University of Southern California Dissertations and Theses

Role of cancer-associated fibroblast secreted annexin A1 in generation and maintenance of prostate cancer stem cells

(USC Thesis Other)

Role of cancer-associated fibroblast secreted annexin A1 in generation and maintenance of prostate cancer stem cells

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

ROLE OF CANCER-ASSOCIATED FIBROBLAST SECRETED ANNEXIN A1 IN
GENERATION AND MAINTENANCE OF PROSTATE CANCER STEM CELLS

by

Lauren Alexandra Rios Geary

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(INTEGRATIVE BIOLOGY OF DISEASE)

AUGUST 2013

Copyright 2013 Lauren Alexandra Rios Geary

Dedication

I would like to dedicate this dissertation to my dad, Lance Izumi, J.D., for not only has he
been a constant supporter of my pursuit of this Degree of Doctor of Philosophy, but he
has also been a teacher and patient listener and source of daily encouragement. Thank
you for the invaluable life lessons that you have helped to teach me along the way and for
making my educational success possible. I will never forget that you put me through high
school at an exceptional institution, essentially launching me toward a greater academic
future than would have otherwise been possible.

I will be praying for you and for your complete healing from prostate cancer. I feel that
along the way in my Ph.D. journey, we have both been challenged with the difficulties,
and as the British say, we have both learned to:

“Keep Calm
&
Carry On”
ii

Acknowledgements
I would sincerely like to thank my advisor Dr. Pradip Roy-Burman for enabling me to
pursue my personal dream and vision of creating not only a better life for myself, but also
for other people. This girl from the trailer park in the backwoods ghetto of Sacramento,
Florin Mobile Estates on Florin Road—known for the highest number of gang-related
deaths within all of Sacramento county, could have chosen an easy out and a less
promising future, like the boys and girls that I grew up with on my street and now reside
in correctional facilities, but I have always believed that I could succeed in the same
capacity as those of my peers who have come from stable backgrounds with strong(er)
financial support and in cultivating my academic talents to reach this stage in my career. I
believe it is my responsibility to use those talents for a greater good than that of my own
satisfaction. I have believed in the promise of this discovery of Annexin A1 in prostate
cancer and I would like to share it with the world for their evaluation and critique, and I
would like to thank my advisor for allowing me to pursue it with my own passion and to
tell the story that I believe is deserved of the research and the potentials for therapeutic
strategies that it holds. I know that we haven’t always seen eye-to-eye on the end point,
but I am confident now, at the end of journey, as we both believe in the impact of the
research that we will find a great and rewarding conclusion to this chapter of our careers
that has spanned the last five years.

And of course, I cannot forget about all of the lab members who have come and gone
through the lab. Thank you for putting up with my quirky sense of humor and tough, no-
iii

nonsense attitude. I especially am fond of the memories of birthday celebrations,
lunches—thank you Kumkum—and Fashion Fridays—thank you Ari (TWIN!) and Helty
(intrepid photographer). And also, thank you Lisa Doumak and Michele King, my friends
across the hall. You are both wonderful friends! Thank you for picking me to be your
unofficial, official mascot!

Additionally, I am grateful to my committee members, Dr. Ebrahim Zandi, Dr. Gregor
Adams and Dr. David Hinton for their steadfast devotion to my success as a graduate
student and for their invaluable counsel to my research. Thank you for your confidence in
me. It has immensely helped me to believe in myself. I especially acknowledge Dr. Zandi
for his continued belief in my abilities as evidenced by the words of confidence and
researchers that have been sent my way to instruct on isolating proteins from conditioned
media. I also acknowledge his laboratory for assistance in providing materials and
training for silver staining. A very special acknowledgement is made to Dr. Kevin Nash,
for his eagerly and thoroughly provided guidance in understanding and characterizing the
pathology of my prostate glandular structures from the in vivo experiments. Thank you
for your dedication to giving my project your full attention for long periods of time and
your overall advice to writing my dissertation…and for the Odwalla juice at Starbucks, I
am always hungry and thus it was much appreciated. Furthermore, I would like to thank
Lora Barsky of the FACS Core for all her technical assistance across many sorts.

iv

What has really helped me to endure the hardships faced in these last six years, from 4
a.m. spheroid-passaging to all-nighters, a torn lumbar disc and persistent plantar fasciitis
is the continued love and support of my family, friends and significant persons in my life.
I am especially grateful to all of my family members who have sent numerous prayers for
strength and success my way. Thank you to my loving parents, April Izumi and Lance
Izumi. I could not have faced this long race without your guidance and love. Thank you
Dad, for seeing the love in two women from a (scary) trailer park, patiently enduring the
neighbors, joining our family and building a better future for both of us. As I entered into
the last stretch of my graduate program, you have bravely entered into your battle with
prostate cancer. For that reason I have poured extra effort into the conclusion of this body
of work and I am dedicating this dissertation to you.

Thank you to all of my friends for your support and the confidence that you have shared
in me. You have all had a significant impact on my life here at USC. I will never forget
the encouragement, support and happy memories that we have shared along the way.
Sarah, Christine, Yvonne, and Ben. You have become a part of my family and I hope that
we will continue to create new memories. Sarah, thank you for your close friendship
throughout the years, the shared venting, exercise and delicious dinners. It is amazing that
we met the first day of interviews, it seems so long ago. Christine, you are one of my
very dearest friends and I could not have faced this final portion of the process without
your help and kind support. I am looking forward to more shopping, laughter and fun
together! Thank you for your template for dissertation formatting, you really saved me!
v

Yvonne, you have always been so sweet and kind to me and I really appreciate it. Thank
you for your readiness to spend time with me! Ben, you are missed in L.A., but I hope
that in the future I will see you. It was a pleasure to share this experience with all of you.
You are all dear to my heart and I hold close all the dinners (and drinking), shopping,
coffee, conversations and adventures we have shared. I sincerely hope that we will
maintain the close relationships that we were able to foster here in the years to come and
enjoy the fruits of our labor together! I am looking forward to seeing where we all end up
in life and remembering our experiences in grad school as we look back. Finally, thank
you to Dr. Vejas Skripkus for your constant generosity, compassion, caring and patience,
especially as the journey got rough. I will not forget your words of advice, support and
encouragement.

Lastly, and most importantly of all, I thank my Lord and God for giving me the strength
and talents necessary to succeed in the completion of this study.
vi

vii

Table of Contents

Dedication ................................................................................................................... ii

Acknowledgements ...................................................................................................iii

List of Tables ............................................................................................................. ix

List of Figures............................................................................................................. x

Abbreviations ..........................................................................................................xiii

Abstract..................................................................................................................xviii

Chapter 1: Introduction ............................................................................................ 1
1.1 Prostate Adenocarcinoma .......................................................................... 1
1.2 CSCs, EMT & Tumor Microenvironment: Implications for Research...... 6
1.3 Cancer Stem Cells.................................................................................... 11
1.4 Epithelial-to-Mesenchymal Transition .................................................... 19
1.5 Tumor Microenvironment........................................................................ 23
1.6 Annexin A1 and Pathobiological Roles................................................... 25
1.7 Significance.............................................................................................. 32

Chapter 2: Mouse Cell Culture Model for EMT Induction by Secreted
Factors ................................................................................................... 35
2.1 Abstract.................................................................................................... 35
2.2 Introduction.............................................................................................. 37
2.3 Results...................................................................................................... 41
2.4 Discussion................................................................................................ 53

Chapter 3: Annexin A1 Secreted from Cancer-Associated Fibroblasts Promotes
Cancer Stem Cell-like Properties........................................................ 57
3.1 Abstract.................................................................................................... 57
3.2 Introduction.............................................................................................. 58
3.3 Results...................................................................................................... 61
3.4 Discussion................................................................................................ 99

Chapter 4: Concluding Remarks & Future Directions...................................... 107
4.1 Summary................................................................................................ 107
4.2 General Discussion ................................................................................ 111
4.3 Limitations of the Study......................................................................... 115
4.4 Future Research ..................................................................................... 119
Chapter 5: Materials and Methods ...................................................................... 123
5.1 Experimental Animals ........................................................................... 123
5.2 Isolation of Cell Cultures and Cell Sorting............................................ 125
5.3 Cell Culture and Assays for Spheroid Formation.................................. 126
5.4 Conditioned Media and AnxA1 Ligands ............................................... 128
5.5 Renal Grafting........................................................................................ 131
5.6 Immunostaining and Western Blots....................................................... 133
5.7 PCR Analysis......................................................................................... 135
5.8 Statistical Analysis................................................................................. 136

References............................................................................................................... 137
viii

List of Tables

Table 5.1 Antibody List ................................................................................... 134

Table 5.2 PCR Primer List............................................................................... 135

ix

List of Figures
Figure 1.1 Annexin A1 and the “vicious cycle” of CSC generation and maintenance. 10
Figure 1.2 Prostate stem cell lineages..................................................................... 17
Figure 2.1 Prostaspheres in 3D culture ................................................................... 41
Figure 2.2 Transcriptional profiles of E4, primary, and cE1, recurrent, cell lines....... 42

Figure 2.3 cE1 3-D co-culture assay with fibroblasts of the model ........................... 43
Figure 2.4 cE1 3-D assay with TGF- β1 and TNF α for EMT induction...................... 45

Figure 2.5 Acceleration of CAF CM induced EMT by TGF- β1 and TNF α................ 46

Figure 2.6 Culture of FACS-isolated SC
hi
, SC
med
and SC
none
cE1 cells...................... 47

Figure 2.7 EMT transcriptional profile of cE1 SC
hi
and SC
med
TGF- β1 or CAF CM
treated cells.......................................................................................... 48

Figure 2.8 Phenotypic changes consistent with EMT in cE1 SC
hi
and SC
med
culture
with CAF CM ...................................................................................... 49

Figure 2.9 Ammonium sulfate precipitation of conditioned media proteins............... 51

Figure 2.10 Identification of CAF CM fraction enriched for EMT-inducing protein
activity................................................................................................. 52

Figure 3.1 CAFs induce EMT in SC
med
PCa cells in vitro ........................................ 63

Figure 3.2 SC
med
cells treated with secreted CAF CM factors show upregulation of
EMT/stem cell transcription factors in vitro............................................ 66

Figure 3.3 Expression of EMT/stem cell transcription factors increases after 14 days
in SC
med
via CAF CM treatment ............................................................ 67

Figure 3.4 AnxA1 is secreted from CAF ................................................................ 69

Figure 3.5 AnxA1 transcription levels in NPF versus CAF ...................................... 70

Figure 3.6 Effect of α-AnxA1 neutralizing antibody (nAb) on spheroids and EMT
of cE1 cells .......................................................................................... 72
x

Figure 3.7 Gross morphology of cE1 grafts under renal capsule............................... 73
Figure 3.8 SC
med
cells treated with CAF CM and Ac2-26 undergo de-differentiation
under the renal capsule of NOD.SCID male mice in vivo......................... 74

Figure 3.9 Significant increase in p63
+
cells in Ac2-26 treated grafts indicates
de-differentiation of SC
med
cells............................................................. 74

Figure 3.10 AnxA1 treatment of LSC
hi
cells of the cPten
-/-
L model in vitro ................ 76
Figure 3.11 Spheroid enrichment by AnxA1 mimetic and recombinant forms............. 77

Figure 3.12 Analysis of pathways associated with AnxA1......................................... 78

Figure 3.13 Twist expression after treatment of LSC
hi
spheroids with Ac2-26 ............ 79

Figure 3.14 Proposed schematic for AnxA1 mode of action in prostate cancer
microenvironment ................................................................................ 80

Figure 3.15 Drawing of different glandular morphologies observed in LSC
hi/med
......... 85
Figure 3.16 LSC
hi
graft gross morphology................................................................ 86

Figure 3.17 Incidence and types of glandular structures seen in LSC
hi
grafts............... 87
Figure 3.18 Histological images of representative untreated control, Ac2-26 treated
and rAnxA1 treated LSC
hi
grafts............................................................ 88

Figure 3.19 Induction by AnxA1 for maintenance of stem cell biological activity....... 89
Figure 3.20 LSC
med
graft gross morphology ............................................................. 90

Figure 3.21 Incidence and types of glandular structures seen in LSC
med
grafts ............ 91
Figure 3.22 Histological images of representative untreated control, Ac2-26 treated
and rAnxA1 treated LSC
med
grafts ......................................................... 92

Figure 3.23 Induction by AnxA1 for de-differentiation ............................................. 93
Figure 3.24 Incidence and size of glandular structures in LSC
hi
and LSC
med
grafts ...... 94

xi

Figure 3.25 Comparison of p63
+
and CK8
+
cells between glandular morphologies in
AnxA1 treated LSC
hi
and LSC
med
grafts ................................................. 95

Figure 3.26 Histologies of acinar glandular structures in AnxA1 treated LSC
hi
and
LSC
med
grafts........................................................................................ 96

Figure 3.27 AnxA1 supports p63
+
cells and proliferation, and therefore may support
the stem cell niche ................................................................................ 97

Figure 4.1 AnxA1 promotes both EMT and differentiation processes, leading to
de novo generation of CSCs and maintenance of existing CSCs............. 119

Figure 5.1 Ammonium sulfate precipitation schematic .......................................... 129
Figure 5.2 Generation of murine full-length recombinant AnxA1 protein ............... 130
Figure 5.3 Required materials and bench procedure for renal capsule
transplantation.................................................................................... 131

Figure 5.4 Isolation of grafts after 10 weeks ......................................................... 132

xii

Abbreviations
AC2-26 N-terminal derived, 25 amino acid mimetic of annexin a1
AKT protein kinase B
ALXR lipoxin A4 receptor
ANXA1 annexin a1
ANOVA analysis of variant
AR androgen receptor
AS ammonium sulfate
BSA bovine serum albumin
BLI bioluminescence imaging
CAF cancer-associated fibroblast
CD49F integrin α-6
cDNA complementary DNA
CK5 cytokeratin 5
CK8 cytokeratin 8
cKO conditional knockout
CM conditioned medium
CRE cre recombinase derived from the P1 bacteriophage
CRPC castration-resistant prostate cancer
CSC cancer stem cell
DAB 3,3'-diaminobenzidine
DAPI 4',6-diamidino-2-phenylindole
xiii

DMEM Dulbecco’s modified Eagle’s medium
DNASE I deoxyribonuclease I
ECM extracellular matrix
EDTA ethylene diamine tetraacetic acid
EGFR-TK epidermal growth factor receptor tyrosine kinase
EMT epithelial-to-mesenchymal transition
ERK extracellular signal-regulated kinase
ESC embryonic stem cell
F12 Ham’s F12 media
FACS fluorescence-activated cell sorting
FBS fetal bovine serum
FITC fluorescein isothiocyanate
FPR formyl peptide receptor
FPR-RS formyl peptide receptor related protein
GAP GTPase activating protein
GAPDH glyceraldehyde 3-phosphate dehydrogenase
GC glucocorticoid
GFP green fluorescent protein
GS goat serum
H&E hematoxylin and eosin
HEPES 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid
HGFR-TK hepatocyte growth factor receptor tyrosine kinase
xiv

IACUC Institutional Animal Care and Use Committee
IF immunofluorescence
IHC immunohistochemistry
IP immunoprecipitation
IPA Ingenuity Systems, Inc., Pathway Analysis
IPTG isopropylthio- β-D-galactoside
JNK c-Jun N-terminal kinase
Ki67 nuclear protein associated with cellular proliferation
KO knockout
LIN lineage specific for endothelial, hematopoietic cells
MAPK mitogen-activated protein kinase
MEM minimum essential medium
MET mesenchymal-to-epithelial transition
MMP matrix metalloproteinase
mRNA messenger RNA
NEAA non-essential amino acids
NF- κB nuclear factor kappa-light-chain-enhancer of activated B

NOD.SCID non-obese diabetic/severe combined immunodeficiency

NPF normal prostate fibroblast
OCT optimal cutting temperature compound
P63 tumor protein p63, aka transformation-related protein 63
PAK p21-activated kinase
xv

PBS phosphate buffered saline
PBS-T phosphate buffered saline-triton X-100
PCA prostate cancer
PI3K phosphoinositide 3-kinase
PIP3 phosphatidylinositol (3,4,5)-trisphosphate
PCR polymerase chain reaction
PDGFR-TK platelet derived growth factor receptor tyrosine kinase
PE phycoerythrin
PFA paraformaldehyde
PrEGM basal media for culturing prostate epithelial cells
PKC protein kinase c
PTEN phosphatase and tensin homologue
qRT-PCR real time quantitative reverse transcription-polymerase
chain reaction
rANXA1 recombinant murine annexin a1
rhEGF recombinant human epidermal growth factor
RHOB ras homologue family member b
RIPA radio immunoprecipitation assay
ROCK Rho-associated coiled-coil-forming kinase
RPMI Roswell Park Memorial Institute medium
RT room temperature
SCA-1 stem cell antigen 1
xvi

TBS tris-buffered saline
TBS-T tris-buffered saline-tween-20
TGF- β1 transforming growth factor-beta 1
TGF- βRII transforming growth factor-beta receptor type 2
TIC tumor-initiating cell
TNF α tumor necrosis factor α
TRITC tetramethyl rhodamine isothiocyanate
TRPM7 transient receptor potential cation channel, subfamily M,
member 7
UGSM urogenital sinus mesenchyme
WB western blot
WT wild-type
WTBS tris-buffered saline-triton X-100
xvii

Abstract
Annexin A1 (AnxA1), a phospholipid-binding protein and known regulator of
glucocorticoid-induced inflammatory signaling, has been implicated for its role in human
cancers, including head and neck cancer, colon, breast and prostate cancers. To determine
the role of AnxA1 found to be secreted by the cancer-associated fibroblasts (CAF) of
prostatic adenocarcinoma, we used primary cultures of CAFs, primary epithelial cultures
and an epithelial cell line (cE1), all derived from the cPten
-/-
L mouse model of prostate
adenocarcinoma in a series of experiments described here. The results of this dissertation
point to a novel function of AnxA1 in prostate tumorigenesis and the establishment of a
Cancer Cell-EMT-CSC lineage. It is projected that the presence of AnxA1 in the tumor
microenvironment contributes to tumor stem cell activity via two separate but
complementary pathways, first by induction of a de-differentiation process leading to
generation of basal stem-like cells from a subpopulation of cancer epithelial cells, and,
second, by enhancing the proliferation and differentiation of the CSCs.

The cE1 cell line was used as a model system and after fractionation into subpopulations
based on the levels of cell surface expression of Sca-1 (S) and CD49f (C) antigens: SC
hi
,
SC
med
and SC
none
, it was found that SC
med
were preferentially susceptible to epithelial-to-
mesenchymal transition (EMT) after treatment with CAF conditioned medium (CM) and
AnxA1 mimetic (Ac2-26) protein, in vitro. Following EMT induction, these cells
acquired stem cell-like properties similar to intrinsic properties of the SC
hi
group based
on up-regulation of stem cell transcription factors (i.e. Oct4), as well as upregulation of
xviii

EMT transcription factors (i.e. Snail and Twist) and mesenchymal markers (i.e.
Vimentin). Gain of stem cell-like biological activity in vivo by SC
med
after EMT
induction by AnxA1 was demonstrated, using renal capsule transplantation technique, by
gain of ability for formation of prostatic glandular structures with high expression of
basal cell marker, p63, consistent with prostate stem/CSCs.

Primary prostate cancer epithelial cell cultures from the model were used to further
define the role of AnxA1 on cells from the CSC niche. Primary cultures of Lin
-
(Ter119,
CD31, CD45; L) SC subpopulations—LSC
hi
and LSC
med
—responded to treatment with
Ac2-26 and murine recombinant full-length AnxA1 (rAnxA1) in vitro through the
formation of greater numbers of spheroids and increased complexity in Matrigel assays.
In vivo, LSC
hi
cells treated with Ac2-26 and rAnxA1 generated more complex, more
abundant glandular structures than untreated controls, showed increase of p63
+
cells,
possessed greater cell numbers and greater proliferation as indicated by overall cell
counts and Ki67
+
staining, and activation of pErk1/2. In vivo, LSC
med
Ac2-26 and
rAnxA1 treated cells displayed significantly enhanced prostatic glandular formation over
controls, significantly higher p63
+
and Ki67
+
cells, three-fold increase in the ratio of basal
to luminal p63
+
:CK8
+
cells (P < 0.05) and similarly, activation of pErk1/2. LSC
hi
cells,
which already contain enrichment for cells with glandular structure formation abilities,
appear to be maintained and more prolific after AnxA1 treatment. LSC
med
cells, which do
not possess the ability to form glandular structures in vivo, were induced for stem cell-
like ability to form glandular structures after AnxA1 treatment and, thus, confirm results
xix

xx

from the cE1 cell line that cells that are not cancer stem cell-like can be induced to
become CSCs.

Chapter 1:
Introduction
“Tumors destroy man in a unique and appalling way, as flesh of his own flesh
which has somehow been rendered proliferative, rampant, predatory and
ungovernable. They are the most concrete and formidable of human maladies, yet
despite more than 70 years of experimental study they remain the least
understood.”
—Francis Peyton Rous, cancer virologist, Nobel lecture, 1966
1.1 Prostate Adenocarcinoma
1.1.1 Prostate Cancer in Man
Cancer remains one of the most relevant topics in disease and therapy in the modern
world. Despite an increasing wealth of knowledge and emerging clinical trials and
therapeutics, within the United States alone, an estimated diagnosis of 1.6 million new
cases of cancer and 580,350 cancer deaths are projected to occur in both men and women
within the year 2013 (Siegel, Naishadham, & Jemal, 2013). Adding to that statistic,
cancer claims as many as 7.6 million lives worldwide each year, with 12.7 million new
cases, and is the leading cause of death in men and women greater than 40 years of age in
developed countries (Jemal et al., 2011). In the United States, data compiled by the
American Cancer Society estimates 238,590 new cases of diagnosed prostate cancer with
an associated 29,700 deaths in men in 2013. According to the Centers for Disease Control
and Prevention (CDC), continuing in 2013 from their report in 2011, aside from non-
melanoma skin cancer, prostate cancer is the most prevalent solid cancer among men in
the western world, wherein approximately 1 in 3 (28%) men aged 50 years or greater will
be diagnosed with the disease at any stage, and 10% of which men will succumb to
1

disease mortality. At all ages, prostate cancer is the second leading cause of death, below
heart disease, in the United States. Interestingly, the state with the highest number of new
cases of prostate cancer is California, with almost three-fold higher incidence than any
other state (Jemal et al., 2011; Siegel et al., 2013).

Despite the apparent mortalities due to cancer, there has been a significant decrease in the
annual number of cancer related deaths over the last 5 years (1.8% in men and 1.5% in
women). Eliciting greater hope and promise is the 20% decline in the cancer death rate
since 1991, which was recorded at a peak rate of 215.1 deaths per 100,000 population.
Translated this decline in the rate of deaths due to cancer has averted approximately 1.18
million deaths in both men and women due to cancer. Similarly, the incidence of prostate
cancer diagnosis reached its peak in 1991 at approximately 250 per 100,000 population.
Since then it has declined to a recorded 153 per 100,000 population in 2009. Jointpoint
trend analysis projects an annual 1.9% decrease in incidence for prostate cancer, although
wide year to year fluctuations are observed and are likely attributable to the standards for
prostate-specific antigen testing for the detection of prostate cancer (Jemal et al., 2011;
Siegel et al., 2013).

In approximately 30% of primary tumors and 63% of metastatic tumors, the tumor
suppressor gene phosphatase and tensin homologue (Pten) is found to be mutated and/or
repressed functionally, thereby implicating Pten mutation as one of the most common
genetic alterations leading to prostate cancer (Dahia, 2000; Nikitin, Matoso, & Roy-
2

Burman, 2007; Sellers & Sawyers, 2002; Suzuki et al., 1998). Additionally, Pten-
controlled signaling pathways are found to be frequently aberrant in human prostate
cancers, leading to downstream accumulation of Pten targets, specifically PIP3. One
major physiological role of Pten is control of cell cycle activities through its phosphatase
activity and subsequent antagonism of the PI3K/AKT pathway. Loss of function of Pten
has been shown to result in constitutive over-activity of the PI3K/AKT pathway in both
human cancer cell lines and mouse models of prostatic adenocarcinoma (Cantley & Neel,
1999; Sellers & Sawyers, 2002; H. Sun et al., 1999; Wang et al., 2003).

1.1.2 Mouse Model for Human Prostate Cancer
Mouse models have been an essential tool for studying disease progression and
pathobiology of prostate cancer. Several well-documented models exist, including the
transgenic conditional Pten deletion and TMPRSS2/ERG mouse models, both of which
employ an androgen-regulated mode of oncogenesis. (For the most recent review of
mouse models of prostate cancer, see Wu, X., et al (2013). Current mouse and cell
models in prostate cancer research. Endocrine-Related Cancer 20,R115-R170. (X. Wu,
Gong, Roy-Burman, Lee, & Culig, 2013).) The conditional Pten deletion mouse model,
reported for its biological relevance with incidence of both invasive carcinoma and
metastasis was generated in collaboration between our laboratory group and others as
reported in (Wang et al., 2003; X. Wu et al., 2001; Zhong, Saribekyan, Liao, Cohen, &
Roy-Burman, 2006).

3

The conditional Pten deletion mouse model, abbreviated here as cPten
-/-
, faithfully
mimics the course of human prostate cancer through its progression to a metastatic stage.
Under normal circumstances, wild type mice do not develop murine prostatic
intraepithelial neoplasia (mPIN) until late in their lives (12-16 months) and do not
progress toward prostate cancer. In the conditional Pten deletion mouse model, however,
homozygous, prostate epithelial cell specific deletion of Pten leads to adenocarcinoma in
all lobes of the mouse prostate by 12 weeks of age. In these mice the disease progresses
from epithelial hyperplasia at 4 weeks of age to mPIN (6-9 weeks of age), and then to
invasive adenocarcinoma from 9 to 12 weeks of age. In the final stage, lymphovascular
invasion of prostate carcinoma cells is observed in conditional Pten null mice (Wang et
al., 2003; X. Wu et al., 2001). This mimics the disease course in humans, which similarly
progresses through defined stages of hyperplasia, PIN, prostate cancer in situ, to invasive
adenocarcinoma and then to metastatic carcinoma (Bubendorf et al., 2000; Cher, 2001).

Additional benefits of the cPten
-/-
mouse model are the histopathological similarities
between secretive epithelial cell of origin in cPten
-/-
tumors and human tumors, initial
response to androgen ablation followed by recurrence driven by androgen-independent
growth and gene expression profiles following Pten deletion that are seen in human
prostate cancers (Wang et al., 2003). Our laboratory extended the application of the
conditional Pten deletion mouse model by the addition of the firefly luciferase gene
under Cre/loxP mediated expression, with the rationale that cells in which Cre
recombinase was active would also be subject to activation of a reporter allele,
4

5

Luciferase, thereby creating the cPten
-/-
L mouse model for non-invasive, live
bioluminescence imaging (BLI) of animals burdened with tumors (Liao et al., 2007). For
the remainder of this manuscript, the conditional Pten deletion mouse model that has
been applied within our laboratory and for my dissertation study experiments shall be
abbreviated as cPten
-/-
L.
1.2 Cancer Stem Cells, Epithelial-to-Mesenchymal Transition and
Tumor Microenvironment: Implications for Research
1.2.1 Converging Topics: Interrelation Between Cancer Stem Cells, Epithelial-to-
Mesenchymal Transition and the Tumor Microenvironment
Many factors of the tumor microenvironment have been shown to exert biological effects,
including metastatic capacity, on malignant tumor cells. The tumor microenvironment
constitutes a heterogeneous niche of mesenchymal cell types, and factors derived from
inflammatory immune cells, endothelial cells, pericytes and cancer-associated fibroblasts
(CAFs) are reported to promote metastasis (Hanahan & Weinberg, 2011; Joyce &
Pollard, 2009; McAllister & Weinberg, 2011; Pietras & Ostman, 2011). Metastasis
accounts for a majority of cancer-related deaths in carcinomas concomitant with organ
failure (G. P. Gupta & Massague, 2006). In the last decade, emerging evidence supports
the notion that tumor microenvironmental factors induce epithelial-to-mesenchymal
transition (EMT) which confers stem cell-like traits and mediates pro-metastatic
properties of tumor cells (Kalluri & Weinberg, 2009).

EMT is a key biological process whereby de-differentiation leads to conversion of
sessile, polarized epithelial cells to motile, fibroblastoid cells displaying mesenchymal
markers (Huber, Kraut, & Beug, 2005; Kalluri & Weinberg, 2009). In cancers, tumor
cells that have undergone EMT lose epithelial polarity and marker expression, gain
resistance to chemotherapy and apoptosis, and acquire stem cell gene expression profiles
and multipotential differentiation (Al-Hajj, Wicha, Benito-Hernandez, Morrison, &
6

Clarke, 2003; Giannoni et al., 2010; P. B. Gupta et al., 2009; Kalluri & Weinberg, 2009;
Mani et al., 2008; Scheel & Weinberg, 2012; Thiery, 2002; Thiery, Acloque, Huang, &
Nieto, 2009).

1.2.2 Hypotheses and Rationale
From this body of evidence, there is a suggested convergence between the tumor
microenvironment, EMT and cancer stem cells (CSCs). I will further explore these
convergent topics in the following sections and attempt to lead a discussion that will
introduce my research in support of my dissertation hypotheses. My overarching
dissertation hypothesis is: cancer-associated fibroblasts (CAFs) in the tumor
microenvironment secrete paracrine factors which:
1. help expand the cancer stem cell subpopulation, and
2. induce an epithelial-to-mesenchymal-like transition to replenish the cancer stem
cell subpopulation.
In essence, cells that are not cancer stem-like cells can be induced to become cancer stem
cells. Early evidence as described in Chapter 2 and Chapter 3, Section 3.3.1,
(upregulation of EMT/stem cell transcription factors and biological activity) provided the
basis for formulating this initial hypothesis.

Following identification of a CAF-secreted factor, Annexin A1 (AnxA1), I further
defined a secondary hypothesis related to the specific role of this soluble protein on the
tumorigenicity of prostate cancer epithelial cells. This hypothesis states that AnxA1:
7

1. mediates epithelial-to-mesenchymal transition in susceptible subpopulations of
epithelial prostate tumor cells, and
2. promotes cancer stem cell-like properties.
Based on the literary findings of the field supporting the notion that protein factors from
the tumor microenvironment may provide heterogeneous cancer cells with the necessary
stimulation to become tumor-initiating or cancer stem/progenitor cells and additional
supportive evidence outlined in Chapter 3, Section 3.3.2 and Section 3.3.3, I rationalized
that cells could be enhanced by exposure to AnxA1 for gain of cancer stem cell-like
biological properties (Figure 1.1).

Both hypotheses follow the understated notion that within the tumor epithelial
population, there exist subpopulations of cells with differing potentials for EMT and stem
cell characteristics. As I discuss in Chapter 3, the understated notion here reflects the data
from experimental models of induction through EMT to gain of CSC characteristics in
vitro and enhanced tumorigenicity and presence of cancer stem/progenitor cells in vivo.
My tertiary hypothesis therefore asserts that definable subpopulations of prostate cancer
epithelial cells:
1. possess the properties of cancer stem cell-like cells, and
2. may be susceptible to epithelial-to-mesenchymal transition.
This final hypothesis describes the paradigm wherein the tumor microenvironment
initiates a vicious cycle in which cancer stem cells are a perpetually renewing population.
8

Microenvironmental cues can induce cancer cells to become cancer stem cells, which
give rise to more cancer cells that are capable of being transformed to cancer stem cells.

Annexin family members are emerging as important mediators of tumor
microenvironmental effects, including EMT in cancer (L. H. Lim & Pervaiz, 2007;
Mussunoor & Murray, 2008). (For more information on Annexin family members refer to
Gerke, V., Creutz, C.E., & Moss, S.E. (2005). Annexins: Linking Ca
2+
Signaling to
Membrane Dynamics. Nature 6,449-461.) (Gerke, Creutz, & Moss, 2005) Within this
body of research, I will shed further light on the role of one specific Annexin family
member, AnxA1, as one such microenvironmental cue on prostate cancer development
and the “vicious cycle” of CSCs and their propagation (as depicted in Figure 1.1).
9

Figure 1.1 Annexin A1 and the “vicious cycle” of CSC generation and maintenance. Annexin
A1 secreted by cancer-associated fibroblasts may be involved in both in vivo generation and
maintenance of cancer stem cells.

CSC
TAC
TDC
CAF

De-differentiation
Self-renewal
Differentiation
Stroma
Epithelium

T-LSC
hi

AnxA1
10

1.3 Cancer Stem Cells
1.3.1 “Discovery” of Cancer Stem Cells: Where Do They Come From?
Within the last decade, well recognized evidence from studies of a broad spectrum of
solid tumors in brain, breast, colon, lung, liver, pancreas, ovarian, head and neck,
melanoma and prostate cancers (Al-Hajj et al., 2003; Collins, Berry, Hyde, Stower, &
Maitland, 2005; Fang et al., 2005; Ferrandina et al., 2008; Kim et al., 2005; Li et al.,
2007; Ma et al., 2007; Prince et al., 2007; S. K. Singh et al., 2004), following the
pioneering work in the hematopoietic system (Bonnet & Dick, 1997), have provided
evidence supporting the existence of a population of “tumor-initiating” cells (TICs), also
termed cancer stem cells (CSCs). These CSCs comprise only a small percentage of the
total tumor cell population; however, are notably recognized for their ability to generate
new tumors in vivo. Independently of this research, normal tissue stem cells have become
well defined by their dual capacity to renew themselves and simultaneously give rise to
committed progeny through the process of asymmetrical division (Reya, Morrison,
Clarke, & Weissman, 2001). CSCs share the properties of self-renewal and differentiation
with normal stem cells, prompting the question of whether similarities between the two
are indicative of a similar point of origin. Two competing theories describe the alternate
pathways by which either (a) normal stem cells may accumulate successive mutations
throughout the life of an adult organism and give rise to cancer stem cells or (b) fully
differentiated cancer cells may undergo a process of de-differentiation to give rise to
tumor-initiating progenitor cells. One model proposes intrinsic susceptibilities or
functional properties within cells that later arise as CSCs, indicating that disease
11

progression was their unalterable fate all along. An alternative hypothesis proposed for
the origin of CSCs, suggests that a heterogeneous population of cancer cells may receive
extracellular cues from the tumor microenvironment to acquire stem cell-like properties
(Rosen & Jordan, 2009).

Original evidence for CSCs and their debated origin has come from rigorous
identification, isolation and testing of hematopoietic stem cells (HSCs). Early research
into a cell of origin from leukemia has paved the way for understanding three key aspects
of putative CSCs: (1) similarities between self-renewal of normal stem cells and CSCs,
(2) the two possibilities of origin of CSCs, and (3) the clonogenicity of CSCs as rare cells
within a tumor that are capable of driving formation, proliferation and growth of tumors.
Because HSCs and their differentiated lineages are well characterized, they have lent
themselves as a highly suitable model system for studying the similarities between
normal and cancer stem cells (Reya et al., 2001). In leukemia cells, pathways known for
regulating normal stem cell development and self-renewal, such as Notch, Sonic
hedgehog and Wnt signaling pathways, have been found to be associated with
oncogenesis (Taipale & Beachy, 2001).

More recently, literature from the last few years has implicated an association between
normal stem cell transcription factors and cancer cells with the properties of self-renewal
and metastatic potential. Of significant note is the oncogenic expression of Oct4, which is
strongly correlated with overall poor prognosis and chemoresistance in ovarian and breast
12

cancers, and has also been found to be expressed in human lung adenocarcinomas, and is
used clinically as a diagnostic marker in primary and metastatic carcinomas (Gidekel,
Pizov, Bergman, & Pikarsky, 2003; Karoubi, Gugger, Schmid, & Dutly, 2009; Liu et al.,
2011; Samardzija, Quinn, Findlay, & Ahmed, 2012; Sung, Jones, Beck, Foster, & Cheng,
2006). The importance of Oct4 as a marker of tumorigenicity and cancer stem cell-like
properties is further indicated in the analysis presented in Chapter 3 on transcription
profiles of cancer cells that have undergone an EMT transition in vitro.

Functionally, CSCs have been defined by their ability to efficiently seed tumors when
injected in limiting dilutions into immunocompromised host mice. Results from limiting
dilution xenograft and transplantation experiments, as well as subpopulation enrichment
for markers associated with normal tissue stem cells support the overall notion that a
small, definable subset of cancer cells are capable of tumorigenic stem cell properties
(Al-Hajj et al., 2003; Cho & Clarke, 2008; Mani et al., 2008; Park, Bergsagel, &
McCulloch, 1971). Recently research now implicates epithelial-to-mesenchymal
transition (EMT), which is a studied mechanism accepted for initiating metastases of
epithelial cancers (Rosen & Jordan, 2009; Santisteban et al., 2009; Thiery et al., 2009).

Initial support for the theory of a tumor-initiating cell type has been pioneered by
research into breast cancer. Within the last decade a CD44
high
CD24
low
putative mammary
stem cell subpopulation has been isolated and demonstrated to reconstitute glandular
mammary structures in cleared murine mammary fat pads (Moraes et al., 2007;
13

Shackleton et al., 2006; Stingl et al., 2006). The same criteria used to isolate these
CD44
high
CD24
low
stem cells has repeatedly been found to enrich for cancer stem cells,
which are able to reconstitute the original tumor phenotype in vivo (Al-Hajj et al., 2003;
Mani et al., 2008). Moreover, Mani et al. (2008), demonstrated that cells displaying both
the CD44
high
CD24
low
surface phenotype and enhanced metastatic abilities could be
enriched for from normal human and mouse mammary cells after induction through an
EMT (Morel et al., 2008; Santisteban et al., 2009). Further evidence of the potential link
between cancer stem cells and EMT is the generation of a mesenchymal morphology in
cells enriched for CD44
high
CD24
low
by both isolation of the normal stem cell population
from human and murine mammary epithelial cells and induction of normal mammary
epithelial cells by EMT. Also significant were the findings that human mammary
epithelial cells (HMLEs) that underwent EMT gained greater spheroid-forming capacity,
increased chemotherapeutic drug resistance and multipotential differentiation, whereas
the CD44
low
CD24
high
cells, which comprised a majority of the unsorted original epithelial
cell population, did not display any of the aforementioned stem cell-like characteristics
(Mani et al., 2008).

Recent findings in breast cancer research have been applicable to further understanding
the molecular etiology of prostate cancer. A putative prostate cancer stem cell population
has been identified from human tumors as CD44
+
CD133
+
α
2
β
1
high
and is shown to be
highly enriched for stem cell regulatory markers Oct4, Bmi1 and β-catenin (Collins et al.,
2005; Hurt & Farrar, 2008; Patrawala et al., 2006). When isolated from both primary and
14

established prostate cancer cell lines, including LNCaP and DU145,
CD44
+
CD133
+
α
2
β
1
high
cells show enhanced invasiveness and tumorigenicity, whereas the
remaining non- CD44
+
CD133
+
α
2
β
1
high
cell population does not display invasiveness and
has diminished tumorigenic potential. A recent publication by Klarmann et al. (2009)
(Klarmann et al., 2009) has further correlated the expression patterns of genes known to
be regulators of EMT with these isolated invasive prostate CSCs.

In summary of the competing hypotheses regarding the point of origin of CSCs (intrinsic
and extrinsic models), while it is plausible to suggest that cancer stem cells arise from
long-lived normal stem cells, having accumulated sequential genetic mutations over time,
and allowing for a slow progression towards abnormal and unregulated cell growth (Ponti
et al., 2005; Reya et al., 2001); alternatively, and as surmounting evidence from the
studies of EMT in cancer cells suggests, differentiated cancer cells may undergo de-
differentiation into a more mesenchymal phenotype and acquire the capacities for self-
renewal, multipotential differentiation and invasiveness. These carcinoma cells that have
undergone EMT become motile, stem cell-like, resistant to chemotherapy and apoptosis,
and are thought to be responsible for seeding distant metastases in the body (Al-Hajj et
al., 2003; Mani et al., 2008). Tumor metastasis is, in part, facilitated by EMTs and further
supports the hypothesis that the EMT process may generate CSCs (Thiery, 2003).

Even with the wealth of information that has been built in the last decade, it will remain
to be exciting as this field of research continues to expand in answering two very relevant
15

questions to cancer researchers: (1) where within the original tissue do cancer stem cells
originate and (2) what extent does EMT play in the generation of CSCs? The goal of my
research is to reveal how induction by EMT enriches for putative CSCs as defined by
their functional self-renewal, differentiation and tumorigenic abilities.

1.3.2 Cancer Stem Cells in the Human and Mouse Prostate
To accomplish my research on the potential Cancer Cell-EMT-CSC lineage of prostate
CSCs, it was necessary to define a prostate tumor epithelial subpopulation with stem cell-
like properties and successfully isolate and experimentally assay these cells. As murine
prostate specific markers differ from those described in humans, the classification of
murine-specific putative normal prostate stem cell and putative prostate cancer stem cell
populations has been critical. The prostate is composed of three epithelial cell types:
luminal cells, the differentiated secretory cells that are the major cell type;
neuroendocrine cells, rare and morphologically heterogeneous cells present in both the
basal and luminal cell layers; and basal cells, the undifferentiated, nonsecretory cells that
may contain the progenitor cells of the prostate (Nikitin et al., 2007; Nikitin, Nafus,
Zhou, Liao, & Roy-Burman, 2009). Basal cells are identifiable by their expression of
cytokeratins (CK) 5 and 14 and p63 (a member of the p53 family). Luminal cells express
CK8 and CK18 and a heterogeneous population of transit-amplifying cells display co-
expression of CK5, CK14, CK8 and CK18. Neuroendocrine cells are rare and are
distinguished by their expression of chromogranin A and synaptophysin, and absence of
AR expression (Figure 1.2).
16

Figure 1.2 Prostate stem cell lineages. Supportive evidence indicates that prostate stem cells
reside in the basal epithelial cell layer, and can be distinguished from differentiated progeny by
high cell surface expression levels of Sca-1 and α6-integrin (CD49f), as well as intracellular
expression of p63. Basal cells give rise to more terminally differentiated progeny. Transit-
amplifying cells are an intermediate cell type between basal and terminally differentiated cells.
Secretory luminal cells and rare neuroendocrine cells are the differentiated cells of the prostate
epithelia. Transit-amplifying and luminal cells are characterized by expression of cytokeratin 8
(CK8). Adapted from (Nikitin et al., 2007).

The proximal region has been proposed to be the probable location of the prostate stem
cell niche because it contains a high number of label-retaining cells, which express stem
cell specific markers, are resistant to androgen ablation and have a greater capacity to
regenerate prostate tissue in growth assays as compared to cells located in the distal
regions (Burger et al., 2005; Goto et al., 2006; Tsujimura et al., 2002). There is increasing
evidence that cell markers such as Sca-1 (stem cell antigen-1), laminin receptor α6-
integrin (CD49f), CD133 (prominin), CD44 and CD117 (c-kit, stem cell factor receptor)
can be used to enrich for stem cells of the mouse prostate (Lawson et al., 2005; Leong,
Wang, Johnson, & Gao, 2008). Exciting new evidence is that a single mouse prostate
stem cell defined by a Sca-1
+
CD133
+
CD44
+
CD117
+
phenotype and grafted under the
Prostate
Stem Cell

Sca-1
α6-integrin
CD133
Notch1
Bcl-2

p63
CK5
CK14
Basal
Cell
Sca-
α6-integrin

p63
Luminal Cell
Synaptophysin
Chromogranin A
CK8
CK18
AR
p63
CK5
CK8
CK19
(AR)
Transit
Amplifying
Cell
Neuroendocrine Cell

Differentiation
17

renal capsule can generate secretion-producing prostatic ducts consisting of basal,
luminal and neuroendocrine cells (Leong et al., 2008). Following our previous
publications and the publications of other researchers using the cPten
-/-
L mouse model for
prostate adenocarcinoma, we support the Lin
-
Sca-1
+
CD49f
+
subpopulation to be enriched
for cells with stem cell-like properties in the mouse prostate (Goldstein, Huang, Guo,
Garraway, & Witte, 2010; Liao, Adisetiyo, Liang, & Roy-Burman, 2010a; Mulholland et
al., 2009).

18

1.4 Epithelial-to-Mesenchymal Transition
The epithelial-to-mesenchymal transition (EMT) is a de-differentiation process by which
differentiated epithelial cells may alter their phenotype and, even, lineage specificity and
generate new mesodermal tissues. As epithelial cells undergo EMT they lose their
epithelial phenotype, characterized by desmosomes, adherens junctions and tight cell
junctions and an apical-basolateral axis of polarity, and gain fibroblastic appearance and
motility in conjunction with a mesenchymal protein expression profile (Thiery et al.,
2009). This process has been notably described during embryogenesis, as in gastrulation
in the induction of endodermal and mesodermal fates (Hay, 1995; Perez-Pomares &
Munoz-Chapuli, 2002; Thiery et al., 2009; Thiery & Sleeman, 2006), in which mesoderm
is generated by EMTs to later give rise to multiple tissues, including muscle and bone.
The reverse process, termed mesenchymal-epithelial transition (MET), has also been
demonstrated to play an equally important role in the development of the embryo. For
example, during the later stages of embryogenesis, mesoderm tissue undergoes MET to
give rise to certain epithelial organs, including the kidneys and ovaries (Davies, 1996).

1.4.1 EMT in Development, Wound Healing and Cancer: Similar Programs
During development, sequential rounds of EMT are necessary for proper fate
determination of cell types within an organism and for proper structural patterning of the
internal organs. These rounds of EMT are referred to as primary, secondary and tertiary
and take place as defined events with specific conserved signaling pathways. In primary
and tertiary EMT, as well as in wound healing (a process which requires transient EMT
19

to allow epithelial cells to sever their adhesion to each other and the basement membrane
and gain migratory abilities to the purpose of reconstructing the wound site) the
expression of members of the TGF β superfamily of proteins drive EMT through the
induction of the transcription factors Snail1 and Snail2 (also termed Slug), in vertebrates
(Thiery et al., 2009). The induction of EMT results in the succession of key molecular
events that ultimately allow for physical disassociation of epithelial cells and subsequent
migration away from the site of origin. Transcriptional direction under Snail1 and Snail2,
specifically leads to direct repression of E-Cadherin (an epithelial-specific cell-cell
adherens junction protein) and loss of epithelial markers and cell-cell adhesion, and gain
in expression of RhoB and Vimentin (cytoskeletal proteins) and matrix
metalloproteinases (MMPs), which leads to cytoskeletal changes and degradation of the
basement membrane, respectively, ultimately allowing for mesenchymal-like weak focal
adhesion contacts, leading edge asymmetry and changes in cell motility (Hanahan &
Weinberg, 2011; Kalluri & Weinberg, 2009; Thiery et al., 2009). TGF- β1 is also highly
expressed in tumors. Reciprocal activation by stromal cells and epithelial cells reinforces
TGF- β1 secretion and thereby the EMT program. In the context of cancer, this reciprocal
activation process plays a pivotal role in enabling cancer cells to gain malignant, invasive
and, finally, metastatic traits (Giannoni et al., 2010). This relatively recently accepted
feature of carcinomas and the incidence of EMT has prompted the adoption of the
phraseology: “Tumors: wounds that do not heal” (Dvorak, 1986). TGF- β1 is certainly an
important mediator of EMT in cancer, however, the majority of the signaling pathways
20

implicated in triggering EMT, including MAPK, NF- κB and TGF- β1, converge at the
induction of the Snail genes and repression of E-Cadherin (Thiery et al., 2009).

1.4.2 Associations Between EMT and CSCs and Metastasis
The reported concordance between the acquisition of CSC traits with EMT
transdifferentiation programs in certain model systems of cancers has suggested the
importance of EMT to not only confer physical dissemination and motility to cancer cells
but also the ability to self-renew and establish, at ectopic sites of dissemination, clonal
expansion of cell populations that maintain the ability to self-renew and progeny that lack
self-renewing ability (Al-Hajj et al., 2003; Brabletz, Jung, Spaderna, Hlubek, & Kirchner,
2005; Mani et al., 2008; Morel et al., 2008; A. Singh & Settleman, 2010). Ability to
initiate new tumors and an unlimited number of progeny are a critical step in metastasis
and contain the features of self-renewal and multi-potential differentiation ascribed to
normal stem cells. Indeed, EMT profiles in clinical pathologies are known to correlate
with higher histological grades, more aggressive tumor subtypes, such as basal-like breast
carcinoma (Sarrio et al., 2008), poorer disease prognosis and worse outcomes for patient
survival (Thiery et al., 2009).

Following these reported associations between CSCs and EMT, recent discoveries in
EMT and MET have become of great importance to cancer research and, more
specifically, to the process that leads a tumor toward metastasis. Additional to clinical
pathological findings, many important and pleiotropically acting transcription factors and
21

chemokines, including Snail, Twist, Zeb, VEGF and TGF- β1, have been found to be
activated during EMT and in cancers and confer many phenotypic traits that are common
to malignant cancers and have been implicated as promoters of invasiveness. Their
overall effect on neoplastic cells includes increased motility and extracellular matrix
remodeling capabilities, resistance to apoptosis and resistance to chemotherapy (Cheng et
al., 2007; Comijn et al., 2001; Hartwell et al., 2006; Huber et al., 2005; Mani et al., 2007;
Oft, Akhurst, & Balmain, 2002; Peinado, Olmeda, & Cano, 2007; Savagner et al., 2005;
J. Yang et al., 2004). The process of tumor metastasis, which has been shown to be
enabled by cancer cells undergoing EMT in breast and colon cancers (Bates & Mercurio,
2003; Mani et al., 2008; Santisteban et al., 2009), has also raised the possibility that such
disseminated cancer cells may have gained the ability to self-renew, allowing them to
survive in the circulation and reconstitute a tumor resembling the original source in a
secondary tissue site.

22

1.5 Tumor Microenvironment
As mentioned in the preceding section for their role in facilitating EMT, stromal
compartment cell types secrete growth factors that serve to enhance the tumorigenicity of
the surrounding neoplastic epithelia. Instrumental to the influence of this compartment of
the tumor microenvironment are stromal fibroblasts, which have been reported to secrete
various soluble factors, such as growth factors and inflammatory cytokines, produce
extracellular matrix proteins and MMPs and proteins that regulate and promote
angiogenesis (De Wever & Mareel, 2003; Kalluri & Zeisberg, 2006; Liotta & Kohn,
2001; Silzle, Randolph, Kreutz, & Kunz-Schughart, 2004).

Reflecting on the discussion regarding conferral of metastatic abilities on cancer cells
through EMT and the role of the tumor stroma in promoting EMT, some model systems
of human carcinomas have reported the requirement of stromal components for
metastatic dissemination of tumor cells. Necessary presence of activated fibroblasts in
prostate and breast carcinoma stimulate growth, motility and invasive behaviors of tumor
cells (Chung, Baseman, Assikis, & Zhau, 2005; Kaminski et al., 2006; Studebaker et al.,
2008).

As the field continues to expand, new key mediators of tumorigenicity are being
discovered and characterized, with recent emphasis on juxtacrine/paracrine products of
the tumor microenvironment that are capable of influencing the neoplastic epithelia. I
now introduce one particular member of the Annexin family of proteins, AnxA1, which I
23

will later emphasize from my experimental results as an important fibroblast-secreted,
soluble factor in induction of EMT and conferral of CSC traits to the cancer epithelial
cells of the murine cPten
-/-
L model of prostate adenocarcinoma.
24

1.6 Annexin A1 and Main Pathobiological Roles
Annexins comprise a family of proteins that bind to specific phospholipids in a calcium-
dependent manner (Crompton, Moss, & Crumpton, 1988; Crumpton & Dedman, 1990).
Members of this protein family are structurally-related through the presence of a
conserved C-terminal core domain comprised of a 70 amino acid motif repeated into four
α-helices (Gerke & Moss, 2002). The calcium-binding and phospholipid-binding
sequences are encoded for in the conserved C-terminus (40-60% homology), whereas an
N-terminus of varying length and sequence is unique to each family member and confers
the specific biological activity of individual annexins (Raynal & Pollard, 1994). Annexin
A1, a 37kDa protein, was the first characterized member of the family (previously known
as macrocortin (Blackwell et al., 1980) and lipocortin-1 (Miele, Cordella-Miele,
Facchiano, & Mukherjee, 1988)) and was originally reported for its anti-phospolipase
activity after glucocorticoid induction (Blackwell et al., 1980; Miele et al., 1988).
Glucocorticoids (GC), also known as corticosteroids, are powerful anti-inflammatory
prescription drugs, and AnxA1 has been shown to be a direct mediator of the effects of
glucocorticoid-inducible resolution of inflammation (Perretti & Flower, 1996). Numerous
additional studies show that both recombinant AnxA1 and AnxA1-derived N-terminal
peptides possess a wide range of anti-inflammatory properties, which include inhibition
of phospholipase A2 (PLA
2
), cyclooxygenase-2 (Cox-2) and nitric oxide, stimulation of
anti-inflammatory cytokine interleukin-10 (IL-10), activation of formyl peptide (FPR)
and lipoxin A
4
receptors to synergistically inhibit neutrophil and monocyte
transmigration, inhibition of leukocyte adhesion, induction of apoptosis and clearance of
25

inflammatory cells (Gavins, Yona, Kamal, Flower, & Perretti, 2003; Getting, Flower, &
Perretti, 1997; Parente & Solito, 2004).

1.6.1 Cytosolic Location and Modification, Secretion
Consistent with its role in the resolution of inflammation, AnxA1 is found in high
abundance in lung, bone marrow, intestine, lymphatic tissue and reproductive tracts, with
its highest concentration being found in the seminal fluid of the prostate (150 μg/ml)
(Christmas, Callaway, Fallon, Jones, & Haigler, 1991). Both the conformation and
biological activity of the protein may be further modified during post-translational
processing by phosphorylation of specific serine, tyrosine and threonine residues,
glycosylation, acetylation and lipidation. In addition to the regulatory region, the N-
terminus also contains the sites for phosphorylation and proteolysis (Raynal & Pollard,
1994). Various signal transducing kinases, such as EGF receptor tyrosine kinase (EGFR
TK), platelet-derived growth factor receptor kinase (PDGFR-TK), hepatocyte growth
factor receptor kinase (HGFR TK), TRPM7 channel kinase and protein kinase C (PKC)
are known to phosphorylate AnxA1 and contribute to its roles in proliferation. Depending
upon the kinase for which AnxA1 is the target, AnxA1 may act as either a promoter or
suppressor of proliferation (Hsiang, Tunoda, Whang, Tyson, & Ornstein, 2006; Skouteris
& Schroder, 1996).

AnxA1 does not possess a recognized signal sequence for targeting to the endoplasmic
reticulum and processing through the classic secretory pathway; however, its high
26

concentration in seminal fluid of the human prostate gland originally evidenced a
selective mechanism of secretion (Christmas et al., 1991). It is now understood that
AnxA1 follows cell-specific novel manners of secretion from many cell types and is
externalized from the cell during cell activation, such as by neutrophil adhesion to
endothelial cell monolayers, or glucocorticoid stimulation (D'Acquisto, Perretti, &
Flower, 2008). In neutrophils, gelatinase granules store high levels (>60%) of
cytoplasmic AnxA1 for extrusion upon activation (Perretti et al., 1996). Cells that do not
store AnxA1 in granules display another distinct secretory pathway. In macrophages and
pituitary folliculo-stellate cells, AnxA1 is exported by the ATP-binding cassette A1
(ABC-A1) transporter system or ATP-sensitive K
+
channels (Chapman, Epton,
Buckingham, Morris, & Christian, 2003; Morris et al., 2002; Payne, Morris, Solito, &
Buckingham, 2005; Wein et al., 2004). In both leukocytes and pituitary cells
phosphorylation on Ser
27
is necessary for protein export and secretion. This action is
directed by Ca
2+
-dependent isoforms of PKC, whose activity is stimulated by
glucocorticoids. Subsequent translocation of the serine
27
-phosphorylated species of
AnxA1 to the plasma membrane occurs at specific lipid domains that allow for secretion,
and is dependent on phosphatidylinositol 3-kinase (PI3K) and MAPK (Croxtall,
Choudhury, & Flower, 2000; John et al., 2003; Solito et al., 2006; Solito et al., 2003;
Yazid et al., 2010). Extracellular Ser
27
-AnxA1 undergoes a conformational change in the
presence of ≥ 1 mM Ca
2+
causing exposure of the N-terminal domain from inside the pore
created by the four repeated motifs of the core domain and, thereby, binding to its
receptor at the cell surface (Rosengarth & Luecke, 2003; Solito et al., 2006). Binding of
27

AnxA1 to the receptor activates downstream signaling via phosphorylation of the
mitogen-associated protein kinases (MAPK) Erk1/2 and, to lesser extent p38 and
Jnk/Sapk, and this ligand-receptor interaction may occur in a juxtacrine, autocrine or
paracrine manner depending on the producing cells (in inflammatory conditions,
juxtacrine interaction with the receptor is most plausible) (Chatterjee et al., 2005; Hayhoe
et al., 2006; Lange, Starrett, Goetsch, Gerke, & Rescher, 2007; Perretti & D'Acquisto,
2009; Tagoe et al., 2008).

1.6.2 Interactions with Receptor
Through its specific N-terminal sequence, AnxA1 directly interacts with a family of G-
protein coupled receptors, the formyl peptide receptors (FPRs), which includes FPR1,
FPR2 (also known as ALXR, which is the recently-renamed FPR-like 1) and FPR3
(formerly FPR-like 2) in humans and Fpr1 and Fpr-related proteins (Fpr-rs) 1-7 in mouse
(Ernst et al., 2004; Perretti et al., 2002). Endogenous and recombinant full-length AnxA1
specifically bind to human FPR2/ALXR and murine Fpr-rs1 (and, in a selected study,
have demonstrated up to ~30% activity through murine Fpr1), whereas the bioactive N-
terminal-derived peptide comprised of amino acids 2-26, Ac2-26, has been shown to
activate all members of the human FPR family and the murine Fpr1 and Fpr-rs1, and may
additionally activate Fpr-rs2 (D'Acquisto, Perretti et al., 2008; Ernst et al., 2004; Gavins
et al., 2003; Hayhoe et al., 2006; John, Gavins, Buss, Cover, & Buckingham, 2008). The
members of this family were first studied for their effects on human neutrophil and
monocyte trafficking (Schiffmann, Corcoran, & Wahl, 1975) and have since become
28

understood to encompass a more diverse array of biological roles due to their promiscuity
in ligand binding, allowing for both N-formyl-peptides (e.g. fMLF) and non-formylated
peptides (e.g. AnxA1 and lipoxin A
4
) (Chiang et al., 2006), and expression in non-
hematopoietic cells including endocrine cells of the thyroid and adrenal cortex, Kupffer
cells, smooth muscle cells and epithelial cells (John et al., 2008). In the mouse, the
divergence and expansion of this receptor family is less well-understood; however, it
does seem evident that Fpr1 represents the murine equivalent of human FPR (Le,
Murphy, & Wang, 2002) and that Fpr-rs1 and Fpr-rs2 are orthologous to human
FPR2/ALXR (Gao, Chen, Filie, Kozak, & Murphy, 1998).

1.6.3 Differential Effects on Cancer in Various Tissues
Deregulation of AnxA1 may lead to a number of human diseases as reported in Fragile X
Syndrome (H. T. Sun, Cohen, & Kaufmann, 2001) and cystic fibrosis (Tsao, Meyer,
Chen, Rosenthal, & Hu, 1998) where AnxA1 metabolism has been shown to be defective,
as well as in autoimmune disorders including Crohn’s disease and systemic lupus
erythematosus which present with autoantibodies to AnxA1 (Podgorski, Goulding, Hall,
Flower, & Maddison, 1992; Stevens, Smith, & Rampton, 1993). Clear observations
indicate that AnxA1 could also serve as an important diagnostic marker for a variety of
tumors and indicate that AnxA1 may have regulatory roles in the progression of cancer.
AnxA1 protein levels are increased in pituitary, breast, pancreatic and bladder cancers
and reduced in esophageal, head and neck and prostate cancers, however, even among the
same tissue types, AnxA1 has been found to have differential expression in certain
29

cancers (L. H. Lim & Pervaiz, 2007). For example, high AnxA1 expression is correlated
with poor prognosis in noninvasive ductal carcinoma and carcinoma in situ (Ahn,
Sawada, Ro, & Nicolson, 1997), but loss of expression is demonstrated in invasive ductal
carcinoma and ductal carcinoma in situ (Shen et al., 2006), and metastatic breast tumors
have been reported with either elevated levels or loss of expression of AnxA1 (Ahn et al.,
1997; Shen et al., 2005). Conflicting reports of AnxA1 effects on proliferation and
metastasis-promoting EMT processes in breast cancer has also been reported in different
studies using human breast tumor cell lines (de Graauw et al., 2010; Khau et al., 2011;
Maschler et al., 2010). One possible explanation for the differential expression of AnxA1
in human breast cancers is the status of the estrogen receptor, although this still remains
to be further studied (Cao et al., 2008). Another hormonally regulated organ where
overexpression of AnxA1 is commonly revealed is pituitary, in which AnxA1 secretion
plays a crucial pathophysiological role in the release of ACTH and other pituitary
hormones under normal biological circumstances (Taylor, Christian, Morris, Flower, &
Buckingham, 1997; Taylor, Cowell, Flower, & Buckingham, 1995).

1.6.4 Understanding the Role of AnxA1 in Prostate Cancer
Interestingly, in androgen-responsive prostate cancers, AnxA1 loss of expression from
the ductal epithelial cells has been well-documented (Kang et al., 2002; Patton, Chen,
Joseph, & Yang, 2005; Paweletz et al., 2000; Smitherman, Mohler, Maygarden, &
Ornstein, 2004; W. Xin, Rhodes, Ingold, Chinnaiyan, & Rubin, 2003). This is in contrast
to AnxA1 expression status in other hormone responsive tumor types (e.g. breast,
30

pituitary). Prostatic stromal cells also display expression of AnxA1 (Christmas et al.,
1991) and, to date, studies have not yet focused on the role of stromal-produced AnxA1
or whether AnxA1 protein levels in this compartment of the prostate gland are altered
during tumorigenesis or in response to changes in androgren levels (e.g. in hormone
refractory castration-resistant prostate cancer (CRPC)). In ductal carcinoma in situ and
invasive carcinoma breast tumors, stromal AnxA1 expression was positively correlated
with infiltration of both epithelial and stromal cells (Khau et al., 2011).
31

1.7 Significance
As most of this research is preliminary within the field of prostate cancer, clear and
significant associations between induction by EMT and, first, the generation of a
mesenchymal phenotype with a more stem-like phenotype and, second, establishment of
functionally and molecularly defined prostate cancer stem cells must be drawn before the
mechanism by which EMT enriches for putative CSCs may be studied. Research findings
confirming EMT as an active process in prostate cancer progression to metastatic cancer
and, furthermore, connecting the tumor-initiating population within prostatic neoplasia
with cells having undergone an EMT will be invaluable in understanding and preventing
the occurrence of metastatic prostate cancer. As the fields addressing CSCs, EMT and
prostate cancer continue to expand, identification of inducers of EMT in vivo and
definition of the signaling pathways involved may be envisioned to have clear clinical
relevance. Delineating the potential role of EMT in the generation of prostate CSCs and
metastatic cancer will be to the benefit of identifying potential therapeutic targets with
the aim of preventing the spread of metastatic cancer by blocking an implicated process
by which cancer cells may gain metastatic malignancy within the body.

An important step in understanding these biological processes in cancer and finding clear
therapeutic targets for better disease management is the discovery of relevant mediators
of tumorigenic biological processes, including the induction of EMT and acquisition of
stem cell traits by cancer cells. An emergent critical factor has been discovered to be
AnxA1. CAF-secreted AnxA1 in the prostate tumor microenvironment is a novel topic.
32

The discovery and characterization of CAF-secreted AnxA1 also supports the primary
hypothesis (Section 1.2.2) that CAF-secreted factors influence and/or maintain the CSC
niche. My dissertation depicts modes by which cancer cells exposed to AnxA1 from
CAFs gain stem cell and EMT properties consistent with enhanced tumorigenicity. These
properties of CSCs and EMT are discussed in Section 1.3.1. Responses to AnxA1 are
also driven by cancer cell heterogeneity and emphasize the susceptibility of certain
cancer cell subpopulations to develop malignant traits.

Owing to the complexity of AnxA1 regulation in various tissues and even in the same
tissue type within human cancer, further elucidation of the role of AnxA1 in the
development and maintenance of cancer remains to be studied. Of particular importance
is the role of AnxA1 in prostate cancer progression from primary to CRPC. Certainly it is
an important molecule in the normal prostate as evidenced by its abundance, greater than
that found elsewhere in the human body, and changes in its expression should have a
profound impact on the pathobiology of neoplastic prostate cells. We have previously
described evidence that CAFs derived from the stromal compartment of prostate tumors
secrete factors that enhance both the stemness and growth potentials of the CSCs in
primary prostate cancer using a conditional Pten knockout mouse model of prostatic
adenocarcinoma (Liao, Adisetiyo et al., 2010a). I have now identified AnxA1 as one of
the pertinent secreted factors. CAF conditioned medium- (CM) derived full-length
AnxA1, murine recombinant full-length AnxA1 (rAnxA1) and N-terminal mimetic
peptide Ac2-26 treatment of both primary epithelial tumor cells and cE1 cell line led to
33

up-regulation of EMT transcription factors and stem cell transcription factors in cE1 in
vitro and enhanced glandular structure formation of primary cells and cE1 cells in vivo.
34

Chapter 2:
Mouse Cell Culture Model for EMT Induction by Secreted
CAF Factors
2.1 Abstract
Fibroblastic stromal cells from prostate tumors (CAFs) have been shown to enhance the
tumorigenic potential of CSCs. Accordingly, signaling factors from the tumor
microenvironment may play an important role in the progression of carcinoma, and, as
recent findings indicate, in the induction of EMT and stem-like properties. Epithelial cell
lines derived from the cPten
-/-
L mouse model of prostate adenocarcinoma could be used
to evaluate and characterize factors that can support or accelerate EMT induction to
cancer epithelial cells, and further define putative factors that may facilitate de-
differentiation to CSC phenotypes and biology. To establish such a model system, (a)
examination of the extent to which prostate cancer cells induced to undergo EMT in vitro
exhibit stem cell characteristics and (b) identification of signaling molecules and
associated mechanisms involved between CAFs and cancer cells that can induce EMT
and contribute to the generation of CSCs were necessary.

We found that a malignant epithelial cell line (cE1) derived from the recurrent,
castration-resistant prostate cancer (CRPC) tumor of the model could be further
fractionated into three different subpopulations based on the high, medium and low levels
of cell surface expressions of Sca-1 (S) and CD49f (C) antigens: SC
hi
, SC
med
and SC
none
,
35

respectively. Of these fractions, the SC
med
population was relatively more sensitive to
undergo EMT-like changes when exposed to TGF-β1 than the other fractions. SC
hi

fraction displayed increased ability to form spheroids in 3-D cultures akin to the CSCs
isolated from the prostate tumors of the model. SC
none
cells appeared to represent the bulk
of the cells with reduced propensity for EMT or spheroid formation. When SC
med
cells
were treated with conditioned medium (CM) from either CAF cultures or normal prostate
fibroblast (NPF) cultures, extensive EMT-like changes were induced by CM from CAFs
and much less by that from NPFs. An ammonium sulfate (AS) precipitation technique for
enrichment of CAF CM derived proteins and the development of in vitro 3-D biological
assays for protein activity using the EMT-sensitive fraction of the cE1 cell line, SC
med
,
were designed to create an assayable system for identifying EMT-inducing CAF secreted
factors. Characterization of the cancer cells following induction with CAF CM and AS-
enriched CAF CM protein fractions revealed an EMT transcriptional profile, marked by
high expression of EMT transcription factors (Snail, Slug and Twist) and stem cell
transcription factors (Oct4, Sox2 and Nanog) and reduced expression of epithelial
cadherins junction protein E-Cadherin, that enabled identification of AnxA1 as a
potential CAF mediator of Cancer Cell-EMT-CSC lineage.
36

2.2 Introduction
Interactions with the tumor microenvironment have been shown to play an important role
in the progression of carcinoma and in the induction of EMT (Giannoni et al., 2010).
Recently we have shown that secreted factors from one of the main components of the
stromal compartment of the tumor microenvironment, the CAFs, can enhance the stem
cell properties of the putative CSCs in vivo (Liao, Adisetiyo et al., 2010a). Therefore, it
follows to determine if whether CAFs, through the secretion of signaling factors that
support the induction of an EMT, may act to generate a lineage of EMT-CSC cancer cells
that are capable of self-renewal and multi-potential differentiation. Identification and
characterization of these pro-tumorigenic fibroblastic and cancer interactions may shed
further light on our understanding of the progression of prostate cancer and lend new
ideas for approaching the treatment of prostate cancer, most significantly the recurrent
phase for which there is currently no cure.

2.2.1 Cell Lines Derived from the cPten
-/-
L Mouse Model of Prostate
Adenocarcinoma
Several mouse model cell lines are available from both primary, androgen dependent
(AD), prostate cancer cells and recurrent, castration-resistant, prostate cancer cells.
Characterization of the AD cell lines, E2 and E4 and the CRPC, cell lines cE1 and cE2
was co-commitment with my experiments to establish an ideal epithelial model system
for studying EMT in prostate cancer. Thus, the published characterization of the AD and
37

CRPC cell lines and my analyses from 2-D monolayer and 3-D Matrigel assays for a cell
line as a model system were mutually beneficial (Liao, Liang et al., 2010).

2.2.2 Known Inducers of EMT from the Tumor Microenvironment
Induction of EMT within prostate cancer cells will most promisingly follow after recently
published findings in EMT in both breast and colon cancer research (Bates & Mercurio,
2003; Santisteban et al., 2009). In these model systems, the role of cellular components of
the immune system as contributors to EMT has begun to shed light on some of the
signaling factors that may be involved in the acceleration of malignant cancer formation
and metastasis in vivo. Both tumor necrosis factor- α (TNF α) and TGF- β1 have been
characterized as potent inducers of EMT (Bates & Mercurio, 2003; Bhowmick et al.,
2001; Ellenrieder et al., 2001; Fujimoto, Sheng, Shao, & Beauchamp, 2001; Lehmann et
al., 2000; Oft, Heider, & Beug, 1998; Portella et al., 1998). TGF- β1 has been found to be
abundantly expressed in many epithelial tumors and in the surrounding stroma, acting as
an active stimulator of growth and invasion during tumor progression (Hsu, Huang,
Hafez, Winawer, & Friedman, 1994). Another feature of the stroma surrounding many
solid tumors is the seeming recruitment and proliferation of macrophages, which have
been shown to produce growth factors that promote angiogenesis, as demonstrated in
both human breast and colon cancers (Etoh, Shibuta, Barnard, Kitano, & Mori, 2000;
Leek et al., 1996; Sunderkotter, Steinbrink, Goebeler, Bhardwaj, & Sorg, 1994). In a
recent study using an EMT model of colon carcinoma, TNF α produced by activated
macrophages was demonstrated to significantly accelerate TGF- β-induced EMT and
38

stimulate a rapid burst of ERK activation and increased MAPK activity (Bates &
Mercurio, 2003).

Physiologically, it is relevant to assess the role of the pro-inflammatory cytokines TGF-
β1 and TNF α on EMT induction in the CRPC prostate cancer cell lines. In in vivo
conditions, the secretion of TGF- β1 is elevated during the progression from early PIN to
progressed adenocarcinoma as the surrounding stromal cells upregulate their production
of this cytokine (Oft et al., 1996; Y. A. Yang et al., 2002). The inflammatory components
of the tumor microenvironment—the hematopoietic system derived leukocyte population,
which includes macrophages—are the major producers of TNF α. Notably, TNF α,
secreted by infiltrating macrophages, in concert with the production of TGF- β1, from
within the stromal fibroblast compartment that is adjacent to the neoplastic epithelia,
serve to further stimulate the pro-tumorigenic role that cytokines play during later tumor
progression. This occurs in several ways. The neoplastic cells and tumor-associated
fibroblasts and macrophages produce an array of cytokines and chemokines, including
large amounts of TGF- β1 and TNF α, which are chemoattractants for infiltrating
granulocytes, mast cells, macrophages, fibroblasts and endothelial cells, all of which
additionally secrete proteolytic enzymes, chemokines and mitogens. This mitogenic
potential of the infiltrating tumor microenvironment helps to stimulate angiogenesis,
induce fibroblast migration and enable tumor growth via mitogenesis or tumor spread via
metastasis (Coussens & Werb, 2002; Etoh et al., 2000; Hsu et al., 1994; Leek et al., 1996;
Sunderkotter et al., 1994). While other cytokines are produced by the tumor
39

microenvironment, TGF- β1 and TNF α have been implicated as important pro-
inflammatory promoters of tumor growth as both regulate a cascade of other downstream
cytokines, matrix metalloproteinases (MMPs) and pro-angiogenic factors (Balkwill,
2002; Bhowmick et al., 2001; Coussens & Werb, 2002; Torisu et al., 2000). Recent
evidence has further linked these two cytokines to EMT, with TGF- β1 as a key inducer
and TNF α serving as a TGF- β-mediated accelerator of EMT (Bates & Mercurio, 2003;
Bhowmick et al., 2001; Ellenrieder et al., 2001; Fujimoto et al., 2001; Lehmann et al.,
2000; Oft et al., 1998; Portella et al., 1998). Co-treatment using both cytokines TGF- β1
and TNF α has previously been shown to induce EMT in human colonic organoids in as
little as four days, as compared to treatment with TGF- β1 alone, which requires a 7 day
time span to induce EMT (Bates & Mercurio, 2003).
40

2.3 Results
2.3.1 Validation of the cE1 Model for EMT
Figure 2.1 Prostaspheres in 3D culture. Cells from two of the cell lines (E2 and cE1; E4 not
shown) derived from the cPten
-/-
L model were grown in Matrigel for 14 days and assayed for
their ability to grow spheroids in culture. Brightfield images were taken at 100x magnification.

Cell lines, E2 and E4, and cE1, derived from the primary and recurrent epithelial tumors
of the cPten
-/-
L mouse model of prostate adenocarcinoma, respectively, were used to
validate a murine cancer cell line model of EMT (Figure 2.1). Epithelial cells from the
cell lines were grown in vitro in Matrigel to assay for the ability of prostate cancer cells
to undergo EMT based on induction by signaling factors from the tumor
microenvironment.

Initial optimization experiments of Matrigel growth factor concentration and potency of
culturing media revealed variation between E2 and E4, and cE1. cE1 were capable of
forming spheroids in Matrigel that could be followed for up to 14 days in culture. E2 and
E4 did not maintain spheroids, and growth observed in culture after 14 days resembled

cE1 recurrent prostate cancer
mouse cell line, 100x
E2 primary prostate tumor
mouse cell line, 100x
41

that of monolayer fibroblast cultures. In addition, whereas cE1 did not show readiness to
undergo EMT-like morphological or molecular changes in the absence of treatment, both
E2 and E4 rapidly underwent proliferation and apparent EMT due to high Matrigel
growth factor concentration (>10 mg/ml) or higher percentage of fetal bovine serum
(FBS; >2%) in the culturing media as assessed by growth rate, filamentous outgrowths of
spheroids and high expression of EMT transcription factors and mesenchymal markers
(Figure 2.2).

Figure 2.2 E4 cells show
high expression of mesenchymal markers (Vimentin and N-Cadherin) and EMT-associated
transcription factors (Snail, Slug and Twist). cE1 cells have higher expression of epithelial
marker, E-Cadherin, and low expression levels of mesenchymal markers and EMT transcription
factors.

Our previous publications have shown that PC-3 human prostate cancer cell lines and
primary murine prostate cancer cells from the cPten
-/-
L model exposed to CAF secreted
factors could display phenotypic morphological changes consistent with EMT induction,
in vitro. As a final confirmation that cE1 cells would be the most relevant cell line for
Transcriptional profiles of E4, primary, and cE1, recurrent, cell lines.
0.0
0.5
1.0
1.5
2.0
2.5
Snail E-Cadherin N-Cadherin Vimentin Slug Twist
Normalized Mean Ratio to β-actin
E4
cE1
42

43

proof-of-principle experiments to study EMT in our mouse model system in vitro, cE1
cells were co-cultured with NPFs or CAFs in 3-D Matrigel assays and monitored over a
period of 11-14 days for similar EMT-like morphological changes as those previously
described in our human prostate cancer cell line and murine primary prostate cancer cell
systems (Lim, Chuong, et al. 2011; Liao, Adisetiyo, et al. 2010). In these co-culture
experiments, greater EMT phenotypic changes were observed in cE1 cells co-cultured
with CAFs than with NPFs or castration-resistant prostate cancer (CRPC)-CAFs (Figure
2.3).
Figure 2.3
s or
CRPC-CAFs, which form spheroids or combination of spheroids and monolayer sheets of cells
with epithelial morphology, respectively. Brightfield images were taken at 100x magnification.

cE1 3-D co-culture assay with fibroblasts of the model. cE1 cells co-cultured with
CAFs display marked mesenchymal morphology compared to cE1 cells co-cultured with NPF
cE1 + CRPC-CAF, Day 11
cE1 + CAF, Day 11
cE1 + NPF, Day 11
cE1 + CAF, Day 3
cE1 + NPF, Day 3
cE1 + CRPC-CAF, Day 3

Primary cultures of NPF, CAF and CRPC-CAF, derived from the cPten
-/-
L mouse model
of prostate adenocarcinoma, were used for both co-culture and CM experiments.

2.3.2 In vitro TGF β/TNF α-mediated EMT experiments using murine cell lines
derived from the cPten
-/-
L model.
High expression of TGF- β1 in the tumor microenvironment has been well reported (Oft
et al., 1996; Y. A. Yang et al., 2002). To ascertain if the effects of CAF on the tumor
epithelial cells were due to this molecule, I was interested to treat cE1 cells with TGF- β1
and other known associated proteins, i.e. TNF α (Bates & Mercurio, 2003). I tested
whether stimulation by TNF α could accelerate TGF- β1-mediated EMT in prostaspheres
from mouse prostate cancer cell lines, similar to the results shown by Bates, et al., 2003,
in colon carcinoma cells. In order to demonstrate that cells exposed to physiologically
elevated levels of these cytokines undergo EMT, as would be expected under tumorigenic
conditions, I assayed for the response of cE1 cells in 3-D Matrigel culture to treatment
with 10 ng TGF- β1 and 10 ng TNF α alone and in combination (Figure 2.4). cE1 cells
appeared to display more morphological characteristics of EMT in response to TGF- β1
and even more so by co-stimulation with TGF- β1 and TNF α. However, these changes
were not strikingly significant under observation by microscopy. Assessment was
accomplished by examining morphologic changes associated with EMT. In this model
system, filamentous outgrowths of spheroids, followed by loss of spheroid cohesion and
resulting monolayer growth of cells with mesenchymal phenotype were taken as an
indication of EMT. The epithelial phenotype that is characterized by tight cell-cell
44

junctions and basement-membrane polarity is essential for cells to maintain 3-D spheroid
formation. This phenotype was replaced during supposed EMT by a mesenchymal
phenotype, in which tight cell-cell interactions were lost, cells gained increased motility
and diminished polarity and rapidly diffused from spheroid structures.
Figure 2.4 cE1 3-D assay with TGF- β1 and TNF α for EMT induction. cE1 cells were grown in
Matrigel for a short period of 6 days to assay for rapid induction of EMT with the synergistic
cytokines TGF- β1 and TNF α. More EMT-like morphology was observed in cE1 cells co-treated
with TGF- β1 and TNF α after 6 days than with either cytokine alone. Brightfield images were
taken at 100x magnification.

To conduct a more physiologically relevant test of the effects of TGF- β1 and TNF α in the
tumor microenvironment on the prostate cancer epithelial cells, I added CAF CM to each

Day 1, 100x
Day 1, 100x
Day 6, 100x
Treatment with 10ngTGF β
Treatment with 10ngTNF α
Day 6, 100x
Day 6, 100x
Treatment with 10ngTGF β + 10ngTNF α
Day 1, 100x
45

treatment group (Figure 2.5). With the addition of CAF CM, and therefore additional
secreted factors of the tumor microenvironment, changes consistent with an induction
through EMT appeared prominent by brightfield observation. Based upon these
observations the addition of CAF CM and its constituent secreted factors appeared to
have a greater affect on the induction of EMT than TGF- β1 alone, or TGF- β1 plus TNF α.

cE1 + CAF CM cE1 + CAF CM + TGF 
cE1 + CAF CM + TNF α cE1 + CAF CM + TGF  + TNF α
50 um

Figure 2.5 Acceleration of CAF CM induced EMT by TGF- β1 and TNF α. Synergistic effects of
treatment with both TGF- β1 and TNF α on cE1 cells grown in 3-D Matrigel assays in the
presence of CAF CM. After 7 days of treatment, the most extensive growth, EMT morphology and
reduction of spheroid growth was observed in cE1 cells co-treated with TGF- β1 and TNFα in the
presence of CAF CM. Brightfield images were taken at 40x magnification.

The next step was determined to be the isolation of potent CAF-secreted EMT-inducing
protein(s). To set a standard for biological function, transcriptional changes of markers
46

associated with EMT after treatment by both TGF- β1 and CAF CM were assayed. Rather
than assaying for changes to the heterogeneous total cE1 cell population, I used
fluorescence-activated cell sorting (FACS) to isolate more homogeneous subpopulations
of cE1 cells based on cell surface marker expression levels of Sca-1 (S) and CD49f (C),
and thereby capture more prolific changes within homogeneous cell groups in response to
treatment. After FACS, sorted cE1 SC
hi
, SC
med
and SC
none
subpopulations were cultured
in tissue culture treated Petri plates and observed for phenotypic differences (Figure 2.6).
Figure 2.6 Culture of FACS-isolated SC , SC and SC cE1 cells. cE1 cells were isolated
based on cell surface expression levels of Sca-1 and CD49f by FACS. Post-isolation, SC
hi
, SC
med

and SC
none
subpopulations were grown in 2-D for clonogenicity. SC
hi
, SC
med
and SC
none
were
observed to grow at different rates with distinctly different morphologies. Brightfield images were
taken at 100x magnification.

hi med none
SC
hi
SC
med
SC
med
SC
none
2D Culture, P0 100x
47

48

Next I assayed for EMT in subpopulations of the cE1 cell line, SC
hi
and SC
med
, in
response to CAF CM and TGF- β1. The changes occurring in cells undergoing EMT were
additionally examined by real time quantitative reverse transcription polymerase chain
reaction (qRT-PCR) to test for changes in expression levels of known genes associated
with both a loss of epithelial phenotype and a gain of mesenchymal phenotype and
activity. TGF- β1 treated SC
med
cells had the most mesenchymal expression pattern
overall, with slightly higher expression levels of N-Cadherin and Vimentin and lower
expression of E-Cadherin. CAF CM treated SC
med
showed higher expression of EMT
transcription factors Snail and Twist.
Figure 2.7 EMT transcriptional profile of cE1 SC and SC TGF- β1 or CAF CM treated
cells. FACS-sorted cE1 cells, SC
hi
and SC
med
,

were grown in Matrigel for 11 days and assayed for
expression levels of E-Cadherin, N-Cadherin, Vimentin, and EMT transcription factors Snail and
Twist. TGF- β1 treated SC
med
cells had the most mesenchymal expression pattern of low E-
Cadherin, high N-Cadherin and Vimentin, and high Snail and Twist. CAF CM treated SC
med

showed high expression of Snail and Twist, and relatively higher expression of Vimentin
compared to SC
hi
.

hi med
Twist
0.000
0.001
0.002
0.003
TGF β Hi TGF β Med CAF Hi CAF Med
Mean Ratio to b-Actin
Vimentin
0.00
0.05
0.10
0.15
0.20
TGF β Hi TGF β Med CAF Hi CAF Med
M e an R atio to b-A ctin
N-Cadherin
TGF β Med CAF Hi CAF Med
0.00
0.01
0.02
0.03
TGF β Hi
M ean R atio to b -A ctin
E-Cadherin
0
0.03
0.06
0.09
TGF β Hi TGF β Med CAF Hi CAF Med
M e a n R a tio to b -A ctin
Snail
0.0000
0.0001
0.0002
0.0003
TGF β Hi TGF β Med CAF Hi CAF Med
Mean Ratio to b-Actin
Reiteration of the greater EMT-like morphological changes observed in cE1 cells due to
CAFs in fractionated subpopulations of cE1 cells treated with CM rather than co-culture
(Figure 2.3) revealed that (a) certain subpopulations of cancer epithelial cells appear to be
susceptible to EMT by CAF factors and (b) the factors that are capable of inducing EMT
in these subpopulations are present within the CM and do not require the presence of
fibroblast cells in culture in order to influence the epithelial cells (i.e. do not require
reciprocal cell-to-cell interactions for biological effect to be seen) and therefore appear to
be secreted (Figure 2.8).
SC
HI
SC
MED
cE1
cE1 +
CAF CM
cE1 +
NPF CM
cE1
cE1 +
CAF CM
cE1 +
NPF CM

Figure 2.8 Phenotypic changes consistent with EMT in cE1 SC
hi
and SC
med
culture with CAF
CM. The greatest observed amount of EMT-like morphology was seen in SC
med
cE1 cells grown
with CAF CM treatment after 14 days in 3-D Matrigel assays. Representative brightfield images
of the observed morphologies are shown.
49

2.3.4 Identification of Annexin A1 as a CAF-derived EMT-inducing Soluble Factor
in Biological Assays
CM from the CAF, but not NPF, was sufficient to induce morphological changes and
molecular characteristics consistent with induction of EMT in cE1 cells in vitro (Figure
2.8; see Chapter 3 for more detail). From this I concluded that the CAF secrete positive
regulators of EMT and stemness. To identify such potential mediators of EMT from the
CAF, NPF CM and CAF CM were fractionated using several rounds of ammonium
sulfate (AS) precipitation followed by biological assays for protein activity using the
three-dimensional Matrigel culture system on cE1 SC
med
cells. After a 14-day treatment
period, enriched fractions of the CAF CM led to EMT-like morphological changes,
statistically significantly decreased number of spheroids formed (P < 0.05), and up-
regulation of EMT transcription factors and stem cell transcription factors similar to the
experimental results described in further detail in Chapter 3 (Figures 2.9 and 2.10). To
identify the protein of interest, positive fractions from ammonium sulfate purification
were subjected to band separation by gel electrophoresis followed by mass spectrometry
(MS) of selected bands. MS analysis implicated the phospholipid binding protein, AnxA1
(which shall be further discussed in Chapter 3), in the band detectably present in the CAF
CM fractions but not in the NPF CM fractions. Western blot against AnxA1 confirmed its
abundance in the CAF CM fractions compared to levels detected in the NPF CM
fractions, as shown in Chapter 3, Figure 3.4.

50

51

Figure 2.9 Ammonium sulfate precipitation of conditioned media proteins. A, Potential
mediators of EMT-CSC lineage present in the CAF CM have been enriched using several rounds
of ammonium sulfate precipitation followed by in vitro biological assays for protein activity
affecting the EMT-sensitive fraction of the cE1 cell line, SC
med
, using the three-dimensional
Matrigel culture system. B, The AS-enriched fraction from the CAF CM (40%) that had the most
extensive EMT-like morphology after 14 days of culture also had a significantly reduced number
of spheroids (P < 0.05) compared to vehicle control and NPF control. Brightfield images were
taken at 40x magnification.

CAF Flowthrough
60% 50% 40% 30% 20%
CAF CM NPF CM PBS Control CAF Heat Inactive
NPF
CAF
SCMED Spheroids Day 14
0
100
200
300
400
500
No. of Spheroids
PBS Heat Flow
p<0.05
p<0.05
p<0.05
60% 50% 40% 30% 20% CAF NPF H.I.
A
B
Ammonium Sulfate Precipitation of Conditioned Media

PBS
EMT Transcription Factors
0
5
10
15
20
25
30
PBS 6.3ul 20% AS 30% AS 40% AS 50% AS 60% AS
Snail
Twist
Slug
Snail & E-Cadherin
0
2
4
6
8
10
12
14
16
PBS 6.3ul 20% AS 30% AS 40% AS 50% AS 60% AS
52

Figure 2.10 Identification of CAF CM fraction enriched for EMT-inducing protein activity.
Biological activity for each CAF CM fraction was assayed using the in vitro spheroid formation
assay with cE1 SC
med
. Biological activity resulted in greater EMT-like morphological changes
versus spheroid growth correlative with upregulation of EMT and stem cell associated
transcription factors. In agreement with the data in the previous figure, CAF CM 40% fraction
treated SC
med
cells displayed the highest propensity toward EMT based on quantitative real time
PCR analysis of EMT transcription factors (Snail, Slug and Twist), stem cell transcription factors
(Oct4, Sox2 and Nanog) and decrease in E-Cadherin. Ratios are expressed as fold change
normalized to β-actin expression.

Snail
E-Cadherin
Stem Cell Markers
0
5
10
15
20
25
30
PBS 63ul 20% AS 30% AS 40% AS 50% AS 60% AS
Oct4
Sox2
Nanog
2.4 Discussion
2.4.1 Use of the cE1 Cell Line as a Model System for EMT
Proof-of-principle experiments for CAF-secreted factors focused my work on the cell
line, cE1, derived from an epithelial tumor of the cPten
-/-
L mouse model of prostate
adenocarcinoma. Epithelial cells from the cell line were grown in vitro in Matrigel for
spheroid formation to assay for the ability of prostate cancer cells to undergo EMT in the
presence of signaling factors from the tumor microenvironment.

The use of the CRPC cell line, cE1, from the model, versus the primary tumor cell lines
E2 and E4, was based upon optimization experiments (data not shown) of the 3-D
Matrigel culture system of spheroid growth and induction of EMT, and published
findings from tumors grown from subcutaneously injected cells in vivo (Liao, Liang et
al., 2010). cE1 did not show readiness to undergo EMT-like morphological and
molecular changes in the absence of treatment, whereas both E2 and E4 rapidly
underwent proliferation and EMT as assessed by growth rate, filamentous outgrowths of
spheroids and high expression of EMT transcription factors and mesenchymal markers.
This in vitro observation of biological variance between E2 and cE1 is also reflected by
our findings from the in vivo tumorigenicity of these cells lines. The differences between
the phenotypes of the subcutaneous tumors formed by E2, E4 and cE1 are highly
noticeable. While E2, E4 tumors appear as carcinosarcomas (epithelial-to-mesenchymal),
cE1 tumors are adenocarcinomas. Thus, while E2 and E4 cells appear to be prone to
strong conversion to a putative “hybrid” phenotype in vivo, cE1 cells are not under
53

similar conditions (Liao, Liang et al., 2010). Therefore, following these observations, the
cell line chosen to best observe changes from an epithelial to mesenchymal phenotype in
vitro was cE1. All of the experiments of the cell line model system for this dissertation
study have been completed using this CRPC cell line.

Use of the Matrigel system for observation of EMT in vitro was based upon previous
published works that showed clear transformation of PC-3 human prostate cancer cells in
response to EMT induction by BMP7 (M. Lim, Chuong, & Roy-Burman, 2011). Both
spatial and temporal processes were believed to be better modeled in a 3-D culture
system that contained physiologically relevant factors, such as an extracellular matrix
provided by the Matrigel, than in a monolayer in tissue culture plates or in non-
attachment plates.

2.4.2 Synergy Between TGF- β1 and TNF α in Promoting EMT
TGF- β1 and TNF α converge through the NF- κB and MAPK/ERK pathways in the
literature surrounding their role in cancer and in EMT of colon carcinoma cells (Bates &
Mercurio, 2003). But perhaps more importantly is the fact that both cytokines positively
regulate Snail expression. It is well understood that TGF- β1 induces EMT through
transcriptional control of Snail (Thiery et al., 2009). TNF α has been shown to stabilize
Snail during EMT by protecting it from protein degradation. This effect is mediated by
the protein CSN2, which is stimulated upon TNF α activation of NF- κB. CSN2 prevents
phosphorylation and ubiquitylation of Snail by inhibiting its association with the protein
54

kinase GSK-3 β, responsible for its phosphorylation and eventual degradation (Y. Wu et
al., 2009). Perhaps the synergy between TGF- β1 and TNF α can be explained as
“constitutive” upregulation of Snail, leading to a continuous signal for driving EMT in
the co-presence of both cytokines.

2.4.3 Ammonium Sulfate Isolation of Annexin A1 from CAF CM
Other secreted proteins may interact with AnxA1, forming biologically active complexes
that act through either a stimulatory or inhibitory manner in inducing the CAF-mediated
EMT-like morphological and transcriptional changes in vitro. Evidence that suggests that
these interactions may exist comes from polyacrylamide gel analysis of AnxA1 in NPF
and CAF CM AS precipitated fractions of secreted proteins. MS revealed a silver-stained
band present in all fractions from CAF CM and implicated the phospholipid binding
protein AnxA1. Presence of this band is easily detectable in CAF CM fractions and
faintly detectable in only the NPF CM 30% fraction. Biological activity that resulted in a)
greater EMT-like morphological changes versus spheroid growth and b) upregulation of
EMT and stem cell associated transcription factors correlated with the CAF CM fractions
that have the strongest AnxA1 silver-stained bands. The NPF CM fraction that contains a
faint AnxA1 band does not induce the EMT-like morphological or transcriptional
changes (data not shown) that are observed from the CAF CM fractions. Silver stained
polyacrylamide gel analysis instead revealed the presence of another band present in the
NPF CM AnxA1-detectable fraction, which was absent within all of the CAF CM
fractions. This band has yet to be analyzed by MS and may represent a partner protein
55

that binds with AnxA1 to form an inactive or inhibitory protein complex that can explain
the inactivity of AnxA1 in NPF CM.

Conversely, other partners that are stimulatory or act as positive regulators of AnxA1
EMT-associated function may exist within the CAF CM and have yet to be identified. An
approach to identifying not only inhibitory partners, but also positive regulators of
AnxA1 is co-immunoprecipitation and use of this technique should be explored in future
analyses of NPF and CAF CM.

In conclusion, from the body of research described in this chapter, I would like to draw
the significance that tumor-derived prostate epithelial cells that undergo an EMT in vitro
are able to generate a population of cells with gene expression profiles indicative of stem
cell characteristics and that, therefore, may be capable of acting as stem-like progenitor
cells. In final summary, this research may provide evidence to demonstrate that prostate
cancer stem cells may be generated in vivo by induction through an EMT due to the
combined actions of the cytokines TFG- β1 and TNF α, and/or other factors, which are
secreted at higher than normal physiological concentrations by the surrounding tumor
microenvironment, especially in the later stages of cancer progression. Further research
should be conducted following the completion of the experiments outlined in this chapter
to elucidate the signaling mechanisms by which such cytokine inducers are able to lead to
EMT and de-differentiation of cancer cells.
56

Chapter 3:
Annexin A1 Secreted from Cancer-Associated Fibroblasts
Influence Prostate Cancer Cells to Gain Stem Cell-like
Properties
3.1 Abstract
Previously described evidence that CAFs derived from the stromal compartment of
prostate tumors secrete factors that enhance both the stemness and growth potentials of
CSCs of the primary prostate tumor in cPten
-/-
L mice prompted identification of AnxA1
as a prominent CAF-secreted factor. CAF CM-derived AnxA1 and N-terminal mimetic
peptide Ac2-26 treatment of the cPten
-/-
L derived cE1 cell line led to up-regulation of
EMT transcription factors as well as stem cell transcription factors in vitro. AnxA1,
shown to be a positive regulator of EMT in vitro, may generate cells with stem cell-like
properties that give rise to multi-differentiated prostatic glandular structures in vivo. In
parallel, AnxA1 enhanced the CSC potential of primary epithelial tumor cells. This was
shown in vitro through increased size and incidence rate of spheroid formation.
Furthermore, Ac2-26 and murine recombinant full-length AnxA1 (rAnxA1) treatment of
both primary epithelial tumor cells and cell lines led to enhanced glandular structure
formation and statistically significant increases in basal marker p63 expression in vivo,
compared to controls (P < 0.05). Treatment is associated with pErk1/2 expression. The
combined results of these experiments indicate that AnxA1 is capable of promoting de
novo generation of CSCs via EMT and maintenance of an existing CSC population.
57

3.2 Introduction
Annexins comprise a family of calcium-dependent phospholipid-binding proteins
(Crompton et al., 1988; Crumpton & Dedman, 1990). Each family contains conserved C-
terminal sequences for the calcium-binding and phospholipid-binding domains (40-60%
homology), whereas an N-terminus of varying length and sequence is unique to each
family member and confers the specific biological activity (Raynal & Pollard, 1994).
AnxA1, a 37kDa protein and the first characterized member of the family, was originally
reported for its anti-phospholipase and anti-inflammatory activities after glucocorticoid
induction (Blackwell et al., 1980; Miele et al., 1988). Subsequent studies showed that
both recombinant AnxA1 and AnxA1-derived N-terminal peptides could mimic
endogenous anti-inflammatory properties of AnxA1 (Gavins et al., 2003; Getting et al.,
1997; Parente & Solito, 2004). In addition to the regulatory region, the N-terminus also
contains the sites for phosphorylation (Raynal & Pollard, 1994) by various signal
transducing kinases, such as EGFR-TK, PDGFR-TK, HGFR-TK, TRPM7 and PKC
(Hsiang et al., 2006; Skouteris & Schroder, 1996).

AnxA1 does not possess a recognized signal sequence for targeting to the endoplasmic
reticulum (Christmas et al., 1991); however, it is now understood to follow cell-specific
novel manners of secretion (D'Acquisto, Perretti et al., 2008). In cells that do not store
AnxA1 in granules (such as in neutrophils), AnxA1 is exported by the ABC-A1
transporter system or ATP-sensitive K
+
channels in response to phosphorylation on Ser
27
,
which is necessary for protein export and secretion (Chapman et al., 2003; Morris et al.,
58

2002; Payne et al., 2005; Perretti et al., 1996; Wein et al., 2004). Phosphorylation on
Ser
27
is directed by Ca
2+
-dependent isoforms of PKC, and subsequent translocation of the
serine
27
-phosphorylated species of AnxA1 to the plasma membrane occurs at specific
lipid domains that allow for secretion (Croxtall et al., 2000; John et al., 2003; Solito et al.,
2006; Solito et al., 2003; Yazid et al., 2010). Extracellular Ser
27
-AnxA1 undergoes a
conformational change in the presence of ≥ 1 mM Ca
2+
causing exposure of the N-
terminal domain from inside the pore created by the four repeated motifs of the core
domain and, thereby, binding to its receptor (Rosengarth & Luecke, 2003; Solito et al.,
2006). Binding of AnxA1 to the receptor activates downstream signaling via
phosphorylation of the MAPK Erk1/2 (Hayhoe et al., 2006; Lange et al., 2007; Perretti &
D'Acquisto, 2009; Tagoe et al., 2008). Through its specific N-terminal sequence, AnxA1
directly interacts with the FPRs, a family of G-protein coupled receptors, which includes
FPR1, FPR2 (also known as ALXR) and FPR3 in humans and Fpr1 and Fpr-related
proteins (Fpr-rs) 1-7 in mouse (Ernst et al., 2004; Perretti et al., 2002). Endogenous and
recombinant full-length AnxA1 specifically bind to FPR2/ALXR and murine Fpr-rs1 (up
to 30% activity has been demonstrated for Fpr1), whereas the bioactive N-terminal-
derived peptide comprised of amino acids 2-26, Ac2-26, has been shown to activate all
members of the human FPR family and the murine Fpr1 and Fpr-rs1, and may
additionally activate Fpr-rs2 (D'Acquisto, Perretti et al., 2008; Ernst et al., 2004; Gavins
et al., 2003; Hayhoe et al., 2006).

59

Regarding tissue distribution, AnxA1 is found in high abundance in lung, bone marrow,
intestine, lymphatic tissue and reproductive tracts and, interestingly, at highest
concentration in the seminal fluid of the prostate (150 μg/ml) (Christmas et al., 1991). It
is also found to have differential expression in certain cancers (L. H. Lim & Pervaiz,
2007). In prostate cancers, AnxA1 loss of expression from the ductal epithelial cells was
reported (Kang et al., 2002; Patton et al., 2005; Paweletz et al., 2000; Smitherman et al.,
2004; W. Xin et al., 2003). Prostatic stromal cells also display expression of AnxA1
(Christmas et al., 1991). In ductal carcinoma in situ and invasive carcinoma breast tumors,
stromal AnxA1 expression was positively correlated with infiltration of both epithelial
and stromal cells (Khau et al., 2011).

60

3.3 Results
3.3.1 Effect of CAFs on Prostate Cancer Cells
We have already demonstrated that CAFs may stimulate the proliferation and
tumorigenicity of prostate cancer in vivo. In order to explore our previous observation of
paracrine/juxtacrine interaction between CAF and CSCs (Liao, Adisetiyo et al., 2010a),
we chose a prostate tumor cell line as well as CAF or normal prostate fibroblasts (NPF)
(S. Yang et al., 2008), all from the same mouse model to constitute a test system. The
tumor cell line, cE1, which was derived from a castration-resistant prostate cancer
(CRPC) of the cPten
-/-
L model (Liao et al., 2007) was described earlier (Liao, Liang et al.,
2010). We observed that co-culture of NPF and CAF had a differential effect on the
epithelial cells. The cE1 cells co-cultured with CAF displayed morphological changes
similar to EMT as compared to cE1 cells either grown under control conditions without
co-culture or cE1 cells co-cultured with NPF (Figure 3.1A).

Considering the presence of phenotypic heterogeneity, even in a cell line, cE1 cells were
subjected to FACS to enrich for subpopulations of putative tumor-initiating/progenitor
cells based on cell surface expression levels of stem cell antigen-1 (Sca-1) and laminin
receptor α6 (CD49f). Our rationale for this analysis was based upon a recent body of
published work whereby prostate cancer putative stem cells have been described. The
Lineage (Lin) negative (CD31/CD45/TER119)
-
/Sca-1
+
/CD49f
+
(LSC) phenotype within
prostatic tissue cells was determined to contain cells with properties of self-renewal and
differentiation in vitro (Lawson et al., 2005; Mulholland et al., 2009; W. Xin et al., 2003)
61

and, after further enrichment for Lin
-
/Sca-1
hi
/CD49f
hi
(LSC
hi
) cells based on high
expression levels of Sca-1 and CD49f, generation of prostate glandular structures in vivo
(Liao, Adisetiyo et al., 2010a). Subpopulations of cE1 cells displaying high, medium or
no expression of Sca-1 and CD49f (SC
hi
, SC
med
or SC
none
,

respectively) were recovered
(Figure 3.1B). Lineage markers were not necessary for selection by FACS for cE1 cells
because contaminating non-epithelial cell types were removed during derivation of the
cell line (Liao, Liang et al., 2010).

62

Figure 3.1 CAFs induce EMT in SC
med
PCa cells in vitro. A, phase contrast images of
representative spheroids from tumor cell line, cE1, after co-culturing with either control media,
NPFs or CAFs. Bar 100 μm. B, cE1 cells were FACS sorted into subpopulations based on high,
medium and no expression of Sca-1 (S) and CD49f (C; SC
hi
, SC
med
and SC
none
, respectively). C,
Sorted cells grown in post-FACS 3-D culture displayed different morphologies and response to
CAF CM. cE1 SC
hi
and SC
med
spheroids increased in size in the presence of CAF CM. Phase
contrast images show gross spheroid morphology and incidence of EMT-like phenotypic changes
in CAF CM treated cE1 SC
med
cells. Sections of spheroids were analyzed by
coimmunofluorescence (co-IF) using antibodies against the basal/transit-amplifying cell marker
CK5 (red) and luminal cell marker CK8 (green). 4’,6-Diamidino-2-phenylindole (DAPI) was
used to label cell nuclei. Phase contrast, bar 100 μm; IF, bar 50 μm. D, Grown in maintenance
media, SC
hi
and SC
med
displayed distinguishing molecular characteristics as assessed by
expression levels of stem cell transcription factors Oct4, Nanog, Sox2 and Snail. Ratios are
expressed as fold change normalized to β-Actin expression. *, P < 0.05; **, P < 0.001; ***, P <
0.001.

63

64

Figure 3.1 Continued

We described the formation of spheroids after co-culture of Lin
-
Sca-1
+
cells from the
primary cPten
-/-
L tumors (T-LS) embedded in a Matrigel matrix with either NPFs or
CAFs that were grown in inserts. T-LS spheroids produced in the co-culture with CAFs
were larger in size and were composed of multiple basal and luminal cell layers as
opposed to those spheroids generated with NPFs or without stromal cells. After further
fractionation of the T-LS cells into T-LSC
hi
and T-LSC
med
fractions (hereafter referred to
as LSC
hi
and LSC
med
, respectively), we found that spheroid-forming cells were largely
segregated with the LSC
hi
subpopulation (Liao, Adisetiyo et al., 2010a). We applied a
Control Snail (Day 14)
0.00
0.20
0.40
0.60
0.80
1.00
1.20
SCHI SCMED
p<0.01
cE1
cE1 +
CAF
100 um
cE1 +
NPF
A C
Hi
0.7%
Med
70.7%
None
7.1%
cE1
Sca
-1
CD49f
B
SC
HI
SC
MED
SC
NONE
Control Oct4 (Day 14)
0.00
0.20
0.40
0.60
0.80
1.00
1.20
SCHI SCMED
p<0.01
D
Control
Control Nanog (Day 14)
0.00
0.20
0.40
0.60
0.80
1.00
1.20
SCHI SCMED
p<0.001
CAF CM
Control Sox2 (Day 14)
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
SCHI SCMED
Control
CAF CM
p<0.001
similar approach to 3-D co-culture of the tumor cell line derived SC
hi
and SC
med
cell
fractions. SC
hi
and SC
med
contained cells that showed spheroid-forming ability, and the
growth of the spheroids from both SC
hi
and SC
med
were stimulated by conditioned
medium (CM) collected from CAF. The spheroids formed from SC
med
, however,
appeared to contain increased number of CK5 and CK8 double positive cells, which
might be indicative of the detection of more transit-amplifying cells, relative to those
from the SC
hi
cells (Figure 3.1C). The results of qRT-PCR analyses of SC
hi
and SC
med

indicated some characteristic differences between these two subpopulations. As shown in
Figure 3.1D, SC
hi
cells grown under control culture conditions have significantly higher
basal expression levels of stem cell transcription factors Oct4 (P < 0.01), Nanog (P <
0.001) and Sox2 (P < 0.001 ), and transcription factor Snail (P < 0.01) as compared to
SC
med
cells. Encouraged by these differences in basal levels, we then proceeded to
determine the expression levels of Oct4, Sox2, Nanog, EMT transcription factors Snail,
Slug and Twist, epithelial marker E-Cadherin and mesenchymal markers N-Cadherin and
Vimentin in SC
hi
, SC
med
and SC
none
cells following exposure to CM from either CAF or
NPF cultures. We noted extensive EMT-like phenotypic changes induced in SC
med
cells
by CM from CAFs and much less by that from NPFs (see results in Chapter 2).
Correspondingly, EMT’ed SC
med
cells displayed a significant up-regulation of Oct4 (P <
0.05), Sox2 (P < 0.001), Snail (P < 0.001), Slug (P < 0.01), Twist (P < 0.01), N-Cadherin
(P < 0.001), Vimentin (P < 0.05 ) and significant decrease in E-Cadherin (P < 0.01) as
compared to SC
hi
and SC
none
, and up-regulation of Nanog (P < 0.001) as compared to
SC
hi
(Figure 3.2). These changes were tested over a 14 day period, wherein expression
65

levels of these transcription factors and markers significantly increased in SC
med
cells
from day 7 to day 14 time points (Figure 3.3). In comparison, SC
hi
expression levels
remained remarkably unchanged (data not shown).

Figure 3.2 SC
med
cells treated with secreted CAF CM factors show upregulation of EMT/stem
cell transcription factors in vitro. SC
hi
, SC
med
and SC
none
cells were subjected to control media,
NPF CM or CAF CM for 14 days. Induction of EMT in response to CAF CM in cE1 SC
med
cells
was confirmed by qRT-PCR using EMT transcription factor markers Snail, Slug and Twist, stem
cell transcription factors Oct4, Sox2 and Nanog, epithelial marker E-Cadherin, and mesenchymal
markers N-Cadherin and Vimentin. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Snail
0.0E+00
5.0E-03
1.0E-02
1.5E-02
2.0E-02
2.5E-02
SCHI SCMED SCNONE
Mean Ratio to b-Actin
Twist
0.0E+00
1.0E-02
2.0E-02
3.0E-02
4.0E-02
5.0E-02
6.0E-02
7.0E-02
SCHI SCMED SCNONE
Mean Ratio to b-Actin
Oct4
0.0E+00
5.0E-02
1.0E-01
1.5E-01
2.0E-01
2.5E-01
3.0E-01
SCHI SCMED SCNONE
Mean Ratio to b-Actin
Sox2
0.0E+00
2.0E-02
4.0E-02
6.0E-02
8.0E-02
1.0E-01
1.2E-01
SCHI SCMED SCNONE
Mean Ratio to b-Actin
N-Cadherin
0.0E+00
5.0E-02
1.0E-01
1.5E-01
2.0E-01
2.5E-01
3.0E-01
3.5E-01
SCHI SCMED SCNONE
Mean Ratio to b-Actin
Nanog
0.0E+00
2.0E-02
4.0E-02
6.0E-02
8.0E-02
SCHI SCMED SCNONE
Mean Ratio to b-Actin
Slug
0.0E+00
1.5E-01
3.0E-01
4.5E-01
SCHI SCMED SCNONE
Mean Ratio to b-Actin
***
**
**
*
***
***
***
E-Cadherin
0.0E+00
2.0E-02
4.0E-02
6.0E-02
8.0E-02
1.0E-01
SCHI SCMED SCNONE
Mean Ratio to b-Actin
**
Vimentin
0.0E+00
3.0E-03
6.0E-03
9.0E-03
1.2E-02
1.5E-02
1.8E-02
SCHI SCMED SCNONE
Mean Ratio to b-Actin
* p<0.05
** p<0.01
*** p<0.001
*
Media
NPF CM
CAF CM
66

Snail Expression
0.0E+00
6.0E+00
1.2E+01
1.8E+01
SCMED D.7 SCMED D.14

Figure 3.3 Expression of EMT/stem cell transcription factors increases after 14 days in SC
med

via CAF CM treatment. Significant up-regulation of EMT and stem cell transcription factors, as
well as mesenchymal markers, and decrease in E-Cadherin in SC
med
cells exposed to CAF CM
was observed between 7 and 14 days of treatment, compared to control. Ratios are expressed as
fold change normalized to β-Actin expression. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Nanog Expression
0.0E+00
5.0E+01
1.0E+02
1.5E+02
2.0E+02
SCMED D.7 SCMED D.14
p<0.001
p<0.001
Slug Expression
0.0E+00
1.0E+01
2.0E+01
3.0E+01
4.0E+01
5.0E+01
SCMED D.7 SCMED D.14
p<0.01
p<0.001
Twist Expression
0.0E+00
2.0E+00
4.0E+00
6.0E+00
SCMED D.7 SCMED D.14
p<0.05
p<0.01
Oct4 Expression
0.0E+00
2.0E+01
4.0E+01
6.0E+01
SCMED D.7 SCMED D.14
p<0.05
p<0.01
p<0.001
Sox2 Expression
0.0E+00
4.0E+01
8.0E+01
1.2E+02
SCMED D.7 SCMED D.14
p<0.001
p<0.001
p<0.001
N-Cadherin Expression
0.0E+00
2.0E+01
4.0E+01
6.0E+01
8.0E+01
SCMED D.7 SCMED D.14
p<0.05
E-Cadherin Expression
0.0E+00
6.0E-01
1.2E+00
1.8E+00
SCMED D.7 SCMED D.14
p<0.01
p<0.05
p<0.01
67

On the basis of these findings, we inferred that while SC
hi
did display certain molecular
characteristics of stem/progenitor cells, such as spheroid-formation capacity, SC
med
cells
appeared to be enriched with a proliferating tumor cell fraction that was especially
susceptible to CAF-induced EMT. More importantly, it appeared that the factors that are
responsible for induction of this biological effect were contained within the conditioned
medium prepared from CAF, indicating the presence of paracrine-acting secreted
molecules.

3.3.2 A Search for Responsible CAF Factors Identifies AnxA1
To identify secreted factor(s) from the CAFs that are capable of inducing EMT and
possibly contributing toward an EMT-CSC lineage, we decided to reduce the complexity
of the CAF CM using fractionation techniques. Proteins from both NPF and CAF CM
were salted out into five fractions following ammonium sulfate (AS) precipitation: 20%,
30%, 40%, 50% and 60%. Increased EMT-like morphology and gene expression level
changes consistent with previous experiments, as shown in Figure 3.1 and Figure 3.2,
were seen when SC
med
cells were exposed to CAF CM AS fractions, particularly at the
30%-40% AS cut (see results in Chapter 2). Silver stain visualization of the proteins after
fractionation of NPF and CAF CM presented detectable levels of a dark band in both
CAF CM and CAF CM fractions, but which appeared weak or absent in the NPF CM and
NPF CM fractions. We selected this band as a potential candidate for mass spectrometry
(MS), an analysis that pointed to the phospholipid binding protein, AnxA1 (Figure 3.4A).
68

Western blot using an antibody against AnxA1 confirmed its abundance in the CAF CM
fractions as compared to the NPF CM fractions (Figure 3.4B).

Figure 3.4 AnxA1 is secreted from CAF. A, AS precipitated NPF and CAF CM protein fractions
were run on an electrophoresis gel for silver stain. A pronounced band was found to be present in
the CAF CM and CAF CM AS fractions, but was not detected in the NPF CM or NPF CM AS
fractions. CAF CM AS fraction bands were removed and analyzed using mass spectrometry (MS).
AnxA1 was implicated from the results of MS. B, Western blot against AnxA1 confirmed the
presence of AnxA1 in stromal cultures. α-AnxA1 stained positive for two predicted bands in blots
in NPF, NPF CM fractions, CAF and CAF CM fractions. Both bands were more prevalent in
CAF and CAF CM fractions compared to NPF, and the upper band that was more prevalent in
CAF CM fractions was correlated with biological activity (as described in biological function
assays in Chapter 2). Highest detection of AnxA1 was found in the 30-40% AS cut of CAF CM. C,
Phospho-serine specific antibody recognizing the amino acid motif ± 3 amino acids surrounding
p-Ser
27
against NPF and CAF lysates revealed more p-Ser
27
–AnxA1 in CAF upper band. D,
Serine phosphorylator, PKC δ is upregulated in CAF. *, P < 0.01.

Phosphorylation of AnxA1 on a specific serine residue (S
27
) by the Ca
2+
-dependent
kinase PKC has been extensively reported to be necessary for extracellular secretion of
the protein (Croxtall et al., 2000; John et al., 2003; Solito et al., 2006; Solito et al., 2003;
Regulators of ANXA1
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
NPF CAF
N orm al M ean R atio to β-A ctin
Pten
Trpm7
PKC β
PKC δ
PKC ε
Regulators of ANXA1
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
NPF CAF
Normalized Mean Ratio to β-Actin
Pten
Trpm7
PKC β
PKC δ
PKC ε
200
116
97.4
66
45
31
21.5
14.4
6.5
CM 20% 30% 40% 50% 60% CM 20% 30%40%50%60%
NPF A
NPF
α-Anxa1
L
CAF
CAF3
ysate
NPF
20%CM
NPF
30%CM
NPF
40%CM
NPF
50%CM
NPF
60%CM
*
B
p<0.003
CAF1
CAF2
K-562
39 kDa
35 kDa
39 kDa
35 kDa
39 kDa
35 kDa
NPF K-562 NPF2 CAF1 CAF4 CAF2
α-pSer
27
-
Anxa1
α-Anxa1
α-β-Actin
C D
69

Yazid et al., 2010). The presence of phospho-serine within the context of the specific
amino acid recognition motif necessary for PKC mediated serine phosphorylation was
assayed by western blot in CAF and NPF cultures. Multiple CAF cultures were shown to
possess higher protein expression levels of phospho-Ser
27
-AnxA1 than NPF cultures, in
addition to higher total abundance of AnxA1 (Figure 3.4C). Known protein
phosphorylators of AnxA1 were also assayed for up-regulation in CAFs versus NPFs.
Calcium-dependent subtypes of protein kinase C (PKC) were proposed to be responsible
for phosphorylation on S
27
(Solito et al., 2003). Transcriptional expression levels of Pten,
Trpm7 and selected PKC subtypes, revealed that PKC δ is significantly up-regulated in
CAFs (Figure 3.4D; P < 0.01). Unexpectedly, AnxA1 mRNA was found to be
significantly less abundant in CAF than in NPF, shown in Figure 3.5 (P < 0.001), which
suggests possible altered translational regulation of AnxA1 protein in CAF compared to
NPF or a feedback mechanism due to higher protein stability in CAF. Interestingly,
increase in intracellular protein levels and subsequent increase in secreted species of
AnxA1 has been correlated with minimal repletion to decrease of RNA expression levels
(D'Acquisto, Paschalidis et al., 2008; Philip, Flower, & Buckingham, 1997).

Figure 3.5 AnxA1 transcription levels in NPF versus CAF. AnxA1 RNA is decreased in CAF
cultures, compared to NPF (P < 0.001).
AnnexinA1 RNA Levels
0.0
0.2
0.4
0.6
0.8
1.0
1.2
NPF CAF
Mean R atio to b-A ctin
AnnexinA1 RNA Levels
0.0
0.2
0.4
0.6
0.8
1.0
1.2
NPF CAF
Norm alized Mean Ratio to β-Actin
p<0.001
*
70

3.3.3 Effect of AnxA1 on cE1 Cells in vitro
Neutralizing antibody (nAb) specific to the active N-terminal portion of the AnxA1
protein was incubated with cE1 cells treated with enriched fractions of CAF CM. Dose-
dependent response to increasing concentrations of AnxA1 nAb added to culture media
showed a significant decrease in the number of spheroids greater than 50 microns (P <
0.01 at 5 μg/ml; P < 0.001 at highest dose, 20 μg/ml; Figure 3.6A). Similarly, exposing
cE1 cells to AnxA1 nAb at the optimal dose, 5 μg/ml, and in the presence of CAF CM
led to a significant decrease in spheroid number in both SC
hi
and SC
med
cells grown in 3-
D culture (P < 0.01 and P < 0.05, respectively). Therefore, addition of the nAb negated
the increase in spheroid number observed when SC
hi
and SC
med
cells were subjected to
growth in the presence of CAF CM (Figure 3.6B). In further support of the ability of
AnxA1 nAb to repress the effects of AnxA1 on spheroid proliferation and EMT-like
properties of treated cells, total cE1 cells were allowed to grow in Matrigel for 14 days in
the presence of CAF CM, AS-enriched CAF CM fractions, or N-terminal mimetic (Ac2-
26) with, or without, addition of AnxA1 nAb. Addition of AnxA1 nAb to cE1 cells
grown in the presence of CAF CM, AS-enriched CAF CM fractions, or Ac2-26 led to
reduced expression of Snail, Slug, Twist and Oct4, indicating the importance of this
protein in promoting EMT and stemness (Figure 3.6C, D). The most pronounced, and
significant, decreases in expression level were consistently observed for Snail, Twist and
Oct4 in cells exposed to AnxA1 nAb in the presence of CAF CM, AS-enriched CAF CM
fractions and Ac2-26, suggesting possible specificity of AnxA1 predominantly upon
Snail, Twist and Oct4 dependent pathways (see P values in Figure 3.6D).
71

cE1 Spheroids Day 14
0
20
40
60
80
100
120
140
PrEGM CAF CM CAF CM + nAb
No. of Spheroids
cE1 + CAF CM 40% AS + Anxa1 nAb
0
100
200
300
400
500
600
700
0 ug/ml 5 ug/ml 10 ug/ml 20 ug/ml
Mean No. Spheroids > 25 um
Day 11 Rx <50 um
Day 11 Rx >50 um
p < 0.01
p < 0.01
p < 0.001
**
*
*
p<0.05
•
p

Figure 3.6 Effect of α-AnxA1 neutralizing antibody (nAb) on spheroids and EMT of cE1 cells.
A, Optimization of nAb in the presence of AnxA1 enriched CAF CM AS fraction. B, Addition of α-
AnxA1 nAb to cE1 SC
hi
and SC
med
cells reduced the number of spheroids grown in the presence of
CAF CM. •, P < 0.05; *, P < 0.01, **, P < 0.001. C and D, Unsorted cE1 cells showed increase
in expression of Snail, Slug, Twist, and Oct4 after treatment with CAF CM, AnxA1 enriched CAF
CM AS fraction from two CAF cultures and Ac2-26. cE1 cells treated with α-AnxA1 nAb in the
presence of CAF CM, AnxA1 enriched CAF CM AS fractions and Ac2-26 had reduced expression
of Snail, Slug, Twist and Oct4. •, P < 0.05; *, P < 0.01; **, P < 0.001.

3.3.4 AnxA1 Treated cE1 SC
med
Cells Undergo De-differentiation in vivo
Next, to substantiate the in vitro findings of AnxA1 induction of EMT in vivo, the SC
med

expressing subpopulation of the cE1 cell line was used in renal capsule transplantation
technique. Control, CAF CM- or Ac2-26-treated cE1 SC
med
cells were admixed with
UGSM and grafts were transplanted under the renal capsule of non-obese diabetic/severe
combined immunodeficient (NOD.SCID) male mice (Figure 3.7).
β-actin
E-Cadherin
Snail
Slug
Twist
Oct4
Sox2
Media
w/V.C.
CAF CM
40%
CAF CM
40%+nAb
CAF CM
+ nAb
SChi <0.01
*
C D
A B
SCmed
cE1 Day 14 Expression
0
1
2
3
4
5
6
Control CAF1 CM CAF1 CM +
α-AnxA1
CAF1 CM
40%
CAF1 CM
40% + α-
AnxA1
CAF2 CM
40%
CAF2 CM
40% + α-
AnxA1
Ac2-26 50 μMAc2-26 10 μM
+ α-AnxA1
10 μM
Ac2-26 50 μM
+ α-AnxA1
10 μM
ΔΔCt Normalized to β-Actin
Snail
Slug
Twist
Oct4
•
*
•
*
**
•
*
**
•
*
**
•
*
•
*
*
•
•
•
•
•
72

Figure 3.7 Gross morphology of cE1 grafts under renal capsule. Mean graft size was measured
by diameter and found to be significantly increased in Ac2-26 treated versus control. *, P <
0.001.

The incidence of glandular structure formation was 75% when SC
med
cells were treated
with CAF CM and 100% when treated with Ac2-26; however, no glandular structures
were detected in control treated SC
med
mixed with UGSM. Staining for androgen receptor
(AR) and proliferation marker Ki67 confirmed that the structures seen in CAF CM- and
Ac2-26-treated grafts were composed of highly proliferating prostate cancer cells (Figure
3.8). Additionally, significant increases in graft size (Figure 3.7), number of p63
+
cells
per graft area (P < 0.05) and ratio of p63
+
to CK8
+
cells per graft area (P < 0.05) were
observed in Ac2-26 treated SC
med
cE1 cells (Figure 3.9).
4 mm 4 mm 4 mm 4 mm
3747 3749 3750 3748
Mean Graft Size
0
1
2
3
4
5
6
Control CAF CM Rx Ac1-25 Rx
Graft D iam eter in m m
* p<0.001
3743 3725 3742
Untreated
Control
CAF CM
Treated
Ac2-26
Treated
3 mm 2.5 mm 2.5 mm
Ac2-26 Rx
3 mm 5 mm 4 mm
3746 3744 3745
73

A
p63 AR
Untreated Control SC
me

Figure 3.8 SC
med
cells treated with CAF CM and Ac2-26 undergo de-differentiation under the
renal capsule of NOD.SCID male mice in vivo. Hematoxylin and eosin (H&E) and
immunohistochemistry (IHC) images of representative untreated, CAF CM treated and Ac2-26
treated cE1 SC
med
cells reveal the formation of glandular structures in CAF CM treated and Ac2-
26 treated grafts. Treated grafts possessed more basal and proliferating prostate cancer cells as
analyzed by IHC for p63, CK8, AR, and Ki67. Images at 400x magnification.

Figure 3.9 Significant increase in p63
+
cells in Ac2-26 treated grafts indicates de-
differentiation of SC
med
cells. B, Incidence of grafts with detectable glandular structures. C,
Basal cell expansion was quantified from the number of p63
+
cells per total cells per graft area
and the ratio of p63
+
cells to CK8
+
cells per total cells per graft area. *, P < 0.05.

p63 Postive Cells/Graft Area
0
5
10
15
20
25
30
35
Control CAF CM Ac1-25
Mean %
p63 + : CK8 + Cells/Graft Area
0
10
20
30
40
50
60
Control CAF CM Ac1-25
Mean %
p<0.05
*
p<0.05
*
C B
Chart: # Grafts with
Glandular Structures
SC
med
control
SC
med
+ CAF CM
0/3
3/3
Glandular Structures
SC
med
+ Ac2-26 3/4
Ac 2-26 Ac 2-26
d
+ UGSM
CK8 H&E Ki67
CAF CM Treated SC
med
+ UGSM
Ac2-26 Treated SC
med
+ UGSM
400x
74

The combined results of SC
med
transplantation when treated with either CAF CM or Ac2-
26 suggest that AnxA1 alone may be able to recapitulate the stimulation provided by
CAFs in glandular structure formation. This experiment establishes proof-of-principle
findings that support AnxA1 as one of the significant secreted factors responsible for the
glandular-forming stimulation provided by CAFs on the adenocarcinoma cells of the
mouse prostate.

3.3.5 Effect of AnxA1 on Primary Tumor LSC
hi/med
Cells from the cPten
-/-
L Model in
vitro
We proceeded to test the effect of AnxA1 on sorted LSC subpopulations from the
cPten
-/-
L model. As previously described, co-culture of LSC
hi
cells with CAF led to
increasing biological complexity, indicated by formation of multiple basal and luminal
layers, as well as immunofluorescent detectable co-expression of basal and luminal cell
markers (Liao, Adisetiyo et al., 2010a). In the present study, we further describe that
spheroids formed after co-culture with CAFs could generate acinar-type projections, an
observation which was seen in the previous study but was not reported (Liao, Adisetiyo
et al., 2010a). Similarly, spheroids formed after treatment of LSC
hi
cells with AS-
enriched CAF CM fractions, Ac2-26 and rAnxA1 grew complex acinar structures.
Expansion of both basal and luminal cell layers after CAF co-culture as well as after
treatment with sources of AnxA1 was revealed with IF (Figure 3.10), consistent with our
75

published findings of the effects on spheroid proliferation and differentiation by factors
purportedly secreted from CAFs in co-cultures (Liao, Adisetiyo et al., 2010a).

Figure 3.10 AnxA1 treatment of LSC
hi
cells of the cPten
-/-
L model in vitro. A, Phase contrast,
hematoxylin and eosin (H&E), and immunofluorscence (IF) images show representative LSC
hi

spheroid morphologies under influence of co-culture with UGSM or CAF and treatment with
AnxA1 enriched CAF CM AS fraction or Ac2-26. Spheroids from 3 cell cultures of UGSM and
CAF co-culture, 3 cultures of AnxA1 enriched CAF CM AS fraction treated, and 5 cell cultures of
Ac2-26 treated LSC
hi
were evaluated for this purpose. Coimmunofluorescence was performed
using antibodies against the basal cell marker p63 (red) and luminal cell marker CK8 (green)
and DAPI (blue), to label cell nuclei.

LSC
hi
cells co-cultured with CAFs or treated with AS-enriched CAF CM fractions and
both mimetic and full-length forms of the AnxA1 protein formed greater numbers of
spheroids than LSC
hi
cells co-cultured with UGSM. Treatment with Ac2-26 and rAnxA1
led to the most significant increase in numbers of spheroids formed. Treatment with Ac2-
CPPL LSC
hi
+
50 μM Ac2-26
CPPL LSC
hi
+
CAF 40% AS
CPPL LSC
hi
+
CAF
CPPL LSC
hi
+
UGSM
Brightfield
PP LSC
+
+
UGSM
H&E
Immunofluoresence
200x
100x
200x
400x
200x 400x 400x
A
76

26 led to a twenty-fold increase over UGSM control (P < 0.01), and rAnxA1 led to a
twenty eight-fold increase (P < 0.05 ) (Figure 3.11A, C).

Figure 3.11 Spheroid enrichment by AnxA1 mimetic and recombinant forms. Spheroids were
counted for 5 independent cultures of Ac2-26 treated LSC
hi
and 3 independent cultures of rAnxA1
LSC
hi
and LSC
med
, respectively. rAnxA1 treatment led to increased numbers of spheroids formed
at increasing concentrations in LSC
hi
and LSC
med
cells. Representative brightfield and graphs are
shown.

While LSC
med
cells virtually lack the ability to form spheroids in vitro, it was worthwhile
to test the effect of AnxA1 on the spheroid-forming ability of these cells. Subsequently,
LSC
hi
and LSC
med
cells were assayed for their response to AnxA1 compared to no
treatment (control). Increasing concentration of rAnxA1 added to culturing media
resulted in the formation of greater numbers of spheroids from LSC
hi
and LSC
med
cells
CPPL LSC
hi
Day 21
0
500
1000
1500
2000
2500
UGSM 5µM Ac1‐25 50µM Ac1‐25
No. of Spheroids/Well
hi
CPPL LSC
> 50 µM
> 100 µM
> 300 µM
p<0.001
***
p<0.001
***
p<0.001
***
p<0.01
*
CPPL LSC
hi
Day 14
0
20
40
60
80
100
120
140
160
Control rAnxa1 5 μM rAnxa1 0.5 μM
No. of Spheroids >25 μm
<50 μm
>50 μm
CPPL LSC
med
Day 14
0
50
100
150
200
250
300
Control rAnxa1 5 μM rAnxa1 0.5 μM
No. of Spheroids >25 μm
<50 μm
>50 μm
Vehicle Control rAnxA1 5 μM rAnxA1 .5 μM
(Day 14)
CPPL LSC
me
A
d
(Day 14)
B
C
Vehicle Control rAnxA1 5 μM rAnxA1 .5 μM
50 μM Ac2-26 5μM Ac2-26
77

than in the presence of media alone. Remarkably significant increase in spheroid number
was observed in LSC
med
cells inoculated with rAnxA1 (Figure 3.11C; P < 0.05).

Recent analyses have established that downstream signaling from interaction of soluble
AnxA1 with its receptor FPR2/ALXR in human and Fpr-rs1 in mouse involves activation
of phosphorylation of Erk1/2 (pErk1/2). Both mimetic and full-length proteins are able
to bind to the receptor (Hayhoe et al., 2006). As shown in Fig. 3.12B, after exposure to
increasing concentrations of Ac2-26, LSC
hi
spheroids effaced increasing pErk1/2
activation. pErk1/2 activation could be reversed by addition of the AnxA1 nAb in the
presence of Ac2-26. Other MAP kinases, p38 and Jnk/Sapk did not show up-regulation of
phosphorylation in Ac2-26 treated spheroids relative to control (data not shown).

Figure 3.12 Analysis of pathways associated with AnxA1. AnxA1 R
x
increases pErk1/2
activation & TGF- β1 in LSC
hi
and cE1 spheroids. B, 3.3-fold pErk1/2 activation and 2-fold
Erk1/2 activation was detected by western blot analysis in LSC
hi
spheroids after treatment with
increasing concentrations of Ac2-26. pErk1/2 was reduced in the presence of Ac2-26 with α-
AnxA1 nAb. C, α-TGF β western blots revealed close to two-fold increase in TGF β after treatment
of cE1 spheroids with increasing concentrations of Ac2-26. TGF β expression was reduced after
α-AnxA1 nAb was combined with Ac2-26 treatment. D, qRT-PCR confirmed activation of MAPK
and TGF β pathways, as shown by significant increase in MAPK and TGF βRII expression after
addition of Ac2-26 to cE1 cells grown in 3-D culture. Control is media with IgG. •, P < 0.05.
MAPK
0
0.5
1
1.5
2
2.5
Control Ac2-26 50 μM Ac2-26 10 μM +
α-AnxA1 10 μM
Ac2-26 50 μM +
α-AnxA1 10 μM
ΔΔCt Normalized to β-Actin
TGF βRII
0
0.5
1
1.5
2
2.5
3
Control Ac2-26 50 μMAc2-26 10 μM +
α-AnxA1 10 μM
Ac2-26 50 μM +
α-AnxA1 10 μM
ΔΔCt Normalized to β-Actin
TGF- β1
β-actin
Media
Control
Ac2-26 5µM
+ nAb 10µM
Media
w/IgG
Ac2-26
5µM
Ac2-26
25µM
Ac2-26
50µM
1.78 0.89 1.01 1 1.4 1
UGSM Ac2-26 5µM
+ nAb 10µM
Media
w/V.C.
Ac2-26
5µM
Ac2-26
50µM
pErk1/2

Erk1/2
+ Ac2-26
25µM
β-actin
•
•
•
D
C
B
78

A recently reported basal-like breast cancer model was used to evaluate the consequence
of AnxA1 in the promotion of EMT and metastasis. In the study, AnxA1 contributed to
EMT and may stimulate an autocrine loop of the TGF β receptor (de Graauw et al., 2010).
Following these reported findings, cE1 cells dosed with increasing concentration of Ac2-
26 showed a two-fold increase in TGF- β1 at the 50 μM dose (physiological level, 150
μg/ml). This result is further supported by the evidence that increase in TGF- β1 from
treatment with Ac2-26 could be reversed by simultaneous addition of AnxA1 nAb
(Figure 3.12C). Increase in MAPK and TGF β activity was supported by qRT-PCR results
shown in Figure 3.12D using cE1 cells grown in Matrigel culture with or without Ac2-26
for 14 days (MAPK, P < 0.05; TGF βRII, P < 0.05). Twist protein levels were also tested
after dose response to Ac2-26. Twist expression was additionally found to be two-fold
up-regulated at 50 μM Ac2-26 and could be reversed after addition of AnxA1 nAb
(Figure 3.13).

Figure 3.13 Twist expression after treatment of cE1 spheroids with Ac2-26. After normalization
to β-actin, Twist expression was found to be increased in cE1 spheroids treated with Ac2-26 at
5 μM, 25 μM and 50 μM concentrations. Twist expression was reduced in the presence of Ac2-26
by nAb to AnxA1.

Twist
β-actin
Media
Control
Ac2-26 5µM +
nAb 10µM
Media
w/IgG
Ac2-26
5µM
Ac2-26
25µM
Ac2-26
50µM
1.9 0.85 1.3 1.3 1 1
79

Together, these biological findings suggest two possible modes of mechanism for
AnxA1-induced EMT and stemness properties. As proposed in the diagram in Figure
3.14, soluble AnxA1 may either act as an EMT promoter via the TGF β pathway,
potentially leading to downstream activation of EMT-driver Snail (mentioned in detail in
Chapter 2, Section 2.4.2), or AnxA1 may directly bind to its receptor and activate
FPR2/ALXR (murine Fpr-rs1) to lead to stable upregulation of total Erk1/2 levels and
subsequent phosphorylation to pErk1/2 (Figure 3.14).

Figure 3.14 Proposed schematic for AnxA1 mode of action in prostate cancer
microenvironment. Diagram generated using Ingenuity Systems, Inc. software (IPA) depicting
interactions between AnxA1, AnxA1-S
27
-phosphorylator PKC, TGF β, and murine AnxA1 FPR
receptor Fpr-rs1. Based on published literature, IPA pathway analysis and results of the current
study, we propose a model for AnxA1 secretion from CAFs and effect on prostate tumor epithelial
cells. TGF β increases recruitment of PKC species in CAFs, leading to phospho-S
27
-AnxA1
secretion. Secreted AnxA1 influences prostate tumor epithelial cells in two ways. AnxA1 leads to
upregulation of TGF β, and AnxA1 physically binds with its receptor Fpr-rs1 to activate
phosphorylation of Erk1/2. Intersection of AnxA1 and TGF β pathways may regulate EMT,
whereas pErk1/2 pathway signaling may influence proliferation and differentiation.
Erk1/2
FPR
80

3.3.6 Glandular Morphologies and Histopathological Examination of AnxA1
Treated LSC
hi/med
Cells from Renal Grafts
In the normal murine prostate, similarly as in the normal human prostate, the whole organ
is comprised of pairs of lobes, each of which is organized into a series of branching ducts
lined with epithelial cells. Three epithelial types comprise the ductal branches of the
prostate. These cell types are luminal, neuroendocrine and basal cells. Luminal cells are
differentiated, functionally active secretory cells and constitute the major cell type within
the mouse prostate. Neuroendocrine cells are rare, express synaptophysin and
chromogranin A, and can be found within both the luminal and basal layers. Basal cells
are undifferentiated, nonsecretory cells and purportedly contain the progenitor cells of the
prostate based on experiments demonstrating prostate regression, atrophy and
regeneration in response to successive rounds of androgen deprivation therapy in rats
(English, Drago, & Santen, 1985; English, Santen, & Isaacs, 1987; Hsing, Kadomatsu,
Bonham, & Danielpour, 1996; Kyprianou & Isaacs, 1988; Nikitin et al., 2009; Rouleau,
Leger, & Tenniswood, 1990).

The normal prostate is composed of glands and stroma. The glands are seen in cross
section to be rounded to irregularly branching. These glands represent the terminal
tubular portions of long tubuloalveolar glands that radiate from the urethra. The glands
are lined by two cell layers: an outer low cuboidal layer and an inner layer of tall
columnar secretory epithelium. These cells project inward as papillary projections
(fingerlike projections that interdigitate with the epithelium). The fibromuscular stroma
81

between the glands accounts for about half of the volume of the prostate in humans, but is
distinctly less abundant in mice (Collins & Maitland, 2006; Roy-Burman, Wu, Powell,
Hagenkord, & Cohen, 2004). Concerning cancer, in human pathologies, the most
common morphologic type is “acinar type” where tumor is thought to arise from or
recapitulate prostatic acini. Other uncommon variants exist including ductal,
neuroendocrine, and mucinous types among others. The acinar type (usual type) is
characterized by back-to-back proliferation of small to intermediate sized tumor acini
with scant to moderate intervening stroma (Collins & Maitland, 2006).

In this dissertation, I describe histopathologies from the grafted tumor epithelial cells
recovered after 10 week transplantation under the renal capsule of non-obese
diabetic/severe combined immunodeficient (NOD.SCID) male mice. Tumor epithelial
cells were shown to be capable of forming glandular structures. In my observations, I
noted three distinct morphological subtypes. In collaboration with Dr. Kevin Nash,
Professor of Pathology, I will describe the basic histological features of these three
subtypes and refer to them in this chapter in reference to their morphology.

The most commonly observed glandular structure in the present study is composed of
low cuboidal epithelial cells. These glandular structures are characterized by their
elongated nuclei and low cytoplasmic profile. These structures were found to range
greatly in size but could comprise sections of the graft greater than 100 microns, and p63
+

cells represented 30-50% of the cells within structures composed of this morphological
82

subtype. Similar, yet distinct in their nuclear profile, smaller glandular structures
comprised solely of simple cuboidal epithelium were also observed. These cells differed
from the low cuboidal epithelial cells in that they appeared uniform in size and shape.
Additionally uniform, their nuclei remained spherical and centrally located within the
cytoplasm of the cell. Further histological analysis revealed that these simple cuboidal
epithelial cells had the highest incidence of p63
+
cells out of all three subtypes, ranging
between 70-80%, but they could display up to 100% p63
+
staining in some serial tissue
sections. The incidence of detection was lower for this subtype and their size was
reduced, ranging between 50 to 80 microns. Perhaps most interestingly, was the 100%
positive correlation between incidence of simple cuboidal epithelial glandular structures
and “acinar” glandular structures (defined and described in detail below).

The final morphological subtype observed was complex, composed of mostly high
columnar epithelia, and with basolaterally located spherical nuclei and prominent
cytoplasm, which were organized into acinar units and were surrounded by endothelial
cells and recruited blood vessels. As a note to the recruitment of vasculature within the
graft, this was observed in many grafts which had high proliferative indexes, as assessed
by Ki67 staining, and total cell counts per graft area. In high columnar glandular
structures, p63
+
staining was significantly lower than that observed in the other two
subtypes, wherein p63
+
cells comprised an average of 17% of the total cells. While some
of the low cuboidal glandular structures were noted in size—the largest was observed up
to 300 microns—high columnar glandular structures were consistently equal to or greater
83

than 300 microns, ranging between 300 to 1500 microns in size, and quantification
analysis revealed these structures to be the largest among all detectable glandular
structures in LSC
hi
and LSC
med
grafts (depicted in Figure 3.24). Within these portions of
the graft, multiple, densely packed acinar formations with small lumens were consistently
noted, in contrast to the glandular structures formed by the two other subtypes, which
were not densely located to each other and formed large hollow lumens.

As mentioned above, other cell types were not only present within areas of the grafts
comprised of high columnar glandular structures, but they appeared to be semi-organized.
Blood vessels were observed at high frequency and traversed between acinar units.
Stromal cells were also arranged around the epithelia. In all grafts in which high
columnar, “acinar,” glandular structures were observed, simple cuboidal glandular
structures were present in close relationship, and located to one end of the graft. Overall,
the high columnar glandular structures displayed some polarity in their formation, in
which simple cuboidal glandular structures could be found at the “originating” end of the
graft, proximal to the body of the graft and the renal cortex, and apoptotic cells were
present at the end most distal to the simple cuboidal structures. Of the three subtypes, the
high columnar, “acinar-type,” glandular structures appear to resemble the functional
prostate the most. Evidence supporting this observation, verified in collaboration with Dr.
Nash, is their formation of detectably branched, compound acini by high columnar
epithelium that appear to be connected through ductal regions and are supported by the
presence of simple cuboidal epithelium, endothelial blood vessels and stromal cells.
84

Figure 3.15 Depictual drawing of “acinar-type” versus “glandular-type” structures formed in
renal capsule grafts in vivo. Author’s rendition of histological difference between the two
morphologies as viewed under microscopy. Morphological differences were defined in
collaboration as described in the text to clarify the author’s meaning in using the words “acinar”
and “glandular” in order to specify a distinct phenotype.

3.3.7 Effect of AnxA1 on LSC
hi
Cells in vivo: Stimulation of Glandular Structure
Formation, Differentiation and Proliferation
LSC
hi
cells were treated with rAnxA1, Ac2-26 or media control and admixed with
UGSM for grafting and transplantation under the renal capsule of male NOD.SCID mice
(Figure 3.16).
85

Figure 3.16 LSC
hi
graft gross morphology. G1362 lost most of the graft during surgical
transplantation. G1361 graft was found attached to, but outside of the renal capsule, and became
admixed with fat and muscle cells. These two grafts were removed from further analysis.

Glandular structures were detected in 6/6 (100%) of the rAnxA1 treated LSC
hi
grafts, 3/5
(60%) of the Ac2-26 treated grafts and 2/5 (40%) of untreated controls (Figure 3.17A).
Of the structures noted in AnxA1 stimulated LSC
hi
cells, glandular structures occurred as
phenotypically low cuboidal glandular, simple cuboidal glandular and acinar types. Low
cuboidal glandular and acinar structures were found to be composed of both basal and
luminal cells, and simple cuboidal glandular had a much relatively higher proportion of
basal cells. Interestingly, LSC
hi
rAnxA1 treated grafts were most stimulated to form both
acinar and glandular structures, which were found to co-occur in 4/6 (67%) of the grafts.
LSC
hi
untreated and rAnxA1 treated structures displayed all three phenotypic
G1360 G1361 G1362 G1363 G1364
G1357
Group#2: LSC
hi
+ rAnxa1 5 μM
Group#1: LSC
hi hi
Control
G1365 G1366 G1367 G1368
Group#3: LSC + Ac2-26 50 μM
G1371 G1373 G1375 G1376
G1392 G1393 G1394
Group#3: LSC
hi
+ Ac2-26 50 μM
G1395
86

morphologies, whereas Ac2-26 treated grafts only displayed incidence of low cuboidal
glandular structures (Figure 3.17A, B).

Figure 3.17 Incidence and types of glandular structures seen in LSC
hi
grafts. A, Chart
depicting the incidence and type of glandular structures (prostatic glandular or acinar) scored in
LSC
hi
untreated control, Ac2-26 treated and rAnxA1 treated grafts. B, Representative images of
H&E of the types of structures detected in rAnxA1 treated grafts. Magnification at 400x. C, IHC
analysis of AnxA1 ligand-receptor pErk1/2 downstream activation in glandular structures.
Representative images of pErk1/2 expression in LSC
hi
untreated control, Ac2-26 treated and
rAnxA1 treated grafts are shown. Five to six glandular structures from tissue sections per group
were stained for pErk1/2 expression.

Representative histological analyses of all three groups revealed that Ac2-26 and rAnxA1
stimulated LSC
hi
cells formed grafts containing both p63
+
and CK8
+
cells. As confirmed
by Ki67, Cre and AR staining, these indeed represented proliferating tumor cells from the
mouse model (Figure 3.18).
A
Chart: # Grafts with
Glandular Structures
LSC
Types of structures in rAnxA1 treated. B
hi
control
LSC
hi
+ Ac2-26
2/5
3/5
Glandular Structures
LSC
hi
+ rAnxa1 6/6
LSC
hi
control
LSC
hi
+ Ac2-26
1/5
0/5
Acinar Structures
LSC
hi
+ rAnxa1 4/6
C
Control Ac2-26 rAnxa1
pErk1/2 Expression
Acinar Simple Cuboidal

Low Cuboidal
87

A
p63 AR
Untreated Control LSC

Figure 3.18 Histological images of representative untreated control, Ac2-26 treated and
rAnxA1 treated LSC
hi
grafts. Tissue sections were analyzed by H&E for basic histology and IHC
for expression of basal cell marker p63, luminal cell marker CK8, androgen receptor (AR) and
Ki67. Magnification at 400x.

Contrastingly, LSC
hi
untreated control grafts were mostly comprised of Ki67
-
, CK8
+
cells
indicating less-proliferative, differentiated cells. Additionally, treatment with Ac2-26 and
rAnxA1 led to a significantly greater increase in p63
+
cells (Ac2-26, P < 0.01; rAnxA1, P
< 0.001) and ratio of p63
+
to CK8
+
cells per graft area (Ac2-26, P < 0.05; rAnxA1, P <
0.05), compared to controls (Figure 3.19). In rAnxA1 and Ac2-26 treated grafts where
glandular structures were detected, there was an increased number of structures compared
to controls, displaying larger areas of the graft covered by glandular formation. The size
and density of glandular structures varied between treated and control grafts, as well as
hi
+ UGSM
CK8 Ki67 H&E
Ac2-26 Treated LSC
hi
+ UGSM
rAnxA1 Treated LSC
hi
+ UGSM
400x
88

between structure types. Typically, AnxA1 (Ac2-26 and rAnxA1) stimulated cells formed
larger structures than in controls, multiple structures per graft were present and in the
case of acinar structures, remarkably higher cellular density was observed (Figure 3.24
and Figure 3.25).

We next analyzed expression of pErk1/2 following induction through AnxA1-mediated
receptor activation. Glandular structures possessed pErk1/2 positive foci, indicating a
possible stable up-regulation of the Erk1/2 pathway due to stimulation by AnxA1, as
similarly seen in vitro (shown in Figure 3.12). Controls contained significantly less
pErk1/2 positive foci (Figure 3.17C and Figure 3.19). pErk1/2 expression was positively
correlated with p63 expression (Figure 3.19).

Figure 3.19 Induction by AnxA1 for maintenance of stem cell biological activity. Charts for
calculated percentage of p63
+
cells, ratio of p63
+
to CK8
+
cells and ratio of pErk1/2
+
to Ki67
+

cells detected in the grafts. Potential to undergo de-differentiation in response to AnxA1 was
assessed by increase in p63 expression compared to CK8. AnxA1 ligand-receptor activation was
assessed by increased level of downstream activation of pErk1/2. Statistical significance is
indicated by *, P < 0.05; •, P < 0.01; **, P < 0.001.

The results of renal capsule transplantation of AnxA1 treated LSC
hi
cells faithfully
reiterated the effects of CAF co-transplantation with the CSCs in our previous publication.
As in our earlier study, CAF or in the present situation, CAF secreted factor, generated
% p63+
0
5
10
15
20
25
30
35
Control Ac2-26 rAnxA1
%+ cells/graft
% p63+ : CK8+
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Control Ac2-26 rAnxA1
Ratio %+ cells/graft
pErk1/2+ : Ki67+
0
0.5
1
1.5
2
2.5
3
Control Ac2-26 rAnxA1
Ratio %+ cells/graft
•
**
*
*
89

grafts with 100% incidence of glandular structures, whereas untreated controls exhibited
20%-25% incidence (Liao, Adisetiyo et al., 2010a).

3.3.8 Effect of AnxA1 on LSC
med
Cells in vivo
LSC
med
cells were prepared as described for LSC
hi
cells (see Section 3.3.7) for renal
capsule transplantation procedure (Figure 3.20).

Figure 3.20 LSC
med
graft gross morphology.

LSC
med
cells were stimulated to form prostatic low cuboidal glandular structures in
response to AnxA1 in 4/4 (100%) of rAnxA1 treated grafts and 4/4 (100%) of Ac2-26
treated grafts. Glandular structures were detected in 0/4 (0%) of untreated controls.
Group#5: LSC
med
Control Group#6: LSC
med
G1391
+ Ac2-26 50 μM
G1377 G1378 G1379 G1384 G1385 G1386 G1390
Group#4: LSC
med
+ rAnxa1 5 μM
G1369 G1387 G1388 G1389
90

Acinar glandular structures were present in 2/4 (50%) of rAnxA1 treated grafts, 1/4
(25%) of Ac2-26 treated grafts and 0/4 (0%) of control grafts. Simple cuboidal glandular
structures were detected in only one of the rAnxA1 treated grafts (Figure 3.21A, B).

Figure 3.21 Incidence and types of glandular structures seen in LSC
med
grafts. A, Chart
depicting the incidence and type of glandular structures (prostatic glandular or acinar) scored in
LSC
med
untreated control, Ac2-26 treated and rAnxA1 treated grafts. B, Representative images of
H&E of the types of structures detected in rAnxA1 treated grafts. Black arrow indicates simple
cuboidal structure found in one graft (LSC
med
+ rAnxA1). Magnification at 400x. C, IHC analysis
of AnxA1 ligand-receptor pErk1/2 downstream activation in glandular structures. Representative
images of pErk1/2 expression in LSC
med
untreated control, Ac2-26 treated and rAnxA1 treated
grafts are shown. Five to six glandular structures from tissue sections per group were stained for
pErk1/2 expression.

Histological analysis of LSC
med
grafts confirmed the presence of AR
+
/Cre
+
/Pten
-
prostate
epithelial cells and revealed more p63
+
, CK8
+
and Ki67
+
staining in rAnxA1 and Ac2-26
treated grafts than in controls, with significantly elevated expression of p63 in AnxA1
treated grafts than in controls (Ac2-26, P < 0.001; rAnxA1, P < 0.01); but, overall
slightly lower expression of p63 in AnxA1 treated grafts in LSC
med
than in LSC
hi
(Figure
3.22 and Figure 3.23). The ratio of p63
+
:CK8
+
cells was found to be higher in Ac2-26 and
significantly higher in rAnxA1 treated grafts (P < 0.05) than in untreated controls, in
B Types of structures in rAnxA1 treated. A
Chart: # Grafts with
Glandular Structures
LSC
med
control
LSC
med
+ Ac2-26
0/4
4/4
Glandular Structures
LSC
med
+ rAnxa1 4/4
LSC
med
control
LSC
med
+ Ac2-26
0/4
1/4
Acinar Structures
LSC
med
+ rAnxa1 2/4
C
Control Ac2-26 rAnxa1
pErk1/2 Expression
Acinar Simple Cuboidal

Low Cuboidal
91

which p63
+
cells were virtually absent (Figure 3.23). AnxA1 treatment stimulated LSC
med

cells to form glandular structures compared to controls ( μm: Ac2-26, P < 0.001; rAnxA1,
P < 0.05), and in acinar structures, noticeably higher cellular density was observed.
Multiple structures per graft were more rarely observed in all groups compared to LSC
hi
,
with the notable and significant exception of those grafts containing acinar glandular
structures (Ac2-26, P < 0.01; rAnxA1, P < 0.05). Low cuboidal glandular structures in
both AnxA1-treated groups were smaller than in LSC
hi
; however, acinar structures
observed were larger in LSC
med
than in LSC
hi
grafts leading to overall greater increase in
size of glandular structures in LSC
med
over LSC
hi
AnxA1-treated grafts (NS; Figure 3.24).

Figure 3.22 Histological images of representative untreated control, Ac2-26 treated and
rAnxA1 treated LSC
med
grafts. Tissue sections were analyzed by H&E for basic histology and
IHC for expression of basal cell marker p63, luminal cell marker CK8, androgen receptor (AR)
and Ki67. Magnification at 400x.

A
p63 AR
Untreated Control LSC
med
+ UGSM
CK8 Ki67 H&E
Ac2-26 Treated LSC
med
+ UGSM
rAnxA1 Treated LSC
med
+ UGSM
400x
92

Analysis of pErk1/2 expression confirmed findings in the LSC
hi
grafts, wherein pErk1/2
expression was positively associated with glandular structures and p63 expression.
pErk1/2 expression was, therefore, found to be stably higher expressed in Ac2-26 and
rAnxA1 treated than control LSC
med
grafts (Figure 3.21C and Figure 3.23).

Figure 3.23 Induction by AnxA1 for de-differentiation. Charts for calculated percentage of
p63
+
cells, ratio of p63
+
to CK8
+
cells and ratio of pErk1/2
+
to Ki67
+
cells detected in the grafts.
Potential to undergo de-differentiation in response to AnxA1 was assessed by increase in p63
expression compared to CK8. AnxA1 ligand-receptor activation was assessed by increased level
of downstream activation of pErk1/2. Statistical significance is indicated by *, P < 0.05; •, P <
0.01; **, P < 0.001.

% p63+ : CK8+
0
0.2
0.4
0.6
0.8
1
1.2
1.4
Control Ac2-26 rAnxA1
Ratio %+ cells/graft
% p63+
0
5
10
15
20
25
Control Ac2-26 rAnxA1
% + cells/graft
*
**
•
pErk1/2+ : Ki67+
0
1
2
3
4
5
Control Ac2-26 rAnxA1
Ratio %+ cells/graft
93

Mean No. Acinar Glandular Structures
0
20
40
60
80
100
120
140
LSChi rAnxA1 LSCmed
Ac2-26
LSCmed
rAnxA1
Mean No. Glandular Structures/Graft
*
0
5
10
15
20
25
30
35
40
Control Ac2-26 rAnxA1 Control Ac2-26 rAnxA1
*
•
•
Mean No. Glandular Structures/Graft
•
LSChi LSCmed

Figure 3.24 Incidence and size of glandular structures in LSC
hi
and LSC
med
grafts. AnxA1
treated LSC
hi/med
form more glandular structures than controls. rAnxA1 treated LSC
med
form more
acinar glandular structures than rAnxA1 treated LSC
hi
(P < 0.05) and Ac2-26 treated LSC
med
(P
< 0.01). AnxA1 treated LSC
hi
appear to be more predisposed to form larger structures with less
variation, and AnxA1 treated LSC
med
gain the ability to form glandular structures with more
resulting variation in size, where acinar structures are the largest. Statistical significance is
indicated by *, P < 0.05; •, P < 0.01; **, P < 0.001.

As shown in Figure 3.24, significant differences in size and number of glandular
structures were observed in AnxA1-treated LSC
hi
and LSC
med
grafts, compared to
controls. In further analysis of the differences between prostatic low cuboidal glandular
structures and acinar glandular structures, comparison of the percentages of CK8
+
cells
and ratios of percentages of p63
+
to CK8
+
cells in both LSC
hi
and LSC
med
, reveal that in
acinar structures there is a significant expansion in the number of CK8
+
cells, leading to
lower relative p63
+
:CK8
+
ratios, but accounting for the significantly greater number and
size of glandular structures detected in these groups (Figure 3.24 and Figure 3.25).
Mean Size Acinar Glandular Structures
0
200
400
600
800
1000
1200
1400
1600
LSChi rAnxA1 LSCmed
Ac2-26
LSCmed
rAnxA1
Mean μm/glandular structure
0
100
200
300
400
500
600
Control Ac2-26 rAnxA1 Control Ac2-26 rAnxA1
LSChi LSCmed
Mean μm/glandular structure
Mean Size Acinar Glandular Structures
0
200
400
600
800
1000
1200
1400
1600
LSChi rAnxA1 LSCmed
Ac2-26
LSCmed
rAnxA1
Mean μm/glandular structure
0
100
200
300
400
500
600
Control Ac2-26 rAnxA1 Control Ac2-26 rAnxA1
Mean μm/glandular structure
*
*
**
LSChi LSCmed
94

Figure 3.25 Comparison of p63
+
and CK8
+
cells between glandular morphologies in AnxA1
treated LSC
hi
and LSC
med
grafts. AnxA1 treated LSC
hi/med
acinar glandular structures show
significant expansion of CK8
+
prostate tumor cells compared to low cuboidal glandular
structures, and therefore have lower ratios of p63
+
:CK8
+
cells. Statistical significance is
indicated by *, P < 0.05; •, P < 0.01; **, P < 0.001.

These results can be seen in the immunohistochemical stainings from LSC
hi
and LSC
med
,
where LSC
hi
treated with rAnxA1 and LSC
med
treated with Ac2-26 and rAnxA1 form
significant numbers of acinar structures with clear and distinct histopathologies from the
low cuboidal glandular structures, which are represented in pathological stainings in
Figure 3.18 and Figure 3.22. As seen in Figure 3.26, difference in acinar histology is
most distinctly noticeable for p63 and CK8 expression analyses, where CK8
+
cells are
found to be abundantly present and p63
+
cells occur in rare foci and simple cuboidal
glandular structures (see further description of results from histology stainings in Section
3.3.6).
LSC
hi
+ rAnxA1 % p63+
0
2
4
6
8
10
12
14
16
18
20
Acinar Low
Cuboidal
Combined
%+ cells/graft
LSC
hi
+ rAnxA1 % CK8+
0
10
20
30
40
50
60
70
80
90
Acinar Low
Cuboidal
Combined
%+ cells/graft
LSC
hi
+ rAnxA1 % p63+ : CK8+
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Acinar Low
Cuboidal
Combined
Ratio %+ cells/graft
•
**
LSC
med
+ rAnxA1 % p63+ : CK8+
0
0.2
0.4
0.6
0.8
1
1.2
1.4
Acinar Low
Cuboidal
Combined
Ratio %+ cells/graft
LSC
med
+ rAnxA1 % p63+
0
5
10
15
20
25
Acinar Low
Cuboidal
Combined
%+ cells/graft
A
B
LSC
med
+ rAnxA1 % CK8+
0
10
20
30
40
50
60
70
80
Acinar Low
Cuboidal
Combined
%+ cells/graft
*
*
95

p63 AR CK8 Ki67 H&E
rAnxA1 Treated LSC
hi
+ UGSM

Figure 3.26 Histologies of acinar glandular structures in AnxA1 treated LSC
hi
and LSC
med

grafts. AnxA1 treated LSC
hi/med
acinar glandular structures are more prolific than cuboidal
structures, have more CK8
+
prostate tumor cells, possess p63
+
foci and have a lower resulting
percentage of p63
+
:CK8
+
cells.

The results from quantification of p63
+
and CK8
+
cells, and their ratios in the various
morphological glandular structures, support the possibility that AnxA1 treated cells have
undergone different de-differentiation and differentiation pathways, leading to the
different phenotypes observed in the histology. All three morphological subtypes are seen
in rAnxA1 treated LSC
hi/med
grafts and, moreover, it is more common to note acinar
glandular structures in rAnxA1 treated, than Ac2-26 treated, LSC
hi/med
grafts,
demonstrating the potency of the full-length protein. These findings, taken in
combination with the histological observation of 100% positive correlation between
rAnxA1 Treated LSC
med
+ UGSM
Ac2-26 Treated LSC
med
+ UGSM
400x
96

detection of high p63 expressing simple cuboidal glandular structures with detection of
low p63/high CK8 expressing acinar glandular structures and significant expansion of
cell number (indicated by significant increase in size and number of acinar structures, and
CK8
+
cells), suggest that in vivo AnxA1 full-length protein has the capacity to generate
multiple differentiation schemes, and support the expansion of differentiated cells and the
maintenance of cells with high expression levels of basal/progenitor cell marker p63.

Figure 3.27 AnxA1 supports p63
+
cells and proliferation, and therefore may support the stem
cell niche. Incidence of simple cuboidal glandular structures with highest expression levels of
basal/progenitor marker, p63, is 100% positively correlated with detection of acinar glandular
structures. Acinar glandular structures display the highest levels of cell expansion and size of
resulting structures formed. It is possible to speculate that AnxA1 is responsible for both the high
expression levels of p63 and the significant proliferation of acinar glandular structures. These
observations are not seen in untreated LSC
med
controls and are reduced in LSC
hi
controls,
supporting the action of AnxA1 in conferring biological activities seen in treated grafts.

A
low high
low high
Cuboidal
basal cells
↑% p63+
↓% p63+
B
Stem(-like)
cells
p63
pErk1/2
Stem(-like)
cells
↑% acini
97

Figure 3.27 illustrates the correlation between simple cuboidal glandular structures with
high levels of p63 expression and large acinar glandular structures.

To sum up, LSC
med
do respond to CAF-secreted signaling factors, possibly through
induction of EMT to generate cells with stem cell-like properties that are capable of
forming complex glandular structures; however, LSC
hi
appear to contain more of the
intrinsic properties for glandular structure formation and stimulation by CAF-secreted
AnxA1 in vivo.

98

3.4 Discussion
In this chapter, we sought to identify secreted factors from the CAFs that can induce
EMT and contribute to the generation of cancer stem cells and progression to
adenocarcinoma and metastasis. We have demonstrated in a previous publication that
CAFs may stimulate the proliferation and tumorigenicity of prostate cancer in vivo.
Therefore, based upon these prior observations, we proposed that the CAF generate
signals that positively regulate prostate cancer. In order to test this hypothesis, primary
cultures of NPF, CAF and UGSM were used for both co-culture and CM experiments to
assay for the ability of prostate cancer cells to undergo EMT in the presence of signaling
factors from the tumor microenvironment. Pioneering work in this manuscript evidences
that CM from the CAF is sufficient to induce EMT de-differentiation of prostate cancer
epithelial cells from the cell line in vitro and in vivo. From this we conclude that the CAF
secrete positive regulators of EMT and stemness.

EMT is a well recognized mechanism for initiating metastasis of epithelial cancers.
During development and wound healing, the function of EMT as a de-differentiation
process allows otherwise fated epithelial cells to alter their lineage specificity and gain a
more motile, plastic, and even stem-like phenotype. Shared attributes between EMT and
CSCs are being uncovered and extracellular signals from the microenvironment which
regulate EMT are being investigated to define their role in a potential Cancer Cell-EMT-
CSC lineage (Al-Hajj et al., 2003; Mani et al., 2008; Santisteban et al., 2009; Thiery et al.,
99

2009). Considering the possible role of EMT in the development of cancer
stem/progenitor cells, we included EMT markers in gene expression analyses.

Proof-of-principle experiments for this manuscript were done using the androgen
depletion-independent cell line cE1. The rationale for this decision was based on
published in vivo observations and unpublished in vitro findings regarding the
epithelial/mesenchymal properties of both androgen- dependent and independent cell
lines derived from the cPten
-/-
L model. Cell lines from the “E” series (E2, E4) were
established from an androgen-dependent tumor and xenografts in NOD.SCID mice
inoculated with these cells formed tumors that have been pathologically characterized as
carcinosarcoma (epithelial-mesenchymal), which is additionally reflected by their
“hybrid” phenotype and higher expression levels of mesenchymal rather than epithelial
cell markers. Unpublished in vitro work indicates that E2 and E4 do not undergo EMT
upon stimulation with the potent EMT-inducer TGF β. In contrast, tumors derived from
the CRPC “cE” series were determined to be histologically like adenocarcinoma, display
higher levels of epithelial markers and show appreciable induction through EMT by
TGF- β1 (Chapter 2; (Liao, Liang et al., 2010)).

To determine whether subpopulation(s) of tumor epithelial cells were most susceptible to
stimulation from CAFs in vitro, first, cE1 cells were assayed for their differential
response to treatment by CAF CM and CAF CM protein fractions enriched for AnxA1,
and second, primary tumor epithelial cells were similarly exposed to CAF CM and CAF
100

CM fractions. In the cE1 cells, expression levels of Oct4, Sox2, Nanog, EMT
transcription factors Snail, Slug and Twist, epithelial marker E-Cadherin and
mesenchymal markers N-Cadherin and Vimentin were assessed in SC
hi
, SC
med
and SC
none

cells that were exposed to CM from either CAF or NPF cultures and monitored for
morphological changes over a period of 14 days. SC
med
cells were preferentially
susceptible to EMT after treatment with CAF CM based on (1) extensive EMT-like
phenotypic changes induced by CM from CAFs, and (2) significant up-regulation of stem
cell and EMT transcription factors, mesenchymal markers and down-regulation of the
epithelial marker, E-Cadherin.

CAFs differentially affect spheroid-forming efficiency when co-cultured with CSCs from
the primary tumor epithelial cells of the cPten
-/-
L model. Compared to both LSC
med
and
LSC
none
cells, which did not display a high propensity to form spheroids in vitro, LSC
hi

cells readily formed spheroids, indicating the possible existence of CSCs within this
fraction of the tumor, and were stimulated two-fold by the presence of CAFs in regard to
spheroid formation (Liao, Adisetiyo, Liang, & Roy-Burman, 2010b). In continuation of
this characterization of the differential effect of CAFs on prostate cancer cells, we
decided to next test whether CAF-derived AnxA1 directly affects the ability of the CSCs
to form spheroids, as shown in the results in Section 3.3.5.

To substantiate the in vitro findings of AnxA1’s biological activity on the different
subpopulations, all three AnxA1 treated cell types—EMT-susceptible cE1 SC
med
and
101

spheroid proliferation-induced LSC
hi
and LSC
med
—were tested in vivo using renal
capsule transplant technique (Liao, Adisetiyo et al., 2010a). Combined in vitro and in
vivo results from the cell line suggest that cE1 SC
med
cells may be able to undergo a de-
differentiation process in response to soluble secreted factors from the CAF CM and
AnxA1. Strong evidence in support of this theory is the higher incidence of p63
+

structures after treatment with CAF CM and AnxA1 mimetic peptide. The majority of the
SC
med
cells represent transit-amplifying and luminal differentiated cells and produce
grafts with high expression of luminal cell marker CK8 and low expression of basal cell
marker p63 as seen in the controls for the experiment. However, treated SC
med
cells give
rise to glandular structures with both basal and luminal layers, with expansion of the
basal compartment. SC
med
de-differentiation due to exposure to AnxA1 may be mediated
through EMT, as observed in vitro. The observation that AnxA1 mimetic peptide was
faithfully able to generate glandular structures histopathologically similar to those formed
after treatment with CAF CM suggests that AnxA1 may be an important secreted
molecule in this process. As shown in in vitro experiments in this study AnxA1 is a
positive regulator of induction through EMT. AnxA1-induced EMT in SC
med
cells may
generate cells with stem cell-like properties that may give rise to multi-differentiated
prostatic glandular structures in vivo.

Two distinct phenotypes were observed after treatment of primary prostate
adenocarcinoma cells with AnxA1. Typical prostaspheres and atypical acini were able to
grow in culture and in parallel, glandular structures and acinar structures developed in
102

vivo. Using the cE1 cell line, acinar spheroids were not observed and the formation of
acini in vivo was limited to one graft. LSC
hi
cells were capable of being stimulated by
AnxA1 both in vitro and in vivo to generate spheroids and glandular structures with acini,
respectively. In vitro, both rAnxA1 and Ac2-26 could induce acini in spheroids. In vivo
rAnxA1 appeared to be a more potent inducer of acinar formation than Ac2-26 in LSC
hi

grafts as four-fold more grafts were observed with acini from treatment with rAnxA1,
and two-fold more in LSC
med
. Furthermore, acinar structure formation in LSC
med
cells
was only present after cells were exposed to AnxA1, both Ac2-26 and rAnxA1. Similarly
rAnxA1 treatment was correlated with greater incidence of acinar formation than
treatment with Ac2-26 and these findings are significant in LSC
med
(P < 0.01); however,
future studies should include larger animal cohorts.

One possibility for the more diverse array of effects seen after treatment with the full-
length versus the mimetic form of AnxA1 in LSC
hi
and LSC
med
grafts is the presence of
the core domain in the full-length—a pore shaped by the 4 α-helical repeats which create
a “doughnut-like” ring around the N-terminal domain and may, thus, serve in physically
interacting with an adjacent receptor, another AnxA1 monomer or mediating protein
(Gerke et al., 2005). Another possibility is that the binding efficiency of Ac2-26 may not
be as stable without the core domain. The other possibility is that the promiscuity of
mimetic Ac2-26 allows for the activation of additional downstream pathways of Fpr1 and
Fpr-rs1, whereas the restrictive interaction of the full-length leads to activation of
signaling downstream of Fpr-rs1 only.
103

Our findings using primary cultures of LSC
med
cells from the mouse model differ
somewhat from the results of the cell line and indicate that follow-up in further studies
should be pursued. Most prominent is the outcome of the renal capsule transplantation
experiment. While the incidence of multiple glandular structure formation and glandular
size, exclusive of acinar structure size, were lower in LSC
med
cells compared SC
med
cells,
when LSC
med
AnxA1-treated cells were compared to LSC
med
control cells, an increase in
glandular structure formation was noted after treatment with both mimetic and full-length
protein. This finding does support that AnxA1 is able to stimulate the formation of
complex, multi-layer glandular structures in the Sca-1
med
/CD49f
med
subpopulation of the
tumors. LSC
med
cells should be investigated in future works for indication of induction
through EMT.

Analyses using a neutralizing antibody against AnxA1 indicate a possible mode of
mechanism for AnxA1 predominantly upon Snail, Twist and Oct4 dependent pathways
through induction of EMT. Current study of EMT and stem cell transcription factor
expression after inoculation with AnxA1 is limited to the cell line cE1. This effect of
AnxA1 on induction of EMT should be further explored in additional cell lines and in the
primary cultures. According to preliminary results of western blotting and IHC (data not
shown), expression of Snail and Oct4 in primary cultures of LSC
hi
cells may also be
affected by presence of AnxA1. These analyses should be extended to epithelial
subpopulations in the cPten
-/-
L model and additional models of prostate adenocarcinoma
104

to thoroughly investigate the relationship between AnxA1, EMT and stem cell-like
properties in prostate cancer.

AnxA1 has been implicated as both a pro-tumorigenic and anti-tumorigenic factor in a
variety of cancers including breast, pancreatic, bladder, esophageal, head and neck and
prostate (L. H. Lim & Pervaiz, 2007). We have shown that interactions between AnxA1
and different effectors, such as the TGF β pathway and the murine equivalent to
FPR2/ALXR, Fpr-rs1, may mediate differential effects of AnxA1 treatment, as well as
response differences between subpopulations. Our findings with the TGF β pathway and
signaling activation downstream of Fpr-rs1 may suggest distinct molecular mechanisms
by which AnxA1 exerts its effects, but future work should address the importance of
these pathways in AnxA1-mediated EMT and stem cell-like properties. It is possible to
speculate that post-translational modifications may also allow for differences in AnxA1
behavior across human neoplasms and experimental analysis of kinases and other known
regulators could reveal clues as to the tissue-specific effects of AnxA1 in varying cancers.

Overall, these findings support the importance of CAFs from the microenvironment in
regulating the tumorigenic and stem cell-like potentials of the cancer cells. Key players in
tumor-microenvironment cross-talk may be implicated using methods presented in this
study. Identification of CAF-secreted factors may present potential therapeutic targets
and strategies to interfere with these molecular interactions between the tumor and the
105

microenvironment should include careful analyses of tumor subpopulations and their
individual responses to the targeted secreted factors.
106

Chapter 4:
Concluding Remarks & Future Directions
“Cancer can take away all of my physical abilities. It cannot touch my
mind, it cannot touch my heart, and it cannot touch my soul.”
—Jim Valvano, American college basketball coach, ESPY Awards speech, 1993
4.1 Summary
Cancer disease progression is a multi-faceted process. Many factors both within and
without the tumor cells contribute to the evolution of the cancer and contribute toward a
potential metastatic outcome. Behind the sequence of events that lead a tumor into
metastasis are the roles of cancer stem cells (CSCs), epithelial-to-mesenchymal transition
(EMT) and the tumor microenvironment. Cancer stem cells display properties of normal
stem cells with regard to self-renewal, multi-lineage differentiation potential and
increased resistance to drug toxicity.

As outlined in Chapter 1, scholars have shown that EMT generates cells that have the
characteristics of cancer stem cells. Successful transdifferentiation through an EMT
process results in transcriptional upregulation of known stem cell transcription factors,
notably Oct4 and miR-200, expression of cell surface stem cell markers, and the ability to
seed new tumors and recapitulate the different cell types of the original tumor in vivo.
EMT also confers motility to otherwise stationary epithelial cells by changing cell-cell
focal adhesions and attachment to the basement membrane. Repression of cadherins leads
to dissociation of cadherin-catenin intercellular junctions, while increased expression of
107

MMPs degrades the extracellular matrix (ECM), culminating in increased cell movement.
Metastasis occurs when cells that are capable of surviving outside their tumor
microenvironment and possess the ability to seed new tumors gain motility, invade past
the basement membrane and intravasate into the blood circulation. Because of the
similarities between the properties of cancer stem cells, cells that have undergone an
EMT and metastatic cancer cells, one widely accepted proposal is the generation of
cancer stem cells and metastatic cancer cells through EMT. Many different factors have
been shown to contribute to and promote EMT, including factors produced by the various
cell types of the tumor microenvironment. One of the most prominent cell types of the
microenvironment are stromal fibroblasts, which in normal physiology provide regulatory
cues to the epithelial cells they surround and maintain. In cancer, the CAFs secrete
factors which promote angiogenesis, growth and invasion of cancer cells, EMT and, in
proposal, support the generation of CSCs.

As this dissertation demonstrates, cancer-associated fibroblasts (CAFs) regulate prostate
cancer stem cell biology, in particular the genesis and/or maintenance of CSCs. I
conclude, based on the results in Chapter 2, that tumor-derived prostate epithelial cells
that undergo an EMT in vitro are able to generate a population of cells with gene
expression profiles indicative of stem cell characteristics and that may, therefore, be
capable of acting as stem-like progenitor cells. The early research in Chapter 2 provided
clues to an origin for prostate cancer stem cells which may be generated in vivo by
induction through an EMT due to secreted factors from the tumor microenvironment. As
108

shown in Chapter 2, CAFs do secrete factors that promote stem cell characteristics in the
prostate epithelial tumor cells and AnxA1 is identified as one of the prominent factors.

In Chapter 3 of my study, I further sought to characterize the role of AnxA1 on the
prostate cancer cells and CSCs. I found that AnxA1 is capable of both promoting de novo
generation of CSCs via EMT and maintenance of an existing CSC population.
Experimental results from Chapter 3 describe evidence that supports this conclusion. As
shown in in vitro experiments in this study, AnxA1 is a positive regulator of induction
through EMT. AnxA1-induced EMT in SC
med
cells may generate cells with stem cell-like
properties that give rise to multi-differentiated prostatic glandular structures in vivo.
Primary prostate cancer cells—LSC
hi/med
—respond to AnxA1 treatment via enhanced
biological activity and evidence of an induction through EMT. Primary cell responses in
agreement with enhanced biological activity are defined in vitro through formation of
greater numbers of spheroids and increased complexity (seen in both LSC
hi
/LSC
med
) and
in vivo through generation of more and larger glandular structures, with more complexity
and greater pErk1/2 activation, and a displayed increase in p63
+
cells to CK8
+
cells in
LSC
hi
(Ac2-26, P < 0.05; rAnxA1, P < 0.05). LSC
med
may be most susceptible to
undergo EMT and gain stem cell-like properties. In vitro, LSC
med
gain ability to form
spheroids and in vivo, LSC
med
gain (a) ability to form detectable glandular structures
compared to controls, (b) multi-potential differentiation and significantly higher p63
+

basal/transit-amplifying cells than CK8
+
luminal cells (rAnxA1, P < 0.05), and (c)
formation of significantly larger glandular structures compared to controls (Ac2-26, P <
109

0.001; rAnxA1, P < 0.05) and treated LSC
hi
cells (NS) as well as greater number of
pErk1/2
+
cells.

110

4.2 General Discussion
It is possible to speculate that post-translational modifications also allow for the
differences in AnxA1 behavior across human neoplasms and experimental analysis of
kinases and other known regulators could reveal clues as to the tissue-specific effects of
AnxA1 in varying cancers. For example, decreased proteolysis of AnxA1 in conjunction
with increased activity of known phosphorylator PKC of Ser
27
residue on AnxA1 within
CAFs could possibly account for the elevated levels of secreted AnxA1 in prostate
cancer.

Two possibilities exist for this Ser
27
directed secretion of AnxA1 across the plasma
membrane. Pharmacological studies support that it is exported by an ABC transporter
(Chapman et al., 2003). Speculation based on the amphipathic properties of AnxA1
suggest that it may additionally adopt different conformations, with or without a
chaperone, to pass through membrane pores or channels (Payne et al., 2005). Although
the secreted form is indistinguishable from intracellular AnxA1 with regard to its amino
acid composition, two-dimensional gel mobility and phospholipid-binding activity, it
does possess an acetylated (blocked) amino terminus that is resistant to Edman
degradation (Christmas et al., 1991).

More of the diversity of AnxA1-mediated effects may be mediated by three possible
modes of action for binding to its receptor. In the case of AnxA1-induced phagocytosis of
apoptotic cells, AnxA1 physically interacts with the “eat me” signal phosphatidylserine
111

that is exposed on the outer layer of the plasma membrane (Gerke & Moss, 2002) and can
do so in one of three ways. Upon membrane-binding of the core domain, the exposed N-
terminus of AnxA1 can (1) bind to a second membrane (Bitto, Li, Tikhonov, Schlossman,
& Cho, 2000; Rosengarth, Gerke, & Luecke, 2001), (2) dimerize to another molecule of
AnxA1 and thereby pull two membranes together (Lambert, Gerke, Bader, Porte, &
Brisson, 1997) or (3) bind to other proteins, such as S100, to form complexes that create
a junction between two membranes (Mailliard, Haigler, & Schlaepfer, 1996; Seemann,
Weber, & Gerke, 1996).

In the case of AnxA1 binding to the FPR2/ALX receptor one possible reason for the
more diverse array of effects seen after treatment with the full-length versus the mimetic
Ac2-26 peptide is the support of the N-terminal domain of the full-length by the pore
from the 4 α-helical repeats, which create a “doughnut-like” ring around the N-terminal
domain, and may, thus, stabilize physical interaction of the N-terminus of the full-length
protein with the bound receptor and with an adjacent receptor or protein. The other
possibility is that the stabilized interaction of the full-length leads to greater activation of
signaling downstream of Fpr-rs1 (and Fpr1 by up to 30%), whereas the lower stability
and promiscuity of mimetic Ac2-26 leads to activation of all three murine receptors and
additional downstream pathways, albeit with less potent constitutive activity.

Another significant result of this dissertation work is the clear association between Oct4
expression and a Cancer Cell-EMT-CSC lineage. Recent reports have implicated Oct4 as
112

an important prognostic tool for detection of aggressive tumor subtypes in a variety of
human cancers. Additionally, Oct4 has been suggested to be a crucial player in
maintaining stem cell characteristics of cancer cells and chemoresistance in both human
cancers and animal models; however, to date the exact role of Oct4 and its expression
patterns in human cancer have remained largely undefined. I hope that my research may
shed some added light on the importance of this stem cell regulatory gene and
transcription factor in the progression and metastatic transformation of prostate cancer
and other solid cancers.

In my model system using cE1 subpopulations to define intrinsic differences among
prostate epithelial cancer cells, I found that the subpopulation with the highest expression
levels of Sca-1 (S) and CD49f (C), SC
hi
—previously reported to be enriched for the
putative CSCs of the cPten
-/-
L model of prostate adenocarcinoma (Liao, Adisetiyo et al.,
2010a), also displayed marked expression of Oct4, and, furthermore, an SC
med

subpopulation of differentiated cancer cells could be induced to upregulate Oct4
expression through an EMT. This demonstrates the potential plasticity of cancer cells to
alter their genetic programming in response to external environmental cues.

The potential implications of these findings for cancer therapy are significant. To date,
although published literature has identified purified subsets of solid cancer cells that are
capable of forming new tumors (Al-Hajj et al., 2003; Mani et al., 2008), the
phenotypically heterogeneous cancer cells of solid tumors are treated equally in clinical
113

measures (Reya et al., 2001). Conventional therapies target tumors by killing mainly cells
with limited proliferative potential; however, CSCs would remain viable in the body after
the course of therapy and would be capable of re-seeding the tumor. In this instance, the
tumor shrinks initially in response to therapeutic treatment, but then grows back. If by
identifying CSCs through functional definitions, such as high Oct4 expression, and CSCs
are able to be targeted specifically, then therapies may be much more successful in fully
eradicating the CSCs and the tumors.

Tangentially connected to this is the observation that cancer cells possess plasticity. The
goal in CSC research for therapeutics has been to identify and isolate the subset of cancer
cells that can lead to metastatic disease, but what if, as my research shows, heterogeneous
populations of cancer cells that appear differentiated may undergo plastic conversion to a
more de-differentiated CSC state? What then? This harks back to the initial question
proposed in the introduction in Section 1.3.1 in Chapter 1: are cancer stem cells a rare
subset of the overall cancer cell population or do all heterogeneous cancer cells possess
the ability to respond to extrinsic environmental cues to become cancer stem cells? If the
results of this dissertation study are accepted in support of the extrinsic model for CSCs,
as I believe that they will be, then a proposed new therapeutic approach would be to not
only target the identified CSCs, but also, to target the essential environmental cues that
could, if left unaltered, allow for the emergence of new pools of CSCs during the course
of chemotherapy (this proposal is further explored in this chapter, Section 4.4.2).
114

4.3 Limitations of the Study
4.3.1 Confirmation that CAF Secrete Higher Concentrations of Phospho-Ser
27
-
AnxA1 than NPF
In assaying for presence of the phospho-Ser
27
species of AnxA1 within the fractions of
the CAF versus NPF CM ammonium sulfate enriched fractions, I discovered that I was
unable to detect phosphorylation signal within the fractions of the CM. Two plausible
possibilities exist that explain this result and still affirm the findings from the CAF
lysates. The first possibility is due to the lower detection limit of the anti-phospho-Ser
27

motif antibody. It is evident that within the CAF lysates the phospho-Ser
27
–AnxA1
species represents only a fraction of the total. This would be expected given the fact that
AnxA1 plays intracellular roles and that not all of the AnxA1 protein translated within
the CAF cells would be destined for phosphorylation or that all molecules of AnxA1
within a cell would have already been phosphorylated for secretion at any given time.
Support for this supposition comes from my work on the western blots, wherein the
amount of AnxA1 present within the CM has been shown to be remarkably less than the
total amount present within the lysates. Continuing along this line of logic, and taking
into further consideration the lower abundance of AnxA1 within enriched fractions of
CM versus the lysates, it is highly possible that the concentration conditions of the
antibody and/or protein loaded into polyacrylamide gel lanes were not sensitive enough
to detect the phospho-Ser
27
–AnxA1 species in the CAF CM.

115

The second very plausible explanation is degradation leading to loss of phosphorylation
signals within the CAF CM, since they have not been stabilized with the addition of
phosphatase inhibitors. The resultantly lost or reduced phosphorylation signal would be
even more difficult to detect given the lower relative sensitivity of the anti-phospho-Ser
27

antibody versus the anti-AnxA1 antibody. Support for this possible theory regarding the
lack of detectable phospho signal is the difference in overall staining using the anti-
phospho-Ser
27
antibody on CM blots versus lysate blots. Even after doubling the
concentration of the anti-phospho-Ser
27
antibody and incubating with blots for 48 hours
at 4° C, followed by development with secondary antibody and chemiluminescent
detection of the hydrogen peroxide substrate, no detectable signal of any kind anywhere
on any of the blots was found. Running the lysates from the same NPF and CAF cultures
used for the attempted CM detection westerns, the anti-phospho-Ser
27
antibody bound to
not only my protein of interest, but also two large proteins present in selected CM
fractions in high quantity. Because the anti-phospho-Ser
27
antibody recognizes all
proteins with the specific signal recognition motif necessary for PKC phosphorylation on
Serine, other proteins of different size were predicted by a Cell Signaling Technology
database program. With the lack of phosphorylation signal in general on CM blots, I
believe that the phosphorylation signal has been lost during the process of collecting and
concentrating the CM.

There are three possible experiments that could answer the question of whether
phosphorylation signals have been lost due to reduced sensitivity or experimental
116

technique. The first is the use of a general anti-phospho-Ser antibody to verify if total
phosphorylation signals within the CM are lost after (1) concentration and (2) ammonium
sulfate precipitation. Secondly, addition of phosphatase inhibitor to cell culture at the
beginning of conditioning or at the point of conditioned media collection may stabilize
phosphorylation signals. One reason for degradation in the non-stabilized CM is that
phosphorylation may have been degraded by soluble phosphatases from lysed cells and
cell debris that remained soluble and therefore present in the CAF CM after collection.
During concentration these diluted phosphatases would have then been put in higher
concentration with secreted proteins with phosphorylation. Thus at concentrations that
were more favorable for reaction kinetics these phosphatases may have been able to
degrade phosphorylation signal from AnxA1 and other soluble CM proteins. Conditioned
media must be concentrated in order to load enough protein per lane to be detectable by
western blot. The amount of phosphatase inhibitor that would be necessary for 1x final
concentration may become excessive. Therefore, I would like to test if I could add
phosphatase inhibitor at a concentration such that after Amicon filter concentration, the
inhibitor would then be at a final 1x concentration in order to preserve the signal.

Finally, the third proposal is to immunoprecipitate total AnxA1 from NPF and CAF CM
using A/G beads and antibody for AnxA1 followed by western for anti-phospho-Ser
27
.
This approach would validate whether the phosphorylated species is present at detectable
levels within the CM. It would be inferred from positive detection of phospho-Ser
27
–
117

AnxA1 that this species is secreted into the CM. I would additionally confirm successful
precipitation of total AnxA1 protein after immunoprecipitation (IP) with western.

4.3.2 Comparison of the Effects of Murine Endogenous, Mimetic N-terminal-
Derived and Murine Recombinant Forms of Annexin A1 on Prostate Cancer Cells
in vitro
Relevant work may still be completed on the in vitro cultures of the LSC
hi
and LSC
med

cells with or without treatment by Ac2-26 or rAnxA1. At the time of passaging primary
cultures for making collagen grafts for the renal capsule transplant experiments, I
collected frozen, lysis buffered and 4% paraformaldehyde (PFA) fixed samples from each
control and treatment group for analysis by qRT-PCR, western and immunohistochemical
staining, respectively. Two independent groups of primary cells were cultured and
ultimately used for grafting, thus, these serve as independent repeat samples for the
aforementioned techniques. Possible analyses that I may run should substantiate the work
that has already been done with the cE1 cell line. I would propose to run qRT-PCR
analysis for Oct4, Snail, Slug, Twist, TGF βRII and MAPK. Western blot should confirm
the results from earlier culture analyses for pErk1/2 and TGF- β1 (see results in Chapter 3,
Section 3.3.5). And, finally, using immunofluorescence I may analyze p63
+
and CK8
+

cells within control and treated spheroids for the purpose of basal cell expansion
quantification based on ratio of p63
+
to CK8
+
cells.

118

4.4 Future Research
4.4.1 Future Research: Contribution to the Field
It is my fervent hope that other scientists in the field of cancer research will pursue their
own unique interest in this body of work. For future research, a proposed model for
AnxA1 mechanism of action in prostate cancer stem cell biology is illustrated below.

Figure 4.1 AnxA1 promotes both EMT and differentiation processes, leading to de novo
generation of CSCs and maintenance of existing CSCs. Diagram generated using Ingenuity
Systems, Inc. software (IPA) depicting interactions between AnxA1, AnxA1-S
27
-phosphorylator
PKC, TGF β, and murine AnxA1 FPR receptor Fpr-rs1. Based on published literature, IPA
pathway analysis and results of the current study, we propose a model for AnxA1 secretion from
CAFs and effect on prostate tumor epithelial cells. TGF β increases recruitment of PKC species in
CAFs, leading to phospho-S
27
-AnxA1 secretion. Secreted AnxA1 influences prostate tumor
epithelial cells in two ways. AnxA1 leads to upregulation of TGF β, and AnxA1 physically binds
with its receptor Fpr-rs1 to activate phosphorylation of Erk1/2. Intersection of AnxA1 and TGF β
pathways may regulate EMT, whereas pErk1/2 pathway signaling may influence proliferation
and differentiation.

Erk1/2
FPR
Differentiation
Proliferation
EMT, “stemness”
119

Figure 4.1 depicts the dual biologies seen in AnxA1-treated prostate cancer cells from the
conditional Pten deletion mouse model. Adapted from the Abstract, AnxA1 from CAFs
contributes to tumor stem cell activity via two separate but complementary pathways:
1) induction of a de-differentiation process leading to generation of basal stem-like cells,
and
2) enhancement of the proliferation and differentiation of the cancer stem cells.
As explained in Chapter 1, Section 1.2.2, in the hypothesis rationale, my findings in this
dissertation body of work implicate AnxA1 in a paradigm in which cells that are not
cancer stem cells can be induced to gain cancer stem cell-like properties, and cells that
already possess the properties of cancer stem cells can be enhanced for their biological
activities (i.e., activation, proliferation and differentiation). Future research examining
AnxA1 in cancer biology should focus on further understanding the differences between
these two processes and whether different subpopulations of tumor cells are more
susceptible to one process or another. I will suggest the possible hypothesis that both
LSC
hi
and LSC
med
subpopulations may respond through both TGF β-supported and
FPR/ERK1/2-supported mechanisms, but in each subpopulation, one response may be
more favorable than the other. For example, LSC
med
cells show activation of Erk1/2, and
therefore activation of the AnxA1 receptor Fpr-rs1; however, they may respond
predominantly via an AnxA1-mediated EMT involving TGF- β1 or TGF βRII, predicted
by their lower basal expression levels of EMT/stem cell transcription factors. Thus,
LSC
med
may be more “primed” to undergo EMT than LSC
hi
.

120

4.4.2 Relevance of the Work to the Clinical Treatment of Prostate Cancer
Cancer cells are always evolving, developing resistance to chemotherapies, and prove to
be a difficult target. In contrast, fibroblastic cells of the tumor microenvironment are less
evolving and therefore are a better target for precise drug delivery. Chapters 2 and 3 of
this dissertation illustrate that AnxA1 is an important factor coming from the prostate
stroma, and has been shown to support the prostate cancer stem cell niche. AnxA1
supports the prostate cancer stem cell niche through maintenance and de novo generation
of basal stem-like cancer cells in vivo, therefore therapies interrupting this process could
prove to be a better approach to eradicating prostate cancers. Moreover, it would be
important to target both stromal-derived AnxA1 (i.e. phospho-Ser
27
-AnxA1) and the
prostate cancer cells as a more effective strategy for eliminating the primary cancer and
preventing recurrence.

Many pleiotropically acting protein factors have been identified from the tumor
microenvironment. A question arises of “How many of these chemokines, cytokines and
growth factors, etc. are needed?” Is there such a thing as an essential “secretome” of the
tumor microenvironment that would confer such specific changes to the epithelial cells
that they would always in that given tumor microenvironment be stimulated to become
more stem cell-like. The question that is really being asked is: “Which component would
be the most beneficial to target given these conditions of the tumor microenvironment
that support the generation of stem cell properties within the cancer cells?” Actually,
there may not be one most essential component. Rather, which one could be more
121

essential for the generation of cancer stem/tumor-initiating cells and when treated appears
to have conferred the most stimulus to that effect.

As described in this study (for further reading, see Chapter 2, Section 2.4.2, and Chapter
3, Section 3.3.5), AnxA1, TGF- β1 and TNF α, all derived from the tumor
microenvironment, may be found to have a significant synergistic effect on pro-
tumorigenic processes, including EMT driven by a resulting constitutive upregulation of
Snail expression and activity. It would be worthwhile from a clinical setting to test the
potential of AnxA1 to determine if it is necessary and sufficient to generate cancer stem
cells.
122

Chapter 5:
Materials and Methods
5.1 Experimental Animals
All animal studies were pre-approved by and used in accordance with the University of
Southern California Institutional Animal Care and Use Committee (IACUC) guidelines.
Mice were housed in autoclave sterilized filter-capped microisolator cages, with five
mice per cage, and fed autoclaved food and given filtered water ad libitum, unless
otherwise noted. Pups were weaned at 21 days old, when they were subsequently tagged
and genotyped. To euthanize the mice, they were placed in a standard CO
2
chamber
attached to a pressurized CO
2
tank. The mice were exposed to the gas for 5 minutes to
attain complete asphyxia. Inhalation of CO
2
was followed by cervical dislocation to
ensure sacrifice of the animals.

5.1.1 Conditional Pten Deletion Mouse Model (cPten
-/-
L)
The conditional Pten deletion mouse model (cPten
-/-
L) used in the current work was
described previously (Liao et al., 2007).

5.1.2 Immunodeficient NOD.SCID Mice
For renal tissue grafting, non-obese diabetic/severe combined immunodeficient
(NOD.SCID) mice, purchased from National Cancer Institute (Frederick, MD), were
used. Renal grafting surgical procedures and post-operative care were pre-approved by
123

and used in accordance with the USC IACUC protocol under Dr. Pradip Roy-Burman.
Pre- and post-surgery, mice were housed in autoclave sterilized filter-capped
microisolator cages, up to five mice per cage and fed autoclaved food. Prior to surgery
and after two weeks post-surgery, mice were given filtered water ad libitum. Two weeks
immediately post-surgery, mice were given sterilized filtered water with 3%
Trimethoprim/sulfamethoxazole oral suspension, USP (40 mg/200 mg per 5 mL; Hi-Tech
Pharmacal Co., Inc.) antibiotic solution in sterilized water bottles, changed every three
days. Mice were supplemented with HydroGel Recovery Gel for Rodents (ClearH2O
Maine) at 1 oz. per cage, twice per week, for the duration of the post-operative recovery
procedure.

124

5.2 Isolation of Cell Cultures and Cell Sorting
5.2.1 Fibroblast Cultures
Urogenital sinus mesenchyme (UGSM) cells were isolated following published
procedures (L. Xin, Ide, Kim, Dubey, & Witte, 2003; L. Xin, Lukacs, Lawson, Cheng, &
Witte, 2007). Normal prostate fibroblasts (NPFs) and CAFs were isolated as previously
described (S. Yang et al., 2008).

5.2.2 Epithelial Cultures
The cE1 cell line was generated as recently published in (Liao, Liang et al., 2010).
Primary cultures of epithelial cells were isolated from single cell suspensions of minced
prostate tumors from cPten
-/-
L mice as described in (Liao et al., 2007).

5.2.3 Cell Sorting
For fluorescence-activated cell sorting (FACS), cells were stained with biotinylated
antibodies against lineage, or “Lin” (CD31, CD45 and TER119; BD Biosciences; 0.1
mg/10
6
cells), followed by Brilliant Violet 421–conjugated streptavidin (Biolegend; 0.1
mg/10
6
cells), fluorescein isothiocyanate (FITC)-conjugated SCA-1 antibody (Biolegend;
0.1 mg/10
6
cells), and phycoerythrin (PE)-conjugated CD49f antibody (Biolegend; 0.1
mg/10
6
cells). Stained cells were then examined using BD FACSAria Cell Sorting
System with BD FACSDiva Software (BD Biosciences).

125

5.3 Cell Culture and in vitro Assays for Spheroid Formation
5.3.1 cE1 2D Cell Culture Conditions
cE1 2-D cultures were grown in a maintenance medium that contained Dulbecco’s
modified Eagle’s medium (DMEM; Gibco), 10% fetal bovine serum (FBS; Gemini), 25
μg/ml bovine pituitary extract (Invitrogen), 5 μg/ml insulin (Sigma-Aldrich), and 6 ng/ml
recombinant human epidermal growth factor (rhEGF; Invitrogen).

5.3.2 cE1 and LSC 3D Cell Culture Conditions
For cE1 3-D culture experiments, cE1 cells were counted and suspended in 1:1 Matrigel
(BD Biosciences; Cat. No.: 354234)/culturing medium in a total volume of 250 μL. The
mixture was placed in a well of a 24-well plate, solidified at 37° C, and then cultured in
medium for 7 or 14 days. Culturing medium for cE1 3-D assays was prepared as
described above, except for substitution of 1% FBS and absence of bovine pituitary
extract. Media was changed every 3 days. For primary epithelial cell culturing, sorted
prostate cells were counted and suspended in 1:1 Matrigel/PrEGM (Lonza) in a total
volume of 250 μL. The mixture was placed in a well of a 24-well plate, solidified at 37°
C, and then cultured in PrEGM for 21 days. Media was changed every 3 days. Spheroids
were counted at 14 and 21 days after plating. For passaging, spheroids formed in Matrigel
were digested in 500 μl of dispase (BD Biosciences) at 37° C for 1 hour followed by
treatment with DMEM/F12 medium (Invitrogen) containing collagenase (Sigma; 1
mg/mL), hyaluronidase (Sigma; 1 mg/mL), and DNase I (Sigma; 1 μg/mL) for 1 hour and
then in 1X TrypLE (Invitrogen) for 10 minutes. After passing through a 40- μm filter (BD
126

Biosciences), cells were counted and re-plated or grafted for renal capsule
transplantation.

127

5.4 Conditioned Media and AnxA1 Ligands
5.4.1 Collection of Conditioned Media
NPF and CAF CM were prepared by 24 hour incubation of serum free DMEM/5 μg/ml
insulin with confluent stromal cultures. Collected media was centrifuged at 300 x g for 5
minutes to remove any contaminating cells and debris. Media was normalized by protein
quantification using Bradford reagent (Bio-Rad) in a Benchmark Plus Microplate
Spectrophotometer (Bio-Rad) and compared to number of fibroblast cells per plate at
time of collection. Fibroblast cells were counted using a Beckman Coulter cell counter.
Following subsequent concentration using Amicon Ultra-15 3K Centrifugal Filter Units
(Millipore) CM was used to treat epithelial cells at a concentration of 0.04 mg/ml or ratio
of 10,000 to 1 fibroblast cells to epithelial cells.

5.4.2 Ammonium Sulfate Precipitation of Conditioned Media Proteins
Ammonium sulfate (AS) CM fractions were prepared following the basic procedure
outlined in (Burgess, 2009). AS powder (Sigma) was added to NPF and CAF CM, and
allowed to reach equilibrium for 30 minutes at 4°C, to pellet out insoluble proteins. After
centrifugation at 10,000 x g for 10 min. at 4°C, more AS was added to raise the amount
of AS in the supernatant 10%. This process was repeated to produce 20%, 30%, 40%,
50% and 60% insoluble fractions. Pelleted proteins were solubilized in 1x PBS and
dialyzed overnight, followed by concentration in Amicon Ultra-15 3K Centrifugal Filter
Units. AS fractions were concentrated for treatment of epithelial cells at 0.02 mg/ml in
500 μl media.
128

Figure 5.1 Ammonium sulfate precipitation schematic.

5.4.3 AnxA1 Ligands
Murine recombinant AnxA1 protein was produced as an N-terminal 6xHis tag fusion
protein in bacteria. Full-length mouse AnxA1 cDNA was purchased from Invitrogen
(Clone No.: 3590168; Cat. No.: FL1002). The cDNA was subcloned into the pET-100/D-
TOPO vector (Invitrogen) for N-terminal His tag in-frame, and AnxA1-His fusion protein
was expressed in bacteria BL21 Star™ (DE3) (Invitrogen) and induced by 1 mM
isopropylthio- β-D-galactoside (IPTG; Invitrogen). The fusion protein was extracted from
bacterial pellet with B-PER 6xHis Fusion Protein Purification Kit using nickel-chelated
resin columns and proprietary included buffers (Thermo Scientific; Cat. No.: 78100).
Purified protein was then passed through Pierce High Capacity Endotoxin Removal Spin
20% AS
30% AS
40% AS
50% AS
60% AS
129

Columns (Thermo Scientific; Cat. No.: 88276) and washed in buffer containing 50 mM
Tris-HCl, 2M NaCl, pH 7.0 through centrifugation using Amicon Ultra-15 3K
Centrifugal Filter Units (Millipore). Peptide Ac2-26 (acetyl-
AMVSEFLKQAWFIENEEQEYVQTVK-OH trifluoroacetate salt; M
r
3089) was
synthesized by Bachem (Cat. No.: H-3196) using solid-phase stepwise synthesis. Purity
was more than 94% as assessed by high performance liquid chromatography (data
supplied by manufacturer).
Figure 5.2 Generation of murine full-length recombinant AnxA1 protein. Following 4 hour
IPTG induction of expression through the pET-100/D-TOPO vector in BL21 Star™ (DE3)
bacteria, bacteria were harvested and lysed. Lysis supernatant was run through a nickel-chelated
resin column from Thermo Pierce to purify His-tagged recombinant proteins. After elution of
His-tagged proteins bound to the column, elution fractions were purified of bacterial endotoxins
and presence of purified rAnxA1 was verified with polyacrylamide gel electrophoresis by
Coomassie Blue (Bio-Rad) staining and western blot against AnxA1.

Sup. Flowthru Wash1(1) Wash 1(2) Wash2(1) Wash2(2) Elut1 Elut2
98
64
50
36
17
148
98
64
50
36
17
148
38
32
88
129
215
+ - + - + - + -
IPTG
- - + + - - + +
Anxa1
0 2
98
64
50
36
17
55
40
35
25
15
70
100
130
170 148
+ - + - + - + -
IPTG
- - + + - - + +
Anxa1
4 6
α-Anxa1
50 50
α-Anxa1
98
64
50
36
17
148
α-Anxa1
Elution1 Elution2
Endotoxin Removal
98
64
50
36
17
148
Elution1 Elution2
130

5.5 Renal Grafting
5.5.1 Preparation of Collagen Grafts
Epithelial cells (10
4
) were mixed with stromal cells (10
4
) in 70 μL neutralized rat tail
collagen type I (BD Biosciences) and placed in the middle part of a well of a 12-well
plate. Sham grafts did not contain cells, but were volume adjusted by addition of stromal
cell culturing media. The grafts were cultured at 37° C overnight in a medium described
in (Liao, Adisetiyo et al., 2010a) before transplanting under the renal capsule of 8- to 12-
week-old male NOD.SCID mice.

5.5.2 Surgical Procedure
For full details for the renal capsule surgical procedure and recommended and required
list of materials see Lukacs, R.U., et al (2010). Isolation, cultivation and characterization
of adult murine prostate stem cells. Nature Protocols 5(4),702-713. (Lukacs, Goldstein,
Lawson, Cheng, & Witte, 2010) (Figure 5.3).

Figure 5.3 Required materials and bench procedure for renal capsule transplantation. Divded
workstations for shaving of fur, sterilization of skin, kidney extraction, transplantation and
recovery are needed for successful surgical operation.

Aseptic technique must be strictly adhered to in order to ensure animal survival. To
minimize risk of bacterial infection, fur was clipped and skin was subsequently
Ear clips, clipper,
70% EtOH, swabs
Fur trimmer,
sutures, heat pad,
isoflurane, tools
Forceps, scissors,
grafts, dissection
microscope
Recovery station:
heat pad/lamp
131

disinfected using alternating scrubs of betadine/chlorohexadine and 70% ethanol,
repeated three times. All instruments were autoclave sterilized. Typical injectable
anesthesia was prepared with ketamine and xylazine at final concentrations of 25 mg/ml
and 1 mg/ml, respectively. Animal loss was additionally minimized by diluting xylazine,
to a final concentration of 0.1 mg/ml, in combination with administration of inhalant
anesthesia applied by isoflurane-soaked gauze in a 15 ml conical tube.

5.5.3 Post-Operative Recovery of Grafts
After 10 weeks, each animal was sacrificed and weighed, and both kidneys were isolated.
Grafted and non-grafted kidneys were weighed for unusual differences in size due to
surgical procedure and ectopic presence of tumor grafts. The kidney with the graft from
each animal was cut with a scalpel and arranged for pictures of gross morphology. The
kidney with the graft was then isolated and fixed (Figure 5.4).

Figure 5.4 Isolation of grafts after 10 weeks. Spleen, pancreas, kidney and location of graft are
noted.
Sham
Group#7: Sham Control
Kidney
Pancreas
Spleen
Graft
132

5.6 Immunostaining and Western Blots
5.6.1 Immunohistochemical Staining
Spheroids in Matrigel (BD Biosciences) were covered by optimal cutting temperature
compound (Sakura), sectioned to 8- μm thickness at -20°C, and stained using either
hematoxylin and eosin (H&E) or immunofluorescence (IF) following previously
established protocols (Liao, Adisetiyo et al., 2010a; Liao et al., 2007). Renal grafts were
fixed with 4% paraformaldehyde and paraffin embedded, sectioned to 5- μm thickness
and stained with either H&E or immunohistochemistry (IHC) using a modified avidin-
biotin complex technique (Liao et al., 2007). Primary antibodies against p63 (1:100;
Abcam), CK8 (1:50; TROMA-1 antibody; Developmental Studies Hybridoma Bank,
University of Iowa), CK5 (1:1000, Covance), androgen receptor (AR; 1:200; Santa Cruz
Biotechnology), Ki67 (1:400; Vector Laboratories), CRE (1:1000; Covance), Pten
(1:200; Cell Signaling Technology), pErk1/2 (1:400; Cell Signaling Technology), TGF-
β1 (1:500; Santa Cruz Biotechnology), or AnxA1 (1:500; Invitrogen) were used. All
microscopy images were captured with Spot Advanced software (Spot™ Imaging
Solutions). Image quantification was completed using ImageJ software (National
Institutes of Health). Neutralizing antibody (anti-N19) against N-terminal residues of
AnxA1 was obtained from Santa Cruz Biotechnology.

5.6.2 Western Blotting
For western blot analysis, whole cell lysates were prepared by addition of RIPA buffer
(Sigma) with proteinase inhibitor (Roche) and phosphatase inhibitor cocktails II and III
133

(Sigma) at a 1X final concentration (manufacturer’s protocols). TGF- β1, pErk1/2 and
AnxA1 antibodies for immunostaining were rated suitable for western blot. Erk1/2
(1:100; Cell Signaling Technology), β-Actin (1:500; Santa Cruz Biotechnology), and
Twist (1:200; Santa Cruz Biotechnology) were additionally used for western blot. Anti-
phospho-serine PKC substrate antibody (1:500) was obtained as a generous sample from
Cell Signaling Technology.
Antibody Manufacturer Dilution
p63 (Cat.# ab735) Abcam 1:100 (IHC); 1:400 (IF)
CK8 (TROMA-1-c) Developmental Studies
Hybridoma Bank,
University of Iowa
1:50 (IHC); 1:50 (IF)
CK5 (Cat.# C-7785) Covance 1:1000 (IHC); 1:1000 (IF)
AR (Cat.# sc-816) Santa Cruz Biotechnology 1:200 (IHC)
Ki67 (Cat.# VP-RM04) Vector Laboratories 1:400 (IHC)
Cre Covance 1:1000 (IHC)
Pten (Cat.# 9188) Cell Signaling Technology 1:200 (IHC)
pErk1/2 (Cat.# 4370) Cell Signaling Technology 1:400 (IHC); 1:500 (WB)
Erk1/2 (Cat.# 9102) Cell Signaling Technology 1:100 (WB)
TGF- β1 (Cat.# sc-146) Santa Cruz Biotechnology 1:500 (WB)
AnxA1 (Cat.# 713400) Invitrogen 1:500 (WB)
Twist (Cat.# sc-6269) Santa Cruz Biotechnology 1:200 (WB)
pSer PKC Substrate
(Cat.# 6967)
Cell Signaling Technology 1:500 (WB)
β-Actin (Cat.# sc-1616-R) Santa Cruz Biotechnology 1:500 (WB)
AnxA1 Neutralizing
antibody (Cat.# sc-1923-R)
Santa Cruz Biotechnology 10 μg/ml
Table 5.2 Antibody List. Antibodies and their dilutions for immunohistochemistry (IHC),
immunofluorescence (IF), western blot (WB) and cell culture neutralizing antibody assays are
listed.
134

5.7 PCR Analyses
Extraction of total cellular RNA, reverse transcription reaction and quantitative real-time
PCR were performed and analyzed following methods in (Liao, Adisetiyo et al., 2010a).
Primer sets are listed in Table 5.2.
Primer Name Primer Sequences (5’ to 3’) Annealing T
m

Snail F ACCCACACTGGTGAGAAGC 58°C This study
Snail R GACCAAGGCTGGAAGGAGTC 58°C This study
Slug F CACAGTTATTATTTCCCCATATCT 53°C This study
Slug R GCAGTCTCTCCTCTTCGTCA 56°C This study
Twist F CTCGGACAAGCTGAGCAAG 56°C (He et al., 2009)
Twist R ACGGAGAAGGCGTAGCTGAG 56°C (He et al., 2009)
Oct4 F CTCACCCTGGGCGTTCTCT 58°C (Liao, Liang et
al., 2010)
Oct4 R AGGCCTCGAAGCGACAGA 58°C (Liao, Liang et
al., 2010)
Sox2 F GCAGTACAACTCCATGACCA 54°C This study
Sox2 R CTCGGACTTGACCACAGAG 54°C This study
Nanog F CTGAGCTATAAGCAGGTTAAGAC 53°C This study
Nanog R CAGATGCGTTCACCAGATAG 53°C This study
E-Cadherin F TCCAGGAACCTCCGTGATG 55°C This study
E-Cadherin R GGGTAACTCTCTCGGTCCAG 55°C This study
N-Cadherin F AGCCTGGGACGTATGTGATG 55°C This study
N-Cadherin R ATGTTGGGTGAAGGTGTGCT 55°C This study
Vimentin F CAAGTCCAAGTTTGCTGACCT 56°C This study
Vimentin R TCTTCCATCTCACGCATCTG 56°C This study
Pten F GCAATCCTCAGTTTGTGGTC 55°C This study
Pten R TCCTCTGGTCCTGGTATGAA 55°C This study
Trpm7 F AGTTGCTTGGAAAGGGTCTT 56°C This study
Trpm7 R GCCACACCTGTATTGACTCC 55°C This study
PKCβ F CCAAGACCATCAAGTGTTCC 55°C This study
PKCβ R CACCCTCTTCCTGGCTTAGT 56°C This study
PKCδ F GTAGTGAGGAGGAGGCAAAG 54°C This study
PKCδ R CGTGGTTCTTGATGTAGTGG 54°C This study
PKCε F AGCCCCTAAAGACAATGAAG 54°C This study
PKCε R CTATGACACCCCAGATGAAA 53°C This study
AnxA1 F GAACTGAAGGGTGACATTGA 53°C This study
AnxA1 R GGCAAAGAGAGATTCCATACT 53°C This study
MAPK F AGTCCTTTTGAGCACCAGAC 54°C This study
MAPK R TTGATGCCAATGATGTTCTC 54°C This study
TGF βRII F TCTCCACAGTGACCACACTC 56°C This study
TGF βRII R TAGACATCCGTCTGCTTGAA 54°C This study
β-Actin F AGTGTGACGTTGACATCCGT 56°C (Liao, Liang et
al., 2010)
β-Actin R CTTGCTGATCCACATCTGCT 55°C (Liao, Liang et
al., 2010)
Table 5.2 PCR Primer List. PCR primer sequences for real time quantitative reverse
transcription polymerase chain reaction are listed.
135

5.8 Statistical Analysis
All data are presented as means ± SD. Differences between individual groups were
analyzed by paired t test or χ
2
, as appropriate, using Excel (Microsoft
®
Office) and InStat
(GraphPad Software, Inc.
©
) software. P values of < 0.05 were considered to be
statistically significant.
136

References

Ahn, S. H., Sawada, H., Ro, J. Y., & Nicolson, G. L. (1997). Differential expression of
annexin I in human mammary ductal epithelial cells in normal and benign and malignant
breast tissues. Clin Exp Metastasis, 15(2), 151-156.
Al-Hajj, M., Wicha, M. S., Benito-Hernandez, A., Morrison, S. J., & Clarke, M. F.
(2003). Prospective identification of tumorigenic breast cancer cells. Proc Natl Acad Sci
U S A, 100(7), 3983-3988.
Balkwill, F. (2002). Tumor necrosis factor or tumor promoting factor? Cytokine Growth
Factor Rev, 13(2), 135-141.
Bates, R. C., & Mercurio, A. M. (2003). Tumor necrosis factor-alpha stimulates the
epithelial-to-mesenchymal transition of human colonic organoids. Mol Biol Cell, 14(5),
1790-1800.
Bhowmick, N. A., Ghiassi, M., Bakin, A., Aakre, M., Lundquist, C. A., Engel, M. E., et
al. (2001). Transforming growth factor-beta1 mediates epithelial to mesenchymal
transdifferentiation through a RhoA-dependent mechanism. Mol Biol Cell, 12(1), 27-36.
Bitto, E., Li, M., Tikhonov, A. M., Schlossman, M. L., & Cho, W. (2000). Mechanism of
annexin I-mediated membrane aggregation. Biochemistry, 39(44), 13469-13477.
Blackwell, G. J., Carnuccio, R., Di Rosa, M., Flower, R. J., Parente, L., & Persico, P.
(1980). Macrocortin: a polypeptide causing the anti-phospholipase effect of
glucocorticoids. Nature, 287(5778), 147-149.
Bonnet, D., & Dick, J. E. (1997). Human acute myeloid leukemia is organized as a
hierarchy that originates from a primitive hematopoietic cell. Nat Med, 3(7), 730-737.
Brabletz, T., Jung, A., Spaderna, S., Hlubek, F., & Kirchner, T. (2005). Opinion:
migrating cancer stem cells - an integrated concept of malignant tumour progression. Nat
Rev Cancer, 5(9), 744-749.
137

Bubendorf, L., Schopfer, A., Wagner, U., Sauter, G., Moch, H., Willi, N., et al. (2000).
Metastatic patterns of prostate cancer: an autopsy study of 1,589 patients. Hum Pathol,
31(5), 578-583.
Burger, P. E., Xiong, X., Coetzee, S., Salm, S. N., Moscatelli, D., Goto, K., et al. (2005).
Sca-1 expression identifies stem cells in the proximal region of prostatic ducts with high
capacity to reconstitute prostatic tissue. Proc Natl Acad Sci U S A, 102(20), 7180-7185.
Burgess, R. R. (2009). Protein precipitation techniques. Methods Enzymol, 463, 331-342.
Cantley, L. C., & Neel, B. G. (1999). New insights into tumor suppression: PTEN
suppresses tumor formation by restraining the phosphoinositide 3-kinase/AKT pathway.
Proc Natl Acad Sci U S A, 96(8), 4240-4245.
Cao, Y., Li, Y., Edelweiss, M., Arun, B., Rosen, D., Resetkova, E., et al. (2008). Loss of
annexin A1 expression in breast cancer progression. Appl Immunohistochem Mol
Morphol, 16(6), 530-534.
Chapman, L. P., Epton, M. J., Buckingham, J. C., Morris, J. F., & Christian, H. C. (2003).
Evidence for a role of the adenosine 5'-triphosphate-binding cassette transporter A1 in the
externalization of annexin I from pituitary folliculo-stellate cells. Endocrinology, 144(3),
1062-1073.
Chatterjee, B. E., Yona, S., Rosignoli, G., Young, R. E., Nourshargh, S., Flower, R. J., et
al. (2005). Annexin 1-deficient neutrophils exhibit enhanced transmigration in vivo and
increased responsiveness in vitro. J Leukoc Biol, 78(3), 639-646.
Cheng, G. Z., Chan, J., Wang, Q., Zhang, W., Sun, C. D., & Wang, L. H. (2007). Twist
transcriptionally up-regulates AKT2 in breast cancer cells leading to increased migration,
invasion, and resistance to paclitaxel. Cancer Res, 67(5), 1979-1987.
Cher, M. L. (2001). Mechanisms governing bone metastasis in prostate cancer. Curr
Opin Urol, 11(5), 483-488.
Chiang, N., Serhan, C. N., Dahlen, S. E., Drazen, J. M., Hay, D. W., Rovati, G. E., et al.
(2006). The lipoxin receptor ALX: potent ligand-specific and stereoselective actions in
vivo. Pharmacol Rev, 58(3), 463-487.
138

Cho, R. W., & Clarke, M. F. (2008). Recent advances in cancer stem cells. Curr Opin
Genet Dev, 18(1), 48-53.
Christmas, P., Callaway, J., Fallon, J., Jones, J., & Haigler, H. T. (1991). Selective
secretion of annexin 1, a protein without a signal sequence, by the human prostate gland.
J Biol Chem, 266(4), 2499-2507.
Chung, L. W., Baseman, A., Assikis, V., & Zhau, H. E. (2005). Molecular insights into
prostate cancer progression: the missing link of tumor microenvironment. J Urol, 173(1),
10-20.
Collins, A. T., Berry, P. A., Hyde, C., Stower, M. J., & Maitland, N. J. (2005).
Prospective identification of tumorigenic prostate cancer stem cells. Cancer Res, 65(23),
10946-10951.
Collins, A. T., & Maitland, N. J. (2006). Prostate cancer stem cells. Eur J Cancer, 42(9),
1213-1218.
Comijn, J., Berx, G., Vermassen, P., Verschueren, K., van Grunsven, L., Bruyneel, E., et
al. (2001). The two-handed E box binding zinc finger protein SIP1 downregulates E-
cadherin and induces invasion. Mol Cell, 7(6), 1267-1278.
Coussens, L. M., & Werb, Z. (2002). Inflammation and cancer. Nature, 420(6917), 860-
867.
Crompton, M. R., Moss, S. E., & Crumpton, M. J. (1988). Diversity in the
lipocortin/calpactin family. Cell, 55(1), 1-3.
Croxtall, J. D., Choudhury, Q., & Flower, R. J. (2000). Glucocorticoids act within
minutes to inhibit recruitment of signalling factors to activated EGF receptors through a
receptor-dependent, transcription-independent mechanism. Br J Pharmacol, 130(2), 289-
298.
Crumpton, M. J., & Dedman, J. R. (1990). Protein terminology tangle. Nature, 345(6272),
212.
139

D'Acquisto, F., Paschalidis, N., Raza, K., Buckley, C. D., Flower, R. J., & Perretti, M.
(2008). Glucocorticoid treatment inhibits annexin-1 expression in rheumatoid arthritis
CD4+ T cells. Rheumatology (Oxford), 47(5), 636-639.
D'Acquisto, F., Perretti, M., & Flower, R. J. (2008). Annexin-A1: a pivotal regulator of
the innate and adaptive immune systems. Br J Pharmacol, 155(2), 152-169.
Dahia, P. L. (2000). PTEN, a unique tumor suppressor gene. Endocr Relat Cancer, 7(2),
115-129.
Davies, J. A. (1996). Mesenchyme to epithelium transition during development of the
mammalian kidney tubule. Acta Anat (Basel), 156(3), 187-201.
de Graauw, M., van Miltenburg, M. H., Schmidt, M. K., Pont, C., Lalai, R., Kartopawiro,
J., et al. (2010). Annexin A1 regulates TGF-beta signaling and promotes metastasis
formation of basal-like breast cancer cells. Proc Natl Acad Sci U S A, 107(14), 6340-
6345.
De Wever, O., & Mareel, M. (2003). Role of tissue stroma in cancer cell invasion. J
Pathol, 200(4), 429-447.
Dvorak, H. F. (1986). Tumors: wounds that do not heal. Similarities between tumor
stroma generation and wound healing. N Engl J Med, 315(26), 1650-1659.
Ellenrieder, V., Hendler, S. F., Boeck, W., Seufferlein, T., Menke, A., Ruhland, C., et al.
(2001). Transforming growth factor beta1 treatment leads to an epithelial-mesenchymal
transdifferentiation of pancreatic cancer cells requiring extracellular signal-regulated
kinase 2 activation. Cancer Res, 61(10), 4222-4228.
English, H. F., Drago, J. R., & Santen, R. J. (1985). Cellular response to androgen
depletion and repletion in the rat ventral prostate: autoradiography and morphometric
analysis. Prostate, 7(1), 41-51.
English, H. F., Santen, R. J., & Isaacs, J. T. (1987). Response of glandular versus basal
rat ventral prostatic epithelial cells to androgen withdrawal and replacement. Prostate,
11(3), 229-242.
140

Ernst, S., Lange, C., Wilbers, A., Goebeler, V., Gerke, V., & Rescher, U. (2004). An
annexin 1 N-terminal peptide activates leukocytes by triggering different members of the
formyl peptide receptor family. J Immunol, 172(12), 7669-7676.
Etoh, T., Shibuta, K., Barnard, G. F., Kitano, S., & Mori, M. (2000). Angiogenin
expression in human colorectal cancer: the role of focal macrophage infiltration. Clin
Cancer Res, 6(9), 3545-3551.
Fang, D., Nguyen, T. K., Leishear, K., Finko, R., Kulp, A. N., Hotz, S., et al. (2005). A
tumorigenic subpopulation with stem cell properties in melanomas. Cancer Res, 65(20),
9328-9337.
Ferrandina, G., Bonanno, G., Pierelli, L., Perillo, A., Procoli, A., Mariotti, A., et al.
(2008). Expression of CD133-1 and CD133-2 in ovarian cancer. Int J Gynecol Cancer,
18(3), 506-514.
Fujimoto, K., Sheng, H., Shao, J., & Beauchamp, R. D. (2001). Transforming growth
factor-beta1 promotes invasiveness after cellular transformation with activated Ras in
intestinal epithelial cells. Exp Cell Res, 266(2), 239-249.
Gao, J. L., Chen, H., Filie, J. D., Kozak, C. A., & Murphy, P. M. (1998). Differential
expansion of the N-formylpeptide receptor gene cluster in human and mouse. Genomics,
51(2), 270-276.
Gavins, F. N., Yona, S., Kamal, A. M., Flower, R. J., & Perretti, M. (2003). Leukocyte
antiadhesive actions of annexin 1: ALXR- and FPR-related anti-inflammatory
mechanisms. Blood, 101(10), 4140-4147.
Gerke, V., Creutz, C. E., & Moss, S. E. (2005). Annexins: linking Ca2+ signalling to
membrane dynamics. Nat Rev Mol Cell Biol, 6(6), 449-461.
Gerke, V., & Moss, S. E. (2002). Annexins: from structure to function. Physiol Rev,
82(2), 331-371.
Getting, S. J., Flower, R. J., & Perretti, M. (1997). Inhibition of neutrophil and monocyte
recruitment by endogenous and exogenous lipocortin 1. Br J Pharmacol, 120(6), 1075-
1082.
141

Giannoni, E., Bianchini, F., Masieri, L., Serni, S., Torre, E., Calorini, L., et al. (2010).
Reciprocal activation of prostate cancer cells and cancer-associated fibroblasts stimulates
epithelial-mesenchymal transition and cancer stemness. Cancer Res, 70(17), 6945-6956.
Gidekel, S., Pizov, G., Bergman, Y., & Pikarsky, E. (2003). Oct-3/4 is a dose-dependent
oncogenic fate determinant. Cancer Cell, 4(5), 361-370.
Goldstein, A. S., Huang, J., Guo, C., Garraway, I. P., & Witte, O. N. (2010).
Identification of a cell of origin for human prostate cancer. Science, 329(5991), 568-571.
Goto, K., Salm, S. N., Coetzee, S., Xiong, X., Burger, P. E., Shapiro, E., et al. (2006).
Proximal prostatic stem cells are programmed to regenerate a proximal-distal ductal axis.
Stem Cells, 24(8), 1859-1868.
Gupta, G. P., & Massague, J. (2006). Cancer metastasis: building a framework. Cell,
127(4), 679-695.
Gupta, P. B., Onder, T. T., Jiang, G., Tao, K., Kuperwasser, C., Weinberg, R. A., et al.
(2009). Identification of selective inhibitors of cancer stem cells by high-throughput
screening. Cell, 138(4), 645-659.
Hanahan, D., & Weinberg, R. A. (2011). Hallmarks of cancer: the next generation. Cell,
144(5), 646-674.
Hartwell, K. A., Muir, B., Reinhardt, F., Carpenter, A. E., Sgroi, D. C., & Weinberg, R.
A. (2006). The Spemann organizer gene, Goosecoid, promotes tumor metastasis. Proc
Natl Acad Sci U S A, 103(50), 18969-18974.
Hay, E. D. (1995). An overview of epithelio-mesenchymal transformation. Acta Anat
(Basel), 154(1), 8-20.
Hayhoe, R. P., Kamal, A. M., Solito, E., Flower, R. J., Cooper, D., & Perretti, M. (2006).
Annexin 1 and its bioactive peptide inhibit neutrophil-endothelium interactions under
flow: indication of distinct receptor involvement. Blood, 107(5), 2123-2130.
142

He, W., Dai, C., Li, Y., Zeng, G., Monga, S. P., & Liu, Y. (2009). Wnt/beta-catenin
signaling promotes renal interstitial fibrosis. J Am Soc Nephrol, 20(4), 765-776.
Hsiang, C. H., Tunoda, T., Whang, Y. E., Tyson, D. R., & Ornstein, D. K. (2006). The
impact of altered annexin I protein levels on apoptosis and signal transduction pathways
in prostate cancer cells. Prostate, 66(13), 1413-1424.
Hsing, A. Y., Kadomatsu, K., Bonham, M. J., & Danielpour, D. (1996). Regulation of
apoptosis induced by transforming growth factor-beta1 in nontumorigenic rat prostatic
epithelial cell lines. Cancer Res, 56(22), 5146-5149.
Hsu, S., Huang, F., Hafez, M., Winawer, S., & Friedman, E. (1994). Colon carcinoma
cells switch their response to transforming growth factor beta 1 with tumor progression.
Cell Growth Differ, 5(3), 267-275.
Huber, M. A., Kraut, N., & Beug, H. (2005). Molecular requirements for epithelial-
mesenchymal transition during tumor progression. Curr Opin Cell Biol, 17(5), 548-558.
Hurt, E. M., & Farrar, W. L. (2008). Cancer stem cells: the seeds of metastasis? Mol
Interv, 8(3), 140-142.
Jemal, A., Bray, F., Center, M. M., Ferlay, J., Ward, E., & Forman, D. (2011). Global
cancer statistics. CA Cancer J Clin, 61(2), 69-90.
John, C. D., Christian, H. C., Morris, J. F., Flower, R. J., Solito, E., & Buckingham, J. C.
(2003). Kinase-dependent regulation of the secretion of thyrotrophin and luteinizing
hormone by glucocorticoids and annexin 1 peptides. J Neuroendocrinol, 15(10), 946-957.
John, C. D., Gavins, F. N., Buss, N. A., Cover, P. O., & Buckingham, J. C. (2008).
Annexin A1 and the formyl peptide receptor family: neuroendocrine and metabolic
aspects. Curr Opin Pharmacol, 8(6), 765-776.
Joyce, J. A., & Pollard, J. W. (2009). Microenvironmental regulation of metastasis. Nat
Rev Cancer, 9(4), 239-252.
143

Kalluri, R., & Weinberg, R. A. (2009). The basics of epithelial-mesenchymal transition. J
Clin Invest, 119(6), 1420-1428.
Kalluri, R., & Zeisberg, M. (2006). Fibroblasts in cancer. Nat Rev Cancer, 6(5), 392-401.
Kaminski, A., Hahne, J. C., Haddouti el, M., Florin, A., Wellmann, A., & Wernert, N.
(2006). Tumour-stroma interactions between metastatic prostate cancer cells and
fibroblasts. Int J Mol Med, 18(5), 941-950.
Kang, J. S., Calvo, B. F., Maygarden, S. J., Caskey, L. S., Mohler, J. L., & Ornstein, D. K.
(2002). Dysregulation of annexin I protein expression in high-grade prostatic
intraepithelial neoplasia and prostate cancer. Clin Cancer Res, 8(1), 117-123.
Karoubi, G., Gugger, M., Schmid, R., & Dutly, A. (2009). OCT4 expression in human
non-small cell lung cancer: implications for therapeutic intervention. Interact Cardiovasc
Thorac Surg, 8(4), 393-397.
Khau, T., Langenbach, S. Y., Schuliga, M., Harris, T., Johnstone, C. N., Anderson, R. L.,
et al. (2011). Annexin-1 signals mitogen-stimulated breast tumor cell proliferation by
activation of the formyl peptide receptors (FPRs) 1 and 2. FASEB J, 25(2), 483-496.
Kim, C. F., Jackson, E. L., Woolfenden, A. E., Lawrence, S., Babar, I., Vogel, S., et al.
(2005). Identification of bronchioalveolar stem cells in normal lung and lung cancer. Cell,
121(6), 823-835.
Klarmann, G. J., Hurt, E. M., Mathews, L. A., Zhang, X., Duhagon, M. A., Mistree, T., et
al. (2009). Invasive prostate cancer cells are tumor initiating cells that have a stem cell-
like genomic signature. Clin Exp Metastasis, 26(5), 433-446.
Kyprianou, N., & Isaacs, J. T. (1988). Activation of programmed cell death in the rat
ventral prostate after castration. Endocrinology, 122(2), 552-562.
Lambert, O., Gerke, V., Bader, M. F., Porte, F., & Brisson, A. (1997). Structural analysis
of junctions formed between lipid membranes and several annexins by cryo-electron
microscopy. J Mol Biol, 272(1), 42-55.
144

Lange, C., Starrett, D. J., Goetsch, J., Gerke, V., & Rescher, U. (2007). Transcriptional
profiling of human monocytes reveals complex changes in the expression pattern of
inflammation-related genes in response to the annexin A1-derived peptide Ac1-25. J
Leukoc Biol, 82(6), 1592-1604.
Lawson, D. A., Xin, L., Lukacs, R., Xu, Q., Cheng, D., & Witte, O. N. (2005). Prostate
stem cells and prostate cancer. Cold Spring Harb Symp Quant Biol, 70, 187-196.
Le, Y., Murphy, P. M., & Wang, J. M. (2002). Formyl-peptide receptors revisited. Trends
Immunol, 23(11), 541-548.
Leek, R. D., Lewis, C. E., Whitehouse, R., Greenall, M., Clarke, J., & Harris, A. L.
(1996). Association of macrophage infiltration with angiogenesis and prognosis in
invasive breast carcinoma. Cancer Res, 56(20), 4625-4629.
Lehmann, K., Janda, E., Pierreux, C. E., Rytomaa, M., Schulze, A., McMahon, M., et al.
(2000). Raf induces TGFbeta production while blocking its apoptotic but not invasive
responses: a mechanism leading to increased malignancy in epithelial cells. Genes Dev,
14(20), 2610-2622.
Leong, K. G., Wang, B. E., Johnson, L., & Gao, W. Q. (2008). Generation of a prostate
from a single adult stem cell. Nature, 456(7223), 804-808.
Li, C., Heidt, D. G., Dalerba, P., Burant, C. F., Zhang, L., Adsay, V., et al. (2007).
Identification of pancreatic cancer stem cells. Cancer Res, 67(3), 1030-1037.
Liao, C. P., Adisetiyo, H., Liang, M., & Roy-Burman, P. (2010a). Cancer-associated
fibroblasts enhance the gland-forming capability of prostate cancer stem cells. Cancer
Res, 70(18), 7294-7303.
Liao, C. P., Adisetiyo, H., Liang, M., & Roy-Burman, P. (2010b). Cancer stem cells and
microenvironment in prostate cancer progression. Horm Cancer, 1(6), 297-305.
Liao, C. P., Liang, M., Cohen, M. B., Flesken-Nikitin, A., Jeong, J. H., Nikitin, A. Y., et
al. (2010). Mouse prostate cancer cell lines established from primary and postcastration
recurrent tumors. Horm Cancer, 1(1), 44-54.
145

Liao, C. P., Zhong, C., Saribekyan, G., Bading, J., Park, R., Conti, P. S., et al. (2007).
Mouse models of prostate adenocarcinoma with the capacity to monitor spontaneous
carcinogenesis by bioluminescence or fluorescence. Cancer Res, 67(15), 7525-7533.
Lim, L. H., & Pervaiz, S. (2007). Annexin 1: the new face of an old molecule. FASEB J,
21(4), 968-975.
Lim, M., Chuong, C. M., & Roy-Burman, P. (2011). PI3K, Erk signaling in BMP7-
induced epithelial-mesenchymal transition (EMT) of PC-3 prostate cancer cells in 2- and
3-dimensional cultures. Horm Cancer, 2(5), 298-309.
Liotta, L. A., & Kohn, E. C. (2001). The microenvironment of the tumour-host interface.
Nature, 411(6835), 375-379.
Liu, C. G., Lu, Y., Wang, B. B., Zhang, Y. J., Zhang, R. S., Chen, B., et al. (2011).
Clinical implications of stem cell gene Oct-4 expression in breast cancer. Ann Surg,
253(6), 1165-1171.
Lukacs, R. U., Goldstein, A. S., Lawson, D. A., Cheng, D., & Witte, O. N. (2010).
Isolation, cultivation and characterization of adult murine prostate stem cells. Nat Protoc,
5(4), 702-713.
Ma, S., Chan, K. W., Hu, L., Lee, T. K., Wo, J. Y., Ng, I. O., et al. (2007). Identification
and characterization of tumorigenic liver cancer stem/progenitor cells. Gastroenterology,
132(7), 2542-2556.
Mailliard, W. S., Haigler, H. T., & Schlaepfer, D. D. (1996). Calcium-dependent binding
of S100C to the N-terminal domain of annexin I. J Biol Chem, 271(2), 719-725.
Mani, S. A., Guo, W., Liao, M. J., Eaton, E. N., Ayyanan, A., Zhou, A. Y., et al. (2008).
The epithelial-mesenchymal transition generates cells with properties of stem cells. Cell,
133(4), 704-715.
Mani, S. A., Yang, J., Brooks, M., Schwaninger, G., Zhou, A., Miura, N., et al. (2007).
Mesenchyme Forkhead 1 (FOXC2) plays a key role in metastasis and is associated with
aggressive basal-like breast cancers. Proc Natl Acad Sci U S A, 104(24), 10069-10074.
146

Maschler, S., Gebeshuber, C. A., Wiedemann, E. M., Alacakaptan, M., Schreiber, M.,
Custic, I., et al. (2010). Annexin A1 attenuates EMT and metastatic potential in breast
cancer. EMBO Mol Med, 2(10), 401-414.
McAllister, S. S., & Weinberg, R. A. (2011). Tumor-host interactions: a far-reaching
relationship. J Clin Oncol, 28(26), 4022-4028.
Miele, L., Cordella-Miele, E., Facchiano, A., & Mukherjee, A. B. (1988). Novel anti-
inflammatory peptides from the region of highest similarity between uteroglobin and
lipocortin I. Nature, 335(6192), 726-730.
Moraes, R. C., Zhang, X., Harrington, N., Fung, J. Y., Wu, M. F., Hilsenbeck, S. G., et al.
(2007). Constitutive activation of smoothened (SMO) in mammary glands of transgenic
mice leads to increased proliferation, altered differentiation and ductal dysplasia.
Development, 134(6), 1231-1242.
Morel, A. P., Lievre, M., Thomas, C., Hinkal, G., Ansieau, S., & Puisieux, A. (2008).
Generation of breast cancer stem cells through epithelial-mesenchymal transition. PLoS
One, 3(8), e2888.
Morris, J. F., Christian, H. C., Chapman, L. P., Epton, M. J., Buckingham, J. C., Ozawa,
H., et al. (2002). Steroid effects on secretion from subsets of lactotrophs: role of folliculo-
stellate cells and annexin 1. Arch Physiol Biochem, 110(1-2), 54-61.
Mulholland, D. J., Xin, L., Morim, A., Lawson, D., Witte, O., & Wu, H. (2009). Lin-Sca-
1+CD49fhigh stem/progenitors are tumor-initiating cells in the Pten-null prostate cancer
model. Cancer Res, 69(22), 8555-8562.
Mussunoor, S., & Murray, G. I. (2008). The role of annexins in tumour development and
progression. J Pathol, 216(2), 131-140.
Nikitin, A. Y., Matoso, A., & Roy-Burman, P. (2007). Prostate stem cells and cancer.
Histol Histopathol, 22(9), 1043-1049.
Nikitin, A. Y., Nafus, M. G., Zhou, Z., Liao, C. P., & Roy-Burman, P. (Eds.). (2009).
Prostate stem cells and cancer in animals: Humana Press.
147

Oft, M., Akhurst, R. J., & Balmain, A. (2002). Metastasis is driven by sequential
elevation of H-ras and Smad2 levels. Nat Cell Biol, 4(7), 487-494.
Oft, M., Heider, K. H., & Beug, H. (1998). TGFbeta signaling is necessary for carcinoma
cell invasiveness and metastasis. Curr Biol, 8(23), 1243-1252.
Oft, M., Peli, J., Rudaz, C., Schwarz, H., Beug, H., & Reichmann, E. (1996). TGF-beta1
and Ha-Ras collaborate in modulating the phenotypic plasticity and invasiveness of
epithelial tumor cells. Genes Dev, 10(19), 2462-2477.
Parente, L., & Solito, E. (2004). Annexin 1: more than an anti-phospholipase protein.
Inflamm Res, 53(4), 125-132.
Park, C. H., Bergsagel, D. E., & McCulloch, E. A. (1971). Mouse myeloma tumor stem
cells: a primary cell culture assay. J Natl Cancer Inst, 46(2), 411-422.
Patrawala, L., Calhoun, T., Schneider-Broussard, R., Li, H., Bhatia, B., Tang, S., et al.
(2006). Highly purified CD44+ prostate cancer cells from xenograft human tumors are
enriched in tumorigenic and metastatic progenitor cells. Oncogene, 25(12), 1696-1708.
Patton, K. T., Chen, H. M., Joseph, L., & Yang, X. J. (2005). Decreased annexin I
expression in prostatic adenocarcinoma and in high-grade prostatic intraepithelial
neoplasia. Histopathology, 47(6), 597-601.
Paweletz, C. P., Ornstein, D. K., Roth, M. J., Bichsel, V. E., Gillespie, J. W., Calvert, V.
S., et al. (2000). Loss of annexin 1 correlates with early onset of tumorigenesis in
esophageal and prostate carcinoma. Cancer Res, 60(22), 6293-6297.
Payne, C. M., Morris, J. F., Solito, E., & Buckingham, J. C. (2005). Modulators of the
SUR2B/Kir6.1 ATP-sensitive K+channel regulate annexin 1 release in the TtT/GF
folliculostellate cell line. Paper presented at the 24th Joint Meeting of the British
Endocrine Societies.
Peinado, H., Olmeda, D., & Cano, A. (2007). Snail, Zeb and bHLH factors in tumour
progression: an alliance against the epithelial phenotype? Nat Rev Cancer, 7(6), 415-428.
148

Perez-Pomares, J. M., & Munoz-Chapuli, R. (2002). Epithelial-mesenchymal transitions:
a mesodermal cell strategy for evolutive innovation in Metazoans. Anat Rec, 268(3), 343-
351.
Perretti, M., Chiang, N., La, M., Fierro, I. M., Marullo, S., Getting, S. J., et al. (2002).
Endogenous lipid- and peptide-derived anti-inflammatory pathways generated with
glucocorticoid and aspirin treatment activate the lipoxin A4 receptor. Nat Med, 8(11),
1296-1302.
Perretti, M., Croxtall, J. D., Wheller, S. K., Goulding, N. J., Hannon, R., & Flower, R. J.
(1996). Mobilizing lipocortin 1 in adherent human leukocytes downregulates their
transmigration. Nat Med, 2(11), 1259-1262.
Perretti, M., & D'Acquisto, F. (2009). Annexin A1 and glucocorticoids as effectors of the
resolution of inflammation. Nat Rev Immunol, 9(1), 62-70.
Perretti, M., & Flower, R. J. (1996). Measurement of lipocortin 1 levels in murine
peripheral blood leukocytes by flow cytometry: modulation by glucocorticoids and
inflammation. Br J Pharmacol, 118(3), 605-610.
Philip, J. G., Flower, R. J., & Buckingham, J. C. (1997). Glucocorticoids modulate the
cellular disposition of lipocortin 1 in the rat brain in vivo and in vitro. Neuroreport, 8(8),
1871-1876.
Pietras, K., & Ostman, A. (2011). Hallmarks of cancer: interactions with the tumor
stroma. Exp Cell Res, 316(8), 1324-1331.
Podgorski, M. R., Goulding, N. J., Hall, N. D., Flower, R. J., & Maddison, P. J. (1992).
Autoantibodies to lipocortin-1 are associated with impaired glucocorticoid
responsiveness in rheumatoid arthritis. J Rheumatol, 19(11), 1668-1671.
Ponti, D., Costa, A., Zaffaroni, N., Pratesi, G., Petrangolini, G., Coradini, D., et al. (2005).
Isolation and in vitro propagation of tumorigenic breast cancer cells with stem/progenitor
cell properties. Cancer Res, 65(13), 5506-5511.
149

Portella, G., Cumming, S. A., Liddell, J., Cui, W., Ireland, H., Akhurst, R. J., et al. (1998).
Transforming growth factor beta is essential for spindle cell conversion of mouse skin
carcinoma in vivo: implications for tumor invasion. Cell Growth Differ, 9(5), 393-404.
Prince, M. E., Sivanandan, R., Kaczorowski, A., Wolf, G. T., Kaplan, M. J., Dalerba, P.,
et al. (2007). Identification of a subpopulation of cells with cancer stem cell properties in
head and neck squamous cell carcinoma. Proc Natl Acad Sci U S A, 104(3), 973-978.
Raynal, P., & Pollard, H. B. (1994). Annexins: the problem of assessing the biological
role for a gene family of multifunctional calcium- and phospholipid-binding proteins.
Biochim Biophys Acta, 1197(1), 63-93.
Reya, T., Morrison, S. J., Clarke, M. F., & Weissman, I. L. (2001). Stem cells, cancer,
and cancer stem cells. Nature, 414(6859), 105-111.
Rosen, J. M., & Jordan, C. T. (2009). The increasing complexity of the cancer stem cell
paradigm. Science, 324(5935), 1670-1673.
Rosengarth, A., Gerke, V., & Luecke, H. (2001). X-ray structure of full-length annexin 1
and implications for membrane aggregation. J Mol Biol, 306(3), 489-498.
Rosengarth, A., & Luecke, H. (2003). A calcium-driven conformational switch of the N-
terminal and core domains of annexin A1. J Mol Biol, 326(5), 1317-1325.
Rouleau, M., Leger, J., & Tenniswood, M. (1990). Ductal heterogeneity of cytokeratins,
gene expression, and cell death in the rat ventral prostate. Mol Endocrinol, 4(12), 2003-
2013.
Roy-Burman, P., Wu, H., Powell, W. C., Hagenkord, J., & Cohen, M. B. (2004).
Genetically defined mouse models that mimic natural aspects of human prostate cancer
development. Endocr Relat Cancer, 11(2), 225-254.
Samardzija, C., Quinn, M., Findlay, J. K., & Ahmed, N. (2012). Attributes of Oct4 in
stem cell biology: perspectives on cancer stem cells of the ovary. J Ovarian Res, 5(1), 37.
150

Santisteban, M., Reiman, J. M., Asiedu, M. K., Behrens, M. D., Nassar, A., Kalli, K. R.,
et al. (2009). Immune-induced epithelial to mesenchymal transition in vivo generates
breast cancer stem cells. Cancer Res, 69(7), 2887-2895.
Sarrio, D., Rodriguez-Pinilla, S. M., Hardisson, D., Cano, A., Moreno-Bueno, G., &
Palacios, J. (2008). Epithelial-mesenchymal transition in breast cancer relates to the
basal-like phenotype. Cancer Res, 68(4), 989-997.
Savagner, P., Kusewitt, D. F., Carver, E. A., Magnino, F., Choi, C., Gridley, T., et al.
(2005). Developmental transcription factor slug is required for effective re-
epithelialization by adult keratinocytes. J Cell Physiol, 202(3), 858-866.
Scheel, C., & Weinberg, R. A. (2012). Cancer stem cells and epithelial-mesenchymal
transition: concepts and molecular links. Semin Cancer Biol, 22(5-6), 396-403.
Schiffmann, E., Corcoran, B. A., & Wahl, S. M. (1975). N-formylmethionyl peptides as
chemoattractants for leucocytes. Proc Natl Acad Sci U S A, 72(3), 1059-1062.
Seemann, J., Weber, K., & Gerke, V. (1996). Structural requirements for annexin I-
S100C complex-formation. Biochem J, 319 ( Pt 1), 123-129.
Sellers, W. R., & Sawyers, C. A. (Eds.). (2002). Somatic genetics of prostate cancer:
Oncogenes and tumor suppressors. Philadelphia: Lippincott Williams & Wilkins.
Shackleton, M., Vaillant, F., Simpson, K. J., Stingl, J., Smyth, G. K., Asselin-Labat, M.
L., et al. (2006). Generation of a functional mammary gland from a single stem cell.
Nature, 439(7072), 84-88.
Shen, D., Chang, H. R., Chen, Z., He, J., Lonsberry, V., Elshimali, Y., et al. (2005). Loss
of annexin A1 expression in human breast cancer detected by multiple high-throughput
analyses. Biochem Biophys Res Commun, 326(1), 218-227.
Shen, D., Nooraie, F., Elshimali, Y., Lonsberry, V., He, J., Bose, S., et al. (2006).
Decreased expression of annexin A1 is correlated with breast cancer development and
progression as determined by a tissue microarray analysis. Hum Pathol, 37(12), 1583-
1591.
151

Siegel, R., Naishadham, D., & Jemal, A. (2013). Cancer statistics, 2013. CA Cancer J
Clin, 63(1), 11-30.
Silzle, T., Randolph, G. J., Kreutz, M., & Kunz-Schughart, L. A. (2004). The fibroblast:
sentinel cell and local immune modulator in tumor tissue. Int J Cancer, 108(2), 173-180.
Singh, A., & Settleman, J. (2010). EMT, cancer stem cells and drug resistance: an
emerging axis of evil in the war on cancer. Oncogene, 29(34), 4741-4751.
Singh, S. K., Hawkins, C., Clarke, I. D., Squire, J. A., Bayani, J., Hide, T., et al. (2004).
Identification of human brain tumour initiating cells. Nature, 432(7015), 396-401.
Skouteris, G. G., & Schroder, C. H. (1996). The hepatocyte growth factor receptor
kinase-mediated phosphorylation of lipocortin-1 transduces the proliferating signal of the
hepatocyte growth factor. J Biol Chem, 271(44), 27266-27273.
Smitherman, A. B., Mohler, J. L., Maygarden, S. J., & Ornstein, D. K. (2004). Expression
of annexin I, II and VII proteins in androgen stimulated and recurrent prostate cancer. J
Urol, 171(2 Pt 1), 916-920.
Solito, E., Christian, H. C., Festa, M., Mulla, A., Tierney, T., Flower, R. J., et al. (2006).
Post-translational modification plays an essential role in the translocation of annexin A1
from the cytoplasm to the cell surface. FASEB J, 20(9), 1498-1500.
Solito, E., Mulla, A., Morris, J. F., Christian, H. C., Flower, R. J., & Buckingham, J. C.
(2003). Dexamethasone induces rapid serine-phosphorylation and membrane
translocation of annexin 1 in a human folliculostellate cell line via a novel nongenomic
mechanism involving the glucocorticoid receptor, protein kinase C, phosphatidylinositol
3-kinase, and mitogen-activated protein kinase. Endocrinology, 144(4), 1164-1174.
Stevens, T. R., Smith, S. F., & Rampton, D. S. (1993). Antibodies to human recombinant
lipocortin-I in inflammatory bowel disease. Clin Sci (Lond), 84(4), 381-386.
Stingl, J., Eirew, P., Ricketson, I., Shackleton, M., Vaillant, F., Choi, D., et al. (2006).
Purification and unique properties of mammary epithelial stem cells. Nature, 439(7079),
993-997.
152

Studebaker, A. W., Storci, G., Werbeck, J. L., Sansone, P., Sasser, A. K., Tavolari, S., et
al. (2008). Fibroblasts isolated from common sites of breast cancer metastasis enhance
cancer cell growth rates and invasiveness in an interleukin-6-dependent manner. Cancer
Res, 68(21), 9087-9095.
Sun, H., Lesche, R., Li, D. M., Liliental, J., Zhang, H., Gao, J., et al. (1999). PTEN
modulates cell cycle progression and cell survival by regulating phosphatidylinositol
3,4,5,-trisphosphate and Akt/protein kinase B signaling pathway. Proc Natl Acad Sci U S
A, 96(11), 6199-6204.
Sun, H. T., Cohen, S., & Kaufmann, W. E. (2001). Annexin-1 is abnormally expressed in
fragile X syndrome: two-dimensional electrophoresis study in lymphocytes. Am J Med
Genet, 103(1), 81-90.
Sunderkotter, C., Steinbrink, K., Goebeler, M., Bhardwaj, R., & Sorg, C. (1994).
Macrophages and angiogenesis. J Leukoc Biol, 55(3), 410-422.
Sung, M. T., Jones, T. D., Beck, S. D., Foster, R. S., & Cheng, L. (2006). OCT4 is
superior to CD30 in the diagnosis of metastatic embryonal carcinomas after
chemotherapy. Hum Pathol, 37(6), 662-667.
Suzuki, H., Freije, D., Nusskern, D. R., Okami, K., Cairns, P., Sidransky, D., et al. (1998).
Interfocal heterogeneity of PTEN/MMAC1 gene alterations in multiple metastatic
prostate cancer tissues. Cancer Res, 58(2), 204-209.
Tagoe, C. E., Marjanovic, N., Park, J. Y., Chan, E. S., Abeles, A. M., Attur, M., et al.
(2008). Annexin-1 mediates TNF-alpha-stimulated matrix metalloproteinase secretion
from rheumatoid arthritis synovial fibroblasts. J Immunol, 181(4), 2813-2820.
Taipale, J., & Beachy, P. A. (2001). The Hedgehog and Wnt signalling pathways in
cancer. Nature, 411(6835), 349-354.
Taylor, A. D., Christian, H. C., Morris, J. F., Flower, R. J., & Buckingham, J. C. (1997).
An antisense oligodeoxynucleotide to lipocortin 1 reverses the inhibitory actions of
dexamethasone on the release of adrenocorticotropin from rat pituitary tissue in vitro.
Endocrinology, 138(7), 2909-2918.
153

Taylor, A. D., Cowell, A. M., Flower, R. J., & Buckingham, J. C. (1995). Dexamethasone
suppresses the release of prolactin from the rat anterior pituitary gland by lipocortin 1
dependent and independent mechanisms. Neuroendocrinology, 62(5), 530-542.
Thiery, J. P. (2002). Epithelial-mesenchymal transitions in tumour progression. Nat Rev
Cancer, 2(6), 442-454.
Thiery, J. P. (2003). Epithelial-mesenchymal transitions in development and pathologies.
Curr Opin Cell Biol, 15(6), 740-746.
Thiery, J. P., Acloque, H., Huang, R. Y., & Nieto, M. A. (2009). Epithelial-mesenchymal
transitions in development and disease. Cell, 139(5), 871-890.
Thiery, J. P., & Sleeman, J. P. (2006). Complex networks orchestrate epithelial-
mesenchymal transitions. Nat Rev Mol Cell Biol, 7(2), 131-142.
Torisu, H., Ono, M., Kiryu, H., Furue, M., Ohmoto, Y., Nakayama, J., et al. (2000).
Macrophage infiltration correlates with tumor stage and angiogenesis in human malignant
melanoma: possible involvement of TNFalpha and IL-1alpha. Int J Cancer, 85(2), 182-
188.
Tsao, F. H., Meyer, K. C., Chen, X., Rosenthal, N. S., & Hu, J. (1998). Degradation of
annexin I in bronchoalveolar lavage fluid from patients with cystic fibrosis. Am J Respir
Cell Mol Biol, 18(1), 120-128.
Tsujimura, A., Koikawa, Y., Salm, S., Takao, T., Coetzee, S., Moscatelli, D., et al. (2002).
Proximal location of mouse prostate epithelial stem cells: a model of prostatic
homeostasis. J Cell Biol, 157(7), 1257-1265.
Wang, S., Gao, J., Lei, Q., Rozengurt, N., Pritchard, C., Jiao, J., et al. (2003). Prostate-
specific deletion of the murine Pten tumor suppressor gene leads to metastatic prostate
cancer. Cancer Cell, 4(3), 209-221.
Wein, S., Fauroux, M., Laffitte, J., de Nadai, P., Guaini, C., Pons, F., et al. (2004).
Mediation of annexin 1 secretion by a probenecid-sensitive ABC-transporter in rat
inflamed mucosa. Biochem Pharmacol, 67(6), 1195-1202.
154

Wu, X., Gong, S., Roy-Burman, P., Lee, P., & Culig, Z. (2013). Current mouse and cell
models in prostate cancer research. Endocr Relat Cancer, 20(4), R155-R170.
Wu, X., Wu, J., Huang, J., Powell, W. C., Zhang, J., Matusik, R. J., et al. (2001).
Generation of a prostate epithelial cell-specific Cre transgenic mouse model for tissue-
specific gene ablation. Mech Dev, 101(1-2), 61-69.
Wu, Y., Deng, J., Rychahou, P. G., Qiu, S., Evers, B. M., & Zhou, B. P. (2009).
Stabilization of snail by NF-kappaB is required for inflammation-induced cell migration
and invasion. Cancer Cell, 15(5), 416-428.
Xin, L., Ide, H., Kim, Y., Dubey, P., & Witte, O. N. (2003). In vivo regeneration of
murine prostate from dissociated cell populations of postnatal epithelia and urogenital
sinus mesenchyme. Proc Natl Acad Sci U S A, 100 Suppl 1, 11896-11903.
Xin, L., Lukacs, R. U., Lawson, D. A., Cheng, D., & Witte, O. N. (2007). Self-renewal
and multilineage differentiation in vitro from murine prostate stem cells. Stem Cells,
25(11), 2760-2769.
Xin, W., Rhodes, D. R., Ingold, C., Chinnaiyan, A. M., & Rubin, M. A. (2003).
Dysregulation of the annexin family protein family is associated with prostate cancer
progression. Am J Pathol, 162(1), 255-261.
Yang, J., Mani, S. A., Donaher, J. L., Ramaswamy, S., Itzykson, R. A., Come, C., et al.
(2004). Twist, a master regulator of morphogenesis, plays an essential role in tumor
metastasis. Cell, 117(7), 927-939.
Yang, S., Pham, L. K., Liao, C. P., Frenkel, B., Reddi, A. H., & Roy-Burman, P. (2008).
A novel bone morphogenetic protein signaling in heterotypic cell interactions in prostate
cancer. Cancer Res, 68(1), 198-205.
Yang, Y. A., Dukhanina, O., Tang, B., Mamura, M., Letterio, J. J., MacGregor, J., et al.
(2002). Lifetime exposure to a soluble TGF-beta antagonist protects mice against
metastasis without adverse side effects. J Clin Invest, 109(12), 1607-1615.
155

156

Yazid, S., Leoni, G., Getting, S. J., Cooper, D., Solito, E., Perretti, M., et al. (2010).
Antiallergic cromones inhibit neutrophil recruitment onto vascular endothelium via
annexin-A1 mobilization. Arterioscler Thromb Vasc Biol, 30(9), 1718-1724.
Zhong, C., Saribekyan, G., Liao, C. P., Cohen, M. B., & Roy-Burman, P. (2006).
Cooperation between FGF8b overexpression and PTEN deficiency in prostate
tumorigenesis. Cancer Res, 66(4), 2188-2194.

Abstract (if available)

Abstract Annexin A1 (AnxA1), a phospholipid-binding protein and known regulator of glucocorticoid-induced inflammatory signaling, has been implicated for its role in human cancers, including head and neck cancer, colon, breast and prostate cancers. To determine the role of AnxA1 found to be secreted by the cancer-associated fibroblasts (CAF) of prostatic adenocarcinoma, we used primary cultures of CAFs, primary epithelial cultures and an epithelial cell line (cE1), all derived from the cPten-/-L mouse model of prostate adenocarcinoma in a series of experiments described here. The results of this dissertation point to a novel function of AnxA1 in prostate tumorigenesis and the establishment of a Cancer Cell-EMT-CSC lineage. It is projected that the presence of AnxA1 in the tumor microenvironment contributes to tumor stem cell activity via two separate but complementary pathways, first by induction of a de-differentiation process leading to generation of basal stem-like cells from a subpopulation of cancer epithelial cells, and, second, by enhancing the proliferation and differentiation of the CSCs. ❧ The cE1 cell line was used as a model system and after fractionation into subpopulations based on the levels of cell surface expression of Sca-1 (S) and CD49f (C) antigens: SCʰⁱ, SCᵐᵉᵈ and SCⁿᵒⁿᵉ, it was found that SCᵐᵉᵈ were preferentially susceptible to epithelial-to-mesenchymal transition (EMT) after treatment with CAF conditioned medium (CM) and AnxA1 mimetic (Ac2-26) protein, in vitro. Following EMT induction, these cells acquired stem cell-like properties similar to intrinsic properties of the SCʰⁱ group based on up-regulation of stem cell transcription factors (i.e. Oct4), as well as upregulation of EMT transcription factors (i.e. Snail and Twist) and mesenchymal markers (i.e. Vimentin). Gain of stem cell-like biological activity in vivo by SCᵐᵉᵈ after EMT induction by AnxA1 was demonstrated, using renal capsule transplantation technique, by gain of ability for formation of prostatic glandular structures with high expression of basal cell marker, p63, consistent with prostate stem/CSCs. ❧ Primary prostate cancer epithelial cell cultures from the model were used to further define the role of AnxA1 on cells from the CSC niche. Primary cultures of Lin⁻ (Ter119, CD31, CD45

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Study of bone morphogenetic protein-2 and stromal cell derived factor-1 in prostate cancer

PDF

Exploration of the roles of cancer stem cells and survivin in the pathogenesis and progression of prostate cancer

PDF

Study of the role of bone morphogenetic proteins in prostate cancer progression

PDF

The noncanonical role of telomerase in prostate cancer cells: exploring a non-telomeric signaling role for telomerase protein (TERT) in a cancer cell line

PDF

Homologous cell systems for the study of progression of androgen-dependent prostate cancer to castration-resistant prostate cancer

PDF

The cancer stem-like phenotype: therapeutics, phenotypic plasticity and mechanistic studies

PDF

Studies of murine prostate cancer stem / progenitor cells

PDF

Bimodal effects of bone morphogenetic proteins in prostate cancer

PDF

Role of a novel transmembrane protein, TMEM56, in tumorigenic growth of human PC3 prostate cancer cell line

PDF

Identification of CBP/FOXM1 as a molecular target in triple negative breast cancer

PDF

Long-term expansion of human alveolar epithelial cells as a novel model system to study lung disease progression in vitro

PDF

Role of endoplasmic reticulum chaperones GRP78 and GRP94 in regulating unfolded protein response and cell survival

PDF

Role of the bone marrow niche components in B cell malignancies

PDF

Harnessing the power of stem cell self-renewal pathways in cancer: dissecting the role of BMI-1 in Ewing’s sarcoma initiation and maintenance

PDF

The role of LGR5 in the pathogenesis of Ewing sarcoma: a marker of aggressive disease and a contributor to the malignant phenotype

PDF

Immunomodulatory and regenerative potential of amniotic fluid stem cells as a treatment strategy for pulmonary fibrosis

PDF

The role of endoplasmic reticulum chaperones in regulating hematopoietic stem cells and hematological malignancies

PDF

Jagged1 functions downstream of Twist1 in the specification of the coronal suture and the formation of a boundary between osteogenic and non-osteogenic cells

PDF

Novel roles for Maf1 in embryonic stem cell differentiation and adipogenesis

Asset Metadata

Creator Geary, Lauren Alexandra Rios (author)

Core Title Role of cancer-associated fibroblast secreted annexin A1 in generation and maintenance of prostate cancer stem cells

School Keck School of Medicine

Degree Doctor of Philosophy

Degree Program Integrative Biology of Disease

Publication Date 02/01/2014

Defense Date 06/19/2013

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

Tag Adenocarcinoma,annexin A1,cancer stem cells,EMT,OAI-PMH Harvest,prostate cancer

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the USC Libraries (provenance)

Advisor Roy-Burman, Pradip (committee chair), Adams, Gregor B. (committee member), Hinton, David R. (committee member), Zandi, Ebrahim (committee member)

Creator Email lgeary@usc.edu,lgeary83@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-327884

Unique identifier UC11292768

Identifier etd-GearyLauren-20130619.pdf (filename),usctheses-c3-327884 (legacy record id)

Legacy Identifier etd-GearyLauren-20130619.pdf

Dmrecord 327884

Document Type Dissertation

Format application/pdf (imt)

Rights Geary, Lauren Alexandra Rios

Type texts

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

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA