions, translocations
and other structural rearrangements. Like polyploidy, structural changes can affect gene copy
number, but may also facilitate the formation of abnormal gene fusion products. Changes in
chromosome structure are generally influenced by two major DNA repair mechanisms:
homologous recombination (HR) and non-homologous recombination (NHR) (reviewed by
Hastings et al. (Hastings et al., 2009)). Whereas HR describes the normal exchange or repair of
13
identical DNA sequences, NHR does not require the identical template for such exchange or
repair. Indeed many cancers show dysregulated DNA repair pathways such as those affected by
BRCA1 (Khanna and Jackson, 2001). However, such promiscuous DNA repair can also be
potentiated through various features associated with genomic sequence architecture including
GC- and AT-content (Abeysinghe et al., 2003; Chuzhanova et al., 2003), Alu motifs (Batzer and
Deininger, 2002), alpha-satellite DNA (Grady et al., 1992), and telomeric sequences (Gisselsson
et al., 2005; Mai, 2010; Murnane, 2006). High-level genomic amplifications, complex structural
changes and deletions have also been associated with the Breakage-Fusion-Bridge (BFB) cycle
(Jones, 2005), characterized by the self-perpetuating cycle involving chromatid breaks and
fusions, triggered by dicentric and ring chromosome rupture during anaphase resolution.
Excessive telomere shortening in cancers has been shown to lead to BFB cycles and, eventually
to generalized genome instability and leading to either to cell death or crisis (Mai, 2010). More
recently, another mechanism for the complex structural rearrangements seen in cancers has been
proposed – chromothripsis (Stephens et al., 2011). Literally translated to mean “chromosome
shattering”, this rare event is believed to result in the shuffling of the genome in one cataclysmic
event whereby the DNA becomes fragmented then reassembled. This event may explain the
pattern of single copy losses, and unique fusions in transcriptional orientations that cannot be
adequately explained based on current DNA repair mechanisms. Though still controversial, deep
sequencing confirms that such a reshuffling event occurs, and is more compelling in tumours
whose genomes remain diploid. Nevertheless, it is evident that through various mechanisms,
genomic regions can experience local copy-number changes (ie. focal amplifications, gains or
deletions) in the absence of ploidy changes. The non-clonality of such changes whether N-CIN
or S-CIN, may also be applicable to specific genes or chromosomal loci. Indeed, these issues are
discussed by Bayani et al., 2007, whereby a distinction between structural chromosomal
instability, S-CIN; and numerical instability N-CIN was proposed.
1.7 Defining Chromosomal Instability in Ovarian Carcinoma
Karyotypic abnormalities are a common feature among epithelial cancers including OCa (Bayani
et al., 2002; Shridhar et al., 2001; Taetle et al., 1999a; Taetle et al., 1999b; Thompson et al.,
1994a; Thompson et al., 1994b; Thompson et al., 1996). The aCGH findings of the TCGA study
(Cancer Genome Atlas Research Network, 2011), not surprisingly, failed to reveal new insight
14
with respect to the pattern of recurrent copy-number changes among OCas. Both classical and
molecular cytogenetic methods have identified the recurrent net gains/amplifications of 1q, 3q 6p
7q, 8q, 12p, 20p and 20q; and net losses of 4p, 4q, 5q, 6q, 8p, 9p, 9q, 11p, 11q ,13q, 14q, 15q,
16p, 16q, 17p, 17q, 18p, 18q, 19p, 19q and 22q (Bayani et al., 2002; Cancer Genome Atlas
Research Network, 2011; Shridhar et al., 2001; Taetle et al., 1999a; Taetle et al., 1999b). The
TCGA studies identified focal amplifications of CCNE1, MYC, MECOM, ZMYND8, IRF2BP2,
ID4, PAX8, TERT and deletions of PTEN, RB1 and NF1. Advances in molecular cytogenetics
(reviewed by Bayani et al., 2007) have helped our understanding of the chromosomal
contributions to carcinogenesis. aCGH has proven to be invaluable for identifying genomic loci
that may contain putative oncogenes or tumour suppressor genes (reviewed by Gorringe and
Campbell, 2009) (Figure 1.2A). However, the major limitation of this technique is that it cannot
accurately reflect the level of chromosomal or genomic heterogeneity that exists. Thus,
information regarding the range of copy-number for a given locus or gene is lost due to the
diluting effects of dominant clones or by high rates of heterogeneity (Bayani and Squire, 2007).
Moreover, the inability of aCGH to accurately assess ploidy also limits our understanding of the
mechanisms that influence the observed karyotypic changes (Albertson et al., 2003; Tan et al.,
2007). It has been previously demonstrated by Bayani et al. 2002, that the karyotypes of OCas
are both structurally and numerically abnormal (Figure 1.2B). Despite the comprehensive
karyotypic analyses of OCas (Bayani et al., 2002; Shridhar et al., 2001; Taetle et al., 1999a;
Taetle et al., 1999b; Thompson et al., 1994a; Thompson et al., 1994b; Thompson et al., 1996),
there have been no studies that distinguish between the CIN of these two forms of
chromosomal change in OCa, and the implications for copy-number change in those gene
affected by such CIN.
15
Figure 1. 2 Numerical and Structural Chromosomal Aberrations in Ovarian Carcinoma.A) Pattern of chromosomal gains (green) and losses (red) across 711 OCas as determined by the Progenetix CGH database (http://www.progenetix.net) , shows non-random whole and partial chromosomal imbalances. B) Spectral karyotyping (SKY) analysis (Bayani et al., 2002), of a metaphase spread from an untreated primary OCa shows both numerical and complex structural chromosomal aberrations. Shown from left to right are the fluorescent images, where each different chromosome is represented by its raw fluorescence; inverted DAPI image; and example of numerical and structural aberrations involving chromosome 3. In this example, six chromosomes 3 were identified: two normal chromosomes 3 and four involved in duplicate structural rearrangements involving chromosomes 2 and 8. Also shown below is the karytoypic description.
16
1.8 The Current Status of Ovarian Cancer Biomarkers: Kallikrein
6 (KLK6) a Promising Ovarian Cancer Biomarker
Undoubtedly, the success of any treatment is dependent on the early detection of the cancer
before the disease can progress and metastasize. The current status of OCa biomarkers is
promising, but still largely inadequate, as there are no biomarkers for detection or screening
(Bast et al., 2009). Therefore, it is imperative to discover cancer biomarkers that are informative
across various clinical parameters. Though the scientific literature regarding putative biomarkers
is voluminous, the past two years has seen the U.S Food and Drug Administration (FDA)
approve very few additional OCa biomarkers to CA-125. These new additions include HE4, the
biomarker panel OVA1™ (Vermillion Inc.), which includes Transthyretin (TT or prealbumin),
Apolipoprotein A-1 (Apo A-1), Beta2-Microglobulin (Beta2M), Transferrin (Tfr) and CA-125
which is integrated with a proprietary algorithm; and the Risk of Ovarian Malignancy Algorithm
(ROMA™) (Fujirebio Diagnostics) which includes CA-125 and HE4 (Moore et al., 2009).
Non-FDA-approved biomarker panels and algorithms have also made their way into clinical
settings and met with promising results, including the Risk of Malignancy Index (RMI) which
improves specificity by combining CA-125 together with menopausal status and imaging scores
(Jacobs et al., 1990); while others, including OvaCheck® (Correlogic Systems Inc.), which
combines CA-125 with seven other markers; and OvaSure™ (LabCorp) (Visintin et al., 2008),
which also includes CA-125 along with leptin, prolactin, osteopontin, IGF2, and macrophage
inhibitory factor, have been met with either continued controversy or have been subsequently
discontinued.
The gold standard serum biomarker for OCa continues to be the carbohydrate antigen, CA-125
(MUC16) a member of the Mucin family of glycoproteins (Kufe, 2009); which is routinely used
to monitor the relative antigen levels following surgery and treatment. This high-molecular
weight transmembrane glycoprotein is expressed by the coelomic- and Müllerian-derived
epithelium and is expressed in cells of the fallopian tube, endometrium and endocervix (Jacobs
and Bast, 1989); while not expressed by normal OSE (Kabawat et al., 1983). Indeed, CA-125,
has no prognostic significance when measured at the time of diagnosis, but appears to have a
high correlation with survival when measured one month after the third course of chemotherapy
for patients with stage III or stage IV disease. As such, CA-125 is valuable in the follow-up and
17
restaging of patients who have elevated CA-125 levels from the time of diagnosis (Hogberg and
Kagedal, 1992; Hoskins, 1993; Mogensen, 1992). However, while elevated CA-125 levels
indicate a high probability of epithelial ovarian cancer, negative CA-125 levels cannot be used to
exclude the presence of residual disease (Makar et al., 1992). Moreover, CA-125 levels can also
be elevated in other malignancies or benign gynecologic problems such as endometriosis (Bast et
al., 1998) and poses a problem for those 20% of ovarian cancers that do not exhibit elevated CA-
125 levels. HE4 (human epididymis protein 4), a low molecular weight glycoprotein, was
discovered by transcriptional profiling to be elevated and expressed in 90% of OCa (Hellstrom et
al., 2003). Expressed by the epididymis in the male reproductive tract, HE4 is also secreted by
the Müllerian epithelia of the female reproductive tract, but not by normal OSE (Drapkin et al.,
2005). To date, HE4 has shown greater sensitivity and specificity than CA-125 for detecting
early-staged OCas, as HE4 tends not to be highly associated with benign lesions, making it a
better marker for malignancy (Hellstrom et al., 2003; Moore et al., 2009). Several studies have
shown that the combined use of CA-125 and HE4 (Montagnana et al., 2011; Moore et al., 2009)
can increase the accuracy of OCa diagnosis and discriminate between OCa and ovarian
endometriotic cysts (Drapkin et al., 2005), though much larger-scaled studies are required to
determine whether this combination of markers is suitable for the screening of asymptomatic
women. The OVA1™ panel utilizes the observed decrease of TT, Tfr and Apo A-1, in addition
to the observed increase in and Beta2M; together with CA-125 into an ovarian malignancy risk
index score. This non-screening serum test however, was approved for use only in women for
whom surgery was already scheduled for an ovarian mass, and for whom consultation with an
oncologist has not yet been ordered, or for which independent radiological assessment fails to
indicate malignancy. Finally ROMA™ (Fujirebio Diagnostics), integrates HE4 and CA-125
with menopausal status, in addition to clinical and radiologic evaluation; into a numerical score
intended to aid in assessing whether premenopausal or postmenopausal women presenting with
an ovarian adnexal mass has a high or low likelihood of malignancy. While promising,
prognostic and predictive biomarkers remain elusive and likely requires the integration of serum,
genomic and transcriptomic markers.
18
1.8.1 The Kallikreins (KLK) and Kallkrein 6 (KLK6): A Promising Biomarker for Ovarian Cancer
The KLKs, which include the tissue kallikrein gene (KLK1) and kallikrein-related peptidase
genes (KLK2-KLK15), encode for secreted serine proteases with trypsin- or chymotrypsin-like
activities (Diamandis et al., 2000a; Yousef et al., 2000). In normal physiology, the KLK genes
are expressed in various tissues (Shaw and Diamandis, 2007) and are involved in proteolytic
cascades (Eissa et al., 2010; Emami and Diamandis, 2008; Kishibe et al., 2007; Komatsu et al.,
2003; Michael et al., 2006). The KLK proteins also cleave a number of substrates (reviewed by
Borgono and Diamandis, 2004). More recently, the KLKs have also been implicated in
tumourigenic events (Mo et al., 2010; Pampalakis et al., 2009) and postulated to play a role in
the pathogenesis of Alzheimer’s disease (AD) (Ashby et al., 2010; Yousef et al., 2003a). It is
apparent, therefore, that KLKs play diverse roles in physiology and pathobiology. The 15
protein-encoding gene members share common features (Harvey et al., 2000; Yousef et al.,
2000). In addition to maintaining similar exon sizes, the catalytic triad (His, Asp and Ser) is
conserved (Schmidt et al., 2004; Sotiropoulou et al., 2009). Each gene encodes a single-chain
prepro-enzyme, which is subsequently processed to an enzymatically inactive proKLK and later
cleaved into its active form (Sotiropoulou et al., 2009). The entire family maps to an
approximately 350kb region at chromosome 19q13.3-13.4 (Harvey et al., 2000; Yousef et al.,
2000) (Figure 1.3) in humans. In addition to the 15 protein-encoding members, there is at least
one KLK pseudogene located between KLKs 2 and 4 (Gan et al., 2000; Yousef et al., 2004a).
One of the most distinguishing characteristics of the human KLK gene family is the fact that all
members localize to the same chromosomal region in a contiguous manner, unlike other trypsin
and trypsin-like proteases, whose gene-family members map to other genomic loci (Pavlopoulou
et al., 2010; Puente and Lopez-Otin, 2004).
KLK expression can be generally described as restrictive (Borgono and Diamandis, 2004;
Clements, 2008; Clements et al., 2004; Yousef and Diamandis, 2001). In the human KLK
cluster, a specific preponderance for prostate-specific expression is seen among KLK15, KLK3
(PSA), KLK2 and KLK4 (Clements et al., 2004; Shaw and Diamandis, 2007; Yousef and
Diamandis, 2001). Diverse mechanisms influencing gene or protein expression have been
investigated including methylation (Cho et al., 2009; Huang et al., 2007; Li et al., 2001; Lu et al.,
2009; Pampalakis and Sotiropoulou, 2006; Sidiropoulos et al., 2005). However, the vast majority
19
of work has shown the influence of hormones on KLK expression (Borgono and Diamandis,
2004; Paliouras and Diamandis, 2007; Zarghami et al., 1997). The advent of high-throughput
sequencing has also enabled the identification of single nucleotide polymorphisms (SNPs) for
each family member (Goard et al., 2007). Post-transcriptionally, microRNAs (miRNAs)
(Barbarotto et al., 2008; Lee et al., 1993) have been shown to be important players in the
regulation of protein synthesis (Chow et al., 2008; Jung et al., 2010; Kojima et al., 2010; Sehic et
al., 2010; White et al., 2010a; White et al., 2010b; Wissmann et al., 2007; Yousef, 2008). Post-
translationally, activated KLK proteins are controlled by endogenous inhibitors, and may also
lose activity through internal cleavage (Borgono and Diamandis, 2004). Amazingly, despite the
similarities between the family members, they are sufficiently unique to encompass the gamut of
normal and abnormal biological processes.
Figure 1.3. The Human Kallikrein Locus at Chromosome 19q13.3/q13.4. Shown is the organization and relative phylogenetic relationship of the human KLK locus, the transcriptional orientations and relative relationship (not drawn to scale) of the KLK gene family members based on the studies by Harvey et al., 2000; Pavlopoulou et al., 2010; and Yousef et al., 2001. Members with similar colours show more relatedness.
1.8.2 Kallikrein 6: Historical Perspectives and Genetics
Kallikrein 6 (KLK6), the HUGO Gene Nomenclature Committee (HGNC) Database – approved
name and symbol, has previously been identified by a number of aliases, including PRSS9, Bssp,
Neurosin, serine protease 9, Protease M, serine protease 19, PRSS18, SP59, and ZYME
(http://www.genenames.org/index.html). Although the classical KLKs (KLK1, 2 and 3), were
already known, KLK6 was the first of the expanded members to be identified. It was originally
20
recognized as Protease M, through a cDNA screen of primary and metastatic breast cancer cell
lines (Anisowicz et al., 1996).
With seven exons and six introns, KLK6 spans approximately 11.0 kb (NCBI Reference
Sequence: NG_011825.1), mapping between KLK5 and KLK7 (Harvey et al., 2000; Yousef et
al., 2004b; Yousef and Diamandis, 2001; Yousef et al., 1999) (Figure 1.3 and Figure 1.4A). The
first two exons are untranslated, with the remaining five exons and position of the catalytic triad
consistent with the other family members. No mutations in the coding exons of KLK6 have
been reported within the Sanger’s Catalogue of Somatic Mutations in Cancer (COSMIC)
http://www.sanger.ac.uk/perl/genetics/CGP/cosmic), International Cancer Genome Consortium
(http://www.icgc.org/), nor by sequencing of the coding regions of primary cancers and controls
(Shan et al., 2007). Likely evolving from an ancestral trypsin-like-KLK gene, KLK6, together
with KLK14, KLK5, KLK10 and KLK15, represents one of the earliest KLKs to come on scene
(Pavlopoulou et al., 2010).
Defined as common variations at a single nucleotide position, where the least common allele is
present in at least 1% of a given population (Collins et al., 1998), SNPs occur approximately one
in every 150-300 bp in the human genome (Brookes, 1999; Kruglyak and Nickerson, 2001;
Ladiges et al., 2004). Within the locus previously analysed by Goard et al. 2007 and Yousef et
al., 2004, over 110 SNPs on the positive and negative strands now span the gene (Figure 1.4A),
and continue to change as more genomic sequences become available. Six SNPs (of the positive
strand) have been localized to coding or splice sites and include: rs77760094; rs111672933;
rs1701950; rs113724718; rs61469141; rs111738447. Two synonymous SNPs are found in exon
6, resulting in no change in the amino acid. However, the four remaining SNPs have a missense
functional class, resulting in an amino acid change, whose significance requires investigation.
For many of these SNPs as well as those detected throughout the gene, validation of the
frequency has still yet to be determined.
1.8.3 KLK6 mRNA
KLK6 possesses a number of alternative transcripts encoding for the full-length KLK6 protein
(Kurlender et al., 2005; Yousef et al., 2004b), in addition to those predicted to produce truncated
forms (http://www.ncbi.nlm.nih.gov/genbank/) (Figure 1.4B). The KLK6 “classical transcript”
(also called Isoform A; NM_002774.3; GenBank Accession U62801) was originally detected in
21
a normal mammary myoepithelial cell strain (Anisowicz et al., 1996), then from the human colon
adenocarcinoma cell line (COLO 201) (Yamashiro et al., 1997), and later from Alzheimer's
disease brain tissues (Little et al., 1997). The longest reported classic transcript cDNA clone
spans 1,526 nucleotides (Anisowicz et al., 1996), yielding a predicted 244 amino acid protein
(Figure 1.4B and C).
The classical transcript utilizes the promoter (P1) located in exon 1, while transcriptional variants
(AY318867 and AY318869) utilize a second promoter (P2) starting in a 20 bp region within the
first intron of KLK6 (Pampalakis et al., 2004), thus lacking exon 1 (Figure 1.4B). A third
promoter (P3), has also been identified in intron 2 and contains no untranslated exons
(Christophi et al., 2004; Pampalakis et al., 2004). Variant transcript 1 (AY318869) has a size of
1517 bp, while transcript variant 2 (AY318867) has a size of 1503 bp (Christophi et al., 2004;
Pampalakis et al., 2004). The ability to generate multiple mRNAs from a given gene increases
protein diversity, and is a reflection of the inherent biological complexities and specificities of
different cell types (Landry et al., 2003). The significance of these KLK6 variant transcripts has
yet to be determined (Tan et al., 2006), however it is evident that there is the propensity for
some tissues to express more than one transcript (Christophi et al., 2004; Kurlender et al., 2005;
Pampalakis et al., 2004; Yousef et al., 2004b). Pampalakis et al. (Pampalakis et al., 2004),
reported the presence of transcript heterogeneity; with tissues of the CNS overwhelmingly
expressing the classical transcript over the variants AY318869, AY318867 and AY279383. In
contrast, breast, spleen, uterus, and skeletal muscle predominantly expressed the variants
AY318869 and AY318867, with either very little or non-detectable levels of the classical
transcript. It has also been shown that primary ovarian carcinomas exhibit transcript
heterogeneity (Shan et al., 2007). In transcript-specific PCR, ovarian cancers were shown to
express the classic transcript, alternative transcript (AY318869), and in a small number of cases,
the splice variant for AY279383. The alternative transcript, AY318869, comprised the major
species of ovarian cancer-derived KLK6 mRNA (Shan et al., 2007). With the exception of these
studies, the majority of expression analyses do not specify the KLK6 transcript being detected.
Therefore, the significance of variant transcripts has not been fully appreciated.
RNA expression of KLK6 in normal human tissues has been confirmed (Figure 1.4B), with
tissues of the central nervous system (CNS) unequivocally showing the highest expression of
RNA transcripts, followed closely by tissues of the breast, kidney and uterus; salivary gland,
22
thymus, spleen, thyroid and testis (Anisowicz et al., 1996; Christophi et al., 2004; Harvey et al.,
2000; Komatsu et al., 2007c; Little et al., 1997; Pampalakis et al., 2004; Shaw and Diamandis,
2007; Yamashiro et al., 1997; Yousef et al., 2004b; Yousef et al., 1999; Yousef et al., 2003b).
Cells of the ovary, lung, liver, bone marrow and pituitary show comparatively the least amount
of KLK6 mRNA. In diseased states, increased KLK6 mRNA expression has been identified in
ovarian carcinomas (Anisowicz et al., 1996; Ni et al., 2004; Rosen et al., 2005; Shan et al., 2007;
Shaw and Diamandis, 2007 ; Tanimoto et al., 2001; White et al., 2009a; White et al., 2009b),
breast (Anisowicz et al., 1996; Wang et al., 2008), uterine (Santin et al., 2005), pancreatic
(Ruckert et al., 2008), colorectal (Henkhaus et al., 2008; Kim et al., 2011; Ogawa et al., 2005),
gastric (Nagahara et al., 2005), skin (Klucky et al., 2007), and urinary bladder cancers (Shinoda
et al., 2007). In contrast to various cancers, the decrease of KLK6 mRNA has been identified in
the brain tissues of patients with AD and PD (Ashby et al., 2010; Ogawa et al., 2000). The
physiological implications of these findings will be discussed later.
1.8.4 KLK6 Protein
KLK6 encodes a single-chain enzymatically inactive prepro-enzyme of 244 amino acids
(Anisowicz et al., 1996; Little et al., 1997; Yamashiro et al., 1997; Yousef et al., 1999)(Figure
1.4C). This prepro-enzyme is subsequently processed to an inactive proKLK through the
removal of the predicted 16 amino acid signal peptide (Ala (A) ↓ Glu (E)) and passed through
the endoplasmic reticulum. The proKLK is secreted to the extracellular space and becomes
converted into its activated mature peptidase by the proteolytic cleavage of the five amino acid,
(Lys (K) ↓ Leu (L)), activation peptide (Debela et al., 2008; Yousef et al., 1999). The catalytic
triad, which characterizes serine proteases, is conserved in KLK6 (ie. His62, Asp106 and Ser197)
(Yousef et al., 1999). Furthermore, the presence of Asp at position 191 suggests that KLK6 has
trypsin-like activity. The activation of KLK6 is likely to involve proteinases endowed with
trypsin-like properties, thus the KLKs may be able to activate themselves or each other (Debela
et al., 2008; Yousef et al., 1999). Critical to understanding the physiological role of KLK6 lies in
discovering its substrates. Magklara et al. (Magklara et al., 2003) demonstrated the degradation
of casein, collagen type I, collagen type IV, fibrinogen as well as synthetic peptides
corresponding to the N-terminal region of amyloid precursor protein (APP) by KLK6; while the
chymotrypsin substrate AAPF-AMC was not cleaved; again confirming the lack of
chymotrypsin-like activity for KLK6 (Ghosh et al., 2004).
23
Interrogation of adult and fetal tissues, and various biological fluids by Shaw et al., 2007
reported that the highest expression of KLK6 is found in the cell lysate of adult brain and in the
CSF by ELISA. Adult spinal cord showed the second greatest KLK6 expression, at roughly 50%
that of the brain. All other adult tissues, possessed significantly lower concentrations (<10% of
the brain). In contrast, fetal stomach showed the greatest expression of KLK6 among all fetal
tissues tested, but still less than 50% of the adult brain; followed by fetal kidney and skin.
Though not tested by Shaw et al., 2007; KLK6 has not been detected in fetal brain (Scarisbrick
et al., 2001). By immunohistochemistry (IHC) (Petraki et al., 2001), tissues of the central and
peripheral nervous system showed a range of immunoreactivity; with intense staining in the
peripheral nerves, but weak nerve cell staining within the CNS; strong staining in the epithelium
of the choroid plexus; and moderate expression among Purkinje, stellate and glial cells. Others
have similarly reported the lack or barely detectable levels of KLK6 in the normal epithelium of
the ovary (Ni et al., 2004; Petraki et al., 2003; Petraki et al., 2006). For most normal tissues and
biological fluids analysed, there is general concordance between RNA and protein levels.
However, due to the issues of transcript heterogeneity, it is unclear whether the mRNA detected
represents translatable protein. Nevertheless bulk extracted brain tissue and CNS fluids maintain
the highest KLK6 mRNA and protein expression in the adult; while most other tissues that
expressed KLK6 mRNA, also showed some degree of protein expression. In OCas, increases in
KLK6 protein levels have been identified, not only in tumour cells (Bayani et al., 2011; Bayani
et al., 2008b; Diamandis et al., 2000b; Diamandis et al., 2000c; Hoffman et al., 2002; Koh et al.,
2011; Kountourakis et al., 2008; Luo et al., 2006; Ni et al., 2004; Petraki et al., 2006; Prezas et
al., 2006; Shan et al., 2007; Yousef et al., 1999; Yousef et al., 2003b; Zheng et al., 2007), but
also in the tumour stroma (Seiz et al., 2012); as well as in renal cancers (Gabril et al., 2010),
gliomas (Strojnik et al., 2009), lung carcinomas (Nathalie et al., 2009; Singh et al., 2008),
salivary gland tumours (Darling et al., 2006), pancreatic ductal carcinomas (Ruckert et al., 2008)
and uterine cancers (Santin et al., 2005). In the case of Alzheimer’s and Parkinson’s Disease,
decreased KLK6 has been observed (Ashby et al., 2010; Menendez-Gonzalez et al., 2008;
Scarisbrick, 2008; Zarghooni et al., 2002); while skin disorders such as psoriasis have also
reported changes in KLK6 levels (Komatsu et al., 2007a; Komatsu et al., 2007b).
24
Figure 1.4. Summary of KLK6 Properties A) Genomic organization of KLK6. Shown are the seven exons (I-VII) with untranslated exons shaded in grey, and coding exons in green. The sizes of each exon and intron are indicated. The distribution of both simple tandem-, di- and trinucleotide repeats are also shown below. Relative SNP distribution across the gene is shown. Green SNPs indicate those identified as missense or synonymous, near splice sites; while Red SNPs indicate those located in coding exons. Blue SNPs indicate those located in non-coding regions and introns. B) Organization of KLK6 RNA transcripts. Shown are a number of identified RNA transcripts as described in the text. The asterisk(*) indicates the transcriptional start site and the downward arrowhead indicates the stop site. Predicted untranslated exons are shaded in grey, predicted coding exons are shown in green. Right) Shown is the relative expression of KLK6 RNA in normal adult human tissues based on several studies Human adult brain shows the greatest KLK6 transcript expression among all tissues in the human body as compared to ovary, lung and liver which barely shows any expression or expression at the level of detection for PCR or northern analysis. C) Organization of KLK6 protein. Shown is the amino acid sequence of the full-length KLK transcript. Amino acids shaded in: Blue indicates the signal peptide sequence; Green indicates the propeptide sequence; Fusia indicates the autolysis loop; and amino acids coloured in Red
25
are those in the catalytic triad. Lower Left) Summary of other protein features including disulfide bonds (red), substrate binding sites (blue) and a glycosylation site (green). The amino acids of the catalytic triad are indicated in triangles. Lower Right) Shown is the relative expression of KLK6 protein detected in both normal adult and fetal tissues as detected by ELISA. Similar to RNA findings, normal adult brain shows the highest KLK6 protein expression. The majority of tissues show low to non-detectable levels when compared to the adult brain. In fetal tissues, stomach shows the greatest expression followed by kidney and skin. Bottom) KLK6-specific immunohistochemical analyses of normal ovarian surface epithelium and ovarian carcinoma. Normal ovarian surface epithelium shows very low/barely detectable levels of KLK6, in contrast to a representative ovarian cancer showing strong immunoreactivity to KLK6 antibodies.
26
1.8.5 Role of KLK6 in Disease
The expression of the KLK genes in a variety of tissues (Shaw and Diamandis, 2007), and their
proteolytic abilities enable them to participate in a variety of physiological and pathobiological
processes (Borgono and Diamandis, 2004; Sotiropoulou et al., 2009). The observed
overexpression or underexpression of KLKs, have helped to elucidate their role in these cellular
processes when the normal balance of their expression is shifted. In the following sections, we
present the experimental evidence linking KLK6 to numerous cellular processes and consider the
integrative effect this could have in the context of diseases such as cancer or neurodegeneration,
which are summarized in Figure 1.5.
Figure 1.5. Consequences of Dysregulated KLK 6 Expression in Cancer and Neurodegenerative Diseases. Experimental evidence discussed in the text implicates KLK6 in many diverse molecular pathways. In cancers, the increase of KLK expression can be oncogenic, leading to increases in inflammation, enhanced signaling and evasion of apoptosis; which can drive cancer progression to enhance pathways involved with cellular proliferation, invasion and metastasis. However, studies in breast cancer indicate a tumour suppressor role for KLK6. Loss of KLK6 expression in breast, and possibly renal cancers, results in the loss of this tumour suppressive function, leading to cancer progression. For multiple sclerosis, experimental evidence suggests that increased KLK6 levels are associated with inflammation, which can also enhance signaling and evasion of apoptosis for inflammatory cells with sustained signaling. Loss of KLK6 in the case of AD and PD leads to a shift in the normal turnover of neural proteins, leading to increased accumulation of plaque-forming fragments.
27
1.8.5.1 Inflammation, Immunity and Skin Pathophysiology
A number of KLK members have been implicated in the normal physiological functioning of the
skin which is tightly regulated for innate immunity and inflammation (reviewed by Sotiropoulou
and Pampalakis, 2010). A large body of work has shown the role of KLKs in skin desquamation,
thus forming the basis for their study in various skin disorders (Borgono et al., 2007; Komatsu et
al., 2005; Komatsu et al., 2003). Indeed, elevated levels of KLK6, KLK8 and KLK13 have been
identified among psoriasis and atopic dermatitis patients (Kishibe et al., 2007; Komatsu et al.,
2007b; Komatsu et al., 2005; Naldi, 2010). Work by Scarisbrick et al. (Scarisbrick, 2008;
Scarisbrick et al., 2006; Scarisbrick et al., 2008), suggested that KLK6 drives the dysregulated
inflammatory processes in the CNS and can alter immune cell survival by inducing the
upregulation of the pro-survival protein Bcl-XL and inhibition of the pro-apoptotic Bcl-2 family
member protein, Bim (Scarisbrick et al., 2011).
1.8.5.2 Neurodegeneration
The highest expression of KLK6 in normal tissues is in the CNS. Particular interest in the role of
KLK6 in the brain comes from its identification (as Zyme) by Little et al., 1997. As such, its role
in both normal and diseased brains has been actively pursued. Alzheimer’s disease (AD), the
most common form of dementia (Ferri et al., 2005; Kalaria et al., 2008), is characterized by the
accumulation of amyloid β(Aβ), a proteolytic fragment of the amyloid precursor protein (APP)
(Zhang et al., 2011). Little et al. 1997, observed in vitro that co-expression of KLK6 cDNA with
APP cDNA in HEK cells resulted in an abundance of truncated amyloidogenic fragments,
implicating a role for KLK6 in the normal turnover of such proteins in the brain. These
observations are in keeping with observations by IHC in normal and AD brain sections showing
the decreased expression of KLK6 in AD brains over controls (Ashby et al., 2010; Ogawa et al.,
2000). Similarly, KLK6 levels in cerebrospinal fluid, blood and tissues from AD patients
revealed a similar trend with an average two-fold decrease of KLK6 in brain tissue extracts
compared to controls (Menendez-Gonzalez et al., 2008; Mitsui et al., 2002; Zarghooni et al.,
2002). Like AD, Parkinson’s disease (PD) is also a neurodegenerative disorder but is
characterized by the accumulation of insoluble alpha-synuclein (α-synuclein) (Shulman et al.,
2011), whereby KLK6 fails to degrade α-synuclein, contributing to the pathogenesis of PD,
possibly through the disrupted trafficking of KLK6 (Tatebe et al., 2010).
28
Unlike AD and PD, KLK6 levels in patients with Multiple Sclerosis (MS) are elevated (Hebb et
al., 2010; Scarisbrick, 2008; Scarisbrick et al., 2006; Scarisbrick et al., 2011; Scarisbrick et al.,
2008) and linked to inflammatory pathways. MS is a neurodegenerative disorder whereby the
myelin sheath, that protects nerve cells, is damaged. Damage to the myelin sheath is driven
primarily by inflammation (Stadelmann et al., 2011). Indeed, several studies (Hebb et al., 2010;
Scarisbrick, 2008; Scarisbrick et al., 2011) have documented significant increases in serum-
KLK6 in patients with MS, as well as those experiencing a secondary progressive disease course.
We have already alluded to the role of KLK6 in inflammation above, and many of the KLK6-
related studies undertaken in MS point to dysregulated inflammatory pathways. In addition to its
contribution to sustained activation of inflammatory signaling, KLK6 also degrades myelin
proteins (Bernett et al., 2002; Scarisbrick et al., 2002) and has been detected in elevated levels in
inflammatory demyelinating lesions of both viral and autoimmune MS models, as well as in MS
pathogenic lesions (Blaber et al., 2002; Scarisbrick et al., 2002).
1.8.5.3 Tumourigenesis
We already mentioned the observed overexpression of KLK6 transcript and protein in a variety
of cancers such as ovarian carcinomas (Anisowicz et al., 1996; Bayani et al., 2011; Bayani et al.,
2008b; Diamandis et al., 2000b; Diamandis et al., 2000c; Hoffman et al., 2002; Koh et al., 2011;
Kountourakis et al., 2008; Luo et al., 2006; Ni et al., 2004; Petraki et al., 2006; Prezas et al.,
2006; Rosen et al., 2005; Shan et al., 2007; Shaw and Diamandis, 2007 ; Tanimoto et al., 2001;
White et al., 2009a; White et al., 2009b; Yousef et al., 1999; Yousef et al., 2003b; Zheng et al.,
2007), breast (Anisowicz et al., 1996; Wang et al., 2008), uterine (Santin et al., 2005),
pancreatic (Ruckert et al., 2008), colorectal (Henkhaus et al., 2008; Kim et al., 2011; Ogawa et
al., 2005), gastric (Nagahara et al., 2005), skin (Klucky et al., 2007), urinary bladder cancers
(Shinoda et al., 2007), gliomas (Strojnik et al., 2009), lung (Nathalie et al., 2009; Singh et al.,
2008) and salivary gland tumours (Darling et al., 2006). Focus is now being diverted to
revealing the details of KLK6’s role in promoting tumourigenesis.
The majority of investigations regarding the consequences of KLK6 overexpression have been
linked to functions supported by in vitro studies demonstrating the ability of KLK6 to degrade
fibrinogen and various collagens, which constitute the basement membrane (Ghosh et al., 2004;
Magklara et al., 2003). These findings have significant implications for tissue remodeling,
29
tumour invasion and migration, with several studies showing that the decrease of KLK6 protein,
either through the use of inhibitors or siRNAs (Ghosh et al., 2004; Henkhaus et al., 2008; Kim et
al., 2011; Nagahara et al., 2005; Shinoda et al., 2007), results in the concomitant reduction in
invasive potential and cellular proliferation. The KLK6-siRNA treatment of colon cancer cell
lines (Kim et al., 2011) showed variable decrease in cell proliferation, though some cell lines
remained similar to controls; while Prezas et al., 2006, showed an increase in tumour burden and
invasive potential in ovarian cancers transfected with an expression vector inducing KLKs 4, 5, 6
and 7, though it is unclear what level of contribution to the observed tumourigenicity each
member made. The ectopic Klk6 expression in mouse keratinocyte cell lines by Klucky et al.,
2007 induced spindle-like morphology and enhanced proliferation, migration, and invasive
capacity. Furthermore, a reduction of E-cadherin protein levels at the cell membrane and of β-
catenin nuclear translocation was also detected in Klk6-expressing mouse keratinocytes and
human HEK293 cells transfected with a KLK6 expression plasmid. Cell-adhesion defects were
also observed in the presence of KLK6, which were rescued in the presence of inhibitor of
metalloproteinase (TIMP)-1 and TIMP-3.
Cell adhesion defects not only characterize morphological changes, but mark alterations in other
molecular pathways including apoptosis and angiogenesis. In keeping with findings by
Scarisbrick et al., 2011, describing that the overexpression of KLK6 in hematopoietic cells leads
to the activation of pro-survival pathways; Kim et al. 2011, found that suppression of KLK6 was
associated with the activation of caspase-8 and caspase-3, as well as the up-regulation of p21cip.
Anoikis, defined as the process by which an anchorage-dependent cell undergoes programmed
cell death as a result of detachment from the surrounding extracellular matrix (Frisch and
Screaton, 2001; Nagaprashantha et al., 2011), is another mechanism that cancers alter in order to
progress to metastatic disease. In a study by Kupferman et al. (Kupferman et al., 2007), anoikis-
resistant oral squamous carcinoma cell lines were generated under detached growth conditions.
cDNA expression profiling identified a 6.2-fold increase in KLK6, among others, in anoikis-
resistant cell lines compared to anoikis-sensitive cells under detached conditions. Based on these
findings, this novel association of KLK6 in anti-apoptotic pathways should be further
investigated. Cancer progression also involves the formation of additional vasculature. Aimes et
al. (Aimes et al., 2003) identified the expression of genes, including KLK6, in endothelial cells
which appear to be regulated, in part, by the nature of the substratum associated with these cells.
30
The role of KLK6 in intracellular signaling has also been actively studied in the context of
protease-activated receptors (PARs) (Hollenberg et al., 2008; Oikonomopoulou et al., 2006a;
Oikonomopoulou et al., 2006c). PARs 1-4 are members of the G-protein–coupled family of
receptors which are subject to activation through the proteolytic cleavage of the receptor N-
terminal sequence to reveal an activating tethered ligand (Hollenberg and Compton, 2002;
Macfarlane et al., 2001). These receptors have been implicated in various physiological roles,
including platelet activation, the modulation of endothelial and vascular smooth muscle cell
function, inflammatory responses to joint injury and tumour cell growth and metastasis
(Oikonomopoulou et al., 2010). Studies by Oikonomopoulou et al. (Oikonomopoulou et al.,
2006a; Oikonomopoulou et al., 2006b; Oikonomopoulou et al., 2006c), demonstrated the ability
of KLK6, as well as other KLK members, to cleave the synthetic peptides representing the
cleavage-activation sequences of PAR1, PAR2 and PAR4, and either activating or disarming the
receptor. Though the majority of work points to pathways associated with inflammation, cancer
investigators are increasingly aware of the link and high degree of cross-talk with inflammation
in contributing to pathogenesis (Aggarwal et al., 2009; Coussens and Werb, 2002; Gonda et al.,
2009). Thus, activation of PARs by KLK6 in the context of carcinogenesis, would feed into
known oncogenic signaling pathways such as PI3K and RAS (Castellano and Downward, 2010).
Due to the associated overexpression in virtually most epithelial cancers, the vast majority of
evidence supports an oncogenic role for KLK6. However, the original cloning of KLK6
(Anisowicz et al., 1996), implicated KLK6 as a putative class II tumour suppressor gene due to
its loss of expression in metastatic breast cancer. Work performed by Pampalakis et al., 2009,
demonstrated the inactivation of KLK6 in metastatic breast cancer due to epigenetic events. The
functional consequences of such inactivation, was further investigated to show that KLK6 may
have a protective role against breast cancer progression by inhibiting the epithelial-to-
mesenchymal transition (EMT). These findings suggest that a role for KLK6-tissue specificity.
1.8.6 Clinical Utility of KLK6 in Ovarian Cancer
The use of KLK3/PSA as a diagnostic and prognostic marker in prostate cancer has led to the
evaluation of KLK6 as a putative biomarker for many of the neoplasms and diseased states.
Since KLK6 is secreted, it has great potential as a biomarker that can be evaluated in patient
specimens obtained by less invasive means, such as blood or urine. To date, the majority of
31
clinical investigations have been performed retrospectively from tissue specimens, with a
handful also looking at KLK6 levels from paired biological fluids.
In cancer, an overwhelming number of studies show the overall increase in KLK6 mRNA and
protein expression when compared to their respective normal tissues, providing the basis for its
evaluation across clinical parameters. The vast majority of KLK6 cancer-related studies
encompass OCa (Bayani et al., 2011; El Sherbini et al., 2011; Hoffman et al., 2002; Koh et al., ;
Rosen et al., 2005; Shan et al., 2007; White et al., 2009a; White et al., 2009b; Zheng et al.,
2007). KLK6 appears to be overexpressed across all subtypes of OCa (Kountourakis et al.,
2008; Ni et al., 2004; White et al., 2009a; White et al., 2009b), with serous and undifferentiated
cancers generally showing greater percentages of cases with KLK6 positivity (Hoffman et al.,
2002; Shan et al., 2007). Because of some of the limitations of CA-125, KLK6 has been assessed
for its ability as a viable OCa biomarker. Rosen et al. (Rosen et al., 2005), who showed that of
the approximately 22% of ovarian cancers which were negative for CA-125 expression, all were
positive for other markers including KLK6. Along these lines, sseveral studies have consistently
shown that the combination of KLK6 with CA-125 enhances their individual diagnostic power
(El Sherbini et al., 2011; Hoffman et al., 2002; Koh et al., 2011; White et al., 2009a). These
findings are consistent with KLK6 RNA expression studies by Shan et al., 2007 who also
showed that KLK6 expression was significantly associated with late stage (stage III/IV) disease,
higher tumour grade, suboptimal debulking, and serous and undifferentiated histotypes.
Furthermore, univariate Cox regression showed that KLK6-positive patients showed an increase
risk of relapse and death, in contrast to KLK6-negative patients. Progression free survival (PFS)
and overall survival (OS) continued to be significant with positive KLK6 expression when
treated as a continuous variable. Kaplan-Meier analysis revealed a significant association
between KLK6 expression status and both progression-free survival (PFS) and overall survival
(OS); which was in contrast to those generated based on CA-125 alone. The value of KLK6
RNA levels to tumour recurrence was also investigated by White et al. 2009 showing that the
expression levels of KLK6 and KLK13 were significantly increased in invasive cancers as
compared to normal ovarian tissues. The high expression was also shown to be a poor prognostic
indicator and indicative of a shorter recurrence-free survival. Based on these observations, KLK6
has been included in large ovarian cancer biomarker validation efforts by the Early Detection
Research Network (EDRN) and Specialized Programs of Research Excellence (SPORE) (Cramer
32
et al., 2011; Zhu et al., 2011). KLK6 ranking among the top 10-20 (of 49 markers).
Specifically, it ranked 9th across all cases and general population controls in terms of sensitivity
(95% confidence interval), and 15th in Area Under the Curve (AUC – 95% confidence interval).
However, for early staged cases and general population controls, KLK6 ranked 29th and 28th for
sensitivity (95% confidence interval) and AUC (95% confidence interval), respectively. The
results of these studies clearly indicate the need for identifying novel biomarkers, to be utilized
individually and in panels, for both early detection and monitoring, since none of the biomarkers
tested performed any better than CA-125 alone.
1.8.7 Regulation of KLK6 Expression and Activity
The biology of an organism relies on the regulation of a gene at the DNA, mRNA, and protein
levels, but also relies on the interplay between these factors and tissue-specific
microenvironment. A number of genomic features that influence KLK6 expression have been
alluded to above and shown in Figure 1.6, and include the finding that no overt sequence
mutations in the coding regions of KLK6 have been found to date that results in its aberrant
expression. However, the presence of SNPs within both coding and non-coding regions of the
gene offers the possibility for uncovering genotype-phenotype associations as more genome
sequences become available.
33
Figure 1.6. Mechanisms of Regulation of KLK6 Expression. Shown is a summary of the various modes of regulation identified for KLK6 as described in the text.
Several studies have documented the concomitant dose-dependent increase in KLK6 gene by the
in vitro exposure of cell lines to androgens and estrogens (Paliouras and Diamandis, 2007;
Yousef et al., 1999). However, others (Shan et al., 2007) have shown minimal effects on KLK6
expression upon stimulation. Interestingly, significant hormone responses were detected in
breast cancer cell lines (Paliouras and Diamandis, 2007; Yousef et al., 1999), while ovarian
cancer cell lines showed non-significant responses (Shan et al., 2007), suggesting that factors
such as cell-type specificity plays a role in regulation. To date, no estrogen response elements
(EREs) have been identified within 6kb upstream of the proximal KLK6 promoter (Paliouras and
Diamandis, 2007), which is consistent with previous observations that classical hormone
response elements (HREs) have not been identified (Yousef and Diamandis, 2001). This
implicates a more complexed integration of mechanisms.
Christophi et al., 2004 examined a region of 500 bp upstream from the start site of the human
and mouse KLK6 genes and found 10 cis-acting regulatory sequences which were common to
both, and were of interest in terms of inflammatory responses in the CNS. Some of the candidate
34
genes associated with these cis-acting regulatory elements include; v-Myb, Ets-family member
ELF-2 (NERF1), sterol regulatory element binding protein 1 and 2, and ELK-1. Later, similar
integrative in silico studies in breast cancer cell lines also identified the putative ELK-1 binding
site, in addition to an E-box and AP-1 binding site and several Sp-1 sites (Pampalakis and
Sotiropoulou, 2006). Finally, KLK6 has also been identified as a target for the vitamin D
analog, EB1089, in studies where KLK6 expression was induced upon treatment with EB1089,
in squamous carcinomas (Lin et al., 2002), colon cancer (Palmer et al., 2003)and the T47D breast
cancer cell line (Pampalakis et al., 2006). Further examination of the KLK6 proximal promoter,
indicated predicted multiple vitamin D response-element consensus sequences (Pampalakis et al.,
2006).
Epigenetic mechanisms have also been extensively examined by Pampalakis and colleagues
(Pampalakis et al., 2006; Pampalakis et al., 2009; Pampalakis and Sotiropoulou, 2006).
Sequence analysis (Pampalakis and Sotiropoulou, 2006) found no CpG islands, defined by the
parameters of a length of more than 200bp, and with a C+G content more than 50%, or with an
observed/expected ratio greater than 0.6; either upstream or within exon-intron sequences of
KLK6. Experimentally, treatment of breast cancer cell lines with 5-aza-2’deoxycytidine (5-aza-
dC), resulted in the induction of KLK6 gene and protein expression offering the possibility that
non-CpG island cytosines are involved in the regulation of transcription. Indeed, these findings
were confirmed (Pampalakis et al., 2008; Pampalakis et al., 2009; Pampalakis and Sotiropoulou,
2006), where specific CpG dinucleotides were subjected to methylation in breast cancer cell
lines, resulting in the loss of KLK6 expression. Interestingly, the in silico comparison of these
CpGs across other mammalian KLK6 orthologs showed conservation of these dinucleotides
(Pampalakis et al., 2008). Fifteen CpG dinucleotides located within the human P1 transcriptional
start site of the classic transcript were analysed. In breast cancer cell lines lacking KLK6
expression, CpG dinucleotides at positions -72, -64, -56, -53, -35, +3 and +14 were completely
methylated. Similarly, those cell lines showing KLK6 expression possessed unmethylated CpG
dinucleotides.
As discussed above, KLK6 possesses several mRNA variants and promoter sites, however the
significance and frequency of such variants has not been thoroughly investigated; nor has it been
determined whether a functional protein results or what physiological role such isoforms may
play (Tan et al., 2006). However, the translation of RNA to protein provides another level for
35
gene regulation. In recent years, microRNAs (miRNAs) have been implicated in contributing to
disease and malignancies (Croce, 2009). miRNAs represent a class of non-coding RNAs that
range in size from 19 to 25 nucleotides, originally described in C. elegans (Lee et al., 1993). The
mechanism of regulation is mediated through degrees of complementarity to the target mRNA 3′
untranslated region (UTR). Perfect complementarity results in the cleavage and degradation of
the target mRNA; whereas less than perfect pairing, represses the translation process (Barbarotto
et al., 2008). In silico analyses of publicly accessible databases show that a number of miRNAs
are predicted to target KLK6, among them the members of the hsa-let-7 family of miRNAs
(Chow et al., 2008). Experimentally, Chow et al., 2008, demonstrated the decrease of KLK6
expression upon transient transfection of hsa-let-7f to the breast cancer cell line MDA-MB-468.
36
1.9 Rationale and Hypothesis
Kallikrein 6 (KLK6) is a putative biomarker for ovarian cancer whose expression is frequently
elevated in the tumour tissues and fluids derived from these patients. However the mechanisms
influencing the observed overexpression are currently unclear. Ovarian cancers show recurrent
patterns of genomic alteration leading to patterns of genomic instability, reflected by
chromosomal gains and losses resulting in both numerical and structural rearrangements.
Consequently, the expression of genes within these genomic regions may also be affected; either
directly by alteration to the KLK genomic region, or by the expression of other genes regulating
the cluster. Therefore, we hypothesize that the observed overexpression of KLK6 is influenced
by its copy-number in ovarian cancer, and that copy-number changes in other genes influencing
its post-transcriptional mechanisms serve to further modulate its expression. Although whole-
genome copy-number studies have been conducted for OCa, there have been no specific studies
focusing on KLK copy number or genes potentially regulating them.
1.10 Objectives
• To distinguish between numerical and structural chromosomal instability in ovarian
cancers utilizing molecular cytogenetic techniques and the implications for copy-
number change in those affected genomic regions.
• To determine the copy-number status of the KLK locus in ovarian cancers and cell lines
and the chromosomal mode of such copy-number change (ie. numerical or structural
changes) using cytogenomic techniques.
• To identify whether the entire locus experiences such copy-number changes or whether
individual gene members are affected.
• To determine the frequency and range of KLK locus changes in ovarian cancers and its
relationship to protein expression, specifically KLK6.
• To determine whether copy-number changes of the KLK locus are clinically relevant.
37
• To investigate other mechanisms affected by chromosomal aberrations which modulate
the observed protein overexpression of KLK6, namely the role of miRNAs in
influencing KLK6 expression.
38
Chapter 2
Distinct Patterns of Numerical and Structural Chromosomal Instability Characterize Sporadic Ovarian
Cancer
Chapter 2 includes the published manuscript entitled “Distinct Patterns of Structural and Numerical Chromosomal Instability Characterize Sporadic Ovarian Cancer” Bayani J, Paderova J, Murphy J, Rosen B, Zielenska M, Squire JA. Neoplasia. 2008. 10:1057-65.
J.Bayani contributed to the development and experimental design of the study and manuscript; established primary cultures and cytogenetic preparations; performed all experiments, analyses and interpretation. J.Paderova provided technical support of cytogenetic preparations and confirmation of karyotypic nomenclature. J.Murphy and B.Rosen provided patient material and associated clinical information. M.Zielenska contributed to the development and design of the study and approved the manuscript. JA Squire contributed to the development and design of the study, manuscript and granting support.
39
2 Distinct Patterns of Structural and Numerical Chromosomal Instability Characterize Sporadic Ovarian Cancer
2.1 Introduction Ovarian cancer is the fourth leading cause of cancer death among North American women, and
the leading cause of death from a gynecological malignancy. The karyotypes of these tumours
are characterized as aneuploid with complex chromosomal aberrations (Bayani et al., 2002;
Bernardini et al., 2005; Shridhar et al., 2001; Taetle et al., 1999a; Taetle et al., 1999b). The
features of many epithelial tumours, including OCa, are the presence of structural changes and
aneuploidy, arising as a consequence of CIN. CIN has been classically defined as the rate of
whole chromosomal gains and losses (Lengauer et al., 1997; Rajagopalan et al., 2003), and has
been a useful means for assessing genomic heterogeneity. CIN has also been less frequently
used to describe the presence of structural aberrations (Lengauer et al., 1997; Muleris et al.,
2008). More complex genomic changes in tumours arise as a consequence of chromosomal
instability (CIN), which can generate both numerical (N-CIN) and structural chromosomal
instability (S-CIN) (Bayani et al., 2007). The observed high levels of aneuploidy and structural
complexity in these tumours suggests errors in DNA repair, mitotic segregation errors and
dysregulation of cell cycle checkpoints (Cahill et al., 1998; Kops et al., 2005) may generate N-
CIN. However the mechanisms that influence structural changes and gross aneuploidy are
distinct. Structural rearrangements appear to be influenced by abnormal DNA repair pathways,
that result in errors in both homologous and non-homologous end-joining of double stranded
DNA and may contribute to S-CIN (Hoeijmakers, 2001) (and reviewed in Bayani et al. 2007).
The formation of structural rearrangements may also come about through telomere-mediated
events, where critically short telomeres are recognized as DNA breaks, capable of recombining
either homologously or non-homologously when DNA-repair pathways are compromised and
telomerase is activated (Stewart et al., 1999). In contrast to structural changes, aneuploidy arises
through various mechanisms predominantly influenced by the dysregulation of cell cycle
checkpoints and mitotic segregation errors; or some combination of all (Cahill et al., 1998; Nigg,
2002).
40
Tetraploidy is a common feature of many carcinomas and sarcomas and is believed to be an
initial stage in the development of aneuploidy, arising either through disruption of chromosome
segregation during mitosis (Stewart et al., 1999) or through failure of cytokinesis (Andreassen et
al., 2001). The tetrapoloidization event not only results in the doubling of genomic content, but
also in the doubling of centrosomes (Andreassen et al., 2001; Skoufias et al., 2001). Aneuploidy
can also arise from mechanisms independent of a tetraploid intermediate, suggesting that the
components of the mitotic machinery possess aberrant function, such as the loss of centrosome
duplication control, leading to abnormal centrosome amplification and creating, multipolar
spindles and unequal genomic segregation (Saunders, 2005).
Cytogenetic analyses of primary OCas have identified tumours with a range of numerical
change, from near-diploid to highly aneuploid abnormal karyotypes. Moreover, within most
reported studies some tumours also have many chromosomes with a high degree of structural
aberration; whilst others, exhibited relatively few, simple rearrangements (Bayani et al., 2002;
Shridhar et al., 2001; Taetle et al., 1999a; Taetle et al., 1999b). Defining these varying
characteristics in greater detail, using primary OCa samples will provide insights concerning the
mechanisms responsible for genomic diversity in these tumours. Recognition of such
mechanisms associated with each class of chromosomal aberration will be beneficial in
understanding the nature of karyotypic progression, as well as avenues for future therapeutic
strategies in ovarian carcinoma. Using integrative molecular cytogenetic analyses including
Spectral Karyotyping (SKY), Comparative Genomic Hybridization (CGH) and interphase
fluorescence in situ hybridization (FISH), ten sporadic OCa specimens derived from seven
patients, previously described by our group (Bayani et al., 2002), were assessed for features of
N-CIN and S-CIN; in addition to centrosome amplification.
2.2 Materials and Methods
2.2.1 Patient Specimens and Cell Lines
Ten tumours from seven sporadic ovarian cancer patients, with informed consent were obtained
from The University Health Network, Toronto, during the years 2000-2001. All tumours, except
for OCA27A/B and OCA714, were previously described by Bayani et al., 2002. These patients
had no previous family history for ovarian or breast cancer, and all specimens were obtained at
the first surgery prior to chemotherapeutic treatment. For three patients, two distinct tumour
41
samples were obtained. OCA21A/OCA21B and OCA27A/OCA27B are paired
primary/metastasis samples. For OCA15A/B tumours were obtained from the right and left
ovaries, respectively. A normal female control fibroblast (passage 9) was also maintained in
culture and used as a technical control for hybridization efficiency and normal mitotic
segregation.
2.2.2 Peptide Nucleic Acid (PNA) Fluorescence in situ Hybridization (FISH) and Spectral Karyotyping Analysis
Cell culture and chromosome preparation: Primary tissues were obtained at the time of surgery
and promptly minced and collagenase treated (Invitrogen, Burlington, Canada) for 24 hours at
37°C in a CO2 incubator. The following day, the disaggregated tissue was processed for short-
term culture and then prepared for cytogenetic harvest as previously described (Bayani et al.,
2002). A normal control fibroblast culture was maintained and processed similarly.
Peptide Nucleic Acid (PNA) Fluorescence in situ Hybridization (FISH): For chromosomal
instability studies, interphase FISH was performed using peptide nucleic acid (PNA) probes
directly labeled with either FITC or Rhodamine for centromeres 2, 4, 7 and 8 (Applied
Biosystems, Foster City, CA.). Pan-centromeric PNA probes (Applied Biosystems, Foster City,
CA.) were also used to identify the presence of multi-centromeric chromosomes. The standard
technique for PNA FISH was carried out as previously published by Wan et al., (Wan et al.,
1999). Two hundred nuclei were assessed in a coded fashion by two independent observers.
Generation of Instability Index: Each tumour specimen was assessed using all four PNA
centromere probes. An instability index was generated based on published criteria by Lengauer
et al. 1997. For each specimen, including a normal fibroblast, two hundred nuclei were counted.
The normal fibroblast sample served as a technical control for the overall hybridization
efficiency for each centromere probe used; and was found to be greater than 96%, with the
average background noise across all centromere probes tested to be 3.6% ((2+3.5+4+4)/4)). For
the tumour specimens, the prevalent clones for each centromere probe tested were initially
established by karyotypic (SKY) analysis, and the percentage of cells containing greater or less
than that clonal population were enumerated. In keeping with guidelines of clonality (Shaffer and
Tommerup, 2005), if it was determined that the prevalent signal count for that centromere was
not consistent with the karytoype analysis, the interphase FISH results were used in our analysis.
42
The identification of populations greater than 30% were also considered as an additional clone
and excluded from the enumeration. For example in OCA3, the total percentage of cells
possessing greater or less than 4 signals for centromere 2 was 52%; for centromere 4, the total
percentage of cells possessing greater or less than 2 signals per cell was 24.5%; for centromere 7,
the total percentage of cells possessing greater or less than 4 signals per cell was 37%; and for
centromere 8, the total percentage of cells possessing greater or less than 3 signals per cell was
22%. Thus the average CIN index is 33.9 ((52+24.5+37+22)/4).
Spectral Karyotyping (SKY): SKY analysis was performed on metaphase preparations using the
SKY Paints™ according to the manufacturer’s instructions (Applied Spectral Imaging, Carlsbad,
CA) and as described by previously by Bayani et al., 2002. Karyotypes derived by SKY analysis
reflect the clonal changes determined from at least ten metaphase spreads where possible. A gain
of a chromosome was described when identified in at least two metaphase spreads, a loss when
identified in three or more cells and a chromosomal rearrangement when identified in two or
more cells. The karyotype descriptions were assigned according to the guidelines set forth by the
International System for Cytogenetic Nomenclature 2005 (ISCN) (Shaffer and Tommerup,
2005). Breakpoint enumeration was performed on intact metaphase spreads that were
analysable.
2.2.3 Centrosome Immunostaining
For all cases, cells were grown onto chambered glass slides and cultured to approximately 70%
confluency. The cells were fixed with a 4% paraformaldehyde solution and processed for
fluorescence immunohistochemistry with mouse monoclonal anti-γ-tubulin antibodies (Sigma-
Aldrich, Oakville, Canada) as previously described (Al-Romaih et al., 2003). Centrosome
signals were evaluated in 200 non-overlapping cells. For each specimen, the percentage of cells
possessing greater than 2 centrosomes per cell were considered abnormal or if the signal size and
shape were different from the size and shape in control cells (Al-Romaih et al., 2003; Pihan et
al., 1998). The slides were viewed and imaged using a Nikon Labophot-2 fluorescent
microscope, and analyzed using the Vysis Quips SmartCapture™ imaging system (Vysis,
Downers Grove, Il).
43
2.3 Results
2.3.1 Numerical Chromosomal Instability: N-CIN
To determine N-CIN levels within the study group, interphase FISH was performed using
centromere–specific probes from four representative chromosomes. Centromere probes for
chromosomes 2, 4, 7 and 8 were chosen for this study based on their propensity for net gain
(chromosomes 7 and 8), loss (chromosome 4) or tendency for little change (chromosome 2) as
determined by previous cytogenetic and CGH studies conducted by this group and others
(Bayani et al., 2002; Bernardini et al., 2005; Shridhar et al., 2001; Taetle et al., 1999a; Taetle et
al., 1999b). Table 2.1 summarizes the results of interphase chromosomal instability findings
using the PNA probes, with representative images from OCA5, OCA 27B and OCA714 in
Figure 2.1. The raw scoring data is detailed in Table 2.2. An N-CIN index was generated for
each tumour as described in the Materials and Methods. Analysis based on the N-CIN indices
revealed two distinct ranges of instability; those with a low N-CIN index ranging from 7.0 to
21.0, and another group of tumours with a higher N-CIN index ranging from 24.9 to 54.9. The
copy number status of these tumours, as determined by previous SKY analysis (Bayani et al.,
2002), revealed those cases with lower instability indices possessed karyotypes in the diploid or
near-diploid range. This was in contrast to the second group where the instability indices were
associated with triploid and tetraploid karyotypes. The tumour that showed the greatest
instability was OCA5 (Figure 2.1A), a near-triploid tumour with an instability index of 54.9.
The results of the individual chromosomes revealed a range of centromere signals from 1 to as
many as 9 signals per cell, for the four chromosomes tested. OCA3 displayed the next greatest
instability index of 33.9. Like OCA5, OCA3 was characterized as near-triploid; however the
range of instability was not as extensive as OCA5. Among the near-tetraploid tumours,
OCA27A/B, paired primary and metastatic tumours, possessed instability indices of 24.9 and
32.7 respectively, for which the FISH and SKY findings for OCA27B are illustrated in Figure
2.1B. OCA15A/B are pair samples showing similar instability indices of 10.7 and 12.7
respectively. Karyotype analysis revealed a hypodiploid karyotype with numerous structural
aberrations similar to both specimens. OCA714 possessed an instability index of 7.0 showing a
diploid karyotype as confirmed by SKY analysis (Figure 2.1C). When the N-CIN indices were
plotted in ascending order (Figure 2.2), the resulting trend revealed that numerical instability
increased in association with transitions in ploidy, specifically there was very little instability
44
among the more diploid tumours, while near-tetraploid tumours showed a moderate to high level
of instability and near-triploid tumours showing the greatest level of instability.
2.3.2 Structural Chromosomal Instability: S-CIN
To determine S-CIN levels, the number of clonal breakpoints were enumerated for each case
(Table 2.1) based on previously published SKY analysis (Bayani et al. 2002 Supplementary
Table A), as well as the inclusion of three new specimens from two patients. SKY analysis of
OCA27A/B revealed similar tetraploid karyotypes between the primary and metastatic tumour
with complex aberrations. The composite karyotype for both OCA27A/B is summarized in
Table 2.3. In the case of OCA714, SKY revealed a highly rearranged karyotype, where almost
every chromosome was involved in a complex aberration (Table 2.3 and Figure 2.1C).
Numerous complex translocations involved more than three different chromosomes and showed
an extremely high number of breakpoints (n=83) (Table 2.1, Figure 2.1C) against a diploid
background.
The enumeration of the number of clonal aberrations and breakpoints showed that diploid and
near-diploid tumours possessed elevated DNA breakage events resulting in complex
translocations, inversions and deletions, than whole chromosomal gains and losses (Table 2.1,
Figure 2.3 and see Bayani et al., 2002 Supplementary Table A). Non-diploid tumours, on the
other hand showed fewer incidences of clonal DNA breakage events, but were associated with
copy number changes that were reflected by the wider range of chromosomes identified per cell.
Moreover, the findings of several duplicate structural changes in only the tetraploid and triploid
cases (Table 2.1 and see Bayani et al. 2002 Supplementary Table B), and illustrated by the
karyotypic changes in OCA27A/B including i(8)(q10)X2, i(14)(q10)X2,
der(19)t(19;20)(p11;p11)X2, der(22)t(11;22)(q14;p11)X2, suggests these rearrangements
occurred as an early event during the diploid stage.
Non-clonal rearrangements were a consistent feature for all the tumours, with diploid and triploid
tumours showing a slightly greater frequency of non-clonal aberrations over tetraploid tumours
(data not shown). However due to the differences in mitotic indices of each specimen and the
number of analyzable metaphases, a precise enumeration of the total number of non-clonal
rearrangements could not be accurately assessed.
45
2.3.3 Multicentric Chromosomes
The presence of chromosomal aberrations such as multicentric chromosomes, are additional
hallmarks for genomic instability mediated by Bridge-Fusion-Breakage (BFB) events or critical
telomere shortening (Gisselsson and Hoglund, 2005; Murnane, 2006). Thus to determine whether
these structures were present in these tumours, FISH using pan-centromeric PNA probes was
performed and revealed the presence of dicentric chromosomes or ring chromosomes containing
multiple pan-centromeric sequences as both clonal and non-clonal changes. Of the tumours
analyzed, OCA8 was the only tumour that did not show evidence of multi-centric chromosomes.
The configurations of multicentromeric chromosomes varied and were identified as ring
chromosomes as in the case of OCA5, telomeric fusions as detected in OCA714 as well as more
typical dicentric configurations as in OCA21B and OCA27B (Figure 2.1D).
Table 2.1 Summary of N-CIN Indices, S-CIN Events and Centrosomal Aberrations in Primary Ovarian Carcinomas
Case Subtype Ploidy N-CIN Index
Percentage of cells with abnormal
Centrosomes
S-CIN Index
OCA714 Serous 2n- 7.0 20.0 83 OCA15A Endometrioid 2n- 10.7 5.5 58 OCA15B Endometrioid 2n- 12.7 14.0 53
OCA8 Pap.Serous 2n+ 21.0 20.0 21 OCA21A Serous 4n/2n 7.6 15.0 26 OCA21B Serous 2n 7.4 11.5 26 OCA27A Serous 4n- 24.9 16.5 29 OCA27B Serous 4n- 32.7 19.0 29
OCA3 Pap.Serous 3n 33.9 20.5 43 OCA5 Clear Cell 3n+ 54.9 24.0 15
46
Figure 2.1. Genomic Instability Findings in Primary Ovarian Cancers by Interphase FISH, SKY and Centrosomal Analysis. A. Genomic Instability findings for OCA5. Shown is the histogram generated by interphase FISH analysis using centromere probes 2, 4, 7 and 8. The X axis represents the number of signals per cell and the Y axis shows the percentage of cells. Interphase analysis identified a large distribution of signals per cell outside the established near triploid genome. An example of the interphase FISH is shown using centromeres 7 (red) and 8 (green) illustrating the variability of centromere signals per cell. Centrosome staining using γ-tubuln (red) illustrate the presence of supernumerary centrosomes present during cell division and interphase. A representative SKY karyotype is also illustrated showing the presence of duplicate rearrangements as well as unique rearrangements. B. Genomic Instability findings for OCA27B. Shown is the histogram OCA27B. A representative interphase FISH image for centromeres 7 and 8 is also shown as well as centrosome staining showing a tripolar metaphase and supernumerary centrosomes. A representative SKY karytoype below illustrates the near tetraploid karyotype and duplicate markers. C. Genomic Instability findings for OCA714. The histogram for the diploid tumour OCA714 shows over 95% of cells with two signals for each centromere tested, with a representative interphase FISH image for centromeres 7 and 8 shown. Supernumerary centrosomes were also detected in a proportion of cells. The SKY karyotype below reveals the structural complexity of the genome, despite its relatively non-aneuploid status. D. Examples of multi-centric chromosomes using pan-centromeric FISH probes. Shown are examples from OCA5, OCA714, OCA21B and OCA27B found to contain multi-centric chromosomes, including the presence of a ring chromosome in OCA5 (inset) and telomere fusions in OCA714 (inset) as well as typical dicentric chromosomes in OCA21B and OCA27B (inset).
47
Figure 2.2. Increasing Copy-Number Instability is Associated with Changes in Ploidy. Based on the generated CIN indices, increasing chromosomal instability was associated with changes in ploidy. The X axis represents the cases sorted in ascending order based on N-CIN indices while the Y axis represents the N-CIN indices. * OCA21A was found to possess a tetraploid karyotype, but contained many normal contaminating normal cells by karyotype analysis. OCA21B possessed a diploid karyotype, but also contained many normal contaminating cells by karyotype analysis
2.3.4 Centrosome Aberrations and Association with Increasing N-CIN
The percentage of cells with an aberrant centrosome was evaluated in all the specimens by
fluorescent immunohistochemistry using anti-γ-tubulin (Table 2.1). The most abnormal specimen
was OCA5 with a 24% of cells showing abnormal centrosome configuration or number (Figure
2.1A). The corresponding instability for this case, as previously discussed, was also the highest.
Although mitotic figures were not frequent, abnormal mitoses were detected and shown in Figure
2.1B for OCA 27B. OCA3 and OCA27A/B also exhibited frequent centrosome aberrations
(Table 2.1) with percentages of 20.5, 16.5 and 19.0 respectively.
While the observed range of cells containing centrosomal aberrations was between 5.5% - 24.0%
across all cases, increasing N-CIN indices were generally associated with the increasing
frequency of cells with aberrant centrosomes as summarized in Figure 2.3. One case, which
failed to follow the general trend, was OCA714, where the relative N-CIN was low (7.0),
although the incidence of centrosome abnormalities (20%) was high.
48
2.4 Discussion The karyotypes of OCas are characterized by aneuploidy and structurally complex chromosomal
rearrangements (Bayani et al., 2002; Shridhar et al., 2001; Taetle et al., 1999a; Taetle et al.,
1999b). Chromosomal instability (CIN) has traditionally been assessed by the rate of whole
chromosomal copy number changes (Lengauer et al., 1997) typically by interphase FISH
analysis using centromere specific probes. With the exception of a few recent studies (Gorringe
et al., 2005; Grigorova et al., 2004; Muleris et al., 2008), the potential significance and extent of
structural changes have been cursory. In this study we investigated the relationship between the
observed karyotypic complexity and ploidy changes observed by structural instability, S-CIN;
and copy number-driven chromosomal instability, N-CIN; respectively; with features associated
with aberrant mitotic progression, in untreated sporadic OCas.
The interphase FISH study and the establishment of N-CIN indices revealed two distinct groups:
Group 1 possessed a low range of N-CIN scores (7.0 to 21.0), characterized as near-diploid; and
Group 2 which possessed a higher range of N-CIN scores (24.0 to 54.9), which were
characterized by near-triploid and near-tetraploid karyotypes. The trend observed by the N-CIN
indices was also seen, although more subtly, when the enumeration of abnormal centrosomes
were assigned to each case. The increasing frequencies of cells with abnormal centrosomes were
observed to be associated with the progression from a 2n cell complement to 2n±; then to 4n and
4n±, through to 3n± (Figure 2.3). The apparent transition to the tetraploidization event and
subsequent reduction to pseudo-tetraploidy and (near) triploidy, implicates errors in
chromosomal segregation and cytokinesis. This is in keeping with centrosome studies showing
the increased occurrence of mitotic segregation anomalies associated with centrosome
amplification (Doxsey, 2002; Pihan and Doxsey, 1999). However, the presence of centrosome
amplification does not always commit cells to multipolarity (Wang et al., 2004), since
centrosome coalescent functions may still remain intact resulting in normal bi-polar division
(Quintyne et al., 2005; Saunders, 2005). This may explain the findings of increased centrosome
number in some diploid tumours, but showing relatively low-level N-CIN. In this situation, the
presence of aberrant centrosome number or shape/configuration may simply be an early indicator
of a general dysregulation of mitotic pathways. Unlike aneuploid genomes (ie. 3n±, 4n±) which
possess sufficient genomic material to survive the losses of several whole chromosomes through
missegreation and multipolarity; gross missegregation of a diploid cell may yield daughter cells
49
lacking the genomic material necessary for adequate survival into the next cell cycle.
Interestingly Rebacz et al. (Rebacz et al., 2007), recently identified griseofulvin as an inhibitor of
centrosomal clustering and demonstrated that induction in tumour cell lines cells containing extra
centrosomes, but maintaining bi-polar division, resulted in loss of centrosomal clustering and
multipolar metaphases leading to cell death. Treatment of normal cells with only two
centrosomes showed no affect on cell viability and centrosomal functioning. This suggests that
tumour cells exhibiting centrosomal amplification and maintaining centrosomal clustering
functions, observed by bi-polar division (or relatively low N-CIN), can be selectively targeted.
When the extent of S-CIN was investigated through the enumeration of clonal breakpoints
(resulting from translocations, inversions, deletions), we found diploid and near-diploid tumours
(OCA 8, OCA 15A/B, OCA718) possessed more occurrences of chromosomal breakage (Table
2.1). In particular, OCA714 displayed the greatest number of chromosomal breakages (n=83
breakages), amidst a diploid background (Table 2.1). These results suggested significant errors
in DNA repair occurred before gross changes in copy number. Although chromosome breakage
events (clonal and non-clonal aberrations) were observed as an ongoing event in the non-diploid
cases, we found many complex rearrangements occurring as duplicates in near-tetraploid or near-
triploid cases, indicating these rearrangements occurred prior to the tetraploidization of the
genome. The presence of non-clonal chromosomal rearrangements across all tumour specimens
indicated that errors in DNA repair was an ongoing event in the evolution of the tumour genome,
however it was noted that the diploid and triploid tumours showed slightly more non-clonal
changes which were generally more structurally complex. The presence of multicentric
chromosomes in all cases, except for OCA8 further supports the notion that impaired DNA
repair is an early event. Studies have shown the presence of such structures are mediated by the
BFB cycle, described as a cycle involving chromatid breaks and fusions triggered by dicentric
and ring chromosome rupture during anaphase resolution (Jones, 2005). This self-perpetuating
process gives rise to amplifications (HSRs, ladder amplifications), complex chromosomal
rearrangements, inverted repeats, interstitial deletions and large duplications (Lim et al., 2005;
Selvarajah et al., 2006). The occurrence of telomere fusions in some cases also implicate
telomere shortening as a mechanism for generating the observed genomic instability (Vukovic et
al., 2007) and has been previously described (Gisselsson and Hoglund, 2005). The most striking
result of the S-CIN study was the finding of numerous DNA breakage events that were
50
independent of copy number changes. The limitation of such analyses however is in the ability
of obtaining sufficient metaphases that are analyzable from short-term primary cultures.
Interestingly, the analysis revealed paired (OCA15A/B) and primary/metastatic (OCA21A/B and
OCA27A/B) specimens showed similar N-CIN and S-CIN indices, suggesting the existence of a
putative clonal progenitor. These observations demonstrate that analysis of recurrent and
metastatic samples derived from the same patients, can provide insightful illustrations of the
adaptive potential of tumour genomes during disease progression and in response to treatment.
The molecular cytogenetic analysis of these tumours revealed several aspects of N-CIN and S-
CIN, namely that: 1) sporadic OCas can show a wide range of N-CIN; 2) increasing N-CIN and
changes in ploidy were associated with the increased occurrence of centrosome aberrations; and
3) that S-CIN occurs more frequently in the diploid stage of karyotypic evolution implicating a
more prominent role for impaired DNA repair pathways early in tumour progression. A re-
evaluation of our previous molecular cytogenetic data in primary osteosarcoma and cell lines
(Bayani et al., 2003; Selvarajah et al., 2007; Zielenska et al., 2001) have also shown this
interplay between N-CIN and S-CIN; and are recapitulated in current studies (unpublished
observations). Based on these findings, we suggest that both forms of chromosomal aberrations
can arise dependently and/or independently of each other, with one form more prevalent than the
other at different times within the evolution of the tumour genome (Figure 2.3). Thus a delicate
balance between N-CIN and S-CIN processes will influence the observed levels of structural and
numerical change in a tumour.
BRCA1 is strongly associated with OCa (Prat et al., 2005), and has been shown to act as a
regulator of DNA damage, repair and transcription, and likely has a role in maintaining genomic
stability (Deng, 2006). There is increasing evidence (reviewed by Deng, 2006), suggesting the S-
CIN damage response function of BRCA1, can act through RB and p53. Interestingly, BRCA1
mouse models in breast cancer (Deng and Wang, 2003; Weaver et al., 2002) have demonstrated
patterns of aneuploidy and centrosomal amplification similar to the N-CIN findings presented in
this study. In keeping with these observations, Bae et al. (Bae et al., 2005), demonstrated a role
for BRCA1 in regulating the expression of genes implicated in the mitotic spindle checkpoint,
chromosome segregation, centrosome function, cytokinesis, and the progression into and through
mitosis resulting in multi-nucleated cells and failed cytokinesis.
51
The concept of tumour ploidy as an indicator of disease aggression and progression, in ovarian
cancer has been previously studied primarily by flow or image cytometric analyses (Kimmig et
al., 2002; Kristensen et al., 2003; Resnik et al., 1997). The extensive use of high-throughput
microarray analyses have shown the average genomic changes in these tumours include whole
and partial chromosomal gains and losses, as well as focal regions of amplification and deletion
(Bernardini et al., 2005; Prasad et al., 2008; Zhang et al., 2007). More recently, these regions of
genomic imbalance have been linked to the differential expression of microRNAs (miRNAs) -
now emerging as an important regulator of protein expression (Barbarotto et al., 2008; Calin and
Croce, 2007). However, while providing detailed information regarding the net copy-number
imbalances, the drawback of net/bulk genomic-based analysis is the inference of the tumour
ploidy and rearrangement status, in the absence of any parallel interphase or metaphase analyses.
This poses a limitation, since there is increasing interest in the karyotypic heterogeneity of
tumours as a measure of genome (in)stability, which is lost when performing such bulk-based
experiments (reviewed by Bayani et al., 2007). Such bulk-based studies appear to ignore the
potentially significant contributions of poly-clonality, which is often considered the source of
“noise” in array analyses and consequently filtered out statistically. Certainly as an extreme
example, it has been demonstrated that rare cell populations such as cancer stem cells, have a
profound and significant impact on disease recurrence and treatment resistance (Ailles and
Weissman, 2007; Dean et al., 2005), despite its relatively small population.
In summary, the results of our study have demonstrated both aspects of CIN, that is copy number
change and structural change, should be distinguished. Thus we propose to differentiate between
these two aspects of chromosomal instability as numerical chromosomal instability (N-CIN),
characterized by the rate of copy number changes; and structural chromosomal instability (S-
CIN), characterized by the extent of structural change and complexity. Indeed we have shown
the presence of S-CIN can occur independently of gross copy number or ploidy changes, and that
interphase FISH analysis, for copy-number enumeration, alone is not sufficient to classify a
given specimen as chromosomally unstable. The significance of such a distinction lies in the
mechanisms that mediate the observed changes, providing viable avenues for therapeutic
intervention.
52
Figure 2.3. Interplay between N-CIN, S-CIN, Centrosomal Aberrations and Ploidy. Illustrated are the N-CIN, S-CIN and Centrosomal Aberrations histograms showing the relationship with CIN indices and ploidy. In this figure, S-CIN and centrosome aberrations were plotted relative to increasing N-CIN indices, which were previously shown in Figure 2.2. As shown here and in Figure 2.2, during the progression from 2n± to 4n± to 3n±, the N-CIN indices of the tumours tested increased (blue). When the S-CIN events were plotted, we observed more predominant S-CIN events in the diploid tumours (green) with lower N-CIN indices. Fewer structural events appear to occur in tetraploid and triploid cancers, while their N-CIN indices increase. A less prominent trend is seen for the frequency of supernumerary and abnormal centrosomes (red); however the cells with more centrosomal alterations were those that were near-tetraploid and near-triploid. The bottom figure illustrates the interplay between numerical and structural changes as it relates to changes in ploidy. At initiation, the diploid genome undergoes DNA-damaging events resulting in the accumulation of structural rearrangements. Such rearrangements result in accompanying changes in gene expression, affecting all pathways including the cell cycle and mitosis. Early indicators of cell cycle and mitotic dysregulation can include centrosomal amplification and low-leveled chromosomal polysomy, while maintaining overall diploidy (bi-polar division) during tumour expansion. At some critical point during tumour progression, the failure of cytokinesis occurs resulting in the doubling of the genome as well as centrosomes. This event provides the genomic content and mitotic machinery to undergo gross numerical changes yielding tumour genomes which can assume near-tetraploid or near-triploid content. While gross numerical changes are occurring, structural events are ongoing and may once again become the predominant genomic event once a stable ploidy level has been obtained. It can be predicted that another failed cytokinetic event can occur and the cycle repeated, since 6n± tumours and 5n± have been observed in many cancers.
53
Table 2.2. Raw FISH Scoring Values
Cen 0 1 2 3 4 5 6 7 8 >9 Percentage of Non‐Clonal Cells
CIN Index
Control 2 0 0 98 1 1 0 0 0 0 0 2 4 0 0.5 96.5 2 1 0 0 0 0 0 3.5 3.6 7 0 2.5 96 1 0.5 0 0 0 0 0 4 8 0 2.5 96 1 0.5 0 0 0 0 0 4 OCA3 2 0 0 25.5 24 48 2.5 0 0 0 0 52 4 0 8 74.5 14 2.5 0 0 0 0 0 24.5 33.9 7 0 0.5 16 20 64 0.5 0 0 0 0 37 8 0 1 19 78 1 0.5 0.5 0 0 0 22 OCA5 2 0 1.5 11.5 13 60 6 7 1 0 0 40 4 0 5 22 24 29 14 1 4 0 1 71 54.9 7 0 0.5 20 3 31 31.5 4.5 1 0 4.5 33.5 8 0 0.5 22 15 25 24 7 1 0.5 5 75 OCA8 2 0 0 15.5 80 4.5 0 0 0 0 0 20 4 0 3.5 43.5 34.5 16 1 0.5 0.5 0.5 0 22 21 7 0 2 92 3 3 0 0 0 0 0 8 8 0 0 21 66 9 0.5 3 0.5 0 0 34 OCA15A 2 0 1 92.5 2.5 3 0.5 0.5 0 0 0 7.5 4 0 3 85.5 2.5 9 0 0 0 0 0 14.5 10.7 7 0 0.5 89.5 3 5.5 0.5 1 0 0 0 10.5 8 0 1 89.5 3 6 0 0.5 0 0 0 10.5 OCA15B 2 0 4.5 89 3 3.5 0 0 0 0 0 11 4 0 3.5 83 4 8.5 1 0 0 0 0 17 12.7 7 0 0.5 90 7.5 2 0 0 0 0 0 10 8 0 0.5 87.5 5 7.5 0 0 0 0 0 13 OCA21A 2 0 0 85.5 1 13 0 0.5 0 0 0 1.5 4 0 0 85.5 5.5 6 0.5 1.5 0.5 0 0.5 8.5 7.6 7 0 4.5 88 1.5 5.5 0 0 0.5 0 0 6.5 8 0 8 80 4.5 6 0.5 0 0.5 0.5 0 14 OCA21B 2 0 4 91 3 2 0 0 0 0 0 7 4 0 3.5 81.5 4 9.5 0.5 0 0.5 0.5 0 9 7.4 7 0 4 82.5 4 8 0.5 0 0.5 0.5 0 9.5 8 0 1 89 2.5 7 0 0 0.5 0 0 4 OCA27A 2 0 0 45* 4 9 35 4.5 0 1 1.5 20 4 0 3 36.5* 22.5 32.5 3 0.5 0 1 1 31 24.9 7 0 0 48* 8 41.5 2 0.5 0 0 0 10.5 8 0 1 47* 9 15 12.5 6.5 6.5 1.5 1 38 OCA27B 2 0 1 10* 13.5 60 11.5 1.5 1 1.5 0 30 4 0 1.5 10* 18.5 60 9.5 0.5 0 0 0 30 32.7 7 0 1 14* 20 62 2.5 0.5 0 0 0 24 8 0 0 13* 12 39 20 10 3.5 0.5 1 47 OCA714 2 0 4 90 4.5 1.5 0 0 0 0 0 10 4 0 5 89 3.5 2 0.5 0 0 0 0 11 7 7 0 1 96 1 1 0 1 0 0 0 4 8 0 0 97 0.5 2 0 0 0 0.5 0 3
54
Shown are the tabulated data for each centromere (Cen) tested in control fibroblasts and primary tumours. Two hundred nuclei were enumerated and the values reflect the percentage of cells containing the specified numbers of signals per cell. Bolded values shaded in grey correspond to the number of signals corresponding to the dominant clone established by karyotype analysis (SKY). In situations where the percentage was equal to or exceeded 30%, this was considered another clonal population. In addition, cells containing a diploid count corresponding to normal contamination (*) as determined by karyotypic analysis were also excluded. All other cell populations were considered as non-clonal which were averaged across all centromere probes used and expressed as the N-CIN Index.
Table 2.3. Composite SKY Karyotypes of OCA27A/B and OCA714
Case SKY Karyotype
OCA27A/B 76-80<4n>,XXXX,der(X)t(X;8)(p22;q22),-1,del(1)(p34),der(2)dup(p11p15), der(2)t(2;17)(p11;11),del(4)(q12),der(5)t(1;5)(p11;q11),-6,i(8)(q10)X2,der(8) t(2;8)(p11;q11),-9,der(9)t(8;9)(q33;q11),der(12)t(12;22)(p11;q11),-13,-13, der(13)t(X;13)(q11;p11),der(13)t(8;13)(p12;p11),-14,i(14)(q10)X2,-16,-16, der(5;16)(q22;?),-17,-17,i(17)(q10),der(19)t(19;20)(p11;p11)X2,-20,-20,-22, -22,der(22)t(11;22)(q14;p11)X2{cp5}.
OCA714 41-46,X,-X,der(1)(13qter→13q21::dup1p21→p22→q11::5p11→5p14),der(2) t(2;3)(q14;?),der(2)(8qter→q21::2p16→q23::20?→20>::3?→3?),der(3)(3qter→3q10→3q10::5?→5?),i13(q10),der(4)(2?→2?::4p15→4q24::2?→2?::?→?),5,der(5) (14qter→14q21::5p12→15q34::8?→8?),der(6)t(6;11)(q10;q13),der(6)(pter→6q10::22?→22?::14q21→14qter),der(7)t(7;19)(q11;q10),der(7)del(p?)del(q?),+der(7) del(p21)del(q31),der(8)(8pter→8q10::13q21→13q34::6?→6?),i(9)(q10),der(9) t(6;9)(?q21:q10),der(10)t(8;10)(?;p14),der(11)(14q?→?::11p12→11q10::2?→2?),der(12)t(12;21)(q14;?),der(13)(3pter→p11::13p13→13q32::15q24→15q26), der(13)t(3;13)(p13;q10),der(14)(14p13→14q24::9?→?::14q24→14qter),-16, der(16)t(16;20)(q22;q13),-17,-18,der(19)(19pter→19q13.3::2q22→2q32),-22.
55
Chapter 3
Impact of Cytogenetic and Genomic Aberrations of the Kallikrein Locus in Ovarian Cancer
This Chapter represents the published manuscript entitled “Impact of Cytogenetic and Genomic Aberrations of the Kallikrein Locus in Ovarian Cancer” Bayani J, Paliouras M, Planque C, Shan SJ, Graham C, Squire JA, Diamandis EP. 2008. Mol Oncol. 3:250-60.
J.Bayani contributed to the development and experimental design of the study and manuscript; established primary cultures and maintained cell lines; identified, extracted and labeled BAC clones for FISH; performed SKY analyses and interpretation; performed aCGH experiments and interpretation; and designed primers and performed PCR. M. Paliouras and C.Planque created KLK-specific probes by long-PCR and cloning; performed KLK-specific ELISA and contributed to the manuscript. SJ. Shan provided technical assistance with the KLK-specific ELISA. C.Graham provided technical support for cell line maintenance; BAC extraction, labeling and quality control. JA Squire contributed to the development and design of the study and manuscript; in addition to granting support. EP. Diamandis contributed to the development and design of the study and manuscript; in addition to granting support.
56
3 Impact of Cytogenetic and Genomic Aberrations of the Kallikrein Locus in Ovarian Cancer
3.1 Introduction
Chromosomal changes, including copy number and structural rearrangements, are a hallmark of
many cancers, since they may significantly alter gene function. Classical cytogenetic and more
contemporary high-resolution integrative genomic strategies have identified genomic signatures
that are common and specific to many tumour types. These findings facilitated identification of
candidate tumour suppressor genes and oncogenes, as well as specific translocations as genomic
biomarkers. Serum biomarkers, such as prostate-specific antigen (PSA), also known as kallikrein
3 (KLK3), have already been shown to be important for screening and monitoring of prostate
cancer (Borgono and Diamandis, 2004). The overexpression of other members of the kallikrein
family in ovarian, breast and other cancers (Borgono et al., 2004) makes this protease family an
attractive source of new biomarkers. However, little is known about the relationship between
chromosomal and genomic aberrations of this gene family and protein expression. The human
tissue kallikreins (KLKs) are a family of secreted serine proteases. All 15 genes are located in
tandem on chromosome 19q13.4 and share, on average, 30–50% sequence identity (Borgono and
Diamandis, 2004; Borgono et al., 2004; Yousef and Diamandis, 2001). The KLKs are aberrantly
expressed in several cancer types; most strikingly, 12 KLKs are concurrently overexpressed in
ovarian cancer (Borgono and Diamandis, 2004). Accumulating evidence suggests that these
kallikreins are potential prognostic and diagnostic cancer biomarkers (Borgono and Diamandis,
2004; Diamandis and Yousef, 2002; Paliouras and Diamandis, 2007; Yousef and Diamandis,
2001) and that several may also be involved in cancer progression (Anisowicz et al., 1996;
Cohen et al., 1992; Ghosh et al., 2004; Magklara et al., 2003) In order to more thoroughly define
the roles of KLKs in cancer pathophysiology and clinical management, it is important to
investigate the mechanisms that influence their dysregulation in cancer. Somatically acquired
chromosomal aberrations are a major mechanism for gene activation. This is well-known in
hematologic malignancies, where specific and recurrent chromosomal translocations result in
oncogenic fusion transcripts.
Recent data indicate that common epithelial tumours, such as prostate cancer, may also harbor
recurrent translocations (Tomlins et al., 2006; Tomlins et al., 2005). Changes in gene expression
57
may be regulated by epigenetic modifications, as well as by gene copy number and chromosomal
translocations (Mitelman et al., 2007). The karyotypes of many carcinomas including ovarian
(Bayani et al., 2002; Rao et al., 2002; Taetle et al., 1999a; Taetle et al., 1999b), breast (Adeyinka
et al., 2000; Pandis et al., 1995; Rummukainen et al., 2001) and prostate cancers (Beheshti et al.,
2000; Wang et al., 2005) are complex and are characterized by numerical and structural
rearrangements. However, it is clear that more advanced molecular cytogenetic and genomic
mapping methods may be required to delineate the precise chromosomal rearrangements that
may influence the expression of specific loci. Chromosome 19 and the 19q13 region, which
harbors the KLK family of genes, are a site of frequent rearrangements in neoplasms
(http://cgap.nci.nih.gov/Chromosomes/Mitelman) and a site for copy-number imbalances in
breast and ovarian cancer, as determined by metaphase (mCGH) and array comparative genomic
hybridization (aCGH) (http://www.progenetix.net/cgi-bin/pgHome.cgi). Due to the known
overexpression of the KLK genes in ovarian, breast and prostate cancers, the objective of this
study was to investigate the copy number and structural rearrangement status of the KLK locus
by fluorescence in situ hybridization (FISH) and/or high-resolution oligonucleotide array
comparative genomic hybridization (aCGH), in ovarian, breast and prostate cancer cell lines and
ovarian cancer tumours.
3.2 Materials and Methods
3.2.1 Patient Specimens and Cell Lines
The ovarian cancer cell line CAOV-3; the breast cancer cell lines T47D, MCF-10A, MDA-MB-
468, MCF-7, BT474; and the prostate cancer cell lines LNCaP and 22RV1 were obtained from
the American Type Culture Collection (ATCC, Manassas, VA) and maintained according to the
product specifications. Excess ascites fluid from ovarian cancer patients, collected for routine
diagnostic testing, was used for short-term cultures. All patient specimens were collected and
handled with protocols approved by the Research Ethics Boards of the University Health
Network and Mount Sinai Hospital (Toronto). The ascites cultures were maintained in alpha-
MEM, supplemented with 15% (v/v) fetal calf serum (Invitrogen Canada, Burlington, ON,
Canada), 1% (v/v) penicillin–streptomycin (Invitrogen) and 1% (v/v) L-glutamine (Invitrogen).
In addition, heparinized whole blood from a normal female donor was cultured with
58
phytohemagglutinin for 72 h in a CO2 incubator in RPMI 1640 medium, supplemented with 20%
(v/v) fetal calf serum, 1% (v/v) L-glutamine and 1% (v/v) penicillin–streptomycin.
3.2.2 Fluorescence in situ Hybridization (FISH)
Cell line, patient and control lymphocyte cultures were prepared for cytogenetic analysis with
0.1 mg/mL Colcemid (Invitrogen) for 2–3 h, followed by osmotic swelling in KCl (0.075 M) and
fixed in methanol:acetic acid (3:1), as previously described (Bayani and Squire, 2004c). For
translocation studies, DNA from the BAC clones RP11- 76F7 and RP11-10I11 was extracted by
standard methods. These clones have been previously shown to include the following kallikrein
genes: KLK3, KLK2, KLK4, KLK5, KLK6, KLK7, KLK8, KLK9, KLK10, KLK11, KLK12, KLK13
and KLK14 (Yousef et al., 2000), (Figure 3.1). Other BAC clones used in this study include
RP11-716O8, RP11-451E12 and RP11-46I12. The extracted BAC DNA was directly labeled
with Spectrum Orange (Vysis/Abbott Laboratories, Des Plaines, IL) by nick translation using the
Vysis Nick Translation Kit, according to the manufacturer’s instructions. FISH to normal human
lymphocytes confirmed the genomic location to the 19q13.3/ 19q13.4 region. Approximately 300
ng of labeled probes were precipitated in excess human Cot-1 DNA (Invitrogen) and sonicated
salmon sperm DNA (Roche) and resuspended in a 50% formamide, 10% dextran sulfate and 2X
SSC hybridization buffer (DAKO, Mississauga, ON, Canada). The probes were heat-denatured
and hybridized to pepsin-treated and denatured normal human metaphase chromosomes, as
described previously (Bayani and Squire, 2004b). Following a rapid-wash technique consisting
of one wash in 0.4X SSC and 0.3% NP- 40 at 72oC for 3min, followed by a 5-min wash at room
temperature in 2X SSC and 0.1% NP-40, slides were mounted in DAPI/Antifade (Vector
Laboratories, Burlington, ON, Canada) and visualized with a Zeiss Axioskop fluorescence
microscope (Carl Zeiss Canada). For each cell line and patient specimen, metaphase preparations
were hybridized with both Spectrum Green-labeled, whole chromosome 19 paint (Vysis) and the
Spectrum Orange-labeled BAC probes and processed as described above. In each case, at least
10 metaphase spreads were analyzed. To determine the copy number and mapping status of
specific KLKs in a subset of tumours, PCR products of genomic fragments of KLK2, KLK4,
KLK6 and KLK13 were cloned into TOPO TA cloning vectors (Invitrogen). The DNAs were
extracted, labeled directly by nick translation and hybridized, as described above using either,
Spectrum Green, Spectrum Orange or the aqua-fluorescing-DEAC (Applied Biosystems).
59
3.2.3 Spectral Karyotyping
Metaphase preparations (described above) were pre-treated and hybridized with SKY Paints
(Applied Spectral Imaging, Carlsbad, CA) and processed as described elsewhere (Bayani and
Squire, 2002). The images were collected using a Zeiss Axioplan fluorescence microscope (Carl
Zeiss Canada) and processed using the Spectral Karyotyping Image Capturing and Analysis
system (ASI). Karyotypic descriptions were assigned according to the guidelines of the
International System for Human Cytogenetic Nomenclature (Shaffer and Tommerup, 2005). For
sequential SKY to FISH experiments, the coverslip was removed and the previously SKYed
slide washed in 0.3% NP-40/0.1X SSC for 5 minutes and dehydrated in an ethanol series. The
slide was then denatured in 70% formamide/2X SSC at 72 ˚C for 3 minutes and dehydrated.
Heat-denatured probe, as described above, was applied to the slide and allowed to hybridize
overnight at 37 ˚C. The slides were washed as described above.
3.2.4 Array Comparative Genomic Hybridization (aCGH)
The Agilent Human Genome 244K microarray platform was used (Agilent Technologies,
Inc.,Palo Alto, CA, USA cat. G4411B) containing 236,381 unique 60-mer oligonucleotide
features. The array design included a total of 5045 features used as internal controls. Features
were based on the UCSChg17 (NCBI build 35, May 2004) Build. DNA was extracted by
standard phenol:chloroform methods, RNase-treated and resuspended in sterile RNase/ DNase-
free water (Sigma Canada). The direct labeling of DNAs, probe evaluation, clean-up,
hybridization to the array and post-hybridization washes were carried out using reagents and
equipment, according to Agilent protocols (http://www.chem.agilent.com). Briefly, at least 3 μg
of male genomic DNA reference (Promega) and 3 μg of test genomic DNA samples were
digested with AluI (12.5 units) and RsaI (12.5 units) (Promega), then verified using the DNA
1000 LabChip Kit and Agilent 2100 Bioanalyzer, according to the manufacturer’s instructions
(Agilent Technologies). Individual reference and experimental samples were then filtered using
the Qiaquick PCR Cleanup Kit (Qiagen Inc., Hilden, Germany). Labeling reactions were
performed with 2 μg of purified digested DNA using the Invitrogen Bioprime labeling kit,
according to the manufacturer’s directions in a 50-μL volume, consisting of a modified dNTP
pool of 120 mM each of dATP, dGTP, dCTP, 60 mM dTTP, and 60 mM of either Cy5-dUTP for
60
the experimental sample or Cy3-dUTP for the 46, XY male reference (Perkin Elmer Life and
Analytical Sciences, Woodbridge, ON, Canada). ‘‘Dye-swaps’’ were carried out for all
experiments. Labeled targets were subsequently filtered using a Centricon YM-30 filter
(Millipore, Bedford, MA, USA). Experimental and reference targets for each hybridization were
pooled, mixed with 50 μg of human Cot-1 DNA (Invitrogen), 50 μL of Agilent 10X blocking
agent and 250 μL of Agilent 2X hybridization buffer. Prior to hybridization, the 500-μL
hybridization mixtures were denatured at 100˚C for 1.5 minutes and incubated at 37˚C for 30
minutes. The sample was applied to the array using an Agilent microarray hybridization chamber
and hybridization was carried out for 40 hours at 65˚C in a rotating oven (Robbins Scientific,
Sunnyvale, CA) at 20 rpm. The arrays were washed for 5 minutes at room temperature in 0.5X
SSPE/0.005% NLS, followed by 3 minutes at 37˚C in 0.1X SSPE/0.005% NLS. Slides were
dried using the Agilent drying solution and scanned using an Agilent 2565AA DNA microarray
scanner (Agilent Technologies). The CGH Analytics software version 3.4 (Agilent
Technologies) was used to analyze the aCGH data. Copy number aberrations were objectively
detected in replicate (dye-swap) experiments using an aberration calling method based on
computing significance scores for all genomic intervals. According to the Agilent CGH
Analytics software, a ratio of 0 represents the presence of two copies within the tumour/ test
genome, and a ratio of 1 represents the net gain of two additional copies (i.e. net gain of four
copies in the tumour/ test at 2n), hence the ratio of 0.5 suggests the net gain of one copy, based
on a normal diploid genome (i.e. three copies at 2n or five copies at 4n).
3.2.5 Measurement of KLKs by ELISA
We have previously described our ELISA methodologies for all KLKs. Briefly, the assays are
based on ‘‘sandwich’’-type ELISA principles with one antibody used for capture and one for
detection. More details on procedures and performance can be found elsewhere (Shaw and
Diamandis, 2007). All assays are highly specific, with no cross-reactivity from other KLKs. For
KLK analysis of cell lines, we used supernatants collected after 7 days of culture. No steroid
hormonal stimulation was used. Ascites fluid was analyzed after centrifugation, to remove cells.
Primary cultures of ascites cells were maintained for 1–7 days and the supernatants were
collected for analysis. Also, cell pellets were lysed and analyzed in a similar fashion. When solid
61
tumour was available (one case, OCA19), tumour tissue was homogenized and KLKs extracted
in a lysis buffer containing NP-40 surfactant. All KLK values were expressed as μg/L.
62
Figure 3.1 Summary of FISH findings of the KLK locus in cancer cell lines and ovarian cancer patients. (A) Schematic illustrating the BAC clones used in this study. (B) Summary of FISH studies in cell line and tumors. Chromosome 19 was analyzed using the paint probe WCP19 and BACs that cover most of the 325-kb KLK genomic region using dual color FISH. Closed circles indicate the presence of the KLK locus located at the resident chromosome 19q site. Open circles indicate the presence of the KLK locus associated with either an intra-chromosomally or interchromosomal structural rearrangement. Also indicated for each analysis is the ploidy established by chromosomal counts. MCF10A and 22RV1 did not show copy number changes or involvement of translocation. In 70% of cells, T47D also showed a normal diploid pattern, but 30% showed a whole chromosomal gain of 19. LNCAP showed four copies of chromosome 19 with four copies of the KLK locus, showing no net gain over ploidy, but two copies over a normal diploid cell. The remaining cell lines and tumors displayed net gains of the KLK locus by unbalanced translocation
63
3.3 Results
3.3.1 Fine Structural Analysis of Chromosomal Rearrangements
Affecting 19q13 in Cancer Cell Lines
Following the confirmatory mapping of the RP11-76F7 and RP11-10I11 BAC clones to normal
human lymphocyte metaphase spreads (data not shown), a dual-color FISH strategy was used to
identify the presence of translocations of the 19q13 region in various cancer cell lines. For each
case, the combined Spectrum Orange-labeled 19q13 BACs (RP11-76F7 and RP11-10I11) were
hybridized with a Spectrum Green-labeled whole chromosome 19 painting probe (WCP19).
Rearrangements were identified when the red KLK FISH signals (and the corresponding WCP19
signal), were associated with chromosomes that did not contiguously hybridize with the whole
chromosome 19 paint. Using this strategy, numerical and structural rearrangements involving the
KLK locus were identified in the ovarian cancer cell line CAOV-3 and the breast cancer cell
lines, MDA-MB-468, MCF-7 and BT474 (Figure 3.1B). No structural changes involving 19q
were identified in the prostate cancer cell lines 22RV1 and LNCaP, consistent with previous
SKY analysis (Beheshti et al., 2000) or the breast cancer cell lines MCF10A and T47D. In
addition, for these cell lines, there was no change in copy number relative to the ploidy of the
genome. Thus, 22RV1 and MCF10A maintained two copies of the KLK locus within the diploid
genome and LNCaP possessed four copies of the KLK locus against a tetraploid genome.
However, T47D, with a pseudo-triploid karyotype, was shown to possess a population of cells
with an additional chromosome 19. Approximately 70% of cells enumerated possessed two
copies of the KLK locus within chromosome 19, with the remaining 30% of cells possessing
three copies of the KLK locus, due to a gain of chromosome 19. Sequential SKY and FISH
analysis was performed on CAOV-3, MDA-MB-468 and MCF7 (Figure 3.2) to determine the
position of rearrangements relative to the KLK genes and to assign copy numbers for the gene
cluster in each cell line. CAOV-3 was shown to be a hypertriploid line showing many complex
structural rearrangements (Figure 3.2A). Two KLK signals were located at their usual 19q
location, whilst another two KLK loci were involved in a duplicated complex, unbalanced
translocation involving chromosomes 19, 2 and 13. These paired rearrangements suggest that this
structural abnormality occurred prior to tetraploidization. Similarly, in MDA-MB-468, two KLK
signals were present at their usual chromosome 19q location, along with an aberrant KLK signal,
64
due to an unbalanced translocation with chromosome 20 (Figure 3.2B). MCF-7, a hypotriploid
line, possessed three unbalanced translocations and only one apparently normal chromosome 19.
One of the rearrangements, the der(19)t(17::11::19), appears to be a derivative of the
der(19)t(11;19) and the other rearrangement is an unbalanced translocation with portions of 12q
translocated adjacent to the KLK locus (Figure 3.2C). The hypertetraploid BT-474 cell line
showed two chromosomes 19 containing the KLK locus, and three additional unbalanced
translocations, each containing the KLK locus, consistent with previously published SKY
findings (Kytola et al., 2000). The unbalanced rearrangements included a der(7)t(7:19),
der(10)t(10;19), and der(8)t(8::20::19) (data not shown).
65
Figure 3.2. KLK status in cancer cell lines by sequential SKY and FISH. Shown are representative metaphases of CAOV-3 (A), MDA-MB-468 (B) and MCF-7 (C) by SKY and FISH using the KLK containing BACs (red). The affected chromosomes are indicated by arrows and shown in the inset.
66
3.3.2 Chromosomal Rearrangements of 19q13 in Ovarian Cancer
Patient Samples
To determine whether the genomic and chromosomal alterations affecting the KLK and the
19q13 region in cancer cell lines were also present in ovarian cancer patient tumours, FISH
analysis was performed on cytogenetic preparations from three representative ascites fluid
samples from ovarian cancer patients. In all three samples, copy number imbalances of the KLK
region were detected (Figure 3.1B). In Patient 1, two copies of the KLK locus were identified at
the resident chromosome 19 site. An additional signal was detected in an unbalanced
rearrangement. For Patient 2, the presence of different cell populations based on chromosome
number (ploidy) were identified, however FISH identified the presence of at least one
chromosome 19 containing the KLK locus at its usual location, and the presence of up to two
additional KLK signals involved in translocations (Figure 3.3A, B). In addition, inserted portions
of chromosome 19 were detected throughout the karyotype. Spectral karyotyping (SKY) of this
patient sample revealed a karyotype possessing complex structural changes, as illustrated in
Figure 3.3C. Aberrations involving chromosome 19 were identified in at least six chromosomes
with one reflecting an intact 19, consistent with initial FISH results. Patient 3 also displayed
genomic heterogeneity, possessing at least two populations of cells, primarily an abnormal
diploid population (~30%) and a tetraploid population (~60%). The tetraploid population was
found to be similar to the diploid clone, suggesting that it arose from duplication of a diploid
progenitor following failed cytokinesis. FISH identified the presence of five KLK signals within
the tetraploid population, and three signals within the diploid population, yielding a net gain of
the KLK locus per cell. Four of the five signals were shown to result from two isochromosomes
19(q) (Figure 3.3D,E arrow) with the remaining signal localized at the resident chromosome 19q
(Figure 3.3F). Other additions and insertions of chromosome 19 material were also seen within
the genome (Figure 3.3F) and confirmed by SKY analysis (Figure 3.3G). In addition, five
primary, untreated ovarian cancers, previously described (Bayani et al., 2002), by SKY and
mCGH were analyzed to determine whether specific alterations leading to KLK imbalances and
structural rearrangements were also present (Figure 3.1B). The BAC RP11-716O8, mapping to
19q11/12, was used to identify affected chromosomes. In all five tumours, copy number
imbalances, either due to whole copy number gains of chromosome 19 or unbalanced
translocations were identified. Moreover, complex structural rearrangements, including a ring
67
chromosome in OCA5 (Figure 3.3H, I) were shown to include the KLK locus, as well as the
detection of i(19)(q) in OCA19. In many cases, the heterogeneity of these short-term cultures
was revealed, with some tumours showing low-level copy number variation due to clonal and
non-clonal changes, as shown by the analysis of interphase nuclei (OCA21A; Figure 3.3J).
68
Figure 3.3 Status of the KLK locus in patient ascites. In each case, WCP19 (green) was hybridized with BACs for the KLK locus (red) to metaphase preparations. (A–C) Patient ascites 2. (A and (B) show the presence of chromosome 19 material (green) dispersed throughout the genome (A), with one KLK signal mapping to the resident chromosome 19, with the remaining two copies involved in translocations (B) (arrows). (C) Integrative FISH and SKY analysis of affected chromosomes. Inverted DAPI banding in shows the banding pattern of the metaphase chromosomes hybridized with the SKY paints (left) with FISH of corresponding chromosomes showing chromosome 19 paint (green) and the KLK locus (red) from A/B (left). By SKY complex rearrangements involving two or more chromosomal partners can be seen through the change of color along the length of a contiguous chromosome. The RGB (red–green–blue) display color for chromosome 19 is bright green, with a normal chromosome 19 (top–bottom), followed by five abnormal chromosomes involved in structural rearrangements with other chromosomal partners. The corresponding FISH using the KLK and whole chromosome 19 painting probes confirmed that structurally rearranged chromosomes by SKY also included the KLK locus. (D–G) Patient ascites 3 shows the presence of a KLK signal at the resident chromosome 19 and the presence two isochromosomes 19q (i(19q)) (D,E). The net result is the gain of KLK copy number due to these isochromosomes. (F) Integrative SKY and FISH analysis revealed the involvement of chromosome 19 in translocations with chromosome 3, but without involvement of the KLK locus. (G) Representative SKY metaphase of patient ascites 3. (H,I) Sporadic and untreated primary cancer OCA5. Shown in red is the KLK locus and a BAC clone for the 19q12 region, shown in green revealing the net gain of the KLK locus through whole gains of chromosome 19 and the presence of a multi-centric (green) ring chromosome containing the KLK locus (red). (I) Integrative SKY and FISH of OCA5. SKY and FISH confirms that no other chromosomes are involved in structural rearrangements on the three chromosomes 19, but the possibility exists for other chromosomal fragments in the ring chromosome. (J) Interphase nuclei of OCA21A shows variability in copy numbers of the KLK locus from cell to cell.
69
3.3.3 High-Resolution aCGH and KLK-specific FISH Analysis Confirms
Copy-Number Imbalances
Since the RP11-76F7 and RP11-10I11 BACs span the entire KLK family locus, aCGH using the
Agilent oligonucleotide platform was performed on the BT474 cell line to determine the copy
number status of each KLK. Array CGH of BT-474 showed a net gain of 19q12-19qter, with a
ratio of approximately 0.5 (Figure 3.4), suggesting an average copy number of five per cell.
These findings are consistent with initial FISH findings using the combined BAC probes (Figure
3.1, Table 3.1). To confirm the aCGH findings from this study and others (Neve et al., 2006), a
multi-color KLK-specific FISH strategy was followed in a subset of lines (Figure 3.4, Table 3.1).
Figure 3.4 illustrates the KLK-specific FISH findings in BT474, which confirm the copy number
gain and the movement of the entire KLK family locus in unbalanced translocations. Based on
these aCGH findings and to those published by others (Shinoda et al., 2007), it was also
determined that among the cell lines analyzed, no specific breakpoint locus within 4 Mb
centromeric to the KLK locus was identified, with most imbalances arising from whole copy
number alterations of 19q. FISH using BAC clones within the 19q13.3 and 19q11/19q12 regions
confirmed these observations (data not shown).
Table 3.1: KLK-Specific FISH: Analysis of Cell Lines
Specimen Number of Copies per Cell
KLK2 KLK4 KLK6 KLK13
NHL (2n) 2 2 2 2
MCF-10A (2n) 2 2 2 2
T47D (2n and pseudo 3n) 2
3
2
3
2
3
2
3
BT474 (4n+) 5 5 5 5
MDA-MB-468 (2n+) 3 3 3 3
NHL= Normal human lymphocytes; 2n= normal diploid genome.
70
Figure 3.4 aCGH and FISH validation of copy number imbalances of the KLK locus in BT-474. (A,B) aCGH for chromosome 19. Regions of gain are highlighted in green with regions of loss indicated by red, and no net change in yellow. (B) Magnification of the KLK locus shows an average ratio approaching +0.5 indicating a net gain of the region. Due to the ploidy of the genome, the net gain suggests one copy over ploidy, thus1+ 4n =5 copies of the KLK locus. Side panels: FISH Validation of individual KLKs in BT474. A three-color FISH approach was used for KLK2 (green), KLK4 (red) and KLK6 (aqua) to BT474 metaphase spreads. Bottom panel: KLK13 validation in BT474. In this three-color experiment, KLK2 (green), KLK6 (blue) and KLK13 (red) were co-hybridized and also shown to co-localize and be present in copy numbers consistent with aCGH findings.
3.3.4 Protein Expression and Relationship to Copy-Number and
Structural Rearrangements in Cancer Cell Lines and Ovarian
Specimens
KLK protein levels were measured prospectively in all cell line tissue culture supernatants and
ascites fluid; and retrospectively, in the primary ovarian cancer specimens, either in cultured cell
supernatants and/or cell lysates, or in tumour extracts (Table 3.2). Of the breast cancer lines,
MCF-10A, a near normal breast epithelial cell line produced relatively very low levels of all
KLKs, with highest expression detected for KLK10. Similar findings were found for T47D, with
low protein levels for all KLKs tested, with the exception of KLK9 and KLK10. Neither line
possessed structural rearrangements involving the KLK locus, nor were there any copy number
71
changes detected in MCF-10A. T47D, as discussed earlier, did possess a population of cells
(30%) containing the whole gain of chromosome 19. The remaining breast and ovarian lines
showed variable KLK protein expression (higher levels of KLK proteins are shown in brackets).
CAOV-3 (KLKs 5, 6, 7, 8, 9, 10); MDA-MB-468 (KLK 5, 6, 7, 8, 9, 10); BT-474 (KLK9);
MCF-7 (KLKs 6, 7, 9, 11); of which all were associated with copy number changes of the KLK
locus due to unbalanced translocations. ELISA results of the prostate cancer cell lines LNCaP
and 22RV1 revealed significant amounts of only KLKs 2, 3 and 9, consistent with previous
findings (Borgono and Diamandis, 2004; Yousef and Diamandis, 2001). The cytogenetic results
of these lines revealed normal copy number for 22RV1 and no net change of copy number for
LNCaP, based on the ploidy status of the cell. Moreover, no structural change of the KLK locus
was identified among these two lines. Analysis of prospective and retrospective ovarian
specimens showed that patient ascites 1 contained high levels of KLKs 1, 5, 6, 7, 8, 9, 10, 11;
patient ascites 2 contained high levels of KLKs 5, 6, 7, 8, 10, 11; patient ascites 3 contained high
levels of KLKs 5, 6, 7, 8, 10, 11 (italic and bold values indicate levels > 10 μg/L). ELISA results
of retrospective ovarian cancer samples were associated with high levels of the KLKs 5, 6, 7, 8,
9, 10 and 11. In all cases, copy number changes were associated with whole chromosomal gains
of chromosome 19 and/or unbalanced translocations of the KLK locus, suggesting that in breast
and ovarian cancers, copy number gains of the KLK locus, particularly through unbalanced
translocations, may be a mechanism for the high expression of KLKs 5, 6, 7, 8, 9, 10 and 11.
72
Table 3.2 ELISA findings for cancer cell lines and patient specimens1
1. Data for KLKs 13, 14 and 15 are not shown, since the values were low in all cases. All values are in µg/L ± standard deviation. KLK12 was not measured; no ELISA assay available; specimen description is given under ‘methods’. 2. Indicated in brackets is the ploidy of the cell, as determined by cytogenetic analysis. 3. sup, supernatant
Specimen KLK1 KLK2 KLK3 KLK4 KLK5 KLK6 KLK7 KLK8 KLK9 KLK10
KLK11
CaOV3(3n+)2 <0.2 <0.005 <0.005 <0.2 227±17 14.5±0.75
3.9±0.27 12.4±0.47 3.2±0.3 50±16 <0.2
MCF‐10A(2n) <0.2 <0.005 <0.005 <0.2 0.25±0.04 <0.2 0.4±0.06 <0.2 <0.5 2.8±1.3 <0.2T47D (2n+, pseudo 3n) <0.2 <0.005 <0.005 <0.2 <0.05 <0.2 <0.2 0.2±0.02 13.8±2.6 0.7±0.04 0.08±0.02 MDA‐MB‐468 (2n+) <0.2 <0.005 <0.005 <0.2 640±29 103±14 1.3±0.2 2.1±0.06 39±2.4 21±3.5 0.21±0.15 BT‐474 (4n+) <0.2 <0.005 <0.005 <0.2 <0.05 <0.2 <0.2 <0.2 13±0.2 <0.05 0.11±0.02 MCF‐7 (3n‐) <0.2 <0.005 <0.005 <0.2 <0.05 0.3±0.0
7 0.44±0.19 <0.2 13±2.5 <0.05 2.3±0.6
LNCaP (4n) <0.2 28±7.0 410±23 0.13±0.09 <0.05 <0.2 <0.2 <0.2 16±5.0 <0.05 <0.222RV1 (2n) <0.2 5±2.5 68±50 <0.2 <0.05 <0.2 <0.2 <0.2 9±4.8 <0.05 <0.2Ascites Patient 1 (2n) 6.0 <0.005 <0.005 <0.2 3.8 84 0.62 5.0 1.2 17 97Patient 2 (3n/6n) 0.75 <0.005 <0.005 <0.2 41 388 2.5 81 0.65 65 94Patient 3 (2n+/4n+) 1.3 <0.005 <0.005 <0.2 11.9 80 9.4 14 0.2 35 90OCA5 (3n+) ‐1 week culture sup3
<0.2 <0.005 <0.005 <0.005 <0.05 0.4 0.3 0.2 5.8 0.5 <0.2
OCA8 (2n+) ‐1 week culture sup
<0.2 <0.005 <0.005 <0.005 2.4 1.3 2.4 0.8 7.2 7.2 2.1
OCA19 (4n) ‐tumor tissue extract
<0.2 <0.005 <0.005 <0.005 0.6 0.1 <0.2 0.1 0.8 0.2 0.8
OCA21 (4n) –pelleted cultured cancer cells
<0.2 <0.005 <0.005 <0.005 17 16 34 11 2.2 30 64
Patient OCA27 (4n) ‐1 week culture sup
<0.2 <0.005 <0.005 <0.005 14 18 4.7 34 2.6 20 40
73
3.4 Discussion
The human tissue KLKs are a family of secreted serine proteases which are often co-expressed
within the same tissues and thought to participate in enzymatic cascades (Borgono et al., 2004).
The KLKs are aberrantly expressed in several cancer types and they are potential biomarkers for
diagnosis, prognosis and prediction of therapy response (Borgono et al., 2007; Diamandis and
Yousef, 2002; Paliouras et al., 2007; Yousef and Diamandis, 2001). The regulation of gene
expression appears to be influenced by steroid hormones (Paliouras and Diamandis, 2007) and
epigenetic changes (Pampalakis et al., 2006; Pampalakis and Sotiropoulou, 2006; Sidiropoulos et
al., 2005); however, the mechanisms that mediate aberrant KLK expression in cancer have not
been entirely delineated. For many neoplasms, altered gene expression is linked to gene copy
number, as in the case of PTEN in prostate cancer (Yoshimoto et al., 2006) and EGFR in lung
cancer. To date, there has been no cytogenetic study that has examined the copy number status or
the possible translocation of KLK gene members, in association with their relative protein
expression in ovarian, breast or prostate cancer cell lines or tissues. Recently, Shinoda et al.
reported the overexpression of KLK5 in urinary bladder carcinoma cell lines, which were
associated with changes in gene copy number (Shinoda et al., 2007), and Ni et al. (Ni et al.,
2004) demonstrated by Southern blot and RT-PCR analysis that the overexpression of KLK6 in a
subset of ovarian cancers was associated with copy number gains. In this study, we analyzed the
copy number, positional mapping status and protein level of the KLK genes in breast, ovarian
and prostate cancer cell lines. Additionally, since many KLKs are overexpressed in ovarian
cancers (Borgono et al., 2007; Diamandis and Yousef, 2002; Yousef and Diamandis, 2001), a
cohort of primary tumours and ascites samples was subjected to detailed cytogenetic and
genomic analyses. FISH and SKY analysis of the eight ovarian cancers studied revealed that the
KLK region was frequently involved in copy number gains in the form of unbalanced
translocations. Similarly, FISH analysis of all three patient ascites samples and five ovarian
carcinomas identified the presence of copy number changes of the KLK locus due to unbalanced
rearrangements. Analysis of the KLK cluster by aCGH for the breast cancer cell line BT474
showed that imbalance of a large region of chromosome 19 was associated with KLK copy
number increases. To determine whether rearrangements took place within the KLK gene locus,
74
cluster-specific FISH for KLK2, KLK4, KLK6 and KLK13 were used to confirm the contiguous
presence of five copies of each of these genes. These findings suggested that copy number
changes usually affect the entire KLK locus (approx. 300 kb) and that elevated expression is not
associated with genomic disruption of the cluster. Similar findings were revealed for the other
cell lines tested (Table 3.1). A major advantage of aCGH is the ability to ascertain the loci that
are involved in breakage, resulting in copy number change, revealing translocation breakpoints
(Selvarajah et al., 2006). Our aCGH data, as well as those of others (Neve et al., 2006), have
shown that the majority of the copy number transition sites in these cell lines occur centromeric
to the KLK locus at 19q12, a region closely associated with the centromere. This suggests that in
a subset of tumours the copy number and unbalanced translocation events likely occur at the
level of the whole chromosome arm. FISH analysis of this region confirmed these observations,
however, also implicate the presence of more subtle genomic aberrations between 19q12 and a 4-
Mb region centromeric to the KLK locus, in BT474, MDA-MB-468 and CAOV-3 cells. These
data draw attention to the possibility that chromosomal alterations, or gains in the genomic
vicinity of the KLK locus, may indirectly contribute towards elevating the transcript and protein
levels of kallikreins. However, we consider it less likely that a classical oncogenic fusion event
involving the KLK cluster itself is taking place, since no common translocation partner or
recurrent rearrangement was observed in the cell lines and ovarian tumours with 19q13
alterations. KLK protein analysis revealed a relationship between copy number gains of the KLK
locus and KLK gene family members. The most consistent data were seen in the prospective
analysis of ascites fluid, whereby high levels of KLKs 5, 6, 7, 8, 10 and 11 were observed (Table
3.2). Similar elevations were also detected in CAOV-3 and MDA-MB-468 (Table 3.2),
suggesting that copy number changes may influence the expression of genes in this region. The
retrospective analysis of cryopreserved primary ovarian cancers and primary cultures also
revealed elevations in KLK protein, although at lower levels, in comparison to freshly obtained
specimens. Human chromosome 19, although being one of the smallest chromosomes, is the
most gene-rich (Venter et al., 2001) and possesses the highest number of CpG islands (Craig and
Bickmore, 1994). The lack of constitutional rearrangements involving chromosome 19 appears
to correlate to the finding of selection against translocations involving gene-dense chromosomes
(Bickmore and Teague, 2002). Since chromosome 19 possesses a high CpG content,
transcriptional regulation by methylation-dependent mechanisms may be compromised as a
result of the translocation event. Indeed, studies have demonstrated the transcriptional silencing
75
of genes due to the introduction of heterochromatin. Changes in the spatial orientation of
chromosomal domains within the nucleus may also influence the extent of gene transcription
(Cremer et al., 2000; Cremer et al., 2004; Sadoni et al., 1999). Furthermore, the extent of
chromatin compaction at the translocation site/region could enhance or hinder the access of
transcription factors or other regulatory binding sites, offering an attractive mechanism for the
aberrant expression of KLKs. The translocation event can also alter the proximity or sensitivity
to hormone responsive elements, which may have important implications for breast, ovarian and
prostate carcinogenesis. The current lack of KLK6 knock-out murine models precludes our
ability to ascertain whether KLK6 has “driver” characteristics (Stratton et al., 2009) or is merely
a “passenger”. In conclusion, in this preliminary analysis of cell lines and patient specimens, we
have demonstrated that there is an association between copy number changes and changes in
protein expression of certain KLK family members, and that these copy number imbalances
primarily result from unbalanced translocations. Moreover, we have shown in this initial survey
that these events are extremely common and may be characteristic of late stage and progressive
ovarian cancer, however additional studies are necessary. The careful examination of early stage
disease and other malignancies will also be required, to determine whether these genomic
changes have prognostic or predictive value.
76
Chapter 4
Genomic Instability and Copy-Number Heterogeneity of Chromosome 19q, Including the Kallikrein Locus, in
Ovarian Carcinomas
This Chapter represents the published manuscript entitled “Genomic Instability and Copy-Number Heterogeneity of Chromosome 19q, Including the Kallkrein Locus, in Ovarian Carcinomas” by Bayani J, Marrano P, Graham C, Zheng Y, Li L, Katsaros D, Lassus H, Butzow R, Squire JA, Diamandis EP. 2011. Mol. Oncol. 1:48-60.
J.Bayani contributed to the development and experimental design of the study and manuscript; identified, extracted and labeled BAC clones for FISH; performed FISH to TMAs and paraffin sections; performed and interpreted all IHC. P.Marrano provided technical support for IHC. C.Graham provided technical support for BAC extraction, labeling and quality control. Y.Zheng and L.Li performed statistical analyses. D.Katsaros, H. Lassus and R.Butzow provided patient specimens and clinical information. JA. Squire contributed to the development and design of the study and manuscript. EP. Diamandis contributed to the development and design of the study and manuscript; in addition to granting support.
77
4 Genomic Instability and Copy-Number Heterogeneity of Chromosome 19q, Including the Kallikrein Locus, In Ovarian Carcinomas
4.1 Introduction
Ovarian carcinoma (OCa) is a heterogeneous disease, reflected by distinct histopathological
subtypes, and clinically characterized according to stage and grade (reviewed by Bast et al.,
2009). A growing body of evidence suggests distinct molecular pathways may mediate the
development and characterization of low-grade and high-grade serous tumours; among them, the
relative differences in genomic instability and mutations of PTEN and KRAS in low-grade
tumours; and mutations and aberrant expression of TP53 among high-grade tumours (Bast et al.,
2009; Shih Ie and Kurman, 2004). Genomically, OCas exhibit complex numerical and structural
chromosomal alterations (Bayani et al., 2002; Pejovic et al., 1992a; Pejovic et al., 1990; Pejovic
et al., 1992b; Shridhar et al., 2001; Taetle et al., 1999a; Taetle et al., 1999b; Thompson et al.,
1994a; Thompson et al., 1994b), and are prone to chromosomal instability (CIN) (Bayani et al.,
2008b; Gorringe et al., 2005). CIN is mediated by various mechanisms (reviewed by Bayani et
al., 2007), resulting in chromosomal abnormalities which can be broadly classified as numerical
chromosomal instability (N-CIN) or structural chromosomal instability (S-CIN) (Bayani et al.,
2007). The consequences of gross numerical and structural chromosomal rearrangements
include changes in gene sequence, gene/locus-specific dosage (Frohling and Dohner, 2008), as
well as epigenetics (Sadikovic et al., 2008); which all contribute to the pathogenesis of this
disease. Chromosome 19q is a site of frequent rearrangements
(http://cgap.nci.nih.gov/Chromosomes/Mitelman), and copy number imbalances
(http://www.progenetix.net/) in many neoplasms, including OCa. Classical cytogenetic analyses
(Taetle et al., 1999a; Taetle et al., 1999b), and more recent molecular cytogenetic findings
(Micci et al., 2009; Tsao et al., 2001) suggest that the non-random numerical and structural
alterations of chromosome 19 play an important role in ovarian carcinogenesis.
In a small pilot study, we recently investigated the role of copy-number of the kallikrein (KLK)
genes in a series of cancer cell lines and primary ovarian cancers shown to possess KLK6
overexpression (Bayani et al., 2008b). In all primary ovarian cancer specimens, a net gain of the
entire KLK locus (19q13.3/4) was identified, either by the whole gains of chromosome 19 or
78
through unbalanced translocations involving breakpoints centromeric to 19q13.3/4, suggesting a
role of copy-number in the observed elevated KLK6 protein levels. The KLKs are a family of 15
genes and 1 pseudogene located in tandem on chromosome 19q13.3/4 (Yousef et al., 2000),
which encode trypsin-like serine proteases. They cleave a variety of substrates including MMPs,
IGFBPs, fibronectins and collagens (Borgono and Diamandis, 2004) and seem to be involved in
many of the pathways deemed the hallmarks of cancer (Hanahan and Weinberg, 2000, 2011).
Due to the widespread and successful use of PSA (also known as KLK3) as a biomarker for the
detection and monitoring of prostate cancer (Lilja et al., 2008), the potential for other KLKs as
biomarkers in ovarian (Kim et al., 2001; Kyriakopoulou et al., 2003; Shan et al., 2007; Yousef et
al., 2003b) and other cancers has been actively pursued (Inoue et al., 2010; Li et al., 2009b;
Mavridis and Scorilas, 2010; Nathalie et al., 2009; Pettus et al., 2009; Planque et al., 2005;
Shinoda et al., 2007). The observed differential expression of the KLK genes has also fueled
numerous studies investigating the mechanism regulating their expression, including hormone
stimulation (Lai et al., 2009; Shan et al., 2007; Shaw and Diamandis, 2007) differential
methylation (Pampalakis et al., 2008; Pampalakis et al., 2009; Pampalakis and Sotiropoulou,
2006; Sidiropoulos et al., 2005) and microRNAs (miRNAs) (Chow et al., 2008; White et al.,
2010a), with few studies investigating copy-number (Bayani et al., 2008b; Ni et al., 2004;
Shinoda et al., 2007). These variable findings suggest that the KLKs may be regulated by
multiple mechanisms.
Among the KLK gene family, KLK6 has been demonstrated to be a promising biomarker for OCa
(Borgono et al., 2006; Hoffman et al., 2002; Obiezu et al., 2001; Shan et al., 2007; White et al.,
2009b; Yousef et al., 2003b). Recent studies by our group (Shan et al., 2007) have shown up to
57-fold higher KLK6 protein expression in cancer specimens compared to normal tissues; with
univariate and multivariate survival analyses showing that patients deemed KLK6-positive, are at
an increased risk of relapse, in comparison to KLK6-negative patients. A significant association
with progression-free survival (PFS) and overall survival (OS) was also demonstrated.
While the assessment of KLK6 as a novel biomarker in OCa continues, the mechanisms leading
to the observed protein overexpression of KLK6 are still unclear (Borgono et al., 2006). Our
previous findings (Bayani et al., 2008b) suggest that copy number may contribute to such
overexpression, and the numerical and structural complexity of OCa karyotypes influences copy
number change. Thus, in this investigation, we have expanded the study to specifically
79
investigate the overall frequency of KLK locus-specific copy-number changes, and to correlate
these findings to KLK6 expression in a series of untreated serous ovarian carcinomas.
Furthermore our experimental design enabled us to examine the frequency of structural and
numerical CIN on 19q, and its implications for understanding ovarian cancer pathogenesis.
4.2 Materials and Methods
4.2.1 Patient Specimens
Seventeen formalin-fixed paraffin embedded (FFPE) sections were obtained from the
Department of Gynecology and Obstetrics, University of Turin, Italy; and an ovarian tissue
microarray (TMA), consisting of 102 patients was obtained from the Department of Pathology at
the Helsinki University Central Hospital (C2 block); and were collected with consent according
to the guidelines of research ethics boards from all institutions. All specimens were classified
histologically as serous tumours derived from patients naïve to chemotherapy. The tissues were
fixed according to standard procedures and the TMA was constructed as previously described
(Lassus and Butzow, 2007). FFPE normal ovarian tissues were obtained from US Biomax
(Rockville, MD, catalog number HuFPT076).
4.2.2 Fluorescence in situ Hybridization (FISH)
For copy-number and instability studies, the following bacterial artificial chromosome (BAC)
clones mapping to three regions of chromosome 19q were obtained from The Centre for Applied
Genomics (Toronto, Canada): 19q12 (RP11-888D2, RP11-197B9, RP11-716O8 and RP11-
1107F24, 19q13.2 (RP11-67A5, RP11-264N23, RP11-825A10), and 19q13.3/4 (RP11-288H1,
RP11-10I11, RP11-615L12). The 19q13.3/4 overlapping BACs span the entire KLK locus to
include KLK1 at the most centromeric end to span all KLK gene family members; and including
CD33 at the telomeric end. To confirm that KLK6 was represented in the overlapping BAC
clones RP11-10I11 and RP11-615L12, KLK6- specific PCR was performed using the following
primer set designed using Primer 3 (http://frodo.wi.mit.edu/): KLK6 forward
GGGGTCCTTATCCATCCACT and KLK6 reverse cagtcgcatctgctgttcat, to yield a 90bp band.
Using the Illustra Hot Start Master Mix (GE Healthcare Life Sciences), 100 ng of BAC DNA
were tested, in addition to proper positive and negative controls. DNA extracted from the BAC
80
clones was labelled with either Spectrum Green (Vysis/Abbott Laboratories, Des Plaines, IL),
Spectrum Orange (Vysis/Abbott), or blue-fluorescing DEAC (Applied Biosystems) by nick
translation using the Vysis Nick Translation Kit (Vysis/Abbott), according to the manufacturer’s
instructions. FISH to normal human lymphocytes (Bayani and Squire, 2004b) confirmed the
genomic location of all BACs. All 19q12 overlapping clones (RP11-888D2, RP11-197B9,
RP11-716O8 and RP11-1107F24) were labeled with Spectrum Green. Similarly, all 19q13.2
clones were labeled with DEAC; and all 19q13.3/4 clones which span the KLK locus, were
labeled with Spectrum Orange. Approximately 300 ng of each labeled DNA were precipitated in
excess Human Cot-1 DNA (Invitrogen, Canada), sonicated salmon sperm DNA (Roche, Canada)
and resuspended in a 50% formamide/10% dextran sulfate/2X SSC hybridization buffer (DAKO,
Mississauga, ON, Canada) to a final volume of 35μl (Bayani and Squire, 2004b). All tissues
were baked overnight at 56oC. The following day, the slides were de-waxed in xylene and
dehydrated in 100% ethanol and pre-treated prior to co-denaturation and hybridization with the
probe cocktail. The following day, the slides were processed in a wash of 0.3% NP-40/0.4XSSC
for 2 minutes at 72oC and a wash of 0.1%NP-40/2XSSC for 5 minutes at RT. The slides were
rinsed in 1XPBS, mounted in a DAPI/Antifade medium (Vectashield/ Vector Laboratories
Canada), and visualized at 60X with a Zeiss Axioskop fluorescence microscope (Carl Zeiss
Canada).
4.2.3 Measurement of Chromosomal Instability
For each case/core at least 50-100 tumour nuclei were scored for the presence of each signal. A
normal diploid cut-off was established using normal tissues, resulting in at least 65% of cells
with 2 signals for each probe and 35% containing 1 signal (due to tissue sectioning). Greater
than 40% of cells with 1 signal indicated the presence of a population with a net loss. The
presence of additional populations were noted when the frequency of positive cells exceeded
10% per population (or cumulatively exceeding 10% if several populations were detected).
Using a modification of chromosomal instability index previously published (Bayani et al.,
2008a) the designation of low, medium or high chromosomal instability (CIN) was assigned
when 1 or 2 populations of cells were identified (low), 2-3 populations of cells were identified
(medium), or greater than 3 populations of cells or amplification (high) were detected.
Numerical changes of 19q: Whole copy-number changes of 19q were identified when the
81
percentage of signals per loci along 19q were similar. Structural changes of 19q: Evidence of
structural rearrangement on 19q was identified when the percentage of signals per cell, for each
of the loci tested, were discordant by greater than 10%.
4.2.4 Immunohistochemistry (IHC)
Immunohistochemical analyses using a mouse monoclonal (1:150), and rabbit poly-clonal
(1:1000) antibody with high specificity for KLK6 was performed as described previously by
Petraki et al. (Petraki et al., 2001; Petraki et al., 2006). IHC for the detection of both wildtype
and mutant p53 protein was performed using a monoclonal antibody (DO-7, 1:25 Dako Canada,
Mississauga, ON.) according to the manufacturer’s instructions. Since wildtype p53 possesses a
short half-life, its presence in minute amounts is below the level of detection for IHC (Vojtesek
et al., 1992). Thus, the accumulation of p53 protein detected by IHC reflects aberrant p53 forms.
Briefly, tissue sections were dewaxed in xylene, dehydrated in 100% ethanol and rinsed in water.
Antigen retrieval was performed by pressure cooking in 0.01M citrate buffer for 30 minutes.
The slides were rinsed in water and treated for 30 minutes in 0.3% hydrogen peroxide/methanol.
The slides were rinsed in water and processed using the ImPRESS Detection Kit (Vector
Laboratories). Following a 20-minute incubation with normal horse serum, both KLK6
antibodies diluted in DAKO diluents, or p53 antibody, were applied to the slides for 1 hour at
room temperature. The slides were incubated with the appropriate secondary anti-rabbit and
anti-mouse antibodies for 30 minutes, washed, treated with the polymer reagent, and detected
with standard DAB solution for 5 minutes. The slides were stained with hematoxylin, and
finally mounted in Permount (Fisher Scientific Canada). No antibody controls were also
performed. The slides were scanned using the Aperio ScanScope (Aperio Technologies Inc, San
Diego, CA) for visualization and image acquisition. Standard hematoxylin and eosin-stained
slides were also scanned for visualization. KLK6 positivity was assessed according to overall
staining intensity (1=low, 2=medium, 3=high) compared to surface epithelial cells from a normal
ovarian tissue control and in “no antibody” controls. Similarly, p53 expression was assessed
against the surface epithelial cells from a normal ovarian tissue control and “no antibody”
controls, and scored according to staining intensity and cell distribution (1=normal; negative/low
staining with <20% cells positive; 2=moderate; moderate staining with >50% cells positive;
3=strong; strong positivity with >50% cells with strong staining).
82
4.2.5 Statistical Methods
The associations between outcome measurements: 19q-rearrangement, KLK copy-number,
chromosomal instability and p53 immunohistochemical intensity and clinical parameters were
examined by ANOVA, two-sample t-test and Chi-square test, as deemed appropriate. Cox
survival analysis was applied to disease free survival (DFS) and overall survival outcomes both
univariately and multivariately. In the multivariate survival analyses, only the clinical parameters
that were shown to be independent were included for adjustment. All statistical analyses were
performed using SAS software (version 9.2; SAS Institute). Two-sided P-values for statistical
significance were set at <0.05.
4.3 Results
4.3.1 Patient Cohort
The entire patient cohort consisted of 119 serous ovarian carcinoma tissues from patients naïve
to chemotherapy. The final number of patients available for detailed analysis was 81, due either
to poor tissue quality or lack of sufficient clinical data. The median age at diagnosis was 56.2
years. Tumours were staged according to the International Federation of Gynecology and
Obstetrics (FIGO) criteria and graded according to Day et al. (Day et al., 1975). Thirteen
tumours were classified as stage I, 6 as stage II, 50 as stage III and 16 as stage IV, with one
unknown stage. Four cases were grade 0 and included samples with evidence of malignancy, 22
were grade I, 15 were grade II and 40 cases were grade III. Response was classified as
complete, partial, progressive or stable at follow-up within 6 months. The median follow-up
time was 72 months.
4.3.2 The KLK Locus (19q13.3/13.4) is Subject to Frequent Copy-
Number Alterations and Chromosomal Instability
All 19q bacterial artificial chromosome (BAC) clones used in this study were individually FISH-
mapped to normal human metaphase chromosomes (data not shown) to ensure the proper
mapping location and absence of cross-hybridization to other chromosomes. The multi-colour
FISH strategy is illustrated in Figure 4.1A, with 19q12 probes, near the centromere labeled in
83
green; 19q13.2 probes, approximately 11MB telomeric to the 19q12 BACs labeled in blue; and
the KLK locus at 19q13.3/4, a further 10MB telomeric to the 19q13.2 BACs, labeled in red.
KLK6-specific PCR confirmed the presence of KLK6 in BAC clones RP11-10I11 and RP11-
615L12 (Figure 4.1A). Of the 119 specimens available, 81 were successfully analysed by FISH
for copy-number changes of the KLK locus. Across all cases, 14 showed 2 copies of the KLK
locus (17.3%); 24 showed a net loss of the locus (29.6%), 41 (50.6%) showed a net gain of the
KLK locus, and 2 (2.5%) showed high-level amplification (Figure 4.1B, Table 4.1). These
findings indicated that 82.7% of cases exhibited copy-number alterations (ie. gain, loss, or
amplification) at the 19q13.3/4 locus, containing the KLK genes. Moreover, because three
genomic regions on 19q were enumerated (19q12, 19q13.2 and 19q13.3/4), we could assess
whether such copy-number gains or losses were due to whole gains of 19q or through structural
rearrangements along 19q. The presence of structural alterations on 19q were identified when
the percentage of signals per cell, per 19q locus scored, was discordant. In contrast, when there
were similar frequencies of signals per cell across all genomic loci, the 19q arm was considered
intact (Figure 4.1C). Of the 14 cases which showed two copies of the KLK locus per cell, 4
(28.6%) were involved in a structural rearrangement on 19q and the remaining 10 cases (71.4%)
showed no indication of structural rearrangement of 19q. Of the 24 cases showing a net loss of
the KLK locus, 18 (75%) were a consequence of structural rearrangements of 19q, while the
remaining 6 (25%) were due to the entire loss of the 19q arm. Twenty-five of the cases showing
net gains of the KLK locus (61.0%) were a result of structural aberrations of 19q, while 16 of the
cases (39%) showed gains of the KLK locus through whole 19q arm copy-number gains. Finally,
the two cases which possessed amplification of the KLK locus, resulted from structural
aberrations of 19q (100%). Examples of these different classes of genomic changes are
illustrated in Figure 4.1C.
84
Figure 4.1. KLK Copy-Number Analysis of Serous Ovarian Carcinomas by FISH and Immunohistochemistry. A. Multi-colour FISH strategy. Shown is a normal chromosome 19 with the mapping location of the BAC clones used and PCR confirmation for the presence of KLK6 in BAC clones RP11-1OI11 and RP11-615L12. Also shown are “+” (positive) and “-” (negative) control lanes. The 19q12 locus is identified by green signals, 19q13.2 by blue signals and 19q13.3/4 (KLK locus) by red signals. B. Left panel: Distribution of copy-number changes of the KLK locus in 81 cases shows that the KLK locus (19q13.3) is subject to frequent copy-number changes which are dominated by gains/amplifications. Right panel: Classification of the presence or absence of 19q rearrangement reveals that 19q is frequently subject to structural aberrations. C. Representative images of IHC for p53 and KLK6 including hematoxylin/eosin (H&E), together with corresponding FISH images depicting each of the 4 classes of copy-number changes of the KLK locus and the two rearrangement classes. The 19q12 locus is identified by green signals, 19q13.2 by blue signals and 19q13.3/4 by red signals. The corresponding histograms show the range in number of signals per cell for each genomic locus. Rearranged cases were identified when the frequency of signals for each of the genomic loci were discordant. Cases were identified as not possessing rearrangements on 19q when the frequency of signals across all genomic loci tested per cell were similar.
It was evident from these results that the copy-number changes were often heterogeneous. In
many cases a predominant clone was present and accompanied with lesser populations, all
contributing to the net gain or loss of the locus. The extent of numerical CIN of the KLK locus
was assessed by enumerating the number of populations (greater than 10% in frequency). In all
cases in which only 2 copies of the KLK locus were identified, low-level CIN was observed; that
is, there was presence of one or two populations of cells. Similarly, cases showing the net loss of
the locus were also characterized primarily as low-level CIN. However, unlike 2-copied and net-
85
deleted cases, the observed net gains of the KLK locus were characterized predominantly by
moderate-CIN (up to 3 populations of cells) and high-level CIN (>3 populations), particularly
among those cases where 19q rearrangements were detected. In those net KLK gained cases
resulting from whole 19q arm gains, the level of CIN ranged from low to moderate (Table 4.1).
The correlation of KLK locus copy number to clinical parameters is summarized in Table 4.1.
Across age, stage, grade, response to treatment, KLK6-specific expression and extent of
chromosomal instability, statistical significance was shown only for chromosomal instability.
Low instability was associated with cases showing only 2 copies of the KLK locus, with copy-
number gains and losses associated with increasing instability (p<0.001).
Table 4.1: Clinical characteristics of patients stratified by KLK copy number:
KLK Copy Number 2 Copies (n=14) Net Loss (n=24) †Net Gain (n=43) p-value* Age (years) Mean (SD) 57.1 (16.1) 51.5 (10.9) 58.6 (13.3) 0.113 Stage I/II 2 (14%) 7 (29%) 10 (24%) 0.582
III/IV 12 (86%) 17 (71%) 32 (76%) Grade 0/I 6 (43%) 8 (33%) 12 (28%) 0.575
II/III 8 (57%) 16 (67%) 31 (72%) Response Complete 11 (79%) 16 (67%) 20 (48%) 0.271
Partial/Stable 2 (14%) 6 (25%) 15 (36%) Progression 1 (7%) 2 (8%) 7 (17%)
Instability Low 14 (100%) 18 (75%) 14 (33%) <0.001 Mod/High 0 (0%) 6 (25%) 29 (67%)
p53-IHC Normal 5 (36%) 7 (29%) 10 (24%) 0.672 Mod/Strong 9 (64%) 17 (71%) 32 (76%)
KLK6-IHC Low/Mod 4 (29%) 11 (46%) 12 (28%) 0.301 Strong 10 (71%) 13 (54%) 31 (72%)
* p value from t-test and Chi-square test.
† Net Gain includes amplified cases
86
4.3.3 Structural Alterations of Chromosome 19q are Associated with
Age and Grade
When the tumours were classified as either possessing 19q rearrangements or without
rearrangements, 61.7% (50/81) possessed 19q rearrangements and the remaining 38.3% (31/81)
had no apparent gross structural rearrangements (Figure 4.1B). The ranges of copy-number
gains and losses for the genomic loci analysed were primarily low. Losses were typically
characterized by the loss of one copy from the primary clonal populations; and gains were
characterized by one to three extra copies compared to the primary clonal population. As
mentioned previously, the KLK locus was amplified in 2 of the rearranged cases; and 19q13.2
amplification was seen in 9/50 (18%) of rearranged cases (or 9/81 (11.1%) of total cases). In 6
of the 9 cases, 19q13.2 was amplified in 100% of cells scored, whereas the remaining 3 cases
showed amplification of 19q13.2 as an additional population to polysomy for the locus,
indicating genomic heterogeneity. Amplification of 19q12 was also detected in 2/50 (4%) of
rearranged cases, but as additional populations to polysomy for the locus (Figure 4.1C).
When 19q rearrangement status was correlated to the same clinical parameters (Table 4.2), grade
was found to be significant (p=<0.001), with 58% of the low-grade tumours showing no 19q
rearrangements. This was in contrast to the high-grade tumours in which 84% of cases possessed
19q rearrangements. Age was also significant (p=0.003), revealing that 19q rearrangements
occurred more frequently in older patients.
Table 4.2: Clinical characteristics of patients stratified by 19q-rearrangement:
19q-Rearranged No (n=31) Yes (n=50) p-value*
Age (years) Mean (SD) 50.8 (13.3) 59.6 (12.3) 0.003 Stage I/II 10 (33%) 9 (18%) 0.119
III/IV 20 (67%) 41 (82%) Grade 0/I 18 (58%) 8 (16%) <0.001
II/III 13 (42%) 42 (84%) Response Complete 20 (65%) 27 (55%) 0.618
Partial/Stable 7 (23%) 16 (33%) Progression 4 (13%) 6 (12%)
* p value from t-test and Chi-square test.
87
4.3.4 Serous Ovarian Carcinomas Show a Range of KLK6 and p53
Expression
Since KLK6 protein overexpression has been observed in ovarian carcinomas,
immunohistochemical analysis was performed, revealing a range of KLK6 overexpression.
Normal ovarian surface epithelium expressed KLK6 at a barely detectable level by IHC,
consistent with previous findings (Ni et al., 2004; Petraki et al., 2001). Positive controls were
derived from formalin fixed paraffin embedded (FFPE) sections with accompanying KLK6
quantification by ELISA from tissue extracts, as previously described (Shan et al., 2007). All 81
patient specimens showed positive cytoplasmic staining for KLK6, ranging from low-level to
very high expression (Figure 4.1C). The comparison of the overall intensity of KLK6 staining to
KLK-locus copy-number showed no clear relationship (Table 4.1, Figure 4.1C). Strong
expression of KLK6 was found in the majority of cases with two copies of the locus (10/14); as
was also the case for specimens with a net gain of the locus (31/43). Cases with a net loss of the
locus showed roughly equal numbers of specimens with low/moderate and strong staining (11/24
and 13/14, respectively). In the two cases which were shown to be amplified, strong cytoplasmic
staining was shown in both cases.
Immunohistochemical analysis of p53 was successful in 80/81 specimens. Normal ovarian tissue
showed negative to very weak reactivity for p53 with less than 20% of cells showing weak
positivity, corresponding to the normally short half-life of wildtype p53 (Vojtesek et al., 1992).
Twenty-two cases (27.5%) were identified as having reactivity consistent with normal p53
protein expression; whereas abnormal p53 was identified in 22 cases (27.5%) showing moderate
staining, and in 36 cases (45%) showing strong reactivity, indicating abnormal p53
accumulation. When p53 IHC was correlated to age, stage, grade, response, KLK copy-number,
chromosomal instability and KLK6 IHC results, only grade was found to be statistically
significant (p=0.040) with higher grade tumours associated with stronger p53 staining.
4.3.5 Univariate and Multivariate Analyses
When hazard ratios (HR) estimated from the Cox regression model for disease-free survival
(DFS) were calculated at both univariate and multivariate levels (Table 4.3), response to
chemotherapy was highly significant (complete response HR=1; partial response/stable disease
88
HR=3.19; progression, HR = 6.96, p=<0.001) for both univariate and multivariate analyses.
Similar data were seen for stage and grade (Table 4.3). Age was significant only in univariate
analyses (p=0.026). Although hazard ratios based on KLK copy number (2 copies, HR =1; Net
loss, HR = 0.78; Net gain, HR = 1.2, p=0.441), or KLK6 IHC (low, HR=1; Mod/High, HR 1.9,
p=0.557) were not statistically significant on univariate and multivariate analyses, Kaplan-Meier
curves for DFS displayed a trend for those patients with a net loss of the locus to have a
relatively better DFS over patients with 2 copies or gains of the locus (Figure 4.2A). Hazard
ratios for DFS revealed that 19q rearrangement status was marginally significant (no
rearrangement, HR=1; 19q rearrangement, HR=1.68; p=0.078) in univariate analysis, but not
significant in multivariate analysis. Furthermore, p53 IHC was also not significant on univariate
(p=0.458) or multivariate (p=0.554) for DFS.
For overall survival (OS) (Table 4.4), chemotherapy response was highly significant in both
univariate and multivariate analysis (p<0.001), as was grade (p<0.001 and p=0.002, respectively)
and age (p<0.001 and p=0.021, respectively). Stage was only significant in univariate analysis
(p=0.009). Weak associations based on 19q-rearrangement and p53 protein status were seen
only in univariate analyses (p=0.068 and p=0.077, respectively). Neither KLK copy-number or
KLK6-specific IHC were significant for OS on univariate or multivariate analyses However, as
demonstrated for DFS, Kaplan Meier curves (Figure 4.2B) for 19q-rearrangement, KLK copy-
number and p53 expression, show a trend for better overall survival in patients who had no
rearrangements in 19q; who had a net loss of the KLK locus or with normal p53 expression.
89
Figure 4.2. Kaplan Meier Curves for Disease Free Survival (DFS) and Overall Survival (OS). Shown are the Kaplan Meier curves (unadjusted) based on 19q rearrangement status, KLK copy number status, KLK locus chromosomal instability and p53-IHC. For both DFS and OS, patients with no 19q rearrangement had a better outcome than those with disruptions of 19q. Similarly, a trend could be seen for better DFS and OS in patients with a net loss of the KLK locus.
90
Table 4.3. Hazard ratio (HR) estimated from the Cox regression model for disease-free survival (DFS).
Univariate Multivariate* HR P HR P
Age (years) 1.02 0.026 1 0.792 Stage I/II 1 0.002 1 0.011
III/IV 3.65 3.13 Grade 0/I 1 0.001 1 0.001
II/III 3.58 4.19 Response Complete 1 <0.001 1 <0.001
Partial/Stable 3.19 3.29 Progression 6.96 6.64
KLK6-IHC Low 1 0.557 1 0.879 Mod/High 1.19 0.95 19q-rearranged No 1 0.078 1 0.756
Yes 1.68 0.9 KLK Copy-Number 2 Copies 1 0.441 1 0.812
Net Loss 0.78 0.78 Net Gain 1.2 0.94
Instability Low 1 0.302 1 0.677 Mod/High 1.33 1.13
p53-IHC Normal 1 0.458 1 0.554 Mod/Strong 1.27 1.22 * Multivariate Cox model for 19q-rearranged, KLK copy-number, chromosomal instability, and p53-IHC includes: stage, grade and chemotherapy response. Mod = moderate .
91
Table 4.4. Hazard ratio (HR) estimated from the Cox regression model for overall survival (OS). Univariate Multivariate* HR P HR P
Age 1.04 <0.001 1.03 0.021 Stage I/II 1 0.009 1 0.237
III/IV 2.94 1.75 Grade 0/I 1 <0.001 1 0.002
II/III 5.22 4.03 Response Complete 1 <0.001 1 <0.001
Partial/Stable 2.89 2.83 Progression 13.82 11.92
KLK6-IHC Low 1 0.498 1 0.574 Mod/High 1.23 1.2 19q-rearranged No 1 0.068 1 0.909
Yes 1.75 1.04 KLK Copy-Number 2 Copies 1 0.109 1 0.718
Net Loss 0.59 0.7 Net Gain 1.26 0.93
Instability Low 1 0.163 1 0.512 Mod/High 1.49 1.22
p53-IHC Normal 1 0.077 1 0.545 Mod/Strong 1.88 1.25 * Multivariate Cox model for 19q-rearranged, KLK copy-number, chromosomal instability, and p53-IHC includes: age, grade and chemotherapy response. Mod = moderate.
4.4 Discussion
Ovarian carcinoma is an aggressive disease with a poor outcome. The search for biomarkers, not
only for diagnosis, but also for better prediction of outcome and response to therapy is of
paramount importance. Although CA-125 is used for monitoring disease recurrence following
surgery and chemotherapy, it lacks specificity as an early detection diagnostic biomarker (Bast et
al., 2009). Potential new biomarkers for OCa include members of the kallikrein gene family
(KLK) which maps to 19q13.3/4 (Borgono et al., 2004; Yousef et al., 2000). Among them,
overexpression of KLK6 protein, as detected by tissue extract enzyme-linked immunosorbent
assay (ELISA) has been demonstrated to be an important prognostic marker (Shan et al., 2007).
92
The mechanisms leading to the observed KLK6 overexpression are still unclear (Borgono et al.,
2004), though hormone stimulation (Lai et al., 2009; Lawrence et al., 2010; Shan et al., 2007;
Shaw and Diamandis, 2007), differential methylation (Pampalakis et al., 2008; Pampalakis et al.,
2006; Pampalakis et al., 2009; Sidiropoulos et al., 2005) and regulation by microRNAs
(miRNAs) (Chow et al., 2008; White et al., 2010b) have been investigated. Based on our
previous preliminary findings (Bayani et al., 2008b), we have further investigated the role of
KLK copy-number in 81 untreated serous ovarian cancers using a three-colour FISH approach
targeting loci on 19q, including the KLK locus, which harbours KLK6, and performing KLK6-
specific immunohistochemistry (IHC) to investigate the relationship between KLK copy number
and KLK6 protein expression. In this way, we were also able to determine whether such copy-
number imbalances were the result of whole chromosomal gains of 19q, or disruption of 19q; as
alluded to in early and more recent studies implicating abnormalities of chromosome 19
(particularly 19q13), in ovarian cancer pathogenesis (Micci et al., 2009; Taetle et al., 1999a;
Taetle et al., 1999b; Thompson et al., 1994a; Tsao et al., 2001).
Of the 81 tumours, we found that the KLK locus, using BAC clones confirmed to contain KLK6,
was subject to changes in copy number (gain, loss or amplification) in 82.7% of cases, with
gains/amplifications (53.1%) being more common over losses (29.6%). Only 2.5% of cases
showed high-level amplification and 17.3% harboured only 2 copies. Consistent with our
previous cytogenetic data (Bayani et al., 2008b), we found that the gains/amplifications of the
KLK locus resulted primarily from structural rearrangements of 19q. The same observation was
made for tumours showing net loss of the KLK locus, and this was in contrast to tumours where
only two copies of the KLK locus were identified. Many carcinomas (Al-Romaih et al., 2003;
Bayani et al., 2003; Maire et al., 2009; Vukovic et al., 2007) including ovarian, show karyotypic
heterogeneity both numerically and structurally (Bayani et al., 2002; Bayani et al., 2008a;
Gorringe et al., 2005) and such on-going chromosomal instability (CIN) and genomic variability
is believed to drive tumour progression (Li et al., 2009a). Indeed, our analyses revealed CIN and
copy-number heterogeneity of the KLK locus (p<0.001), illustrated by the observed number of
subpopulations of cells with greater or fewer KLK locus signals, in addition to the primary clonal
population (Figure 4.1). Hazard ratio and Cox regression analyses for DFS based on KLK copy-
number status showed no significant correlation in univariate and multivariate analysis (p=0.441
and p=0.812, respectively). However, Kaplan-Meier analysis suggested a trend for increasing
93
DFS among those cases with net loss of the KLK locus; whereas, patients with 2 copies, or those
who were shown to possess gains/amplifications had similar DFS.
KLK6-specific IHC analyses revealed overexpression in all cases ranging from weak to intense
immunoreactivity. Although there was no correlation between copy-number and protein
expression (p=0.301), there was a trend suggesting that those tumours with a gain of the locus
were more likely to overexpress KLK6. When only two copies of the locus were identified, the
majority of cases expressed KLK6 levels with equal frequencies of moderate to low expression.
This suggests that copy-number could contribute somewhat to protein expression, but is not the
primary regulatory mechanism, as exemplified by the cases with two copies, or loss of one copy,
but still exhibiting strong KLK6 expression; or cases with gain/amplification of the locus and
relatively moderate expression. No relationship with respect to DFS (p=0.557), was observed
based on KLK6 IHC expression, in keeping with the results of another study (Kountourakis et
al., 2008), but in contrast to Shan et al., 2007, suggesting ELISA-based measurements of protein
levels may be more informative. Thus, the results of the KLK6 immunohistochemical and copy-
number analyses point to additional mechanisms contributing to KLK6 protein overexpression
Interestingly, when tumours were stratified as 19q rearranged or non-19q rearranged, 61.7% of
cases showed structural abnormalities at 19q; which were strongly associated with tumour grade
(p<0.001). Fewer low-grade tumours possessed structural rearrangements of 19q (16%) as
compared to their high-grade counterparts (84%). Moreover, correlation to DFS and OS showed
that patients with 19q rearrangements had a shorter DFS and worse OS, which were of borderline
significance (p=0.078 and p=0.068 respectively) in univariate analyses. These combined
findings provide some credence to the two pathway model for ovarian tumourigenesis, where
type I ovarian tumours are classified as low-grade, slow-growing, with low-level and increasing
CIN; and type II tumours, are classified as higher grade, more rapidly growing, with high CIN,
and with a much lower 5-year survival (Ricciardelli and Oehler, 2009; Shih Ie and Kurman,
2004). Our study demonstrated that tumours classified as low-grade (grade 0/I) had fewer
structural and range of numerical abnormalities of chromosome 19 than high-grade tumours
(grade II/III) (Table 4.2); supporting the notion that changes on chromosome 19q, or the
molecular pathways influencing the observed numerical and structural changes on chromosome
19q, may play a role in OCa pathogenesis. Indeed, Type I tumours more frequently possess
mutations in KRAS and PTEN, and are subject to greater microsatellite instability (MSI), whereas
94
Type II tumours frequently possess mutations in p53 and BRCA1/2, and are subject to CIN. Our
p53 IHC findings are consistent with this model, with high-grade tumours more frequently
showing overexpression for p53 over low-grade tumours (p=0.040). The contribution of TP53 to
chromosomal instability has been well-documented (Fukasawa, 2005; Tomasini et al., 2008).
Thus, the concurrent findings of genomic instability and p53 overexpression, in high-grade cases
is consistent with previous reports (Blegen et al., 2000; Ceccaroni et al., 2004; McManus et al.,
1996). Although some studies have shown correlation of p53 expression and overall poor
outcome and survival (Lassus and Butzow, 2007), our analyses showed only a weak association
with poor overall survival (p=0.077), likely due to our smaller sample size; or choice of
antibodies, since the DO-7 clone detects both mutant and wildtype p53.
To our knowledge, this is the first report suggesting that both structural and numerical
abnormalities of chromosome 19 distinguish low-grade and high-grade cancers. Further analyses
on a larger cohort of tumours using the specific markers distinguishing Type I and Type II
tumours (Ricciardelli and Oehler, 2009; Shih Ie and Kurman, 2004) is warranted to determine
whether the preliminary observations seen here are associated. Indeed, early classical
cytogenetic studies have demonstrated OCa karyotypes are numerically and structurally
complex, with a propensity for the non-random gain and loss of specific chromosomes, and the
apparent preferential sites of chromosomal breakage (Taetle et al., 1999a; Taetle et al., 1999b),
including 1p1*, 1q1*, 1p2*, 1q2*, 1p3*, 1q3, 3p1*, 1q4*, 6q1*, 6p2, 6q2, 7p1, 7p2*, 11p1*,
11q2*, 12p1, 12q2*, 13p1, and 19q1 (where the asterisk (*) denotes the major band associated
with the chromosomal rearrangement). These early studies have implicated a role for
chromosome 19 in OCa pathogenesis (Pejovic et al., 1992a; Pejovic et al., 1990; Taetle et al.,
1999a; Taetle et al., 1999b; Thompson et al., 1994a; Thompson et al., 1994b). Improvements in
molecular cytogenetic analyses by way of multi-colour karyotyping (Bayani and Squire, 2004a),
readily revealed the cytogenetic complexity and heterogeneity of OCas (Bayani et al., 2002; Rao
et al., 2002), further refining copy-number changes and preferential sites of chromosome
breakage. Based on bulk high-throughput DNA analyses of OCas (derived largely from aCGH
experiments), chromosome 19 has a relatively equal frequency of copy-number gain and loss of
approximately 10% (reviewed by Gorringe et al., 2009), with allelic imbalances showing more
preferential gain in most of 19p and 19q (Gorringe et al., 2007). Very little information,
however, regarding the nature, frequency and heterogeneity of structural alterations leading to
95
local copy-number imbalances for chromosome 19 exists on a per cell basis. Moreover,
deciphering the link between the observed global patterns of ploidy change and chromosomal
instability with clinical factors, are becoming increasingly more relevant (Swanton et al., 2009).
The chromosome 19 landscape was recently revisited by Micci et al., 2009 who undertook an
extensive microdissection and reverse-FISH approach to identify the critical regions and loci on
chromosome 19 that are subject to copy number alterations in ovarian carcinomas; showing the
frequent involvement of chromosome 19q13 in unbalanced translocations both intra- and
interchromosomally. We have extended this observation to examine the 19q13 sub-bands:
19q13.2 and 19q13.3/4, demonstrating genomic instability and copy-number heterogeneity exists
within this major band. As well, Tsao et al. (Tsao et al., 2001) virally immortalized human
ovarian surface epithelial cells to identify early chromosomal events leading to tumourigenesis,
and found that all immortalized cell lines showed gains of 19q13, and the whole gain of
chromosome 19 prior to immortalization, re-enforcing the idea that chromosome 19 plays an
important role in early tumourigenesis and in maintaining tumourigenicity. A number of genes
subject to copy number changes, in addition to the KLK locus, have also been identified on 19q
(Thompson et al., 1996), including ACTN4 (Yamamoto et al., 2009), AKT2 (Bellacosa et al.,
1995; Nakayama et al., 2006) and cyclin E (Etemadmoghadam et al., 2009). The amplification
of AKT2 (19q13.2) has been demonstrated to be an important indicator of poor prognosis
(Bellacosa et al., 1995) and implicated in the pathogenesis of high-grade ovarian cancers
(Nakayama et al., 2007; Nakayama et al., 2006). Nine of our cases (11.1%) showed
amplification of the 19q13.2 region, which we speculate may reflect AKT2 amplification, since
one of the BACs used for the 19q13.2 region (RP11-67A5), includes AKT2. Moreover, our
observations in these chemotherapy-naïve tumours indicate that these changes likely result from
on-going tumourigenic processes rather than induced changes from radiation or
chemotherapeutic agents.
Chromosome 19 is unique, possessing the greatest number of genes, gene families, CpG island
density, and also contains a high level of repeat elements despite being among the smallest
chromosomes (Grimwood et al., 2004). Thus, it is not surprising that defects in DNA repair
pathways associated with OCa (ie. BRCA1/2) may be manifested in chromosomes with greater
potential for damage or epigenetic modification when protective pathways have been
compromised. Additionally, of the over 700 microRNAs (miRNAs) listed in the miRBase
96
(http://www.mirbase.org), chromosome 19q also harbours the highest number of miRNAs, with
greater than 60 of the approximately 90 miRNAs located on chromosome 19 mapping to 19q
(specifically at 19q13.4). miRNAs, which have been shown to be important regulators of gene
expression (reviewed by Barbarotto et al., 2008 and implicated in ovarian cancer (Nam et al.,
2008; Zhang et al., 2006; Zhang et al., 2008) can potentially regulate hundreds of gene targets,
thus the consequences of genomic changes at 19q can have far reaching implications than simply
the expression of genes mapping to chromosome 19. In fact, the KLK findings in this study
suggest that other mechanisms, in addition to copy-number, influence the observed
overexpression of KLK6, which may include miRNAs. Recently Chow et al., 2008,
demonstrated KLK6 and KLK10 were regulated by miRNAs in a breast cancer cell line.
Preliminary findings by our group (Bayani et al., In press) show that the expression of a number
of miRNAs predicted to regulate KLK6, are decreased in ovarian cancer cell lines and primary
tumours as compared to normal ovarian miRNAs. Certainly, the role of miRNAs to protein
expression may explain the lack of correlation between KLK copy-number and KLK6 protein
expression. Since the primary mechanism of gene regulation by miRNAs is mediated by the
inhibition of translation, rather than the degradation of RNA (Barbarotto et al., 2008); the
observed increases in KLK6 RNA transcript influenced by copy-number and/or combinations of
the aforementioned mechanisms, may not reflect the actual level of protein ultimately expressed.
This could explain the observation for tumours with gains of the KLK locus and relatively
moderate/low KLK6 IHC overexpression, wherein these tumour cells may continue to express
those miRNAs regulating (and inhibiting) KLK6 translation. Similarly, for tumours with the
observed net loss of the KLK locus and strong KLK6 protein expression, the miRNAs regulating
KLK6 may be diminished in these cells, permitting translation of the up-regulated KLK6
transcript. Interestingly, a few of these miRNAs predicted to regulate KLK6, as well as other
KLKs, are also located on chromosome 19q.
In summary we have demonstrated that the KLK locus at 19q13.3/4 is subject to high genomic
instability and copy number heterogeneity, mediated by structural rearrangements of 19q.
Moreover, structural rearrangements on 19q are associated with tumour grade, and may be
associated with, or a marker of the differential pathogenesis distinguishing low-grade and high-
grade serous cancers. While KLK6 copy number, through the enumeration of the KLK locus,
does not appear to directly regulate the observed KLK6 overexpression per se, it is one
97
contributing factor. Lastly, the unique genomic properties of chromosome 19 suggest that the
observed instability of 19q and the genes mapping to this location, including KLK gene members
warrant further investigation.
98
Chapter 5
Modulation of KLK6 Protein Expression in the Ovarian Cancer Cell Line OVCAR-3 by miRNA Let-7a
This Chapter represents data from the accepted manuscript entitled “Copy-Number and Expression Alterations of miRNAs in the Ovarian Cancer Cell Line OVCAR-3: Impact on Kallikrein 6 Protein Expression”. Bayani J, Kuzmanov U, Saraon P; Fung, WA, Soosaipillai A, Squire JA and Diamanids EP Clinical Chemistry published November 7, 2012 as doi:10.1373/clinchem.2012.193060 (Publish-Ahead-of-Print).
J.Bayani contributed to the development and experimental design of the study and manuscript; maintained cultures; identified, extracted and labeled BAC clones for FISH; performed, analysed and interpreted all FISH, SKY and aCGH; performed miRNA/mRNA extraction and miRNA profiling; performed transient transfections and KLK-specific-ELISA. U.Kuzmanov contributed to the development and experimental design of the study; assisted with transient transfections and quality control. P.Saraon performed and provided technical support PCR validation of miRNA profiling results. WA.Fung assisted with KLK-specific ELISA and quality control. A.Soosaipillai provided technical support and production of KLK-specific antibodies; JA. Squire contributed to the development and design of the study and manuscript. EP. Diamandis contributed to the development and design of the study and manuscript; in addition to granting support.
99
5 Modulation of KLK6 Protein Expression in the Ovarian Cancer Cell Line OVCAR-3 by miRNA Let-7a
5.1 Introduction
Ovarian cancer (OCa) continues to be the leading cause of death from a gynecological
malignancy for North American women. The largely asymptomatic nature of the disease
contributes significantly to its lethality. There currently are no markers for early diagnosis or for
screening (Bast et al., 2009), therefore identifying novel biomarkers to generate meaningful
proteomic signatures is essential. One promising biomarker comes from the Kallikrein (KLK)
gene family of serine proteases (Borgono and Diamandis, 2004). Kallikrein 6 (KLK6) is
frequently overexpressed at the mRNA and protein levels in tissues and fluids derived from OCa
patients; and implicated in a variety of normal physiological processes. These include
extracellular matrix remodeling, maintenance of protein clearance in the brain, and activating
inflammatory pathways and signaling cascades (reviewed by Bayani and Diamandis, 2011). The
overexpression of KLK6 in OCas at the mRNA and protein levels has been associated with poor
prognosis and the experimental evidence shows this overexpression enhances malignancy
(Bayani and Diamandis, 2011). Prognostically, several studies have consistently shown that the
combination of KLK6 with CA-125 enhances their diagnostic power (El Sherbini et al., 2011;
Shan et al., 2007; White et al., 2009a). Recently, we demonstrated that the KLK locus at
19q13.3/4, to which all members of the KLK gene family contiguously map to, are subject to
copy-number changes in OCa (Bayani et al., 2011; Bayani et al., 2008b). Our cytogenetic
studies (Bayani et al., 2008b), revealed that the entire locus was involved in the copy-number
change, rather than individual members. Our initial observations have recently been confirmed
by array studies (Beroukhim et al., 2010), showing a consistent copy-number ratio across the
entire locus in several cancer types, including OCa. The highly aneuploid and structurally
abnormal OCa karyotypes also demonstrated that the locus was subject to copy-number
heterogeneity (Bayani et al., 2002; Bayani et al., 2008a), reflecting genomic and chromosomal
instability often not revealed by the averaging algorithms of these high-throughput genome-wide
technologies (Albertson et al., 2003). Although we demonstrated that copy-number gains of
KLK6 were associated with increased KLK6 protein expression, a number of cancers with either
100
two copies and/or a subpopulation containing one copy of the KLK locus (in these predominantly
polyploidy cancers), exhibited protein levels comparable to cases with extra copies or
amplification of the locus. Conversely, a portion of cancers with populations of copy-number
gains or amplifications, exhibited KLK6 protein expression deemed as moderate to low.
Therefore, these data suggest that in addition to copy-number, other mechanisms influence the
observed overexpression of KLK6 protein in OCa, particularly at the post-transcriptional level.
Despite the fact that a number of mechanisms have been studied, it is an area that remains
unclear and in need of further study (reviewed by Bayani and Diamandis, 2011). A growing
body of evidence demonstrates that microRNAs (miRNAs) are key players in regulating and
fine-tuning protein expression in both normal and diseased tissues (reviewed by Schickel et al.,
2008). miRNAs comprise a class of non-coding RNAs. Excised from a 6-10 nucleotide hairpin
precursor RNA (pre-miRNA), miRNAs are transcribed from a larger primary transcript (pri-
miRNA) and processed in a similar fashion as many of the siRNAs to yield a mature miRNA.
Perfect complementarity between the miRNA and the target gene mRNA 3′ untranslated region
(UTR), results in the cleavage and degradation of the target mRNA; whereas less than perfect
pairing represses the translation process. It is in this elegant fashion that miRNAs can target
different mRNAs, increasing the diversity of gene regulation. Alterations in miRNAs have been
analysed in many tumour types including OCa (reviewed by Dahiya and Morin, 2010), and have
also been shown to be affected by copy-number (Croce, 2009; Dahiya and Morin, 2010; Zhang
et al., 2008). Recent bioinfomatic and experimental findings by Chow et al., 2008; and White et
al., 2010, suggest that the KLK genes are also subject to regulation by miRNAs. In this study,
we examined the contribution of copy-number, of both the KLK locus and of miRNAs predicted
to target KLK6, to the observed protein expression of KLK6 in a representative KLK6-
overexpressing ovarian cancer cell line OVCAR-3.
5.2 Materials and Methods
5.2.1 Fluorescence in situ Hybridization (FISH), Spectral Karyotyping
(SKY) and Array Comparative Genomic Hybridization (aCGH)
The OVCAR-3 cell line was obtained from the American Type Culture Collection (ATCC) and
maintained according the manufacturer’s instructions FISH: For copy-number studies, bacterial
101
artificial chromosome (BAC) clones mapping to three regions of chromosome 19q (19q12,
19q13.2, the KLK locus at 19q13.3/4) were obtained from The Centre for Applied Genomics
(Toronto, Canada) and differentially labeled as described previously by Bayani et al., 2011.
19q12 probes were labeled with Spectrum Green, all 19q13.2 probes were labeled with DEAC;
and all 19q13.3/4 clones which span the KLK locus, were labeled with Spectrum Orange.
Metaphase spreads from OVCAR-3 were prepared by standard cytogenetic techniques, as
described by Bayani and Squire, 2004, denatured and hybridized with the 19q multi-probe
cocktail (Bayani et al., 2011). The slides were mounted in a DAPI/Antifade medium
(Vectashield/ Vector Laboratories Canada) and visualized at 60X with a Zeiss Axioskop
fluorescence microscope (Carl Zeiss Canada). SKY: Metaphase preparations were pre-treated
and hybridized with SKY Paints (Applied Spectral Imaging, Carlsbad, CA) and processed as
described by Bayani and Squire, 2002. The images were collected using a Zeiss Axioplan
fluorescence microscope (Carl Zeiss Canada) and processed using the Spectral Karyotyping
Image Capturing and Analysis system (ASI). Karyotypic descriptions were assigned according to
the guidelines of the International System for Human Cytogenetic Nomenclature (Shaffer and
Tommerup, 2005). aCGH: The Agilent Human Genome 244K microarray platform was used
(Agilent Technologies, Inc.,Palo Alto, CA, USA cat. G4411B) containing 236,381 unique 60-
mer oligonucleotide features. The array design included a total of 5,045 features used as internal
controls. The features were based on the UCSChg17 (NCBI build 35, May 2004) Build. DNA
was extracted by standard phenol:chloroform methods, RNase-treated and resuspended in sterile
RNase/ DNase-free water (Sigma Canada). The direct labeling of DNAs, probe evaluation,
clean-up, hybridization to the array and post-hybridization washes were carried out using
reagents and equipment, according to Agilent protocols (http://www.chem.agilent.com) and as
previously described by Bayani et al., 2008. The CGH Analytics software version 3.4 (Agilent
Technologies) was used to analyze the aCGH data. Copy number aberrations were objectively
detected in replicate (dye-swap) experiments using an aberration calling method based on
computing significance scores for all genomic intervals.
5.2.2 microRNA Profiling
miRNAs were extracted using the mirVana™ miRNA Isolation Kit (Ambion, Life Technologies)
according to the manufacturer’s instructions and stored at -80˚C until ready for use.
102
Quantification of miRNA expression was performed by reverse-transcriptase real-time PCR
using the TaqMan® MicroRNA v1.0 system (Applied Biosystems Inc. (ABI), Life
Technologies) and compared to normal ovarian derived miRNA (Ambion Inc, Life
Technologies). The TaqMan® Low Density Array (TLDA) Human miRNAs Panel and
Multiplex RT pools of eight 48-miRNA sets (ABI) were used for global miRNA expression
analysis by qRT-PCR. Briefly, miRNAs were converted into specific cDNA derived from mature
miRNAs and quantified using the TaqMan® MicroRNA Assay. The TLDA card contains 365
lyophilized human TaqMan® miRNA sequences plus three small nucleolar RNAs (RU6B,
RNU4 and RNU44) as endogenous controls. Data were quantified and analyzed using Sequence
Detection System (version 2.3) (ABI, Life Technologies). miRNAs relative expression was
normalized against endogenous controls and normal ovarian miRNAs according to the
following: 2−ΔCT, where ΔCT=(CTmiRNA−CTsnoRNAs)(Schmittgen et al., 2004).
5.2.3 Identification of miRNAs Predicted to Target KLK6
To identify putative miRNAs predicted to target the 3’untranslated region (UTR) of KLK6, three
public databases were accessed: the miRNA Registry at the Sanger Institute
(http://microrna.sanger.ac.uk/sequences/) (Griffiths-Jones et al., 2008), Memorial Sloan-
Kettering Cancer Center’s miRNA database (http://www.microrna.org/microrna/home.do) (Betel
et al., 2010), and TargetScan (http://www.targetscan.org/) (Grimson et al., 2007). Predicted
miRNAs common to all three databases were considered candidates for further study.
5.2.4 Transient Transfection of miRNAs to OVCAR-3
For transient transfection studies, the Pre-miR™ miRNA Precursor Kit (Ambion, Life
Technologies) was utilized according to the manufacturer’s instructions. Briefly, the hsa-let-7a-
5p precursor miRNA (PM10050, Ambion, Life Technologies), which corresponds to identical
mature miRNAs derived from hsa-let-7a-1, hsa-let-7a-2, and hsa-let-7a-3 loci; and a scrambled
miRNA negative control (Cat. No. 17110, Life Technologies) were prepared at a final
concentration of 6.25 μM. OVCAR-3 cells were trypsinized and adjusted to a final
concentration of 1 x 105 cells/ml. In six-well tissue culture plates, 2.4 ml of OVCAR-3 cells
(performed in duplicate) were combined with the transfection mixture comprised of the siPORT
103
NeoFX transfection agent (Ambion, Life Technologies) and Opti-Mem I Medium (Ambion, Life
Technologies), as outlined by the manufacturer. The cultures were allowed to incubate at 37˚C in
a CO2 incubator for 24 hours, after which the culture medium was replaced with fresh normal
growth medium. Biological replicates for each condition were prepared. After 72 hours, the
media was collected and stored at -80˚C, the cells counted by trypan blue staining and the cell
pellet stored for later RNA extraction using the mirVana™ miRNA Isolation Kit (Ambion, Life
Technologies).
5.2.5 Quantitative Reverse Transcription Polymerase Chain Reaction
(qRT-PCR)
Quantification of mature hsa-let-7a, using the TaqMan® miRNA Assays (ABI, ID RT000377
and TM000377), including a control small nuclear RNA, RNU44 (ABI, ID RT001094 and
TM001094) were performed according to the manufacturer’s instructions. Briefly, 10ng of
extracted miRNAs were reverse transcribed using RT-specific primers, then subjected to PCR
using miRNA or control-specific TaqMan® system. As described above, the data were
quantified and analyzed using Sequence Detection System (version 2.3) (Life Technologies).
miRNAs relative expression was normalized against endogenous controls according to the
following: 2−ΔCT, where ΔCT=(CTmiRNA−CTsnoRNAs)(Schmittgen et al., 2004). The assays were
performed in triplicate for each sample.
5.2.6 Measurement of KLK6 and KLK10 by ELISA
We have previously described our ELISA methodologies for all KLKs. Briefly, the assays are
based on ‘‘sandwich’’-type ELISA principles with one antibody used for capture and one for
detection. A monoclonal-monoclonal ELISA configuration for the immunodetection of KLKs 6
and 10 was utilized as follows. White polystyrene microtiter plates were coated with 500-ng/well
monoclonal antibody in 100 μL coating buffer (50 mmol/L Tris- HCl, pH 7.8) overnight. The
plates were washed twice in wash buffer (10 mmol/L Tris-HCl, pH 7.4, containing 150 mmol/L
NaCl and 0.5 mL/L Tween 20), to which 50 μL of recombinant protein calibrators or samples,
diluted in a 6% BSA solution, were added to the wells, along with 50 μL assay buffer (6% BSA
104
containing 25 mL/L normal mouse serum, 100 mL/L normal goat serum, and 10 g/L bovine
IgG). The plates were incubated for 2 hours with continuous shaking and washed 6 times as
described above. 100 μL/well of biotinylated monoclonal detection antibody (50 ng) diluted in
assay buffer were added to each well, incubated for 1 hour, and washed the plates 6 times. 100
μL (5 ng) alkaline phosphatase–conjugated streptavidin diluted in 6% BSA was added to each
well, incubated for 15 minutes with continuous shaking, and washed. 100 μL of diflunisal
phosphate solution (0.1 mol/L Tris-HCl, pH 9.1, containing 1 mmol/L diflunisal phosphate, 0.1
mol/L NaCl, and 1 mmol/L MgCl2) were added to each well, and incubated for 10 minutes with
continuous shaking. Finally, 100 μL developing solution (1 mmol/L Tris, 0.4 mol/L NaOH, 2
mmol/L TbCl3, and 3 mmol/L EDTA) was added to each well and mixed for 1 minute.
Fluorescence was measured with a time-resolved fluorometer (Perkin-Elmer); and the calibration
and data reduction analyses were performed automatically, as described (Christopoulos and
Diamandis, 1992).
5.3 Results
5.3.1 Identification of Predicted miRNAs Regulating KLK6 and miRNA
Profiling
To identify the predicted miRNAs targeting the 3’UTR of the KLK6 gene, a survey of three
publically accessible databases: the miRNA Registry at the Sanger Institute
(http://microrna.sanger.ac.uk/sequences/), Memorial Sloan-Kettering Cancer Center’s miRNA
database (http://www.microrna.org/microrna/home.do), and TargetScan
(http://www.targetscan.org/), was conducted. From these three databases, the most commonly
predicted miRNAs for KLK6 were from the members of the hsa-let-7 family of miRNAs,
namely hsa-let-7a, hsa-let-7b, hsa-let-7c, hsa-let-7d and hsa-let-7e (Figure 5.1A). miRNA
profiling of the KLK6-overexpressing OVCAR-3 using the TLDA arrays, showed the
dysregulation of these miRNAs as compared to normal ovarian-derived miRNA, with the
decrease in expression of hsa-let-7 family members. Shown in Table 5.1 are the log10
transformed expression ratios for the miRNAs on the TLDA array that are predicted to target
KLK6 as identified by the database searches. In addition, the miRNA expression of those
predicted by White et al., 2010 to target KLK6 are also included in Table 5.1. Of the 15
105
miRNAs predicted to target KLK6 (in all three or in two of three databases, and for which
expression data was available on the TLDA array, as well as those predicted by White et al.,
2010) 10 of these miRNAs showed decreased expression. Expression profiling of the other
predicted mRNAs (ie. identified only in one database), showed a similar trend, with 13 of 24
miRNAs showing decreased expression. Validation of the miRNA findings by mature hsa-let-
7a-specific qPCR confirmed the relative low-leveled expression of hsa-let-7a in OVCAR-3
compared to the normal ovarian miRNA control (Figure 5.1B).
To test whether these candidate miRNAs could affect protein expression, hsa-let-7a-5p was
transiently transfected into OVCAR-3. Mature hsa-let-7a is encoded by three different loci in
humans: hsa-let-7a-1 (9q22.32), hsa-let-7a-2 (11q24.1), and hsa-let-7a-3 (22q.13.3) (Boyerinas
et al., 2010). Following a 72-hour transfection with either hsa-let-7a-5p; or a scrambled
miRNA control, the amount of secreted KLK6 was determined by ELISA in the media.
Additionally, qRT-PCR for mature hsa-let-7a in the hsa-let-7a-5p-transfected cell line confirmed
the 6-fold increase of the mature transcript over the scrambled-transfected control (data not
shown). After adjusting for the final cell number, an average decrease of 25-35% in the level of
KLK6 protein in the hsa-let-7a-5p treated cells was detected, in comparison to the scrambled-
transfected control. Based on previous studies showing the modulation of KLK6 and 10 protein
expression by hsa-let-7f (Chow et al., 2008; White et al., 2010b), we surveyed the databases
from miRNAs predicted to target KLK10 and their expressions (Table 5.2). Of the miRNAs
identified in all three databases, three of four showed decreases in expression, including hsa-let-
7b, hsa-miR-214 and hsa-miR-485-5p (Table 5.2). Of the 20 miRNAs identified in two
databases, 11 also showed a relative decrease in expression. Based on these database algorithms,
among the hsa-let-7 family members, hsa-let-7b is more strongly predicted to target KLK10,
than hsa-let-7a, hsa-let-7c, hsa-let-7d or hsa-let-7e; with hsa-let-7a identified in two databases as
a putative regulating miRNA. Nevertheless, we tested whether hsa-let-7a could also affect
KLK10 protein expression. ELISA for KLK10 in the hsa-let-7a-5p transfected line showed an
8%-15% decrease in secreted KLK10 protein as compared to the control (Figure 5.1C).
106
Figure 5.1 . miRNA Profiling of OVCAR-3 and miRNAs Predicted to Target KLK6. A) The Venn diagram depicts the results of a search using the publically accessible databases including the miRNA Registry at the Sanger Institute (http://microrna.sanger.ac.uk/sequences/), Memorial Sloan-Kettering Cancer Center’s miRNA database (http://www.microrna.org/microrna/home.do) and TargetScan (http://www.targetscan.org/). miRNAs common to all databases are shown in red. The expression of those miRNAs present on the TLDA microarray are also depicted and reflect the relative expression to normal ovarian miRNAs (performed in duplicate). B) hsa-let-7a specific RT-PCR confirming the relative decrease in expression of hsa-let-7a OVCAR-3 (performed in duplicate) compared to normal ovary. C) Results of KLK6 and KLK10 ELISA of supernatants from hsa-let-7a- transfected OVCAR-3 compared to the negative control OVCAR-3 showing the decrease in KLK6 and KLK10 expression upon transient transfection. Error bars represent the standard deviation of replicates.
107
Table 5.1. Expression values of miRNAs predicted to target KLK6 in OVCAR-3
Shown above are the log10 ratios of duplicate TLDA miRNA profiling for OVCAR-3 when compared to miRNAs derived from normal ovarian tissue. † = those miRNAs previously reported by White et al. (White et al., 2010b) to be predicted to target KLK6. Negative values reflect the relative loss of expression and positive values reflect the relative gain in expression.
Database miRNA Average Expression (Range) Error (±) Target Scan hsa -let-7a† -0.69 (-0.74/-0.64) 0.05 Sanger hsa -let-7b† -1.23 (-1.29/-1.16) 0.06 Sloan Kettering hsa -let-7c -0.27 (-0.27/-0.26) 0.00
hsa -let-7d† -0.23 (-0.37/-0.09) 0.14 hsa -let-7e† -0.43 (-0.54/-0.32) 0.11 hsa-mir-98† 0.32(0.26/0.38) 0.06 hsa-mir-608 0.35 -
Target Scan hsa -let-7f† -0.36 (-0.21/-0.50) 0.15 Sloan Kettering hsa -let-7g -0.76 (-0.78/-0.74) 0.02 Target Scan hsa-mir-575 -1.76 (-2.07/-1.45) 0.31 Sanger hsa-mir-659 -1.66 (-1.68/-1.63) 0.03
hsa-mir-639 1.44 (1.54/1.34) 0.10 hsa-mir-555 0.13 - hsa-mir-296 0.04 (0.05/0.03) 0.01
Sanger hsa-mir-17-3p -0.25 (-0.30/-0.20) 0.05 hsa-mir-17-5p 0.81 (0.73/0.89) 0.08 hsa-mir-551b 1.17 (1.26/1.07) 0.09 hsa-mir-200a 1.67 (1.73/1.61) 0.06 hsa-mir-200b 2.75 (2.74/2.77) 0.02 hsa-mir-193a -1.16 (-1.17/-1.14) 0.01 hsa-mir-193b -0.76 (-0.77/-0.75) 0.01 hsa-mir-18a 1.34 (1.32/1.37) 0.03
hsa-mir-324-3p -0.25 (-0.27/-0.24) 0.02 hsa-mir-324-5p 0.30 (0.30/0.29) 0.01
hsa-mir-132 -2.04 (-2.23/-1.85) 0.19 hsa-mir-554 0.47 -
Target Scan hsa-mir-576 0.39 (0.61/0.17) 0.22 hsa-mir-342 -0.44 (-0.47/-0.42) 0.03 hsa-mir-601 -0.19 (-0.53/0.15) 0.34 hsa-mir-654 -2.61(-2.67/-2.55) 0.06 hsa-mir-140† -1.71(-1.71/-1.72) 0.00 hsa-mir-630 -0.45 (0.05/-0.96) 0.50 hsa-mir-520a -0.49 (-0.68/-0.31) 0.19 hsa-mir-503 -0.89 (-0.77/-1.01) 0.12 hsa-mir-548a 0.56 (0.65/0.47) 0.09 hsa-mir-548b 0.36 - hsa-mir-125a -0.97 (-0.96/-0.96) 0.01 hsa-mir-125b -1.28 (-1.24/-1.32) 0.04
108
Table 5.2. Expression values of miRNAs predicted to target KLK10 in OVCAR-3
Database miRNA Average Expression (Range) Error (±) Target Scan hsa-let-7b† -1.23 (-1.29/-1.16) 0.06 Sloan Kettering hsa-miR-214† -4.74 (-4.22/-5.27) 0.52 Sanger hsa-miR-224† 1.11 (1.10/1.13) 0.02
hsa-miR-485-5p -2.61 (-2.67/-2.55) 0.06 Sanger hsa-let-7a† -0.69 (-0.74/-0.64) 0.05 Target Scan hsa-let-7c -0.27 (-0.27/-0.26) 0.00
hsa-let-7d -0.23 (-0.37/-0.09) 0.14 hsa-let-7e -0.43 (-0.54/-0.32) 0.11
hsa-miR-148b 0.54 (0.52/0.55) 0.01 hsa-miR-152 -1.06 (-1.00/-1.12) 0.06 hsa-miR-197 -0.94 (-1.05/-0.83) 0.11 hsa-miR-326 0.38 (0.26/0.50) 0.12 hsa-miR-98† 0.32 (0.26/0.38) 0.06
Target Scan hsa-miR-1 -1.91 (-1.93/-1.90) 0.02 Sloan Kettering hsa-miR-143† -4.42 (-4.48/-4.36) 0.06
hsa-miR-18a 1.34 (1.32/1.37) 0.03 hsa-miR-192 -0.66 (-0.76/-0.56) 0.10 hsa-miR-193b -0.76 (-0.77/-0.75) 0.01 hsa-miR-206† 0.40 (0.33/0.46) 0.07 hsa-miR-215 0.71 (0.65/0.77) 0.06 hsa-miR-510 -0.14 (0.09/-0.38) 0.24
hsa-miR-515-3p -3.00 (-0.14/-5.87) 2.87 hsa-miR-613 0.40 (0.03/0.77) 0.37 hsa-miR-646 0.12 (0.33/-0.08) 0.21
White et al., 2010 hsa-miR-125b -1.28 (-1.24/-1.32) 0.04 hsa-miR-140 -1.71 (-1.71/-1.72) 0.00 hsa-miR-149 0.10 (0.07/0.13) 0.03 hsa-miR-432 -2.92 (-2.96/-2.87) 0.05
†White et al., 2010
Shown above are the log10 ratios of duplicate TLDA miRNA profiling for OVCAR-3 when compared to miRNAs derived from normal ovarian tissue. miRNAs indicated in bold represent those miRNAs also predicted to target KLK6. † = those miRNAs previously reported by White et al., 2010 to be predicted to target KLK10. Negative values reflect the relative loss of expression and positive values reflect the relative gain in expression.
109
5.3.2 OVCAR-3 Cytogenomics
Spectral karyotyping (SKY) analysis of OVCAR-3 showed a hyper-triploid cell line (59-70
chromosomes) with numerous simple and complex structural rearrangements consistent with
previous reports (http://www.ncbi.nlm.nih.gov/sky/skyweb.cgi). Structural rearrangements
involving chromosome 19 were identified, in addition to two apparently normal copies of
chromosome 19. A complex chromosome described as a der(22)(16;19;22) was shown to contain
a large homogeneously staining region (hsr) indicative of locus amplification (Figure 5.2A).
aCGH of the cell line revealed a high-level amplification spanning 19q11 to 19q13.2 and focal
amplification at 19q13.4 (Figure 5.2B). These were consistent with other aCGH findings for
this cell line (Beroukhim et al., 2010). Using a multi-colour 19q probe set previously described
(Bayani et al., 2011), we confirmed the copy-number findings identified by aCGH.
Amplification of 19q11 (green) and 19q13.2 (blue) were identified on the der(22), as seen by the
co-localization of FISH signals. Single hybridization signals, representing the KLK locus, were
seen only on the two normal chromosomes 19 (red) (Figure 5.2C). aCGH ratios for the KLK
locus suggested its net loss, reflecting the limitations of CGH to account for ploidy status of the
cell line and the possibility of non-clonal losses and gains of chromosome 19. Thus, at most,
there are two copies of the KLK locus in this triploid cell line.
Since copy-number has been shown to be a contributing mechanism to changes in miRNA
expression (Croce, 2009; Dahiya and Morin, 2010; Zhang et al., 2008), we assessed the copy-
number status of the three hsa-let-7a genes which encode for an identical mature hsa-let-7a
miRNA (Boyerinas et al., 2010): hsa-let-7a-1 (9q22.32), hsa-let-7a-2 (11q24.1), and hsa-let-7a-
3 (22q.13.3) using the aCGH data (Figure 5.3). Both hsa-let-7a-2 and hsa-let-7a-3 mapped to
loci generating ratios indicative of an overall net loss, while hsa-let-7a-1 indicated ratios
indicative of two copies. Since OVCAR-3 possesses a triploid karyotype, with both clonal and
non-clonal changes of these chromosomes; there is the presence of at least two copies of hsa-let-
7a-1; and at least one, but not more than two copies, each of hsa-let-7a-2 and hsa-let-7a-3;
which is consistent with the SKY findings. Indeed, when we analysed the structural
rearrangements of those chromosomes to which these miRNAs mapped, we found that while the
majority of metaphases possessed these structural rearrangements affecting these loci, for some,
only 50% (5/10) of metaphases possessed these derivative chromosomes (Figure 5.3). For hsa-
110
let-7a-1 (and hsa-let-7f-1) mapping to 9q22.3, two normal chromosomes 9 were detected in all
cells (10/10), however the der(9)t(8;9) was only detected in 60% of metaphases with the loss of
material telomeric to 9q12. For hsa-let-7a-2, mapping at 11q24.1; no normal chromosomes 11
were identified, however 80% (8/10) of metaphases possessed the der(11)t(11;14) rearrangement
accounting for copies of hsa-let-7a-2 at the telomeric end of the chromosome. However, the
der(11)t(11;16;18) rearrangement, present in only 60% (6/10) of metaphases possesses a deletion
telomeric to 11q23. Together, these account for the net deletion of the locus. Finally, hsa-let-7a-
3 which maps to 22q13.2, showed a net loss, contributed primarily by the combination of losses
of the locus in the following rearrangements: i(22)(p10) (8/10 metaphases); and
der(22)t(11;8;16) (5/10 metaphases) and the presence of the locus in the der(22)t(16;22) (7/10
metaphases) and der(22)t(16;19;22) (9/10 metaphases). Table 5.3 summarizes the net-copy
number changes for the miRNAs predicted to target KLK6, in at least two of the databases and
for those identified by White et al., 2010, by combined aCGH and cytogenetic analysis, and
accounting for the ploidy of the cell line. For each mature miRNA, there is a general
concordance in the association between overepxression of the mature miRNA and net copy-
number gains of their genomic mapping locations; or decreased expression associated with net
copy-number losses of their associated genomic loci. The exceptions to these findings are hsa-
let-7c and hsa-let-7f-2 which exhibited the relative decrease in expression, despite possessing net
gains of their loci; while hsa-miR-639 showed a relative increase in expression though there was
a net loss in copy-number. Similarly, Table 5.4 summarizes the net-copy number changes by
combined aCGH and cytogenetic analysis for the miRNAs predicted to target KLK10 in at least
two of the databases and for those identified by White et al., (2010), showing a similar trend of
positive association between the copy-number status of the miRNA and its expression, with the
exception of seven which showed an inverse association.
111
Table 5.3. Summary of miRNA Expression and Copy-Number Changes of miRNAs Predicted to Target KLK6
†White et al. , 2010.
Shown are the combined expression and cytogenomic analyses of miRNAs predicted to target KLK6. miRNAs indicated in bold represent those miRNAs identified in all three publically accessible databases as described in the Methods. † = those miRNAs previously reported by White et al., 2010 to be predicted to target KLK6. Loss=net decrease in copy-number. Gain=net increase in copy number. Balanced=net copy number of 2/no change in copy number.
aCGH miRNA ExpressionDatabase miRNA Locus Net Copy Number Mature miRNA Target Scan Sloan Kettering Sanger
hsa-let-7a-1† hsa-let-7a-2† hsa-let-7a-3†
9q22.32 11q24.1 22q13.3
Balanced Loss Loss
Decreased
hsa-let-7b† 22q13.3 Loss Decreased hsa-let-7c 21q21.21 Gain Decreased hsa-let-7d† 9q22.32 Balanced Decreased hsa-let-7e† 19q13.3 Loss Decreased hsa-miR-98† Xp11.22 Gain Increased Target Scan Sloan Kettering
hsa-let-7f-1† hsa-let-7f-2†
9q22.32 Xp11.22
Balanced Gain
Decreased
hsa-let-7g 3p21.1 Balanced Decreased Target Scan hsa-miR-296 20q13.32 Gain Increased Sanger hsa-miR-575 4q21.22 Balanced Decreased hsa-miR-639 19p13.2 Loss Increased hsa-miR-659 22q13.1 Loss Decreased hsa-miR-140† 16q22.1 Loss Decreased
112
Table 5.4 Summary of miRNA Expression and Copy-Number Changes of miRNAs Predicted to Target KLK10 aCGH miRNA ExpressionDatabase miRNA Locus Net Copy Number Mature miRNA Target Scan hsa-let-7b† 22q13.3 Balanced Decreased Sloan Kettering hsa-miR-214† 1q24.3 Gain Decreased Sanger hsa-miR-224† Xq28 Balanced Increased hsa-miR-485-5p 14q32.3 Balanced Decreased Sanger hsa-let-7a-1† 9q22.32 Balanced Decreased Target Scan hsa-let-7a-2† 11q2.1 Loss Decreased hsa-let-7a-3† 22q13.3 Loss Decreased hsa-let-7c 21q21.21 Gain Decreased hsa-let-7d 9q22.32 Balanced Decreased hsa-let-7e 19q13.3 Loss Decreased hsa-miR-148b 12q13.13 Gain Increased hsa-miR-152 7q21.32 Gain Decreased hsa-miR-197 1p13.3 Loss Decreased hsa-miR-326 11q13.4 Amplified Increased hsa-miR-98† Xp11.22 Gain Increased Target Scan hsa-miR-1-1 20q13.33 Gain Decreased Sloan Kettering hsa-miR-1-2 18q11.2 Loss Decreased hsa-miR-143† 5q32 Loss Decreased hsa-miR-18a 13q31.1 Loss Increased hsa-miR-192 11q13.1 Gain Decreased hsa-miR-193b 16p13.2 Loss Decreased hsa-miR-206† 6p12.2 Balanced Increased hsa-miR-215 1q41 Gain Increased hsa-miR-510 Xq27.3 Loss Decreased hsa-miR-515-3p 19q13.42 Amplified Decreased hsa-miR-613 12p13.1 Gain Increased hsa-miR-646 20q13.3 Gain Increased White et al. 2010. hsa-miR-125b-1† 11q24.1 Loss Decreased hsa-miR-125b-2† 21q21.1 Gain Decreased hsa-miR-140† 16q22.1 Loss Decreased hsa-miR-149† 2q37.3 Gain Increased hsa-miR-432† 14q32.2 Balanced Decreased
†White et al,. 2010.
Shown are the combined expression and cytogenomic analyses of miRNAs predicted to target KLK10. miRNAs indicated in bold represent those miRNAs also predicted to target KLK6. † = those miRNAs previously reported by White et al., 2010, to be predicted to target KLK10. Loss=net decrease in copy-number. Gain=net increase in copy number. Balanced=net copy number of 2/no change in copy number.
113
Figure 5.2 Cytogenomic Analysis of OVCAR-3. A) Spectral Karyotyping (SKY) shows the presence of two normal chromosomes 19 (green) and a structurally abnormal der(22) comprised of chromosomes 16 (orange), 19 (green) and 22 (pink). B) Array Comparative Genomic Hybridization (aCGH) results of chromosome 19 in OVCAR-3 shows the copy-number amplifications, gains and losses along the chromosome. An enlargement of the KLK locus at 19q13.3/4 shows the net copy-number loss of the entire KLK locus. Green, blue and red circles beside the ideogram depict the location of the multicolour FISH probes used to verify the copy-number status of KLK6 as described in the Methods. C) Locus-specific FISH to OVCAR-3 metaphases confirm the amplification of 19q12 (green signals) on the der(22) and single copies at the normal chromosome 19 position (left panel); amplification of 19q13.3 (blue signals) on the der(22) and single copies at the normal mapping position (middle panel); single copy of the KLK locus (red signals) only on the normal chromosomes 19.
114
Figure 5.3. Cytogenomic Analysis of hsa-let-7a Family Members in OVCAR-3. A) hsa-let-7a-1 mapping to chromosome 9q22.32 together in a cluster with hsa-let-7f-1 and hsa-let-7d shows a 1:1 ratio in copy number indicating two copies. Cytogenetic analysis of chromosome 9 confirms the presence of two normal chromosomes 9 and one der(9)t(9;8), with a breakpoint centromeric to the mapping location of the miRNA cluster. B) hsa-let-7a-2 mapping to chromosome 11q24.1 shows the net loss in copy number suggesting that no more than two copies for the locus exists. Cytogenetic analysis for chromosome 11 shows the presence of two abnormal chromosomes 11 as a der(11:14) in 8/10 cells, accounting for the copies of the miRNA; and the presence of a der(11)t(11;16;18) present in 6/10 cells with a deletion of the terminal region of chromosome 11 containing hsa-let-7a-2. C) hsa-let-7a-3 mapping to chromosome 22q13.3 also showing the net loss of the locus, with no more than two copies present. Cytogenetic analyses for chromosome 22 shows the presence of several structural aberrations involving chromosome 22, with such structural aberrations resulting in the loss of the locus.
115
5.4 Discussion
In this study, we investigated the role of miRNAs in the regulation of KLK6 protein expression
in a representative ovarian cancer cell line OVCAR-3, and how genomic instability and copy-
number may contribute to this mode of regulation. The serine protease, KLK6, is significantly
overexpressed in OCa tumours and ascetic fluids, and associated with a poor prognosis
(reviewed by Bayani and Diamandis, 2011). We and others, have shown the prognostic utility of
KLK6 to enhance, and in some cases out-perform CA-125 (El Sherbini et al., 2011; Koh et al.,
2011; Shan et al., 2007; White et al., 2009a). Therefore, its potential as an OCa biomarker
continues to be evaluated, ranking among the top 10-20 putative biomarkers in large-scale
multicenter studies (Cramer et al., 2011; Zhu et al., 2011). Increasing evidence suggest that
KLK6 contributes significantly to tumourigenesis beyond the functions extracellular matrix
degradation and tissue remodeling, including the participation in inflammatory processes, altered
immune cell survival and in enhanced intracellular signaling via activation of G-protein–coupled
receptor family members: protease-activated receptors (PARs) (reviewed by Bayani and
Diamandis, 2011). Therefore, the observed increase in KLK6 expression in cancers signifies the
dysregulation of many pathways, especially those deemed the hallmarks of cancer (Hanahan and
Weinberg, 2000, 2011). Although several mechanisms have been investigated including
methylation, and the response to steroid hormones, inconsistencies within and between various
cancers renders our understanding of KLK6 regulation still largely unknown (Bayani and
Diamandis, 2011). Recently, we reported that serous OCas exhibit copy-number heterogeneity
due to genomic instability of the KLK locus and showed a general concordance for
overexpression of KLK6 protein with copy-number (Bayani et al., 2011; Bayani et al., 2008b).
However, a number of cancers with either two copies, or with a combined subpopulation
containing one copy of the KLK locus, showed protein levels comparable to cases with extra
copies or amplification of the locus. Conversely, a portion of cancers with populations of copy-
number gains or amplifications, exhibited KLK6 protein expression deemed as moderate to low.
Therefore, we explored other mechanisms known to modulate protein expression which are also
affected by copy-number, namely microRNAs (miRNAs).
miRNAs form an important class of regulators that participate in diverse biological functions
including development, cell proliferation, differentiation, and apoptosis, and may function as
116
oncogenes or tumour suppressors (reviewed by Schickel et al., 2008). Thus, a decrease in the
expression of a miRNA results in the permissive expression of target genes that promote
tumourigenesis; and the overexpression of miRNAs limits the expression of target genes that
could inhibit tumourigenesis (Esquela-Kerscher and Slack, 2006). The complexity in this form
of regulation is demonstrated by the fact that a miRNA can target multiple mRNAs, and each
mRNA may be targeted by different miRNAs (Esquela-Kerscher and Slack, 2006; Griffiths-
Jones et al., 2008). Studies by Chow et al., 2008 and White et al., 2010 have recently shown that
protein levels of various members of the KLK family are modulated by miRNAs in ovarian,
breast and renal carcinomas. Their bioinfomatical approaches (Chow et al., 2008; White et al.,
2010a; White et al., 2010b) have identified candidate miRNAs predicted to target the members
of the KLK family, and our own survey of the publically accessible databases reported here are in
agreement with their findings. The consensus of these results indicates that members of the hsa-
let-7 family of miRNAs are strong candidates for targeting specific KLKs, namely KLK6, and to
a lesser extent, KLK10 (Figure 5.1, Table 5.1, Table 5.2). Our TLDA profiling of the KLK6-
overexpressing OCa cell line, OVCAR-3, showed the majority of miRNAs predicted to target
KLK6 were decreased in expression (Table 5.1). All hsa-let-7 family members present on the
array showed the decreased expression of their mature miRNA compared to miRNAs derived
from normal ovary.
In addition to being KLK6-expressing, OVCAR-3 also satisfied the copy-number criteria for the
KLK locus. Our previous work in primary tumours indicated that when lost, the KLK locus
rarely exist clonally as one copy (Bayani et al., 2011; Bayani et al., 2008b). While a significant
portion of tumour cells could possess one copy, an admixture of tumour cells containing two or
more copies were also seen (Bayani et al., 2011; Bayani et al., 2008b). Other cytogenetic studies
support the observation that chromosome 19 is typically involved in both clonal and non-clonal
numerical and structural aberrations (Bayani et al., 2008b; Micci et al., 2009), and that the vast
majority of OCas are polyploidy rather than diploid (Bayani et al., 2002; Bayani et al., 2008a;
Bayani et al., 2008b; Taetle et al., 1999b). Therefore the averaging algorithms of aCGH studies
that report the frequent net loss of the locus (Cancer Genome Atlas Research Network, 2011;
Gorringe and Campbell, 2009), fail to take into account ploidy and copy-number heterogeneity
(Albertson et al., 2003). This falsely leads to the assumption that one copy of the locus exists
when designated as “lost” or “deleted”. FISH, using our multi-colour probe cocktail, showed
117
only two copies of the KLK locus at the normal mapping location (19q13.3/4). Moreover, SKY
analysis of the cell line showed OVCAR-3 to be structurally and numerically abnormal, with a
near triploid genome, consistent with reported karyotypic changes for this cell line at the
National Center for Biotechnology Information (NCBI) SKY/M-FISH and CGH Database
(http://www.ncbi.nlm.nih.gov/sky/). Although chromosomal abnormalities involving
chromosome 19 were identified (Figure 5.2A), the KLK locus was not subject to these structural
rearrangements, residing only at its normal mapping location (Figure 5.2C). Complementing the
karyotypic analyses, aCGH of OVCAR-3 confirmed the net loss of the entire KLK locus,
including KLK6 (Figure 5.2B), consistent with the finding of two copies against a triploid
background and accounting for the random losses of chromosome 19. Therefore, OVCAR-3
represents the KLK6-overexpressing and diploid/net loss copy-number scenario observed in our
primary tumour study.
Because copy-number has been known to play a role in miRNA expression (Croce, 2009; Dahiya
and Morin, 2010; Zhang et al., 2008), we also assessed the copy-number status of the hsa-let-7
family members commonly predicted to target KLK6, as well as other candidate miRNAs (Table
5.3). Indeed for both KLK6 and KLK10, there was good concordance between the copy-number
status and the expression of the miRNA (Table 5.3 and Table 5.4). The hsa-let-7 members map
to regions subjected to frequent copy number changes (Beroukhim et al., 2010; Cancer Genome
Atlas Research Network, 2011), with the net copy number changes exhibited in this cell line
representative of the genomic profiles seen in most OCas. With the exception of hsa-let-7f-2 and
hsa-let-7c, we found that expression of miRNAs were associated with their copy-number status.
The combined aCGH and karyotypic analyses of hsa-let-7a, which are encoded by three distinct
genes and are ultimately processed to result in identical mature hsa-let-7a (hsa-let-7a-1
(9q22.32), hsa-let-7a-2 (11q24.1), and hsa-let-7a-3 (22q.13.3)); showed the overall net loss of
genomic material (Figure 5.3). Because hsa-let-7a has been identified in many OCa miRNA
profiling and functional studies to be an important member of the OCa miRNA signature
(reviewed by Dahiya et al., 2010), and since the mature form results from three distinct copies
within the normal human genome, hsa-let-7a was chosen as a candidate for modulating KLK6
protein expression. Indeed when hsa-let-7a miRNA was transiently transfected into OVCAR-3,
there was a decrease in KLK6 protein in the hsa-let-7a-transfected cell line supernatant, as
compared to the scrambled-transfected control (Figure 5.1C). These findings were consistent
118
with those (Chow et al., 2008; White et al., 2010a; White et al., 2010b) showing the decrease in
KLK6 expression by hsa-let-7f , which possesses the identical seed sequence to hsa-let-7a
(Boyerinas et al., 2010 ) and binding site on the KLK6 3’ UTR (Chow et al., 2008; White et al.,
2010a; White et al., 2010b). The biological consequences of lowered KLK6 protein levels in
ovarian tumour cells has been shown experimentally to result in reduced migration, proliferation
and malignant potential (Bayani and Diamandis, 2011). Together with the experimental
evidence of hsa-let-7a’s tumour suppressive functions (Boyerinas et al., 2010), it supports an
important physiological role for KLK6 during pathogenesis. Bioinfomatical analyses by White
et al., 2010, as well as ours (Table 4.2) identified hsa-let-7a as a possible modulator for KLK10,
thus we also performed ELISA for KLK10 and found that hsa-let-7a was able to decrease
KLK10 protein expression but to a lower extent than for KLK6. Like KLK6, KLK10
overexpression has also been reported in OCa, and associated with worse prognosis in late-
staged disease (El Sherbini et al., 2011; Koh et al., 2011).
These findings add to the growing number of mRNA targets for the hsa-let-7 members, including
those genes implicated in many facets of tumourigenesis such as: MYC (Helland et al., 2011;
Nadiminty et al., 2011), RAS (Johnson et al., 2005), HMGA2 (Shell et al., 2007), Dicer
(Tokumaru et al., 2008), RAB40C (Yang et al., 2011b), E2F1, E2F3 (Bueno et al., 2010), E2F2,
CCND2 (Dong et al., 2010), and STAT (Wang et al., 2010). The observed decrease in
expression of the let-7 genes in many cancers (Boyerinas et al., 2010) is consistent with the
observed cellular dedifferentiation and re-expression of oncofetal genes seen in tumourigenesis
(Boyerinas et al., 2010; Shell et al., 2007). An ancient and highly conserved gene family, 13
members have been identified among humans (hsa-let-7a-1, 7a-2, 7a-3, 7b, 7c, 7d, 7e, 7f-1, 7f-2,
7g, 7i, hsa-mir-98, and hsa-mir-202) with identical seed sequences (Boyerinas et al., 2010). It is
intriguing to consider that these specific miRNAs exist in multiple copies because they perform
functions necessary to the survival of the organism. These findings also add to the current
understanding of KLK protein expression, and linking KLK6 to fundamental physiological
pathways mediated by this highly conserved and ancient family of miRNAs. The KLK gene
family also has an extensive phylogenetic history, with KLK6 being an ancestral member
(Pavlopoulou et al., 2010). Moreover, the bioinfomatic predictions performed to date suggest
members of the hsa-let-7 family may preferentially target KLK6 (Chow et al., 2008; White et al.,
2010a; White et al., 2010b), and KLK10 to a lesser extent (Table 5.3). The down regulation of
119
hsa-let-7a has been consistently reported among OCas (Dahiya and Morin, 2010), and its relative
expression shown to have an impact on survival outcomes of patients based on their
chemotherapy (Lu et al., 2011). The dysregulation of hsa-let-7a expression has profound
biological consequences that can be expressed through the changes in the levels of target
proteins, making such proteins putative biomarkers. Identifying and verifying the putative targets
of this and other miRNAs can provide the opportunity to discover relevant biomarkers and
protein-based signatures that could enhance genomically and transcriptomically-derived
signatures and delineate important clinical subgroups. Currently limited by the availability of
specific and sensitive antibodies for protein analyses, advances in mass-spectrometric technology
may help overcome this obstacle (Kulasingam et al., 2010).
In this study, we have demonstrated the hsa-let-7 family member, hsa-let-7a, is a contributing
modulator of KLK6 protein expression that is independent of the KLK6 copy-number status.
Additionally, we demonstrated that hsa-let-7a can also weakly affect the protein expression of
KLK10. Our cytogenomic analyses showed the strong contribution of copy-number and miRNA
expression in this representative OCa cell line. Moreover, the data present here and supported by
others (Chow et al., 2008; White et al., 2010a; White et al., 2010b), links the KLKs to a
fundamentally important regulator of gene expression essential to growth and development
(Boyerinas et al., 2010). With the continued validation of other hsa-let-7a targets, it is possible
that a clinically significant proteomic signature, including KLK6, can be developed to improve
the diagnostic and predictive needs so urgently needed in ovarian carcinoma.
120
Chapter 6
Summary and Future Directions
121
6 Summary and Future Directions
6.1 Summary
Ovarian carcinoma remains the leading cause of death from a gynecological malignancy. The
vague and lack of specific symptoms permits the virtually unnoticed progression of the disease;
such that at the time of diagnosis, the cancer has almost always metastasized. Although the FDA
has recently approved HE4, OVA1™ and the ROMA™ algorithm to the only other approved
marker CA-125, the lack of additional biomarkers for diagnosis contributes to the high mortality
rate. Prostate specific antigen (PSA/KLK3), the most well-known cancer biomarker, belongs to a
family of serine proteases – the Kallikreins (KLKs). This 15-member gene family holds great
potential as biomarkers for other cancers including OCa. Specifically, KLK6 has shown promise
as a biomarker for predicting disease free survival (DSF) and OS (reviewed by Bayani and
Diamandis, 2011), as it has been shown to be over-expressed at the mRNA and protein levels in
serous OCas as well as in the ascites fluids and serum of these patients. Aside from its ability to
degrade components of the extracellular matrix and basement membrane, KLK6 has been linked
to a variety of biological processes deemed the hallmarks of cancer (Hanahan and Weinberg,
2000, 2011). The mechanisms to the observed overexpression of KLK6 have been investigated
but are still poorly understood and should consider the integrative contributions at the level of
DNA, RNA and protein.
This current body of work examined the contributions of the observed karyotypic complexity
that characterizes OCa genomes, to the overexpression of a potentially important biomarker for
OCa – KLK6. It was systematically demonstrated for the first time in the published manuscript
entitled, “Distinct patterns of structural and numerical chromosomal instability characterize
sporadic ovarian cancer”, that the karyotypes of sporadic OCas experience a range of
chromosomal instability; both numerically (N-CIN) and structurally (S-CIN) (Bayani et al.,
2008a). Utilizing both interphase and metaphase FISH-based techniques, it was shown that
diploid tumours showed low-levels of N-CIN compared to triploid and tetraploid cancers. The
relationship of numerical instability and ploidy was highlighted by the finding that
supernumerary centrosomes were similarly identified most frequently among the triploid and
tetrapoloid cancers. This is consistent with the notion that mitotic segregation errors are
122
associated with centrosome amplification (Doxsey, 2002). The enumeration of breakpoints,
aided by the use of SKY, revealed that diploid cancers showed a higher number of structural
aberrations than triploid or tetrapoloid cancers. Though the presence of clonal and non-clonal
breakage events were on-going, many of the stable structural changes appeared to occur at the
diploid stage, since duplicate structural rearrangements were often present in the polyploid
cancers. These observations are in keeping with the significant role of the BRCA genes in OCa.
Recalling that expression profiling showed that sporadic cancers segregated into BRCA1- or
BRCA2-like cancers (Jazaeri, 2009; Jazaeri et al., 2002), and that hyper-methylation of BRCA1 is
a frequent event in sporadic cancers (Cancer Genome Atlas Research Network, 2011), the
finding that structural rearrangements appear to precede numerical changes supports the primary
role of the BRCA genes in the repair of DNA double stranded breaks. Secondarily, evidence
suggests that BRCA1 may additionally contribute to numerical and ploidy changes, via RB and
TP53 pathways (Deng, 2006). This initial study demonstrated the ability for OCa genomes to
undergo local copy-number changes independent of ploidy.
Having established the interplay between N-CIN and S-CIN among OCa cancers, the copy-
number status of the KLK locus in a series of cancer cell lines, primary OCa tumours and ascites
was determined and compared to KLK protein levels, in the published manuscript “Impact of
cytogenetic and genomic aberrations of the kallikrein locus in ovarian cancer” (Bayani et al.,
2008b). Utilizing molecular cytogenetic techniques including KLK-specific FISH, aCGH and
SKY, together with ELISA, it was first to demonstrate that the entire locus experiences copy-
number changes rather than individual KLK-family members and, by enlarge, the entire locus is
involved in unbalanced translocations without disruption of the cluster. However, in prostate
cancer there have been two reports describing extremely rare fusion transcripts between ETV4
(17q21) and KLK2 (Hermans et al., 2008), as well as ETV1 (7p21) and KLK2 (Pflueger et al.,
2011). Interestingly, the orientation of KLK2 and ETV4, and the resulting fusion product cannot
be explained by a single chromosome translocation occurrence; however, the recent proposal of
the chromothripsis model (Stephens et al., 2011), may offer some explanation of this rare event.
However, caution should be taken with respect to this model, as there is no concrete evidence to
suggest that the extremely complexed inter- and intra-chromsomal rearrangements detected by
DNA sequencing (Stephens et al., 2011) truly arises from a single catastrophic event (Righolt
and Mai, 2012). Unfortunately, no cytogenetic analysis was performed on these prostate cancer
123
patients to elucidate the possible chromosomal events leading to the fusion transcript.
Nevertheless, this initial KLK study also demonstrated that regions on 19q12 and 19q13 were
common sites of chromosomal breakage. In this small cohort of various cancer cell lines and
primary cancers, there was general concordance with the increase of KLK copy-number with
protein expression, however, this was not always the case, suggesting that tissue specificity and
other regulators of gene expression were at play. Interrogation of primary OCas similarly showed
that copy-number increases were also due to unbalanced translocations and more importantly,
was first to show that the KLK locus experiences copy-number heterogeneity.
Following this pilot study, the copy-number status was assessed in a larger and clinically
annotated series of untreated serous OCas in the published manuscript “Genomic instability and
copy-number heterogeneity of chromosome 19q, including the kallikrein locus, in ovarian
carcinomas” (Bayani et al., 2011). The purpose was to determine the range of KLK copy-
number, the extent of copy-number heterogeneity and instability, its relationship to KLK6
expression and whether such copy-number status was clinically relevant. A three-colour FISH
approach to a clinically-annotated serous OCa TMA, was the first comprehensive study to
examine KLK copy-number. Differentially labeled FISH probes for 19q12, 19q13.2 and the KLK
locus at 19q13.3/4 were utilized. Eighty-one patients were analysed to reveal that contrary to the
amassed aCGH findings, the KLK locus frequently experienced more gains/amplifications
(51.3%) than losses (29.6%). This confirms the diluting effects of contaminating normal cells,
normalization algorithms of aCGH experiments and the inability of aCGH to account for ploidy
(Albertson et al., 2003). The KLK locus was also found to be subjected to high copy-number
heterogeneity (p<0.001), particularly in cases with a net gain of the locus, supporting our earlier
observations (Bayani et al., 2008b). Moreover, these KLK locus gains were typically due to
structural rearrangements along 19q, also consistent with the previous findings (Bayani et al.,
2008b). Co-relations to clinical data showed that KLK copy-number status was not associated
with age, stage, grade or response to treatment. KLK6-specfic IHC showed a weak correlation to
copy-number indicating that other mechanisms together with copy-number drives the observed
protein over-expression. Interestingly, when 19q rearrangement status was analysed, older
women (mean 59.6 years) were more likely to have these gross structural rearrangements on 19q
than younger women (mean 50.8 years) (p = 0.003). Furthermore, grade 0/I tumours were more
likely to not have gross structural changes on 19q than higher grade tumours (II/II) (p<0.001),
124
suggesting that disruptions of chromosome 19 may confer a more aggressive phenotype. Indeed
several studies have shown that aberrations of chromosome 19 may be important in early OCa
tumorigenesis (Micci et al., 2009; Tsao et al., 2001). Moreover, the unique genomic properties
of chromosome 19, such as the high GC content, high density of genes and gene families, as well
as harbouring the largest number of miRNAs within the human genome (Grimwood et al., 2004),
supports the notion for a more malignant phenotype associated with numerical and structural
disruption of chromosome 19. Although these cancers were not classified as being Type I or
Type II (Kurman and Shih Ie, 2011), perhaps such changes on 19q are an additional
characteristic of Type I and Type II cancers. Kaplan Meier curves for DFS and OS reveal a
trend towards better DFS and OS in patients with a net loss of the KLK locus or in patients
without a 19q rearrangement. This could support the notion that the loss of KLK copy-number
may still contribute, in some way, to the decrease in transcript and protein; and that structural
disruption of 19q affects the expression of genes on chromosome 19q and/or genes associated
with their rearranged chromosomal partners. These results indicate that KLK copy-number
status may be clinically relevant, though further investigation is required.
The results of the aforementioned study demonstrated the modest contributing role of the
chromosomal changes to KLK6 copy number and to KLK6 protein expression. However, the
role of other genes influencing KLK6 protein expression had yet to be investigated. In the
submitted manuscript entitled “Modulation of KLK6 Protein Expression in the Ovarian Cancer
Cell Line OVCAR-3 by miRNA Let-7a” the role of miRNAs in regulating KLK6 protein
expression was explored utilizing an OCa cancer cell line model. Molecular cytogenomic
techniques including FISH, SKY and aCGH defined the karytoypic landscape of OVCAR-3.
With a triploid genome, the KLK locus was present in only two copies, indicating a net loss
relative to ploidy. In fact, this cell line underscores the limitation of aCGH and its interpretation.
In the absence of karyotypic analyses, the observed decrease in aCGH ratio is consistent with a
one-copy loss, leading most to believe that only one copy of the locus exists (ie. relative to a
normal diploid cell); when in reality two copies of the locus was found to be present against a
background of low-level numerical change for chromosome 19 and a triploid genome. Thus, this
cell line is representative of most OCas. The recent studies of the role of miRNAs in cancer
have highlighted the importance of regulation at the post-transcription level (Schickel et al.,
2008). To this end, we investigated the potential role of miRNAs in regulating KLK6 protein
125
expression. miRNA profiling of OVCAR-3 showed the increased and decreased expression of
miRNAs consistent with other published reports (reviewed by Dahiya et al. (Dahiya and Morin,
2010)). miRNA prediction databases identified the hsa-let-7 family of miRNAs to be strong
candidates of regulating miRNAs for KLK6; with those hsa-let-7 family members showing
decreased expression in OVCAR-3. Indeed, hsa-let-7 has been shown to be frequently decreased
in expression in OCa (Dahiya and Morin, 2010). Several studies have shown that regions of
copy-number change in OCas house miRNAs that are subjected to concomitant changes in
expression (Croce, 2009; Dahiya and Morin, 2010; Zhang et al., 2008). aCGH and karyotypic
analysis by SKY, confirmed the decrease in expression of these putative regulating miRNAs
were associated with the net loss of copy-number. Previous studies (Chow et al., 2008; White et
al., 2010a; White et al., 2010b) demonstrated the ability for hsa-let-7f to directly target the 3’-
UTR of KLK6. Thus, with hsa-let-7a possessing the identical seed sequence that characterized
this miRNA family, the ability of this hsa-let-7 member to affect KLK6 expression was assessed.
Precursor hsa-let-7a was transiently transfected to OVCAR-3 and shown to reduce KLK6
expression in the culture supernatant as compared to the scrambled control. Moreover, it was
demonstrated that this miRNA could also weakly affect the expression of another KLK member,
KLK10. Therefore, it has been demonstrated that KLK6 expression is affected not only by its
copy-number, but also by the copy-number and subsequent expression of miRNAs predicted to
target it. Moreover, the post-transcriptional regulation by miRNA can act to both greatly affect
protein expression or to fine-tune protein expression, independent of RNA expression.
In conclusion, the studies summarized here support the hypothesis that the observed
overexpression of KLK6 protein in OCa is mediated in part by its copy-number, but also
by miRNAs targeting it post-transcriptionally; and such changes in copy-number of the
KLK locus and regulating miRNAs are subjected to chromosomal instability that are
defined by numerical and structural alterations in the OCa genome. Clinically, patients
with a net loss of the locus experience a trend for better DFS and OS; as do patients
without structural alterations along chromosome 19q.
126
6.2 Future Directions
This body of work has demonstrated that KLK6 expression, in serous OCa, is influenced at the
genomic level and post-transcriptionally, by miRNAs; both by the copy-number KLK6 and by
the copy-number of the regulating miRNAs. Moreover, we have demonstrated that these genes
experience copy-number heterogeneity mediated by the underlying genomic instability that
characterizes these cancers. These findings have added to the current understanding of KLK6
protein expression, integrating regulation at the level of DNA and RNA; not only of KLK6, but
of the miRNA(s) regulating it. However, because miRNAs can potentially target hundreds of
mRNAs, there exists the opportunity to identify additional and new biomarkers with clinical
significance. Therefore, elevated KLK6 levels may coincide with the differential modulation of
other proteins and together serve as important biological signatures in OCa.
To this end, improvements in mass spectrometry (MS) now make it possible to identify
differentially expressed proteins (Kulasingam and Diamandis, 2008). It can be envisioned that
these comparative proteomic technologies can identify the differentially expressed proteins
regulated by changes in these miRNAs. One such technique, termed Stable isotope labeling by
amino acids in cell culture (SILAC) (Ong et al., 2002) allows for the detection of differences in
protein abundance between two (or more) samples. In SILAC, the cells are grown in media,
incorporating either “light” or “heavy” essential amino acids (typically lysine and arginine);
analogous to fluorescently labeled “normal” vs. fluorescently labeled “tumour/test” in array
comparative genomic hybridization (aCGH) or expression studies. When proteins from different
conditions are combined and analyzed simultaneously by MS, heavy and light-labeled peptides
of the same proteins will generate spectra that are slightly different due to a shift in mass caused
by the heavy isotopes, enabling relative quantification (Cox et al., 2009).While SILAC has only
been utilized by a handful of groups (Baek et al., 2008; Cox et al., 2009; Ebner and Selbach,
2011; Huang et al., 2012; Lossner et al., 2011; Mestdagh et al., 2010; Ong et al., 2002; Selbach
et al., 2008; Vinther et al., 2006; Yan et al., 2011; Yang et al., 2010); no one to date has
systematically used this method for OCas. Therefore, there is the opportunity to identify these
differentially expressed and putative biomarkers, as related to the currently accepted biomarkers,
CA-125 and HE4; and in addition to promising biomarkers such as KLK6.
127
Indeed, preliminary experiments performed by our group (Bayani et al., 2012. unpublished
observations), OVCAR-3 cells that were differentially labeled in culture with either “heavy”
isotopic arginine and lysine, or “light” isotopic arginine and lysine; were transiently transfected
with hsa-let-7a, as described in Chapter 5. ELISA confirmed the expected decrease in KLK6
protein expression in the culture media of the light-labeled-hsa-let-7a-transfected OVCAR-3
cells in comparison to the heavy-labeled-scramble-transfected OVCAR-3 cells. From these
cultures, total protein was extracted from the cell pellets and equal amounts were combined and
processed for MS. The resulting mass spectra can then be analysed using the freely available
MaxQuant Software (htpp://smsquant.sourceforge.net/)(Cox et al., 2009). It is expected that a
number of predicted targets will be modulated by the introduction of hsa-let-7a, and could serve
as putative OCa biomarkers. However, the assay is limited by a number of factors. Firstly, it
should be kept in mind that the miRNA targets listed in the aforementioned databases constitute
in silico predictions, and require independent validation. Therefore, it is conceivable for this
method to identify new miRNA targets while providing some supporting data to exclude others.
Secondly, the reliable identification of proteins by MS, depends heavily on the multi-stepped and
careful sample preparation and handling; as well as operation of the equipment by technical
experts. Thirdly, the variable inefficiencies of the transient transfection methods between
experiments could result in discrepancies during validation, thus a stable transfection approach
would provide a reliable platform for consequent validation experiments. Moreover, the model
presented here should be repeated not only in this cell line, but among other OCa cell lines and
other miRNAs to generate a more comprehensive miRNA-based differential signature. Only
then, and through biomarker candidate selection criteria (Kulasingam and Diamandis, 2008;
Kulasingam et al., 2010), can these putative biomarkers be tested rigorously in clinically-
annotated retrospective specimens. It is in this way that leading-edge proteomic techniques may
be integrated into the now mature and growing (epi)genomics and transcriptomic data presently
available.
Finally, the data linking miRNAs to KLK6 described in this body of work, may also reveal
another biological function for the increased mRNA KLK6; not only in cancers, but during
normal physiological and diseased states where copy-number is not affect - whereby the
increased KLK6 mRNA expression may serve as a natural miRNA “sponge” to sequester
miRNAs from other mRNA targets (Ebert and Sharp, 2010). Therefore, the presence of
128
functional, variant and non-coding KLK6 transcripts (Bayani and Diamandis, 2011) could serve
as miRNA decoys to modulate seemingly unrelated biological processes to those of already
associated with KLK6. To date, there is no research examining the role of KLK6 mRNA in this
context.
Despite the fact that the KLK locus was identified over a decade ago, the contributions to both
normal and disease physiology have yet to be fully elucidated. Clearly these future studies could
prove to be very interesting and moving our understanding of the KLK family forward.
129
References
Abeysinghe, S.S., Chuzhanova, N., Krawczak, M., Ball, E.V., and Cooper, D.N. (2003). Translocation and gross deletion breakpoints in human inherited disease and cancer I: Nucleotide composition and recombination-associated motifs. Hum Mutat 22, 229-244.
Adeyinka, A., Kytola, S., Mertens, F., Pandis, N., and Larsson, C. (2000). Spectral karyotyping and chromosome banding studies of primary breast carcinomas and their lymph node metastases. Int J Mol Med 5, 235-240.
Aggarwal, B.B., Vijayalekshmi, R.V., and Sung, B. (2009). Targeting inflammatory pathways for prevention and therapy of cancer: short-term friend, long-term foe. Clin Cancer Res 15, 425-430.
Ailles, L.E., and Weissman, I.L. (2007). Cancer stem cells in solid tumors. Curr Opin Biotechnol 18, 460-466.
Aimes, R.T., Zijlstra, A., Hooper, J.D., Ogbourne, S.M., Sit, M.L., Fuchs, S., Gotley, D.C., Quigley, J.P., and Antalis, T.M. (2003). Endothelial cell serine proteases expressed during vascular morphogenesis and angiogenesis. Thromb Haemost 89, 561-572.
Al-Romaih, K., Bayani, J., Vorobyova, J., Karaskova, J., Park, P.C., Zielenska, M., and Squire, J.A. (2003). Chromosomal instability in osteosarcoma and its association with centrosome abnormalities. Cancer Genet Cytogenet 144, 91-99.
Albertson, D.G., Collins, C., McCormick, F., and Gray, J.W. (2003). Chromosome aberrations in solid tumors. Nat Genet 34, 369-376.
Andreassen, P.R., Lohez, O.D., Lacroix, F.B., and Margolis, R.L. (2001). Tetraploid state induces p53-dependent arrest of nontransformed mammalian cells in G1. Mol Biol Cell 12, 1315-1328.
Anisowicz, A., Sotiropoulou, G., Stenman, G., Mok, S.C., and Sager, R. (1996). A novel protease homolog differentially expressed in breast and ovarian cancer. Mol Med 2, 624-636.
Ashby, E.L., Kehoe, P.G., and Love, S. (2010). Kallikrein-related peptidase 6 in Alzheimer's disease and vascular dementia. Brain Res, 1-10.
Auersperg, N., Wong, A.S., Choi, K.C., Kang, S.K., and Leung, P.C. (2001). Ovarian surface epithelium: biology, endocrinology, and pathology. Endocr Rev 22, 255-288.
Bae, I., Rih, J.K., Kim, H.J., Kang, H.J., Haddad, B., Kirilyuk, A., Fan, S., Avantaggiati, M.L., and Rosen, E.M. (2005). BRCA1 regulates gene expression for orderly mitotic progression. Cell Cycle 4, 1641-1666.
Baek, D., Villen, J., Shin, C., Camargo, F.D., Gygi, S.P., and Bartel, D.P. (2008). The impact of microRNAs on protein output. Nature 455, 64-71.
Barbarotto, E., Schmittgen, T.D., and Calin, G.A. (2008). MicroRNAs and cancer: profile, profile, profile. Int J Cancer 122, 969-977.
Bast, R.C., Jr., Hennessy, B., and Mills, G.B. (2009). The biology of ovarian cancer: new opportunities for translation. Nat Rev Cancer 9, 415-428.
130
Bast, R.C., Jr., Xu, F.J., Yu, Y.H., Barnhill, S., Zhang, Z., and Mills, G.B. (1998). CA 125: the past and the future. Int J Biol Markers 13, 179-187.
Batzer, M.A., and Deininger, P.L. (2002). Alu repeats and human genomic diversity. Nat Rev Genet 3, 370-379.
Bayani, J., Brenton, J.D., Macgregor, P.F., Beheshti, B., Albert, M., Nallainathan, D., Karaskova, J., Rosen, B., Murphy, J., Laframboise, S., et al. (2002). Parallel analysis of sporadic primary ovarian carcinomas by spectral karyotyping, comparative genomic hybridization, and expression microarrays. Cancer Res 62, 3466-3476.
Bayani, J., and Diamandis, E.P. (2011). Review: The physiology and pathobiology of human kallikrein-related peptidase 6 (KLK6). Clin Chem Lab Med.
Bayani, J., Kuzmanov, U., Saraon, P., Fung, W.A., Soosaipillai, A., Squire, J.A., and Diamandis, E.P. (In press). Copy-Number and Expression Alterations of miRNAs in the Ovarian Cancer Cell Line OVCAR-3: Impact on Kallikrein 6 Protein Expression. Clinical Chemistry.
Bayani, J., Marrano, P., Graham, C., Zheng, Y., Li, L., Katsaros, D., Lassus, H., Butzow, R., Squire, J.A., and Diamandis, E.P. (2011). Genomic instability and copy-number heterogeneity of chromosome 19q, including the kallikrein locus, in ovarian carcinomas. Mol Oncol 5, 48-60.
Bayani, J., Paderova, J., Murphy, J., Rosen, B., Zielenska, M., and Squire, J.A. (2008a). Distinct patterns of structural and numerical chromosomal instability characterize sporadic ovarian cancer. Neoplasia 10, 1057-1065.
Bayani, J., Paliouras, M., Planque, C., Shan, S.J., Graham, C., Squire, J.A., and Diamandis, E.P. (2008b). Impact of cytogenetic and genomic aberrations of the kallikrein locus in ovarian cancer. Mol Oncol 2, 250-260.
Bayani, J., Selvarajah, S., Maire, G., Vukovic, B., Al-Romaih, K., Zielenska, M., and Squire, J.A. (2007). Genomic mechanisms and measurement of structural and numerical instability in cancer cells. Semin Cancer Biol 17, 5-18.
Bayani, J., and Squire, J. (2004a). Multi-color FISH techniques. Curr Protoc Cell Biol Chapter 22, Unit 22 25.
Bayani, J., and Squire, J.A. (2002). Spectral karyotyping. Methods Mol Biol 204, 85-104.
Bayani, J., and Squire, J.A. (2004b). Fluorescence in situ Hybridization (FISH). Curr Protoc Cell Biol Chapter 22, Unit 22 24.
Bayani, J., and Squire, J.A. (2004c). Preparation of cytogenetic specimens from tissue samples. Curr Protoc Cell Biol Chapter 22, Unit 22 22.
Bayani, J., and Squire, J.A. (2007). Application and interpretation of FISH in biomarker studies. Cancer Lett 249, 97-109.
Bayani, J., Zielenska, M., Pandita, A., Al-Romaih, K., Karaskova, J., Harrison, K., Bridge, J.A., Sorensen, P., Thorner, P., and Squire, J.A. (2003). Spectral karyotyping identifies recurrent complex rearrangements of chromosomes 8, 17, and 20 in osteosarcomas. Genes Chromosomes Cancer 36, 7-16.
131
Beheshti, B., Karaskova, J., Park, P.C., Squire, J.A., and Beatty, B.G. (2000). Identification of a high frequency of chromosomal rearrangements in the centromeric regions of prostate cancer cell lines by sequential giemsa banding and spectral karyotyping. Mol Diagn 5, 23-32.
Bellacosa, A., de Feo, D., Godwin, A.K., Bell, D.W., Cheng, J.Q., Altomare, D.A., Wan, M., Dubeau, L., Scambia, G., Masciullo, V., et al. (1995). Molecular alterations of the AKT2 oncogene in ovarian and breast carcinomas. Int J Cancer 64, 280-285.
Benedet, J.L., Bender, H., Jones, H., 3rd, Ngan, H.Y., and Pecorelli, S. (2000). FIGO staging classifications and clinical practice guidelines in the management of gynecologic cancers. FIGO Committee on Gynecologic Oncology. Int J Gynaecol Obstet 70, 209-262.
Bernardini, M., Lee, C.H., Beheshti, B., Prasad, M., Albert, M., Marrano, P., Begley, H., Shaw, P., Covens, A., Murphy, J., et al. (2005). High-resolution mapping of genomic imbalance and identification of gene expression profiles associated with differential chemotherapy response in serous epithelial ovarian cancer. Neoplasia 7, 603-613.
Bernett, M.J., Blaber, S.I., Scarisbrick, I.A., Dhanarajan, P., Thompson, S.M., and Blaber, M. (2002). Crystal structure and biochemical characterization of human kallikrein 6 reveals that a trypsin-like kallikrein is expressed in the central nervous system. J Biol Chem 277, 24562-24570.
Beroukhim, R., Mermel, C.H., Porter, D., Wei, G., Raychaudhuri, S., Donovan, J., Barretina, J., Boehm, J.S., Dobson, J., Urashima, M., et al. (2010). The landscape of somatic copy-number alteration across human cancers. Nature 463, 899-905.
Betel, D., Koppal, A., Agius, P., Sander, C., and Leslie, C. (2010). Comprehensive modeling of microRNA targets predicts functional non-conserved and non-canonical sites. Genome Biol 11, R90.
Bickmore, W.A., and Teague, P. (2002). Influences of chromosome size, gene density and nuclear position on the frequency of constitutional translocations in the human population. Chromosome Res 10, 707-715.
Blaber, S.I., Scarisbrick, I.A., Bernett, M.J., Dhanarajan, P., Seavy, M.A., Jin, Y., Schwartz, M.A., Rodriguez, M., and Blaber, M. (2002). Enzymatic properties of rat myelencephalon-specific protease. Biochemistry 41, 1165-1173.
Blegen, H., Einhorn, N., Sjovall, K., Roschke, A., Ghadimi, B.M., McShane, L.M., Nilsson, B., Shah, K., Ried, T., and Auer, G. (2000). Prognostic significance of cell cycle proteins and genomic instability in borderline, early and advanced stage ovarian carcinomas. Int J Gynecol Cancer 10, 477-487.
Borgono, C.A., and Diamandis, E.P. (2004). The emerging roles of human tissue kallikreins in cancer. Nat Rev Cancer 4, 876-890.
Borgono, C.A., Kishi, T., Scorilas, A., Harbeck, N., Dorn, J., Schmalfeldt, B., Schmitt, M., and Diamandis, E.P. (2006). Human kallikrein 8 protein is a favorable prognostic marker in ovarian cancer. Clin Cancer Res 12, 1487-1493.
Borgono, C.A., Michael, I.P., and Diamandis, E.P. (2004). Human tissue kallikreins: physiologic roles and applications in cancer. Mol Cancer Res 2, 257-280.
132
Borgono, C.A., Michael, I.P., Komatsu, N., Jayakumar, A., Kapadia, R., Clayman, G.L., Sotiropoulou, G., and Diamandis, E.P. (2007). A potential role for multiple tissue kallikrein serine proteases in epidermal desquamation. J Biol Chem 282, 3640-3652.
Boyd, J., Sonoda, Y., Federici, M.G., Bogomolniy, F., Rhei, E., Maresco, D.L., Saigo, P.E., Almadrones, L.A., Barakat, R.R., Brown, C.L., et al. (2000). Clinicopathologic features of BRCA-linked and sporadic ovarian cancer. Jama 283, 2260-2265.
Boyerinas, B., Park, S.M., Hau, A., Murmann, A.E., and Peter, M.E. (2010). The role of let-7 in cell differentiation and cancer. Endocr Relat Cancer 17, F19-36.
Brookes, A.J. (1999). The essence of SNPs. Gene 234, 177-186.
Bueno, M.J., Gomez de Cedron, M., Laresgoiti, U., Fernandez-Piqueras, J., Zubiaga, A.M., and Malumbres, M. (2010). Multiple E2F-induced microRNAs prevent replicative stress in response to mitogenic signaling. Mol Cell Biol 30, 2983-2995.
Cahill, D.P., Lengauer, C., Yu, J., Riggins, G.J., Willson, J.K., Markowitz, S.D., Kinzler, K.W., and Vogelstein, B. (1998). Mutations of mitotic checkpoint genes in human cancers [see comments]. Nature 392, 300-303.
Calin, G.A., and Croce, C.M. (2007). Chromosomal rearrangements and microRNAs: a new cancer link with clinical implications. J Clin Invest 117, 2059-2066.
Cancer Genome Atlas Research Network (2011). Integrated genomic analyses of ovarian carcinoma. Nature 474, 609-615.
Cass, I., Baldwin, R.L., Varkey, T., Moslehi, R., Narod, S.A., and Karlan, B.Y. (2003). Improved survival in women with BRCA-associated ovarian carcinoma. Cancer 97, 2187-2195.
Castellano, E., and Downward, J. (2010). Role of RAS in the regulation of PI 3-kinase. Curr Top Microbiol Immunol 346, 143-169.
Ceccaroni, M., Chieco, P., Alboni, C., De Laco, P., Pagano, K., Ceccarelli, C., Santini, D., Taroni, B., and Pelusi, G. (2004). p53 expression, DNA ploidy and mitotic index as prognostic factors in patients with epithelial ovarian carcinoma. Tumori 90, 600-606.
Cho, N.Y., Kim, J.H., Moon, K.C., and Kang, G.H. (2009). Genomic hypomethylation and CpG island hypermethylation in prostatic intraepithelial neoplasm. Virchows Arch 454, 17-23.
Chow, T.F., Crow, M., Earle, T., El-Said, H., Diamandis, E.P., and Yousef, G.M. (2008). Kallikreins as microRNA targets: an in silico and experimental-based analysis. Biol Chem 389, 731-738.
Christophi, G.P., Isackson, P.J., Blaber, S., Blaber, M., Rodriguez, M., and Scarisbrick, I.A. (2004). Distinct promoters regulate tissue-specific and differential expression of kallikrein 6 in CNS demyelinating disease. J Neurochem 91, 1439-1449.
Christopoulos, T.K., and Diamandis, E.P. (1992). Enzymatically amplified time-resolved fluorescence immunoassay with terbium chelates. Anal Chem 64, 342-346.
Chuzhanova, N., Abeysinghe, S.S., Krawczak, M., and Cooper, D.N. (2003). Translocation and gross deletion breakpoints in human inherited disease and cancer II: Potential involvement of repetitive sequence elements in secondary structure formation between DNA ends. Hum Mutat 22, 245-251.
133
Cibula, D., Widschwendter, M., Majek, O., and Dusek, L. (2011). Tubal ligation and the risk of ovarian cancer: review and meta-analysis. Hum Reprod Update 17, 55-67.
Clements, J.A. (2008). Reflections on the tissue kallikrein and kallikrein-related peptidase family - from mice to men - what have we learnt in the last two decades? Biol Chem 389, 1447-1454.
Clements, J.A., Willemsen, N.M., Myers, S.A., and Dong, Y. (2004). The tissue kallikrein family of serine proteases: functional roles in human disease and potential as clinical biomarkers. Crit Rev Clin Lab Sci 41, 265-312.
Cohen, P., Graves, H.C., Peehl, D.M., Kamarei, M., Giudice, L.C., and Rosenfeld, R.G. (1992). Prostate-specific antigen (PSA) is an insulin-like growth factor binding protein-3 protease found in seminal plasma. J Clin Endocrinol Metab 75, 1046-1053.
Collins, F.S., Brooks, L.D., and Chakravarti, A. (1998). A DNA polymorphism discovery resource for research on human genetic variation. Genome Res 8, 1229-1231.
Coussens, L.M., and Werb, Z. (2002). Inflammation and cancer. Nature 420, 860-867.
Cox, J., Matic, I., Hilger, M., Nagaraj, N., Selbach, M., Olsen, J.V., and Mann, M. (2009). A practical guide to the MaxQuant computational platform for SILAC-based quantitative proteomics. Nat Protoc 4, 698-705.
Craig, J.M., and Bickmore, W.A. (1994). The distribution of CpG islands in mammalian chromosomes. Nat Genet 7, 376-382.
Cramer, D.W., Bast, R.C., Jr., Berg, C.D., Diamandis, E.P., Godwin, A.K., Hartge, P., Lokshin, A.E., Lu, K.H., McIntosh, M.W., Mor, G., et al. (2011). Ovarian cancer biomarker performance in prostate, lung, colorectal, and ovarian cancer screening trial specimens. Cancer Prev Res (Phila) 4, 365-374.
Cramer, D.W., Hutchison, G.B., Welch, W.R., Scully, R.E., and Ryan, K.J. (1983). Determinants of ovarian cancer risk. I. Reproductive experiences and family history. J Natl Cancer Inst 71, 711-716.
Cremer, T., Kreth, G., Koester, H., Fink, R.H., Heintzmann, R., Cremer, M., Solovei, I., Zink, D., and Cremer, C. (2000). Chromosome territories, interchromatin domain compartment, and nuclear matrix: an integrated view of the functional nuclear architecture. Crit Rev Eukaryot Gene Expr 10, 179-212.
Cremer, T., Kupper, K., Dietzel, S., and Fakan, S. (2004). Higher order chromatin architecture in the cell nucleus: on the way from structure to function. Biol Cell 96, 555-567.
Croce, C.M. (2009). Causes and consequences of microRNA dysregulation in cancer. Nat Rev Genet 10, 704-714.
Crum, C.P., Drapkin, R., Miron, A., Ince, T.A., Muto, M., Kindelberger, D.W., and Lee, Y. (2007). The distal fallopian tube: a new model for pelvic serous carcinogenesis. Curr Opin Obstet Gynecol 19, 3-9.
Dahiya, N., and Morin, P.J. (2010). MicroRNAs in ovarian carcinomas. Endocr Relat Cancer 17, F77-89.
Darling, M.R., Jackson-Boeters, L., Daley, T.D., and Diamandis, E.P. (2006). Human kallikrein 6 expression in salivary gland tumors. J Histochem Cytochem 54, 337-342.
134
Day, T.G., Jr., Gallager, H.S., and Rutledge, F.N. (1975). Epithelial carcinoma of the ovary:prognostic importance of histologic grade. Natl Cancer Inst Monogr 42, 15-21.
Dean, M., Fojo, T., and Bates, S. (2005). Tumour stem cells and drug resistance. Nat Rev Cancer 5, 275-284.
Debela, M., Beaufort, N., Magdolen, V., Schechter, N.M., Craik, C.S., Schmitt, M., Bode, W., and Goettig, P. (2008). Structures and specificity of the human kallikrein-related peptidases KLKs 4, 5, 6, and 7. Biol Chem 389, 623-632.
Deng, C.X. (2006). BRCA1: cell cycle checkpoint, genetic instability, DNA damage response and cancer evolution. Nucleic Acids Res 34, 1416-1426.
Deng, C.X., and Wang, R.H. (2003). Roles of BRCA1 in DNA damage repair: a link between development and cancer. Hum Mol Genet 12 Spec No 1, R113-123.
Diamandis, E.P., and Yousef, G.M. (2002). Human tissue kallikreins: a family of new cancer biomarkers. Clin Chem 48, 1198-1205.
Diamandis, E.P., Yousef, G.M., Luo, L.Y., Magklara, A., and Obiezu, C.V. (2000a). The new human kallikrein gene family: implications in carcinogenesis. Trends Endocrinol Metab 11, 54-60.
Diamandis, E.P., Yousef, G.M., Soosaipillai, A.R., and Bunting, P. (2000b). Human kallikrein 6 (zyme/protease M/neurosin): a new serum biomarker of ovarian carcinoma. Clin Biochem 33, 579-583.
Diamandis, E.P., Yousef, G.M., Soosaipillai, A.R., Grass, L., Porter, A., Little, S., and Sotiropoulou, G. (2000c). Immunofluorometric assay of human kallikrein 6 (zyme/protease M/neurosin) and preliminary clinical applications. Clin Biochem 33, 369-375.
Dong, Q., Meng, P., Wang, T., Qin, W., Qin, W., Wang, F., Yuan, J., Chen, Z., Yang, A., and Wang, H. (2010). MicroRNA let-7a inhibits proliferation of human prostate cancer cells in vitro and in vivo by targeting E2F2 and CCND2. PLoS One 5, e10147.
Doxsey, S. (2002). Duplicating dangerously: linking centrosome duplication and aneuploidy. Mol Cell 10, 439-440.
Drapkin, R., von Horsten, H.H., Lin, Y., Mok, S.C., Crum, C.P., Welch, W.R., and Hecht, J.L. (2005). Human epididymis protein 4 (HE4) is a secreted glycoprotein that is overexpressed by serous and endometrioid ovarian carcinomas. Cancer Res 65, 2162-2169.
Ebert, M.S., and Sharp, P.A. (2010). Emerging roles for natural microRNA sponges. Curr Biol 20, R858-861.
Ebner, O.A., and Selbach, M. (2011). Whole cell proteome regulation by microRNAs captured in a pulsed SILAC mass spectrometry approach. Methods Mol Biol 725, 315-331.
Eissa, A., Amodeo, V., Smith, C.R., and Diamandis, E.P. (2010). Kallikrein-related peptidase-8 (KLK8) is an active serine protease in human epidermis and sweat and is involved in a skin barrier proteolytic cascade. J Biol Chem.
135
El Sherbini, M.A., Sallam, M.M., Shaban, E.A., and El-Shalakany, A.H. (2011). Diagnostic Value of Serum Kallikrein-Related Peptidases 6 and 10 Versus CA125 in Ovarian Cancer. Int J Gynecol Cancer 21, 625-632.
Emami, N., and Diamandis, E.P. (2008). Human kallikrein-related peptidase 14 (KLK14) is a new activator component of the KLK proteolytic cascade. Possible function in seminal plasma and skin. J Biol Chem 283, 3031-3041.
Esquela-Kerscher, A., and Slack, F.J. (2006). Oncomirs - microRNAs with a role in cancer. Nat Rev Cancer 6, 259-269.
Etemadmoghadam, D., deFazio, A., Beroukhim, R., Mermel, C., George, J., Getz, G., Tothill, R., Okamoto, A., Raeder, M.B., Harnett, P., et al. (2009). Integrated genome-wide DNA copy number and expression analysis identifies distinct mechanisms of primary chemoresistance in ovarian carcinomas. Clin Cancer Res 15, 1417-1427.
Ewald, B., Sampath, D., and Plunkett, W. (2008). Nucleoside analogs: molecular mechanisms signaling cell death. Oncogene 27, 6522-6537.
Fathalla, M.F. (1971). Incessant ovulation--a factor in ovarian neoplasia? Lancet 2, 163.
Fearon, E.R., and Vogelstein, B. (1990). A genetic model for colorectal tumorigenesis. Cell 61, 759-767.
Ferri, C.P., Prince, M., Brayne, C., Brodaty, H., Fratiglioni, L., Ganguli, M., Hall, K., Hasegawa, K., Hendrie, H., Huang, Y., et al. (2005). Global prevalence of dementia: a Delphi consensus study. Lancet 366, 2112-2117.
Frisch, S.M., and Screaton, R.A. (2001). Anoikis mechanisms. Curr Opin Cell Biol 13, 555-562.
Frohling, S., and Dohner, H. (2008). Chromosomal abnormalities in cancer. N Engl J Med 359, 722-734.
Fukasawa, K. (2005). Centrosome amplification, chromosome instability and cancer development. Cancer Lett 230, 6-19.
Gabril, M., White, N.M., Moussa, M., Chow, T.F., Metias, S.M., Fatoohi, E., and Yousef, G.M. (2010). Immunohistochemical analysis of kallikrein-related peptidases in the normal kidney and renal tumors: potential clinical implications. Biol Chem 391, 403-409.
Gan, L., Lee, I., Smith, R., Argonza-Barrett, R., Lei, H., McCuaig, J., Moss, P., Paeper, B., and Wang, K. (2000). Sequencing and expression analysis of the serine protease gene cluster located in chromosome 19q13 region. Gene 257, 119-130.
Ghosh, M.C., Grass, L., Soosaipillai, A., Sotiropoulou, G., and Diamandis, E.P. (2004). Human kallikrein 6 degrades extracellular matrix proteins and may enhance the metastatic potential of tumour cells. Tumour Biol 25, 193-199.
Gilks, C.B., and Prat, J. (2009). Ovarian carcinoma pathology and genetics: recent advances. Hum Pathol 40, 1213-1223.
Gisselsson, D., and Hoglund, M. (2005). Connecting mitotic instability and chromosome aberrations in cancer--can telomeres bridge the gap? Semin Cancer Biol 15, 13-23.
Gisselsson, D., Lv, M., Tsao, S.W., Man, C., Jin, C., Hoglund, M., Kwong, Y.L., and Jin, Y. (2005). Telomere-mediated mitotic disturbances in immortalized ovarian epithelial cells
136
reproduce chromosomal losses and breakpoints from ovarian carcinoma. Genes Chromosomes Cancer 42, 22-33.
Goard, C.A., Bromberg, I.L., Elliott, M.B., and Diamandis, E.P. (2007). A consolidated catalogue and graphical annotation of dbSNP polymorphisms in the human tissue kallikrein (KLK) locus. Mol Oncol 1, 303-312.
Gonda, T.A., Tu, S., and Wang, T.C. (2009). Chronic inflammation, the tumor microenvironment and carcinogenesis. Cell Cycle 8, 2005-2013.
Gorringe, K.L., and Campbell, I.G. (2009). Large-scale genomic analysis of ovarian carcinomas. Mol Oncol 3, 157-164.
Gorringe, K.L., Chin, S.F., Pharoah, P., Staines, J.M., Oliveira, C., Edwards, P.A., and Caldas, C. (2005). Evidence that both genetic instability and selection contribute to the accumulation of chromosome alterations in cancer. Carcinogenesis 26, 923-930.
Gorringe, K.L., Jacobs, S., Thompson, E.R., Sridhar, A., Qiu, W., Choong, D.Y., and Campbell, I.G. (2007). High-resolution single nucleotide polymorphism array analysis of epithelial ovarian cancer reveals numerous microdeletions and amplifications. Clin Cancer Res 13, 4731-4739.
Grady, D.L., Ratliff, R.L., Robinson, D.L., McCanlies, E.C., Meyne, J., and Moyzis, R.K. (1992). Highly conserved repetitive DNA sequences are present at human centromeres. Proc Natl Acad Sci U S A 89, 1695-1699.
Griffiths-Jones, S., Saini, H.K., van Dongen, S., and Enright, A.J. (2008). miRBase: tools for microRNA genomics. Nucleic Acids Res 36, D154-158.
Grigorova, M., Staines, J.M., Ozdag, H., Caldas, C., and Edwards, P.A. (2004). Possible causes of chromosome instability: comparison of chromosomal abnormalities in cancer cell lines with mutations in BRCA1, BRCA2, CHK2 and BUB1. Cytogenet Genome Res 104, 333-340.
Grimson, A., Farh, K.K., Johnston, W.K., Garrett-Engele, P., Lim, L.P., and Bartel, D.P. (2007). MicroRNA targeting specificity in mammals: determinants beyond seed pairing. Mol Cell 27, 91-105.
Grimwood, J., Gordon, L.A., Olsen, A., Terry, A., Schmutz, J., Lamerdin, J., Hellsten, U., Goodstein, D., Couronne, O., Tran-Gyamfi, M., et al. (2004). The DNA sequence and biology of human chromosome 19. Nature 428, 529-535.
Gudmundsson, J., Johannesdottir, G., Arason, A., Bergthorsson, J.T., Ingvarsson, S., Egilsson, V., and Barkardottir, R.B. (1996). Frequent occurrence of BRCA2 linkage in Icelandic breast cancer families and segregation of a common BRCA2 haplotype. Am J Hum Genet 58, 749-756.
Hanahan, D., and Weinberg, R.A. (2000). The hallmarks of cancer. Cell 100, 57-70.
Hanahan, D., and Weinberg, R.A. (2011). Hallmarks of cancer: the next generation. Cell 144, 646-674.
Harvey, T.J., Hooper, J.D., Myers, S.A., Stephenson, S.A., Ashworth, L.K., and Clements, J.A. (2000). Tissue-specific expression patterns and fine mapping of the human kallikrein (KLK) locus on proximal 19q13.4. J Biol Chem 275, 37397-37406.
137
Hastings, P.J., Lupski, J.R., Rosenberg, S.M., and Ira, G. (2009). Mechanisms of change in gene copy number. Nat Rev Genet 10, 551-564.
Hebb, A.L., Bhan, V., Wishart, A.D., Moore, C.S., and Robertson, G.S. (2010). Human kallikrein 6 cerebrospinal levels are elevated in multiple sclerosis. Curr Drug Discov Technol 7, 137-140.
Helland, A., Anglesio, M.S., George, J., Cowin, P.A., Johnstone, C.N., House, C.M., Sheppard, K.E., Etemadmoghadam, D., Melnyk, N., Rustgi, A.K., et al. (2011). Deregulation of MYCN, LIN28B and LET7 in a molecular subtype of aggressive high-grade serous ovarian cancers. PLoS One 6, e18064.
Hellstrom, I., Raycraft, J., Hayden-Ledbetter, M., Ledbetter, J.A., Schummer, M., McIntosh, M., Drescher, C., Urban, N., and Hellstrom, K.E. (2003). The HE4 (WFDC2) protein is a biomarker for ovarian carcinoma. Cancer Res 63, 3695-3700.
Henkhaus, R.S., Gerner, E.W., and Ignatenko, N.A. (2008). Kallikrein 6 is a mediator of K-RAS-dependent migration of colon carcinoma cells. Biol Chem 389, 757-764.
Hermans, K.G., Bressers, A.A., van der Korput, H.A., Dits, N.F., Jenster, G., and Trapman, J. (2008). Two unique novel prostate-specific and androgen-regulated fusion partners of ETV4 in prostate cancer. Cancer Res 68, 3094-3098.
Hinds, L., and Price, J. (2010). Menopause, hormone replacement and gynaecological cancers. Menopause Int 16, 89-93.
Hoeijmakers, J.H. (2001). Genome maintenance mechanisms for preventing cancer. Nature 411, 366-374.
Hoffman, B.R., Katsaros, D., Scorilas, A., Diamandis, P., Fracchioli, S., Rigault de la Longrais, I.A., Colgan, T., Puopolo, M., Giardina, G., Massobrio, M., et al. (2002). Immunofluorometric quantitation and histochemical localisation of kallikrein 6 protein in ovarian cancer tissue: a new independent unfavourable prognostic biomarker. Br J Cancer 87, 763-771.
Hogberg, T., and Kagedal, B. (1992). Long-term follow-up of ovarian cancer with monthly determinations of serum CA 125. Gynecol Oncol 46, 191-198.
Hollenberg, M.D., and Compton, S.J. (2002). International Union of Pharmacology. XXVIII. Proteinase-activated receptors. Pharmacol Rev 54, 203-217.
Hollenberg, M.D., Oikonomopoulou, K., Hansen, K.K., Saifeddine, M., Ramachandran, R., and Diamandis, E.P. (2008). Kallikreins and proteinase-mediated signaling: proteinase-activated receptors (PARs) and the pathophysiology of inflammatory diseases and cancer. Biol Chem 389, 643-651.
Hoskins, W.J. (1993). Surgical staging and cytoreductive surgery of epithelial ovarian cancer. Cancer 71, 1534-1540.
Huang, T.C., Saharabuddhe, N.A., Kim, M.S., Getnet, D., Yang, Y., Peterson, J.M., Ghosh, B., Chaerkady, R., Leach, S.D., Marchionni, L., et al. (2012). Regulation of Lipid Metabolism By Dicer Revealed Through SILAC Mice. J Proteome Res.
138
Huang, W., Zhong, J., Wu, L.Y., Yu, L.F., Tian, X.L., Zhang, Y.F., and Li, B. (2007). Downregulation and CpG island hypermethylation of NES1/hK10 gene in the pathogenesis of human gastric cancer. Cancer Lett 251, 78-85.
Inoue, Y., Yokobori, T., Yokoe, T., Toiyama, Y., Miki, C., Mimori, K., Mori, M., and Kusunoki, M. (2010). Clinical Significance of Human Kallikrein7 Gene Expression in Colorectal Cancer. Ann Surg Oncol.
Jacobs, I., and Bast, R.C., Jr. (1989). The CA 125 tumour-associated antigen: a review of the literature. Hum Reprod 4, 1-12.
Jacobs, I., Oram, D., Fairbanks, J., Turner, J., Frost, C., and Grudzinskas, J.G. (1990). A risk of malignancy index incorporating CA 125, ultrasound and menopausal status for the accurate preoperative diagnosis of ovarian cancer. Br J Obstet Gynaecol 97, 922-929.
Jazaeri, A.A. (2009). Molecular profiles of hereditary epithelial ovarian cancers and their implications for the biology of this disease. Mol Oncol 3, 151-156.
Jazaeri, A.A., Yee, C.J., Sotiriou, C., Brantley, K.R., Boyd, J., and Liu, E.T. (2002). Gene expression profiles of BRCA1-linked, BRCA2-linked, and sporadic ovarian cancers. J Natl Cancer Inst 94, 990-1000.
Johnson, S.M., Grosshans, H., Shingara, J., Byrom, M., Jarvis, R., Cheng, A., Labourier, E., Reinert, K.L., Brown, D., and Slack, F.J. (2005). RAS is regulated by the let-7 microRNA family. Cell 120, 635-647.
Jones, R.N. (2005). McClintock's controlling elements: the full story. Cytogenet Genome Res 109, 90-103.
Jung, M., Schaefer, A., Steiner, I., Kempkensteffen, C., Stephan, C., Erbersdobler, A., and Jung, K. (2010). Robust microRNA stability in degraded RNA preparations from human tissue and cell samples. Clin Chem 56, 998-1006.
Kabawat, S.E., Bast, R.C., Jr., Bhan, A.K., Welch, W.R., Knapp, R.C., and Colvin, R.B. (1983). Tissue distribution of a coelomic-epithelium-related antigen recognized by the monoclonal antibody OC125. Int J Gynecol Pathol 2, 275-285.
Kalaria, R.N., Maestre, G.E., Arizaga, R., Friedland, R.P., Galasko, D., Hall, K., Luchsinger, J.A., Ogunniyi, A., Perry, E.K., Potocnik, F., et al. (2008). Alzheimer's disease and vascular dementia in developing countries: prevalence, management, and risk factors. Lancet Neurol 7, 812-826.
Karst, A.M., and Drapkin, R. (2010). Ovarian cancer pathogenesis: a model in evolution. J Oncol 2010, 932371.
Khanna, K.K., and Jackson, S.P. (2001). DNA double-strand breaks: signaling, repair and the cancer connection. Nat Genet 27, 247-254.
Kim, H., Scorilas, A., Katsaros, D., Yousef, G.M., Massobrio, M., Fracchioli, S., Piccinno, R., Gordini, G., and Diamandis, E.P. (2001). Human kallikrein gene 5 (KLK5) expression is an indicator of poor prognosis in ovarian cancer. Br J Cancer 84, 643-650.
Kim, J.T., Song, E.Y., Chung, K.S., Kang, M.A., Kim, J.W., Kim, S.J., Yeom, Y.I., Kim, J.H., Kim, K.H., and Lee, H.G. (2011). Up-regulation and clinical significance of serine protease kallikrein 6 in colon cancer. Cancer 117, 2608-2619.
139
Kimmig, R., Wimberger, P., Hillemanns, P., Kapsner, T., Caspari, C., and Hepp, H. (2002). Multivariate analysis of the prognostic significance of DNA-ploidy and S-phase fraction in ovarian cancer determined by flow cytometry following detection of cytokeratin-labeled tumor cells. Gynecol Oncol 84, 21-31.
King, M.C., Marks, J.H., and Mandell, J.B. (2003). Breast and ovarian cancer risks due to inherited mutations in BRCA1 and BRCA2. Science 302, 643-646.
Kishibe, M., Bando, Y., Terayama, R., Namikawa, K., Takahashi, H., Hashimoto, Y., Ishida-Yamamoto, A., Jiang, Y.P., Mitrovic, B., Perez, D., et al. (2007). Kallikrein 8 is involved in skin desquamation in cooperation with other kallikreins. J Biol Chem 282, 5834-5841.
Klucky, B., Mueller, R., Vogt, I., Teurich, S., Hartenstein, B., Breuhahn, K., Flechtenmacher, C., Angel, P., and Hess, J. (2007). Kallikrein 6 induces E-cadherin shedding and promotes cell proliferation, migration, and invasion. Cancer Res 67, 8198-8206.
Koh, S.C., Razvi, K., Chan, Y.H., Narasimhan, K., Ilancheran, A., Low, J.J., and Choolani, M. (2011). The association with age, human tissue kallikreins 6 and 10 and hemostatic markers for survival outcome from epithelial ovarian cancer. Arch Gynecol Obstet 284, 183-190.
Kojima, K., Fujita, Y., Nozawa, Y., Deguchi, T., and Ito, M. (2010). MiR-34a attenuates paclitaxel-resistance of hormone-refractory prostate cancer PC3 cells through direct and indirect mechanisms. Prostate 70, 1501-1512.
Komatsu, N., Saijoh, K., Kuk, C., Liu, A.C., Khan, S., Shirasaki, F., Takehara, K., and Diamandis, E.P. (2007a). Human tissue kallikrein expression in the stratum corneum and serum of atopic dermatitis patients. Exp Dermatol 16, 513-519.
Komatsu, N., Saijoh, K., Kuk, C., Shirasaki, F., Takehara, K., and Diamandis, E.P. (2007b). Aberrant human tissue kallikrein levels in the stratum corneum and serum of patients with psoriasis: dependence on phenotype, severity and therapy. Br J Dermatol 156, 875-883.
Komatsu, N., Saijoh, K., Otsuki, N., Kishi, T., Micheal, I.P., Obiezu, C.V., Borgono, C.A., Takehara, K., Jayakumar, A., Wu, H.K., et al. (2007c). Proteolytic processing of human growth hormone by multiple tissue kallikreins and regulation by the serine protease inhibitor Kazal-Type5 (SPINK5) protein. Clin Chim Acta 377, 228-236.
Komatsu, N., Saijoh, K., Toyama, T., Ohka, R., Otsuki, N., Hussack, G., Takehara, K., and Diamandis, E.P. (2005). Multiple tissue kallikrein mRNA and protein expression in normal skin and skin diseases. Br J Dermatol 153, 274-281.
Komatsu, N., Takata, M., Otsuki, N., Toyama, T., Ohka, R., Takehara, K., and Saijoh, K. (2003). Expression and localization of tissue kallikrein mRNAs in human epidermis and appendages. J Invest Dermatol 121, 542-549.
Kops, G.J., Weaver, B.A., and Cleveland, D.W. (2005). On the road to cancer: aneuploidy and the mitotic checkpoint. Nat Rev Cancer 5, 773-785.
Kountourakis, P., Psyrri, A., Scorilas, A., Camp, R., Markakis, S., Kowalski, D., Diamandis, E.P., and Dimopoulos, M.A. (2008). Prognostic value of kallikrein-related peptidase 6 protein expression levels in advanced ovarian cancer evaluated by automated quantitative analysis (AQUA). Cancer Sci 99, 2224-2229.
140
Kristensen, G.B., Kildal, W., Abeler, V.M., Kaern, J., Vergote, I., Trope, C.G., and Danielsen, H.E. (2003). Large-scale genomic instability predicts long-term outcome for women with invasive stage I ovarian cancer. Ann Oncol 14, 1494-1500.
Kruglyak, L., and Nickerson, D.A. (2001). Variation is the spice of life. Nat Genet 27, 234-236.
Kufe, D.W. (2009). Mucins in cancer: function, prognosis and therapy. Nat Rev Cancer 9, 874-885.
Kulasingam, V., and Diamandis, E.P. (2008). Strategies for discovering novel cancer biomarkers through utilization of emerging technologies. Nat Clin Pract Oncol 5, 588-599.
Kulasingam, V., Pavlou, M.P., and Diamandis, E.P. (2010). Integrating high-throughput technologies in the quest for effective biomarkers for ovarian cancer. Nat Rev Cancer 10, 371-378.
Kupferman, M.E., Patel, V., Sriuranpong, V., Amornphimoltham, P., Jasser, S.A., Mandal, M., Zhou, G., Wang, J., Coombes, K., Multani, A., et al. (2007). Molecular analysis of anoikis resistance in oral cavity squamous cell carcinoma. Oral Oncol 43, 440-454.
Kurlender, L., Borgono, C., Michael, I.P., Obiezu, C., Elliott, M.B., Yousef, G.M., and Diamandis, E.P. (2005). A survey of alternative transcripts of human tissue kallikrein genes. Biochim Biophys Acta 1755, 1-14.
Kurman, R.J., and Shih Ie, M. (2011). Molecular pathogenesis and extraovarian origin of epithelial ovarian cancer--shifting the paradigm. Hum Pathol 42, 918-931.
Kyriakopoulou, L.G., Yousef, G.M., Scorilas, A., Katsaros, D., Massobrio, M., Fracchioli, S., and Diamandis, E.P. (2003). Prognostic value of quantitatively assessed KLK7 expression in ovarian cancer. Clin Biochem 36, 135-143.
Kytola, S., Rummukainen, J., Nordgren, A., Karhu, R., Farnebo, F., Isola, J., and Larsson, C. (2000). Chromosomal alterations in 15 breast cancer cell lines by comparative genomic hybridization and spectral karyotyping. Genes Chromosomes Cancer 28, 308-317.
Ladiges, W., Kemp, C., Packenham, J., and Velazquez, J. (2004). Human gene variation: from SNPs to phenotypes. Mutat Res 545, 131-139.
Lai, J., Myers, S.A., Lawrence, M.G., Odorico, D.M., and Clements, J.A. (2009). Direct progesterone receptor and indirect androgen receptor interactions with the kallikrein-related peptidase 4 gene promoter in breast and prostate cancer. Mol Cancer Res 7, 129-141.
Landen, C.N., Jr., Birrer, M.J., and Sood, A.K. (2008). Early events in the pathogenesis of epithelial ovarian cancer. J Clin Oncol 26, 995-1005.
Landry, J.R., Mager, D.L., and Wilhelm, B.T. (2003). Complex controls: the role of alternative promoters in mammalian genomes. Trends Genet 19, 640-648.
Lassus, H., and Butzow, R. (2007). The classification of p53 immunohistochemical staining results and patient outcome in ovarian cancer. Br J Cancer 96, 1621-1622; author reply 1623-1624.
Lawrence, M.G., Lai, J., and Clements, J.A. (2010). Kallikreins on steroids: structure, function, and hormonal regulation of prostate-specific antigen and the extended kallikrein locus. Endocr Rev 31, 407-446.
141
Ledermann, J.A., Marth, C., Carey, M.S., Birrer, M., Bowtell, D.D., Kaye, S., McNeish, I., Oza, A., Scambia, G., Rustin, G., et al. (2011). Role of molecular agents and targeted therapy in clinical trials for women with ovarian cancer. Int J Gynecol Cancer 21, 763-770.
Lee, R.C., Feinbaum, R.L., and Ambros, V. (1993). The C. elegans heterochronic gene lin-4 encodes small RNAs with antisense complementarity to lin-14. Cell 75, 843-854.
Lee, Y., Miron, A., Drapkin, R., Nucci, M.R., Medeiros, F., Saleemuddin, A., Garber, J., Birch, C., Mou, H., Gordon, R.W., et al. (2007). A candidate precursor to serous carcinoma that originates in the distal fallopian tube. J Pathol 211, 26-35.
Lengauer, C., Kinzler, K.W., and Vogelstein, B. (1997). Genetic instability in colorectal cancers. Nature 386, 623-627.
Li, B., Goyal, J., Dhar, S., Dimri, G., Evron, E., Sukumar, S., Wazer, D.E., and Band, V. (2001). CpG methylation as a basis for breast tumor-specific loss of NES1/kallikrein 10 expression. Cancer Res 61, 8014-8021.
Li, L., McCormack, A.A., Nicholson, J.M., Fabarius, A., Hehlmann, R., Sachs, R.K., and Duesberg, P.H. (2009a). Cancer-causing karyotypes: chromosomal equilibria between destabilizing aneuploidy and stabilizing selection for oncogenic function. Cancer Genet Cytogenet 188, 1-25.
Li, X., Liu, J., Wang, Y., Zhang, L., Ning, L., and Feng, Y. (2009b). Parallel underexpression of kallikrein 5 and kallikrein 7 mRNA in breast malignancies. Cancer Sci 100, 601-607.
Liehr, T., Starke, H., Weise, A., Lehrer, H., and Claussen, U. (2004). Multicolor FISH probe sets and their applications. Histol Histopathol 19, 229-237.
Lilja, H., Ulmert, D., and Vickers, A.J. (2008). Prostate-specific antigen and prostate cancer: prediction, detection and monitoring. Nat Rev Cancer 8, 268-278.
Lim, G., Karaskova, J., Beheshti, B., Vukovic, B., Bayani, J., Selvarajah, S., Watson, S.K., Lam, W.L., Zielenska, M., and Squire, J.A. (2005). An integrated mBAND and submegabase resolution tiling set (SMRT) CGH array analysis of focal amplification, microdeletions, and ladder structures consistent with breakage-fusion-bridge cycle events in osteosarcoma. Genes Chromosomes Cancer 42, 392-403.
Lin, R., Nagai, Y., Sladek, R., Bastien, Y., Ho, J., Petrecca, K., Sotiropoulou, G., Diamandis, E.P., Hudson, T.J., and White, J.H. (2002). Expression profiling in squamous carcinoma cells reveals pleiotropic effects of vitamin D3 analog EB1089 signaling on cell proliferation, differentiation, and immune system regulation. Mol Endocrinol 16, 1243-1256.
Little, S.P., Dixon, E.P., Norris, F., Buckley, W., Becker, G.W., Johnson, M., Dobbins, J.R., Wyrick, T., Miller, J.R., MacKellar, W., et al. (1997). Zyme, a novel and potentially amyloidogenic enzyme cDNA isolated from Alzheimer's disease brain. J Biol Chem 272, 25135-25142.
Loeb, L.A., and Christians, F.C. (1996). Multiple mutations in human cancers. Mutat Res 350, 279-286.
Lossner, C., Meier, J., Warnken, U., Rogers, M.A., Lichter, P., Pscherer, A., and Schnolzer, M. (2011). Quantitative proteomics identify novel miR-155 target proteins. PLoS One 6, e22146.
142
Lu, C.Y., Hsieh, S.Y., Lu, Y.J., Wu, C.S., Chen, L.C., Lo, S.J., Wu, C.T., Chou, M.Y., Huang, T.H., and Chang, Y.S. (2009). Aberrant DNA methylation profile and frequent methylation of KLK10 and OXGR1 genes in hepatocellular carcinoma. Genes Chromosomes Cancer 48, 1057-1068.
Lu, L., Schwartz, P., Scarampi, L., Rutherford, T., Canuto, E.M., Yu, H., and Katsaros, D. (2011). MicroRNA let-7a: a potential marker for selection of paclitaxel in ovarian cancer management. Gynecol Oncol 122, 366-371.
Luo, L.Y., Soosaipillai, A., Grass, L., and Diamandis, E.P. (2006). Characterization of human kallikreins 6 and 10 in ascites fluid from ovarian cancer patients. Tumour Biol 27, 227-234.
Lynch, H.T., Casey, M.J., Snyder, C.L., Bewtra, C., Lynch, J.F., Butts, M., and Godwin, A.K. (2009). Hereditary ovarian carcinoma: heterogeneity, molecular genetics, pathology, and management. Mol Oncol 3, 97-137.
Lynch, H.T., and de la Chapelle, A. (2003). Hereditary colorectal cancer. N Engl J Med 348, 919-932.
Macfarlane, S.R., Seatter, M.J., Kanke, T., Hunter, G.D., and Plevin, R. (2001). Proteinase-activated receptors. Pharmacol Rev 53, 245-282.
Macville, M., Veldman, T., Padilla-Nash, H., Wangsa, D., O'Brien, P., Schrock, E., and Ried, T. (1997). Spectral karyotyping, a 24-colour FISH technique for the identification of chromosomal rearrangements. Histochem Cell Biol 108, 299-305.
Magklara, A., Mellati, A.A., Wasney, G.A., Little, S.P., Sotiropoulou, G., Becker, G.W., and Diamandis, E.P. (2003). Characterization of the enzymatic activity of human kallikrein 6: Autoactivation, substrate specificity, and regulation by inhibitors. Biochem Biophys Res Commun 307, 948-955.
Mai, S. (2010). Initiation of telomere-mediated chromosomal rearrangements in cancer. J Cell Biochem 109, 1095-1102.
Maire, G., Yoshimoto, M., Chilton-MacNeill, S., Thorner, P.S., Zielenska, M., and Squire, J.A. (2009). Recurrent RECQL4 imbalance and increased gene expression levels are associated with structural chromosomal instability in sporadic osteosarcoma. Neoplasia 11, 260-268, 263p following 268.
Makar, A.P., Kristensen, G.B., Bormer, O.P., and Trope, C.G. (1992). CA 125 measured before second-look laparotomy is an independent prognostic factor for survival in patients with epithelial ovarian cancer. Gynecol Oncol 45, 323-328.
Mavridis, K., and Scorilas, A. (2010). Prognostic value and biological role of the kallikrein-related peptidases in human malignancies. Future Oncol 6, 269-285.
McManus, D.T., Murphy, M., Arthur, K., Hamilton, P.W., Russell, S.E., and Toner, P.G. (1996). p53 mutation, allele loss on chromosome 17p, and DNA content in ovarian carcinoma. J Pathol 179, 177-182.
Menendez-Gonzalez, M., Castro-Santos, P., Suarez, A., Calatayud, M.T., Perez-Pinera, P., Martinez, M., Ribacoba, R., and Gutierrez, C. (2008). Value of measuring plasmatic levels of neurosin in the diagnosis of Alzheimer's disease. J Alzheimers Dis 14, 59-67.
143
Mestdagh, P., Bostrom, A.K., Impens, F., Fredlund, E., Van Peer, G., De Antonellis, P., von Stedingk, K., Ghesquiere, B., Schulte, S., Dews, M., et al. (2010). The miR-17-92 microRNA cluster regulates multiple components of the TGF-beta pathway in neuroblastoma. Mol Cell 40, 762-773.
Micci, F., Weimer, J., Haugom, L., Skotheim, R.I., Grunewald, R., Abeler, V.M., Silins, I., Lothe, R.A., Trope, C.G., Arnold, N., et al. (2009). Reverse painting of microdissected chromosome 19 markers in ovarian carcinoma identifies a complex rearrangement map. Genes Chromosomes Cancer 48, 184-193.
Michael, I.P., Pampalakis, G., Mikolajczyk, S.D., Malm, J., Sotiropoulou, G., and Diamandis, E.P. (2006). Human tissue kallikrein 5 is a member of a proteolytic cascade pathway involved in seminal clot liquefaction and potentially in prostate cancer progression. J Biol Chem 281, 12743-12750.
Mikaelsdottir, E.K., Valgeirsdottir, S., Eyfjord, J.E., and Rafnar, T. (2004). The Icelandic founder mutation BRCA2 999del5: analysis of expression. Breast Cancer Res 6, R284-290.
Miller, S.M., Roussi, P., Daly, M.B., and Scarpato, J. (2010). New strategies in ovarian cancer: uptake and experience of women at high risk of ovarian cancer who are considering risk-reducing salpingo-oophorectomy. Clin Cancer Res 16, 5094-5106.
Mitelman, F., Johansson, B., and Mertens, F. (2007). The impact of translocations and gene fusions on cancer causation. Nat Rev Cancer 7, 233-245.
Mitsui, S., Okui, A., Uemura, H., Mizuno, T., Yamada, T., Yamamura, Y., and Yamaguchi, N. (2002). Decreased cerebrospinal fluid levels of neurosin (KLK6), an aging-related protease, as a possible new risk factor for Alzheimer's disease. Ann N Y Acad Sci 977, 216-223.
Mo, L., Zhang, J., Shi, J., Xuan, Q., Yang, X., Qin, M., Lee, C., Klocker, H., Li, Q.Q., and Mo, Z. (2010). Human kallikrein 7 induces epithelial-mesenchymal transition-like changes in prostate carcinoma cells: a role in prostate cancer invasion and progression. Anticancer Res 30, 3413-3420.
Mogensen, O. (1992). Prognostic value of CA 125 in advanced ovarian cancer. Gynecol Oncol 44, 207-212.
Montagnana, M., Danese, E., Giudici, S., Franchi, M., Guidi, G.C., Plebani, M., and Lippi, G. (2011). HE4 in ovarian cancer: from discovery to clinical application. Adv Clin Chem 55, 1-20.
Moore, R.G., McMeekin, D.S., Brown, A.K., DiSilvestro, P., Miller, M.C., Allard, W.J., Gajewski, W., Kurman, R., Bast, R.C., Jr., and Skates, S.J. (2009). A novel multiple marker bioassay utilizing HE4 and CA125 for the prediction of ovarian cancer in patients with a pelvic mass. Gynecol Oncol 112, 40-46.
Muleris, M., Chalastanis, A., Meyer, N., Lae, M., Dutrillaux, B., Sastre-Garau, X., Hamelin, R., Flejou, J.F., and Duval, A. (2008). Chromosomal instability in near-diploid colorectal cancer: a link between numbers and structure. PLoS ONE 3, e1632.
Murnane, J.P. (2006). Telomeres and chromosome instability. DNA Repair (Amst) 5, 1082-1092.
144
Nadiminty, N., Tummala, R., Lou, W., Zhu, Y., Zhang, J., Chen, X., Devere White, R.W., Kung, H.J., Evans, C.P., and Gao, A.C. (2011). MicroRNA let-7c Suppresses Androgen Receptor Expression and Activity via Regulation of Myc Expression in Prostate Cancer Cells. J Biol Chem 287, 1527-1537.
Nagahara, H., Mimori, K., Utsunomiya, T., Barnard, G.F., Ohira, M., Hirakawa, K., and Mori, M. (2005). Clinicopathologic and biological significance of kallikrein 6 overexpression in human gastric cancer. Clin Cancer Res 11, 6800-6806.
Nagaprashantha, L.D., Vatsyayan, R., Lelsani, P.C., Awasthi, S., and Singhal, S.S. (2011). The sensors and regulators of cell-matrix surveillance in anoikis resistance of tumors. Int J Cancer 128, 743-752.
Nakayama, K., Nakayama, N., Jinawath, N., Salani, R., Kurman, R.J., Shih Ie, M., and Wang, T.L. (2007). Amplicon profiles in ovarian serous carcinomas. Int J Cancer 120, 2613-2617.
Nakayama, K., Nakayama, N., Kurman, R.J., Cope, L., Pohl, G., Samuels, Y., Velculescu, V.E., Wang, T.L., and Shih Ie, M. (2006). Sequence mutations and amplification of PIK3CA and AKT2 genes in purified ovarian serous neoplasms. Cancer Biol Ther 5, 779-785.
Naldi, L. (2010). Scoring and monitoring the severity of psoriasis. What is the preferred method? What is the ideal method? Is PASI passe? facts and controversies. Clin Dermatol 28, 67-72.
Nam, E.J., Yoon, H., Kim, S.W., Kim, H., Kim, Y.T., Kim, J.H., Kim, J.W., and Kim, S. (2008). MicroRNA expression profiles in serous ovarian carcinoma. Clin Cancer Res 14, 2690-2695.
Nathalie, H.V., Chris, P., Serge, G., Catherine, C., Benjamin, B., Claire, B., Christelle, P., Briollais, L., Pascale, R., Marie-Lise, J., et al. (2009). High kallikrein-related peptidase 6 in non-small cell lung cancer cells: an indicator of tumour proliferation and poor prognosis. J Cell Mol Med 13, 4014-4022.
Neve, R.M., Chin, K., Fridlyand, J., Yeh, J., Baehner, F.L., Fevr, T., Clark, L., Bayani, N., Coppe, J.P., Tong, F., et al. (2006). A collection of breast cancer cell lines for the study of functionally distinct cancer subtypes. Cancer Cell 10, 515-527.
Ni, X., Zhang, W., Huang, K.C., Wang, Y., Ng, S.K., Mok, S.C., Berkowitz, R.S., and Ng, S.W. (2004). Characterisation of human kallikrein 6/protease M expression in ovarian cancer. Br J Cancer 91, 725-731.
Nigg, E.A. (2002). Centrosome aberrations: cause or consequence of cancer progression? Nat Rev Cancer 2, 815-825.
Obiezu, C.V., Scorilas, A., Katsaros, D., Massobrio, M., Yousef, G.M., Fracchioli, S., Rigault de la Longrais, I.A., Arisio, R., and Diamandis, E.P. (2001). Higher human kallikrein gene 4 (KLK4) expression indicates poor prognosis of ovarian cancer patients. Clin Cancer Res 7, 2380-2386.
Ogawa, K., Utsunomiya, T., Mimori, K., Tanaka, F., Inoue, H., Nagahara, H., Murayama, S., and Mori, M. (2005). Clinical significance of human kallikrein gene 6 messenger RNA expression in colorectal cancer. Clin Cancer Res 11, 2889-2893.
145
Ogawa, K., Yamada, T., Tsujioka, Y., Taguchi, J., Takahashi, M., Tsuboi, Y., Fujino, Y., Nakajima, M., Yamamoto, T., Akatsu, H., et al. (2000). Localization of a novel type trypsin-like serine protease, neurosin, in brain tissues of Alzheimer's disease and Parkinson's disease. Psychiatry Clin Neurosci 54, 419-426.
Oikonomopoulou, K., Diamandis, E.P., and Hollenberg, M.D. (2010). Kallikrein-related peptidases: proteolysis and signaling in cancer, the new frontier. Biol Chem 391, 299-310.
Oikonomopoulou, K., Hansen, K.K., Saifeddine, M., Tea, I., Blaber, M., Blaber, S.I., Scarisbrick, I., Andrade-Gordon, P., Cottrell, G.S., Bunnett, N.W., et al. (2006a). Proteinase-activated receptors, targets for kallikrein signaling. J Biol Chem 281, 32095-32112.
Oikonomopoulou, K., Hansen, K.K., Saifeddine, M., Vergnolle, N., Tea, I., Blaber, M., Blaber, S.I., Scarisbrick, I., Diamandis, E.P., and Hollenberg, M.D. (2006b). Kallikrein-mediated cell signalling: targeting proteinase-activated receptors (PARs). Biol Chem 387, 817-824.
Oikonomopoulou, K., Hansen, K.K., Saifeddine, M., Vergnolle, N., Tea, I., Diamandis, E.P., and Hollenberg, M.D. (2006c). Proteinase-mediated cell signalling: targeting proteinase-activated receptors (PARs) by kallikreins and more. Biol Chem 387, 677-685.
Ong, S.E., Blagoev, B., Kratchmarova, I., Kristensen, D.B., Steen, H., Pandey, A., and Mann, M. (2002). Stable isotope labeling by amino acids in cell culture, SILAC, as a simple and accurate approach to expression proteomics. Mol Cell Proteomics 1, 376-386.
Orr, G.A., Verdier-Pinard, P., McDaid, H., and Horwitz, S.B. (2003). Mechanisms of Taxol resistance related to microtubules. Oncogene 22, 7280-7295.
Pal, T., Permuth-Wey, J., Kapoor, R., Cantor, A., and Sutphen, R. (2007). Improved survival in BRCA2 carriers with ovarian cancer. Fam Cancer 6, 113-119.
Paliouras, M., Borgono, C., and Diamandis, E.P. (2007). Human tissue kallikreins: the cancer biomarker family. Cancer Lett 249, 61-79.
Paliouras, M., and Diamandis, E.P. (2007). Coordinated steroid hormone-dependent and independent expression of multiple kallikreins in breast cancer cell lines. Breast Cancer Res Treat 102, 7-18.
Palmer, H.G., Sanchez-Carbayo, M., Ordonez-Moran, P., Larriba, M.J., Cordon-Cardo, C., and Munoz, A. (2003). Genetic signatures of differentiation induced by 1alpha,25-dihydroxyvitamin D3 in human colon cancer cells. Cancer Res 63, 7799-7806.
Pampalakis, G., Arampatzidou, M., Amoutzias, G., Kossida, S., and Sotiropoulou, G. (2008). Identification and analysis of mammalian KLK6 orthologue genes for prediction of physiological substrates. Comput Biol Chem 32, 111-121.
Pampalakis, G., Diamandis, E.P., and Sotiropoulou, G. (2006). The epigenetic basis for the aberrant expression of kallikreins in human cancers. Biol Chem 387, 795-799.
Pampalakis, G., Kurlender, L., Diamandis, E.P., and Sotiropoulou, G. (2004). Cloning and characterization of novel isoforms of the human kallikrein 6 gene. Biochem Biophys Res Commun 320, 54-61.
146
Pampalakis, G., Prosnikli, E., Agalioti, T., Vlahou, A., Zoumpourlis, V., and Sotiropoulou, G. (2009). A tumor-protective role for human kallikrein-related peptidase 6 in breast cancer mediated by inhibition of epithelial-to-mesenchymal transition. Cancer Res 69, 3779-3787.
Pampalakis, G., and Sotiropoulou, G. (2006). Multiple mechanisms underlie the aberrant expression of the human kallikrein 6 gene in breast cancer. Biol Chem 387, 773-782.
Pandis, N., Jin, Y., Gorunova, L., Petersson, C., Bardi, G., Idvall, I., Johansson, B., Ingvar, C., Mandahl, N., Mitelman, F., et al. (1995). Chromosome analysis of 97 primary breast carcinomas: identification of eight karyotypic subgroups. Genes Chromosomes Cancer 12, 173-185.
Papageorgiou, T., and Stratakis, C.A. (2002). Ovarian tumors associated with multiple endocrine neoplasias and related syndromes (Carney complex, Peutz-Jeghers syndrome, von Hippel-Lindau disease, Cowden's disease). Int J Gynecol Cancer 12, 337-347.
Pavlopoulou, A., Pampalakis, G., Michalopoulos, I., and Sotiropoulou, G. (2010). Evolutionary History of Tissue Kallikreins. Plos One 5.
Pejovic, T., Heim, S., Mandahl, N., Baldetorp, B., Elmfors, B., Floderus, U.M., Furgyik, S., Helm, G., Himmelmann, A., Willen, H., et al. (1992a). Chromosome aberrations in 35 primary ovarian carcinomas. Genes Chromosomes Cancer 4, 58-68.
Pejovic, T., Heim, S., Orndal, C., Jin, Y.S., Mandahl, N., Willen, H., and Mitelman, F. (1990). Simple numerical chromosome aberrations in well-differentiated malignant epithelial tumors. Cancer Genet Cytogenet 49, 95-101.
Pejovic, T., Himmelmann, A., Heim, S., Mandahl, N., Floderus, U.M., Furgyik, S., Elmfors, B., Helm, G., Willen, H., and Mitelman, F. (1992b). Prognostic impact of chromosome aberrations in ovarian cancer. Br J Cancer 65, 282-286.
Petraki, C.D., Karavana, V.N., and Diamandis, E.P. (2003). Human kallikrein 13 expression in normal tissues: an immunohistochemical study. J Histochem Cytochem 51, 493-501.
Petraki, C.D., Karavana, V.N., Skoufogiannis, P.T., Little, S.P., Howarth, D.J., Yousef, G.M., and Diamandis, E.P. (2001). The spectrum of human kallikrein 6 (zyme/protease M/neurosin) expression in human tissues as assessed by immunohistochemistry. J Histochem Cytochem 49, 1431-1441.
Petraki, C.D., Papanastasiou, P.A., Karavana, V.N., and Diamandis, E.P. (2006). Cellular distribution of human tissue kallikreins: immunohistochemical localization. Biol Chem 387, 653-663.
Pettus, J.R., Johnson, J.J., Shi, Z., Davis, J.W., Koblinski, J., Ghosh, S., Liu, Y., Ravosa, M.J., Frazier, S., and Stack, M.S. (2009). Multiple kallikrein (KLK 5, 7, 8, and 10) expression in squamous cell carcinoma of the oral cavity. Histol Histopathol 24, 197-207.
Pfisterer, J., and Ledermann, J.A. (2006). Management of platinum-sensitive recurrent ovarian cancer. Semin Oncol 33, S12-16.
Pfisterer, J., Plante, M., Vergote, I., du Bois, A., Hirte, H., Lacave, A.J., Wagner, U., Stahle, A., Stuart, G., Kimmig, R., et al. (2006). Gemcitabine plus carboplatin compared with carboplatin in patients with platinum-sensitive recurrent ovarian cancer: an intergroup
147
trial of the AGO-OVAR, the NCIC CTG, and the EORTC GCG. J Clin Oncol 24, 4699-4707.
Pflueger, D., Terry, S., Sboner, A., Habegger, L., Esgueva, R., Lin, P.C., Svensson, M.A., Kitabayashi, N., Moss, B.J., MacDonald, T.Y., et al. (2011). Discovery of non-ETS gene fusions in human prostate cancer using next-generation RNA sequencing. Genome Res 21, 56-67.
Pihan, G.A., and Doxsey, S.J. (1999). The mitotic machinery as a source of genetic instability in cancer. Semin Cancer Biol 9, 289-302.
Pihan, G.A., Purohit, A., Wallace, J., Knecht, H., Woda, B., Quesenberry, P., and Doxsey, S.J. (1998). Centrosome defects and genetic instability in malignant tumors. Cancer Res 58, 3974-3985.
Planque, C., de Monte, M., Guyetant, S., Rollin, J., Desmazes, C., Panel, V., Lemarie, E., and Courty, Y. (2005). KLK5 and KLK7, two members of the human tissue kallikrein family, are differentially expressed in lung cancer. Biochem Biophys Res Commun 329, 1260-1266.
Prasad, M., Bernardini, M., Tsalenko, A., Marrano, P., Paderova, J., Lee, C.H., Ben-Dor, A., Barrett, M.T., and Squire, J.A. (2008). High definition cytogenetics and oligonucleotide aCGH analyses of cisplatin-resistant ovarian cancer cells. Genes Chromosomes Cancer.
Prat, J., Ribe, A., and Gallardo, A. (2005). Hereditary ovarian cancer. Hum Pathol 36, 861-870.
Prezas, P., Arlt, M.J., Viktorov, P., Soosaipillai, A., Holzscheiter, L., Schmitt, M., Talieri, M., Diamandis, E.P., Kruger, A., and Magdolen, V. (2006). Overexpression of the human tissue kallikrein genes KLK4, 5, 6, and 7 increases the malignant phenotype of ovarian cancer cells. Biol Chem 387, 807-811.
Puente, X.S., and Lopez-Otin, C. (2004). A genomic analysis of rat proteases and protease inhibitors. Genome Res 14, 609-622.
Quintyne, N.J., Reing, J.E., Hoffelder, D.R., Gollin, S.M., and Saunders, W.S. (2005). Spindle multipolarity is prevented by centrosomal clustering. Science 307, 127-129.
Rafnar, T., Benediktsdottir, K.R., Eldon, B.J., Gestsson, T., Saemundsson, H., Olafsson, K., Salvarsdottir, A., Steingrimsson, E., and Thorlacius, S. (2004). BRCA2, but not BRCA1, mutations account for familial ovarian cancer in Iceland: a population-based study. Eur J Cancer 40, 2788-2793.
Rajagopalan, H., Nowak, M.A., Vogelstein, B., and Lengauer, C. (2003). The significance of unstable chromosomes in colorectal cancer. Nat Rev Cancer 3, 695-701.
Rao, P.H., Harris, C.P., Yan Lu, X., Li, X.N., Mok, S.C., and Lau, C.C. (2002). Multicolor spectral karyotyping of serous ovarian adenocarcinoma. Genes Chromosomes Cancer 33, 123-132.
Rasheed, Z.A., and Rubin, E.H. (2003). Mechanisms of resistance to topoisomerase I-targeting drugs. Oncogene 22, 7296-7304.
Rebacz, B., Larsen, T.O., Clausen, M.H., Ronnest, M.H., Loffler, H., Ho, A.D., and Kramer, A. (2007). Identification of griseofulvin as an inhibitor of centrosomal clustering in a phenotype-based screen. Cancer Res 67, 6342-6350.
148
Resnik, E., Trujillo, Y.P., and Taxy, J.B. (1997). Long-term survival and DNA ploidy in advanced epithelial ovarian cancer. J Surg Oncol 64, 299-303.
Ricciardelli, C., and Oehler, M.K. (2009). Diverse molecular pathways in ovarian cancer and their clinical significance. Maturitas 62, 270-275.
Righolt, C., and Mai, S. (2012). Shattered and stitched chromosomes-chromothripsis and chromoanasynthesis-manifestations of a new chromosome crisis? Genes Chromosomes Cancer 51, 975-981.
Risch, H.A., McLaughlin, J.R., Cole, D.E., Rosen, B., Bradley, L., Fan, I., Tang, J., Li, S., Zhang, S., Shaw, P.A., et al. (2006). Population BRCA1 and BRCA2 mutation frequencies and cancer penetrances: a kin-cohort study in Ontario, Canada. J Natl Cancer Inst 98, 1694-1706.
Roach, J.C., Glusman, G., Smit, A.F., Huff, C.D., Hubley, R., Shannon, P.T., Rowen, L., Pant, K.P., Goodman, N., Bamshad, M., et al. (2010). Analysis of genetic inheritance in a family quartet by whole-genome sequencing. Science 328, 636-639.
Rosen, D.G., Wang, L., Atkinson, J.N., Yu, Y., Lu, K.H., Diamandis, E.P., Hellstrom, I., Mok, S.C., Liu, J., and Bast, R.C., Jr. (2005). Potential markers that complement expression of CA125 in epithelial ovarian cancer. Gynecol Oncol 99, 267-277.
Ruckert, F., Hennig, M., Petraki, C.D., Wehrum, D., Distler, M., Denz, A., Schroder, M., Dawelbait, G., Kalthoff, H., Saeger, H.D., et al. (2008). Co-expression of KLK6 and KLK10 as prognostic factors for survival in pancreatic ductal adenocarcinoma. Br J Cancer 99, 1484-1492.
Rummukainen, J., Kytola, S., Karhu, R., Farnebo, F., Larsson, C., and Isola, J.J. (2001). Aberrations of chromosome 8 in 16 breast cancer cell lines by comparative genomic hybridization, fluorescence in situ hybridization, and spectral karyotyping. Cancer Genet Cytogenet 126, 1-7.
Sadikovic, B., Al-Romaih, K., Squire, J.A., and Zielenska, M. (2008). Cause and consequences of genetic and epigenetic alterations in human cancer. Curr Genomics 9, 394-408.
Sadoni, N., Langer, S., Fauth, C., Bernardi, G., Cremer, T., Turner, B.M., and Zink, D. (1999). Nuclear organization of mammalian genomes. Polar chromosome territories build up functionally distinct higher order compartments. J Cell Biol 146, 1211-1226.
Salvador, S., Gilks, B., Kobel, M., Huntsman, D., Rosen, B., and Miller, D. (2009). The fallopian tube: primary site of most pelvic high-grade serous carcinomas. Int J Gynecol Cancer 19, 58-64.
Santin, A.D., Diamandis, E.P., Bellone, S., Soosaipillai, A., Cane, S., Palmieri, M., Burnett, A., Roman, J.J., and Pecorelli, S. (2005). Human kallikrein 6: a new potential serum biomarker for uterine serous papillary cancer. Clin Cancer Res 11, 3320-3325.
Saunders, W. (2005). Centrosomal amplification and spindle multipolarity in cancer cells. Semin Cancer Biol 15, 25-32.
Scarisbrick, I.A. (2008). The multiple sclerosis degradome: enzymatic cascades in development and progression of central nervous system inflammatory disease. Curr Top Microbiol Immunol 318, 133-175.
149
Scarisbrick, I.A., Blaber, S.I., Lucchinetti, C.F., Genain, C.P., Blaber, M., and Rodriguez, M. (2002). Activity of a newly identified serine protease in CNS demyelination. Brain 125, 1283-1296.
Scarisbrick, I.A., Blaber, S.I., Tingling, J.T., Rodriguez, M., Blaber, M., and Christophi, G.P. (2006). Potential scope of action of tissue kallikreins in CNS immune-mediated disease. J Neuroimmunol 178, 167-176.
Scarisbrick, I.A., Epstein, B., Cloud, B.A., Yoon, H., Wu, J., Renner, D.N., Blaber, S.I., Blaber, M., Vandell, A.G., and Bryson, A.L. (2011). Functional role of kallikrein 6 in regulating immune cell survival. PLoS One 6, e18376.
Scarisbrick, I.A., Isackson, P.J., Ciric, B., Windebank, A.J., and Rodriguez, M. (2001). MSP, a trypsin-like serine protease, is abundantly expressed in the human nervous system. J Comp Neurol 431, 347-361.
Scarisbrick, I.A., Linbo, R., Vandell, A.G., Keegan, M., Blaber, S.I., Blaber, M., Sneve, D., Lucchinetti, C.F., Rodriguez, M., and Diamandis, E.P. (2008). Kallikreins are associated with secondary progressive multiple sclerosis and promote neurodegeneration. Biol Chem 389, 739-745.
Schickel, R., Boyerinas, B., Park, S.M., and Peter, M.E. (2008). MicroRNAs: key players in the immune system, differentiation, tumorigenesis and cell death. Oncogene 27, 5959-5974.
Schmidt, A.E., Ogawa, T., Gailani, D., and Bajaj, S.P. (2004). Structural role of Gly(193) in serine proteases: investigations of a G555E (GLY193 in chymotrypsin) mutant of blood coagulation factor XI. J Biol Chem 279, 29485-29492.
Schmittgen, T.D., Jiang, J., Liu, Q., and Yang, L. (2004). A high-throughput method to monitor the expression of microRNA precursors. Nucleic Acids Res 32, e43.
Schrock, E., Veldman, T., Padilla-Nash, H., Ning, Y., Spurbeck, J., Jalal, S., Shaffer, L.G., Papenhausen, P., Kozma, C., Phelan, M.C., et al. (1997). Spectral karyotyping refines cytogenetic diagnostics of constitutional chromosomal abnormalities. Hum Genet 101, 255-262.
Sehic, A., Risnes, S., Khan, Q.E., Khuu, C., and Osmundsen, H. (2010). Gene expression and dental enamel structure in developing mouse incisor. Eur J Oral Sci 118, 118-130.
Seiz, L., Dorn, J., Kotzsch, M., Walch, A., Grebenchtchikov, N.I., Gkazepis, A., Schmalfeldt, B., Kiechle, M., Bayani, J., Diamandis, E.P., et al. (2012). Stromal cell-associated expression of kallikrein-related peptidase 6 (KLK6) indicates poor prognosis of ovarian cancer patients. Biol Chem 393, 391-401.
Selbach, M., Schwanhausser, B., Thierfelder, N., Fang, Z., Khanin, R., and Rajewsky, N. (2008). Widespread changes in protein synthesis induced by microRNAs. Nature 455, 58-63.
Selvarajah, S., Yoshimoto, M., Maire, G., Paderova, J., Bayani, J., Squire, J.A., and Zielenska, M. (2007). Identification of cryptic microaberrations in osteosarcoma by high-definition oligonucleotide array comparative genomic hybridization. Cancer Genet Cytogenet 179, 52-61.
Selvarajah, S., Yoshimoto, M., Park, P.C., Maire, G., Paderova, J., Bayani, J., Lim, G., Al-Romaih, K., Squire, J.A., and Zielenska, M. (2006). The breakage-fusion-bridge (BFB)
150
cycle as a mechanism for generating genetic heterogeneity in osteosarcoma. Chromosoma 115, 459-467.
Shaffer, L.G., and Tommerup, N. (2005). An International System for Human Cytogenetic Nomenclature (2005) (Basel , Switzerland, Karger Publishers).
Shan, S.J., Scorilas, A., Katsaros, D., and Diamandis, E.P. (2007). Transcriptional upregulation of human tissue kallikrein 6 in ovarian cancer: clinical and mechanistic aspects. Br J Cancer 96, 362-372.
Shaw, J.L., and Diamandis, E.P. (2007). Distribution of 15 human kallikreins in tissues and biological fluids. Clin Chem 53, 1423-1432.
Shell, S., Park, S.M., Radjabi, A.R., Schickel, R., Kistner, E.O., Jewell, D.A., Feig, C., Lengyel, E., and Peter, M.E. (2007). Let-7 expression defines two differentiation stages of cancer. Proc Natl Acad Sci U S A 104, 11400-11405.
Shih Ie, M., and Kurman, R.J. (2004). Ovarian tumorigenesis: a proposed model based on morphological and molecular genetic analysis. Am J Pathol 164, 1511-1518.
Shinoda, Y., Kozaki, K., Imoto, I., Obara, W., Tsuda, H., Mizutani, Y., Shuin, T., Fujioka, T., Miki, T., and Inazawa, J. (2007). Association of KLK5 overexpression with invasiveness of urinary bladder carcinoma cells. Cancer Sci 98, 1078-1086.
Shridhar, V., Lee, J., Pandita, A., Iturria, S., Avula, R., Staub, J., Morrissey, M., Calhoun, E., Sen, A., Kalli, K., et al. (2001). Genetic analysis of early- versus late-stage ovarian tumors. Cancer Res 61, 5895-5904.
Shulman, J.M., De Jager, P.L., and Feany, M.B. (2011). Parkinson's disease: genetics and pathogenesis. Annu Rev Pathol 6, 193-222.
Siddik, Z.H. (2003). Cisplatin: mode of cytotoxic action and molecular basis of resistance. Oncogene 22, 7265-7279.
Sidiropoulos, M., Pampalakis, G., Sotiropoulou, G., Katsaros, D., and Diamandis, E.P. (2005). Downregulation of human kallikrein 10 (KLK10/NES1) by CpG island hypermethylation in breast, ovarian and prostate cancers. Tumour Biol 26, 324-336.
Singh, J., Naran, A., Misso, N.L., Rigby, P.J., Thompson, P.J., and Bhoola, K.D. (2008). Expression of kallikrein-related peptidases (KRP/hK5, 7, 6, 8) in subtypes of human lung carcinoma. Int Immunopharmacol 8, 300-306.
Skoufias, D.A., Andreassen, P.R., Lacroix, F.B., Wilson, L., and Margolis, R.L. (2001). Mammalian mad2 and bub1/bubR1 recognize distinct spindle-attachment and kinetochore-tension checkpoints. Proc Natl Acad Sci U S A 98, 4492-4497.
Sogaard, M., Kjaer, S.K., and Gayther, S. (2006). Ovarian cancer and genetic susceptibility in relation to the BRCA1 and BRCA2 genes. Occurrence, clinical importance and intervention. Acta Obstet Gynecol Scand 85, 93-105.
Sotiropoulou, G., Pampalakis, G., and Diamandis, E.P. (2009). Functional roles of human kallikrein-related peptidases. J Biol Chem 284, 32989-32994.
Stadelmann, C., Wegner, C., and Bruck, W. (2011). Inflammation, demyelination, and degeneration - recent insights from MS pathology. Biochim Biophys Acta 1812, 275-282.
151
Stephens, P.J., Greenman, C.D., Fu, B., Yang, F., Bignell, G.R., Mudie, L.J., Pleasance, E.D., Lau, K.W., Beare, D., Stebbings, L.A., et al. (2011). Massive genomic rearrangement acquired in a single catastrophic event during cancer development. Cell 144, 27-40.
Stewart, Z.A., Leach, S.D., and Pietenpol, J.A. (1999). p21(Waf1/Cip1) inhibition of cyclin E/Cdk2 activity prevents endoreduplication after mitotic spindle disruption. Mol Cell Biol 19, 205-215.
Stratton, M.R., Campbell, P.J., and Futreal, P.A. (2009). The cancer genome. Nature 458, 719-724.
Strojnik, T., Kavalar, R., Zajc, I., Diamandis, E.P., Oikonomopoulou, K., and Lah, T.T. (2009). Prognostic impact of CD68 and kallikrein 6 in human glioma. Anticancer Res 29, 3269-3279.
Swanton, C., Nicke, B., Schuett, M., Eklund, A.C., Ng, C., Li, Q., Hardcastle, T., Lee, A., Roy, R., East, P., et al. (2009). Chromosomal instability determines taxane response. Proc Natl Acad Sci U S A 106, 8671-8676.
Taetle, R., Aickin, M., Panda, L., Emerson, J., Roe, D., Thompson, F., Davis, J., Trent, J., and Alberts, D. (1999a). Chromosome abnormalities in ovarian adenocarcinoma: II. Prognostic impact of nonrandom chromosome abnormalities in 244 cases. Genes Chromosomes Cancer 25, 46-52.
Taetle, R., Aickin, M., Yang, J.M., Panda, L., Emerson, J., Roe, D., Adair, L., Thompson, F., Liu, Y., Wisner, L., et al. (1999b). Chromosome abnormalities in ovarian adenocarcinoma: I. Nonrandom chromosome abnormalities from 244 cases. Genes Chromosomes Cancer 25, 290-300.
Tan, D.S., Lambros, M.B., Natrajan, R., and Reis-Filho, J.S. (2007). Getting it right: designing microarray (and not 'microawry') comparative genomic hybridization studies for cancer research. Lab Invest 87, 737-754.
Tan, O.L., Whitbread, A.K., Clements, J.A., and Dong, Y. (2006). Kallikrein-related peptidase (KLK) family mRNA variants and protein isoforms in hormone-related cancers: do they have a function? Biol Chem 387, 697-705.
Tanimoto, H., Underwood, L.J., Shigemasa, K., Parmley, T.H., and O'Brien, T.J. (2001). Increased expression of protease M in ovarian tumors. Tumour Biol 22, 11-18.
Tatebe, H., Watanabe, Y., Kasai, T., Mizuno, T., Nakagawa, M., Tanaka, M., and Tokuda, T. (2010). Extracellular neurosin degrades alpha-synuclein in cultured cells. Neurosci Res 67, 341-346.
Thompson, F.H., Emerson, J., Alberts, D., Liu, Y., Guan, X.Y., Burgess, A., Fox, S., Taetle, R., Weinstein, R., Makar, R., et al. (1994a). Clonal chromosome abnormalities in 54 cases of ovarian carcinoma. Cancer Genet Cytogenet 73, 33-45.
Thompson, F.H., Liu, Y., Emerson, J., Weinstein, R., Makar, R., Trent, J.M., Taetle, R., and Alberts, D.S. (1994b). Simple numeric abnormalities as primary karyotype changes in ovarian carcinoma. Genes Chromosomes Cancer 10, 262-266.
Thompson, F.H., Nelson, M.A., Trent, J.M., Guan, X.Y., Liu, Y., Yang, J.M., Emerson, J., Adair, L., Wymer, J., Balfour, C., et al. (1996). Amplification of 19q13.1-q13.2
152
sequences in ovarian cancer. G-band, FISH, and molecular studies. Cancer Genet Cytogenet 87, 55-62.
Tobias, D.H., Eng, C., McCurdy, L.D., Kalir, T., Mandelli, J., Dottino, P.R., and Cohen, C.J. (2000). Founder BRCA 1 and 2 mutations among a consecutive series of Ashkenazi Jewish ovarian cancer patients. Gynecol Oncol 78, 148-151.
Tokumaru, S., Suzuki, M., Yamada, H., Nagino, M., and Takahashi, T. (2008). let-7 regulates Dicer expression and constitutes a negative feedback loop. Carcinogenesis 29, 2073-2077.
Tomasini, R., Mak, T.W., and Melino, G. (2008). The impact of p53 and p73 on aneuploidy and cancer. Trends Cell Biol 18, 244-252.
Tomlins, S.A., Mehra, R., Rhodes, D.R., Smith, L.R., Roulston, D., Helgeson, B.E., Cao, X., Wei, J.T., Rubin, M.A., Shah, R.B., et al. (2006). TMPRSS2:ETV4 gene fusions define a third molecular subtype of prostate cancer. Cancer Res 66, 3396-3400.
Tomlins, S.A., Rhodes, D.R., Perner, S., Dhanasekaran, S.M., Mehra, R., Sun, X.W., Varambally, S., Cao, X., Tchinda, J., Kuefer, R., et al. (2005). Recurrent fusion of TMPRSS2 and ETS transcription factor genes in prostate cancer. Science 310, 644-648.
Tothill, R.W., Tinker, A.V., George, J., Brown, R., Fox, S.B., Lade, S., Johnson, D.S., Trivett, M.K., Etemadmoghadam, D., Locandro, B., et al. (2008). Novel molecular subtypes of serous and endometrioid ovarian cancer linked to clinical outcome. Clin Cancer Res 14, 5198-5208.
Trimbos, J.B., Parmar, M., Vergote, I., Guthrie, D., Bolis, G., Colombo, N., Vermorken, J.B., Torri, V., Mangioni, C., Pecorelli, S., et al. (2003). International Collaborative Ovarian Neoplasm trial 1 and Adjuvant ChemoTherapy In Ovarian Neoplasm trial: two parallel randomized phase III trials of adjuvant chemotherapy in patients with early-stage ovarian carcinoma. J Natl Cancer Inst 95, 105-112.
Tsao, S.W., Wong, N., Wang, X., Liu, Y., Wan, T.S., Fung, L.F., Lancaster, W.D., Gregoire, L., and Wong, Y.C. (2001). Nonrandom chromosomal imbalances in human ovarian surface epithelial cells immortalized by HPV16-E6E7 viral oncogenes. Cancer Genet Cytogenet 130, 141-149.
Venter, J.C., Adams, M.D., Myers, E.W., Li, P.W., Mural, R.J., Sutton, G.G., Smith, H.O., Yandell, M., Evans, C.A., Holt, R.A., et al. (2001). The sequence of the human genome. Science 291, 1304-1351.
Vilar, E., and Gruber, S.B. (2010). Microsatellite instability in colorectal cancer-the stable evidence. Nat Rev Clin Oncol 7, 153-162.
Vinther, J., Hedegaard, M.M., Gardner, P.P., Andersen, J.S., and Arctander, P. (2006). Identification of miRNA targets with stable isotope labeling by amino acids in cell culture. Nucleic Acids Res 34, e107.
Visintin, I., Feng, Z., Longton, G., Ward, D.C., Alvero, A.B., Lai, Y., Tenthorey, J., Leiser, A., Flores-Saaib, R., Yu, H., et al. (2008). Diagnostic markers for early detection of ovarian cancer. Clin Cancer Res 14, 1065-1072.
153
Vojtesek, B., Bartek, J., Midgley, C.A., and Lane, D.P. (1992). An immunochemical analysis of the human nuclear phosphoprotein p53. New monoclonal antibodies and epitope mapping using recombinant p53. J Immunol Methods 151, 237-244.
Vukovic, B., Beheshti, B., Park, P., Lim, G., Bayani, J., Zielenska, M., and Squire, J.A. (2007). Correlating breakage-fusion-bridge events with the overall chromosomal instability and in vitro karyotype evolution in prostate cancer. Cytogenet Genome Res 116, 1-11.
Wan, T.S., Martens, U.M., Poon, S.S., Tsao, S.W., Chan, L.C., and Lansdorp, P.M. (1999). Absence or low number of telomere repeats at junctions of dicentric chromosomes. Genes Chromosomes Cancer 24, 83-86.
Wang, Q., Hirohashi, Y., Furuuchi, K., Zhao, H., Liu, Q., Zhang, H., Murali, R., Berezov, A., Du, X., Li, B., et al. (2004). The centrosome in normal and transformed cells. DNA Cell Biol 23, 475-489.
Wang, S.M., Mao, J., Li, B., Wu, W., and Tang, L.L. (2008). [Expression of KLK6 protein and mRNA in primary breast cancer and its clinical significance]. Xi Bao Yu Fen Zi Mian Yi Xue Za Zhi 24, 1087-1089.
Wang, Y., Lu, Y., Toh, S.T., Sung, W.K., Tan, P., Chow, P., Chung, A.Y., Jooi, L.L., and Lee, C.G. (2010). Lethal-7 is down-regulated by the hepatitis B virus x protein and targets signal transducer and activator of transcription 3. J Hepatol 53, 57-66.
Wang, Y., Xue, H., Cutz, J.C., Bayani, J., Mawji, N.R., Chen, W.G., Goetz, L.J., Hayward, S.W., Sadar, M.D., Gilks, C.B., et al. (2005). An orthotopic metastatic prostate cancer model in SCID mice via grafting of a transplantable human prostate tumor line. Lab Invest 85, 1392-1404.
Weaver, Z., Montagna, C., Xu, X., Howard, T., Gadina, M., Brodie, S.G., Deng, C.X., and Ried, T. (2002). Mammary tumors in mice conditionally mutant for Brca1 exhibit gross genomic instability and centrosome amplification yet display a recurring distribution of genomic imbalances that is similar to human breast cancer. Oncogene 21, 5097-5107.
White, N.M., Bui, A., Mejia-Guerrero, S., Chao, J., Soosaipillai, A., Youssef, Y., Mankaruos, M., Honey, R.J., Stewart, R., Pace, K.T., et al. (2010a). Dysregulation of kallikrein-related peptidases in renal cell carcinoma: potential targets of miRNAs. Biol Chem 391, 411-423.
White, N.M., Chow, T.F., Mejia-Guerrero, S., Diamandis, M., Rofael, Y., Faragalla, H., Mankaruous, M., Gabril, M., Girgis, A., and Yousef, G.M. (2010b). Three dysregulated miRNAs control kallikrein 10 expression and cell proliferation in ovarian cancer. Br J Cancer 102, 1244-1253.
White, N.M., Mathews, M., Yousef, G.M., Prizada, A., Fontaine, D., Ghatage, P., Popadiuk, C., Dawson, L., and Dore, J.J. (2009a). Human kallikrein related peptidases 6 and 13 in combination withCA125 is a more sensitive test for ovarian cancer than CA125 alone. Cancer Biomark 5, 279-287.
White, N.M., Mathews, M., Yousef, G.M., Prizada, A., Popadiuk, C., and Dore, J.J. (2009b). KLK6 and KLK13 predict tumor recurrence in epithelial ovarian carcinoma. Br J Cancer 101, 1107-1113.
154
Wissmann, M., Yin, N., Muller, J.M., Greschik, H., Fodor, B.D., Jenuwein, T., Vogler, C., Schneider, R., Gunther, T., Buettner, R., et al. (2007). Cooperative demethylation by JMJD2C and LSD1 promotes androgen receptor-dependent gene expression. Nat Cell Biol 9, 347-353.
Yamamoto, S., Tsuda, H., Honda, K., Onozato, K., Takano, M., Tamai, S., Imoto, I., Inazawa, J., Yamada, T., and Matsubara, O. (2009). Actinin-4 gene amplification in ovarian cancer: a candidate oncogene associated with poor patient prognosis and tumor chemoresistance. Mod Pathol 22, 499-507.
Yamashiro, K., Tsuruoka, N., Kodama, S., Tsujimoto, M., Yamamura, Y., Tanaka, T., Nakazato, H., and Yamaguchi, N. (1997). Molecular cloning of a novel trypsin-like serine protease (neurosin) preferentially expressed in brain. Biochim Biophys Acta 1350, 11-14.
Yan, G.R., Xu, S.H., Tan, Z.L., Liu, L., and He, Q.Y. (2011). Global identification of miR-373-regulated genes in breast cancer by quantitative proteomics. Proteomics 11, 912-920.
Yang, D., Khan, S., Sun, Y., Hess, K., Shmulevich, I., Sood, A.K., and Zhang, W. (2011a). Association of BRCA1 and BRCA2 mutations with survival, chemotherapy sensitivity, and gene mutator phenotype in patients with ovarian cancer. Jama 306, 1557-1565.
Yang, Q., Jie, Z., Cao, H., Greenlee, A.R., Yang, C., Zou, F., and Jiang, Y. (2011b). Low-level expression of let-7a in gastric cancer and its involvement in tumorigenesis by targeting RAB40C. Carcinogenesis 32, 713-722.
Yang, Y., Chaerkady, R., Kandasamy, K., Huang, T.C., Selvan, L.D., Dwivedi, S.B., Kent, O.A., Mendell, J.T., and Pandey, A. (2010). Identifying targets of miR-143 using a SILAC-based proteomic approach. Mol Biosyst 6, 1873-1882.
Yoshimoto, M., Cutz, J.C., Nuin, P.A., Joshua, A.M., Bayani, J., Evans, A.J., Zielenska, M., and Squire, J.A. (2006). Interphase FISH analysis of PTEN in histologic sections shows genomic deletions in 68% of primary prostate cancer and 23% of high-grade prostatic intra-epithelial neoplasias. Cancer Genet Cytogenet 169, 128-137.
Yousef, G.M. (2008). microRNAs: a new frontier in kallikrein research. Biol Chem 389, 689-694.
Yousef, G.M., Borgono, C.A., Michael, I.P., and Diamandis, E.P. (2004a). Cloning of a kallikrein pseudogene. Clin Biochem 37, 961-967.
Yousef, G.M., Borgono, C.A., White, N.M., Robb, J.D., Michael, I.P., Oikonomopoulou, K., Khan, S., and Diamandis, E.P. (2004b). In silico analysis of the human kallikrein gene 6. Tumour Biol 25, 282-289.
Yousef, G.M., Chang, A., Scorilas, A., and Diamandis, E.P. (2000). Genomic organization of the human kallikrein gene family on chromosome 19q13.3-q13.4. Biochem Biophys Res Commun 276, 125-133.
Yousef, G.M., and Diamandis, E.P. (2001). The new human tissue kallikrein gene family: structure, function, and association to disease. Endocr Rev 22, 184-204.
Yousef, G.M., Kishi, T., and Diamandis, E.P. (2003a). Role of kallikrein enzymes in the central nervous system. Clin Chim Acta 329, 1-8.
155
Yousef, G.M., Luo, L.Y., Scherer, S.W., Sotiropoulou, G., and Diamandis, E.P. (1999). Molecular characterization of zyme/protease M/neurosin (PRSS9), a hormonally regulated kallikrein-like serine protease. Genomics 62, 251-259.
Yousef, G.M., Polymeris, M.E., Yacoub, G.M., Scorilas, A., Soosaipillai, A., Popalis, C., Fracchioli, S., Katsaros, D., and Diamandis, E.P. (2003b). Parallel overexpression of seven kallikrein genes in ovarian cancer. Cancer Res 63, 2223-2227.
Zarghami, N., Grass, L., and Diamandis, E.P. (1997). Steroid hormone regulation of prostate-specific antigen gene expression in breast cancer. Br J Cancer 75, 579-588.
Zarghooni, M., Soosaipillai, A., Grass, L., Scorilas, A., Mirazimi, N., and Diamandis, E.P. (2002). Decreased concentration of human kallikrein 6 in brain extracts of Alzheimer's disease patients. Clin Biochem 35, 225-231.
Zhang, L., Huang, J., Yang, N., Greshock, J., Liang, S., Hasegawa, K., Giannakakis, A., Poulos, N., O'Brien-Jenkins, A., Katsaros, D., et al. (2007). Integrative genomic analysis of phosphatidylinositol 3'-kinase family identifies PIK3R3 as a potential therapeutic target in epithelial ovarian cancer. Clin Cancer Res 13, 5314-5321.
Zhang, L., Huang, J., Yang, N., Greshock, J., Megraw, M.S., Giannakakis, A., Liang, S., Naylor, T.L., Barchetti, A., Ward, M.R., et al. (2006). microRNAs exhibit high frequency genomic alterations in human cancer. Proc Natl Acad Sci U S A 103, 9136-9141.
Zhang, L., Volinia, S., Bonome, T., Calin, G.A., Greshock, J., Yang, N., Liu, C.G., Giannakakis, A., Alexiou, P., Hasegawa, K., et al. (2008). Genomic and epigenetic alterations deregulate microRNA expression in human epithelial ovarian cancer. Proc Natl Acad Sci U S A 105, 7004-7009.
Zhang, Y.W., Thompson, R., Zhang, H., and Xu, H. (2011). APP processing in Alzheimer's disease. Mol Brain 4, 3.
Zheng, Y., Katsaros, D., Shan, S.J., de la Longrais, I.R., Porpiglia, M., Scorilas, A., Kim, N.W., Wolfert, R.L., Simon, I., Li, L., et al. (2007). A multiparametric panel for ovarian cancer diagnosis, prognosis, and response to chemotherapy. Clin Cancer Res 13, 6984-6992.
Zhu, C.S., Pinsky, P.F., Cramer, D.W., Ransohoff, D.F., Hartge, P., Pfeiffer, R.M., Urban, N., Mor, G., Bast, R.C., Jr., Moore, L.E., et al. (2011). A framework for evaluating biomarkers for early detection: validation of biomarker panels for ovarian cancer. Cancer Prev Res (Phila) 4, 375-383.
Zielenska, M., Bayani, J., Pandita, A., Toledo, S., Marrano, P., Andrade, J., Petrilli, A., Thorner, P., Sorensen, P., and Squire, J.A. (2001). Comparative genomic hybridization analysis identifies gains of 1p35 approximately p36 and chromosome 19 in osteosarcoma. Cancer Genet Cytogenet 130, 14-21.
156
Summary of Publications Distinct patterns of structural and numerical chromosomal instability characterize sporadic ovarian cancer. Bayani J, Paderova J, Murphy J, Rosen B, Zielenska M, Squire JA. Neoplasia. 2008 Oct;10(10):1057-65. Impact of cytogenetic and genomic aberrations of the kallikrein locus in ovarian cancer. Bayani J, Paliouras M, Planque C, Shan SJ, Graham C, Squire JA, Diamandis EP. Mol Oncol. 2008 Oct;2(3):250-60. Epub 2008 Jul 22. Genomic instability and copy-number heterogeneity of chromosome 19q, including the kallikrein locus, in ovarian carcinomas. Bayani J, Marrano P, Graham C, Zheng Y, Li L, Katsaros D, Lassus H, Butzow R, Squire JA, Diamandis EP. Mol Oncol. 2011 Feb;5(1):48-60. Epub 2010 Aug 11. The physiology and pathobiology of human kallikrein-related peptidase 6 (KLK6). Bayani J and Diamandis EP. Clin Chem Lab Med. 2011 Nov 3;50(2):211-33. Stromal cell-associated expression of kallikrein-related peptidase 6 (KLK6) indicates poor prognosis of ovarian cancer patients. Seiz L, Dorn J, Kotzsch M, Walch A, Grebenchtchikov NI, Gkazepis A, Schmalfeldt B, Kiechle M, Bayani J, Diamandis EP, Langer R, Sweep FC, Schmitt M, Magdolen V. Biol Chem. 2012 Apr 1;393(5):391-401. Modulation of KLK6 Protein Expression in the Ovarian Cancer Cell Line OVCAR-3 by miRNA Let-7a. Bayani J, Kuzmanov U, Saraon P; Fung, WA, Soosaipillai A, Squire JA and Diamanids EP Submitted to Clin Chem. July 2012. The Kallikrein Related Peptidases Editors: Schmitt M and Magdolen V Publisher: De Gruyter In Press Chapter 18: Expression of KLKs under (Patho-) Physiological Conditions Bayani, J, Petraki, KD, Dimitromanolakis A, Milou V, Diamandis EP and Schmitt M Chapter 25: Genomic Instability of the KLK Locus in Various Cancers Bayani, J and Diamandis E.P.