<%BANNER%>

Characterization of Antiestrogen Function in MCF7-Derived Mammospheres

Permanent Link: http://ufdc.ufl.edu/UFE0041922/00001

Material Information

Title: Characterization of Antiestrogen Function in MCF7-Derived Mammospheres
Physical Description: 1 online resource (147 p.)
Language: english
Creator: Ao, Ada
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: alpha, antiestrogen, breast, cancer, estrogen, mammosphere, mcf7, receptor, tamoxifen
Biochemistry and Molecular Biology (IDP) -- Dissertations, Academic -- UF
Genre: Medical Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The emerging cancer stem cell (CSC)/ tumor-initiating cell (TIC) hypothesis seeks to explain cancer persistence and recurrence. It supposes that a small subgroup of cells within tumors is multipotent and has chemoresistant features, which contributes to the heterogeneous and regenerative phenotype of recurring tumors. In breast cancer, previous studies have shown the TIC subgroup can be propagated in-vitro from both primary tumors and established cell lines by culturing the cells as 'mammospheres'. The goal of this project is to characterize the antiestrogen response of mammospheres (MCF7S) derived from estrogen receptor(ER)-alpha positive mammary epithelial MCF7 cells, and to determine if there is enrichment of putative TICs after antiestrogen challenge. It is shown here that MCF7S cell proliferation was decreased by antiestrogen treatments. Sphere formation was affected by a selective estrogen receptor modulator (SERM), but not by a selective estrogen receptor down-regulator (SERD). Using the sphere formation assay, we determined that the sphere forming ability returned after drug removal and that there was no significant decrease in sphere formation frequency. Surprisingly, stable ER? knockdown MCF7S cells were more sensitive to SERM and exhibited diminished sphere formation frequency. Treatment of ER-alpha knockdown cells with SERD did not significantly change cell proliferation, sphere formation or frequency. In summary, the data showed that a stable fraction of potential TICs remained after antiestrogen challenge. The results also demonstrated that ER-alpha is not essential for MCF7S survival, but may mediate antiestrogen function. Further studies on SERMs may be warranted as SERMs may target potential TICs that use alternative mitogenic pathways.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Ada Ao.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Lu, Jianrong.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041922:00001

Permanent Link: http://ufdc.ufl.edu/UFE0041922/00001

Material Information

Title: Characterization of Antiestrogen Function in MCF7-Derived Mammospheres
Physical Description: 1 online resource (147 p.)
Language: english
Creator: Ao, Ada
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: alpha, antiestrogen, breast, cancer, estrogen, mammosphere, mcf7, receptor, tamoxifen
Biochemistry and Molecular Biology (IDP) -- Dissertations, Academic -- UF
Genre: Medical Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The emerging cancer stem cell (CSC)/ tumor-initiating cell (TIC) hypothesis seeks to explain cancer persistence and recurrence. It supposes that a small subgroup of cells within tumors is multipotent and has chemoresistant features, which contributes to the heterogeneous and regenerative phenotype of recurring tumors. In breast cancer, previous studies have shown the TIC subgroup can be propagated in-vitro from both primary tumors and established cell lines by culturing the cells as 'mammospheres'. The goal of this project is to characterize the antiestrogen response of mammospheres (MCF7S) derived from estrogen receptor(ER)-alpha positive mammary epithelial MCF7 cells, and to determine if there is enrichment of putative TICs after antiestrogen challenge. It is shown here that MCF7S cell proliferation was decreased by antiestrogen treatments. Sphere formation was affected by a selective estrogen receptor modulator (SERM), but not by a selective estrogen receptor down-regulator (SERD). Using the sphere formation assay, we determined that the sphere forming ability returned after drug removal and that there was no significant decrease in sphere formation frequency. Surprisingly, stable ER? knockdown MCF7S cells were more sensitive to SERM and exhibited diminished sphere formation frequency. Treatment of ER-alpha knockdown cells with SERD did not significantly change cell proliferation, sphere formation or frequency. In summary, the data showed that a stable fraction of potential TICs remained after antiestrogen challenge. The results also demonstrated that ER-alpha is not essential for MCF7S survival, but may mediate antiestrogen function. Further studies on SERMs may be warranted as SERMs may target potential TICs that use alternative mitogenic pathways.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Ada Ao.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Lu, Jianrong.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041922:00001


This item has the following downloads:


Full Text





CHARACTERIZATION OF ANTIESTROGEN RESPONSE IN MCF7-DERIVED
MAMMOSPHERES




















By

ADA SI NGA AO


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2010

































@2010 Ada Si Nga Ao



























To piling higher and deeper









ACKNOWLEDGMENTS

I cannot do justice to all those who helped me to reach this point. I extend my

thanks to all my friends for their sympathetic ear and understanding throughout this

endeavor. Thanks to Nicole, Amanda, Shermi, Star, Michele, Julia, Zhou, Joeva,

Carlos, and Heather. I sincerely thank my family for their support for everything I've

ever done in my life, even when they don't necessarily agree with my pursuits. Special

thanks to my cousins, especially Amy and Diane, for keeping me grounded. My

deepest thanks and appreciation to Dr. Jude Deeney, Dr. Emma Heart, Dr. Lina Moitoso

de Vargas, Dr. John Flanagan, Dr. Ann-Marie Richards, Liping Hu, Dr. Lihan Liu, Dr.

Rita Avancini, and Dr. Isabel Chiu for being valuable role models during my early days

in a lab.

I would like to thank Dr. Jianrong Lu for the opportunity to perform my graduate

training in his lab. Thanks to my labmates: Sushama for commiserating with me, and

Tong for displaying an almost Zen-like calm through everything that boggles the mind.

To Tommy, I have to admit I don't know you too well but I wish you the best for the

future. I would also like to thank my advisory committee (Drs. Linda Bloom, Jo rg

Bungert, Harry Nick, and Brent Reynolds) for all their helpful comments throughout my

training, especially Dr. Nick and Dr. Bloom for their patience during a difficult part of the

process.

Finally, a special thanks to Will for his continued love, encouragement, and

friendship. It is a privilege to share his dark, twisted sense of humor and a joy to know

his inner goodness. He has been right beside me through all the ups and downs, and

for that I'm very grateful.









TABLE OF CONTENTS

page

A C KN O W LEDG M ENTS ............................. ............... ........................................ 4

L IS T O F T A B L E S ............................................................................................................ 8

LIS T O F F IG U R E S .................................................................. 9

LIST O F A BBR EV IATIO NS .............................................. 11

A B S T R A C T ................................................................................................................... 1 3

CHAPTER

1 IN T R O D U C T IO N .................................................... .......... 15

Introduction to Mammary Gland Development............................... 15
Breast Cancer.............................. ......... 16
Types of Breast Cancer..................... ....................... .................... 17
Ductal Carcinom a......................................... ............... 17
Lobular C arcinom a .............................................. ..... ....................... 17
Other Types of Breast Cancer..................................... ..... 18
E stroge n R ece pto r A lpha ..................................................... ............... ............ 19
Estrogen Receptor Alpha Signaling.................................... 20
G enom ic Function .............................................. 20
Non-Genomic Function ...................... ......... .................... 22
C ell of O rigin ........................................................................... ....... ................. 23
Clonal Evolution M odel...................................................... ... ............... 23
The Cancer Stem Cell (CSC) Hypothesis............................ ............... 23
Tumor Heterogeneity and CSC Hypothesis................... ....... 24
Implications of Each Model on Breast Cancer Recurrence ............................ 25
Estrogen receptor alpha, stem cells and breast cancer ........................... 26
Contribution of CSC/TIC to antiestrogen resistance ............................... 28
CSC Isolation and Detection .................. .............. .......... 31
Other Examples of CSC Enrichment ...................................................... 32

2 GENERAL MATERIALS AND METHODS ........... .... .................................. ..... 36

Cell Culture .............. .. ... .......... ... ................. 36
MCF7S Cell Proliferation and Sphere Formation Assay .............................. ... 37
Serial Passaging ...... ...................... .......... ........ 37
Cell Proliferation Assay ..... .................................. .... .... ............ 37
Sphere Form action Assay ........ ........................................................ ............ 38
RNA Isolation .............................. ......................................38
Reverse Transcriptase PCR (RT-PCR) ...................... ................................ 39
R e a l-T im e P C R ............................................................................................. 3 9









Total Protein Isolation .................. .......... ...................... ... ............. 40
Nuclear Extraction ................ ...... .......... ......... 40
Immunoblot Analysis........................................... ........ 41
Immunofluorescence..................................... .............. ..... 42
short-hairpin RNA (shRNA) Vector Construction .................. .......... 43
shRNA Oligo Design and Cloning ........................................ 43
Plasmid DNA MiniPrep .................... ............ .................... .... ......... 44
Retrovirus Production and Transduction of Target Cells................................... 45
Transient Transfection.......................................... 45
Retroviral Infection for Stable Integration ...... ...... ................ ............... 45
Selection and Enrichment of Single Clones...................... ............ 46
Analysis of CD44 Expression................................................. 47
BrdU/Propidium Iodide Cell Proliferation Assay ......................................... 47
Annexin V-PE Apoptosis Detection using BD Pharmingen Kit ...... ........................ 48
Statistical Analysis ...... .. ................................................... ............................ 48
In V ivo Tum origenic Assay ...................... ....... ......... .............................. 48
Preparation of Mice ............................................... 49
Preparation of Tumor Cells............................ .................... 49
Mouse Injections .................... .............. ............... 49

3 CHARACTERIZING MAMMOSPHERES DERIVED FROM MCF7 PARENTAL
C E L L S .............. ..... ............ ................. ................................................... 5 4

Introduction ................... .. ......... ................ 54
Results ................................ ......... ............... ....... .................... 55
Characterization of Mammospheres (MCF7S) Derived from MCF7 Cells........ 55
Expression of Putative Breast Tumorigenic Marker CD44 in MCF7S Cells...... 56
ERa Status and Stability in MCF7S Cells ............. .. .................. 57
D discussion ............. ................................................................................. 58

4 THE EFFECTS OF ANTIESTROGEN ON MAMMOSPHERE FORMATION.......... 66

Introduction ................... ........ ............... 66
Results ......... .. ...... ......... ......................... ............... 68
M CF7S Response to Antiestrogens ......................................................... ........ 68
Antiestrogen Efficacy in Parental MCF7P and MCF7S ................ ............... 69
Sphere Formation Frequency Following Antiestrogen Challenge.................. 71
MCF7S Cell Proliferation under Long-Term Antiestrogen Treatment ............... 72
MCF7S Cell Cycle Analysis ........................................ ... .............. 73
M C F7S A poptosis A ssay.................................................. 74
D is c u s s io n ..................................................................... 7 5

5 THE ROLE OF ESTROGEN RECEPTOR ALPHA ON MAMMOSPHERE
F O R M A T IO N .......... ......... ........... .... .................... ............... 8 7

Introd uctio n .................................... ................ ............... 8 7
Results.........n a ........................ 88.. .....8
R e s u lts .................................................................................................................... 8 8









Proliferation of ERa Knockdown M CF7S....................................................... 88
Antiestrogen Response of ERa Knockdown MCF7S...................... ........ 88
Sphere Formation Frequency of ERa Knockdown MCF7S Following
Antiestrogen Challenge .............................. .. .. ........... .............. 89
Using SERDs and SERMs to Mimic shER Effects in MCF7S.......................... 90
D iscu ss io n .......... ......... .................. ...... .................................... 9 1

6 CONCLUSIONS AND FUTURE DIRECTIONS .................... .................. 98

Conclusions and Discussion ............. ............. ......................... 98
Future D directions .......... ............. ................. ................................. 102

APPENDIX

A SUPPLEMENTAL FIGURES ......................................... 104

C h a p te r 3 .............. ................. .............................................................. 10 4
C h a p te r 4 .............. ................. .............................................................. 10 5
Chapter 5 ............... ........ ......................... 112

B BRD8 FUNCTION DURING EARLY CARDIAC DEVELOPMENT .................... 113

In tro d u ctio n .................................................................................. .............. ........ 1 13
Overview of Mammalian Heart Development .................................. 113
Transcription Factors Regulating Heart Formation .............. ........ ...... ......... 114
Specification of Myocardial Progenitor Cells ............... .... ................ 114
Differentiation of Precursors into Cardiomyocytes ...................................... 115
Downstream Transcriptional Networks Associated With Heart Formation ..... 115
General Introduction to Brd8 ........ ........................................................ .. ...... 116
Materials and Methods................ ................................. 116
Im munohistochemistry......................................................... 116
W hole Mount In Situ Hybridization (W HISH) ............................. 117
Riboprobe synthesis ....... .............................. ...... 117
Mouse embryo preparation ......... ...................... ... .................. 118
W hole m ount in situ hybridization .............. ............. ................ ...... ... 118
RT-PCR Primers for Brd8 Allele Expression ............... ........ ......... 119
Results and Discussion ....................................... 119

LIST O F R EFER EN C ES ...................................................... ............... 130

BIOGRAPHICAL SKETCH .............. .. ........ ................. 147









LIST OF TABLES


Table page

2-1 shRNA Oligos designed for retrovirus-mediated knockdown ........................... 52

2-2 Antibodies used for Immunoblotting (IB), Immunofluorescence (IF), or Flow
Cytom etry (FC) ........................... ................ ........ .. ..... .......... 53

2-3 Prim ers used for real-tim e RT-PC R ............................................. .... .. ............... 53

3-1 Data for CD44-FITC signal quantification using flow cytometer....................... 64

4-1 Growth kinetics of MCF7S long term expansion under antiestrogen
challenge. ................ ......... ............... ... ......... ............. 84

A-1 Additional data for CD44-FITC signal quantification using flow cytometer
(Figure A -1). .......... .. .... ..... ........... ................... 104

A-2 Data for BrdU/PI cell cycle analysis MCF7S cells treated 72 hours with
antiestrogens (Figure A -5) ......................................................... ..... ....... 110









LIST OF FIGURES


Figure page

1-1 E strogen receptor isoform s...................................................... .... ................. 34

1-2 Mammary gland differentiation hierarchy........ .......................... ............. .. 34

1-3 Potential relationship between mammary differentiation hierarchy and the
develop ent of tum or subtypes............................................... ... .................. 35

2-1 Tissue culture scheme for MCF7S proliferation assay and sphere formation
a s s a y ..................... .. .. ......... .. .. ......... ....................................... 5 0

2-2 Cloning vector information for microRNA-adapted retroviral vector................ 51

3-1 MCF7-derived mammospheres (MCF7S) growth curve .................................. 61

3-2 Phase contrast microscopy of mammospheres derived from MCF7 parental
ce lls ......... ......... ......... .................................. ........................... 6 1

3-3 Enriching MCF7S from MCF7P by serial passage in mammosphere media...... 62

3-4 Comparing CD44 expression in MCF7P vs. MCF7S using flow cytometer ....... 63

3-5 ERa expression in MCF7P and MCF7S.......... ... ..... ...... ................... 65

4-1 Mammosphere formation in the presence of antiestrogens.............................. 78

4-2 Cell proliferation of MCF7S in the presence of antiestrogens............................ 79

4-3 MCF7 adherent culture response to 4-hydroxytamoxifen (4-OHT) with
various culture m edia. ..................... .... ................................................ 80

4-4 Immunoblot of cytoplasmic and nuclear fractions from MCF7P and MCF7S
cells treated for 72 hours with 4-hydroxytamoxifen (4-OHT) or ICI 182780
(ICI) to characterize ERa stability. .............. ....... .... ......... ............. 81

4-5 Sphere formation frequency of MCF7S after antiestrogen challenge using 4-
hydroxytamoxifen (4-OHT) or ICI 182780 (ICI)................................... ... 82

4-6 Long-term expansion of MCF7S in the presence of antiestrogens. The lines
are expressed on a semilog graph and slope of each line was calculated as
log expansion for each condition .............................................. 83

4-7 Propidium iodide (PI) cell cycle analysis................................ ..................... 85

4-8 Annexin V apoptosis assay for antiestrogen treated MCF7S ........................... 86









5-1 Knockdown of ERa (shER) in MCF7S cells................................ ... ................ 93

5-2 Representative phase-contrast images of MCF7S, control shRNA MCF7S,
and shER clone 7 exposed to antiestrogens for 6 days................................... 94

5-3 Antiestrogen response of ERa knockdown in MCF7S cells.............................. 95

5-4 Sphere formation frequency of antiestrogen treated ERa knockdown MCF7S
ce lls. ......... .... .............. .................................. ........................... 96

5-5 SERDs and SERMs combination treatment to mimic shER effects in MCF7S... 97

A-1 Additional figures comparing CD44-expression in MCF7P and MCF7S........... 104

A-2 Real time RT-PCR of MCF7P for estrogen response gene TFF1.................. 105

A-3 Real time RT-PCR of MCF7S for estrogen response gene TFF1..................... 106

A-4 Time course of CTSD gene expression induced with 100 nM 173-estradiol
(E2) ............................................. ........... 107

A-5 BrdU/PI cell cycle analysis for MCF7S treated with antiestrogens for 72
h o u rs ......... .... .............. ................................ ........................... 10 9

A-6 Individual Annexin V experimental data for antiestrogen treated MCF7S at 48
hours and 72 hours summarized in Figure 5-5 ............. ................................ 111

A-7 Representative phase-contrast images of shER C7 knockdown (left) and
control shRNA (right) mammospheres. ................ ............. ............... 112

B-1 Gene trap vector inserted into mouse genome to generate Brd8 null allele..... 122

B-2 Brd8 gene-trapped mouse embryos and expression of wild type and mutant
a lle le s ......... ................................................. ............................... 1 2 3

B-3 Endogenous Brd8 expression in wild-type E10.5 mouse embryos using
whole mount in situ hybridization (W MISH) ...... ...... .. .... .................. 124

B-4 Hematoxylin and eosin stained sections of Brd8 mutant embryos (E10.5)
compared to wild-type littermates ....... ...................... .............. 125

B-5 PECAM whole mount immunohistochemistry comparing vascularization of
Brd8 mutant and stage matched wild-type embryos (E8.5).............................. 127

B-6 Whole mount in situ hybridization (WMISH) characterizing expression of key
cardiac development genes.................. .......... ... ..... ............... 128











4-OHT

AE

AKT

AP-1

BrdU

BSA

CSC

EGF

ERa

ERE

ERK

ERRy

FGF

ICI

MAPK

MaSC

MCF7P

MCF7S

NFKB

PI

P13K

PKA

PKC

Poly-HEMA


LIST OF ABBREVIATIONS

4-Hydroxytamoxifen

Antiestrogen

Protein Kinase B

Activator Protein-1

Bromodeoxyuridine, 5-bromo-2-deoxyuridine

Bovine Serum Albumin

Cancer Stem Cell

Epidermal Growth Factor

Estrogen Receptor Alpha

Estrogen Response Element

Extracellular Signal-Regulated Kinase

Estrogen Related Receptor Gamma

Fibroblast Growth Factor

ICI 182780 (Fulvestrant, FASLODEX)

Mitogen-Activated Protein Kinase

Mammary Stem Cell

MCF7 Parental

MCF7 Spheroid Culture (mammospheres)

Nuclear Factor kappa B

Propidium Iodide

Phosphoinositide 3-Kinase

Protein Kinase A

Protein Kinase C

Poly (2-hydroxyethyl methacrylate)









RT Reverse Transcriptase

RT-PCR Reverse Transcriptase-Polymerase Chain Reaction

SERD Selective Estrogen Receptor Down-regulators

SERM Selective Estrogen Receptor Modulators

SFA Sphere Formation Assay

shER shRNA knockdown Estrogen Receptor Alpha

shRNA short hairpin RNA

SP-1 Specificity Protein-1

TAM Tamoxifen

TIC Tumor Initiating Cell









Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy


CHARACTERIZATION OF ANTIESTROGEN RESPONSE IN MCF7-DERIVED
MAMMOSPHERES

By

Ada Si Nga Ao

August 2010

Chair: Jianrong Lu
Major: Medical Sciences--Biochemistry and Molecular Biology

The emerging cancer stem cell (CSC)/ tumor-initiating cell (TIC) hypothesis seeks

to explain cancer persistence and recurrence. It supposes that a small subgroup of

cells within tumors is multipotent and has chemoresistant features, which contributes to

the heterogeneous and regenerative phenotype of recurring tumors. In breast cancer,

previous studies have shown the TIC subgroup can be propagated in-vitro from both

primary tumors and established cell lines by culturing the cells as "mammospheres".

The goal of this project is to characterize the antiestrogen response of mammospheres

(MCF7S) derived from estrogen receptor (ERa)-positive mammary epithelial MCF7

cells, and to determine if there is enrichment of putative TICs after antiestrogen

challenge. It is shown here that MCF7S cell proliferation was decreased by

antiestrogen treatments. Sphere formation was affected by a selective estrogen

receptor modulator (SERM), but not by a selective estrogen receptor down-regulator

(SERD). Using the sphere formation assay, we determined that the sphere forming

ability returned after drug removal and that there was no significant decrease in sphere

formation frequency. Surprisingly, stable ERa knockdown MCF7S cells were more









sensitive to SERM and exhibited diminished sphere formation frequency. Treatment of

ERa knockdown cells with SERD did not significantly change cell proliferation, sphere

formation or frequency. In summary, the data showed that a stable fraction of potential

TICs remained after antiestrogen challenge. The results also demonstrated that ERa is

not essential for MCF7S survival, but may mediate antiestrogen function. Further

studies on SERMs may be warranted as SERMs may target potential TICs that use

alternative mitogenic pathways.









CHAPTER 1
INTRODUCTION

Introduction to Mammary Gland Development

The mammary gland is unusual namely because it does not fully develop until

puberty in response to systemic hormones and local growth factors. The mammary

gland is a tree-like structure made up of multiple cell lineages that form the structural

units of the gland. Each individual glandular off-shoot from the main branching duct is

termed the terminal duct lobular unit (TDLU) in human, or terminal end bud (TEB) in

rodents (Howard and Gusterson, 2000; Sternlicht, 2006). The TDLU is comprised of

terminal ductules with a bilayer of polarized luminal epithelial cells lining the inside

surrounded by a layer of contractile myoepithelial cells, topped with a cluster of alveoli

cells (Visvader, 2009). Multiple TDLUs composes each of the 15-20 lobes in a human

breast, with each lobe imbedded in adipose tissue (Morrison et al., 2008).

In rodents, a simpler analogous structure is the terminal end bud (TEB), and has

been more extensively studied. The TEB is made up of luminal-restricted cells called

body cells, surrounded by a layer of bipotent progenitors called cap cells. The TEB

undergoes dynamic expansion and restructuring during puberty and pregnancy directed

by physiological cues. The body cells develop into milk-producing alveoli cells during

pregnancy. The cap cells also undergo differentiation into luminal or myoepithelial

lineage during pregnancy (Howard et al., 2000; Smalley and Ashworth, 2003).

The breast undergoes additional transformation post-pregnancy and lactation

called involution, in which the TDLUs or TEBs shrink back to pre-pregnancy state. The

process can occur multiple times in a woman's lifetime, giving credence to a potential

reservoir of lineage-restricted progenitor cells that sustain multiple rounds of breast









remodeling. A potential stem-cell hierarchy in postnatal breast tissue has gathered

support, and may be a source of aberrant progenitor cells that increase breast cancer

tumorigenic potential in an analogous fashion to hematopoietic cancers (Villadsen,

2005; Smalley et al., 2003). However, there is debate concerning the permanent or

transient nature of this pool, and also the localization of these stem cells within the

mammary gland structure.

Breast Cancer

Breast cancer has been one of the more persistent diseases in the United States

for women. It has an age-adjusted incidence rate of 122.8 per 100,000 women each

year (Altekruse et al., 2009). Despite advances in early detection and diagnosis, there

has been little change in mortality rate once the tumor metastasizes. However, the first

5 years is generally favorable (NCI, SEER). Current treatments can eliminate the bulk

of the tumor and reduce tumor mass, but it often returns with a more aggressive

phenotype. The question of tumor recurrence has vexed researchers as they try to

identify the root cause. It was first elegantly termed the "seed and soil" hypothesis,

proposed by Stephen Paget in 1889. Most studies have focused on genetic or

molecular factors that may promote tumorigenic phenotypes such as evasion of cell

death, self-renewal, increased proliferation, resistance to anti-growth signal,

angiogenesis, tissue invasion and metastasis (Hanahan and Weinberg, 2000). There

are recent attempts to find and to target the "seed" that sustains tumor initiation as the

means to long-term tumor suppression. This phenomenon shows a shift in focus in

cancer biology: from studying the deregulation of normal cellular functions to searching

for a specific cell type equipped with tumorigenic cellular processes by default.









Types of Breast Cancer

Breast cancer classification is based on criteria such as the invasiveness, point of

origin, proliferative potential, and hormone receptor status. These pathological

parameters were designed to evaluate tumor progress for diagnosis and prognosis.

The information provided by such assessments guide the formation of treatment plans.

Breast cancer types fall within four general categories: ductal carcinoma in situ (DCIS),

lobular carcinoma in situ (LCIS), invasive ductal carcinoma (IDC), and invasive lobular

carcinoma (ILC). Some tumors may contain a combination of these types. There are

also rarer subtypes that are classified within each category (American Cancer Society,

2009).

Ductal Carcinoma

DCIS is the most common type of non-invasive breast cancer and has the best

prognosis, as they usually contain hormone-receptor positive cells that may respond to

anti-hormone therapy (ACS, 2009). The cancer cells are typically of luminal origin and

are located inside the duct, but have not invaded surrounding tissue or lymph nodes.

IDC is the invasive variety of ductal carcinoma when the cancer cells invade the

duct wall and begins to spread into the fatty tissue of the breast and lymph nodes (ACS,

2009). They may then spread to other parts of the body through the lymphatic system

and bloodstream. IDC is the most common type of breast cancer, but it can be further

partitioned into other invasive breast cancer types. As metastasized cells, they are

typically less responsive to treatments, fast growing, and have worse prognosis.

Lobular Carcinoma

LCIS may be classified as non-invasive breast cancer, but patients with this

cancer type have a higher risk of developing invasive cancer. The cancer cells are









localized to the lobular glands and also contain hormone receptor positive cells that may

be responsive to treatment (ACS, 2009).

ILC, like IDC, is the invasive version of the in situ cancer type and are difficult to

treat and to track progression using standard detection methods like mammograms. It

is a rarer subtype of invasive breast cancer, making up approximately 10% of all

invasive breast cancer (ACS, 2009).

Other Types of Breast Cancer

There are less common types of breast cancer that do not readily conform to the

above subgroups. They have a more complex pathology and are not fully understood.

A few of these types are briefly described as examples. They are chiefly fast growing,

invasive cancer cells that require aggressive therapy to control.

Inflammatory breast cancer (IBC) is a rare form of invasive breast cancer that may

be mistaken for an infection. It elicits an inflammatory response due to blockage of

lymph vessels in the skin by cancer cells. IBC tends to grow quickly and aggressively.

They are insensitive to antihormone and are treated with chemotherapy or radiation

(ACS, 2009).

Triple negative breast cancers do not express estrogen, progesterone, or HER2

receptors. This type of cancer is especially aggressive as there is no regulatory

signaling from growth receptors, and they are completely non-responsive to receptor-

mediated inhibitors. It usually requires aggressive chemotherapy to check growth,

although results are temporary and recurrence rate is high (ACS, 2009).

Medullary carcinoma is considered a subtype of IDC with cells derived from milk

duct cells. It has a well-defined boundary between tumor and normal tissue as

abnormally large cancer cells are surrounded by immune cells along the periphery. The









prognosis for this cancer type is better than most invasive breast cancer and may be

successfully treated with standard therapies (ACS, 2009).

Estrogen Receptor Alpha

There are two estrogen receptor isoforms, ERa and ER3. They are members of

the nuclear receptor superfamily, and function as ligand-activated transcription factors.

The two proteins are encoded on different chromosomes, have distinct localization

pattern, and subtle differences in ligand binding affinity and structure (Figure 1-1). In

general terms, the two isoforms are present in the breast and have antagonistic

functions. ER3 can heterodimerize with ERa and inhibit activation of ERa target genes

(Jones et al., 1999; Kuiper et al., 1998; Zhu et al., 2006; Ariazi et al., 2006). Both

receptors contain five distinct domains (Green and Carroll, 2007). From the N-terminus,

domain A/B contains the transcriptional activation function 1 (AF-1) region that

modulates receptor activity. Domain C contains the DNA binding domain (DBD)

composed of cysteine-rich zinc fingers. Domain D is a hinge region, followed by

domains E and F at the C-terminus. The E domain is the ligand binding domain (LBD)

and also contains the second activation function region (AF-2). F domain is a variable

region between the 2 ER isoforms. As the MCF7 cell line does not express ER3

significantly (Lindberg et al., 2010; Paruthiyil et al., 2004), ER3 will be omitted in this

introduction.

In mammalian tissues, ERa is known to be endogenously expressed at high levels

in the mammary gland, male and female reproductive tracts, bone, the cardiovascular

system, and parts of the brain. Knockout studies in mice and ERa-deficiency studies in

humans have reported phenotypes including infertility and inhibition of normal









physiological changes during puberty (Couse and Korach, 1999; Curtis Hewitt et al.,

2000). Present knowledge maintains that ERa has a general role in modulating cell

proliferation during embryonic development, but it is not essential. However, the

receptor has a profound role in postnatal development, especially during puberty and

pregnancy as the mammary gland undergoes extensive remodeling.

Estrogen Receptor Alpha Signaling

The unliganded form of ERa is sequestered within the nuclei in inhibitory

complexes composed of heat shock protein (Hsp90) chaperones (Fliss et al., 2000).

Upon ligand binding and activation, ERa dissociates from the inhibitory complex and

may regulate cellular proliferation and target gene expression through direct or indirect

DNA interaction, or participate in non-genomic mitogenic signaling (Ariazi et al., 2006;

Bjornstrom and Sjoberg, 2005).

Genomic Function

The classical function of ERa is mainly through genomic means as the receptor

regulates target gene expression. In the classic genomic model, the receptors dimerize

upon ligand binding and interact directly with DNA on the promoter region of target

genes. The receptor binds preferentially on conserved cis-element sequences called

the Estrogen Response Element (ERE) (Hayashi and Yamaguchi, 2008). Recent

genome wide ERa-DNA interaction studies have uncovered novel binding sites and new

classes of ERa regulated genes (Carroll et al., 2006; Lin et al., 2007).

ERa may also bind DNA without activation by ligands. Ligand-independent

binding is typically induced by receptor phosphorylation, and mediated by growth factor

activated kinase cascades (Dudek and Picard, 2008; Lannigan, 2003). These kinases









activate ERa by phosphorylating specific amino acid residues. Eight specific residues

have been reported and each is associated with a specific kinase cascade (Hayashi et

al., 2008). Cell cycle regulated cyclin A/cdk2 complex phosphorylates S104/106.

MAPK is known to phosphorylate S118 (Atanaskova et al., 2002). Akt (or PKB), in

complex with P13K, can phosphorylate S167 (Stoica et al., 2003; Campbell et al., 2001).

These phosphorylation sites are located within the activation function 1 (AF-1) region,

which resides in the A/B domain, of the receptor. Phosphorylation of this region was

shown to upregulate ERa transcriptional activity by inducing structure conformational

changes that promotes coactivator interactions. PKA is able to phosphorylate S236,

located in the core DNA binding domain, and inhibit dimerization (Chen et al., 1999).

p38 MAPK is known to phosphorylate T311 in the C-terminal domain (Lee and Bai,

2002). The Src kinase phosphorylates the receptor at Y537 (Arnold et al., 1995). A

novel phosphorylation site at the extreme C-terminus has recently been identified at

S559, and it is phosphorylated by protein kinase CK2 (Williams et al., 2009).

ERa can activate target gene expression indirectly through protein-protein

interactions with other cofactors via the AF-1 or AF-2 region (Baek et al., 2002; Shang

et al., 2000; Jakacka et al., 2001; Metivier et al., 2003). Such interactions are generally

modulated by conformational changes that accompany ligand binding or

phosphorylation of ERa and its protein partner; a few examples include AP-1, SP-1,

STAT and NFKB (McDonnell and Norris, 2002). Recently discovered cofactors include

epigenetic chromatin modifiers such as BRM, GCN5, and PCAF (Green et al., 2007). A

complex, dynamic picture concerning ERa mediated gene regulation is emerging. It

supports step-wise recruitment of transcription factors to cis-elements, and ordered









assembly of the transcription initiation complex at those regions (Metivier et al., 2006;

Shang et al., 2000).

Non-Genomic Function

ERa can also influence cell proliferation through cytoplasmic signaling. These

pathways are associated with estrogenic functions that act as part of an extranuclear

kinase cascade through membrane associated ERa. These signaling cascades include

G-protein coupled receptors (Pedram et al., 2006, 2007), cell membrane ion channel,

tyrosine kinase c-Src, ERK1/2, p38, JNK, PKA, PKC, P13K, and Notch pathway (Rizzo

et al., 2008; Fu and Simoncini, 2008). The presence of extranuclear ERa signaling is

controversial (Losel et al., 2003). This is mainly due to numerous crosstalk between

ERa and kinase cascades as discussed earlier for ligand-independent genomic effects.

These networks make it difficult to experimentally test and confirm estrogenic action in

the non-genomic context. Other issues under discussion include whether membrane

associated ERa is the classic receptor tethered to the membrane by adaptor protein or

is a variant protein. An amino-terminus truncated isoform of the receptor (ER46) has

been implicated as such a variant in human endothelial cells (Li et al., 2003). A variant

form of metastatic tumor antigen 1 (MTA1) was shown to sequester classical ERa in the

cytoplasm and enhanced non-genomic function (Kumar et al., 2002). Present

understanding of non-genomic function of ERa is incomplete, but it proposes an

integrated approach to understanding the full range of ERa molecular mechanisms.









Cell of Origin


Clonal Evolution Model

The cell-of-origin question has been much debated in cancer biology. Can a

complex, heterogeneous tumor originate from a single cell? Or are there many points of

origin to account for the various cell types found? The clonal evolution hypothesis

supports both single and multicellular origin of cancer cells (Shipitsin and Polyak, 2008;

Campbell and Polyak, 2007; Shackleton et al., 2009). Clonal evolution is natural

selection applied to tumorigenesis, and loosely fits the "seed and soil" hypothesis. The

basic framework of clonal evolution is a hierarchical model where a single cell can give

rise to the whole tumor and continuously transform as it adapts to its environment. At

the top of the pyramid is a single transformed cell. It may acquire additional

reproductive advantage with each cell division through inherent genetic instability

characteristic of tumor cells. These abnormalities may be acquired through mutation, or

general protein variations due to stochastic gene expression (Raj and van

Oudenaarden, 2008) that optimize survival in its particular environment. Thus, under

favorable conditions, a single transformed cell may give rise to a tumor by growing

quickly and aggressively. There may still be multiple subclones generated throughout

this process, but they would not be the dominant cell type. If the environment is less

favorable, the condition may select for multiple subclones that compete for dominance

and contribute to heterogeneity.

The Cancer Stem Cell (CSC) Hypothesis

The cancer stem cell (CSC)/ tumor initiating cell (TIC) hypothesis represents a

paradigm shift in considering the nature of the "seed" (Lobo et al., 2007). The

fundamental change is instead of a cell that outgrows and out-competes other cells as









proposed by clonal evolution; it suggests a small reservoir of dedicated founder cells

reside within tumors. These founder cells are believed to be capable of self-renewal

through asymmetric cell division that is analogous to somatic stem cells by sharing

certain genetic features (Visvader and Lindeman, 2008). In essence, the CSC/TIC

hypothesis suggests a predictable, orderly cause for tumor initiation instead of a more

random act through natural selection or genetic drift (made more favorable by the

unstable genome of cancer cells). It also dictates that tumor initiation originates from

the top of a hierarchical system, whereas in clonal evolution a tumor can arise to

multiple points of origin (Visvader et al., 2008).

Tumor Heterogeneity and CSC Hypothesis

The cancer stem cell (CSC)/tumor-initiating cells (TIC) hypothesis gained

prominence when they were first isolated in hematopoietic malignancies (Bonnet and

Dick, 1997). The hypothesis supposes there is a rare subpopulation of cells (< 2%) with

stem-like features within tumors. The characteristics include self-renewal, the ability to

differentiate, and resistance to therapy. These features can confer tumorigenic

properties to TICs and may explain recurrence after therapy.

However, CSCs themselves can evolve significantly from their original pool during

disease progression (Rosen and Jordan, 2009; Hwang-Verslues et al., 2009). It is

equally possible that a tumor may contain both stem-like cells and genomically unstable

clones, both vying for dominance (Campbell et al., 2007). To complicate the issue,

there may yet be a graduated pool of tumorigenic lineage-restricted progenitor cells

(Lim et al., 2009). These considerations nullify much of CSC hypothesis' appeal since it

resembles the clonal evolution model. The CSC model thus represents tumorigenic









potential in a given group of cancer cells, not their fate as each cell would still undergo

significant adaptive changes (Shackleton et al., 2009).

To reconcile the seemingly contradictory viewpoints is to admit the two models are

not mutually exclusive. A tumor is a distinct and complex entity that remains dependent

upon its host. It may have originated from a founder cell, but it needs to adapt to its

host environment in order to propagate itself. The xenograph model typically used to

evaluate tumorigenicity or CSCs may not provide the optimal microenvironment for the

pool of varying CSCs, but simply select for a small fraction. If true, this indicates the

potential pool and frequency of tumor-initiating cells can be larger than present

estimates (Kelly et al., 2007; Kern and Shibata, 2007; Alison et al., 2009). It is

reasonable to suggest tumor-initiating cells can undergo clonal evolution and evolve into

a more efficient "seed", which may then develop into a complex solid tumor. Therefore,

it is counterproductive to be limited by semantics when a broad range, integrated

approach is required to target a tumorigenic pool rather than single cells. In addition, it

demonstrates the necessity for a more precise vocabulary to be introduced.

Implications of Each Model on Breast Cancer Recurrence

The CSC hypothesis supports targeting therapy towards the most tumorigenic

cells. The reasoning is to provide more effective treatment and inhibit tumor recurrence,

which tends to produce cancer cells with an aggressive phenotype. This idea assumes

that tumorigenic cells are at the top of a hierarchy that gives rise to non-tumorigenic

cells at the lower tiers. Therefore eliminating the highly tumorigenic subpopulation

would render the remaining cells more susceptible to standard antihormonal or

chemotherapy treatments.









Under the clonal evolution model, the large pool of heterogeneous tumor cells

should be the target. This is also applicable to the potentially mixed CSC population. It

poses a bigger challenge as it will be necessary to identify common elements in the

tumorigenic pool for effective therapeutic designs. Pinning down a malignant signature

in these cells (regardless of origin) can be a daunting task. However, there is progress

in that direction as tumorigenic cells share many molecular cascades with stem cells

and vice versa. A number of oncogenic and self-renewal signaling pathways associated

with breast CSCs have been identified. A few examples include the Wnt pathway

(Chen et al., 2007; Rappa and Lorico), Hedgehog (Liu et al., 2006), Notch (Farnie et al.,

2007; Dontu, Jackson, et al., 2004), and receptor tyrosine kinases (Vera-Ramirez et al.,

2010; Farnie et al., 2007; Rappa et al.), NFKB (Zhou et al., 2008), PTEN/mTOR (Zhou

et al., 2007), and CDKI (Pei et al., 2009; Liu et al., 2009). Ongoing research strives to

understand pathways in stem cell maintenance and their mis-regulation, which may

contribute insight into oncogenic pathways. These studies have yielded therapeutic

developments that target breast CSC as well as other types of cancers. These include

molecular signatures (Liu et al., 2007), immunotherapy (Morrison et al., 2008),

nutritional phytochemicals (Kakarala et al., 2009), and high-throughput drug screening

(Gupta et al., 2009).

Estrogen receptor alpha, stem cells and breast cancer

The heterogeneous nature of CSCs/TICs themselves suggest varied cell-of-origin.

They may be derived from multipotent cells, from the range of lineage-restricted

progenitor cells that arise over the course of differentiation, or from transformed cells

that acquired stem-like features. Within the framework of ERa-positive breast cancer









recurrence, it is possible for ERa-positive cancer cells to acquire stem-like properties

which allow for perpetual self-renewal, evasion from antiestrogen inhibition and

recurrence of hormone-sensitive tumors. This may account for the approximately 30%

recurrence hazard after 5 years of initial antiestrogen treatment (Dignam et al., 2009;

EBCTCG, 2005). ERa-expressing cells with highly tumorigenic features may shed light

on this observation. If such a population exist, then it is worthwhile to characterize its

response to antiestrogen in order to evaluate antiestrogen efficacy in this context.

Developmental studies have yielded some clues. Breast epithelial cells undergo a

differentiation hierarchy similar to that of hematopoietic cells (Figure 1-2). The source

starts with the undifferentiated mammary stem cells (MaSCs). They are maintained in

the mammary tissue via self-renewal and differentiate to bi-potent progenitors and

onward to the two major cell subtypes: myoepithelial and luminal progenitors. These

short-termed stem cells are maintained throughout life and contribute to hormone-

dependent remodeling of the breast during puberty and pregnancy (Vargo-Gogola and

Rosen, 2007). However, it is unknown whether hormone stimulation of progenitor cells

is ERa dependent as evidence are contradictory. Mouse mammary stem cells have

been demonstrated as ERa negative, but are hormone responsive (Asselin-Labat et al.,

2006, 2010; Booth and Smith, 2006). There is contrary evidence as well, with in vivo

evidence of a small number of slow cycling cell that are ERa positive (Clarke et al.,

2005; Smith, 2005). The ongoing debate continues as one side argues that ERa

positive cells found during fetal development only serve to regulate ERa negative stem

cell maintenance via paracrine mechanisms (LaMarca and Rosen, 2008), while the

opposition maintains ERa positive progenitor cells not only proliferate and differentiate









in response to estrogen but also produce paracrine factors that influence nearby ERa

negative cells (Clarke et al., 2005; Dontu, EI-Ashry, et al., 2004; Booth et al., 2006;

Wicha, 2008). The population of ERa positive progenitor cells is believed to be present

only during a small developmental window, and these cells are necessary for early

mammary gland development.

The idea that these committed progenitors give rise to TICs is gaining ground.

Accumulating evidence from genetic profiles of mammary stem cells have been

compared to human tumor datasets (Lim et al., 2009; Raouf et al., 2008). These

studies demonstrated that aberrant luminal progenitors with limited estrogen receptor

expression are found to correlate with basal-like tumors (Lim et al., 2009). The

implications are that the various breast cancer subtypes may have corresponding

partners in the stem cell hierarchy (Figure 1-3).

As additional information concerning progenitor transformation emerges, it will

generate a richer picture for understanding tumor initiation and a wider array of tools for

their precise identification. Further support is emerging as gene expression profiling

using patient data confirms that traditional histological classification is a crude

prognostic indicator, as even ER-positive tumors can have varying degrees of

malignancy which molecular profiling can discern (Sotiriou et al., 2003, 2006).

Molecular profiling information would be useful in diagnosing and tailoring the most

effective treatment plan for each patient.

Contribution of CSC/TIC to antiestrogen resistance

Antiestrogens have been the treatment of choice for ERa positive breast

carcinomas since the 1970s. These tumors tend to be estrogen dependent and









blocking estrogen signaling with antiestrogens has proven effective. Patients with

antiestrogen responsive tumors generally have the best prognosis and survive the

longest after initial diagnosis. However, resistance to antiestrogen therapy has

substantially hindered long-term use of antiestrogen compounds. It is interesting to

consider if a potential ERa positive stem-like cancer cell may be antiestrogen sensitive

as well. One of the key criteria of a CSC/TIC is resistance to chemotherapy and

radiation (Phillips et al., 2006; Grimshaw et al., 2008; Fillmore and Kuperwasser, 2007;

Li et al., 2008). So, does ERa signaling supersede self-renewal pathways? If it does

not, it may explain why prolonged antiestrogen treatment can generate non-responsive

ERa positive cells. It would confirm the presence of an intrinsically different cellular

subgroup that is responsible for tumorigenesis, and this subgroup may utilize alternative

signaling pathways. However, if ERa signaling supersedes self-renewal signaling, then

one may suppose antiestrogen therapy can be effective against less differentiated ERa

positive cells as well as against fully differentiated luminal cells. The culmination of

specific cofactor expression and activity patterns, ligand availability, receptor

conformation, and growth factor signaling within each particular cell type determine their

response to antiestrogen inhibition. It is conceivable that the stem-like cells at various

differentiation stages would have varying degrees of antiestrogen sensitivity. Further

characterization of antiestrogens in stem-like cells may provide a new perception into

the development of antiestrogen resistance.

To focus on the role of ERa in this study, only antiestrogens relating directly with

ERa will be further discussed. Other forms of antiestrogen therapy not related to the

receptor, such as aromatase inhibitors, will be omitted. There are two major classes of









antiestrogens that bind directly to ERa. The first class is selective estrogen receptor

modulators (SERMs). They are classic antagonists specific for the estrogen receptor,

although they may have agonist effects in a cell specific manner. The agonist activity

induced by SERMs has been attributed to ERa conformation changes upon SERM

binding that may recruit coactivators in certain cell types (Jordan, 2007). SERMs have

a triphenylethylene chemical structure and a higher binding affinity to ERa compared to

173-estrodiol (Clarke et al., 2003). Tamoxifen (TAM) was the first FDA approved

compound for use against hormone-dependent breast cancer in pre- and

postmenopausal women. It is a triphenylethylene that is metabolized in the liver by

cytochrome P450 enzymes to the metabolically active form 4-hydroxytamoxifen (4-

OHT) (Jordan, 2007). 4-OHT has a stronger binding affinity for ERa's LBD than TAM.

Many other SERMs have been developed over the years, the most promising of which

is raloxifene, a benzothiophene derivative, as it has no known agonist activity (Lewis

and Jordan, 2005). However, raloxifen and other benzothiophene derivatives have

lower binding affinity to ERa and have a short half-life (Jordan, 2007).

The second class of compounds is selective estrogen receptor down-regulators

(SERDs). They are often referred to as pure antiestrogens because they have no

known agonist effects. SERDs function under a completely different mechanism from

SERMs. They impair receptor dimerization and induce conformational changes that

expose ERa's hydrophobic surface. The unnatural conformation destabilizes the

receptor and promotes degradation via the ubiquitination pathway (Ariazi et al., 2006).

Fulvestrant (ICI 182780) is the most widely studied drug of the SERDs. Clinical trials

comparing fulvestrant to TAM have concluded that there were no significant differences









in efficacy between the two drugs. Fulvestrant has become another option for patients

that developed resistance to other lines of therapy; however clinical data has not shown

it to be better than other therapies (Ariazi et al., 2006).

CSC Isolation and Detection

The premise of the CSC/TIC model has been keenly applied towards the study of

breast cancer recurrence. The initial toolbox consisted of BrdU label retentions and

Hoechst dye efflux to isolate a slow-cycling "side-population". This method was used in

mouse mammary stem cell studies and it was noted that less differentiated cells

overexpress multi-drug resistant transporters (Welm et al., 2002). The tools have since

expanded as additional molecular markers were identified. A tumorigenic subpopulation

was identified and isolated as CD44+CD24-Lin- (AI-Hajj et al., 2003) in human breast

carcinomas. CD44 and CD24 are not the only stem-like markers proposed; they are just

the most commonly used in breast cancer studies. These markers have formed the

basis for isolation of mammary stem cells in mice (Stingl et al., 2006; Regan and

Smalley, 2007), rat (Zucchi et al., 2007), and humans (Liu et al., 2006). These markers

were adopted and combined with neural stem cell culture technique to generate an in-

vitro culture system that enriches for stem/progenitor cells from human mammary tissue

(Liu et al., 2005), referred to as mammosphere culture. It basically is growing cells at

low density in serum-free, defined media under suspension condition to generate clonal

spheres. The mammosphere culture method has also consistently propagated

tumorigenic cultures from established cell lines (Ponti et al., 2005; Fillmore and

Kuperwasser, 2008). A recent study provided preliminary evidence correlating

differential cellular adhesion to stem-like/mesenchymal properties (Walia and Elble,

2010). Although the underlying mechanism is not understood, it indicates that cell









adhesion and culture environment can drive cells towards a stem-like/mesenchymal

phenotype. This may provide an explanation for the effectiveness of TIC enrichment in

mammosphere culture.

The debate concerning the genuine utility of the CD44 marker is ongoing. Several

studies have noted that CD44 expression is not unique to stem cell (Shipitsin et al.,

2007; Hwang-Verslues et al., 2009; Wright, Robles, et al., 2008) and that there is a

significant level of heterogeneity within CD44-positive cell fractions (Shipitsin et al.,

2007; Pece et al., 2010; Park et al., 2010). However, CD44 expression has been

consistently linked to a more progenitor-like tumorigenic subpopulation (Shipitsin et al.,

2007; Sheridan et al., 2006; Abraham et al., 2005), and has potential prognostic value

(Liu et al., 2007). The relative enrichment is dependent on cellular context. Cells

isolated in this manner have expressed some stemness properties, but their relative

abundance in a given population and their function require further study (Zucchi et al.,

2008; Wright, Calcagno, et al., 2008).

A new addition to the arsenal is the identification of aldehyde dehydrogenase 1

(ALDH1) that is overexpressed in tumors, particularity in the highly tumorigenic and

undifferentiated fraction (Ginestier et al., 2007). ALDH1 has been further evaluated in

established cell lines and initial results were confirmed (Charafe-Jauffret et al., 2009).

This marker has also been applied to other cell types as CD44 has been. Its validity as

a CSC marker has likewise been contested because ALDH1 expression varies widely

depending on cell type and microenvironment (Neumeister and Rimm, 2009).

Other Examples of CSC Enrichment

Results gathered using the existing markers are imperfect, and will surely be re-

evaluated when new methods become available. In addition to markers mentioned









above, CD133 (Wright, Calcagno, et al., 2008) has also been used to identify

stem/progenitor fractions from epithelial cells in other tissues including colon

(Vermeulen et al., 2008), bladder (Chan et al., 2009), and prostate (Shi et al., 2007;

Wang et al., 2009). Similar issues concerning cell-of-origin and heterogeneity is also

being discussed in these models.

The overall picture emerging from CSC/TIC identification is the obvious paucity of

available biomarkers and the limited application of each. It has not constrained

research in this area because, regardless of the real existence of cancer stem cells, it is

useful to isolate and understand the more persistent cell types in solid tumors. The

main obstacle has been the difficulty in isolating those fractions. The tools currently

being developed will contribute to understanding the rarer subgroups and help develop

long-term strategies for disease management.











A/B core D F
region DBD region LBD region
184 250 311 547 595
ER(a 100% 100% 10 0% /100%


lAB 214 26d


ER3


4n8 t9I


27/ 4 97 26% 8 22%


Figure 1-1. Estrogen receptor isoforms (Ariazi et al., 2006). Amino acid sequence
identity is shown as percentage homology relative to ERa. Abbreviations:
DBD is DNA binding domain. LBD is ligand binding domain.






ER l ER


C R C V'a CW- CT
C(?~cV nW~4c


AhmjtI"
hunmal
pfmhnito


-I*K p 'u cKTa

. cODA fpCAm' MUO
ERcrw Er


Vr r B- Ct O



CD49' Et -A C24 C0133
ER PM EABB


.I ER 0 ER
I


common
pmwi~r


- Aserow


"CD49 a CC4'Coar c


aflrolar c.U


CQ"C3C9f M24~CE~t~l
WIap~

coa~r &wCA


Figure 1-2. Mammary gland differentiation hierarchy (Visvader, 2009). Model of
mammary differentiation hierarchy with mouse primary cell surface markers in
blue, and human primary cell surface markers in red. The common
progenitor cell is believed to be a potential bipotent cell restricted to
mammary development.


r"











ubty m.


"Nfrma14ko"
Sl


B,-I
(lvirn ta' vvyri'o'




J


Luminal B


cxnmwM
pMOginWOr


UwrwA



t
fe
I duc@fofe


Iaofar cT


MY06pithowmycpdhis
myoopIxri.af
-o n


Figure 1-3. Potential relationship between mammary differentiation hierarchy and the
development of tumor subtypes (Visvader, 2009). A hypothetical model for
the development of various breast cancer subtypes from a hypothetical
mammary differentiation hierarchy. It supposes less differentiated tumor
subtypes are derived from oncogenic transformation of lineage-restricted
progenitor cells, while highly differentiated subtypes are derived from fully
differentiated luminal cell types.









CHAPTER 2
GENERAL MATERIALS AND METHODS

Cell Culture

MCF7 cell from ATCC were grown as monolayer culture in DMEM media

(MediaTech) supplemented with 10% bovine calf serum (Hyclone), 1% L-glutamine

(MediaTech), and 1% penicillin-streptomycin solution (MediaTech). They were cultured

in tissue-culture grade petri dishes. To generate MCF7 mammospheres (MCF7S),

MCF7 parental cells were first washed with PBS. They were trypsinized with 0.05%

trypsin/EDTA (MediaTech) for 5 minutes. The trypsin was quenched with whole DMEM

media, stained with 0.4% solution trypan blue (Gibco) for 5 minutes to exclude dead

cells, and manually counted using a hemocytometer to calculate cell density according

to the formula: cell number count (average) x dilution factor x 104 = cells/ml. The cells

were diluted to 5000 cells per ml in PBS and washed twice. The cell pellet was

resuspended in defined mammosphere media and seeded onto Poly-HEMA coated

dishes to prevent attachment. Poly-HEMA working solution was a 2X solution diluted

from 10X stock solution in 95% ethanol. The 10X Poly-HEMA stock solution was

composed of 0.12 g/ml Poly-HEMA dissolved in 95% ethanol at 60C.

MCF7 mammospheres (MCF7S), derived from mammosphere culture selected

MCF7 cells, were maintained in 50:50 DMEM/F12 media (MediaTech) supplemented

with 20 ng/ml EGF (Sigma), 10 ng/ml bFGF (Sigma), 5ug heparin (Sigma), 1%

penicillin-streptomycin solution (MediaTech), 1% L-glutamine (MediaTech), and 1X B-27

supplement (Invitrogen). Cells were trypsinized every 6-7 days and passage at 10,000

cells per ml onto Poly-HEMA coated dishes. For frozen cell stocks, MCF7S were









dissociated with trypsin and resuspend in desired amount of mammosphere culture

media containing 10% DMSO, and stored at -80"C.

MCF7S Cell Proliferation and Sphere Formation Assay

Serial Passaging

For each passage, the MCF7S cells were spun down from suspension at 700 rpm

for 6 minutes and the supernatant was removed. The cell pellet was resuspended in 50

pil 0.05% trypsin/EDTA for 2 minutes, and then actively dissociated by pipetting up and

down. The digestion was quenched by adding 150 iil of mammosphere media. 20 iil of

cell suspension was taken for trypan blue viability assay and counted. The cell

suspension was diluted to 10,000 cells per ml and plated with desired antiestrogen or

vehicle on Poly-HEMA coated tissue culture dishes. The cells were incubated

undisturbed for 7 days before repeating dissociation and cell counting.

The fold change was calculated as final cell density / initial cell density. The fold

change for each passage is used to calculate cell expansion. To calculate cell

expansion, the fold change for each passage was multiplied by the cell density in each

preceding passage. The final results were expressed on a semilog graph and the slope

was calculated as log expansion to determine growth rate changes.

Cell Proliferation Assay

MCF7S cells were spun down from suspension at 700 rpm for 6 minutes and the

supernatant was removed. The cell pellet was resuspended in 50 iil 0.05%

trypsin/EDTA for 2 minutes, and then actively dissociated by pipetting up and down.

The digestion was quenched by adding 150 iil of mammosphere media. 20 iil of cell

suspension was taken for trypan blue viability assay and counted. Cell suspension was









diluted to 5,000 cells per ml and was plated with desired drug or vehicle on Poly-HEMA

coated tissue culture dishes. The cells were incubated undisturbed for 2, 4, or 6 days

before harvesting for viable cell count and replating for sphere formation assay (Figure

2-1).

Sphere Formation Assay

Directly after viable cell count for the cell proliferation assay described above, the

cell suspension was diluted in 2 ml of mammosphere media to a cell density of 5000

cells per ml. For each treatment condition, 200 iil of the cell suspension was allocated

per well for one column (8 wells) of a 96-well plate with no Poly-HEMA coating. There

was an estimated 1000 cells per well at this point. 100 iil of the cell suspension in each

well was diluted to 2 adjacent wells, which contained 100 iil of mammosphere media.

The result was a final 1:1 dilution in 16 wells; each well contained an estimated 500

cells. The estimated total number of cells seeded was 8000 cells per condition (Figure

2-1). Plated cells were incubated undisturbed for 6 days. Phase contrast pictures were

then taken at 50x magnification and sphere size and number were manually evaluated

from the images using ImageJ software. Sphere formation frequency was calculated

using the formula: (number of spheres/number of total cells plated) x 100%.

RNA Isolation

Samples were collected and homogenized by vortex in 0.5 ml Trizol reagent

(Invitrogen) to obtain total RNA. 0.1 ml of chloroform was added to each homogenized

samples. The samples were then centrifuged at 12,000 x g for 15 minutes at 4C to

separate aqueous: phenol-chloroform phase. The aqueous phase was extracted from

each sample. The RNA was precipitated with the addition of 75% isopropanol (v/v), and









centrifuged maximum speed at 4C for 15 minutes. The RNA pellets were washed with

1 ml of 75% ethanol and centrifuged to remove supernatant. The pellets were air-dried

for 5 minutes before dissolving in sterile-filtered TE (10mM Tris pH 8, 1mM EDTA) and

stored at -80"C.

Reverse Transcriptase PCR (RT-PCR)

Approximately 1 pg of total RNA was added to a reaction cocktail containing

DEPC-treated ddH20, 2.5 piM dNTP, and 5 nM random hexamer primers to 16 iil final

volume. The mixture was incubated at 70C for 5 minutes, and quenched quickly on

ice. 2 pil of 10X RT buffer (NEB M-MuLV), 1 pil RNase Inhibitor (Promega), and 1 pil M-

MuLV Reverse Transcriptase (NEB) were added to the reaction to a final volume of 20

pil, which was incubated at 42C for 1 hour. The reaction was heat-inactivated at 95C

for 5 minutes, and was diluted 1:5 dilution with ddH20. 1-2 iil of diluted template was

used for real-time PCR.

Real-Time PCR

Each real-time PCR reaction was composed of the following: 1-2 iil cDNA

generated from RT-PCR, 1 iil of 5 piM primer mix working solution, 8 iil ddH20, and 10

pil 2X Sybr Green PCR Master Mix (Applied Biosystems). Triplicates were done for

each reaction and results were expressed as relative quantitation normalized to beta-

actin expression as control. Duplicates of reactions with no template were also included

on real-time PCR plate for each run as negative control. Gene expression differences

larger than 2-fold are considered to be significant. The thermal cycling parameters were

as follows: 95C for 10 minutes, 40 cycles of 95C for 15 sec for denaturing step and 60-

64C for 60 sec for primer extension, and a melting curve analysis was performed at the









end of each run. StepOne (48-well), or StepOnePlus (96-well) Real-time PCR

machines (Applied Biosystems) were used for data collection. Primers used were listed

in Table 2-3.

Total Protein Isolation

Total protein lysates were isolated from MCF7 or MCF7S cell by first washing cells

in cold PBS twice, then adding 50-200 iil lysis buffer (50mM Tris pH 7.5, 1mM EDTA,

1% (v/v) SDS, 1% 2-mercaptoethanol, 20mM dithiothreitol). The samples were boiled

for 10 minutes, then 6X sample buffer (4x Tris-SDS pH 6.8, 30% glycerol, 10 % SDS,

0.6M dithiothreitol, 0.012% bromophenol blue) was added. The samples were either

stored at -20C. The samples were boiled again for 5 minutes prior to loading onto

SDS-PAGE gel for analysis.

Nuclear Extraction

Cells were collected after antiestrogen treatment in cold PBS and rinsed twice.

The cells were resuspended in 400 iil hypotonic buffer (10 mM HEPES pH7.9, 10 mM

KCI, 2 mM MgCI2, 0.1 mM EDTA, 0.5 mM Dithiothreitol, 1X protease inhibitor cocktail),

and incubated 15 minutes on ice. 30 iil of 10% NP-40 was added to each sample and

then vortex 10 times for 3-5 seconds each. The samples were centrifuged for 1 minute

at 4C. The cytosolic supernatant was collected in pre-chilled microcentrifuge tubes for

each sample, and stored at -80"C for later use. Each nuclear pellet was resuspended in

40 iil nuclear extraction buffer (50 mM HEPES ph7.9, 50 mM KCL, 500 mM NaCI, 0.1

mM EDTA, 0.5 mM Dithiothreitol, 1X protease inhibitor cocktail). The samples were

incubated at 4C with rocking for 30 minutes. The nuclear fractions were collected in

pre-chilled microcentrifuge tubes after centrifuging at max speed for 1 minute at 4C.









The nuclear fraction was diluted 1:1 with hypotonic buffer before storage at -80C for

later use.

Immunoblot Analysis

Protein lysate and nuclear extracts were analyzed by immunoblot. The samples

were resolved by first adding the appropriate amount of 6X sample buffer and boiled for

5 minutes. The samples were loaded onto 10% Tris-HCI polyacrylamide separating

gels /4% stacking gel at 1mm thickness for electrophoration in 1X running buffer (25

mM Tris, 190mM glycine, 0.2% SDS). The gel was then electrotransferred onto

polyvinylidene fluoride (PVDF) membrane using Trans-Blot Semi-Dry Electrophoretic

Transfer Cell (BioRad) in 1X transfer buffer (20 mM Tris, 192mM glycine, 10%

methanol). Membranes were stained with Fast Green (0.1% Fast Green FCF, 50%

methanol, 10% acetic acid) for 5 minutes at room temperature to ensure transfer and

equal loading. Stained membranes were washed 2-3 times in TBST (30mM Tris pH

7.5, 200mM NaCI, 0.1% (v/v) Tween-20), and incubated in 3% (w/v) non-fat dry milk in

TBST blocking solution for 30 minutes at room temperature. Blocked membranes were

rinsed in TBST before probing with diluted primary antibody in TBST for 30 minutes at

room temperature, or overnight at 4C. The membranes were then washed 3 times at

room temperature in TBST for 5 minutes each. They were next incubated in 1:10000

diluted peroxidase-conjugated secondary antibodies in TBST for 30 minutes at room

temperature. The membranes were then washed 3 times at room temperature in TBST

for 5 minutes each. Bound antibodies were detected by applying Pierce ECL substrate

solution (Thermo Scientific) and exposing the membrane to X-ray film.









Immunofluorescence

Cells were attached to glass cover slips prior to immunofluorescence staining

procedure. Adherent cells were grown directly onto glass cover slips and treated with

antiestrogens. Mammosphere cells were dissociated with trypsin, quenched with

media, centrifuged to concentrate cells to 50 iil media. The cell suspension was then

gently applied onto glass cover slips in 6-well tissue culture plate, and centrifuged at

3000 rpm for 4 minutes to attach cells onto cover slips. The cover slips were rinsed with

PBS twice, and then 3.7% formaldehyde/PBS was added to cover slips to fix the cells

for 5 minutes at room temperature. The cells were rinsed with 0.1% NP-40/PBS 3

times. The cell membranes were permeablized for 15 minutes at room temp with 0.1%

NP-40/PBS. Next, they were incubated with 1:50 diluted primary antibodies in 3%

BSA/0.1 % NP-40/PBS for 1 hour at room temperature. They were then rinsed 3 times

with 0.1% NP-40/PBS for 5 minutes at room temperature. They were incubated with

1:500 diluted flurophore-conjugated secondary antibodies in 3% BSA/0.1% NP-40/PBS

for 1 hour at room temperature in the dark. The cover slips were rinsed 3 times with

0.1% NP-40/PBS for 5 minutes at room temperature, followed by a quick rinse with

water, and then counterstained with 200 pg/ml of Hoechst 33342 for 5 minutes at room

temperature. The cover slips were rinsed with water, and allowed to air dry in the dark

for 30 minutes. The cover slips were mounted onto glass slides using Fluoromount-G

(Southern Biotech) and allowed to dry overnight at room temperature in the dark. Glass

slides were stored at 4C.









short-hairpin RNA (shRNA) Vector Construction

shRNA Oligo Design and Cloning

Oligos for shRNA construction were designed using shRNA psm2 designer at

RNAi Central (http://katahdin.cshl.edu/siRNA/RNAi.cgi?type=shRNA). The accession

number NM_000125.2 was entered for estrogen receptor alpha shRNA design. In order

to generate a high-fidelity oligo, it was broken into two fragments when ordering from

Invitrogen (Table 2-1). The breaks in the two fragments were designed with

overlapping, complimentary loop regions so they anneal and extend during PCR into the

full length oligo. The desiccated oligos were dissolved in TE buffer and combined to a

final working solution of 1 pM. 5 pM Mir30 PCR primers were used for amplification

using Phusion High-Fidelity DNA Polymerase (Finnzymes). Mir30 primer sequences

were as follows: forward 5'-aagccctttgtacaccctaagcct-3' and reverse 5'-

actggtgaaactcacccagggatt-3'.

The PCR was performed by mixing 1 pl of 1 pM mixed oligos, 1 pl of 5 pM Mir30

PCR primers, 1 pl of Phusion polymerase, 2 pl of 10 mM dNTP mix, and distilled water

to a final volume of 20 pl. The reaction was first denatured at 98C for 30 seconds,

followed by 35 cycles of 1) denatuaration at 98C for 10 seconds, 2) annealing at 60C

for 30 seconds, and 3) extension at 72C for 30 seconds. The reaction included a final

round of extension at 72C for 5 minutes.

After PCR, the fragment was gel purified using QIAquick Gel Extraction Kit

(Qiagen). The purified fragment was digested with Xhol and EcoRI restriction enzymes

for at least 2 hours at 37C. The retrovirus vector pLMP was also digested in parallel.

The pLMP vector sequence and information can be found in Figure 2-2

(Openbiosystems). The enzymes were heat inactivated at 65C for 30 minutes. The









digested fragment and vector were mixed at a 7:1 ratio, respectively and ligated with 1

pil of T4 DNA ligase and 1X ligase buffer (NEB) in 10 iil final volume for 1 hour -

overnight. The ligated DNA was transformed into DH5a competent E. Coli cells by

heat-shock at 37C for 45 seconds. The cells were allowed to recover in LB, incubated

at 37C for 30 minutes with agitation. The cells were plated onto 50 pg per ml ampicillin

LB agar plates and incubated overnight at 37C for selection of positive clones. Clones

were manually picked and amplified in 5 ml LB culture with 50 pg per ml ampicillin

selection overnight at 37C with agitation.

Plasmid DNA MiniPrep

1 ml of the overnight culture was transferred to a microcentrifuge tube. The

bacterial cells were centrifuged at 8000 rpm for 1 minute. The cell pellets were

resuspended with 200 iil P1 Buffer (50 mM Tris-CI pH 8.0, 10 mM EDTA) each. The

cells were then lysed by adding 200 iil P2 Lysis Buffer (200mM NaOH, 1% SDS w/v) to

each tube and mixed by inversion. Next, 200 iil P3 Neutralization Buffer (3 M

potassium acetate) were added to each tube and mixed by inversion. The samples

were centrifuged at 14000 rpm for 5 minutes to pellet precipitated proteins. 0.5 ml of

the supernatant was transferred to a new tube and reserved. The protein pellet was

discarded. 1 ml of 100% ethanol was added to the reserved supernatant and mixed to

precipitate DNA. The precipitated DNA was pelleted by centrifugation at 14000 rpm for

10 minutes. The supernatant was discarded. The DNA pellets were washed once in 1

ml 70% ethanol and the supernatants were removed. The DNA pellets were air dried

for 5 to10 minutes. The DNA was dissolved in 30 to 50 iil TE (1 M Tris-CI, 0.5 M EDTA









pH 8) containing 20 rig/ml RNase A. Purified DNA was sequenced to confirm shRNA

oligo insertion prior to retrovirus production.

Retrovirus Production and Transduction of Target Cells

Transient Transfection

To generate retroviral particles, the shRNA vector (Figure 2-2) containing the

specific knockdown sequence was transiently transfected into transformed HEK293

cells called Phoenix. The Phoenix cells overexpress retroviral proteins Gag-Pol and

Env to facilitate packaging of shRNA into retroviral particles and increase their

production. Phoenix cells were maintained in DMEM media (MediaTech) supplemented

with 10% bovine calf serum (Hyclone), 1% L-glutamine (MediaTech), and 1% penicillin-

streptomycin (MediaTech). Cells were seeded at ~60% confluency 24 hours prior to

transfection. They were transfected using Fugene 6 Transfection Reagent (Roche) by

first diluting 2 iil of the transfection reagent in 100 iil PBS. 1 ig of DNA was also diluted

in 100 il PBS. The two diluted reagents were combined and incubated at room

temperature for 30 minutes to form a cationic lipid-mediated transfection complex. The

complex was added directly to cells dropwise. The cells were then incubated from 12

hours to overnight before switching to fresh media. The cells were incubated for an

additional 48 hours. The virus-containing media was collected and passed through a 45

Iim filter to exclude cell debris. The viral media was aliquoted and used immediately to

infect target cells. Excess aliquots were stored at -80C, or disposed after bleaching.

Retroviral Infection for Stable Integration

Target cells were trypsinized and plated 24 hours prior to retroviral infection.

Adherent cells were infected by replacing culture media with the infection cocktail, which









consisted of 1:1 viral media: culture media and 4 pig/ml polybrene. The cells were

incubated for 24 hours, and then the infection cocktail was replaced with fresh media.

The cells were incubated for an additional 24 hours.

In order to infect suspension cells, they were first centrifuged to remove culture

media. The cell pellet was resuspended in the infection cocktail and plated on Poly-

HEMA coated tissue culture dishes. The cells were incubated for 24 hours, and then

the infection cocktail was replaced with fresh media. The cells were incubated for an

additional 24 hours.

After the 24 hours of recovery in fresh media, the cells were treated with 2 pg/ml

puromycin dihydrochloride (Cellgro) to begin selection of transformed cell. Uninfected

target cells were treated in parallel to estimate selection completion, which is typically

complete 48 hours after addition of puromycin. The selection media was replaced with

fresh culture media after selection and transformed cells were allowed to expand to

desired density. The cells were then trypsinized and dissociated into a uniform

suspension and aliquoted. The transformed cell stocks were stored in freezing media

(bovine serum albumin containing 10% v/v DMSO) at -80C, maintained as polyclonal

culture, or further selected for monoclonal culture.

Selection and Enrichment of Single Clones

To select single clones in mammosphere culture, the spheres were dissociated

with trypsin to a single cell suspension. The cell suspension was first diluted to 1000

cells per ml, and then distributed 200 iil per well into 2 columns of a 96-well plate. The

2 columns were serially diluted 100 iil down the remaining columns of the plate, which

also contained 100 iil per well. The final dilution was < 5 cells per well in 200 iil. The









cells were incubated and media was added or replaced as needed until single colonies

can be visualized. Each clone was manually picked using micropipet tips into each well

of a 24-well plate (Poly-HEMA coated). The clones were allowed to expand until

sufficient cells were obtained to ascertain knockdown using immunoblot analysis.

Analysis of CD44 Expression

For each sample, approximated 1x106 cells were trypsinized to obtain single cell

suspension. The cells were washed twice in cold PBS + 1% BSA. The cells were

incubated at 4C with either 1:25 diluted CD44-FITC antibody or isotype-FITC control

antibody in PBS + 1% BSA for 45 minutes in the dark. After incubation, the cells were

washed twice in cold PBS + 1% BSA. The cell pellet was resuspended in 400 iil cold

PBS + 1% BSA for flow cytometry analysis within 1 hour.

BrdU/Propidium Iodide Cell Proliferation Assay

An asynchronous MCF7S culture of 1 x 106 cells were trypsinized and cells were

pulsed with 10 piM BrdU for 2.5 hours. Duplicates of samples without BrdU pulse were

included as BrdU staining controls. They were washed in cold PBS, and fixed overnight

with cold 70% ethanol at 4C in the dark. The samples were stored at -20C, protected

from light, until BrdU staining. On the day of staining, the cells were centrifuged and the

ethanol was removed. They were then washed once in PBS + 1% BSA. To denature

the DNA and expose BrdU antigen, the cells were treated for 30 minutes in 2 M

HCI/0.5% Triton-X 100 at room temperature. The cells were centrifuged and the

supernatant was removed, and the cells were washed once in 0.1 M Na2B407 pH 8.5.

The supernatant was removed after centrifugation, and the cells were incubated with

1:50 dilution anti-BrdU-FITC antibody in 0.5% (v/v) Tween-20/1% BSA/PBS containing









100 pig/ml (w/v) RNase A. The samples were incubated in the dark at room

temperature for 1 hour. The samples were washed twice in PBS + 1% BSA and

resuspended in PBS + 10 pig/ml propidium iodide (Sigma). The samples were

incubated at room temperature for 15 minutes in the dark, and kept on ice for less than

1 hour prior to flow cytometry. PI cell cycle modeling and data generation was

performed using ModFit LT software.

Annexin V-PE Apoptosis Detection using BD Pharmingen Kit

Cells were washed 2 times in cold PBS, and then resuspended in 1X hypotonic

binding buffer (BD Pharmingen) at 1x106 cells per ml. 100 pl (1x105 cells) were

transferred to a culture tube. 5 pl Annexin V-PE antibodies and 5 pl 7-amino-

actinomycin D (7-AAD) were added to each tube. Then the tubes were incubated at

room temperature for 15 minutes in the dark. Next, 400 pl 1X hypotonic binding buffer

was added to each tube. The samples were processed in a flow cytometer within 1

hour. Cells treated with 7-AAD alone, Annexin V alone, and unstained cells were used

as controls to set up compensation and quadrants.

Statistical Analysis

Statistical analyses were derived from at least three independent experiments.

Error bars for three independent experiments were presented as standard error of the

mean (SEM), and statistically significant differences were determined using the

Student's t-test.

In Vivo Tumorigenic Assay

The following in vivo mouse injection protocol was adopted from Brian Morrison,

advised by Dr. Alejandro Lopez, QIMR, Griffith University, AU.









Preparation of Mice

A colony of NOD. Cg-RagltmlMO/ml2rgtmwlI/lSzJ mice (The Jackson Laboratory, Bar

Harbor, Maine, USA) stock number 007799, were maintained and 5 to 8 weeks old

littermates were used for the experiment, when possible. 173-estradiol slow releasing

pellets were implanted into the mice one day before injection.

Preparation of Tumor Cells

First, cell stocks were thawed in 37C water bath and cells were washed once in

RPMI. Then, the media was removed and the cells were washed with PBS. Next, the

cells were resuspended in PBS and a viability assay was performed using trypan blue

and Countess cell counter. For each injection, the cells were mixed in PBS 1:1 with

Matrigel (BD Biosciences) to a final cell concentration of 2 x 106 in 100 pil.

Mouse Injections

The inoculation area of the mice was cleaned and sterilized with ethanol. The

cells were mixed and drawn into a cold 1cc syringe without a needle to prevent cell

damage. A 26.5 gauge needle was used to inject 2 x 106 cells subcutaneously (s.c.)

into the lower right flank of the mice. Mice were monitored twice a week for overall

health. Tumor diameters were measured with a digital caliper 4 weeks after injection

and once a week thereafter. The tumor volume in mm3 was calculated using the

formula: Volume = width2 x length x 0.5.

















Growth -tlet i -.ti.:..eis used:
Curve 1) SERM
2) Pule a. lies lio.en
IC/ 1827

/ ,4 \




Replate-500
cells/well (final)
Sphere In 96-well plate
Formation
Examine frequency
of sphere formation
in surviving cells


Figure 2-1. Tissue culture scheme for MCF7S proliferation assay and sphere formation
assay. The growth curve was performed by first seeding viable MCF7S cell
suspension at clonal density and cultured in the presence of 4-
hydroxytamoxifen or ICI 182780, then incubate for 2, 4, or 6 days. Viable
cells were counted at each time point, and related in fresh media at a
defined cell density (-500 cell per well) in a 96-well plate. The cells were
incubated for 6 days and sphere formation frequency was assessed.











SgrA 7716 Cr'CCGG jG
NdeI 7344 CA'TA TG








MS
0 7








Acl 4191 GT'mkAC
Sall 4190 G'TCGA C
Pad 4163- TTAAT'TAA


Bgill 1410 A'GATC T
8t.4 / HpaI 1520 GTT'AAC
5' miR30 1528
PspXI 1535 vC'TCCA Gb
SnHoT -1535 C'TCGA G
EcoRI 547 G'AATT C
3'minR30 -1553
V-LMP .4el 1786 A'CCGG T
4 b -- I Blpl-1949-GC'TLAGC
J BsI 2252- C'GTAC_G
SRsI 2312 CG'GwCCG
----- SacI 2410 CC_GC'GG
SPspOMI 2962 G'GGCC C
01, -Apal 2966 GGGCC'C
SrwrlI -3000 C'CTAG_G
GFP BsaA 3165 AC"GTr
--- Pml 3165 CAC'GTG
A arl 3188 CACCTGCnnnn'nnnn
Pf\ IMI 3302 CCAn_ nn'nTGG
\ Nol 3433 CCATG G
BsfXI 3434 CCAn nnnI'nTGG
BtgZI 3556 GCGATGDnnnnnDann 'nnn
Fall 3944 AAGnnannCTTnnnnnnon annn'


B IMal


I O-i -iR, UP


Xho 1 EcoR1


Figure 2-2. Cloning vector information for microRNA-adapted retroviral vector. A)
Vector map and unique restriction sites of MSCV-LMP cloning vector. B)
Xhol-EcoR1 cloning site for shRNAmir expression using retroviral 5' LTR and
PSI (LP) promoter. PGK promoter (Ppgk) drives expression of selection
cassettes. Puror cassette allows for selection of stable integrates. IRES-GFP
served as marker for stable integration. Abbreviations: LTR is long terminal
repeats. MiR is microRNA. Ppgk is phosphoglycerate kinase promoter. Puro
is puromycin. IRES is internal ribosome entry site. GFP is green
fluorescence protein.


,5'mi3Fl0


T'30'i










Table 2-1. shRNA Oligos designed for retrovirus-mediated knockdown.
Oligo Gene Sequence Start
Name Targeted Position
shER-1F ERa TGCTGTTGACAGTGAGCGAAGGGAGAATGTTGAAACACAATA 1143
GTGAAGCCACAGATGTA

shER-1R ERa TCCGAGGCAGTAGGCAGAGGGAGAATGTTGAAACACAATACA 1143
TCTGTGGCTTCACTA

shER-3F ERa TGCTGTTGACAGTGAGCGCGGAGTTTGTGTGCCTCAAATCTA 1689
GTGAAGCCACAGATGTA

shER-3R ERa TCCGAGGCAGTAGGCAAGGAGTTTGTGTGCCTCAAATCTACA 1689
TCTGTGGCTTCACTA

Control Tbx2 TGCTGTTGACAGTGAGCGAGCCAAGTATATCCTGCTGATGTA 724
shRNA-F GTGAAGCCACAGATGTA

Control Tbx2 TCCGAGGCAGTAGGCAGGCCAAGTATATCCTGCTGATGTACA 724
shRNA-R TCTGTGGCTTCACTA









Table 2-2. Antibodies used for Immunoblotting (IB), Immunofluorescence (IF), or Flow
Cytometry (FC)
Antibody Species Isotype Company Catalogue Use
Number
Alexa Fluor 488- Rabbit Goat-lgG Molecular Probes A11034 IF
FITC (Invitrogen)
BrdU-FITC Mouse-lgG eBioscience 11-6071-41 FC
(PRB-1)
CD44-FITC Human Mouse-lgG BD Pharmingen 555478 FC
(G44-26)
ERa (HC-20) Human Rabbit-lgG Santa Cruz Biotechnology SC-543 IB
ERa (MC-20) Human Rabbit-lgG Santa Cruz Biotechnology SC-542 IF
HRP-Donkey Mouse Donkey-lgG Jackson ImmunoResearch 715-035- IB
anti mouse 150

HRP-Donkey Rabbit Donkey-lgG Jackson ImmunoResearch 711-035- IB
anti rabbit 152

Iso-FITC Mouse-lgG2b Molecular Probes MG2b01 FC
LSD1 Human Rabbit-lgG Cell Signaling C69G12 IB
Tubulin Human Mouse-lgG Sigma T9026 IB




Table 2-3. Primers used for real-time RT-PCR
Name Forward Reverse Tm (C)

hERa CCGGCATTCTACAGGCCA TCGGTCTTTTCGTATCCCAC 60


hERp TGTCTGCAGCGATTACGCA GCGCCGGTTTTTATCGATT 60


hTFF1 GTACACGGAGGCCCAGACAGA AGGGCGTGACACCAGGAAA 64


hCTSD CTGCACAAGTTCACGTCCAT ACTGGGCGTCCATGTAGTTC 60









CHAPTER 3
CHARACTERIZING MAMMOSPHERES DERIVED FROM MCF7 PARENTAL CELLS

Introduction

Despite the assumption that breast cancer cell lines are relatively homogeneous,

the cells in tissue culture are highly dynamic and heterogeneous in reality (Lacroix and

Leclercq, 2004; Burdall et al., 2003). It has been demonstrated that tissue culture

conditions can select and enrich for different dominant cell types from the original

source (Osborne et al., 1987; Ince et al., 2007). One of the key points in the cancer

stem cell (CSC)/tumor initiating cell (TIC) hypothesis has been that there exists an

inherently more tumorigenic cell population in a particular tumor. This population is

believed to be more resistant to treatment and may reseed the tumor at a later date,

thus contributing to tumor recurrence. Previous studies have taken advantage of the

innate heterogeneity in tissue culture and used mammosphere culture techniques to

enrich for potential TICs from established cell lines. This included MCF7 cells (Ponti et

al., 2005), which is an ERa positive breast cancer epithelial cell line. The

mammosphere-enriched culture contained cells that were characterized as less

differentiated, more resistant to conventional antitumor treatments, and more

tumorigenic (Phillips et al., 2006; Fillmore et al., 2008).

In the context of tumor recurrence following antiestrogen treatment, It has been

hypothesized that ERa positive TICs may be responsible for antiestrogen resistance

and tumor recurrence in ERa expressing cancers (Dontu, EI-Ashry, et al., 2004). This

idea has been controversial because evidence for the existence and role of ERa

positive stem cells in primary culture were contradictory. This is not surprising as this

cell group may exist in a very limited window during mammary development, as









discussed in Chapter 1. The group of ERa positive progenitor cell suits the

requirements for TICs as they are sufficiently de-differentiated to maintain stem-like

features such as self renewal and bipotency. However, they may express functional

hormone receptors that can respond to antihormonal treatments.

To test this idea, we first generated mammospheres from MCF7 cells and

determined their ERa status. MCF7 was chosen because there are few ERa positive

cell lines currently available for study. Also, the relationship between MCF7-derived

mammospheres (MCF7S) and potential TICs has been supposed by earlier studies.

The human breast tumorigenic marker CD44 antigen expression (AI-Hajj et al., 2003)

has been correlated with MCF7S sphere formation frequency and stem-ness markers

like Oct4 has also been associated with MCF7S (Ponti et al., 2005). Therefore, MCF7S

may serve as a model for studying antiestrogen response in potential TICs in ERa

breast cancer.

Results

Characterization of Mammospheres (MCF7S) Derived from MCF7 Cells

MCF7-derived mammospheres were derived by first plating MCF7 parental cells

(MCF7P) at 5000 cells per ml cell density in defined serum-free media under

suspension conditions as described in Materials and Methods. The MCF7S cells have

a doubling time of 2.5-3 days under these conditions (Figure 3-1). As they divide, each

cell formed increasingly larger spheroids and extended microspikes (Figure 3-2A;

Figure 3-2B). The microspikes are presumed to be microfilaments that the spheroid

cells use to detect nutrient availability in the surrounding environment. Eight days after

initial plating, spheres were typically over 50 microns in diameter and formed a









multicellular mass (Figure 3-2C; Figure 3-2D). The culture was maintained without

additional growth factor supplementation or culture media change over the course of a

week from initial plating until next passage. Initial attempts of growth factor

supplementation resulted in the swelling of cells and subsequent lysis.

The mammospheres were serially passage to evaluate their growth kinetics as

they transformed from MCF7 parental cells. MCF7P cells were seeded at a density of

10000 cells per ml at each passage, and viable cells were counted using trypan blue

exclusion after 1 week in culture. The cells were thus tracked for 6 weeks and the data

was plotted as fold change in cell density between final cell count and initial seeding

(Figure 3-3). Two independent experiments were performed and they demonstrated

that MCF7 parental cells undergo a rapid expansion in proliferation during the first 2-3

weeks in mammosphere culture. However, as the selection progresses a dominating

subgroup emerges with stable growth kinetic. After 3 weeks, the initial MCF7 parental

cell culture transformed into the MCF7S culture enriched with an alternate dominant

subgroup and has predictable growth rate of roughly 200-fold change per passage.

Expression of Putative Breast Tumorigenic Marker CD44 in MCF7S Cells

The CD44 antigen was first described as an integral cell membrane glycoprotein

with a role in cellular attachment to the extracellular matrix via hyaluronate (Aruffo et al.,

1990). Subsequent studies have demonstrated a key role for CD44 during development

and tumor formation as a necessary component for cellular migration and invasion (Jin

et al., 2006; Schmits et al., 1997; Godar et al., 2008). CD44 expression was identified

as a marker for breast cancer cell tumorigenicity (AI-Hajj et al., 2003), and has been

used to isolate TICs in mammosphere cultures derived from both primary (Dontu et al.,









2003) and established cell lines (Fillmore et al., 2008; Ponti et al., 2005; Cariati et al.,

2008).

This marker was used to further characterize MCF7-derived mammospheres and

to determine if there is similar enrichment. Previous studies used CD44 expression to

sort for potentially tumorigenic cells prior to mammosphere culture (Fillmore et al., 2007;

Shipitsin et al., 2007; Grimshaw et al., 2008; Li et al., 2008). It was necessary to

confirm that mammosphere culture alone can enrich for a similar subgroup, even if the

efficiency is lower. MCF7 parental and MCF7S cells were both fixed and stained for

CD44 antigen as described in Materials and Methods. The cells were assayed using

flow cytometry to quantify the relative percentages of CD44 positivity in the samples.

Isotype controls were included for both cell types to account for background signal. As

shown in Figure 3-3 and Table 3-1, there is an approximately 59.92% of MCF7S cells

expressing CD44 while only 1.31% of MCF7 parental cells expressed the marker. The

experiment was repeated and MCF7S CD44 expression percentage was reproducible

(Figure A-1 and Table A-1). However, MCF7 parental cell can have up to 17-25%

CD44 expression after long-term culture (Figure A-1 and Table A-1).

ERa Status and Stability in MCF7S Cells

It is unknown whether ERa expression and stability may be linked to

mammosphere formation. Previous studies have argued that mammosphere formation

is associated with ERa negative, basal cell types rather than ERa positive, luminal

subgroups (Sleeman et al., 2007; Asselin-Labat et al., 2007). More recent evidence

argued that a stem cell hierarchy exists during mammary stem cell development that

supports the notion of a lineage-restricted, ERa positive progenitor cell (Villadsen et al.,









2007). With these conflicting viewpoints, ERa expression in MCF7S cells was

compared to MCF7 parental cell to determine if mammosphere formation alters ERa

expression. The RNA was extracted from both cell types and a cDNA library was

generated from total RNA using reverse transcription. Semi-quantitative real time RT-

PCR showed comparable levels of both ER isoforms (alpha and beta) for MCF7P and

MCF7S (Figure 3-5A). ER expression levels in both samples were normalized to beta-

actin expression (Figure 3-5A). It appeared MCF7S displayed lower levels of ER

isoforms than MCF7P. The difference in ERa expression was less than 2-fold and was

not considered significant. The difference in ER3 expression was 2.06 and was

considered to be significant, but the biological relevant of the difference was unknown.

Indirect immunofluorescence was used to determine if there were variations in ERa

protein stability at the single cell level. For a negative control, no primary antibody was

used and the cells were stained with secondary antibody only. ERa negative HEK293

cells were used as negative controls during initial experiments. As shown in Figure 3-

5B, ERa was primarily a nuclear protein in both MCF7P and MCF7S. All MCF7 cells

were positive for ERa.

Discussion

It is possible to take advantage of innate heterogeneity in established breast

cancer cell lines to establish a mammosphere culture. However, the efficiency and

reproducibility is quite variable. Only a single cell line was used in this study, as

attempts at generating mammospheres from other cell lines had been unsuccessful.

MDA-MB-231 (human ERa negative breast cancer epithelial), Eph4 (mouse non-

tumorigenic ERa positive mammary epithelial), MCF10A (human non-tumorigenic ERa









negative breast epithelial), BT474 (human ERa negative breast cancer epithelial), and

ZR-75-1 (human ERa positive breast cancer epithelial) have either failed to survive in

mammosphere culture outright, or failed to expand in subsequent passages.

The reproducibility of MCF7-derived mammospheres hints that unknown factors

lend MCF7 additional flexibility to environmental changes. This feature may be due to

the selection of a subgroup with favorable traits, or possibly to the active transformation

of the population in response to extracellular conditions. There is insufficient

information at this time to support either conjecture. However, one may hypothesize

that both scenarios are in play. This is because if there is indeed a dominant subgroup

that survived the selection process, one may expect a higher percentage of CD44

positive cells than the roughly 60% that exists in mammosphere culture as the dominant

subgroup would expand with each additional passage. As there is no clear dominance,

one may suppose that a mixed population still exists and has shifted to a new

equilibrium. The newly emerged dominant population in the MCF7S culture may have a

distinguishing molecular profile (Kok et al., 2009). However, these differences are not

fully characterized in this study.

In summary, the parental MCF7 cell population possessed a predictable growth

rate once it acclimated to mammosphere culture conditions (Figure 3-3). The resulting

MCF7S culture showed greater consistency in CD44 expression than MCF7 parental

cells (Figure 3-4). These two observations indicated that there was a shift in equilibrium

as the adherent MCF7 parental cells adapted to mammosphere culture. However, the

exact mechanism for the shift is unknown. The presumed differences that existed

between the MCF7S and MCF7P cultures may affect hormone sensitivity of the two cell









types. This hypothesis was first assessed by evaluating ERa expression in the two

groups. As shown in Figure 3-5, there was no significant difference in ERa mRNA

expression or protein stability. However, there was a significant difference in ER3

expression that may have further relevant but was not probed further.

This did not offer sufficient explanation for ERa requirement in mammosphere

formation or maintenance. It also precluded assumptions relating antiestrogen

response in the two groups. These considerations were further studied and discussed

in Chapter 4 and 5.




























Figure 3-1. MCF7-derived mammospheres (MCF7S) growth curve (n=2). The average
viable cell density of MCF7S was assessed every 48 hours after initial
seeding at 2500 cell per ml.


Figure 3-2. Phase contrast microscopy of mammospheres derived from MCF7 parental
cells. Representative images of: A) 2 days after initial seeding, B) 4 days
after initial seeding with microspikes indicated by arrows, C) 8 days after
initial seeding, and D) Hoechst nuclei staining of mammospheres 8 days after
initial seeding to show mammosphere are multicellular. Scale bar = 100
microns.


1200000
n=2
1000000

800000

. 600000

400000

200000

0
0 2 4 6


Day 2












Day 8


D














= 200
E
IA



r 100
S150





0
o


1 2 3 4 5 6


Passage


Figure 3-3. Enriching MCF7S from MCF7P by serial passage in mammosphere media
(n=2). Each passage is 7 days. Fold change is expressed as final cell
density/initial cell density. The average of two independent experiments is
represented on the graph.




















Side Scatter


B

: 0.4% Key
MCF7P ISO
A : MCF7P CD44-FITC 1
f .. ....... MCF7P CD44-FITC 2
: f- <-- 1.31% MCF7S ISO
.. .. i MCF7S CD44-FITC 1
MCF7S CD44-FITC 2



-,, < 59.92%






100 10 02 l3 l4
CD44 FITC

Figure 3-4. Comparing CD44 expression in MCF7P vs. MCF7S using flow cytometer.
A) Representative scatter plot and gating of FACS sorted cells. B)
Representative overlay histogram of CD44-FITC and Iso-FITC staining for
MCF7P and MCF7S. Technical duplicates were shown for the experiment.
Isotype staining control (Iso-FITC) showed negligible staining for both sample
groups. MCF7P cells showed an averaged 1.31% CD44-positive staining.
MCF7S cells showed an averaged 59.92% CD44-positive staining.
Percentages were calculated from 30000 cells.









Table 3-1. Data for CD44-FITC signal quantification using flow cytometer.
Sample Gated Events Background % Gated % Total (Not
Subtracted gated)
Events
MCF7P Iso 28329 112 0.4 0.37
MCF7P CD44-1 28537 375 1.31 1.42
MCF7P CD44-2 28548 427 1.5 1.42
MCF7S Iso 29504 24 0.08 0.08
MCF7S CD44-1 28922 17330 59.92 57.77
MCF7S CD44-2 29413 19419 66.02 64.73

Table 3-1 data represents additional experiment to evaluate CD44 expression in
MCF7P and MCF7S (Figure 3-4). A total of 30000 cells were counted for each sample.
Gated events represent cell fraction used for analysis (R1) based on forward and side
scatter pattern as shown in Figure A-1A. Background subtracted events represented
CD44-FITC signal levels above Iso-FITC control. % Gated is percentage ratio of gated
events / background subtracted events. % Total is percentage ratio of gated events /
30000 total events.













1.2


o E
o.8
.. 0.8 n=2
X WC
S0.6 U ERa
0 ER3
E 0.4

0.2

0


MCF7P


MCF7S


B MCF7P C MCF7S



ERa ERat








Negative Negative
Control Control





Figure 3-5. ERa expression in MCF7P and MCF7S. A) Semi-quantitative real-time RT-
PCR for ERa and ER3 expression in MCF7P and MCF7S cells. The data is
shown as average of two independent experiments. Each experiment was
performed with technical triplicates. Both ERa and ER3 expression was
normalized to beta-actin expression. Technical duplicate negative control (no
cDNA template) was included in the experiment. B and C) Indirect
immunofluorescence for ERa protein (green) was performed on MCF7P and
MCF7S (B and C, respectively). DAPI (blue) counterstain was used to
indicate nuclear region. Immunofluorescence negative control was performed
by omitting primary anti-ERa antibody. Abbreviations: MCF7P is MCF7
parental. MCF7S is MCF7 mammosphere culture. ERa is estrogen receptor
alpha. ER3 is estrogen receptor beta.









CHAPTER 4
THE EFFECTS OF ANTIESTROGEN ON MAMMOSPHERE FORMATION

Introduction

If the mammosphere culture has indeed selected or transformed the parental

culture, one may expect MCF7S to display altered behavior when compared to MCF7P.

In previous studies, mammosphere cells were shown to be resistant to

chemotherapeutic agents such as 5-fluorouracil and paciltaxel (Fillmore et al., 2008)

when compared to parental cells. The observation implied that there were intrinsic

properties in mammosphere cells that lend additional resistance to cytotoxic agents,

rather than acquiring resistance from extended exposure.

We questioned if such resistance is an indication of potential TICs enrichment in

mammosphere culture, and if it is applicable to antiestrogens. We were interested in

studying MCF7S antiestrogen response, which may contribute new insights into cancer

recurrence in ERa positive tumors. We utilized the sphere forming ability of MCF7S

cells to query their tumor-initiating potential, and to quantify possible TIC enrichment.

The sphere formation assay was also used to assess the bulk culture's tumorigenic

potential following antiestrogen challenge. Pharmacological inhibition by antiestrogens

also served to elucidate the role of ERa in mammosphere formation and maintenance.

The rationale for the sphere formation assay was that tumorigenic cells have a

higher probability of survival in mammosphere culture, while the non-tumorigenic

population is less likely to survive under the same conditions. Therefore, it would be

possible to estimate the tumor-initiating potential of a cell population by quantifying the

sphere forming frequency of the MCF7S population. While it is true that not every

mammosphere is derived from a TIC, previous studies have suggested the









mammosphere-forming population to be more tumorigenic in general (Ponti et al., 2005;

Li et al., 2008; Grimshaw et al., 2008; Fillmore et al., 2008; Cariati et al., 2008). The

reaction of this population to antiestrogen treatments will yield information useful for

clinical applications.

In addition, we studied the long-term expansion of MCF7S cells under

antiestrogen challenge to determine the growth kinetics of potential TICs. If the MCF7S

culture is indeed enriched for potential TICs, there would be a stable subpopulation that

maintains the culture over many passages while under antiestrogen challenge. The

expansion of this subgroup would remain relatively constant over time and would not be

affected by antiestrogen treatment. However, if there is merely selection of an

antiestrogen resistant subgroup it would be reflected in the growth kinetics of the bulk

culture over time. For example, if the population had acquired resistance then the cell

number is expected to increase at later passages as the dominant subgroup continues

to expand. Inversely, if the population becomes quiescent or senescent then cell

number would decrease over time.

Two classes of antiestrogens were used in this study. The first is selective

estrogen receptor modulator (SERM), which is the classic ERa antagonist. The

particular compound used was 4-hydroxytamoxifen (4-OHT), which is an active

metabolite of tamoxifen. The second is selective estrogen receptor down-regulator

(SERD), which induces ERa degradation and impairs receptor dimerization. ICI 182780

(ICI), also known as Fulvestrant or FASLODEX was the SERD chosen for this study.









Results


MCF7S Response to Antiestrogens

The initial treatment of MCF7S with antiestrogens produced an interesting

phenotype. MCF7S were seeded at 5000 cells per ml and treated with vehicle or

various dilutions of 4-OHT and ICI. The cells were incubated for 6 days and spheres

above 50 microns in diameter were scored. As shown in Figure 4-1, there was a

marked difference in sphere formation ability when the cells were exposed to the two

different antiestrogens. Figure 4-1A illustrated the loss of sphere formation upon

treatment with at least 2.5 piM 4-OHT, which caused cells to form disordered

aggregates. Quantification of sphere formation is shown in Figure 4-1B. Samples

exposed to vehicle controls, 10 piM 173-Estradiol, 1 piM 4-OHT, or two different

concentrations of ICI did not exhibit sphere formation disruption and remained as tightly

packed spheroids. Three different concentrations of 4-OHT were tested (1, 2.5 and 5

piM) and sphere formation inhibition was evident at concentrations of 2.5 piM or above

(Figure 4-1B). This result suggested MCF7S can respond to antiestrogen treatments

and that ERa antagonism has a significantly different effect than ERa reduction.

Proliferation study was used to further characterize antiestrogen effects on

MCF7S cells. The cells were plated at 5000 cells per ml and treated with antiestrogens

on the day of seeding. The cells were then counted every 48 hours for 6 days to

determine the proliferation rate of these cells in the presence of antiestrogens. Figure

4-2 demonstrated the effects varied amounts of SERM /SERD had on MCF7S

proliferation. It showed MCF7S responded to antiestrogens in a dose dependent

manner. Proliferation of MCF7S cells were not significantly affected by 1 piM 4-OHT,









while 2.5 piM 4-OHT significantly decreased (p<0.05) MCF7S proliferation by about

50%. MCF7S proliferation was further decreased by 5 piM 4-OHT, which may be signs

of cytotoxic effects (Figure 4-2A). The effects of 1 piM ICI treatment were comparable to

that of 2.5 pM 4-OHT for the inhibition of MCF7S proliferation (Figure 4-2B). MCF7S

cells treated with 0.5 piM ICI showed a decrease in cell proliferation, but it is not

significant when compared to 0.1% DMSO control (Figure 4-2B).

Antiestrogen Efficacy in Parental MCF7P and MCF7S

MCF7P cells were then treated with 4-OHT to determine if they were similarly

affected by antiestrogens as MCF7S and to confirm drug efficacy. MCF7P cell are

generally maintained in complete DMEM medium that contains a full complement of

steroidal hormones, thus dulling the effects of antiestrogens. Therefore, MCF7P cells

were cultured under adherent conditions in complete DMEM, steroid-free DMEM, and

mammosphere media to compare 4-OHT effects on proliferation.

The cells were seeded using complete DMEM media onto three 6-well plates and

allowed to attach for several hours. After which, two of the plates were washed with

PBS and the media was replaced with either steroid-free (SF) media or mammosphere

media. The cells were incubated for 48 hours in their respective media, and then either

0.1% DMSO or 2.5 piM of 4-OHT was added to the cells and the time course

commenced. The cells were counted with trypan blue exclusion every 48 hours for 6

days to assess MCF7P proliferation. The experiment was repeated in Figure 4-3C

using 6 days incubation time point because that is when significant differences were

detectable.









Figure 4-3 showed there were no major differences in 4-OHT response

regardless of the culture media used. Figure 4-3A indicated no significant differences in

proliferation between MCF7P cells cultured in complete DMEM or steroid-free (SF)

media, and there were little differences in cell number in response to 2.5 piM 4-OHT.

Nor were there a significant difference between cells cultured in steroid-free (SF) media

or mammosphere (sphere) media as shown in Figure 4-3B and Figure 4-3C. In

summary, MCF7P was unaffected by differences in culture media when grown under

adherent conditions and were not significantly affected by 4-OHT under these

conditions. The results indicated a significant difference in antiestrogen response

between MCF7S and MCF7P.

It was necessary to confirm drug efficacy in the two cell groups in order to

appraise their respective response. Nuclear extraction was performed for both cell

types after antiestrogen treatment for 48 hours to determine if there is a difference in

ERa stability or localization. ICI is a known SERD, which gave a clear confirmation of

drug efficacy in both MCF7S and MCF7P as ICI decreased overall ERa protein levels

(Figure 4-4). 4-OHT has been implicated as an ERa stabilizer in previous studies

(Wijayaratne and McDonnell, 2001; Marsaud et al., 2003). The results in Figure 4-4

verified 4-OHT effect to be consistent in both cell groups with the increase in ERa

observed in the nuclear fraction.

ERa transcriptional function was examined using real-time RT-PCR to evaluate

the expression of Trefoil Factor 1 (TFF1), a known estrogen response gene. TFF1

expression in MCF7P cells was predictably upregulated by 173-estradiol (E2) treatment,

and the gene was maximally expressed 12-24 hours following E2 addition (Figure A-









2A). MCF7P showed noticeable reduction of TFF1 expression upon pre-treatment with

antiestrogens (Figure A-2B, A-2C). TFF1 expression in MCF7S cells were increase 12-

24 hours after E2 addition, and followed a similar trend as MCF7P (Figure A-3A).

However, data for antiestrogen inhibition of TFF1 induction were inconsistent and

overall changes were too low (less than 2-fold) to be considered significant (Figure A-

3B, A-3C). CTSD was another estrogen response gene tested (Figure A-4), but it did

not showed significant upregulation upon E2 treatment and was not further evaluated.

This may reflect inherent heterogeneity in gene expression profile upon antiestrogen

treatments, or ERa was no longer responsive at the transcriptional level.

Sphere Formation Frequency Following Antiestrogen Challenge

Previous studies have correlated mammosphere formation with TIC enrichment

(Grimshaw et al., 2008; Fillmore et al., 2008). Therefore, sphere formation frequency

was quantified after antiestrogen challenge as an indicator of tumor-initiation potential.

Figure 2-1 offers the general workflow for the study. Following cell count for the

evaluation of MCF7S antiestrogen response (Figure 4-2), the surviving cells were

related at approximately 500 cells per well in a 96-well plate without drugs to

determine sphere formation efficiency after antiestrogen challenge. Spheres larger than

50 microns in diameter were scored for sphere formation frequency.

As shown in Figure 4-5, there was no statistically significant decrease in sphere

forming frequency following antiestrogen treatments with the exception of 2 days 0.5 piM

ICI treatment (p=0.028) in Figure 4-5B. There was a statistically significant increase in

sphere forming frequency with 0.1 piM 4-OHT (p=0.0004) and 2.5 piM 4-OHT (p=0.014,

4 days), (p=0.021, 6 days) treatments (Figure 4-5A), however the biological significant









of this observation is unknown. These results suggested that antiestrogens did not

statistically decrease sphere formation potential of MCF7S cells, and that less than 10%

of bulk MCF7S cells were responsible for sphere formation. In addition, the evidence

supported the notion of a relatively stable subpopulation in MCF7S was resistant to

acute antiestrogen treatment and survived to perpetuate sphere formation.

MCF7S Cell Proliferation under Long-Term Antiestrogen Treatment

According to the TIC hypothesis, TICs remain and persist in the bulk population.

This infinite growth serves to "reseed" a tumor after initial treatments and contributes to

overall malignancy. However, a subtle detail should be noted as the TIC hypothesis

does not stipulate the TICs would ever become the dominant population through

endless expansion. The theory argues that potential TICs only grow and divide

sufficiently to maintain themselves, and the growth of the bulk population is dependent

on non-TICs. This reasoning contends that cells in the bulk population were more

dynamic and their growth can be altered due to environmental causes. Therefore, if

MCF7S does indeed contain a subpopulation of TICs, then it is possible to examine this

property through long-term serial passage. This method, which is analogous to that

used in neural stem cells characterization (Reynolds and Rietze, 2005), can also probe

the possibility of antiestrogen effects on long term mammosphere culture self-renewal

property.

This was done by seeding viable cells at 10000 cells per ml at each passage. The

cells were treated either with vehicle or antiestrogen at the time of seeding, and

incubated for 7 days which constitutes one passage. Viable cells were counted at the

end of each passage. The fold change in cell number for each passage was used to

calculate potential expansion of the population if all the cells, instead of a fraction, were









passage. The long term growth kinetics can be determined in this manner and allowed

for comparison between different antiestrogen treatments.

As shown in Figure 4-6 and Table 4-1, 0.1% DMSO treated MCF7S did not

influence long term cell expansion. There was a small decrease in growth for 1 piM and

2.5 piM 4-OHT treated MCF7S, but there were no significant differences between the

two drug concentrations. 5 piM 4-OHT and ICI treated cells suffered the most dramatic

decrease in long term proliferation and expansion. However, the differences were not

statistically significant.

Long term serial passage further delineated the marked differences in mechanism

for the two classes of antiestrogens, which helped to clarify some conclusions drawn

from Figure 4-5. The minimal decrease in growth kinetics for 1 piM and 2.5 piM 4-OHT

treated MCF7S helped to explain the increase in sphere formation frequency in those

samples. This information suggested that indeed not every sphere was derived from a

potential TIC, but the overall health and proliferation of a cell contributed to sphere

formation frequency. Paradoxically, decrease in long-term cell expansion did not

significantly decrease sphere formation frequency, as indicated by 5 piM 4-OHT and ICI

treated cells (Figure 4-5 and Figure 4-6). The results indicated sphere formation

potential of the MCF7S bulk population was not decreased by antiestrogens, even when

the long term growth rate was slowed.

MCF7S Cell Cycle Analysis

To further study the growth kinetics of antiestrogens treated MCF7S, cell cycle

analysis was performed using bromodeoxyuridine (BrdU)/propidium iodide (PI) staining.

MCF7S cells were dissociated and treated with 2.5 piM 4-OHT or 1 piM ICI for either 48









or 72 hours. The cells were then fixed in 70% ethanol and BrdU/PI staining was

performed as described in Chapter 2. BrdU staining was largely unsuccessful due to

technical hurdles and data was obtained for only a single experiment (Figure A-5).

BrdU staining indicated a 10-14% increase in G1 phase retentions, with a

corresponding decrease in S phase and G2 phase cell cycle for antiestrogen-treated

cells compared to vehicle-treated control. Cell cycle data were primarily obtained from

the analysis and modeling of propidium iodide (PI) histogram using ModFit LT software.

As shown in Figure 4-7, cells treated with 2.5 piM 4-OHT had a significant fraction

(p=0.0009) of cells retained in G1phase (75.3%) and significant reduction (p=0.00006)

in G2 phase (7.39%) 48 hours after treatment when compared to vehicle treated control.

There was significant reduction (p=0.0003) in S phase (18.9%) 72 hours following drug

introduction. MCF7S cells treated with 1 piM ICI had a significant reduction (p=0.0009)

in S phase (13.3%) cell cycle 48 hours after addition of the compound. Cellular growth

arrest was more pronounced at 72 hours as a significant (p=0.0003) percentage of cells

stalled at G1 phase (78.8%) with a correspondingly significant (p=0.00005) decrease of

cells in S phase (13.4%), compared to vehicle treated control. These results showed

growth arrest for 2.5 piM 4-OHT treated cells 48 hours after drug addition, while 1 piM ICI

showed maximal growth arrest 72 hours after drug treatment.

MCF7S Apoptosis Assay

Apoptosis assay using Annexin V staining was performed to account for observed

cell loss after antiestrogen treatment. As shown in Figure 4-8, there was a statistically

significant (p=0.025) increase in apoptosis for 2.5 piM 4-OHT treated cells 72 hours of

drug treatment. However, there was no significant apoptosis observed upon 1 piM ICI









treatment. There was significant quantitative variation between experiments, but the

trend was consistent within individual experiments as 2.5 piM 4-OHT treated MCF7S

consistently displayed a statistically higher percentage of dead cells than those treated

with 1 piM ICI (Figure A-6). However, the data generated from Annexin V assay may

not be biologically significant as attempts at other cell death assays (TUNEL assay,

trypan blue) failed to detect significant changes in cell death 48 or 72 hours after

antiestrogen treatment. Therefore, it is possible that the loss of viable cells is due to

acute necrosis induced within the first 24 hours of antiestrogen treatment.

Discussion

MCF7S cells possessed a unique disposition that altered their response to

antiestrogens when compared to MCF7P cells (Figure 4-3). The initial assumption that

MCF7S cells may be more resistant to antiestrogen than MCF7P was dispelled, as the

inverse appeared to be true. This observation may be the result of optimized growth

condition in MCF7P adherent culture that rendered the cells less responsive to

antiestrogens. It is also possible that a higher degree of heterogeneity was maintained

in adherent MCF7P culture, which resulted in a higher background and hindered

antiestrogen response detection. The MCF7S culture was enriched for a narrower

range of cells grown under more stressful conditions, which may contribute to the

distinct antiestrogen response profile when compared to the original population.

The data suggests that mammosphere formation is not dependent upon ERa

status as SERD treated cells retain sphere forming ability (Figure 4-1). However,

MCF7S cells are responsive to antiestrogen effects and ERa may play a role in cell

proliferation kinetics (Figure 4-2 and Figure 4-6). Typically less than 10% of the ERa









positive cells can form spheres larger than 50 microns (Figure 4-5). These cells

appeared to be more resistant to acute SERM treatments, as the normal recommended

dosage is 1 piM for 4-OHT to inhibit proliferation which had little effect on MCF7S cells

(Figure 4-2). SERD induced growth inhibition was consistent with published studies,

which utilized 1 piM to demonstrate efficacy (Figure 4-2). If mammosphere culture truly

enriched for tumor-initiating cells, then these results suggest tumorigenicity may not be

dependent on hormone receptor status; nor does it indicate hormone receptor status as

a straightforward prognostic factor.

Interestingly, the sphere formation frequency did not significantly decrease after

antiestrogen challenge (Figure 4-5). This signified that the same percentage of

surviving cells were resistant to acute antiestrogen treatment. However, not all

surviving cells were capable of sphere formation after drug removal or may contribute to

tumor persistence. The two classes of antiestrogens produced different characteristics

in surviving cells. The conclusions concerning antiestrogen action suggested ERa

antagonism may result in acute cell death (Figure 4-8) and perhaps growth arrest as a

secondary result (Figure 4-7), while ERa destabilization may resulted mainly in growth

inhibition (Figure 4-7). The observations for 4-OHT may be consistent with the sphere

formation phenotype (Figure 4-1) because cell membranes are disrupted during

necrosis and would inhibit cell-cell adhesion. In the case of ICI induced growth arrest

(Figure 4-7), sphere formation phenotype and frequency remained undisturbed (Figure

4-1 and Figure 4-5).

This theory was supplemented by long term serial passage studies, which

confirmed that the two antiestrogens influenced MCF7S growth in different manners









(Figure 4-6). 4-OHT did not significantly change the growth kinetics of MCF7S cells,

which is consistent with the increase in cell death (Figure 4-8). The signs of growth

arrest shown in Figure 4-7 did not appear to contribute significantly to long term cell

expansion, which led to the conclusion that 4-OHT sensitive MCF7S cells may undergo

growth arrest in addition to cell death. This situation was not observed in ICI treated

MCF7S as the compound was found to be more effective against cell expansion (Figure

4-6 and Figure 4-7), then stimulating cell death (Figure 4-8). This indicated that while

mammosphere formation does not require ERa, the receptor can control proliferative

potential in individual cells by inhibiting growth or cell death. One hypothesis is ERa

has an alternative role for a fraction of the mammosphere forming cells, which may

account for differential response to antiestrogens within the MCF7S population.

As discussed in Chapter 1, ERa can influence cell proliferation through both

genomic and non-genomic pathways. The receptor functions are modulated by factors

such as ligand availability and binding, cofactor recruitment, and mitogenic signaling.

Therefore, the observation that the two classes of antiestrogens exert different effects

on MCF7s is to be expected. It is also possible that 4-OHT has non-ERa targets, or 4-

OHT may generate proliferative effects in cells through alternative ERa signaling. In

this culture model, after the removal of 4-OHT, the cells resume sphere formation and

are expected to still be ERa positive. These features suggested that antiestrogen had

no significant effect on tumor initiating potential of ERa positive breast cancer cells.

However, antiestrogen may continue to suppress tumor growth and prevent recurrence

over an extended period of time.











Untreated


ii.


10pM E2




.I


0.1% DMSO





-,I


2.5pM 4-OHT


.E"a...1
~i, .


10
9
U 8
C

0
o


1 3



0

S DMSO 0.5uM ICI luM CI luM40HT 2.5uM 40HT 5uM 40HT


Figure 4-1. Mammosphere formation in the presence of antiestrogens. A)
Representative phase-contrast images of MCF7S cells in the presence of
antiestrogens or vehicle controls for 7 days. B) Quantification of >50 microns
diameter MCF7S spheres scored from images in A). The percentage on y-
axis was calculated as (number of >50 micron spheres / total number of cells
seeded) X 100%. Abbreviations: E2 is 173-estrodiol. 4-OHT is 4-
hydroxytamoxifen. ICI is pure antiestrogen ICI 182780. Scale bar = 100
microns.


0.1% EtOH


49


- Y:.


U t


1pM ICI


0


';;""


;;iiiiii. ....,,,iiir~iii~











1200000

1000000 n=4

S800000
S-4-0.1% DMSO
r 600000 I-4-OHT
E1pM 4- OHT
400000 2.5pM 4-OHT
T -X- T5p M 4-OHT
J 200000

0
0 2 4 6
*p<0.05 Days
B
1200000

1000000 n=4
E
S800000

S--00000.1% DMSO
600000
---0.5 IM ICI
S400000 M I

u 200000



0 2 4 6
*p<0.05 Days


Figure 4-2. Cell proliferation of MCF7S in the presence of antiestrogens. A) Cell
proliferation of 4-OHT treated MCF7S derived from four independent
experiments. B) Cell proliferation of ICI treated MCF7S derived from four
independent experiments. Abbreviations: 4-OHT is 4-hydroxytamoxifen. ICI
is pure antiestrogen ICI 182780. Error bars are calculated standard error of
the mean (S.E.M.). Statistical analysis was performed using pair Student's t-
test.













1400000

1200000 n=1

1000000 U DMEM+0.1%DMSO

800000 U DMEM+2.5uM 40HT
I a SF+0.1%DMSO
@ 600000 -
Om SF+2.5uM40HT
400000 *- sphere+0.1% DM50

200000 sphere+2.5uM 40HT


days


days days


Figure 4-3. MCF7 adherent culture response to 4-hydroxytamoxifen (4-OHT) with
various culture media. A) Comparing 4-OHT response in MCF7P cultured in
complete (DMEM), steroid free (SF) and mammosphere (sphere) media. B)
Comparing 4-OHT response in MCF7P cultured in steroid-stripped (SF) vs.
mammosphere (sphere) media. C) Comparing 4-OHT response in MCF7P
cultured for 6 days in SF vs. sphere media. Abbreviations: 4-OHT is 4-
hydroxytamoxifen.


900000
800000
700000
600000
500000
400000
300000
200000
100000
0


n=1


5 SF+0.1% DMSO
4 SF+2.5uM 40HT
0 sphere+0.1% DMSO
sphere+2.5uM 40HT




days


300000

250000

200000

a 150000

100000

50000



SF+0.1%DMSO SF+2.SuM4-OHT sphere+0.1% sphere+2.5uM
6 days DMSO 4-OHT
n=l














0.1% DMSO 2.' il 140HT liM ICI
C N C N C N


a-ER

a-tubuin

a-LSDL


MCF7 Parental ERa Loci.l;z 4-ion (n-3)


'p<0.D5 "ODMSO0 C O1DJMSO


5uM40HT 0 2 5uM40HT HN


0.1% DMSO :11i 140HT
C N C N


a-ER

a-tubulin

a-LSD1


MVCF7S ERa Localization (n-3)



25






*p 5 DMSO C 1%DrSO N 2u4 T C 2 hT C 5uM40HT N 1IICI C uM ICI N


Figure 4-4. Immunoblot of cytoplasmic and nuclear fractions from MCF7P and MCF7S
cells treated for 72 hours with 4-hydroxytamoxifen (4-OHT) or ICI 182780
(ICI) to characterize ERa stability. A) Immunoblot for MCF7P. B)
Immunoblot for MCF7S. a-tubulin was used as loading control for
cytoplasmic (C) fraction. a-LSD1 was used as loading control for nuclear (N)
fraction. Densitometry quantification of three independent experiments is
shown below a representative immunoblot. Statistically analysis was
performed using paired Student's t-test.


luM C1 (


lIM ICI
C N












14

12 n=4



8 0O.1% DMSO
8,
1 M IM 4-OHT
w 6
2.,-D pr %.14 OHT
4-,
S4 5pM4-OHT

2

0
days 4days 6days
*p<0.05

B
14

12 n=4

10




h y 6 0.512M ICI











Figure 4-5. Sphere formation frequency of MiF7S after antiestrogen challenge using 4-
hydroxytamoxifen (4-OHT) or ICI 182780 (ICI). A) Sphere formation
frequency of 4-OHT treated MGF7S derived from four independent
experiments. B) Sphere formation frequency of ICI treated MCF7S derived
from four independent experiments. Plating efficiency was calculated as
(number of sphere >50 microns diameter/ total number of cells plated) x
100%. Statistically analysis was performed using paired Student's t-test.





































1.OE+18

1.OE+16 -

1.OE+14 n=4
E


/ -W- MCF7S +2.5 pM 4-OHT
1.OE+10
m MCF7S +1pM ICI
1.OE+08

1.OE+06

1.OE+04 ---- .
0 1 2 3 4 5 6
Passage


Figure 4-6. Long-term expansion of MCF7S in the presence of antiestrogens. The lines
are expressed on a semilog graph and slope of each line was calculated as
log expansion for each condition (Table 4-1). A) Average expansion derived
from four independent experiments of MCF7S treated with antiestrogens or
vehicle compared to untreated control. B) Comparing average expansion of
vehicle control against 2.5 pM 4-OHT and 1 pM ICI treated MCF7S using data
set from A). Abbreviations: 4-OHT is 4-hydroxytamoxifen. ICI is pure
antiestrogen ICI 182780. Error bars are shown as calculated standard error
of the mean (S.E.M.).


1.0E+18

1.0E+16

1.OE+14

1.0E+12

1.OE+10

1.0E+08

1.0E+06

1.0E+04


n=4
-0-MCF7S

---MCF7S +0.1% DMSO

-- -MCF7S +1M 4-OHT

MCF7S +2.5pM 4-OHT

- MCF7S +5M 4-OHT

MMCF7S +0.5pM ICI

-- MCF7S +1M ICI


l/a


1.E+4 C-CFS+.1ADV


0 1 2 3 4 5 6
Passage


-


I I I









Table 4-1. Growth kinetics of MCF7S long term expansion under antiestrogen
challenge.
Sample Line Slope (log) R2
MCF7S 2.380 0.999
MCF7S +0.1% DMSO 2.321 0.999
MCF7S +1 pM 4-OHT 2.046 0.999
MCF7S +2.5pM 4-OHT 2.019 0.999
MCF7S +5pM 4-OHT 1.790 0.994
MCF7S +0.5pM ICI 1.861 0.999
MCF7S +1 pM ICI 1.574 0.997

Table 4-1 represents the slope of lines represented in Figure 4-6A. The slope was
calculated a log expansion for each condition.












90 -

80 -

70

60
4,V
o50

U
4

40
0 -
. 30 -

20

10


* 48h 0.1% DMSO

* 48h luM ICI

S48h 2.5uM 40HT


0,,O.


* p<0.001
**p<0.0001

n=4


0 -


G2


m 72h0.1% DMSO

* 72h luM ICI

S72h2.5uM 40HT


* p<0.001
~p
n=4


Figure 4-7. Propidium iodide (PI) cell cycle analysis. A) Cell cycle analysis of 48 hours
antiestrogen treated MCF7S derived from four independent experiments. B)
Cell cycle analysis of 72 hours antiestrogen treated MCF7S derived from four
independent experiments. PI cell cycle analysis and modeling were
performed using ModFit LT software. Abbreviations: 4-OHT is 4-
hydroxytamoxifen. ICI is pure antiestrogen ICI 182780. Error bars were
expressed as standard error of the mean (S.E.M.). Statistical analysis was
performed using paired Student's t-test.


0 -












50

40 -

30

a 20
10 -

0
0.1% DMSO 2.5iM4- 1pM ICI 0.15" DMSO 2.5pN
OHT OH-
n=3
p<0.05 48h 72
*p~0.0


Figure 4-8. Annexin V apoptosis assay for antiestrogen treated MCF7S. The average
of three independent experiments is shown. Abbreviations: 4-OHT is 4-
hydroxytamoxifen. ICI is pure antiestrogen ICI 182780. Error bars were
expressed as standard error of the mean (S.E.M.). Statistical analysis was
performed using paired Student's t-test.









CHAPTER 5
THE ROLE OF ESTROGEN RECEPTOR ALPHA ON MAMMOSPHERE FORMATION

Introduction

Evidence gathered through pharmacological inhibition in Chapter 4 suggested

mammosphere formation did not require ERa. However, the receptor can potentially

exert significant command on proliferation in a fraction of the bulk mammosphere

culture. A molecular biological approach was undertaken to confirm pharmacological

results and to delineate the role of ERa more specifically. Stable ERa knockdown

clones (shER) were generated using shRNA integration to further examine the

receptor's role in mammosphere culture.

As previously discussed in Chapter 1, there may be changes in ERa signaling

upon antiestrogen binding that may promote misleading interpretation of the results.

ERa has a role in multiple signaling pathways such as NFKB, Notch, and Hedgehog

(Chapter 1). These pathways have been implicated in mammosphere culture

maintenance as well as TIC enrichment (Murohashi et al., 2009; Liu et al., 2006; Zhou

et al., 2008). Therefore, it is necessary to clarify ERa function in MCF7S to ascertain its

role in the context of potential TICs.

The different mechanisms of growth inhibition produced by the two classes of

antiestrogens have been described in previous studies (Fan et al., 2006; Osipo et al.,

2003; Shaw et al., 2006). It is of further interest to study antiestrogen effects on ERa

knockdown cells as it may compliment studies done on hormone receptor negative cell

lines (Jeng et al., 1994; Lazennec and Katzenellenbogen, 1999; Keen et al., 2003),

which were insensitive to antiestrogen treatment. These studies have noted ectopic re-

introduction of hormone receptor restored antiestrogen sensitivity. If the inverse is true









for MCF7S cells, then one may expect shER cells to be more resistant to antiestrogens,

to have higher sphere formation frequency and higher tumorigenic potential.

Results

Proliferation of ERa Knockdown MCF7S

The generation of shER clones was described in detail in Chapter 2. The clones

were evaluated by immunoblot and several knockdown clones were identified (Figure 5-

1A) with significant ERa knockdown. The proliferation of the clones was compared to

the bulk MCF7S cells to determine potential differences in growth rate. As shown in

Figure 5-1 B, there is a decrease in cell doubling for all 3 clones with clone 11 growing

the slowest of the three. However, clones 7 and 9 had comparable doubling rates and

the difference from MCF7S was not significant.

All three shER clones were capable of sphere formation, which further confirmed

ERa was not required for mammosphere formation or maintenance (Figure 5-2). Visual

inspection of the shER clonal spheres did not reveal significant differences in size or

frequency (Figure A-7). The clones can be continuously passage as MCF7S, although

experiments were done using cells cultured for less than three passages.

Antiestrogen Response of ERa Knockdown MCF7S

The shER clones were tested for their antiestrogen response as described in

Chapter 2 and Chapter 4. Control shRNA MCF7S cells were likewise treated as

controls. Figure 5-2 showed the sphere formation phenotype of MCF7S, control shRNA

MCF7S, and shER clone 7. Both MCF7S and control shRNA MCF7S showed similar

sphere formation inhibition by 2.5 piM 4-OHT. These results were consistent with data

from Figure 4-1A. However, shER clone 7 underwent massive cell loss judging by the









presence of cell debris. The surviving cells formed tightly packed spheroids. As

expected, 1 piM ICI treatment did not affect sphere formation phenotype or significantly

affect proliferation.

The proliferation of knockdown clones under antiestrogens was further

characterized in the same manner as described previously in Chapter 2 and 4. The

clones were evaluated for only one time point after treatment (6 days) because Figure

4-2 showed it was when the most significant differences were detectable. The results

were expressed as % Cell Survival, which represented the cell density of antiestrogen

treated samples relative to vehicle treated control for each shRNA used, in order to

normalize results for comparison. Figure 5-3 summarized the results, in which control

shRNA MCF7S response to antiestrogen corroborated with data shown in Chapter 4.

Surprisingly, there was a heightened sensitivity to 2.5 pM 4-OHT treatment in all three

clones. 1 pM ICI efficacy was significantly reduced in clones 7 and 9, as the compound

had no significant effect on cell density. This was an indication that ERa was indeed

sufficiently removed from the cell culture to nullify SERD effects.

Interpretation of antiestrogen effects on clone 11 should be considered with

caution as its slower doubling time (Figure 5-1 B) may be a contributing factor to its

antiestrogen response. Its proliferation was affected by 4-OHT in a similar manner as

the other two clones. However, it did not show the same degree of resistance to ICI

treatment, which may be the result of its slower growth rate.

Sphere Formation Frequency of ERa Knockdown MCF7S Following Antiestrogen
Challenge

The sphere formation frequency of the knockdown clones were similarly evaluated

as in Figure 4-5. The clones were related in fresh media without treatment after 6









days of antiestrogen challenge. The resulting spheres were scored for their size (> 50

microns diameter) and their total number. Data in Figure 5-4 showed the same lack of

change in sphere formation frequency for control shRNA spheres as shown in Figure 4-

5. This was proven in terms of both total frequency and sphere diameter (>50 microns).

For knockdown clones, the sphere formation frequency of vehicle and ICI treated cells

were likewise stable in both MCF7S and control shRNA MCF7S. A noticeable reduction

in sphere formation frequency was identified in knockdown cell when treated with 4-

OHT. This result was consistent with the low number of surviving cells after initial

antiestrogen challenge. These observations indicated that a fraction of MCF7S that

initially escaped acute 4-OHT-induced apoptosis were sensitized after ERa knockdown.

The implication is the potential TIC pool in shER MCF7S shrank.

Using SERDs and SERMs to Mimic shER Effects in MCF7S

The increased potency observed for 4-OHT in shER MCF7S cells was puzzling as

the compound was expected to lose its effect upon ERa knockdown. In order to

address clinical relevance, we attempted to mimic knockdown effect using a

combination of SERD and SERM. MCF7S cell were subjected to daily dosing with fresh

media containing various concentrations and combinations of antiestrogens. The data

summarized in Figure 5-5 poses several points for further considerations. First, it

indicated that pharmacologically induced ERa degradation was insufficient to promote

the level of cell death seen in 4-OHT treated shER cells. Second, continuous high dose

4-OHT exposure was required to prompt a similar decrease in cell number as 4-OHT

treated shER MCF7S. Third, there were significant additive effects with combinatorial

treatments. Finally, the response was dose dependent. In general, it would require a









constant, cytotoxic level of antiestrogens to achieve the same effect as a single high

dose of SERM in shER cells. This would suggest ERa saturation can inhibit MCF7S

proliferation.

Discussion

ERa is proving to be a paradoxical factor in MCF7S cells. It is not required for

sphere formation and maintenance; however its concentration in the cell has vast

impact on cell proliferation. One possible explanation is the lower concentration of ERa

allowed a single high dose of 4-OHT to fully saturate all available receptors in the bulk

shER population, and so increase the appearance of drug potency. This suggests that

recurrence in some cases of hormone-receptor positive tumor are not due to resistance,

but because high receptor levels prevented optimal inhibition by ERa antagonists. This

resulted in cells evading antiestrogen effects that theoretically were still responsive to

treatment, thus resulting in recurrence and disease advancement. Standard

antiestrogen treatments are not optimized for such heterogeneity that naturally exists.

Rather, lower doses are preferred due to side effects that disrupt patient's quality of life.

Another possibility to consider is 4-OHT binds alternative targets as the

concentration of its preferred binding partner decreased. It is conceivable that the high

4-OHT dosage antagonized not only ERa, but also saturated other targets such as

Estrogen-Related Receptor beta (ERR3) and gamma (ERRy) (Coward et al., 2001;

Tremblay et al., 2001). Additional targets for 4-OHT binding may exist, as it is generally

true for pharmacological agents. It has been suggested that 4-OHT mediated apoptosis

through both ERa dependent and independent pathways such as oxidative stress and

activation of stress kinases (Obrero et al., 2002). The present study provided evidence









that mammosphere formation and potential TIC enrichment respond to 4-OHT in a

related manner.












shRN l Clones
MCF7S 7 9 11


ta-ER

a-tubulin

Rel. Intensity


n


1 0.13 0.06 0.04


Figure 5-1. Knockdown of ERa (shER) in MCF7S cells. A) Immunoblot comparing
ERa levels of three independent shRNA clones with bulk MCF7S culture.
Relative densitometry reduction for shRNA clones were shown as relative to
MCF7 sample. B) Growth curve of shER clones compared to bulk MCF7S
culture averaged from two independent experiments.


shER Clone Proliferation (n=2)

1000000
900000
800000
700000 --sEC
E 600000 -- MCF7S
c. 500000 -- -f-shER C7
a 400000 -
300000 C
200000- shER C11
100000
0 ,
0 2 Days 4 6
Days











2.5 pM 4-C


MCF7S. .
W. .. .-*.
0 .. '


)HT


1 pl


a.
I i *
*0
V*


M IC


.5

3 ;'


0 t,

1. -
L .. '. '*
,. .' ,, 4' .e

<. f

L 1 *' i .::a Ai, .
./


Figure 5-2. Representative phase-contrast images of MCF7S, control shRNA MCF7S,
and shER clone 7 exposed to antiestrogens for 6 days. 0.1% DMSO did not
affect sphere formation phenotype for all three cell groups. 4-OHT treatment
inhibited sphere formation MCF7S and control shRNA cells. The shER C7
cells responded to 4-OHT with increase in cell debris, but surviving cells
formed tightly packed spheroids. ICI treated cells all retained sphere
formation phenotype. Abbreviations: 4-OHT is 4-hydroxytamoxifen. ICI is
pure antiestrogen ICI 182780. Scale bar = 182 microns.























94


0.1% DMSO


Control
shRNA








shER C7































B
120 -

100

80

60

a 40

20


0.1% 2.5uM 4- luM ICI 0.1% 2.5uM4- luMICI
DMSO OHT DMSO OHT
n=3
'p<0.05 shER-C9 control shRNA


S 120

100

80

S60

40

20

0
0.1%DMSO 2.5uM4- luM ICI 0.1%DMSO 2.5uM4- luM ICI
OHT OHT

n=2
shER-C11 control shRNA



Figure 5-3. Antiestrogen response of ERa knockdown in MCF7S cells. Data for clone 7
(A) and clone 9 (B) were generated from three independent experiments.
Data for clone 11 (C) were derived from two independent experiments. 4-
OHT is 4-hydroxytamoxifen. ICI is pure antiestrogen ICI 182780. Error bars
were expressed as standard error of the mean (S.E.M.). Statistical analysis
was performed using paired Student's t-test.


120 *
100-

W 80 -
60-
40 -
20 -
0
0.1% 2.5uM4- luMICI 0.1% 2.5uN
DMSO OHT DMSO OH]

*p<0.05
3 shER-C7 control s
n=3















25
*
20










0.1% 2.5uM 4 luM IC 0.1% 2.5uM 4- luM ICl
DMSO OHT DMSO OHT
*p<0.05
shER-C7 control shRNA
B 15
10








0 --
ST Total














K U >50um



0.1% 2.5uM4- luM CI 0.1% 2.5uM4- luMICI
DMSO OHT DMSO OHT
*p


















n=3
*p 18
16
14









O 8
O6 m Total







2
0 -














0.1% 2.5uM4- lu ICI 0.1% 2.5uM4- luMICI
DMSO OHT DMSO OHT
n=3









shER-C9 control shRNA













Figure 5-4. Sphere formation frequency of antiestrogen treated ERa knockdown
MCF7S cells. Three independent experiments were performed for clones 7
(A) and 9 (B). Clone 11 (C) experiments were repeated twice. Abbreviations:
4-OHT is 4-hydroxytamoxifen. ICl is pure antiestrogen ICl 182780. Error

bars were expressed as standard error of the mean (S.E.M.). Statistical
analysis was performed using paired Student's t-test.










120%

100%

80%



6 40% -

20% -

20%
0.1%DMSO lm 4-OHT 2.5pM 1pM ICI 1PM ICI 1pM ICl
4-OHT +1pM 40HT +2.5pM
'p<0.05 40HT
n=3

Figure 5-5. SERDs and SERMs combination treatment to mimic shER effects in
MCF7S. Media containing the indicated antiestrogens were administered
daily for 6 days. The data was derived from three independent experiments.
Abbreviations: 4-OHT is 4-hydroxytamoxifen. ICI is pure antiestrogen ICI
182780. All antiestrogen treated MCF7S showed significantly decreased
viable cells compared to 0.1% DMSO treated cells. Error bars were
expressed as standard error of the mean (S.E.M.). Statistical analysis was
performed using paired Student's t-test.









CHAPTER 6
CONCLUSIONS AND FUTURE DIRECTIONS

Conclusions and Discussion

The CSC/TIC hypothesis captured the imagination of cancer biologists by

proposing TICs as a recognizable target for the study of tumor recurrence. It has been

generally presumed the TICs did not express hormone receptors. This assumption is

colored by a strict interpretation of mammary stem cell hierarchy in which MaSCs

belonged to the basal subtype while hormone sensitive cells were grouped in the

luminal category (Smalley et al., 2003; Villadsen, 2005). This assertion has limited the

scope of cancer biology by superimposing one discipline onto another. Emerging

studies are probing the role of hormone receptor positive progenitor cells and hormone

response in hormone receptor negative stem cells (Lim et al., 2009; Asselin-Labat et al.,

2010; Clarke et al., 2005; Booth et al., 2006; Raouf et al., 2008), with the expectation to

define the function of hormones and hormone receptors during development. There

may emerge some common links that will enhance oncogenic studies.

In this study, evidence was presented for the role of ERa as a mediator of

antiestrogen effects in the context of MCF7-derived mammospheres. MCF7S was

shown to be phenotypically distinct from MCF7P and express higher levels of putative

tumorigenic marker CD44. ERa expression and endogenous protein levels were not

significantly difference between the two cell groups. The receptors can activate known

estrogen response gene TFF1 in both cell types, although gene expression is highly

variable in MCF7S cells. The antiestrogen response in MCF7S was significantly

different from MCF7P as MCF7S displayed dose dependent growth inhibition upon

acute antiestrogen exposure, while MCF7P showed minimal sensitivity to antiestrogens.









However, the sphere formation frequency in MCF7S was unperturbed following drug

removal. Profound differences in response to SERM versus SERD were discernable in

MCF7S culture. The long term growth kinetics were strongly affected by SERD

treatments, while that of SERM was not significantly different from control and may

contribute to increased sphere formation frequency after antiestrogen challenge.

Mammosphere formation and maintenance was shown to be ERa independent using

pharmacological and molecular means. However, ERa knockdown clones showed

vastly different antiestrogen response. shER cells became resistant to ICI treatment

and sphere formation frequency was unaffected, as expected. Surprisingly, the

knockdown cells became highly sensitive to 4-OHT induced growth inhibition and

sphere formation frequency was drastically reduced. These two observations

suggested bulk MCF7S culture is composed of a mixed population with ERa

potentiating divergent proliferative pathways. Additional pharmacological inhibition data

in Figure 5-5 suggested there were ERa positive MCF7S subclones that were only

partially inhibited by the antagonist 4-OHT. One explanation may be the receptor's

concentration is particularly high in that fraction and a single dose of 4-OHT was

insufficient as some active receptors remain and signal cell growth. Another possibility

is that 4-OHT has non-ERa targets that also regulate MCF7S proliferation. This is not

unexpected as 4-OHT resistant subclones may have alternate proliferation signaling

pathways that can compensate for ERa inhibition.

The idea that alternative mitogenic pathways alter antiestrogen response and

allow for anchorage-independent growth is not new. Previous studies have been done

where MCF7 was maintained under long-term steroid deprivation (Martin et al., 2003,









2005; Chan et al., 2002). It was noted that MAPK/ERK1/ERK2 and P13K activity are

upregulated under these conditions and contributed to hormone-independent growth,

while the cells remained antiestrogen responsive. Considering the fact that

mammosphere culture is maintained in growth factor supplemented media, it is possible

that mitogenic pathways upregulated under these conditions contributed to the

observed MCF7S antiestrogen response. MAPK pathway upregulation would also allow

for MCF7 to grow as anchorage-independent spheroids (Fukazawa et al., 2002;

Thottassery et al., 2004). Recent studies have shown prolong ERK activation was

required to mediate 4-OHT may induced cell death, and effects of ERK phosphorylation

can be mitigated by ERa activation (Zheng et al., 2007; Zhou et al., 2007). If true, this

pathway would account for 4-OHT hypersensitivity observed in shER MCF7S. A

proposed model would be ERK activation via growth factor stimulation in shER MCF7S

combined with 4-OHT induce apoptosis through ERa independent mechanism (Obrero

et al., 2002) result in uncontrolled cell death as ERa was taken out of a signaling loop

that may regulate ERK activation (Hutcheson et al., 2003; Britton et al., 2006). These

signaling pathways would undoubtedly affect downstream gene expression and amplify

effects at the genomic level.

An additional consideration is how key tumorigenic characteristics may drive

mammosphere formation. Of the six key features designated as "hallmarks" of cancer

(Hanahan et al., 2000), four can be applied to mammosphere formation in MCF7S cells.

They are self-sufficiency in growth signals, insensitivity to antigrowth signals, evading

apoptosis, and limitless replicative potential (Hanahan et al., 2000). MCF7S cells were

able to form tightly pack spheroid cultured in a rudimentary media under suspension


100









conditions, and displayed indefinite self-renewal capacity. While many establish breast

cancer cells lines shared the above characteristics under normal culture conditions, only

MCF7 was able to survive in mammosphere culture (Chapter 1). It would be of general

interest to probe for intrinsic cellular differences that may account for this phenomenon,

and establish which of the aforementioned key features significantly contribute to

mammosphere formation.

There are major gaps in knowledge regarding cellular response to hormones, and

the role of steroid hormone receptors in cellular signaling under different cellular

context. Of course, it is a daunting task to catalogue all possible cell signaling

combinations. However, it may be worth the effort for it provides the basis for disease-

relevant gene regulation studies.

Mammosphere culture provided a model for further characterization of tumor

recurrence. However, it is uncertain if mammosphere cells may be accurately termed

CSCs/TICs. It is probable that there is no definable population to target, and tumor

recurrence depend upon a mixed subclone population with the evolutionally advantage

to survive. The current experimental approach of isolating specific subgroup may aid

mechanistic studies; but it is inadequate in providing a complete picture of tumor

development. As evident in this study, there are heterogeneous subgroups that can be

enriched under mammosphere forming conditions that respond markedly different from

adherent culture. It appears adherent culture tends to homogenized cell population as it

provides a rich environment for the survival of most cells, rather than selecting for the

hardier portion that are most likely to be malignant. Nevertheless, this portion may still


101









contain antiestrogen responsive cells that can be eliminated if treatment regiments are

optimized.

Future Directions

In vivo tumorigenic studies are ongoing to confirm MCF7S tumorigenicity. If in

vitro experiments are translatable, it is expected that MCF7S will be more tumorigenic

than MCF7P. Antiestrogen-treated MCF7S cells will be expected to have similar

tumorigenic frequencies, and 4-OHT treated shER clones will have the lowest

frequency. Of course, there are reasons to suppose in vivo data will not match those

from in vitro studies. The injected cells will receive estrogen supplement by means of

an implanted pellet in the mouse, and the cells will no longer receive growth factors.

Therefore, it is unknown if MCF7S will revert to MCF7P phenotype under these

conditions and display no differences in tumorigenic ability. There is the additional

concern that the microenvironment at the injection site would alter cellular signaling that

translates into confusing results where no conclusions may be drawn.

Cytoplasmic signaling in MCF7S is the next logical area of investigation. A rich

source of information concerning mitogenic and estrogen receptor crosstalk can be

gathered with further study, which can translate to a deeper understanding into both

genomic and non-genomic ERa functions. Of specific interest is ERK phosphorylation

as it appears to be an important branch point in cytoplasmic and ligand-independent

ERa functions that may mediate antiestrogen response.

Finally, a multi-pronged approach has long been advocated as the most efficient

means of cancer control. However, such a rigorous treatment regimen may be

prohibitive for patients both in terms of cost, time investment, and quality of life.


102









Therefore, recent efforts towards optimizing the potency and the innovative usage of

currently available drugs are vital for the future of cancer treatment. This study has

provided evidence for extending SERM treatment beyond the current recommendation

of 5 years. The results from this study have shown the tumor initiating potential of

estrogen receptor positive cells were unaffected by antiestrogen treatments (Chapter 4).

However, the evidence also suggested antiestrogens can inhibit proliferation and

potentially suppress tumor recurrence (Chapter 4 and 5). The data also suggested 4-

hydroxytamoxifen may have additional non-ERa targets that warrant further study

(Chapter 5). An important question remains: whether antiestrogen resistant cells and

TICs are the same. The tools currently available are insufficient to provide solid

experimental proof. However, the enthusiasm generated in the area will surely drive

future technical improvements.


103









APPENDIX A
SUPPLEMENTAL FIGURES

Chapter 3


S-



o


Side Scatter


Figure A-1. Additional figures comparing CD44-expression in MCF7P and MCF7S. A)
Scatter plot and gating of FACS sorted cells. B) Overlay histogram of CD44-
FITC and Iso-FITC staining for MCF7P and MCF7S. Technical duplicates
were shown for the experiment. Isotype staining control (Iso-FITC) showed
negligible staining for both sample groups. MCF7P cells showed an
averaged 21.4% CD44-positive staining. MCF7S cells showed an averaged
57.97% CD44-positive staining. Percentages were calculated from 30000
cells.


Table A-1. Additional data for CD44-FITC signal quantification using flow cytometer
(Figure A-1).
Background % Total (Not
Sample Gated Events Subtracted % Gated ga
Eventsgated)
MCF7P Iso 29366 394 1.34 1.31
MCF7P CD44-1 29403 5192 17.66 17.31
MCF7P CD44-2 29378 7486 25.48 24.95
MCF7S Iso 29317 59 0.2 0.2
MCF7S CD44-1 29333 17283 58.92 57.61
MCF7S CD44-2 29205 16654 55.51 56.49


104












Table A-1 data represents additional experiment to evaluate CD44 expression in
MCF7P and MCF7S. A total of 30000 cells were counted for each sample. Gated
events represent cell fraction used for analysis (R1) based on forward and side scatter
pattern as shown in Figure A-1A. Background subtracted events represented CD44-
FITC signal levels above Iso-FITC control. % Gated is percentage ratio of gated
events/ background subtracted events. % Total is percentage ratio of gated events /
30000 total events.



Chapter 4


Figure A-2. Real time RT-PCR of MCF7P for estrogen response gene TFF1. A) Time
course of TFF1 expression induced with 100 nM 3-estrodiol (E2). B) and C)
Inhibition of E2 induced TFF1 expression after 1 hour antiestrogens pre-
treatment. All TFF1 expression normalized to beta-actin expression.
Abbreviations: 4-OHT is 4-hydroxytamoxifen. ICI is ICI 182780. All TFF1
expression was normalized to beta-actin expression. Error bars represented
technical triplicates. Each graph is n=1.


105


140

. 120

100


E
- 60

S40

20


1 2 4 8 12 24
Hours E2 Induction


14
o 12
10
g 8
6
4
2

Untreated E2 E2+40HT E2+ICI
12 Hours


Za




1


0.3

0
Untreated E2 E2+40HT E2+1CI
24 Hours












































10

S8

o 6
E

4


















z
2

0
Untreated E2 E2+40HT E2+ICI
12 Hours





6






c 3

2
6~~ ~ ~~ ---------------------------------









Untreated E2 E2+40HT E2+ICI
24 Hours


2 -


1.6 -
14
12
1
c 08
0.6
04

0 -
0


Untreated


E2 E2+40HT E2+1CI
24 Hours


Figure A-3. Real time RT-PCR of MCF7S for estrogen response gene TFF1. A) Time

course of TFF1 expression induced with 100 nM 173-estradiol (E2), TFF1

expression levels were shown relative to untreated cells. B) TFF1 expression


106


45

40

35
S30

z 25

20

15-

10


0
1 12 24
Hours E2 Induction


1.4

o 12







0.4

02

0

Untreated E2 E2+40HT E2+CI
12Hours


1.4

5 12

1



0.6

S0.4

0.2



Untreated E2 E2+40HT E2+ICI
24 Hours










after 12 hours induction with 100 nM 173-estradiol (E2) following 1 hour
antiestrogen pre-treatment. C) TFF1 expression after 24 hours induction with
100 nM 173-estradiol (E2) following 1 hour antiestrogen pre-treatment.
Abbreviations: 4-OHT is 4-hydroxytamoxifen. ICI is ICI 182780. All TFF1
expression was normalized to beta-actin expression. Error bars represented
technical triplicates. Each graph is n=1.


Figure A-4. Time course of CTSD gene expression induced with 100 nM 173-estradiol
(E2). CTSD expression levels were shown relative to untreated cells, and
normalized to beta actin expression. Error bars represented technical
triplicates. The graph is n=1.


107


CTSD (100 nM E2)
1.4
1.2
N 1
4 0.8
4 0.6
r 0.4
0.2
0
MCF7S+E2 MCF7S+E2 MCF7S+E2 MCF7S+E2 MCF7S+E2 MCF7S+E2
lh 4h 8h 12h 24h 48h












A +BrdU
o AAD9009.004 AA909099.004





3. "





0 200 400 600 800 1000
FL2-W




0 200 400 600 800 1000
FL2-A


B -BrdU

a AA090909.001 __AA909009.001












200 400 600 800 1000 .
...o
CI'

o ..I
o I I,

0 200 400F.00 800 1000




S0 200 400 600 800 1000
FL2-A



















108











AA09009.005


0 200 400 600 800 1000
FL2-A


2.5
AA090909.006










0 200 400 600 800 1000
FL2-W


pM 40HT
o0---------


AA0090909006


. a .. '4b .. 6k .. S ..1
200 400 600 800 1000
FL2-A


Figure A-5. BrdU/PI cell cycle analysis for MCF7S treated with antiestrogens for 72
hours. A) 0.1% DMSO treated control pulsed with 10piM BrdU for 2.5 hours.
Total events were shown as R2 fraction in scatter plot on the left. B) No BrdU
pulse control MCF7S cells. Gated events were shown in scatter plot on the
left. C) 1 pM ICI treated MCF7S cells pulsed with 10p M BrdU for 2.5 hours.
Gated events were shown in scatter plot on the left. D) 2.5 piM 4-OHT treated
MCF7S cells pulsed with 10piM BrdU for 2.5 hours. Gated events were
shown in scatter plot on the left. All samples were analyzed by flow
cytometer. Abbreviations: 4-OHT is 4-hydroxytamoxifen. ICI is pure
antiestrogen ICI 182780. 30000 cells were counted for each sample. Data
represents n=1.


109


1iM ICI
0.--


FL2-W


D







W-

0.









Table A-2. Data for BrdU/PI cell cycle analysis MCF7S
antiestroqens (Figure A-5).


cells treated 72 hours with


Sample Gated Background % Gated % Total (Not
Events (S Subtracted gated)
Phase) Events
-BrdU 17 27495 0.06 0.06
0.1% DMSO +BrdU 10474 28417 36.86 34.91
2.5 piM 4-OHT +BrdU 7323 28056 26.1 24.41
1 piM ICI +BrdU 7652 27844 27.48 25.51


Table A-2 data represents experiment data for BrdU/PI cell cycle analysis in MCF7S
cells treated with antiestrogens for 72 hours (Figure A-5). A total of 30000 cells were
counted for each sample. Gated events represent BrdU incorporation during S phase of
the cell cycle (Figure A-5 right). Background subtracted events represented cell fraction
(R2) deemed unicellular based on scatter plot gating (Figure A-5 left). % Gated is
percentage ratio of gated events / background subtracted events. % Total is
percentage ratio of gated events / 30000 total events.


110












60

50

40
0
M 30

20

10

0
0.1% DMSO 2.5uM luMICI 0.1%DMSO 2.5uM luM ICI
40HT 40HT

48h 72h


50
45
40
35
30
Q)25
, 20
15
10
5


.1%DMSO 25uM lu IC 0.1%DMSO 2.5uM luMIC
40HT 40HT

48h 72h


25

20

15

~e 10

5

0
0.1% DMSO 2.5uM luM ICI 0.1% DMSO 2.5uM luM IC
40HT 40HT

48h 72h


Figure A-6. Individual Annexin V experimental data for antiestrogen treated MCF7S at
48 hours and 72 hours summarized in Figure 5-5. Each graph represented
one experiment (n=l). Total percentage of apoptosis was shown.
Abbreviations: 4-OHT is 4-hydroxytamoxifen. ICI is pure antiestrogen ICI
182780.


111










Chapter 5


Cnntrnl chRNA


Figure A-7. Representative phase-contrast images of shER C7 knockdown (left) and
control shRNA (right) mammospheres. There were no observable
differences in sphere size or formation frequency between the two cell types.
Scale bars = 182 microns


112









APPENDIX B
BRD8 FUNCTION DURING EARLY CARDIAC DEVELOPMENT

Introduction

Congenital heart defects are common conditions that affect a large portion of the

population. Although medical advances have greatly extended the life of affected

individuals, the underlying causes of these conditions are not fully characterized.

Genetic and molecular approaches have identified a number of key regulators that form

complex signaling networks directing each stage of heart development. In addition,

there is emerging evidence for epigenetic regulation of developmental pathways. This

includes the possible "priming" of promoter regions for activation via incorporation of

histone variants by SWR1 chromatin remodeling complexes. To this end, we propose a

SWR1 subunit, Brd8, as a novel protein involved in cardiac development. The goal of

the project is to gain a better understanding of Brd8's role in cardiogenesis by studying

the morphological characteristics of Brd8 mutant mice, identifying specific genes

affected in Brd8 mutants, and investigating the mechanism by which Brd8 affects

downstream gene expression.

Overview of Mammalian Heart Development

Major events in heart development are shown in the Figure 1, as reviewed in

(Srivastava, 2006). Cardiac progenitor cells from the mesoderm form the primary and

secondary heart field in the cardiogenic region. The first heart field (FHF) is localized

along the anterior part of the embryos, while the second heart field (SHF) is located

posterior to the FHF. The cardiogenic region has undergone patterning and a ventral

midline has been established, which guides the formation of the primitive heart tube

from cells in the FHF. The first heart field differentiates into two symmetrical regions


113









that extend from the midline; this structure is known as the cardiac crescent. The

crescent has specific cell populations that will form the future left ventricle (LV) and atria

(A) of the heart. The two halves of the crescent later converge along the midline. This

forms the primitive heart tube, containing two distinct cell populations destined to be the

ventricle (V) or the two halves of the atria (RA and LA). The SHF differentiates at this

stage, and cells from the SHF migrate into both ends of the heart tube forming the

future right ventricle (RV), the outflow tract or conotruncus (CT), and parts of the atria.

The SHF also supports a population of non-mesodermal cells called the neural crest,

which contribute to the septation of the heart and valve formation in late heart

development. The heart tube then undergoes rightward looping, segments of the aortic

arches are specified, and the chambers are formed by the asymmetric proliferation of

various regions. Septation, valve formation, and maturation of the aortic arches

complete heart development (Srivastava, 2006).

Transcription Factors Regulating Heart Formation

Specification of Myocardial Progenitor Cells

All tissue types originate from the epiblast, derived from the inner cell mass of the

post-implantation blastocyst. During gastrulation (E6.5 in mouse), the primitive streak

forms on the posterior region of the embryo, the primary germ layers (ectoderm,

endoderm, mesoderm) are defined and enter the primitive streak (Tam and Loebel,

2007). The earliest mesodermal genes are MesP1/2 and Fgf8, which are expressed

during gastrulation (Abu-lssa and Kirby, 2007; Brand, 2003). Induction of Nkx2.5

expression by upstream signaling induces differentiation of mesoderm germ cells into

cardiomyocyte precursors (Xu and Baldini, 2007; Srivastava and Olson, 2000). These


114









upstream signals include members of the Bone Morphogenetic Protein family (BMP)

(Prall et al., 2007), as well as Sonic Hedgehog (Shh), Fibroblast Growth Factor (FGF),

Wnt and Notch proteins (Srivastava, 2006; Kimelman, 2006; Dunwoodie, 2007).

Differentiation of Precursors into Cardiomyocytes

A number of other induction signals trigger differentiation of precursors into mature

cardiomyocytes, and further specification as the first and second heart fields. The first

heart field (FHF) is characterized by activators such as GATA4 (Pu et al., 2004; Xin et

al., 2006), which interacts with Nkx2.5 (Pu et al., 2004; Tanaka et al., 1999) and other

downstream regulatory factors during chamber formation. Cells in the secondary heart

field (SHF) differentiate at a later stage. They are first marked by the expression of Isll

(Dodou et al., 2004; Park et al., 2006) and Foxhl (Xu et al., 2007; von Both et al.,

2004). The upstream factors in both heart fields interact with a multitude of downstream

targets. These interactions are important for specifying polarity and asymmetry,

organization of multiple cell types, inducing proliferation and formation of defined

cardiac structures (Dunwoodie, 2007).

Downstream Transcriptional Networks Associated With Heart Formation

The plethora of downstream transcriptional targets overlap and cross-regulate,

creating a complex regulatory network. Many families of transcription factors contribute

to the development of specific cardiac structure. Each family contains numerous

isoforms, giving rise to an enormous variety of combinational signaling. Their

interactions as activating or repressive complexes direct and fine tune each step in the

development of a four-chamber heart from the two heart fields. This complex network

has been intensely studied and a few are well-characterized. Some examples are the

Tbx family (Plageman and Yutzey, 2005; Stennard and Harvey, 2005; Naiche et al.,


115









2005), Gata-4,-5,-6 (Pu et al., 2004; Patient and McGhee, 2002), Mef2 family

(Karamboulas et al., 2006; Black and Olson, 1998), HANDs (Risebro et al., 2006; Firulli,

2003), and certain Nkx2 (Tanaka et al., 1999; Heathcote et al., 2005; Pabst et al.,

2000).

General Introduction to Brd8

Brd8 protein was originally identified as a coactivator for thyroid receptors and

retinoic acid receptors (Monden et al., 1997, 1999). It contains two bromo domains,

which are generally associated with acetylated-histone binding. Brd8 is a subunit of

various chromatin remodeling complexes, including HAT complexes and histone

exchange complexes such as Domino (Raisner and Madhani, 2006; Cai et al., 2003). It

has also been suggested to be a cell differentiation switch (Benevolenskaya et al.,

2005). Tentative evidence suggests that Brd8 is a downstream target of Tbx5 (Mori et

al., 2006), a key regulator in cardiomyocyte differentiation and chamber formation.

Taken together these data suggest that Brd8 can be mediating differentiation signals

induced by nuclear receptors and thereby affecting downstream gene regulation by

interacting with trans-acting cofactors. Brd8 may also be organ-specific as its

expression is controlled by organ-specific transcription factors.

Materials and Methods

Methods not previously described in Chapter 2 are described below.

Immunohistochemistry

Immunohistochemistry was performed as follows, the embryos are dissected away

from maternal tissue. They were then rinsed in PBS pH 7.3 (137mM NaCI, 2.7mM KCI,

4.3mM Na2HPO4-7H20, 1.4mM KH2PO4) and fixed overnight with 4% paraformaldehyde

(PFA) (w/v) at 4C. Next day, they were washed twice in PBS, dehydrated in a


116









methanol series and bleached with 5% hydrogen peroxide (v/v) in methanol for 2 hours

to neutralize endogenous peroxidase activity. They were rehydrated in a methanol

series to PBS, blocked with 1.5% sheep serum in PBS, and incubated with primary

antibody overnight at 4C. Samples were washed extensively in PBS, and HRP-

conjugated secondary antibody applied overnight at 4C. They were washed as

previously described after primary antibody incubation. The samples were incubated

with HRP-conjugated secondary antibody for 1 hour at room temperature. Signal was

detected using (3,3'-diaminobenzidine tetrahydrochloride) DAB Plus substrate kit

(Zymed). Once sufficient signal was achieved, samples were washed in PBS, postfixed

in 4% PFA and 0.1% glutaraldehyde (w/v) in PBS, washed again in PBS, cleared in

50% and 70% glycerol in PBS. The samples were stored at 4C until images were

captured.

Whole Mount In Situ Hybridization (WHISH)

Riboprobe synthesis

Riboprobes were generated from plasmids containing the cDNA gene of interest

and bacteriophage T7/T3 promoters. Digoxigenin (DIG)-labeled riboprobes were

synthesized from 10 pg linearized DNA template, 2 pl DIG RNA labeling mix (Roche), 4

pl of 5X transcription buffer, 100 mM DTT, 1 pl RNase inhibitor, and 1 pl of appropriate

RNA polymerase to synthesis antisense/sense strand in a 20 pl reaction. The reaction

was incubated at 37C for 2 hours, followed by DNase treatment. The reaction was

stopped by the addition of 2 pl 0.5M EDTA pH 8. The DIG-labeled riboprobe was

pelleted via ethanol precipitation with the aid of LiCI. The pellet was resuspended in

DEPC-treated water and stored at -20C until hybridization.


117









Mouse embryo preparation

Dissected mouse embryos at desired time points were treated as follows prior to

hybridization: fixed in 4% PFA at room-temperature (RT) for 20 minutes, and then

dehydrated through a methanol series and stored at -80C until hybridation.

Whole mount in situ hybridization

In-Situ hybridization was performed as follows: embryos were rehydrated through

methanol series to PBT (PBS+0.1% Tween-20), bleached in 6% hydrogen peroxide for

1 hour, washed in PBT, treated with 10 pg/ml Proteinase K for 6-8 minutes, washed in 2

mg/ml glycine for 10 minutes, and rinsed in PBT. The samples were then postfixed in

4% PFA/0.2% glutaraldehyde (w/v) for 20 minutes, blocked in prehybridation solution

(50% formamide, 5X SSC, 0,1% Tween-20, 0.1% SDS, 50 pig/ml heparine, 50 pig/ml t-

RNA, 60mM citric acid) at 70C for 1 hour, and hybridized overnight at 70C in fresh

prehybridization buffer containing 0.2 pg/ml DIG-labeled probe. On the next day,

samples were extensively washed at 65C with wash solutions (50% formamide, 5X

SSC, 60-24mM citric acid, 1-0.2% SDS, and 0.1% Tween-20), blocked in 2%

Boehringer blocking reagent (w/v) and 10% sheep serum (v/v) dissolved in Maleic Acid

Buffer (MAB). The samples were then incubated overnight in 1:2000 pre-blocked, AP-

conjugated anti-DIG antibody at 4C. On the following day, samples were washed at

length with MAB, with a final overnight wash in MAB at 4C. Finally, samples were

washed in alkaline phosphatases buffer (NTMT) and the signal was detected using BM

Purple AP Substrate (Roche). Once sufficient signal was obtained, samples were

washed in PBT, postfixed in 4% PFA/ 0.1% glutaraldehyde for 1 hour, washed again in

PBT, and stored at 4C until images of the stained embryos are photographed.


118









RT-PCR Primers for Brd8 Allele Expression

Primers used to confirm gene trap vector insertion is illustrated in Figure B-1 B.

The sequences were as follows: exon 1 (Brd-1: 5'-AGGAGTGGGGATCAGAACTG-3'),

beta-geo of gene-trap insert (Beta-geo: 5'-GTATCGGCCTCAGGAAGATCG-3'), and

exon 5 (Brd-ex-5: 5-CTCAGCAGTCAGTTTGCGAAC-3'). Primer set Brd-1 and Brd-ex-

5 amplified the wild type allele, while primer set Brd-1 and Beta-geo amplified the

mutant allele. Brd8"' ES cell cDNA (XE487) were reversely transcribed using beta-geo

specific primers, and was included to show the expected size of mutant bands.

Sample cDNA were obtained from E10.5 embryos following homogenization in

Trizol to extract total RNA, and RT-PCR as described in Chapter 2.

Results and Discussion

The initial evaluation of Brd8 gene-trapped whole mouse embryos included gross

morphological characterization and assessing Brd8 allelic expression level. The Brd8

mutant mouse line demonstrated obvious embryonic phenotype such as developmental

retardation and lethality by E10.5 (Figure B-2A). Semi-quantitative RT-PCR showed the

Brd8 allele was not completely depleted in mutants, but low residual expression

remained in both ES cell clone XE487 and Brd8 homozygous mutant embryos at E10.5

(Figure B-2B and Figure B-2C). Endogenous Brd8 expression in wild type E10.5

embryos was assessed using whole mount in situ hybridization to determine Brd8

localization. As shown in Figure B-3, endogenous Brd8 was ubiquitously expressed

throughout the embryos with the exception of the cardiac region. The evidence

suggested Brd8 was an important cofactor during embryogenesis. However, the data

implied the role of Brd8 is generalized and may not be organ specific. It was possible

that the cardiac defects observed were the indirect result of generalized growth arrest.


119









The observed abnormal cardiac pathology (Figure B-4) may be attributed to the heart

being one of the earliest organs to be fully developed and thus was one of the first

organs affected by growth inhibition. The failures of the mutant embryo to turn and the

neural tube to close were additional indicators of early growth inhibition (Figure B-2A

and B-4).

The malformation of essential organs could explain the early lethality of Brd8

mutants. Further evaluation of Brd8 mutant embryos included whole mount

immunohistochemistry using PECAM antibody to compare vasculature development of

Brd8 mutant embryos and stage matched wild type embryos. As shown in Figure B-5,

Brd8 mutant embryos showed less robust PECAM signal in the neural, cardiac and

dorsal regions. The failure of the heart tube to undergo normal rightward looping was

evident in the Brd8 mutant embryo (Figure B-5B and 5C). The mutant embryo

displayed an enlargement of the ventricular regions of the heart tube, which suggested

that cardiac patterning occurred independently of positional cues. It also implied that

cells in the heart tube were already programmed to become specific cardiac

components very early in development.

The Brd8 mutant embryos showed cardiac-specific morphological defects such as

cardiac edema, abnormal ventricular enlargement and insufficient rightward looping of

the heart tube. The expression levels and localization of specific cardiac differentiation

markers were assessed by whole mount in situ hybridization to determine if the

abnormal cardiac pathology had any bearing on cardiac cell fate determination. As

shown in Figure B-6, Brd8 mutant embryonic heart tube retained a similar molecular

signature as stage matched wild type embryos. Hybridization of Tbx5 (Figure B-6A),


120









Nkx2.5 (Figure B-6B), and Mlc2v (Figure B-6C) in mutant embryos all showed a similar

expression localization and pattern to wild type embryos. However, there were obvious

ventricular enlargement and lack of rightward looping as mentioned earlier. These

results indicated that while the patterning of the heart tube was fixed during early

development, cardiac morphogenesis still required differential positional cues

throughout the process to direct proper formation. Without which, the heart was

physiologically defective and resulted in embryonic lethality.

The preliminary data provided here demonstrated a generalized, but essential role

for Brd8 during mammalian development. Even though the Brd8 allele may not be fully

silenced (Figure B-2B and 2C), it was enough to cause developmental arrest and

physiological defects that resulted in lethality. However, Brd8 did not appear to affect

early specification or differentiation of mesodermal cells to cardiomyocytes. This would

suggest the cardiac defects observed were secondary effects initialed by Brd8

downregulation. Additional cardiac molecular markers may be used to further evaluate

cardiac patterning and dissect this assumption. It is unclear how Brd8 expression may

influence embryonic development. It may be speculated that the ubiquitous expression

of Brd8 (Figure B-3) would affect multiple regulators. The deregulation of these

regulators culminated into an unsustainable combination that initiated cell death. These

factors may include Sonic Hedgehog, BMP, Wnt, Notch, or FGF. It can be of interest to

study Brd8 effects on known signaling pathways and cellular proliferation. This course

of inquiry can offer additional insight into the effects of Brd8 in mammalian

development.


121










pGTOlxf
A L2 pUC backbone
IEn2 inlrl )C=IrS7iii bPIC)--


B
Brd-1 Brd-ex-5
Wild Type Brd8

Brd-1 Beta-geo
BrdS Mutant I Bela-geo I I E
Not to scale


Figure B-1. Gene trap vector inserted into mouse genome to generate Brd8 null allele.
A) Design of gene trap vector used (SIGTR). Abbreviations: Intron is 1.5 kb
of Mouse En2 intron 1. SA is splice acceptor of mouse En2 exon 2. P-geo is
fusion of P-galactosidase and neomycin transferase. pA is SV40
polyadenylation signal. B) Diagram of gene trap vector insertion into Brd8
intron 2 and relative position of primers used to evaluate insertation. 5' primer
for both mutant and wild type is Brd-1. 3' primer for wild type Brd8 is Brd-ex-
5. 3' primer for Brd8 mutant knock-in vector is Beta-geo. Diagram is not to
scale.


122





























C








Figure B-2. Brd8 gene-trapped mouse embryos and expression of wild type and mutant
alleles. A) Bright field microscopy image of wild type (right) and mutant Brd8
(left) embryos at E9.5. Brd8 mutant embryos displayed overall growth
retardation and cardiac edema (white arrow). B) Semi-quantitative RT-PCR
to assay Brd8 allelic expression in cDNA extracted from Brd8 mutant (Brd8-'-)
and Brd8 heterozygous (Brd8 /-) E10.5 embryos. The primers were located
as described in Figure B-1B. The cDNA were amplified for 35 cycles and
performed in duplicate. Abbreviations: XE 487 is Brd8-' ES cells cDNA
control. Mutant (Mut) and wild type (VT) PCR product bands are indicated
on the right. C) Beta-actin semi-quantitative RT-PCR normalization controls
were shown for mutant Brd8 E10.5 embryos cDNA (left), and Brd8
heterozygous (Brd8 +/-) cDNA (right). PCR cycle numbers are indicated
below PCR product band.


123










5 hours


1 hour


sense


sense antisense


antisense


10 hours 23 hours











sense antisense sense antisense



Figure B-3. Endogenous Brd8 expression in wild-type E10.5 mouse embryos using
whole mount in situ hybridization (WMISH). E10.5 wild type embryos were
hybridized to Brd8 antisense probe and compared to litter matched embryos
hybridized to sense probe as control for 1, 5, 10 or 23 hours. There was no
signal for sense probe, as expected. Brd8 specific antisense probe indicated
Brd8 was expressed throughout the embryo, except for the cardiac region
(red circle).


124










wld



























.- a



*K-





j


mutant


Figure B-4. Hematoxylin and eosin stained sections of Brd8 mutant embryos (E10.5)
compared to wild-type littermates. Brd8 mutant embryo (right) displayed
general growth retardation as indicated by neural tube defect (black arrow).
showed possible signs of cardiac bifidia (red point). The respective sections
were compared to E10.5 wild type sections (left). The samples were
processed by Jim Richardson's lab at University of Texas Southwestern
Medical Center, Molecular Pathology Core. The sections were evaluated by
Dr. Hideko Kasahara at University of Florida, Department of Physiology and
Functional Genomics.


125








A PECAM E8.5 WT














B PECAM E9.5 Brd8-/-


126




























Figure B-5. PECAM whole mount immunohistochemistry comparing vascularization of
Brd8 mutant and stage matched wild-type embryos (E8.5). A) Wild type E8.5
embryo immunostain for PECAM viewed from various angles. Black arrow
denotes embryonic heart tube. The embryo showed robust vasculature
formation in neural, heart and dorsal regions. The heart tube showed normal
rightward looping, as expected at this embryonic stage. B) Brd8 mutant
(Brd8 --) E9.5 embryo immunostain for PECAM viewed from various angles.
Black arrow denotes embryonic heart tube. The embryonic heart tube
showed incomplete rightward looping. Vascularization of the mutant embryo
also appeared to be deficient in the neural and dorsal regions. However,
vascularization of the yolk sac was robust. C) Left side view and comparison
of PECAM immunostain E8.5 wild type (left) and E9.5 Brd8 mutant (right)
embryos. Black arrows indicate embryonic heart tube. The comparison
demonstrates mutant embryo retarded development in the neural, heart tube,
and dorsal regions.


127

















































Figure B-6. Whole mount in situ hybridization (WMISH) characterizing expression of
key cardiac development genes. A) E8.5 wild type compared to E10.5 Brd8
mutant embryos hybridized to Tbx5 probe. The embryos were shown from
the left (left image) and ventral angle (right image). Tbx5 is specific for
ventricle region and it is seen in the two samples. The Brd8 mutant embryos
display an enlarged heart tube. B) E8.5 wild type compared to E10.5 Brd8
mutant embryos hybridized to Nkx2.5 probe. The embryos were shown from
the left (left image) and ventral angle (right image). Nkx2.5 is specific for the


128









entire embryonic heart tube. Both embryos showed positive signal for Nkx2.5
in the heart tube region. Wild-type embryo showed stronger signal in the
inflow and outflow tract at E8.5, and weaker signal at atrial and ventricle
regions. Brd8 mutant embryo showed robust signal for the entire heart tube.
Defective rightward looping and heart tube enlargement was observed in the
mutant embryo. C) E8.5 wild type compared to E10.5 Brd8 mutant embryos
hybridized to Mlc2v probe. The embryos are shown from ventral view. Mlc2v
is specific for ventricular region of the heart tube. Both embryos showed
positive signal in the ventricular region. Brd8 mutant embryo (right) showed
abnormal enlargement of the heart tube, compared to stage matched wild
type embryo (left).


129









LIST OF REFERENCES


Abraham, B. K., Fritz, P., McClellan, M., Hauptvogel, P., Athelogou, M., and Brauch, H.
(2005). Prevalence of CD44+/CD24-/low cells in breast cancer may not be
associated with clinical outcome but may favor distant metastasis. Clin Cancer
Res 11, 1154-9.

Abu-lssa, R., and Kirby, M. L. (2007). Heart Field: From Mesoderm to Heart Tube. Annu
Rev Cell Dev Biol. Available at:
http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=
Citation&list_uids=17456019.

ACS (2009). Breast Cancer Detailed Guide.

AI-Hajj, M., Wicha, M. S., Benito-Hernandez, A., Morrison, S. J., and Clarke, M. F.
(2003). Prospective identification of tumorigenic breast cancer cells. Proceedings
of the National Academy of Sciences of the United States of America 100, 3983-
8.

Alison, M. R., Islam, S., and Lim, S. M. (2009). Number crunching in the cancer stem
cell market. Breast Cancer Res 11, 302.

Altekruse, S., Kosary, C., Krapcho, M., Krapcho, N., Aminou, R., Waldron, W., Ruhl, J.,
Howlader, N., Tatalovich, Z., Cho, H., et al. (2009). SEER Cancer Statistics
Review, 1975-2007 (National Cancer Institute).

Ariazi, E. A., Ariazi, J. L., Cordera, F., and Jordan, V. C. (2006). Estrogen receptors as
therapeutic targets in breast cancer. Current topics in medicinal chemistry 6, 181-
202.

Arnold, S. F., Obourn, J. D., Jaffe, H., and Notides, A. C. (1995). Phosphorylation of the
human estrogen receptor on tyrosine 537 in vivo and by src family tyrosine
kinases in vitro. Mol. Endocrinol 9, 24-33.

Aruffo, A., Stamenkovic, I., Melnick, M., Underhill, C. B., and Seed, B. (1990). CD44 is
the principal cell surface receptor for hyaluronate. Cell 61, 1303-1313.

Asselin-Labat, M. L., Shackleton, M., Stingl, J., Vaillant, F., Forrest, N. C., Eaves, C. J.,
Visvader, J. E., and Lindeman, G. J. (2006). Steroid hormone receptor status of
mouse mammary stem cells. Journal of the National Cancer Institute 98, 1011-4.

Asselin-Labat, M. L., Sutherland, K. D., Barker, H., Thomas, R., Shackleton, M., Forrest,
N. C., Hartley, L., Robb, L., Grosveld, F. G., van der Wees, J., et al. (2007).
Gata-3 is an essential regulator of mammary-gland morphogenesis and luminal-
cell differentiation. Nature cell biology 9, 201-9.


130









Asselin-Labat, M., Vaillant, F., Sheridan, J. M., Pal, B., Wu, D., Simpson, E. R., Yasuda,
H., Smyth, G. K., Martin, T. J., Lindeman, G. J., et al. (2010). Control of
mammary stem cell function by steroid hormone signalling. Nature advance
online publication. Available at: http://dx.doi.org/10.1038/nature09027 [Accessed
April 26, 2010].

Atanaskova, N., Keshamouni, V. G., Krueger, J. S., Schwartz, J. A., Miller, F., and
Reddy, K. B. (2002). MAP kinase/estrogen receptor cross-talk enhances
estrogen-mediated signaling and tumor growth but does not confer tamoxifen
resistance. Oncogene 21, 4000-8.

Baek, S. H., Ohgi, K. A., Rose, D. W., Koo, E. H., Glass, C. K., and Rosenfeld, M. G.
(2002). Exchange of N-CoR corepressor and Tip60 coactivator complexes links
gene expression by NF-kappaB and beta-amyloid precursor protein. Cell 110,
55-67.

Benevolenskaya, E. V., Murray, H. L., Branton, P., Young, R. A., and Kaelin, W. G.
(2005). Binding of pRB to the PHD protein RBP2 promotes cellular differentiation.
Mol Cell 18, 623-35.

Bjornstrom, L., and Sjoberg, M. (2005). Mechanisms of estrogen receptor signaling:
convergence of genomic and nongenomic actions on target genes. Molecular
endocrinology (Baltimore, Md 19, 833-42.

Black, B. L., and Olson, E. N. (1998). Transcriptional control of muscle development by
myocyte enhancer factor-2 (MEF2) proteins. Annu Rev Cell Dev Biol 14, 167-96.

Bonnet, D., and Dick, J. E. (1997). Human acute myeloid leukemia is organized as a
hierarchy that originates from a primitive hematopoietic cell. Nature medicine 3,
730-7.

Booth, B. W., and Smith, G. H. (2006). Estrogen receptor-alpha and progesterone
receptor are expressed in label-retaining mammary epithelial cells that divide
asymmetrically and retain their template DNA strands. Breast Cancer Res 8,
R49.

von Both, I., Silvestri, C., Erdemir, T., Lickert, H., Walls, J. R., Henkelman, R. M.,
Rossant, J., Harvey, R. P., Attisano, L., and Wrana, J. L. (2004). Foxhl is
essential for development of the anterior heart field. Dev Cell 7, 331-45.

Brand, T. (2003). Heart development: molecular insights into cardiac specification and
early morphogenesis. Dev Biol 258, 1-19.

Britton, D. J., Hutcheson, I. R., Knowlden, J. M., Barrow, D., Giles, M., McClelland, R.
A., Gee, J. M. W., and Nicholson, R. I. (2006). Bidirectional cross talk between
ERalpha and EGFR signalling pathways regulates tamoxifen-resistant growth.
Breast Cancer Res. Treat 96, 131-146.


131









Burdall, S., Hanby, A., Lansdown, M., and Speirs, V. (2003). Breast cancer cell lines:
friend or foe? Breast Cancer Res 5, 89-95.

Cai, Y., Jin, J., Tomomori-Sato, C., Sato, S., Sorokina, I., Parmely, T. J., Conaway, R.
C., and Conaway, J. W. (2003). Identification of new subunits of the multiprotein
mammalian TRRAP/TIP60-containing histone acetyltransferase complex. J Biol
Chem 278, 42733-6.

Campbell, L. L., and Polyak, K. (2007). Breast tumor heterogeneity: cancer stem cells or
clonal evolution? Cell cycle (Georgetown, Tex 6, 2332-8.

Campbell, R. A., Bhat-Nakshatri, P., Patel, N. M., Constantinidou, D., Ali, S., and
Nakshatri, H. (2001). Phosphatidylinositol 3-kinase/AKT-mediated activation of
estrogen receptor alpha: a new model for anti-estrogen resistance. The Journal
of biological chemistry 276, 9817-24.

Cariati, M., Naderi, A., Brown, J. P., Smalley, M. J., Pinder, S. E., Caldas, C., and
Purushotham, A. D. (2008). Alpha-6 integrin is necessary for the tumourigenicity
of a stem cell-like subpopulation within the MCF7 breast cancer cell line.
International journal of cancer 122, 298-304.

Carroll, J. S., Meyer, C. A., Song, J., Li, W., Geistlinger, T. R., Eeckhoute, J., Brodsky,
A. S., Keeton, E. K., Fertuck, K. C., Hall, G. F., et al. (2006). Genome-wide
analysis of estrogen receptor binding sites. Nat. Genet 38, 1289-1297.

Chan, C. M. W., Martin, L., Johnston, S. R. D., Ali, S., and Dowsett, M. (2002).
Molecular changes associated with the acquisition of oestrogen hypersensitivity
in MCF-7 breast cancer cells on long-term oestrogen deprivation. J. Steroid
Biochem. Mol. Biol 81, 333-341.

Chan, K. S., Espinosa, I., Chao, M., Wong, D., Ailles, L., Diehn, M., Gill, H., Presti, J.,
Chang, H. Y., van de Rijn, M., et al. (2009). Identification, molecular
characterization, clinical prognosis, and therapeutic targeting of human bladder
tumor-initiating cells. Proceedings of the National Academy of Sciences of the
United States of America 106, 14016-21.

Charafe-Jauffret, E., Ginestier, C., lovino, F., Wicinski, J., Cervera, N., Finetti, P., Hur,
M. H., Diebel, M. E., Monville, F., Dutcher, J., et al. (2009). Breast cancer cell
lines contain functional cancer stem cells with metastatic capacity and a distinct
molecular signature. Cancer research 69, 1302-13.

Chen, D., Pace, P. E., Coombes, R. C., and Ali, S. (1999). Phosphorylation of Human
Estrogen Receptor alpha by Protein Kinase A Regulates Dimerization. Mol. Cell.
Biol. 19, 1002-1015.


132









Chen, M. S., Woodward, W. A., Behbod, F., Peddibhotla, S., Alfaro, M. P., Buchholz, T.
A., and Rosen, J. M. (2007). Wnt/beta-catenin mediates radiation resistance of
Scal+ progenitors in an immortalized mammary gland cell line. Journal of Cell
Science 120, 468-477.

Clarke, R., Liu, M. C., Bouker, K. B., Gu, Z., Lee, R. Y., Zhu, Y., Skaar, T. C., Gomez,
B., O'Brien, K., Wang, Y., et al. (2003). Antiestrogen resistance in breast cancer
and the role of estrogen receptor signaling. Oncogene 22, 7316-39.

Clarke, R. B., Spence, K., Anderson, E., Howell, A., Okano, H., and Potten, C. S.
(2005). A putative human breast stem cell population is enriched for steroid
receptor-positive cells. Developmental biology 277, 443-56.

Couse, J. F., and Korach, K. S. (1999). Estrogen receptor null mice: what have we
learned and where will they lead us? Endocr. Rev 20, 358-417.

Coward, P., Lee, D., Hull, M. V., and Lehmann, J. M. (2001). 4-Hydroxytamoxifen binds
to and deactivates the estrogen-related receptor gamma. Proceedings of the
National Academy of Sciences of the United States of America 98, 8880-4.

Curtis Hewitt, S., F Couse, J., and S Korach, K. (2000). Estrogen receptor transcription
and transactivation: Estrogen receptor knockout mice what their phenotypes
reveal about mechanisms of estrogen action. Breast Cancer Res 2, 345-352.

Dignam, J., Dukic, V., Anderson, S., Mamounas, E., Wickerham, D., and Wolmark, N.
(2009). Hazard of recurrence and adjuvant treatment effects over time in lymph
node-negative breast cancer. Breast Cancer Research and Treatment 116, 595-
602.

Dodou, E., Verzi, M. P., Anderson, J. P., Xu, S., and Black, B. L. (2004). Mef2c is a
direct transcriptional target of ISL1 and GATA factors in the anterior heart field
during mouse embryonic development. Development 131, 3931-3942.

Dontu, G., Abdallah, W. M., Foley, J. M., Jackson, K. W., Clarke, M. F., Kawamura, M.
J., and Wicha, M. S. (2003). In vitro propagation and transcriptional profiling of
human mammary stem/progenitor cells. Genes & development 17, 1253-70.

Dontu, G., EI-Ashry, D., and Wicha, M. S. (2004). Breast cancer, stem/progenitor cells
and the estrogen receptor. Trends in endocrinology and metabolism: TEM 15,
193-7.

Dontu, G., Jackson, K., McNicholas, E., Kawamura, M., Abdallah, W., and Wicha, M.
(2004). Role of Notch signaling in cell-fate determination of human mammary
stem/progenitor cells. Breast Cancer Res 6, R605-R615.

Dudek, P., and Picard, D. (2008). Genomics of signaling crosstalk of estrogen receptor
alpha in breast cancer cells. PloS one 3, e1859.


133









Dunwoodie, S. L. (2007). Combinatorial signaling in the heart orchestrates cardiac
induction, lineage specification and chamber formation. Semin Cell Dev Biol 18,
54-66.

Effects of chemotherapy and hormonal therapy for early breast cancer on recurrence
and 15-year survival: an overview of the randomised trials (2005). Lancet 365,
1687-1717.

Fan, M., Yan, P. S., Hartman-Frey, C., Chen, L., Paik, H., Oyer, S. L., Salisbury, J. D.,
Cheng, A. S., Li, L., Abbosh, P. H., et al. (2006). Diverse Gene Expression and
DNA Methylation Profiles Correlate with Differential Adaptation of Breast Cancer
Cells to the Antiestrogens Tamoxifen and Fulvestrant. Cancer Res 66, 11954-
11966.

Farnie, G., Clarke, R. B., Spence, K., Pinnock, N., Brennan, K., Anderson, N. G., and
Bundred, N. J. (2007). Novel Cell Culture Technique for Primary Ductal
Carcinoma In Situ: Role of Notch and Epidermal Growth Factor Receptor
Signaling Pathways. J. Natl. Cancer Inst. 99, 616-627.

Fillmore, C., and Kuperwasser, C. (2007). Human breast cancer stem cell markers
CD44 and CD24: enriching for cells with functional properties in mice or in man?
Breast Cancer Res 9, 303.

Fillmore, C. M., and Kuperwasser, C. (2008). Human breast cancer cell lines contain
stem-like cells that self-renew, give rise to phenotypically diverse progeny and
survive chemotherapy. Breast Cancer Res 10, R25.

Firulli, A. B. (2003). A HANDful of questions: the molecular biology of the heart and
neural crest derivatives (HAND)-subclass of basic helix-loop-helix transcription
factors. Gene 312, 27-40.

Fliss, A. E., Benzeno, S., Rao, J., and Caplan, A. J. (2000). Control of estrogen receptor
ligand binding by Hsp90. The Journal of Steroid Biochemistry and Molecular
Biology 72, 223-230.

Fu, X., and Simoncini, T. (2008). Extra-nuclear signaling of estrogen receptors. IUBMB
Life 60, 502-510.

Fukazawa, H., Noguchi, K., Murakami, Y., and Uehara, Y. (2002). Mitogen-activated
protein/extracellular signal-regulated kinase kinase (MEK) inhibitors restore
anoikis sensitivity in human breast cancer cell lines with a constitutively activated
extracellular-regulated kinase (ERK) pathway. Mol. Cancer Ther 1, 303-309.

Ginestier, C., Hur, M. H., Charafe-Jauffret, E., Monville, F., Dutcher, J., Brown, M.,
Jacquemier, J., Viens, P., Kleer, C. G., Liu, S., et al. (2007). ALDH1 is a marker
of normal and malignant human mammary stem cells and a predictor of poor
clinical outcome. Cell stem cell 1, 555-67.


134









Godar, S., Ince, T. A., Bell, G. W., Feldser, D., Donaher, J. L., Bergh, J., Liu, A., Miu, K.,
Watnick, R. S., Reinhardt, F., et al. (2008). Growth-inhibitory and tumor-
suppressive functions of p53 depend on its repression of CD44 expression. Cell
134, 62-73.

Green, K. A., and Carroll, J. S. (2007). Oestrogen-receptor-mediated transcription and
the influence of co-factors and chromatin state. Nat Rev Cancer 7, 713-722.

Grimshaw, M. J., Cooper, L., Papazisis, K., Coleman, J. A., Bohnenkamp, H. R.,
Chiapero-Stanke, L., Taylor-Papadimitriou, J., and Burchell, J. M. (2008).
Mammosphere culture of metastatic breast cancer cells enriches for tumorigenic
breast cancer cells. Breast Cancer Res 10, R52.

Gupta, P. B., Onder, T. T., Jiang, G., Tao, K., Kuperwasser, C., Weinberg, R. A., and
Lander, E. S. (2009). Identification of selective inhibitors of cancer stem cells by
high-throughput screening. Cell 138, 645-59.

Hanahan, D., and Weinberg, R. A. (2000). The hallmarks of cancer. Cell 100, 57-70.

Hayashi, S., and Yamaguchi, Y. (2008). Estrogen signaling pathway and hormonal
therapy. Breast cancer (Tokyo, Japan) 15, 256-61.

Heathcote, K., Braybrook, C., Abushaban, L., Guy, M., Khetyar, M. E., Patton, M. A.,
Carter, N. D., Scambler, P. J., and Syrris, P. (2005). Common arterial trunk
associated with a homeodomain mutation of NKX2.6. Hum Mol Genet 14, 585-
93.

Howard, B. A., and Gusterson, B. A. (2000). Human breast development. J Mammary
Gland Biol Neoplasia 5, 119-137.

Hutcheson, I. R., Knowlden, J. M., Madden, T. A., Barrow, D., Gee, J. M., Wakeling, A.
E., and Nicholson, R. I. (2003). Oestrogen receptor-mediated modulation of the
EGFR/MAPK pathway in tamoxifen-resistant MCF-7 cells. Breast cancer
research and treatment 81, 81-93.

Hwang-Verslues, W. W., Kuo, W. H., Chang, P. H., Pan, C. C., Wang, H. H., Tsai, S. T.,
Jeng, Y. M., Shew, J. Y., Kung, J. T., Chen, C. H., et al. (2009). Multiple lineages
of human breast cancer stem/progenitor cells identified by profiling with stem cell
markers. PloS one 4, e8377.

Ince, T. A., Richardson, A. L., Bell, G. W., Saitoh, M., Godar, S., Karnoub, A. E.,
Iglehart, J. D., and Weinberg, R. A. (2007). Transformation of Different Human
Breast Epithelial Cell Types Leads to Distinct Tumor Phenotypes. Cancer Cell
12, 160-170.


135









Jakacka, M., Ito, M., Weiss, J., Chien, P. Y., Gehm, B. D., and Jameson, J. L. (2001).
Estrogen receptor binding to DNA is not required for its activity through the
nonclassical AP1 pathway. J. Biol. Chem 276, 13615-13621.

Jeng, M. H., Jiang, S. Y., and Jordan, V. C. (1994). Paradoxical regulation of estrogen-
dependent growth factor gene expression in estrogen receptor (ER)-negative
human breast cancer cells stably expressing ER. Cancer Lett 82, 123-128.

Jin, L., Hope, K. J., Zhai, Q., Smadja-Joffe, F., and Dick, J. E. (2006). Targeting of
CD44 eradicates human acute myeloid leukemic stem cells. Nat. Med 12, 1167-
1174.

Jones, P. S., Parrott, E., and White, I. N. (1999). Activation of transcription by estrogen
receptor alpha and beta is cell type- and promoter-dependent. The Journal of
biological chemistry 274, 32008-14.

Jordan, V. C. (2007). Chemoprevention of breast cancer with selective oestrogen-
receptor modulators. Nature reviews 7, 46-53.

Kakarala, M., Brenner, D. E., Korkaya, H., Cheng, C., Tazi, K., Ginestier, C., Liu, S.,
Dontu, G., and Wicha, M. S. (2009). Targeting breast stem cells with the cancer
preventive compounds curcumin and piperine. Breast cancer research and
treatment. Available at: internal-pdf://Spices kill BCSC Wicha 2009-
0573672704/Spices kill BCSC Wicha 2009.pdf.

Karamboulas, C., Dakubo, G. D., Liu, J., De Repentigny, Y., Yutzey, K., Wallace, V. A.,
Kothary, R., and Skerjanc, I. S. (2006). Disruption of MEF2 activity in
cardiomyoblasts inhibits cardiomyogenesis. J Cell Sci 119, 4315-21.

Keen, J. C., Yan, L., Mack, K. M., Pettit, C., Smith, D., Sharma, D., and Davidson, N. E.
(2003). A novel histone deacetylase inhibitor, scriptaid, enhances expression of
functional estrogen receptor alpha (ER) in ER negative human breast cancer
cells in combination with 5-aza 2'-deoxycytidine. Breast Cancer Res. Treat 81,
177-186.

Kelly, P. N., Dakic, A., Adams, J. M., Nutt, S. L., and Strasser, A. (2007). Tumor growth
need not be driven by rare cancer stem cells. Science (New York, N.Y 317, 337.

Kern, S. E., and Shibata, D. (2007). The fuzzy math of solid tumor stem cells: a
perspective. Cancer research 67, 8985-8.

Kimelman, D. (2006). Mesoderm induction: from caps to chips. Nat Rev Genet 7, 360-
72.


136









Kok, M., Koornstra, R. H., Margarido, T. C., Fles, R., Armstrong, N. J., Linn, S. C., Van't
Veer, L. J., and Weigelt, B. (2009). Mammosphere-derived gene set predicts
outcome in patients with ER-positive breast cancer. The Journal of pathology
218, 316-26.

Kuiper, G. G., Lemmen, J. G., Carlsson, B., Corton, J. C., Safe, S. H., van der Saag, P.
T., van der Burg, B., and Gustafsson, J. A. (1998). Interaction of estrogenic
chemicals and phytoestrogens with estrogen receptor beta. Endocrinology 139,
4252-63.

Kumar, R., Wang, R., Mazumdar, A., Talukder, A. H., Mandal, M., Yang, Z., Bagheri-
Yarmand, R., Sahin, A., Hortobagyi, G., Adam, L., et al. (2002). A naturally
occurring MTA1 variant sequesters oestrogen receptor-[alpha] in the cytoplasm.
Nature 418, 654-657.

Lacroix, M., and Leclercq, G. (2004). Relevance of breast cancer cell lines as models
for breast tumours: an update. Breast Cancer Res. Treat 83, 249-289.

LaMarca, H. L., and Rosen, J. M. (2008). Minireview: hormones and mammary cell fate-
-what will I become when I grow up? Endocrinology 149, 4317-4321.

Lannigan, D. A. (2003). Estrogen receptor phosphorylation. Steroids 68, 1-9.

Lazennec, G., and Katzenellenbogen, B. S. (1999). Expression of human estrogen
receptor using an efficient adenoviral gene delivery system is able to restore
hormone-dependent features to estrogen receptor-negative breast carcinoma
cells. Mol. Cell. Endocrinol 149, 93-105.

Lee, H., and Bai, W. (2002). Regulation of Estrogen Receptor Nuclear Export by
Ligand-lnduced and p38-Mediated Receptor Phosphorylation. Mol. Cell. Biol. 22,
5835-5845.

Lewis, J. S., and Jordan, V. C. (2005). Selective estrogen receptor modulators
(SERMs): mechanisms of anticarcinogenesis and drug resistance. Mutation
research 591, 247-63.

Li, L., Haynes, M. P., and Bender, J. R. (2003). Plasma membrane localization and
function of the estrogen receptor a variant (ER46) in human endothelial cells.
Proceedings of the National Academy of Sciences of the United States of
America 100, 4807-4812.

Li, X., Lewis, M. T., Huang, J., Gutierrez, C., Osborne, C. K., Wu, M. F., Hilsenbeck, S.
G., Pavlick, A., Zhang, X., Chamness, G. C., et al. (2008). Intrinsic resistance of
tumorigenic breast cancer cells to chemotherapy. Journal of the National Cancer
Institute 100, 672-9.


137









Lim, E., Vaillant, F., Wu, D., Forrest, N. C., Pal, B., Hart, A. H., Asselin-Labat, M. L.,
Gyorki, D. E., Ward, T., Partanen, A., et al. (2009). Aberrant luminal progenitors
as the candidate target population for basal tumor development in BRCA1
mutation carriers. Nature medicine 15, 907-13.

Lin, Z., Reierstad, S., Huang, C., and Bulun, S. E. (2007). Novel estrogen receptor-
alpha binding sites and estradiol target genes identified by chromatin
immunoprecipitation cloning in breast cancer. Cancer Res 67, 5017-5024.

Lindberg, K., Strom, A., Lock, J. G., Gustafsson, J., Haldosen, L., and Helguero, L. A.
(2010). Expression of estrogen receptor beta increases integrin alphal and
integrin betal levels and enhances adhesion of breast cancer cells. J. Cell.
Physiol 222, 156-167.

Liu, M., Casimiro, M. C., Wang, C., Shirley, L. A., Jiao, X., Katiyar, S., Ju, X., Li, Z., Yu,
Z., Zhou, J., et al. (2009). p21CIP1 attenuates Ras- and c-Myc-dependent breast
tumor epithelial mesenchymal transition and cancer stem cell-like gene
expression in vivo. Proceedings of the National Academy of Sciences of the
United States of America. Available at: internal-pdf://p21cip1 block EMT to BCSC
via Ras c-myc PNAS 2009-0980524288/p21cip1 block EMT to BCSC via Ras c-
myc PNAS 2009.pdf.

Liu, R., Wang, X., Chen, G. Y., Dalerba, P., Gurney, A., Hoey, T., Sherlock, G., Lewicki,
J., Shedden, K., and Clarke, M. F. (2007). The prognostic role of a gene
signature from tumorigenic breast-cancer cells. The New England journal of
medicine 356, 217-26.

Liu, S., Dontu, G., Mantle, I. D., Patel, S., Ahn, N. S., Jackson, K. W., Suri, P., and
Wicha, M. S. (2006). Hedgehog signaling and Bmi-1 regulate self-renewal of
normal and malignant human mammary stem cells. Cancer research 66, 6063-
71.

Liu, S., Dontu, G., and Wicha, M. S. (2005). Mammary stem cells, self-renewal
pathways, and carcinogenesis. Breast Cancer Res 7, 86-95.

Lobo, N. A., Shimono, Y., Qian, D., and Clarke, M. F. (2007). The biology of cancer
stem cells. Annual review of cell and developmental biology 23, 675-99.

Losel, R. M., Falkenstein, E., Feuring, M., Schultz, A., Tillmann, H., Rossol-Haseroth,
K., and Wehling, M. (2003). Nongenomic steroid action: controversies, questions,
and answers. Physiol. Rev 83, 965-1016.

Marsaud, V., Gougelet, A., Maillard, S., and Renoir, J. (2003). Various phosphorylation
pathways, depending on agonist and antagonist binding to endogenous estrogen
receptor alpha (ERalpha), differentially affect ERalpha extractability, proteasome-
mediated stability, and transcriptional activity in human breast cancer cells. Mol.
Endocrinol 17, 2013-2027.


138









Martin, L. A., Farmer, I., Johnston, S. R., Ali, S., Marshall, C., and Dowsett, M. (2003).
Enhanced estrogen receptor (ER) alpha, ERBB2, and MAPK signal transduction
pathways operate during the adaptation of MCF-7 cells to long term estrogen
deprivation. The Journal of biological chemistry 278, 30458-68.

Martin, L., Farmer, I., Johnston, S. R. D., Ali, S., and Dowsett, M. (2005). Elevated
ERK1/ERK2/estrogen receptor cross-talk enhances estrogen-mediated signaling
during long-term estrogen deprivation. Endocr. Relat. Cancer 12 Suppl 1, S75-
84.

McDonnell, D. P., and Norris, J. D. (2002). Connections and Regulation of the Human
Estrogen Receptor. Science 296, 1642-1644.

Metivier, R., Reid, G., and Gannon, F. (2006). Transcription in four dimensions: nuclear
receptor-directed initiation of gene expression. EMBO reports 7, 161-7.

Metivier, R., Penot, G., Hubner, M. R., Reid, G., Brand, H., Kos, M., and Gannon, F.
(2003). Estrogen receptor-alpha directs ordered, cyclical, and combinatorial
recruitment of cofactors on a natural target promoter. Cell 115, 751-763.

Monden, T., Kishi, M., Hosoya, T., Satoh, T., Wondisford, F. E., Hollenberg, A. N.,
Yamada, M., and Mori, M. (1999). p120 acts as a specific coactivator for 9-cis-
retinoic acid receptor (RXR) on peroxisome proliferator-activated receptor-
gamma/RXR heterodimers. Mol Endocrinol 13, 1695-703.

Monden, T., Wondisford, F. E., and Hollenberg, A. N. (1997). Isolation and
characterization of a novel ligand-dependent thyroid hormone receptor-
coactivating protein. J Biol Chem 272, 29834-41.

Mori, A. D., Zhu, Y., Vahora, I., Nieman, B., Koshiba-Takeuchi, K., Davidson, L., Pizard,
A., Seidman, J. G., Seidman, C. E., Chen, X. J., et al. (2006). Tbx5-dependent
rheostatic control of cardiac gene expression and morphogenesis. Dev Biol 297,
566-86.

Morrison, B. J., Schmidt, C. W., Lakhani, S. R., Reynolds, B. A., and Lopez, J. A.
(2008). Breast cancer stem cells: implications for therapy of breast cancer.
Breast Cancer Res 10, 210.

Murohashi, M., Hinohara, K., Kuroda, M., Isagawa, T., Tsuji, S., Kobayashi, S.,
Umezawa, K., Tojo, A., Aburatani, H., and Gotoh, N. (2009). Gene set
enrichment analysis provides insight into novel signalling pathways in breast
cancer stem cells. British journal of cancer. Available at: internal-pdf://Murahashi
M, BrJC 2009, TGFbeat NFkappa etc ID as upreg in TIC-0053868804/Murahashi
M, BrJC 2009, TGFbeat NFkappa etc ID as upreg in TIC.pdf.

Naiche, L. A., Harrelson, Z., Kelly, R. G., and Papaioannou, V. E. (2005). T-box genes
in vertebrate development. Annu Rev Genet 39, 219-39.


139










Neumeister, V., and Rimm, D. (2009). Is ALDH1 a good method for definition of breast
cancer stem cells? Breast cancer research and treatment. Available at: internal-
pdf://Neumeister V, BCRT 2009, ALDH1 as marker for BCSC rebuttal-
2214998528/Neumeister V, BCRT 2009, ALDH1 as marker for BCSC
rebuttal. pdf.

Obrero, M., Yu, D. V., and Shapiro, D. J. (2002). Estrogen receptor-dependent and
estrogen receptor-independent pathways for tamoxifen and 4-hydroxytamoxifen-
induced programmed cell death. The Journal of biological chemistry 277, 45695-
703.

Osborne, C. K., Hobbs, K., and Trent, J. M. (1987). Biological differences among MCF-7
human breast cancer cell lines from different laboratories. Breast Cancer
Research and Treatment 9, 111-121.

Osipo, C., Gajdos, C., Liu, H., Chen, B., and Jordan, V. C. (2003). Paradoxical Action of
Fulvestrant in Estradiol-lnduced Regression of Tamoxifen-Stimulated Breast
Cancer. J. Natl. Cancer Inst. 95, 1597-1608.

Pabst, O., Herbrand, H., Takuma, N., and Arnold, H. H. (2000). NKX2 gene expression
in neuroectoderm but not in mesendodermally derived structures depends on
sonic hedgehog in mouse embryos. Dev Genes Evol 210, 47-50.

Park, E. J., Ogden, L. A., Talbot, A., Evans, S., Cai, C. L., Black, B. L., Frank, D. U., and
Moon, A. M. (2006). Required, tissue-specific roles for Fgf8 in outflow tract
formation and remodeling. Development 133, 2419-33.

Park, S. Y., Gonen, M., Kim, H. J., Michor, F., and Polyak, K. (2010). Cellular and
genetic diversity in the progression of in situ human breast carcinomas to an
invasive phenotype. The Journal of clinical investigation. Available at: internal-
pdf://Park SY, J Clin Invest 2010, genetics heterogenity in primary breast cancer
and CSC-1604248320/Park SY, J Clin Invest 2010, genetics heterogenity in
primary breast cancer and CSC.pdf.

Paruthiyil, S., Parmar, H., Kerekatte, V., Cunha, G. R., Firestone, G. L., and Leitman, D.
C. (2004). Estrogen Receptor {beta} Inhibits Human Breast Cancer Cell
Proliferation and Tumor Formation by Causing a G2 Cell Cycle Arrest. Cancer
Res 64, 423-428.

Patient, R. K., and McGhee, J. D. (2002). The GATA family (vertebrates and
invertebrates). Curr Opin Genet Dev 12, 416-22.

Pece, S., Tosoni, D., Confalonieri, S., Mazzarol, G., Vecchi, M., Ronzoni, S., Bernard,
L., Viale, G., Pelicci, P. G., and Di Fiore, P. P. (2010). Biological and Molecular
Heterogeneity of Breast Cancers Correlates with Their Cancer Stem Cell
Content. Cell 140, 62-73.


140









Pedram, A., Razandi, M., and Levin, E. R. (2006). Nature of functional estrogen
receptors at the plasma membrane. Molecular endocrinology (Baltimore, Md 20,
1996-2009.

Pedram, A., Razandi, M., Sainson, R. C., Kim, J. K., Hughes, C. C., and Levin, E. R.
(2007). A conserved mechanism for steroid receptor translocation to the plasma
membrane. The Journal of biological chemistry 282, 22278-88.

Pei, X. H., Bai, F., Smith, M. D., Usary, J., Fan, C., Pai, S. Y., Ho, I. C., Perou, C. M.,
and Xiong, Y. (2009). CDK inhibitor p18(INK4c) is a downstream target of
GATA3 and restrains mammary luminal progenitor cell proliferation and
tumorigenesis. Cancer cell 15, 389-401.

Phillips, T. M., McBride, W. H., and Pajonk, F. (2006). The response of CD24(-
/low)/CD44+ breast cancer-initiating cells to radiation. Journal of the National
Cancer Institute 98, 1777-85.

Plageman, T. F., and Yutzey, K. E. (2005). T-box genes and heart development: putting
the "T" in heart. Dev Dyn 232, 11-20.

Ponti, D., Costa, A., Zaffaroni, N., Pratesi, G., Petrangolini, G., Coradini, D., Pilotti, S.,
Pierotti, M. A., and Daidone, M. G. (2005). Isolation and in vitro propagation of
tumorigenic breast cancer cells with stem/progenitor cell properties. Cancer
research 65, 5506-11.

Prall, O. W., Menon, M. K., Solloway, M. J., Watanabe, Y., Zaffran, S., Bajolle, F.,
Biben, C., McBride, J. J., Robertson, B. R., Chaulet, H., et al. (2007). An Nkx2-
5/Bmp2/Smadl negative feedback loop controls heart progenitor specification
and proliferation. Cell 128, 947-59.

Pu, W. T., Ishiwata, T., Juraszek, A. L., Ma, Q., and Izumo, S. (2004). GATA4 is a
dosage-sensitive regulator of cardiac morphogenesis. Dev Biol 275, 235-44.

Raisner, R. M., and Madhani, H. D. (2006). Patterning chromatin: form and function for
H2A.Z variant nucleosomes. Curr Opin Genet Dev 16, 119-24.

Raj, A., and van Oudenaarden, A. (2008). Nature, nurture, or chance: stochastic gene
expression and its consequences. Cell 135, 216-26.

Raouf, A., Zhao, Y., To, K., Stingl, J., Delaney, A., Barbara, M., Iscove, N., Jones, S.,
McKinney, S., Emerman, J., et al. (2008). Transcriptome analysis of the normal
human mammary cell commitment and differentiation process. Cell stem cell 3,
109-18.


141









Rappa, G., and Lorico, A. Phenotypic characterization of mammosphere-forming cells
from the human MA-11 breast carcinoma cell line. Experimental Cell Research In
Press, Corrected Proof Available at:
http://www.sciencedirect.com/science/article/B6WFC-4Y4PBH6-
3/2/00fa5c9a4f47afcdaaf8ea5be9276647 [Accessed April 30, 2010].

Regan, J., and Smalley, M. (2007). Prospective isolation and functional analysis of stem
and differentiated cells from the mouse mammary gland. Stem cell reviews 3,
124-36.

Reynolds, B. A., and Rietze, R. L. (2005). Neural stem cells and neurospheres--re-
evaluating the relationship. Nat. Methods 2, 333-336.

Risebro, C. A., Smart, N., Dupays, L., Breckenridge, R., Mohun, T. J., and Riley, P. R.
(2006). Hand1 regulates cardiomyocyte proliferation versus differentiation in the
developing heart. Development 133, 4595-606.

Rizzo, P., Miao, H., D'Souza, G., Osipo, C., Yun, J., Zhao, H., Mascarenhas, J., Wyatt,
D., Antico, G., Hao, L., et al. (2008). Cross-talk between Notch and the Estrogen
Receptor in Breast Cancer Suggests Novel Therapeutic Approaches. Cancer
Res 68, 5226-5235.

Rosen, J. M., and Jordan, C. T. (2009). The increasing complexity of the cancer stem
cell paradigm. Science (New York, N.Y 324, 1670-3.

Schmits, R., Filmus, J., Gerwin, N., Senaldi, G., Kiefer, F., Kundig, T., Wakeham, A.,
Shahinian, A., Catzavelos, C., Rak, J., et al. (1997). CD44 regulates
hematopoietic progenitor distribution, granuloma formation, and tumorigenicity.
Blood 90, 2217-2233.

Shackleton, M., Quintana, E., Fearon, E. R., and Morrison, S. J. (2009). Heterogeneity
in cancer: cancer stem cells versus clonal evolution. Cell 138, 822-9.

Shang, Y., Hu, X., DiRenzo, J., Lazar, M. A., and Brown, M. (2000). Cofactor dynamics
and sufficiency in estrogen receptor-regulated transcription. Cell 103, 843-52.

Shaw, L., Sadler, A., Pugazhendhi, D., and Darbre, P. (2006). Changes in oestrogen
receptor-[alpha] and -[beta] during progression to acquired resistance to
tamoxifen and fulvestrant (Faslodex, ICI 182,780) in MCF7 human breast cancer
cells. The Journal of Steroid Biochemistry and Molecular Biology 99, 19-32.

Sheridan, C., Kishimoto, H., Fuchs, R. K., Mehrotra, S., Bhat-Nakshatri, P., Turner, C.
H., Goulet, R., Badve, S., and Nakshatri, H. (2006). CD44+/CD24- breast cancer
cells exhibit enhanced invasive properties: an early step necessary for
metastasis. Breast Cancer Res 8, R59.


142









Shi, X., Gipp, J., and Bushman, W. (2007). Anchorage-independent culture maintains
prostate stem cells. Developmental biology 312, 396-406.

Shipitsin, M., Campbell, L. L., Argani, P., Weremowicz, S., Bloushtain-Qimron, N., Yao,
J., Nikolskaya, T., Serebryiskaya, T., Beroukhim, R., Hu, M., et al. (2007).
Molecular definition of breast tumor heterogeneity. Cancer cell 11, 259-73.

Shipitsin, M., and Polyak, K. (2008). The cancer stem cell hypothesis: in search of
definitions, markers, and relevance. Laboratory investigation; a journal of
technical methods and pathology 88, 459-63.

Sleeman, K. E., Kendrick, H., Robertson, D., Isacke, C. M., Ashworth, A., and Smalley,
M. J. (2007). Dissociation of estrogen receptor expression and in vivo stem cell
activity in the mammary gland. The Journal of cell biology 176, 19-26.

Smalley, M., and Ashworth, A. (2003). Stem cells and breast cancer: A field in transit.
Nat Rev Cancer 3, 832-844.

Smith, G. H. (2005). Label-retaining epithelial cells in mouse mammary gland divide
asymmetrically and retain their template DNA strands. Development (Cambridge,
England) 132, 681-7.

Sotiriou, C., Neo, S., McShane, L. M., Korn, E. L., Long, P. M., Jazaeri, A., Martiat, P.,
Fox, S. B., Harris, A. L., and Liu, E. T. (2003). Breast cancer classification and
prognosis based on gene expression profiles from a population-based study.
Proceedings of the National Academy of Sciences of the United States of
America 100, 10393-10398.

Sotiriou, C., Wirapati, P., Loi, S., Harris, A., Fox, S., Smeds, J., Nordgren, H., Farmer,
P., Praz, V., Haibe-Kains, B., et al. (2006). Gene Expression Profiling in Breast
Cancer: Understanding the Molecular Basis of Histologic Grade To Improve
Prognosis. J. Natl. Cancer Inst. 98, 262-272.

Srivastava, D. (2006). Making or breaking the heart: from lineage determination to
morphogenesis. Cell 126, 1037-48.

Srivastava, D., and Olson, E. N. (2000). A genetic blueprint for cardiac development.
Nature 407, 221-6.

Stennard, F. A., and Harvey, R. P. (2005). T-box transcription factors and their roles in
regulatory hierarchies in the developing heart. Development 132, 4897-910.

Sternlicht, M. D. (2006). Key stages in mammary gland development: the cues that
regulate ductal branching morphogenesis. Breast Cancer Res 8, 201.


143









Stingl, J., Eirew, P., Ricketson, I., Shackleton, M., Vaillant, F., Choi, D., Li, H. I., and
Eaves, C. J. (2006). Purification and unique properties of mammary epithelial
stem cells. Nature 439, 993-7.

Stoica, G. E., Franke, T. F., Moroni, M., Mueller, S., Morgan, E., lann, M. C., Winder, A.
D., Reiter, R., Wellstein, A., Martin, M. B., et al. (2003). Effect of estradiol on
estrogen receptor-alpha gene expression and activity can be modulated by the
ErbB2/PI 3-K/Akt pathway. Oncogene 22, 7998-8011.

Tam, P. P., and Loebel, D. A. (2007). Gene function in mouse embryogenesis: get set
for gastrulation. Nat Rev Genet 8, 368-81.

Tanaka, M., Chen, Z., Bartunkova, S., Yamasaki, N., and Izumo, S. (1999). The cardiac
homeobox gene Csx/Nkx2.5 lies genetically upstream of multiple genes essential
for heart development. Development 126, 1269-80.

Thottassery, J. V., Sun, Y., Westbrook, L., Rentz, S. S., Manuvakhova, M., Qu, Z.,
Samuel, S., Upshaw, R., Cunningham, A., and Kern, F. G. (2004). Prolonged
extracellular signal-regulated kinase 1/2 activation during fibroblast growth factor
1- or heregulin betal-induced antiestrogen-resistant growth of breast cancer cells
is resistant to mitogen-activated protein/extracellular regulated kinase kinase
inhibitors. Cancer Res 64, 4637-4647.

Tremblay, G. B., Bergeron, D., and Giguere, V. (2001). 4-Hydroxytamoxifen is an
isoform-specific inhibitor of orphan estrogen-receptor-related (ERR) nuclear
receptors beta and gamma. Endocrinology 142, 4572-4575.

Vargo-Gogola, T., and Rosen, J. M. (2007). Modelling breast cancer: one size does not
fit all. Nature reviews 7, 659-72.

Vera-Ramirez, L., Sanchez-Rovira, P., Ramirez-Tortosa, C. L., Quiles, J. L., Ramirez-
Tortosa, M. C., Alvarez, J. C., Fernandez-Navarro, M., and Lorente, J. A. (2010).
Gene-expression profiles, tumor microenvironment, and cancer stem cells in
breast cancer: Latest advances towards an integrated approach. Cancer Treat
Rev. Available at: http://www.ncbi.nlm.nih.gov/pubmed/20299155 [Accessed
March 29, 2010].

Vermeulen, L., Todaro, M., de Sousa Mello, F., Sprick, M. R., Kemper, K., Perez Alea,
M., Richel, D. J., Stassi, G., and Medema, J. P. (2008). Single-cell cloning of
colon cancer stem cells reveals a multi-lineage differentiation capacity.
Proceedings of the National Academy of Sciences of the United States of
America 105, 13427-32.

Villadsen, R., Fridriksdottir, A. J., Ronnov-Jessen, L., Gudjonsson, T., Rank, F.,
LaBarge, M. A., Bissell, M. J., and Petersen, O. W. (2007). Evidence for a stem
cell hierarchy in the adult human breast. The Journal of cell biology 177, 87-101.


144









Villadsen, R. (2005). In search of a stem cell hierarchy in the human breast and its
relevance to breast cancer evolution. APMIS 113, 903-921.

Visvader, J. E. (2009). Keeping abreast of the mammary epithelial hierarchy and breast
tumorigenesis. Genes & development 23, 2563-77.

Visvader, J. E., and Lindeman, G. J. (2008). Cancer stem cells in solid tumours:
accumulating evidence and unresolved questions. Nature reviews 8, 755-68.

Walia, V., and Elble, R. C. (2010). Enrichment for breast cancer cells with
stem/progenitor properties by differential adhesion. Stem Cells Dev. Available at:
http://www.ncbi.nlm.nih.gov/pubmed/20222827 [Accessed March 22, 2010].

Wang, X., Kruithof-de Julio, M., Economides, K. D., Walker, D., Yu, H., Halili, M. V., Hu,
Y. P., Price, S. M., Abate-Shen, C., and Shen, M. M. (2009). A luminal epithelial
stem cell that is a cell of origin for prostate cancer. Nature 461, 495-500.

Welm, B. E., Tepera, S. B., Venezia, T., Graubert, T. A., Rosen, J. M., and Goodell, M.
A. (2002). Sca-l(pos) cells in the mouse mammary gland represent an enriched
progenitor cell population. Dev. Biol 245, 42-56.

Wicha, M. S. (2008). Cancer stem cell heterogeneity in hereditary breast cancer. Breast
Cancer Res 10, 105.

Wijayaratne, A. L., and McDonnell, D. P. (2001). The human estrogen receptor-alpha is
a ubiquitinated protein whose stability is affected differentially by agonists,
antagonists, and selective estrogen receptor modulators. J. Biol. Chem 276,
35684-35692.

Williams, C. C., Basu, A., EI-Gharbawy, A., Carrier, L. M., Smith, C. L., and Rowan, B.
G. (2009). Identification of four novel phosphorylation sites in estrogen receptor
alpha: impact on receptor-dependent gene expression and phosphorylation by
protein kinase CK2. BMC Biochem 10, 36.

Wright, M. H., Calcagno, A. M., Salcido, C. D., Carlson, M. D., Ambudkar, S. V., and
Varticovski, L. (2008). Brcal breast tumors contain distinct CD44+/CD24- and
CD133+ cells with cancer stem cell characteristics. Breast Cancer Res 10, R10.

Wright, M. H., Robles, A. I., Herschkowitz, J. I., Hollingshead, M. G., Anver, M. R.,
Perou, C. M., and Varticovski, L. (2008). Molecular analysis reveals
heterogeneity of mouse mammary tumors conditionally mutant for Brcal.
Molecular cancer 7, 29.

Xin, M., Davis, C. A., Molkentin, J. D., Lien, C. L., Duncan, S. A., Richardson, J. A., and
Olson, E. N. (2006). A threshold of GATA4 and GATA6 expression is required for
cardiovascular development. Proc Natl Acad Sci U S A 103, 11189-94.


145









Xu, H., and Baldini, A. (2007). Genetic pathways to mammalian heart development:
Recent progress from manipulation of the mouse genome. Semin Cell Dev Biol
18, 77-83.

Zheng, A., Kallio, A., and Harkonen, P. (2007). Tamoxifen-induced rapid death of MCF-
7 breast cancer cells is mediated via extracellularly signal-regulated kinase
signaling and can be abrogated by estrogen. Endocrinology 148, 2764-77.

Zhou, J., Wulfkuhle, J., Zhang, H., Gu, P., Yang, Y., Deng, J., Margolick, J. B., Liotta, L.
A., Petricoin, E., and Zhang, Y. (2007). Activation of the PTEN/mTOR/STAT3
pathway in breast cancer stem-like cells is required for viability and maintenance.
Proceedings of the National Academy of Sciences of the United States of
America 104, 16158-63.

Zhou, J., Zhang, H., Gu, P., Bai, J., Margolick, J. B., and Zhang, Y. (2008). NF-kappaB
pathway inhibitors preferentially inhibit breast cancer stem-like cells. Breast
cancer research and treatment 111, 419-27.

Zhou, J. H., Yu, D. V., Cheng, J., and Shapiro, D. J. (2007). Delayed and persistent
ERK1/2 activation is required for 4-hydroxytamoxifen-induced cell death. Steroids
72, 765-77.

Zhu, B. T., Han, G. Z., Shim, J. Y., Wen, Y., and Jiang, X. R. (2006). Quantitative
structure-activity relationship of various endogenous estrogen metabolites for
human estrogen receptor alpha and beta subtypes: Insights into the structural
determinants favoring a differential subtype binding. Endocrinology 147, 4132-50.

Zucchi, I., Astigiano, S., Bertalot, G., Sanzone, S., Cocola, C., Pelucchi, P., Bertoli, G.,
Stehling, M., Barbieri, O., Albertini, A., et al. (2008). Distinct populations of tumor-
initiating cells derived from a tumor generated by rat mammary cancer stem
cells. Proceedings of the National Academy of Sciences 105, 16940-16945.

Zucchi, I., Sanzone, S., Astigiano, S., Pelucchi, P., Scotti, M., Valsecchi, V., Barbieri,
O., Bertoli, G., Albertini, A., Reinbold, R. A., et al. (2007). The properties of a
mammary gland cancer stem cell. Proceedings of the National Academy of
Sciences of the United States of America 104, 10476-81.


146









BIOGRAPHICAL SKETCH

Ada Ao was born in 1981 in Macau to Chi Seng Ao and Lai Meng Chan. They

immigrated to Boston, Massachusetts in 1990. She completed secondary education at

Boston Latin School in Boston, Massachusetts in 1999. Ada then attended Brandeis

University from 1999-2003, where she majored in biology and biochemistry. After

graduating with a Bachelor of Science in biology, she was employed for 2 years as a

research technician at Boston University Medical Center under the guidance of Dr. Jude

Deeney, studying the effects of fatty acids on insulin secretion in rat islets. In 2005, Ada

entered the Interdisciplinary Program in Biomedical Sciences (IDP) at the University of

Florida to pursue a graduate degree in the Department of Biochemistry and Molecular

Biology. She joined the laboratory of Dr. Jianrong Lu in May 2006, and qualified for

PhD candidacy September 2007. Her tenure has included a diverse course of studies,

such as the characterization of cardiac defects in mouse embryos, the study of

Estrogen Related Receptors on hypoxic gene regulation, and the characterization of

antiestrogen effects on MCF7-derived mammospheres.


147





PAGE 1

1 CHARACTERIZATION OF ANTIESTROGEN REPONSE IN MCF7 DERIVED MAMMOSPHERES By ADA SI NGA AO A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2010

PAGE 2

2 2010 A da Si Nga Ao

PAGE 3

3 To piling higher and deeper

PAGE 4

4 ACKNOWLEDGMENTS I cannot do justice to all those who helped me to reach this point. I extend my thanks to all my friends for their sympat hetic ear and understanding throughout this endeavor. Thanks to Nicole, Amanda, Shermi, Star, Michele, Julia, Zhou, Joeva Carlos, and Heather ever done in my life, even when they ecessar il y agree with my pursuit s Special thanks to my cousins, especially Amy and Diane, for keeping me grounded. My deepest thanks and appreciation to Dr. Jude Deeney, Dr. Emma Heart, Dr. Lina Moitoso de Vargas, Dr. John Flanagan, Dr. Ann Marie Richar ds, Liping Hu Dr. Lihan Liu Dr. Rita Avancini, and Dr. Isabel Chiu for being valuable role models during my early days in a lab I would like to thank Dr. Jianrong Lu for the opportunity to perform my graduate training in his lab. Thanks to my labmat es: Sushama for commiserating with me, and Tong for displaying an almost Zen like calm through everything that boggles the mind future. I would also like to thank my adv isory committee (Drs. Linda Bloom, J rg Bungert, Harry Nick, and Brent Reynolds) for all their helpful comments throughout my training, especially Dr. Nick and Dr. Bloom for their patience during a difficult part of the process. Finally, a special thanks to Will for hi s continued love, encouragement and friendship. It is a privilege to share his dark, twisted sense of humor and a joy to know his inner goodness. He has been right beside me through all the ups and downs a nd

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 8 LIST OF FIGURES ................................ ................................ ................................ .......... 9 LIST OF ABBREVIATIONS ................................ ................................ ........................... 11 ABSTRACT ................................ ................................ ................................ ................... 13 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 15 Introduction to Mammary Gland Development ................................ ........................ 15 Breast Cancer ................................ ................................ ................................ ......... 16 Types of Breast C ancer ................................ ................................ .................... 17 Ductal Carcinoma ................................ ................................ ............................. 17 Lobular Carcinoma ................................ ................................ ........................... 17 Other Types of Breast Cancer ................................ ................................ .......... 18 Estrogen Receptor Alpha ................................ ................................ ........................ 19 Estrogen Receptor Alpha Signaling ................................ ................................ .. 20 Genomic Function ................................ ................................ ............................ 20 Non Genomic Function ................................ ................................ .................... 22 Cell of Origin ................................ ................................ ................................ ........... 23 Clonal Evolution Model ................................ ................................ ..................... 23 The Cancer Stem Cell (CSC) Hypothesis ................................ ......................... 23 Tumor Heterogeneity and CSC Hypothesis ................................ ...................... 24 Implications of Each Model on Breast Cancer Recurrence .............................. 25 Estrogen receptor alpha, stem cells and breast cancer ............................. 26 Contribution of CSC/TIC to antiestrogen resistance ................................ .. 28 CSC Isolation and Detection ................................ ................................ ............ 31 Other Examples of CSC Enrichment ................................ ................................ 32 2 GENERAL MATERIALS AND METHODS ................................ .............................. 36 Cell Culture ................................ ................................ ................................ ............. 36 MCF7S Cell Proliferation and Sphere Formation Assay ................................ ......... 37 Serial Passaging ................................ ................................ .............................. 37 Cell Proliferation Assay ................................ ................................ .................... 37 Sphere Formation Assay ................................ ................................ .................. 38 RNA Isolation ................................ ................................ ................................ .......... 38 Reverse Transcript ase PCR (RT PCR) ................................ ................................ .. 39 Real Time PCR ................................ ................................ ................................ ....... 39

PAGE 6

6 Total Protein Isolation ................................ ................................ ............................. 40 Nuclear Extraction ................................ ................................ ................................ .. 40 Immunoblot Analysis ................................ ................................ ............................... 41 Immunofluorescence ................................ ................................ ............................... 42 short hairpin RNA (shRNA) Vector Construction ................................ .................... 43 shRNA Oligo Design and Cloning ................................ ................................ .... 43 Plasmid DNA MiniPrep ................................ ................................ ..................... 44 Retrovirus Production and Transduction of Target Cells ................................ ......... 45 Transient Transfection ................................ ................................ ...................... 45 Retroviral Infection for Stable Integration ................................ ......................... 45 Selection and Enrichment of Single Clones ................................ ...................... 46 Analysis of CD44 Expression ................................ ................................ .................. 47 BrdU/Propidium Iodide Cell Proliferation Assay ................................ ...................... 47 Annexin V PE Apoptosis Detection using BD Pharmingen Kit ................................ 48 Statistical Analysis ................................ ................................ ................................ .. 48 In Vivo Tumorigenic Assay ................................ ................................ ..................... 48 Preparation of Mice ................................ ................................ .......................... 49 Preparation of Tumor Cells ................................ ................................ ............... 49 Mouse Injections ................................ ................................ .............................. 49 3 CHARACTERIZING MAMMOS PHERES DERIVED FROM MCF7 PARENTAL CELLS ................................ ................................ ................................ .................... 54 Introduction ................................ ................................ ................................ ............. 54 Results ................................ ................................ ................................ .................... 55 Characterization of Mammospheres (MCF7S) Derived from MCF7 Cells ........ 55 Expression of Putative Breast Tumorigenic Marker CD44 in MCF7S Cells ...... 56 ER Status and Stability in MCF7S Cells ................................ ......................... 57 Discussion ................................ ................................ ................................ .............. 58 4 THE EFFECTS OF ANTIESTROGEN ON MAMMOSPHERE FORMATIO N .......... 66 Introduction ................................ ................................ ................................ ............. 66 Results ................................ ................................ ................................ .................... 68 MCF7S Response to Antiestrogens ................................ ................................ 68 Antiestrogen Efficacy in Parental MCF7P and MCF7S ................................ .... 69 Sphere Formation Frequency Following Antiestrogen Challenge ..................... 71 MCF7S Cell Proliferation under Long Term Antiestrogen Treatment ............... 72 MCF7S Cell Cycle Analysis ................................ ................................ .............. 73 MCF7S Apoptosis Assay ................................ ................................ .................. 74 Discussion ................................ ................................ ................................ .............. 75 5 THE ROLE OF ESTROGEN RECEPTOR ALPHA ON MAMMOSPHERE FORMATION ................................ ................................ ................................ .......... 87 Introduction ................................ ................................ ................................ ............. 87 Results ................................ ................................ ................................ .................... 88

PAGE 7

7 Proliferation of ER Knockdown MCF7S ................................ .......................... 88 Antiestrogen Response of ER Knockdown MCF7S ................................ ........ 88 Sphere Formation Frequency of ER Knockdown MCF7S Following Antiestrogen Chal lenge ................................ ................................ ................. 89 Using SERDs and SERMs to Mimic shER Effects in MCF7S ........................... 90 Discussion ................................ ................................ ................................ .............. 91 6 CONCLUSIONS AND FUTURE DIRECTIONS ................................ ...................... 98 Conclusions and Discussion ................................ ................................ ................... 98 Future Directions ................................ ................................ ................................ .. 102 APPENDIX A SUPPLEMENTAL FIGURES ................................ ................................ ................ 104 Chapter 3 ................................ ................................ ................................ .............. 104 Chapter 4 ................................ ................................ ................................ .............. 105 Chapter 5 ................................ ................................ ................................ .............. 112 B BRD8 FUNCTION DURING EARLY CARDIAC DEVELOPMENT ........................ 113 Introduction ................................ ................................ ................................ ........... 113 Overview of Mammalian Heart Development ................................ ........................ 113 Transcription Factors Regulating Heart Formation ................................ ............... 114 Specification of Myocardial Progenitor Cells ................................ .................. 114 Differentiation of Precursors into Cardiomyocytes ................................ .......... 115 Downstream Transcriptional Networks Associated With Heart Formation ..... 115 General Introduction to Brd8 ................................ ................................ ................. 116 Materials and Me thods ................................ ................................ .......................... 116 Immunohistochemistry ................................ ................................ .................... 116 Whole Mount In Situ Hybridization (WHISH) ................................ .................. 117 Riboprobe synthesis ................................ ................................ ................ 117 Mouse embryo preparation ................................ ................................ ...... 118 Whole mount in situ hybridization ................................ ............................ 118 RT PCR Primers for Brd8 Allele Expression ................................ .................. 119 Results and Discussion ................................ ................................ ......................... 119 LIST OF REF ERENCES ................................ ................................ ............................. 130 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 147

PAGE 8

8 LIST OF TABLES Table page 2 1 shRNA Oligos desig ned for retrovirus mediated knockdown. ............................. 52 2 2 Antibodies used for Immunoblotting (IB), Immunofluorescence (IF), or Flow Cytometry (FC) ................................ ................................ ................................ ... 53 2 3 Primers used for real time RT PCR ................................ ................................ .... 53 3 1 Data for CD44 FITC signal quantification using flow cytometer. ......................... 64 4 1 Gro wth kinetics of MCF7S long term expansion under antiestrogen challenge. ................................ ................................ ................................ ........... 84 A 1 Additional data for CD44 FITC signal quantification using flow cytometer (Figure A 1). ................................ ................................ ................................ ..... 104 A 2 Data for BrdU/PI cell cycle analysis MCF7S cells treated 72 hours with antiestrogens (Figure A 5). ................................ ................................ ............... 110

PAGE 9

9 LIST OF FIGURES Figure page 1 1 Estrogen receptor isoforms ................................ ................................ ................. 34 1 2 Mammary gland differentiation hierarchy ................................ ........................... 34 1 3 Potential re lationship between mammary differentiation hierarchy and the development of tumor subtypes ................................ ................................ .......... 35 2 1 Tissue culture scheme for MCF7S proliferation assay and sphere formation assay.. ................................ ................................ ................................ ................ 50 2 2 Cloning vector information for mic roRNA adapted retroviral vector .................... 51 3 1 MCF7 derived m ammospheres (MCF7S) growth curve ................................ ..... 61 3 2 Phase contrast microscopy of mammospheres d erived from MCF7 parental cells ................................ ................................ ................................ .................... 61 3 3 Enriching MCF7S from MCF7P by serial passage in mammo sphere media. ..... 62 3 4 Comparing CD44 expression in MCF7P vs. MCF7S using flow cytometer ....... 63 3 5 ER expression in MCF7P and MCF7S ................................ ............................ 65 4 1 Mammosphere formation i n the presence of antiestrogens ................................ 78 4 2 Cell proliferation of MCF7 S in the presence of antiestrogens ............................. 79 4 3 MCF7 adherent culture response to 4 hydroxytamoxifen (4 OH T) with various culture media. ................................ ................................ ........................ 80 4 4 Immunoblot of cytoplasmic and nuclear fractions from MCF7P and MCF7S cells treated for 72 hours with 4 hydroxytamoxifen (4 OHT) or ICI 182780 (ICI) to characterize ER stability ................................ ................................ ...... 81 4 5 Sphere formation frequency of MCF7S after antiestrogen challenge using 4 hydroxytamoxi fen (4 OHT) or ICI 182780 (ICI) ................................ ................... 82 4 6 Long term expansion of MCF7S in the presence of antiestrogens. The lines are expressed on a semilog graph and slope of each line was calculated as log expansion for each condition ................................ ................................ ........ 83 4 7 Propidium iod ide (PI) cell cycle analysis ................................ ............................. 85 4 8 Annexin V apoptosis assay f or antiestrogen treated MCF7S ............................. 86

PAGE 10

10 5 1 Kn ockdown of ER (shER) in MCF7S cells ................................ ....................... 93 5 2 Representative phase contrast images of MCF7S, control shRNA MCF7S, and shER clone 7 expo sed to antiestrogens for 6 days ................................ ...... 94 5 3 Antiestrogen response of ER knockdown in MCF7S cells ................................ 95 5 4 Sphere formation frequency of antiestrogen treated ER knockdown MCF7S cells ................................ ................................ ................................ ................... 96 5 5 SERDs and SERMs combination treatment to mimic shER effects in MCF7S ... 97 A 1 Additional figures comparing CD44 ex pression in MCF7P and MCF7S. ......... 104 A 2 Real time RT PCR of MCF7P for estrogen response gene TFF1 .................... 105 A 3 Real time RT PCR of MCF7S for es trogen r esponse gene TFF1 ..................... 106 A 4 Time course of CTSD gene expression induced with 100 nM 17 estradiol (E2) ................................ ................................ ................................ ................... 107 A 5 BrdU/PI cell cycle analysis for MCF7S treated with antiestrogens for 72 hours ................................ ................................ ................................ ............... 109 A 6 Individual Annexin V experimental data for antiestrogen treated MCF7S at 48 hours and 72 hours summarized in Figure 5 5. ................................ ................ 111 A 7 Representative phase contrast images of shER C7 knockdown (left) and control shRNA (right) mammospheres. ................................ ............................ 112 B 1 Gene trap vector inserted into mouse geno me to generate Brd8 null allele ..... 122 B 2 Brd8 gene trapped mouse embryos and expression of wild type and mutant alleles ................................ ................................ ................................ ............... 123 B 3 Endogenous Brd8 expression in wild type E10.5 mouse embryos using whole mount in situ hybridization (WMISH) ................................ ...................... 124 B 4 Hematoxylin and eosin stained sections of Brd8 mutant embryos (E10.5) compar ed to wild type littermates ................................ ................................ .... 125 B 5 PECAM whole mount immunohistochemistry comparing vascularization of Brd8 mutant and stage mat ched wild type embryos (E8.5) ............................. 127 B 6 Whole mount in situ hybridization (WMISH) characterizing expression of key cardiac development genes ................................ ................................ ............. 128

PAGE 11

11 LIST OF ABBREVIATIONS 4 OHT 4 Hydroxytamoxif en AE Anti estrogen AKT Protein Kinase B AP 1 Activator Protein 1 BrdU Bromodeoxyuridine, 5 bromo 2 deoxyuridine BSA Bovine Serum Albumin CSC Cancer Stem Cell EGF Epidermal Growth Factor ER Estrogen Receptor Alpha ERE Estrogen Response Element ERK Extracel lular Signal Regulated Kinase ERR Estrogen Related Receptor Gamma FGF Fibroblast Growth Factor ICI ICI 182780 (Fulvestrant FASLODEX ) MAPK Mitogen Activated Protein Kinase MaSC Mammary Stem Cell MCF7P MCF7 Parental MCF7S MCF7 Spheroid Culture (mammospher es) NF B Nuclear Factor kappa B PI Propidium Iodide PI3K Phosphoinositide 3 Kinase PKA Protein Kinase A PKC Protein Kinase C Poly HEMA Poly (2 hydroxyethyl methacrylate)

PAGE 12

12 RT Reverse Transcriptase RT PCR Reverse Transcriptase Polymerase Chain Reaction SERD S elective Estrogen Receptor Down regulators SERM Selective Estrogen Receptor Modulators SFA Sphere Formation Assay shER shRNA knockdown Estrogen Receptor Alpha shRNA short hairpin RNA SP 1 Specificity Protein 1 TAM Tamoxifen TIC Tumor Initiating Cell

PAGE 13

13 Ab stract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy CHARACTERIZATION OF ANTIESTROGEN RESPONSE IN MCF7 DERIVED MAMMOSPHERES B y Ada Si Nga Ao August 2010 Chair: Jianrong Lu Major: Medical Sciences -Biochemistry and Molecular Biology The emerging cancer stem cell (CSC)/ tumor initiating cell (TIC) hypothesis seeks to explain cancer persistence and recurrence. It supposes that a small subgroup of cells within tumors is multipotent and has chemoresistant features, which contributes to the heterogeneous and regenerative phenotype of recurring tumors. In breast cancer, previous studies have shown the TIC subgroup can be pro pagate d in vitro from both primary tumors and established cell lines The goal of this project is to characterize the antiestrogen response of mammospheres positive mammary ep ithelial MCF7 cells, and to determine if there is enrichment of putative TICs after antiestrogen challenge. It is shown here that MCF7S cell proliferation was decreased by antiestrogen treatments. Sphere formation was affected by a selective estrogen rec eptor modulator (SER M), but not by a selective estrogen receptor down regulator (SERD) Using the sphere formation assay, we determined that the sphere forming ability returned after drug removal and that there was no significant decrease in sphere format cells were more

PAGE 14

14 sensitive to SERM and exhibited diminished sphere formation frequency Treatment of knockdo wn cells with SERD did not significant ly change cell proliferation, sphere formation or frequency. In summary, the data showed that a stable fraction of potential TICs remained aft er antiestrogen challenge. The r esults also demonstrated that not essential for MCF7S survival, but may mediate antiestrogen fu nction. F urther studies on SERMs may be warranted as SERMs may target potential TICs that use alt ernative mitogenic pathways

PAGE 15

15 CHAPTER 1 INTRODUCTION Introduction to Mammary Gland Development The mammary gland is unusual namely because it does not fully develop until puberty in r esponse to systemic hormones and local growth factors. The mammary gland is a tree like structure made up of multiple cell lineages that form the structural units of the gland. Each individual glandular off shoot from the main branching duct is termed th e terminal duct lobular unit (TDLU) in human, or terminal end bud (TEB) in rodents (Howard and Gust erson, 2000; Sternlicht, 2006) The TDLU is comprised of terminal ductules with a bilayer of polarized luminal epithelial cells lining the inside surrounded by a layer of contractile myoepithelial cells, topped with a cluster of alveoli cells (Visvader, 2009) Multiple TDLUs composes each of the 15 20 lobes in a human breast, with each lobe imbedded in adipose tissue (Morrison et al., 2008) In rodents, a simpler analogous structure is the terminal end bud (TEB), and has been more extensively studied. The TEB is made up of luminal restricted cells call ed b ody cells, surrounded by a layer of bipotent progenitors called cap cells. The TEB undergoes dynamic expansion and restructuring during puberty and pregnancy directed by physiological cues. The body cells develop into milk producing alveoli cells during pregnancy. The cap cells also undergo differentiation into luminal or myoepithelial lineage during pregnancy (Howard et al., 2000; Smalley and Ashworth, 2003) The breast undergoes additional transformation post pregnancy and lactation called involution, in which the TDLUs or TEBs shrink back to pre pregnancy state. The process can occur multiple times in a woman's lifetime, giving credence to a potential reservoir of lineage restricte d progenitor cells that sustain multiple rounds of breast

PAGE 16

16 remodeling. A potential stem cell hierar chy in postnatal breast tissue has gathered support, and may be a source of abe rrant progenitor cells that increas e breast cancer tumorigenic potential in an analogous fashion to hematopoietic cancers (Villadsen, 2005; Smalley et al., 2003) However, there is debate concerning the permanent or transient nature of this pool, and also the localization of these stem cells within the mammary gland structure. Breast Cancer Breast cancer has been one of the more persistent diseases in the United States for women. It has an age adjusted incidence rate of 12 2.8 per 100,000 women each year (Altekruse et al., 2009) Despite advances in early detection and diagnosis, t here has been little change i n mortality rate once the tumor metastasizes However, the first 5 years is generally favorable (NCI SEER) C urrent treatment s can eliminate the bulk of the tumor and reduce tumor mass, but it often returns with a more aggressive phenotype. The question of tumor recurrence has vexed researchers as they try to identify the root cause. It was first elegantly term ed the "s eed and soil" hypothesis, proposed by Stephen Paget in 1 889. M ost studies have focused on genetic o r molecular factors that may promote tumorigenic phenotypes such as evasion of cell death, self renewal increased proliferation resistance t o ant i growth signal, angiogenesis, tissue invasion and metastasis (Hanahan and Weinberg, 2000) There are recent attempts to find and to target the "seed" that sustains tumor initiation as the means to long term tumor suppression. This phenomenon shows a shift in focus in cancer biology: from studying the deregulation of normal cellular functions to searching for a specific cell type equipped with tumorigenic c ellular processes by default.

PAGE 17

17 Types of Breast Cancer Breast cancer classification is based on criteria such as the invasi veness, point of origin proliferative potential, and hormone receptor status. These pathological parameters were designed to evaluate tumor progress for diagnosis and prognosis. The information provided by such assessment s guide the formation of treatment plans Breast cancer types fall within four general categories : ductal carcinoma in situ (DCIS), lobular carcinoma in situ (LCIS), invasive ductal carcinoma (IDC), and invasive lobular carcinoma (ILC). Some tumors may contain a combination of these types. There are also rarer subtypes that are classified within each category (American Cancer Society 2009 ) Ductal C arcinoma DCIS is the most common type of non invasive breast cancer and has the best prognosis, as they usually contain hormone receptor positive cell s that may respo nd to anti hormone therapy (ACS, 2009) The cancer cell s are typically of luminal origin and are located inside the duct, but have not invaded surrounding tissue or lymph nodes. IDC is the invasive variety of ductal carcino ma when the cancer cells invade the duc t wall and begins to spread into the fatty tissue of the breast and lymph nodes (ACS, 2009) They may then spread to other parts of the body t hrough the lymphatic system and bloodstream. IDC is the most common type of breast cancer, but it can be further partitioned into other invasive breast cancer types. As metastasized cells, they are typically less responsive to treatments fast growing, a nd have worse prognosis. Lobular C arcinoma LCIS may be classified as non invasive breast cancer, but patients with this cancer type have a higher risk of developing invasive cancer. The cancer cells are

PAGE 18

18 localized to the lobular glands and also contain hor mone receptor positive cells that may be responsive to treatment (ACS, 2009) ILC, like IDC, is the invasive version of the in situ cancer typ e and are difficult to treat and to track progression using standard detection methods like mammograms It is a rarer subtype of invasive breast cancer, making up approximately 10% of all invasive breast cancer (ACS, 2009) Other Types of Breast C ancer There are less common types of breast cancer that do not readily conform to the ab ove subgroups. They have a more complex pathology and are no t fully understood. A few of these types are briefly described as examples. They are chiefly fast growing, invasive cancer cells that require aggressive therapy to control. Inflammatory breast cancer (IBC) is a rare form of invasive breast cancer that ma y be mistaken for an infection. It elicits an inflammatory response due to blockage of lymph vessels in the skin by cancer cells. IBC tends to grow quickly and aggressively. They are insensitive to antihormone and are treated with chemotherapy or radiat ion (ACS, 2009) Triple negative breast cancers do not express estrogen, progesterone, or HER2 receptors. This type of cancer is especially aggressive as there is no regulatory signaling from growth receptors, and they are completely non responsive to receptor mediated inhibitors. It usually requires aggressive chemotherapy to check growth, although results are temporary and recurrence rate i s high (ACS, 2009) Medullary carcinoma is considered a subtype of IDC wi th cells derived from milk duct cells It has a well defined boundary between tumor and normal tissue as abnormally large cancer cells are surrounded by immune cells along the periphery. The

PAGE 19

19 prognosis for this cancer type is better than most invasive breast cancer and may be successfully treated with standard therapies (ACS, 2009) Estrogen Receptor Alpha There are two estrogen receptor isoforms, ER and ER They are member s of the nuclear receptor super family and function as ligand activated transcription factors The two proteins are encoded on different chromosomes, have distinct localization pattern and subtle differences in ligand binding af finity and structure (Figure 1 1) In general terms, the two isoforms are present in the breast and have antagonistic functions. ER can heterodimerize with ER and inhibit activation of ER target genes (Jones et al., 1999; Kuiper et al ., 1998; Zhu et al., 2006; Ariazi et al., 2006) Both receptors contain five distinct domains (Green and Carroll, 2007) From the N terminus, domain A/B contains the transcriptional activation function 1 (AF 1) region that modulates receptor activity. Domain C contains the DNA binding domain (DBD) composed of cysteine rich zinc fingers. Domain D is a hinge region, followed by domains E and F at the C ter minus. The E domain is the ligand binding domain (LBD) and also contains the second activation function region (AF 2). F domain is a variable region between the 2 ER isoforms. As the MCF7 cell line does not express ER significantly (Lindberg et al., 2010; Paruthiyil et al., 2004) ER will be omitted in this intr oduction. In mammalian tissues, ER is known to be endogenously expressed at high levels in the mammary gland, male and female reproductive tracts, bone, the cardiovascular system, and parts of the bra in. Knockout studies in mice and ER deficiency studie s in humans have reported phenotypes including infertility and inhibition of normal

PAGE 20

20 physiological changes during puberty (Couse and Korach, 1999; Curtis Hewitt et al., 2000) Present knowledge maintains that ER has a general role in modulating cell proliferation during embryonic development, but it is not essential. However, the receptor has a profound role in postnatal development, especially during puberty and pregnancy as the mammary gland und ergoes extensi ve remodeling. Estrogen Receptor Alpha Signaling The unligand ed form of ER is sequestered within the nuclei in inhibitory complex es composed of heat shock protein ( Hsp90 ) chaperones (Fliss et al., 2000) Upon ligand binding and activation, ER dissociates from the inhibitory complex and may regulate cellular proliferation and target gene expression th rough direct or indirect DNA interaction or participat e in non genomic mitogenic signaling (Ariazi et al., 2006; Bjornstrom and Sjoberg, 200 5) Genomic F unction The classical function of ER is mainly through genomic means as the receptor regulates target gene expression. In the classic genomic model, the receptors dimerize upon ligand binding and interact directly with DNA on the promoter region of target genes. The re ceptor binds prefere ntially on conserved cis element sequence s called the Es trogen Response Element (ERE) (Hayashi and Yamaguchi, 2008) Recent genome wide ER D NA interaction studies have uncovered novel binding sites and new classes of ER regulated genes (Carroll et al., 2006; Lin et al., 2007) ER may also bind DNA without activation by ligands. Ligand independent binding is typically induced by receptor phosphorylation, and mediated by growth factor activated kinase cascades (Dudek and Picard, 2008; Lannigan, 2003) These kinases

PAGE 21

21 activate ER by phosphorylating sp ecific amino acid residues. Eight specific resid ues have been reported and each is associated with a specific kinase cascade (Hayashi et al., 2008) Cell cycle regulated cyclin A/cdk2 comple x phosphorylates S104/106. MAPK is known to phosphorylate S118 (Atanaskova et al., 2002) Akt (or PKB), in complex with PI3K, can phosphorylate S167 (Stoica et al., 2003; Campbell et al., 2001) These phosphorylation sites are located within the activ ation function 1 (AF 1) region which resides in the A/B domain, of the receptor Phosphorylation of this region was shown to upregulate ER transcriptional activity by inducing structure conformational changes that promotes coactivator interactions PKA is able to phosphorylate S236, located in the core DNA binding domain, and inhibit dimerization (Chen et al., 1999) p 38 MAPK is known to phosphorylate T311 in the C terminal domain (Lee and Bai, 2002) The Src kinase phosphorylates the receptor at Y537 (Arnold et al., 1995) A novel phosphorylation site at the extreme C terminus has recently bee n identified at S559, and it is phosphorylated by protein kinase CK2 (Williams et al., 2009) ER can activate target gene expression indirectly through pro tein protein interactions with other cofactors via the AF 1 or AF 2 region (Baek et al., 2002; Shang et al., 2000; Jakacka et al., 2001; Mtivier et al., 2003) Such interactions are generally modulated by conformation al changes that acc ompany ligand bi nding or phosphorylation of ER and its protein partner ; a few examples include AP 1, SP 1 STAT and NF B (McDonnell and Norris, 2002) Rec ently discovered cofactors include epigenetic chromatin modifiers such as BRM, GCN5, and PCAF (Green et al., 2007) A complex, dynamic picture concerning ER mediated gene regulation is emerging. It supports step wise recruitment of transcription factors to cis elements, and ordered

PAGE 22

22 assembly of the transc ription initiation complex at those regions (Metivier et al., 2006; Shang et al., 2000) Non Genomic F unction ER can also influence cell proliferation through cytoplasmic signaling The s e pathways are associated with estrogenic functions that act a s part of a n extranuclear kinase cascade through membrane associated ER These signaling cascades include G protein coupled receptors (Pedram et al., 2006, 2007) cell membrane ion channel, tyrosine kinase c Src, ERK1/2, p38, JNK, PKA, PKC, PI3K and Notch pathway (Rizzo et al., 2008; Fu and Simoncini, 2008) The presence of extranuclear ER signaling is controversial (Losel et al., 2003) This is mainly due to numerous crosstalk between ER and kinase cascades as discussed earlier for ligand independent genomic effects. These networks make it difficult to experimentally test and confirm estrogenic action in the non genomic context. Other issues under discussion include whether membrane associated ER is the classic receptor tethered to the membrane by adaptor protein or is a variant protein A n amino terminus truncated iso form of the receptor ( ER46 ) has been implicated as such a variant in human endothelial cells (Li et al., 2003) A variant form of metastatic tumor antigen 1 (MTA1) was shown to sequester classical ER in the cytoplasm and enhanced non genomic function (Kumar et al., 2002) Present understanding of non genomic function of ER is incomplete, but it propo ses an integrated approach to understanding the full range of ER molecular mechanisms.

PAGE 23

23 Cell of Origin Clonal Evolution Model The cell of origin question has been much d ebated in cancer biology. Can a complex, heterogeneous tumor originate from a single cell? Or are there many point s of origin to account for the various cell types found? The clonal evolution hypothesis supports both single and multicellular origin of cancer cells (Shipitsin and Polyak, 2008; Campbell and Polyak, 2007; Shackleton et al., 2009) Clonal evolution is natural se l ection applied to tumorigenesis and loosely fits the "seed and soil" hypothesis. The basic framework of clonal evolution is a hierarchical model where a single cell can give rise to the whole tumor and continuously transform as it adapts to its environm ent At the top of the pyramid is a single transformed cell. It may acquire additional reproductive advantage with each cell division through inherent genetic instability characteristic of tumor cells. These abnormalities may be acquired through mutatio n, or general protein variations due to stochastic gene expression (Raj and van Oudenaarden, 2008) that optimize sur vival in its particular environment Th us, under favorable conditions, a single transformed cell may give rise to a tumor by gro wing quickly and aggressively. There may still be multiple subclones generated throughout this process but they would not be the dominant cell type. If the environm ent is less favorable, the condition may select for multiple subclones that compete for dominance and contribute to heterogeneity. The Cancer Stem Cell (CSC) Hypothesis The cancer stem cell (CSC)/ tumor initiating cell (TIC) hypothesis represents a parad igm shift in considering the natur e of the "seed" (Lobo et al., 2007) Th e fundamental change is instead of a cell that outgrows and ou t competes other cel ls as

PAGE 24

24 proposed by clonal evolution; it suggests a small reservoir of dedicated founder cells reside within tumors. These founder cells are believed to be capable of self renewal through asymmetric cell divis i on that is analogous to som atic stem cells by s haring certain genetic features (Visvader and Lindeman, 2008) In essence, the CSC/TIC hypothesis suggests a predictable, orderly cause for tumor initiatio n instead of a more random act through natural selection or genetic drift (made more favorabl e by the unstable genome of cancer cell s ). It also dictates that tumor initiation originates from the top of a hierarchical system, whereas in clonal evolution a tumor can a rise to multiple points of origin (Visvader et al., 2008) Tumor Heterogeneity and CSC Hypothesis The cancer stem cell (CSC)/tumor initiating cells (TIC) hypothesis gained prominence when they were first isolated in hematopoietic malignancies (Bonnet and Dick, 1997) The hypothesis sup poses there is a rare subpopulation of cells (< 2%) with stem like features within tumors. The characteristics include self renewal, the ability to differentiate, and resistance to therapy. These features can confer tumorigenic properties to TICs and may explain recurrence after therapy. However, CSCs themselves can evolve significantly from their original pool during disease progression (Rosen and Jordan, 2009; Hwang Verslues et al., 2009) It is equally possible that a tumor may contain both stem like cells and genomically unstable clones both vying for dominance (Campbell et al., 2007) To complicate the issue, there may yet be a graduated pool of tumorigenic lineage restricted progenitor cells (Lim et al., 2009) These considerations nullify m resembles the clonal evolution model. The CSC model thus represents tumorigenic

PAGE 25

25 potential in a given grou p of cancer cells, not their fate as each cell would still undergo significant adaptive changes (Shackleton et al., 2009) To reconcile the s eemingly contradictory viewpoints is to admit the two mode ls are not mutually exclusive. A tumor is a distinct and complex entity that remains d ependent upon its host. It may have originated from a founder cell, but it needs to adapt to its host environm ent in order to propagate itself. T he xenograph model typically used to evaluate tumorigenicity or CSCs may not provide the optimal microenvironment for the pool of varying CSCs, but simply select for a small fraction. If true, this indicates the potenti al pool and freque ncy of tumor initiating cells can be larger than present estimates (Kelly et al., 2007; Kern and Shibata, 2007; Alison et al., 2009) It is reasonable to suggest tumor initiating cells can undergo clonal evolution and evolve into develop into a complex solid tumor. Therefore, it is counterproductive to be limited by semantics when a broad range, integrated approach is required to target a tumorigenic pool rather than single cells In addition, it demonstrates the necessity for a more pr ecise vocabulary to be introduced. Implications of Each Model on Breast Cancer Recurrence T he CSC hypothesis support s targeting th erapy towards the most tumorigenic cells The reasoning is to provide more effective treatment and inhibit tumor recurrence, which tends to produce cancer cells with an aggressive phenotype. This idea assumes that tumorigenic cell s are at the top of a hierarchy that give s rise to non tumorigenic cells at the lower tiers Therefore eliminating the highly tumorigenic subpopulati on would render the remaining cells more susceptible to standard antihormonal or chemotherapy treatments.

PAGE 26

26 Under the clonal evolution model, the large pool of heterogeneous tumor cell s should be the target. This is also applicable to the potentially mixe d CSC population. It pose s a bigger challenge as it will be necessary to ident ify common elements in the tumorigenic pool for effective therapeutic designs Pinning down a malignant signature in these cells (regardless of origin) can be a daunting task. However, there is progress in that direction as tumorigenic cells share many molecular cascades with stem cells and vice versa. A number of oncogenic and self renewal signaling pathways associated with breast CSCs have been identified. A few examples in clude the Wnt pathway (Chen et al., 2007; Rappa and Lorico) Hedgehog (Liu et al., 2006) Notch ( Farnie et al., 2007; Dontu, Jackson, et al., 2004) and receptor tyrosine kinases (Vera Ramirez et al., 2010; Farnie et al., 2007; Rappa et al.) NF B (Zhou et al., 2008) PTEN/mTOR (Zhou et al., 2007) and CDKI (Pei et al., 2009; Liu et al., 2009) Ongoing research strives to understand pathways in stem cell maintenance and their mis regulation, which may contribute in sight into oncogenic pathways. T hese studies have yielded therapeutic developments that target breast CSC as well as other types of cancers. These include molecular signatures (Liu et al., 2007) immunoth erapy (Morrison et al., 2008) nutritional phytochemicals (Kakarala et al., 2009) and high throughput drug screening (Gupta et al., 2009) Estrogen receptor alpha, stem cells and breast cancer The heterogeneous nature of CSCs/TICs themselves suggest varied cell of origin. They may be derived from multipotent cells, from the range of lineage restricted progenitor cells that arise over the course of differentiation or from transformed cells that ac quired stem like features Within the framework of ER positive breast cancer

PAGE 27

27 recurrence, it is possible f or ER positive cancer cells to acquire stem like properties which allow for perpetual self renewal, evasion from antiestrogen inhibition and recurre nce of hormone sensitive tumors. This may account for the approximately 30% recurrence hazard after 5 years of initial antiestrogen treatment (Dignam et al., 2009; EBCTCG, 2005) ER expressing cells with highly tumorigenic features may shed light on this observation. If such a population exist, then it is worthwhile to characterize its response to antiestrogen in order to evaluate antiestrogen efficacy in this context. Developmenta l studies have yielded some clues B reast epithelial cells undergo a differentiation hierarchy similar to that of hematopoietic cells (Figure 1 2) The source starts with the undifferentiated mammary stem cells (MaSCs). They are maintained in the mammar y tissue via self renewal and differentiate to bi potent progenitors and onward to the two major cell subtypes: myoepithelial and luminal progenitors. These short termed stem cells are maintained throughout life and contribute to hormone dependent remodel ing of the breast during puberty and pregnancy (Vargo Gogola and Rosen, 2007) However, it is unknown whether hormone stimulation of progenitor cells is ER dependent as evidence are contradictory. Mouse mammary stem cells have been demonstrated as ER negative, but are hormone responsive (Asselin Labat et al., 2006, 2010; Booth and Smith, 2006) Th ere is contrary evidence as well, with in vivo evidence of a small number of slow cycling cell tha t are ER positive (Clarke et al., 2005; Smith, 2005) The ongoing debate continues as one side argues th at ER positive cells found duri ng fetal development only serve to regulate ER negative stem cell maintenance v ia paracrine mechanisms (LaMarca and Rosen, 2 008) while the opposition maintain s ER positive progen itor cells not only proliferate and differentiate

PAGE 28

28 in resp onse to estrogen but also produce paracrine factors that influence nearby ER negative cells (Clark e et al., 2005; Dontu, El Ashry, et al., 2004; Booth et al., 2006; Wicha, 2008) The population of ER positive progenitor cells is believed to be present only during a small developmental window, and these cells are necessary for early mammary gland development. The idea that these committed progenitors give rise to TICs is gaining ground A ccumulating evidence from genetic profiles of mammary stem cells have been compared to human tumor datasets (Lim et al., 2009; Raouf et al., 2008) These studies demonstrated that aberrant luminal progenitors with limited estrogen receptor expression are found t o correlate with basal like tumors (Lim et al., 2009) The implications are that the various breast c ancer subtypes may have corresponding partners in the stem cell hierarchy (Figure 1 3) As additional information concerning prog enitor transformation emerges, it will generate a richer picture for understanding tumor initiation and a wider array of tools for their precise identification. Further support is emerging as gene expression profiling using patient data confirms that trad itional histological classification is a crude prognostic indicator, as even ER positive tumors can have varying degrees of malignancy which molecular profiling can discern (Sotiriou et al., 2003, 2006) Molecular profiling information would be useful in diagnosing and tailoring the most effective treatment plan for each patient. Contribution of CSC /T IC to antiestrogen resistance Antiestrogens have been the treatment of choice for ER positive breast carcinomas since the 1970s. These tumors tend to be estrogen dependent and

PAGE 29

29 blocking estrogen signaling with antiestrogens has proven effective. Patients with antiestrogen responsive tumors generally have the best prognosis and survive the longes t after initial diagnosis. However, resistance to antiestrogen therapy has substantially hindered long term use of antiestrogen compounds. It is interesting to c onsider if a potential ER positive stem like cancer cell may be antiestrogen sensitive as well. One of the key cr iteria of a CSC/TIC is resistance to chemotherapy and radiation (Phillips et al., 2006; Grimshaw et al., 2008; Fillmore and Kuperwasser, 2007; Li et al., 2008) So, does ER signaling supersede self renewal pathways? If it do es not it may explain why prolong ed antiestrogen treatment can generate non responsive ER positive cells. It would confirm the presence of an intrinsically different cellular subgroup that is respons ible for tumorigenesis and this subgroup may utilize alternative signaling pathways However, if ER signaling s upersedes self renewal signaling then one may suppose antiestrogen therapy can be effective against less differentiated ER positive cell s as well as against fully differentiated luminal cells. The culmination of specific cofactor expression and activity patterns, ligand availability, receptor conformation, and growth factor signaling w ithin each particular cell type determine their response t o antiestrogen inhibition. It is conceivable that the stem like cells at various differentiation stages would have varying degrees of antiestrogen sensitivity. Further characterization of antiestrogens in stem like cell s may provide a new perception into the development of antiestrogen resistance. To focus on the role of ER in this study, o nly antiestrogens relating directly with ER will be further discussed Other forms of antiestrogen therapy not related to the receptor, such as aromatase inhibitors, will be omitted. There are two major classes of

PAGE 30

30 antiestrogens that bind directly to ER The first class is selective estrogen receptor modulators (SERMs). They are classic antagonists specific for the estrogen receptor, although they may have agonist effects in a cell specific manner. The agonist activity induced by SERMs has been attributed to ER conformation changes upon SERM binding that may recruit coactivators in certain cell types (Jordan, 2007) SERMs have a triphen yle thy lene chemical structure and a higher binding affinity to ER compared to 17 estrodiol (Clarke et al., 2003) Tamoxifen (TAM) was the first FDA approved compound for use against hormone dependent breast cancer in pre an d postmenopausal women. It is a triphenylethylene that is metabolized in the liver by cytochrome P450 enzymes t o the metabolically active form 4 hydroxytamoxifen (4 OHT) (Jordan, 2007) 4 OHT has a strong er binding affinity for ER LBD than TAM Many other SERMs have been developed over the years, the most promising of which is raloxifene, a benzothiophene derivative as it has no known agonist activity (Lewis and Jordan, 2005) However, raloxifen and other benzothiophene derivatives have lower binding affinity to ER and have a short half life (Jordan, 2007) The second class of compounds is select ive estrogen receptor down regulators (SERDs). They are often referred to as pure antiestrog ens because they have no known agonis t effects SERDs function under a completely different mechanism from SERMs. They impair receptor dimerization and induce con formational changes that expose ER receptor and promotes degradation via the ubiquitination pathway (Ariazi e t al., 2006) Fulvestrant (ICI 182780) is the most widely studied drug of the SERDs Clinical trials comparing fulvestrant to TAM have concluded that t here were no significant differences

PAGE 31

31 in efficacy between the two drugs. Fulvestrant has become anothe r option for patients that developed resistance to other lines of therapy; however clinical data has not shown it to be better than other therapies (Ariazi et al., 2006) CSC Isolation and Detection The premise of the CSC/TIC model has been keenly applied towards the study of breast cancer recurrence. The initial toolbox consisted of BrdU label retentions and Hoechst dye efflux to isolate a slo w cycling "side population". This method was used in mouse mammary stem cell studies and it was noted that less differentiat ed cells overexpress multi drug resistant transporters (Welm et al., 2002) The tools have since expanded as additional molecular markers were identified. A tumorigen ic subpopulation was identified and isolated as CD44+CD24 Lin (Al Hajj et al., 2003) in human breast carcinomas CD44 and CD24 are not the only stem like markers proposed; they are just the most commonly used in breast cancer studies These markers have formed the basis fo r isolation of mammary stem cells in mice (Stingl et al., 2006; Regan and Smalley, 2007) rat (Zucchi et al., 2007) and humans (Liu et al., 2 006) These markers were adopted and combined with neural stem cell culture technique to generate an in vitro culture system that enriches for stem/progenitor cells from human mammary tissue (Liu et al., 2005) referred to as mammosphere culture. It basically is growing cells at low density in serum free, defined media under suspension condition to generate clonal spheres The mammosphere culture method h as also consistently propagate d tumorigenic cultures from established cell lines (Ponti et al., 200 5; Fillmore and Kuperwasser, 2008) A recent study provided preliminary evidence correlating differential cellular adhesion to stem like/mesenchymal properties (Walia and Elble, 2010) Although the underlying mechani sm is not understood, it indicates that cell

PAGE 32

32 adhesion and culture environment can drive cells towards a stem like/mesenchymal phenotype. This may provide an explanation for the effectiveness o f TIC enrichment in mammosphere culture. The debate concerning the genuine uti lity of the CD44 marker is ongoing Several studies have noted that CD44 expression i s not unique to stem cell (Shipitsin et al., 2007; Hwang Verslues et al., 2009; Wright, Robles, et al., 2008) and tha t there is a significant level of hete rogeneity within CD44 positive cell fractions (Shipitsin et al., 2007; Pece et al., 2010; Park et al., 2010) However, CD44 expression has been consistently linked to a more progenitor like tumorigenic subpopulation (Shipitsin et al., 2007; Sheridan et al., 2006; Abraham et al., 2005) and has potential prognostic value (Liu et al., 2007) The relative enrichment is dependent on cellular context. Cells isolate d in this manner have express ed some stemness properties, but their relative abundan ce in a given population and their function require further study (Zucchi et al., 2008; Wright, Calcagno, et al., 2008) A new addition to the arsenal is the identification of aldehyde dehydrogenase 1 (ALDH1) that is ove rexpressed in tumors, particularity in the highl y tumorigenic and undifferentiated fraction (Ginestier et al., 2007) ALDH1 has been further evaluated in established cell lines a nd initial results were c onfirmed (Charafe Jauffret et al., 2009) This marker has also been applied to other cell types as CD44 has been. Its validity as a CSC marker has likewis e been contested because ALDH1 expression varies widely depending on cell type and microenvironment (Neumeister and Rimm, 2009) Other Examples of CSC Enr ichment Results gathered using the existing markers are imperfect, and will s urely be re evaluated when new methods become available. In addition to markers mentioned

PAGE 33

33 above CD133 (Wright, Calcagno, et al., 2008) has also been used to identify stem/progenitor fractions fr om epithelial cells in other tissues including colon (Vermeulen et al., 2008) bladder (Chan et al., 2009) and prostate (Shi et al., 2007; Wang et al., 2009) Similar issue s concerning cell of origin and heterogeneity is also being d iscusse d in these models. The overall pictur e emerging from CSC/TIC identification is the obvious paucity of available biomarkers and the li mited application of each. It has not constrained research in this area because regardless of the real existence of cancer stem cell s it is useful to isolate and understand the more persistent cell types in solid tumors The main obstacle has been t he difficulty in isolating those fractions. The tools currently being developed will contribute to understanding the rarer subgroup s and help develop long term st rategies for disease management.

PAGE 34

34 Figure 1 1. Estrogen receptor isoforms (Ariazi et al., 2006) Amino acid sequence identity is shown a s percentage homology relative to ER Abbreviations: DBD is DNA binding domain. LBD is ligand binding domain. Figure 1 2 Mammary gland differentiation hierarchy (Visvade r, 2009) Model of mammary differentiation hierarchy with mouse primary cell surface markers in blue, and human primary cell surface markers in red. The common progenitor cell is believed to be a potential bipotent cell restricted to mammary development

PAGE 35

35 Figure 1 3 Potential relationshi p between mammary differentiation hierarchy and the development of tumor subtypes (Visvader, 2009) A hypothetical model for the development of various breast cancer subtypes from a hypothetical mammary differentiation hierarchy. It supposes less differentiated tumor subtypes are derived from oncogenic transformation of lineage restricted progenitor cell s, while highly differentiated subtypes are derived from fully differentiated luminal cell types.

PAGE 36

36 CHAPTER 2 GENERAL MATERIALS AN D METHODS Cell Culture MCF7 cell from ATCC were grown as monolayer culture in DMEM media (MediaTech) supplemented with 10% bovin e calf serum (Hyclone), 1% L glutamine (MediaTech), and 1% penicillin streptomycin solution (MediaTech). They were cultured in tissue culture grade petri dishes. To generate MCF7 mammospheres (MCF7S), MCF7 parental cells were first washed with PBS. They were trypsinized with 0.05% trypsin/EDTA (MediaTech) for 5 minutes. The trypsin was quenched with whole DMEM media, stained with 0.4% solution trypan blue (Gibco) for 5 minutes to exclude dead cells, and manually counted using a hemocytometer to calculat e cell density according to the formula: cell number count (average) x dilution factor x 10 4 = cells/ml The cells were diluted to 5000 cells per ml in PBS and washed twice. The cell pellet was resuspended in defined mammosphere media and seeded onto Pol y HEMA coated dishes to prevent attachment. Pol y HEMA working solution was a 2X so lution diluted from 10X stock s olution in 95% ethanol. The 10X Poly HEMA stock solution was composed of 0.12 g/ml Poly HEMA dissolved in 95% ethanol at 60 C. MCF7 mammosphere s (MCF7S), derived from mammosphere culture selected MCF7 cells, were maintained in 50:50 DMEM/F12 media (MediaTech) supplemented with 20 ng/ml EGF (Sigma), 10 ng/ml bFGF (Sigma), 5ug heparin (Sigma), 1% penicillin streptomycin solution (MediaTech), 1% L glutamine (MediaTech), and 1X B 27 supplement (Invitrogen). Cells were trypsinized every 6 7 days and passaged at 10,000 cells per ml onto Poly HEMA coated dishes. For frozen cell stocks, MCF7S were

PAGE 37

37 dissociated with trypsin and resuspe nd in desired amount of mammosphere culture media containing 10% DMSO, and stored at 80 C. MCF7S Cell Proliferation and Sphere Formation Assay Serial Passaging For each passage, the MCF7S cells were spun down from suspension at 700 rpm for 6 minutes and the supernatant was removed. The cell pellet was resuspended in 50 l 0.05% trypsin/EDTA for 2 minutes, and then actively dissociated by pipetting up and down. The digestion was quenched by adding 150 l of mammosphere media. 20 l of cell suspension was taken for trypan blue viability assay and counted. The cell suspension was diluted to 10,000 cells per ml and plated with desired antiestrogen or vehicle on Poly HEMA coated tissue culture dishes. The cells were incubated undisturbed for 7 days before repeating dissociation and cell counting. The fold change was calculated as final cell de nsity / initial cell density. The fold change for each passage is used to calculate cell expansion. To calculate cell expansion, the fold change for each passage was multiplied by the cell densit y in each preceding passage. The final results were expres sed on a semilog graph and the slope was calculated as log expansion to determine growth rate changes. Cell Proliferation A ssay MCF7S cells were spun down from suspension at 700 rpm for 6 minutes and the supernatant was removed. The cell pellet was re suspended in 50 l 0.05% trypsin/EDTA for 2 minutes, and then actively dissociated by pipetting up and down. The digestion was quenched by adding 150 l of mammosphere media. 20 l of cell suspension was taken for trypan blue viability assay and counted. Cell suspensio n was

PAGE 38

38 diluted to 5,000 cells per ml and was plated with desired drug or vehicle on Poly HEMA coated tissue culture dishes. The cells were incubated undisturbed for 2, 4, or 6 days before harvesting for viable cell count and replating for sphere formation assay (Figure 2 1) Sphere Formation A ssay Directly after viable cell count for the cell proliferation assay described above, the cell suspension was diluted in 2 ml of mammosphere media to a cell density of 5000 cells per ml. For each treatment condition 200 l of the cell suspension was allocated per well for one column (8 wells) of a 96 well plate with no Poly HEMA coating. There was an estimated 1000 cells per well at this point. 100 l of the cell suspension in each well was diluted to 2 adjacent w ells, which contained 100 l of mammosphere media. The result was a final 1:1 dilution in 16 wells; each well contained an estimated 500 cells. The estimated total number of cells seeded was 8000 cells per condition (Figure 2 1) Plated cells were incub ated undisturbed for 6 days. Phase contrast pictures were then taken at 50x magnification and sphere size and number were manual ly evaluated from the images using ImageJ software. Sphere formation frequency was calculated using the formula: (number of sp heres/number of total cells plated) x 100%. RNA Isolation Samples were collected and homogenized by vortex in 0.5 ml Trizol reagent (Invitrogen) to obtain total RNA. 0.1 ml of chloroform was added to each homogenized samples. The samples were then cen trifuged at 12,000 x g for 15 minutes at 4 C to separate aqueous: phenol chloroform phase. The aqueous phase was extracted from each sample. The RNA was precipitated with the addition of 75% isopropanol (v/v), and

PAGE 39

39 centrifuged maximum speed at 4 C for 15 minutes. The RNA pellets were washed with 1 ml of 75% ethanol and centrifuged to remove supern atant. The pellets were air dried for 5 minutes before dissolving in sterile filtered TE (10mM Tris pH 8, 1mM EDTA) and stored at 80 C. Reverse Transcriptase P CR (RT PCR) Approximately 1 g of total RNA was added to a reaction cocktail containing DEPC treated ddH2O, 2.5 M dNTP, and 5 nM random hexamer primers to 16 l final volume. The mixture was incubated at 70 C for 5 minutes, and quenched quickly on ice. 2 l of 10X RT buffer (NEB M MuLV), 1 l RNase Inhibitor (Promega), and 1 l M MuLV Reverse Transcriptase (NEB) were added to the reaction to a final volume of 20 l, which was incubated at 42 C for 1 hour. The reaction was heat inactivated at 95 C for 5 minutes, and was diluted 1:5 dilution with ddH2O. 1 2 l of diluted template was used for real time PCR. Real Time PCR Each real time PCR reaction was composed of the following: 1 2 l cDNA generated from RT PCR, 1 l of 5 M primer mix working solution, 8 l ddH2O, and 10 l 2X Sybr Green PCR Master Mix (Applied Biosystems). Triplicates were done for each reaction and results were expressed as relative quantitation normalized to beta actin expression as control. Duplicates of reactions with no template were also included on real time PCR plate for each run as negative control Gene expression differences larger than 2 fold are considered to be significant. The thermal cycling parameters were as follows: 95 C for 10 minutes, 40 cycles of 95 C for 15 sec for denaturing step and 60 64 C for 60 sec for primer extension, and a melting curve analysis was performed at the

PAGE 40

40 end of each run. StepOne (48 well), or StepOnePlus (96 well) Real time PCR machines (Applied Biosystems) were used for data collection. Pr imers used were listed in Table 2 3. Total Protein Isolation Total protein lysates were isolated from MCF7 or MCF7S cell by first washing cells in cold PBS twice, then adding 50 200 l lysis buffer (50mM Tris pH 7.5, 1mM EDTA, 1% (v/v) SDS, 1% 2 mercapto ethanol, 20mM dithiothreitol). The samples were boiled for 10 minutes, then 6X sample buffer (4x Tris SDS pH 6.8, 30% glycerol, 10 % SDS, 0.6M dithiothreitol, 0.012% bromophenol blue) was added. The samples were either stored at 20 C. The samples were boiled again for 5 minutes prior to loading onto SDS PAGE gel for analysis. Nuclear Extraction Cells were collected after antiestrogen treatment in cold PBS and rinsed twice. The cells were resuspended in 400 l hypotonic buffer (10 mM HEPES pH7.9, 10 mM KCl, 2 mM MgCl 2 0.1 mM EDTA, 0.5 mM Dithiothreitol, 1X protease inhibitor cocktail), and incubate d 15 minutes on ice. 30 l of 10% NP 40 was a dded to each sample and then vortex 10 times for 3 5 seconds each. The samples were centrifuged for 1 minute at 4 C. The cytosolic supernatant was collected in pre chilled microcentrifuge tubes for each sample, and stored at 80 C for later use. Each nuclear pellet was resuspended in 40 l nuclear extraction buffer (50 mM HEPES ph7.9, 50 mM KCL, 500 mM NaCl, 0.1 mM EDTA, 0.5 mM Dithiothreitol, 1X protease inhibitor cocktail). The samples were incubated at 4 C with rocking for 30 minutes. The nuclear fractions were collected in pre chilled microcentrifuge tubes after centrifuging at max speed for 1 minute at 4 C.

PAGE 41

41 The nuclear fraction was diluted 1:1 with hypotonic buffer before storage at 80 C for later use. Immunoblot Analysis Protein lysate and nuclear extracts were analyzed by immunoblot. The samples were resolved by first adding the appropriate amount o f 6 X sample buffer and boil ed for 5 minutes. The samples were loaded onto 10% Tris HCl polyacrylamide separating gels /4% stacking gel at 1mm thic kness for electrophoration in 1X running buffer (25 mM Tris, 190m M glycine, 0.2% SDS). The gel wa s then electro transferred onto polyvinylidene fluoride (PVDF) membrane using Trans Blot Semi Dry Electrophore tic Transfer Cell (BioRad) in 1X transfer buffer (20 mM Tris, 192mM glycine, 10% methanol). Membranes were stained with Fast Green (0.1% Fast Green FCF, 50% met hanol, 10% acetic acid) for 5 minutes at room temperature to ensure transfer and equal loading Stained membranes were washed 2 3 times in TBST (30mM Tris pH 7.5, 200mM NaCl, 0.1% (v/v) Tween 20), and incubated in 3% (w/v) non fat dry milk in TBST blockin g solution for 30 minutes at room temperature. Blocked membranes were rinsed in TBST before probing with diluted primary antibody in TBST for 30 minutes at room temperature, or overni room temperature in TBST for 5 minutes each. They were next incubated in 1:10 000 diluted peroxidase conjugated secondary antibodies in TBST for 30 minutes at room temperature. The membranes were then washed 3 times at room temperature in TBST for 5 minutes each. Bound antibodies were detected by applying Pierce ECL substrate solution (Thermo Scientific) and exposing the membrane to X ray film.

PAGE 42

42 Immunofluorescence Cells were attached to glass cover sli ps prior to immunofluorescence staining procedure. Adherent cells were grown directly onto glass cover slips and treated with antiestrogens. Mammosphere cells were dissociated with trypsin, quenched with media, centrifuged to concentrate cells to 50 l media. The cell suspension was then gently applied onto glass cover slips in 6 well tissue culture plate, and centrifuged at 3000 rpm for 4 minutes to attach cells onto cover slips. The cover slips were rinsed with PBS twice, and then 3.7% formaldehyd e/PBS was added to cover slips to fix the cells for 5 minutes at room temperature. The cells were rinsed with 0.1% NP 40/PBS 3 times. The cell membranes were permeablized for 15 minutes at room temp with 0.1% NP 40/PBS. Next, they were incubated with 1: 50 diluted primary antibodies in 3% BSA/0.1% NP 40/PBS for 1 hour at room temperature. They were then rinsed 3 times with 0.1% NP 40/PBS for 5 minutes at room temperature. They were incubated with 1:500 diluted flurophore conjugated secondary antibodies i n 3% BSA/0.1% NP 40/ PBS for 1 hour at room temperature in the dark. The cover slips were rinsed 3 times with 0.1% NP 40/PBS for 5 minutes at room temperature, follow ed by a quick rinse with water, and then counterstained with 200 g/ml of Hoechst 33342 f or 5 minutes at room temperature. The cover slips were rinsed with water, and allowed to air dry in the dark for 30 minutes. The cover slips were mounted onto glass slides using Fluoromount G (Southern Biotech) and allowed to dry overnight at room temper ature in the dark. Glass slides were stored at 4 C.

PAGE 43

43 short hairpin RNA (shRNA) Vector Construction shRNA Oligo Design and C loning Oligos for shRNA construction were designed using shRNA psm2 designer at RNAi Central ( http://katahdin.c shl.edu/siRNA/RNAi.cgi?type=shRNA ). The accession number NM_000125.2 was entered for estrogen receptor alpha shRNA design. In order to generate a high fidelity oligo, it was broken into two fragments when ordering from Invitrogen (Table 2 1) The break s in the two fragments were designed with overlapping, complim entary loop regions so they anneal and extend during P CR into the full length oligo. The desiccated oligos were dissolved in TE buffer and combined to a final working solution of 1 Mir30 PCR primers were used for amplification using Phusion High Fidelity DNA Polymerase (Finnzymes) Mir30 primer sequences were as follows: forward 5' aagccctttgtacaccctaagcct 3' and reverse 5 actggtgaaactcacccagggatt 3'. The PCR was perfor followed by 35 cycles of 2) for 30 seconds, and 3) included a final round of extension After PCR, the f ragment was gel purified using Q IAquick Gel Extr action Kit (Qiagen). The purified fragment was digested with XhoI and EcoRI restriction enzymes The pLMP vector sequence and informatio n can be found in Figure 2 2 (

PAGE 44

44 digested f ragment and vector were mixed at a 7:1 ratio, respectively and ligated with 1 l of T4 DNA ligase and 1X ligase buffer (NEB) in 10 l final volume for 1 hour overn heat onto LB Plasmid DNA MiniPrep 1 ml of the overnight culture was transferred to a microcentrifuge tube. The bacterial cells were centrifuged at 8000 rpm for 1 minute. The cell pellets were resuspended with 200 l P1 Buffer (50 mM Tris Cl pH 8.0, 10 mM EDTA) each. The cells were then lysed by adding 200 l P2 Lysis Buffer (200mM NaOH, 1% SDS w/v) to each tube and mixed by inversion. Next, 200 l P3 Neutralization Buffer (3 M potassium acetate) were added to each tube and mixed by inversion. The samples were centrifuged at 14000 rpm for 5 minutes to pelle t precipitated proteins. 0.5 ml of the supernatant was transferred to a new tube and reserved. The protein pellet was discarded. 1 ml of 100% ethanol was added to the reserved supernatant and mixed to precipitate DNA. The precipitated DNA was pelleted by centrifugation at 14000 rpm for 10 minutes. The supernatant was discarded. The DNA pellets were washed once in 1 ml 70% ethanol and the supernatants were remove d. The DNA pellets were air dried for 5 to10 minutes. The DNA was dissolved in 30 to 50 l TE (1 M Tris Cl, 0.5 M EDTA

PAGE 45

45 pH 8) containing 20 g/ml RNase A. Purified DNA was sequenced to confirm shRNA oligo insertion prior to retrovirus production. Retrovirus Production and Transduction of Target Cells Transient Transfection To generate retrovi ral particles, the shRNA vector (Figure 2 2) containing the specific knockdown sequence was transiently transfected into transformed HEK293 cells called Phoenix. The Phoenix cells overexpress retroviral proteins Gag Pol and Env to facilitate packaging of shRNA into retroviral particles and increase their production. Phoenix cells were maintained in DMEM media (MediaTech) supplemented with 10% bovine calf serum (Hyclone), 1% L glutamine (MediaTech), and 1% penicillin streptomycin (MediaTech). Cell s were s eeded at ~60% confluency 24 hours prior to transfection. They were transfected using Fugene 6 Transfection Reagent (Roche) by first diluting 2 l of the transfection reagent in 100 l PBS. 1 g of DNA was also diluted in 100 l PBS. The two diluted reag ents were combined and incubate d at room temperature for 30 minutes to form a cationic lipid mediated transfection complex. The complex was added directly to cells dropwise. The cells were then incubated from 12 hours to overnight before switching to fre sh media. The cells were incubated for an additional 48 hours. The virus containing media was collected and passed through a 45 m filter to exclude cell debris. The viral media was aliquot ed and used immediately to infect target cells. Excess aliquots were stored at 80 C, or disposed after bleaching. Retroviral Infection for Stable Integration Target cells were tryp s inized and plated 24 hours prior to retroviral infection. Adherent cells were infected by replacing culture media with the infection cocktail, which

PAGE 46

46 consisted of 1:1 viral media: culture media and 4 g/ml polybrene. The cells were incubated for 24 hours, and then th e infection cocktail was replaced with fresh media. The cells were incubated for an additional 24 hours. In order to infect suspension cells, they were first centrifuged to remove culture media. The cell pellet was resuspended in the infection cocktail a nd plated on Poly HEMA coated tissue culture dishes. The cells were incubated for 24 hours, and then the infection cocktail was replaced with fresh media. The cells were incubated for an additional 24 hours. After the 24 hours of recover y in fresh media, the cells were treated with 2 g/ml puromycin dihydrochloride (Cellgro) to begin selection of transformed cell. Uninfected target cells were treated in parallel to estimate selection completion, which is typically complete 48 hours after addition of puro myc in. The s election media was replaced with fresh culture media after selection and transformed cells were allowed to expand to desired density. The cells were then try p sin i zed and dissociated into a uniform suspension and aliquot ed The transformed ce ll stocks were stored in freezing media (bovine serum albumin containing 10% v/v DMSO) at 80 C, maintained as polyclonal culture, or further selected for monoclonal culture Selection and Enrichment of Single C lones To select single clones in mammosphere culture, the spheres were dissociated with trypsin to a single cell suspension. The cell sus pension was first diluted to 1000 cells per ml, and then distributed 200 l per well into 2 columns of a 96 well plate. The 2 columns were serially diluted 100 l down the remaining columns of the plate, which also contain ed 100 l per well. The final di lution was < 5 cells per well in 200 l. The

PAGE 47

47 cells were incubated and media was added or replaced as needed until single colonies can be visualized. Each clone was manually pic ked using micropipet tips into each well of a 24 well plate (Poly HEMA coated) The clones were allowed to expand until sufficient cells were obtain ed to ascertain knockdown using immunoblot analysis. Analysis of CD44 Expression For each sample, approximated 1x10 6 cells were trypsinized to obtain single cell suspension. The cells were washed twice in cold PBS + 1% BSA. The cells were incubated at 4 FITC antibody or isotype FITC control antibody in PBS + 1% BSA for 45 minutes in the dark. After incubation, the cells were washed twice in cold PBS + 1% BSA. The cell pellet was resuspended in 400 l cold PBS + 1% BSA fo r flow cytometry analysis within 1 hour. BrdU/Propidium Iodide Cell Proliferation Assay An asynchronous MCF7S culture of 1 x 10 6 cells w e re trypsin iz ed and cells were pulsed with 10 M BrdU for 2.5 hours. Duplicates of samples without BrdU pulse were incl uded as BrdU staining controls. They were washed in cold PBS, and fixed overnight with cold 70% ethanol at 4 C in the dark. The samples were stored at 20 C, protected from light, until BrdU staining. On the day of staining, the cells were centrifuged a nd the ethanol was removed. They were then washed once in PBS + 1% BSA. To denature the DNA and expose BrdU antigen, the cells were treated for 30 minutes in 2 M HCl/0.5% Triton X 100 at room temperature. The cells were centrifuged and the supernatant w as removed, and the cells were washed once in 0.1 M Na 2 B 4 O 7 pH 8.5. The supernatant was removed after centrifugation, and the cells were incubated with 1:50 dilution anti BrdU FITC antibody in 0.5% (v/v) Tween 20/1% BSA/PBS containing

PAGE 48

48 100 g/ml (w/v) RNase A. The samples were incubated in the dark at room temperature for 1 hour. The samples were washed twice in PBS + 1% BSA and resuspended in PBS + 10 g/ml propidium iodide (Sigma). The samples were incubated at room temperature for 15 m inutes in the dark, and kept on ice for less than 1 hour prior to flow cytometry. PI cell cycle modeling and data generation was performed using ModFit LT software. Annexin V PE Apoptosis Detection using BD Pharmingen Kit Cell s were washed 2 times in co ld PBS, and then resuspended in 1X hypotonic binding buffer (BD Pharmingen) at 1x10 6 cells per ml. 100 amino actinomycin D (7 AAD) were added to each tube. Then the tubes were incubated at ding buffer was added to each tube. The samples were processed in a flow cytometer within 1 hour. Cells treated with 7 AAD alone, Annexin V alone, and unstained cells were used as controls to set up compensation and quadrants. Statistical Analysis Sta tistical analyses were derived from at least three i ndependent experiments. Error bars for three independent experiments were presented as stan dard error of the mean (SEM), and statistically significant differences were determined using the te st. In Vivo Tumorigenic Assay The following in vivo mouse injection protocol was adopted from Brian Morrison, advised by Dr. Alejandro Lopez, QIMR, Griffith University, AU.

PAGE 49

49 Preparation of M ice A colony of NOD.Cg Rag 1tm1Mom Il2rg tm1Wjl /SzJ mice (The Jacks on Laboratory, Bar Harbor, Maine, USA) stock number 007799, were maintained and 5 to 8 weeks old littermates were used for the experiment, when possible 17 estradiol slow releasing pellets were i mplanted into the mice one day before injection. Preparation of Tumor C ells First, cell stocks were thawed in 37 C water bath and cells were washed once in RPMI Then, the media was removed and the cells were wash ed with PBS. Next, the cells were resuspended in PBS and a viability assay was performed usi ng trypan blue and Countess cell counter For each injection, the cells were mixed in PBS 1:1 with M atrigel (BD Biosciences) to a final cell concentration of 2 x 10 6 in 100 l. Mouse Injections The inoculation area of the mice was cleaned and sterilized with ethanol The cells were mixed and draw n into a cold 1cc syringe without a needle to prevent cell damage. A 26.5 gauge needle was used to inject 2 x 10 6 cells subcuta neously (s.c.) into the lower right flank of the mice. Mice were monitored twice a week for overall health. Tumor diameter s were measured with a digital caliper 4 weeks after injection and once a week thereafter. The tumor volume in mm 3 was calculated u sing the formula: Volume = width 2 x length x 0.5.

PAGE 50

50 Figure 2 1. Tissue culture scheme for MCF7S proliferation assay and sphere formation assay The growth curve was performed by first seeding viable MCF7S cell suspension at clonal density and cultu red in the presence of 4 hydroxytamoxifen or ICI 182780, then incubate for 2, 4, or 6 days. Viable cells were counted at each time point, and replated in fresh media at a defined cell density (~500 cell per well) in a 96 well plate. The cells were incuba ted for 6 days and sphere formation frequency was assessed.

PAGE 51

51 A B Figure 2 2. C loning vector information for microRNA adapted retroviral vector A) Vector map and unique restriction sites o f MSCV LMP cloning vector. B) Xho1 EcoR1 cloning s ite for shRNAmir expression using retroviral and PSI ( ) promoter. PGK promoter (Ppgk) drives expression of selection cassettes. Puro r cassette allows for selection of stable integrates. IRES GFP served as marker for stable integration Abbrevia tions: LTR is long terminal repeats. MiR is microRNA. Ppgk is phosphoglycerate kinase promoter. Puro is puromycin. IRES is internal ribosome entry site. GFP is green fluorescence protein.

PAGE 52

52 Table 2 1. shRNA O ligos designed for retrovirus mediated knockdown Oligo Name Gene Targeted Sequence Start Position shER 1F ER TGCTGTTGACAGTGAGCGAAGGGAGAATGTTGAAACACAATA GTGAAGCCACAGATGTA 1143 shER 1R ER TCCGAGGCAGTAGGCAGAGGGAGAATGTTGAAACACAATACA TCTGTGGCTTCACTA 1143 shER 3F ER TGCTGTTGACAGTGAGCGCGGAGTTTGTGTGCCTCAAATCTA GTGAAGCCACAGATGTA 1689 shER 3R ER TCCGAGGCAGT AGGCAAGGAGTTTGTGTGCCTCAAATCTACA TCTGTGGCTTCACTA 1689 Control shRNA F Tbx2 TGCTGTTGACAGTGAGCGAGCCAAGTATATCCTGCTGATGTA GTGAAGCCACAGATGTA 724 Control shRNA R Tbx2 TCCGAGGCAGTAGGCAGGCCAAGTATATCCTGCTGATGTACA TCTGTGGCTTCACTA 724

PAGE 53

53 Table 2 2. Antibodies u sed for Immunoblotting (IB), Immunofluorescence (IF), or Flow Cytometry (FC) Antibody Species Isotype Company Catalogue Number Use Alexa Fluor 488 FITC Rabbit Goat IgG Molecular Probes (Invitrogen) A11034 IF BrdU FITC (PRB 1) Mouse IgG eBioscience 11 6071 41 FC CD44 FITC (G44 26) Human Mouse IgG BD Pharmingen 555478 FC (HC 20) Human Rabbit IgG Santa Cruz Biotechnology SC 543 IB (MC 20) Human Rabbit IgG Santa Cruz Biotechnology SC 542 IF HRP Donkey anti mouse Mouse Donkey IgG Jackson ImmunoResea rch 715 035 150 IB HRP Donkey anti rabbit Rabbit Donkey IgG Jackson ImmunoResearch 711 035 152 IB Iso FITC Mouse IgG2b Molecular Probes MG2b01 FC LSD1 Human Rabbit IgG Cell Signaling C69G12 IB Tubulin Human Mouse IgG Sigma T9026 IB Table 2 3 Primers used for real time RT PCR Name Forward Reverse Tm ( C) hER CCGGCATTCTACAGGCCA TCGGTCTTTTCGTATCCCAC 60 TGTCTGCAGCGATTACGCA GCGCCGGTTTTTATCGATT 60 hTFF1 GTACACGGAGGCCCAGACAGA AGGGCGTGACACCAGGAAA 64 hCTSD CTGCACAAGTTCACGTCCAT ACTGGGCGTCCATGTAGTTC 60

PAGE 54

54 CHAPTER 3 CHARACTERIZING MAMMO SPHERES DERIV ED FROM MCF7 PARENTAL CELLS Introduction Despite the assumption that breast cancer cell lines are relatively homogeneous, the cells in tissue culture are highly dynamic and heterogeneous in reality (Lacroix and Leclercq, 2004; Burdall et al., 2003) It has been demonstrated that tissue culture conditions can select and enrich for different dominant ce ll type s from the original source (Osborne et al., 1987; Ince et al., 2007) One of the key point s in the cancer stem cell (CSC)/tumor initiating cell (TIC) hypothesis has been that the re exists an inherently more tumorigenic cell population in a particular tumor. This population is believed to be more resistant to treatment and may reseed the tumor at a later date, thus con tributing to tumor recurrence. Previous studies have take n advantage of the innate heterogeneity in tissue culture and used mammosphere culture techniques to enrich for potential TICs from established cell lines This included MC F7 cells (Ponti et al., 2005) which is an ER positive breast cancer epithelial cell line. The mammosphere enriched culture contained cells that were characterized as less diff erentiated more resistant to conventional antitumor treatments and more tumorigenic (Phillips et al., 2006; Fillmore et al., 2008) In the context of tumor recurrence following antiestrogen t reatment, It has been hypothesize d that ER positive TIC s may be responsible for antiestrogen resistance and tumor recurrence in ER expressing cancers (Dontu, El Ashry, et al., 2004) This id ea has been controversial because evidence for the existence and role of ER positive stem cells in primary culture were contradictory This is not surprising as this cell group may exist in a very limited window during mammary development, as

PAGE 55

55 dis cussed in Chapter 1. The group of ER positive progenitor cell suits the requirements for TICs as they are sufficiently de differentiated to maintain stem like features such as self ren ewal and bipotency However, they may express functional hormone rece ptors that can respond to antihormonal treatments. To test this idea, we first generated mammospheres from MCF7 cells and determined their ER status MCF7 was chosen because there are few ER positive cell lines currently available for study. Also, th e relationship between MCF7 derived mammospheres (MCF7S) and potential TICs has been supposed by earlier studies The human breast tumorigenic marker CD44 antigen expression (Al Hajj et al., 2003) has been correlated with MCF7S sphere formation frequency and stem ness markers like Oct4 has also been associated with MCF7S (Ponti et al., 2005) Therefore, MCF7S may serve as a model for studying antiestrogen response in potential TICs in ER breast cancer. Results Characteri zation of Mammospheres (MCF7S) D erived from MCF7 Cells MCF7 deri ved mammospheres were derived by first plating MCF7 parental cells (MCF7P) at 5000 cells per ml cell density in defined serum free media under suspension conditions as described in Materials and Methods The MCF7S cells have a doubling time of 2.5 3 days under these conditions ( Figure 3 1 ). As they divide, each cell form ed increasingly larger spheroids and extended microspikes ( Figure 3 2A; Figure 3 2 B). The microspikes are presumed to be microfilaments that the spheroid cells use to detect nutrient avai lability in the surroundi n g e nvironment. Eight days after initial plating, sphere s were typically over 50 microns in diameter and formed a

PAGE 56

56 multicellular mass (Figure 3 2C; Figure 3 2 D). The culture wa s maintained without additional growth factor suppleme ntation or culture media change over the course of a week from initial plating until next passage. Initial attempts of growth factor supplementation resulted in the swelling of cells and subsequent lysis. The mammosphere s were serially passaged to eval uate their growth kinetics as they transforme d from MCF7 parental cells. MCF7P cells were seeded at a density of 10000 cells per ml at each passage, and viable cells were counted using trypan blue exclusion after 1 week in culture. The cells were thus tr acked for 6 weeks and the data was plotted as fold change in cell density between final cell count and initial seeding (Figure 3 3 ). Two independent experiments were performed and they demonstrated that MCF7 parental cells undergo a rapid expansion in pro liferation during the first 2 3 weeks in mammosphere culture. However, as the se lection progresses a do minating subgroup emerges with stable growth kinetic After 3 weeks, the initial MCF7 parental cell culture transformed into the MCF7S culture enriched with a n alternate dominant subgroup and has predictable growth rate of roughly 200 fold change per passage. Expression of Putative Breast Tumorigenic Marker CD44 in MCF7S Cells The CD44 antigen was first described as an integral cell membrane glycoprotein with a role in cellular attachment to the extracellular matrix via hyaluronate (Aruffo et al., 1990) Subsequent studies have demonstrated a key role for CD44 during develo pment and tumor formation as a necessary component for cellular migration and invasion (Jin et al., 2006; Schmits et al., 1997; Godar et al., 2008) CD44 expression was identified as a marker for breast cancer cell tumorigenicity (Al Hajj et al., 2003) and has been used to isolate TICs in mammosphere cultures derived from both primary (Dontu et al.,

PAGE 57

57 2003) and established cell lines (Fillmore et al., 2008; Ponti et al., 2005; Cariati et al., 2008) This marker was u sed to further characterize MCF7 derived mammospheres and to determine if there is similar enrichment. Previous studi es used CD44 expression to sort for potentially tumorigenic cells prior to mammosphere culture (Fillmore et al., 2007; Shipitsin et al., 2007; Grimshaw et al., 2008; Li et al., 2008) I t was necessary to confirm that mammosphere culture alone can enrich for a similar subgroup, even if the efficiency is lower MCF7 parental and MCF7S cells were both fixed and stained for CD44 antigen as described in Materials and Methods. The cells were assayed using flow cytometry to quantify the relative percentages of C D44 positivity in the samples. Isotype controls were included for both cell types to account for background signal. As shown in Figure 3 3 and Table 3 1 there is an approximately 59. 92 % of MCF7S cells express ing CD44 while only 1 .31 % of MCF7 parental cells expressed the marker The experiment was repeated and MCF7S CD44 expression percentage was reproducible (Figure A 1 and Table A 1 ). However, MCF7 parental cell can have up to 17 2 5% C D44 expression after long term culture (Figure A 1 and Table A 1 ). ER Status and Stability in MCF7S Cells It is unknown whether ER expression and stability may be linked to mammosphere formation. Previous studies have argued that mammosphere formation is associated with ER negative, basal cell types rather than ER posi tive, luminal subgroups (Sleeman et al., 2007; Asselin Labat et al., 2007) More recent evidence argued that a stem cell hierarchy exists during mammary stem cell development that support s the notion of a lineage restricted, ER positive progenitor cell (Villadsen et al.,

PAGE 58

58 2007) With these conflicting viewpoints, ER expression in MCF7S cells was compared to MCF7 parental cell to determine if mammosphere formation alters ER expression. The RNA was extracted from both cell types and a cDNA library was g enerated from total RNA using reverse transcription Semi quantitative real time RT PCR showed comparable levels of both ER isoforms (alpha and beta) for MCF7P and MCF7S (Figure 3 5A). ER expression levels in both sample s were normalized to beta actin expression (Figure 3 5 A). It appeared MCF7S displayed lower levels of ER isoforms than MCF7P. T he difference in ER expression was less than 2 fold and w as not considered significant. The difference in ER expression was 2.06 and was considered to be significant, but the biological relevant of the difference was unknown. Indirect immunofluorescence was used to determine if there were variation s in ER p rotein stability at the single cell level. For a negative control, no primary antibody was used and the cells were stained with secondary antibody only. ER negative HEK293 cells were used as negative control s during initial experiments As shown in Fig ure 3 5 B, ER was primarily a nuclear protein in both MCF7P and MCF7S. All MCF7 cells were positive for ER Discussion It is possible to take advantage of innate heterogeneity in established breast cancer cell lines to establish a mammosphere culture. However, the efficiency and rep roducibility is quite variable. Only a single cell line wa s used in this study, as attempts at generating mammospheres from other cell lines had been unsuccessful MDA MB 231 ( human ER negative breast cancer epithelial) Eph4 (mouse non tumorigenic ER positive mammary epithelial) MCF10A (human non tumorigenic ER

PAGE 59

59 negative breast epithelial) BT474 (human ER negative breast cancer epithelial) and ZR 75 1 (human ER positive breast cancer epithelial) have either failed t o survive in mammosphere culture outright, or failed to expand in subsequ ent passages The reproducibility of MCF7 derived mammospheres hints that unknown factors lend MCF7 additional flexibility to environmental changes. This feature may be due to the selection of a subgroup with favorable traits or possibly to the active transformation of the population in respon se to extracellular conditions There is insufficient information at this time to support either conjecture. However, one may hypothesize that both scenarios are in play. This is because if there is indeed a dominant subgroup that survive d the selection process, one may expect a higher percentage of CD44 positive cells than the roughly 60% that exists in mammosphere culture as the dominant subgroup would expand with each additional passage As there is no clear dominance, one may suppose that a mixed population still exists and has shifted to a new equilibrium The newly emerged dominant population in the MCF7S culture may have a distingui shing molecular profile (Kok et al., 2009) However, the se differences are not fully characterized in this study. In summary, t he parental MCF7 cell popu lation possessed a pred ictable growth rate once it a cclimated to mammosphere culture conditions (Figure 3 3 ) T he resulting MCF7S culture showed greater consistency in CD44 expression than MCF7 parental cells (Figure 3 4 ) These two observations indicate d that there was a shift in equilibrium as the adherent MC F7 parental cells adapt ed to mammosphere culture. However, the exact mechanism for the shift is unknown. The presumed differences that exist ed between the MCF7S and MCF7P cultures may affect hormo ne sensitivity of the two cell

PAGE 60

60 types. This hypothesis was first assessed by evaluating ER expression in the two groups. As shown in Figure 3 5 there wa s no significant difference in ER mRNA ex pression or protein stability. However, there was a significant difference in ER expression that may have further relevant but was not probed furth er. This did not offer sufficient explanation for ER requirement in mammosphere formation or maintenance. It also precluded assumptions relating antiestrogen response in the two groups. These con siderations were further studied and discussed in Chapte r 4 and 5

PAGE 61

61 Figure 3 1. MCF7 derived mammospheres ( MCF7S ) growth curve (n=2). The average viable ce ll density of MCF7S was assessed every 48 hours after initial seeding at 2500 cell per ml. A B C D Figure 3 2 Phase contrast microscopy o f m ammospheres derived from MCF7 parental cells Representative images of: A) 2 days after initial se eding, B) 4 days after initial seeding with microspikes indicated by arrows, C) 8 days after initial seeding, and D) Hoechs t nuclei staining of mammospher es 8 days after initial seeding to show mammosphere are multicellular Scale bar = 100 microns

PAGE 62

62 Figure 3 3 Enriching MCF7S from MCF7P by serial passage in mammosphere media (n=2) Each passage is 7 days. Fold change is expressed as final cell density/initial cell density. The average of two independent experiments is represented on the graph.

PAGE 63

63 A B Fi gure 3 4 Comparing CD44 exp ression in MCF7P vs. MCF7S using flow cytometer A) Representative sc atter plot and gating of FACS sorted cells. B) Representative overlay histogram of CD44 FITC and Iso FITC staining for MCF7P and MCF7S. Technical duplicates were shown f or the experiment. Isotype staining control (Iso FITC) showed negligible staining for both sample groups. MCF7P cells sh owed an averaged 1.31% CD44 positive staining. MCF7S cells showed an averaged 59.92% CD44 positive staining. Percentages wer e calculated from 30000 cells.

PAGE 64

64 Table 3 1. Data for CD44 FITC signal quantification using flow cytometer. Sample Gated Events Background Subtracted Events % Gated % Total (Not gated) MCF7P Iso 28329 112 0.4 0.37 MCF7P CD44 1 28537 375 1.31 1.42 MCF7P CD44 2 28548 427 1.5 1.42 MCF7S Iso 29504 24 0.08 0.08 MCF7S CD44 1 28922 17330 59.92 57.77 MCF7S CD44 2 29413 19419 66.02 6 4.73 Table 3 1 data represents addi ti onal experiment to evaluate CD44 expression in MCF7P and MCF7S (Figure 3 4) A total of 30000 cells were counted for each sample. Gated events represent cell fraction used for analysis (R1) based on forward and side scatter pattern as shown in Figure A 1A. Background subtracted events represented CD44 FITC signal levels above Iso FITC control. % Gated is percentage ratio o f gated events / background subtracted events % Total is percentage ratio of gated events / 30000 total events.

PAGE 65

65 A B C Figure 3 5 ER expression in MCF7P and MCF7S. A) Semi quantitative real time RT PCR for ER and ER expression in MCF7P and MCF7S cells The data is shown as average of two independent experiments. Each experiment was performed with technical triplicates. Both ER and ER expression was normalized to beta actin expression. Technical duplicate negative control (no cDNA template) w as included in the experiment. B and C) Indirect immunofluorescence for ER protein (green) was performed on MCF7P and MCF7S (B and C, re spectively). DAPI (blue) counterstain was use d to indicate nuclear region. Immunofluorescence n egative control was performed by omitting primary anti ER antibody. Abbreviations: MCF7P is MCF7 parental. MCF7S is MCF7 mammosphere culture. ER is estrog en receptor alpha. ER is estrogen receptor beta.

PAGE 66

66 CHAPTER 4 THE EFFECTS OF ANTIE STROGEN ON MAMMOSPHE RE FORMATION Introduction If the mammosphere culture has indeed selected or transformed the parental culture one may expect MCF7 S to display altered be havior when compared to MCF7P. In previous studies, mammos phere cells were shown to be resistant to chemotherapeutic agents such as 5 fluorouracil and pacil taxel (Fillmore et al., 2008) when compar ed to parental cells. The observation implied that there were intrinsic properties in mammosphere cells that lend additional resistance to cytoto xic agents, rather than acquiring resistance from exten ded exposure We question ed if such resistance is an indication of p otential TICs enrichment in mammosphere culture and if it is applicable to antiestrogens We were interested in studying MCF7S antiestrogen response which may contribute new insights in to cancer recurrence in ER positive tumors We utilize d the sphere forming ability of MCF7S cells to query their tumor initiating potential and to quantify possible TIC enrichment. The sphere formation assay was also used to assess tumorigenic potential follo wing antiestrogen challenge. Pharmacological inhib ition by antiestrogens also serve d to elucidate the role of ER in mammosphere formation and maintenance. The rationale for the sphere formation assay was that tumorigenic cells have a higher probability of survival in mammosphere culture, while the non tumorigenic population is less likely to survive under the sa me conditions. Therefore, it would be possible to estimate the tumor initiating potential of a cell population by quantifying the sphere formin g frequency of the MCF7S population. While it is true that not every mammosphere is derived from a TIC, previous studies have suggested the

PAGE 67

67 mammos phere forming population to be more tumorigenic in general (Ponti et al., 2005; Li et al., 2008; Grimshaw et al., 2008; Fillmore et al., 2008; Cariati et al., 2008) The reaction of this population to antiestrogen treatment s will yield information useful for clinical applications. In addition, we studied the long term expansion of MCF7S cells under antiestrogen challenge to determine the growth kinetics of potential TICs. I f the MCF7S culture is indeed enriched for poten tial TICs, there would be a stable sub population that maintains the culture over many passages while under antiestrogen challenge. The expansion of this subgroup would remain relati vely constant over time and would not be affected by antiestrogen treatmen t. However, if there is merely selection of an antiest rogen resistant subgroup it would be reflected in the growth kinetics of the bulk culture over time. For example, if the population had acquired r e sistance then the cell number is expected to increase at later passages as the dominant subgroup continues to expand Inversely, if the population become s quiescent o r senescent then cell number would decrease over time. Two cla sses of antiestrogens were used in this study The first is selective estroge n r eceptor m odulator (SERM), which is the classic ER antagonist The particular compound used wa s 4 hydro xytamoxife n (4 OHT) which i s an active metabolite of tam oxi fen. The second is s elec tive estrogen receptor d own regulator (SERD), which induces ER degradation and impairs receptor dimerization ICI 182780 (ICI) also kno wn as Fulvestrant or FASLODEX wa s the SERD chosen for this study.

PAGE 68

68 Results MCF7S Response to Antiestrogens The initial treatment of MCF7S with antiestrogens prod uced an interestin g phenotype. MCF7S were seeded at 5000 cells per ml and treated with vehicle or various dil utions of 4 OHT and ICI. The cells were incubated for 6 days and spheres above 50 microns in diameter were scored. As shown in Figure 4 1, there wa s a marked diff erence in sphere formation ability when the cells were exposed to the two diff erent antiestrogen s Figure 4 1A illustrated the loss of sphere formation upon treatment with at least 2.5 M 4 OHT which caused cells to form disordered aggregates Quantific ation of sphere formation is shown in Figure 4 1B. S ampl es exposed to vehicle controls, 10 M 17 Estra diol, 1 M 4 OHT, or two different concentrations of ICI did not exhibit sphere formation disruption and remain ed as tightly packed spheroids Three di fferent concentrations of 4 OHT were tested (1, 2.5 and 5 M) and sphere formation inhibition was evident at concentrations of 2.5 M or above (Figure 4 1B ). This result suggested MCF7S can respond to antiestrogen treatments and that ER antagonism has a significantly dif ferent effect than ER reduction Proliferation study was used to further characterize antiestrogen ef fects on MCF7S cells. The cells were plated at 5000 cells per ml and treated with antiestrogens on the day of seeding. The cells were then counted every 48 hours for 6 days to determine the proliferation rate of these cell s in the presence of antiestrogens. Figu re 4 2 demonstrated the effects varied amounts of SERM / SERD had on MCF7S proliferation. It showed MCF7S responded to antiestr oge ns in a dose dependent manner. Proliferation of MCF7S cells were not significantly affected by 1 M 4 OHT

PAGE 69

69 while 2.5 M 4 OHT significantly decreased (p<0.05) MCF7S proliferation by about 50% MCF7S proliferation was further decreased by 5 M 4 OHT which may be signs of cytotoxic effects (Figure 4 2A). The effects of 1 M ICI treatment were compara ble to that of 2.5 M 4 OHT for the inhibition of MCF7S proliferation (Figure 4 2B). MCF7S cells treated with 0.5 M ICI showed a decrease in cell proliferation, but it is not significant when compared to 0.1% DMSO control (Figure 4 2B). Antiestrogen Ef ficacy in Parental MCF7P and MCF7S MCF7P cells were the n treated with 4 OHT to determine if they were similarly affected by antiestrogens as MCF7S and to confirm drug efficacy. MCF7P cell are generally maintained in complete DMEM medium that contains a f ull complement of steroidal hormones, thus dulling the effects of antiestrogens. Therefore, MCF7P cells were cultured under adhere nt conditions in complete DMEM, steroid free DMEM, and mammosphere media to compare 4 OHT effects on proliferation. The cel ls were seeded using complete DMEM media onto three 6 well plates and allowed to attach for several hours. After which, two of the plates were washed with PBS and the media was replaced with either steroid free (SF) media or mammosphere media. The cell s were incubated for 48 hours in their respective media, and then either 0.1% DMSO or 2.5 M of 4 OHT was added to the cells and the time course commence d The cells were counted with trypan blue exclusion every 48 hours for 6 days to assess MCF7P proliferation. The experiment was repeated in Figure 4 3C using 6 days incubation time point bec ause that is when significant differences were detectable.

PAGE 70

70 Figure 4 3 showed there were no major differences in 4 OHT response regardless of the culture media used Figure 4 3 A indicated no significant differences in proliferation between MCF7P cells cul tured in complete DMEM or steroid free (SF) media, and there were little differences in cell number in response to 2.5 M 4 OHT. Nor were there a significant difference between cells cultured in steroid free ( SF) media or mammosphere (sphe re) media as shown in Figure 4 3 B and Figure 4 3C In summary MCF7P wa s un affected by difference s in culture media when grown under a dherent conditions and were not significantly affected by 4 OH T under these conditions. The results indicate d a significant difference in antiestrogen response between MCF7S and MCF7P. It was necessary to confirm drug efficacy in the two cell groups in o rder to appraise their respective response. Nuclear extraction was performed for both cell types after antiestrogen treatment for 48 hours to determine if there is a difference in ER stability or localization. ICI is a known SERD, which gave a clear confirmation of drug efficacy in both MCF7S and MCF7P as ICI decreased overall ER protein levels (Figure 4 4 ). 4 OHT has been implicated as an ER stabilizer in previous studies (Wijayaratne and McDonnell, 2001; Marsaud et al., 2003) The results in Figure 4 4 verified 4 OHT effec t to be consistent in both cell groups with the increase in ER observed in the nuclear fraction ER transcriptional function was examined using real time RT PCR to evaluate the expression of Trefoil Factor 1 (TFF1), a known estrogen response gene. TFF1 expression in MCF7P cells was predictably upregulated by 17 estradiol (E2) treatment and the gene was maximally expressed 12 24 hours following E2 addition (Figure A

PAGE 71

71 2A ). MCF7P showed noticeable reduction of TFF1 expression up on pre treatment with antiestrogens (Figure A 2B, A 2C) TFF 1 expression in MCF7S cells were increase 12 24 hours after E2 addition, and followed a similar trend as MCF7P (Figure A 3A). However, data for antiestrogen inhibition of TFF1 induction were inconsistent and overall changes were too low (less than 2 fold) to be considered si gnifica nt (Figure A 3B, A 3C). CTSD was another estrogen response gene tested (Figure A 4), but it did not showed significant u pregulation upon E2 treatment and was not further evaluated. This may reflect i nherent heterogeneity in gene expression profile upon a ntiestrogen treatments or ER was no longer responsive at the transcriptional level Sphere Formation Frequency Following Antiestrogen Challenge Previous studies have correlated mammosphere formation with TIC enri chment (Grimshaw et al., 2008; Fillmore et al., 2008) Therefore, sphere formation frequency was quantified after antiestrogen challenge as an indicator of tumor initiation potential. Figure 2 1 offers the general workflow for the study. Following cell count for the evaluation of MCF7S antiestrogen response (Figure 4 2), the surviving cells were replated at approximately 500 cells per well i n a 96 well plate without drugs to determine sphere formation efficiency after antiestro gen challenge. Spheres larger than 50 microns in diameter were scored for sphere formation frequency. As shown in Figure 4 5, there was no statistically significant decrease in sphere forming frequency following antiestrogen treatments with the exception of 2 days 0.5 M ICI treatment (p=0.028) in Figure 4 5B There was a statistically significant increase in sphere forming frequency with 0.1 M 4 OHT (p=0.0004) and 2.5 M 4 OHT (p=0.014, 4 days) (p=0.021, 6 days) treatments (Figur e 4 5A) however the biological signi ficant

PAGE 72

72 of this observation is unknown These results suggested that antiestrogens did not statistically decrease sphere formation potential of MCF7S cells, and that less than 10% of bulk MCF7S cells were responsible for sphere formation In addition, the evidence supported the notion of a relatively stable subpopulation in MCF7S wa s resistant to acute ant iestrogen treatment and survived to perpetuate sphere formation MCF7S Cell Proliferation under Long T erm Antiestrogen Treatment According to the TIC h ypothesis, TICs remain an d persist in the bulk population. This infinite g contributes to overall malig nancy. However, a subtle detail should be noted as the TIC hypothesis does not stipulate t he TICs would ever become the dominant population through endless expansion. The theory argues that potential TICs only grow and divide sufficiently to maintain themselves and the growth of the bulk population is depende nt on non TICs This reasoning co ntends that cells in the bulk pop ulation were more dynamic and their growth can be a ltered due to environmental causes Therefore, if MCF7S does indeed conta in a subpopulation of TICs then it is possible to examine this property through long t erm serial passage. This method, which is analogous to that used in neural stem cells characterization (Reynolds and Rietze, 2005) can also probe the possibility of antiestrogen effects on long term mammosphere culture self renewal property This was done by seeding viable cells at 10000 cells per ml at each passage. The cells were treated either with vehicle or antiestrogen at the time of seeding, and incubate d fo r 7 days which constitutes one passage. Viable cells were counted at the end of each passage. The fold change in cell number for each passage was used to calculate potential expansion of the population if all the cells, instead of a fraction, were

PAGE 73

73 passag ed. The long term growth kinetics can be determined in this manner and allow ed for comparison between different antiestrogen treatments. As shown in Figure 4 6 and Table 4 1 0.1% DMSO treated MCF7S did not influence long term cell expansion. There wa s a small decrease in growth for 1 M and 2.5 M 4 OHT treated MCF7S, but there were no significant difference s between the two drug concentrations. 5 M 4 OHT and ICI treated cells suffer ed the most dramatic decrease in long ter m proliferation and expansion However, the differences were not statistically significant Long term serial passage further delineated the marked differences in mechanism for the two classes of antiestrogens, which helped to clarify some conclusions drawn from Figure 4 5. The minimal decrease in growth kinetics for 1 M and 2.5 M 4 OHT treated MCF7S helped to explain the increase in sphere formati on frequency in those samples. This information suggested that indeed not every sphere was derived from a potential TIC, but the overall health and proliferation of a cell c ontributed to sphere formation frequency. Paradoxically, decrease in long term cell expansion did not significantly decrease sphere formation frequency, as indicated by 5 M 4 OHT a nd ICI treated cells (Figure 4 5 and Figure 4 6) The results indicated s phere formation p otential of the MCF7S bulk population was not decreased by antiestrogens, even when the long term growth rate was slowed MCF7S Cell Cycle Analysis To further study the growth kinetics of antiestrogens treated MCF7S, cell cycle analysis was performed using bromodeoxyuridine (BrdU)/ propidium iodide (PI) staining. MCF7S cells were dissociated and treated with 2.5 M 4 OHT or 1 M ICI for either 48

PAGE 74

74 or 72 hours. The cells were then fixed in 70% ethanol and BrdU/PI staining was performed as described in Chapter 2. BrdU staining was largely unsuccessful due to technical hurdles and data was obtained for only a single exper iment (Figure A 5 ). BrdU staining indicated a 10 14% increase in G1 phase retentions, with a corresponding decrease in S phase and G2 ph ase cell cycle for antiestrogen treated cells compared to vehicle treated control. C ell cycle data were primaril y obtained from the analysis and mod eling of propidium iodide ( PI ) histogram using ModFit LT software. As shown in Figure 4 7, cell s treated with 2.5 M 4 OHT had a significant fraction (p=0.0009) of cells retained in G1 phase (75.3%) and significant reduction (p=0.00006) in G2 phase ( 7.39%) 48 hours after treatment when compared to vehicle treated control The re was significant reduction (p=0.0003) in S phase (18.9%) 72 hours following drug introduction. MCF7S cells treated with 1 M ICI had a significant reduction (p=0.0009) in S phase (13.3%) cell cycle 48 hours after addition of the compound C ellular growth arrest was more pronounced at 72 hours a s a significant (p=0.0003) percentage of cells stalled at G1 phase (78.8%) with a correspondingly significant (p=0.00005) decrease of cells in S phase (13.4%), compared to vehicle treated control T hese results showed growth arrest for 2.5 M 4 OHT treate d cells 48 hours after drug addition, while 1 M ICI showed maximal growth arrest 72 hours after drug treatment MCF7S Apoptosis Assay Apoptosis assay using A nnexin V staining was performed to account for observed cell loss after antiestrogen treatmen t. A s shown in Figure 4 8, there was a statistically significant (p=0.025) increase in apoptosis for 2.5 M 4 OHT treated cells 72 hours of drug treatment. Howeve r, there was no significant apoptosis observed upon 1 M ICI

PAGE 75

75 treatment. There was significant qu antitative variation between experiments, but the trend was consistent within individual experiments as 2.5 M 4 OHT treated MCF7S consistently display ed a statistically higher percentage of dead cells than those treated with 1 M ICI (Figure A 6 ). Howeve r, t he data generated from Annexin V assay may not be biologically significant as attempts at other cell death assays (TUNEL assay, trypan blue) failed to detect significant changes in cell death 48 or 72 hours after antiestrogen treatment. Therefore, it is possible that the loss of viable cells is due to acute necrosis induced within the first 24 hours of antiestrogen treatment. Discussion MCF7S cells possess ed a unique disposition that altered their response to antiestrogens when compared to MCF7P cells (Figure 4 3) The initial assumption that MCF7S cells may be more resistant t o antiestrogen t han MCF7P was dispelled as the inverse appeared to be true This observation may be the result of optimized growth condition in MCF7P adherent culture that rend ered the cells less responsive to antiestrogens. It is also possible that a higher degree of heterogeneity was maintained in adherent MCF7P culture, which resulted in a higher background and hindered antiestrogen response detection T he MCF7S culture was enriched for a narrow er range of cells grown under more stressful conditions, which may contribute to the distinct antiestrogen response profile when compared to the originial population. The data suggests that mammosphere formation is not dependent upo n ER status as SERD treated cells retain sphere forming ability (Figure 4 1 ) However, MCF7S cells are responsive to antiestrogen effects and ER may play a role in cell proliferation kinetics (Figure 4 2 and Figure 4 6 ) Typically less than 10 % of the ER

PAGE 76

76 positive cells can form spheres larger than 50 microns (Figure 4 5) These cells appear ed to be more resistant to acute SERM treatments as the normal recommended dosage is 1 M for 4 OHT to inhibit proliferation which had little effect on MCF7S cells (Fi gure 4 2) SERD induced growth inhibition was consis tent with published studies, which utilized 1 M to demonstrate efficacy (Figure 4 2) If mammosphere culture truly enriched for tumor initiating cells, then these results suggest tumorigenicity may not be depen dent on hormone receptor status; nor does it indicate hormone receptor status a s a stra ightforward prognostic factor. Interestingly, the sphere formation frequency did not significantly decrease after anties trogen challenge (Figure 4 5) This si gnified that the same percentage of surviving cells were resistant t o acute antiestrogen treatment. However, not all surviving cells were capable of sphere formation after drug removal or may contribute to tumor persistence. T he two classes of antiestrog ens produced different characteristics in surviving cells. The conclusions concerning antiestrogen action suggested ER antagonism may result in acute cell death (Figure 4 8) and perhaps growth arrest as a secondary result (Figure 4 7) while ER destabil ization may resulted mainly in growth inhibition (Figure 4 7) The observations for 4 OH T may be consistent with the sphere formation phenotype (Figure 4 1) because cell membr ane s are disrupted during necrosis and would inhibit cell cell adhesio n. In the case of ICI induced growth arrest (Figure 4 7) sphere formation phenotype and frequency remain ed undisturb ed (Figure 4 1 and Figure 4 5) This theory wa s supplemented by long term serial passage studies, which confirm ed that t he two antiestrogen s influ ence d MCF7S growth in different manners

PAGE 77

77 (Figure 4 6) 4 OHT did not significantly change the growth kinetics of MCF7S cells which is consistent with the increase in cell death (Figure 4 8) The signs of growth arrest shown in Figure 4 7 did not appear t o contribute significantly to long term cell expansion, which led to the conclusion that 4 OHT sensitive MCF7S cells may undergo growth arr est in addition to cell death This situation was not observed in ICI treated MCF7S as the compound was found to be more effective against cell expansion (Figure 4 6 and Figure 4 7) then stimulating cell death (Figure 4 8) This indicate d that while mammosphere formati on does not require ER the receptor can control proliferative potential in individual cells by inhi biting growth or cell death One hypothesis is ER has a n alternative role for a fraction of the mammo sphere forming cells which may account for differential response to antiestrogens within the MCF7S population As discussed in Chapter 1, ER can inf luence cell proliferati on through both genomic and non genomic pathways. The receptor function s are modulated by factors such as ligand availability and binding, cofactor recruitment, and mitogenic signaling Therefore, the observation that the two class es of antiestrogens exert different effec ts on MCF7s is to be expected. It is also possible that 4 OHT has non ER targets, or 4 OHT may generate proliferative effects in cells through alternative ER signaling. In this culture model, after the removal of 4 OHT, the cells resume sphere formation and are expected to still be ER positive These features suggested th at antiestrogen had no significant effect on tumor initiating potential of ER positive breast cancer cells. However, antiestrogen may continue to suppress tumor growth and prevent recurrence over an extended period of time.

PAGE 78

78 A B Figure 4 1 Mam mosphere formation in the presence of antiestrogens. A) Representative p hase contrast images of MCF7S cel ls in the presence of antiestrogens or vehicle controls for 7 days. B) Quantification of >50 microns diameter MCF7S spheres scored from images in A). The percentage on y axis was calculated as ( number of >50 micron spheres / total number of cells seeded ) X 100%. Abbreviations: E2 is 17 estrodiol. 4 OHT is 4 hydroxytamoxifen. ICI is pure antiestrogen ICI 182780. Scale bar = 100 microns.

PAGE 79

79 A B Figure 4 2. Cell proliferation of MCF7S in the presence of antiestrogens. A) Cell proliferation of 4 OHT treated MCF7S derived from fou r independent experiments B) Cell proliferation of ICI treated MCF7S derived from four independent experiments Abbreviations: 4 OHT is 4 hydroxytamoxifen. ICI is pure antiestrogen ICI 182780. Error bars are calculated standa rd error of the mean (S.E. test.

PAGE 80

80 A B C Figure 4 3 MCF7 adherent culture response to 4 hydroxytamoxifen (4 OHT) with various culture media A) Comparing 4 OHT response in MCF7P cultured in complete (DMEM), steroid free ( SF ) and mammosphere (sphere) media. B ) Comparing 4 OHT response in MCF7P cultured in steroid s tripped (SF) vs. mammosphere (sphere) media C) Comparing 4 OHT response in MCF7P cultured for 6 days in SF vs. sphere media Abbreviations: 4 OHT is 4 hydroxytamoxifen.

PAGE 81

81 A B Figure 4 4 Immunoblot of cytoplasmic and nuclear fractions from MCF 7P and MCF7S cells treated for 72 hours with 4 hydroxytamoxifen (4 OHT) or ICI 18 2780 (ICI) to c hara cterize ER stability A ) Immunoblot fo r MCF7P. B) Immunoblot for MCF7S. tubulin was used as loading control for cytoplasmic (C) fraction. LSD1 was used as loading control for nuclear (N) fraction. Densitometry quantification of three independent experime nts is shown below a representative immunoblot. Statistically analysis was tes t.

PAGE 82

82 A B Figure 4 5. Sphere formation fr equency of MCF7S after antiestrogen challenge using 4 hydroxytamoxifen (4 OHT) or ICI 182780 (ICI) A) Sphere formation frequency of 4 OHT treated MCF7S derived from four independent experiments. B) Sphere formation frequency of ICI treated MCF7S derived from four independent experiments. Plating efficiency was calculated as (number of sphere >50 microns diameter/ total number of cells plated) x 100%. test.

PAGE 83

83 A B Figure 4 6 Long term expansion of MCF7S in the presence of antiestrogens. The lines are expressed on a semilog graph a nd slope of each line was calculated as log expansion for each condition (Table 4 1) A) Average expansion derived from four independent experiments of MCF7S treated with antiestrogens or vehicle compared to untreated control. B) Comparing average exp ansion of vehicle control against 2.5 M 4 OHT and 1 M ICI treated MCF7S using data set from A). Abbreviations: 4 OHT is 4 hydroxytamoxifen. ICI is pure antiestrogen ICI 182780. Error bars are shown as calculated standard error of the mean (S.E.M.).

PAGE 84

84 Table 4 1. Growth kinetics of MCF7S long term expansion under antiestrogen challenge Sample Line Slope (log) R2 MCF7S 2.380 0.999 MCF7S +0.1% DMSO 2.321 0.999 MCF7S +1 M 4 OHT 2.046 0.999 MCF7S +2.5 M 4 OHT 2.019 0.999 MCF7S +5 M 4 OHT 1.790 0.994 MCF7S +0.5 M ICI 1.861 0.999 MCF7S +1 M ICI 1.574 0.997 Table 4 1 represents the slope of lines represented in Figure 4 6A. The slope was calculated a log expansion for each condition.

PAGE 85

85 A B Figure 4 7. Propidium iodide (PI) cell cycle analysis. A) Cell cycle analysis of 48 hours antiestrogen treated MC F7S derived from four independent experiments. B) Cell cycle analysis of 72 hours antiestrogen treated M CF7S derived from four independent experiments. PI cell cycle analysis and modeling were p erformed using ModFit LT software. Abbreviations: 4 OHT is 4 hydroxytamoxifen. ICI is pure antiestrogen ICI 182780. Error bars were expressed as standard error of the mean (S.E.M.). Statistical analysis was test.

PAGE 86

86 Figure 4 8. Annexin V apoptosis assay for antiestrogen treated MCF7S. The a verage of three independent experiments is shown. Abbreviations: 4 OHT is 4 hydroxytamoxifen. ICI is pure antiestrogen ICI 182780. Error bars were expressed as standard error of the mean (S.E.M.). Statistical analysis was test.

PAGE 87

87 CHAPTER 5 THE ROLE OF ESTROGEN RECEPTOR ALPHA ON MA MMOSPHERE FORMATION Introduction Evidence gathered through pharmacological inhibition in Chapter 4 suggested mammos phere formation did not require ER However, the receptor can potentially exert significant command on proliferation in a fraction of the bulk mammosphere culture. A molecular biological approach was undertaken to confirm pharmacological results and to delineate the role of ER more specifically. Stable ER knockdown clones (shER) were generated using shRNA integration to further examine the receptor s role in mammosphere culture. As previously discussed in Chapter 1 there may be changes in ER signa ling upon antiestrogen binding that may promote misleadi ng interpretation of the results ER has a role in m ultiple signaling pathways such as NF B Notch and He dgehog (Chapter 1). These pathways have been implicated in mammosphere culture maintenance as well as TIC enrichment (Murohashi et al., 2009; Liu et al., 2006; Zhou et al., 2008) Therefore, it is necessary to clarify ER function in MCF7S to ascertain i ts role in the context of potential TICs The different mechanisms of growth inhibition produced by the two classes of antiestrogens have b een described in previous studies (Fan et al., 2006; Osipo et al., 2003; Shaw et al., 2006) It is of further i nterest t o s tudy antiestrogen effects on ER knockdown cells as it may compliment studies done on hormone receptor negative cell lines (Jeng et al., 1994; Lazennec and Katzenellenbogen, 1999; Keen et al., 2003) which were insensitive to antiestrogen treatment These studies have noted ectopic re intr oduction of hormone receptor res tore d antiestrogen sensitivity If the inverse is true

PAGE 88

88 for MCF7S cells, then one may expect shER cells to be more resistant to antiestrogens, to have higher sphere formation frequency and higher tumorigenic potential Results Proliferation of ER Knockdown MCF7S The generation of shER clones was described in detail in Chapter 2. The clones were evaluated by immunoblot and several knockdown clones were identified (Figure 5 1A) with significant ER knockdown. The proliferation of the clones was c ompared to the bulk MCF7S cells to determine potential differences in growth rate. As show n in Figure 5 1B, there is a decrease in cell doubling for all 3 clo nes with clone 11 g rowing the slowest of the three. However, clones 7 and 9 had comparable doubl ing rate s and the difference from MCF7S was not significant. All three shER clones were capable of sphere forma tion, which further confirmed ER was not required for mammosphere formation or maintenance (Figure 5 2) Visual inspection of the shER clonal spheres did not reveal significant differences in size or frequency ( Figure A 7 ). The clones can be continuously passage d as MCF7S, althoug h ex periments were done using cells cultured for less than three passages Antiestrogen Response of ER Knockdown MCF7S The shER clones were tested for their antiestrogen response as described in Chapter 2 and Chapter 4. Control shRNA MCF7S cells wer e likew ise treat ed as control s Figure 5 2 showed the sphere formation phenotype of MCF7S, control shRNA MCF7S, and shER clone 7. Both MCF7S and control shRNA MCF7S showed similar sphere formation inhibition by 2.5 M 4 OHT. These results were consistent with data from Figure 4 1A However, shER clone 7 underwent massive cell loss judging by the

PAGE 89

89 presence of cell debris. The surviving cells formed tightly packed spheroids. As expected, 1 M ICI treatment did not affect sphere formation phenotype or significan tly affect proliferation The proliferation of knockdown clones under antiestrogens was further characterized in the same manner as described previously in Chap t er 2 and 4 The clones were evaluated for only one time point af ter treatment (6 days) becau se Figure 4 2 showed it was when the most signi ficant differences were detectable The results were expressed as % Cell Survival which represented the cell density of antiestrogen treated samples relative to vehicle treated control for each shRNA used i n order to no rmalize results for comparison. Figure 5 3 summarized the results, in which control shRNA MCF7S response to antiestrogen corroborate d with data shown in Chapter 4. OHT treatment in all three significantly reduced in clones 7 and 9 as the compound had no signific ant effect on cell density This was an was indeed sufficie ntly removed from the cell culture to nullify SERD effects. Interpretation of antiestrogen effects on clone 11 should be considered with caution as its slower doubling time (Figure 5 1B) may be a contributing factor to it s antiestrogen response. Its proliferation was affected by 4 OHT in a similar manner as the ot her two clones. However, it did not show the same degree of resistance to ICI treatment which may be the result of its slower growth rate. Sphere Formation Frequency of ER Knockdown MCF7S Following Antiestrogen Challenge The sphere formation frequency of the knockdown clones were similarly evaluated as in Figure 4 5. The clones were replated in fresh media without treatment after 6

PAGE 90

9 0 days of antiestrogen challenge. The res ult ing spheres were scored for their size (> 50 microns diameter) and their total number. Data in Figure 5 4 showed the same lack of change in sphere formation frequency for control shRNA spheres as shown in Figure 4 5 This was proven in terms of both t otal frequency an d sphere diameter (>50 microns). For knockdown clones, the sphere formation frequency of vehicle and ICI treated cell s were likewise stable in b oth MCF7S and control shRNA MCF7S A noticeable reduction in sphere formation frequency was i dentified in knockdown cell when treated with 4 OHT. This result was consistent with the low number of surviving cells after initial antiestrogen challenge. These observations indicated that a fraction of MCF7S th at initially escaped acute 4 OHT induced apoptosis were sensitized after ER knockdown. The implication is the potential TIC pool in shER MCF7S shrank Using SERDs and SERMs to Mimic shER Effects in MCF7S The increase d potency observed for 4 OHT in shER M CF7S cells was puzzling as the compound was expected to lose its effect upon ER knockdown. In order to address clinical relevance, we attempted to mimic knockdown effect using a combination of SERD and SERM MCF7S cell were subjected to daily dosing with fresh media containing various concentrations and combinations of anti estrogen s The data summarized in Figure 5 5 poses several points for further considerations. First, it indicated that pharmacologically induced ER degradation was insufficient to promote the level of cell death seen in 4 OHT treated shE R cells. Second continuous high dose 4 OHT exposure was required to prompt a similar decrease in cell number as 4 OHT treated shER MCF7S Third, there were significant a dditive effects with combinatori al treatment s Finally, the respons e was dose dependent. In genera l it would require a

PAGE 91

91 constant, cytotoxic level of antiestrogens to achieve the same effect as a single hig h dose of SERM in shER cells. This would suggest ER saturation can inhibit MCF7S proliferation. Discussion ER is proving to be a paradoxical factor in MCF7S cells. It is not required for sphere formation and maintenance; however its concentration in the cell has vast impact on cell proliferation. One possible explanation is the lower concentration of ER allowed a single high dose of 4 OHT to fully saturate all available receptors in the bulk shER population and so increase the appearance of drug potency. This suggests that recurrence in some cases of hormone receptor positiv e tumor are not due to resistance but because high receptor levels prevented opti mal inhibition by ER antagonists. This resulted in cells evading antiestrogen effects that theoretically were still responsive to treatment thus resulting in recurrence and disease advancement Standard antiestrogen treatments are not optimized for such hetero geneity that naturally exists. Rather, lower doses are preferred due to side effects that disrupt s quality of life. Another pos sibility to consider is 4 OHT binds alternative targets as the concentration of its prefer red binding partner decreased. I t is conceivable that the high 4 OHT dosage antagonized not only ER but also saturated other target s such as Estrogen Related Receptor beta (ERR ) and gamma (ERR ) (Coward et al., 2001; Tremblay et al., 2001) Additional targets for 4 OHT binding may exist, as it is generally tr ue for pharmacological agents. It has been suggested that 4 OHT mediated apoptosis through both ER dependent and independent pathways suc h as oxidative stress and activation of stress kinases (Obrero et al., 2002) The present study provided evidence

PAGE 92

92 that mammosphere formation and potential TIC enrichment respond to 4 O HT in a related manner

PAGE 93

93 A B Figure 5 1. Knockdown of ER (shER) in MCF7S cells. A) Immunoblot comparing ER levels of three independent shRNA c lones with bulk MCF7S culture. Relative densitometry reduction for shRNA clones were shown as relative to MCF7 sample. B) Growth curve of shER clones compared to b ulk MCF7S culture averaged from two independent experiments

PAGE 94

94 Figure 5 2. Representative phase contrast imag es of MCF7S, control shRNA MCF7S, and shER clone 7 exposed to antiestrogens for 6 days 0.1% DMSO did not affect sphere formation phenotype for all three cell groups. 4 OHT treatment inhibited sphere formation MCF7S and control shRNA cells. The shER C7 cells responded to 4 OHT with increase in cell debris, but surviving cells formed tightly packed spheroids. ICI treated cells all retained sphe re formation phenotype. Abbreviations: 4 OHT is 4 hydroxytamoxifen. ICI is pure antiestrogen ICI 182780. Scale bar = 182 microns.

PAGE 95

95 A B C Figure 5 3 Antiestrogen response of ER knockdown in MCF7S cells. Data for clone 7 (A) and clone 9 (B) were generated from three independent experiments. Data for clone 11 (C) were derived from two independent experiments. 4 OHT is 4 hydroxytamoxifen. ICI is pure antiestrogen ICI 182780. Error bars were expressed as standard error of the mean (S.E.M.). Stat istical analysis test

PAGE 96

96 A B C Figure 5 4 Sphere formation frequency of antiestrogen treated ER knockdown MCF7S cells. Three independent experiments were performed for clones 7 (A) and 9 (B) Clone 11 (C) experiments were repeated twice. Abbreviations: 4 OHT is 4 hydroxytamoxifen. ICI is pure antiestrogen ICI 182780. Error bars were expressed as standard error of the mean (S.E.M.). Statistical test

PAGE 97

97 Figure 5 5. SERDs and SERMs combination treatment t o mimic shER effects i n MCF7S. Media containing the indicated antiestrogens were administered daily for 6 days. The data was derived from three independent experiments Abbreviations: 4 OHT is 4 hydroxytamoxifen. ICI is pure antiestrogen ICI 182780. All antiestrogen treated MCF7S showed significantly decreased viable cells compared to 0.1% DMS O treated cells Error bars were expressed as standard error of the mean (S.E.M.). Statistical analysis was test.

PAGE 98

98 CHAPTER 6 CONCLUSION S AND FUTURE DIRECTION S Conclusions and Discussion The CSC/TIC hypothesis captured th e imagi nation of cancer biologists by proposing TICs as a recognizable target for the study of tumor recurrence It has been ge nerally pre sumed the TICs did not express ho rmone receptors. This assumption is colored by a strict interpretat ion of mammary s tem cell hierarchy in which MaSCs belonged to the basal subtype while hormone sensitiv e cells were grouped in the luminal category (Smalley et al., 2003; Villadsen, 2005) T his assertion has limited the scope of cancer biology by superimposing one discipline onto another Emerging studies are probing the role of hormone rece ptor positive progeni tor cells and hormone response i n hormone receptor negative stem cells (Lim et al., 2009; Assel in Labat et al., 2010; Clarke et al., 2005; Booth et al., 2006; Raouf et al., 2008) with the expectation to define the function of hormones and hormone receptors during development. There may emerge some common links that will enhance oncogenic studies. In this study, evidence was presented for the role of ER as a mediator of antiestrogen effects in the context of MCF7 derived mammospheres. MCF7S was shown to be phenoty pically distinct from MCF7P and express higher levels of putative tumorigenic marker CD44. ER expression and endogenous protein levels were not significantly difference between the two cell groups. The receptors can activate known estrogen response gene TFF1 in both cell types, although gene expression is highly variable in MCF7S cells. The antiestrogen response in MCF7S was significantly di fferent f rom MCF7P as MCF7S displayed dose dependent growth inhibition upon acute antiestrogen exposure while MCF7P showed minimal sensitivity to antiestrogens.

PAGE 99

99 However, the sphere formation frequency in MCF7S was unperturbed following drug removal Pro found differences in response to SERM versus SERD were discernable in MCF7S culture. The long term growth kinetics were strongly affected by SERD treatments, while that of SERM was no t significantly different from control and may contribute to increased s phere formation frequency after antiestrogen challenge. Mammosphere formation and maintenance was shown to be ER independent using pharma cological and molecular means. However, ER knockdown clones showed vastly different antiestrogen response. shER cells became resistant to ICI treatment and sphere formation frequency was unaffected as expec ted. Surprisingly t he knockdown cells became highly sensitive to 4 OHT induced growth inhibition and sphere formation frequency was drastically reduced. These two observations suggested bulk MCF7S culture is composed of a mixed population with ER potentiating divergent pro liferative pathways Additional pharmacological inhibition data in Figure 5 5 suggested there were ER positive MCF7S subclones that were only partially inhibited by the antagonist 4 OHT. One explanation may be concentration is particu larl y high in that fraction and a single dose of 4 OHT wa s insufficient as some active receptors remain and signal cell growth Another possibility is that 4 OHT has non ER targets that also regulate MCF7S proliferation. This is not unexpected as 4 OHT resi stant subclones may have alternate proliferation signaling pathways that can compensate for ER inhibition The idea that alternative mitogenic pathways alter antiestrogen response and allow for anchorage independent growth is not n ew. Previous studies have been done where MCF7 was maintained under long term steroid deprivation (Martin et al., 2003,

PAGE 100

100 2005; Chan et al., 2002) I t was noted that MAPK/ERK1/ERK2 and PI3K activity are upregulated under these conditions and contributed to hormone independent growth, while the cells remain ed antiestr ogen responsive. Considering the fact that mammosphere culture is maintained in growth factor supplemented media, it is possible that mitogenic pathways upregulated under these conditions contributed to the observed MCF7S antiestrogen response. MAPK path way upregulation would also allow for MCF7 to grow as anchorage independent spheroids (Fukazawa et al., 2002; Thottassery et al., 2004) Recent studies have shown prolong ERK activation wa s required to mediate 4 OHT may induced cell death, and effects of ERK phosphorylation can be mitigated by ER activation (Zheng et al., 2007; Zhou et al., 2007) If true, this pathway would account for 4 OHT hypersensitivity obser ved in shER MCF7S. A proposed model would be ERK act i vation via g rowth factor stimulation in shER MCF7S combined with 4 OHT induce apoptosis through ER independent mechanism (Obrero et al., 2002) result in uncontrolled cell death as ER was taken out of a signaling loo p that ma y regulate ERK activation (Hutcheson et al., 2003; Britton et al., 2006) These signaling pathways would undoubt edly affect downstream gene expression an d amplify effects at the genomic level. An additional consideration is how key tumorigenic characteristics ma y drive (Hanahan et al., 2000) four can be applied to mammosphere formation in MCF7S cells. They are self sufficiency in growth signals, insensitivity to antigrowt h signals, evading apoptosis, and limitless replicative potential (Hanahan et al., 2000) MCF7S cells were able to form tightly pack spheroid cultured in a rudimentary media under suspension

PAGE 101

101 conditions, and displayed indefinite self renewal capacity. While many establish breast cancer cells lines shared the above characteristics under normal culture conditions, only MCF7 was able to survive in mammosphere culture (Chapter 1). It would be of general interest to probe for intrinsic cellular differences that may account for this phenomenon, and establish which of the aforementioned key features significantly contribute to mammosphere formation. There are major gaps in knowledge regarding cellular response to hormones, and the role of steroid hormone receptors in cellular signal ing under diff erent cell ular context Of course, it is a daunting task to catalogue all possible cell signaling combina tions. However, it may be worth the effort for it provides the basis for disease relevant gene regulation studies Mammosphere culture provided a model for further characterization of tumor recurrence. However, it is uncertain if mammosphere cells may b e accurately termed CSCs/TICs. It is probable that there is no definable population to t arget, and tumor recurrence depend upon a mixed subclone population with the evolutionally advantage to survive. The current experimental approach of isolating specif ic subgroup may aid mechanistic studies; but it is inadequate in providing a complete picture of tumor development As e vident in this study, there are heterogeneous subgroups that can be enriched under mammosphere forming conditions that respond markedly different from adherent culture. It appears adherent culture tends to homogenized cell population as it provides a rich environment for the survival of most cells, rather than selecting for the hardier portion that ar e most likely to be malignant. Never theless, this portion may still

PAGE 102

102 contain antiestrogen responsive cell s that can be eliminated if treatment regiments are optimized. Future Directions In vivo tumorigeni c studies are ongoing to confirm MCF7S tumorigenicity. If in vitro experiments are trans latable, it is expected that MCF7S will be more tumorigenic than MCF7P. Antiestrogen treated MCF7S cells will be expected to have similar tumorigenic frequencies and 4 OHT treated shER clones w ill have the lowest frequency. Of course, there are reasons to suppose in vivo data will not match those from in vitro studies. The injected cells will receive estrogen supplement by means of an implanted pellet in the mous e, and the cells will no longer receive growth factors. Therefore, it is unknown if MCF7S w ill revert to MCF7P phenotype under these conditions and display no difference s in tumorigenic ability There is the additional concern that the microenvironment at the injection site would alter cellular signaling that translates into confusing results w here no conclusions may be drawn Cytoplasmic signaling in MCF7S is the next logical area of investigation. A rich source of information concerning mitogenic and estrogen receptor crosstalk can be gathered with further study, which can translate to a de eper understanding into both genomic and non genomic ER functions. Of specific interest is ERK phosphorylation as it appears to be an important branch point in cytoplasmic and ligand independent ER function s that may mediate antiestrogen response. Final ly, a multi pronged approach has long been advocated as the most efficient means of cancer control. However, such a rigorous treatment regimen may be prohibitive for patients both in terms of cost, time investment, and quality of life.

PAGE 103

103 Therefore, recent efforts towards optimizing the potency and the innovative usage of currently available drugs are vital for t he future of cancer treatment. This study has provided evidence for extending SERM treatment beyond the current recommendation of 5 years. The res ults from this study have shown the tumor initiating potential of estrogen receptor positive cells were unaffected by antiestrogen treatments (Chapter 4). However, the evidence also suggested antiestrogens can inhibit proliferation and potentially suppres s tumor recurrence (Chapter 4 and 5). The data also suggested 4 hydroxytamoxifen may have additional non ER targets that warrant further study (Chapter 5). An importa nt question remains: whether antiestrogen resistant cells and TIC s are the same The tools currently availab le are insufficient to provide solid experimental proof However, the enthusiasm gene rated in the area will surely drive future technical improvements.

PAGE 104

104 APPENDIX A SUPPLE MENTAL FIGURES Chap t er 3 A B Figur e A 1. Additional figures comparing CD44 expression in MCF7P and MCF7S A) Scatter plot and gating of FACS sorted cells. B) Ove rlay histogram of CD44 FITC and Iso FITC staining for MCF7P and MCF7S. Technical duplicates were shown for the experiment. Isotype staining control (Iso FITC) showed negligible staining for both sample groups. MCF7P cells showed an averaged 21.4% CD44 p ositive staining. MCF7S cells showed an averaged 57.97% CD44 positive staining. Percentages were calculated from 30000 cells. Table A 1. Additional data for CD44 FITC signal quantification using flow cytometer (Figure A 1) Sample Gated Events Backgr ound Subtracted Events % Gated % Total (Not gated) MCF7P Iso 29366 394 1.34 1.31 MCF7P CD44 1 29403 5192 17.66 17.31 MCF7P CD44 2 29378 7486 25.48 24.95 MCF7S Iso 29317 59 0.2 0.2 MCF7S CD44 1 29333 17283 58.92 57.61 MCF7S CD44 2 29205 16654 55.51 56 .49

PAGE 105

105 Table A 1 data represents addi ti onal experiment to evaluate CD44 expression in MCF7P and MCF7S. A total of 30000 cells were counted for each sample. Gated events represent cell fraction used for analysis (R1) based on forward and side scatter patte rn as shown in Figure A 1A. Background subtracted events represented CD44 FITC signal levels above Iso FITC control. % Gated is percentage ratio o f gated events / background subtracted events % Total is percentage ratio of gated events / 30000 total eve nts. Chapter 4 A B C Figure A 2. Real time RT PCR of MCF7P fo r estrogen response gene TFF1. A) Time course of TFF1 expression induced with 100 nM estrodiol (E2). B) and C) Inhibition of E2 induced TFF1 expression after 1 hour antiestrogens pre treatment. All TFF1 expression normalized to beta actin expression. Abbreviations: 4 OHT is 4 hydroxytamoxifen. ICI is ICI 182780. All TFF1 expression was normalized to beta actin expression. Error bars represented technical triplicates. Each gr aph is n=1.

PAGE 106

106 A B C Figure A 3 Real time RT PCR of MCF7S for es trogen response gene TFF1. A) Time course of TFF1 expression induced with 100 nM 17 estra diol (E2), TFF1 expression levels were shown relative to untreated cells. B) TFF1 expression

PAGE 107

107 after 12 hours induction with 100 nM 17 estradiol (E2) following 1 hour antiestrogen pre treatment. C) TFF1 expression after 24 hours induction with 100 nM 17 estradiol (E2) following 1 hour antiestrogen pre treatment. Abbreviations: 4 OHT is 4 hydroxytamoxifen. ICI is ICI 182780. All TFF1 expression was normalized to beta actin expression. Error bars repr esented technical triplicates. Each graph is n=1. Figure A 4. T ime course of CTSD gene expression induced with 100 nM 17 estradiol (E2). CTSD expression levels were shown relative to untreated cells and normalized to beta actin expression Error bars represented technical triplicates. The graph is n=1.

PAGE 108

108 A B

PAGE 109

109 C D Figure A 5 Brd U/PI c ell cycle analysis for MCF7S treated with antiestrogens for 72 hours. A) 0.1% DMSO treated control pulsed with 10 M BrdU for 2.5 hours. Total events were shown as R2 fraction in scatter plot on t he left. B) No BrdU pulse control MCF7S cells. Gated events were shown in scatter plot on the left. C) 1 M ICI treated MCF7S cells pulsed with 10 M BrdU for 2.5 hours. Gated events were shown in scatter plot on the left. D) 2.5 M 4 OHT treated MCF7S cells pulsed with 10 M BrdU for 2.5 hours. Gated events were shown in scatter plot on the left. All samples were analyzed by flow cytometer. Abbreviations: 4 OHT is 4 hydroxytamoxifen. ICI is pure antiestrogen ICI 182780. 30000 cells were count ed for each sample Data represents n=1.

PAGE 110

110 Table A 2. Data for BrdU/PI cell cycle analysis MCF7S cells treated 72 hours with antiestrogens (Figure A 5) Sample Gated Events (S Phase) Background Subtracted Events % Gated % Total (Not gated) BrdU 17 27495 0. 06 0.06 0.1% DMSO +BrdU 10474 28417 36.86 34.91 2.5 M 4 OHT +BrdU 7323 28056 26.1 24.41 1 M ICI +BrdU 7652 27844 27.48 25.51 Table A 2 data represents experiment data for BrdU/PI cell cycle analysis in MCF7S cells treated with antiestrogens for 72 hours (Figure A 5) A total of 30000 cells were counte d for each sample. Gated events represent BrdU incorporation during S phase of the cell cycle (Figure A 5 right). Background subtracted e vents represented cell fraction (R2) deemed unicellular based on scatter plot gating (Figure A 5 left) % Gated is p ercentage ratio of gated events / background subtracted events. % Total is percentage ratio of gated events / 30000 total events.

PAGE 111

111 Figure A 6 Individual Annexin V experimental data for antiestrogen treated MCF7S at 48 hours and 72 hours summari zed in Figure 5 5 Each graph represented one experiment ( n=1 ) Total percentage of apoptosis was shown. Abbreviations: 4 OHT is 4 hydroxytamoxifen. ICI is pure antiestrogen ICI 182780.

PAGE 112

112 Chapter 5 Figure A 7 Representative phase contrast images of shER C7 knockdown (left) a nd control shRNA (right) mammospheres. There were no observable differences in sphere size or formation frequency between the two cell types. Scale bars = 182 microns

PAGE 113

113 APPENDIX B BRD8 FUNCTION DURING EARLY CARDIAC DEVELOPMEN T Introduction Congenital heart defects are common conditions that affect a large portion of the population. Although medical advances have greatly extended the life of affected individuals, the underlying causes of these conditions are not fully charact erized Genetic and molecular approaches have identified a number of key regulators that form complex signaling networks directing e ach stage of heart development. In addition, there is emerging evidence for epigenetic regulation of de velopmental pathway s This histone variants by SWR1 chromatin remodeling complexes. To this end, we propose a SWR1 subunit, Brd8, as a novel protein involved in cardia c development. Th e goal of the n cardiogenesis by s tudy ing the morphological characteristic s of Brd8 mutant mice, identifying specific genes affected in Brd8 mutants, and investigating the mechanism by which Brd8 af fects downstream gene expression. Overview of Mammalian Heart D evelopment Major events in heart development are shown in the Figure 1, as reviewed in (Sr ivastava, 2006) Cardiac progenitor cells from the mesoderm form the primary and secondary heart field in the cardiogenic region. The first heart field (FHF) is localized along the anterior part of the embryos, while the second heart field (SHF) is loca ted posterior to the FHF. The cardiogenic region has undergone patterning and a ventral midline has been established, which guides the formation of the primitive heart tube from cells in the FHF. The first heart field differentiates into two symmetrical regions

PAGE 114

114 that extend from the midline; this structure is known as the cardiac crescent. The crescent has specific cell populations that will form the future left ventricle (LV) and atria (A) of the heart. The two halves of the crescent later converge alon g the midline. This forms the primitive heart tube, containing two distinct cell populations destined to be the ventricle (V) or the two halves of the atria (RA and LA). The SHF differentiates at this stage, and cells from the SHF migrate into both ends of the heart tube forming the future right ventricle (RV), the outflow tract or conotruncus (CT), and parts of the atria. The SHF also supports a population of non mesodermal cells called the neural crest, which contribute to the septation of the heart an d valve formation in late heart development. The heart tube then undergoes rightward looping, segments of the aortic arches are specified, and the chambers are formed by the asymmetric proliferation of various regions. Septation, valve formation, and mat uration of the aortic arches complete heart development (Srivastava, 2006) Transcription F actor s Regulating H eart F ormation Specification of Myocardial Progenitor C ells All tissue types originate from the epiblast, derived from the inner cell mass of the post implantation blastocyst. During gastrulation (E6.5 in mouse), the primitive streak forms on the posterior region of the embryo, the p rimary germ layers (ectoderm, endoderm, mesoderm) are defined and enter the primitive streak (Tam and Loebel, 2007) The earliest mesodermal genes are MesP1/2 and Fgf8, which are expressed during gastrulation (Abu Issa and Kirby, 2007; Brand, 2003) Induction of Nkx2.5 expression by upstream signaling induces differentiation of mesoderm germ cells into cardiomyocyte precursors (Xu and Baldini, 2007; Srivastava and Olson, 2000) These F ig. 2 from [1]

PAGE 115

115 upstream signals include members of the Bone Morphogeneti c Protein family (BMP) (Prall et al., 2007) as well as Sonic Hedgehog (Shh), Fibroblast Growth Factor (FGF), Wnt and Notch proteins (Srivastava, 2006; Kimelman, 2006; Dunwoodie, 2007) Differentiation of P recursors into C ardiomyocytes A number of other induction signals trigger differentiation of precursors into mature cardiomyocytes, and further specification as the first and second heart fields. The first heart field (FHF) is ch aracterized by activators such as GATA4 (Pu et al., 2004; Xin et al., 2006) which interacts with Nkx2.5 (Pu et al., 2004; Tanaka et al., 1999) and other downstream regulatory factors during chamber formation. Cells in the secondary heart field (SHF) differentiate at a later stage. They are first marked by the expression of Isl1 (Dodou et al., 2004; Park et al., 2006) and Foxh1 (Xu et al., 2007; von Both et al., 2004) The upstream factors in both heart fields interact with a multitude of downstream targets. These interactions are important for specifying polarity and asymmetry, organization of multiple cell ty pes, inducing proliferation and formation of defined cardiac structures (Dunwoodie, 2007) Downstream Transcriptional Networks Associated With Heart F ormation The plethora of downstream transcriptional targets overlap and cross regulate, creating a complex regulatory network. Many families of transcription factors contribute to the development of specific cardiac structure. Each family contain s numerous isoforms, giving rise to an enormous variety of combinational signaling. Their interactions as activating or repressive complexes direct and fine tune each step in the development of a four chamber heart from the two heart fields. This complex network has been intensely studied and a few are well characterized. Some examples are the Tbx family (Plageman and Yutzey, 2005; Stennard and Harvey, 2005; Naiche et al.,

PAGE 116

116 2005) Gata 4, 5, 6 (Pu et al., 2004; Patient and McGhee, 2002) Mef2 family (Karamboulas et al., 2006; Black and Olson, 1998) HANDs (Risebro et al ., 2006; Firulli, 2003) and certain Nkx2 (Tanaka et al., 1999; Heathcote et al., 2005; Pabst et al., 2000) General Introduction to Brd8 Brd8 protein was originally identified as a coactivator for thyroid receptors and retinoic acid receptors (Monden et al., 1997, 1999) It contain s two bromo domains, which are generally associated wit h acetylated histone binding. Brd8 is a subunit of various chromatin remodeling complexes, including HAT complexes and histone exchange com plexes such as Domin o (Raisner and Madhani, 2006; Cai et al., 2003) It has also been suggested to be a cell differentiation switch (Benevolenskaya et al ., 2005) Tentative evidence suggests that Brd8 is a downstream target of Tbx5 (Mori et al., 2006) a key regulator in cardiomyocyte differentiation and c ham ber formation. Taken together these data suggest that Brd8 can be mediating differentiation signals induced by nuclear receptors and thereby affect ing downstream gene regulation by interacting with trans acting cofactors. Brd8 may al so be organ specif ic as its expression is controlled by organ specific transcription factors. Materials and Methods Methods not previously described in Chapter 2 are described below. Immunohistochemistry Immunohistochemistry wa s performed as follows, the embryos are disse cted away from maternal tissue. They were then rinsed in PBS pH 7.3 (137mM NaCl, 2.7mM KCl, 4.3mM Na 2 HPO 4 7H 2 O, 1.4mM KH 2 PO 4 ) and fixed overnight with 4% paraformaldehyde (PFA) (w/v) at 4 C. Next day they were washed twice in PBS, dehydrated in a

PAGE 117

117 metha nol series and bleached with 5% hydrogen peroxide (v/v) in methanol for 2 hours to neutralize endogenous peroxidase a ctivity. They were rehydrated in a methanol series to PBS, blocked with 1.5% sheep serum in PBS, and incubated with primary antibody overn ight at 4 C. Samples were washed extensively in PBS, and HRP conjugated secondary antibody applied overnight at 4 C. They were washed as previously described after primary antibody incubati on. The samples were incubated with HRP conjugated secondary antibody for 1 hour at room temperature. Signal was detected diaminobenzidine tetrahydrochloride) DAB Plus substrate kit (Zymed). Once sufficie nt signal was achieved, samples were washed in PBS, postfixed in 4% PFA and 0.1% glutaraldehyde (w/v) in PBS, w ashed again in PBS, cleared in 50% and 70% glycerol in PBS. The samples were stored at 4 C until images were captured. Whole Mount In Situ Hybridization (WHISH) Riboprobe synthesis Riboprobes were generated from plasmids containing the cDNA gene of interest and bacteriophage T7/T3 promoters. Digoxigenin (DIG) labeled rib oprobes were synthe sized from 10 g linearized DNA template, 2 l DIG RNA labeling mix (Roche), 4 l of 5X tra nscription buffer, 100 mM DTT, 1 l RNase inhibitor, and 1 l of appropriate RNA polymerase to synthesis antisense/sense strand in a 20 l reaction. The reaction wa s incubated at 37 C for 2 hours, follow ed by DNase treatment. The reaction was stopped by the addition of 2 l 0.5M EDTA pH 8. The DIG labeled riboprobe was pelleted via ethanol precipitation with the aid of LiCl. The pel let wa s resuspended in DEPC treated water and sto red at 20 C until hybri dization.

PAGE 118

118 Mouse e mbryo preparation Dissected mouse embryos at desired time points were treated as follows prior to hybridization: fixed in 4% PFA at room temperature (RT) for 20 minutes and then dehydrated through a methanol seri es and stored at 80 C until hybridation. Whole mount in situ hybridization In Situ hybridization wa s performed as follo ws: embryos were rehydrated through methanol series to PBT (PBS+0.1% Tween 20), bleached in 6% hydrogen peroxide for 1 hour, washed in PBT, treated with 10 g/ml Proteinase K for 6 8 minutes, washed in 2 mg/ml glycine for 10 minutes, and rinsed in PBT. The samples were then postfixed in 4% PFA/0.2% glutaraldehyde (w/v) for 20 minutes, blocked in prehybridation solution (50% formamide, 5 X SSC, 0,1% Tween 20, 0.1% SDS, 50 g/ml heparine, 50 g/ml t RNA, 60mM citric acid) at 70 C for 1 hour, and hybridized overnight at 70 C in fresh prehybr idization buffer containing 0.2 g/ml D IG labeled probe. On the next day, samples were extensively washed at 65 C with wash solutions (5 0% formamide, 5X SSC, 60 24mM citric acid, 1 0.2% SDS, and 0.1% Tween 20) blocked in 2% Boehringer blocking reagent (w/v) and 10% sheep serum (v/v) dissolved in Maleic Acid Buffer ( MAB ). The samples were then incubated overnight in 1:2000 pre blocked, AP conjugated anti DIG antibody at 4 C. On the following day, samples were washed at length with MAB, with a final overnight wash in MAB at 4 C. Finally, samples were washed in alkaline phosphatases buffer (NTMT) and the signal was detected using BM Purple AP Substrate (Roche) Once suffi cient signal was obtained, samples were washed in PBT, postfixed in 4% PFA/ 0.1% glutaral dehyde for 1 hour, washed again in PBT and stored at 4 C until images of the stained embryos are photographed.

PAGE 119

119 RT PCR Primers for Brd8 Allele E xpression Primers used to confirm gene trap vector insertion is illustrated in Figure B 1B. The sequences were as follows: exon 1 (Brd AGGAGTGGGGATCAGAACTG beta geo of gene trap insert (Beta GTATCGGCCTCAGGAAGATCG exon 5 (Brd ex 5: 5 CTCAGCAGTCAGTTTGC GAAC mer set Brd 1 and Brd ex 5 amplified the wild type allele, while pri mer set Brd 1 and Beta geo a mplified the mutant allele. Brd8 +/ ES cell cDNA (XE487) were reversely transcribed u sing beta geo specific primers, and was included to show the expected size of mutant bands. Sample cDNA were obtained from E10.5 embryos following homogenization in Trizol to extract total RNA, and RT PCR as described in Chapter 2. Result s and Discussion The initial evaluation of Brd8 gene trapped whole mouse embr yos included gross morphological characterization a nd assessing Brd8 allelic expression level The Brd8 mutant mouse line demonstrated obvious embryonic phenotype such as developmental retardation and lethality by E10.5 (Figure B 2A) Semi quantitative R T PCR showed the Brd8 allele was not completely depleted in mutants, but low residual expression remained in both ES cell clone XE487 and Brd8 homozygous mutant embryos at E10.5 (Figure B 2B and Figure B 2C). Endogenous Brd8 expression in wild type E10.5 embryos was assessed using whole mount in situ hybridization to determine Brd8 localization. As shown in Figure B 3, endogenous Brd8 was ubiquitously expressed throughout the embryos with the ex ception of the cardiac region. The evidence suggested Brd8 w as an important cofactor during embryogenesis. However, the data implied the role of Brd8 is generalized and may not be organ specific. It was possible that the cardiac defects observed were the indirect result of generalized growth arrest

PAGE 120

120 The observed abnormal cardiac pathology (Figure B 4) may be attributed to the heart being one of the earliest organs to be fully developed a nd thus was one of the first organs affected by growth inhibition. The failures of the mutant embryo to turn and the neural tub e to close were additional indicator s of early growth inhibition (Figure B 2A and B 4). The malformation of essential organs could explain the early lethality of Brd8 mutants Further evaluation of Brd8 mutant embryos included whole mount immunohistoche mistry using PECAM antibody to compare vasculature development of Brd8 mutant embryos and stage matched wild type embryos. As shown in Figure B 5, Brd8 mutant embryos showed less robust PECAM signal in the neural, cardiac and dorsal regions. T he failure of the heart tube to undergo normal rightward looping was ev ident in the Brd8 mutant embryo (Figure B 5B and 5C) The mutant embryo displayed an enlargement of the ventricular regions of the heart tube, which suggested that cardiac patterning occurred ind ependently of positional cues. It also implied that cells in the heart tube were already programmed to become specific cardiac components very early in development. T he Brd8 mutant embryos sh owed cardiac specific morphological defects such as cardiac ed ema, abnormal ventricular enlargement and insufficient rightward looping of the heart tube. The expression levels and localization of specific cardiac differentiation markers were assessed by whole mount in situ hybridization to determine if the abnormal cardiac pathology had any bearing on ca rdiac cell fate determination. As shown in Figure B 6, Brd8 mutant embryonic heart tube retained a similar molecular signature as stage matched wild type embryos. Hybridization of Tbx5 (Figure B 6A),

PAGE 121

121 Nkx2.5 (Figure B 6B), and Mlc2v (Figure B 6C) in mutant embryos all showed a similar expression localization and pattern to wild type embryos. However, there were obvious ventricular enlargement and lack of rightward looping as mentioned earlie r. These results indicate d that while the patterning of the heart tube was fixed during early development, cardiac morphogenesis still required differential positional cues throughout the process to direct proper formation. Without which, the heart was physiologically defective a nd resulted in embryonic lethality. The preliminary data provided here demonstrated a generalized, but essential role for Brd8 during mammalian development. Even though the Brd8 allele may not be fully silenced (Figure B 2B and 2C), it was enough to cau se developmental arrest and physiological defects that resulted in let hality. However, Brd8 did not appear to affect early specification or differentiation of mesodermal cells to cardiomyocytes. This would suggest the cardiac defects observed were second ary effects initialed by Brd8 downregulation. Additional cardiac molecular markers may be used to further evaluate cardiac patterning and dissect this assumption. It is unclear how Brd8 expression may influence embryonic development. It may be speculate d that the ubiquitous expression of Brd8 (Figure B 3) would affect multiple regulators. The deregulation of these regulators culminated into an unsustainable combination that initiated cell death. These factors may include Sonic Hedgehog, BMP, Wnt, Notch or FGF. It can be of interest to study Brd8 effects on known signaling pathways and cellular proliferation. This course of inquiry can offer additional insight into the effects of Brd8 in mammalian development.

PAGE 122

122 A B Figure B 1. Gene trap vector in serted into mouse genome to generate Brd8 null allele. A) Design of gene trap vector used (SIGTR). Abbreviations: Intron is 1. 5 kb of Mouse En2 intron 1. SA is splice acceptor of mouse En2 exon 2. geo is galactosidas e and neomycin transferase. pA is SV40 polyadenylation signal. B) Diagram of gene trap vector insertion into Brd8 intron 2 for both mutant and wil d type is Brd ex t knock in vector is Beta geo. Diagram is not to scale.

PAGE 123

123 A B C Figure B 2. Brd8 gene trapped mouse embryos and expression of wild type and mutant alleles. A) Bri ght field microscopy image of wild type (right) and mutant Brd8 (left) embryos at E9.5. Brd8 mutant embryos displayed overall growth retardation and cardiac edema (white arrow). B) Semi quantitative RT PCR to assay Brd8 allelic exp ression in cDNA extract ed from Brd8 mutant (Brd8 / ) and Brd8 heterozygous (Brd8 +/ ) E10.5 embryos. The p rimer s were located as described in Figure B 1B. The cD NA were amplified for 35 cycles and performed in duplicate. Abbreviations: XE 487 is Brd8 / ES cells cDNA control. Mutant (Mut) and wild type (WT) PCR product bands ar e indicated on the right. C) B eta actin semi quantitative RT PCR normalization controls were shown for mutant Brd8 E10.5 embryos cDNA (left), and Brd8 heterozygous (Brd8 +/ ) cDNA (right). PCR cycle numbers are indicated below PCR product band.

PAGE 124

124 Figure B 3. Endogenous Brd8 expression in wild type E10.5 mouse embryos using whole mount in situ hybridization (WMISH) E10.5 wild type embryos were hybridized to Brd8 antisense probe and compared to l itter matched embryos hybri dized to sense probe as control for 1, 5, 10 or 23 hours. There was no signal for sense probe, as expected. Brd8 specific antisense probe indicated Brd8 was express ed throughout the embryo, except for the cardiac region (red ci rcle).

PAGE 125

125 Figure B 4. Hematoxylin and eosin stained sections of Brd8 mutant embryos (E10.5) compared to wild type littermates. Brd8 mutant embryo (right) displayed general growth retardation as indicated by neural tube defect (black arrow). It showed possible signs of cardiac bifidia (red point). The respective sections were compared to E10.5 wild type sections (left). The samples were processed by Jim Richardson ab at University of Texas Southwestern Medical Center, Molecular Pathology Core. The sections were evaluated by Dr. Hideko Kasahara at University of Florida, Department of Physiology and Functional Genomics.

PAGE 126

126 A B

PAGE 127

127 C Figure B 5. PECAM whole mount immunohistochemistry comparing vascularization of Brd8 mutant and stage matched wild ty pe embryos (E8 .5) A) Wild type E8.5 embryo immunostain for PECAM viewed from various angles. Black arrow denotes embryonic heart tube. The embryo showed robust vasculature formation in neural, heart and dorsal regions. The heart tube showed normal ri ghtward looping, as expected at this embryonic stage. B) Brd8 mutant (Brd8 / ) E9.5 embryo immunostain for PECAM viewed from various angles. Black arrow denotes embryonic heart tube. The embryonic heart tube showed incomplete rightward looping. Vascu larization of the mutant embryo also appeared to be deficient in the neural and dorsal region s. However, vascularization of the yolk sac was robust. C) Left side view and comparison of PECAM immunostain E8.5 wild type (left) and E9.5 Brd8 mutant (right) embryos. Black arrows indicate embryonic heart tube. The comparison demonstrates mutant embryo retarded development in the neural, heart tube, and dorsal regions.

PAGE 128

128 A B C Figure B 6. Whole mount in situ hybridization (WMISH) characterizing expre ssion of key car diac development genes. A) E8.5 wild type compared to E10.5 Brd8 mutant embryos hybridized to Tbx5 probe. The embryos were shown from the left (left image) and ventral angle (right image). Tbx5 is specific for ventricle region and it is seen in the two samples. The Brd8 mutant embryos display an enlarged heart tube. B) E8.5 wild type compared to E10.5 Brd8 mutant embryos hybridized to Nkx2.5 probe. The embryos were shown from the left (left image) and ventral angle (right image). Nkx2 .5 is specific for the

PAGE 129

129 entire embryonic heart tube. Both embryo s showed positive signal for Nkx2.5 in the heart tube region. Wild type embryo showed stronger signal in the inflow and outflow tract at E8.5, and weaker signal at atrial and ventricle region s. Brd8 mutant embryo showed robust signal for the entire heart tube. Defective rightward looping and heart tube enlargement was observed in the mutant embryo. C ) E8.5 wild type compared to E10.5 Brd8 mutant embryos hybridized to Mlc2v probe The embry os are shown from ventral view. Mlc2v is specific for ventricular region of the heart tube. Both embryos showed positive signal in the ventricular region. Brd8 mutant embryo (right) showed abnormal enl argement of the heart tube, compared to stage matche d wild type embryo (left)

PAGE 130

130 LIST OF REFERENCES Abraham, B. K., Fritz, P., McClellan, M., Hauptvogel, P., Athelogou, M., and Brauch, H. (2005). Prevalence of CD44+/CD24 /low cells in breast can cer may not be associated with clinical outcome but may favor distant metastasis. Clin Cancer Res 11 1154 9. Abu Issa, R., and Kirby, M. L. (2007). Heart Field: From Mesoderm to Heart Tube. Annu Rev Cell Dev Biol. Available at: http://www.ncbi.nlm.nih.go v/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt= Citation&list_uids=17456019. ACS (2009). Breast Cancer Detailed Guide. Al Hajj, M., Wicha, M. S., Benito Hernandez, A., Morrison, S. J., and Clarke, M. F. (2003). Prospective identification of tumorigenic b reast cancer cells. Proceedings of the National Academy of Sciences of the United States of America 100 3983 8. Alison, M. R., Islam, S., and Lim, S. M. (2009). Number crunching in the cancer stem cell market. Breast Cancer Res 11 302. Altekruse, S., K osary, C., Krapcho, M., Krapcho, N., Aminou, R., Waldron, W., Ruhl, J., Howlader, N., Tatalovich, Z., Cho, H., et al. (2009). SEER Cancer Statistics Review, 1975 2007 (National Cancer Institute). Ariazi, E. A., Ariazi, J. L., Cordera, F., and Jordan, V. C (2006). Estrogen receptors as therapeutic targets in breast cancer. Current topics in medicinal chemistry 6 181 202. Arnold, S. F., Obourn, J. D., Jaffe, H., and Notides, A. C. (1995). Phosphorylation of the human estrogen receptor on tyrosine 537 in v ivo and by src family tyrosine kinases in vitro. Mol. Endocrinol 9 24 33. Aruffo, A., Stamenkovic, I., Melnick, M., Underhill, C. B., and Seed, B. (1990). CD44 is the principal cell surface receptor for hyaluronate. Cell 61 1303 1313. Asselin Labat, M. L., Shackleton, M., Stingl, J., Vaillant, F., Forrest, N. C., Eaves, C. J., Visvader, J. E., and Lindeman, G. J. (2006). Steroid hormone receptor status of mouse mammary stem cells. Journal of the National Cancer Institute 98 1011 4. Asselin Labat, M. L ., Sutherland, K. D., Barker, H., Thomas, R., Shackleton, M., Forrest, N. C., Hartley, L., Robb, L., Grosveld, F. G., van der Wees, J., et al. (2007). Gata 3 is an essential regulator of mammary gland morphogenesis and luminal cell differentiation. Nature cell biology 9 201 9.

PAGE 131

131 Asselin Labat, M., Vaillant, F., Sheridan, J. M., Pal, B., Wu, D., Simpson, E. R., Yasuda, H., Smyth, G. K., Martin, T. J., Lindeman, G. J., et al. (2010). Control of mammary stem cell function by steroid hormone signalling. Nature advance online publication Available at: http://dx.doi.org/10.1038/nature09027 [Accessed April 26, 2010]. Atanaskova, N., Keshamouni, V. G., Krueger, J. S., Schwartz, J. A., Miller, F., and Reddy, K. B. (2002). MAP kinase/estrogen receptor cross talk en hances estrogen mediated signaling and tumor growth but does not confer tamoxifen resistance. Oncogene 21 4000 8. Baek, S. H., Ohgi, K. A., Rose, D. W., Koo, E. H., Glass, C. K., and Rosenfeld, M. G. (2002). Exchange of N CoR corepressor and Tip60 coacti vator complexes links gene expression by NF kappaB and beta amyloid precursor protein. Cell 110 55 67. Benevolenskaya, E. V., Murray, H. L., Branton, P., Young, R. A., and Kaelin, W. G. (2005). Binding of pRB to the PHD protein RBP2 promotes cellular dif ferentiation. Mol Cell 18 623 35. Bjornstrom, L., and Sjoberg, M. (2005). Mechanisms of estrogen receptor signaling: convergence of genomic and nongenomic actions on target genes. Molecular endocrinology (Baltimore, Md 19 833 42. Black, B. L., and Olso n, E. N. (1998). Transcriptional control of muscle development by myocyte enhancer factor 2 (MEF2) proteins. Annu Rev Cell Dev Biol 14 167 96. Bonnet, D., and Dick, J. E. (1997). Human acute myeloid leukemia is organized as a hierarchy that originates fr om a primitive hematopoietic cell. Nature medicine 3 730 7. Booth, B. W., and Smith, G. H. (2006). Estrogen receptor alpha and progesterone receptor are expressed in label retaining mammary epithelial cells that divide asymmetrically and retain their tem plate DNA strands. Breast Cancer Res 8 R49. von Both, I., Silvestri, C., Erdemir, T., Lickert, H., Walls, J. R., Henkelman, R. M., Rossant, J., Harvey, R. P., Attisano, L., and Wrana, J. L. (2004). Foxh1 is essential for development of the anterior heart field. Dev Cell 7 331 45. Brand, T. (2003). Heart development: molecular insights into cardiac specification and early morphogenesis. Dev Biol 258 1 19. Britton, D. J., Hutcheson, I. R., Knowlden, J. M., Barrow, D., Giles, M., McClelland, R. A., Gee, J. M. W., and Nicholson, R. I. (2006). Bidirectional cross talk between ERalpha and EGFR signalling pathways regulates tamoxifen resistant growth. Breast Cancer Res. Treat 96 131 146.

PAGE 132

132 Burdall, S., Hanby, A., Lansdown, M., and Speirs, V. (2003). Breast can cer cell lines: friend or foe? Breast Cancer Res 5 89 95. Cai, Y., Jin, J., Tomomori Sato, C., Sato, S., Sorokina, I., Parmely, T. J., Conaway, R. C., and Conaway, J. W. (2003). Identification of new subunits of the multiprotein mammalian TRRAP/TIP60 con taining histone acetyltransferase complex. J Biol Chem 278 42733 6. Campbell, L. L., and Polyak, K. (2007). Breast tumor heterogeneity: cancer stem cells or clonal evolution? Cell cycle (Georgetown, Tex 6 2332 8. Campbell, R. A., Bhat Nakshatri, P., Pa tel, N. M., Constantinidou, D., Ali, S., and Nakshatri, H. (2001). Phosphatidylinositol 3 kinase/AKT mediated activation of estrogen receptor alpha: a new model for anti estrogen resistance. The Journal of biological chemistry 276 9817 24. Cariati, M., N aderi, A., Brown, J. P., Smalley, M. J., Pinder, S. E., Caldas, C., and Purushotham, A. D. (2008). Alpha 6 integrin is necessary for the tumourigenicity of a stem cell like subpopulation within the MCF7 breast cancer cell line. International journal of can cer 122 298 304. Carroll, J. S., Meyer, C. A., Song, J., Li, W., Geistlinger, T. R., Eeckhoute, J., Brodsky, A. S., Keeton, E. K., Fertuck, K. C., Hall, G. F., et al. (2006). Genome wide analysis of estrogen receptor binding sites. Nat. Genet 38 1289 12 97. Chan, C. M. W., Martin, L., Johnston, S. R. D., Ali, S., and Dowsett, M. (2002). Molecular changes associated with the acquisition of oestrogen hypersensitivity in MCF 7 breast cancer cells on long term oestrogen deprivation. J. Steroid Biochem. Mol. Biol 81 333 341. Chan, K. S., Espinosa, I., Chao, M., Wong, D., Ailles, L., Diehn, M., Gill, H., Presti, J., Chang, H. Y., van de Rijn, M., et al. (2009). Identification, molecular characterization, clinical prognosis, and therapeutic targeting of human bladder tumor initiating cells. Proceedings of the National Academy of Sciences of the United States of America 106 14016 21. Charafe Jauffret, E., Ginestier, C., Iovino, F., Wicinski, J., Cervera, N., Finetti, P., Hur, M. H., Diebel, M. E., Monville, F. Dutcher, J., et al. (2009). Breast cancer cell lines contain functional cancer stem cells with metastatic capacity and a distinct molecular signature. Cancer research 69 1302 13. Chen, D., Pace, P. E., Coombes, R. C., and Ali, S. (1999). Phosphorylatio n of Human Estrogen Receptor alpha by Protein Kinase A Regulates Dimerization. Mol. Cell. Biol. 19 1002 1015.

PAGE 133

133 Chen, M. S., Woodward, W. A., Behbod, F., Peddibhotla, S., Alfaro, M. P., Buchholz, T. A., and Rosen, J. M. (2007). Wnt/beta catenin mediates radiation resistance of Sca1+ progenitors in an immortalized mammary gland cell line. Journal of Cell Science 120 468 477. Clarke, R., Liu, M. C., Bouker, K. B., Gu, Z., Lee, R. Y., Zhu, Y., Skaar, T. C., Gomez, B., O'Brien, K., Wang, Y., et al. (2003). Antiestrogen resistance in breast cancer and the role of estrogen receptor signaling. Oncogene 22 7316 39. Clarke, R. B., Spence, K., Anderson, E., Howell, A., Okano, H., and Potten, C. S. (2005). A putative human breast stem cell population is enriched for steroid receptor positive cells. Developmental biology 277 443 56. Couse, J. F., and Korach, K. S. (1999). Estrogen receptor null mice: what have we learned and where will they lead us? Endocr. Rev 20 358 417. Coward, P., Lee, D., Hull, M. V., and Lehmann, J. M. (2001). 4 Hydroxytamoxifen binds to and deactivates the estrogen related receptor gamma. Proceedings of the National Academy of Sciences of the United States of America 98 8880 4. Curtis Hewitt, S., F Couse, J., and S Korach, K. (2000). Es trogen receptor transcription and transactivation: Estrogen receptor knockout mice what their phenotypes reveal about mechanisms of estrogen action. Breast Cancer Res 2 345 352. Dignam, J., Dukic, V., Anderson, S., Mamounas, E., Wickerham, D., and Wolm ark, N. (2009). Hazard of recurrence and adjuvant treatment effects over time in lymph node negative breast cancer. Breast Cancer Research and Treatment 116 595 602. Dodou, E., Verzi, M. P., Anderson, J. P., Xu, S., and Black, B. L. (2004). Mef2c is a di rect transcriptional target of ISL1 and GATA factors in the anterior heart field during mouse embryonic development. Development 131 3931 3942. Dontu, G., Abdallah, W. M., Foley, J. M., Jackson, K. W., Clarke, M. F., Kawamura, M. J., and Wicha, M. S. (20 03). In vitro propagation and transcriptional profiling of human mammary stem/progenitor cells. Genes & development 17 1253 70. Dontu, G., El Ashry, D., and Wicha, M. S. (2004). Breast cancer, stem/progenitor cells and the estrogen receptor. Trends in en docrinology and metabolism: TEM 15 193 7. Dontu, G., Jackson, K., McNicholas, E., Kawamura, M., Abdallah, W., and Wicha, M. (2004). Role of Notch signaling in cell fate determination of human mammary stem/progenitor cells. Breast Cancer Res 6 R605 R615. Dudek, P., and Picard, D. (2008). Genomics of signaling crosstalk of estrogen receptor alpha in breast cancer cells. PloS one 3 e1859.

PAGE 134

134 Dunwoodie, S. L. (2007). Combinatorial signaling in the heart orchestrates cardiac induction, lineage specification an d chamber formation. Semin Cell Dev Biol 18 54 66. Effects of chemotherapy and hormonal therapy for early breast cancer on recurrence and 15 year survival: an overview of the randomised trials (2005). Lancet 365 1687 1717. Fan, M., Yan, P. S., Hartman Frey, C., Chen, L., Paik, H., Oyer, S. L., Salisbury, J. D., Cheng, A. S., Li, L., Abbosh, P. H., et al. (2006). Diverse Gene Expression and DNA Methylation Profiles Correlate with Differential Adaptation of Breast Cancer Cells to the Antiestrogens Tamoxif en and Fulvestrant. Cancer Res 66 11954 11966. Farnie, G., Clarke, R. B., Spence, K., Pinnock, N., Brennan, K., Anderson, N. G., and Bundred, N. J. (2007). Novel Cell Culture Technique for Primary Ductal Carcinoma In Situ: Role of Notch and Epidermal Gro wth Factor Receptor Signaling Pathways. J. Natl. Cancer Inst. 99 616 627. Fillmore, C., and Kuperwasser, C. (2007). Human breast cancer stem cell markers CD44 and CD24: enriching for cells with functional properties in mice or in man? Breast Cancer Res 9 303. Fillmore, C. M., and Kuperwasser, C. (2008). Human breast cancer cell lines contain stem like cells that self renew, give rise to phenotypically diverse progeny and survive chemotherapy. Breast Cancer Res 10 R25. Firulli, A. B. (2003). A HANDful of questions: the molecular biology of the heart and neural crest derivatives (HAND) subclass of basic helix loop helix transcription factors. Gene 312 27 40. Fliss, A. E., Benzeno, S., Rao, J., and Caplan, A. J. (2000). Control of estrogen receptor liga nd binding by Hsp90. The Journal of Steroid Biochemistry and Molecular Biology 72 223 230. Fu, X., and Simoncini, T. (2008). Extra nuclear signaling of estrogen receptors. IUBMB Life 60 502 510. Fukazawa, H., Noguchi, K., Murakami, Y., and Uehara, Y. ( 2002). Mitogen activated protein/extracellular signal regulated kinase kinase (MEK) inhibitors restore anoikis sensitivity in human breast cancer cell lines with a constitutively activated extracellular regulated kinase (ERK) pathway. Mol. Cancer Ther 1 3 03 309. Ginestier, C., Hur, M. H., Charafe Jauffret, E., Monville, F., Dutcher, J., Brown, M., Jacquemier, J., Viens, P., Kleer, C. G., Liu, S., et al. (2007). ALDH1 is a marker of normal and malignant human mammary stem cells and a predictor of poor clin ical outcome. Cell stem cell 1 555 67.

PAGE 135

135 Godar, S., Ince, T. A., Bell, G. W., Feldser, D., Donaher, J. L., Bergh, J., Liu, A., Miu, K., Watnick, R. S., Reinhardt, F., et al. (2008). Growth inhibitory and tumor suppressive functions of p53 depend on its rep ression of CD44 expression. Cell 134 62 73. Green, K. A., and Carroll, J. S. (2007). Oestrogen receptor mediated transcription and the influence of co factors and chromatin state. Nat Rev Cancer 7 713 722. Grimshaw, M. J., Cooper, L., Papazisis, K., Co leman, J. A., Bohnenkamp, H. R., Chiapero Stanke, L., Taylor Papadimitriou, J., and Burchell, J. M. (2008). Mammosphere culture of metastatic breast cancer cells enriches for tumorigenic breast cancer cells. Breast Cancer Res 10 R52. Gupta, P. B., Onder, T. T., Jiang, G., Tao, K., Kuperwasser, C., Weinberg, R. A., and Lander, E. S. (2009). Identification of selective inhibitors of cancer stem cells by high throughput screening. Cell 138 645 59. Hanahan, D., and Weinberg, R. A. (2000). The hallmarks of c ancer. Cell 100 57 70. Hayashi, S., and Yamaguchi, Y. (2008). Estrogen signaling pathway and hormonal therapy. Breast cancer (Tokyo, Japan) 15 256 61. Heathcote, K., Braybrook, C., Abushaban, L., Guy, M., Khetyar, M. E., Patton, M. A., Carter, N. D., S cambler, P. J., and Syrris, P. (2005). Common arterial trunk associated with a homeodomain mutation of NKX2.6. Hum Mol Genet 14 585 93. Howard, B. A., and Gusterson, B. A. (2000). Human breast development. J Mammary Gland Biol Neoplasia 5 119 137. Hutc heson, I. R., Knowlden, J. M., Madden, T. A., Barrow, D., Gee, J. M., Wakeling, A. E., and Nicholson, R. I. (2003). Oestrogen receptor mediated modulation of the EGFR/MAPK pathway in tamoxifen resistant MCF 7 cells. Breast cancer research and treatment 81 81 93. Hwang Verslues, W. W., Kuo, W. H., Chang, P. H., Pan, C. C., Wang, H. H., Tsai, S. T., Jeng, Y. M., Shew, J. Y., Kung, J. T., Chen, C. H., et al. (2009). Multiple lineages of human breast cancer stem/progenitor cells identified by profiling with s tem cell markers. PloS one 4 e8377. Ince, T. A., Richardson, A. L., Bell, G. W., Saitoh, M., Godar, S., Karnoub, A. E., Iglehart, J. D., and Weinberg, R. A. (2007). Transformation of Different Human Breast Epithelial Cell Types Leads to Distinct Tumor Ph enotypes. Cancer Cell 12 160 170.

PAGE 136

136 Jakacka, M., Ito, M., Weiss, J., Chien, P. Y., Gehm, B. D., and Jameson, J. L. (2001). Estrogen receptor binding to DNA is not required for its activity through the nonclassical AP1 pathway. J. Biol. Chem 276 13615 13 621. Jeng, M. H., Jiang, S. Y., and Jordan, V. C. (1994). Paradoxical regulation of estrogen dependent growth factor gene expression in estrogen receptor (ER) negative human breast cancer cells stably expressing ER. Cancer Lett 82 123 128. Jin, L., Hope K. J., Zhai, Q., Smadja Joffe, F., and Dick, J. E. (2006). Targeting of CD44 eradicates human acute myeloid leukemic stem cells. Nat. Med 12 1167 1174. Jones, P. S., Parrott, E., and White, I. N. (1999). Activation of transcription by estrogen receptor alpha and beta is cell type and promoter dependent. The Journal of biological chemistry 274 32008 14. Jordan, V. C. (2007). Chemoprevention of breast cancer with selective oestrogen receptor modulators. Nature reviews 7 46 53. Kakarala, M., Brenner, D. E., Korkaya, H., Cheng, C., Tazi, K., Ginestier, C., Liu, S., Dontu, G., and Wicha, M. S. (2009). Targeting breast stem cells with the cancer preventive compounds curcumin and piperine. Breast cancer research and treatment. Available at: internal pdf:// Spices kill BCSC Wicha 2009 0573672704/Spices kill BCSC Wicha 2009.pdf. Karamboulas, C., Dakubo, G. D., Liu, J., De Repentigny, Y., Yutzey, K., Wallace, V. A., Kothary, R., and Skerjanc, I. S. (2006). Disruption of MEF2 activity in cardiomyoblasts inhibit s cardiomyogenesis. J Cell Sci 119 4315 21. Keen, J. C., Yan, L., Mack, K. M., Pettit, C., Smith, D., Sharma, D., and Davidson, N. E. (2003). A novel histone deacetylase inhibitor, scriptaid, enhances expression of functional estrogen receptor alpha (ER) in ER negative human breast cancer cells in combination with 5 aza 2' deoxycytidine. Breast Cancer Res. Treat 81 177 186. Kelly, P. N., Dakic, A., Adams, J. M., Nutt, S. L., and Strasser, A. (2007). Tumor growth need not be driven by rare cancer stem ce lls. Science (New York, N.Y 317 337. Kern, S. E., and Shibata, D. (2007). The fuzzy math of solid tumor stem cells: a perspective. Cancer research 67 8985 8. Kimelman, D. (2006). Mesoderm induction: from caps to chips. Nat Rev Genet 7 360 72.

PAGE 137

137 Kok, M., Koornstra, R. H., Margarido, T. C., Fles, R., Armstrong, N. J., Linn, S. C., Van't Veer, L. J., and Weigelt, B. (2009). Mammosphere derived gene set predicts outcome in patients with ER positive breast cancer. The Journal of pathology 218 316 26. Ku iper, G. G., Lemmen, J. G., Carlsson, B., Corton, J. C., Safe, S. H., van der Saag, P. T., van der Burg, B., and Gustafsson, J. A. (1998). Interaction of estrogenic chemicals and phytoestrogens with estrogen receptor beta. Endocrinology 139 4252 63. Kuma r, R., Wang, R., Mazumdar, A., Talukder, A. H., Mandal, M., Yang, Z., Bagheri Yarmand, R., Sahin, A., Hortobagyi, G., Adam, L., et al. (2002). A naturally occurring MTA1 variant sequesters oestrogen receptor [alpha] in the cytoplasm. Nature 418 654 657. Lacroix, M., and Leclercq, G. (2004). Relevance of breast cancer cell lines as models for breast tumours: an update. Breast Cancer Res. Treat 83 249 289. LaMarca, H. L., and Rosen, J. M. (2008). Minireview: hormones and mammary cell fate what will I bec ome when I grow up? Endocrinology 149 4317 4321. Lannigan, D. A. (2003). Estrogen receptor phosphorylation. Steroids 68 1 9. Lazennec, G., and Katzenellenbogen, B. S. (1999). Expression of human estrogen receptor using an efficient adenoviral gene deli very system is able to restore hormone dependent features to estrogen receptor negative breast carcinoma cells. Mol. Cell. Endocrinol 149 93 105. Lee, H., and Bai, W. (2002). Regulation of Estrogen Receptor Nuclear Export by Ligand Induced and p38 Mediat ed Receptor Phosphorylation. Mol. Cell. Biol. 22 5835 5845. Lewis, J. S., and Jordan, V. C. (2005). Selective estrogen receptor modulators (SERMs): mechanisms of anticarcinogenesis and drug resistance. Mutation research 591 247 63. Li, L., Haynes, M. P ., and Bender, J. R. (2003). Plasma membrane localization and Proceedings of the National Academy of Sciences of the United States of America 100 4807 4812. Li, X., Lewis, M. T., Huang, J., Gutierrez, C., Osborne, C. K., Wu, M. F., Hilsenbeck, S. G., Pavlick, A., Zhang, X., Chamness, G. C., et al. (2008). Intrinsic resistance of tumorigenic breast cancer cells to chemotherapy. Journal of the National Cancer Institute 100 672 9

PAGE 138

138 Lim, E., Vaillant, F., Wu, D., Forrest, N. C., Pal, B., Hart, A. H., Asselin Labat, M. L., Gyorki, D. E., Ward, T., Partanen, A., et al. (2009). Aberrant luminal progenitors as the candidate target population for basal tumor development in BRCA1 mutati on carriers. Nature medicine 15 907 13. Lin, Z., Reierstad, S., Huang, C., and Bulun, S. E. (2007). Novel estrogen receptor alpha binding sites and estradiol target genes identified by chromatin immunoprecipitation cloning in breast cancer. Cancer Res 67 5017 5024. Lindberg, K., Strm, A., Lock, J. G., Gustafsson, J., Haldosn, L., and Helguero, L. A. (2010). Expression of estrogen receptor beta increases integrin alpha1 and integrin beta1 levels and enhances adhesion of breast cancer cells. J. Cell. Ph ysiol 222 156 167. Liu, M., Casimiro, M. C., Wang, C., Shirley, L. A., Jiao, X., Katiyar, S., Ju, X., Li, Z., Yu, Z., Zhou, J., et al. (2009). p21CIP1 attenuates Ras and c Myc dependent breast tumor epithelial mesenchymal transition and cancer stem cell like gene expression in vivo. Proceedings of the National Academy of Sciences of the United States of America. Available at: internal pdf://p21cip1 block EMT to BCSC via Ras c myc PNAS 2009 0980524288/p21cip1 block EMT to BCSC via Ras c myc PNAS 2009.pdf. Liu, R., Wang, X., Chen, G. Y., Dalerba, P., Gurney, A., Hoey, T., Sherlock, G., Lewicki, J., Shedden, K., and Clarke, M. F. (2007). The prognostic role of a gene signature from tumorigenic breast cancer cells. The New England journal of medicine 356 21 7 26. Liu, S., Dontu, G., Mantle, I. D., Patel, S., Ahn, N. S., Jackson, K. W., Suri, P., and Wicha, M. S. (2006). Hedgehog signaling and Bmi 1 regulate self renewal of normal and malignant human mammary stem cells. Cancer research 66 6063 71. Liu, S., Dontu, G., and Wicha, M. S. (2005). Mammary stem cells, self renewal pathways, and carcinogenesis. Breast Cancer Res 7 86 95. Lobo, N. A., Shimono, Y., Qian, D., and Clarke, M. F. (2007). The biology of cancer stem cells. Annual review of cell and develo pmental biology 23 675 99. Losel, R. M., Falkenstein, E., Feuring, M., Schultz, A., Tillmann, H., Rossol Haseroth, K., and Wehling, M. (2003). Nongenomic steroid action: controversies, questions, and answers. Physiol. Rev 83 965 1016. Marsaud, V., Goug elet, A., Maillard, S., and Renoir, J. (2003). Various phosphorylation pathways, depending on agonist and antagonist binding to endogenous estrogen receptor alpha (ERalpha), differentially affect ERalpha extractability, proteasome mediated stability, and t ranscriptional activity in human breast cancer cells. Mol. Endocrinol 17 2013 2027.

PAGE 139

139 Martin, L. A., Farmer, I., Johnston, S. R., Ali, S., Marshall, C., and Dowsett, M. (2003). Enhanced estrogen receptor (ER) alpha, ERBB2, and MAPK signal transduction pathw ays operate during the adaptation of MCF 7 cells to long term estrogen deprivation. The Journal of biological chemistry 278 30458 68. Martin, L., Farmer, I., Johnston, S. R. D., Ali, S., and Dowsett, M. (2005). Elevated ERK1/ERK2/estrogen receptor cross talk enhances estrogen mediated signaling during long term estrogen deprivation. Endocr. Relat. Cancer 12 Suppl 1 S75 84. McDonnell, D. P., and Norris, J. D. (2002). Connections and Regulation of the Human Estrogen Receptor. Science 296 1642 1644. Meti vier, R., Reid, G., and Gannon, F. (2006). Transcription in four dimensions: nuclear receptor directed initiation of gene expression. EMBO reports 7 161 7. Mtivier, R., Penot, G., Hbner, M. R., Reid, G., Brand, H., Kos, M., and Gannon, F. (2003). Estro gen receptor alpha directs ordered, cyclical, and combinatorial recruitment of cofactors on a natural target promoter. Cell 115 751 763. Monden, T., Kishi, M., Hosoya, T., Satoh, T., Wondisford, F. E., Hollenberg, A. N., Yamada, M., and Mori, M. (1999). p120 acts as a specific coactivator for 9 cis retinoic acid receptor (RXR) on peroxisome proliferator activated receptor gamma/RXR heterodimers. Mol Endocrinol 13 1695 703. Monden, T., Wondisford, F. E., and Hollenberg, A. N. (1997). Isolation and charac terization of a novel ligand dependent thyroid hormone receptor coactivating protein. J Biol Chem 272 29834 41. Mori, A. D., Zhu, Y., Vahora, I., Nieman, B., Koshiba Takeuchi, K., Davidson, L., Pizard, A., Seidman, J. G., Seidman, C. E., Chen, X. J., et al. (2006). Tbx5 dependent rheostatic control of cardiac gene expression and morphogenesis. Dev Biol 297 566 86. Morrison, B. J., Schmidt, C. W., Lakhani, S. R., Reynolds, B. A., and Lopez, J. A. (2008). Breast cancer stem cells: implications for therapy of breast cancer. Breast Cancer Res 10 210. Murohashi, M., Hinohara, K., Kuroda, M., Isagawa, T., Tsuji, S., Kobayashi, S., Umezawa, K., Tojo, A., Aburatani, H., and Gotoh, N. (2009). Gene set enrichment analysis provides insight into novel signalling p athways in breast cancer stem cells. British journal of cancer. Available at: internal pdf://Murahashi M, BrJC 2009, TGFbeat NFkappa etc ID as upreg in TIC 0053868804/Murahashi M, BrJC 2009, TGFbeat NFkappa etc ID as upreg in TIC.pdf. Naiche, L. A., Harre lson, Z., Kelly, R. G., and Papaioannou, V. E. (2005). T box genes in vertebrate development. Annu Rev Genet 39 219 39.

PAGE 140

140 Neumeister, V., and Rimm, D. (2009). Is ALDH1 a good method for definition of breast cancer stem cells? Breast cancer research and tre atment. Available at: internal pdf://Neumeister V, BCRT 2009, ALDH1 as marker for BCSC rebuttal 2214998528/Neumeister V, BCRT 2009, ALDH1 as marker for BCSC rebuttal.pdf. Obrero, M., Yu, D. V., and Shapiro, D. J. (2002). Estrogen receptor dependent and es trogen receptor independent pathways for tamoxifen and 4 hydroxytamoxifen induced programmed cell death. The Journal of biological chemistry 277 45695 703. Osborne, C. K., Hobbs, K., and Trent, J. M. (1987). Biological differences among MCF 7 human breas t cancer cell lines from different laboratories. Breast Cancer Research and Treatment 9 111 121. Osipo, C., Gajdos, C., Liu, H., Chen, B., and Jordan, V. C. (2003). Paradoxical Action of Fulvestrant in Estradiol Induced Regression of Tamoxifen Stimulated Breast Cancer. J. Natl. Cancer Inst. 95 1597 1608. Pabst, O., Herbrand, H., Takuma, N., and Arnold, H. H. (2000). NKX2 gene expression in neuroectoderm but not in mesendodermally derived structures depends on sonic hedgehog in mouse embryos. Dev Genes E vol 210 47 50. Park, E. J., Ogden, L. A., Talbot, A., Evans, S., Cai, C. L., Black, B. L., Frank, D. U., and Moon, A. M. (2006). Required, tissue specific roles for Fgf8 in outflow tract formation and remodeling. Development 133 2419 33. Park, S. Y., G onen, M., Kim, H. J., Michor, F., and Polyak, K. (2010). Cellular and genetic diversity in the progression of in situ human breast carcinomas to an invasive phenotype. The Journal of clinical investigation. Available at: internal pdf://Park SY, J Clin Inve st 2010, genetics heterogenity in primary breast cancer and CSC 1604248320/Park SY, J Clin Invest 2010, genetics heterogenity in primary breast cancer and CSC.pdf. Paruthiyil, S., Parmar, H., Kerekatte, V., Cunha, G. R., Firestone, G. L., and Leitman, D. C. (2004). Estrogen Receptor {beta} Inhibits Human Breast Cancer Cell Proliferation and Tumor Formation by Causing a G2 Cell Cycle Arrest. Cancer Res 64 423 428. Patient, R. K., and McGhee, J. D. (2002). The GATA family (vertebrates and invertebrates). C urr Opin Genet Dev 12 416 22. Pece, S., Tosoni, D., Confalonieri, S., Mazzarol, G., Vecchi, M., Ronzoni, S., Bernard, L., Viale, G., Pelicci, P. G., and Di Fiore, P. P. (2010). Biological and Molecular Heterogeneity of Breast Cancers Correlates with Thei r Cancer Stem Cell Content. Cell 140 62 73.

PAGE 141

141 Pedram, A., Razandi, M., and Levin, E. R. (2006). Nature of functional estrogen receptors at the plasma membrane. Molecular endocrinology (Baltimore, Md 20 1996 2009. Pedram, A., Razandi, M., Sainson, R. C., K im, J. K., Hughes, C. C., and Levin, E. R. (2007). A conserved mechanism for steroid receptor translocation to the plasma membrane. The Journal of biological chemistry 282 22278 88. Pei, X. H., Bai, F., Smith, M. D., Usary, J., Fan, C., Pai, S. Y., Ho, I C., Perou, C. M., and Xiong, Y. (2009). CDK inhibitor p18(INK4c) is a downstream target of GATA3 and restrains mammary luminal progenitor cell proliferation and tumorigenesis. Cancer cell 15 389 401. Phillips, T. M., McBride, W. H., and Pajonk, F. (200 6). The response of CD24( /low)/CD44+ breast cancer initiating cells to radiation. Journal of the National Cancer Institute 98 1777 85. Plageman, T. F., and Yutzey, K. E. (2005). T box genes and heart development: putting the "T" in heart. Dev Dyn 232 1 1 20. Ponti, D., Costa, A., Zaffaroni, N., Pratesi, G., Petrangolini, G., Coradini, D., Pilotti, S., Pierotti, M. A., and Daidone, M. G. (2005). Isolation and in vitro propagation of tumorigenic breast cancer cells with stem/progenitor cell properties. Ca ncer research 65 5506 11. Prall, O. W., Menon, M. K., Solloway, M. J., Watanabe, Y., Zaffran, S., Bajolle, F., Biben, C., McBride, J. J., Robertson, B. R., Chaulet, H., et al. (2007). An Nkx2 5/Bmp2/Smad1 negative feedback loop controls heart progenitor specification and proliferation. Cell 128 947 59. Pu, W. T., Ishiwata, T., Juraszek, A. L., Ma, Q., and Izumo, S. (2004). GATA4 is a dosage sensitive regulator of cardiac morphogenesis. Dev Biol 275 235 44. Raisner, R. M., and Madhani, H. D. (2006). Pa tterning chromatin: form and function for H2A.Z variant nucleosomes. Curr Opin Genet Dev 16 119 24. Raj, A., and van Oudenaarden, A. (2008). Nature, nurture, or chance: stochastic gene expression and its consequences. Cell 135 216 26. Raouf, A., Zhao, Y., To, K., Stingl, J., Delaney, A., Barbara, M., Iscove, N., Jones, S., McKinney, S., Emerman, J., et al. (2008). Transcriptome analysis of the normal human mammary cell commitment and differentiation process. Cell stem cell 3 109 18.

PAGE 142

142 Rappa, G., and Lorico, A. Phenotypic characterization of mammosphere forming cells from the human MA 11 breast carcinoma cell line. Experimental Cell Research In Press, Corrected Proof Available at: http://www.sciencedirect.com/science/article/B6WFC 4Y4PBH6 3/2/00fa5c9a 4f47afcdaaf8ea5be9276647 [Accessed April 30, 2010]. Regan, J., and Smalley, M. (2007). Prospective isolation and functional analysis of stem and differentiated cells from the mouse mammary gland. Stem cell reviews 3 124 36. Reynolds, B. A., and Rietze, R. L. (2005). Neural stem cells and neurospheres -re evaluating the relationship. Nat. Methods 2 333 336. Risebro, C. A., Smart, N., Dupays, L., Breckenridge, R., Mohun, T. J., and Riley, P. R. (2006). Hand1 regulates cardiomyocyte proliferation versus d ifferentiation in the developing heart. Development 133 4595 606. Rizzo, P., Miao, H., D'Souza, G., Osipo, C., Yun, J., Zhao, H., Mascarenhas, J., Wyatt, D., Antico, G., Hao, L., et al. (2008). Cross talk between Notch and the Estrogen Receptor in Breast Cancer Suggests Novel Therapeutic Approaches. Cancer Res 68 5226 5235. Rosen, J. M., and Jordan, C. T. (2009). The increasing complexity of the cancer stem cell paradigm. Science (New York, N.Y 324 1670 3. Schmits, R., Filmus, J., Gerwin, N., Senaldi, G., Kiefer, F., Kundig, T., Wakeham, A., Shahinian, A., Catzavelos, C., Rak, J., et al. (1997). CD44 regulates hematopoietic progenitor distribution, granuloma formation, and tumorigenicity. Blood 90 2217 2233. Shackleton, M., Quintana, E., Fearon, E. R ., and Morrison, S. J. (2009). Heterogeneity in cancer: cancer stem cells versus clonal evolution. Cell 138 822 9. Shang, Y., Hu, X., DiRenzo, J., Lazar, M. A., and Brown, M. (2000). Cofactor dynamics and sufficiency in estrogen receptor regulated transc ription. Cell 103 843 52. Shaw, L., Sadler, A., Pugazhendhi, D., and Darbre, P. (2006). Changes in oestrogen receptor [alpha] and [beta] during progression to acquired resistance to tamoxifen and fulvestrant (Faslodex, ICI 182,780) in MCF7 human breast cancer cells. The Journal of Steroid Biochemistry and Molecular Biology 99 19 32. Sheridan, C., Kishimoto, H., Fuchs, R. K., Mehrotra, S., Bhat Nakshatri, P., Turner, C. H., Goulet, R., Badve, S., and Nakshatri, H. (2006). CD44+/CD24 breast cancer cells exhibit enhanced invasive properties: an early step necessary for metastasis. Breast Cancer Res 8 R59.

PAGE 143

143 Shi, X., Gipp, J., and Bushman, W. (2007). Anchorage independent culture maintains prostate stem cells. Developmental biology 312 396 406. Shipitsi n, M., Campbell, L. L., Argani, P., Weremowicz, S., Bloushtain Qimron, N., Yao, J., Nikolskaya, T., Serebryiskaya, T., Beroukhim, R., Hu, M., et al. (2007). Molecular definition of breast tumor heterogeneity. Cancer cell 11 259 73. Shipitsin, M., and Pol yak, K. (2008). The cancer stem cell hypothesis: in search of definitions, markers, and relevance. Laboratory investigation; a journal of technical methods and pathology 88 459 63. Sleeman, K. E., Kendrick, H., Robertson, D., Isacke, C. M., Ashworth, A., and Smalley, M. J. (2007). Dissociation of estrogen receptor expression and in vivo stem cell activity in the mammary gland. The Journal of cell biology 176 19 26. Smalley, M., and Ashworth, A. (2003). Stem cells and breast cancer: A field in transit. N at Rev Cancer 3 832 844. Smith, G. H. (2005). Label retaining epithelial cells in mouse mammary gland divide asymmetrically and retain their template DNA strands. Development (Cambridge, England) 132 681 7. Sotiriou, C., Neo, S., McShane, L. M., Korn, E. L., Long, P. M., Jazaeri, A., Martiat, P., Fox, S. B., Harris, A. L., and Liu, E. T. (2003). Breast cancer classification and prognosis based on gene expression profiles from a population based study. Proceedings of the National Academy of Sciences of t he United States of America 100 10393 10398. Sotiriou, C., Wirapati, P., Loi, S., Harris, A., Fox, S., Smeds, J., Nordgren, H., Farmer, P., Praz, V., Haibe Kains, B., et al. (2006). Gene Expression Profiling in Breast Cancer: Understanding the Molecular Basis of Histologic Grade To Improve Prognosis. J. Natl. Cancer Inst. 98 262 272. Srivastava, D. (2006). Making or breaking the heart: from lineage determination to morphogenesis. Cell 126 1037 48. Srivastava, D., and Olson, E. N. (2000). A genetic blu eprint for cardiac development. Nature 407 221 6. Stennard, F. A., and Harvey, R. P. (2005). T box transcription factors and their roles in regulatory hierarchies in the developing heart. Development 132 4897 910. Sternlicht, M. D. (2006). Key stages i n mammary gland development: the cues that regulate ductal branching morphogenesis. Breast Cancer Res 8 201.

PAGE 144

144 Stingl, J., Eirew, P., Ricketson, I., Shackleton, M., Vaillant, F., Choi, D., Li, H. I., and Eaves, C. J. (2006). Purification and unique prope rties of mammary epithelial stem cells. Nature 439 993 7. Stoica, G. E., Franke, T. F., Moroni, M., Mueller, S., Morgan, E., Iann, M. C., Winder, A. D., Reiter, R., Wellstein, A., Martin, M. B., et al. (2003). Effect of estradiol on estrogen receptor alp ha gene expression and activity can be modulated by the ErbB2/PI 3 K/Akt pathway. Oncogene 22 7998 8011. Tam, P. P., and Loebel, D. A. (2007). Gene function in mouse embryogenesis: get set for gastrulation. Nat Rev Genet 8 368 81. Tanaka, M., Chen, Z., Bartunkova, S., Yamasaki, N., and Izumo, S. (1999). The cardiac homeobox gene Csx/Nkx2.5 lies genetically upstream of multiple genes essential for heart development. Development 126 1269 80. Thottassery, J. V., Sun, Y., Westbrook, L., Rentz, S. S., Manu vakhova, M., Qu, Z., Samuel, S., Upshaw, R., Cunningham, A., and Kern, F. G. (2004). Prolonged extracellular signal regulated kinase 1/2 activation during fibroblast growth factor 1 or heregulin beta1 induced antiestrogen resistant growth of breast cancer cells is resistant to mitogen activated protein/extracellular regulated kinase kinase inhibitors. Cancer Res 64 4637 4647. Tremblay, G. B., Bergeron, D., and Giguere, V. (2001). 4 Hydroxytamoxifen is an isoform specific inhibitor of orphan estrogen rece ptor related (ERR) nuclear receptors beta and gamma. Endocrinology 142 4572 4575. Vargo Gogola, T., and Rosen, J. M. (2007). Modelling breast cancer: one size does not fit all. Nature reviews 7 659 72. Vera Ramirez, L., Sanchez Rovira, P., Ramirez Tort osa, C. L., Quiles, J. L., Ramirez Tortosa, M. C., Alvarez, J. C., Fernandez Navarro, M., and Lorente, J. A. (2010). Gene expression profiles, tumor microenvironment, and cancer stem cells in breast cancer: Latest advances towards an integrated approach. C ancer Treat Rev. Available at: http://www.ncbi.nlm.nih.gov/pubmed/20299155 [Accessed March 29, 2010]. Vermeulen, L., Todaro, M., de Sousa Mello, F., Sprick, M. R., Kemper, K., Perez Alea, M., Richel, D. J., Stassi, G., and Medema, J. P. (2008). Single cel l cloning of colon cancer stem cells reveals a multi lineage differentiation capacity. Proceedings of the National Academy of Sciences of the United States of America 105 13427 32. Villadsen, R., Fridriksdottir, A. J., Ronnov Jessen, L., Gudjonsson, T., Rank, F., LaBarge, M. A., Bissell, M. J., and Petersen, O. W. (2007). Evidence for a stem cell hierarchy in the adult human breast. The Journal of cell biology 177 87 101.

PAGE 145

145 Villadsen, R. (2005). In search of a stem cell hierarchy in the human breast and i ts relevance to breast cancer evolution. APMIS 113 903 921. Visvader, J. E. (2009). Keeping abreast of the mammary epithelial hierarchy and breast tumorigenesis. Genes & development 23 2563 77. Visvader, J. E., and Lindeman, G. J. (2008). Cancer stem c ells in solid tumours: accumulating evidence and unresolved questions. Nature reviews 8 755 68. Walia, V., and Elble, R. C. (2010). Enrichment for breast cancer cells with stem/progenitor properties by differential adhesion. Stem Cells Dev. Available at: http://www.ncbi.nlm.nih.gov/pubmed/20222827 [Accessed March 22, 2010]. Wang, X., Kruithof de Julio, M., Economides, K. D., Walker, D., Yu, H., Halili, M. V., Hu, Y. P., Price, S. M., Abate Shen, C., and Shen, M. M. (2009). A luminal epithelial stem cell that is a cell of origin for prostate cancer. Nature 461 495 500. Welm, B. E., Tepera, S. B., Venezia, T., Graubert, T. A., Rosen, J. M., and Goodell, M. A. (2002). Sca 1(pos) cells in the mouse mammary gland represent an enriched progenitor cell populat ion. Dev. Biol 245 42 56. Wicha, M. S. (2008). Cancer stem cell heterogeneity in hereditary breast cancer. Breast Cancer Res 10 105. Wijayaratne, A. L., and McDonnell, D. P. (2001). The human estrogen receptor alpha is a ubiquitinated protein whose sta bility is affected differentially by agonists, antagonists, and selective estrogen receptor modulators. J. Biol. Chem 276 35684 35692. Williams, C. C., Basu, A., El Gharbawy, A., Carrier, L. M., Smith, C. L., and Rowan, B. G. (2009). Identification of fo ur novel phosphorylation sites in estrogen receptor alpha: impact on receptor dependent gene expression and phosphorylation by protein kinase CK2. BMC Biochem 10 36. Wright, M. H., Calcagno, A. M., Salcido, C. D., Carlson, M. D., Ambudkar, S. V., and Var ticovski, L. (2008). Brca1 breast tumors contain distinct CD44+/CD24 and CD133+ cells with cancer stem cell characteristics. Breast Cancer Res 10 R10. Wright, M. H., Robles, A. I., Herschkowitz, J. I., Hollingshead, M. G., Anver, M. R., Perou, C. M., an d Varticovski, L. (2008). Molecular analysis reveals heterogeneity of mouse mammary tumors conditionally mutant for Brca1. Molecular cancer 7 29. Xin, M., Davis, C. A., Molkentin, J. D., Lien, C. L., Duncan, S. A., Richardson, J. A., and Olson, E. N. (20 06). A threshold of GATA4 and GATA6 expression is required for cardiovascular development. Proc Natl Acad Sci U S A 103 11189 94.

PAGE 146

146 Xu, H., and Baldini, A. (2007). Genetic pathways to mammalian heart development: Recent progress from manipulation of the mo use genome. Semin Cell Dev Biol 18 77 83. Zheng, A., Kallio, A., and Harkonen, P. (2007). Tamoxifen induced rapid death of MCF 7 breast cancer cells is mediated via extracellularly signal regulated kinase signaling and can be abrogated by estrogen. Endoc rinology 148 2764 77. Zhou, J., Wulfkuhle, J., Zhang, H., Gu, P., Yang, Y., Deng, J., Margolick, J. B., Liotta, L. A., Petricoin, E., and Zhang, Y. (2007). Activation of the PTEN/mTOR/STAT3 pathway in breast cancer stem like cells is required for viabili ty and maintenance. Proceedings of the National Academy of Sciences of the United States of America 104 16158 63. Zhou, J., Zhang, H., Gu, P., Bai, J., Margolick, J. B., and Zhang, Y. (2008). NF kappaB pathway inhibitors preferentially inhibit breast can cer stem like cells. Breast cancer research and treatment 111 419 27. Zhou, J. H., Yu, D. V., Cheng, J., and Shapiro, D. J. (2007). Delayed and persistent ERK1/2 activation is required for 4 hydroxytamoxifen induced cell death. Steroids 72 765 77. Zhu, B. T., Han, G. Z., Shim, J. Y., Wen, Y., an d Jiang, X. R. (2006). Quantitative structure activity relationship of various endogenous estrogen metabolites for human estrogen receptor alpha and beta subtypes: Insights into the structural determinants favoring a differential subtype binding. Endocrino logy 147 4132 50. Zucchi, I., Astigiano, S., Bertalot, G., Sanzone, S., Cocola, C., Pelucchi, P., Bertoli, G., Stehling, M., Barbieri, O., Albertini, A., et al. (2008). Distinct populations of tumor initiating cells derived from a tumor generated by rat mammary cancer stem cells. Proceedings of the National Academy of Sciences 105 16940 16945. Zucchi, I., Sanzone, S., Astigiano, S., Pelucchi, P., Scotti, M., Valsecchi, V., Barbieri, O., Bertoli, G., Albertini, A., Reinbold, R. A., et al. (2007). The pro perties of a mammary gland cancer stem cell. Proceedings of the National Academy of Sciences of the United States of America 104 10476 81.

PAGE 147

147 BIOGRAPHICAL SKETCH Ada A o was born in 1981 in Macau to Chi Seng Ao and Lai Meng Chan. They immigrated to Bosto n, Massachusetts in 1990. She completed secondary education at Boston Latin School in Boston, Massachusetts in 1999. Ada then attended Brandeis University from 1 999 2003, where she majored in biology and b iochemistry. After graduating with a Bachelor of Science in b iology, she was employed for 2 years as a research technician at Boston University Medical Center under the guidance of Dr. Jude Deeney, studying the effects of fatty acids on insulin secretion in rat islets. In 2005, Ada entered the Interdis ciplinary Program in Biomedical Sciences (IDP) a t the University of Florida to pursue a graduate degree in the Department of Biochemistry and Molecular Biology. She joined the laboratory of Dr. Jianrong Lu in May 2006, and qualified for PhD candidacy Sept ember 2007. Her tenure has included a diverse course of studies, such as the characterization of cardiac defects in mouse embryos, the study of Estrogen Related Receptors on hypoxic gene regulation, and the characterization of antiestrogen effects on MCF7 derived mammospheres.