Investigating the effects of, the isoflavone, Phenoxodiol on Prostate

Transcription

Investigating the effects of, the isoflavone, Phenoxodiol on Prostate
INVESTIGATING THE CYTOTOXIC
EFFECTS OF THE ISOFLAVONE,
PHENOXODIOL, ON PROSTATE
CANCER CELLS.
This thesis is presented for the degree of
Doctor of Philosophy
The University of Western Australia
School of Anatomy, Physiology and Human Biology
Simon David Mahoney, BSc Hons
2012
Supervised by:
Professor Arunasalam Dharmarajan
Professor Michael Millward
i
THIS THESIS IS DEDICATED TO MY FAMILY, MY FRIENDS AND MY SUPERVISORS, WHO’S
SUPPORT WAS UNWAVERING.
ii
DECLARATION
The work presented within this thesis was completed between March 2005 and June 2012 in
the School of Anatomy, Physiology and Human Biology at the University of Western Australia. I
hereby declare that all work presented is entirely my own, unless explicitly stated otherwise.
All contributions by others are formally disclosed and duly acknowledged.
Simon Mahoney
1st July 2012
iii
ACKNOWLEGEMENTS
I would like to acknowledge the support of the following people without whom my thesis
would not have been possible.
My supervisors Professor Arun Dharmarajan and Professory Michael Millward
Mr Greg Cozens, Dr Peter Mark, Dr Kathy Heel and Ms Susan Hisheh for their technical
expertise and support.
My fellow students; Clare Berry, Lloyd White, Jeremy Drake, Bijanka Franklin, Brilliana Von
Katterfeld, Mats Hellstrom, Jill Muhling, Hannah Crabb, Margaret Pollet, Melissa Berg, Thea
Shavlakadze, Kasie Jeffrey for their expertise and support.
My Family, Friends and anyone I may have missed for all their support and help throughout the
process and Kerry for helping me through the last obstacles.
iv
CONTENTS
Declaration .................................................................................................................... iii
Acknowlegements ....................................................................................................... iv
List of Equations ............................................................................................................ xi
List of Figures ................................................................................................................ xi
List of Tables ............................................................................................................... xiv
List of Abbreviations .................................................................................................... xv
Thesis Layout............................................................................................................... xix
Abstract ....................................................................................................................... xix
1.
Chapter One: Introduction ................................................................................. 1
2.
Chapter Two: Literature Review ......................................................................... 4
2.1.
Cancer ........................................................................................................................... 4
2.1.1.
Common Cancers and Prostate Cancer In Males in Australia ............................... 4
2.1.2.
Cancer As a Disease In Australia ........................................................................... 6
2.2.
Phenoxodiol .................................................................................................................. 7
2.2.1.
Flavanoids and Isoflavones ................................................................................... 8
2.2.2.
Phenoxodiols Reported Method Of Action ......................................................... 10
2.2.3.
Synergistic Activty ............................................................................................... 13
2.2.4.
Treatment With Phenoxodiol.............................................................................. 13
2.3.
Carcinogenesis / Cancer Development ....................................................................... 14
2.3.1.
Evasion Of Apoptosis .......................................................................................... 17
2.3.2.
Self-Sufficiency In Growth Signals ....................................................................... 17
2.3.3.
Insensitivity To Anti-Growth Signals ................................................................... 18
2.3.4.
Limitless Replicative Potential ............................................................................ 19
2.3.5.
Sustained Angiogenesis....................................................................................... 20
v
2.3.6.
Tissue Invasion And Metastasis ........................................................................... 20
2.3.7.
Pluripotent Stem Cell Differentiation and Development .................................... 21
2.3.8.
Gap Junctional Intercellular Connections ............................................................ 22
2.4.
The Prostate ................................................................................................................ 23
2.4.1.
Testing For Prostate Cancer ................................................................................ 25
2.4.2.
Phenotypic Progression of Prostate Cancer ........................................................ 25
2.4.3.
Treatment of Prostate Cancer ............................................................................. 27
2.4.4.
Chemotherapy and Prostate Cancer ................................................................... 28
2.5.
Apoptosis ..................................................................................................................... 29
2.5.1.
Morphology of Apoptotic Cells ............................................................................ 31
2.5.2.
Apoptotic Signalling ............................................................................................. 33
2.5.3.
Extrinsic Pathways ............................................................................................... 34
2.5.4.
Intrinsic Pathways ................................................................................................ 35
2.5.5.
Caspases .............................................................................................................. 36
2.5.6.
Apoptosome ........................................................................................................ 38
2.5.7.
Bcl-2 Family.......................................................................................................... 38
2.5.8.
X-Linked Inhibitor of Apoptosis Protein .............................................................. 39
2.5.9.
Apoptosis Inducing Factor ................................................................................... 40
2.5.10.
Reactive Oxygen Species ..................................................................................... 41
2.6.
The Cell Cycle ............................................................................................................... 42
2.6.1.
Cyclins, Cyclin Dependant Kinases ....................................................................... 43
2.6.2.
p21WAF1/p53 ......................................................................................................... 43
2.6.3.
Ki-67 ..................................................................................................................... 44
vi
2.6.4.
2.7.
3.
c-Myc ................................................................................................................... 45
Wnt Signalling ............................................................................................................. 45
2.7.1.
β-catenin ............................................................................................................. 46
2.7.2.
Androgen Receptor and β-catenin ...................................................................... 47
2.7.3.
sFRP4 ................................................................................................................... 48
Chapter Three: Materials and Methods ............................................................ 50
3.1.
Tissue Culture.............................................................................................................. 50
3.2.
Cell culture .................................................................................................................. 50
3.2.1.
Subculture And Counting Of Cells ....................................................................... 51
3.2.2.
Cell Cryopreservation and Storage ..................................................................... 51
3.2.3.
LNCaP Cells .......................................................................................................... 52
3.2.3.1
3.2.4.
3.2.4.1
3.2.5.
3.2.5.1
3.3.
LNCaP media formulation ............................................................................... 53
DU145 Cells ......................................................................................................... 53
DU145 Media Formulation ............................................................................. 53
PC3 Cells .............................................................................................................. 54
PC3 Media Formulation .................................................................................. 54
Drug Dilutions ............................................................................................................. 54
3.3.1.
Phenoxodiol ........................................................................................................ 54
3.3.2.
Phenoxodiol Working Stock ................................................................................ 55
3.3.3.
Phenoxodiol Working Solution ........................................................................... 55
3.3.4.
Phenoxodiol Treatment ...................................................................................... 57
3.3.5.
Docetaxel ............................................................................................................ 57
3.3.6.
Docetaxel Working Stock and Working Solution ................................................ 57
3.4.
Optimisation of Proliferation ...................................................................................... 58
vii
3.4.1.
Determination of Cell Seeding Concentrations ................................................... 58
3.4.2.
Phenoxodiol MTS Assay ....................................................................................... 59
3.4.3.
Phenoxodiol Docetaxel Isobologram ................................................................... 61
3.4.4.
Phenoxodiol And Caspase Inhibition ................................................................... 64
3.4.5.
Phenoxodiol and Docetaxel Combination Treatment ......................................... 65
3.4.6.
Phenoxodiol and Purified sFRP4 Protein Combination Treatment ..................... 66
3.5.
Reactive Oxygen Species Detection............................................................................. 67
3.6.
Acidity Analysis ............................................................................................................ 69
3.7.
Apoptosis Analysis Assays ........................................................................................... 69
3.7.1.
DNA Extraction .................................................................................................... 69
3.7.2.
3’-End Labelling DNA Fragmentation Analysis .................................................... 71
3.7.3.
Annexin-V-Fluos Propidium Iodide Flow Cyometry............................................. 72
3.7.4.
Sybr Gold Fragmentation Analysis....................................................................... 75
3.7.5.
JC-1 Mitochondrial Potential Assay ..................................................................... 76
3.7.6.
Caspase-3 Activity Assay...................................................................................... 78
3.8.
Cell Cycle Analysis ........................................................................................................ 79
3.8.1.
Cell Cycle Preparation.......................................................................................... 80
3.8.2.
Cell Cycle Flow Cytometry ................................................................................... 81
3.8.3.
Cell Cycle Data Analysis ....................................................................................... 85
3.9.
Assessment of RNA Expression ................................................................................... 87
3.9.1.
RNA Extraction ..................................................................................................... 88
3.9.2.
RNA Integrity ....................................................................................................... 89
3.9.3.
DNAse Treatment ................................................................................................ 89
viii
3.9.4.
Reverse Transcriptase Polymerase Chain Reaction ............................................ 90
3.9.5.
Post PCR Clean-Up .............................................................................................. 91
3.9.6.
Primer Design ...................................................................................................... 92
3.9.7.
Real Time Quantitative PCR Analysis .................................................................. 93
3.9.8.
Gel Electrophoresis and Extraction ..................................................................... 96
3.9.9.
qPCR Standard Production .................................................................................. 97
3.10.
3.10.1.
Protein Extraction ............................................................................................. 101
3.10.2.
Bradford Protein Quantification Assay ............................................................. 102
3.10.3.
SDS-PAGE Western Blot Analysis ...................................................................... 103
3.10.4.
Immunoblotting ................................................................................................ 107
3.10.5.
Quantification of Western Blot Analysis ........................................................... 109
3.11.
4.
Assessment of Protein Expression ........................................................................ 100
Statistical Analysis ................................................................................................. 110
Chapter Four: The Cytotoxic Effects of Phenoxodiol On The Prostate Cancer Cell
Lines; LNCaP, DU145 and PC3 ................................................................................ 111
5.
4.1.
Introduction .............................................................................................................. 111
4.2.
Aims........................................................................................................................... 114
4.3.
Methodology ............................................................................................................. 114
4.4.
Results ....................................................................................................................... 118
4.5.
Discussion.................................................................................................................. 134
Chapter Five: Cell Death Signalling In Response To Phenoxodiol Treatment .... 138
5.1.
Introduction .............................................................................................................. 138
5.2.
Aims........................................................................................................................... 142
ix
5.3.
Methodology ............................................................................................................. 142
5.4.
Results ....................................................................................................................... 145
5.5.
Discussion .................................................................................................................. 166
6.
Chapter Six: The Effects of Phenoxodiol On The Cell Cycle .............................. 172
6.1.
Introduction ............................................................................................................... 172
6.2.
Aims ........................................................................................................................... 177
6.3.
Methodology ............................................................................................................. 177
6.4.
Results ....................................................................................................................... 179
6.5.
Discussion .................................................................................................................. 197
7.
Chapter Seven: Phenoxodiol in Combination with Docetaxel ......................... 203
7.1.
Introduction ............................................................................................................... 203
7.2.
Aims ........................................................................................................................... 207
7.3.
Methodology ............................................................................................................. 207
7.4.
Results ....................................................................................................................... 210
7.5.
Discussion .................................................................................................................. 225
8.
Chapter Eight: General Discussion.................................................................. 230
8.1.
Discussion .................................................................................................................. 230
8.2.
Conclusion ................................................................................................................. 239
8.3.
Limitations ................................................................................................................. 242
9.
10.
Bibliography .................................................................................................. 244
Appendices ................................................................................................ 261
x
LIST OF EQUATIONS
Equation 1 Phenoxodiol Working Stock Equation ...................................................................... 55
Equation 2 Phenoxodiol Working Solution Equation ................................................................. 56
Equation 3 Docetaxel Stock And Working Solutions .................................................................. 58
LIST OF FIGURES
Figure 1: Burden of Disease in Australia In 2003 .......................................................................... 6
Figure 2 Phenoxodiol Molecular Structure ................................................................................... 8
Figure 3 Production of Equol From Daidzein ................................................................................ 9
Figure 4 Phenoxodiols Reported Method of Action As Adapted From Silasi et al. 2009 ........... 12
Figure 5: Acquired Capabilities of Cancer Adapted from Hanahan et al 2000 and Trosko et al
2004 ............................................................................................................................................ 16
Figure 6: Stem Cell Theory Adapted from Trosko et al 2004 ...................................................... 22
Figure 7 Characteristic Progression of Prostate Cancer From Early to Late Stage ..................... 26
Figure 8 Development Of Resistance To Chemotherapeutic Treatment Adapted from
Johnstone, Ruefli and Lowe 2002 ............................................................................................... 29
Figure 9 Simple Apoptotic Signalling Adapted From Riedl and Salvesen 2007 .......................... 37
Figure 10 MTS Assay Setup ......................................................................................................... 60
Figure 11 96 Well isobologram Setup ......................................................................................... 62
Figure 12 LNCaP Sybr Gold Visualised DNA Fragmentation With DNA Ladder, Low Weight DNA
Fragmentation Is Visible ............................................................................................................. 76
Figure 13 Cell Populations Multi-Gated In FACS Diva Software For acquisition And Analysis ... 83
Figure 14 Cell Cycle Populations For Analysis Aquired Using Gated Populations In FACS Diva
Software ...................................................................................................................................... 84
Figure 15 Example Of Cell Population Gating For Analysis In FlowJo Software ......................... 85
xi
Figure 16 Watson versus Dean-Jett-Fox Cell Cycle Analysis In FlowJo Software ........................ 86
Figure 17 Watson Pragmatic Cell Cycle Analysis of PC3 Cells Undergoing Phenoxodiol
Treatment .................................................................................................................................... 87
Figure 18 Example Of A Cyclin-D1 Melt Curve ............................................................................ 96
Figure 19 qPCR Standards Cycling Assessment ........................................................................... 98
Figure 20 Standards Analysis by Corbett Rotor-Gene 6000 Software. ....................................... 99
Figure 21 qPCR Full Run Standards and Samples ........................................................................ 99
Figure 22 Standards and Samples Analysis By Corbett Rotor-Gene 6000 Software ................. 100
Figure 23 Cell Proliferation Rates At 6, 24 and 48 hours .......................................................... 119
Figure 24 Cell Proliferation Rates After 24 and 48 Hours Of Phenoxodiol Treatment ............. 121
Figure 25 pH Changes In Phenoxodiol Treated Culture Media ................................................. 122
Figure 26 JC-1 Analysis of Mitochondrial Membrane Potential Over 24 and 48 Hours of
Treatment .................................................................................................................................. 125
Figure 27 Fluorescent Analysis of Caspase-3 Activity after Phenoxodiol Treatment Over 24 and
48 Hours .................................................................................................................................... 127
Figure 28 3'-End Labelling Apoptotic Analysis Post Phenoxodiol Treatment over 24 and 48
Hours ......................................................................................................................................... 129
Figure 29 An Example of LNCaP 3’-end Labelling Qualitative DNA Laddering After Exposure to
Film ............................................................................................................................................ 130
Figure 30 Annexin V-Fluos / Propidium Iodide Double Staining Analysis of Prostate Cancer Cells
Post Phenoxodiol Treatment over 24 and 48 Hours ................................................................. 133
Figure 31 AIF mRNA Expression Analysis of Prostate Cancer Cells Post Phenoxodiol Treatment
Over 24 And 48 Hours ............................................................................................................... 146
Figure 32 Bax mRNA Expression Analysis of Prostate Cancer Cells Post Phenoxodiol Treatment
Over 24 And 48 Hours ............................................................................................................... 148
Figure 33 Bcl-xL mRNA Expression Analysis of Prostate Cancer Cells Post Phenoxodiol
Treatment Over 24 And 48 Hours ............................................................................................. 150
xii
Figure 34 Caspase-3 mRNA Expression Analysis of Prostate Cancer Cells Post Phenoxodiol
Treatment Over 24 And 48 Hours ............................................................................................. 152
Figure 35 xIAP mRNA Expression Analysis of Prostate Cancer Cells Post Phenoxodiol Treatment
Over 24 And 48 Hours ............................................................................................................... 154
Figure 36 AIF Protein Level Analysis Of Prostate Cancer Cells Post Phenoxodiol Treatment Over
24 And 48 Hours........................................................................................................................ 156
Figure 37 Bax Protein Level Analysis Of Prostate Cancer Cells Post Phenoxodiol Treatment Over
24 And 48 Hours........................................................................................................................ 157
Figure 38 Bcl-xL Protein Level Analysis Of Prostate Cancer Cells Post Phenoxodiol Treatment
Over 24 And 48 Hours ............................................................................................................... 159
Figure 39 xIAP Protein Level Analysis Of Prostate Cancer Cells Over 24 and 48 Hours Post
Phenoxodiol Treatment Over 24 And 48 Hours ........................................................................ 161
Figure 40 Caspase Inhibition Treatment With 10µM Z-VAD-FMK (CI) and 30µM Phenoxodiol
(PXD) Over 48 hours.................................................................................................................. 163
Figure 41 Nitric Oxide Formation In Prostate Cancer Cells Over 24 and 48 Hours Post
Phenoxodiol Treatment Measured Via Griess Assay ................................................................ 165
Figure 42 LNCaP Cell Cycle Analysis After 24 and 48 Hours Of 10µM and 30µM Phenoxodiol
Treatment ................................................................................................................................. 180
Figure 43 DU145 Cycle Analysis After 24 and 48 Hours Of 10µM and 30µM Phenoxodiol
Treatment ................................................................................................................................. 182
Figure 44 PC3 Cell Cycle Analysis After 24 and 48 Hours Of 10µM and 30µM Phenoxodiol
Treatment. ................................................................................................................................ 184
Figure 45 c-Myc mRNA Expression Analysis of Prostate Cancer Over 24 and 48 Hours Post
Phenoxodiol Treatment. ........................................................................................................... 186
Figure 46 Cyclin-D1 mRNA Expression Analysis of Prostate Cancer Cells Over 24 and 48 Hours
Post Phenoxodiol Treatment. ................................................................................................... 188
xiii
Figure 47 Ki-67 mRNA Expression Analysis of Prostate Cancer Cells Over 24 and 48 Hours Post
Phenoxodiol Treatment. ............................................................................................................ 190
Figure 48 p21 mRNA Expression Analysis of Prostate Cancer Over 24 and 48 Hours Post
Phenoxodiol Treatment. ............................................................................................................ 192
Figure 49 Active β-catenin Protein Level Analysis of Prostate Cancer Cells Over 24 and 48 Hours
Post Phenoxodiol Treatment. .................................................................................................... 194
Figure 50 sFRP4 Protein Level Analysis of Prostate Cancer Cells Over 24 and 48 Hours Post
Phenoxodiol Treatment ............................................................................................................. 196
Figure 51 Docetaxel Response Curve Measured After 48 Hours of Treatment. ....................... 212
Figure 52 LNCaP Phenoxodiol / Docetaxel Isobologram Measured After 48 Hours of Treatment
................................................................................................................................................... 213
Figure 53 DU145 Phenoxodiol / Docetaxel Isobologram Measured After 48 Hours Of
Treatment. ................................................................................................................................. 214
Figure 54 PC3 Phenoxodiol / Docetaxel Isobologram Measured After 48 Hours Of Treatment.
................................................................................................................................................... 215
Figure 55 10µM Phenoxodiol 100nM Docetaxel Combination Therapy Measured After 48
Hours Of Treatment................................................................................................................... 218
Figure 56 30µM Phenoxodiol 100nM Docetaxel Combination Therapy Measured After 48
Hours Of Treatment................................................................................................................... 221
Figure 57 30µM PXD And 500pg/mL sFRP4 Protein Combination Therapy After 48 Hours. .... 224
LIST OF TABLES
Table 1: Predicted Prostate Cancer Rates 2011-2020, Adapted From Begg, 2007 ....................... 5
Table 2 The 10 Most Commonly Occuring Cancer In Australia 2007, Adapted From Begg 2007 . 5
Table 3 Phenoxodiol Treatment Calculations Per mL Of Media ................................................. 56
Table 4: Seeding Concentrations And Volumes .......................................................................... 59
Table 5 Isobologram Setup For Vehicle Control And Treatment Groups.................................... 63
xiv
Table 6 Caspase Inhibition and Phenoxodiol Treatment Solutions ............................................ 65
Table 7 sFRP4 & Phenoxodiol Treatment Solutions ................................................................... 67
Table 8 3'-End Labelling Reaction Mixture ................................................................................. 71
Table 9 Cell Staining With Annexin-V/Propidium Iodide (AV/PI) Double Stain .......................... 73
Table 10 Cell Staining Setup For AV/PI ....................................................................................... 74
Table 11 AV/PI Labelling Solutions ............................................................................................. 74
Table 12 Propidium Iodide Staining Solution Per 30mL ............................................................. 81
Table 13 RQ1 DNase Treatment Reaction Components............................................................. 90
Table 14 RT PCR Reaction Components ..................................................................................... 91
Table 15 Primer Sequences, Product Size And Annealing Temperature .................................... 93
Table 16 Immolase Taq qPCR Reaction Mix ............................................................................... 95
Table 17 qPCR Standards Serial Dilution .................................................................................... 98
Table 18 Bradford Protein Standards Setup ............................................................................. 102
Table 19 SDS-Page/Acrylamide Gel Recipe............................................................................... 104
Table 20 Example Of Protein Sample Preparation ................................................................... 105
Table 21 Antibody Concentrations ........................................................................................... 108
Table 22 Immunoblotting Protocol........................................................................................... 109
Table 23 Phenoxodiol/Docetaxel Concentration Combinations For An Isobologram .............. 208
LIST OF ABBREVIATIONS
Abbreviation
Extended Form
µ, µm, µM
Micro, micrometer, micromolar
AAT
Androgen Ablation Therapy
AIF
Apoptosis Inducing Factor
APS
Ammonium Persulfate
AR
Androgen Receptor
ARE
Antioxidant Response Element
ATCC
American Type Culture Collection
xv
AV, AVF, Annexin-V
Annexin-V-Fluorescein
Bax
Bcl-2-associated X Protein
Bcl-2
B-cell Lymphoma 2
Bcl-xL
B-cell lymphoma extra large
BLAST
Basic Logical Alignment Search Tool
BSA
Bovine Serum Albumin
BP
Base Pair
BPH
Benign Prostatic Hyperplasia
Cat#
Catalogue Number
Cdk
Cyclin Dependant Kinase
CdkI
Cyclin Dependant Kinase Inhibitor
cDNA
Complementary DNA
CHAPS
3-[(3-Cholamidopropyl)dimethylammonio]-1-propanesulfonate
CI
Caspase Inhibitor (Z-VAD-FMK)
CNS
Central Nervous System
CRPC
Castrate Resistant Prostate Cancer
DALY
Disability Adjusted Life Years
ddATP
Dideoxyadenosine Triphosphate
ddH2O
Double distilled water / also deionised water
DEAN
Diethylamine NONOate diethylammonium salt
DEPC
Diethylpyrocarbonate
DHT
5-alpha-dihydrotestosterone
DISC
Death-Inducing Signalling Complex
DJF
Dean-Jett-Fox
DMDC
Dimethyl dicarbonate
DMSO
Dimethyl Sulfoxide, (CH3)2SO
DNA
Deoxyribonucleic Acid
DOC
Docetaxel
DTT
Dithiothreitol
EDTA
Ethylenediaminetetraacetic Acid
EGTA
Ethyleneglyocoltetraacetic Acid
ER
Estrogen Receptor
FACS
Fluorescence-Activated Cell Sorting
FBS
Fetal Bovine Serum
xvi
FCCP
Carbonylcyanide-4-trifluoromethoxyphenylhydrazone
FLIP
Flice Like Inhibitory Protein
FLUOS
Fluorescein
FSC, -A, -H
Forward Scatter, -Area, -Height
FZD
Frizzled Receptor
g
RCF (see below) = 0.00001118rN2 r=rotational radius cm
N=rotating speed in revolutions per minute (RPM)
GIJC
Gap Junctional Intercellular Connections
GSK3-β
Glycogen Synthase Kinase 3 - Beta
HEPES
4-(2-Hydroxyethyl)piperazine-1-ethanesulfonic acid
HRP
Streptavidin-Horse Radish Peroxidase
HRPC
Hormone refractory prostate cancer
LEF
Lymphoid Enhancer Factor
LRP5/LRP6
Low Density Lipoprotein Receptor-Related Protein
MEM
Minimum Essential Medium
M, mM
Molar, millimolar
M-MLV
Moloney Murine Leukaemia Virus
mRNA
Messenger Ribonucleic Acid
MTS
3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)2-(4-sulfophenyl)-2H-tetrazolium
NED
N-1-naphthylethylenediamine dichloride
nm, nM
Nanometre, nanomolar
NO
Nitric Oxide
PAGE
Polyacrylamide Gel Electrophoresis
PARP
poly(ADP-ribose) polymerase
PBS
Phosphate Buffer Solution
PC2
Physical Containment Level 2
PCR
Polymerase Chain Reaction
PES
Phenazine Ethosulfate
PI
Propidium Iodide
PI_DNA, -A, -H, -W
Propidium Iodide, -Area, -Height, -Width
PIPES
1,4-piperazinediethanesulfonic acid
PMSF
Phenylmethylsulfonyl fluoride
PS
Phosphatidylserine
PSA
Prostate Specific Antigen
xvii
PXD
Phenoxodiol
QPCR
Quantitative Polymerase Chain Reaction
RCF
Relative Centrifugal Force = 0.00001118rN2
Redox
Reduction Oxidation
RIPA
Radioimmunoprecipitation
RNA
Ribonucleic Acid
ROS
Reactive Oxygen Species
RT PCR
Reverse Transcriptase Polymerase Chain Reaction
SDS
Sodium Dodecyl Sulfate
SDS-PAGE
Sodium Dodecyl Sulfate Polyacrylamide Gel Electrophoresis
shRNA
Short Hairpin Ribonucleic Acid
SNP
Sodium Nitroprusside
SOD
Superoxide Dismutase
SSC, -A, -H
Side Scatter, -Area, -Height
TAE
Tris-acetate EDTA
TBS-T
Tris Buffered Saline with Tween
TCF
T-cell factor
TE
Tris EDTA
TEMED
N,N,N',N'-Tetramethylethylenediamine
TNF-α
Tumour Necrosis Factor Alpha
TRIS
2-Amino-2-(hydroxymethyl)-1,3-propanediol
TRIS-HCL
2-Amino-2-(hydroxymethyl)-1,3-propanediol with a pH set by
Hydrochloric acid addition
TRUS
Transrectal ultrasound
TURP
Transurethral Resection of Prostate
Versus
Versus
xIAP
X-linked Inhibitor of Apoptosis Protein
Z-DEVD-R110
Rhodamine 110 bis-(N-CBZ-L-aspartyl-L-glutamyl-L-valyl-L-aspartic
acid amide)
Z-VAD-FMK
Carbobenzoxy-valyl-alanyl-aspartyl-(O-methyl)fluoromethylketone
xviii
THESIS LAYOUT
This thesis comprises of 10 chapters divided into; Introduction, Literature Review,
Methodology, Four experimental chapters, General Discussion, Bibliography and Appendices.
Each experimental chapter is divided into; an introduction, brief methodology, results and
discussion section. The general discussion section is divided into discussion, conclusion and
limitations of the study.
ABSTRACT
In this study we investigated the cytotoxic effects of the isoflavone molecule Phenoxodiol, on
the prostate cancer cell lines LNCaP, DU145 and PC3. LNCaP cells represented a hormone
responsive early stage prostate cancer while DU145 and PC3 cells represented late stage
hormone refractory and chemoresistant prostate cancer. We investigated the cytotoxic effects
induced in these cell lines over a period of 24 or 48 hours of treatment with 10µM and 30µM
doses of phenoxodiol. Cells exhibited significant cytotoxicity and mitochondrial depolarisation
in response to phenoxodiol treatment. The study established LNCaP and DU145 cells
responded apoptotically to treatment, while PC3 cells respond necrotically. After confirming
cytotoxic responses the underlying signalling mechanisms were investigated which revealed
that phenoxodiol was not inducing cell death via an increase in Caspase-3 activity, in two of
three cell lines, and the apoptotic machinery of the cells responded in a varied manner
between cell lines. Phenoxodiol was revealed to work via Caspase independent actions and the
effect of the mitochondrial signalling molecules was not consistently altered in response to
treatment indicating phenoxodiol did not directly target intrinsic or extrinsic caspase signalling
pathways. We then investigated the effects of phenoxodiol on the cell cycle where treatment
groups exhibited a significant increase in G1 and S phase populations and a corresponding
decrease in G2 phase populations. The underlying signalling was investigated and it was
determined that p21WAF1 expression was increased consistently between the three cell lines in
xix
response to treatment. Finally we investigated the effects of phenoxodiol in combination with,
the G2 phase inhibitor, docetaxel and purified sFRP4 protein, a known agonist of the Wnt/βcatenin growth regulation pathway. Phenoxodiol exhibited the ability to interfere with
docetaxel treatment as it prevented cells from reaching G2 phase, where docetaxel is
effective. Phenoxodiol exhibited the ability to increase the effectiveness of sFRP4 with
combination therapy significantly increasing the effectiveness of the two compounds
compared to individual responses. Phenoxodiol exhibits significant cytotoxicity, inducing
caspase independent cell death in the prostate cancer cell lines LNCaP, DU145 and PC3. All
three cell lines had significantly decreased viable cell populations, after only 48 hours of
treatment, through G1 and S phase cell cycle arrest. By targeting non classical apoptotic
pathways and successfully inducing cell cytotoxicity even in the most chemoresistant cell lines
and coupled with the reported ability of high tolerance of orally ingested phenoxodiol and few
reported side effects; Phenoxodiol represents a strong, effective, potential treatment for all
stages of prostate cancer.
xx
1. CHAPTER ONE: INTRODUCTION
Prostatic adenocarcinoma is the second most commonly diagnosed malignancy in men and is
the most common cause of death in men in many western countries. Like other cancers,
prostate cancer has numerous clinical states ranging from a hormone-naïve clinically localised
primary tumour to lethal androgen-independent metastases (Arnold and Isaacs 2002).
Regulation of prostate growth is mediated via androgens and the corresponding androgen
receptor (AR) which regulates the transcription of target survival and apoptosis genes. Latephase metastatic disease is often androgen independent arising from either increased AR
expression and enhanced nuclear localisation of the AR, AR mutation resulting in a more
promiscuous receptor or the presence of alternative survival pathways (Bcl-2 upregulation)
that circumvent the AR (Feldman and Feldman 2001; Litvinov et al. 2003).
Epidemiological studies have consistently shown an inverse association between isoflavone
intake and the risk of cancer (Beecher 2003; Brown et al. 2005). In vitro mechanistic studies on
isoflavones provide insight into potential modes of anti-cancer action ranging from cell cycle
arrest and apoptosis induction to anti-angiogenic and anti-proliferative effects (Middleton et
al. 2000; Brown et al. 2005). Phenoxodiol, 2H-1-benzopyran-7-0,1,3-(4-hydroxyphenyl), is an
isoflavone derivative that has been shown to elicit cytotoxic effects against a broad range of
human cancers (Aguero et al. 2005; Axanova et al. 2005; Brown et al. 2005; Silasi et al. 2009).
Previous reports have indicated that phenoxodiol is a topoisomerase II inhibitor, inhibits
sphingosine kinase activity, downregulates transcription of angiogenic matrix metalloprotease
2 and other markers of angiogenesis, and can also induce apoptosis in chemoresistant ovarian
cancer cells by regulating anti-apoptotic signalling pathways (Constantinou et al. 2003;
Kamsteeg et al. 2003; Sapi et al. 2004; Aguero et al. 2005; Axanova et al. 2005; Brown et al.
2005; Gamble et al. 2005; Alvero et al. 2006). Phenoxodiol has been reported to inhibit cell
1
proliferation in a wide range of human cancer cell lines and induce G1 cell cycle arrest as
opposed to the G2-M arrest seen in similar flavanoid compounds like genistein (Alhasan et al.
1999; Aguero et al. 2005). It is postulated that phenoxodiol elicits its global anticancer activity
by modulating the sphingomyelin pathway (De Luca et al. 2005; Morre et al. 2007). Recent
data, however, have demonstrated that phenoxodiol is able to enhance perforin induced cell
death elicited by T-cells in both in vitro and in vivo models of colon cancer (Georgaki et al.
2009). Recent evidence has demonstrated that phenoxodiol augments the anti-cancer activity
of cisplatin against the DU145 prostate cancer cell line both in in vitro and in vivo studies. An
intracellular cisplatin accumulation assay showed a 35% (p<0.05) increase in the uptake of
cisplatin when cells were treated with a combination of 1μM cisplatin and 5μM phenoxodiol,
resulting in a 300% (p<0.05) increase in DNA adducts, hence explaining the sensitisation effect
(McPherson et al. 2009).
The mainstay of primary prostate treatment options is combined androgen ablation and
taxane therapy; while this approach is curative in early stage cases late stage treatment often
results in refractory disease development (Montero et al. 2005). Treatment options for
patients with late stage metastatic prostate carcinoma rely on the premise that androgeninsensitive prostate carcinoma cells retain their basic cellular apoptotic machinery to undergo
programmed cell death, however increased tumour-cell resistance to apoptosis is an
underlying molecular reason contributing to disease progression and chemo-resistance
(Gleave et al. 2005). Apoptosis, the process of physiological cell death, is a focal area for the
study of cancer cell death. Apoptotic signalling is a mechanism of ordered cell death used by
the body to remove cells with damaged genetic material, and as such has evolved into a highly
complex and multi-channelled pathway to prevent accidental cell death from occurring.
Cancerous cells are able to prevent these messages of apoptosis from activating their
intracellular pro-apoptotic pathways. The ability of cancer cells to evade apoptosis is an
2
essential hallmark of cancer, and a key objective of cancer therapy is to restore cellular
sensitivity to apoptosis (Hanahan and Weinberg 2000).
In this study we investigated and quantified the cytotoxic effects of phenoxodiol treatment in
the prostate cancer cell lines LNCaP, DU145 and PC3 which represent early through late stage
prostate cancer. We investigated the type of cytotoxic response, the underlying cell death
signalling machinery, the effects on the cell cycle and the cell cycle signalling and the combined
effect of phenoxodiol and docetaxel and phenoxodiol and sFRP4, a Wnt/β-catenin cell growth
pathway agonist.
3
2. CHAPTER TWO: LITERATURE
REVIEW
2.1.
CANCER
Cancer describes a range of diseases in which abnormal cells proliferate and spread out of
control. Common descriptions for cancer types are; benign or unlikely to harm the host,
malignant or harmful to the host and metastatic where a malignant tumour has spread into
other tissues in the body and implanted, growing in size. Other terms for cancer include
tumours and neoplasms, although these terms can also be used as descriptive terms for noncancerous growths. Carcinomas are malignant tumours arising from an epithelial origin while
sarcomas are malignant tumours arising from a connective tissue origin (Stevens 2009).
2.1.1. COMMON CANCERS AND PROSTATE CANCER IN MALES IN
AUSTRALIA
In 2007, there were more than 62,000 new cases of cancer diagnosed in males, which when
age-standardised equated to about 595 cases per 100,000 males. While age-standardised rates
of all cancers combined for males have risen since 1998, it is expected that they will steady at
about 568 new cases per 100,000 males between 2011 and 2020. This is primarily due to the
anticipated steadying of rates of prostate cancer incidence, coupled with decreasing rates in
lung cancer in males, which accounts for a further 9–10% of cases. When taking into account
expected changes in the population structure, this will translate into about 85,000 new cases
in males expected to be diagnosed in 2020 (Begg 2007).
4
TABLE 1: PREDICTED PROSTATE CANCER RATES 2011-2020, ADAPTED FROM BEGG, 2007
Year
0–24 years
25–44 years
45–64 years
65–84 years
85+ years
Count
Rate
Count
Rate
Count
Rate
Count
Rate
Count
Rate
2011
35
0.9
30
0.9
7,750
276.8
11,460
891.6
1,660
1131.1
2012
35
0.9
30
0.9
8,060
286.1
12,150
902.6
1,750
1130.8
2013
35
1.0
30
1.0
8,420
296.3
12,790
913.8
1,850
1130.7
2014
40
1.0
35
1.0
8,820
306.6
13,400
923.2
1,930
1130.3
2015
40
1.0
35
1.0
9,210
316.5
14,030
932.3
2,020
1129.2
2016
40
1.1
35
1.1
9,640
326.3
14,660
940.9
2,100
1128.2
2017
45
1.1
40
1.1
10,070
336.3
15,250
945.2
2,160
1127.1
2018
45
1.2
40
1.2
10,490
346.6
15,890
951.2
2,220
1126.2
2019
45
1.2
40
1.2
10,930
357.7
16,550
958.9
2,280
1125.3
2020
50
1.2
45
1.2
11,350
368.6
17,220
966.5
2,340
1124.4
TABLE 2 THE 10 MOST COMMONLY OCCURING CANCER IN AUSTRALIA 2007, ADAPTED FROM BEGG
2007
Males
Site/type
Cases
Females
ASR
CI (95%)
Site/type
Cases
ASR
CI (95%)
Prostate
19,403
182.9
180.3–
185.5
Breast
12,567
109.2
107.3–
111.1
Bowel
7,804
75.2
73.5–76.9
Bowel
6,430
53.4
52.1–54.7
Melanoma
of skin
5,980
57.2
55.7–58.7
Melanoma
of skin
4,362
38.2
37.1–39.4
Lung
5,948
57.9
56.5–59.4
Lung
3,755
31.3
30.3–32.4
Lymphoid
cancers
4,116
39.6
38.4–40.8
Lymphoid
cancers
3,160
26.8
25.9–27.8
Myeloid
cancers
1,859
18.5
17.7–19.4
Uterus
1,942
16.5
15.8–17.3
Kidney
1,716
16.3
15.5–17.1
Unknown
primary
1,401
11.0
10.4–11.6
Bladder
1,644
16.5
15.7–17.3
Thyroid
1,331
12.2
11.6–12.9
Unknown
primary
1,496
14.9
14.2–15.7
Ovary
1,266
10.8
10.2–11.4
Pancreas
1,352
13.1
12.4–13.8
Myeloid
cancers
1,232
10.1
9.5–10.7
All cancers
62,019
595.1
599.8
All cancers
46,349
393.9
397.5
ASR: The rates were standardised to the Australian population as at 30 June 2001 and are
expressed per 100,000 population.
5
Prostate cancer accounts for about 30% of all new cases of cancer diagnosed in males
(excluding non-melanoma skin cancers), and is the second most common cause of cancerrelated death in males after lung cancer. With 19,403 new cases diagnosed in Australia in
2007, prostate cancer is a major public health concern (Begg 2007).
2.1.2. CANCER AS A DISEASE IN AUSTRALIA
In 2003, Cancer over took cardiovascular disease, 19% to 18%, as the biggest burden of disease
and injury in Australia. Four Fifths of that burden was from premature deaths, with Lung,
Colorectal, Breast and Prostate Cancer accounting for the leading majority of burden (Begg
2007).
FIGURE 1: BURDEN OF DISEASE IN AUSTRALIA IN 2003
Burden of disability adjusted life years (DALYs) by broad cause group expressed as: proportions
of total, proportions by sex and proportions due to fatal and non-fatal outcomes, Australia,
2003 (Begg 2007)
6
Total costs to the Australian health system were 2.9 billion or 5.8% of the total health
expenditure for 2003 (Begg 2007). Since 2003 statistics were released however, many newer,
high cost drugs have entered onto the market, examples of which are Herceptin and Avastin.
Their use for common cancers like breast and colorectal, will significantly increase the costs of
conventional cancer treatment as part of the Australian health expenditure.
2.2.
PHENOXODIOL
Phenoxodiol ,[2H-1-benzopyran-7-0, 1,3-(4-hydroxyphenyl)], is a synthetic isoflavone molecule
first isolated from soy beans and now currently undergoing Phase III clinical trials for the
treatment of platinum and taxane refractory ovarian cancer (Morre et al. 2009). Phenoxodiol
has been shown to elicit cytotoxic effects against a broad range of human cancers (Sapi et al.
2004; Aguero et al. 2005; Axanova et al. 2005; Brown et al. 2005; Silasi et al. 2009; Aguero et
al. 2010). An isoflavone derivative, phenoxodiol has been shown to have a greater bioavailability and increased potency/efficacy than its parent compound, which suffered in clinical
trials due to low bio availability (Aguero et al. 2005; Klein et al. 2007). Regardless of
progression in clinical trials, the primary cellular signalling target or targets of phenoxodiol
remain elusive and multiple studies covering various cancer cell types have detected different
methods of action (Straszewski-Chavez et al. 2004; Alvero et al. 2007; Georgaki et al. 2009;
Herst et al. 2009; McPherson et al. 2009; Saif et al. 2009; Aguero et al. 2010; de Souza et al.
2010; Wu et al. 2011).
7
FIGURE 2 PHENOXODIOL MOLECULAR STRUCTURE
The molecular structure of Phenoxodiol represented in 3D as presented at www.novogen.com
2.2.1. FLAVANOIDS AND ISOFLAVONES
In vitro mechanistic studies on isoflavones provide insight into potential modes of anti-cancer
action ranging from cell cycle arrest and apoptosis induction to anti-angiogenic and antiproliferative effects (Middleton et al. 2000; Brown et al. 2005). Strong in vitro evidence exists
for the activity of a variety of isoflavones on hormone sensitive and insensitive prostate cancer
cell lines (Hempstock et al. 1998; Mitchell et al. 2000; Hedlund et al. 2003) and in vivo in rats
(Risbridger et al. 2001). Isoflavones appear to have pleiotropic effects on prostate cancer cells,
including an ability to exert hormonal influences. Phenoxodiol is a synthetic isoflavone
metabolite that is a natural intermediate (dehydroequol, 7,40-dihydroxyisoflav-3-ene) in the
metabolism of daidzein to equol (Joannou et al. 1995). It is cytotoxic in vitro and in vivo in rats
(Constantinou et al. 2003; Mor et al. 2006). It may be capable of re-sensitising platinum- and
taxane-resistant ovarian cancer cells in vitro (Kamsteeg et al. 2003; Sapi et al. 2004; Kluger et
al. 2007) and appears to have antiangiogenic (Gamble et al. 2005) and anti-inflammatory
8
properties (Widyarini et al. 2001). It has improved bioavailability when compared with
genistein and low toxicity in clinical trials (Kelly and Husband 2003; Davies 2005; de Souza et al.
2006).
FIGURE 3 PRODUCTION OF EQUOL FROM DAIDZEIN
Production of Phenoxodiol naturally occurs as a process of Daidzein to Equol formation
(Joannou et al. 1995). Figure adapted from (Jian 2009).
9
2.2.2. PHENOXODIOLS REPORTED METHOD OF ACTION
Phenoxodiol has been reported to elicit diverse cytotoxicity activity via several mechanisms
including induction of G1 arrest via p53-independent p21WAF1 regulation, resulting in loss of
cyclin-D kinase activity in HN12 head and neck cancer cells, strong S-phase arrest in ovarian
cancer cells at high concentrations, restoration of death receptor-refractory cancer cells to
respond to extracellular death signals by re-enabling the transduction of an activated Fas or
TRAIL receptor signal to Caspase-8 due to phenoxodiol induced degradation of Flice Like
Inhibitory Proteins (FLIP) which enables activation of the extrinsic apoptosis pathway (Aguero
et al. 2005; Alvero et al. 2006; Alvero et al. 2007; Alvero et al. 2008; Aguero et al. 2010).
Phenoxodiol also engages the intrinsic apoptosis pathway, which is typically non-functional in
late stage cancer due to overexpression of pro-survival factors such as sphingosine-1
phosphate, which drives the stabilisation of Akt causing an accumulation of XIAP and
subsequent inhibition of Caspase-3 and -9 (Choueiri et al. 2006; Alvero et al. 2008). Recent
data, however, have demonstrated that phenoxodiol is able to enhance perforin induced cell
death elicited by T-cells in both in vitro and in vivo models of colon cancer (Georgaki et al.
2009). Recent evidence has demonstrated that phenoxodiol augments the anti-cancer activity
of cisplatin against the DU145 prostate cancer cell line both in in vitro and in vivo studies. An
intracellular cisplatin accumulation assay showed a 35% (p<0.05) increase in the uptake of
cisplatin when cells were treated with a combination of 1μM cisplatin and 5μM phenoxodiol,
resulting in a 300% (p<0.05) increase in DNA adducts, hence explaining the sensitisation effect
(McPherson et al. 2009).
An early event in the phenoxodiol mechanism of action is the disruption of the
sphingomyelinase pathway resulting in an accumulation of ceramide, which drives the
truncation of Bid by Caspase-2, thereby resulting in Bid translocation to the mitochondria.
Caspase-8 also contributes to Bid formation accompanied by Bax translocation to the
10
mitochondria, both Bax and Bid serve to depolarise the mitochondria, resulting in Cytochrome
c release and the formation of the apoptosome utilising Caspase 9. Key to the mechanism by
which phenoxodiol enables the reactivation of caspase-mediated apoptosis is its ability to
promote X Linked Inhibitor of Apoptosis Protein (XIAP) removal. This is achieved by the release
of SMAC-Diablo and OMI-Htra2 from the mitochondria, which both serve to reduce the
intracellular content of XIAP, thereby allowing the full activation of Caspase-3 and -9. XIAP was
also found to be targeted for degradation by the proteasome in phenoxodiol treated cells,
which also further explains the active removal of intracellular XIAP (Saif et al. 2009). Active
XIAP removal sheds light on the mechanism by which phenoxodiol acts as a chemosensitiser in
refractory ovarian cancer and melanoma as expression of XIAP is linked with chemoresistance
(Kamsteeg et al. 2003; Mahoney 2007).
Phenoxodiol is reported to be a topoisomerase II inhibitor, inhibiting sphingosine kinase
activity, downregulating transcription of angiogenic matrix metalloprotease 2 and other
markers of angiogenesis and inducing apoptosis in chemoresistant ovarian cancer cells by
regulating anti-apoptotic signalling pathways (Constantinou et al. 2003; Kamsteeg et al. 2003;
Aguero et al. 2005; Axanova et al. 2005; Brown et al. 2005; Gamble et al. 2005; Alvero et al.
2006). Phenoxodiol has exhibited the ability to target a subset of NADH oxidases (NOX) that is
thought to be a terminal oxidase primarily expressed on cancer cells (Herst et al. 2007). These
cell surface NADH oxidases form part of a family of ECTO-NOX proteins that play an important
role in red-ox regulation (Yagiz et al. 2007). Phenoxodiol has been reported to inhibit cell
proliferation in a wide range of human cancer cell lines and induce G1 cell cycle arrest as
opposed to the G2-M arrest seen in similar flavanoid compounds like genistein (Alhasan et al.
1999; Aguero et al. 2005). It is postulated that phenoxodiol elicits its global anticancer activity
by modulating the sphingomyelin pathway (De Luca et al. 2005; Moore et al. 2007).
11
Phenoxodiol has been demonstrated to successfully induce cytotoxicity across multiple cancer
cell types, through multiple and varied reported methods of action.
FIGURE 4 PHENOXODIOLS REPORTED METHOD OF ACTION AS ADAPTED FROM SILASI ET AL. 2009
Overexpression of AKT, XIAP and FLIP are linked with cancer progression. XIAP inhibits both
death receptor and mitochondrial-mediated apoptosis, while FLIP predominantly inhibits death
receptor apoptosis. Phenoxodiol initiates a cascade of intracellular events including ceramide
accumulation, caspase 2 activation, and inhibition of sphingosine kinase activity, which results
in reduced intracellularsphingosine-1-phosphate (S1P). Reduced S1P disrupts AKT-p formation
and stabilisation of the AKT protein, thereby promoting AKT, XIAP and FLIP total protein
degradation. Mitochondrial depolarisation results in the release of the serine protease
Omi/HtrA2, which contributes to caspase-mediated cleavage of XIAP and Smac-Diablo which
sequesters and inhibits XIAP. Concomitant with Omi/Hrta2 and Smac-Diablo appearance in the
cytosol, Cytochrome c is also released from the mitochondria thereby causing the activation of
Caspase-9 and subsequent activation of Caspase-3. Due to FLIPs removal, Caspase-8 is also
activated contributing to Caspase-3 activation via the mitochondria. Adapted from (Silasi et al.
2009)
12
2.2.3. SYNERGISTIC ACTIVTY
Synergy with other chemotherapeutic agents has been described using mouse xenograft
models, Alvero and colleagues demonstrated significant synergy of phenoxodiol in
combination with carboplatin (decrease in tumour mass from 6% with carboplatin as
monotherapy to 47% in combination) and paclitaxel (35 – 74%), as well as with gemcitabine
(Alvero et al. 2007; Alvero et al. 2008). Pretreatment with phenoxodiol was found to shift the
IC 50 of each of these drugs downward while co-administration with topotecan also decreased
the IC 50 and have efficacy in 5/9 ovarian cell lines, which were known to be resistant to
topotecan monotherapy (Alvero et al. 2007). By using melanoma microarrays and automated
quantitative analysis technology, (Kluger et al. 2007) demonstrated that pretreatment with
and a novel agent, triphendiol, for the treatment of pancreaticobiliary cancers sensitise
platinum resistant melanoma cells to carboplatin. Moreover, they linked this effect to the
increased cleavage and degradation of XIAP, a direct inhibitor of Caspase-9, -3 and -7 that
leads to decreased initiation and execution of apoptosis and has long been associated with
resistance to chemotherapy in a number of malignancies.
2.2.4. TREATMENT WITH PHENOXODIOL
Phenoxodiol is rapidly absorbed after oral administration and maximum plasma concentration
occurs after 3 hours. In the plasma, phenoxodiol is present almost exclusively (99%) in a
conjugated state with glucuronide and/or sulfate moieties and is highly bound to human
plasma proteins (80 – 95%) with the greatest affinity for albumin. It binds to a lesser extent to
α-1-acid glycoprotein and sex-hormone-binding globulin. The half-life of the drug is 8 – 10 h
and excretion is predominantly urinary in its conjugated form (Silasi et al. 2009). Following
continuous intravenous infusion, plasma concentrations of free plasma phenoxodiol rose
rapidly with an apparent accumulation half-life of 0.17 hours (Howes et al. 2011). Due to the
13
high clearance rate of phenoxodiol it would appear that administration by continuous infusion
or by chronic oral administration may be the optimal modes of administration if it considered
that constant plasma levels are desirable for anti-cancer therapy (Howes et al. 2011).
Following multiple Phase I, II trials it appears that phenoxodiol is a multi-pathway initiator of
apoptosis with broad anti-tumour activity and high specificity for tumour cells and although
clinical trials are still ongoing, phenoxodiol appears to be particularly suited for reversal of
chemo-resistance and its activity is being investigated as a chemo-sensitiser of standard
chemotherapy agents in solid cancers. Phenoxodiol was granted ‘fast track’ status by the US
FDA in its development as a chemosensitiser for platinum and taxane drugs used in the
treatment of recurrent ovarian cancer (Silasi et al. 2009).
2.3.
CARCINOGENESIS / CANCER DEVELOPMENT
Carcinogenesis is the development of cancerous tissue from single cells. It is a multistage
process that assumes cancer is a clonal development from a single cell that has been blocked
from terminal differentiation. Growth signal autonomy, insensitivity to antigrowth signals and
resistance to apoptosis all lead to an uncoupling of the cell’s growth program from signals in
the environment. In 2000, a paper titled The Hallmarks of Cancer condensed and abridged the
common pathways that cancer cells must progress through to develop from benign to
malignant and finally metastatic tumour phenotypes (Hanahan and Weinberg 2000). There
were six potential characteristics cells could acquire to develop into cancerous cells.
1) Evading apoptosis
2) Self-sufficiency in growth signals
3) Insensitivity to anti-growth signals
*Bolded items were investigated in this
thesis.
4) Limitless replicative potential
5) Sustained angiogenesis
14
6) Tissue invasion of metastasis
In 2004 (Trosko et al. 2004) argued the previous six phases were important phenotypic
markers, as well as concepts, but the role of stem cells and cell-cell communication was
equally important in determining cancer development.
7) Pluripotent stem cell differentiation and development
8) Gap junctional intercellular connection
(Trosko et al. 2004) states that cancer cell development must go through initiator, promotion
and finally progression phases. Stem cell development is involved in the initiator phase. The
promotion phase covers potentially reversible or interruptible clonal expansion of the initiated
cell by a combination of growth stimulation and inhibition of apoptosis. When the expanded
initiated cells accrue sufficient mutations and epigenetic alterations to become growth
stimulus independent and resistant to growth inhibitors and apoptosis, to have unlimited
replicative potential and invasive and metastatic phenotypes, then the progression phase has
been achieved. The promotion and progression phase is targeted by conventional cancer
treatment.
15
FIGURE 5: ACQUIRED CAPABILITIES OF CANCER ADAPTED FROM HANAHAN ET AL 2000 AND TROSKO ET
AL 2004
The eight potential characteristics that cancer cells can develop, or develop from, as referred to
in Hanahan et al. 2000 and Trosko et al. 2004.
16
2.3.1. EVASION OF APOPTOSIS
Apoptosis is the programmed physiological cell death of a cell in response to both extrinsic and
intrinsic factors. The ability for prostate cancer cells to avoid apoptotic signalling is considered
fundamental in cancer development. The capacity for a population of cells to expand is not
controlled just by growth rate, but also by attrition rate, and attrition is controlled by
apoptosis (Hanahan and Weinberg 2000). Evasion of apoptosis was one of the first
characteristics of cancer discovered when cells began to no longer typically respond to
chemotherapy treatments as they once did. The evidence from mouse models, cultured cells
and biopsies is suggesting that most, if not all cancers have the ability to evade a majority of
apoptotic signalling. Apoptosis has also been the focus of targeted drug design. The ability to
resensitise tumours to apoptotic signalling is considered critical in modern chemotherapy
design as it would make tumour masses far easier to treat. It would also allow the use of older
drugs, in combination potentially lowering doses needed resulting in reduced side effects,
lower mortality and hopefully decreased recurrence rates (Brown and Wouters 1999; Brown
and Wilson 2003; Brown and Attardi 2005).
2.3.2. SELF-SUFFICIENCY IN GROWTH SIGNALS
Normal cells require mitogenic growth signals to enter into a proliferative state from a
quiescent state, cancer cells exhibit an opposite behaviour with tumour cells generating many
of their own growth signals, reducing their dependence on the stimulation of the normal tissue
environment (Hanahan and Weinberg 2000). Prostate cells have a particular dependence on
the hormone testosterone and are stimulated by the androgen receptor (AR) to proliferate.
The transition to androgen independence is a multifaceted process that involves selection and
outgrowth of cells less dependent on androgen stimulation as well as adaptive upregulation of
genes that help the cancer cells survive and grow after androgen ablation (So et al. 2003).
17
Prostate tumours are composed of various subpopulations of cells that respond differently to
androgen withdrawal therapy and the development of a mutated AR or null AR is characteristic
of late stage prostate cancer where the cells have gained self-sufficiency in growth signals
(Arnold and Isaacs 2002). One signalling pathway that has been implicated in prostate
progression and late stage independence from androgen stimulation is the Wnt family of
proteins. A key pathway of cellular homeostasis and proliferation, aberrant Wnt signalling
activates ⁄ stimulates proteins and respective target genes, which drive prostate cancer
progression. Wnt5a has been implicated in aggressive metastasis and prostate cancer relapse
after prostatectomy and the cell surface frizzled receptors are overexpressed in prostate
cancer. However this overexpression is counterbalanced by the secreted Frizzled Related
Protein family (sFRP), which interact with Frizzled receptors expressed in prostate cancer cells,
and this heterodimer suppresses AR-mediated transactivation (Farooqi et al. 2011).
2.3.3. INSENSITIVITY TO ANTI-GROWTH SIGNALS
In conjunction with insensitivity to apoptosis signalling; prostate cancer cells develop
insensitivity to anti-growth signals. Within a normal tissue, multiple antiproliferative signals
exist to maintain cellular quiescence and tissue homeostasis, including soluble extracellular
growth inhibitory signals and intracellular signals (Hanahan and Weinberg 2000). Anti-growth
signals typically force cells into a quiescent state or further differentiated, post-mitotic cell
type and can affect the expression of integrins and other cell adhesion modelcules, the result
preventing cells from moving past the G1 phase of the cell cycle by affecting Cyclin-D1
expression (Ladha et al. 1998). Avoidance of anti-growth signals is not enough to solely cause
progression of cancer, a reliance on c-Myc or the Wnt/β-catenin pathway in stimulating
mitogenic activity is considered an integral part of the development of insensitivity to antigrowth signals through stimulation of pro-growth pathways (He et al. 1998).
18
2.3.4. LIMITLESS REPLICATIVE POTENTIAL
Even if cells have developed self-sufficiency in growth signals and an ability to evade apoptosis
they are still limited in mitogenic capability by normal cell replication machinery. The Hayflick
limit (or Hayflick Phenomenon) is the number of times a normal cell population will divide
before it stops, presumably because the telomeres shorten to a critical length. Mammalian
cells in a cell culture divide between 40 and 60 times then enter a senescence phase with each
mitotic division effectively shortening the telomeres on the DNA of the cell (Hayflick and
Moorhead 1961). Telomere shortening in humans eventually makes cell division impossible,
and it is presumed to correlate. Maintenance of the length of the telomeric region appears to
prevent genomic instability and the development of cancer. Upon activation of mitogenic
signalling, cells commit to entry into a series of regulated steps allowing completion of the cell
cycle. Cells begin in G1 phase, the time between M and S phases, and before entry into S
phase, where DNA is replicated, must pass through a restriction point (Pardee 1974) that
analyses and attempts to repair DNA damage. After S phase, cells enter G2 phase (the time
between the S and M phases) where cells can repair errors that occurred during DNA
duplication, preventing the propagation of these errors to daughter cells. Finally, the
separation into two daughter cells by chromatid separation occurs and is called M phase
(Senderowicz 2004). The sequence of events in cell cycle progression is highly orchestrated and
depends on the cyclic activation and inactivation of cyclin dependent kinases (CDK), which
govern the progression of the cells from one phase to another. In the event of tumourigenesis,
constitutive mitogenic signalling as well as mutations in tumour suppressor genes and protooncogenes leads to cell cycle deregulation and uncontrolled proliferation (MacLachlan et al.
1995).
19
2.3.5. SUSTAINED ANGIOGENESIS
The growth of new blood vessels, angiogenesis, is critical in the development of the cancer
mass (Berry and Eisenberger 2005; Alvero et al. 2008). Solid tumour masses require large
amounts of energy to continue to replicate and grow at such rapid rates, the oxygen and
nutrients consumed by these cells is crucial and results in nearly all cells residing within 100µm
of a capillary blood vessel (Hanahan and Weinberg 2000). The development of blood vessels in
normal tissue and organs occurs at organogenesis in the foetal life, in comparison angiogenesis
occurs whenever irregular vascularisation signalling occurs, called the angiogenic switch
(Carmeliet and Jain 2011). Tumour-related angiogenesis supports tumour growth and is also a
major obstacle for successful immune therapy as it prevents migration of immune effector
cells into established tumour parenchyma (Hamzah et al. 2008). Thus angiogenesis can protect
the cancer mass from the hosts immune system while providing it with energy and oxygen. The
coordinated expression of pro- and antiangiogenic signalling molecules, and their modulation
by proteolysis, appear to reflect the complex homeostatic regulation of normal tissue
angiogenesis and of vascular integrity (Avraamides et al. 2008).
2.3.6. TISSUE INVASION AND METASTASIS
Tissue invasion and metastasis of cancer relies upon cells having developed the ability to evade
apoptosis as well as ignore the lack of gap junctional responses. Anoikis, the ability of cells to
evade apoptosis when shed from the primary tumour mass, is a key component in metastatic
invasion of tissue (Langley and Fidler 2011). Invasion relies on the angiogenic ability of the cells
and is considered one of the last Approximately one in eight prostate cancer cases
metastasises widely, typically to bone, adjacent soft tissue, liver, and lung and is consistent
with poor prognosis and outcome (Kleeberger et al. 2007). Development of a metastatic
phenotype is the result of an accumulation of many gene dysregulations and is considered a
20
key step in the progression from benign to metastatic and malignant tumour phenotypes
(Hanahan and Weinberg 2000).
2.3.7. PLURIPOTENT
STEM
CELL
DIFFERENTIATION
AND
DEVELOPMENT
The cancer stem cell hypothesis has two central tenets—tumours are derived from
transformation of normal stem cells or their progeny (i.e., progenitor or even differentiated
cells) and every tumour contains a small population of stem-like cells that possess a unique
ability to drive tumour formation and maintain tumour homeostasis (Reya et al. 2001). Current
theory is that stem cells are immortal and only become mortal once they enter into a terminal
differentiation cycle, as seen in stem cell derived tumours that can be serially initiated in mice
including solid tumours (Li et al. 2008).
Thus, stem cells are a self-renewing, immortalised group of cells that have been determined to
exist in nearly all organs and tissues (Reya et al. 2001). Stem cell populations normally make up
to 0.01% of a total cell population, however in cancerous tissues stem cells have been found
from 0.1-0.2% of total cells, an increase of ten to twenty fold. This increase in self renewing
cells that are not easily targeted by conventional chemotherapy, due to their slow cell cycling,
receptor complexes and adaptive ability is one theory used to describe the ability of cancer to
become refractive over time to initially successful treatment (Li et al. 2008).
21
FIGURE 6: STEM CELL THEORY ADAPTED FROM TROSKO ET AL 2004
Showing the characteristics of stem cell theory, Figure 6 demonstrates how slow cycling stem
cells regenerate and differentiate normally. Following initiation stem cells can still differentiate
normally, however clonal expansion and promotion of initiated cells can result in progression
phase, where cancer cells develop malignancy.
2.3.8. GAP JUNCTIONAL INTERCELLULAR CONNECTIONS
Gap junctional intercellular connections (GIJC) are involved in carcinogenesis in various ways.
Most, if not all, tumour cells lack functional GJIC (Trosko et al. 2004). Cancer cells lack the
expression of connexion genes due to the activation of various oncogenes or the loss of
specific tumour suppressor genes, resulting in a cell less affected by the extracellular
environment and less likely to differentiate. Most, if not all, tumour promoting chemicals and
conditions reversibly inhibit GJIC and oncogenes, such as Src, Ras, Raf, have been reported to
down regulate GJIC. Antisense connexin genes transfected in normal cells induces a
tumourigenic phenotype while connexin genes transfected into tumour cells restore growth
control and reduce the tumourigenicity of the cells (King et al. 2000).
22
2.4.
THE PROSTATE
The prostate is a walnut sized gland of the male accessory reproductive system that surrounds
the urethra and ejaculatory duct directly inferior to the urinary bladder (Saladin 2007). It
measures 2x4x3 cm and is an aggregate of 30-50 compound tubuloacinar glands enclosed in a
single fibrous capsule, these glands empty via ~20 pores into the urethral wall. The position of
the prostate immediately anterior to the rectum allows for it to be palpated through the rectal
wall for lumps suggestive of prostate cancer (Saladin 2007). The prostate is approximately ½
glandular, ¼ involuntary muscle, ¼ fibrous tissue and the structure has a dense fibrous capsule
of the prostate that incorporates the prostatic plexuses of nerves and veins (Basmajian and
Slonecker 1989; Moore et al. 2007). The prostate is surrounded by the visceral layer of the
pelvic fascia, forming a fibrous prostatic sheath that is thin anteriorly, continuous
anterolaterally with the puboprostatic ligaments, and dense posteriorly, continuous with the
rectovesical septum, surgically called the fascia of Denonvilliers which separates the prostate
and urinary bladder from the rectum (Basmajian and Slonecker 1989; Moore et al. 2007).
The prostate is a modified portion of the urethral with the glands organised in 3 concentric
groups; mucosal are short and simple and open all around the urethra while the intermediate
or submucosal glands open into the prostatic sinus. The outermost or main glands are long and
branching and envelope the other groups except in front where those of opposite sides are
joined by a nonglandular isthmus and nearly all hypertrophies of the prostate arise from these
mucusol and submucosal glands (Basmajian and Slonecker 1989).
The prostate arterial blood supply is predominantly from branches of the internal iliac artery
including the inferior vesical arteries, internal pudendal and middle rectal arteries (Moore et
al. 2007). The prostatic venous drainage is complicated with the deep dorsal vein of the penis
23
draining the front plexus, the vesical plexus draining superiorly and posteriorly draining into
the internal iliac veins while also potentially anastamosing with the vertebra plexus of veins
and all of this occurring between the fibrous capsule of the prostate and the prostatic sheath
(Basmajian and Slonecker 1989). Due to this varied blood drainage prostate cancer frequently
spreads to the central nervous system (CNS), lower lumbar vertebrae, pelvic bones and femora
(Basmajian and Slonecker 1989). The nervous innervations of the prostate are autonomic, with
parasympathetic efferent fibres arising from pelvic splanchnic nerves (sacral nerves 2, 3 and 4)
while sympathetic efferent fibres come from sacral splanchnic nerves (arising from
sympathetic trunk) and together they help form the inferior hypogastric plexus which supplies
the prostate gland. The lymphatic drainage of the prostate is chiefly into the internal iliac
nodes but some pass to the sacral lymph nodes with implications for metastasising prostate
cancers ability to spread into spinal cord and CNS (Moore et al. 2007).
The prostate gland is histologically divided into three major regions – the peripheral, central
and transitional zones. While most cancers develop in the peripheral zone (68%), fewer
originate in the transitional (24%) or central (8%) zone (McNeal et al. 1988). Each zone consists
of ducts and acini lined by an epithelial sheet. The epithelium consists of a bi-layer of basal
cells beneath the secretory, luminal cells and is interspersed with neuroendocrine cells. The
majority of the basal cells are androgen receptor (AR) negative and consist of self-renewing
stem cells that differentiate into transit amplifying cells (also AR negative) with limited
proliferative capacity. These transit amplifying cells clonally expand, differentiate and migrate
from the basal to the luminal layer where they differentiate to form mature, secretory luminal
cells that are nonproliferative and AR-positive (Litvinov et al. 2003; Uzgare et al. 2004). A
minority of basal stem cells differentiate into neuroendocrine cells that are AR-negative and
secrete specific peptides (Arnold and Isaacs 2002).
24
2.4.1. TESTING FOR PROSTATE CANCER
Common testing for prostate cancer involves the digital rectal exam (DRE), detection of benign
prostatic hyperplasia (BPH) and admission of the prostatic specific antigen (PSA) blood exam.
The position of the prostate surrounding the urethra means that, during conditions of benign
prostatic hyperplasia, a transurethral resection of the prostate (TURP) will be performed and
the tissue analysed for potential cancer pathology. The position of the prostate anterior to the
rectum allows for a trans-rectal ultrasound biopsy (TRUS) to be performed when high PSA
scores indicate a potential for the tissue to become cancerous, the TRUS technique allows for
ultrasound to guide the biopsy needle to the abnormal looking region (Park et al. 2012).
Following biopsy using the TURP or TRUS methods cells are stained and compared to the
Gleason scale, which compares how differentiated cells are, and gives a 1-5 grade (Gleason
1977). Grades between two separate regions are added and result in a Gleason score (also
known as sum or pattern) which results in a prognosis with a minimum score or 2 and a
maximum or 10, the higher the number the higher the chance the cancer has reached
malignant or metastatic phenotypes. The move from benign to metastatic prostate cancer is
complicated but is based around the development of the androgen independent,
undifferentiated and subsequent invasive phenotypes (Feldman and Feldman 2001; So et al.
2003).
2.4.2. PHENOTYPIC PROGRESSION OF PROSTATE CANCER
Prostate cancer progresses from a benign to malignant to metastatic phenotype in conjunction
with a loss of androgen receptor function and a gain of invasive potential (So et al. 2003). The
prostatic epithelium is surrounded by a fibromuscular stroma that contains AR-positive
smooth muscle cells and fibroblasts amongst other cell types. Androgen occupancy of nuclear
25
AR in these stromal cells results in their production and secretion of paracrine growth and
survival factors for the epithelium (Litvinov et al. 2003).
Transformation to a malignant phenotype involves a shift from this paracrine axis to a situation
in which AR signalling directly activates the production of autocrine growth and survival factors
by the prostate cancer cells themselves as well as regulates the production of secretory
proteins by these malignant cells. Thus androgen acts in an autocrine manner as the major
regulator of proliferation and survival in such androgen responsive prostate cancer cells
(Litvinov et al. 2003). It is for this reason that androgen blockade via administration of
leutenizing hormone releasing hormone (LHRH) analogues results in inhibition of tumour
growth and a positive clinical response. Cellular adaptation to low levels of circulating
androgen along with clonal expansion, however, results in the emergence of hormone
refractory tumours with or without metastases (Gao et al. 2001; Isaacs and Isaacs 2004).
Understanding the mechanisms for development of hormone refractory cancer (i.e.
dysregulation of the signalling network) requires knowledge of the cellular organisation and
the cell of origin for the cancer. Prostate cancer is very common in developed countries and is
widely variable in clinical course. Most cases remain confined to the prostate and adjacent soft
tissue and cause no harm. However, approximately one in eight cases metastasises widely,
typically to bone, adjacent soft tissue, liver, and lung (Kleeberger et al. 2007).
FIGURE 7 CHARACTERISTIC PROGRESSION OF PROSTATE CANCER FROM EARLY TO LATE STAGE
26
2.4.3. TREATMENT OF PROSTATE CANCER
Age and hormone are two known factors influencing the incidence of prostate cancer and,
because of that, cancer cells initially respond to androgen withdrawal by undergoing apoptosis
among the hormone-dependent population. However, patients with advanced or metastatic
prostate cancer develop hormone refractory status that becomes fatal because of the growth
of androgen-independent tumour cells and the emergence of tumour clones. Therefore, the
potential cancer chemotherapy to cause apoptosis in metastatic prostate cancer is necessary
and urgent for the clinical treatment (Hotte and Saad 2010; Parente et al. 2012).
Patients with localised disease are candidates for ionizing radiation therapy. Patients with
metastatic carcinoma of the prostate are usually treated with androgen ablation therapy. After
further disease progression, they may be treated with chemotherapy including docetaxel and
doxorubicin (Sweat 2005). Ionizing radiation and doxorubicin are DNA-damaging agents that
induce double-strand breaks in DNA. This leads to the activation ofDNA damage checkpoints.
One of the key proteins in these pathways is the tumour suppressor p53, which triggers cell
cycle arrest and induces the repair of DNA damage, responses needed for cell survival, or
alternatively, apoptosis (Devlin et al. 2008).
The interaction between testosterone and the androgen receptor (AR) is essential for prostate
development. Because AR signalling has also been shown to play a key role in prostate
carcinogenesis, androgen ablation therapy (AAT) is a commonly used form of treatment,
particularly for advanced disease. While AAT leads to significant levels of prostate cancer cell
27
apoptosis, the effect is short-lived and ultimately not curative as most patients develop
androgen-independent disease (D'Antonio et al. 2008).
2.4.4. CHEMOTHERAPY AND PROSTATE CANCER
Currently, only patients who have detectable macroscopic metastatic disease should receive
systemic chemotherapy outside of a clinical trial. As any treatment for advanced disease
remains palliative, patients with advanced prostate cancer are encouraged to participate in
clinical trials (Hotte and Saad 2010). Combined docetaxel (a taxane drug that induces
polymerisation of microtubules and phosphorylation of the Bcl-2 protein) and prednisone is
currently considered the standard of care for men with detectable metastatic disease, based
largely on the simultaneous publication of two large randomised controlled trials comparing
this combination with the previously established standard of mitoxantrone and prednisone
(Parente et al. 2012). (Petrylak et al. 2004; Tannock et al. 2004) simultaneously published
combined trials of docetaxel and prednisone. Men in both trials had clinical evidence of
metastases with or without symptoms and had undergone anti-androgen withdrawal
response. Overall survival was the primary endpoint in both trials. (Tannock et al. 2004)
reported improved survival with docetaxel (every-3-weeks dosing) compared with
mitoxantrone–prednisone, median survival: 18.9 months versus 16.5 months(Petrylak et al.
2004) reported longer survival time with docetaxel–estramustine combination chemotherapy
as compared with mitoxantrone, median survival: 17.5 months versus 15.6 months. Late stage
prostate cancer treatment still remains an attempt at extension of life and is not yet
considered curative (Hotte and Saad 2010).
28
FIGURE 8 DEVELOPMENT OF RESISTANCE TO CHEMOTHERAPEUTIC TREATMENT ADAPTED FROM
JOHNSTONE, RUEFLI AND LOWE 2002
Addition of chemotherapeutic drugs to tumour cells results in interaction between the drug and
intracellular targets, and induction of the primary effect. Classical drug resistance proteins such
as drug efflux pumps can inhibit the primary effect by preventing drug-target interactions.
Depending on the severity of the initial insult, drug-induced damage may result in cytotoxic cell
death or initiate a series of secondary effects mediated by various stress signalling pathways
leading to cell death or cell cycle arrest. Consequently, mutations in these downstream events
can produce multidrug resistance. Adapted from (Johnstone et al. 2002).
2.5.
APOPTOSIS
In classic apoptosis cellular membranes are disrupted, the cytoplasmic and nuclear skeletons
broken down and nuclear material fragmented and within the space of 30 to 120 minutes the
cell is destroyed and engulfed by nearby cells with no inflammatory response (Wyllie et al.
29
1980). When first described in 1972 (Kerr et al. 1972) raised the possibility of apoptosis being a
barrier to cancer after discovering massive apoptosis in populations of rapidly growing
hormone dependent tumour cells following hormone withdrawal. Resistance to apoptosis has
been investigated intensely for the last 20 years with some characteristic pathway mutations
becoming recognised in many different forms of cancer.
The loss of a functional p53 oncogene was described in 1996 (Harris 1996) and is discovered in
well over 50% of all cancers. This causes the removal of one of the key DNA damage sensing
apparatus in the cell and causes significant inability to induce apoptosis in response to DNA
damage, allowing more errors to accumulate rapidly. Abnormalities, including hypoxia and
oncogene hyper-expression, are also funnelled in part via p53 to the apoptotic machinery;
these too are impaired at eliciting apoptotic activity when p53 function is lost (Levine 1997).
Discovery of the anti-apoptotic Bcl-2 gene (Korsmeyer 1992) and later its family of related
signalling molecules further significantly enhanced the knowledge of how cancer cells develop
resistance to apoptotic stimulation.
The ability for cancer cells to ignore anoikis, the process of apoptosis that cells undergo when
shed off their central mass, leads to eventual implantation and metastasis. The process of
evading apoptosis is integral to the assist the development of secondary, metastatic tumours
and is important in the removing the reliance of GJIC to maintain cell integrity. Researchers
became aware that apoptosis, a form of cell death, played a crucial role in a myriad of
physiological and pathological processes (Kerr et al. 1972) Apoptosis is often referred to as a
physiological cell suicide program that is critical for the development and maintenance of
healthy tissues (Deveraux et al. 1999). The mechanisms for apoptosis have been strongly
conserved during evolution and dysregulation of cell death pathways occur in cancer,
30
autoimmune and immunodeficiency diseases and in neurodegenerative (Deveraux et al. 1999).
The process of apoptosis is necessary to remove unwanted cells from multicellular organisms.
It is the mechanism by which specific groups of cells are removed during embryogenesis so
that a particular course of development may be followed. The commitment to apoptosis
involves both signalling and effector stages. Though they vary between cells, the array of
apoptosis induction signals trigger signalling pathways that coalesce, often at the
mitochondria, to activate central effector pathways involving a series of proenzymes, the
caspases (Maguire et al. 2000). When activated caspases are efficient at degrading cellular
processes, and DNA, to the point where normal physiological activity is impossible, resulting in
cellular response and apoptotic phenomena (Hengartner 2000).
2.5.1. MORPHOLOGY OF APOPTOTIC CELLS
Apoptosis occurring in cells is characterised by common distinctive morphological and
molecular features (Kerr et al. 1972). Apoptotic cells will undergo a regulated autodigestion,
which involves the disruption of cytoskeletal integrity, cell shrinkage, nuclear condensation,
activation of endonucleases, blebbing of the cell surface, chromatin condensation and the
formation of apoptotic bodies distinct to apoptosis (Maguire et al. 2000; Birkey Reffey et al.
2001). The result is contraction of the cytoplasm, accompanied by condensation of nuclear
chromatin into several large masses. Caspase mediated events are likely to contribute to the
characteristic morphological changes that result in DNA cleavage at the internucleosomal
region into 185 base pair multiples (Hengartner 2000). After fragmentation, DNA exhibits a
typical condensation at the nuclear margin. The cell forms membrane bound apoptotic bodies
of various sizes, containing organelles and nuclear fragments. These apoptotic bodies provide
a strong stimulus for phagocytosis and are subsequently consumed by their viable
neighbouring cells or specialist phagocytes (Schwartzman and Cidlowski 1993).
31
The process of apoptosis is a highly conserved mechanism in multicellular organisms which
allows for the removal of unwanted cells without an inflammatory response (Deveraux et al.
1999; Hengartner 2000). Once the apoptotic process is triggered cells undergo a series of
molecular and morphologic changes that characterise classical apoptosis. These processes
include; irreversible chromosome condensation (pyknosis 1 ) with nuclear fragmentation
(karyorrhexis2) via inter-nucleosome DNA cleavage, cell shrinkage, formation of multiple
membrane blebs (zeiosis3) and finally the breakdown of the cell into apoptotic bodies
containing organelles and pieces of the degraded nucleus that are phagocytosed by
surrounding cells (van Heerde et al. 2000). Recently, necrosis, once thought of as simply a
passive, unorganised way to die, has emerged as an alternate form of programmed cell death
whose activation might have important biological consequences, including the induction of an
inflammatory response. Autophagy has also been suggested as a possible mechanism for nonapoptotic death despite evidence from many species that autophagy represents a survival
strategy in times of stress. Recent advances have helped to define the function of and
mechanism for programmed necrosis and the role of autophagy in cell survival and suicide
(Edinger and Thompson 2004).
Other apoptotic modifications include early membrane changes such as externalisation of
membrane bound phosphatidylserine which promotes the phagocytosis of the apoptotic
bodies without inducing an inflammatory response (Martin et al. 1995). This early
externalisation is a key step in detecting cells entering early stage apoptosis and several
distinct apoptotic events have been identified on the molecular signalling level following
phosphatidylserine externalisation including; the signalling stimulus for degradation of DNA
1
Pyknosis or karyopyknosis, is the irreversible condensation of chromatin in the nucleus of a cell
undergoing necrosis or apoptosis. It is followed by karyorrhexis
2
Karyorrhexis The destructive fragmentation of the nucleus of a dying cell whereby its chromatin is
distributed irregularly throughout the cytoplasm
3
Zeiosis is the term used to describe the formation of blebs in a cell. A bleb is an irregular bulge in the
plasma membrane of a cell caused by localised decoupling of the cytoskeleton from the plasma
membrane.
32
into fragments of multiples of ~200 base pairs, the proteolytic cleavage of poly(ADP-ribose)
polymerase (PARP) and cytoskeleton components separation (van Heerde et al. 2000).
2.5.2. APOPTOTIC SIGNALLING
Apoptosis is regulated at many levels, including the initiation, transduction, amplification and
execution stages and the mutations that disrupt each stage have been detected in tumour
cells. Because mutations in cancers necessarily produce a selective advantage to emerging
tumour cells, the identification of mutated components and their frequency of mutation
highlight critical regulatory points in survival and proliferative processes. The fact that
apoptosis is disabled at distant stages in different tumour types suggest that its critical control
points are context dependent. Identification of these control points singles out distinct “sites
of attack” for targeting by novel chemotherapeutic drugs (Johnstone et al. 2002).
On the molecular level, the cell death program can be divided into three parts: initiation,
execution, and termination of apoptosis. Apoptosis is initiated by a variety of stimuli, including
growth factor withdrawal (“death by neglect”), UV or -irradiation, chemotherapeutic drugs,
and death receptor signals. In most cases the execution phase is characterised by membrane
inversion and exposure of phosphatidylserine, blebbing (zeiosis), fragmentation of the nucleus,
chromatin condensation, and DNA degradation. In the termination phase, “apoptotic bodies”
are engulfed by phagocytes resulting in no inflammatory response (Hengartner 2000).
While mutations in cancer cells often target regulators of the intrinsic apoptotic pathway
(indirect) such as p53 and the Bcl-2 related proteins, alterations that disrupt apoptosis
downstream of the mitochondria have been reported (Petronilli et al. 2001). Silencing of Apaf1 occurs in metastatic melanoma and over expression of IAPs and heat shock proteins (HSP),
which can inhibit Caspase-9 activation, is commonly observed in human tumours (Deveraux et
al. 1999). Postmitochondrial mutations appear less frequently than those targeting upstream
33
components of the apoptotic program, though this could represent greater redundancy in the
downstream pathway or difficulty maintaining cell viability following damage to the
mitochondria. Consistent with this, apoptosis can be induced without the activation of
caspases or while the caspases themselves are prevented from action by a pan caspase
inhibitor such as Z-VAD-FMK (Johnstone et al. 2002).
2.5.3. EXTRINSIC PATHWAYS
To date, it is accepted that every cell type harbors the machinery to commit suicide by
apoptosis. Subsequently, it is acknowledged that apoptosis plays a crucial role in homeostasis
and pathology and the molecular biology and biochemistry of apoptotic death machinery are
far from being completely resolved (van Heerde et al. 2000). Despite apoptosis exhibiting great
diversity in signalling pathways, three functionally distinct phases of apoptosis, common to all
cell types can be distinguished. First, the initiation phase can be induced by death inducing
signals such as Fas ligand (Fas-L) and tumour necrosis factor α (TNF-α), a lack of growth and/or
survival signals, or DNA damage which may devise the cell to prepare for suicide, commonly
these pathways involve the Caspase signalling system (Susin et al. 2000; Cregan et al. 2004;
Dohi et al. 2004; Svingen et al. 2004). The initiation phase will result in the activation of the
second more general, decision phase, in which the cell is still able to make the decision to live.
This phase is characterised, in most cases, by involvement of the mitochondrion. This organelle
provides molecular links between the upstream initiation phase and downstream execution
phase, by releasing apoptosis inducing factor (AIF), Cytochrome c, Caspases-2,-3 and -9. When
the cell is committed to die, and thus the point of no return has been passed, the third phase
the execution phase is activated (Wang et al. 1999). This phase is characterised by the
activation of the downstream or effector caspases which subsequently orchestrate a sequence
of events by their hierarchical activation. These events include the loss of cell junctions, cell
34
shrinkage, chromatin condensation and margination, nuclear pyknosis and fragmentation,
membrane blebbing and disassembly of the cell into membrane-enclosed vesicles called
apoptotic bodies. Several events have been identified on the biochemical level, including the
degradation of DNA into fragments of multiples of ~200 base pairs, the proteolytic cleavage of
poly(ADP-ribose) polymerase (PARP) and cytoskeleton components, and the cell surface
exposure of phosphatidylserine (PS) (van Heerde et al. 2000).
2.5.4. INTRINSIC PATHWAYS
Disruption of the intrinsic pathway is common in cancer cells with the p53 tumour suppressor
gene the most frequently mutated gene in human tumours, and loss of p53 function can both
disable apoptosis and accelerate tumour development in transgenic mice (Ryan et al. 2001).
Function mutations or altered expression of p53 downstream effectors (PTEN, Bax, Bak, and
Apaf-1), or upstream regulators (ATM, Chk2, Mdm2, and p19ARF), occur in human tumours
and as a result, the presence of wild-type p53 does not necessarily indicate that the pathway is
intact and therefore correlating p53 gene integrity with a functional p53 pathway is not always
correct (Schmitt et al. 1998). The B-cell lymphoma 2 (Bcl-2) family members are key regulators
of the intrinsic apoptotic pathway and consist of both pro- and anti-apoptosis members who
expression is commonly altered in tumours (Reed 2000). Mutations or altered expression of
upstream regulators of Bcl-2 proteins are associated with cancer. For example, the Bad-kinase
intrinsic Akt, is positively regulated by various oncoproteins, and negatively regulated by the
PTEN tumour suppressor, amplified Akt and mutated PTEN, have been found with high
frequency in a variety of solid cancers, indicating the importance of intrinsic cell signalling in
regulating tumourigenisis (Datta and Datta 1999).
35
2.5.5. CASPASES
Caspases (Cysteine Aspartases) are the key effector proteases of apoptosis, existing in healthy
cells as inactive precursor molecules (zymogens) called procaspases (Silke et al. 2001;
Donepudi and Grutter 2002). Three major apoptotic pathways have been identified: one
activated by death receptor activation and the other by intrinsic stress induced stimuli and
finally a endoplasmic reticulum specific stress pathway has been determined (Nakagawa et al.
2000; MacFarlane et al. 2002). Mechanisms described for activating caspases include;
noncovalent association with caspase activating proteins, such as Apaf-1 or FAS-L, leading to
autocatalytic cleavage of the procaspase polypeptide at specific aspartic acid residues, or
cleavage at specific aspartic acid residues within the inactive zymogens by other activated
caspases (Zou et al. 2003). The caspases are highly conserved throughout the animal kingdom
(Deveraux et al. 1997), and many function to initiate Cytochrome c release from the
mitochondria.
When added to cytosol environment Cytochrome c initiates an apoptotic program which
involves proteolytic processing and activation of more caspases, and apoptotic-like destruction
of exogenously added nuclei (Deveraux et al. 1997). Caspase functions are subject to
modulation by a family of inhibitor of apoptosis proteins (IAP’s) (Deveraux et al. 1999). Effector
caspases, such as Caspase-3 and Caspase-7 are activated by initiator caspases, such as
Caspase-9 and Caspase-8, through proteolytic cleavage. Once activated, the effector caspases
are responsible for the proteolytic cleavage of a diverse array of structural and regulatory
proteins, resulting in the apoptotic phenotype (Suzuki et al. 2001). The initiator caspases
Caspase-9 and Caspase-8 converge on the activation of the executioner caspases, Caspase-3
and Caspase-7. The critical role of those caspases in transmitting cell death signals is
underscored by the results of many phenotypic gene ablation experiments that vary between
embryonic lethality (Caspases-7 and -8) to perinatal lethality (Caspases-3 and -9) (Riedl and
36
Salvesen 2007). Caspases-3 and -7 are considered to be the major effector caspases within
cells, activated by the initiator caspase cascade, they can feed back to alter upstream caspases
such as Caspase-8 and Caspase-9 allowing some degree of self regulation (Silke et al. 2001).
FIGURE 9 SIMPLE APOPTOTIC SIGNALLING ADAPTED FROM RIEDL AND SALVESEN 2007
The pathways in evidence are the caspase-dependant apoptotic signalling resulting from
extrinsic (DISC) or intrinsic (apoptosome) stimuli. The proteolytic caspase cascade in the
initiation and execution of apoptosis. Both the intrinsic and extrinsic pathways use related
principles in sensing an apoptotic signal and executing apoptosis. In the intrinsic pathway, an
apoptotic stimulus in the cell leads to the assembly of a signalling platform, the apoptosome,
which activates the initiator Caspase-9. An extrinsic apoptotic signal, by contrast, is mediated
by binding of an extracellular ligand to a transmembrane receptor, leading to the formation of
the death-inducing signalling complex (DISC), which is capable of activating the initiator
Caspase-8. Once activated, either Caspase-9 or Caspase-8 cleaves executioner Caspase-3 and
Caspase-7, which represents the execution level of the caspase cascade and leads to apoptosis
of the doomed cell. Negative regulators of the caspase cascade can be found on both levels.
37
Whereas FLIP blocks the activation of the initiator Caspase-8 in the DISC, XIAP can block both
the initiation phase, by inhibiting Caspase-9, and the execution phase, by blocking Caspase-3
and Caspase-7, of the cascade. Adapted from (Riedl and Salvesen 2007).
2.5.6. APOPTOSOME
Upon intrinsic stimulation of apoptosis, the release of Bax and Cytochrome c from the
mitochondria initiates the apoptosis promoting activity of Cytochrome c by allowing
interaction with the Apaf-1 protein. Binding of Cytochrome c to Apaf-1 causes recruitment of
Caspase-9, forming the apoptosome (Slee et al. 1999; Slee et al. 1999; Srinivasula et al. 2001).
The apoptosome is a large molecule that interacts with the executioner caspases, Caspase-3
and Caspase-7, which initiate DNA fragmentation and cause the cell to enter into apoptosis
(Boatright and Salvesen 2003). Caspase-3 can be activated via extrinsic pathways that produce
activated Caspase-9 through Caspase-8 signalling. If intrinsic pathways are activated, the
release of Cytochrome c and apoptosome formation results in Caspase-3 activation (Riedl and
Salvesen 2007).
2.5.7. BCL-2 FAMILY
The mitochondrial pathway of cell death is mediated by Bcl-2 family proteins, a group of antiapoptotic and pro-apoptotic proteins that regulate the passage of small molecules, such as
Cytochrome c, Smac/Diablo, and Apoptosis Inducing Factor (AIF), which activates caspase
cascades, through the mitochondrial transition pore. The activation of caspases is
counteracted by anti-apoptotic molecules of the Bcl-2 family (Bcl-2, Bcl-xL), because these Bcl2 family proteins heterodimerise with proapoptotic members of the Bcl-2 family (Bax, Bak) and
interfere with release of cytochrome c by pore-forming proteins (Bid, Bik) (Gross et al. 1999).
38
B-cell lymphoma-extra large (Bcl-xL) is a transmembrane molecule in the mitochondria. It is
involved in the signal transduction pathway of the FAS-L and is one of several anti-apoptotic
proteins which are members of the Bcl-2 family of proteins. It has been implicated in the
survival of cancer cells with Bcl-xL expression in prostate cancer leading to cell proliferation,
apoptosis inhibition and obvious decreased radiosensitivity in human prostatic carcinoma cells
(Wang et al. 2008). In the absence of activated Akt, Bcl-2 family member Bad forms
heterodimers with Bcl-xL, an antiapoptotic protein that prevents the release of Cytochrome c
from mitochondria (Gross et al. 1999). This complex formation abrogates the antiapoptotic
function of Bcl-xL thereby facilitating apoptotic death via a cytochrome c-dependent pathway.
As a consequence, the dynamic interaction between Bcl-xL and Bad represents a critical
determinant of cell fate downstream of the phosphatidylinositol-3 kinase (PI3K)/Akt cascade,
and may represent an alternative mechanism for cancer cells to evade apoptosis (Cheng et al.
2001; Lee et al. 2008).
Bax is an important multidomain pro-apoptotic proteins essential for the mitochondrial release
of Cytochrome c. The expression of Bax is up regulated by the tumour suppressor protein p53,
and Bax has been shown to be involved in p53-mediated apoptosis (Goodman et al. 1998). In
normal mammalian cells, the majority of Bax is found in the cytosol, but upon initiation of
apoptotic signalling, Bax undergoes a conformational shift, and inserts into organelle
membranes, primarily the outer mitochondrial membrane resulting in Cytochrome c release
(Lee et al. 2008).
2.5.8. X-LINKED INHIBITOR OF APOPTOSIS PROTEIN
X-linked inhibitor of apoptosis protein (xIAP) is part of a family of endogenous caspase
inhibitors that, when up regulated, have been implicated in the early stage development of
39
prostate and breast epithelial carcinoma (Parton et al. 2002; Krajewska et al. 2003). In
conjunction with anti-apoptotic Bcl-2 members (Bcl-2, Bcl-xL) IAP family members exert
significant anti-apoptotic effects through; inhibition of Caspase-3 and -7 activation and
function, inactivation of pro-apoptotic Bcl-2 family members and involvement in signalling
pathways NF-κB and JNK (Levkau et al. 2001; Sanna et al. 2002; Hu et al. 2003; Dubrez-Daloz et
al. 2008). XIAP expression levels are also found to be increased in epithelial ovarian and
prostate cancers, and have been linked to chemoresistance and progression of cells to
metastasis (Krajewska et al. 2003; Berezovskaya et al. 2005; Mahoney 2007).
As opposed to other members of the IAP family, XIAP is the only member capable of direct
inhibition of both the execution and initiation phases of the apoptotic cascade; it inhibits
apoptotic activation by inhibiting caspase-9, and inhibits the execution phase by inhibiting the
effector caspases, Caspase-3 and Caspase-7 (Kluger et al. 2007). As XIAP selectively binds and
inhibits Caspase-3, -7 and inhibits Apaf-1-Cytochrome c-mediated activation of Caspase-9 (the
apoptosome) it is responsible for directly inhibiting much of the apoptotic pathways and is a
considered a key factor in chemoresistance in tumours (Mahoney 2007).
2.5.9. APOPTOSIS INDUCING FACTOR
Apoptosis Inducing Factor (AIF) is one of the apoptogenic molecules released from the
mitochondria upon multiple stimuli with overexpression of AIF causing apoptosis independent
of caspases (Xie et al. 2005). AIF has also been shown to act as a free radical scavenger and
play a role in normal mitochondrial oxidative phosphorylation. Upon stimulus by some death
signals AIF translocates to the nucleus when apoptosis is induced (Susin et al. 1999).
Experiments using recombinant AIF showed that it induced chromatin condensation in isolated
nuclei and large-scale fragmentation of DNA, it induced purified mitochondria to release the
40
apoptogenic proteins Cytochrome c and Caspase-9 and microinjection of AIF into the
cytoplasm of intact cells induces condensation of chromatin, dissipation of the mitochondrial
transmembrane potential, and exposure of phosphatidylserine in the plasma. None of these
effects is prevented by the wide-ranging caspase inhibitor known as Z-VAD-FMK indicating that
AIF is a mitochondrial effector of apoptotic cell death independent of caspase activity (Susin et
al. 1999).
2.5.10.
REACTIVE OXYGEN SPECIES
Chemotherapy and radiation treatments are known to induce oxidative stress in cancer cells,
rapid accumulation of highly reactive molecules, such as, nitric oxide occurs and results in
damage to cell structures and potentially activates the apoptotic cascade. Oxidative stress
induces the expression of antioxidant genes that contain an antioxidant response element in
their promoters; this can result in a triggering of apoptosis or necrosis of various cell types and,
in several instances, can inhibit cell growth and interfere with the cell cycle (Minelli et al.
2009). Cancer cells protect themselves from reactive oxygen species (ROS) by increasing the
amount of proteins like superoxide dismutase (SOD) which quickly responds to ROS production
and removes it (Murphy 2009). In this regard, it has been reported that Bcl-2 reduces
accumulation of reactive oxygen species (ROS) in transfected cells and can protect a variety of
cells from apoptosis induced by oxidative stressors. Conversely, the Bcl-2-related protein Bax
causes increased mitochondrial production of ROS and apoptosis when overexpressed in
cultured cells (Dharmarajan et al. 1999). Aside from Bcl-2, other factors that play a role in
protecting cells from oxidative stress include superoxide dismutases (SOD), which are
responsible for the conversion of superoxide radical anion to peroxide intermediates, and
glutathione peroxidase and catalase, which convert peroxides to water (Dharmarajan et al.
1999).
41
2.6.
THE CELL CYCLE
The cell cycle is composed of G1, S, G2 and M phases, which represent normal function, DNA
replication, organelle replication and mitotic separation respectively (Marieb 2012). Any errors
that may occur at these steps could be catastrophic for normal cell functioning and, as such,
the cell cycle apparatus retains potent signalling molecules that can search for errors and
rapidly induce apoptosis (Cooper 2003). The initial point where this process can occur is
referred to as the G1 restriction point (Pardee 1974).
To maintain normal function, the cell cycle is a potent inducer of cell death through regulation
of DNA content at a variety of restriction points. As the cell reaches a restriction point, the
gene p53 detects and attempts to repair any evident DNA damage; if the damage is too great it
can initiate apoptosis. The p53 gene is a potent tumour suppressor that exerts its functions
through activation of downstream targets, some of which include induction of CDKN1/p21WAF1,
14-3-3y, and REPRIMO for G1-G2 arrest; p53R2 for DNA repair and Bax, Puma, p53AIP1, PERP,
and CD95 for apoptosis. Unfortunately p53 is found mutated in nearly 70% of carcinomas(elDeiry et al. 1993; Scherr et al. 1999; Ryan et al. 2001; Fojo and Bates 2003; Huo et al. 2004).
Considerable evidence indicates that the choice made by p53 to activate cell cycle arrest and
DNA repair pathways, or the apoptosis pathway after DNA damage, is dependent on the
extent of unrepaired or misrepaired double-strand breaks in the DNA (Devlin et al. 2008).
Disabling the p53 pathway enables cells to enter and proceed through the cell cycle under
conditions that increase the frequencies of aneuploidy, gene amplification, deletion and
translocation. This, coupled with loss of p53 dependant apoptosis, increases genetic instability
and is highly selected during cancer progression (Vafa et al. 2002).
42
2.6.1. CYCLINS, CYCLIN DEPENDANT KINASES
Cell cycle progression is intricately regulated by the interactions of cyclin and cyclin-dependent
kinase (Cdk) complexes. Cyclin-D1 is a regulatory subunit of the highly conserved cyclin family
that phosphorylates and, together with sequential phosphorylation by cyclin-E/CDK2,
inactivates the cell-cycle inhibiting function of the retinoblastoma protein. Upon mitogenic
stimulation, the cyclin-D Cdk4/Cdk6 and cyclin-E Cdk2 complexes mediate phosphorylation of
retinoblastoma protein, which induces transcriptionally active E2F thereby ensuring G1-S
transit (Baldin et al. 1993; Roy et al. 2007). Retinoblastoma protein serves as a gatekeeper of
the G1 phase, and passage through the restriction point leads to DNA synthesis, making cyclinD1 expression an important target to regulate cancer cell replication (Fu et al. 2004).
2.6.2. P21 WAF1 /P53
p21WAF1 is a cyclin dependant kinase inhibitor (CdkI) family member, along with p27 and p57,
which interfere with the cyclin dependant kinase-cyclin complex. The group is regulated both
by internal and external signals, with the expression of p21WAF1 under transcriptional control of
the p53 tumour suppressor gene (Srivastava et al. 2007). As an important mediator of
apoptosis, the frequency of somatic mutations in the p21WAF1 gene and family in cancers is very
rare; which underlines the importance of these molecules as promising therapeutic targets.
Both p21 and p27 are upregulated by a variety of regulatory pathways at the transcriptional as
well as post-transcriptional levels, with p21WAF1 transcriptionally upregulated by p53 in
response to DNA damage. p21WAF1 is also activated by various transcription factors and
subsequently mediates growth arrest, senescence and apoptosis in a p53-independent
manner, and p21 mRNA stability can be post transcriptionally regulated by HuR, an RNAbinding protein, in response to stress (Roy et al. 2007; Roy et al. 2008).
43
p21WAF1 is a potent cyclin-dependent kinase inhibitor (CKI). The p21 (CIP1/WAF1) protein binds
to and inhibits the activity of cyclin-CDK2 or -CDK1 complexes, and thus functions as a
regulator of cell cycle progression at G1. The expression of this gene is tightly controlled by the
tumour suppressor protein p53, through which this protein mediates the p53-dependent cell
cycle G1 phase arrest in response to a variety of stress stimuli. This was a major discovery in
the early 1990s that revealed how cells stop dividing after being exposed to damaging agents
such as radiation. In addition to growth arrest, p21 can mediate cellular senescence and one of
the ways it was discovered was as a senescent cell-derived inhibitor (Niculescu et al. 1998;
Vafa et al. 2002). Expression of p21 is mainly dependent on two factors 1) stimulus provided
2) type of the cell. Growth arrest by p21 can promote cellular differentiation. p21 therefore
prevents cell proliferation. p53 exerts its functions mainly through transactivational activity,
including the induction of CDKN1/p21, 14-3-3y, and REPRIMO for G1-G2 arrest; p53R2 for DNA
repair and Bax, Puma, p53AIP1, PERP, and CD95 for apoptosis. Considerable evidence indicates
that the choice made by p53 to activate cell cycle arrest and DNA repair pathways or the
apoptosis pathway after DNA damage is dependent on the extent of unrepaired or misrepaired
double-strand breaks in the DNA (Waldman et al. 1996; Devlin et al. 2008).
2.6.3. KI-67
Ki-67 antigen is present in all proliferating cells (normal and neoplastic) and its evaluation
allows determining growth fraction of cellular population in a relatively easy way. Previous
studies proved the hypothesis that in breast cancer the Ki-67 index is an independent
prognostic factor in both patient survival assessment and disease recurrence. Ki-67 index can
also be used as a predictive factor of neoplastic cell response to certain types of therapy (Koda
et al. 2007). An important indicator of cell cycle pace, Ki-67 is present in all proliferating cells,
normal and neoplastic, and its level of expression indicates the rate of growth of a cell
population. Studies proved the hypothesis that, in breast cancer, the Ki-67 index is an
44
independent prognostic factor in both patient survival assessment and disease recurrence
(Dudderidge et al. 2007).
2.6.4. C-MYC
c-Myc is an important regulator of cell function that was implicated as one of the first
oncogenes. It is known that cell cycle regulation is altered under excess c-Myc expression with
a decrease in time taken to reach the restriction point of G1 (Yin et al. 1999). c-Myc protein
overexpression has been reported to immortalise cells, to reduce their growth factor
requirements and to promote cell cycle progression and genomic instability (Wasylishen and
Penn 2010).
According to a role in the maintenance of the mitotic spindle integrity,
overexpression of c-Myc has been reported to disrupt the spindle checkpoints activated by
taxanes. In colon cancer cell lines, c-Myc amplification has been related to the modulation of
the multiple effects of paclitaxel (Cassinelli et al. 2004). C-Myc cell cycle regulation has been
implicated by varied means; canonical Wnt/β-catenin stimulation, Cyclin-D and Cyclin-E CDK2
stimulation, inhibiting the effect of p27 a CDKI family member amongst others (Mateyak et al.
1999).
2.7.
WNT SIGNALLING
The Wnt family of secreted ligands operates through multiple receptors, to modulate discrete
intracellular linear and integrated signalling pathways in embryonic development, in adults and
in prostate cancer progression. Canonical signalling occurs when a Wnt ligand attaches to a
Frizzled (Fzd) family receptor and to a low density lipoprotein receptor-related protein 5 (LRP5)
or LRP6 coreceptor. This triggers a cytoplasmic Wnt/β-catenin transduction cascade, which
45
mediates β-catenin stability via glycogen synthase kinase 3 beta (GSK3β) and results in a
context dependent transcriptional outcome. Stabilized β-catenin forms multicomponent
transcriptional complex with LEF⁄Tcf and activates downstream targets such as c-Myc (Farooqi
et al. 2011). In the absence of Wnt signals, cytoplasmic β-catenin is sequestered by a complex
consisting of Axin, the tumour suppressor Adenomatous polyposis coli (APC), and GSK3β (Hall
et al. 2006).
Non-canonical Wnt proteins such as Wnt5a, lead to the activation of protein kinase C and
calcineurin with downstream intracellular targets. The Wnt/Calcium pathway has been
reported to control proliferation and have the ability to act as a tumour suppressor (Liang et al.
2003). The activity of Wnt proteins is controlled by soluble extracellular antagonists including
secreted frizzled-related proteins (sFRP), Wnt inhibitory factor-1 (Wif1), Cerberus, and
Dickkopf (DKK) (Hirata et al. 2011). sFRP, WIF-1, and Cerberus act as competitive inhibitors of
the frizzled receptor by sequestering Wnt factors and can, therefore, block both canonical and
non-canonical Wnt pathways. DKK-1, in contrast, binds the Wnt co-receptors LRP 5 and 6 to
block canonical Wnt signalling (Hall et al. 2010).
2.7.1. Β-CATENIN
The protein β-catenin has at least 2 functions of interest in prostate cancer: it participates in
cadherin-mediated adhesion, and it is the "molecular node" of the Wnt canonical signalling
pathway. In the absence of Wnt signals (i.e., when the pathway is inactive), the
serine/threonine kinase glycogen synthase kinase 3 beta (GSK3β) forms complexes with
proteins, which in turn bind to soluble β-catenin and facilitate its phosphorylation (Wan et al.
2012). Phosphorylated β-catenin then binds an E3 ubiquitin ligase then undergoes
proteosomal degradation, preventing the accumulation and transcriptional activity of β46
catenin. When Wnt ligands bind their receptor complex, the resulting activation of the
cytoplasmic protein dishevelled inactivates GSK3β, thereby preventing degradation of soluble
β-catenin and stabilizing it in the cytoplasm (Hirata et al. 2011; Menezes et al. 2012).
Cytoplasmic β-catenin then translocates to the nucleus, where it heterodimerises with
transcription factors of the T-cell factor/lymphoid enhancer–binding factor (TCF/LEF) family
(van Es et al. 2003). Accumulation of soluble β-catenin is therefore critical for activation of Wnt
transcription in the pathway. More recently, another group reported that activation of Wnt/βcatenin signalling is involved in prostate cancer initiation and progression in a mouse model
(Yu et al. 2011). Together, these findings imply that the Wnt canonical pathway is involved in
the pathogenesis of a subgroup of advanced prostate cancers.
2.7.2. ANDROGEN RECEPTOR AND Β-CATENIN
Initiating the normal and cancerous prostate cell cycle progression is the activity of the growth
factor androgen receptor (AR), which is found to be mutated in most late stage prostate
cancers. The Wnt pathway and its interaction with AR have been suspected to play important
roles in prostate cancer (Wang et al. 2008). Initiating the cell cycle progression is the activity of
the growth factor androgen receptor (AR), which is found to be mutated in most late stage
prostate cancers and is widely accepted to be the initiator of prostate cell replication and
differentiation through transcription of mitogenic and anti-apoptotic. Pathways such as the
canonical Wnt/β-catenin signalling pathway drive the cells towards growth and differentiation
while interaction with the anti-apoptotic Bcl-xL prevent the cells from entering into apoptosis
when AR is stimulated (Sun et al. 2008; Wang et al. 2008). Because AR signalling has also been
shown to play a key role in prostate carcinogenesis, androgen ablation therapy is a commonly
used form of treatment, particularly for advanced disease. While ablation therapy leads to
initially significant levels of prostate cancer cell apoptosis, the effect is short-lived and
47
ultimately not curative as most patients develop androgen-independent disease (D'Antonio et
al. 2008). Manipulation of this growth factor by; upregulation of receptors, mutation to a
constitutively active form or a complete loss of AR as a mitogenic requirement are
characteristics of late stage prostate cancer (Peng et al. 2008).
Levels of β-catenin and AR are both increased in hormone refractory prostate cancer, βcatenin is activated by canonical Wnt stimulation, but mutated forms of β-catenin, which can
result in a permanently stabilised protein, have also been detected in prostate cancer. βcatenin has multiple functions that involve both cell adhesion and signal transduction in
response to Wnt ligands (Chesire et al. 2002). In the absence of Wnt ligands, GSK3β complexes
with other proteins and degrades cytoplasmic β-catenin, but when Wnt ligands bind to the
frizzled receptor complex, the resulting activation of the cytoplasmic protein dishevelled
inactivates GSK3β, thereby preventing degradation of β-catenin (Wan et al. 2010; Wan et al.
2012). It is important to note that GSK3β has been linked to prevention of AR transcriptional
activity; so Wnt ligand binding both increases AR expression and β-catenin stability (Li et al.
2008). Upon stimulation, β-catenin translocates to the nucleus where it can drive cell cycle
progression through interaction with T-cell factor (TCF)/lymphoid enhancer factor (LEF)
transcription factors to initiate transcription of target genes such as c-Myc and Cyclin-D1,
driving cell growth. As such, dysregulation of the Wnt, AR or β-catenin pathways can result in a
metastatic, highly replicative phenotype (Wang et al. 2008). It has been postulated that the
ratio of β-catenin/AR might be an important prognostic indicator that may even help define a
subpopulation of men with prostate cancer for individualised management (Wan et al. 2012).
2.7.3. SFRP4
48
Secreted frizzled related protein 4 (sFRP4) is one of a group of proteins that antagonise the
canonical Wnt/β-catenin pathway, decreasing Wnt activity and preventing activated β-catenin
from forming (Drake et al. 2003; Drake et al. 2009). sFRP4 can also decrease invasiveness in
androgen-independent prostate cancer cells and has been shown to be anti-angiogenic and
pro-apoptotic in nature (Muley et al. 2010). Moreover, the correlation between increased
membranous sFRP4 and β-catenin expression in a large human cohort supports evidence for
sFRP4 as a prognostic marker in localised androgen-dependent prostate cancer (Horvath et al.
2004). Unlike other inhibitors of Wnt signalling, sFRP4 appears to affect androgen-dependent
and androgen-independent prostate cancer (Horvath et al. 2007).
49
3. CHAPTER THREE: MATERIALS AND
METHODS
3.1.
TISSUE CULTURE
Cell culture was performed in a tissue culture facility located within a physical containment
level 2 (PC2) grade laboratory in the school of Anatomy, Human Biology and Physiology at the
University of Western Australia. The tissue culture facility utilised biohazard class II hoods and
cell culture incubators operating at 37°C 5% CO2. Incubators were sterilised six monthly to
prevent contamination and water in the 37°C water bath and incubators was treated with antibacterial agent Aqua-clear (Cleanware).
3.2.
CELL CULTURE
All cell lines were cultured in T-75cm2 tissue culture flasks with vented, filtered caps. Two types
were used over the length of the project; one from Sarstedt (cat#83.1813.002) and one from
Corning (cat#3290). Similarly 96 well plates (Sarstedt cat#83.1835.500, Corning cat#3595) and
6 well plates (Sarstedt cat#83.1839.500, Corning cat#3516) were bought from both companies
with no detectable change in cellular adherence to plastic or growth rate. Tissue culture media
was replaced every three days with cells reaching confluence approximately every five to eight
days dependent upon seed rate. Flasks were reused for 4 passages before being replaced by a
new flask. 10cm dishes were supplied by BD Falcon (cat#353003).
50
3.2.1. SUBCULTURE AND COUNTING OF CELLS
All cell lines were subcultured in a similar manner. Cells had their media aspirated, 4ml of 1x
Phosphate Buffer Solution (PBS) was added to the flask then aspirated off. Following the wash
with 1xPBS, 3ml Trypsin 0.5% EDTA (Gibco Cat#25300) was quickly washed over the cells
before being aspirated and another 3ml of Trypsin 0.5% EDTA was added, cells were moved to
a 37°C 5% CO2 incubator for 5 minutes or until the cells had released off the surface. Flasks
were given a gentle tap to encourage cell separation and to assist the release of adherent cells
from the plastic surface. The trypsin wash step noticeably increased the speed at which cells
would successfully release from the flask plastic.
Following trypsinisation, 5ml of complete growth media was added into the flask, to inactivate
the trypsin EDTA mixture, and the mixture was washed up and down along the flask’s base
surface, increasing cell homogenisation and washing off the last adherent cells, before being
transferred to a 15ml Falcon tube (Sarstedt Cat#62.554.205-500CS). The cells were mixed
thoroughly to ensure homogeneity then 20µL was removed and placed into both sides of a
haemocytometer. The inside 5x5 well section of the haemocytometer was counted on both
sides, with the top and right line touching cells counted while the bottom and left touching
cells were discarded. An average of the two numbers was taken and then multiplied x104 to
get the concentration of cells/mL in suspension.
3.2.2. CELL CRYOPRESERVATION AND STORAGE
To ensure adequate numbers of low passage cells, stocks were frozen down regularly for long
term use. Cells were trypsinised and counted as previously described, and then the 15ml
Falcon tube was centrifuged at 1.5krpm/0.4rcf on an Eppendorf 5702 centrifuge for 3 minutes.
51
Following centrifugation the cells were clumped in the bottom of the tube and the
media/trypsin aspirated off. Freezing down solution was added such that the cells were left in
a concentration of 1x106 cell/mL. Freezing down solution was a combination of 90% total
growth media and 10% dimethyl sulfoxide (DMSO, Sigma cat#D2650). Cells were placed in a
freezing down container (Nunc) containing -20°C isoproanol and moved into a -80°C freezer for
24 hours. After being in the -80°C freezer for 24 hours the cells were removed and placed in a
liquid nitrogen long term storage container.
3.2.3. LNCAP CELLS
The LNCaP clone was isolated in 1977 by J.S. Horoszewicz, from a needle aspiration biopsy of
the left supraclavicular lymph node of a 50 year old Caucasian male with confirmed diagnosis
of metastatic prostate carcinoma (Horoszewicz et al. 1980). These cells are responsive to 5alpha-dihydrotestosterone (DHT) which means they contain a functioning androgen receptor
(AR) and are indicative of an early stage prostate cancer cell line even though they are
metastatic in nature (Horoszewicz et al. 1983). The cells do not grow into a uniform monolayer
but grow in clusters that must be broken apart when subculturing. They attach very lightly to
the substrate plastic, will rapidly acidify the media and are a hypertetraploid. They contain
high amounts of excess chromosome numbers with 22% of the cells containing 84
chromosomes, while cells with 86 (20%) and 87 (18%) also occur at high frequencies in culture.
Two forms of androgen receptor exist in LNCaP cells, a 110kDa and 112kDa with the 112kDa
shown to be phosphorylated form (Alimirah et al. 2006).
52
3.2.3.1
LNCAP MEDIA FORMULATION
LNCaP cells (cat# CRL-1740) were sourced from the American Type Culture Collection (ATCC)
and grown in a media comprising of RPMI 1640 (Gibco Cat#21870) supplemented with; 10%
fetal bovine serum (FBS, DKSH Australia Cat#FBS S101), 1% penicillin streptomycin (Gibco
Cat#15070), 1% Glutamax™ (Gibco Cat#35050), 10mM HEPES buffer (Sigma cat#H4034), 4.5g/L
d-Glucose (Sigma cat#G7528) and 1mM sodium pyruvate (Sigma cat#S8636).
3.2.4. DU145 CELLS
DU145 cells were isolated by (Stone et al. 1978) from a grade IV brain metastasis of prostate
cancer in a 69 year old Caucasian male. They are used as a “classical” example of late stage
prostate cancer and have moderate metastatic potential compared to PC3 cells. The cell line is
only weakly positive for acid phosphatase and exhibits very low DHT activity and is considered
androgen receptor (AR) negative. (Alimirah et al. 2006) have since determined that DU145 and
PC3 cells contain AR in measurable levels however, the cells do not respond to AR stimulation
and therefore have a broken pathway. Although they are not truly AR negative as they do
express a form of the receptor, it is a completely inactive form. DU145 cells are a hypotriploid
human cell line.
3.2.4.1
DU145 MEDIA FORMULATION
DU145 cells (cat#HTB-81) were sourced from the ATCC and grown in media comprising of
Minimum Essential Medium Earle’s (MEM, Gibco cat#11090) supplemented with; 10% fetal
bovine serum, 1% penicillin streptomycin, 1% Glutamax™, 1mM sodium pyruvte and 100µM
MEM non-essential amino acids solution (Gibco cat#11140).
53
3.2.5. PC3 CELLS
PC3 cells were isolated by (Kaighn et al. 1979) from a grade IV bone metastasis of prostate
cancer in a 62 year old Caucasian male. They are used as a “classical” example of late stage
prostate cancer and exhibit high metastatic potential. The cell line is only weakly positive for
acid phosphatase and exhibits very low DHT activity and is considered androgen receptor (AR)
negative. (Alimirah et al. 2006) have since determined that DU145 and PC3 cells contain AR in
measurable levels but that the cells do not respond to AR stimulation and therefore have a
broken pathway, but are not truly AR negative as they do express a form of the receptor, just
an inactive form.
3.2.5.1
PC3 MEDIA FORMULATION
PC3 cells (cat#CRL-1435) were sourced from ATCC and grown in media comprising of F-12K
(Kaighns Modification, Gibco cat#21127) supplemented with; 10% fetal bovine serum, 1%
penicillin streptomycin and 1% Glutamax™.
3.3.
DRUG DILUTIONS
3.3.1. PHENOXODIOL
151mg of Phenoxodiol, 2H-1-benzopyran-7-0,1,3-[4-hydroxyphenyl], was supplied from
Novogen Pty Ltd in pink, slightly sticky, powdered form, in an amber glass bottle and was
stored at 4°C in a container with desiccation crystals to prevent moisture build up. Phenoxodiol
was delivered into the cells with DMSO as the carrier vehicle solution; hence all treatments
54
include a vehicle control group and the 10µM concentration has extra carrier added to
equilibrate vehicle control between treatment groups.
3.3.2. PHENOXODIOL WORKING STOCK
Phenoxodiol has a molecular weight of 240.26g/L and the treatment concentrations used in
this thesis were determined to be 10µM and 30µM by Novogen. To get the dose to this level
PXD was diluted at 10mg/mL, in cell culture sterile DMSO, which gave a concentration of
approximately 41600µM.
EQUATION 1 PHENOXODIOL WORKING STOCK EQUATION
3.3.3. PHENOXODIOL WORKING SOLUTION
Following dilution in DMSO, PXD working stock was stored in a covered vial at -20°C for a
maximum of three weeks before being discarded. Phenoxodiol was further diluted to make
working solutions in the complete growth media appropriate to each cell line. Combining 10µL
PXD stock (41600µM) with 990µL media resulting in a working solution with a concentration of
416µM which was then further diluted to reach the final cell concentrations of 10µM and
30µM. Vehicle control was standardised between all samples by addition of appropriately
55
concentrated DMSO in media to make DMSO working solution 10µL of DMSO was added into
990µL appropriate growth media per cell line. Where any dilution was performed a mirror
dilution containing only DMSO was performed to get an equally concentrated vehicle control
solution.
TABLE 3 PHENOXODIOL TREATMENT CALCULATIONS PER ML OF MEDIA
Treatment Group
DMSO Working
Phenoxodiol
Media µL
Solution µL
Working Solution µL
DMSO Control
72.15µL
-
927.75µL
10µM Phenoxodiol
48.11µL
24.04 µL
927.75µL
30µM Phenoxodiol
-
72.15µL
927.75µL
EQUATION 2 PHENOXODIOL WORKING SOLUTION EQUATION
56
3.3.4. PHENOXODIOL TREATMENT
Cells were seeded at the appropriate level and grown for 48 hours. After 48 hours growth
media was aspirated from cells and replaced with appropriate treatment of Phenoxodiol;
Vehicle Control (DMSO), 10µM PXD and 30µM PXD. Following 24 or 48 hours of treatment,
media was removed and centrifuged to retain floating cells, and the experiment carried out.
3.3.5. DOCETAXEL
100mg of Docetaxel, 1,7β, 10β-trihydroxy-9-oxo-5β, 20-epoxytax-11-ene-2α, 4, 13α-triyl 4acetate-2-benzoate-13-{(2R,3S)}-3-[(tert-butoxycarbonyl)amino]-2-hydroxy-3phenylpropanoate}, was purchased (Sigma Aldrich Cat#01885) in a white crystalline powdered
form. Powder was supplied in an amber glass bottle and was stored at -20°C in a container
with desiccation crystals to prevent moisture build up. Docetaxel was delivered into the cells
with DMSO as the carrier vehicle solution in two separate working solutions. All treatments
include two vehicle control groups to equilibrate vehicle control between treatment groups.
3.3.6. DOCETAXEL WORKING STOCK AND WORKING SOLUTION
The following dilution was performed on Docetaxel to produce the two usable working
solutions (1.2378µM and 123.78nM) for use in an isobologram study with Phenoxodiol.
Docetaxel has a molecular weight of 807.88g/L and the concentrations determined for use in
this thesis were 0.1, 1, 5, 10 and 100nM. Docetaxel working solutions were discarded after use
while working stock was kept at -20°C in a covered, sealed, glass bottle for a maximum of two
weeks before being discarded. Where any dilution was performed a mirror dilution containing
only DMSO was performed to get an equally concentrated vehicle control solution.
57
EQUATION 3 DOCETAXEL STOCK AND WORKING SOLUTIONS
3.4.
OPTIMISATION OF PROLIFERATION
MTS proliferation assays were performed to determine appropriate cell seeding rates for 96
well, 6 well and 10cm plate experiments over 48 hours of growth and 48 hours of treatment.
The CellTiter 96® AQueous One Solution Cell Proliferation Assay (Promega cat#G3580) is a
colorimetric method for determining the number of viable cells in proliferation, cytotoxicity or
chemosensitivity assays. The CellTiter 96® AQueous One Solution Reagent contains a tetrazolium
compound
3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-
tetrazolium (MTS) and an electron coupling reagent phenazine ethosulfate (PES). PES has
enhanced chemical stability, which allows it to be combined with MTS to form a stable
solution. MTS absorbance was measured at 492nm on a Labsystems Multiskan RC plate
reader.
3.4.1. DETERMINATION OF CELL SEEDING CONCENTRATIONS
To determine the appropriate number of cells to use in 96 and 6 well plate assays an initial cell
seeding concentration assay was performed, on all three cells lines, to prevent cells from
58
reaching crowding induced senescence at the end of the treatment period (96 hours total
growth). Cells were seeded into clear 96 well plates (Sarstedt cat#83.1835, Corning
cat#CLS3595) with concentrations of; 500, 1000, 1500, 2000, 2500, 3000, 4000 and 5000
cells/100µL. Cells were seeded in the opposite direction to the treatment groups, if the
treatments are in the direction of wells 2-3-4-5 then the wells are seeded B2-C2-D2-E2 and
vice versa, this prevented bias in treatment groups caused by unequal cell volumes due to
poor mixing. After 48 hours of growth all wells had the media aspirated and replaced with
100µL of fresh complete cell media. Plates underwent MTS proliferation analysis after 6, 24
and 48 hours of total growth post initial 48 hour period. Assays were performed by adding a
20µL volume of the CellTiter 96® AQueous One Solution Reagent directly to the 100µL of sample
media and cells, incubating for 3 hours at 37°C 5% CO2, and then recording absorbance at
492nm with a Labsystems Multiskan RC plate reader. The quantity of formazan product as
measured by the amount of 492nm absorbance is directly proportional to the number of living
and metabolising cells in culture.
TABLE 4: SEEDING CONCENTRATIONS AND VOLUMES
Cell Line
96 Well Plate (100µL)
6 Well plate (2ml)
10cm Dish (5ml)
LNCaP
30,000 cells/ml
54,000 cells/ml
100,000 cells/ml
DU145
18,000 cells/ml
32,000 cells/ml
100,000 cells/ml
PC3
25,000 cells/ml
45,000 cells/ml
100,000 cells/ml
3.4.2. PHENOXODIOL MTS ASSAY
Phenoxodiol (PXD) was shown to increase activity of the MTS assay product formazen without
cellular interface, all PXD MTS proliferation assays included n=2 no cell controls for all
treatment groups to remove background absorbance increase caused by PXD. In final analysis
59
no cell controls were averaged and removed from each of the cell samples which varied from
n=4 to n=6. Experimentation began with cells seeded at appropriate rates into clear 96 well
plates (Sarstedt#83.1839, Corning#CLS3516) with a total volume of 100µL per well. All assays
included the following groups; Vehicle control (DMSO), 10µM PXD and 30µM PXD over 24 and
48 hours. Cells were grown for 48 hours in an incubator at 37°C 5% CO2 before the complete
cell media was aspirated and replaced with 100µL of appropriate treatment media then left to
grow at incubator at 37°C 5% CO2 for 24 and 48 hours. Following appropriate growing period
20µL of CellTiter 96® AQueous One Solution was added to each well, mixed gently with a pipette
then left for 3 hours in an incubator at 37°C 5% CO2 then analysed on a Labsystems Multiskan
RC plate reader at 492nm.
Vehicle
Control
10µM
PXD
30µM
PXD
FIGURE 10 MTS ASSAY SETUP
MTS assay setup with cell treatments indicated by yellow and no cell controls indicated by red,
multiple cell lines could be grown on the same plate for 24 or 48 hour treatment however the
two time points could not be performed on the same plate due to risk of contamination.
60
3.4.3. PHENOXODIOL DOCETAXEL ISOBOLOGRAM
An isobologram is a test for properties of interference, additivity or synergy between two or
more chemicals, and was performed between phenoxodiol and the common prostate cancer
treatment Docetaxel, a taxane based anti-microtubulin chemotherapeutic agent. The
isobologram had 24 individual treatment groups with 4 Phenoxodiol (0,5,10,30µM PXD) and 6
Docetaxel (and 0,0.1,1,5,10,100nM DOC) treatment concentrations, compared to each other.
All treatments had a cell n=4 and no cell control of each n=2 resulting in a total of 144 wells
per cell line per time course. Due to this high number only the 48 hour isobologram was
performed to allow for maximal treatment variation. As the outside rows of 96 well plates
have been shown to have a statistically significant difference in light absorption they were
excluded from use resulting in the need for a single plate for each Phenoxodiol treatment
group (4 plates total used). Cells were seeded at appropriate rates in 100µL complete growth
media and left for 48 hours.
61
DOC
Vehicle
Control
0nM
DOC
0.1nM
DOC
5nM
DOC
10nM 100nM
DOC
DOC
FIGURE 11 96 WELL ISOBOLOGRAM SETUP
The 96 well plate setup required for each cell line and Phenoxodiol concentration with yellow
indicating cell containing wells and red indicating no cell control wells. Each cell line had 4
plates with varying levels of Phenoxodiol used; Vehicle Control (DMSO), 5µM, 10µM and 30µM
Phenoxodiol. Cells were treated over a 48 hour period with all treatment groups balanced to
the most concentrated DMSO group (30µM PXD + 100nM DOC).
Following 48 hours of growth cells had their media aspirated and an appropriate media added
to each well. Media was made to control for the DMSO vehicle used to deliver both the PXD
and DOC treatments. In media preparation two sets of DOC working solutions were required
(1.2378µM and 123.78nM) to be able to accurately pipette the volumes required across the
spread of concentrations and as such two vehicle (DMSO) solutions were also required.
62
TABLE 5 ISOBOLOGRAM SETUP FOR VEHICLE CONTROL AND TREATMENT GROUPS
PXD and DOC
PXD
PXD
DOC
DOC
DOC
DOC
Media
Concentrations Vehicle
Working
Vehicle
Working
Vehicle
Working
to total
(µM and nM)
Control
Solution
Control
Solution
Control
Solution
1mL (µL)
(DMSO)
(µL)
1
1 (µL)
2
2 (µL)
(µL)
(416µM)
(DMSO)
(1.2378
(DMSO)
(123.78
(µL)
µM)
(µL)
nM)
P0D0
72.11
-
80.79
-
8.08
-
839.02
P 0 D 0.1
72.11
-
80.79
-
7.27
.81
839.02
P0D1
72.11
-
80.79
-
-
8.08
839.02
P0D5
72.11
-
76.75
4.04
8.08
-
839.02
P 0 D 10
72.11
-
72.71
8.08
8.08
-
839.02
P 0 D 100
72.11
-
-
80.79
8.08
-
839.02
P5D0
60.09
12.02
80.79
-
8.08
-
839.02
P 5 D 0.1
60.09
12.02
80.79
-
7.27
.81
839.02
P5D1
60.09
12.02
80.79
-
-
8.08
839.02
P5D5
60.09
12.02
76.75
4.04
8.08
-
839.02
P 5 D 10
60.09
12.02
72.71
8.08
8.08
-
839.02
P 5 D 100
60.09
12.02
-
80.79
8.08
-
839.02
P 10 D 0
48.07
24.04
80.79
-
8.08
-
839.02
P 10 D 0.1
48.07
24.04
80.79
-
7.27
.81
839.02
P 10 D 1
48.07
24.04
80.79
-
-
8.08
839.02
P 10 D 5
48.07
24.04
76.75
4.04
8.08
-
839.02
P 10 D 10
48.07
24.04
72.71
8.08
8.08
-
839.02
P 10 D 100
48.07
24.04
-
80.79
8.08
-
839.02
P 30 D 0
-
72.11
80.79
-
8.08
-
839.02
P 30 D 0.1
-
72.11
80.79
-
7.27
.81
839.02
P 30 D 1
-
72.11
80.79
-
-
8.08
839.02
P 30 D 5
-
72.11
76.75
4.04
8.08
-
839.02
P 30 D 10
-
72.11
72.71
8.08
8.08
-
839.02
P 30 D 100
-
72.11
-
80.79
8.08
-
839.02
The total number of solutions required were two DOC working and vehicle control solutions,
one PXD working solution and one PXD vehicle control solution. Following 48 hours of
63
treatment, with PXD and DOC, 20µL of CellTiter 96® AQueous One Solution Reagent was added to
each well and mixed gently then incubated at 37°C 5% CO2 for 3 hours before being analysed
on a Labsystems Multiskan RC plate reader at 492nm.
3.4.4. PHENOXODIOL AND CASPASE INHIBITION
In order to understand potential interactions between the caspase signalling pathway and
Phenoxodiol an experiment was performed using the pan caspase inhibitor Z-VAD-FMK
(Carbobenzoxy-valyl-alanyl-aspartyl-(O-methyl)-fluoromethylketone)
purchased
from
BD
Pharmingen Cat#550377. Z-VAD-FMK is a cell permeable general caspase inhibitor that
irreversibly binds to the catalytic site of caspase proteases and inhibits apoptosis. 1mg of ZVAD-FMK was reconstituted in 214µL of DMSO with a resulting stock concentration of 10mM.
A final concentration of 10µM Z-VAD-FMK was used in the study with the maximal PXD dose of
30µM over a 48 hour period. Before use a titration was performed indicating that 100µM ZVAD-FMK was cytotoxic and 100nM Z-VAD-FMK was not sufficiently concentrated to prevent
caspase activation from a UV irradiation source (measured by analysing activated Caspase-3
post irradiation). A final concentration of 10µM Z-VAD-FMK was chosen as it prevented
Caspase-3 from activating post UV irradiation however cells did eventually enter into cell death
when measured after 48 hours. A 100 fold dilution in complete cell media was performed to
get a working solution of 100µM Z-VAD-FMK. Cells were seeded at appropriate levels in clear
96 well plates with cells n=4 and no cell controls n=2 for the treatment groups; Vehicle control,
10µM Caspase Inhibitor, 30µM PXD and 10µM CI + 30µM PXD. Following 48 hours of growth in
a 37°C 5% CO2 incubator the media was aspirated and replaced with the appropriate
treatment media.
64
TABLE 6 CASPASE INHIBITION AND PHENOXODIOL TREATMENT SOLUTIONS
Treatment
Phenoxodiol
Phenoxodiol
Caspase
Caspase
Media µL to
Group
DMSO
Working
Inhbitor
Inhbitor
make 1mL
Working
Solution
DMSO
Working
Solution (µL)
416µM (µL)
Working
Solution
Solution (µL)
100µM (µL)
DMSO
72.15
-
100
-
827.85
72.15
-
-
100
827.85
-
72.15
100
-
827.85
-
72.15
-
100
827.85
Control
10µM
Caspase
Inhibitor (CI)
30µM
Phenoxodiol
10µM CI +
30µM
Phenoxodiol
Following 48 hours of treatment, with PXD and CI, 20µL of CellTiter 96® AQueous One Solution
Reagent was added to each well and mixed gently then incubated at 37°C 5% CO2 for 3 hours
before being analysed on a Labsystems Multiskan RC plate reader at 492nm.
3.4.5. PHENOXODIOL AND DOCETAXEL COMBINATION TREATMENT
In order to further understand the interaction of phenoxodiol and docetaxel an experiment
was performed to analyse pre-treatment of cells with phenoxodiol and subsequent cell
cytotoxicity when both solutions were recombined. Cells would be exposed to either; 48 hours
of 10µM or 30µM phenoxodiol only, 48 hours of 100nM docetaxel only, 48 hours of
10µM/30µM phenoxodiol 100nM docetaxel combination, 48 hours of 10µM/30µM
phenoxodiol and 24 hours of docetaxel or 48 hours docetaxel and 24 hours of 10µM/30µM
65
phenoxodiol. Cells were seeded at appropriate rates in 100µL complete growth media and left
for 48 hours with an n=4 for treatment groups and n=2 for no cell (media only) controls. Media
was then aspirated off and replaced with treatment media (as seen in table 4) with the two 48
hour/24 hour combinations receiving a normal dose of either; 10µM/30µM phenoxodiol or
100nM docetaxel. All treatments were controlled for DMSO vehicle concentration. After 24
hours of treatment the final treatment was given in concentrated form to the treatment wells
that required it, either 100nM docetaxel to the 48 hour PXD 24 hour DOC treatment or
10µM/30µM PXD to the 48 hour DOC 24 hour PXD treatment, again DMSO vehicle was
controlled for at this point. After a further 24 hour incubation, 20µL MTS was added to each
well and incubated in the dark at 37°C 5% CO2 for 3 hours before being analysed on a
Labsystems Multiskan RC plate reader at 492nm.
3.4.6. PHENOXODIOL AND PURIFIED SFRP4 PROTEIN COMBINATION
TREATMENT
In order to assess the potential interaction of Phenoxodiol with the canonical Wnt/β-catenin
pathway and non-canonical Wnt pathway, an experiment was performed utilising the frizzled
receptor antagonist secreted frizzled related protein 4 (sFRP4). Cells would be proliferated for
48 hours then exposed to 48 hours of treatment utilising; vehicle control (PBS+DMSO),
125pg/mL sFRP4, 250ph/mL sFRP4, 500pg/mL sFRP4, 10µM Phenoxodiol, 30µM Phenoxodiol
and a combination of 500pg/mL sFRP4/30µM Phenoxodiol. Cells were seeded at appropriate
rates in 100µL complete growth media and left for 48 hours with an n=4 for treatment groups
and n=2 for no cell (media only) controls. Following 48 hours of growth the complete media
was aspirated off and replaced with treatment media (below) with the working solution
concentration of sFRP4 purified protein being 125µg/mL in PBS. After 48 hours of incubation,
66
20µL MTS was added to each well and incubated in the dark at 37°C 5% CO2 for 3 hours before
being analysed on a Labsystems Multiskan RC plate reader at 492nm.
TABLE 7 SFRP4 & PHENOXODIOL TREATMENT SOLUTIONS
Treatment
DMSO
Phenoxodiol
Group
Working
Solution (µL)
Vehicle
PBS (µL)
sFRP4
Media µL to
Working
Working
1mL
Solution
Solution
416µM (µL)
(125µg/mL)
72.15
-
4
-
923.85
72.15
-
3
1
923.85
72.15
-
2
2
923.85
72.15
-
-
4
923.85
48.11
24.04
4
-
923.85
-
72.15
4
-
923.85
-
72.15
-
4
923.85
Control
(DMSO + PBS)
125pg/mL
sFRP4
250pg/mL
sFRP4
500pg/mL
sFRP4
10µM
Phenoxodiol
30µM
Phenoxodiol
500pg/mL
sFRP4 + 30µM
Phenoxodiol
3.5.
REACTIVE OXYGEN SPECIES DETECTION
Reactive oxygen species (ROS), specifically nitric oxide (NO), detection was performed using a
Griess Reagent System (Promega cat#G2930). The Griess system uses sulfanilamide and N-1naphthylethylenediamine dichloride (NED) under acidic conditions supplied by phosphoric
acid. This assay measures the NO2- which is one of two stable non-volatile breakdown products
from NO production.
67
Cells were seeded at the appropriate rate on a 96 well plate with cell controls of n=4 and no
cell controls n=2 with 4 treatment groups; Vehicle control, 10µM PXD, 30µM PXD, DEAN and
SNP. Following 48 hours of growth cells were treated with appropriate levels of Vehicle
control, Phenoxodiol or DEAN for 24 and 48 hours. DEAN (Diethylamine NONOate
diethylammonium salt, Sigma cat#D5431) a nitric oxide stimulator was used as a positive
control to show that the cells can produce NO with appropriate stimulation. DEAN was in a
stock concentration of 100mM and was diluted 1:1000 in complete cell media to become a
final concentration of 100µM. Sodium Nitroprusside (SNP) was another positive control
utilised to indicate the cells were capable of producing nitric oxide when stimulated and was in
a 10mM stock concentration that was diluted 1:100 in complete media to a final concentration
of 100µM.
Before the assay was run, 50µL of media was removed from each well leaving a volume of
50µL. A nitrite standard reference curve was performed on each assay plate. 1mL of 100µM
nitrite solution was produced by diluting the 0.1M nitrite solution supplied with the assay kit
1:10,000 in complete media. 50µL media was added to rows B10-H10 and B11-H11 and 100µL
of prepared 100µM nitrite solution was added to A10 and A11. Using a 200µL pipette set to
50µL each row had 50µL of sample taken and added into the next row, serially diluting the
solution down until the final row (H) where the product was discarded. The standard curve
wells were left with 50µL of nitrite and media solution in all wells with a concentration scale of
100, 50, 25, 12.5, 6.25, 3.13, 1.56 and 0µM nitrite in rows A-H respectively.
The sulfanilamide and NED solutions were allowed to come up to room temperature for 15
minutes, after being stored at 4°C. 50µL of sulfanilamide solution was added to each sample
and standards well, bringing the total volume per well to 100µL and the plate was covered in
68
foil to protect it from the light, and incubated at room temperature for 10 minutes. Following
incubation a further 50µL of NED solution was added to each well and the plate was again
incubated and covered at room temperature for 10 minutes. Following this final incubation the
plate was analysed on a Labsystems Multiskan RC plate reader at 540nm with increases in
absorbance expression indicating increased NO output.
3.6.
ACIDITY ANALYSIS
Media was analysed post treatment with Phenoxodiol to test whether the drug was altering
the acidity of the media. A pH meter was calibrated with buffer solutions before use and at a
room temperature of 24°C. Post 48 hour treatment media was placed under the probe to
analyse any pH changes between vehicle control and treatment groups.
3.7.
APOPTOSIS ANALYSIS ASSAYS
Following treatment with Phenoxodiol apoptosis and necrosis levels were determined through
five assays; 3’-End Labelling DNA Fragmentation Analysis, Annexin-V-FLUOS/PI Flow
Cytometry, Sybr Gold DNA Fragmentation Analysis, JC-1 Mitochondrial Potential Detection and
Activated Caspase-3 Activity. DNA extraction was required for both 3’-End labelling and Sybr
Gold DNA fragmentation assays.
3.7.1. DNA EXTRACTION
Cells were seeded in 6-well plates, n=4, at appropriate seeding rates, and grown for 48 hours.
Following 48 hours of growth the media was removed and replaced with 2mL of treatment
69
media containing; Vehicle control (DMSO), 10µM PXD or 30µM PXD. After 24 and 48 hours
treatment, the media was removed, placed in a 2mL eppendorf tube and centrifuged at 800g.
The media was then discarded leaving cells behind, while 1mL of 1xPBS was placed on each cell
well. Wells were scraped with a cell scraper (Sarstedt cat#83.1832) and pipetted into the 2mL
eppendorf tube that had their media centrifuged in. Cells were again centrifuged at 800g and
1xPBS supernatant discarded. Cells were then homogenised in 1mL of DNA homogenisation
buffer4. The homogenate then had 62.5µL of 10% SDS added, was vortexed and incubated at
65°C for 30 minutes. 175µL of 8M potassium acetate was then added, vortexed and the
homogenate incubated on ice for a further 60 minutes. Samples were then centrifuged for 10
minutes at 4°C at 12,000rpm in an eppendorf refrigerated centrifuge (Eppendorf cat#5417R)
and the supernatant was transferred to a new tube. 2µL of RNaseA 500µg/mL (Roche Applied
Science cat#10109169001) was added and the mixture then incubated at 37°C for 1 hour.
Equal volumes (500µL) of phenol:chloroform/isoamyl alcohol (1:1) were added and vortexed
vigorously, then centrifuged at for 5 minutes at 4°C at 6000rpm.
Upper aqueous phase was collected and 0.1 (50µL) volumes of 3M Sodium Acetate and 2.5
volumes (1250µL) of cold 100% ethanol were added and precipitated overnight at -80°C.
Following precipitation DNA was centrifuged for 30 minutes at 14,000rpm and the supernatant
removed. The pellet was washed with 100µL of 70% ethanol by vortexing and centrifuging for
3 minutes at 14,000rpm. Ethanol was discarded and the pellet air dried for at least 1 hour in a
class I hood then resuspended in 20µL of dH2O, vortexed and placed in a 37°C incubator for 15
minutes. Solution was vortexed and gently centrifuged to ensure dissolution of DNA pellet into
the dH2O, then a 2µL sample was taken and diluted in 48µL of dH2O for spectrophotometer
analysis on a Nanodrop ND-1000 spectrophotometer (Nanodrop cat#ND-1000) with samples
outside of a 1.8-2 260/280nm range discarded. Samples were stored in a freezer at -80°C.
4
DNA Homogenisation Buffer: 0.1M NaCl, 0.01M EDTA pH 8.0, 0.3M Tris-HCl, pH 8.0, 0.2M Sucrose in
H2O
70
3.7.2. 3’-END LABELLING DNA FRAGMENTATION ANALYSIS
Three prime end labelling (3’-end labelling) was used as a method of quantitative and
qualitative DNA fragmentation analysis. The radioactive phosphate isotope (P32) was bound to
ddATP ([α32P]-ddATP, Amersham cat#PB10233-250uCI) and then an enzyme, terminal
transferase, tagged all fragmented DNA with the isotope/ddATP mix, which was then exposed
to film allowing for a qualitative measure of apoptosis versus necrosis. This was followed by
radiation detection on a Packard Tri-Carb 1500 Liquid Scintillation Counter, allowing a
quantitative measurement of apoptosis.
To perform a 3’-end labelling fragmentation assay DNA was isolated and quantified as
previously described. A total of 1µg of DNA was used per sample (n=4 per treatment) and the
DNA solution diluted to 29µL total volume with ddH2O then added to a 3’-End Labelling
reaction solution.
TABLE 8 3'-END LABELLING REACTION MIXTURE
Reagent
Volume
DNA (1µg in 29µL H2O)
29µL
5X Reaction Buffer5
10µL
CoCl2 (25mM) solution
5µL
[α32P]-ddATP (50µCi; 3.4pmol/µL)
5µL
Terminal Transferase (25 units/µL in 50% glycerol stock diluted 80 fold in water)
1µL
Once the solution was complete samples were; vortexed briefly, quickly centrifuged and then
incubated at 37°C for 60 minutes. The reaction was terminated with the addition of 5µL of
0.25M EDTA and then 2µL of transfer RNA solution (25mg/mL) was added with 12µL 10M
ammonium acetate and 180µL of -20°C 100% ethanol before the solution was vortexed and
5
Terminal Transferase Reaction Buffer (5X): 1M Potassium Cacodylate, 0.125M Tris-HCl, 1.25mg/mL
BSA; pH 6.6
71
precipitated at -80°C for 60 minutes. After the DNA was precipitated samples were centrifuged
at 10,500rcf (Eppendorf Cat#5415C) for 20 minutes and the hot supernatant was discarded
before the pellet was resuspended in 55µL of 1x TE. The precipitation step containing
ammonium acetate and ethanol was repeated with the pellet air dried for 20 minutes, after
the supernatant was discarded, then diluted in 40µL of 1x TE and stored at -20°C overnight.
A 2% agarose-1x TAE gel was prepared and set while samples were thawed and 20µL of sample
combined with 4µL of DNA loading buffer6 and loaded into the gel with the excess solution
stored at -20°C. Samples underwent electrophoresis at 60 volts for between 2.5 and 4 hours in
1x TAE buffer. Following electrophoresis gels were placed into a gel slab dryer for 2 hours or
until the gel had over 90% of it’s liquid content removed. The thin gel was then placed into
plastic wrap and exposed to X-ray film (Kodak) at -80°C for 6-24 hours depending on sample
run.
Following film exposure the gel lanes were cut into individual strips and loaded into
scintillation count tubes with 2mL of liquid scintillant. Tubes were then analysed for
radioactivity using a Packard Liquid Scintillation counter and these counts per minute used to
determine the level of apoptotic fragmentation in samples.
3.7.3. ANNEXIN-V-FLUOS PROPIDIUM IODIDE FLOW CYOMETRY
An early feature of apoptosis is the translocation of the phosphatidylserine (PS) receptor from
the inner to the outer surface of the plasma membrane. Annexin-V-Fluorescein (Annexin-VFLUOS, AVF) is a fluorescence binding protein with a high affinity for externalised
6
DNA Loading Buffer: 30% glycerol, 0.25% bromophenol blue, 0.25% xylene cylanol
72
phosphatidylserine receptor, and therefore a good detector of apoptotic cells. However, AVF
can also enter necrotic/late stage apoptotic cells and bind to the PS exposed on the inner cell
membrane. By staining simultaneously with Propidium Iodide (PI), a DNA stain that fluoresces
once intercalated into double stranded DNA but is impermeable to complete cell membranes,
we can distinguish apoptotic from necrotic cells as apoptotic cells will actively exclude PI until
such a time as their membrane becomes breached. Annexin-V-FLUOS, Propidium Iodide
double staining is able to detect healthy, apoptotic and late stage apoptotic/necrotic cells and
was performed using the Roche Applied Science Annexin-V-FLUOS detection kit cat#1 828 681.
TABLE 9 CELL STAINING WITH ANNEXIN-V/PROPIDIUM IODIDE (AV/PI) DOUBLE STAIN
Cell Type
AVF Staining
PI Staining
Normal Cells
Negative
Negative
Necrotic Cells
Positive
Positive
Apoptotic Cells
Positive
Negative
Cells were seeded in 6-well plates, n=4, at appropriate seeding rates, and grown for 48 hours
with an extra set of 3 wells required for machine setup. Following 48 hours of growth the
media was removed and replaced with 2mL of treatment media containing; Normal Media
(Setup Cells), Vehicle control (DMSO), 10µM PXD or 30µM PXD. After 24 and 48 hours
treatment, the media was removed, placed in a 2mL eppendorf tube and centrifuged at 400g
for 5 minutes. Cells had 1mL of 1xPBS added, then aspirated and centrifuged at 400g for 5
minutes, before 500µL 0.5% Trypsin-EDTA solution was added to each well and the cells
incubated at 37°C 5% CO2 for 5 minutes. 200µL of complete media containing FBS was then
added to inhibit further trypsin activity. Cells were then washed down the face of the 6 well
plate, washing off any stuck cells, before being centrifuged at 400g for 5 minutes. Media was
then removed from the cell pellet and pellet resuspended in 100µL of appropriate labelling
73
solution. Each time AV/PI staining was performed a set of setup cells was included in the
extraction and treatment run, grown in normal media. These cells allowed for analysis
between runs and correct FACS machine setup.
TABLE 10 CELL STAINING SETUP FOR AV/PI
Cell Type
Stain
Control Cells (n=1)
No AVF and No PI
Control Cells (n=1)
AVF Only
Control Cells (n=1)
PI Only
Vehicle Control Cells (n=4)
AVF and PI
Treatment Cells (n=4)
AVF and PI
TABLE 11 AV/PI LABELLING SOLUTIONS
Staining Buffers
Labelling Solution Recipe
No AVF + No PI
Added 100µl Incubation buffer to cell pellet
AVF Only
Mixed 204µl Incubation buffer + 4µl Annexin-V (enough for 1 sample)
Added 100µl to cell pellet
PI Only
Mixed 204µl Incubation buffer7 + 4µl Propidium Iodide8 (enough for 1
sample) Added 100µl to cell pellet
AVF + PI
Mixed 1ml Incubation buffer + 100µl Annexin-V + 100µl Propidium
Iodide (enough for 10 samples) Added 100µl to cell pellet
After 100µL of appropriate buffer was added to each sample, cells were incubated for 15
minutes at 25°C in darkness. Incubation buffer was added with volume depending on cell
pellet size, between 200 and 400µL was used to ensure an suitable event rate through the
7
8
Incubation Buffer: 10mM HEPES/NaOH, pH 7.4, 140mM NaCl, 5mM CaCl2 (keeps for 3 months at 4°C)
PI Solution: Stock solution 50µg/ml in dH2O (PI is light sensitive and was stored at 4°C in the dark)
74
FACSCalibur (BD Biosciences cat#342976) at low speed. The flow cytometer was configured
using a 488nm excitation and a 515nm bandpass filter for fluorescein detection and a 600nm
filter for PI detection. Electronic compensation of the instrument was required to exclude
overlapping of the two emission spectra and was performed by the FACS specialist, Dr Kathy
Heel from the Centre for Microscopy, Characterisation & Analysis, University of Western
Australia. Post experimental graphing was performed using FlowJo software version 7.2.5,
engine
1.999995.
Once
cell
populations
were
gated
appropriately
to
exclude
fragments/doublets/triplets the cells fell within the 3 areas, No staining (live cells), Annexin-VFLUOS only (Early apoptotic cells) and double staining of Annexin-V-FLUOS and Propidium
Iodide (Late stage apoptosis/Necrotic death). The % of cells within each set of gates was then
graphed.
3.7.4. SYBR GOLD FRAGMENTATION ANALYSIS
SYBR Gold staining was performed as a measure of qualitating DNA quality and checking 3’-end
fragmentation laddering. SYBR Gold (Invitrogen cat#S11494) is a DNA intercalating agent that
allows for detection of DNA in a cell. Apoptosis produces a characteristic laddering effect
within the cellular DNA due to initiation of endogenous endonuclease activity on sites between
nucleosomes. Nucleosomes lie approximately 200 base pairs (bp) apart and the result of
endonuclease activation is 180bp internucleosomal fragments and multiples thereof. Analysis
of cellular DNA through an agarose gel provides a characteristic laddering effect if cells are
undergoing apoptosis, or a smear if the cells are necrotic. DNA was extracted as previously
described9 and 1µg of DNA was determined by spectrophotometer then placed into a 2%
agarose Tris-acetate EDTA (TAE) gel with 1µL of Loading Buffer10 (Promega Cat# G1881) per
9
3.7.1 DNA Extraction Page: 69
Loading Buffer: 0.4% orange G, 0.03% bromophenol blue, 0.03% xylene cyanol FF, 15% Ficoll® 400,
10mM Tris-HCl (pH 7.5) and 50mM EDTA (pH 8.0)
10
75
5µL of sample. 40 volts was applied to the gel over a period of 4 hours to allow for high
resolution of the samples before the gel was placed into a container with 1xSYBR Gold11 TAE12
buffer covering the gel completely then placed in the dark for 40 minutes. Following staining
the gel was placed on a Kodak Imager 2000 and excited at 465nm while emission was read at
535nm.
FIGURE 12 LNCAP SYBR GOLD VISUALISED DNA FRAGMENTATION WITH DNA LADDER, LOW WEIGHT
DNA FRAGMENTATION IS VISIBLE
3.7.5. JC-1 MITOCHONDRIAL POTENTIAL ASSAY
At the onset of apoptosis, the mitochondrial membrane is rapidly depolarised (Hong et al.
2004). Accompanying this, 5,5’,6,6’-tetrachloro-1,1’,3,3’-tetraethylbenzimidazolcarbo-cyanine
11
12
SYBR Gold Stock 10,000x made up to 1x in TAE
40 mM Tris-acetate, 1 mM EDTA pH 7.5–8.0
76
iodide (JC-1) incorporates into mitochondria and forms monomers that fluoresce green
(520nm), whereas at high membrane potentials it forms J-aggregates which fluoresce red
(590nm) (Salvioli et al. 1997). Therefore the ratio of aggregated versus monomeric JC-1 gives a
quantitative representation of mitochondrial membrane permeability such that a low
red:green ratio is indicative of apoptosis (Zamzami et al. 2000; Zamzami et al. 2000). The JC-1
mitochondrial assay (Invitrogen, cat#M34152) assesses mitochondrial membrane potential in
live cells and allows for quantification of depolarisation of mitochondria, an event
characteristic of early stage cell death signalling. JC-1 penetrates the cytosol of eukaryotic cells
and exhibits potential-dependent accumulation in mitochondria, indicated by a fluorescence
emission shift from green (~529nm) to red (~590nm). Consequently, mitochondrial
depolarisation is indicated by a decrease in the red/green fluorescence intensity ratio. As a
positive control the kit used carbonylcyanide-4-trifluoromethoxyphenylhydrazone (FCCP) a
potent mitochondrial depolarising agent.
Cells were seeded with appropriate number onto white 96 well fluorescent plates (Greiner BioOne cat#675083) and following 48 hours of growth media was removed and appropriate
treatment used with cell control n=4 and no cell controls n=2. Four treatments were used with;
normal media for FCCP staining, vehicle control (DMSO), 10µM PXD and 30µM PXD over 24
and 48 treatment times. JC-1 staining solution was made up each assay by combining 2.5mM
JC-1 stock solution (Molecular Probes cat#T-3168) 1:75 in media without FBS. FCCP staining
solution was made up using 1:75 JC-1 buffer with 1:66 FCCP (50mM stock) in media without
FBS.
Following treatment with Phenoxodiol the media was removed and either 50µL JC-1 or 50µL of
JC-1/FCCP staining solution was added to each well and incubated at 37°C 5% CO 2 covered to
protect the plate from light, for one hour. Following incubation staining solution was removed
and 200μl PBS with 5% Bovine Serum Albumin (BSA) was added to each well to quench excess
77
fluorescence. The plate was incubated for 5 minutes at 37°C 5% CO2 before the PBS/BSA
solution was removed and 100µL PBS added to each well. The assay was read on a BMG
Labtech Fluostar Optima fluorescent plate reader with the plate chamber set to 37°C and each
well measured in a 5x5 cross-section. A set gain level was used between all cell lines and time
points. Excitation/Emission was measured from the top, green fluorescence was measured at
485nm excitation and 520nm emission while red fluorescence was measured at 544nm
excitation and 590nm emission. Red results were divided by green to get the ratio of red to
green emission, with a low ratio indicated high depolarisation rate and early apoptotic
induction.
3.7.6. CASPASE-3 ACTIVITY ASSAY
Activated Caspase-3 activity was measured using the EnzChek Capase-3 Assay Kit #2 from
Invitrogen (cat#E13184) which contains a rhodamine 110 substrate compound Z-DEVD-R110
which is turned, by activated Caspase-3, into the green fluorescent rhodamine 110. Cells were
plated onto a black fluorescent plate (Greiner Bio-One cat#675086) at the appropriate level
and grown for 48 hours in normal media. Media was aspirated and replaced with treatment
media; DMSO Control, 10µM and 30µM PXD n=4 plus n=2 no cell controls.
Caspase-3 staining dye was prepared with 100µL per sample containing; 20µL 5x Reaction
Buffer13, 0.5µL Dithiothreitol (DTT), 5µL 20x Cell Lysis Buffer14, 0.5µL Z-DEVD-R110 substrate
and 74µL of deionised water. Following 24 and 48 hours of total exposure to treatment media,
media was aspirated from the cells and replaced with 100µL of Caspase-3 staining dye per
sample. 96 well plates were covered with foil and placed on ice to incubate for 30 minutes
13
14
Reaction buffer: 20mL of 50mM PIPES, pH 7.4, 10mM EDTA, 0.5% CHAPS
Cell lysis buffer: 1.5mL of 200mM TRIS, pH 7.5, 2M NaCl, 20mM EDTA, 0.2% Triton-X-100
78
before being incubated at room temperature for a following 30 minutes. The assay was read
on a BMG Labtech Fluostar Optima fluorescent plate reader with the plate chamber set to 37°C
and each well measured in a 5x5 cross-section.
Excitation/Emission was measured from the top with fluorescence occurring when activated
Caspase-3 had converted Z-DEVD-R110 into R110. Cells were measured at 496nm excitation
and 520nm emission with a set gain level between cell lines and time points used.
3.8.
CELL CYCLE ANALYSIS
To determine if Phenoxodiol induced any potent cell cycle effects propidium iodide cell cycle
analysis was performed using a BD FACS Canto II flow cytometer (Becton Dickinson, Cat#
338962). Cells were analysed for DNA content through fluorescence of propidium iodide which
increases in fluorescence by 20 to 30 fold once intercalated into nucleic acids. Cells are
measured with a given fluorescence level and through shape and size analysis G1 content can
be attributed to a given population, G2 is assumed to be between 1.9 and 2.1 fold greater in
DNA fluorescence with S phase cells between the two population peaks. Analysis of forward
scatter (FSC) and side scatter (SSC) area (-A), height (-H) and width (-W) in the cell population
allowed for determination of cell size and cell complexity/granularity respectively. Using
complexity and cell size allowed cell fragments, doublets and triplets to be discarded and
gated out of analysis with a minimum of 10,000 individual cells within gated regions being
counted for significance to be reached.
79
3.8.1. CELL CYCLE PREPARATION
Cells were plated at 1x105 cell/mL with 5mL total media in 100mm plates (BD Falcon
cat#353003) and grown for 48 hours before appropriate treatment was used; Normal media,
vehicle control, 10µM and 30µM PXD (n=4). Cells were grown for 24 and 48 hours then media
was aspirated and centrifuged for 4 minutes at 400g while dishes were coated with 4mL of 1x
PBS and washed. Media was removed from the cell pellet and replaced with 1x PBS aspirated
from cell plates and centrifuged again for 4 minutes at 400g while 3mL of trypsin-EDTA was
placed into each plate and incubated for at least 5 minutes at 37°C 5% CO2.
Once trypsinisation had occurred, determined by a majority of cells separating, 3mL of 1x PBS
was added to each plate and cells washed then placed into tubes containing the cell pellet
from media and 1x PBS wash centrifugation. Cells were centrifuged for 5 minutes at 400g
,leaving a large cell pellet, before media was aspirated leaving just the pellet behind. Cells
were then gently vortexed while a 4mL, -20°C, 70% ethanol 30% water mixture was added
drop-wise into the vortexing cells causing fixation. Cells were stored at 4°C for up to 3 months
in the 70% ethanol solution with longer storage possible although not performed in this
experiment.
Following fixation, on the day of analysis, cells were centrifuged out of the 70% ethanol
solution at 800g for 5 minutes, a higher g rating is required as the ethanol partly crenates the
cells and prevents successful centrifugation at the lower 400g rating previously used. Following
centrifugation ethanol was removed and replaced with 3mL of 1x PBS before being centrifuged
at 400g for 5 minutes. PBS was aspirated and again replaced with 1x PBS and centrifuged for 5
minutes at 400g to ensure complete cell rehydration. Following second and final rehydration
step the cell had 1x PBS removed and replaced by between 400 and 1mL of propidium iodide
80
staining solution before being incubated for 30 minutes at 37°C to allow RNAse activity to
occur.
TABLE 12 PROPIDIUM IODIDE STAINING SOLUTION PER 30ML
Chemical
Volume
Fetal Bovine Serum
600µL
Sodium Citrate
30mg
EDTA
2.3mg
Triton X-100
9µL
Propidium Iodide
1.5mg
RNAse A (10u/µL)
15µL
H2O
29376µL
3.8.2. CELL CYCLE FLOW CYTOMETRY
When cell cycle flow cytometry was performed, negative controls were used to setup the
machine to within parameters to allow for accurate gating, assessment and analysis of
samples. Samples were optimally analysed at between 100 and 200 events per second, greater
events per second were prevented by addition of extra PI buffer and low levels were
prevented by using low initial volumes of buffer to stain the cells. The events per second and
gating used during analysis was critical in increasing resolution sampling. Cells were first
removed from their tubes and, before use in the BD FACS Canto II, each sample was pipetted
through a 30µm nylon mesh, to remove cell clumping, and was then placed into a FACS sample
tube and run through the FACS Canto II on low speed setting which significantly increased cell
resolution over the medium or high speed sampling. The flow cytometer fluoresced cells at
81
488nm excitation and detected at 619nm emission. Cells were detected and recorded in linear
mode.
Once cell fluorescence was detected the laser voltage settings were adjusted to move cells into
the gated area, voltage was adjusted per run dependent upon cell type and moved cells on the
PI_DNA-A (y-axis) versus PI-A (x-axis) graph towards a 45 degree. Once the voltage was set,
cells were gated to exclude doublets, triplets and cell debris using a SSC-A versus FSC-A graph.
FSC-A and FSC-H population counts were determined by the SSC-A versus FSC-A graph and the
populations were then equalised by altering FSC area scaling until each was displaying a close
(≥95%) population count. FSC was then determined to be accurately configured. Once FSC-A
and FSC-H had been configured PI_DNA-A and PI-DNA-H detection was equalised by gating a
SSC-A versus PI_DNA-A graph and placing a gate over the G1 population and using this
population to determine the PI_DNA-A and PI_DNA-H population, which was configured by
altering blue laser scaling until the population counts were close (≥95%) to each other and it
was determined that cells were accurately configured.
82
FIGURE 13 CELL POPULATIONS MULTI-GATED IN FACS DIVA SOFTWARE FOR ACQUISITION AND
ANALYSIS
Gated populations used to determine FSC and PI_DNA for set-up of the FACS Canto II to detect
DNA height and width in cells accurately.
83
FIGURE 14 CELL CYCLE POPULATIONS FOR ANALYSIS AQUIRED USING GATED POPULATIONS IN FACS
DIVA SOFTWARE
FSC and PI_DNA once they have been adjusted by altering FSC Scaling and Blue Laser scaling.
Following successful setup cells were processed through the FACS Canto II until at least 10,000
events lay within the P3 gate. It is evident from the count versus PI_DNA-A graph that the cells
have left shifted along the axis, this did not alter analysis as cell lines would shift left and right
slightly based on sort rate, volume and would be compensated for in FlowJo.
84
3.8.3. CELL CYCLE DATA ANALYSIS
Analysis of cell cycle results was performed using FlowJo software version 7.2.5, engine
1.999995. Cells were entered into a new workspace per experiment and each sample was
gated first with FSC area (x-axis) and SSC area (y-axis) and then sub-gated looking at DNA Area
(x-axis) and DNA Width (y-axis) before a cell cycle analysis was performed on the DNA subgated populations.
FIGURE 15 EXAMPLE OF CELL POPULATION GATING FOR ANALYSIS IN FLOWJO SOFTWARE
Example of cell population gating indicating how the side scatter/forward scatter population
gate can remove doublet cells, triplet cells and small debris fields while DNA width/DNA area
gate can be used to further remove cellular debris.
Once a cell cycle analysis had been performed a Watson (pragmatic) analysis was performed
with draw model sum, draw components and pattern fill turned on. With G2 populations
assumed to be between 1.95 and 2.05 of the G1 population, dependent upon the cell line, and
a serious change in G2 population and overall cell cycle shape, the Watson (pragmatic) analysis
performed better in evaluating treatment effects upon cell cycle distribution, than the DeanJett-Fox (DJF) method. As an analysis the Watson (pragmatic) makes no assumptions about the
shape of the s-phase distribution and fits the s-phase under the graph while the Dean-Jett-Fox
85
analysis assumes that the s-phase can be modelled by a second degree polynomial, requiring a
G2 peak which was lacking in some treatments. Following analysis with FlowJo results were
tabulated in Microsoft Excel 2007 and graphed using column and pie graph options.
WATSON (PRAGMATIC) C ELL CYCLE
DISTRIBUTION
D E A N - J E T T - FO X C E L L C Y C L E
D I ST R I B U T I O N
G1
G2
S
FIGURE 16 WATSON VERSUS DEAN-JETT-FOX CELL CYCLE ANALYSIS IN FLOWJO SOFTWARE
A Watson (pragmatic) versus Dean-Jett-Fox (DJF) analysis of a control cell culture, clearly visible
is the Gaussian distribution of the DJF S-phase as assumed by the G1 and G2 peak distribution
versus the Watson (pragmatic) method of S-phase distribution defined by graph shape.G1
phase populations are represented in green, S phase populations in yellow and G2 phase
populations in blue.
86
FIGURE 17 WATSON PRAGMATIC CELL CYCLE ANALYSIS OF PC3 CELLS UNDERGOING PHENOXODIOL
TREATMENT
The cell cycle of a PC3 cell treated with 30µM Phenoxodiol for 48 hours. The G2 peak is almost
undetectable and only found by constraining samples so that G2 = 1.95-2.05 x G1. Analysis with
DJF fails to determine S and G2 phases successfully, as determining a Gaussian distribution
pattern is impossible without a G2 peak. Therefore DJF analysis cannot be used to on
Phenoxodiol treated samples.
3.9.
ASSESSMENT OF RNA EXPRESSION
Analysis of RNA (ribonucleic acid) expression, targeting genes of interest, was performed
through the use of the Polymerase Chain Reaction (PCR) technique. The technique utilised the
Taq polymerase enzyme to make copies of messenger RNA (mRNA) genes, designated by
primer sequences, and to use these copies to determine changes between control and
treatment sample groups. This technique allowed measurement of expression levels of specific
genes of interest which determined whether phenoxodiol was affecting particular signalling
pathways within the cells. To perform this analysis cells would undergo RNA extraction, DNAse
Treatment, Reverse Transcription PCR (RT-PCR), RT-PCR cleanup and finally qPCR testing.
87
3.9.1. RNA EXTRACTION
Cells underwent RNA extraction as the first step in mRNA analysis. Cells were seeded in 6-well
plates, n=4, at appropriate seeding rates (stated above), and grown for 48 hours. Following 48
hours of growth the media was removed and replaced with 2mL of treatment media
containing; Vehicle control (DMSO), 10µM PXD or 30µM PXD. After 24 and 48 hours treatment,
the media was removed, placed in a 2mL eppendorf tube and centrifuged at 800g. The media
was then discarded leaving cell pellet behind while 1mL of Tri Reagent (Molecular Research
Centre Cat#TR118, now Tri Reagent RT Cat#RT111) was placed into each well. Tri Reagent
works via the Chomczynski method of RNA separation using Guanidinium thiocyanate-phenolchloroform phase separation of DNA, RNA and protein (Chomczynski and Sacchi 1987). This
method takes longer than a column based separation technique but has been shown to
increase RNA purity.
Wells were scraped with a cell scraper (Sarstedt) and pipetted into the 2mL eppendorf tube
that had their media centrifuged in, then mixed vigorously with a 1000µL pipette (Eppendorf
Cat#3121 000.120). Homogenised cells were rested at room temperature for 5 minutes before
the addition of 200µL chloroform (Sigma-Aldrich Cat#472476) was added to each sample.
Tubes were shaken for 30 seconds before resting at room temperature for 10 minutes. Tubes
were then centrifuged for 15 minutes at 12,000g 4°C with the top layer (typically 70% of the
sample) of the phase separated mixture being transferred to a new tube leaving behind a DNA
and protein solution. 500µL of isopropanol was then added to the top phase that had been
removed from each sample and was vortexed then stored overnight at 4°C.
Following overnight storage samples were centrifuged for 10 minutes at 4°C and 12000g. A
white pellet forms at the base of the tube, the supernatant is removed leaving behind the
88
pellet and 1mL of -20°C 75% Ethanol was added to each tube, samples were vortexed then
centrifuged again for 5 minutes at 4°C and 12,000g. Following the 75% ethanol wash step the
ethanol was removed leaving behind a pellet which was air dried in a class 1 biohazard hood
for approximately 5 minutes before the addition of 50µL of DEPC treated water (ribonuclease
free).
3.9.2. RNA INTEGRITY
RNA sample purity and concentration was determined by taking a 2µL aliquot of each sample
for spectrophotometer analysis on a Nanodrop 1000 spectrophotometer (ThermoScientific
cat#ND1000) with samples outside of a 1.7-2 260/280nm range discarded and those within
stored at -80°C. RNA also underwent electrophoresis on a 1.5% agarose-TAE gel with ethidium
bromide or Sybrsafe (Invitrogen Cat#S33102), as the preferred nucleotide fluorescent stain,
added to the gel before it was set in a mould. Five microliters of RNA was combined with 1µL
of 6x Blue/Orange Loading Dye (Promega Cat#G1881) and placed in the gel under a 1X TAE
buffer at 100 volts for 30 minutes. Visualisation of the 28s, 18s, and 5s ribosomal bands was
established under ultra violet light using a Kodak 2000MM Image Station. Samples were stored
at -80°C. Sample purity and concentration was determined by taking a 2µL aliquot of each
sample for spectrophotometer analysis on a Nanodrop spectrophotometer with samples
outside of a 1.7-1.9 260/280nm range discarded.
3.9.3. DNASE TREATMENT
RNA was DNase treated to remove any genomic DNA before use, with an RQ1 RNase-free
DNase treatment (Promega Cat#M6101). The quantity of sample required for 2µg of RNA was
determined with a spectrophotometer then 2µL of RQ1 DNase and 1µL of RQ1 DNase 10X
89
Reaction buffer15 was added to the RNA in 200µL clear Axygen PCR tubes (Fisher Biotec
Cat#PCR-02-C) and made up to a final volume of 10µL of DEPC treated, nuclease free water.
The solution was vortexed and placed in a PTC-100 thermocycler (MJ Research Inc Cat#PTC100 ) at 37°C for 30 minutes. Following DNase treatment 1µL of DNase Stop solution16 was
added to each tube, mixed via vortexing and the tubes returned to the thermal cycler for
another 10 minutes at 65°C to inhibit the activity of the DNase leaving a 2µg solution of RNA
with a volume of 11µL.
TABLE 13 RQ1 DNASE TREATMENT REACTION COMPONENTS
Reagent
Volume
2µg RNA
__µL
RQ1 RNase Free DNase
2µL
RQ1 10x Reaction Buffer
1µL
H2O
Volume to final reaction volume of 10µL
3.9.4. REVERSE TRANSCRIPTASE POLYMERASE CHAIN REACTION
Reverse transcriptase polymerase chain reaction (RT PCR) is a technique which takes purified
RNA and amplifies the low yield RNA whilst also producing a complementary strand referred to
as complementary DNA (cDNA). This strand allows for increased stability of mRNA as well as
the subsequent amplification increasing the ability to perform PCR analysis of expression of
specific mRNA sequences within a sample.
One microgram of DNase treated RNA was processed through RT PCR using a M-MLV Reverse
Transcriptase RNase H Minus, Point Mutant Taq enzyme (Promega Cat#M3682). Briefly, 1g of
15
16
400mM Tris-HCl (pH 8.0), 100mM MgSO4 and 10mM CaCl2.
20mM EGTA (pH 8.0)
90
DNase treated RNA (5.5µL) had 8µL DEPC H2O and 0.5µL Random Primers (Promega Cat#
C1181) added, mixed gently and heated to 70°C for five minutes and incubated on ice for five
minutes. The following components were then added in order to give a total reaction volume
of 25µL.
TABLE 14 RT PCR REACTION COMPONENTS
Reagent
Volume
M-MLV 5x Reaction Buffer17
5µL
10mM dNTP’s18
1.3µL
M-MLV RNase H Minus, Point Mutant
1µL
RNasin19
1µL
DEPC H2O
2.7µL
Following the addition of all components into the appropriate PCR tube the PTC-100
thermocycler was programmed to run for 10 minutes at 25°C, 50 minutes at 55°C and finally
for 15 minutes at 70°C before holding at 4°C, samples were then stored at -20°0C.
3.9.5. POST PCR CLEAN-UP
Following RT PCR a clean-up of excess dNTP’s, reaction salts, M-MLV Taq and unused primers
from the sample cDNA was performed using an UltraClean PCR clean-up kit (MoBio
Laboratories, Inc. Cat#12500). Each PCR reaction (25µL) was treated with a 5-times volume
(125µl) of SpinBind (Guanidine-HCl/Isopropanol), mixed well and transferred to a spin filter
unit and centrifuged at 13000rpm for 30 seconds. The flow-through was discarded, the filter
basket returned to the tube and 300µl of SpinClean (<80% Ethanol solution) added to the filter.
17
At final concentration: 50mM Tris-HCl (pH 8.3), 75mM KCl, 3mM MgCl2, 10mM DTT
Promega Cat#C1145
19
Promega Cat#N2111
18
91
The spin column was then spun once for 30 seconds, with flowthrough discarded, then again
for one minute. The filter basket was then removed to a clean 1.5ml tube and 50µl of Elution
Buffer (10mM Tris, pH 8.0) added directly to the filter. The spin column was centrifuged at
13000 rpm for 30 seconds, the filter basket disposed of, with the resultant cDNA stored in a
freezer at -20°C.
3.9.6. PRIMER DESIGN
Before qPCR can occur primer sets must be designed to bind to the cDNA of the genes of
interest exclusively. Primers are base pair sequences of arginine, guanine, cytosine and
tyrosine that are ~20 base pairs long and are designed to be sequence specific to a gene of
interest. Human nucleotide sequences were fetched from the Pubmed database
(http://www.ncbi.nlm.nih.gov/pubmed/) and primers were checked or designed to bind to the
gene sequence, or sequences if there were multiple available. Primers not available from
published
sources
were
designed
using
the
http://frodo.wi.mit.edu/cgi-bin/primer3/primer3_www.cgi)
online
software
and
rechecked
(Primer
using
3,
local
software. All primers underwent a BLAST (http://blast.ncbi.nlm.nih.gov/Blast.cgi) search check
to ensure no cross reactivity. Primers not available from published works were designed to
span at least one intron/exon boundary with a primer sequence specifically crossing the
boundary to prevent the amplification of genomic DNA and ensure only cDNA replication.
Wherever possible the GC content between primer sets was kept as close as possible to allow
for equivalent annealing temperatures while primer designed with large regions of matching
base pairs were excluded to limit primer dimerisation as much as possible.
92
TABLE 15 PRIMER SEQUENCES, PRODUCT SIZE AND ANNEALING TEMPERATURE
Gene
Sequence
Product Size
Annealing
(Base Pairs)
Temperature
Forward: GATCACGCTGTTGTGAGTGG
AIF
179bp
61°C
218bp
52°C
167bp
61°C
167bp
61°C
322bp
55°C
165bp
62°C
185bp
60°C
225bp
60°C
194bp
52°C
333bp
61°C
181bp
56°C
183bp
56°C
Reverse: TCTTGTGCAGTTGCTTTTGC
β-Catenin
Forward: GATTTGATGGAGTTGGAC
Reverse: TGTTCTTGAGTGAAGGAC
Forward: GCTGGACATTGGACTTCCTC
Bax
Reverse: TCAGCCCATCTTCTTCCAGA
Forward: ACAATGCAGCAGCCGAGAG
Bcl-xL
Reverse: ATGTGGTGGAGCAGAGAAGG
Caspase 3
Forward:
AAGGATCCTTAATAAAGGTATCCATGGAGAACACT
Reverse:
AAAGAATTCCATCACGCATCAATTCCACAATTTCTT
Cyclin-D1
Forward: AACTACCTGGACCGCTTCCT
Reverse: CCACTTGAGCTTGTTCACCA
GAPDH
Forward: CAGAACATCATCCCTGCATCCACT
Reverse: GTTGCTGTTGAAGTCACAGGAGAC
Ki-67
Forward: AGTCAGACCCAGTGGACACC
Reverse: TGCTGCCGGTTAAGTTCTCT
Forward: CTGAAGGTCAAAGGGAATGTG
L19
Reverse: GGACAGAGTCTTGATGATCTC
p21
WAF1
Forward: CCGAAGTCAGTTCCTTGTGG
Reverse: AAGTCGAAGTTCCATCGCTCA
sFRP4
Forward: CGATCGGTGCAAGTGTAAAA
Reverse: GACTTGAGTTCGAGGGATGG
XIAP
Forward: GGGGTTCAGTTTCAAGGACA
Reverse: CGCCTTAGCTGCTCTTCAGT
3.9.7. REAL TIME QUANTITATIVE PCR ANALYSIS
Polymerase chain reaction (PCR) is a cycling of temperatures that results in the amplification of
a specific product and shows if a particular gene is transcriptionally expressed. Quantitative
PCR (qPCR) is similar to PCR but has an additional fluorescent tag which binds to double
93
stranded DNA and provides, fluorometrically, a quantitative profile on the amplified product.
Each annealed primer-cDNA set denatures into single strands within a small temperature
range according to its length, sequence and GC content (Ririe et al. 1997).
Real time quantitative PCR (qPCR) analysis was performed on the clean cDNA samples to
detect and determine levels of gene expression within the cell lines, under treatment and
vehicle control conditions, over 24 and 48 hours of treatment. In a regular PCR the mRNA is
mixed with buffer, magnesium, dNTP’s, Taq enzyme and primers then undergoes repeated
cycles of denaturation of the mRNA, annealing of the primers to the denatured mRNA and
extension, which occur at different temperatures. The product is a specifically amplified region
of mRNA.
Quantitative PCR was initially performed upon samples to provide a viable product which was
size confirmed using gel electrophoresis. Following electrophoresis samples were excised and
purified to produce a saturated PCR positive sample which was serially diluted 10 fold in 1xTE
(pH 8.0) until a set of standards had been produced which were run in duplicate in all reactions
allowing accurate quantitation of qPCR product concentration. Corbett Life Science RG3000
and RG6000 (now Qiagen Rotor-Gene Q) real time PCR thermocyclers were used to perform all
qPCR experiments. Quantitative PCR was conducted using two separate Taq enzymes; iQ Sybr
Green Supermix 2X (Biorad, Cat#170-8882) and Immolase Taq (Bioline Cat#BIO-21046).
iQ Sybr Green based qPCR’s were performed with a reaction mix of; 5μl 2X iQ Sybr Green
Supermix20, 1μl forward primer, 1μl reverse primer, 1μl DEPC H2O and 2μl cDNA. Immolase
20
100mM KCl, 40mM Tris-HCI, pH 8.4, 0.4mM of each dNTP (dATP, dCTP, dGTP, dTTP), iTaq DNA
polymerase 50 units/ml, 6mM MgCl2, SYBR Green I, 20nM fluoresein, and stabilisers.
94
reactions had a reaction mix as seen in Table 14. Mixtures were denatured at 95°C for 10
minutes before cycling at the following steps; 95°C denature for 2 seconds, annealing at
temperate specified in Table 1321 for 15 seconds and extension at 72°C for 10 seconds. Cycling
was performed for between 40 and 45 total cycles after which a stepwise melt curve was
performed.
TABLE 16 IMMOLASE TAQ QPCR REACTION MIX
Component
Volume for 10µL Reaction (µL)
10X Buffer22
1.0
Immolase Taq
0.1
50mM MgCl2
0.5-0.8 Depending on primer set
Sybr Green
0.5
10mM dNTP’s
0.4
Forward Primer
0.4
Reverse Primer
0.4
DEPC H2O
4.7-5.7
cDNA
1.0-2.0
Melt curves were performed for each qPCR reaction by heating the reaction in a 0.5°C
stepwise fashion from 72°C up to 99°C to ensure the exclusive melting of the desired product.
The fluorescence of the sample is plotted against the temperature, with a typical spike in
fluorescence occurring at the temperature at which the product denatures. Accordingly, the
desired product can be identified, as well as any contaminants such as primer dimerisation
(primer-to-primer binding). The melt curve and standards R2 value were used to determine the
quality of the sample run and reactions with high levels of multiple product formation and low
R2 values were not quantified.
21
22
Page Number 61
Proprietary Buffer, reagent mix unavailable.
95
FIGURE 18 EXAMPLE OF A CYCLIN-D1 MELT CURVE
3.9.8. GEL ELECTROPHORESIS AND EXTRACTION
Samples underwent gel electrophoresis against a known 100bp DNA Step Ladder (Promega
Cat#G6951) to confirm product size and further ensure if primer dimerisation did occur at the
same melt temperature as the gene of interest it was detected due to size differences.
Samples underwent electrophoresis on a 1.5% agarose-TAE gel with 5µL per 50mL 10,000x
GelRed (Biotium Cat#41002) nucleotide fluorescent stain added to the gel before it was set in a
mould. Five microliters of sample was combined with 1µL of 6x Blue/Orange Loading Dye
(Promega Cat#G1881) and placed in the gel under a 1x TAE buffer at 100 volts for 30 minutes.
Visualisation of the 100bp DNA Step Ladder bands was established under ultra violet light
using a UV illuminator and the appropriate product band was excised from the gel carefully to
prevent UV irradiation.
Using an Axyprep DNA gel extraction kit (Axygen Biosciences Cat#AP-GX-250) the excised gel
band was placed into a pre weighed 2mL eppendorf tube, and the total weight was calculated
and then removed from tube-only weight to work out gel-only weight. Once the gel weight
was determined the gel extraction kit used an assumption of 100mg = 100µL of reagent
volume. Once weighed, 3x sample volume of buffer DE-A was added to the tube and the
96
gel/buffer solution was heated at 75°C and vortexed until gel was fully solubilised. Once mixed
0.5x DE-A volume of buffer DE-B was added and if the fragment was <400bp then a further 1x
sample volume of isopropanol was added to the tube. An Axyprep spin column was placed into
a new 2mL eppendorf tube and the dissolved gel solution placed into the spin column then
centrifuged at 12,000g for 1 minute. Filtrate was discarded and 500µL of buffer W1 was added
through the spin column, sample was centrifuged at 12,000g for 30 seconds. The filtrate was
discarded again and 700µL of buffer W2 added to the column and centrifuged for a further 30
seconds at 12,000g. The filtrate was again discarded and the column centrifuged for 60
seconds at 12,000g to ensure complete drying of the filter and removal of any residual buffer.
The column was then placed into a new 1.5mL micro-centrifuge tube and 25µL of elutent
pipetted onto the membrane, left to stand for 1 minute then centrifuged at 12,000g for 1
minute. The DNA was stored at -20°C in a post-PCR room for potential sequencing and to be
serially diluted into PCR standards.
3.9.9. QPCR STANDARD PRODUCTION
Standards were made for each gene to enable accurate quantification of products for qPCR
reactions. Tenfold serial dilutions were made from gel extracted DNA standards were then
used simultaneously in qPCRs to establish the relative expression of the cDNA of interest for
each of the primer sets tested. Standards were cycled in duplicate and a standard curve was
created to generate an R2≥0.99, with the experimental samples assigned a value depending on
where they fit on the standard curve. By standardising these results against the human
ribosomal L19 internal control, the amount of product in each sample is quantifiable. Samples
were assigned values via rotor-gene software, Corbett Rotor-Gene 6000 application software,
version 1.7 (Build 87).
97
TABLE 17 QPCR STANDARDS SERIAL DILUTION
Standard
Serial Dilution
A
Gel extracted sample
B
1µL Standard A + 9µL 1x TE
C
1µL Standard B + 9µL 1x TE
D
1µL Standard C + 9µL 1x TE
E
1µL Standard D + 9µL 1x TE
F
1µL Standard E + 9µL 1x TE
G
1µL Standard F + 9µL 1x TE
H
1µL Standard G + 9µL 1x TE
FIGURE 19 QPCR STANDARDS CYCLING ASSESSMENT
Standards were 10 fold dilutions of each other, those with samples within their bounds were
selected as the most appropriate standards for samples. Cycling analysis of standards allowed
inappropriate replicates to be removed before analysis. Y axis contains concentration while X
axis displays at which cycle detectable amplification of product began.
98
FIGURE 20 STANDARDS ANALYSIS BY CORBETT ROTOR-GENE 6000 SOFTWARE.
Standards analysis from the Corbett rotor-gene software uses the standards to design a line of
best fit with an R2≤.99 and apply this to cycling samples. Y axis contains a given count while X
axis displays listed concentrations set by user.
FIGURE 21 QPCR FULL RUN STANDARDS AND SAMPLES
A qPCR experiment displaying both standards and samples with each other, negative samples
are not displayed as they do not cycle.
99
FIGURE 22 STANDARDS AND SAMPLES ANALYSIS BY CORBETT ROTOR-GENE 6000 SOFTWARE
A completed run with samples and standards displayed against the generated line of best fit.
3.10.
ASSESSMENT OF PROTEIN EXPRESSION
Protein expression changes following treatment with Phenoxodiol was determined by analysis
using the western blot technique. Protein analysis by Western blot shows whether a particular
protein is translationally expressed for the epitote that is specific to the particular antibody
used. Western blot analysis was performed using polyacrylamide gel electrophoresis (PAGE) to
separate different size proteins which were then transferred to a nitrocellulose membrane
where primary antibodies were bound to the epitope of proteins followed by a secondary
tagged antibody to visualise proteins of interest. The cytoskeleton protein β-actin was used as
a control standard to prevent variation in results due to loading or concentration differences.
100
3.10.1.
PROTEIN EXTRACTION
Protein was extracted from all cell lines using a Radioimmunoprecipitation (RIPA) Buffer
consisting of 150mM NaCl, 50mM Tris-HCl (pH 7.5), 1% Triton X-100, 0.5% Sodium
Deoxycholate, 0.1% SDS and 0.1mM Phenylmethylsulfonyl fluoride (PMSF) in dH2O. Cells were
seeded into either 6 well, or 10cm dishes, at appropriate concentrations then covered with
2mL of complete cell media and incubated at 37°C 5% CO2 for 48 hours. Media was then
aspirated off and replaced with appropriate volumes of treatment media (2mL per well, in a 6
well plate, or 5mL per 10cm dishes) and incubated for either 24 or 48 hours at 37°C 5% CO2.
RIPA buffer was prepared directly before addition to the wells containing cells for every
extraction by addition in order of; 50mg sodium deoxycholate, 1.5ml 1M NaCl, 100μl of 10%
SDS, 500μl 1M Tris-HCl (pH 7.5) and 100μl Triton X-100 dissolved in 7.8ml of dH2O. Following
preparation buffer was placed into -20°C to chill to as cold as possible without freezing before
use, immediately prior to use 10μl of -20°C 100mM PMSF was added.
To prepare samples for use media was aspirated off the 6-well/10cm dishes and placed into a
2mL/15mL eppendorf tube and centrifuged for 5 minutes at 800g. The media in the tubes was
then aspirated post centrifuge leaving the cell pellet behind. Either 1mL or 3mL of sterile 4°C
1xPBS (pH 7.4) buffer was added to each well/dish and then the well/dish was scraped with a
cell scraper and resulting cell and 1xPBS mixture pipetted into the 2mL/15mL eppendorf tubes
where appropriate, then centrifuged for 5 minutes at 800g. The addition of the 1xPBS and
scrape steps was critical for removing excess FBS from samples before RIPA buffer was added,
as FBS was found to interfere with accurate protein extraction causing the western blot
analysis to fail. Once the cell pellet was in the tube and the supernatant removed, 50µL RIPA
buffer was added to 6-well plate samples or 200µL RIPA buffer added to 10cm dish samples, all
were mixed and cells lysed via vigorous pipetting and vortexing before being transferred to
new 1.5mL eppendorf tubes. Samples were then left to sit on ice at for 30 minutes before
101
being placed into a 4°C centrifuge and centrifuged at maximum speed for 5 minutes.
Supernatant was transferred to a new tube and stored at -80°C for use in protein
quantification and western blot analysis.
3.10.2.
BRADFORD PROTEIN QUANTIFICATION ASSAY
The Bradford protein quantification assay uses a standard curve to help determine
concentration of protein in samples. Protein standards were made by serial dilution of
acetylated Bovine Serum Albumin (BSA 10mg/ml) to create standards ranging from 500 to
100μg/ml protein.
TABLE 18 BRADFORD PROTEIN STANDARDS SETUP
Solution
Concentration of BSA
Components
A
500µg/mL
50µL BSA (10mg/mL) + 950µL
0.01xPBS
B
400µg/mL
760µL A + 190µL 0.01xPBS
C
300µg/mL
750µL B + 250µL 0.01xPBS
D
200µg/mL
600µL C + 300µL 0.01xPBS
E
100µg/mL
400µL D + 400µL 0.01xPBS
Protein samples were diluted 1:20 in 0.01M PBS and 10μl transferred into glass tubes. In
duplicate, 10μl of each Standard was transferred to glass tubes, to which 200μl of protein dye
was added, vortexed and incubate for 5 minutes at room temperature. The colour of the
protein samples was checked against the Standards to ensure their colour would fit within the
standard curve. 0.01M PBS was used to zero the nanodrop spectrophotometer. Standards
were analysed by 2µL addition to the spectrophotometer and using a visible light analysis a
standard curve produced. Sample concentration was determined by adding 2μl to the
102
nanodrop and analysing with the spectrophotometer, and the concentration of protein (μg/ml)
was extrapolated from the standard curve and then multiplied by 20 to remove the initial
dilution factor.
3.10.3.
SDS-PAGE WESTERN BLOT ANALYSIS
Protein analysis was performed using a western blot analysis whereby protein expression is
determined through antibody binding to specific epitopes on target proteins followed by
chemo-luminescent visualisation of the antibody binding. Once visualised, quantification of
samples can occur by the comparative luminosity of treatment versus control accounting for
the β-actin loading control. Sodium Dodecyl Sulfate Polyacrylamide Gel Electrophoresis (SDSPAGE) is the process where protein samples are separated into fragments according to their
molecular weight, this occurs by loading protein into an acrylamide gel and running a current
through the gel. Proteins move through the gel at a speed designated by their molecular
weight, charge and conformation. A Biorad Mini Trans-Blot Electrophoretic Transfer Cell
apparatus (Bio-Rad Cat#170-3930) was used to perform both SDS-PAGE protein separation and
the nitrocellulose membrane transfer.
Glass plates with 1mm spacers, their accompanying smaller glass plates and the 12 well combs
were cleaned with 70% ethanol, allowed to dry then aligned and placed into the stacking gel
locking mechanism. A 12% acrylamide separating gel was poured into the space between the
two glass plates until separating gel was approximately 1.5cm below the top of the glass
plates, 200µL of dH2O was then added to each corner of the separating gel and left for 20
minutes to set. Blotting paper was then used to remove excess dH2O and unset gel, if the
separating gel was not completely flat across the top surface the gel was destroyed, glass
plates cleaned and another separating gel poured. Once the separating gel was correctly
103
formed, a 4% acrylamide stacking gel was poured across the top of the separating gel, with the
12 well comb placed into it, ensuring no air bubbles formed at the base of the comb, to
prevent issues with protein loading.
TABLE 19 SDS-PAGE/ACRYLAMIDE GEL RECIPE
Gel Ingredient
Volume, 12%
Volume, 4% Stacking
Separating Gel
Gel
ddH2O
3.3mL
6.1mL
1.5M Tris-HCl (pH 8.8)
2.5mL
-
0.5M Tris-HCl (pH 8.8)
-
2.5mL
10% SDS
100µL
100µL
30% Acrylamide/Bis23
4.0mL
1.33mL
10% Ammonium Persulfate (APS)
100µL
50µL
Tetramethylethylenediamine
10µL
10µL
(TEMED)24
During separating and stacking gel solution mixing the; dH2O, 1.5M/0.5M Tris-HCl (pH 8.8),
30% Acrylamide/Bis, 10% SDS was added and mixed together and just before the pouring of
the gel solution the N,N,N',N'-Tetramethylethylenediamine (TEMED) and 10% ammonium
persulfate (APS) was added then quickly poured as TEMED and APS cause the gel to set. The
separating and stacking gel solutions were mixed in 15mL falcon tubes and discarded in
incineration containers once set or unused as acrylamide is a neurotoxin and cannot be
discarded safely in another manner. The stacking gel was allowed to set for 20 minutes then
glass plates were removed from gel locking mechanism and placed into the electrophoresis
tank in a new gel electrophoresis locking mechanism. Gels were placed into the
electrophoresis tank with the smaller plates facing each other, and the reservoir between
23
24
Cat# A3574 – Sigma Aldrich
Cat# T9281 – Sigma Aldrich
104
them filled with 4°C electrode buffer25. Electrode Buffer was also added to the electrophoresis
tank to a final depth of 3cm to cover the electrodes within the bottom of the tank and placed
between the two glass plates creating a layer of buffer over the comb/well level.
Three separate samples per treatment group (nine samples total per time point, 18 total
samples per cell line) were thawed and 30 or 50µg of each sample was placed into new 1.5mL
eppendorf tubes. The most dilute sample equalled the largest volume and all other tubes had
0.01M PBS added up to equal the volume. Once the volume was determined an equal volume
of protein loading buffer26 was added to each sample and then gently mixed before being
placed in a 95°C heat block for a maximum of 5 minutes, to denature the protein. Samples
were removed from the heat block and loaded, with a Hamilton microlitre syringe, into
consecutive wells in the set acrylamide stacking gel.
TABLE 20 EXAMPLE OF PROTEIN SAMPLE PREPARATION
Sample
Volume for 30µg
0.01M PBS
Loading Buffer26
Control 1
14µL
6µL
20µL
Control 2
18µL
2µL
20µL
Control 3
12µL
8µL
20µL
PXD 10µM 1
17µL
3µL
20µL
PXD 10µM 2
16µL
4µL
20µL
PXD 10µM 3
19µL
1µL
20µL
PXD 30µM 1
20µL
-
20µL
PXD 30µM 2
12µL
8µL
20µL
PXD 30µM 3
14µL
6µL
20µL
25
25mM Tris-HCl, 200mM Glycine and 0.1% SDS
4% SDS, 2% β-mercaptoethanol, 20% v/v glycerol, 250mM Tris-HCl (pH 6.8) and 0.006%
bromophenolblue
26
105
The samples were loaded, underneath the electrode buffer25, into the stacking gel from rows
3-11 with row two containing 5µL of SeeBlue Plus 2 molecular weight standard (Invitrogen,
Cat# LC5925). After loading the protein samples and ladder, the electrodes were inserted into
the apparatus and the electrical source, with the black electrode connected to the negative
terminal and positive electrode connected to the red terminal. The apparatus was set to a
voltage of 100V until the samples reached the top of the separating gel and had started to
cross between stacking and separating gel. The voltage was then increased to 130V and left
until the 6kDa molecular weight band reached the bottom of the separating gel. Once
separated, the gels were removed from the electrophoresis tank and locking mechanism, the
glass plates were gently prised apart to prevent tearing of the gel. Once apart the stacking gel
was removed from the separating gel and the stacking gel was discarded into an incineration
container.
In preparation for the transfer of proteins to Hybond-C Extra nitrocellulose membrane
(Amersham Cat#RPN303E), 4 pieces of blotting paper and 2 nitrocellulose membranes (one for
each gel/timepoint) were cut to the required size (equal to the separating gel) and, with 6
absorbent pads, soaked in transfer buffer27. The transfer cassettes were layered with; one pad,
one piece of blotting paper, the separating gel, the nitrocellulose membrane, more blotting
paper and 2 more pads, with each component layered consecutively upon the black side of the
rack. The contents were kept saturated with transfer buffer throughout and any air bubbles
were removed from between the gel and membrane with a glass rolling pin. The transfer racks
were closed, locked and inserted into a transfer apparatus with the black side facing the back
of the apparatus. A 2-3cm magnetic stirrer was placed into the bottom of the electrophoresis
tank with the transfer apparatus and an -20°C ice brick placed in the tank. The tank was then
27
800ml dH2O, 3.03g Tris, 14.4g Glycine, 200ml Methanol
106
filled with transfer buffer27 and the electrodes attached negative to black, positive to red and
the powerpack set to produce 100V for 75 minutes.
Following the transfer the gel was discarded into an incineration container, and the
nitrocellulose membrane rinsed quickly in Tris-buffered Saline Tween-2028 solution (TBS-T)
prior to being covered with Ponceau S29 solution for 30 seconds, to reveal all transferred
proteins and ensure no air bubble affected areas. The Ponceau S was returned to the stock
bottle, and the excess removed by repeated washes of dH2O. The membrane was cut into
strips containing the desired range of protein sizes for immunoblotting. The remaining
Ponceau S was removed from the membrane with TBS-T for 5 minutes, and the membranes
stored in sealed plastic bags at 4°C until required for immunoblotting.
3.10.4.
IMMUNOBLOTTING
In order to prevent non-specific antigen binding, membranes were blocked in 5% non-fat milk
powder (Diploma) in TBS-T for one hour at room temperature. Following blocking membranes
were washed three times for five minutes in TBS-T before 2mL of appropriate primary
antibody (Table 21) was added to membrane. Membranes with primary antibodies were
incubated at 4°C overnight with gentle agitation after which membranes were washed with
TBS-T three times for 5 minutes each to remove any unbound antibody. Streptavidin horse
radish peroxidase (HRP) conjugated secondary antibodies were primarily used. Following
primary incubation membranes were washed in TBST and incubated with; XIAP, β-actin AntiMouse HRP (1:10000; Dako), AIF, Bax, Active β-catenin and sFRP4 Anti-Rabbit HRP (1:10000
Dako), Bcl-xL (1:10000 Vector BA-1000) in TBST for one hour at room temperature. After
28
29
Tris Buffered Saline/0.05% Tween-20 pH
2% Ponceau S, 2%Trichloroacetic acid
107
incubation, excess liquid was removed completely with blotting paper, and 1ml each of the
two SuperSignal West Pico Chemiluminescent Substrates (ECL) (Pierce, Cat#34080), were
combined and placed onto the membrane and incubated for 5 minutes. Excess liquid was
again removed completely with blotting paper, and the membrane bagged and imaged on an
ImageStation (Kodak, IS2000MM).
TABLE 21 ANTIBODY CONCENTRATIONS
Protein
1° Antibody Concentration
AIF
1:1000 (Cayman Cat#160773)
2° Antibody Concentration
1:10000 Anti-Rabbit HRP
(Dako)
BAX
1:200 (Abcam Cat#7977)
1:10000 Anti-Rabbit HRP
(Dako)
Β-actin
Β-catenin
1:5000(Sigma Aldrich
1:10000 Anti-Mouse HRP
Cat#A5441)
(Dako)
1:1000 (Millipore Cat#05-601)
1:10000 Anti-Rabbit HRP
(Dako)
Bcl-xL
1:200 (Abcam Cat#7974)
1:10000 Anti-Rabbit
Biotinylated (Vector BA-1000)
sFRP4
1:1000 (Millipore Cat#09-129)
1:10000 Anti Rabbit (Dako)
xIAP
1:250 (BD Transduction
1:10000 Anti-Mouse HRP
Cat#610762)
(Dako)
108
TABLE 22 IMMUNOBLOTTING PROTOCOL
Protein
AIF
Bax
Blocking
β-
β-
actin
Catenin
Bcl-xL
sFRP4
xIAP
5% non-fat milk powder in TBS-T
One hour room temperature with gentle agitation
1°
Antibod
y
Rabbit anti
Rabbit
Mouse
Mouse
Rabbit
Rabbit
Mouse
AIF
anti Bax
anti β-
anti-
anti Bcl-xL
anti
Anti xIAP
(Cayman
(Abcam
actin
active-
(Abcam
sFRP4
(BD
Cat#16077
Cat#7977
(Sigma-
β-
Cat#7974
(Millipore
Transduc
3)
)
Aldrich
catenin
)
Cat#09-
tion
1:1000
1:200
Cat#A54
(Upstate
1:200
129)
Cat#6107
TBS-T
TBS-T
41)
Cat#
TBS-T
1:1000
62)
1:5000
1:1000
TBS-T
1:250 3%
TBS-T
3% Milk
Milk TBS-
TBS-T
T
Overnight at 4°C with gentle agitation
Washing
TBS-T
3 x 5 Minutes with gentle agitation
2°
Antibod
y
1:10000
1:10000
1:10000
1:10000
Anti-
1:10000
1:10000
Anti-
Anti-
Anti-
Anti-
Rabbit
Anti-
Anti-
Rabbit
Rabbit
Mouse
Mouse
Biotinylat
Rabbit
Mouse
HRP
HRP
HRP
HRP
ed
HRP
HRP
(Dako)
(Dako)
(Dako)
(Dako)
(Vector
(Dako)
(Dako)
BA-1000)
One hour at Room Temp with gentle agitation
Washing
TBS-T
3 x 5 Minutes with gentle agitation
3.10.5.
QUANTIFICATION OF WESTERN BLOT ANALYSIS
Western blot protein identification was quantified using the computer program Scion Image
Beta, edition 3B and Kodak 1D Image Analysis Software. The images captured were analysed
109
using the densitometric analysis method where pixel densities are established and
quantitated. Pixel densities of the bands were normalised against β-Actin expression.
3.11.
STATISTICAL ANALYSIS
Statistical significance for all experiments was determined using 2-tailed t-tests assuming
unequal variance using Microsoft Excel 2010, data was compared within time points and not
across time points or across cell lines. Statistical significance established at P≤0.05.
110
4. CHAPTER FOUR: THE CYTOTOXIC
EFFECTS OF PHENOXODIOL ON THE
PROSTATE CANCER CELL LINES;
LNCAP, DU145 AND PC3
4.1.
INTRODUCTION
Adenocarcinoma of the prostate is the second most commonly diagnosed malignancy in men
and is a common cause of cancer mortality in men in many western countries and increasingly
in developing nations (Hsing et al. 2000). In Australia prostate cancer represents the most
significant of all cancers affecting males with a total of 19,403 new incidences in 2007, 31.3%
of all new cancer cases (AIHW 2010). Prostate cancer also represents the second highest rate
of mortality associated with cancer affecting males at 13% behind lung cancer, which had 19%
of all mortality in 2007 (AIHW 2010). Predictions for the future indicate that prostate cancer
rates are projected to increase from 11,191 in 2001 to 16,800 in 2011, with this large expected
increase reflected in the projection for all cancers affecting males (AIHW 2010). Like other
cancers, prostate cancer has numerous clinical states ranging from a hormone-naïve clinically
localised primary tumour to lethal androgen-independent metastases. Regulation of prostate
growth is mediated via androgens and the corresponding androgen receptor (AR) which
regulates the transcription of target survival and apoptosis genes. Late-phase metastatic
disease is often androgen independent arising from; increased AR expression, enhanced
nuclear localisation of the AR, loss of function due to acquisitions in AR mutation resulting in a
more promiscuous receptor, or the presence of alternative survival pathways (Bcl-xL, Wnt
family upregulation) that circumvent the need for a functional androgen receptor (Gleave et
al. 2005).
111
The primary systemic treatment option for early and late stage prostate cancer is androgen
ablation therapy combined with an apoptosis inducing drug, such as Docetaxel or Paclitaxel.
Chemotherapy treatment options for patients with late-phase metastatic prostate carcinoma
rely on the premise that androgen-insensitive prostate carcinoma cells retain their basic
cellular apoptotic machinery to undergo programmed cell death, however increased tumourcell resistance to apoptosis is an underlying molecular reason contributing to disease
progression and chemo-resistance (Miyamoto et al. 2004; Miyamoto et al. 2005). The exact
mechanisms responsible for prostate cancer growth, especially emergence of the androgenindependent phenotype, are still far from fully understood (Feldman and Feldman 2001).
While other therapies, such as radiotherapy and chemotherapy, are available for advanced
prostate cancer, whether these therapies, either alone or combined with hormonal therapy,
significantly prolong patient survival remains controversial (Miyamoto et al. 2004; Miyamoto
et al. 2005). Thus, novel treatment strategies, instead of, or in combination with, androgen
deprivation therapy for advanced prostate cancer need to be developed such as (Petrylak et al.
2004) (Tannock et al. 2004) have shown.
One such novel method currently undergoing intense investigation is the flavanoid compounds
found in the dietary intake of populations with low cancer rates. Flavanoids are important
regulators in plants of biochemical and physiological processes, acting as antioxidants, enzyme
inhibitors, pigments and hormones. Human consumption of flavanoids has long been
recognised to manage anti-inflammatory, antioxidant, antiallergic, hepatoprotective,
antithrombotic,
antiviral
and
anticarcinogenic
activities
(Middleton
et
al.
2000).
Epidemiological studies have consistently shown an inverse association between isoflavone
intake and risk of cancer (Brusselmans et al. 2005). In vitro mechanistic studies on isoflavones
provide insight into modes of anticancer action ranging from cell cycle arrest and apoptosis
induction to angiogenic and antiproliferative effects (Li et al. 2008; Seo et al. 2011). To date,
112
genistein has been shown to demonstrate broad anticancer effects such as receptor tyrosine
kinase and cyclin dependent kinase inhibition (Alhasan et al. 1999; Li and Sarkar 2002).
Phenoxodiol, [2H-1-Benzopyran-7-0,1,3-(4-hydroxyphenyl)], is an isoflavone derivative that has
also been shown to elicit cytotoxic effects against a broad range of human cancers. Currently
undergoing human clinical trials, it has shown promise in patients with recurrent ovarian
cancer where the cancer is refractory or resistant to standard chemotherapy, and in patients
with hormone-refractory prostate cancer (Sapi et al. 2004; Brown and Attardi 2005).
Preliminary studies involving a number of flavanoid derivatives have demonstrated that
Phenoxodiol inhibits cell proliferation of a wide range of human cancer cell lines including,
leukaemia, breast and prostate carcinomas, and is five to twenty times more potent than a
similar compound, Genistein (Aguero et al. 2005). Phenoxodiol has been characterised in
ovarian cancer as affecting key ovarian anti-apoptotic signalling pathways (Kamsteeg et al.
2003) as well as reversing the ability of cells to become resistant to Docetaxel, through over
expression of anti-apoptotic molecules (Sapi et al. 2004). In mammary carcinogenesis
(Constantinou et al. 2003) found that the in vitro activity acted as an inhibitor of cell division
and/or an inducer of cell differentiation whilst short term investigation into Phenoxodiol
activity in prostate cancer by (Axanova et al. 2005) indicated that growth of LNCaP cells, in
monoculture and coculture with osteoblasts, resulted in the downregulation of the cancer
specific enzyme tNADH-Oxidase. Finally (Aguero et al. 2005; Aguero et al. 2010) indicated that
Phenoxodiol induces G1 specific arrest through loss of Cyclin-Dependant Kinase 2 activity by
p53-independent induction of p21WAF1 in a battery of human cell lines, or that phenoxodiol
effects multiple cancer types through prevention of cells from reproducing. Clearly,
phenoxodiol has demonstrated an ability to increase susceptibility of various cancer cell types
to initiate cell cytotoxicity; however the underlying mechanism of action in prostate cancer has
yet to be determined.
113
In this study we investigate Phenoxodiol’s ability to elicit anticancer effects in cells
representative of the clinical stages of prostate cancer development, by directly inhibiting
proliferation, and by eliciting direct cytotoxic effects against androgen-responsive and
androgen-refractory prostate cancer cell lines. We also seek to show that if phenoxodiol elicits
potent cytotoxicity, is it an apoptotic or necrotic response, and is it initiated by classic
apoptotic signalling pathways.
4.2.
AIMS
The aims of this chapter were as follows:
Aim 1: To characterise the cell growth rates of cell lines LNCaP, DU145 and PC3 in vitro.
Aim 2: To determine the potential cytotoxicity of phenoxodiol on in vitro prostate cancer cells.
Aim 3: To determine an effective set of doses of phenoxodiol that inflict cytotoxicity on
prostate cancer cells.
Aim 4: To determine the type of cytotoxic cell response following Phenoxodiol treatment in
prostate cancer cells.
4.3.
METHODOLOGY
The methodology utilised in this chapter is discussed in detail in the materials and methods
chapter (page 50). The goal of this set of experiments was to determine the optimum growth
characteristics for the three prostate cancer cell lines; LNCaP, DU145 and PC3. The cell lines
are representative of early stage androgen receptor wild type (LNCaP), late stage androgen
receptor mutant (DU145) and late stage androgen receptor null (PC3) prostate cancer.
114
Proliferation assays ensured that after 96 hours of culture the cell lines were undergoing
logarithmic cell growth. Cells were subcultured and seeded onto 96 well plates at varying rates
of initial concentration, after an initial 48 hours of subculture the cultures media were
replaced and a further 48 hours of growth allowed before application of 20µL CellTiter 96®
AQueous One MTS dye. After 3 hours of incubation the colour change was analysed on a
Labskan plate reader at 492nm. Increased absorbance rates were consistent with increased
metabolic and proliferation rates. Once initial seeding concentrations ensuring logarithmic
growth rates over 96 hours were established Phenoxodiol was diluted down into a final
concentration of 10µM and 30µM with a DMSO vehicle. Control and 10µM phenoxodiol
solutions had DMSO added to standardise final vehicle concentration with the 30µM
phenoxodiol treatment. Following dilution the cell proliferation rate, under the influence of
phenoxodiol, was measured by seeding 96 well plates at appropriate rates as pre-determined.
After 48 hours of initial culture the media was replaced for either; DMSO vehicle, 10µM or
30µM phenoxodiol solutions. Solutions were placed in wells lacking cells so that any
interaction of phenoxodiol with MTS dye could be assessed. The cells were treated for 24 and
48 hours, after which MTS dye was added to each well, incubated for 3 hours then assessed for
absorbance.
Acidity analysis was performed on cell media after 48 hours of treatment via a pH balanced
probe. Media pH was measured at 37°C, the incubator temperature for the cells in media.
Functional apoptosis assays to detect early and late stage apoptosis in response to
phenoxodiol treatment were performed. The JC-1 mitochondrial depolarisation assay was used
to determine if phenoxodiol induced mitochondrial efflux, an indicator of early apoptosis. Cells
were seeded into white fluorescent 96 well plates and following phenoxodiol treatment the
media was removed and JC-1 fluorescent dye added with positive control cells receiving a
combination of JC-1 and FCCP. After incubation then 20 minutes of 5% BSA solution the dye
115
was replaced with 100µL PBS and the fluorescent plates analysed on a Fluostar Optima plate
reader. Excitation/Emission was measured from above the plate. Green fluorescence was
measured at 485nm excitation and 520nm emission while red fluorescence was measured at
544nm excitation and 590nm emission. Red emission results were divided by green and the
ratio of red to green compared with a low ratio indicated high depolarisation rate and early
apoptotic induction.
The fluorescent Caspase-3 activity assay measured activated Caspase-3 expression and was
performed by seeding and treating cells on black 96 well fluorescent plates. Following
appropriate treatment with phenoxodiol, activated Caspase-3 was determined using an
EnzChek Capase-3 Assay Kit #2® containing a fluorescent compound (Z-DEVD-R110) which is
converted, by activated Caspase-3, into the green fluorescent rhodamine 110. After treatment
the media replaced with a staining solution and incubated on ice for 30 minutes followed by
room temperature for a final 30 minutes. Excitation/Emission was measured from above the
plate. Cells were measured at 496nm excitation and 520nm emission with a set gain level
between cell lines and time points used.
DNA laddering indicative of late stage apoptosis was performed using the 3’-end labelling
assay. DNA was extracted using a phenol/chloroform extraction protocol and one microgram
of DNA was tagged with radioactive P32 then loaded into 2% agarose gel with loading dye, a
low voltage was applied for approximately 4 hours after which the separated DNA gel was
dehydrated. The gel was exposed to film for 24 hours at -80°C and the film developed. The
dehydrated gel was cut into strips representing the lanes of the gel, loading regions were
removed, and liquid scintillant placed in each gel, after which radioactivity was measured using
a liquid scintillation counter. The exposed film provided a qualitative measure of apoptosis
116
through visible DNA laddering while the scintillation counter provided a quantitative measure
of apoptosis with increased radioactivity correlated to increased apoptotic rate.
An Annexin-V-Fluos/Propidium Iodide double stain apoptosis measurement was performed on
live cells which were placed into a FACSCalibur flow cytometer measuring individual cell
fluorescence. AV/PI double stains indicate normal cell function, early apoptosis and late
apoptosis/necrosis. Cells were grown in 6 well plates and treated with phenoxodiol, after
appropriate time the media was removed and centrifuged. Cells were trypsinised off the 6 well
plates, centrifuged in media then washed with warm PBS before AV/PI staining solution was
added. After 15 minutes incubation at 25°C, running solution was added and the cells analysed
on the FACSCalibur flow cytometer.
117
4.4.
RESULTS
Cell proliferation measurements provide an in vitro method of accurately determining cell
growth rates. Figure 23 demonstrates the MTS cell proliferation graphs for prostate cancer cell
lines; LNCaP, DU145 and PC3 over 6, 24 and 48 hours. Cell proliferation was measured at
492nm, analysing the total absorbance of light under the influence of MTS dye, an indicator of
cell metabolism. At 48 hours a visible drop in absorbance represents decreased proliferation in
all cell lines, determining their maximum seed rates. Ninety six well plate seeding rates were
set at; LNCaP: 3000 cells/100µL, DU145: 1500 cells/100µL and PC3: 2500 cells/100µL. This
ensured that after a maximum of 48 hours of treatment (96 hours of total growth) any
decreased proliferative effect that phenoxodiol may have had was not due to a potential
confluence of cells within the 96 well plate confines. After this point all control solutions were
assumed to be at maximum cell proliferation and samples were compared to the control
relative value. Scaling these results 1.8 fold utilised these initial seeding rates for 6 well plate
use; LNCaP; 54,000 cells/mL, DU145: 27,000 cells/mL and PC3: 45,000 cells/mL. When using
10cm plates an initial seeding rate of 100,000 cells/mL was used for all three cell lines, while
not maximising all available space it ensured a T75 flask could be split into more than two
10cm plates.
118
Absorbance (492nm)
LNCaP Cell Proliferation
1.4
1.2
1
0.8
0.6
0.4
0.2
0
6hr
24hr
48hr
500 1000 1500 2000 2500 3000 4000 5000
Cells Seeded /100uL
DU145 Cell Proliferation
Absorbance (492nm)
2.5
2
1.5
6 hrs
1
24hrs
0.5
48hrs
0
300
500 1000 1500 2000 2500 3000
Cells Seeded /100uL
PC3 Cell Proliferation
Absorbance (492nm)
3
2.5
2
6hr
1.5
1
24hr
0.5
48hr
0
500 1000 1500 2000 2500 3000 4000 5000
Cells Seeded /100uL
FIGURE 23 CELL PROLIFERATION RATES AT 6, 24 AND 48 HOURS
* indicates initial cell seeding rate with logarithmic cell growth after 48 hours.
119
Studies have revealed phenoxodiol’s ability to reduce cell proliferation and induce cell
cytotoxicity in cell lines representing various cancer types and mouse xenograft models
(Kamsteeg et al. 2003; Axanova et al. 2005; Aguero et al. 2010). Figure 24 demonstrates the
reduction in prostate cancer cell line proliferation after 24 and 48 hours of phenoxodiol
treatment. Phenoxodiol treatment, in media alone, was found to interact with the MTS dye
and increase the opacity in a dose dependent manner causing a falsely positive increase in cell
metabolism measurements which was accounted for with media only wells. LNCaP cell
treatment with 30µM phenoxodiol significantly decreased cell proliferation, versus control
population, over 24 (p=0.037) and 48 (p=0.01) hours. DU145 cells were significantly reduced in
proliferation rate, versus control population, after 10µM and 30µM treatment concentrations
over 24 (p<0.05, p=0.0032) and 48 (p<0.001, p<0.001) hours of treatment. PC3 cell treatment
with 30µM phenoxodiol significantly decreased cell proliferation, versus control population,
after 24 (p=0.022) hours and both 10µM and 30µM phenoxodiol concentrations after 48
(p<0.05, p=0.002) hours. The resulting proliferation graphs indicated the LNCaP cells were the
least responsive to phenoxodiol treatment with DU145 then PC3 as the most response cell
lines. This result indicated that phenoxodiol was better at arresting cell metabolism in late
stage representative prostate cancer cells (DU145, PC3) than early (LNCaP), utilising the MTS
proliferation assay.
120
% Growth Rate Relative to Control
LNCaP Proliferation & Phenoxodiol Treatment
120
100
80
60
Control
40
10µM PXD
20
30µM PXD
0
24hrs
48hrs
Treatment
% Growth Rate Relative to Control
DU145 Proliferation & Phenoxodiol Treatment
120
100
80
60
Control
40
10µM PXD
20
30µM PXD
0
24hrs
48hrs
Treatment
% Growth Rate Relative to Control
PC3 Proliferation & Phenoxodiol Treatment
120
100
80
60
Control
40
10µM PXD
20
30µM PXD
0
24hrs
48hrs
Treatment
FIGURE 24 CELL PROLIFERATION RATES AFTER 24 AND 48 HOURS OF PHENOXODIOL TREATMENT
* indicates significance relative to control for that time point. All cell populations had two
media only replicates of which the mean absorbance value was removed from the total cell
absorbance.
121
To ensure phenoxodiol was not inducing a reduction in proliferation by acidification or
alkalisation of media the pH of cell culture media was tested. Figure 25 demonstrates the pH
changes induced by 24 and 48 hours of phenoxodiol treatment. No significant differences in
pH were detected after treatment indicating phenoxodiol did not induce cytotoxicity by
acidifying or alkalising the media.
pH
LNCaP Media pH Analysis
8
7
6
5
4
3
2
1
0
Control
10µM PXD
30µM PXD
24 Hours
48 Hours
Time
pH
DU145 Media pH Analysis
8
7
6
5
4
3
2
1
0
Control
10µM PXD
30µM PXD
24 Hours
48 Hours
Time
pH
PC3 Media pH Analysis
8
7
6
5
4
3
2
1
0
Control
10µM PXD
30µM PXD
24 Hours
48 Hours
Time
FIGURE 25 PH CHANGES IN PHENOXODIOL TREATED CULTURE MEDIA
122
The JC-1 mitochondrial depolarisation assay measures induced mitochondrial efflux, an
indicator of early apoptosis, induced by the release of potent pro-apoptotic signalling
molecules such as Cytochrome C and Apoptosis Inducing Factor (AIF) (Juhaszova et al. 2004).
Figure 26 demonstrates the mitochondrial membrane potential change initiated after 24 and
48 hours of 10µM and 30µM phenoxodiol treatment. FCCP positive controls gave an indication
of maximal potential depolarisation rate and were significantly different from all control,
10µM and 30µM phenoxodiol treatments across all cell lines (p<0.001). LNCaP cells displayed a
significant decrease in membrane potential, versus control population, after 24 hours
treatment with 30µM treatment (p=0.012) and with both 10µM and 30µM treatments, versus
control, after 48 hours (p= 0.017, p<0.001). In the 48 hour treatment cohort the 30µM
phenoxodiol group were observed to have almost depolarised as much as the FCCP positive
control, establishing that LNCaP cells had a large increase in mitochondrial depolarisation,
indicative of a susceptibility to apoptotic signalling induced by phenoxodiol treatment. As a cell
line representative of early stage prostate cancer, this was an expected outcome after the lack
of response observed using the MTS absorbance assay.
DU145 mitochondrial membrane potential was significantly decreased at the 24 hour time
point in response to 30µM phenoxodiol treatment versus control (p<0.001) and 10µM
(p=0.0148). Over a 48 hour treatment course the 10µM concentration was significantly
depolarised versus control (p=0.0015) while the 30µM concentration was significantly different
to control (p<0.001) and 10µM (p=0.00485). The DU145 cell response to treatment indicated a
susceptibility to apoptotic signalling in a dose and time dependent manner with 48 hour
treatment time being more effective than the 24 hour time point. DU145 cell depolarisation,
after phenoxodiol treatment, never approached the maximal depolarisation rate indicated by
the FCCP treatments. The comparison indicated a reduced sensitivity to mitochondrial
depolarisation in response to phenoxodiol treatment and therefore reduced early apoptotic
123
signalling. Resistance to apoptotic induction is typical from a cell line indicative of late stage
prostate cancer.
Finally, PC3 mitochondrial membrane potential was significantly decreased at the 24 hour time
point in response to 10µM Phenoxodiol treatment versus control (p=0.03) while the 30µM
treatment was significantly different from control (p=0.0121) and 10µM PXD (p=0.0215). Post
48 hours of treatment the 10µM concentration was significantly different to control (p=0.018)
while the 30µM concentration significantly different to control (p=0.0024) but not to 10µM
(p=0.06). The PC3 cell mitochondrial depolarisation rate in response to treatment was even
less than that of DU145 cells which was expected from the cell line that represents a more
aggressive late stage prostate cancer etiology. Though there was significant increase in PC3
mitochondrial depolarisation versus control, the small overall increase indicated that PC3 cells
were not as susceptible to early apoptotic signalling as either LNCaP or DU145 cells.
124
Flourescence Relative to Control
LNCaP JC-1 Analysis
120
100
80
Control
60
10µM PXD
40
30µM PXD
20
FCCP
0
24 Hours
48 Hours
Treatment
Flourescence Relative to Control
DU145 JC-1 Analysis
120
100
80
Control
60
10µM PXD
40
30µM PXD
20
FCCP
0
24 Hours
48 Hours
Treatment
Flourescence Relative to Control
PC3 JC-1 Analysis
120
100
80
Control
60
10µM PXD
40
30µM PXD
20
FCCP
0
24 Hours
48 Hours
Treatment
FIGURE 26 JC-1 ANALYSIS OF MITOCHONDRIAL MEMBRANE POTENTIAL OVER 24 AND 48 HOURS OF
TREATMENT
* indicates significance relative to control for that time point. ** indicates significance relative
to 10µM phenoxodiol treatment and control for that time point. *** indicates significance
125
relative to 30µM phenoxodiol and control for that time point. Cell depolarisation is the
green/red fluorescence ratio relative to the control.
Caspase-3 is a potent stimulator of apoptosis and the final effector caspase of the caspase
cascade apoptosis pathway. Caspase-3 exists as the constant target of extrinsic and intrinsic
apoptotic cell signalling and is a direct target of anti-apoptotic cell signalling (Janicke et al.
1998; Boatright and Salvesen 2003)). Figure 27 demonstrates activated Caspase-3 expression
measured through fluorescence analysis after 24 and 48 hours of phenoxodiol treatment.
LNCaP Caspase-3 activity was found to significantly decrease after 48 hours treatment with
10µM phenoxodiol treated cells experiencing a significant decrease in Caspase-3 expression
versus DMSO vehicle control (p=0.003) and 30µM PXD treated cells experiencing significant
expression decrease versus control (p<001) and 10µM (p=0.043). The total fluorescence
decrease measured was 18.1% (10µM) and 29.08% (30µM) respectively. DU145 cells had a
small but significant increase in Caspase-3 expression over 48 hours with both 10µM and
30µM (p=0.009, p=0.0123) phenoxodiol treatments expressing more activated Caspase-3. The
measured fluorescence increase was only 4.5% (10µM) and 3.8% (30µM) over the control
population. Such a small increase could potentially have biologically significant effects
however in view of the lack of response from PC3 cells and the decreased expression by LNCaP
cells it might be viewed that this small change, while statistically significant, is not the driving
factor in cell cytotoxicity in DU145 cells. PC3 cells had no significant changes in activated
Caspase-3 expression measured over 24 and 48 hours after 10µM and 30µM phenoxodiol
treatment even though there was a clear anti-proliferative effect demonstrated in the MTS
assay. This data suggests, but does not confirm, that phenoxodiol induces cell cytotoxicity
through a caspase independent manner.
126
Absorbance 520nm
LNCaP Caspase-3 Activity
45000
40000
35000
30000
25000
20000
15000
10000
5000
0
Control
10µM PXD
30µM PXD
24 Hours
48 Hours
Time
Absorbance 520nm
DU145 Caspase-3 Activity
45000
40000
35000
30000
25000
20000
15000
10000
5000
0
Control
10µM PXD
30µM PXD
24 Hours
48 Hours
Time
Absorbance 520nm
PC3 Caspase-3 Activity
45000
40000
35000
30000
25000
20000
15000
10000
5000
0
Control
10µM PXD
30µM PXD
24 Hours
48 Hours
Time
FIGURE 27 FLUORESCENT ANALYSIS OF CASPASE-3 ACTIVITY AFTER PHENOXODIOL TREATMENT OVER
24 AND 48 HOURS
* indicates significance relative to control for that time point. ** indicates significance relative
to 10µM Phenoxodiol treatment and control for that time point. A set gain level between cell
lines and time points was used.
127
Apoptotic DNA fragmentation is a key feature of apoptosis, and is achieved by caspase
activation of endogenous endonucleases which subsequently cleave chromatin DNA into
fragments of around 180 base pairs multiples thereof. The amount of laddering can be used as
a direct measure of the amount of induced apoptosis (McCarthy et al. 1997). Figure 28
demonstrates DNA fragmentation through the quantitative measurement of three prime end
labelling (3’-End Labelling) which measures fragmentation levels by scintillation count.
Laddering can also indicate an apoptotic versus necrotic response and Figure 29 demonstrates
the qualitative measure of 3’-End labelling viewed through exposure to film. In the LNCaP cells
DNA laddering was clearly evident while in the DU145 samples laddering and a slight smear
was visible. PC3 cells had a necrotic smear in the film and no evidence of laddering. Due to the
measured response of PC3 as being necrotic, only the LNCaP and DU145 samples were
measured in the liquid scintillation counter to qualitatively assess their total apoptotic
response. LNCaP cells displayed a significant increase in apoptotic laddering versus control
after 24 hours of treatment with both 10µM (p=0.013) and 30µM (p=003) treatments,
exhibiting increased apoptosis by 35% (10µM) and 39% (30µM) respectively, compared to
control. After 48 hours of treatment with 30µM phenoxodiol the LNCaP cells were determined
to have a statistically significant increase in apoptosis versus control (p=0.044) exhibiting a 63%
increase in apoptotic laddering.
This result confirms the LNCaP cells were sensitive to apoptosis induction as indicated by JC-1
analysis. DU145 cells had a significant increase in apoptosis over 24 hours in response to the
30µM phenoxodiol treatment versus control (p=0.02) and versus 10µM (p=0.017) with an
increase in percentage of apoptosis by 15% versus control. After 48 hours of treatment it was
observed that 10µM phenoxodiol treatment exhibited increased apoptosis versus control
(p=0.009) and a 66% increase in apoptosis. The 30µM phenoxodiol treatment significantly
increased apoptosis versus control (p=0.03) and 10µM phenoxodiol (p<0.05), after 48 hours,
128
with a 700% increase in the 30µM treatment dose versus control indicating that DU145 cells
were very sensitive to apoptosis induced by phenoxodiol treatment after 48 hours of
treatment.
Low MW DNA Labelling (CPM)
LNCaP 3'-End Labelling
2500
2000
1500
Control
1000
10µM PXD
500
30µM PXD
0
24 Hours
48 Hours
Treatment
Low MW DNA Labelling (CPM)
DU145 3'-End Labelling
6000
5000
4000
3000
Control
2000
10µM PXD
1000
30µM PXD
0
24 Hours
48 Hours
Treatment
FIGURE 28 3'-END LABELLING APOPTOTIC ANALYSIS POST PHENOXODIOL TREATMENT OVER 24 AND 48
HOURS
* indicates significance relative to control for that time point. ** indicates significance relative
to 10µM Phenoxodiol treatment and control for that time point. Counts Per Minute (CPM) is
the amount of low molecular weight DNA labelled with radioactive P32 by the ddATP enzyme.
([α32P]-ddATP).
129
DMSO 10µM 30µM
FIGURE 29 AN EXAMPLE OF LNCAP 3’-END LABELLING QUALITATIVE DNA LADDERING AFTER EXPOSURE
TO FILM
The final functional measurement of apoptosis was executed using an Annexin-VFluos/Propidium Iodide (AV/PI) double stain on live cells which were placed into a FACSCalibur
flow cytometer for the measurement of cell fluorescence. No staining indicated normal cell
function, Annexin-V staining indicated early stage apoptosis by the conjugation of Annexin-V to
membrane bound, but externalised phosphatidylserine. Annexin-V only staining represents
early apoptosis as the cells still retain the ability to exclude PI. The AV/PI double stain indicated
late stage apoptosis/necrosis as the AV was bound to externalised phosphatidylserine and the
PI was bound to nuclear DNA which the cells lacked the ability to exclude (Martin et al. 1995).
Figure 30 demonstrates the cytotoxic effects of 24 and 48 hours of 10µM and 30µM
phenoxodiol treatment, as measured by fluorescence activated cell sorting (FACS) analysis also
called flow cytometry.
After 24 hours of treatment LNCaP cells indicated no significant changes in relation to control
populations; however, a high level of early and late apoptotic cell populations was present.
130
After 48 hours of treatment the 10µM phenoxodiol treated cells were significantly different
versus control in the alive (p<0.0008, 64% decreased), early apoptosis (p=0.014, 153%
increase) and late apoptosis (p=0.0005, 332%) cell populations. The 48 hour 30µM
phenoxodiol treatments were significantly different from control in the alive (p<0.05, 50%
decrease) and late apoptotic population (p=0.03, 216% increase) but the early apoptotic
population was not statistically significant from control early apoptotic (p=0.130). LNCaP cells
were initially unresponsive then became very sensitive to phenoxodiol treatment, with the live
cell population decreasing and the early and late apoptotic populations increasing significantly
in response to treatment. The LNCaP cell population were difficult to obtain a 10,000 viable
cell count for each sample due to phenoxodiol’s cytotoxicity.
DU145 cells had significant changes over 24 hours in both 10µM and 30µM phenoxodiol
treatments. The 10µM treatment significantly decreased alive cells versus control (p=0.037,
12.3% decrease), significantly increased early apoptotic (p=0.047, 74% increase) and
significantly increased late apoptotic (p=0.048, 100% increase) cell populations. The 30µM
treatment group had a significant early apoptotic population increase versus control
(p=0.0006, 291% increase) and difference versus 10µM phenoxodiol treatment in early
apoptotic (p=0.003, 125% increase) and late apoptotic (p=0.02, 58% decrease) cell populations.
Over 48 hours the 10µM treatment group exhibited significantly decreased live cell population
(p=0.0002, 17% decrease), significantly increased early apoptotic (p=0.0002, 3000% increase)
and late apoptotic (p=0.0026, 69% increase) cell populations versus control. The 30µM
treatment group was significantly different from both control and 10µM treatments at live cell
population (p<0.001, 37% decrease, p<0.0004, 24.1% decrease), early apoptotic (p<0.001,
5000% increase, p=0.0007, 75% increase) and late apoptotic (p=0.0002, 251% increase,
p=0.0008, 108% increase) cell populations. The DU145 cells increased both early and late
apoptotic populations in a time/dose dependant manner and reflected the decreasing cell
131
viability. Phenoxodiol treatment of DU145 cells did not exhibit as large a decrease in live cell
population after treatment as was observed in LNCaP cells, but was still able to induce
significant cytotoxicity in the prostate cancer cell line.
PC3 cells indicated no significant changes in early apoptotic cell population across the 24 and
48 hour treatment time points. After 24 hours of treatment with 10µM phenoxodiol treatment
a significant difference versus control in both live cell (p=0.0007, 15% decrease) and necrotic
(p=0.0001, 180% increase) was exhibited while the 30µM treatment had a significant
difference versus control in both live cell (p=0.0001, 23.2% decrease) and necrotic (p=0.0007,
272% increase) and a significant difference versus 10µM in necrotic (p=0.03, 33% increase) cell
population. The 48 hour 10µM treatment group had a significant difference versus control in
both live cell (p<0.001, 24.6% decrease) and necrotic (p<0.001, 287% increase) while the 30µM
treatment had a significant difference versus control in both live cell (p<0.001, 34.1%
decrease) and necrotic (p<0.001, 415% increase) and a significant difference versus 10µM in
live cell (p=0.0004, 12.7% decrease) and necrotic (p<0.001, 32.9% increase) cell population.
During 3’-end labelling it was determined that PC3 cells responded to phenoxodiol by a
necrotic response. The results of the AV/PI FACS analysis correlate with only the late stage
apoptotic/necrotic double staining cells increasing in number, indicative of a purely necrotic
response. PC3 cells were less susceptible to phenoxodiol cytotoxicity versus the other cell
lines, with a 34.1% decrease in live cell population after 48 hours of treatment with a 30µM
dose, this is compared to the 37% decrease in viable cell numbers from DU145 and 50%
decrease in LNCaP cells.
132
LNCaP AV/PI FACS Analysis
100
% of Cells
80
60
40
20
0
Control
Alive
10µM PXD 30µM PXD Control 10µM PXD 30µM PXD
24 Hours
48 Hours
Treatment
Early Apoptotic
Late Apoptotic / Necrotic
DU145 AV/PI FACS Analysis
100
% of Cells
80
60
40
20
0
Control
10µM PXD 30µM PXD Control 10µM PXD 30µM PXD
24 Hours
48 Hours
Treatment
Alive
Early Apoptotic
Late Apoptotic/Necrotic
PC3 AV/PI FACS Analysis
100
% of Cells
80
60
40
20
0
Control
10µM PXD 30µM PXD Control 10µM PXD 30µM PXD
24 Hours
48 Hours
Treatment
Alive
Early Apoptotic
Late Apoptotic/Necrotic
FIGURE 30 ANNEXIN V-FLUOS / PROPIDIUM IODIDE DOUBLE STAINING ANALYSIS OF PROSTATE CANCER
CELLS POST PHENOXODIOL TREATMENT OVER 24 AND 48 HOURS
133
* indicates significance relative to control group for that time point i.e. 24 hour Control early
apoptotic versus 24 hour 10µM Phenoxodiol early apoptotic. ** indicates significance relative
to 10µM Phenoxodiol treatment and control for that time point i.e. 48 hour Control 10µM
Phenoxodiol versus 48 Hour 30µM Phenoxodiol Control.
4.5.
DISCUSSION
Determining appropriate initial cell concentration rates ensured logarithmic growth after 96
hours and negated any cellular activity decrease due to cell crowding effects. We demonstrate
here, utilizing the MTS viability assay, that phenoxodiol elicits time- and dose-dependent antiproliferative activity against both androgen-responsive and androgen resistant prostate cancer
cell lines. Initial results from the proliferation assay indicated LNCaP was less susceptible to
phenoxodiol treatment than either DU145 or PC3 cells; however, visualising the cells indicated
the opposite, a majority of LNCaP cells, far more than DU145 or PC3, had been induced into
cytotoxicity and were dead. As we had already accounted for the interaction between
Phenoxodiol and MTS dye there was no indication as to why the MTS assay suggested very
little decrease in proliferative activity yet visualisation of the cells clearly indicated significant
cytotoxicity was occurring. (Wang et al. 2010) has reported an underestimating of antiproliferative effects of cytotoxic compounds due to MTS dye in LNCaP cells than when
compared to other techniques. While the LNCaP proliferation assay was not necessarily
sensitive to the MTS addition, the use of apoptotic assays, 3’end labelling, JC-1 and AV/PI flow
cytometry, ensured accurate cytotoxicity information was obtained. DU145 and PC3 cell lines
indicated dose and time responsive cytotoxicity over 24 and 48 hours treatment with
phenoxodiol. While both cell lines are considered examples of late stage prostate cancer, PC3
is acknowledged as an extremely chemoresistant cell line and correspondingly had the least
decrease in proliferation.
134
Acidity measurements in cell culture media resulted in no significant pH changes detected
between control and treatment groups using total media at 37°C. Variation in cell cytotoxicity
and proliferation decreases were not due to pH changes in the cell media. JC-1 mitochondrial
depolarisation assays indicated depolarisation under the influence of phenoxodiol treatment.
Mitochondrial catastrophe is a potent inducer of cell cytotoxicity and considered an early
apoptotic indicator as it stimulates the intrinsic caspase pathway via the conformational
change of the pro-apoptotic Bcl-2 family members. This causes mitochondrial efflux of more
potent pro-apoptotic Bcl-2 family members such as Bax, resulting in an imbalance between
anti- and pro-apoptotic signalling (Gross et al. 1999; Kuruvilla et al. 2003). FCCP was used as
the positive control and indicated the maximum potential depolarisation of the cells at each
time point and treatment (Kuruvilla et al. 2003). LNCaP cells were very susceptible to cell
death signalling, with 30µM phenoxodiol treatment almost equalling maximal mitochondrial
depolarisation indicated by FCCP. As the cell line indicative of an early stage prostate cancer,
the LNCaP response to phenoxodiol as detected using the JC-1 assay was more accurate than
the MTS assay. DU145 and PC3 both had significant depolarisation, although far more limited
than LNCaP compared to FCCP depolarisation. This was another indication the models of late
stage prostate cancer, known for their chemoresistance, were appropriate choices for this
study. There is currently no curative treatment for late stage prostate cancer (Canfield et al.
2006).
Classic apoptosis, such as that induced by chemotherapeutic agents, can proceed via extrinsic
death-receptor mediated and intrinsic mitochondrial-mediated pathways ultimately resulting
in the activation of Caspase-3 (Asselin et al. 2001; Mahoney 2007). Given that phenoxodiol
induced classic apoptosis in ovarian cancer (Gamble et al. 2005; Alvero et al. 2006), it was
surprising to find that activated Caspase-3 expression remained unchanged in PC3 cells and
even declined in LNCaP cells, while the actual increase in expression in DU145 cells was
135
minimal. Activated Caspase-3 activity assays indicated that phenoxodiol may function through
a caspase independent pathway, as LNCaP levels of activated Caspase-3 were significantly
decreased after 48 hours of phenoxodiol treatment while the cells were undergoing detectable
apoptosis. The lack of consistent signalling indicated that cell cytotoxicity induced by
phenoxodiol was potentially initiated via a caspase independent cell death pathway.
Apoptotic analysis by 3′-end labelling and Annexin-V/PI based flow cytometry validated both
qualitatively and quantitatively that phenoxodiol induced apoptotic cell death in LNCaP and
DU145 cell lines. Apoptosis was detected by DNA fragmentation analysis and Annexin-V
binding to the externalisation of phosphatidylserine only, indicating early stage apoptosis; or in
conjunction with propidium iodide nuclear staining, indicating late stage apoptosis. In contrast,
phenoxodiol significantly decreased the proliferation of PC3 cells in the absence of DNA
laddering and externalisation of phosphatidylserine, indicating a necrotic response that was
qualitatively measured by 3’-end labelling and thereby confirming the response. The use of
only one apoptotic measure, the lack of confirmation of DNA laddering or the use of only
propidium iodide in flow cytometry makes singular technique apoptotic studies flawed.
Annexin-V phosphatidylserine only and AV/PI double staining is required to detect early versus
late apoptosis or necrosis but does not guarantee either. Confirmation with DNA laddering
ensures the outcome with a quick technique that is both qualitative and quantitative, but by
itself DNA laddering cannot indicate early stage apoptosis. FACS analysis established that
phenoxodiol clearly induced; the appearance of early and late stage apoptotic populations in
LNCaP and DU145 cells, the appearance of necrosis in PC3 cells as well as a marked decrease in
viable cells in all cell lines. These data were qualitatively confirmed by the visualisation of
characteristic low molecular weight laddering during 3’-end labelling. All three cell lines
responded to phenoxodiol treatment with a minimum decrease in viable cells of 30% after
only 48 hours treatment, and a variable incidence of early apoptotic and late stage
136
apoptotic/necrotic cell population production. In all modes of analysis, cell death was reliant
on dose and duration of phenoxodiol exposure and yet Caspase-3 was not activated in
phenoxodiol treated PC3 cells. Neither LNCaP nor DU145 cells had any consistent caspase 3
change observed which implicated a non caspase dependant route as the target of apoptotic
induction.
Phenoxodiol exhibits significant cytotoxicity, inducing cell death in the prostate cancer cell
lines LNCaP, DU145 and PC3. All three cell lines exhibited significantly decreased viable cell
populations after only 48 hours of treatment but they also utilised different signalling
pathways than those reported in previous studies (Kamsteeg et al. 2003; Sapi et al. 2004;
Alvero et al. 2006; Mor et al. 2006). This establishes the cell-death inducing capabilities of
phenoxodiol on prostate cancer cells and suggests the induction of apoptotic cell death
through non-classical pathways. By targeting non classical apoptotic pathways and successfully
inducing cell cytotoxicity even in the most chemoresistant cell lines, phenoxodiol displays
potential as a drug for future treatment of prostate cancer.
137
5. CHAPTER FIVE: CELL DEATH
SIGNALLING IN RESPONSE TO
PHENOXODIOL TREATMENT
5.1.
INTRODUCTION
Two of the hallmarks of cancer initiation and progression are the ability of cells to become
insensitive to anti-growth signals and to evade apoptosis. To do this the cells accumulate
increased expression of anti-apoptotic signalling molecules, which prevent activation of the
intrinsic and extrinsic signalling pathways that normally induce cell death (Hanahan and
Weinberg 2000). The accumulation of these factors, coupled with a dysregulation in the cell
cycle, results in a progressively increasing rate of mutation and the movement of cells from
benign to metastatic types. In response to cytotoxic insult from both external and internal
sources, cells will normally initiate machinery, such as caspases, that execute apoptotic cell
death.
Apoptosis is important as it results in cell death with DNA cleavage into unusable fragments
with no resultant inflammatory response. Achieving apoptosis results in fewer side effects
from chemotherapeutic agents and, as such, initiating an apoptotic response becomes the end
point target of all new therapy trials (Brown and Wouters 1999). Though they vary between
cells, the array of apoptosis induction signals commonly trigger signalling pathways that
coalesce at the mitochondria. Once stimulated, the mitochondrial cell walls efflux powerful
apoptotic stimulators such as Cytochrome c and Bax (Lee et al. 2008) to activate central
effector pathways involving a series of proenzymes, the caspases (Maguire et al. 2000). When
activated, caspases are efficient at degrading cellular processes and DNA to the point where
normal physiological activity is impossible, resulting in cellular response and apoptotic
138
phenomena (Hengartner 2000). The mitochondrial pathway of cell death is mediated by the
Bcl-2 family of proteins, a group of anti-apoptotic and pro-apoptotic proteins that regulates
the passage of small molecules such as Cytochrome c, Apoptosis Inducing Factor (AIF) and Bax,
which activates caspase cascades through the mitochondrial transition pore (Korsmeyer 1992;
Korsmeyer et al. 2000). The activation of caspases is counteracted by anti-apoptotic molecules
of the Bcl-2 family (Bcl-2, Bcl-xL), because these Bcl-2 family proteins heterodimerise with proapoptotic members of the Bcl-2 family (Bax, Bak) and interfere with release of Cytochrome c
by inactivating pore-forming proteins (Gross et al. 1999). Cancer cells commonly have the antiapoptotic members of the Bcl-2 family increased in expression, allowing them to prevent
extrinsic anti-growth signalling and avoid intrinsic stimulation of apoptosis (Cory and Adams
2002).
Upon intrinsic stimulation of apoptosis, the release of Bax and Cytochrome c from the
mitochondria initiates the apoptosis promoting activity of Cytochrome c by allowing
interaction with the Apaf-1 protein. Binding of Cytochrome c to Apaf-1 causes recruitment of
Caspase-9, forming the apoptosome (Slee et al. 1999; Srinivasula et al. 2001; Zou et al. 2003).
The apoptosome is a large molecule that interacts with the executioner caspases, Caspase-3
and Caspase-7, which initiate DNA fragmentation and cause the cell to enter into apoptosis
(Boatright and Salvesen 2003). Caspase-3 can be activated via extrinsic pathways that produce
activated Caspase-9 through Caspase-8 signalling. If intrinsic pathways are activated, the
release of Cytochrome c and apoptosome formation results in Caspase-3 activation (Riedl and
Salvesen 2007). During normal cell function Caspase-3 is in an inactive zymogen form. The
zymogen feature of Caspase-3 is necessary because, if unregulated, caspase activity would kill
cells indiscriminately and therefore Caspase-3 activity is strictly limited until it is cleaved by an
initiator caspase after apoptotic signalling (Donepudi and Grutter 2002; Boatright and Salvesen
2003).
139
The Bcl-2 family members are not the only regulatory proteins of apoptosis nor are the
caspases the only method of executing programmed cell death. Members of the Inhibitors of
Apoptosis Proteins (IAP) such as xIAP prevent apoptosis by inactivating Caspase-3 and Caspase9 directly, amongst other roles, and are found in increased expression in chemoresistant
cancer cells (Dohi et al. 2004; Dubrez-Daloz et al. 2008). FLICE inhibitory protein (FLIP) binding
to Caspase-8 can also prevent death receptor mediated apoptosis and has been linked to
phenoxodiol treatment (De Luca et al. 2008). Apoptosis inducing factor (AIF) is involved in
initiating a caspase-independent pathway of apoptosis and can act as an NADH oxidase.
Normally it is found behind the outer membrane of the mitochondria and is therefore
secluded from the nucleus; however, when the mitochondria are damaged, it moves to the
cytosol and to the nucleus. Another AIF function is to regulate the permeability of the
mitochondrial membrane upon apoptosis stimulation, causing caspase dependant pathways to
be activated (Xie et al. 2005). Apoptosis can still happen in cells that lack Cytochrome c, Apaf-1
or caspases with AIF causing apoptotic changes independent of caspase activity (Susin et al.
1999; Li et al. 2000).
X-linked inhibitor of apoptosis protein (xIAP) is part of a family of endogenous caspase
inhibitors that, when up regulated, have been implicated in the early stage development of
prostate and breast epithelial carcinoma (Parton et al. 2002; Krajewska et al. 2003). In
conjunction with anti-apoptotic Bcl-2 members (Bcl-2, Bcl-xL) IAP family members exert
significant anti-apoptotic effects through; inhibition of Caspase-3 and -7 activation and
function, inactivation of pro-apoptotic Bcl-2 family members and involvement in signalling
pathways NF-κB and JNK (Levkau et al. 2001; Sanna et al. 2002; Hu et al. 2003; Dubrez-Daloz et
al. 2008). XIAP expression levels are also found to be increased in epithelial ovarian and
prostate cancers, and have been linked to chemoresistance and progression of cells to
metastasis (Berezovskaya et al. 2005).
140
Separate to direct apoptotic interaction, cytotoxic compounds can cause reactive oxygen
species production, which leads to oxidative stress in cancer cells by rapid accumulation of
highly reactive molecules such as nitric oxide (Murphy 2009). When this occurs it results in
damage to cell structures and potentially activates the apoptotic cascade making it a target for
chemotherapy agents (Dharmarajan et al. 1999). Oxidative stress induces the expression of
antioxidant genes that contain an antioxidant response element in their promoters; this can
result in a triggering of apoptosis or necrosis of various cell types and, in several instances, can
inhibit cell growth and interfere with the cell cycle (Minelli et al. 2009).
After determining that phenoxodiol induced cytotoxicity in prostate cancer cells, the next
focus was to determine the molecular signalling that resulted in cytotoxicity after phenoxodiol
treatment. Molecular signalling that results in apoptosis is generally initiated via extrinsic and
intrinsic pathways that mediate mitochondrial permeability and Bax upregulation. This results
in Caspase-3 activation and subsequent apoptosis. To elucidate the specific mechanism by
which phenoxodiol elicits an anti-proliferative effect in prostate cancer cell lines, we
investigated the interactions of the apoptotic signalling molecules AIF, Bax, Bcl-xL, Caspase-3
and xIAP in prostate cancer cells treated with phenoxodiol. Subsequently this chapter
investigates if caspase signalling is a requirement for phenoxodiol activity and whether
phenoxodiol treatment could induce the production of reactive oxygen species which would
then induce a cytotoxic response via oxidative stress.
141
5.2.
AIMS
The aims of this chapter were as follows:
Aim 1: To determine consistency in cell death induction signalling between the cell lines by
investigation of AIF, Bax, Bcl-xL, Caspase-3 and xIAP expression changes post phenoxodiol
treatment.
Aim 2: To determine if phenoxodiol activity required intrinsic or extrinsic caspase signalling
pathways in LNCaP, DU145 and PC3 cells.
Aim 3: To determine if Phenoxodiol-induced reactive oxygen species generation by
investigating nitric oxide production post phenoxodiol treatment.
5.3.
METHODOLOGY
The methodology utilised in this chapter is discussed in detail in the materials and methods
chapter (page 50). Quantitative PCR (qPCR) analysis and Western blot analysis were performed
on all three cell lines LNCaP, DU145 and PC3, after 24 and 48 hours exposure to 10µM and
30µM phenoxodiol concentrations. Briefly, cells were seeded into 6 well plates and treated
appropriately with phenoxodiol, RNA extraction was performed using the Chomczynski
guanidinium thiocyanate-phenol-chloroform method via a TRI Reagent® extraction kit
(Chomczynski and Sacchi 1987). Purified RNA was analysed for concentration using a
Nandrop™ light spectrophotometer and 1µg was firstly DNAse treated and then converted into
cDNA using a “M-MLV Reverse Transcriptase RNase H Minus, Point Mutant Taq” Promega
enzyme kit with random primers. Removal of excess random primers, reagent and salts was
performed using a Mol-Bio spin column post-pcr purification kit, leaving a final volume of 50µL
cDNA. Each qPCR experiment was performed using 2µL of cDNA per sample. Primers were
142
designed to test for AIF, Bax, Bcl-xL, Caspase-3 and xIAP by using published primers. Primer
spanning of an intron/exon boundary was confirmed using Primer3 and BLAST search engines.
Quantitative PCR (qPCR) was performed by adding 2µL of cDNA to a master mix of primers,
Taq polymerase, dNTP’s, salt and water to bring the final volume to 10µL. qPCR was then
performed on samples for 35 temperature cycles at varying annealing temperatures, on a
Corbett Rotorgene 3000 or 6000, with initial results run against a gel ladder to determine if
primer dimerisation had occurred. Once a successful sample was gel loaded and confirmed, a
gel extraction was performed, purifying the final product that was then serially diluted 10 fold
each step to produce a set of standards. In the full sample qPCR experiments, each sample was
duplicated and compared to five standards which were also duplicated. The standards were 10
fold dilutions of given concentrations and the resulting standard curve then determined final
sample concentrations. Final sample concentrations were compared against the expression of
L19.
Protein level analysis was performed using a Western blot protocol. Briefly, cells were seeded
into six well plates, treated with phenoxodiol appropriately and a RIPA buffer/βmercaptoethanol-based,
whole
cell
lysate,
protein
extraction
performed.
Protein
concentration was determined against a standard curve on a Nanodrop™ light
spectrophotometer and then, with varying concentrations per protein to be detected, loaded
onto a SDS-Page/acrylamide based gel and current applied to the gel. After protein separation
through the SDS-Page/acrylamide gel the proteins were transferred to a nitrocellulose
membrane by the application of voltage to a transfer buffer. Ponceau S was used to stain the
nitrocellulose membrane to determine if transfer was effective, if air bubbles existed or if even
protein loading had occurred. Finally, primary antibodies were added to the membrane and
143
incubated in the appropriate conditions, excess was removed with TBS-T washes and a
secondary antibody conjugated to HRP applied in the appropriate conditions. A
chemiluminescent substrate kit was used to detect antibody binding concentration.
A caspase inhibition assay was performed using the pan-caspase inhibitor Z-VAD-FMK. Briefly,
cells were seeded into 96 well plates and after 48 hours of growth treated with either; DMSO
vehicle control, 30µM phenoxodiol, 10µM Z-VAD-FMK pan caspase inhibitor or a combination
of both 30µM phenoxodiol and 10µM Z-VAD-FMK, in complete media over a period of 48
hours. After 48 hours of treatment 20µL of MTS dye was added to each well and the plates
incubated for 3 hours before the absorbance of light at 492nm was measured on a LabSystems
plate reader.
The Griess assay was performed to determine if phenoxodiol induced the production of nitric
oxide, a potent free radical. Cells were plated into 96 well plates and grown for 48 hours
before the media were replaced with a negative control, DMSO Control, 10µM phenoxodiol,
30µM phenoxodiol, 1000µM DEAN or 100µM SNP containing media. After 24 or 48 hours of
treatment a standard curve was performed on the plate with complete media and 10 fold
dilutions of a 100µM nitrite solution. After the standard curve was prepared 50µL of the
treatment media was aspirated from each well and replaced with 50µL sulfanilamide solution
and then incubated for 10 minutes. After 10 minutes there was an addition of 50µL of NED
solution to each well followed by another 10 minute incubation. After incubation the plate was
scanned on a LabSystems plate reader at 540nm absorbance to determine NO production rate.
144
5.4.
RESULTS
Analysis of mRNA signalling expression by qPCR was performed to determine the extrinsic and
intrinsic apoptotic pathways activated by phenoxodiol treatment. The pro-apoptotic molecule
Apoptosis Inducing Factor (AIF) can induce apoptosis, in a caspase independent manner, by
causing chromatin condensation, interacting with NADH oxidase, inducing DNA fragmentation
and interacting with the mitochondrial membrane, thereby increasing permeability (Xie et al.
2005). Figure 31 demonstrates the quantitative mRNA expression of the AIF, over 24 and 48
hours, post phenoxodiol treatment. LNCaP AIF expression levels were significantly decreased
after 48 hours of treatment with 10µM (p=0.026) and 30µM (p=0.042) phenoxodiol
concentrations. At the 24 hour time point all cell lines expressed no significant difference in
comparison to the DMSO vehicle control levels. This result was also seen after 48 hours for the
DU145 and PC3 cell lines. AIF mRNA expression was determined to not be consistently altered
by phenoxodiol treatment between the prostate cancer cell lines.
145
LNCaP AIF qPCR Expression
10
AIF/L19
8
*
6
*
Control
4
10µM PXD
2
30µM PXD
0
24 Hours
48 Hours
Treatment
AIF/L19
DU145 AIF qPCR Expression
8
7
6
5
4
3
2
1
0
Control
10µM PXD
30µM PXD
24 Hours
48 Hours
Treatment
AIF/L19
PC3 AIF qPCR Expression
35
30
25
20
15
10
5
0
Control
10µM PXD
30µM PXD
24 Hours
48 Hours
Treatment
FIGURE 31 AIF MRNA EXPRESSION ANALYSIS OF PROSTATE CANCER CELLS POST PHENOXODIOL
TREATMENT OVER 24 AND 48 HOURS
* indicates significance relative to control for that time point.
146
Increased expression of the pro-apoptotic molecule Bax results in mitochondrial membrane
permeabilisation, releasing several pro-apoptotic molecules, such as Cytochrome c, which
induce apoptosis through activation of caspases (Cheng et al. 2001). Figure 32 demonstrates
the quantitative mRNA expression of pro-apoptotic molecule Bcl-2-associated X protein (Bax),
over 24 and 48 hours, post phenoxodiol treatment in prostate cancer cells. LNCaP and PC3
cells expressed no significant difference in Bax expression between DMSO vehicle control and
treatment groups over either time points. DU145 cells expressed an increase in Bax signalling
with both 10µM (p<0.05) and 30µM (p=0.037) phenoxodiol treatment groups after 48 hours
but no significant difference in the 24 hour cohort. There was no significant difference
detected between Bax signalling levels after treatment with either phenoxodiol
concentrations. Bax signalling was not found to be consistently altered across the prostate
cancer cells lines in response to phenoxodiol treatment.
147
LNCaP Bax qPCR Expression
25
Bax/L19
20
15
Control
10
10µM PXD
5
30µM PXD
0
24 Hours
48 Hours
Treatment
DU145 Bax qPCR Expression
20
Bax/L19
15
*
Control
10
10µM PXD
5
30µM PXD
0
24 Hours
48 Hours
Treatment
PC3 Bax qPCR Expression
25
Bax/L19
20
15
Control
10
10µM PXD
5
30µM PXD
0
24 Hours
48 Hours
Treatment
FIGURE 32 BAX MRNA EXPRESSION ANALYSIS OF PROSTATE CANCER CELLS POST PHENOXODIOL
TREATMENT OVER 24 AND 48 HOURS
* indicates significance relative to control for that time point.
148
Bcl-xL protects cells from apoptosis by regulating mitochondria membrane potential and
volume by interacting with pro-apoptotic members, Bax or Bim, and subsequently preventing
the release of Cytochrome c and other mitochondrial factors from the intermembrane space
into the cytosol (Sun et al. 2008). Figure 33 demonstrates the quantitative mRNA expression of
the anti-apoptotic molecule Bcl-xL, over 24 and 48 hours, post phenoxodiol treatment in
prostate cancer cells. The expression levels of Bcl-xL were not found to change significantly in
response to treatment with Phenoxodiol at any time point. Bcl-xL mRNA signalling was
consistently unresponsive following exposure to phenoxodiol treatment.
149
LNCaP Bcl-xL qPCR Expression
25
Bcl-xL/L19
20
15
Control
10
10µM PXD
5
30µM PXD
0
24 Hours
48 Hours
Treatment
Bcl-xL/L19
DU145 Bcl-xL qPCR Expression
70
60
50
40
30
20
10
0
Control
10µM PXD
30µM PXD
24 Hours
48 Hours
Treatment
PC3 Bcl-xL qPCR Expression
Bcl-xL/L19
2
1.5
Control
1
10µM PXD
0.5
30µM PXD
0
24 Hours
48 Hours
Treatment
FIGURE 33 BCL-XL MRNA EXPRESSION ANALYSIS OF PROSTATE CANCER CELLS POST PHENOXODIOL
TREATMENT OVER 24 AND 48 HOURS
150
Caspase-3 is a potent downstream apoptotic signalling molecule that, once activated, can
induce DNA fragmentation and other apoptotic events (Janicke et al. 1998; Janicke et al. 1998).
Figure 34 demonstrates the quantitative mRNA expression of the pro-apoptotic molecule
Caspase-3 over 24 and 48 hours, post phenoxodiol treatment in prostate cancer cells. LNCaP
and PC3 cells were not found to significantly change expression of Caspase-3 signalling over
any course of treatment and time points. The 48 hours of treatment, 30µM PXD, PC3 sample
was not significant (p=0.25). After 24 hours treatment DU145 cells indicated no expression
differences over DMSO vehicle control but the 48 hour 10µM (p=0.023) and 30µM (p<0.05)
PXD treatments were found to be significantly increased over DMSO vehicle control. The
Caspase-3 mRNA signalling results corresponded to the Caspase-3 activity assay and indicated
increased expression of Caspase-3 in DU145 cells after 48 hours of treatment with
phenoxodiol. While activated Caspase-3 was seen to decrease (chapter 4) after 48 hours in
LNCaP cells, the signalling was not found to be altered. The previously reported decrease in
activated Caspase-3, but a lack of detected signalling change could be due to the dramatic
decrease in live cell population after 48 hours in the LNCaP cell line. Therefore the LNCaP cells
would need to be tested for activated Caspase-3 between 24 and 48 hours to determine the
period where Caspase-3 signalling and expression is detectably altered but a majority of the
LNCaP cell population has not progressed to late stage apoptosis.
151
LNCaP Caspase-3 qPCR Expression
Caspase-3/L19
50
40
30
Control
20
10µM PXD
10
30µM PXD
0
24 Hours
48 Hours
Treatment
DU145 Caspase-3 qPCR Expression
Caspase-3/L19
0.6
0.5
0.4
*
0.3
Control
0.2
10µM PXD
0.1
30µM PXD
0
24 Hours
48 Hours
Treatment
PC3 Caspase-3 qPCR Expression
Caspase-3/L19
12
10
8
6
Control
4
10µM PXD
2
30µM PXD
0
24 Hours
48 Hours
Treatment
FIGURE 34 CASPASE-3 MRNA EXPRESSION ANALYSIS OF PROSTATE CANCER CELLS POST PHENOXODIOL
TREATMENT OVER 24 AND 48 HOURS
* indicates significance relative to control for that time point.
152
xIAP is a potent inhibitor of apoptosis that interacts with Caspase-3 thereby preventing
downstream apoptotic effects and Caspase-9 signalling (Straszewski-Chavez et al. 2004). Figure
35 demonstrates the quantitative mRNA expression of the anti-apoptotic molecule X-linked
Inhibitor of Apoptosis Protein (xIAP), over 24 and 48 hours, post phenoxodiol treatment in
cells. No significant changes in xIAP mRNA expression levels were detected across any
treatment group or time point. Phenoxodiol does not increase cytotoxicity through depression
of xIAP mRNA signalling in prostate cancer cell lines.
153
xIAP/L19
LNCaP xIAP qPCR Expression
3.5
3
2.5
2
1.5
1
0.5
0
Control
10µM PXD
30µM PXD
24 Hours
48 Hours
Treatment
DU145 xIAP qPCR Expression
5
xIAP/L19
4
3
Control
2
10µM PXD
1
30µM PXD
0
24 Hours
48 Hours
Treatment
PC3 xIAP qPCR Expression
12
xIAP/L19
10
8
6
Control
4
10µM PXD
2
30µM PXD
0
24 Hours
48 Hours
Treatment
FIGURE 35 XIAP MRNA EXPRESSION ANALYSIS OF PROSTATE CANCER CELLS POST PHENOXODIOL
TREATMENT OVER 24 AND 48 HOURS
Western blot protein analysis was performed on whole cell lysate to determine protein level
changes after phenoxodiol treatment for the genes AIF, Bax, Bcl-xL and xIAP. Activated
154
Caspase-3 protein was not detected because an activated Caspase-3 fluorescent activity assay
had already been performed. Cell protein expression was measured using densitometric
analysis after performing a Western blot, with protein expression compared to a standardising
protein, β-actin. AIF protein is a potent pro-apoptotic molecule that can induce apoptosis via a
caspase independent manner through mitochondrial permeabilisation (Susin et al. 1999).
Figure 36 demonstrates quantitative protein levels of the pro-apoptotic AIF protein over 24
and 48 hours post phenoxodiol treatment in prostate cancer cells. LNCaP cells were
determined to have significantly higher AIF protein levels after 24 hours of treatment with
30µM (p=0.015) PXD treatment, while at the 48 hour time point the 10µM PXD treatment was
found to express significantly lower levels of AIF than DMSO vehicle control (p=0.007) and
30µM PXD (p=0.008). The cell line DU145 was found to have no detectable levels of AIF
expression in any time point or group. PC3 cells were found to have significantly lower levels of
AIF protein versus DMSO vehicle control after 24 hours and 10µM of PXD treatment (p=0.036);
while, in the 48 hour treatment group, the 30µM PXD cohort exhibited a strong trend towards
biological significance but was not statistically significant (p=0.08). AIF protein expression was
not determined to consistently change after phenoxodiol treatment across the prostate cancer
cell lines.
155
LNCaP AIF Protein Levels
3.00
AIF/β-actin
2.50
2.00
**
1.50
Control
1.00
10µM PXD
0.50
30µM PXD
0.00
24 Hours
48 Hours
Treatment
PC3 AIF Protein Levels
1
AIF/β-actin
0.8
0.6
0.4
0.2
Control
10µM PXD
*
30µM PXD
0
24 Hours
48 Hours
Treatment
FIGURE 36 AIF PROTEIN LEVEL ANALYSIS OF PROSTATE CANCER CELLS POST PHENOXODIOL TREATMENT
OVER 24 AND 48 HOURS
* indicates significance relative to control for that time point. ** indicates significance relative
to 30µM Phenoxodiol treatment and control for that time point.
Bax is an important pro-apoptotic protein essential for the mitochondrial release of
Cytochome c, inducing cellular apoptosis (Lee et al. 2008). Figure 37 demonstrates quantitative
protein levels of the pro-apoptotic Bax protein over 24 and 48 hours post phenoxodiol
treatment in prostate cancer cells. LNCaP cells were found to have significantly lower levels of
Bax protein after 24 hours of treatment with both 10µM (p=0.024) and 30µM (p≤0.001) PXD
treatment. No significant change was detected in the LNCaP 48 hour treatment group. DU145
cells did not express detectable levels of Bax protein, the second pro-apoptotic protein found
156
to be down regulated to non-detectable levels in DU145 cells. PC3 cells expressed no
significant change in Bax protein expression across any time point or treatment group in
comparison to DMSO vehicle control. Bax protein expression was not determined to alter
consistently across the prostate cancer cell lines following treatment with phenoxodiol.
Bax/β-actin
LNCaP Bax Protein Levels
4
3.5
3
2.5
2
1.5
1
0.5
0
Control
*
10µM PXD
*
30µM PXD
24 Hours
48 Hours
Treatment
Bax/β-actin
PC3 Bax Protein Levels
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
Control
10µM PXD
30µM PXD
24 Hours
48 Hours
Treatment
FIGURE 37 BAX PROTEIN LEVEL ANALYSIS OF PROSTATE CANCER CELLS POST PHENOXODIOL
TREATMENT OVER 24 AND 48 HOURS
* indicates significance relative to control for that time point.
Bcl-xL protein is a potent anti-apoptotic agent that acts through the binding of molecules with
BH3 domain-only molecules in stable mitochondrial complexes, thereby preventing activation
157
of pro-apoptotic pathways such as Bax (Cheng et al. 2001). Figure 38 demonstrates
quantitative protein levels of the anti-apoptotic Bcl-xL protein over 24 and 48 hours post
phenoxodiol treatment in prostate cancer cells. LNCaP cells expressed no significant
differences across 24 hours but, after 48 hours of treatment, both 10µM (p=0.003) and 30µM
(p=0.0099) PXD treatment groups were significantly increased in the expression of Bcl-xL
protein versus DMSO vehicle control. DU145 cells were found to have no significant expression
changes across 24 hours of treatment, although the 30µM PXD treatment group trended
towards biological significance (p=0.08) versus DMSO vehicle control. After 48 hours of
treatment the DU145 30µM PXD treatment group was found to be significantly different to
both the DMSO vehicle control (p=0.013) and the 10µM PXD treatment group (p=0.0018). PC3
cells expressed no significant differences between groups across 24 hours of treatment but,
after 48 hours of treatment, both 10µM (p=0.04) and 30µM (p=0.041) PXD treatment groups
had significantly lower expression of Bcl-xL protein than DMSO vehicle control. Bcl-xL protein
expression was altered in an inconsistent manner between the three cell lines following
phenoxodiol treatment.
158
Bcl-xL/β-actin
LNCaP Bcl-xL Protein Levels
7
6
5
4
3
2
1
0
*
*
Control
10µM PXD
30µM PXD
24 Hours
48 Hours
Treatment
Bcl-xL/β-actin
DU145 Bcl-xL Protein Levels
1.6
1.4
1.2
1
0.8
0.6
0.4
0.2
0
**
Control
10µM PXD
30µM PXD
24 Hours
48 Hours
Treatment
PC3 Bcl-xL Protein Levels
Bcl-xL/β-actin
0.5
0.4
0.3
Control
0.2
10µM PXD
0.1
*
*
30µM PXD
0
24 Hours
48 Hours
Treatment
FIGURE 38 BCL-XL PROTEIN LEVEL ANALYSIS OF PROSTATE CANCER CELLS POST PHENOXODIOL
TREATMENT OVER 24 AND 48 HOURS
* indicates significance relative to control for that time point. ** indicates significance relative
to 10µM Phenoxodiol treatment and control for that time point.
159
X-linked inhibitor of apoptosis protein (xIAP) is part of a family of endogenous caspase
inhibitors that, when up regulated, have been implicated in the early stage development of
prostate and breast epithelial carcinoma (Parton et al. 2002; Krajewska et al. 2003). Figure 39
demonstrates quantitative protein levels of the anti-apoptotic xIAP protein over 24 and 48
hours post phenoxodiol treatment in prostate cancer cells. LNCaP cells were not found to have
a differential expression of xIAP protein across all time points and treatment groups versus
DMSO vehicle control. The DU145 cell population at 24 hours was determined to have a
significant decrease in xIAP protein expression in the 10µM (p=0.048) PXD versus DMSO
vehicle control, while the 30µM PXD treatment group was found to be significantly decreased
in comparison to both DMSO vehicle control (p=0.001) and 10µM PXD (p=0.005). No changes
were detected across the 48 hour treatment group in the DU145 cell population although the
30µM PXD group had a strong trend (p=0.07) towards a decrease in xIAP protein expression
compared to DMSO vehicle control. PC3 cells were not found to have a differential expression
of xIAP protein across all time points and treatment groups versus DMSO vehicle control.
160
xIAP/β-actin
LNCaP xIAP Protein Levels
1.6
1.4
1.2
1
0.8
0.6
0.4
0.2
0
Control
10µM PXD
30µM PXD
24 Hours
48 Hours
Treatment
DU145 xIAP Protein Levels
xIAP/β-actin
1
0.8
0.6
0.4
0.2
*
Control
10µM PXD
**
30µM PXD
0
24 Hours
48 Hours
Treatment
PC3 xIAP Protein Levels
xIAP/β-actin
2
1.5
Control
1
10µM PXD
0.5
30µM PXD
0
24 Hours
48 Hours
Treatment
FIGURE 39 XIAP PROTEIN LEVEL ANALYSIS OF PROSTATE CANCER CELLS OVER 24 AND 48 HOURS POST
PHENOXODIOL TREATMENT OVER 24 AND 48 HOURS
* indicates significance relative to control for that time point. ** indicates significance relative
to 10µM Phenoxodiol treatment and Control for that time point.
161
Following the results of the Caspase-3 activity assay, qPCR and Western blots indicating that
the apoptotic signalling was inconsistent between cell lines, a pan caspase inhibition assay was
performed to determine if phenoxodiol functioned through manipulation of the intrinsic or
extrinsic caspase signalling pathways. Following treatment with both phenoxodiol and a pancaspase inhibitor, an MTS proliferation assay measuring cell proliferation was performed.
Figure 40 demonstrates the effect of a 10µM treatment of pan-caspase inhibitor Z-VAD-FMK
on the proliferation of cells in conjunction with 30µM of phenoxodiol after 48 hours. None of
the cell lines exhibited a significant difference between their respective DMSO vehicle control
group and 10µM caspase inhibitor (CI) only treatments, nor the 30µM PXD and combined
10µM CI 30µM PXD treatment groups. The LNCaP cell 30µM PXD only treatment group was
found to be significantly decreased in proliferation versus DMSO vehicle control (p=0.048) and
the 10µM CI only treatment (p≤0.001). The LNCaP cell combined 10µM CI 30µM PXD
treatment had the same result with a significant decrease in cell proliferation versus DMSO
vehicle control (p=0.0024) and 10µM CI only treatments (p≤0.001). DU145 cells did not follow
the same trend strictly, the 30µM PXD (p<0.05) and combined 10µM CI 30µM PXD (p=0.029)
groups were significantly different versus DMSO vehicle control but not statistically different
to the 10µM CI only treatment. PC3 cells exhibited the same response to caspase inhibition
and phenoxodiol treatment as LNCaP cells. The PC3 cells 30µM PXD group was significantly
different to DMSO vehicle control (p≤0.001) and the 10µM CI only (p=0.0062) groups. The PC3
cells combined 10µM CI 30µM PXD group was significantly different to DMSO vehicle control
(p≤0.001) and the 10µM CI only (p=0.0058) groups.
162
% Cell Growth vs Control
LNCaP Z-VAD-FMK Caspase Inhibition
120
100
**
80
**
Negative
60
Control
40
CI 10µm
20
PXD 30µm
0
CI + PXD
48 Hours
Treatment
% Cell Growth vs Control
DU145 Z-VAD-FMK Caspase Inhibition
120
100
80
*
60
Negative
*
Control
40
CI 10µm
20
PXD 30µm
0
CI + PXD
48 Hours
Treatment
% Cell Growth vs Control
PC3 Z-VAD-FMK Caspase Inhibition
120
100
80
**
60
**
Negative
Control
40
CI 10µm
20
PXD 30µm
0
CI + PXD
48 Hours
Treatment
FIGURE 40 CASPASE INHIBITION TREATMENT WITH 10µM Z-VAD-FMK (CI) AND 30µM PHENOXODIOL
(PXD) OVER 48 HOURS
* indicates significance relative to control for that time point. ** indicates significance relative
to 10µM CI treatment and Control for that time point.
163
Chemotherapy and radiation treatments are known to induce oxidative stress in cancer cells
by rapid accumulation of highly reactive molecules such as nitric oxide (NO). After determining
that pan-caspase inhibition did not affect phenoxodiols ability to prevent the proliferation of
prostate cancer cells, a nitric oxide measurement assay, the Griess assay, was performed to
determine if phenoxodiol induced the production of nitric oxide, a potent free radical which
can induce cell cytotoxicity through oxidative stress. When this occurs it results in damage to
cell structures and potentially activates the apoptotic cascade, making it a target for
chemotherapy agents (Dharmarajan et al. 1999). Figure 41 demonstrates the amount of nitric
oxide produced by phenoxodiol addition to prostate cancer cells over 24 and 48 hours and
includes the positive control nitric oxide stimulators DEAN and SNP. None of the cell lines
exhibited significant changes in nitric oxide levels in comparison to DMSO vehicle control at
either 10µM or 30µM concentrations of phenoxodiol after 24 or 48 hours of treatment. LNCaP
cells exhibited a 1000% increase in NO production versus DMSO vehicle control under the
influence of DEAN (p<0.001) and while SNP effectively quadrupled control (p<0.001) it was
dwarfed by DEAN’s over 1000% increase in NO production and was significantly lower
(p<0.001) than DEAN stimulated production. DU145 and PC3 cell lines displayed a similar result
with a significant increase in NO production over DMSO vehicle control by DEAN stimulated
cells (DU145 p<0.001, PC3 p<0.001). In both PC3 and DU145 cell lines SNP stimulated an
increased production of NO versus DMSO vehicle control (DU145 p<0.001, PC3 p<0.001) and
continued the trend of SNP stimulated cells responding much lower than DEAN stimulated
cells (DU145 p<0.001, PC3 p<0.001). Both DU145 and PC3 cell lines had 1000% increase in NO
production under the influence of DEAN, similar to LNCaP; however, the SNP production was
only a 200% increase in NO production versus DMSO vehicle control whereas the LNCaP cells
had a 400% increase. This suggests the late stage representative cell lines, DU145 and PC3,
were less responsive to SNP stimulation than the early stage representative, LNCaP.
Phenoxodiol does not induce the production of reactive oxygen species in prostate cancer
cells.
164
% NO Production vs Control
LNCaP Nitric Oxide Production Following
Phenoxodiol Treatment
1200
*
Negative
800
Control
**
**
400
10µM PXD
30µM PXD
DEAN
0
24 Hours
48 Hours
SNP
Treatment
% NO Production vs Control
DU145 Nitric Oxide Production Following
Phenoxodiol Treatment
1200
Negative
800
Control
10µM PXD
400
30µM PXD
DEAN
0
24 Hours
48 Hours
SNP
Treatment
% NO Production vs Control
PC3 Nitric Oxide Production Following Phenoxodiol
Treatment
1200
Negative
800
Control
10µM PXD
400
30µM PXD
DEAN
0
24 Hours
48 Hours
SNP
Treatment
FIGURE 41 NITRIC OXIDE FORMATION IN PROSTATE CANCER CELLS OVER 24 AND 48 HOURS POST
PHENOXODIOL TREATMENT MEASURED VIA GRIESS ASSAY
165
* indicates significance relative to control for that time point. ** indicates significance relative
to 10µM DEAN treatment and control for that time point.
5.5.
DISCUSSION
Two of the hallmarks of cancer initiation and progression are the ability of cells to become
insensitive to anti-growth signals and to evade apoptosis. To do this the cells accumulate
increased expression of anti-apoptotic signalling molecules, which prevents activation of the
intrinsic and extrinsic signalling pathways that normally induce cell death (Hanahan and
Weinberg 2000). Here we report that phenoxodiol does not induce cytotoxicity through direct
manipulation of the expression of key elements of the extrinsic and intrinsic cell death
pathways. In the previous chapter we characterised the prostate cancer cells’ response to
phenoxodiol’s treatment and determined that cytotoxicity appears to be partly caused by
significant mitochondrial depolarisation after treatment. This results in a significant decrease
in cell population and increase in cell death via apoptosis and late stage apoptosis. The JC-1
study gave an indication that mitochondrial catastrophe may be occurring, which can affect
AIF, Bax, Bcl-xL, Caspase-3 and xIAP expression.
A key aspect of chemotherapeutic treatments is the ability to induce cytotoxicity in multiple
stages of tumour development. Treatments with a known, consistent method of action
maximise potential treatment candidates and have increased success rates. Successful
treatments, which induce cytotoxicity and target the extrinsic or intrinsic apoptotic signalling
pathways, are focussed on due to the well characterised ability of these pathways to induce
inflammation-free cell death (Brown and Wouters 1999; Fojo and Bates 2003). To elucidate the
mechanism by which phenoxodiol elicits an anti-proliferative effect in prostate cancer cell
lines, we investigated the kinetics of apoptosis induction. In the previous chapter we reported
166
a lack of Caspase-3 activation, or decrease, within the first 48 hours for LNCaP and PC3 cell
lines, with DU145 having a subsequent increase at the 48 hour time point. This was surprising
because we detected significant levels of apoptosis in LNCaP and DU145 cell lines with PC3
cells responding necrotically to treatment, and other studies have indicated the ability of
phenoxodiol to induce apoptosis through PARP-1 activation (Aguero et al. 2010). The
significant increase in Caspase-3 expression in DU145 cells, although minor, was in response to
an increase in Caspase-3 signalling following 48 hours of treatment. This confirmed that, in
some prostate cancer cell lines, phenoxodiol can induce a downstream classic apoptotic
response as previously reported (Kamsteeg et al. 2003; Aguero et al. 2005; Alvero et al. 2006;
Yu et al. 2006).
Prostate cancer has been shown to express Bcl-2 family members (Scherr et al. 1999) as well as
having changes in the ratio of pro- and anti-apoptotic signalling molecules following treatment
(Tang et al. 2006). However, a lack of change has also been previously reported in clonogenic
studies (Tannock and Lee 2001). The over expression of Bcl-xL leads to resistance to
chemotherapeutic drugs and radiotherapy, while depletion of Bcl-xL by shRNA, antisense
cDNA, or antisense oligodeoxynucleotides can impair tumour cell growth, sensitise cells to
apoptosis inducing agents, and increase cell sensitivity to chemotherapy (Wang et al. 2008).
Members of the BH3 family, such as Bcl-2 and Bcl-xL, are anti-apoptotic and maintain
mitochondrial membrane integrity by binding mitochondrial porin channels. These antiapoptotic proteins also protect cells from apoptosis by binding to pro-apoptotic BH3-only
members, such as Bid and Bax (Korsmeyer et al. 2000; Cheng et al. 2001; Cory and Adams
2002; Puthalakath and Strasser 2002). The ratio of pro-apoptotic Bax protein to anti-apoptotic
Bcl-xL can be a determinant of cellular susceptibility to apoptosis. The release of Bax from the
mitochondria in LNCaP, DU145 and PC3 cells was exhibited in the previously reported JC-1
mitochondrial depolarisation assay.
167
In this study we determined that the cytotoxicity of phenoxodiol was not associated with
consistent changes in both pro-apoptotic and anti-apoptotic members of the Bcl-2 family
between the cell lines, contrasting previous studies (Yu et al. 2006). LNCaP displayed a
decreased expression of Bax protein and increased expression of Bcl-xL protein, with cells
attempting to prevent the effect of phenoxodiol induced mitochondrial depolarisation by
mediating the effects of the Bcl-2 family. However, LNCaP cells still exhibited significant
cytotoxicity to treatment, indicating that the classic intrinsic apoptotic pathway might not be
directly responsible for the effect of phenoxodiol treatment. DU145 cells did not express Bax
protein at all, influencing the cell into a strong anti-apoptotic phenotype: yet a detectable
decrease in Bcl-xL protein was determined after treatment, indicating a susceptibility to
phenoxodiol induced mitochondrial depolarisation with resultant mitochondrial efflux. PC3
cells levels of Bax were unchanged following treatment, but the expression of Bcl-xL levels
were significantly reduced, indicating that the ratio of Bcl-xL:Bax had significantly decreased, a
sign of cell death signalling sensitivity from this notoriously chemoresistant cell line (Li et al.
2008).
Modulation of other pro- or anti-apoptotic proteins also remained inconsistent, indicating that
phenoxodiol could be inducing cytotoxicity through means other than the classical intrinsic and
extrinsic cell death pathways, with DU145 cells exhibiting downstream effects of these other
pathways. Phenoxodiol mediated cell death was not linked to the Caspase-independent
pathway of Apoptosis Inducing Factor (AIF). AIF induces caspase independent cell death
primarily through translocation from the mitochondria to the nucleus, where it mediates
chromatin condensation and high-molecular-weight fragmentation of DNA (Hong et al. 2004).
AIF expression was only mediated in the LNCaP cell line, with DU145 not expressing the
protein. Expression in LNCaP cells was found to decrease in response to decreased mRNA
168
signalling, which is counter intuitive to the effects of phenoxodiol as it was expected to
increase this expression to induce the cytotoxic effects that were exhibited. PC3 cells were not
determined to have much AIF signalling alteration except for a small decrease after 24 hours of
treatment, while the 48 hour treatment group remained unchanged.
The lack of a consistent change in gene expression between LNCaP, DU145 and PC3 cell lines
indicates that phenoxodiol does not promote cell death by disrupting the stoichiometry of key
pro- and anti-apoptotic molecules investigated here. The common model for cancer therapy
activity has been that anti-cancer regimens cause apoptotic signalling to occur, often mediated
by p53 and Caspase-3 upregulation (Fojo and Bates 2003). Therefore, cells that are resistant to
apoptosis achieve this through forced over expression of anti-apoptotic molecules such as xIAP
and Bcl-xL family members or p53 mutation (Brown and Attardi 2005). Members of the
Inhibitors of Apoptosis Proteins (IAP), such as xIAP, can prevent apoptosis by inactivating
Caspase-3 and Caspase-9. The results reported in this study are in contrast to other reports,
where it has been demonstrated in ovarian cancer that caspase activation, Bax upregulation
and xIAP degradation are key proteins modulated by phenoxodiol (Sapi et al. 2004; Alvero et
al. 2006; Mor et al. 2006; Kluger et al. 2007). We have previously established a link between
chemoresistance and xIAP expression in refractory ovarian cancer (Amantana et al. 2004). We
detected only decreased xIAP expression in the DU145 cells following 24 hours of treatment.
This decrease coincides with an increase in Caspase-3 mRNA signalling and activated
expression, reinforcing the ability of phenoxodiol to induce DU145 cells into an intrinsically
mediated apoptotic response downstream of the pathway phenoxodiol is interacting with.
The inability of the pan caspase inhibitor, Z-VAD-FMK, to prevent the anti-proliferative activity
of phenoxodiol in all three cell lines reinforces that the phenoxodiol induced mechanism of cell
169
death in prostate cancer is occurring via a caspase-independent pathway. This is in contrast to
many studies that indicate PARP and caspase activation as the effect of phenoxodiol treatment
(Kamsteeg et al. 2003; Aguero et al. 2005). One potential method of induction investigated
was the ability of phenoxodiol to induce accumulation/production of reactive oxygen species
(ROS) in the cell, causing oxidative stress. Mitochondrial swelling and volume increase due to
depolarisation is linked to ROS production, where oxidative stress induces the expression of
specific genes that contain an antioxidant response element (ARE) in their promoters. ARE
promoted genes can trigger apoptosis or necrosis of various cell types and, in several
instances, can inhibit cell growth and interfere with the cell cycle (Juhaszova et al. 2004;
Minelli et al. 2009). The cell cycle modulator p21WAF1 has been implicated in protecting the cell
from NO induced cell death (Yang et al. 2000) and it has also been reported that anti-apoptotic
Bcl-2 family members, such as Bcl-xL, reduce accumulation of reactive oxygen species (ROS) in
transfected cells and can protect a variety of cells from apoptosis induced by oxidative
stressors. Conversely, Bax expression causes increased mitochondrial production of ROS and
apoptosis when over expressed in cultured cells (Dharmarajan et al. 1999). This study
investigated the effects of phenoxodiol in producing nitric oxide (NO), a potent ROS molecule,
after the effect of phenoxodiol on mitochondrial depolarisation was exhibited. We discovered
that, in all cell lines, phenoxodiol did not induce NO production as a response to treatment,
but that all three cell lines retained the ability to have NO production stimulated by DEAN and
SNP addition.
Recent reports indicate that, in addition to apoptosis, cancer cells can be effectively
eliminated through necrosis, mitotic catastrophe and premature senescence, which result in
cell cycle arrest and subsequent cell death signalling (Brown and Wouters 1999; Brown and
Wilson 2003; Tannock et al. 2004). Cells that are non-responsive to treatment through the
classic intrinsic apoptosis pathways, yet have an apoptotic response, must activate this process
170
through other mechanisms such as mitochondrial response and depolarisation, mitotic
catastrophe and DNA degradation (Ruth and Roninson 2000; Brown and Wilson 2003).
Additionally, cells can undergo apoptosis through mechanisms such as Cytochrome c release,
Caspase-8 and Caspase-9 signalling, as well as high-molecular-weight DNA breakdown. We
previously exhibited the ability of phenoxodiol induction to induce efflux of mitochondria and,
therefore, Cytochrome c from the mitochondria into the cytoplasm. Indeed, it would appear
from our studies of cell lines representative of different phases of prostate cancer, that
phenoxodiol elicits an alternative induction of cytotoxicity instead of classical extrinsic and
intrinsic apoptotic signalling.
This study indicates that phenoxodiol exhibits a significant ability to induce cytotoxicity and cell
death in the prostate cancer cell lines LNCaP, DU145 and PC3 through different signalling
pathways than those previously reported in ovarian cancer studies, and clearly demonstrates
the caspase independent induction of this cell death. This study also determined that nitric
oxide production and, therefore, oxidative stress is not altered after treatment. Coupled with
the induction of apoptotic cell death through non-classical pathways, phenoxodiol shows
potential as a drug for future treatment of prostate cancer that is resistant to apoptotic
signalling.
171
6. CHAPTER SIX: THE EFFECTS OF
PHENOXODIOL ON THE CELL CYCLE
6.1.
INTRODUCTION
Two of the hallmarks of cancer are the ability to be self-sufficient in growth signals and to have
infinite replicative potential, which can be initiated through damaging the cell cycle restriction
points, thereby allowing for progression of cancer from benign to metastatic states (Hanahan
and Weinberg 2000). Dysregulated growth rates can initiate cancer as more cell cycle
machinery becomes damaged, mutations are allowed to accumulate, methylation status
changes occur and cells become undifferentiated and progress towards metastasis. The cell
cycle is composed of G1, S, G2 and M phases, which represent normal function, DNA
replication, organelle replication and mitotic separation respectively (Marieb 2012). Any errors
that may occur at these steps could be catastrophic for normal cell functioning and, as such,
the cell cycle apparatus retains potent signalling molecules that can search for errors and
rapidly induce apoptosis (Cooper 2003). The initial point where this process can occur is
referred to as the G1 restriction point (Pardee 1974).
To maintain normal function, the cell cycle is a potent inducer of cell death through regulation
of DNA content at a variety of restriction points. As the cell reaches a restriction point, the
gene p53 detects and attempts to repair any evident DNA damage; if the damage is too great it
can initiate apoptosis. The p53 gene is a potent tumour suppressor that exerts its functions
through activation of downstream targets, some of which include induction of CDKN1/p21WAF1,
14-3-3y, and REPRIMO for G1-G2 arrest; p53R2 for DNA repair and Bax, Puma, p53AIP1, PERP,
and CD95 for apoptosis (Wasylishen and Penn 2010). Unfortunately p53 is found mutated in
nearly 70% of carcinomas. Considerable evidence indicates that the choice made by p53 to
172
activate cell cycle arrest and DNA repair pathways, or the apoptosis pathway after DNA
damage, is dependent on the extent of unrepaired or misrepaired double-strand breaks in the
DNA (Devlin et al. 2008). Disabling the p53 pathway enables cells to enter and proceed
through the cell cycle under conditions that increase the frequencies of aneuploidy, gene
amplification, deletion and translocation. This, coupled with loss of p53 dependant apoptosis,
increases genetic instability and is highly selected during cancer progression (Vafa et al. 2002).
Cell cycle progression is intricately regulated by the interactions of cyclin and cyclin-dependent
kinase (Cdk) complexes. Cyclin-D1 is a regulatory subunit of the highly conserved cyclin family
that phosphorylates and, together with sequential phosphorylation by Cyclin-E/CDK2,
inactivates the cell-cycle inhibiting function of the retinoblastoma protein. Upon mitogenic
stimulation, the Cyclin-D Cdk4/Cdk6 and Cyclin-E Cdk2 complexes mediate phosphorylation of
retinoblastoma protein, which induces transcriptionally active E2F thereby ensuring G1-S
transit (Baldin et al. 1993; Roy et al. 2007). Retinoblastoma protein serves as a gatekeeper of
the G1 phase, and passage through the restriction point leads to DNA synthesis, making CyclinD1 expression an important target to regulate cancer cell replication (Fu et al. 2004).
P21WAF1 is a cyclin dependant kinase inhibitor (CdkI) family member, along with p27 and p57,
which interfere with the cyclin dependant kinase-cyclin complex. The group is regulated both
by internal and external signals, with the expression of p21WAF1 under transcriptional control of
the p53 tumour suppressor gene (Srivastava et al. 2007). As an important mediator of
apoptosis, the frequency of somatic mutations in the p21WAF1 gene and family in cancers is very
rare; which underlines the importance of these molecules as promising therapeutic targets.
Both p21 and p27 are upregulated by a variety of regulatory pathways at the transcriptional as
well as post-transcriptional levels, with p21WAF1 transcriptionally upregulated by p53 in
173
response to DNA damage. p21WAF1 is also activated by various transcription factors and
subsequently mediates growth arrest, senescence and apoptosis in a p53-independent
manner, and p21 mRNA stability can be post transcriptionally regulated by HuR, an RNAbinding protein, in response to stress (Roy et al. 2007; Roy et al. 2008).
An important indicator of cell proliferation rate, Ki-67 is present in all proliferating cells,
normal and neoplastic, and its level of expression indicates the rate of growth of a cell
population. Studies proved the hypothesis that, in breast cancer, the Ki-67 index is an
independent prognostic factor in both patient survival assessment and disease recurrence. The
Ki-67 index can also be used as a predictive factor of neoplastic cell response to certain types
of therapy as well as a measure capable of indicating success of treatment (Koda et al. 2007).
C-Myc is an important regulator of cell function that was implicated as one of the first
oncogenes. It is known that cell cycle regulation is altered under excess c-Myc expression with
a decrease in time taken to reach the restriction point of G1. Cell cycle regulation has been
implicated by varied means; canonical Wnt/β-catenin stimulation, Cyclin-D and Cyclin-E CDK2
stimulation, inhibiting the effect of p27 a CdkI family member amongst others (Mateyak et al.
1999). Conversely, c-Myc expression has also been promoted as a control mechanism for cell
cycle-induced apoptosis and prevention of differentiation in normal functioning cells
(Wasylishen and Penn 2010).
Initiating the normal and cancerous prostate cell cycle progression is the activity of the growth
factor androgen receptor (AR), which is found to be mutated in most late stage prostate
cancers. The AR is widely accepted to be the initiator of prostate cell replication and
differentiation signalling through transcription of mitogenic and anti-apoptotic factors. Cell
signalling pathways, such as the canonical Wnt/β-catenin signalling pathway, drive the cells
174
towards growth and differentiation while interaction with anti-apoptotic pathways, such as the
Bcl-xL pathway, prevents the cells from entering into apoptosis when the AR is stimulated (Sun
et al. 2008; Wang et al. 2008). Because AR signalling has also been shown to play a key role in
prostate carcinogenesis, androgen ablation therapy is a commonly used form of treatment,
particularly for advanced disease. While ablation therapy leads to initially significant levels of
prostate cancer cell apoptosis, the effect is short-lived and ultimately not curative because
most patients develop androgen-independent/hormone refractory disease (D'Antonio et al.
2008). Manipulation of this growth factor by upregulation of receptors, mutation to a
constitutively active form or a complete loss of AR as a mitogenic requirement are
characteristics of late stage prostate cancer (Peng et al. 2008).
Levels of β-catenin and AR are both increased in hormone refractory prostate cancer, βcatenin is activated by canonical Wnt stimulation, but mutated forms of β-catenin, which can
result in a permanently stabilised protein, have also been detected in prostate cancer. βcatenin has multiple functions that involve both cell adhesion and signal transduction in
response to Wnt ligands (Chesire et al. 2002). In the absence of Wnt ligands, GSK3β complexes
with other proteins and degrades cytoplasmic β-catenin, but when Wnt ligands bind to the
frizzled receptor complex, the resulting activation of the cytoplasmic protein dishevelled
inactivates GSK3β, thereby preventing degradation of β-catenin (Wan et al. 2012). It is
important to note that GSK3β has been linked to prevention of AR transcriptional activity; so
Wnt ligand binding both increases AR expression and β-catenin stability (Li et al. 2008). Upon
stimulation, β-catenin translocates to the nucleus where it can drive cell cycle progression
through interaction with T-cell factor (TCF)/lymphoid enhancer factor (LEF) transcription
factors to initiate transcription of target genes such as c-Myc and Cyclin-D1, driving cell
growth. As such, dysregulation of the Wnt, AR or β-catenin pathways can result in a
metastatic, highly replicative phenotype (Wang et al. 2008). It has been postulated that the
175
ratio of β-catenin/AR might be an important prognostic indicator that may even help define a
subpopulation of men with prostate cancer for individualised management (Wan et al. 2012).
Secreted frizzled related protein 4 (sFRP4) is one of a group of proteins that antagonise the
canonical Wnt/β-catenin pathway, decreasing Wnt activity and preventing activated β-catenin
from forming. sFRP4 can also decrease invasiveness in androgen-independent prostate cancer
cells and has been shown to be anti-angiogenic and pro-apoptotic in nature (Muley et al.
2010). Moreover, the correlation between increased membranous sFRP4 and β-catenin
expression in a large human cohort supports evidence for sFRP4 as a prognostic marker in
localised androgen-dependent prostate cancer. Unlike other inhibitors of Wnt signalling, sFRP4
appears to affect androgen-dependent and androgen-independent prostate cancer (Horvath et
al. 2004; Horvath et al. 2007).
In this chapter we explored the ability of phenoxodiol to impact the cell cycle of prostate
cancer cells and investigate the underlying signalling pathways c-Myc, Cyclin-D1, Ki-67 and
p21WAF1. We then investigated whether phenoxodiol alters the expression of the active protein
form of β-catenin, which is known to be important in prostate homeostasis and whose
expression has been implicated in poor prognosis. Finally, we investigated sFRP4 protein
expression, after phenoxodiol treatment because sFRP4 is an inhibitor of the canonical Wnt/βcatenin signalling pathway and a potential inhibitor of prostate cancer cell growth.
176
6.2.
AIMS
The aims of this chapter were as follows:
Aim 1: To determine if phenoxodiol affected the cell cycle phases of the prostate cancer lines
LNCaP, DU145 and PC3.
Aim 2: To determine if signalling pathways c-Myc, Cyclin-D1, Ki-67, p21WAF1 were altered post
phenoxodiol treatment.
Aim 3: To determine if the expression of sFRP4 ligand was altered in response to phenoxodiol
treatment.
6.3.
METHODOLOGY
The methodology utilised in this chapter is discussed in detail in the materials and methods
chapter (page 50). Analysing the population of cells in each cell cycle phase indicated whether
phenoxodiol had a direct effect upon the cell cycle of prostate cancer cells. Briefly, cells were
seeded into 6 well plates and incubated for 48 hours before 24 or 48 hours of treatment media
were applied. Cell media were aspirated and centrifuged and cells were trypsinised, then
placed into the same tube and centrifuged. Cells were washed with PBS and then -20°C 70%
ethanol was added drop wise, while vortexing to fixate the cells. Cells were stored at 4°C then
rehydrated with PBS and stained with propidium iodide before being placed into a FACS Canto
II cytometer, which determined fluorescence and therefore DNA content. FlowJo software was
utilised to further analyse the sample and distinguish quantitative populations of G1, S and G2
phase cells.
177
Once it was determined that phenoxodiol had an impact on the cell cycle it was necessary to
explore the underlying cell cycle and cell proliferation signalling that could be the target of
phenoxodiol treatment. Quantitative PCR (qPCR) analysis was performed on all three cell lines,
LNCaP, DU145 and PC3 after 24 and 48 hours of 10µM and 30µM phenoxodiol treatment. As
previously states, cells were seeded, treated and then RNA extraction was performed using the
Chomczynski guanidinium thiocyanate-phenol-chloroform method (Chomczynski and Sacchi
1987). Purified RNA was analysed and converted into cDNA then cleaned with a post PCR kit.
Each qPCR experiment was performed using 2µL of cDNA per sample and primers were
designed to test for c-Myc, Cyclin-D1, Ki-67 and p21WAF1 by using published primers. Primer
spanning of an intron/exon boundary was confirmed using Primer3 and BLAST search engines.
Quantitative PCR (qPCR) was performed as previously states, using 2µL of cDNA to a master
mix of primers/Taq and then applying 35 temperature cycles at varying annealing
temperatures on a Corbett Rotorgene 3000 or 6000. The standards used were 10 fold dilutions
of a purified gel extract and were given concentrations. The resulting standard curve then
determined final sample concentrations compared against the expression of L19. Protein level
analysis was performed using a Western blot protocol as previously described. Cells were
seeded, treated and protein extracted with a RIPA buffer/β-mercaptoethanol based, whole cell
lysate method. After protein concentration was determined it was loaded onto a SDSPage/acrylamide based gel and current applied to the gel. After protein separation, transfer to
a nitrocellulose membrane was performed and Ponceau red used to determine effective
transfer. Finally, primary antibodies were added to the membrane and incubated in the
appropriate conditions, excess was removed with TBS-T washes and a secondary antibody
conjugated to HRP applied in the appropriate conditions. A chemiluminescent substrate kit
was then used to detect antibody binding concentration and results compared to β-actin
protein.
178
6.4.
RESULTS
In the previous chapter we determined that apoptotic signalling was not consistently altered in
response to phenoxodiol treatment and, therefore, it was not directly targeting one specific
pathway tested. Another of the hallmarks of cancer, the ability to limitlessly replicate, was
investigated through looking at the cell cycle response to phenoxodiol treatment. After 24 or
48 hours of phenoxodiol treatment, cells were stained with propidium iodide and analysed to
determine DNA content, which was then assessed into cell cycle phase populations G1, S and
G2 phase using FlowJo software.
Figure 42 demonstrates the LNCaP cell line cell cycle response to 10µM and 30µM phenoxodiol
treatment over 24 and 48 hours by assessing the cell cycle phase populations differentiated by
DNA content. Phenoxodiol induced significantly decreased G2 phase cell populations versus
DMSO vehicle control, over 24 hours for both 10µM (p<0.001) and 30µM (p<0.001)
phenoxodiol treatments in LNCaP cells. The S phase cell population was found to increase
versus DMSO vehicle control following 24 hours of 10µM (p<0.0021) and 30µM (p=0.0016)
phenoxodiol treatment. The 10µM and 30µM phenoxodiol treatment groups were not
significantly different versus each other after 24 hours.
Phenoxodiol induced significantly decreased G2 phase cell populations over 48 hours, versus
DMSO vehicle control for both 10µM (p<0.001) and 30µM (p=0.0016) treatments in LNCaP
cells. Only the 10µM phenoxodiol treatment was determined to increase the S phase cell
population (p=0.0056) after 48 hours but the 30µM phenoxodiol treatment had a significantly
increased cell population in G1 phase versus both DMSO vehicle control (p=0.0013). In LNCaP
cells phenoxodiol was determined to consistently decrease G2 phase cell population over 24
and 48 hours following treatment with 10µM and 30µM concentrations.
179
LNCaP Cell Cycle Response to 24 Hours of PXD
Treatment
% of Cell Population
100
80
60
Control
40
10µM PXD
20
30µM PXD
0
G1
S
G2
Treatment
LNCaP Cell Cycle Response to 48 Hours of PXD
Treatment
% of Cell Population
100
80
60
Control
40
10µM PXD
20
30µM PXD
0
G1
S
G2
Treatment
FIGURE 42 LNCAP CELL CYCLE ANALYSIS AFTER 24 AND 48 HOURS OF 10µM AND 30µM PHENOXODIOL
TREATMENT
* indicates significance relative to control for that time point. ** indicates significance relative
to 10µM Phenoxodiol treatment and control for that time point.
Figure 43 demonstrates the DU145 cell line cell cycle response to 10µM and 30µM
phenoxodiol treatment over 24 and 48 hours by assessing the cell cycle phase populations
differentiated by DNA content. Following 24 hours of phenoxodiol treatment the DU145 G2
phase cell population was significantly decreased versus DMSO vehicle control in both 10µM
(p=0.0028) and 30µM (p=0.0021) concentrations. Only the 10µM phenoxodiol treatment had
an increased S phase cell population versus DMSO vehicle control (p<0.001) but the 30µM
180
treatment had an increased G1 phase cell population versus DMSO vehicle control (p=0.0038)
and 10µM (p=0.0049).
Phenoxodiol induced significantly decreased G2 phase cell populations in DU145 cells, over 48
hours for both 10µM (p<0.001) and 30µM (P<0.001) treatments versus DMSO vehicle control
and between the treatment concentrations (P<0.001). The S phase cell population was
significantly increased versus DMSO vehicle control in both 10µM (p<0.001) and 30µM
(p<0.001) treatments as well as significantly different between treatments (p<0.001). The G1
phase cell population was significantly decreased versus DMSO vehicle control for the 10µM
treatment (p<0.001) and significantly increased versus DMSO vehicle control (p<0.001) and
versus 10µM (p<0.001) for the 30µM treatment group. In DU145 cells phenoxodiol was
determined to consistently decrease G2 phase cell population over 24 and 48 hours following
treatment with 10µM and 30µM concentrations.
181
DU145 Cell Cycle Response to 24 Hours of PXD
Treatment
% of Cell Population
100
80
60
Control
40
10µM PXD
20
30µM PXD
0
G1
S
G2
Treatment
DU145 Cell Cycle Response to 48 Hours of PXD
Treatment
% of Cell Population
100
80
60
Control
40
10µM PXD
20
30µM PXD
0
G1
S
G2
Treatment
FIGURE 43 DU145 CYCLE ANALYSIS AFTER 24 AND 48 HOURS OF 10µM AND 30µM PHENOXODIOL
TREATMENT
* indicates significance relative to control for that time point. ** indicates significance relative
to 10µM Phenoxodiol treatment and control for that time point.
182
Figure 44 demonstrates the PC3 cell line cell cycle response to 10µM and 30µM phenoxodiol
treatment over 24 and 48 hours by assessing the cell cycle phase populations differentiated by
DNA content. Following 24 hours of phenoxodiol treatment the PC3 G2 phase cell population
was significantly decreased versus DMSO vehicle control in both 10µM (p<0.001) and 30µM
(p=0.0071) concentrations. No significant differences were detected in G1 phase cell
population but S phase cell populations were significantly increased versus DMSO vehicle
control for 10µM (p<0.001) and 30µM (p=0.0022) treatments.
Following 48 hours of treatment the PC3 G2 phase cell population was significantly decreased
versus DMSO vehicle control in both 10µM (p<0.001) and 30µM (p<0.001) treatments as well
as significantly different between the treatments (p<0.001). The PC3 S phase cell population
was significantly increased versus DMSO vehicle control in both 10µM (p<0.001) and 30µM
(p<0.001) and significantly different between treatments (p<0.001). The PC3 G1 phase cell
population was significantly decreased versus DMSO vehicle control in both 10µM (p<0.001)
and 30µM (p<0.022) treatments and significantly different between the treatments (p<0.001).
In PC3 cells phenoxodiol was determined to consistently decrease G2 phase cell population
over 24 and 48 hours following treatment with 10µM and 30µM concentrations.
183
PC3 Cell Cycle Response to 24 Hours of PXD
Treatment
% of Cell Population
100
80
60
Control
40
10µM PXD
20
30µM PXD
0
G1
S
G2
Treatment
PC3 Cell Cycle Response to 48 Hours of PXD
Treatment
% of Cell Population
100
80
60
Control
40
**
10µM PXD
20
30µM PXD
0
G1
S
G2
Treatment
FIGURE 44 PC3 CELL CYCLE ANALYSIS AFTER 24 AND 48 HOURS OF 10µM AND 30µM PHENOXODIOL
TREATMENT.
* indicates significance relative to control for that time point. ** indicates significance relative
to 10µM Phenoxodiol treatment and control for that time point.
184
Once it was determined that phenoxodiol had an impact on the cell cycle, it was necessary to
explore the underlying cell cycle and cell proliferation signalling that could be the target of
phenoxodiol treatment. Quantitative PCR (qPCR) analysis was performed on all three cell lines,
with the housekeeping gene L19 used as a standardising agent, control expression was
designated as one, for ease of graphing.
c-Myc is a potent initiator of cell replication and has been implicated in increasing the rate at
which cells enter S phase (Wasylishen and Penn 2010). Figure 45 demonstrates the
quantitative mRNA expression of c-Myc in cells treated over 24 and 48 hour periods with
phenoxodiol. After 48 hours of 30µM phenoxodiol treatment, PC3 cells were found to
significantly increase the expression of c-Myc versus DMSO vehicle control (p=0.033). Neither
LNCaP nor DU145 cells were found to have any significant changes in c-Myc expression in
response to phenoxodiol treatment.
185
LNCaP c-Myc qPCR Expression
3
c-Myc/L19
2.5
2
Control
1.5
1
10µM PXD
0.5
30µM PXD
0
24 Hours
48 Hours
Treatment
DU145 c-Myc qPCR Expression
3
c-Myc/L19
2.5
2
Control
1.5
1
10µM PXD
0.5
30µM PXD
0
24 Hours
48 Hours
Treatment
PC3 c-Myc qPCR Expression
2.5
c-Myc/L19
2
1.5
Control
1
10µM PXD
0.5
30µM PXD
0
24 Hours
48 Hours
Treatment
FIGURE 45 C-MYC MRNA EXPRESSION ANALYSIS OF PROSTATE CANCER OVER 24 AND 48 HOURS POST
PHENOXODIOL TREATMENT.
* indicates significance relative to control for that time point.
186
Cyclin-D1 is recognised as potent initiator of cell cycle progression from G1 through to S phase
by the Cyclin-D1 Cdk4 complex activating the Cyclin E Cdk2 complex, which results in inhibition
of the cell cycle inhibiting Rb protein (Ladha et al. 1998). Figure 46 demonstrates the
quantitative mRNA expression of the cell cycle regulator gene Cyclin-D1 over 24 and 48 hours
post phenoxodiol treatment in prostate cancer cells. Decreasing Cyclin-D1 expression, in
response to phenoxodiol treatment, could result in quiescent and apoptotically sensitive cells.
LNCaP cells did not have a detectable change in Cyclin-D1 expression level under the influence
of phenoxodiol treatment. DU145 cells were found to have a significant decrease in the
expression of Cyclin-D1 versus DMSO vehicle control (p=0.0071) after 24 hours of treatment
with 30µM phenoxodiol no other changes in expression were detected in the DU145 cell line.
PC3 cells exhibited a similar trend to treatment as the DU145 cells, with both the 10µM
phenoxodiol (p=0.026) and 30µM phenoxodiol (p=0.0011) treatments significantly decreased
in expression versus DMSO vehicle control after a 24 hour period of treatment.
187
LNCaP Cyclin-D1 qPCR Expression
CyclinD1/L19
2.5
2
1.5
Control
1
10µM PXD
0.5
30µM PXD
0
24 Hours
48 Hours
Treatment
CyclinD1/L19
DU145 Cyclin-D1 qPCR Expression
1.6
1.4
1.2
1
0.8
0.6
0.4
0.2
0
Control
10µM PXD
30µM PXD
24 Hours
48 Hours
Treatment
PC3 Cyclin-D1 qPCR Expression
1.2
CyclinD1/L19
1
0.8
0.6
Control
0.4
10µM PXD
0.2
30µM PXD
0
24 Hours
48 Hours
Treatment
FIGURE 46 CYCLIN-D1 MRNA EXPRESSION ANALYSIS OF PROSTATE CANCER CELLS OVER 24 AND 48
HOURS POST PHENOXODIOL TREATMENT.
* indicates significance relative to control for that time point.
188
Ki-67 expression is an effective indicator of rate of proliferation in cells (Koda et al. 2007).
Figure 47 demonstrates the quantitative mRNA expression of the cell proliferation gene Ki-67
over 24 and 48 hours post phenoxodiol treatment in prostate cancer cells. LNCaP cells were
significantly decreased in Ki-67 mRNA signalling over 24 hours with both 10µM (p=0.0043) and
30µM (p=0.0065) phenoxodiol treatments exhibiting decreased mRNA expression versus
DMSO vehicle control. DU145 cells had no significant difference in Ki-67 mRNA expression over
24 and 48 hours of treatment with 10µM and 30µM PXD although a biological trend towards
decreased expression was indicated (p=0.082, p=0.1 respectively). PC3 cells exhibited a
significant decrease in Ki-67 mRNA signalling, over 24 hours, in both 10µM (p=0.034) and
30µM (p=0.036) phenoxodiol treatments while the 48 hour 10µM phenoxodiol treatment
exhibited significantly decreased mRNA expression versus DMSO vehicle control (p<0.05) and
30µM phenoxodiol (p=0.032).
189
LNCaP Ki-67 qPCR Expression
1.2
Ki-67/L19
1
0.8
0.6
Control
0.4
10µM PXD
0.2
30µM PXD
0
24 Hours
48 Hours
Treatment
DU145 Ki-67 qPCR Expression
2.5
Ki-67/L19
2
1.5
Control
1
10µM PXD
0.5
30µM PXD
0
24 Hours
48 Hours
Treatment
Ki-67/L19
PC3 Ki-67 qPCR Expression
1.6
1.4
1.2
1
0.8
0.6
0.4
0.2
0
Control
10µM PXD
30µM PXD
24 Hours
48 Hours
Treatment
FIGURE 47 KI-67 MRNA EXPRESSION ANALYSIS OF PROSTATE CANCER CELLS OVER 24 AND 48 HOURS
POST PHENOXODIOL TREATMENT.
* indicates significance relative to control for that time point. ** indicates significance relative
to 30µM Phenoxodiol treatment and control for that time point.
190
Figure 48 demonstrates the quantitative mRNA expression of the cell proliferation gene
p21WAF1 over 24 and 48 hours post phenoxodiol treatment in prostate cancer cells. P21WAF1 is a
cell cycle inhibiting factor that can prevent the formation of the Cyclin E Cdk2 complex. This
inhibition prevents the progression of the cell through G1 to S phase (Roy et al. 2007). LNCaP
cells exhibited a significant increase in p21 mRNA expression versus DMSO vehicle control over
24 hours of treatment with 10µM phenoxodiol (p=0.0099), and over 48 hours of treatment
with both 10µM (p<0.050) and 30µM phenoxodiol (p=0.011) concentrations. DU145 cells
exhibited a significant increase in p21 mRNA expression versus DMSO vehicle control after 48
hours of treatment with both 10µM (p=0.0048) and 30µM (p=0.028) phenoxodiol
concentrations. PC3 cells exhibited a significant increase in p21 mRNA expression over both 24
and 48 hours of treatment (p=0.042 and p=0.0044 respectively) with the 30µM phenoxodiol
treatment group.
191
LNCaP p21 qPCR Expression
12
p21/L19
10
8
6
Control
4
10µM PXD
2
30µM PXD
0
24 Hours
48 Hours
Treatment
p21/L19
DU145 p21 qPCR Expression
4
3.5
3
2.5
2
1.5
1
0.5
0
Control
10µM PXD
30µM PXD
24 Hours
48 Hours
Treatment
PC3 p21 qPCR Expression
25
p21/L19
20
15
Control
10
10µM PXD
5
30µM PXD
0
24 Hours
48 Hours
Treatment
FIGURE 48 P21 MRNA EXPRESSION ANALYSIS OF PROSTATE CANCER OVER 24 AND 48 HOURS POST
PHENOXODIOL TREATMENT.
* indicates significance relative to control for that time point.
192
β-catenin is the activated downstream protein of the canonical Wnt/β-catenin signalling
pathway that can interact with androgen receptor and, as one from a list of potential targets,
activate Cyclin-D1 and therefore the cell cycle (Yu et al. 2011). Figure 49 demonstrates the
protein expression of active β-catenin over 24 and 48 hours post treatment with phenoxodiol
LNCaP cells exhibited a significant decrease in β-catenin protein versus DMSO vehicle control
after 24 hours of 10µM phenoxodiol treatment (p=0.045) and after 48 hours of treatment with
both 10µM (p=0.023) and 30µM (p=0.046) phenoxodiol. DU145 cells exhibited an increase in
β-catenin expression with the 24 hour 10µM phenoxodiol treatment group increased over
both DMSO vehicle control (p=0.019) and 30µM phenoxodiol (p=0.011), while the 48 hour
30µM phenoxodiol treatment was increased versus DMSO vehicle control (p=0.012). PC3 cells
were similar to the LNCaP cell response with a decrease over 24 hours versus DMSO vehicle
control in both 10µM (p=0.022) and 30µM (p<0.05) phenoxodiol treatment groups.
193
LNCaP active β-catenin Protein Levels
β-catenin/β-actin
1.2
1
0.8
0.6
Control
0.4
10µM PXD
0.2
30µM PXD
0
24 Hours
48 Hours
Treatment
β-catenin/β-actin
DU145 active β-catenin Protein Levels
3.5
3
2.5
2
1.5
1
0.5
0
Control
10µM PXD
30µM PXD
24 Hours
48 Hours
Treatment
β-catenin/β-actin
PC3 active β-catenin Protein Levels
1.4
1.2
1
0.8
0.6
0.4
0.2
0
Control
10µM PXD
30µM PXD
24 Hours
48 Hours
Treatment
FIGURE 49 ACTIVE Β-CATENIN PROTEIN LEVEL ANALYSIS OF PROSTATE CANCER CELLS OVER 24 AND 48
HOURS POST PHENOXODIOL TREATMENT.
* indicates significance relative to DMSO vehicle control for that time point. ** indicates
significance relative to 30µM Phenoxodiol treatment and DMSO vehicle control for that time
point.
194
sFRP4 is an antagonist to the canonical Wnt/β-catenin signalling pathway that interacts with
androgen receptor and can activate Cyclin-D1, and therefore the cell cycle (Horvath et al.
2007; Drake et al. 2009). Figure 50 demonstrates the protein level of sFRP4 over 24 and 48
hours post phenoxodiol treatment in prostate cancer cells. LNCaP cells exhibited a significant
increase in sFRP4 protein expression after treatment with 10µM phenoxodiol over 24
(p=0.0035) and 48 hours (p<0.05) versus control. While the 30µM treatments were not
statistically significant versus control they trended towards biological significance over 24
(p=0.1) and 48 (p=0.066) hours of treatment. DU145 cells exhibited a significant increase in
sFRP4 levels after 24 hours of treatment with 30µM phenoxodiol versus control (p=0.028) and
versus 10µM phenoxodiol (p=0.0072). While no significant difference was determined over 48
hours versus control, a significant difference was detected between the 10µM and 30µM
phenoxodiol treatments (p=0.038). Finally PC3 cells exhibited a significant increase in sFRP4
levels versus control in the 10µM (p=0.021) and 30µM (p=0.008) over 24 hours of treatment.
No significant difference was detected in the 48 hour treatment group of PC3 cells. sFRP4
levels were found to alter after treatment with phenoxodiol with LNCaP, DU145 and PC3 cells
lines exhibiting increases in protein expression following 24 hours of treatment.
195
LNCaP sFRP4 Protein Levels
sFRP4/β-actin
5
4
3
Control
2
10µM PXD
1
30µM PXD
0
24 Hours
48 Hours
Treatment
DU145 sFRP4 Protein Levels
sFRP4/β-actin
0.8
0.6
Control
0.4
10µM PXD
0.2
30µM PXD
0
24 Hours
48 Hours
Treatment
PC3 sFRP4 Protein Levels
sFRP4/β-actin
2
1.5
Control
1
10µM PXD
0.5
30µM PXD
0
24 Hours
48 Hours
Treatment
FIGURE 50 SFRP4 PROTEIN LEVEL ANALYSIS OF PROSTATE CANCER CELLS OVER 24 AND 48 HOURS POST
PHENOXODIOL TREATMENT
* indicates significance relative to DMSO vehicle control for that time point. ** indicates
significance relative to DMSO vehicle control and 10µM Phenoxodiol treatment for that time
point. Ϯ indicates significance relative to 30µM Phenoxodiol only for that point.
196
6.5.
DISCUSSION
Two of the hallmarks of cancer are the ability to be self-sufficient in growth signals and to have
infinite replicative potential, which can be initiated through damaging the cell cycle restriction
points, allowing for progression of cancer from benign to metastatic (Hanahan and Weinberg
2000). Here we report that phenoxodiol induces cell cycle arrest in the G1/S phase of the cell
cycle, with the resultant arrest due to the upregulation of p21WAF1. The cytotoxicity may be due
to downstream signalling of molecules such as Akt and ASK1 (Bott et al. 2005; Xie et al. 2010;
Ahmad et al. 2011). c-Myc is a potent oncogene and expression was found to alter in PC3 cells
in response to phenoxodiol. We also report that the expression of Ki-67 and Cyclin-D1 was
altered after phenoxodiol treatment and that the canonical Wnt pathway protein, β-catenin
levels were decreased in LNCaP and PC3 cells while the Wnt receptor antagonist sFRP4 levels,
were increased in all cell lines.
Upon activation of mitogenic signalling, cells commit to entry into a series of regulated steps
allowing completion of the cell cycle. Cells begin in G1 phase, the time between M and S
phases, and before entry into S phase, where DNA is replicated, must pass through a
restriction point (Pardee 1974) that analyses and attempts to repair DNA damage. After S
phase, cells enter G2 phase (the time between the S and M phases) where cells can repair
errors that occurred during DNA duplication, preventing the propagation of these errors to
daughter cells. Finally, the separation into two daughter cells by chromatid separation occurs
and is called M phase (Senderowicz 2004). The sequence of events in cell cycle progression is
highly orchestrated and depends on the cyclic activation and inactivation of cyclin dependent
kinases (CDK), which govern the progression of the cells from one phase to another. In the
event of tumourigenisis, constitutive mitogenic signalling as well as mutations in tumour
197
suppressor genes and proto-oncogenes leads to cell cycle deregulation and uncontrolled
proliferation (MacLachlan et al. 1995; Roy et al. 2007).
The tumour suppressor p53 is the primary controller of cell cycle activity, which triggers cell
cycle arrest, induces the repair of DNA damage or apoptosis by induction of p21WAF1, p53R2,
Bax and Puma (Devlin et al. 2008). p21WAF1 is a cyclin dependant kinase inhibitor (CdkI) family
member, along with p27 and p57, which interfere with the cyclin dependant kinase-cyclin
complex. The group is regulated both by internal and external signals with the expression of
p21WAF1 under transcriptional control of the p53 tumour suppressor gene (Srivastava et al.
2007). In this study we exhibited the ability of phenoxodiol to induce cell cycle arrest at both
10µM and 30µM concentrations over 24 and 48 hours of treatment, with the resultant cells
exhibiting significantly decreased cell populations of G2 cells and subsequent arrest of the cell
cycle visible in the significant alteration of G1 and S phase cell populations. In all cell lines the
decrease in G2 phase cell population was consistent and resulted in very low cell populations,
while some treatments resulted in high S phase arrest and others in G1 phase arrest, it’s clear
that the method of phenoxodiol induced cell cycle arrest is independent of p53 status, with
LNCaP (p53-wild type), DU145 (p53-mutated) and PC3 (p53-null) cells all representing different
p53 status cell types.
p21WAF1 is a tumour suppressor gene that can induce disruption of the Cyclin-e/Cdk2 complex
and prevent subsequent progression from G1 phase into S phase of the cell cycle (el-Deiry et
al. 1993; Bott et al. 2005). Numerous studies have shown that upregulation of p21WAF1 causes
growth arrest in various cancer models and, though p21WAF1 was initially identified to be
transcriptionally up-regulated by p53 in response to DNA damage, recent studies have shown
that p21WAF1 can also be induced by various transcription factors with subsequent mediation of
198
cell cycle arrest, senescence, and apoptosis in a p53 independent manner (Aguero et al. 2005;
Roy et al. 2008). The ability of isoflavones, and specifically phenoxodiol, to induce cell cycle
arrest has been previously reported with studies indicating that arrest was induced by p21WAF1
stabilisation and expression increase (Aguero et al. 2005; Aguero et al. 2010; Seo et al. 2011).
We investigated the expression of p21WAF1 after the data indicated significant cell cycle arrest
as a response to phenoxodiol treatment. p21WAF1 signalling expression was found to be
significantly increased across all the cell lines in response to treatment, indicating that
activation of p21WAF1 was occurring via a p53 independent manner, with resulting cell cycle
arrest. The induction of cytotoxicity in the cells was independent of caspase activation, as
previously shown, and potentiated by induced mitotic depolarisation. This confirms previous
studies that have indicated that isoflavones induce cell cycle arrest through activation and
stabilisation of p21WAF1 (Aguero et al. 2010; Seo et al. 2011).
The assembly of Cyclin-D1, with its CDK4/6 partners, is a mitogen regulated process occurring
in early G1; with the resultant Cyclin-D1-CDK4/CDK6 complexes promoting G1 progression by
inhibiting the activity of the retinoblastoma protein (Rb), resulting in activation of E2F and
subsequent cyclin/cdk signalling, which enters the cell into the cell cycle (Baldin et al. 1993;
Ladha et al. 1998). Many oncogenic signals induce Cyclin-D1 expression and do so through
distinct DNA sequences in the Cyclin-D1 promoter, including Ras, Src,ErbB2, and β-catenin.
Decreasing the expression of Cyclin-D1, or interference with the cyclin/Cdk complex results in
cell arrest in G1 phase and eventual senescence. We investigated the expression of Cyclin-D1
after treatment with phenoxodiol and determined that DU145 and PC3 both had a significant
decrease in signalling over 24 hours but not over 48, while LNCaP cells did not change
expression of cyclin-D1 signalling. While not a direct target of phenoxodiol treatment in
prostate cancer cells, the role of Cyclin-D1 seems to be tissue and oncogene specific, with
Cyclin-D1 linked to activation of the Wnt/β-catenin signalling pathway (Fu et al. 2004).
199
Ki-67 antigen is present in all proliferating cells (normal and neoplastic) and its evaluation
allows determination of the rate of growth. Ki-67 expression has been shown to have a strong
relationship with Gleason's grading, which has an important correlation with the prognosis of
prostate cancer and, as such, it is an independent predictive factor in both patient survival
assessment and disease recurrence (Koda et al. 2007; Madani et al. 2011). Ki-67 expression
was found to significantly decrease in LNCaP and PC3 cells, while there was a biological trend
towards this in DU145 cells but error margins resulted in no detection of significant alteration.
The decreased Ki-67 signalling expression confirms the cell cycle arrest data and indicates that
prostate cancer cells are undergoing senescence induced cytotoxicity. c-Myc signalling has
been shown to be a proto-oncogene, with regulation of c-Myc expression inducing the
expression of other oncogenes, in response and leading to a neoplastic cell type. It is known
that cell cycle regulation is altered under excess c-Myc expression, with a decrease in time
taken to reach the restriction point of G1 (Wasylishen and Penn 2010). In this study we
determined that c-Myc expression was only altered under the influence of phenoxodiol in PC3
cells after 48 hours of treatment, with signalling potentially being a mechanism to drive the
cell out of arrest.
Another controller of the cell cycle is the canonical Wnt/β-catenin pathway where Wnt growth
signals cause decreased expression of GSK3β protein, which results in an active form of βcatenin translocating to the nucleus and initiating cell cycle progression through multiple
pathways including Cyclin-D1 and c-Myc expression (Zhang et al. 2011). An up regulated
canonical Wnt/β-catenin pathway has multiple functions; it can be potentiated by AR
expression and has been implicated in both cell adhesion and signal transduction in response
to Wnt ligands, resulting in an invasive metastatic phenotype (Chesire et al. 2002). We
investigated the expression of the active form of β-catenin and determined that LNCaP and
PC3 cells had significantly reduced active β-catenin expression after 24 hours of phenoxodiol
200
treatment, with LNCaP continuing on to a significant 48 hour effect, which coincides with a
subsequent decrease in Cyclin-D1 signalling expression in androgen insensitive PC3 cells while
LNCaP expression was unchanged due to the potential interaction of AR receptor. DU145 cells
exhibited an increased expression of active β-catenin after treatment over 24 and 48 hours,
indicating that alteration of canonical Wnt signalling is not a direct effect of phenoxodiol
treatment but a downstream signalling outcome.
Secreted frizzled related protein 4 (sFRP4) is one of a group of proteins that antagonise the
canonical Wnt/β-catenin pathway. The interference with Wnt activity results in decreased
activated β-catenin formation and decreased invasive potential. sFRP4 has been verified to
have an anti-angiogenic ability as well as the ability to induce apoptotic signalling (Drake et al.
2009; Muley et al. 2010). sFRP4 appears to affect androgen-dependent and androgenindependent prostate cancer and has a role as a prognostic marker for androgen dependant
prostate cancer (Horvath et al. 2007). We determined that sFRP4 expression was increased
over 24 hours in PC3 cells, coinciding with decreased active β-catenin levels in that cell line and
increased levels over 24 and 48 hours in LNCaP cells, coinciding with decreased active βcatenin in that cell line. sFRP4 levels were increased in DU145 cells but did not correspond
with β-catenin changes, indicating that DU145 cells might have an alteration in the canonical
Wnt/β-catenin signalling pathway. Phenoxodiol exhibited the potential to regulate canonical
Wnt/β-catenin signalling through increased expression of sFRP4 and subsequent decreased
expression of active β-catenin in two of the cell lines. The functionality of this canonical Wnt/βcatenin antagonism in prostate cancer must be determined before treatment with
phenoxodiol can be utilised for successful targeting of the pathway directly. However, soy
isoflavone interaction with the pathway has previously been exhibited (Liss et al. 2010).
201
In this study we determined that phenoxodiol treatment induced significant cell cycle arrest
across 24 and 48 hours of 10µM and 30µM phenoxodiol treatment. p21WAF1 signalling
expression was found to be significantly increased across all the cell lines in response to
treatment, indicating that activation of p21WAF1 and subsequent cell arrest was occurring via a
p53 independent manner, with induction of cytotoxicity independent of caspase activation.
We determined that c-Myc and Cyclin-D1 expression was not consistently altered but that Ki67 signalling expression was decreased in line with the cell cycle arrest. The interaction of
sFRP4 and active β-catenin implied that phenoxodiol treatment results in an alteration of
sFRP4 and β-catenin signalling though changes are dependent on the cell line. Phenoxodiol
demonstrates an ability in prostate cancer cells to induce significant cytotoxicity in cells by
interacting with p21WAF1 and inducing cell cycle arrest irrespective of p53 status or caspase
pathway interactions. These data indicate that phenoxodiol would be effective as a potential
future treatment modality for both hormone sensitive and hormone refractory prostate
cancer.
202
7. CHAPTER SEVEN: PHENOXODIOL IN
COMBINATION WITH DOCETAXEL
7.1.
INTRODUCTION
Advanced prostate cancer has a 5-year survival of only 30%. Historically, chemotherapy has
been used with palliative intent but unclear survival benefit for these advanced-stage patients
(Montero et al. 2005). The lack of curative measures for late stage prostate cancer and the
ability to develop resistance to treatment have been linked to the existence of cancer stem
cells. The cancer stem cell population acts as an initiator of cancer and exists in low numbers.
Cancer stem cells have the ability to acquire resistance upon exposure to treatment, which
results in the development of a new population of cells that are now resistant to previously
successful treatment (Reya et al. 2001; Trosko et al. 2004). Current practices for hormonerefractory/castrate resistant, metastatic prostate cancer incorporate the use of taxanes.
Docetaxel, in particular, is being incorporated in numerous current clinical trials either as a
single or combination agent against androgen-independent prostate cancer, and it is also
being investigated for its use as a neoadjuvant or adjuvant agent in hormone sensitive, locally
aggressive prostate cancer (Canfield et al. 2006).
Docetaxel and paclitaxel are from a class of anti-cancer agents called taxanes that bind to and
stabilise microtubules causing G2/M cell-cycle arrest and apoptosis. Taxanes have a different
mechanism of action from that of any other class of anti-cancer drugs, namely hyperstabilisation of microtubules (Montero et al. 2005). Taxanes function by targeting the subunit
of the tubulin heterodimer, the key component of cellular microtubules that allow for
chromatid separation during the M cell cycle phase. Although the action and anti-cancer
activity of paclitaxel and docetaxel are much the same, key differences exist clinically;
203
docetaxel shows activity in patients with metastatic solid tumours that are resistant to
paclitaxel (Michaud et al. 2000). The actual mechanisms that lead to cell death remain unclear
but may include activation of intrinsic pathways essential for apoptosis, induction of bcl-2
phosphorylation facilitating apoptosis and inhibition of angiogenesis. Cell death after exposure
to docetaxel appears to involve apoptotic mechanisms, including classic features such as DNA
fragmentation, cell volume shrinkage, and membrane-bound apoptotic bodies (Michaud et al.
2000). Studies have also shown that apoptosis induced by taxanes involves several apoptotic
signal molecules, such as JNK, protein kinase A, c-Raf-1/Ras/Bcl-2, p53/p21WAF1 and mitogenactivated protein kinases (ERK and p38). However, the mode of apoptotic action in different
tumours is far from clear (Gan et al. 2011).
Combination therapies have the potential to increase the effectiveness of drug treatments
while simultaneously increasing quality of life by reducing side effects, lowering effective
dosage rates or by increasing effectiveness of one compound once combined with another.
Successful combination therapies that produce an increase in quality of life for patients, even if
not associated with life extension, are a highly valued outcome of research, even with drugs
that do not directly interact (Kantoff et al. 1999; Parente et al. 2012). An example of this is the
use of Docetaxel with androgen ablation therapy in late stage prostate cancer defined as
castrate resistant prostate cancer (CRPC). To date, docetaxel-based chemotherapy remains the
only treatment that has demonstrated an overall survival benefit in most men with metastatic
CRPC regardless of whether they are symptomatic or have visceral metastases (Hotte and Saad
2010). Despite the initial effectiveness of these treatments, late stage prostate cancer
treatment is still palliative and characterised by androgen independence, metastasis and
poorly differentiated cells.
204
To determine an effective combination therapy accurately, the interaction of the two
compounds is analysed and the effectiveness investigated for synergistic, additive or even
interference effects (Zhao et al. 2004). True synergism is rare as most drugs in combination are
additive. Due to clearance rates of these drugs through the body keeping serum levels at an
effective concentration means that, in an additive environment, one might not be able to
lower effective concentrations, thereby increasing the risk of side effects but theoretically
decreasing the length of time treatment must be carried out (Tallarida 2006). Isobolograms are
one method of determining effective ranges of drug treatments in combination, because
multiple concentrations of drugs can be compared against each other and a profile of effective
dose rates can be constructed for the environment being tested. In prostate cancer surgical
intervention, radiotherapy and chemotherapy are utilised, sometimes in combination, to treat
tumours.
Cytoplasmic and/or nuclear β-catenin can be used as an indicator of activation of the Wnt/βcatenin pathway because it is observed in up to 71% of advanced prostate tumour specimens
(Chesire et al. 2002). Increasing evidence also suggests that canonical Wnt signalling could
function to assist in bone metastasis formation, a progressive state that has a particularly poor
prognosis (Hall et al. 2006). It has been revealed that the Wnt ligand target, the frizzled
receptor, is over expressed in prostate cancer but that this over expression is counterbalanced
by the secreted frizzled related protein (sFRP) family which attempts to suppress AR-mediated
transactivation (Kawano et al. 2009).
Methylation of the sFRP proteins has been suggested as a marker of invasive carcinoma, with a
resulting poor prognosis (Ahmad et al. 2011). Apart from β-catenin degradation, it has been
suggested that sFRP4 inhibition of Wnt signalling causes stabilisation of GSK3β, resulting in p53
205
activation and associated signalling such as p21WAF1, as well as a decreased invasive potential in
androgen-independent prostate cancer cells. sFRP4 has also been indicated to have
antiangiogenic and pro-apoptotic properties in a wide variety of cells (Watcharasit et al. 2002;
Fox and Dharmarajan 2006; Muley et al. 2010). Unlike other inhibitors of Wnt signalling, sFRP4
appears to affect androgen-dependent and androgen-independent prostate cancer (Horvath et
al. 2007).
In this study we seek to investigate the cytotoxic effects of the prostate cancer treatment
docetaxel, and determine a range of treatment concentrations. Potential interactions of
phenoxodiol and docetaxel across multiple concentrations are investigated, searching for
synergistic, additive or interference effects. We also investigate if any cytotoxic effects are
increased through pretreatment with docetaxel or phenoxodiol and, finally, we try to
determine if the Wnt/β-catenin antagonist sFRP4, when combined with phenoxodiol, will have
an interactive effect.
206
7.2.
AIMS
The aims of this chapter were as follows:
Aim 1: To characterise the cytotoxic effect of Docetaxel on prostate cancer cells.
Aim 2: To determine the effect of combination therapy with docetaxel and phenoxodiol on
prostate cancer cells.
Aim 3: To determine if pretreatment with phenoxodiol or docetaxel has an impact on the
effect of subsequent combination therapy.
Aim 4: To determine the effect of combination therapy with phenoxodiol and sFRP4 purified
protein on prostate cancer cells.
7.3.
METHODOLOGY
The methodology utilised in this chapter is discussed in detail in the materials and methods
chapter (page 50). Initially, a dose response curve to docetaxel would be needed to determine
effective dose rates to use in conjunction with phenoxodiol. Briefly, cells were seeded onto 96
well plates at appropriate rates and incubated for 48 hours. Media were then aspirated and
replaced with a 10 fold serially diluted concentration of docetaxel from 10µM to .1nM, with
DMSO as the vehicle control, and all samples having an equal concentration of DMSO vehicle
added. After 48 hours of treatment, 20µL of MTS dye were added and the cells incubated for 3
hours before absorbance was measured at 492nm. Increased absorbance was linked to
increased cell metabolism and, therefore, proliferation. No interaction was noted between
docetaxel and the MTS dye.
207
An isobologram was performed between docetaxel and phenoxodiol; the samples were all
balanced for vehicle concentration. Briefly, cells were seeded into 96 well plates and incubated
for 48 hours; afterwards they had the cell media aspirated and a treatment media applied,
containing any combination listed in Table 23, resulting in 24 different treatments.
TABLE 23 PHENOXODIOL/DOCETAXEL CONCENTRATION COMBINATIONS FOR AN ISOBOLOGRAM
Drug Name
Conc.
Phenoxodiol 0µM
5µM
Docetaxel
.1nM 1nM
0nM
10µM 30µM
5nM
10Nm 100nM
A list of potential concentration combinations of Phenoxodiol and Docetaxel during the
isobologram, resulting in a final quantity of 24 individual treatment types, all balanced for
vehicle control.
The isobologram was performed after 48 hours of incubation with treatment media; 20µL of
MTS dye were then added to each well and incubated for 3 hours. Afterwards media
absorbance was measured at 492nm, with the resulting absorbance indicating cell
proliferation rates within the prostate cancer cell lines.
A pretreatment study was then performed. Briefly, cells were seeded at appropriate rates into
96 well plates and incubated for 48 hours, treatment media were applied to each well and
incubation occurred for another 24 hours. After 24 hours, a set of 96 well plates had 20µL of
MTS dye added, were incubated for 3 hours and absorbance was measured at 492nm on a
plate reader. The other plates had the addition of new treatment compounds added to
appropriate wells, i.e. some cells had exposure to 24 hours of 30µM phenoxodiol then 24
208
hours of exposure to 100nM docetaxel and 30µM phenoxodiol in combination. This tested for
potential induction of pre-treatment sensitivity. After 48 hours of total treatment with the
initial compound concentration, the remaining 96 well plates had 20µL of MTS dye added and
were incubated for 3 hours, followed by measurement of absorbance at 492nm on a plate
reader.
Finally, a Wnt antagonist and phenoxodiol treatment study was performed. Briefly, cells were
seeded into 96 well plates at appropriate rates and incubated for 48 hours, after which cells
were treated with varying concentrations of purified sFRP4 protein up to 500pg/mL, which was
combined with 30µM phenoxodiol treatment. Vehicle consisted of PBS and DMSO
respectively. Following 48 hours of exposure to treatment, cells had 20µL of MTS dye added
and were incubated for 3 hours before absorbance was measured at 492nm on a plate reader.
209
7.4.
RESULTS
Docetaxel is a common treatment for early and late stage prostate cancer, and is highly
cytotoxic as a cell microtubulin stabiliser, causing G2 phase cell cycle blockage in replicating
cells by damaging the mitotic spindle (Michaud et al. 2000). As phenoxodiol is a novel drug for
prostate cancer with an ability to affect the cell cycle, a series of experiments were run to
determine potential synergy, additivity, interference or lack of interaction between each
compound.
Docetaxel is known to induce G2 phase cell cycle arrest by inhibiting the activity of the mitotic
spindle (Montero et al. 2005). Figure 51 demonstrates the cell proliferation rates of prostate
cancer cell lines after 48 hours of treatment with 10 fold dilutions of docetaxel spanning .1nM
to 10µM concentrations. LNCaP cells exhibited a significant increase in cell proliferation versus
DMSO vehicle control after treatment with 1nM docetaxel (p≤0.001). LNCaP cells treated with
10nM, 100nM, 1µM, and 10µM concentrations all exhibited a significant decrease in
proliferation versus control (p=0.033, p≤0.001 for the last 3). While 100nM, 1µM, and 10µM
concentrations were not significantly different to each other they were all significantly
decreased over the 10nM docetaxel concentration (p≤0.001 for all).
DU145 cells had a similar pattern of response to phenoxodiol treatment with .1nM and 1nM
doses significantly increasing cell proliferation versus control (p≤0.001 for all). While 10nM,
100nM, 1µM, and 10µM concentrations exhibited a significant decrease in proliferation versus
control (p≤0.001 for all). There was no significant difference detected between 100nM, 1µM,
and 10µM docetaxel treatments but all three were significantly decreased with respect to the
10nM treatment (p≤0.001 for all).
210
PC3 cells exhibited a similar trend with .1nM and 1nM docetaxel treatments significantly
increasing cell proliferation versus control (p≤0.001 for both), and 10nM, 100nM, 1µM and
10µM docetaxel concentrations significantly decreasing proliferation versus control (p≤0.001
for all). Only the 1µM docetaxel concentration was found to be significantly decreased versus
10nM and 100nM concentrations (p=0.042). All three cell lines exhibited a significant decrease
in proliferation in response to high concentrations of docetaxel, as such the 100nM dose of
docetaxel was chosen as the lowest effective dose over 48 hours of treatment.
211
% Cell Growth versus Control
LNCaP Docetaxel Response over 48 Hours
140
120
100
80
60
40
20
0
Treatment
Control
.1nM
1nM
10nM
100nM
1µM
10µM
% Cell Growth versus Control
DU145 Docetaxel Response over 48 Hours
140
120
100
80
60
40
20
0
Control
Treatment
.1nM
1nM
10nM
100nM
1µM
10µM
% Cell Growth versus Control
PC3 Docetaxel Response over 48 Hours
140
120
100
80
60
40
20
0
Control
Treatment
.1nM
1nM
10nM
100nM
1µM
10µM
FIGURE 51 DOCETAXEL RESPONSE CURVE MEASURED AFTER 48 HOURS OF TREATMENT.
* indicates significance relative to control after 48 hours. ** indicates significance relative to
control and * groups after 48 hours. *** indicates significance relative to control and *, **
groups after 48 hours.
212
Phenoxodiol has exhibited the ability to induce G1/S phase cell cycle arrest, while docetaxels
published method of action is to inhibit the cell cycle at the M phase, resulting in G2 phase
arrest (Michaud et al. 2000; Aguero et al. 2010). After determining the minimum effective
concentration dose an isobologram was performed to look for interaction between
phenoxodiol and docetaxel with varying concentrations of each utilised. Figure 52
demonstrates an isobologram analysis of phenoxodiol and docetaxel treatments, over 48
hours, with a total of 24 individual treatments all standardised to DMSO vehicle control in
LNCaP cells. Isobolograms determine interference, synergy and additivity with significant
interference effects indicated in the follow graphs. Additive effects were evident throughout
the isobologram but, while there was a significant decrease in cell proliferation rate versus
control (0µM PXD, 0nM DOC) over most populations, there was a significant increase in
proliferative rate versus 100nM docetaxel only for that time point (p=0.014) indicating that the
5µM phenoxodiol treatment inhibited the effectiveness of 100nM docetaxel concentration.
These data provide evidence for phenoxodiol interfering with LNCaP cell docetaxel treatment
at a specific concentration combination.
LNCaP Phenoxodiol/Doxetaxel Isobologram
% Cell Growth vs Control
140
120
100
80
60
40
20
0
0nM DOC .1nM DOC 1nM DOC
0µM PXD
5µM PXD
5nM DOC 10nM DOC 100nM DOC
10µM PXD
30µM PXD
FIGURE 52 LNCAP PHENOXODIOL / DOCETAXEL ISOBOLOGRAM MEASURED AFTER 48 HOURS OF
TREATMENT
213
* indicates significant increase in proliferation relative to [0µM PXD, 100nM DOC] treatment
after 48 hours.
Figure 53 demonstrates an isobologram analysis of phenoxodiol and docetaxel treatments
over 48 hours with a total of 24 individual treatments all standardised to DMSO vehicle control
in DU145 cells. Additive effects were evident throughout the isobologram but, while there was
a significant decrease in cell proliferation rate versus control (0µM PXD, 0nM DOC) over most
populations, the indicated samples displayed a significant increase in proliferative rate versus
100nM docetaxel only for that time point (5µM PXD p≤0.001, 10µM PXD p=0.015), indicating
that the 5µM and 10µM concentrations of phenoxodiol inhibited the effectiveness of a 100nM
docetaxel concentration. These data provide evidence for phenoxodiol interfering with DU145
cell docetaxel treatment at a specific concentration combination.
DU145 Phenoxodiol/Docetaxel Isobologram
% Cell Growth vs Control
140
120
100
80
60
40
20
0
0nM DOC .1nM DOC 1nM DOC 5nM DOC 10nM DOC
0µM PXD
5µM PXD
10µM PXD
100nM
DOC
30µM PXD
FIGURE 53 DU145 PHENOXODIOL / DOCETAXEL ISOBOLOGRAM MEASURED AFTER 48 HOURS OF
TREATMENT.
* indicates significant increase in proliferation relative to [0µM PXD, 100nM DOC] treatment
after 48 hours.
214
Figure 54 demonstrates an isobologram analysis of phenoxodiol and docetaxel treatments
over 48 hours with a total of 24 individual treatments all standardised to DMSO vehicle control
in PC3 cells. Additive effects were evident throughout the isobologram but, while there was a
significant decrease in cell proliferation rate versus control (0µM PXD, 0nM DOC) over most
populations, the indicated samples displayed a significant increase in proliferative rate versus
100nM docetaxel only for that time point (p=0.015), indicating that the 5µM phenoxodiol
treatment inhibited the effectiveness of 100nM docetaxel only concentration. These data
provide evidence for phenoxodiol interfering with PC3 cell docetaxel treatment at a specific
concentration combination.
PC3 Phenoxodiol/Docetaxel Isobologram
% Cell Growth vs Control
140
120
100
80
60
40
20
0
0nM DOC .1nM DOC 1nM DOC 5nM DOC 10nM DOC
0µM PXD
5µM PXD
10µM PXD
100nM
DOC
30µM PXD
FIGURE 54 PC3 PHENOXODIOL / DOCETAXEL ISOBOLOGRAM MEASURED AFTER 48 HOURS OF
TREATMENT.
* indicates significant increase in proliferation relative to [0µM PXD, 100nM DOC] treatment
after 48 hours.
Following the isobologram results, a more in depth investigation of potential sensitisation
needed to be performed with pre-treatment of the cells with phenoxodiol or docetaxel and
215
then subsequent addition of the remaining agent. Figure 55 demonstrates a combination
therapy treatment of 10µM phenoxodiol and 100nM docetaxel in prostate cancer cells over 24
and 48 hours, with two samples receiving 24 hours solo drug treatment and then the next 24
hours of combined treatment. LNCaP cells exhibited a significant decrease in proliferation
versus control over 24 hours in the 10µM PXD (p=0.0089) and combination (p<0.001)
treatments, while combination therapy was also significantly lower than the 10µM PXD
treatment (p=0.0096). LNCaP cells exhibited a significant decrease in proliferation versus
control over 48 hours when 10µM PXD (p<0.001), combination (p<0.001) and 48Hr PXD
(p<0.001) treatments were used. The 100nM DOC (p<0.05) and 48Hr DOC (p=0.0035) were all
significantly decreased in proliferation versus 48Hr PXD as well as 10µM PXD combination and
Control treatments. Lastly, 48hr DOC was also significantly decreased versus 100nM DOC
(p=0.022).
DU145 cells exhibited a significant decrease in proliferation over 24 hours with 10µM PXD
(p<0.001), 100nM DOC (p<0.001) and combination (p<0.001) treatments. After 48 hours,
DU145 cells exhibited significantly decreased proliferation versus control in 10µM PXD
(p<0.001) and 48Hr PXD (p<0.001) treatments but exhibited no significant difference between
each other, while 100nM DOC (p=0.022), combination (p=0.027) and 48Hr DOC (p=0.019)
treatments were all significantly decreased versus 48Hr PXD as well as 10µM PXD and Control.
PC3 cells exhibited a significant decrease in proliferation versus control after 24 hours of
treatment with 10µM PXD (p<0.001), 100nM DOC (p<0.001) and combination (p<0.001)
treatments, while 100nM DOC (p=0.0072) and combination (p=0.024) treatments were also
significantly decreased versus 10µM PXD. Following the trend set by LNCaP and DU145 cells,
PC3 cells exhibited a significant decrease in proliferation versus control over 48 hours after
216
treatment with 48Hr PXD (p<0.0011), while 10µM PXD (p=0.0062), 100nM DOC (p=0.0041),
combination (p=0.0026) and 48Hr DOC (p=0.0029) exhibited significantly decreased
proliferation versus 48Hr PXD as well as Control.
* indicates significance relative to control for that time point. ** indicates significance relative
to control and * groups for that time point. *** indicates significance relative to control and *,
** groups for that time point.
The [48Hr 10µM PXD+24Hr 100nM DOC] and [48Hr 100nM DOC+24Hr 10µM PXD] treatments
will be referred to as the 48hr PXD and 48Hr DOC treatments respectively while the [48Hr
10µM PXD+48Hr 100nM DOC] will be referred to as the combination treatment.
217
LNCaP 10µM PXD/DOC Combined Treatment
% Cell Growth vs Control
100
Control
80
10µM PXD
60
100nM DOC
40
10µM PXD + 100nM DOC
20
0
24 Hours
48 Hours
48Hr 10µM PXD+24Hr
100nM DOC
48Hr 100nM DOC+24Hr
10µM PXD
Treatment
DU145 10µM PXD/DOC Combined Treatment
% Cell Growth vs Control
100
Control
80
10µM PXD
60
100nM DOC
40
10µM PXD + 100nM DOC
20
0
24 Hours
48 Hours
48Hr 10µM PXD+24Hr
100nM DOC
48Hr 100nM DOC+24Hr
10µM PXD
Treatment
PC3 10µM PXD/DOC Combined Treatment
% Cell Growth vs Control
100
Control
80
10µM PXD
60
100nM DOC
40
10µM PXD + 100nM DOC
20
0
24 Hours
48 Hours
Treatment
48Hr 10µM PXD+24Hr
100nM DOC
48Hr 100nM DOC+24Hr
10µM PXD
FIGURE 55 10µM PHENOXODIOL 100NM DOCETAXEL COMBINATION THERAPY MEASURED AFTER 48
HOURS OF TREATMENT.
218
Figure 56 Indicates a combination therapy treatment of 30µM phenoxodiol and 100nM
docetaxel in prostate cancer cells, over 24 and 48 hours, with two samples receiving 24 hours
solo drug treatment then the next 24 hours of combined treatment. Phenoxodiol has exhibited
the ability to induce G1/S phase cell cycle arrest, and docetaxel’s published method of action is
to inhibit the cell cycle at the M phase, resulting in G2 phase arrest (Michaud et al. 2000;
Aguero et al. 2010). LNCaP cells did not exhibit a significant change in proliferation versus
control over 24 hours. LNCaP cells exhibited a significant decrease in proliferation versus
control over 48 hours, when 30µM PXD and 48Hr PXD treatments were used (p<0.05,
p<0.033). The 100nM DOC (p=0.012), combination (p=0.0092) and 48Hr DOC (p=0.0090) were
all significantly decreased in proliferation versus 48Hr PXD, as well as 30µM PXD and Control
treatments..
DU145 cells exhibited a significant decrease in cell proliferation versus control in 30µM PXD
(p<0.0034), 100nM DOC (p<0.001) and combination (p<0.001), with both 100nM DOC
(p<0.001) and combination (p<0.005) therapy exhibiting further decreased proliferation versus
30µM PXD. A similar trend occurred in DU145 cells as in LNCaP, with 30µM PXD (p<0.001) and
48Hr PXD (p<0.001) significantly decreased in proliferation versus control but exhibiting no
significant difference between each other, while 100nM DOC, combination and 48Hr DOC
treatments were all significantly decreased versus 48Hr PXD (P<0.001 for all) as well as 30µM
PXD and Control.
Finally, PC3 cells exhibited a significant decrease in cell proliferation versus DMSO vehicle
control over 24 hours for 100nM DOC (p<0.001) and combination therapy (p<0.017), with
100nM DOC also being significantly decreased versus 30µM PXD (p<0.0021) and combined
treatment (p<0.015). Over 48 hours the trend displayed by LNCaP and DU145 continued, PC3
219
cells were significantly decreased in proliferation over time versus control in the 30µM PXD
(p<0.001) and 48Hr PXD (p=0.001) treatments, while 100nM DOC (p=0.0015), combination
(p=0.011) and 48Hr DOC (p=0.0015) were all significantly decreased versus 48Hr PXD as well as
30µM PXD and Control. This data suggests that, in all three cell lines, pre-treatment with
phenoxodiol for 24 hours before the addition of docetaxel to make a combination with
phenoxodiol does not result in a decrease in cell proliferation versus treatment with 30µM
phenoxodiol by itself. These data also suggest that combined treatment with phenoxodiol does
not increase the activity of docetaxel treatment in comparison to 100nM only docetaxel
treatment over 48 hours.
* indicates significance relative to control for that time point. ** indicates significance relative
to control and * groups for that time point.
The [48Hr 30µM PXD+24Hr 100nM DOC] and [48Hr 100nM DOC+24Hr 30µM PXD] treatments
will be referred to as the 48hr PXD and 48Hr DOC treatments respectively while the [48Hr
30µM PXD+48Hr 100nM DOC] will be referred to as the combination treatment.
220
LNCaP 30µM PXD/DOC Combined Treatment
% Cell Growth vs Control
100
Control
80
30µM PXD
60
100nM DOC
40
30µM PXD + 100nM DOC
20
0
24 Hours
48 Hours
48Hr 30µM PXD+24Hr
100nM DOC
48Hr 100nM DOC+24Hr
30µM PXD
Treatment
DU145 30µM PXD/DOC Combined Treatment
% Cell Growth vs Control
100
Control
80
30µM PXD
60
100nM DOC
40
30µM PXD + 100nM DOC
20
0
24 Hours
48 Hours
48Hr 30µM PXD+24Hr
100nM DOC
48Hr 100nM DOC+24Hr
30µM PXD
Treatment
PC3 30µM PXD/DOC Combined Treatment
% Cell Growth vs Control
100
Control
80
30µM PXD
60
100nM DOC
40
30µM PXD + 100nM DOC
20
0
24 Hours
48 Hours
Treatment
48Hr 30µM PXD+24Hr
100nM DOC
48Hr 100nM DOC+24Hr
30µM PXD
FIGURE 56 30µM PHENOXODIOL 100NM DOCETAXEL COMBINATION THERAPY MEASURED AFTER 48
HOURS OF TREATMENT.
221
After determining the effects of pre-treatment with another known cell cycle inhibitor,
Docetaxel, it was decided to test the effect of decreasing Wnt regulatory pathway signalling by
addition of sFRP4 whole protein in cell culture in conjunction with phenoxodiol treatment. This
was to determine the effects of sFRP4 on prostate cancer cells lines and detect any; synergy,
additivity, interference or lack of effect when sFRP4 was used in conjunction with phenoxodiol.
Figure 57 Indicates the effect of the combination of 500pg/mL of secreted frizzled related
protein 4 (sFRP4) and 30µM phenoxodiol treatment, over 48 hours, on prostate cancer cell
lines. Due to the large amount of statistics generated comparing the samples, a general
(p<0.05) will be used to indicate significant differences and only specific differences noted.
LNCaP cells treated with 125pg/mL and 250pg/mL sFRP4 were found to exhibit significantly
reduced proliferation versus control (p<0.05) over 48 hours. LNCaP cells treated with
500pg/mL sFRP4 and 10µM PXD treatments were not determined to be significantly different
to each other but were both significantly reduced in comparison to control, 125pg/mL and
250pg/mL sFRP4 treatments (p<0.05). LNCaP cells treated with 30µM phenoxodiol exhibited
significantly reduced proliferation against control, all sFRP4 concentrations and 10µM
phenoxodiol treatment (p<0.05); and, finally, the combination of 30µM PXD and 500pg/mL
sFRP4 proved to significantly reduce proliferation in comparison to control and all other
treatments (p<0.05), with a large improvement over 30µM phenoxodiol only.
DU145 cells exhibited a significant decrease in proliferation versus control with the 250pg/mL
sFRP4 treatment group (p<0.05), while neither 125pg/mL nor 500pg/mL sFRP4 were
determined to exhibit any significant changes versus control over 48 hours. DU145 cells
treated with either 10µM and 30µM phenoxodiol treatments were significantly decreased in
proliferation in comparison to control and all sFRP4 concentrations (p<0.05) but not versus
222
each other. Finally, the DU145 cells treated with combined sFRP4 and phenoxodiol exhibited a
significant decrease in proliferation versus control and the other treatment groups (p<0.05).
PC3 cells exhibited a significant decrease in cell proliferation versus control, over 48 hours,
with the 10µM phenoxodiol treatment (p<0.05). None of the sFRP4 treatments were
significantly different versus control in the chemoresistant late stage PC3 prostate cancer cells.
PC3 cells treated with 30µM phenoxodiol exhibited significantly reduced proliferation versus
control, all sFRP4 treatments and 10µM phenoxodiol (p<0.05); and finally, similar to the
previous two cell lines, PC3 cells treated with 30µM phenoxodiol and 500pg/mL sFRP4 were
significantly reduced in proliferation in comparison to control and all other treatments. This
study indicates that the combination of 30µM phenoxodiol and 500pg/mL sFRP4 results in a
significant decrease in cell proliferation across all three cell lines. This indicates that sFRP4 may
act as a sensitisation agent to the demonstrated cytotoxic effect of phenoxodiol.
* indicates significance relative to DMSO/PBS vehicle control for that time point.
** indicates significance relative to control and * groups for that time point.
*** indicates significance relative to control, * and ** groups
ϯ is significance relative to all other treatment groups.
223
120
LNCaP PXD/sFRP4 Combined Treatment
Control
% Cell Growth versus Control
100
125pg/ml sFRP4
80
250pg/ml sFRP4
60
500pg/ml sFRP4
40
10uM PXD
20
30uM PXD
0
Treatment
120
DU145 PXD/sFRP4 Combined Treatment
Control
100
% Cell Growth versus Control
30uM PXD +
500pg/ml sFRP4
125pg/ml sFRP4
80
250pg/ml sFRP4
60
500pg/ml sFRP4
40
10uM PXD
20
30uM PXD
0
Treatment
30uM PXD +
500pg/ml sFRP4
PC3 PXD/sFRP4 Combined Treatment
140
Control
% Cell Growth versus Control
120
125pg/ml sFRP4
100
250pg/ml sFRP4
80
60
500pg/ml sFRP4
40
10uM PXD
20
30uM PXD
0
Treatment
30uM PXD +
500pg/ml sFRP4
FIGURE 57 30µM PXD AND 500PG/ML SFRP4 PROTEIN COMBINATION THERAPY AFTER 48 HOURS.
224
7.5.
DISCUSSION
Castrate resistant prostate cancer is defined by disease progression despite androgendeprivation therapy and may present as one or any combination of a continuous rise in serum
levels of prostate-specific antigen (PSA), progression of pre-existing disease or the appearance
of new metastases. Current practices for hormone-refractory/castrate resistant, metastatic
prostate cancer incorporate the use of taxanes (Petrylak et al. 2004; Tannock et al. 2004). In
this study we investigated the effects of docetaxel on the prostate cancer cell lines LNCaP,
DU145 and PC3 to determine effective concentrations that would impact cell proliferation in a
measureable manner. We determined that low doses of docetaxel induced cell growth at .1
and 1nM concentrations. Low doses of cytotoxic drugs have been known to initiate oxidative
stress in cells, which acts as a mitogenic factor instead of inhibitory, as the cell easily
compensates for the low levels of stimulation, resulting in cell progression through the cell
cycle (Minelli et al. 2009). In all three cell lines 10nM and 100nM doses induced significant
decreases in cell proliferation and maximal decreases in proliferation respectively, suggesting
potential effective concentrations to utilise in combination with phenoxodiol.
Taxanes are used in the treatment of advanced prostate cancer and recurrent hormone
refractory and castrate resistant prostate cancer. Docetaxel, as with other taxanes, binds to
the β-tublin subunit in microtubulin, promotes polymerisation of tubulin and disrupts
microtubule dynamics. As a consequence of microtubulin stabilisation, cells become arrested
in G2/M phase and eventually undergo an apoptotic form of cell death. The effects of taxanes
may vary depending on cell type and drug concentration. These observations raised the
question as to what might be the intracellular signalling machinery that controls the apoptotic
differences in response to taxanes in prostate cancer. Studies have shown that apoptosis
induced by taxanes involves several apoptotic signal molecules, such as c-Jun N-terminal
225
kinase (JNK), protein kinase A (PKA), c-Raf-1/Ras/Bcl-2, p53/p21, and mitogen-activated
protein kinases ERK and p38 (Wang et al. 2008). Combined docetaxel and prednisone is
currently considered the standard of care for men with castrate resistant prostate cancer,
based largely on the simultaneous publication of Tannock and Petrylak (2004), where two large
randomised controlled trials compared this combination with the previously established
standard of mitoxantrone and prednisone. To determine an effective combination therapy
accurately, the interaction of two compounds is analysed and the effectiveness investigated
for synergistic, additive or even interference effects using an isobologram chart where a range
of effective doses is compared with each other (Zhao et al. 2004; Tallarida 2006).
In this study we investigated the effects of phenoxodiol combined with docetaxel treatment in
an isobologram format over 48 hours, with 24 different potential combinations per cell line.
We determined that the 5µM phenoxodiol concentration was inducing an interference effect
against the 100nM docetaxel concentration. This was the highest concentration of docetaxel
used and the concentration that exhibited the maximal effect of the compound after 48 hours
of treatment. All three cell lines had significant interference effects caused by phenoxodiol and
docetaxel combination at that concentration while the DU145 cell line also had the 10µM
phenoxodiol dose significantly impacting on the ability of docetaxel to induce cytotoxicity.
To further investigate these phenomena we evaluated pretreatment combinations in
conjunction with complete combinations over 48 hours. We investigated the effects of treating
the cells with vehicle control, 10µM phenoxodiol, 100nM docetaxel or combination 10µM
phenoxodiol/100nM docetaxel for 24 hours. Following this, the two extra sets of cells received
an addition of 10µM phenoxodiol; or 100nM docetaxel giving them a total of 48 hours of
100nM docetaxel and 24 hours of phenoxodiol or 48 hours 10µM phenoxodiol and 24 hours of
226
100nM docetaxel. This was also repeated for 30µM phenoxodiol concentrations to determine
if the increased cytotoxicity of phenoxodiol would overcome the limited interference that had
appeared in the isobolograms. The cell lines indicated that, at all phenoxodiol concentrations
after 48 hours of phenoxodiol and 24 hours of docetaxel treatment, there was no significant
different versus phenoxodiol by itself, other than PC3 10µM samples. This was followed by the
result that, at both 10µM and 30µM concentrations, there was a significant increase in
proliferation versus 100nM docetaxel alone, meaning that phenoxodiol had significantly
impacted the ability of docetaxel to induce cell arrest at the G2/M stage. While synergism and
additivity are the targets of drugs, especially those with different modes of action,
phenoxodiol’s exhibited ability to induce cell arrest at G1 and S phases of the cell cycle
prevents the cells from entering the G2/M region where docetaxel can be effective (Aguero et
al. 2010). Studies have also indicated that increased expression of p21WAF1 corresponds to an
inhibition of docetaxel activity via a p38 dependent signalling pathway and that cell cycle
inhibitors can actually protect the cells from taxane induced cell death during certain periods
(Canfield et al 2006; Gan et al 2011). A better understanding of how such mechanisms work at
the molecular level may have implications in the rational use of isoflavone/taxane based
chemotherapy.
The canonical Wnt/β-catenin signalling pathway promotes transcriptional activity through
activation of downstream target genes such as c-Myc and Cyclin-D1 driving cellular
proliferation (Menezes et al 2012; Watcharasit et al 2002). The Wnt ligand target, the frizzled
receptor, is overexpressed in prostate cancer but that this overexpression is counterbalanced
by the secreted frizzled related protein (sFRP) family which attempt to suppress AR-mediated
transactivation (Kawano et al. 2009). Methylation of the sFRP proteins has been suggested as a
marker of invasive carcinoma with a resulting poor prognosis (Ahmad et al. 2011) and
conversely, a total lack of β-catenin in a prostate cell has also been implicated in metastatic
227
formation (Aaltomaa et al. 2005). Unlike other inhibitors of Wnt signalling, sFRP4 appears to
affect androgen-dependent and androgen-independent prostate cancer (Horvath et al. 2007).
In this study we investigated the effects of purified sFRP4 protein in conjunction with
phenoxodiol, investigating the published abilities of isoflavones to interact with the Wnt
pathway (Li et al. 2008; Liss et al. 2010). Purified protein samples induced a decrease in cell
proliferation in LNCaP and D145 cells but not in PC3 cells. All cell lines exhibited a significantly
decreased proliferation rate when 500pg/mL purified protein was combined with 30µM
phenoxodiol over 48 hours of treatment, beyond that of 30µM phenoxodiol by itself. The
ability for sFRP4 to interact with frizzled results in a stabilisation of the GSK3β molecule, which
has been shown to be critical for soy based molecules, such as isoflavones, to induce cell
toxicity (Liss et al. 2010). This is one manner in which sFRP4 could be sensitising prostate
cancer cells to phenoxodiol treatment; others include the increased expression of p21WAF1 after
frizzled activation (Hall et al. 2010), GSK3β binding to and inactivating IAP members (Li et al.
2008), stabilisation of p53 expression (Watcharasit et al. 2002) or even downregulation of the
AR through GSK3β stabilisation. The Wnt pathway affects many parts of cell homeostasis and
the ability to prevent Wnt ligand activation of β-catenin and, therefore, stabilise GSK3β results
in a signalling situation in which phenoxodiol is more effective.
In this study we demonstrate that the cytotoxic taxane compound, docetaxel, induces cell
death that is attenuated by co-treatment or pre-treatment of cells with phenoxodiol. This
attenuation is associated with the prevention of cells from entering the G2/M phase of the cell
cycle where docetaxel is functional, damaging the spindle fibres and potentially due to p21WAF1
mediated cell survival after docetaxel treatment. Increased expression of p21WAF1, p53 and p38
has been determined to inhibit the ability of docetaxel to induce cytotoxicity (Gan et al. 2011).
228
We also exhibit the ability of sFRP4 protein to increase the effectiveness of phenoxodiol
treatment through alteration of the Wnt/β-catenin signalling pathway. Through stabilisation of
the GSK3β molecule, sFRP4 induces degradation of active β-catenin which causes an increased
sensitivity to isoflavone cytotoxic induction by increasing p21WAF1 expression and decreases
expression of c-Myc, Cyclin-D1 and other potent oncogenes. Phenoxodiol induces significant
cytotoxicity when combined with a Wnt/β-catenin receptor blocker such as sFRP4. This
promotes the concept that combination therapy of a Wnt inhibitor with phenoxodiol might
increase the effectiveness of phenoxodiol and give a subset population of prostate cancer
sufferers a more effective treatment regime.
229
8. CHAPTER EIGHT: GENERAL
DISCUSSION
8.1.
DISCUSSION
Phenoxodiol, [2H-1-Benzopyran-7-0,1,3-(4-hydroxyphenyl)], is an isoflavone derivative that has
also been shown to elicit cytotoxic effects against a broad range of human cancers. Currently
undergoing human clinical trials, it has shown promise in patients with recurrent ovarian
cancer where the cancer is refractory or resistant to standard chemotherapy, and in patients
with hormone-refractory prostate cancer (Sapi et al. 2004; Brown et al. 2005). Preliminary
studies involving a number of flavanoid derivatives have demonstrated that phenoxodiol
inhibits cell proliferation in a wide range of human cancer cell lines including leukaemia, breast
and prostate carcinomas, and is five to twenty times more potent than a similar compound,
Genistein (Aguero et al. 2005). Phenoxodiol has been characterised in ovarian cancer as
affecting key ovarian anti-apoptotic signalling pathways (Kamsteeg et al. 2003) as well as
reversing the ability of cells to become resistant to Docetaxel, through over expression of antiapoptotic molecules (Sapi et al. 2004). In breast cancer phenoxodiol acts as an inhibitor of cell
division (Constantinou et al. 2003), whilst in prostate cancer cells co-cultured with osteoblasts
phenoxodiol downregulated the cancer specific enzyme tNADH-Oxidase (Axanova et al. 2005).
Finally, it has been demonstrated that phenoxodiol induces G1 specific arrest through loss of
Cyclin-Dependant Kinase 2 activity by p53-independent induction of p21WAF1 in a battery of
human cell lines and that phenoxodiol affects multiple cancer types through prevention of cell
replication (Aguero et al. 2005; Aguero et al. 2010).
The ability to induce high rates of cytotoxicity in castrate resistant prostate cancer cells is the
goal of mainstream treatment regimes. Investigating the cytotoxicity of phenoxodiol on
230
prostate cancer cells is driven by the historically low 5-year survival rate for advanced prostate
cancer where chemotherapy is palliative in nature (Kessler and Albertsen 2003). In this study
we initially sought to characterise the effects of phenoxodiol on the prostate cancer cell lines
LNCaP (AR positive/p53 wild type), DU145 (AR negative p53-mutant) and PC3 (AR negative
p53-null). These cell lines represent early to late stage prostate cancers phenotypes and exhbit
altered expression of p53, a commonly mutated gene. We compared the action of phenoxodiol
against molecules regulating some of the key hallmarks of cancer stated by (Hanahan and
Weinberg 2000) and then updated by (Trosko et al. 2004). We targeted a framework of genes
that regulated the cell’s ability to be; self-sufficient in growth signals; insensitive to anti-growth
signals; able to evade apoptosis and able to have limitless replicative potential.
Determining appropriate initial cell concentration rates ensured logarithmic growth
throughout the study and ensured that phenoxodiol elicits time- and dose-dependent antiproliferative activity against both androgen-responsive and androgen-resistant prostate cancer
cell lines. LNCaP cells were found to be less sensitive to phenoxodiol induced cytotoxicity
measured by the MTS proliferation assay than visual analysis suggested. (Wang et al. 2010)
reported an underestimating of anti-proliferative effects of cytotoxic compounds due to MTS
dye in LNCaP cells when compared to other techniques. The use of apoptotic assays 3’-end
labelling, JC-1 and AV/PI flow cytometry ensured accurate cytotoxicity information was
obtained. LNCaP, DU145 and PC3 cells exhibited significant increases in mitochondrial
depolarisation, causing the release of Cytochrome c, Bax and other potent pro-apoptotic
molecules into the cytoplasm, where they could induce cell cytotoxicity. LNCaP cells had the
largest mitochondrial depolarisation response, appropriate for an early stage prostate cancer
model, while DU145 and PC3 cells had less total depolarisation, fitting their characterisation as
chemoresistant late stage prostate cancer cell representatives. The flow cytometry and 3’-end
labelling study determined that LNCaP and DU145 responded apoptotically to treatment while
231
PC3 cells responded necrotically. Cytotoxicity was determined to not be induced by pH
changes in the media.
Classic apoptosis, such as that induced by chemotherapeutic agents, can proceed via extrinsic
death receptor-mediated and intrinsic mitochondrial-mediated pathways ultimately resulting
in the activation of Caspase-3 (Asselin et al. 2001). Phenoxodiol has been shown to induce
classic apoptosis in ovarian cancer (Gamble et al. 2005; Alvero et al. 2006). However, this
current study did not detect activated Caspase-3 expression increase in LNCaP and PC3 cells,
while the actual increase in expression of Caspase-3 in DU145 cells was minimal. The lack of
consistent signalling indicated that cell cytotoxicity induced by phenoxodiol was potentially
initiated via a caspase independent cell death pathway. Phenoxodiol exhibited significant
cytotoxicity, inducing cell death in the prostate cancer cell lines LNCaP, DU145 and PC3 by
apoptotic and necrotic responses. All three cell lines demonstrated significantly decreased
viable cell populations after only 48 hours of treatment as determined by the AV/PI FACS
analysis technique, the JC-1 mitochondrial depolarisation assay indicating a method of
apoptotic induction via intrinsic cytotoxic signalling.
Two of the hallmarks of cancer initiation and progression are the ability of cells to become
insensitive to anti-growth signals and to evade programmed cell death, which is caused by
accumulated increased expression of anti-apoptotic signalling molecules that prevent
activation of the intrinsic and extrinsic signalling pathways that normally induce cell death
(Hanahan and Weinberg 2000). We reported that phenoxodiol does not induce cytotoxicity
through direct manipulation of the expression of key elements of the extrinsic and intrinsic cell
death pathways. Though phenoxodiol induced mitochondrial depolarisation and a significant
232
decrease in cell population, expression of the pro-apoptotic AIF, Bax, Caspase-3 and antiapoptotic Bcl-xL, xIAP were not consistently altered after phenoxodiol treatment.
A key aspect of chemotherapeutic treatments is the ability to induce cytotoxicity reliably over
a range of cells representing multiple stages of development, with a method of action that can
be consistently determined, as this maximises potential treatment candidates and successful
application. Though we detected apoptosis and mitochondrial depolarisation, we determined
that phenoxodiol acted in a caspase-independent manner. When phenoxodiol was combined
with a broad spectrum caspase inhibitor there was no effect on the ability of phenoxodiol to
induce cell death. In this study we determined that the cytotoxicity of phenoxodiol was not
associated with consistent changes in both pro-apoptotic and anti-apoptotic members of the
Bcl-2 family between the cell lines, unlike previous studies (Aguero et al. 2005; Yu et al. 2006).
LNCaP displayed a decreased expression of Bax protein and increased expression of Bcl-xL
protein, with cells attempting to prevent the effect of phenoxodiol-induced mitochondrial
depolarisation by mediating the effects of the Bcl-2 family. However, LNCaP cells still exhibited
significant cytotoxicity to treatment, indicating that the classic intrinsic apoptotic pathway
might not be directly responsible for the effect of phenoxodiol treatment. DU145 cells did not
express Bax protein at all, influencing the cell into a strong anti-apoptotic phenotype; yet a
detectable decrease in Bcl-xL protein was determined after treatment indicating a
susceptibility
to
phenoxodiol-induced
mitochondrial
depolarisation
with
resultant
mitochondrial efflux. This is in conjunction with PC3 cells, whose levels of Bax were unchanged
due to treatment but Bcl-xL levels were significantly reduced, indicating that the ratio of BclxL:Bax had significantly decreased; a sign of cell death signalling sensitivity from this
notoriously chemoresistant cell line (Li et al. 2008).
The lack of a consistent change in gene expression between LNCaP, DU145 and PC3 cell lines
indicates that phenoxodiol does not promote cell death by disrupting the stoichiometry of key
233
pro- and anti-apoptotic molecules studied here in prostate cancer cells. The common model
for cancer therapy activity has been that anti-cancer regimens cause apoptotic signalling to
occur, often mediated by p53 and Caspase-3 upregulation (Fojo and Bates 2003). Therefore,
cells that are resistant to apoptosis achieve this through forced over expression of antiapoptotic molecules such as xIAP and Bcl-xL family members or p53 mutation (Brown and
Attardi 2005). Members of the Inhibitors of Apoptosis Proteins (IAP), such as xIAP, can prevent
apoptosis by inactivating Caspase-3 and Caspase-9. The results reported in this study are in
contrast to other reports where it has been demonstrated in ovarian cancer that caspase
activation, Bax upregulation and xIAP degradation are key proteins modulated by phenoxodiol
(Sapi et al. 2004; Alvero et al. 2006; Mor et al. 2006; Kluger et al. 2007). The inability of the pan
caspase inhibitor, Z-VAD-FMK, to prevent the anti-proliferative activity of phenoxodiol in all
three cell lines reinforces that the phenoxodiol-induced mechanism of cell death in prostate
cancer cell lines is occurring via a caspase-independent pathway. This is in contrast to many
studies that indicate PARP and caspase activation as being the effect of phenoxodiol treatment
(Kamsteeg et al. 2003; Aguero et al. 2005; Aguero et al. 2010).
One potential method of cytotoxic induction investigated was the ability of phenoxodiol to
induce accumulation/production of reactive oxygen species (ROS) in the cell, causing oxidative
stress. Oxidative stress has been linked to mitochondrial depolarisation but we determined
that, in all cell lines, phenoxodiol did not induce nitric oxide production as a response to
treatment, even though all three cell lines retained the ability to have NO production
stimulated by DEAN and SNP addition.
In addition to apoptosis, cancer cells can be effectively eliminated through necrosis, mitotic
catastrophe and premature senescence, which result in cell cycle arrest and subsequent cell
234
death signalling (Brown and Wouters 1999; Tannock and Lee 2001; Brown and Wilson 2003).
Cells that are non-responsive to treatment through the classic intrinsic apoptosis pathways,
yet have an apoptotic response, must activate this process through other mechanisms such as
mitochondrial response and depolarisation, mitotic catastrophe and DNA degradation (Ruth
and Roninson 2000; Brown and Wilson 2003). Two of the hallmarks of cancer are the ability to
be self-sufficient in growth signals and to have infinite replicative potential, which can be
initiated through damaging the cell cycle restriction points, thereby allowing for progression of
cancer from a benign to metastatic state (Hanahan and Weinberg 2000). This study indicates
that phenoxodiol induces cell cycle arrest in the G1 and S phase of the cell cycle, with the
resultant arrest due to the upregulation of p21WAF1 and cytotoxicity due to resultant
downstream apoptotic signalling events. c-Myc and Cyclin-D1 expression was not consistently
altered in response to phenoxodiol treatment but Ki-67 was down regulated in LNCaP and PC3
cells and, though not significant, there was a similar trend in the DU145 cells.
The tumour suppressor p53 is the primary controller of cell cycle activity, which triggers cell
cycle arrest; but is only active in LNCaP cells with DU145 cells expressing a mutated p53 and
PC3 cells being p53-null. p21WAF1 is a cyclin dependant kinase inhibitor (CdkI) family member,
along with p27 and p57, which interfere with the cyclin dependant kinase-cyclin complex. The
ability of isoflavones, and specifically phenoxodiol, to induce cell cycle arrest has been
previously reported with studies indicating that arrest was induced by p21WAF1 stabilisation and
expression increase (Aguero et al. 2005; Aguero et al. 2010; Seo et al. 2011). In this study we
exhibited the ability of phenoxodiol to increase p21WAF1 expression, with the resultant cell
cycle arrest at both 10µM and 30µM concentrations over 24 and 48 hours of treatment. The
cell lines exhibited significantly decreased populations of G2 cells and arrest of the cell cycle,
visible in the significant alteration of G1 and S phase cell populations. It’s clear that the
235
method of phenoxodiol-induced cell cycle arrest is independent of p53 status and caspase
activity.
The canonical Wnt/β-catenin signalling pathway controls cellular interaction with growth
signals, where Wnt signalling causes decreased expression of GSK3β protein which results in an
active form of β-catenin translocating to the nucleus and initiating cell cycle progression
through multiple pathways including Cyclin-D1 and c-Myc expression (Zhang et al. 2011). We
investigated the expression of the active form of β-catenin and determined that LNCaP and
PC3 cells had significantly reduced active β-catenin expression after 24 hours, with LNCaP
continuing on to a significant 48 hour effect, which coincides with a subsequent decrease in
Cyclin-D1 signalling expression in androgen insensitive PC3 cells, while LNCaP expression is
unchanged due to the potential moderation of the AR receptor. DU145 cells exhibited an
increased expression of active β-catenin after treatment over 24 and 48 hours, indicating that
alteration of canonical Wnt signalling is not a direct effect of phenoxodiol treatment but a
downstream signalling outcome.
Secreted frizzled related protein 4 (sFRP4) is one of a group of proteins that antagonise the
canonical Wnt/β-catenin pathway, decreasing Wnt activity and causing decreased activated βcatenin formation and decreased invasive potential. We determined that sFRP4 expression
was increased over 24 hours in PC3 cells coinciding with decreased active β-catenin expression
in that cell line. sFRP4 expression increased over 24 and 48 hours in LNCaP cells, coinciding
with decreased active β-catenin in that cell line. sFRP4 expression was increased in DU145 cells
but did not correspond with β-catenin expression changes, indicating that DU145 cells might
have an modification of the canonical Wnt/β-catenin signalling pathway. Separate from βcatenin degradation it has been suggested that sFRP4 inhibition of Wnt signalling causes
236
stabilisation of GSK3β resulting in p53 activation and associated signalling such as p21WAF1, as
well as a decreased invasive potential in androgen-independent prostate cancer cells.
Addiction of purified sFRP4 protein resulted in a decrease in cell proliferation in LNCaP and
D145 cells but not in PC3 cells. All cell lines exhibited a significantly decreased proliferation
rate when 500pg/mL purified protein was combined with 30µM phenoxodiol over 48 hours of
treatment, beyond that of 30µM phenoxodiol by itself. The ability for purified sFRP4 to interact
with frizzled results in a potential stabilisation of the GSK3β molecule and p21WAF1 expression
alteration which (Liss et al. 2010) has shown to be critical for soy based molecules, like
isoflavones, to induce cell toxicity.
Docetaxel and paclitaxel are from a class of anti-cancer agents called taxanes that bind to and
stabilise microtubules, causing G2/M cell-cycle arrest and cell death (Montero et al. 2005). In
this study we investigated the effects of docetaxel on the prostate cancer cell lines LNCaP,
DU145 and PC3 to determine effective concentrations that would impact cell proliferation in a
measureable manner. In all three cell lines 10nM and 100nM doses induced significant
decreases in cell proliferation and maximal decreases in proliferation respectively, suggesting
potential effective concentrations to utilise in combination with phenoxodiol.
Taxanes are used in the treatment of advanced prostate cancer and recurrent, hormone
refractory and castrate-resistant prostate cancer. Docetaxel, as with other taxanes, binds to
the β-tublin subunit in microtubulin, promotes polymerisation of tublin and disrupts
microtubule dynamics. As a consequence of microtubulin stabilisation, cells become arrested
in G2/M phase and eventually undergo an apoptotic form of cell death. In this study we
investigated the effects of phenoxodiol combined with docetaxel treatment in an isobologram
format over 48 hours, with 24 different potential combinations per cell line. We determined
237
that the 5µM phenoxodiol concentration was inducing an interference effect against the
100nM docetaxel concentration. All three cell lines exhibited significant interference effects
caused by phenoxodiol and docetaxel combination at the specific combination, while DU145
also had the 10µM phenoxodiol dose impacting on the ability of docetaxel to induce
cytotoxicity.
Pretreatment drug combinations in conjunction with complete combinations over 48 hours
were then investigated. The cell lines indicated that, at all phenoxodiol concentrations after 48
hours of phenoxodiol and 24 hours of docetaxel treatment, there was no significant difference
in effect versus phenoxodiol by itself, other than PC3 10µM samples. This is followed by the
result that, at both 10 and 30µM concentrations, there was a significant increase in
proliferation versus 100nM docetaxel alone; indicating that phenoxodiol had significantly
impacted the ability of docetaxel to induce cell arrest at the G2/M stage. While synergism and
additivity are the targets of drugs, especially those with different modes of action,
phenoxodiol’s exhibited ability to induce arrest at G1 and S phases of the cell cycle prevents
the cells from entering the G2/M region, where docetaxel can be effective (Aguero et al.
2010). Studies have also indicated that increased expression of p21WAF1 corresponds to an
inhibition of docetaxel activity via a p38-dependent signalling pathway, and that cell cycle
inhibitors can actually protect the cells from taxane-induced cell death during certain periods
(Canfield et al. 2006; Gan et al. 2011). A better understanding of how such mechanisms work
at the molecular level may have implications in the rational use of isoflavone/taxane based
chemotherapy.
We have exhibited the ability of phenoxodiol to induce cytotoxicity and cell death through a
variety of means in a p53 and caspase-independent manner by arrest of cells in the G1 and S
238
phases of the cell cycle, thereby preventing cells from moving into G2/M phase. The cell
cytotoxicity was partly mediated by mitochondrial efflux, with all cells responding with a
degree of total efflux of factors into the cytoplasm in response to phenoxodiol treatment. The
cell cycle arrest was mediated by expression increases in p21WAF1 and the resultant cytotoxic
induction resulted in an apoptotic and necrotic response, independent of p53 status. We also
demonstrated that pre-treatment or co-treatment of prostate cancer cells with docetaxel and
phenoxodiol actually caused an interference response; and that co-treatment with sFRP4
purified protein and phenoxodiol exhibited a sensitisation effect, potentially caused by the
stabilisation of the GSK3β molecule and a subsequent increase in p21WAF1 signalling.
8.2.
CONCLUSION
We investigated the ability of Phenoxodiol to induce cytotoxicity in prostate cancer cell lines
representing early and late stage prostate cancer. We then determined the extent of induced
cytotoxicity by using multiple qualitative and quantitative measures to characterise the mode
of action. We can conclude that phenoxodiol induces significant cellular cytotoxicity over 24
and 48 hours of treatment with 10µM and 30µM doses of phenoxodiol. We can also state that
phenoxodiol induces apoptosis in LNCaP and DU145 cell lines while PC3 cell lines respond
necrotically, as evidenced by apoptotic DNA fragmentation in LNCaP and DU145 cells and DNA
smearing in PC3, similarly confirmed by AV/PI flow cytometry. We also determined that
Caspase-3 expression is not significantly up regulated in response to phenoxodiol treatment
except in DU145 cells; indicating that phenoxodiol’s method of action in prostate cancer cells
might be caspase independent.
239
We investigated the underlying apoptotic signalling machinery, after phenoxodiol treatment,
examining the link between early and late stage apoptotic signalling molecule expression. We
then determined that phenoxodiol had the ability to induce cytotoxicity through means other
than the direct innervation of intrinsic and extrinsic caspase signalling pathways. The analysis
of the anti-apoptotic Bcl-xL and xIAP molecules and pro-apoptotic AIF, Bax and Caspase-3
molecules did not show evidence of consistent signalling response as a result of phenoxodiol
treatment. Following this, a pan caspase inhibition study indicated that caspase expression was
not critical for phenoxodiol’s method of action; in contrast to previously published data and
important when many cancers have strong anti-apoptotic protein expression inhibiting
signalling through classic extrinsic and intrinsic apoptosis pathways.
We investigated the link between phenoxodiol and the cell cycle by determining the potential
effects on the various prostate cancer cell lines and looking for a common signalling pathway
in response to treatment. We also investigated the response of active β-catenin and sFRP4 to
phenoxodiol treatment after studies concluded evidence of a link between β-catenin
expression and cell cycle in prostate cancer. We determined that phenoxodiol prevented cell
cycle progression in a dose/time dependant manner through arrest in the G1 and S phase of
the cell cycle. This corresponded to an increase in P21WAF1 signalling in all cell lines. Alterations
in c-Myc, Cyclin-D1 and Ki-67 expression were detected, with only Ki-67 indicating a consistent
response. The expression of active β-catenin was down regulated in LNCaP and PC3 cells but
increased in DU145, indicating that phenoxodiol was not directly targeting β-catenin
expression in the canonical Wnt signalling pathway. However, sFRP4 levels were altered in a
manner that was indicative of the activated β-catenin decreases demonstrated. Both LNCaP
and PC3 cells exhibited increased expression of sFRP4 and resultant decreased expression of
active β-catenin protein.
240
We investigated the effects of combination therapy of phenoxodiol and docetaxel looking at
the effective dose of docetaxel, an isobologram of dose/response curves for phenoxodiol and
docetaxel therapy and the use of each as a pre-treatment for sensitisation. We also
investigated the effects of sFRP4 treatment in conjunction with phenoxodiol, targeting the cell
cycle through multiple means of interaction. We determined the docetaxel effective treatment
dose and, at that concentration, phenoxodiol treatment inhibits docetaxel at certain
combinations and concentrations by preventing cells from entering the G2/M phase of the cell
cycle where docetaxel exhibits its cell cycle arrest response. While high rates of cytotoxicity
were evident, we determined that combination therapy of docetaxel and phenoxodiol can
result in an interference reaction as expected. Phenoxodiol is effective at preventing cells from
progressing through G1/S phase, where docetaxel is effective, thereby preventing its ability to
induce G2 cell cycle arrest. Combination therapy with sFRP4 indicated a sensitisation to
phenoxodiol treatment, which was not necessarily synergistic, because sFRP4 didn’t exert a
consistent cytotoxic effect over 48 hours. However, interference with the canonical Wnt/βcatenin pathway assisted the action of phenoxodiol. Therefore, Wnt/β-catenin antagonism
results in stabilisation of GSK3β which assists in the expression of p21WAF1.
In this thesis we provide evidence that the soy based isoflavone, phenoxodiol has the ability to
potentiate cell cytotoxicity through a variety of means including canonical Wnt/β-catenin
signalling interference, Cdk inhibitor p21WAF1 activation, cell cycle arrest independent of p53
status and caspase independent induction of apoptosis. We exhibited the ability of
phenoxodiol to induce significant cytotoxicity in cells, representing early and late stage
prostate cancer, over 24 and 48 hours of treatment with 10µM and 30µM concentrations.
Phenoxodiol activity was determined to be increased when applied in combination therapy
with a Wnt antagonist, sFRP4. This implicates sensitisation of tumours towards phenoxodiolinduced cytotoxicity through antagonism of the Wnt receptor frizzled. Coupled with the
241
reported ability of high tolerance of orally ingested phenoxodiol, and few reported side
effects, Phenoxodiol represents a strong, effective, potential treatment for all stages of
prostate cancer including the currently incurable castrate-resistant late stage prostate cancer.
8.3.
LIMITATIONS
My perceived limitations of the study are as follows. Several grams only of phenoxodiol was
provided once for the entirety of the thesis experimental use. As such the IC50 concentration
was not determined, merely a response to treatment was initially sought and the use of 10µM
and 30µM concentrations was suggested by Novogen Pty Ltd as doses reaching biological
efficacy. The doses were kept at these concentrations due to the 2 week lifespan of
phenoxodiol once diluted in DMSO and higher concentrations would result in a rapid depletion
of the compound. Docetaxel also suffered from a short lifespan once diluted but was available
for purchase once supplies were diminished whereas phenoxodiol was not.
My cell cycle data was completed and provided to Novogen 18 months prior to the publishing
of the 2010 Aguero article which contains data very similar to what I found. Hence I view this
thesis to be a complete work of, at the time of experimental completion, unknown responses
to phenoxodiol as I was not informed about any data produced by other research groups being
provided with phenoxodiol. With the published article of 2010 indicating that PXD affected cell
cycle and P21WAF1 status in prostate cancer cells, we can independently confirm their findings
while also noting that we determined significant increases in S phase and G1 phase arrest.
In vivo experimentation in Nude mice was explored but determined to be difficult to complete
effectively with all three cell lines represented. The use of a 3D ultrasound capable of mapping
242
tumour volume and blood flow was not available within the state and getting access was cost
prohibitive. Animal ethics raised concerns over the use of multiple anaesthesia to measure
tumour growth and regression and thought a larger n value would be more appropriate but
lacking in the ability to track tumour changes over a course of doses.
Silencing work would have been an effective method to deduce the importance of p21WAF1
expression to phenoxodiol activity however the reported p27 and p16 pathway upregulation
and unknown conservation status of p21 WAF1 in various forms of cancer has since indicated
that p21 is not the sole target of phenoxodiol activity and silencing it would not necessarily
have resulted in an inability for phenoxodiol to induce cytotoxicity through cell cycle arrest.
The decreased sensitivity of LNCaP to the MTS proliferation assay, as described by (Wang et al.
2010), was problematic and prevented it as a sole measure of effective proliferation decrease
between cell lines, but still allowed for comparisons within the treatments of the LNCaP cell
line. The MTS assay was primarily utilised for its economy of scale that allowed the 3 cell lines
and multiple treatments to be effectively treated, stained and analysed within short periods of
each other. Other proliferation analysis methods were investigated but tritiated thymidine was
cost prohibitive and radioactive, XTT/MTT assays ran into exactly the same problem and were
more expensive, trypan blue viability staining would be excessively time intensive for the
purpose and the volume of cells/treatments being analysed in one point would have been
decreased. Cell density analysis was limited by access to the equipment which was granted and
then permission was retracted after 2 months of experiments meaning I couldn’t compare or
utilise the data generated by that technique.
243
9. BIBLIOGRAPHY
Aaltomaa, S., V. Karja, P. Lipponen, T. Isotalo, J. P. Kankkunen, M. Talja and R. Mokka (2005).
"Reduced alpha- and beta-catenin expression predicts shortened survival in local
prostate cancer." Anticancer Res 25(6C): 4707-4712.
Aguero, M. F., M. M. Facchinetti, Z. Sheleg and A. M. Senderowicz (2005). "Phenoxodiol, a
novel isoflavone, induces G1 arrest by specific loss in cyclin-dependent kinase 2 activity
by p53-independent induction of p21WAF1/CIP1." Cancer Res 65(8): 3364-3373.
Aguero, M. F., M. Venero, D. M. Brown, M. E. Smulson and L. A. Espinoza (2010). "Phenoxodiol
inhibits growth of metastatic prostate cancer cells." Prostate 70(11): 1211-1221.
Ahmad, I., R. Patel, Y. Liu, L. B. Singh, M. M. Taketo, X. R. Wu, H. Y. Leung and O. J. Sansom
(2011). "Ras mutation cooperates with beta-catenin activation to drive bladder
tumourigenesis." Cell Death Dis 2: e124.
AIHW, Ed. (2010). Cancer in Australia 2010: an overview. Cancer Series. Canberra, AIHW.
Alhasan, S. A., H. Pietrasczkiwicz, M. D. Alonso, J. Ensley and F. H. Sarkar (1999). "Genisteininduced cell cycle arrest and apoptosis in a head and neck squamous cell carcinoma
cell line." Nutr Cancer 34(1): 12-19.
Alimirah, F., J. Chen, Z. Basrawala, H. Xin and D. Choubey (2006). "DU-145 and PC-3 human
prostate cancer cell lines express androgen receptor: implications for the androgen
receptor functions and regulation." FEBS Lett 580(9): 2294-2300.
Alvero, A. B., D. Brown, M. Montagna, M. Matthews and G. Mor (2007). "PhenoxodiolTopotecan Co-Administration Exhibit Significant Anti-Tumor Activity Without Major
Adverse Side Effects." Cancer Biol Ther 6(4).
Alvero, A. B., M. Kelly, P. Rossi, A. Leiser, D. Brown, T. Rutherford and G. Mor (2008). "Antitumor activity of phenoxodiol: from bench to clinic." Future Oncol 4(4): 475-482.
Alvero, A. B., D. O'Malley, D. Brown, G. Kelly, M. Garg, W. Chen, T. Rutherford and G. Mor
(2006). "Molecular mechanism of phenoxodiol-induced apoptosis in ovarian carcinoma
cells." Cancer 106(3): 599-608.
Amantana, A., C. A. London, P. L. Iversen and G. R. Devi (2004). "X-linked inhibitor of apoptosis
protein inhibition induces apoptosis and enhances chemotherapy sensitivity in human
prostate cancer cells." Mol Cancer Ther 3(6): 699-707.
Arnold, J. T. and J. T. Isaacs (2002). "Mechanisms involved in the progression of androgenindependent prostate cancers: it is not only the cancer cell's fault." Endocr Relat
Cancer 9(1): 61-73.
Asselin, E., G. B. Mills and B. K. Tsang (2001). "XIAP regulates Akt activity and caspase-3dependent cleavage during cisplatin-induced apoptosis in human ovarian epithelial
cancer cells." Cancer Res 61(5): 1862-1868.
Avraamides, C. J., B. Garmy-Susini and J. A. Varner (2008). "Integrins in angiogenesis and
lymphangiogenesis." Nat Rev Cancer 8(8): 604-617.
244
Axanova, L., D. J. Morre and D. M. Morre (2005). "Growth of LNCaP cells in monoculture and
coculture with osteoblasts and response to tNOX inhibitors." Cancer Lett 225(1): 3540.
Baldin, V., J. Lukas, M. J. Marcote, M. Pagano and G. Draetta (1993). "Cyclin D1 is a nuclear
protein required for cell cycle progression in G1." Genes Dev 7(5): 812-821.
Basmajian, J. and C. Slonecker (1989). Grant's Method of Anatomy: A clinical Problem-Solving
Approach. Baltimore, Hong Kong, London, Sydney, Williams & Wilkins.
Beecher, G. R. (2003). "Overview of dietary flavonoids: nomenclature, occurrence and intake."
J Nutr 133(10): 3248S-3254S.
Begg, S., . Vos, T,. Barker, B,. Stevenson, C,. Stanley, L,. Lopez, AD,. Ed. (2007). The burden of
disease and injury in Australia 2003. Canberra, AIHW.
Berezovskaya, O., A. D. Schimmer, A. B. Glinskii, C. Pinilla, R. M. Hoffman, J. C. Reed and G. V.
Glinsky (2005). "Increased expression of apoptosis inhibitor protein XIAP contributes
to anoikis resistance of circulating human prostate cancer metastasis precursor cells."
Cancer Res 65(6): 2378-2386.
Berry, W. and M. Eisenberger (2005). "Achieving treatment goals for hormone-refractory
prostate cancer with chemotherapy." Oncologist 10 Suppl 3: 30-39.
Birkey Reffey, S., J. U. Wurthner, W. T. Parks, A. B. Roberts and C. S. Duckett (2001). "X-linked
inhibitor of apoptosis protein functions as a cofactor in transforming growth factorbeta signaling." J Biol Chem 276(28): 26542-26549.
Boatright, K. M. and G. S. Salvesen (2003). "Mechanisms of caspase activation." Curr Opin Cell
Biol 15(6): 725-731.
Bott, S. R., M. Arya, R. S. Kirby and M. Williamson (2005). "p21WAF1/CIP1 gene is inactivated
in metastatic prostatic cancer cell lines by promoter methylation." Prostate Cancer
Prostatic Dis 8(4): 321-326.
Brown, D. M., G. E. Kelly and A. J. Husband (2005). "Flavonoid compounds in maintenance of
prostate health and prevention and treatment of cancer." Mol Biotechnol 30(3): 253270.
Brown, J. M. and L. D. Attardi (2005). "The role of apoptosis in cancer development and
treatment response." Nat Rev Cancer 5(3): 231-237.
Brown, J. M. and G. Wilson (2003). "Apoptosis genes and resistance to cancer therapy: what
does the experimental and clinical data tell us?" Cancer Biol Ther 2(5): 477-490.
Brown, J. M. and B. G. Wouters (1999). "Apoptosis, p53, and tumor cell sensitivity to
anticancer agents." Cancer Res 59(7): 1391-1399.
Brusselmans, K., R. Vrolix, G. Verhoeven and J. V. Swinnen (2005). "Induction of cancer cell
apoptosis by flavonoids is associated with their ability to inhibit fatty acid synthase
activity." J Biol Chem 280(7): 5636-5645.
Canfield, S. E., K. Zhu, S. A. Williams and D. J. McConkey (2006). "Bortezomib inhibits
docetaxel-induced apoptosis via a p21-dependent mechanism in human prostate
cancer cells." Mol Cancer Ther 5(8): 2043-2050.
245
Carmeliet, P. and R. K. Jain (2011). "Principles and mechanisms of vessel normalization for
cancer and other angiogenic diseases." Nat Rev Drug Discov 10(6): 417-427.
Cassinelli, G., R. Supino, V. Zuco, C. Lanzi, A. I. Scovassi, S. C. Semple and F. Zunino (2004).
"Role of c-myc protein in hormone refractory prostate carcinoma: cellular response to
paclitaxel." Biochem Pharmacol 68(5): 923-931.
Cheng, E. H., M. C. Wei, S. Weiler, R. A. Flavell, T. W. Mak, T. Lindsten and S. J. Korsmeyer
(2001). "BCL-2, BCL-X(L) sequester BH3 domain-only molecules preventing BAX- and
BAK-mediated mitochondrial apoptosis." Mol Cell 8(3): 705-711.
Chesire, D. R., C. M. Ewing, W. R. Gage and W. B. Isaacs (2002). "In vitro evidence for complex
modes of nuclear beta-catenin signaling during prostate growth and tumorigenesis."
Oncogene 21(17): 2679-2694.
Chomczynski, P. and N. Sacchi (1987). "Single-step method of RNA isolation by acid
guanidinium thiocyanate-phenol-chloroform extraction." Anal Biochem 162(1): 156159.
Choueiri, T. K., R. Wesolowski and T. M. Mekhail (2006). "Phenoxodiol: isoflavone analog with
antineoplastic activity." Curr Oncol Rep 8(2): 104-107.
Constantinou, A. I., R. Mehta and A. Husband (2003). "Phenoxodiol, a novel isoflavone
derivative, inhibits dimethylbenz[a]anthracene (DMBA)-induced mammary
carcinogenesis in female Sprague-Dawley rats." Eur J Cancer 39(7): 1012-1018.
Cooper, S. (2003). "Reappraisal of serum starvation, the restriction point, G0, and G1 phase
arrest points." FASEB J 17(3): 333-340.
Cory, S. and J. M. Adams (2002). "The Bcl2 family: regulators of the cellular life-or-death
switch." Nat Rev Cancer 2(9): 647-656.
Cregan, S. P., V. L. Dawson and R. S. Slack (2004). "Role of AIF in caspase-dependent and
caspase-independent cell death." Oncogene 23(16): 2785-2796.
D'Antonio, J. M., C. Ma, F. A. Monzon and B. R. Pflug (2008). "Longitudinal analysis of androgen
deprivation of prostate cancer cells identifies pathways to androgen independence."
Prostate 68(7): 698-714.
Datta, B. and R. Datta (1999). "Induction of apoptosis due to lowering the level of eukaryotic
initiation factor 2-associated protein, p67, from mammalian cells by antisense
approach." Exp Cell Res 246(2): 376-383.
Davies, R., Frydenberg, M., Tulloch, A., Kelly, G. (2005). Final Results of a PhaseIb/IIa Study of
Oral Phenoxodiol in Patients with Late-stage, Hormone-Refractory Prostate Cancer.
American Association for Cancer Research. Anaheim, CA.: 1.
De Luca, T., E. Bosneaga, D. M. Morre and D. J. Morre (2008). "Downstream targets of altered
sphingolipid metabolism in response to inhibition of ENOX2 by phenoxodiol."
Biofactors 34(3): 253-260.
De Luca, T., D. M. Morre, H. Zhao and D. J. Morre (2005). "NAD+/NADH and/or CoQ/CoQH2
ratios from plasma membrane electron transport may determine ceramide and
sphingosine-1-phosphate levels accompanying G1 arrest and apoptosis." Biofactors
25(1-4): 43-60.
246
de Souza, P. L., W. Liauw, M. Links, S. Pirabhahar, G. Kelly and L. G. Howes (2006). "Phase I and
pharmacokinetic study of weekly NV06 (Phenoxodiol), a novel isoflav-3-ene, in
patients with advanced cancer." Cancer Chemother Pharmacol 58(4): 427-433.
de Souza, P. L., P. J. Russell, J. H. Kearsley and L. G. Howes (2010). "Clinical pharmacology of
isoflavones and its relevance for potential prevention of prostate cancer." Nutr Rev
68(9): 542-555.
Deveraux, Q. L., E. Leo, H. R. Stennicke, K. Welsh, G. S. Salvesen and J. C. Reed (1999).
"Cleavage of human inhibitor of apoptosis protein XIAP results in fragments with
distinct specificities for caspases." Embo J 18(19): 5242-5251.
Deveraux, Q. L., R. Takahashi, G. S. Salvesen and J. C. Reed (1997). "X-linked IAP is a direct
inhibitor of cell-death proteases." Nature 388(6639): 300-304.
Devlin, H. L., P. C. Mack, R. A. Burich, P. H. Gumerlock, H. J. Kung, M. Mudryj and R. W. deVere
White (2008). "Impairment of the DNA repair and growth arrest pathways by p53R2
silencing enhances DNA damage-induced apoptosis in a p53-dependent manner in
prostate cancer cells." Mol Cancer Res 6(5): 808-818.
Dharmarajan, A. M., S. Hisheh, B. Singh, S. Parkinson, K. I. Tilly and J. L. Tilly (1999).
"Antioxidants mimic the ability of chorionic gonadotropin to suppress apoptosis in the
rabbit corpus luteum in vitro: a novel role for superoxide dismutase in regulating bax
expression." Endocrinology 140(6): 2555-2561.
Dohi, T., K. Okada, F. Xia, C. E. Wilford, T. Samuel, K. Welsh, H. Marusawa, H. Zou, R.
Armstrong, S. Matsuzawa, G. S. Salvesen, J. C. Reed and D. C. Altieri (2004). "An IAPIAP complex inhibits apoptosis." J Biol Chem 279(33): 34087-34090.
Donepudi, M. and M. G. Grutter (2002). "Structure and zymogen activation of caspases."
Biophys Chem 101-102: 145-153.
Drake, J., A. M. Shearwood, J. White, R. Friis, N. Zeps, A. Charles and A. Dharmarajan (2009).
"Expression of secreted frizzled-related protein 4 (SFRP4) in primary serous ovarian
tumours." Eur J Gynaecol Oncol 30(2): 133-141.
Drake, J. M., R. R. Friis and A. M. Dharmarajan (2003). "The role of sFRP4, a secreted frizzledrelated protein, in ovulation." Apoptosis 8(4): 389-397.
Dubrez-Daloz, L., A. Dupoux and J. Cartier (2008). "IAPs: more than just inhibitors of apoptosis
proteins." Cell Cycle 7(8): 1036-1046.
Dudderidge, T. J., S. R. McCracken, M. Loddo, T. R. Fanshawe, J. D. Kelly, D. E. Neal, H. Y. Leung,
G. H. Williams and K. Stoeber (2007). "Mitogenic growth signalling, DNA replication
licensing, and survival are linked in prostate cancer." Br J Cancer.
Edinger, A. L. and C. B. Thompson (2004). "Death by design: apoptosis, necrosis and
autophagy." Curr Opin Cell Biol 16(6): 663-669.
el-Deiry, W. S., T. Tokino, V. E. Velculescu, D. B. Levy, R. Parsons, J. M. Trent, D. Lin, W. E.
Mercer, K. W. Kinzler and B. Vogelstein (1993). "WAF1, a potential mediator of p53
tumor suppression." Cell 75(4): 817-825.
247
Farooqi, A. A., S. Mukhtar, A. M. Riaz, S. Waseem, S. Minhaj, B. A. Dilawar, B. A. Malik, A.
Nawaz and S. Bhatti (2011). "Wnt and SHH in prostate cancer: trouble mongers occupy
the TRAIL towards apoptosis." Cell Prolif 44(6): 508-515.
Feldman, B. J. and D. Feldman (2001). "The development of androgen-independent prostate
cancer." Nat Rev Cancer 1(1): 34-45.
Fojo, T. and S. Bates (2003). "Strategies for reversing drug resistance." Oncogene 22(47): 75127523.
Fox, S. and A. Dharmarajan (2006). "WNT signaling in malignant mesothelioma." Front Biosci
11: 2106-2112.
Fu, M., C. Wang, Z. Li, T. Sakamaki and R. G. Pestell (2004). "Minireview: Cyclin D1: normal and
abnormal functions." Endocrinology 145(12): 5439-5447.
Gamble, J. R., P. Xia, C. N. Hahn, J. J. Drew, C. J. Drogemuller, D. Brown and M. A. Vadas (2005).
"Phenoxodiol, an experimental anticancer drug, shows potent antiangiogenic
properties in addition to its antitumour effects." Int J Cancer.
Gan, L., J. Wang, H. Xu and X. Yang (2011). "Resistance to docetaxel-induced apoptosis in
prostate cancer cells by p38/p53/p21 signaling." Prostate.
Gao, J., J. T. Arnold and J. T. Isaacs (2001). "Conversion from a paracrine to an autocrine
mechanism of androgen-stimulated growth during malignant transformation of
prostatic epithelial cells." Cancer Res 61(13): 5038-5044.
Georgaki, S., M. Skopeliti, M. Tsiatas, K. A. Nicolaou, K. Ioannou, A. Husband, A. Bamias, M. A.
Dimopoulos, A. I. Constantinou and O. E. Tsitsilonis (2009). "Phenoxodiol, an
anticancer isoflavene, induces immunomodulatory effects in vitro and in vivo." J Cell
Mol Med.
Gleason, D. (1977). The Veteran's Administration Cooperative Urologic Research Group:
histologic grading and clinical staging of prostatic carcinoma. Urologic Pathology: The
Prostate. M. Tannenbaum. Philadelphia, Lea and Febiger: 171-198.
Gleave, M., H. Miyake and K. Chi (2005). "Beyond simple castration: targeting the molecular
basis of treatment resistance in advanced prostate cancer." Cancer Chemother
Pharmacol 56 Suppl 1: 47-57.
Goodman, S. B., K. Kugu, S. H. Chen, S. Preutthipan, K. I. Tilly, J. L. Tilly and A. M. Dharmarajan
(1998). "Estradiol-mediated suppression of apoptosis in the rabbit corpus luteum is
associated with a shift in expression of bcl-2 family members favoring cellular
survival." Biol Reprod 59(4): 820-827.
Gross, A., J. M. McDonnell and S. J. Korsmeyer (1999). "BCL-2 family members and the
mitochondria in apoptosis." Genes Dev 13(15): 1899-1911.
Hall, C. L., S. Kang, O. A. MacDougald and E. T. Keller (2006). "Role of Wnts in prostate cancer
bone metastases." J Cell Biochem 97(4): 661-672.
Hall, C. L., H. Zhang, S. Baile, M. Ljungman, S. Kuhstoss and E. T. Keller (2010). "p21CIP-1/WAF1 induction is required to inhibit prostate cancer growth elicited by deficient
expression of the Wnt inhibitor Dickkopf-1." Cancer Res 70(23): 9916-9926.
248
Hamzah, J., M. Jugold, F. Kiessling, P. Rigby, M. Manzur, H. H. Marti, T. Rabie, S. Kaden, H. J.
Grone, G. J. Hammerling, B. Arnold and R. Ganss (2008). "Vascular normalization in
Rgs5-deficient tumours promotes immune destruction." Nature 453(7193): 410-414.
Hanahan, D. and R. A. Weinberg (2000). "The hallmarks of cancer." Cell 100(1): 57-70.
Harris, C. C. (1996). "Structure and function of the p53 tumor suppressor gene: clues for
rational cancer therapeutic strategies." J Natl Cancer Inst 88(20): 1442-1455.
Hayflick, L. and P. S. Moorhead (1961). "The serial cultivation of human diploid cell strains."
Exp Cell Res 25: 585-621.
He, T. C., A. B. Sparks, C. Rago, H. Hermeking, L. Zawel, L. T. da Costa, P. J. Morin, B. Vogelstein
and K. W. Kinzler (1998). "Identification of c-MYC as a target of the APC pathway."
Science 281(5382): 1509-1512.
Hedlund, T. E., W. U. Johannes and G. J. Miller (2003). "Soy isoflavonoid equol modulates the
growth of benign and malignant prostatic epithelial cells in vitro." Prostate 54(1): 6878.
Hempstock, J., J. P. Kavanagh and N. J. George (1998). "Growth inhibition of prostate cell lines
in vitro by phyto-oestrogens." Br J Urol 82(4): 560-563.
Hengartner, M. O. (2000). "The biochemistry of apoptosis." Nature 407(6805): 770-776.
Herst, P. M., J. E. Davis, P. Neeson, M. V. Berridge and D. S. Ritchie (2009). "The anti-cancer
drug, phenoxodiol, kills primary myeloid and lymphoid leukemic blasts and rapidly
proliferating T cells." Haematologica 94(7): 928-934.
Herst, P. M., T. Petersen, P. Jerram, J. Baty and M. V. Berridge (2007). "The antiproliferative
effects of phenoxodiol are associated with inhibition of plasma membrane electron
transport in tumour cell lines and primary immune cells." Biochem Pharmacol 74(11):
1587-1595.
Hirata, H., Y. Hinoda, K. Nakajima, K. Kawamoto, N. Kikuno, K. Ueno, S. Yamamura, M. S.
Zaman, G. Khatri, Y. Chen, S. Saini, S. Majid, G. Deng, N. Ishii and R. Dahiya (2011).
"Wnt antagonist DKK1 acts as a tumor suppressor gene that induces apoptosis and
inhibits proliferation in human renal cell carcinoma." Int J Cancer 128(8): 1793-1803.
Hong, S. J., T. M. Dawson and V. L. Dawson (2004). "Nuclear and mitochondrial conversations
in cell death: PARP-1 and AIF signaling." Trends Pharmacol Sci 25(5): 259-264.
Horoszewicz, J. S., S. S. Leong, T. M. Chu, Z. L. Wajsman, M. Friedman, L. Papsidero, U. Kim, L. S.
Chai, S. Kakati, S. K. Arya and A. A. Sandberg (1980). "The LNCaP cell line--a new model
for studies on human prostatic carcinoma." Prog Clin Biol Res 37: 115-132.
Horoszewicz, J. S., S. S. Leong, E. Kawinski, J. P. Karr, H. Rosenthal, T. M. Chu, E. A. Mirand and
G. P. Murphy (1983). "LNCaP model of human prostatic carcinoma." Cancer Res 43(4):
1809-1818.
Horvath, L. G., S. M. Henshall, J. G. Kench, D. N. Saunders, C. S. Lee, D. Golovsky, P. C. Brenner,
G. F. O'Neill, R. Kooner, P. D. Stricker, J. J. Grygiel and R. L. Sutherland (2004).
"Membranous expression of secreted frizzled-related protein 4 predicts for good
prognosis in localized prostate cancer and inhibits PC3 cellular proliferation in vitro."
Clin Cancer Res 10(2): 615-625.
249
Horvath, L. G., J. E. Lelliott, J. G. Kench, C. S. Lee, E. D. Williams, D. N. Saunders, J. J. Grygiel, R.
L. Sutherland and S. M. Henshall (2007). "Secreted frizzled-related protein 4 inhibits
proliferation and metastatic potential in prostate cancer." Prostate 67(10): 1081-1090.
Hotte, S. J. and F. Saad (2010). "Current management of castrate-resistant prostate cancer."
Curr Oncol 17 Suppl 2: S72-79.
Howes, J. B., P. L. de Souza, L. West, L. J. Huang and L. G. Howes (2011). "Pharmacokinetics of
phenoxodiol, a novel isoflavone, following intravenous administration to patients with
advanced cancer." BMC Clin Pharmacol 11: 1.
Hsing, A. W., L. Tsao and S. S. Devesa (2000). "International trends and patterns of prostate
cancer incidence and mortality." Int J Cancer 85(1): 60-67.
Hu, Y., G. Cherton-Horvat, V. Dragowska, S. Baird, R. G. Korneluk, J. P. Durkin, L. D. Mayer and
E. C. LaCasse (2003). "Antisense oligonucleotides targeting XIAP induce apoptosis and
enhance chemotherapeutic activity against human lung cancer cells in vitro and in
vivo." Clin Cancer Res 9(7): 2826-2836.
Huo, J. X., S. A. Metz and G. D. Li (2004). "p53-independent induction of p21(waf1/cip1)
contributes to the activation of caspases in GTP-depletion-induced apoptosis of
insulin-secreting cells." Cell Death Differ 11(1): 99-109.
Isaacs, J. T. and W. B. Isaacs (2004). "Androgen receptor outwits prostate cancer drugs." Nat
Med 10(1): 26-27.
Janicke, R. U., P. Ng, M. L. Sprengart and A. G. Porter (1998). "Caspase-3 is required for alphafodrin cleavage but dispensable for cleavage of other death substrates in apoptosis." J
Biol Chem 273(25): 15540-15545.
Janicke, R. U., M. L. Sprengart, M. R. Wati and A. G. Porter (1998). "Caspase-3 is required for
DNA fragmentation and morphological changes associated with apoptosis." J Biol
Chem 273(16): 9357-9360.
Jian, L. (2009). "Soy, isoflavones, and prostate cancer." Mol Nutr Food Res 53(2): 217-226.
Joannou, G. E., G. E. Kelly, A. Y. Reeder, M. Waring and C. Nelson (1995). "A urinary profile
study of dietary phytoestrogens. The identification and mode of metabolism of new
isoflavonoids." J Steroid Biochem Mol Biol 54(3-4): 167-184.
Johnstone, R. W., A. A. Ruefli and S. W. Lowe (2002). "Apoptosis: a link between cancer
genetics and chemotherapy." Cell 108(2): 153-164.
Juhaszova, M., D. B. Zorov, S. H. Kim, S. Pepe, Q. Fu, K. W. Fishbein, B. D. Ziman, S. Wang, K.
Ytrehus, C. L. Antos, E. N. Olson and S. J. Sollott (2004). "Glycogen synthase kinase3beta mediates convergence of protection signaling to inhibit the mitochondrial
permeability transition pore." J Clin Invest 113(11): 1535-1549.
Kaighn, M. E., K. S. Narayan, Y. Ohnuki, J. F. Lechner and L. W. Jones (1979). "Establishment
and characterization of a human prostatic carcinoma cell line (PC-3)." Invest Urol
17(1): 16-23.
Kamsteeg, M., T. Rutherford, E. Sapi, B. Hanczaruk, S. Shahabi, M. Flick, D. Brown and G. Mor
(2003). "Phenoxodiol--an isoflavone analog--induces apoptosis in chemoresistant
ovarian cancer cells." Oncogene 22(17): 2611-2620.
250
Kantoff, P. W., S. Halabi, M. Conaway, J. Picus, J. Kirshner, V. Hars, D. Trump, E. P. Winer and N.
J. Vogelzang (1999). "Hydrocortisone with or without mitoxantrone in men with
hormone-refractory prostate cancer: results of the cancer and leukemia group B 9182
study." J Clin Oncol 17(8): 2506-2513.
Kawano, Y., S. Diez, P. Uysal-Onganer, R. S. Darrington, J. Waxman and R. M. Kypta (2009).
"Secreted Frizzled-related protein-1 is a negative regulator of androgen receptor
activity in prostate cancer." Br J Cancer 100(7): 1165-1174.
Kelly, G. E. and A. J. Husband (2003). "Flavonoid compounds in the prevention and treatment
of prostate cancer." Methods Mol Med 81: 377-394.
Kerr, J. F., A. H. Wyllie and A. R. Currie (1972). "Apoptosis: a basic biological phenomenon with
wide-ranging implications in tissue kinetics." Br J Cancer 26(4): 239-257.
Kessler, B. and P. Albertsen (2003). "The natural history of prostate cancer." Urol Clin North
Am 30(2): 219-226.
King, T. J., L. H. Fukushima, T. A. Donlon, A. D. Hieber, K. A. Shimabukuro and J. S. Bertram
(2000). "Correlation between growth control, neoplastic potential and endogenous
connexin43 expression in HeLa cell lines: implications for tumor progression."
Carcinogenesis 21(2): 311-315.
Kleeberger, W., G. S. Bova, M. E. Nielsen, M. Herawi, A. Y. Chuang, J. I. Epstein and D. M.
Berman (2007). "Roles for the stem cell associated intermediate filament Nestin in
prostate cancer migration and metastasis." Cancer Res 67(19): 9199-9206.
Klein, R., D. Brown and A. M. Turnley (2007). "Phenoxodiol protects against Cisplatin induced
neurite toxicity in a PC-12 cell model." BMC Neurosci 8: 61.
Kluger, H. M., M. M. McCarthy, A. B. Alvero, M. Sznol, S. Ariyan, R. L. Camp, D. L. Rimm and G.
Mor (2007). "The X-linked inhibitor of apoptosis protein (XIAP) is up-regulated in
metastatic melanoma, and XIAP cleavage by Phenoxodiol is associated with
Carboplatin sensitization." J Transl Med 5: 6.
Koda, M., M. Sulkowska, L. Kanczuga-Koda, J. Tomaszewski, W. Kucharczuk, T. Lesniewicz, S.
Cymek and S. Sulkowski (2007). "The effect of chemotherapy on Ki-67, Bcl-2 and Bak
expression in primary tumors and lymph node metastases of breast cancer." Oncol Rep
18(1): 113-119.
Korsmeyer, S. J. (1992). "Bcl-2 initiates a new category of oncogenes: regulators of cell death."
Blood 80(4): 879-886.
Korsmeyer, S. J., M. C. Wei, M. Saito, S. Weiler, K. J. Oh and P. H. Schlesinger (2000). "Proapoptotic cascade activates BID, which oligomerizes BAK or BAX into pores that result
in the release of cytochrome c." Cell Death Differ 7(12): 1166-1173.
Krajewska, M., S. Krajewski, S. Banares, X. Huang, B. Turner, L. Bubendorf, O. P. Kallioniemi, A.
Shabaik, A. Vitiello, D. Peehl, G. J. Gao and J. C. Reed (2003). "Elevated expression of
inhibitor of apoptosis proteins in prostate cancer." Clin Cancer Res 9(13): 4914-4925.
Kuruvilla, S., C. W. Qualls, Jr., R. D. Tyler, S. M. Witherspoon, G. R. Benavides, L. W. Yoon, K.
Dold, R. H. Brown, S. Sangiah and K. T. Morgan (2003). "Effects of minimally toxic levels
of carbonyl cyanide P-(trifluoromethoxy) phenylhydrazone (FCCP), elucidated through
251
differential gene expression with biochemical and morphological correlations." Toxicol
Sci 73(2): 348-361.
Ladha, M. H., K. Y. Lee, T. M. Upton, M. F. Reed and M. E. Ewen (1998). "Regulation of exit
from quiescence by p27 and cyclin D1-CDK4." Mol Cell Biol 18(11): 6605-6615.
Langley, R. R. and I. J. Fidler (2011). "The seed and soil hypothesis revisited--the role of tumorstroma interactions in metastasis to different organs." Int J Cancer 128(11): 2527-2535.
Lee, D. H., M. Szczepanski and Y. J. Lee (2008). "Role of Bax in quercetin-induced apoptosis in
human prostate cancer cells." Biochem Pharmacol 75(12): 2345-2355.
Levine, A. J. (1997). "p53, the cellular gatekeeper for growth and division." Cell 88(3): 323-331.
Levkau, B., K. J. Garton, N. Ferri, K. Kloke, J. R. Nofer, H. A. Baba, E. W. Raines and G. Breithardt
(2001). "xIAP induces cell-cycle arrest and activates nuclear factor-kappaB : new
survival pathways disabled by caspase-mediated cleavage during apoptosis of human
endothelial cells." Circ Res 88(3): 282-290.
Li, H., X. Chen, T. Calhoun-Davis, K. Claypool and D. G. Tang (2008). "PC3 human prostate
carcinoma cell holoclones contain self-renewing tumor-initiating cells." Cancer Res
68(6): 1820-1825.
Li, J., M. Xing, M. Zhu, X. Wang, M. Wang, S. Zhou, N. Li, R. Wu and M. Zhou (2008). "Glycogen
synthase kinase 3beta induces apoptosis in cancer cells through increase of survivin
nuclear localization." Cancer Lett 272(1): 91-101.
Li, K., Y. Li, J. M. Shelton, J. A. Richardson, E. Spencer, Z. J. Chen, X. Wang and R. S. Williams
(2000). "Cytochrome c deficiency causes embryonic lethality and attenuates stressinduced apoptosis." Cell 101(4): 389-399.
Li, Y. and F. H. Sarkar (2002). "Gene expression profiles of genistein-treated PC3 prostate
cancer cells." J Nutr 132(12): 3623-3631.
Li, Y., Z. Wang, D. Kong, R. Li, S. H. Sarkar and F. H. Sarkar (2008). "Regulation of
Akt/FOXO3a/GSK-3beta/AR signaling network by isoflavone in prostate cancer cells." J
Biol Chem 283(41): 27707-27716.
Liang, H., Q. Chen, A. H. Coles, S. J. Anderson, G. Pihan, A. Bradley, R. Gerstein, R. Jurecic and S.
N. Jones (2003). "Wnt5a inhibits B cell proliferation and functions as a tumor
suppressor in hematopoietic tissue." Cancer Cell 4(5): 349-360.
Liss, M. A., M. Schlicht, A. Kahler, R. Fitzgerald, T. Thomassi, A. Degueme, M. Hessner and M.
W. Datta (2010). "Characterization of soy-based changes in Wnt-frizzled signaling in
prostate cancer." Cancer Genomics Proteomics 7(5): 245-252.
Litvinov, I. V., A. M. De Marzo and J. T. Isaacs (2003). "Is the Achilles' heel for prostate cancer
therapy a gain of function in androgen receptor signaling?" J Clin Endocrinol Metab
88(7): 2972-2982.
MacFarlane, M., W. Merrison, S. B. Bratton and G. M. Cohen (2002). "Proteasome-mediated
degradation of Smac during apoptosis: XIAP promotes Smac ubiquitination in vitro." J
Biol Chem 277(39): 36611-36616.
252
MacLachlan, T. K., N. Sang and A. Giordano (1995). "Cyclins, cyclin-dependent kinases and cdk
inhibitors: implications in cell cycle control and cancer." Crit Rev Eukaryot Gene Expr
5(2): 127-156.
Madani, S. H., S. Ameli, S. Khazaei, M. Kanani and B. Izadi (2011). "Frequency of Ki-67 (MIB-1)
and P53 expressions among patients with prostate cancer." Indian J Pathol Microbiol
54(4): 688-691.
Maguire, T., P. Harrison, O. Hyink, J. Kalmakoff and V. K. Ward (2000). "The inhibitors of
apoptosis of Epiphyas postvittana nucleopolyhedrovirus." J Gen Virol 81(Pt 11): 28032811.
Mahoney, S., Berry, C., Hisheh, S., Dharmarajan, A. (2007). "X-linked Inhibitor of Apoptosis
Protein Expression and Cisplatin Resistance in Ovarian Cancer." International Journal
of Integrative Biology 1(1): 8.
Marieb, E., . Hoehn, K. (2012). Human Anatomy and Physiology. Sydney, Pearson.
Martin, S. J., C. P. Reutelingsperger, A. J. McGahon, J. A. Rader, R. C. van Schie, D. M. LaFace
and D. R. Green (1995). "Early redistribution of plasma membrane phosphatidylserine
is a general feature of apoptosis regardless of the initiating stimulus: inhibition by
overexpression of Bcl-2 and Abl." J Exp Med 182(5): 1545-1556.
Mateyak, M. K., A. J. Obaya and J. M. Sedivy (1999). "c-Myc regulates cyclin D-Cdk4 and -Cdk6
activity but affects cell cycle progression at multiple independent points." Mol Cell Biol
19(7): 4672-4683.
McCarthy, P. J., S. F. Sweetman, P. G. McKenna and V. J. McKelvey-Martin (1997). "Evaluation
of manual and image analysis quantification of DNA damage in the alkaline comet
assay." Mutagenesis 12(4): 209-214.
McNeal, J. E., E. A. Redwine, F. S. Freiha and T. A. Stamey (1988). "Zonal distribution of
prostatic adenocarcinoma. Correlation with histologic pattern and direction of
spread." Am J Surg Pathol 12(12): 897-906.
McPherson, R. A., P. T. Galettis and P. L. de Souza (2009). "Enhancement of the activity of
phenoxodiol by cisplatin in prostate cancer cells." Br J Cancer 100(4): 649-655.
Menezes, M. E., D. J. Devine, L. A. Shevde and R. S. Samant (2012). "Dickkopf1: a tumor
suppressor or metastasis promoter?" Int J Cancer 130(7): 1477-1483.
Michaud, L. B., V. Valero and G. Hortobagyi (2000). "Risks and benefits of taxanes in breast and
ovarian cancer." Drug Saf 23(5): 401-428.
Middleton, E., Jr., C. Kandaswami and T. C. Theoharides (2000). "The effects of plant flavonoids
on mammalian cells: implications for inflammation, heart disease, and cancer."
Pharmacol Rev 52(4): 673-751.
Minelli, A., I. Bellezza, A. Tucci, M. G. Rambotti, C. Conte and Z. Culig (2009). "Differential
involvement of reactive oxygen species and nucleoside transporters in cytotoxicity
induced by two adenosine analogues in human prostate cancer cells." Prostate 69(5):
538-547.
253
Mitchell, J. H., S. J. Duthie and A. R. Collins (2000). "Effects of phytoestrogens on growth and
DNA integrity in human prostate tumor cell lines: PC-3 and LNCaP." Nutr Cancer 38(2):
223-228.
Miyamoto, H., S. Altuwaijri, Y. Cai, E. M. Messing and C. Chang (2005). "Inhibition of the Akt,
cyclooxygenase-2, and matrix metalloproteinase-9 pathways in combination with
androgen deprivation therapy: potential therapeutic approaches for prostate cancer."
Mol Carcinog 44(1): 1-10.
Miyamoto, H., E. M. Messing and C. Chang (2004). "Androgen deprivation therapy for prostate
cancer: current status and future prospects." Prostate 61(4): 332-353.
Montero, A., F. Fossella, G. Hortobagyi and V. Valero (2005). "Docetaxel for treatment of solid
tumours: a systematic review of clinical data." Lancet Oncol 6(4): 229-239.
Moore, K., A. Agur, A. Dalley, V. Oxorn and M. Moore (2007). Essential Clinical Anatomy.
Baltimore, Philidelphia, Lippincott Williams & Wilkins.
Mor, G., H. H. Fu and A. B. Alvero (2006). "Phenoxodiol, a novel approach for the treatment of
ovarian cancer." Curr Opin Investig Drugs 7(6): 542-548.
Morre, D. J., P. J. Chueh, K. Yagiz, A. Balicki, C. Kim and D. M. Morre (2007). "ECTO-NOX target
for the anticancer isoflavene phenoxodiol." Oncol Res 16(7): 299-312.
Morre, D. J., N. McClain, L. Y. Wu, G. Kelly and D. M. Morre (2009). "Phenoxodiol treatment
alters the subsequent response of ENOX2 (tNOX) and growth of hela cells to paclitaxel
and cisplatin." Mol Biotechnol 42(1): 100-109.
Muley, A., S. Majumder, G. K. Kolluru, S. Parkinson, H. Viola, L. Hool, F. Arfuso, R. Ganss, A.
Dharmarajan and S. Chatterjee (2010). "Secreted frizzled-related protein 4: an
angiogenesis inhibitor." Am J Pathol 176(3): 1505-1516.
Murphy, M. P. (2009). "How mitochondria produce reactive oxygen species." Biochem J 417(1):
1-13.
Nakagawa, T., H. Zhu, N. Morishima, E. Li, J. Xu, B. A. Yankner and J. Yuan (2000). "Caspase-12
mediates endoplasmic-reticulum-specific apoptosis and cytotoxicity by amyloid-beta."
Nature 403(6765): 98-103.
Niculescu, A. B., 3rd, X. Chen, M. Smeets, L. Hengst, C. Prives and S. I. Reed (1998). "Effects of
p21(Cip1/Waf1) at both the G1/S and the G2/M cell cycle transitions: pRb is a critical
determinant in blocking DNA replication and in preventing endoreduplication." Mol
Cell Biol 18(1): 629-643.
Pardee, A. B. (1974). "A restriction point for control of normal animal cell proliferation." Proc
Natl Acad Sci U S A 71(4): 1286-1290.
Parente, P., F. Parnis and H. Gurney (2012). "Emerging and second line therapies for the
management of metastatic castration-resistant prostate cancer: The Australian
perspective." Asia Pac J Clin Oncol 8(1): 31-42.
Park, Y. J., K. H. Bae, B. S. Jin, H. J. Jung and J. S. Park (2012). "Is increased prostatic urethral
angle related to lower urinary tract symptoms in males with benign prostatic
hyperplasia/lower urinary tract symptoms?" Korean J Urol 53(6): 410-413.
254
Parton, M., S. Krajewski, I. Smith, M. Krajewska, C. Archer, M. Naito, R. Ahern, J. Reed and M.
Dowsett (2002). "Coordinate expression of apoptosis-associated proteins in human
breast cancer before and during chemotherapy." Clin Cancer Res 8(7): 2100-2108.
Peng, Y., C. X. Li, F. Chen, Z. Wang, M. Ligr, J. Melamed, J. Wei, W. Gerald, M. Pagano, M. J.
Garabedian and P. Lee (2008). "Stimulation of prostate cancer cellular proliferation
and invasion by the androgen receptor co-activator ARA70." Am J Pathol 172(1): 225235.
Petronilli, V., D. Penzo, L. Scorrano, P. Bernardi and F. Di Lisa (2001). "The mitochondrial
permeability transition, release of cytochrome c and cell death. Correlation with the
duration of pore openings in situ." J Biol Chem 276(15): 12030-12034.
Petrylak, D. P., C. M. Tangen, M. H. Hussain, P. N. Lara, Jr., J. A. Jones, M. E. Taplin, P. A. Burch,
D. Berry, C. Moinpour, M. Kohli, M. C. Benson, E. J. Small, D. Raghavan and E. D.
Crawford (2004). "Docetaxel and estramustine compared with mitoxantrone and
prednisone for advanced refractory prostate cancer." N Engl J Med 351(15): 15131520.
Puthalakath, H. and A. Strasser (2002). "Keeping killers on a tight leash: transcriptional and
post-translational control of the pro-apoptotic activity of BH3-only proteins." Cell
Death Differ 9(5): 505-512.
Reed, J. C. (2000). "Mechanisms of apoptosis." Am J Pathol 157(5): 1415-1430.
Reya, T., S. J. Morrison, M. F. Clarke and I. L. Weissman (2001). "Stem cells, cancer, and cancer
stem cells." Nature 414(6859): 105-111.
Riedl, S. J. and G. S. Salvesen (2007). "The apoptosome: signalling platform of cell death." Nat
Rev Mol Cell Biol 8(5): 405-413.
Ririe, K. M., R. P. Rasmussen and C. T. Wittwer (1997). "Product differentiation by analysis of
DNA melting curves during the polymerase chain reaction." Anal Biochem 245(2): 154160.
Risbridger, G. P., H. Wang, M. Frydenberg and A. Husband (2001). "The in vivo effect of red
clover diet on ventral prostate growth in adult male mice." Reprod Fertil Dev 13(4):
325-329.
Roy, S., M. Kaur, C. Agarwal, M. Tecklenburg, R. A. Sclafani and R. Agarwal (2007). "p21 and
p27 induction by silibinin is essential for its cell cycle arrest effect in prostate
carcinoma cells." Mol Cancer Ther 6(10): 2696-2707.
Roy, S., R. P. Singh, C. Agarwal, S. Siriwardana, R. Sclafani and R. Agarwal (2008).
"Downregulation of both p21/Cip1 and p27/Kip1 produces a more aggressive prostate
cancer phenotype." Cell Cycle 7(12): 1828-1835.
Ruth, A. C. and I. B. Roninson (2000). "Effects of the multidrug transporter P-glycoprotein on
cellular responses to ionizing radiation." Cancer Res 60(10): 2576-2578.
Ryan, K. M., A. C. Phillips and K. H. Vousden (2001). "Regulation and function of the p53 tumor
suppressor protein." Curr Opin Cell Biol 13(3): 332-337.
255
Saif, M. W., E. Tytler, F. Lansigan, D. M. Brown and A. J. Husband (2009). "Flavonoids,
phenoxodiol, and a novel agent, triphendiol, for the treatment of pancreaticobiliary
cancers." Expert Opin Investig Drugs 18(4): 469-479.
Saladin, K. (2007). Anatomy and Physiology: The Unity of Form and Function. New York,
McGraw Hill.
Salvioli, S., A. Ardizzoni, C. Franceschi and A. Cossarizza (1997). "JC-1, but not DiOC6(3) or
rhodamine 123, is a reliable fluorescent probe to assess delta psi changes in intact
cells: implications for studies on mitochondrial functionality during apoptosis." FEBS
Lett 411(1): 77-82.
Sanna, M. G., J. da Silva Correia, Y. Luo, B. Chuang, L. M. Paulson, B. Nguyen, Q. L. Deveraux
and R. J. Ulevitch (2002). "ILPIP, a novel anti-apoptotic protein that enhances XIAPmediated activation of JNK1 and protection against apoptosis." J Biol Chem 277(34):
30454-30462.
Sapi, E., A. B. Alvero, W. Chen, D. O'Malley, X. Y. Hao, B. Dwipoyono, M. Garg, M. Kamsteeg, T.
Rutherford and G. Mor (2004). "Resistance of ovarian carcinoma cells to docetaxel is
XIAP dependent and reversible by phenoxodiol." Oncol Res 14(11-12): 567-578.
Scherr, D. S., E. D. Vaughan, Jr., J. Wei, M. Chung, D. Felsen, R. Allbright and B. S. Knudsen
(1999). "BCL-2 and p53 expression in clinically localized prostate cancer predicts
response to external beam radiotherapy." J Urol 162(1): 12-16; discussion 16-17.
Schmitt, E., A. Steyaert, G. Cimoli and R. Bertrand (1998). "Bax-alpha promotes apoptosis
induced by cancer chemotherapy and accelerates the activation of caspase 3-like
cysteine proteases in p53 double mutant B lymphoma Namalwa cells." Cell Death
Differ 5(6): 506-516.
Schwartzman, R. A. and J. A. Cidlowski (1993). "Apoptosis: the biochemistry and molecular
biology of programmed cell death." Endocr Rev 14(2): 133-151.
Senderowicz, A. M. (2004). "Targeting cell cycle and apoptosis for the treatment of human
malignancies." Curr Opin Cell Biol 16(6): 670-678.
Seo, Y. J., B. S. Kim, S. Y. Chun, Y. K. Park, K. S. Kang and T. G. Kwon (2011). "Apoptotic effects
of genistein, biochanin-A and apigenin on LNCaP and PC-3 cells by p21 through
transcriptional inhibition of polo-like kinase-1." J Korean Med Sci 26(11): 1489-1494.
Silasi, D. A., A. B. Alvero, T. J. Rutherford, D. Brown and G. Mor (2009). "Phenoxodiol:
pharmacology and clinical experience in cancer monotherapy and in combination with
chemotherapeutic drugs." Expert Opin Pharmacother 10(6): 1059-1067.
Silke, J., P. G. Ekert, C. L. Day, C. J. Hawkins, M. Baca, J. Chew, M. Pakusch, A. M. Verhagen and
D. L. Vaux (2001). "Direct inhibition of caspase 3 is dispensable for the anti-apoptotic
activity of XIAP." Embo J 20(12): 3114-3123.
Slee, E. A., C. Adrain and S. J. Martin (1999). "Serial killers: ordering caspase activation events
in apoptosis." Cell Death Differ 6(11): 1067-1074.
Slee, E. A., M. T. Harte, R. M. Kluck, B. B. Wolf, C. A. Casiano, D. D. Newmeyer, H. G. Wang, J. C.
Reed, D. W. Nicholson, E. S. Alnemri, D. R. Green and S. J. Martin (1999). "Ordering the
cytochrome c-initiated caspase cascade: hierarchical activation of caspases-2, -3, -6, -7,
-8, and -10 in a caspase-9-dependent manner." J Cell Biol 144(2): 281-292.
256
So, A. I., A. Hurtado-Coll and M. E. Gleave (2003). "Androgens and prostate cancer." World J
Urol 21(5): 325-337.
Srinivasula, S. M., R. Hegde, A. Saleh, P. Datta, E. Shiozaki, J. Chai, R. A. Lee, P. D. Robbins, T.
Fernandes-Alnemri, Y. Shi and E. S. Alnemri (2001). "A conserved XIAP-interaction
motif in caspase-9 and Smac/DIABLO regulates caspase activity and apoptosis." Nature
410(6824): 112-116.
Srivastava, R. K., Q. Chen, I. Siddiqui, K. Sarva and S. Shankar (2007). "Linkage of curcumininduced cell cycle arrest and apoptosis by cyclin-dependent kinase inhibitor
p21(/WAF1/CIP1)." Cell Cycle 6(23): 2953-2961.
Stevens, A., . Lowe, J,. Scott, I,. (2009). Core Pathology, Mosby Elsevier.
Stone, K. R., D. D. Mickey, H. Wunderli, G. H. Mickey and D. F. Paulson (1978). "Isolation of a
human prostate carcinoma cell line (DU 145)." Int J Cancer 21(3): 274-281.
Straszewski-Chavez, S. L., V. M. Abrahams, E. F. Funai and G. Mor (2004). "X-linked inhibitor of
apoptosis (XIAP) confers human trophoblast cell resistance to Fas-mediated
apoptosis." Mol Hum Reprod 10(1): 33-41.
Sun, A., J. Tang, Y. Hong, J. Song, P. F. Terranova, J. B. Thrasher, S. Svojanovsky, H. G. Wang and
B. Li (2008). "Androgen receptor-dependent regulation of Bcl-xL expression:
Implication in prostate cancer progression." Prostate 68(4): 453-461.
Susin, S. A., E. Daugas, L. Ravagnan, K. Samejima, N. Zamzami, M. Loeffler, P. Costantini, K. F.
Ferri, T. Irinopoulou, M. C. Prevost, G. Brothers, T. W. Mak, J. Penninger, W. C.
Earnshaw and G. Kroemer (2000). "Two distinct pathways leading to nuclear
apoptosis." J Exp Med 192(4): 571-580.
Susin, S. A., H. K. Lorenzo, N. Zamzami, I. Marzo, B. E. Snow, G. M. Brothers, J. Mangion, E.
Jacotot, P. Costantini, M. Loeffler, N. Larochette, D. R. Goodlett, R. Aebersold, D. P.
Siderovski, J. M. Penninger and G. Kroemer (1999). "Molecular characterization of
mitochondrial apoptosis-inducing factor." Nature 397(6718): 441-446.
Suzuki, Y., Y. Nakabayashi, K. Nakata, J. C. Reed and R. Takahashi (2001). "X-linked inhibitor of
apoptosis protein (XIAP) inhibits caspase-3 and -7 in distinct modes." J Biol Chem
276(29): 27058-27063.
Svingen, P. A., D. Loegering, J. Rodriquez, X. W. Meng, P. W. Mesner, Jr., S. Holbeck, A. Monks,
S. Krajewski, D. A. Scudiero, E. A. Sausville, J. C. Reed, Y. A. Lazebnik and S. H.
Kaufmann (2004). "Components of the cell death machine and drug sensitivity of the
National Cancer Institute Cell Line Panel." Clin Cancer Res 10(20): 6807-6820.
Sweat, G. T. (2005). "Guiding prostate cancer treatment choices. Early detection means more
options for more men." Postgrad Med 117(4): 45-50.
Tallarida, R. J. (2006). "An overview of drug combination analysis with isobolograms." J
Pharmacol Exp Ther 319(1): 1-7.
Tang, Y., M. A. Khan, O. Goloubeva, D. I. Lee, D. Jelovac, A. M. Brodie and A. Hussain (2006).
"Docetaxel followed by castration improves outcomes in LNCaP prostate cancerbearing severe combined immunodeficient mice." Clin Cancer Res 12(1): 169-174.
257
Tannock, I. F., R. de Wit, W. R. Berry, J. Horti, A. Pluzanska, K. N. Chi, S. Oudard, C. Theodore, N.
D. James, I. Turesson, M. A. Rosenthal and M. A. Eisenberger (2004). "Docetaxel plus
prednisone or mitoxantrone plus prednisone for advanced prostate cancer." N Engl J
Med 351(15): 1502-1512.
Tannock, I. F. and C. Lee (2001). "Evidence against apoptosis as a major mechanism for
reproductive cell death following treatment of cell lines with anti-cancer drugs." Br J
Cancer 84(1): 100-105.
Trosko, J. E., C. C. Chang, B. L. Upham and M. H. Tai (2004). "Ignored hallmarks of
carcinogenesis: stem cells and cell-cell communication." Ann N Y Acad Sci 1028: 192201.
Uzgare, A. R., Y. Xu and J. T. Isaacs (2004). "In vitro culturing and characteristics of transit
amplifying epithelial cells from human prostate tissue." J Cell Biochem 91(1): 196-205.
Vafa, O., M. Wade, S. Kern, M. Beeche, T. K. Pandita, G. M. Hampton and G. M. Wahl (2002).
"c-Myc can induce DNA damage, increase reactive oxygen species, and mitigate p53
function: a mechanism for oncogene-induced genetic instability." Mol Cell 9(5): 10311044.
van Es, J. H., N. Barker and H. Clevers (2003). "You Wnt some, you lose some: oncogenes in the
Wnt signaling pathway." Curr Opin Genet Dev 13(1): 28-33.
van Heerde, W. L., S. Robert-Offerman, E. Dumont, L. Hofstra, P. A. Doevendans, J. F. Smits, M.
J. Daemen and C. P. Reutelingsperger (2000). "Markers of apoptosis in cardiovascular
tissues: focus on Annexin V." Cardiovasc Res 45(3): 549-559.
Waldman, T., C. Lengauer, K. W. Kinzler and B. Vogelstein (1996). "Uncoupling of S phase and
mitosis induced by anticancer agents in cells lacking p21." Nature 381(6584): 713-716.
Wan, X., W. Ji, X. Mei, J. Zhou, J. X. Liu, C. Fang and W. Xiao (2010). "Negative feedback
regulation of Wnt4 signaling by EAF1 and EAF2/U19." PLoS One 5(2): e9118.
Wan, X., J. Liu, J. F. Lu, V. Tzelepi, J. Yang, M. W. Starbuck, L. Diao, J. Wang, E. Efstathiou, E. S.
Vazquez, P. Troncoso, S. N. Maity and N. M. Navone (2012). "Activation of beta-catenin
signaling in androgen receptor-negative prostate cancer cells." Clin Cancer Res 18(3):
726-736.
Wang, G., J. Wang and M. D. Sadar (2008). "Crosstalk between the androgen receptor and
beta-catenin in castrate-resistant prostate cancer." Cancer Res 68(23): 9918-9927.
Wang, L. G., X. M. Liu, W. Kreis and D. R. Budman (1999). "The effect of antimicrotubule agents
on signal transduction pathways of apoptosis: a review." Cancer Chemother Pharmacol
44(5): 355-361.
Wang, P., S. M. Henning and D. Heber (2010). "Limitations of MTT and MTS-based assays for
measurement of antiproliferative activity of green tea polyphenols." PLoS One 5(4):
e10202.
Wang, R., F. Lin, X. Wang, P. Gao, K. Dong, S. H. Wei, S. Y. Cheng and H. Z. Zhang (2008).
"Suppression of Bcl-xL expression by a novel tumor-specific RNA interference system
inhibits proliferation and enhances radiosensitivity in prostatic carcinoma cells."
Cancer Chemother Pharmacol 61(6): 943-952.
258
Wasylishen, A. R. and L. Z. Penn (2010). "Myc: the beauty and the beast." Genes Cancer 1(6):
532-541.
Watcharasit, P., G. N. Bijur, J. W. Zmijewski, L. Song, A. Zmijewska, X. Chen, G. V. Johnson and
R. S. Jope (2002). "Direct, activating interaction between glycogen synthase kinase3beta and p53 after DNA damage." Proc Natl Acad Sci U S A 99(12): 7951-7955.
Widyarini, S., N. Spinks, A. J. Husband and V. E. Reeve (2001). "Isoflavonoid compounds from
red clover (Trifolium pratense) protect from inflammation and immune suppression
induced by UV radiation." Photochem Photobiol 74(3): 465-470.
Wu, L. Y., T. De Luca, T. Watanabe, D. M. Morre and D. J. Morre (2011). "Metabolite
modulation of HeLa cell response to ENOX2 inhibitors EGCG and phenoxodiol."
Biochim Biophys Acta 1810(8): 784-789.
Wyllie, A. H., J. F. Kerr and A. R. Currie (1980). "Cell death: the significance of apoptosis." Int
Rev Cytol 68: 251-306.
Xie, Q., T. Lin, Y. Zhang, J. Zheng and J. A. Bonanno (2005). "Molecular cloning and
characterization of a human AIF-like gene with ability to induce apoptosis." J Biol Chem
280(20): 19673-19681.
Xie, W., Y. He, D. Huo, Y. Huang and W. Wu (2010). "A mimic of phosphorylated prolactin
inhibits human breast cancer cell proliferation via upregulation of p21 waf1." Med
Oncol 27(4): 1340-1345.
Yagiz, K., L. Y. Wu, C. P. Kuntz, D. James Morre and D. M. Morre (2007). "Mouse embryonic
fibroblast cells from transgenic mice overexpressing tNOX exhibit an altered growth
and drug response phenotype." J Cell Biochem 101(2): 295-306.
Yang, F., A. von Knethen and B. Brune (2000). "Modulation of nitric oxide-evoked apoptosis by
the p53-downstream target p21(WAF1/CIP1)." J Leukoc Biol 68(6): 916-922.
Yin, X. Y., L. Grove, N. S. Datta, M. W. Long and E. V. Prochownik (1999). "C-myc overexpression
and p53 loss cooperate to promote genomic instability." Oncogene 18(5): 1177-1184.
Yu, F., R. N. Watts, X. D. Zhang, J. M. Borrow and P. Hersey (2006). "Involvement of BH3-only
proapoptotic proteins in mitochondrial-dependent Phenoxodiol-induced apoptosis of
human melanoma cells." Anticancer Drugs 17(10): 1151-1161.
Yu, X., Y. Wang, D. J. DeGraff, M. L. Wills and R. J. Matusik (2011). "Wnt/beta-catenin activation
promotes prostate tumor progression in a mouse model." Oncogene 30(16): 18681879.
Zamzami, N., C. El Hamel, C. Maisse, C. Brenner, C. Munoz-Pinedo, A. S. Belzacq, P. Costantini,
H. Vieira, M. Loeffler, G. Molle and G. Kroemer (2000). "Bid acts on the permeability
transition pore complex to induce apoptosis." Oncogene 19(54): 6342-6350.
Zamzami, N., D. Metivier and G. Kroemer (2000). "Quantitation of mitochondrial
transmembrane potential in cells and in isolated mitochondria." Methods Enzymol
322: 208-213.
Zhang, D. Y., H. J. Wang and Y. Z. Tan (2011). "Wnt/beta-catenin signaling induces the aging of
mesenchymal stem cells through the DNA damage response and the p53/p21
pathway." PLoS One 6(6): e21397.
259
Zhao, L., M. G. Wientjes and J. L. Au (2004). "Evaluation of combination chemotherapy:
integration of nonlinear regression, curve shift, isobologram, and combination index
analyses." Clin Cancer Res 10(23): 7994-8004.
Zou, H., R. Yang, J. Hao, J. Wang, C. Sun, S. W. Fesik, J. C. Wu, K. J. Tomaselli and R. C.
Armstrong (2003). "Regulation of the Apaf-1/caspase-9 apoptosome by caspase-3 and
XIAP." J Biol Chem 278(10): 8091-8098.
260
10.APPENDICES
261