University of Southern California Dissertations and Theses

Modulation of Runx proteins by steroid hormone receptors

(USC Thesis Other)

Modulation of Runx proteins by steroid hormone receptors

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content MODULATION OF RUNX PROTEINS BY STEROID HORMONE RECEPTORS
by
Omar Khalid
_____________________________________________________________________
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(PHARMACEUTICAL SCIENCES)
May 2009
Copyright 2009 Omar Khalid
ii
Dedication
I would like to dedicate this thesis to my mother, Nasreen
Khan. She has been there for me since the very beginning.
Without her love I would be nothing.
I would also like to dedictate this thesis to Donna Sir. She
has been a source of support, love and caring for me
always.
iii
Acknowledgements
First and foremost I would like to acknowledge my PI, Dr. Gerhard A. Coetzee.
He has been there for me from the very beginning. He gave me an opportunity that I will
never forget. He has been a great friend and his growing trust in me has helped me to
develop into an independent scientist. His “tough love” and humility has always been a
source of inspiration for me. I respect his work ethic, his creativity, and his mentorship.
Thank you for always having time for me, and thank you for giving me a chance.
Furthermore I would like to acknowledge my collaborating PI, Dr. Baruch
Frenkel. He has also always been there for me and given me the opportunity to work in
his lab as well. Like Gerry he has been a great friend and mentor. His trust has helped me
develop into an independent scientist. Meetings with him were always a joy. I respect
him as a scientist and cherish the time we have had together. I would also like to
acknowledge Dr. Stallcup for all his support and kindness from the very early days in
PIBBS to final days of my graduate carreer. My committee members Dr. Curtis Okamoto
and Dr. Sarah Hamm-Alvarez have always been wonderful in their unwaivering support
and optimism for my PhD. I want to thank all of you for being so wonderful.
I would like to especially acknowledge Dr. Sanjeev Baniwal. Ever since he came
my project took off to the sky. I remember telling him if you come work with our group
“we can go places” and we have. I appreciate all our conversations about work related
and non-work related things. He has been a great friend and mentor to me as well. We
have grown closer with time and developed a great trust that has been helpful in all
aspects of my life. Thank you for choosing USC, and thank you for your friendship.
iv
I would like to thank my fellow labmates for always being there for me. This
includes Howie Shen, Jennifer Prescott, Unnati Jariwala, Steven Pregizer, Jon Cogan,
Artem Barski, Chunli Yan, Lily Talevera, Alyson Walters, and Marcus Wantroba.
However, from my labmates two have become really close friends. First is Dr. Li Jia, he
is the man! Li has always had time to listen to my questions, give me advice, or just hang
out. He is a great guy and he has made my experience here wonderful. I hope I haven’t
annoyed him too much with my silliness Second is Dr. Grant Buchanan. Grant is also the
man! He has been there for me also to listen to my questions, give me advice, or just hang
out as well. I think I did annoy him a bit but he had enough patience to give me a chance.
Another person that I have known from my very early days at CSUF to here at USC is
Daniel J. Purcell. He has always been a wonderful friend and colleague. I always enjoy
my time with him and the Stallcup lab.
v
Table of Contents
Dedication ii
Acknowledgments iii
List of Tables viii
List of Figures xii
Abstract xiii
Chapter 1: Introduction 1
1. Transcriptional Regualtion 1
2. Transcription Factors: Structure and Function 8
2.1 Runx Proteins 10
2.2 Steroid Hormone Receptors 13
3. Transcription Factors: Genes Enconding Runx, ER, and AR 18
4. Transcription Factors: Development and Disease 19
4.1 Runx Proteins in Development and Disease 19
4.2 ER and AR: Development and Disease 22
5. Cofactors 29
5.1 Runx Cofactors 29
5.2 ER and AR Cofactors 30
6. Outline and Scope of this Thesis 32
Chapter 2: Materials and Methods 34
1. Cell Culture 34
2. Transient Transfections: Luciferase Reporter Assays 39
3. Electromobility Shift Assays (EMSA) 44
4. Immunoflourescence 55
5. Co-immunoprecipiation 58
6. Real time RT-PCR 60
7. Chromatin Immunoprecipitation (ChIP) 63
8. Flourescent Recovery After Phtobleaching (FRAP) 68
9. Microarray Data Mining 70
Chapter 3: Modulation of Runx2 by ER! 72
1. Introduction 72
2. Results 76
vi
2.1 ER! and ER", inhibits Runx2 activity in a ligand
specific manner 76
2.2 Mapping of ERa domains in Runx2 Repression 80
2.3 ER! physically interacts with Runx2 83
2.4 Inhibtion of Runx2 by Estradiol in
MC3T3-E1 cells 89
2.5 Synthetic ER ligands inhibit or stimulate Runx2 92
2.6 Negative Correlation between ER! and
Runx2 in BCa 97
2.7 Understanding the mechanism of
ER!/Runx2 interaction 104
3. Discussion 107
Chapter 4: Modulation of Runx2 by AR 111
1. Introduction 111
2. Results 115
2.1 Luciferase assays and EMSA for Runx
proteins in PC3 115
2.2 Runx1 and Runx2 are inhibited by the AR 118
2.3 Runx1 and Runx2 Cellular localization in
PC3 with AR 120
2.4 Runx1 and Runx2 Cellular localization in
Cos7 with AR 120
2.5 Mechanism of Runx2 Inhibition by the AR 124
2.6 Direct Protein-Protein Interaction of
Runx2 and AR 132
2.7 Runx2 does not interact with the A573D
Point Mutant 137
2.8 Inhibition of AR activity by Runx Proteins 137
2.9 Nuclear Runx2 mobility changes in the
Presence of AR 143
2.10 Inhibition of Endogenous Runx Proteins
by the AR 147
2.11 Effects of SARMS on Runx2 Activity
by the AR 152
2.12 Micorray Data Analysis of Runx2 Genes in PCa 157
3. Discussion 159
Chapter 5: Discussion ` 163
1. Summary of Research Findings 163
2. Overview of Findings in the ER/Runx2 Project 163
3. Overview of the Findings from the AR/Runx1,2 Project 168
4. Future Directions 169
4.1 Clinical Setting 169
vii
4.2 Molecular Setting 174
5. Concluding Remarks 175
References 177
viii
List of Tables
Table 2.1 Summary of Endogenous Expression of Runx1, Runx2, AR, and ER 38
Table 2.2 Summary of Primers Used for RT-qPCR 61
Table 3.1 Runx2 and ER! Target Genes 99
ix
List of Figures
1.1. Transcriptional Complex 4
1.2 Runx Proteins, AR& ER 9
1.3 Runt Domain 11
1.4 Ligands for ER & AR 16
1.5 DNA-binding Domain of AR & ER 17
1.6 Anatomy of the Prostate 26
1.7 Bone Metastasis of PCa and BCa 28
1.8 Coactivator/Corepressors of Runx2 and Nuclear Receptors 31
2.1 Luciferase Reporter Assay 41
2.2 Developing the Luciferase Reporter Assay 42
2.3 Overview of electromobility shift Assays (EMSA) 47
2.4 Assessment of Runx2 Activity in C4-2B Cells 49
2.5 Assessment of Runx2 Activity in LNCaP Cells 50
2.6 “Helper Activity” from other cell types. 51
2.7 HeLa NE extract fractions recovers Runx activity in C4-2B cells 52
2.8 AR DNA-binding activity in C4-2B and LNCaP Cells 53
2.9 Nuclear Changes in Epigenetic Marks Upon DHT treatment in C4-2B 57
2.10 An overview of Co-Immunoprecipitation 59
2.11 An example of RTq-PCR 62
2.12 An overview of ChIP for Runx2 65
2.13 Testing Runx2 antibodies for ChIP 66
x
2.14 An example of a FRAP experiment 69
2.15 An example of a 2D Unsupervised Cluster 71
3.1 ER! inhibits human Runx2 77
3.2 ER! inhibits mouse Runx2 78
3.3 Runx2 does not influence ER!’s transcriptional activation activity 79
3.4 Functional mapping of ER! domains in Runx2 repression 82
3.5 Interaction between ER! and Runx2 in Co-IP and GST pull-down assays 85
3.6 Direct interaction of ER!-LBD with Runx2 in Superphysiological
Conditions 86
3.7 Immunoflourescence of ER! and Runx2 87
3.8 Development stage-specific inhibition of Runx2 by E2 in osteoblasts 90
3.9 Various effects of SERM-bound ER! on Runx2 94
3.10 Modulation of Runx2 target genes in MCF7 cells by SERMS 96
3.11 Meta-analysis of the correlation between ER! and Runx2 in
BCa biopsies 101
3.12 Correlation between ER! and Runx2 target genes in BCa biopsises 102
3.13 Correlation between ER! and Runx2 target genes in BCa biopsises 103
3.14 Immunoprecipitation of ER! by Runx2 with Mnase and EtBR 105
4.1 Luciferase Reporter Assay and EMSA for Runx Proteins in PC3 Cells 116
4.2 Runx1 and Runx2 are inhibited by the AR 118
4.3 Immunoflourescence of PC3 cells with Runx1 or Runx2, and AR 121
4.4 Immunoflourescence of Cos7 cells with Runx1 or Runx2, and AR 122
4.5 Models of how AR may be inhibiting Runx2 126
xi
4.6 Activation/Repression Curves of AR and Runx1 or Runx2 127
4.7 Inhibition of Runx in PC3 cells by different AR domains 129
4.8 Inhibition of Runx in PC3 cells by the AR DBD 130
4.9 Delta AF1 inhibition of Runx2 in Cos7 Cells 131
4.10 Overview of GST pull-down assays 133
4.11 Runx2 directly interacts with the AR DBD 134
4.12 Full AR DBD is necessary for the interaction with Runx2 135
4.13 Runt-PST of Runx2 interacts with the AR DBD 136
4.14 A573D mutant AR doesnot interact with or affect Runx2 DNA binding 139
4.15 AR activity is inhibited by Runx1 and Runx2 141
4.16 Transactivationally dead Runx2 can inhibit AR activity 143
4.17 FRAP experiments of GFP-Runx2 in the Presence of AR and vice versa 145
4.18 Model depicting the mechanism of mutual inhibition of AR and Runx2 146
4.19 Runx2 activity is inhibited by AR in MC3T3-E1 Cells 148
4.20 Runx2 activity is inhibited by AR in SAOS-2 Cells 149
4.21 Runx2 activity is not inhibited by AR in ROS Cells 150
4.22 Decrease in Runx2 Occupancy on the OSE2 site by DHT treatment 151
4.23 Inhibition of Runx activity in PC3 cells by T877A AR mutant and OHF 154
4.24 Inhibition of Runx1 and Runx2 activity in Cos7 cells by T877A
AR mutant and OHF 155
4.25 FRAP experiments with GFP-Runx2 and T877A AR mutant with OHF 156
4.26 Correlation of KLK3 with Runx2 Target Genes in PCa Tumor Samples 158
5.1 Cartoon depicting the development of ER negative breast cancer 166
xii
5.2 Immunopreicipitation of Runx2 with Coactivators ` 167
5.3 Cartoon of the strategy possible in developing drugs Based
on AR/Runx2 intx 172
5.4 LNCaP cells overexpressing Runx2 Undergoing Cell Death 173
xiii
Abstract
Estrogen receptor ! (ER !) and androgen receptor (AR) are master transcription factors
in the breast and prostate, respectively. They are commonly known in development of
sexual characteristics. However, both ER! and AR have been known to be involved in
breast cancer (BCa) and prostate cancer (PCa) progression, respectively. The Runx
family of transcription factors plays a role in hematopoiesis (Runx1), skeletogenesis
(Runx2) and neurogenesis (Runx3). In addition, Runx proteins inhibit cell cycle
progression, and have been assigned tumor suppressor roles in various contexts. Because
both BCa and PCa cells metastasize to bone at high frequency, investigators have
interrogated the possibility that they share characteristics with osteoblasts. Indeed, BCa
and PCa cells were found to have “osteomimetic” properties, including expression of
Runx2 and Runx2-target genes otherwise expressed by osteoblasts. Provoked by the
reported physical interaction between AR and Runx2, we initiated a study to test whether
ER! and AR might promote BCa and PCa progression (respectively) by inhibiting
Runx2 activity through direct protein-protein interactions. We report here that ER! and
AR both inhibit Runx2 transcriptional activity via interaction through specific domains in
a receptor specific and ligand specific manner. In addition, it was revealed to a lesser
extent that AR was able to strongly inhibit Runx1 transcriptional activity, but ER! was
not. Immunohistochemistry analyses revealed that both ER! and Runx2 colocalize in the
nucleus in specific subnuclear domains as well as AR and Runx2. An inverse relationship
was observed of Runx2 target genes and ER! target genes in BCa patient tumor samples.
Similarly an inverse relationship was also observed of Runx2 target genes and AR target
genes in PCa patient tumor samples, suggesting that the molecular biology that we
observe in vitro may be occurring within patients. Molecular characterization of
ER!/Runx2 and AR/Runx2 interaction and the functional consequences of this
interaction may form the basis for novel therapeutic approaches to treat both BCa and
PCa patients.
1
Chapter 1: Introduction
1. Transcriptional Regulation
Transcription is a basic molecular mechanism responsible for specialization of
cells. Cells have specific functions that enable organisms to carry out necessary
biological processes that are important for survival. These processes include the immune
response, reformation of bone, and several others that are essential to almost every
organism. An overview of transcription is given below.
There are both cis-elements and trans-acting factors involved in transcription (80).
Trans-acting factors are proteins in the cell nucleus that bind to cis-elements, referred to
as transcription factor response elements. These proteins regulate the transcription of
genes and modulate cell function. Cis elements may be considered as promoters or
enhancers. Promoters occur near (within about 100 bp) the transcriptional start site (80).
The proximal promoters of genes contain either a TATA box or down-stream
promoter element (DPE) each within 25-30 bp of the transcriptional start site (80). The
TATA binding protein (TBP) binds to the TATA box and begins the process of
transcription by unwinding the DNA, whereas DPE where TFIID binds and is associated
usually with TATA-less promoters. TBP with other factors form TFIID, which is a part
of the RNA Polymarase II (RNA Pol II) pre-initiation complex (PIC) (80). PIC is made
up of six general transcription factors (GTF) TFIIA, TFIIB, TFIID, TFIIE, TFIIF, and
TFIIH. PIC helps position RNA Pol II at the transcriptional start site, unwinds the DNA,
and positions the DNA in the RNA Pol II active site for transcription. Nevertheless, all
genes do not have a TATA box these regions are known as TATA-less promoters.
2
Instead of using a TATA-box they have an initiator element. TFIID is able to contact the
initiator element and thereby form both PIC and recruit RNA Pol II without a TATA-box
[As reviewed in (144)] (Figure 1.1).
Enhancers, silencers, and insulators are types of cis-acting regulatory elements
involved in regulation of transcription. Enchancers increase the basal level of
transcription by being recognized by tissue-specific transcription factors (trans-acting
factors) that promote assembly of the RNA Pol II complex. Enhancers are orientation
independent, meaning that enhancers can be bound by transcription factors with the
sequence in 5' to 3' or 3' to 5' direction (7). Enhancers are usually composed of many of
the transcription factor binding sites in a cluster (7). Enhancers can be very far away
from the transcription start site from hundreds of kb pairs upstream of a promoter, in an
intron of the gene, and even at after the 3' end of the gene (7). Distance is irrelevant to
enhancers possibly due to DNA-looping, where a far away enhancer is brought close to a
promoter by forming a loop in the DNA (7). Silencers perform the opposite function of
enhancers and are involved in reducing transcriptional levels. Silencers, like enhancers,
can be orientation independent and far away from the promoter element (96). These types
of silencers are termed "classical silencers." Nevertheless, some silencers have been
discovered that are position-dependent and have been termed "negative regulatory"
elements (96). Silencers serve as binding sites for repressors and can recruit negative
cofactors. In some cases the same transcription factors involved in enhancer function can
be recruited in another context for repressor functions as has been shown with Runx
proteins for example (14). Insulators, on the other hand, are involved in regulating both
3
enhancer and silencer activity. Insulators are elements that block the communication
between promoters and enhancers (enhancer activity), and prevent the spread of
repressive chromatin (silencer activity) (80).
Recently, the concept of transcriptional hubs have emerged. It has been shown
that two distant genes on different chromosomes can come close together to form a hub
for transcription, and has been termed chromosome “kissing.” Nunez and colleagues
fluorescently labeled two genes known to be activated by hormone treatment, and upon
hormone treatment saw that these two genes came in close proximity to each other in the
nucleus (64). Suggesting, that .hubs for transcription are formed where cofactors,
transcription factors, and DNA come into complex.
4
Figure 1.1. Transcriptional Complex
Figure1.1 Legend. Summary of the transcriptional complex necessary to initiate
transcription. “Activator” proteins (transcription factors) bind to enhancer elements and
interact with a coactivator complex that associates with GTFs and RNA Pol II to initiate
transcription. In this study the “activators” of interest are Runx proteins, androgen
receptor and estrogen receptor !.
5
CpG islands are regions of DNA that are encriched with CG dinucleotides and are
in or near 40-70% of promoters within mammals (123). A formal definition of CpG
islands is a region of at least 200 bp, with greater than 50% CG content and a 60% or
higher observed/expected ratio of CpG’s (140). CpG islands typically occur at
housekeeping genes (33), but may be altered in disease (99). An increase in the
methylation status of CpG islands on gene promoters have been known to decrease
expression levels of that gene. For example promoter hypermethylation of the Runx3
gene correlates with a subsequent decrease in expression of Runx3 (99).
Clearly there are many critical steps involved in transcription. All of these steps
work in concert for cell growth and survival. Therefore, regulation of transcription is
fundamental in every eukaryotic organism. Regulation mainly occurs by tight control of a
class of proteins known as transcription factors. Trans-acting transcription factors are
DNA-binding proteins that bind to specific DNA elements and can recruit GTF’s (either
directly or through adaptor proteins) to allow for transcription (80). Thus, controlling
which genetic information gets converted from DNA to RNA. Transcription factors are
involved in control of development, response to intercellular signals, response to
environmental stressors, and cell cycle control (80).
Another level of control for transcription is how accessible the DNA is, which has
a lot to do with DNA packaging. DNA within eukaryotes is packaged with histone
proteins to form chromatin (86). Chromatin is packaged from DNA in a very precise
manner. DNA (147 bp in length) is wrapped around a histone octamer to form
nucleosomes, which are essentially ‘beads on a string’ (36).The nucleosomes are
6
subsequently condensed with the addition of histone H1 to form a 30 nm fiber. These
fibers are subsequently found in different condensation states. Highly condensed
chromatin is indicative of heterochromatin, an inaccesible region of DNA where
transcription is most likely not occurring (36). However, this heterochromatin may be
able to spread open and become accessible for transcription known as euchromatin (93).
The DNA wrapping the nucleosomes is looser within euchromatin compared to
heterochromatin, thereby allowing accessibility RNA Pol II and the various complexes
pertinent in transcription. However, all euchromatin is not transcribed and therefore may
be converted to heterochromatin to protect these regions of DNA (93).
Histones are the major proteins within chromatin structure. There are five major
histones: H1 (linker histone), H2A, H2B, H3, and H4 (93). Nucleosomes are comprised
of two H2A-H2B dimers and an H3-H4 tetramer (93). The core histones have long N-
terminal extensions, not packaged within the nucleosome, which are the sites of many
post-translational modifications. These modifications include acetylation, methylation,
phosphorylation, ubiquitination, sumoylation, and ADP-ribosylation (114). Proteins such
as CARM1, SET proteins, EZH2, PRMT proteins, and SUV4-20H1or H2 are only a
small number in a long list of proteins involved in methylation of various lysine and
arginine residues that can be mono, di, or trimethylated (137). The methylation marks are
indicative of both active and inactive transcriptional states. A class of proteins known as
Jumanji proteins can reverse methylation marks (43). Another series of histone
modification enzymes include histone acetyltransferases (HATs) that add acetyl groups to
tails, and histone deacetylases (HDACs) that remove residues from the tail. It has been
7
observed that acetylation is an activating modification leading to an openness in
chromatin structure, and deacetylation the opposite (89).
Another mechanism of altering chromatin structure is by ATP-dependent
remodeling by a class of proteins known as SWI/SNF (Switch/sucrose non fermentable).
SWI/SNF family of protein were orginially discovered in yeast but have human
homologs that function in similar ways. SWI/SNF type proteins have ATPase activity
stimulated by the presence of DNA (138). This activity remodels the chromatin structure
by disrupting histone-DNA interactions allowing the DNA to unwrap or ‘loosen’ from
the histone octamer (138). SWI/SNF proteins may also alter nucleosome positioning.
(138)
The loss of histones or the use of histone variants is another method of altering
DNA accessibility and altering transcriptional activity. Histone H2A can be replaced by
histone H2AZ which destabilizes the nucleus or H3 can be replaced by H3.3 which is
known to associated with activation of transcription (111). Recently, it has be reported
that histone loss can occur in regulatory regions and coding regions of actively
transcribed genes (150).
Overall there are several layers of transcriptional regulation are in place to ensure
that appropriate gene expression occurs. Many steps have evolved to regulate
transcription and coordinate the appropriate gene expression in a spatial-temporal specific
manner. As well as, presence of tissue-specific transcription factors to also ensure
appropriate gene expression. All these layers of complexity also increase the potential of
8
disease, since only one of the many steps involved in this transcriptional process needs to
be perturbed to result in the beginnings of disease.
2. Transcription factors: Structure and Function
Transcription factors are key proteins involved in regulation of transcription rates.
Transcription factors often have structural domains specific for a particular functions. For
example, transcription factors may have a DNA-binding domain to allow for binding to
specific DNA elements as well as other cofactors (63, 65). Many transcription factors
may also have activation domains and/or repression domains (63, 65). Other proteins can
bind to both of these domains to regulate transcription. Transcription factors known as
steroid hormone receptors have a specialized domain known as a ligand-binding domain
(REF). The ligand-binding domain binds ligand and opens the protein structure allowing
control of the activity of the trancription factor (63, 65). Two transcription factor types
form the main focus in the work to be reported in this thesis: 1) Runx proteins, and 2)
Steroid Hormone Receptors (primarily estrogen and androgen receptors) (Figure 1.2).
9
Figure 1.2. Runx proteins, Androgen Receptor and Estrogen Receptor
Figure 1.2 Legend. Domains present in Runx proteins, androgen receptor, and estrogen
receptor !. All Runx proteins share the runt homology domain. Unlike Runx proteins, the
androgen receptor (AR) and estrogen receptor (ER) ! have ligand binding domains.

RHD
RHD
RHD
DBD
DBD
LBD
LBD
NTD
NTD
Runx1
Runx2
Runx3
AR
ER!
A
B
1-539 540-647 648-919
1-180 181-265 266-595
1-64 65-181 182-415
NTD
NTD
NTD
CTD
CTD
CTD
1-176 177-303 304-596
1-87 88-204 205-480
10
2.1 Runx Proteins
There are three types of Runx proteins, Runx1, Runx2, and Runx3 (69). All share
a 128 amino acid stretch within the center of the linear polypeptide, known as the runt
domain. This domain binds to specific Runx elements in DNA. Runx elements usually
have a consensus binding motif of 5’-ACCACA-3’ (83). Structurally the runt domain is
characterized as a S-type immunoglobulin fold (5). This fold has two antiparallel _ sheets
with a total of seven to nine strands packed on one another, with the connections between
the strands provided by varying loop regions (5). It is mainly a _ sheet motif that forms a
type of glove around the DNA (Figure 1.3). All Runx proteins isoforms have the capacity
to bind to the same DNA element. They therefore regulate specific gene expression
depending on which Runx protein isoform is present in a particular cell type.
Additionally, a further level of specificity can be achieved by variability in the activation
and repression domains of particular isoforms that result in different coactivator and/or
corepressor recruitments on a particular promoter (69).
11
Figure 1.3 Runt Domain
Figure 1.3 Legend. Cartoon representation of the Runt domain. This domain has an S-
type immunoglobulin fold that forms a “glove” around DNA. The arrows represent the
several beta sheets and the cylindrical box represents a alpha helix present in the N-
terminal region of the domain. All Runx proteins share this domain (5).
12
Runx proteins localize with the nucleus using two domains : 1) a nuclear
localization signal (NLS) and 2) the nuclear matrix targeting signal (NMTS) (54). The
NLS is a stretch of 9 amino acids C-terminal to the Runt domain, and the NMTS is a 31
amino acid stretch located at the end of the Runx protein (54). The NMTS structurally
forms a helix-loop-helix structure and directs Runx proteins to subnuclear foci. This
localization is essential for transcriptional activity (165). Mutations or deletions of the
NMTS alters Runx2 spatial distribution within the nucleus, causing Runx2 to be more
mobile and unable to form specific nuclear foci (164). The mechanism of NMTS nuclear
foci formation is not clear but may involve the interaction of other proteins that bind near
or at the NMTS (164).
The activation domains of Runx proteins coordinate the binding of other proteins
that can enhance their transcriptional ability. The primary activation domain of Runx
proteins is a proline-serine-threonine (PST) rich region adjacent to the NMTS (143). The
PST region is known to bind to several coactivators that enhance Runx activity. If the
PST region is fused to a heterologous DNA-binding domain it can activate the
corresponding DNA promoters (143). There is another activation domain at the N-
terminal region of Runx proteins, but this domain by itself cannot potentiate the
activation if tethered to a heterologous DNA-binding domain (143). Unique to Runx2 is a
stretch of polyglutamines and polyalinines in the N-terminal region, known as the QA
domain. Polymorphisms of the QA region exist yet the exact function of these
polymorphisms is unclear. Some speculate that it may have some capacity to alter the
overall activity of Runx2 (30).
13
Repression domains are also present on Runx proteins. Therefore, Runx proteins,
in a context dependent manner, can either activate or repress transcription. This may
depend on the types and amounts of other proteins, which interact with Runx. All Runx
proteins have an inhibitor domain (ID) near the C-terminal end (155), which functions to
recruit corepressors. Regions of the N-terminus can inhibit Runx proteins ability to bind
DNA (66). The Runt domain itself has also been known to bind to corepressors and other
transcription factors that can alter the Runx protein-DNA binding (53). Therefore, there
are multiple regions involved in transcriptional repression or activation that coordinate
Runx proteins to either behave as transcriptional activators or repressors.
2.2 Steroid Hormone Receptors
Steroid hormone receptors (SHR) also have a DNA binding domain, as well as
activation and repression domains. Unique to SHRs is their ability to bind specific
ligands to a region known as the ligand binding domain (153). This allows a molecular
‘on/off’ switch for SHR. Particular to the work reported in this thesis are two types of
receptors: 1) estrogen receptors (ER), and 2) the androgen receptor (AR). Historically
SHRs were discovered much earlier than Runx proteins and greater overall
structural/functional information is known about SHRs compared to Runx proteins (28).
AR and ER are in a class of SHRs known as Type I, in which the receptors are in
an “off” position when ligand is not present (56). The AR is instead associated with heat
shock proteins sequestered in the cytoplasm, which does not allow binding to its DNA
motif (92). When AR’s ligand, dihydroxytestosterone (DHT, Figure 1.4), is present AR
14
then translocates to the nucleus and binds to specific elements known as AREs (androgen
response elements). ER is constitutively nuclear, and upon binding to it’s ligand, estradiol
(E2, Figure 1.4) it binds to it’s DNA motif known as EREs (estrogen response elements) .
Therefore, when ligand is added these receptors come in the “on” position and can bind
to their cognate DNA motifs recruiting several coactivators and/or coreprossors. (97).
The DNA-binding domain allows these receptors to bind DNA. The DNA binding
domain has a great deal of similarity in the ER isoforms ! and " (90). The ER! and "
share 97% homology in their DBD and consist of two zinc finger motifs that coordinate
the binding of DNA (90). Each zinc finger consists of four cysteine residues, which
chelates a zinc ion in the center (Figure 1.5B). Residues found in the D-box (distal zinc
finger) participate in homodimerization, while residues found in the P-box (proximal zinc
finger) interact specifically with DNA and in contact with a palindromic response
element (125, 126). ER! DBD behaves as a monomer in solution but forms dimers when
bound to DNA (125, 126).
The androgen receptor (AR) DBD shares a 56% homology to that of the ER!
DBD (34). Similar to the ER DBD it has two zinc fingers each consisting of four cysteine
residues that chelates a zinc ion in the center of each (Figure 1.5A). The AR DBD also
coordinates both dimerization and binding to its DNA element, via the D-box and P-box,
respectively (94). (148).
The other motif unique to SHRs and critical in the ‘on/off’ mechanism of these
receptors is the ligand binding domain (LBD). The sequence similarity between ER" to
15
ER! is 59% (90). Even though there is low sequence similarity of LBDs in general
throughout SHRs, there overall structure is similar. The LBD has a hormone binding site,
an interface to allow for homo or heterodimerization, as well as coactivator and
corepressor interaction.
The ER LBD consists of 11! helices and for comparison with other LBDs that
have 12 helices the nomenclature goes from H1-H12, with H2 missing (87, 130). The ER
LBD is structurally considered an a-helical sandwich. A ‘sandwich’ structure is organized
in three layers with H5, H6, H8, H9 as the central layer, with one side having H1 and H3
and the other side containg H7, H10, and H11 (87, 130). The ligand pocket is closed on
one side by an antiparallel "-sheet and on the other side by H12 known to be involved in
the activation function (AF-2) (87, 130).
Similar to the ER LBD the AR LBD contains 11! helices forming a sandwich
like structure as well (119).. The ligand for AR (DHT) is bound in a pocket of helices H3,
H5, and H10 different than the ER LBD (119). Unlike the ER LBD, the AR LBD has a
bipartite type structure with a flexible and a rigid region (119). However, similar to the
ER LBD, H12 closes the binding pocket and is known to be involved in activation
function (AF-2) as well (119). These subtle differences within the two LBDs allows for
selectivity of different ligands that ensures the proper transactivation function of each
receptor.
16
Figure 1.4. Ligands for ER and AR
Figure 1.4 Legend. Estradiol (E2) and dihyroxytestosterone (DHT), ligands for estrogen
receptor ! and androgen receptor respectively.
OH
HO
E2
OH
O
DHT
17
Figure 1. 5. DNA-binding Domain of AR & ER
G
G
Q K
T
C
L
I
C C
C
G
S
K
G
D
E
A
S
C
T
L
A
H
Y
Zn
C
C C
C
Zn
R L
L R K Y K Q K G E A A F R K F
P
S
V
R
A
S
N
D
P
T N
I K
D R
K R
F
P- box D- box 2+ 2+
T R
Y
C
A
V
C C
C
E
G
K
N
D
Y
A
S
G
Y
S
W
V
G
H
Y
Zn
S F R K F A
E
D N H G Q I Y M
C
C C
C
Q
A
T S
Zn
R
Q
I K
D R
K R
N
L
T
P
A
N
Q
D- box
P- box 2+
2+
A
Figure 1.5 Legend. DNA-binding Domain of the Androgen receptor and Estrogen
Receptor. A. DNA-binding domain of the androgen receptor. B. DNA-binding
domain of estrogen receptor !. (57, 73, 98, 107, 120, 122, 142, 145)
B
18
Both the DBD and LBD are highly structured regions of SHRs. However, an important
region where no crystallization studies have been done, and is highly variable among
SHRs is known as the N-terminal domain (NTD). The NTD is known to be random coil
and flexible (65). Spectroscopic analysis has shown that the NTD in conjunction with
other interacting proteins can acquire both secondary and tertiary structures (65).
Nevertheless, the types of structural changes that may occur in the presence of DNA are
specific to the promoter and in conjunction to coactivators present. AF regions in the
NTD provide docking sites for coactivators and corepressors (65). The large NTD of the
AR contains two activation functions AF1, and AF5 (6). ER! also contains an AF1
region in the NTD, which is only weakly involved in transactivation. Tthis region is
much shorter in the " form. In both isoforms the main transactivation potential resides in
the LBD AF2 region.
3. Transcription Factors: Genes encoding Runx proteins, ERs and AR
The RUNX genes for RUNX1, RUNX2, and RUNX3 are located on
chromosomes 21q22.12, 6p21, and 1p36.1 in humans, respectively (69). All RUNX
genes have two promoters P1 (distal) and P2 (proximal) resulting in different isoforms
(69). CpG island methylation may be a form of regulation of these RUNX genes, since
CpG islands are both upstream and downstream of the gene (69). This will allow for
tissue specificity of what RUNX genes are transcribed. The ERs are encoded by two
genes: ! on chromosome 6q25.1 and " on chromosome 14 (136). The AR is encoded on
the X chromosome at Xq11-12 (10). The AR transcript itself is stabilized by the
19
presence of ligand. Unlike the AR, the ER is degraded by the presence of its ligand.
However, the degradative process for the ER is dependent on other proteins present
within the cell and not necessarily the ER transcript itself. Studies have shown that p160s
may help in stability of the ER (129).
4. Transcription Factors: Development and Disease
Transcription factors play a fundamental role in cell fate, and development.
However, since these proteins are necessary in the developmental processes it isn’t
surprising that alteration of the activity of these proteins by mutation, changes in cellular
environment, etc. can lead to the onset of various diseases including cancer.
4.1 Runx Proteins in Development and Disease
The three types of Runx proteins are essential in development and differentiation
of cells. In general, Runx1 is necessary for hematopoiesis, Runx2 for osteogenesis by
osteoblast differentiation, and Runx3 for neuronal and gastrointestinal development.
However, several of the Runx proteins overlap in function and may work in conjunction
to determine the cellular status (156).
Runx1 knockout mice are embryonically lethal at day 11.5 to 13.5 of gestation
and have a block in definitive hematopoiesis (151). Analysis of the blood and the fetal
liver shows no formation of erythroblasts (immediate precursor to erthrocytes that are the
most common type of red blood cell) or megakaryocytes (a bone marrow cell responsible
for the production of blood platelets necessary for normal blood clotting). Embroyonic
20
stem (ES) cells lacking Runx1 cannot differentiate into definitive erythroid (red blood
cell like) and myeloid (bone marrow like) cells in vitro (151). However, heterozygote
studies have shown that primitive erythropoiesis (red blood cell production) appears to be
normal (162). Therefore, Runx1 is crucial in the later steps of hematopoetic
differentiation. Studies have also shown that Runx1 is necessary for the differentiation of
hematopoietic cells from a population of endothelial cells in the embryo as well (40).
Therefore, Runx1 may regulate the development of hematopoetic stem cells from
endothelial cells (40). Runx1 regulates several genes necessary in the myeloid and
lymphoid differentiation including IL3, CSF2, CSF1R, CD4, and Tcrd (101).
Since Runx1 normally functions in hematopoiesis it should be expected that
perturbation of this vital transcription factor in humans can lead to several diseases with
relation to blood. Loss of function mutations in Runx1 are involved in acute myeloid
leukemia (AML). Runx proteins historically have also been known as AML proteins for
there association with leukemia. Chromosomal translocations of Runx1 occur in
approximately 50% of the leukemia cases (118). Two major types of translocations are
known as Runx1-ETO (chromosome eight to twenty-one translocation) and ETV6-Runx1
(85, 132). Both types of translocations repress the ability of Runx1 to function. Another
type of disorder known as familial platelet disorder is also known to develop into AML
(84). This disorder is also due to loss of function mutations in Runx1. Also gain of
function mutations in Runx1 may be involved in initiating leukemias as has been
postulated in acute megakaryoblastic leukemias with down syndrome patients that have
an extra copy of Runx1 (159).
21
Runx2 like Runx1 is very important in differentiation of stem cells. However,
Runx2 functions mainly in bone by influencing osteoblast differentiation from
mesenchymal stem cells. Runx2 knockout mice are unable to form bone and die soon
after birth (55). Runx2 is necessary for bone integrity and functions in maintaining the
balance of bone formation (osteoblasts) versus bone resorption (osteoclasts) (55). Runx2
is required for mesenchymal condensation, which is mediated by extracellular matrix and
cell adhesion molecules (55). These condensations are necessary so cartilage can
differentiate and form bone. Mesenchymal stem cells that lack Runx2 can differentiate
into chondrocytes and adipocytes, but not osteoblasts (55). Runx2 target genes are known
as critical markers in osteoblast differentiation,. tThe classical target is osteocalcin that
recently has been implicated in bone metabolism (68). There are several other genes such
as collagens and MMPs (matrix metallo proteins) that are targets for Runx2 and
necessary in osteoblast differentiation (39).
Haploinsufficiency in humans of Runx2 is involved in cleinocranial dystosis
(CCD) (100). Individuals with CCD usually have deformities within the collar bones and
the cranium (100). CCD patients have underdeveloped bones and joints, they are shorter
in stature than family members without CCD, they have supernumerary teeth (extra
teeth), and they may have a bulging forehead (100). Runx2 has been also implicated in
cancer metastasis (108). Runx2 is prevalent in both breast and prostate cancers yet the
exact mechanisms of Runx2 in these cancers is not clear (108). The studies presented in
this thesis will address some of the aspects of Runx2 in both breast and prostate cancer.
22
Runx3 is expressed in several cell types including epithelial cells, mesenchymal
cells and blood cells. However, a majority of the expression is within dorsal root ganglion
neurons, epithelial cells in the gastrointestinal tract, and hematopoietic cells (9, 44).
Therefore, Runx3 functions in the physiology of several aspects of an organism. Runx3
has been shown to function in lineage specific differentiation of neurons as well T-cell
development (44, 156).
Runx3 knockout mice have been shown to have an increase in cellular
proliferation and decreased apoptosis, leading to the development of polyps and a
“hyperplastic phenotype” in the gastrointestinal tract (9). In patients suffering with gastric
cancer it has been found that Runx3 function is inactivated by deletion or gene silencing
due to promoter hypermethylation during early as well as late stages of this disease (99).
Runx proteins are important to help in the renewal of cells in various biological
responses, which an overview was given above. All three Runx proteins are involved in
differentiation and have been shown to be involved in different types of cancer. Certain
mechanistic aspects of how Runx proteins function in cancer have been discovered which
include translocations, deletions, and mutations that alter the function of these proteins.
4.2 ER and AR: Development and Disease
Both ER and AR are necessary in the development of females or males,
respectively (149, 160). The receptors are necessary in the development of the
reproductive tract, the neuroendocrine, skeletal and cardiovascular systems. Their f
23
unctions are crucial during puberty. Alteration of the function of either receptor has also
been implicated in cancer (149, 160).
The two types of ERs (! and ") vary in their cellular distrubtion. ER! is found in
the uterus, mammary gland, testis, pituitary, liver, kidney, heart, and skeletal muscles. On
the other hand, ER" is expressed in ovary and prostate. ER! and " are equally expressed
in the epididymis, thyroid, adrenals, gonad and various regions of the brain. Knockout
mice have been developed for each of the two types. When ER! is knocked out both
males and females are infertile, when ER" is knocked out only females become
“subfertile” . In males ER! has been shown to be involved in the development of testes,
aggression, and involved in mating behavior, with ER" disruption having no functional
consequence. In females the ER! knockout affects the uterus’ sensitivity to estradiol and
there is no ovulation, ER" knockout does not seem to have as strong of an effect. In
regard to bone, ER! knockout mice are shorter than wildtype in both males and female
yet with females there is a decreased bone diameter and males have a decrease in density.
ER" knockout females surprisingly have increased bone density and with males there is
no affect. Only in ER! and " double knockouts is there a deleterious affect in the
cardiovascular system after vascular injury [as reviewed in (149)].
ER function can have an impact in disease. Women after menopause (when there
is a sharp decrease in estrogen) tend to develop osteoporotic fractures, and the protective
effects of the ER on bone are lost (59). Also, there have been implications in the increase
in obesity with the loss of estrogen and ER function (149). Breast cancers are either ER
24
positive or negative, the majority being ER positive. In these cases ER! overexpression
is associated with increased cell proliferation (22, 147, 152). The effect of ER" in cancer
is still not well understood (136). There are several endocrine therapies that involve the
use of selective estrogen receptor modulators (SERMS) that act as ER! antagonists and
may help in the degradation of the ER!. Breast cancers may be or become ER! negative
(22, 147, 152). A loss of ER! results in an increase in metastasis and a poor prognosis in
breast cancer patients. ER! has also been implicated in ovarian cancer, colon cancer,
prostate cancer and endometrial cancer (16, 38, 70, 154). For example, the loss of ER"
has been associated with colon cancer and ER" specific agonists have been used for
treatment (38).
The AR, similar to the ER, is involved in the development of primary sexual
characteristics of sexual organ development as well as secondary sexual characteristics of
sexual traits outside of the organ development (160). AR knockout mice (ARKO) abolish
the development of masculine organs and instead exhibit female-like external organs
(160). The male reproductive organs of the seminal vesicles, vas deferens, epididymis
and prostate were not present in ARKO mice, however small testes were present (160).
This is very similar to what occurs in males that suffer from AIS (androgen insensitivity
syndrome) or Tfm (testicular feminization) (160). ARKO male mice show no aggressive
behaviors and do not have atypical male aggressiveness. Outside of the obvious sexual
affects of abolishing the AR there are affects on the bone as well (160). ARKO mice have
increased bone turnover compared to wildtype mice, with bone resorption greater than
25
formation (160). Implicating an increase in osteoclast activity, and thus leading to a
decreased overall bone volume (trabecular and cortical) compared to wildtype (160).
AR is critical in the development and maintenance of the prostate. The prostate is
composed of four major regions: 1) peripheral zone, 2) central zone, 3) transition zone,
and 4) the stroma (Figure 1.6) (20). The transition zone is where 34% of prostate cancers
develop as well as a non-life threatening condition known as benign prostatic hyperplasia
(BPH) (4, 128). In the peripheral zone is where 64% of the prostate cancers develop (20).
Androgens and estrogens may both play roles in the development of BPH (127).
However, in prostate cancer androgens and AR activity seem to be primary factors (49).
Increase in AR activity leads to proliferative effects within the prostate (49). One of the
classic biomarkers indicative of prostate cancer is the AR target gene PSA (prostate
specific antigen) measuring both the activity of the AR signaling-axis, and prostate organ
size (49). Similar to the ER, selective androgen receptor modulators (SARMS) are used
as ablation therapies in addition to or in place of castration. Nevertheless, recurrence of
prostate cancer inevitably occurs due to mutations of the AR, increases in AR and
coactivator levels, and epigenetic changes allowing greater accessibility for AR target
genes (49).
26
Figure 1.6. Anatomy of the Prostate.
Figure 1.6 Legend Depicted are the central, transitional, and peripheral zone of the
prostate in relation to the urethra and bladder (20).
Bladder
Urethra Central Zone
Transition Zone
Peripheral Zone
Prostate
27
Unlike ER and breast cancer, there are no AR negative tumors. However, similar
to breast cancer, prostate cancer has the ability to metastasize to bone. In prostate cancer
there is the development of lesions and increased bone deposition upon metastasis
(Figure 7). In breast cancer bone metastasis leads to an osteolytic phenotype (Figure 1.7).
There is a connection between the ER, AR and bone, not only in cancer, but in normal
aspects of bone turnover and bone growth. The main regulatory protein involved in the
bone is Runx2, therefore there may be a link between SHRs and Runx proteins as shown
in this thesis. .
28
Figure 1.7. Bone Metastasis of PCa (prostate cancer) and BCa (breast cancer)
Figure 1.7 Legend. PCa bone metastasis of PCa cells (red spheres) interacting with
osteoblasts (stars) to alter osteoblast activity and lead to the development of lesions on
the bone. BCa bone metastasis of BCa cells (turquoise spheres) interact with osteoclasts
(hexagons) to alter osteoclast activity and lead to the development of an osteolytic
phenotype. This is a simplified model representing an array of complex signaling that is
occurring.
29
5. Cofactors
Transcription factors do not work in isolation. They work with a series of proteins
that help in modulating them in a positive or negative way. A brief overview of cofactors
is provided here. There are several definitions that exist for cofactors; the one used here
will be at its simplest, any protein that has the ability to modulate transcription without
binding to DNA itself (a coactivator or corepressor) These cofactors alter the way
transcription factors are bound to DNA and the conformation of these transcription
factors as well. Or these cofactors may affect the DNA and histone complex itself to
modify the transcriptional activity.
5.1 Runx Cofactors
The main cofactor for Runx proteins is the protein known as CBF" (37). CBF"
itself does not bind DNA but is known to heterodimerize with Runx proteins by binding
to the runt domain. This alters the conformation of the runt domain to allow Runx
proteins to bind DNA with an affinity 6 to 10 times greater than in the absence of the
CBF" (37). CBF" knock-out mice have a similar phenotypes as the respective Runx
knockout mice. CBF" translocations have also been known to be involved in cancers
such as the CBF"-MYH11 present in leukemia (75). Additionally, several coactivators
and corepressor bind and modulate Runx activity (104, 133). For example, Runx2 binds
to p300, CBP, MOZ, and MORF all histone acetyltransferases (HATs) which enhances
Runx2 transcriptional activity (Figure 8A) (104, 133). Corepressors, such as
30
histone deacetylases (HDACs) Groucho/TLE (proteins associated with HDACs) (Figure
1.8A) (48, 155). Runx2 has also been shown to interact with proteins involved in the
proteasome degradation pathway, thereby inhibiting Runx2 activity (3). Involvement of
cofactors in Runx function is a continually growing in which a few are described here.
5.2 ER and AR Cofactors
Much more is known about the cofactors of ER and AR function. These cofactors
include the p160s class of coactivators such as GRIP1 (42). GRIP1 enhances
transcriptional activity by acetylation of histones to allow the DNA to be more accessible.
GRIP1 is considered a primary coactivator. Other coactivators include CARM1, which is
a histone methyltransferase. CARM1 is considered a secondary coactivator, because it
coactivates in conjunction with GRIP1 and not alone (17). P300 and CBP are general
coactivators for ER, AR and Runx proteins (Figure 1.8B). There are several corepressor
for ER and AR, including HDACs (as with Runx proteins) and associated proteins
HDAC complexes such as NCOR1 and NCOR2 (Figure 1.8B) (31). Complex formation
of either coactivators or corepressors that mediate the changes in transcriptional activity
of ER and AR, mediated transcriptional modulation. In both breast and prostate cancers
several coactivators and corepressor gene expressions are altered, thereby modulating ER
and AR activity respectively.
31
Figure 1.8. Coactivator/Corepressors of Runx2 and Nuclear Receptors
Figure 1.8 Legend. A. The coactivator and corepressor known to bind to Runx2 and
thereby modulate transcription. B. The coactivator and corepressor known to bind both
the androgen receptor and estrogen receptor !.
HRE
Runx Element
Runx2
CBF!
p300
Runx2
HDACs
TLE
NR NR
GRIP1
CARM1
p300
NR
NCOR
Moz
HRE
Runx Element
A
B
32
6. Outline and Scope of this Thesis
As discussed above several coactivators and corepressor are involved in
modulating transcription factor activity. However, transcription factors themselves can
interact with each other to alter their activity. They are known as coregulators or
transcriptional collaborators. It has been shown that ER interacts with NFkB to alter
NFkB activity (110). The GR has been shown to interact with AP1 to modulate AP1
activity (21). Several transcription factors have been shown to interact with Runx
proteins to modulate their activity and this includes SMADS, Twist, AP1, Lef1,
PPARgamma, Oct-1, C/EBP, Dlx proteins, Msx2, Ets1, Menin1, and VDR (71).
The work presented in this thesis focuses on the capacity of Runx proteins, in
particular Runx1 and Runx2, to interact with, and be modulated by the activity of steroid
hormone receptors ER! and AR. As presented here, the modulation of Runx1 and Runx2
activity occurs in both a ligand and receptor specific manner. This finding has significant
biological impact. Since Runx proteins do not have a molecular switch like the ER or
AR, a new layer of control is imposed by its steroid receptor interactions. The overall
working hypothesis in this thesis is that SHRs (ER or AR) repress or activate Runx
proteins (Runx1 or Runx2) upon treatment of a particular ligand. This is achieved
through direct protein-protein interactions. The results presented in this thesis can
potentially influence perspectives in both development and disease in which both Runx
proteins and SHRs are present.
There were two main overall projects conducted in this thesis to address the
hypothesis mentioned above. One part examined how ER! and " modulated
33
Runx1 and Runx2 activity, and the other examined how AR modulated Runx1 and Runx2
activity. I report different outcomes dependent on the transcription factors examined. In
summary the data presented will show that Runx2 activity can be strongly inhibited by
ER! and AR, whereas Runx1 activity was inhibited by the AR only. The physiological
implications of these two major findings are discussed in the rest of this thesis.
34
Chapter 2: Material and Methods
Materials and methods are presented here in one unified chapter since similar
methods were used in both projects (Chapters 3 & 4) within this thesis. Also the
development of these techniques led to discovery of novel concepts unrelated to the main
projects presented in this thesis.
1. Cell Culture
Several cell lines were used to assess the activity of endogenous steroid hormone
receptors and Runx proteins, as well as those for the overexpression of these proteins. A
majority of the cell lines used were derived from prostate cancer and breast cancer
patients, which are described below. All cells were maintained in a 37
o
C incubator with
5% CO
2
, in medium containing no antibiotics. The presence of antibiotics masks both
bacterial and mycoplasma contamination.
PC3 cells used were derived from a bone metastasis of prostate adenocarcinoma
from a 62-year-old Caucasian male and developed in Dr. Tilley’s lab (12) . PC3 cells
were maintained in RPMI-1640 and 5% fetal bovine serum (FBS). They were grown in a
T-75 flask and at confluency they were passaged at a ratio of one to five. PC3 cells have
an approximate doubling time of 24hrs.
LNCaP cells used were derived from a lymph node metastatic lesion of human
prostate adenocarcinomas from 50 year-old Caucasian male obtained from ATCC.
LNCaP cells were maintained in RPMI-1640 and 5% FBS. They were grown in a T-75
35
flask and at confluency they were passaged at a ratio of one to five. LNCaP cells have
and approximate doubling time of 60hrs.
C4-2B cells used were derived from LNCaP cells that were grown in castrated
nude mice and metastasized to bone. These cells were obtained from Labcorp. C4-2B
cells were maintained in RPMI-1640 and 5% FBS. They were grown in a T-75 flask and
at confluency they were passaged at a ratio of one to five. C4-2B cells have an
approximate doubling time of 48 hrs.
MDA-MB-231 cells used were derived from a mammary gland metastatic breast
adenocarcinoma from a 51 year-old Caucasian female and obtained from ATCC. These
cells were maintained in Dulbeco Modified Eagle Medium (DMEM) and 5% FBS. They
were grown in a T-75 flask and at confluency were passaged at a ratio of one to ten.
MDA-MB-231 cells have an approximate doubling time of 24hrs.
MCF7 cells used were derived from a mammary gland breast adenocarcinoma
from a 69 year-old Caucasian female and obtained from ATCC. These cells were
maintained in DMEM and 5% FBS. They were grown in a T-75 flask and at confluency
were passaged at a ratio of one to ten. MCF7 cells have an approximate doubling time of
29hrs.
T47D cells used were derived from ductal carcinoma of the breast from a 54 year-
old female patient and were obtained from ATCC. These cells were maintained in
DMEM and 5% FBS. They were grown in a T-75 flask and at confluency were passaged
at a ratio of one to ten. These cells have an approximate doubling time of 34hrs.
MC3T3-E1 cells used were derived from a mouse osteoblast lineage and obtained from
36
ATCC. These cells were maintained in aMEM and 10%FBS. These cells were grown in a
T-75 flask and at confluency were passaged at a ratio of one to ten. These cells have an
approximate doubling time of 18hrs. To support development of the osteoblast
phenotype, cells were treated with 50 !g/ml ascorbic acid and 10 nM ß-glycerophosphate
commencing at confluence to induce osteoblast differentaition. Also Runx-reporter
osteoblasts were generated by stable transfection of MC3T3-E1 cells with the
6XOSE2–luc reporter construct using the calcium phosphate coprecipitation method and
hygromycin (100 ng/ml) as the selection drug. Resistance to hygromycin was conferred
by cotransfection of the pCEP plasmid (1:15 molar ratio).
Cos7 cells are an SV-40 transformed cell line derived from monkey kidney
fibroblasts and obtained from ATCC. These cells were maintained in DMEM and 5%
FBS. These cells were grown in a T-75 flask and at confluency were passaged at a ratio
of one to ten. Cells were discarded after 15 passages of a cell stock due to alterations in
signaling as assessed from transient transfections (protocol described below). Cos7 cells
have an approximate doubling time of 24 hours.
Trypsinization of cells was done in the following manner. Upon confluency the
cells were washed with PBS and trypsinized with 0.5% Trypsin form 5-10 minutes in a
37
o
C incubater. Cells were resuspended with the media specific to each cell type and
replated in the appropriate ratio in a subsequent flask.
37
A summary is provided in Table 2.1 of cell type used and AR, ER!, Runx1 and
Runx2 status. This gives an overview of how these cells may be useful as a tool in
assessing Runx activity in conjunction with steroid receptor activity.
38
Table 2.1 Summary of endogenous expression of active
Runx1, Runx2, AR, and ERa.
Cell Type Runx1 Runx2 AR ERa
PC3
+ + - -
LNCaP
- - + +
C4-2B
- - + +
MDA-MB-231
- + - -
MCF7
- + - +
T47D
- + + +
MC3T3-E1
+ + + +
Cos7
- - - -
39
2. Transient Transfections: Luciferase Reporter Assays
Transient transfections were done with luciferase reporters for the androgen
response element (ARE), estrogen response element (ERE), or six times osteoblast
specific element 2 (6XOSE2). The ARE reporters we used in our transfections consisted
of the PSA540 reporter, Probasin reporter, and MMTV reporter (graciously provided by
Dr. R.J. Matusik’s group at the Vanderbilt Prostate Cancer Center, Nashville TN). The
ERE reporter used was the MMTV reporter with the ARE elements replaced by ERE
elements (graciously provided by Dr. Stallcup’s group at USC, Los Angeles CA). The
6XOSE2 reporter was derived from the OSE2 element upstream of the osteocalcin gene
(graciously provided by Dr. Karsenty’s group at Columbia University, New York NY).
The ARE reports on androgen receptor activity, the ERE reports on estrogen receptor
activity, and 6XOSE2 reports on Runx1 and Runx2 activity.
Cos7 cells were seeded at a density of 10,000 per well in 96-well plates (3mm in
diameter) and grown in 5% charcoal-stripped serum (CSS)-containing phenol red-free
DMEM for 24 h to remove any steroids present with the cells. The indicated plasmids in
each experiment (50 ng reporter, 52 ng total) were transfected using ‘LTX and Plus
reagent’ according to manufacture’s specification (Invitrogen, Carlsbad, CA) (Figure
2.1). Equimolar amounts of the empty vectors pCDNA 3.1 and pSG5 were used as
controls, and the pCAT-basic promoter-less plasmid was used as ‘filler’ DNA to ensure
that the total amount of DNA transfected is the same in all wells. Under these conditions
the total amounts of promoter equivalents (molar) and DNA mass (ng) were the same in
all wells of a particular experiment. As an internal control, 0.01 ng of tk-Renilla-luc or
40
CMV-renilla-luc (Promega, Madison, WI) was used for correction of transfection
efficiency to ensure that the basal promoter of the renilla had no effect in our transfection
assays, example is shown in Figure 2.2 of setting up the appropriate transfection
conditions. After transfection cells are lysed with passive lysis buffer (Promega) for 15
minutes and read on a luminometer (BMG Labtech Floustar Optima, Durham NC)
T47D, MCF7, and MDA-MB-231 cells were seeded at a density of 20,000 per
well in 96-well plates and grown in 5%-CSS phenol red free DMEM for 48hrs. PC3 cells
were seeded at a density of 15,000 cells per well in 96-well plates and grown in 5%-CSS
phenol red free RPMI-1640 for 24hrs. Transfections were done in a manner similar to
described above.
LNCaP cells were seeded at a density of 50,000 cells per well in 12 well plates
(10.5mm in diameter) and grown in 5%-CSS phenol red free RPMI-1640 for 48 hrs.
Transfections were done in a similar manner as described above but scaled by a factor of
15 for the amount of DNA transfected, since there is approximately a 15 fold increase in
surface area when scaling up from a 96-well plate to a twelve well plate.
41
Figure 2.1 Luciferase Reporter Assay
Figure 2.1 Legend. An overview of luciferase reporter activity. Firefly and Renilla
luciferase reporters are incubated with lipofectamine and then overlayed in the cell type
of interest that has been plated in 96 wells 24 hrs prior to transfection. After incubation
with lipofectamine, cells are overlayed with media containing the appropriate ligand.
Cells are subsequently lysed with passive lysis buffer and read on a luminometer plate
reader.
96 Well Plate
Cells
Lipofectamine
Firefly Luciferase
Cells are lysed and read on a 96 well
plate reader.
42
Figure 2.2 Developing the Luciferase Reporter Assay
Figure 2.2 Legend. An example of luciferase activity (6XOSE2-luciferase transfected in
Cos7 cells) tested and corrected for renilla. The renilla activity does not significantly
change in each condition and the amount of Runx2 (R2) construct also did not have an
affect in altering renilla activity (C). The coexpression of NTD/DBD of the AR,
LBD/DBD of the AR, and wt AR modulated Runx2 activity similarly in both Runx2
concentrations (A,B).
43
1 ng of Runx2 transfected with 1ng of each construct
0
5000
10000
15000
20000
25000
30000
R2 R2 + DHT R2 + AR R2 + AR +
DHT
R2 +
NTD/DBD
R2 +
NTD/DBD
+ DHT
R2 +
DBD/LBD
R2 +
DBD/LBD
+ DHT
mock mock +
DHT
RLU
uncorrected
corrected
Renilla Luciferase Activity
0
1000
2000
3000
4000
5000
6000
7000
8000
R2
R2 + DHT
R2 + AR
R2 + AR + DHT
R2 + NTD/DBD
R2 + NTD/DBD + DHT
R2 + DBD/LBD
R2 + DBD/LBD + DHT
mock
mock + DHT
RLU
0.5 ng of RUNX2
1.0 ng of RUNX2
0.5 ng of Runx2 transfected with 1ng each construct
0
2000
4000
6000
8000
10000
12000
14000
16000
R2 R2 + DHT R2 + AR R2 + AR +
DHT
R2 +
NTD/DBD
R2 +
NTD/DBD
+ DHT
R2 +
DBD/LBD
R2 +
DBD/LBD
+ DHT
mock mock +
DHT
RLU
uncorrected
corrected
A
B
C
44
3. Electromobility Shift Assays (EMSA)
Whole cell extract (WCE) preparation for EMSA was prepared in the following
manner. Cells were grown in 10 cm (diameter) plates to confluency in respective growth
medium. Subsequently, cells were scraped with a cell lifter and centrifuged at 3000 rpm
for 5 minutes at 4
o
C. Cell pellets were then resuspended 50-100 !l (1.5X the packed
volume) in lysis buffer (100 mM Hepes, 500 mM KCl, 5 mM MgCl
2
, 0.5 mM EDTA,
28% glycerol, pH 7.5) with freshly added 0.1 mM Na
3
VO
4
, 1mM PMSF and 1% protease
inhibitor cocktail (Sigma, St. Louis MO). The cells were subsequently lysed by passing
the cells through a 1 cc insulin syringe and centrifuged at 14,000 rpm for 10 minutes at
4
o
C to remove any insoluble components. The supernatant was aliquoted, snap frozen in
liquid nitrogen, and stored at -80
o
C until use.
EMSA was performed with 10 !g of WCE (as determined by a Bradford assay)
and 100 fmoles of an end-labeled 21 base pair oligonucleotides probe containing the
OSE2 site when assessing Runx protein complex binding Probe was prepared as
described in Luppen et. al 2003 (77). The WCE was incubated on ice for 10 minutes with
100 mM KCl, 1µg salmon sperm, unlabeled oligonucleotides, and antibodies as
indicated. The final reaction was done in 20 mM Hepes buffer (pH 7.5) containing 50
MM KCl, 1mM MgCl
2
, 2mM EDTA, and 2.8% glycerol for 10 minutes on ice followed
by 15 minutes at room temperature with the labeled probe. The complexes that were
formed were resolved in 0.25X TBE native polyacrylamide (6%) gel containing 5%
glycerol as described Luppen et. al. 2003 (77) (Figure 2.3). EMSA was
45
performed in a similar manner using an AREIII probe when assessing AR protein
complex binding.. The AREIII probe was designed based on the PSA enhancer region
where it is well known that AR binds (personal communiciation Dr. Li Jia)
In attempts to identify the DNA binding activity of Runx proteins in prostate
cancer cell lines and improve the EMSA protocol, extracts from PC3, C4-2B, and LNCaP
cells were examined. PC3 cells had Runx1 and Runx2 described in Chapter 4. C4-2B and
LNCaP cells did not have any Runx proteins that could bind DNA, yet when incubated
with PC3 extracts there was an increase in Runx binding, greater than PC3 cells alone as
shown in Figure 2.4 and Figure 2.5 respectively. This data suggests that there is some
“helper” activity present in PC3 extracts which enables the Runx present in C4-2B and
LNCaP to bind DNA. To further understand this “helper” activity for Runx proteins, C4-
2B cells were incubated with extracts from various cell types (Figure 2.6). All the
extracts that were incubated with C4-2B extracts could rescue the Runx DNA binding
activity of C4-2B cells. Furthermore, it was possible to obtain nuclear extracts of HeLa
cells that had been fractionated using a P11 cation exchange column (fractions were
graciously provided from Woojin An’s group). When testing these fractions with C4-2B
extracts the protein(s) present at 0.5M and 0.8M NaCl elution were able to rescue C4-2B
Runx activity strongly compared to the background of the elution alone (Figure 2.7).
These data suggest that there was the need of a missing “helper” activity to restore C4-2B
Runx activity. This project was not further continued due to the development of other
projects within the lab, but these data helped in establishing, and developing the EMSA
protocol to assess endogenous Runx protein binding to DNA in prostate cancer cell lines.
46
To establish the EMSA protocol for AR binding to DNA in prostate cancer cells,
EMSAs were done with LNCaP and C4-2B WCE against an androgen response element
(AREIII) probe (Figure 2.8). EMSA using AREIII as a probe revealed no differences in
activity from WCE of LNCaP and C4-2B cells, indicating that the AR’s ability to bind to
its cognate DNA response element wasn’t different in etiher cell type (Figure 2.8).
47
Figure 2.3 Overview of electromobility shift assay (EMSA)
Figure 2.3 Legend. Whole cell extract (WCE) was obtained from the cell type of interest
and incubated with probe alone, or probe and antibody. Probe alone is shown in the first
lane. If the probe bound to any protein or protein complex in the WCE a shift would
occur as shown in the second lane. When the WCE and probe were incubated with a
specific antibody and there was a supershift as shown in the second lane then a specific
protein against that antibody was in the complex. Of main interest to this project was the
probe based on the osteoblast specific element, both wt and mutant oligonucleotides
sequence of the sense strand are given above.
48
WCE
WCE
6% Polyacrylamide Gel
WCE
antibody
antibody
Probe
Wt OSE2 sense:
CTGCAATCACCAACCACAGCA (21)
mut OSE2 sense:
CTGCAATCACCAGCAGCAGCA (21)
49
Figure 2.4 Assessment of Runx2 activity in C4-2B.
Figure 2.4 Legend. Assessment of Runx2 activity in C4-2B cells as well as recovery by
an unknown “helper” activity. Labeled oligonucleotides corresponding to the OSE2
probe were incubated with WCE (10 !g/reaction) obtained from PC3 or C4-2B cells,
separately or in combination as indicated. Unlabeled wild-type (Wt) and mutant (Mu)
probes were used for competition reaction as indicated to assess specific Runx protein
binding. Extracts were also boiled (B) to ensure that salts in the buffers were not involved
in altering Runx activity.
wt mut
10x 100x 10x 100x
wt mut
10x 100x 10x 100x
PC3AO PC3AO & C4-2B
1 2 1+2 1B+2 1+2B
1 1+2
1 = probe (OSE2) and PC3AO extract, B=Boiled
2 = probe (OSE2) and C4-2B extract, B=Boiled
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Lanes
50
Figure 2.5 Assessment of Runx2 activity in LNCaP cells
Figure 2.5 Legend. Assessment of Runx2 activity in LNCaP cells as well as recovery by
an unknown “helper” activity. Labeled oligonucleotides corresponding to the OSE2
probe were incubated with WCE (10 !g/reaction) obtained from PC3 or LNCaP cells,
separately or in combination as indicated. Unlabeled wild-type (Wt) and mutant (Mu)
probes were used for competition reaction as indicated to assess specific Runx protein
binding. Extracts were also boiled (B) to ensure that salts in the buffers were not involved
in altering Runx activity.
wt mut
10x 100x 10x 100x
wt mut
10x 100x 10x 100x
PC3AO PC3AO & LNCaP
1 2 1+2 1B+2 1+2B
1 1+2
1 = probe (OSE2) and PC3AO extract, B=Boiled
2 = probe (OSE2) and LNCaP extract, B=Boiled
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Lanes
51
Figure 2.6 “Helper activity” from other cell types.
Figure 2.6 Legend. EMSA against the OSE2 probe was performed on C4-2B WCE
separately or in combination with WCE extract obtained from various cell types as
indicated. All cell types were able to ameliorate the inhibited DNA binding activity of
Runx proteins in C4-2B cells.
0 1 2 1+2 3 1+3 4 1+4 5 1+5
0 = OSE2 probe
1 = C4-2B
2 = CV1
3 = Hela
4 = HL-60
5 = MCF7
52
Figure 2.7 HeLa NE extract fractions recovered Runx activity in C4-2B cells.
Figure 2.7 Legend. HeLa NE fractions obtained from a P11 cation exchange column
when incubated with C4-2B WCE and the OSE2 probe. C4-2B Runx DNA binding
activity was strongly recovered at 0.5 and 0.8M NaCl elutions, as shown above. There
was nothing in the flow through of the column that was able to “help” Runx DNA
binding in C4-2B cells. Proteins from the HeLa NE fractions themselves did not bind to
the OSE2 probe.
0.1 0.1+2 0.3 0.3+2 0.5 0.5+2 0.85 0.85+2 1 1+2 1.5 1.5+2 0 2 FT FT+2 NE NE+2
0 = probe (OSE2)
2 = probe and C4-2B extract
0.1 to 1.5 are M concentrations of NaCl elution of a P11 Cellulose phosphate Fibrous Cation
Exchange Column of Hela NE dialyzed with WCE buffer
C4-2B WCE and HeLa NE
53
Figure 2.8 AR DNA-binding activity in C4-2B and LNCaP Cells.
Figure 2.8 Legend. Labeled oligonucleotides corresponding to the sequence of ARE III at
the PSA enhancer were incubated with WCE (10 !g/reaction) obtained from LNCaP or
C4-2B cells treated with 10nM DHT or EtOH vehicle for 4 hours. Unlabeled wild-type
(Wt) and mutant (Mu) probes were used for competition reaction as indicated. The
specific DNA-protein complexes are indicated by the arrowhead. The probe sequences
are: 5’-CTCTGGAGGAACATATTGTATCGATTGT-3’ (Wild-type probe);
5’CTCTGGAGCAAAATATTCTAACGATTGT-3’ (Mutant probe).
54
55
4. Immunofluorescence
Cells were grown on a 18 mm
2
coverslips in six well plates (VWR) for 24 h using
5% CSS-containing phenol red-free DMEM. Cells were transfected with 100 ng each of
the described plasmids or endogenous proteins/markers were examined, and subsequently
treated with vehicle or 10nM of appropriate hormone for 24 h. Cells were then fixed
with 95% methanol for 15 min and permeabilized with 1% saponin (Sigma, St. Louis
MO). Proteins were visualized with the respective primary antibodies and secondary
antibodies conjugated to either rhodamine or a fluorescein tag. Cells were mounted using
Vectashield Hard Set mounting medium with DAPI which intercalates with the DNA to
allow it to be visualized by fluorescent microscopy (Burlingame, CA). Cells were viewed
using a LSM 510 Zeiss confocal microscope at 60X magnification. Images were
processed using the default settings on Image J for colocalization finder, surface plot
diagrams, and RG color picker. Using the default settings on the colocalization finder we
quantitated the percentage colocalization where both red and green pixels overlapped to
yield yellow.
An example of the confocal microscopy is presented in Figure 2.9. C4-2B cells
were examined for AR and several histone modifications after either vehicle treated or
DHT treated conditions (activated AR condition). The H3K4 monomethylation
(monoK4) and H3K4 trimethylation (triK4) histone marks were altered upon DHT
treatment throught the entire cell nucleus (Figure 2.9). Interestingly, the monoK4 mark
moved to the nuclear periphery upon DHT treatment, while the triK4 mark moved from
the nuclear periphery and diffused everywhere in the nucleus (Figure 2.9). This suggests
56
that there are nuclear zones and during transcriptional activation (such as DHT treatment
to activate the AR) these zones are altered.
57
Figure 2.9 Nuclear Changes in Epigenetic Marks Upon DHT treatment in C4-2B
Figure 2.9 Legend.C4-2B cells in either vehicle (EtOH) or DHT conditions were probed
with antibodies against the AR and the histone modification marks for histone H3
described on the left. Co-localization of the AR (red) and sites with each histone
modification (green) are observed in the merged images
ACH3 EtOH
AR
Marker Nucleus R&G Merge RBG Merge
ACH3 DHT
MonoK4 EtOH
MonoK4 DHT
diK4 EtOH
diK4 DHT
triK4 EtOH
triK4 DHT
triK27 EtOH
triK27 DHT
.
58
5. Co-immunoprecipitation
Cells were seeded at a density of 300,000 per well in six-well plates and grown in
5% CSS-containing phenol red-free medium for 24 h. Cells were transfected when
indicated, and 24 h later they were lysed in a 50 mM Tris-HCl buffer (pH 7.4) containing
150 mM NaCl, 1mM EDTA, 1% Triton X-100, and fresh Protease Inhibitors Cocktail
(1%; Sigma, St. Louis, MO). After homogenization by passing ten times through a 1 cc
microfine insulin syringe, lysates were cleared from insoluble cellular debris by
centrifugation at 14,000 rpm for 5 min in a bench-top microfuge, and 15% of the lysate
solution was set aside to assess input. The remaining lysate was immunoprecipitated
with approximately 3 !g of the specified antibody and 30 µl Protein-G beads (Amersham
Biosciences Freiburg,

Germany), washed three times for 5 min each with the same buffer,
followed by centrifugation at 4,000 rpm for 1 min. Thirty µl of the
immunoprecipitate/bead suspension were mixed with 20 µl of 2.5 x ‘sample buffer’ (50%
Glycerol, 10% SDS, 1M Tris-HCl pH 6.8, 25% b-Mercaptoethanol, 0.5% Bromopenol
blue) and boiled for 5 min. Input and immunoprecipitates were analyzed by Western blot
analysis (Figure 2.10).
59
Figure 2.10 An overview of co-immunoprecipitation.
Figure 2.10 Legend. Cells were lysed and extract was incubated with appropriate
antibody overnight at 4
o
C to immunoprecipitate the protein of interest and any other
proteins complexed to it. Antibody/protein complexes were subsequently pulled down
with protein G agarose beads and washed then eluted with the appropriate buffers. Eluted
proteins were assayed by SDS-PAGE and immunoblotted against a protein different from
the immunoprecipitated protein to assess if those two proteins complexed with each
other.
Lyse Cells Add antibody Add beads
Wash and Elute SDS-PAGE Immunoblot
60
6. Real time RT-PCR
Total RNA was isolated using Aurum Total RNA kit (Bio-Rad, Hercules, CA) following
the manufacturer’s recommendations. One microgram of total RNA was reverse-
transcribed with reverse transcriptase (Invitrogen) and the cDNA subjected to real-time
PCR amplification (RT-qPCR) using IQ SYBR Green (Bio-Rad). PCR primers were
designed using the Primer3 program [(http://frodo.wi.mit.edu/cgi-
bin/primer3/primer3_www.cgi) and are listed in Table 2.2. Copy number was determined
using a standard curve (Figure 2.11A). The primers used gave a single peak during the
melt curve to ensure there was a single product forming (Figure 2.11B).
61
Table 2.2 Summary of Primers Used for RT-qPCR
Gene Forward Primer Reverse Primer
mOC 5’-CGGCCCTGAGTCTGACAAA-3’ 5’-
GCCGGAGTCTGTTCACTACCT
T-3’
Luciferase 5’-CCAGGGATTTCAGTCGATGT-3’ 5-
AATCTCACGCAGGCAGTTCT-
3
mRANKL 5’-CGCTCTGTTCCTGTACCTTCG-3’ 5-AGGCTTGTTTCATCCTG-3’
mMMP9 5’-TACAGGGCCCCTTCCTTACT-3’ 5’-
CTGACGTGGGTTACCTCTGG-
3’
mL10A 5’-CGCCGCAAGTTTCTGGAGAC-3’ 5’-
CTTGCCAGCCTTGTTTAGGC-
3’
mOC
promoter
5’-CTAATTGGGGGTCATGTGCT-3’ 5’-
CCAGCTGAGGCTGAGAGAGA
-3’
mInsulin
promoter
5’-TAGCACCAGGCAAGTGTTTG-3’ 5’-
CTGCTTGCTGATGGTCTCTG-
3’
PSA Intron 5 5’-GGTGCGAGAGGGAAGAAAGG-3 5’-
CACCATCCACAGTTCCTGACT
C-3’
62
Figure 2.11 An example of RTq-PCR.
Figure 2.11 Legend. RT-qPCR samples are shown above. A standard curve was
necessary to ensure that PCR amplification was in a linear range (A). A melt curve was
also performed to ensure that a single peak formed suggesting a single PCR product (B).
A
B
63
7. Chromatin Immunoprecipitation (ChIP)
Cells used for ChIP analyses were plated in 150 mm (diameter) dishes and grown
in the appropriate media to confluency. If it were necessary, cells were supplemented
with 50 !g/ml ascorbic acid and 10 nM ß-glycerophosphate commencing at confluence to
induce differentiation. Two days prior to treatment (4 or 24 hrs) with vehicle or 10 !M
DHT, cells are grown in 5%-CSS phenol red free media.
After treatment, media were removed and cells were treated for 10 minutes at
room temperature with 1% formaldehyde (Sigma, St. Louis, MO). Cells were washed
with PBS containing 0.25mM glycine to stop the cross-linking of protein/DNA
complexes. Cells are subsequently washed with ice-cold PBS and harvested by scraping
with PBS containing 1X prostease inhibitors. Cells were pelleted and resuspended with
SDS Lysis Buffer (1% SDS, 10mM EDTA, 50mM Tris-HCl pH8.1, 2X protease
inhibitors). Cell lysates were sonicated and centrifuged at 14,000g for 10 minutes to
remove any cellular debris. From this 10 !l was stored at -20
o
C for input analysis, and
100ul was used per IP. Each sample for IP was diluted with 900ul of dilution buffer
(0.01% SDS, 1.1% Triton X 100, 1.2 mM EDTA pH 8.1, 16.7 mM Tris-HCl pH 8.1, 167
mM NaCl, 1X protease inhibitors). To each sample, 40 !l of Protein G sepharose bead
slurry was added and incubated with rotation at 4
o
C for preclearing of any non-specific
proteins that bind to Protein-G sepharose beads. Beads were pelleted and supernatant was
used for IP with 3 !g Runx2 antibody (Santa Cruz Biotechnology, Inc., Santa Cruz, CA,
SC-10758, M70) or 3 !g rabbit IgG (Santa Cruz SC-2027). Samples were
immunoprecipitated overnight for at 4
o
C on a rotator.
64
Subsequently, 40 !l of Protein G slurry was added and incubated for 1hour at 4
o
C
on a rotator. Beads were pelleted and supernatant was removed. The beads were washed
on a 4
o
C rotator for 5 minutes with 1ml of the following buffers: Low Salt Buffer (0.1%
SDS, 1% Troton X 100, 2mM EDTA, 20 mM Tris-HCl, 150 mM NaCl), High Salt Buffer
(0.1% SDS, 1% Troton X 100, 2mM EDTA, 20 mM Tris-HCl, 500 mM NaCl), LiCl
Buffer (0.25 M LiCl, 1% Igepal CA-630, 1% Deoxycholate, 1 mM EDTA, 10mM Tris-
HCl), two times with TE (10 mM Tris-HCl, 1mM EDTA). After the wash-incubations the
beads were eluted twice at 15 minutes each with 250 !l of elution buffer (1% SDS, 0.1M
NaHCO
3
) and combined. At this step the input samples were also resuspended with 500
!l of elution buffer. To each elution 20 !l of 5M NaCl was added and then incubated at
65
o
C overnight to reverse cross-linking. Next 10ul 0.5M EDTA, 20 !l 1M Tris-HCl pH
6.5, and 2 !l of 10mg/ml Proteinase K were added. Samples were incubated at 45
o
C for 1
hour. DNA was recovered by phenol/Chloroform extraction and ethanol precipitated
overnight using 2 !l of tRNA as a carrier. DNA was pelleted, washed with 70% ethanol
and resuspended in 100 !l of water. Subsequently RT-qPCR was used to assess if the site
of interest was chromatin immunoprecipitated. All RT-qPCR data obtained were
corrected by input.
A summary of the steps involved in ChIP is presented in Figure 2.12. To assess
which cell lines as well as which Runx2 antibody was best for ChIP analyses, several
control experiments were done to determine the best system to analyze Runx2 occupancy
(Figure 2.13). It was determined that using the M70 antibody with MC3T3-E1 cells gave
the best results when compared to a non-specific site (Figure 2.13a).
65
Figure 2.12 An overview of ChIP for Runx2
Figure 2.12 Legend. An overview of ChIP for Runx2. Cell were grown to confluency and
allowed to differentiate if necessary. Cells were lysed after protein-DNA cross-linking
and chromatin was extracted. Extracted chromatin was sonicated until the samples were
completely homogenized and the liquid was clear. Sonicated chromatin was
immunoprecipitated with Runx2 antibody. Protein-DNA complexes were reversed and
DNA was isolated. Isolated DNA was probed by PCR for specific promoter regions
reported to bind Runx2.
66
Figure 2.13 Testing Runx2 antibodies for ChIP
Figure 2.13 Legend. MC3T3-E1, PC3, and SAOS-2 cells were chromatin
immunoprecipitated for Runx2 and assayed for specific Runx2 sites and a non-specific
site. Antibodies used for ChiP were C19, S19, and M70 Runx2 antibodies obtained from
Santa Cruz Biotechnologies. MC3T3-E1 cells where assayed ChIP at a specific
osteocalcin promoter region and non-specific insulin region, the M70 and S19 antibody
performed equally well in showing occupancy, while the C19 did not show any occupany
(A). Similar results were obtained from PC3 and SAOS-2 cells when assessing
occupancy upstream of the IBSP gene (Runx2 specific site) and a non-specific site (PSA
intron 5) (B,C).
67
MC3T3E1
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
Insulin OC
C19
MC3T3E1
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
Insulin OC
S19
MC3T3E1
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
Insulin OC
M70
SAOS-2
0
0.05
0.1
0.15
0.2
0.25
PSA Intron 5 IBSP 10kb
upstream
M70
SAOS-2
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
PSA Intron 5 IBSP 10kb
upstream
C19
SAOS-2
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
PSA Intron 5 IBSP 10kb
upstream
S19
PC3
0
0.05
0.1
0.15
0.2
0.25
PSA Intron 5 IBSP 10kb
upstream
M70
PC3
0
0.05
0.1
0.15
0.2
0.25
PSA Intron 5 IBSP 10kb
upstream
C19
PC3
0
0.1
0.2
0.3
0.4
0.5
0.6
PSA Intron 5 IBSP 10kb
upstream
S19
A
B
C
68
8. Flourescent Recovery After Photobleaching (FRAP)
Cos7 cells were plated at 40,000 cells in an 8 well Lab-Tek Chambered
coverglass with cover (each well is 15 mm in diameter). Cells were grown in phenol red
free DMEM-5% CSS for 24 hrs prior to transfection. Cos7 cells were transfected with the
appropriate GFP construct and co-expression of any other proteins as (described in each
respective experiment) similar to the transfection procedure described above, only scaled
to cover the surface volume of each chambered well. After transfection, cells were grown
in phenol red free DMEM-5% CSS for 24 hrs. One hour prior to the FRAP experiment
the media was removed and substituted with DMEM-5%CSS containing the appropriate
steroid ligand.
Cells were viewed using a LSM 510 Zeiss confocal microscope at 60X
magnification. The cells were observed at a wavelength of 488 nm at 4% intensity for 15
seconds, prior to bleaching for 5 seconds with an 80% intensity. Subsequently images
were captured at each second for one to three minutes at 488nm (4% intensity) to assess
the recovery in photobleaching. All data obtained was normalized to the average starting
intensity and data was corrected if there was any loss due to background photobleaching
that occurred when the confocal image was captured. An example of FRAP is given in
Figure 2.14. A region of the GFP transfected cell is photobleached and allowed to recover
overtime (Figure 2.14A). The intestiny of the floursecence is measured and plotted
relative to the pre-bleaching intensity and controlled for background loss in intensity
(Figure 2.15B).
69
Figure 2.14 An example of a FRAP experiement.
Figure 2.14 Legend. Cos7 cells were transfected with a GFP-cosntruct of interest and
bleached for 5 seconds. A circular region was bleached (red circle in A) and the region
recovered fluorescence overtime. That recovery was measured throughout time (B). The
recovery reached a plateau lower than the original intensity designating two fractions, a
mobile fraction that was able to move back to the bleached location and recover intensity,
and an immobile fraction that was trapped in that region and was not able to recover in
fluorescent intensity. All intensity was normalized to the original intensity and subtracted
for any background loss due to capturing the images. An LSM 510 Zeiss confocal
microscope at 60X magnification was used to capture all images.
GFP-Construct
0
0.2
0.4
0.6
0.8
1
1.2
0 20 40 60 80 100 120 140 160
A
B
Mobile Fraction
Bleach
Time (s)
70
9. Microarray Data Mining
Normalized expression data was obtained from Oncomine. Runx2 target genes
from the literature were subjected to an unsupervised 2-dimensional hierarchical cluster
analysis using the software Jmp V 5.0 (http://www.jmp.com). ER! target genes were
mapped using Heatmap V 1.0 (http://quertermous.stanford.edu/heatmap.htm) to the
corresponding Runx2 cluster. A similar procedure was used to analyze Runx2 target
genes in prostate cancer samples. An example of 2-D cluster analysis is presented in
Figure 2.15.
71
Figure 2.15 An example of a 2D unsupervised cluster.
Figure 2.15 Legend. Gene expression data of samples were analyzed using a JMP 5.0
program in an unsupervised cluster. The Y axis represents that different samples and the
X axes represents different genes. Clustering was done in a unsupervised manner using
the default settings two dimensionally. Thereby, genes that expressed in a similar manner
were grouped together along the X axis, and samples that had similar overall expression
of the different genes were grouped together on the Y axis. Cladograms on the X and Y
axis delineate which groups of samples or genes respectively were similar. In this
example there were two main branches of samples as well as genes, showing two patches
of red (higher expressed genes) and green (lower expressed genes).
72
Chapter 3: Modulation of Runx2 by ER!
1. Introduction
The mammalian runt-related gene family encodes three transcription factors that
play pivotal roles in lineage-specific cell growth and differentiation in a variety of cell
types. Whereas Runx1 is implicated in definitive hematopoiesis and Runx3 in gut and
nervous system development (19), Runx2 is a master transcription factor controlling
osteoblast differentiation (124). Runx2-null mice lack osteoblasts and fail to form bone
(61, 102). Low bone mass (LBM) has been observed in Runx2 heterozygous mice (121),
mice missing one of two Runx2 isoforms (157), as well as transgenic mice whose
osteoblasts express a dominant negative (DN) form of Runx2 under the control of the
osteocalcin (OC) promoter (25). Runx2 has also been shown to act downstream of bone
anabolic agents, including BMPs, PTH and estrogens (67, 79, 81). Surprisingly,
however, transgenic mice, whose osteoblasts over-express Runx2 under the control of the
!1(I) collagen promoter also have LBM leading to spontaneous fractures (35, 76);
conversely, transgenic mice expressing a Runx2-DN under the control of the same
promoter have high bone mass and are protected against ovariectomy-induced bone loss
(79). Conceivably, therefore, pro-skeletal agents may inhibit Runx2 under some
conditions in order to prevent bone loss.
The runt-related proteins also play important roles in cancer. Runx proteins act in
many systems as tumor suppressors since they positively promote cell differentiation
(109). However, they may also have oncogenic potential in other settings (46).
73
Threrefore, dependent on context, they can function as either tumor suppressors or
promoters (8).These dual potential roles of Runx proteins (13) have confounded the field
for many years and remain intense topics of investigation (45-47). Runx2 deficiency has
been recently shown to promote immortalization and tumorigenesis (166). This is
consistent with earlier studies of Runx1 and Runx3, in which inactivation of the former
(via chromosomal translocation) and epigenetic silencing of the latter (via DNA
methylation) were found associated with hematologic and gastric cancers, respectively
(46). On the other hand, Runx2 has been implicated in cancer progression (8), including
breast cancer bone metastatasis (2). Possibly, transient inactivation of Runx2 promotes
tumorigenesis, and its subsequent reactivation supports the metastatic phenotype.
Like Runx2, estrogens play critical roles in bone metabolism and carcinogenesis.
Loss of estrogen function, at menopause or during anti-estrogen therapy for the
management of hormone-driven malignancy, increases the risk for osteoporotic fractures,
cardiovascular disease and dementia (74, 82, 158). On the other hand, estrogens have
been associated with carcinogenesis (106), as hormone replacement therapy for
postmenopausal women increases the risk for breast, endometrial and ovarian cancer
(105). ER! is also known to have pro-proliferative effects on breast epithelial cells and is
a well-established risk factor for breast cancer (27). On the other hand, ER!-negative
breast tumors have a very poor prognosis, implying a tumor suppressive effect of the
transcription factor in certain settings (116). Just as is the case for Runx proteins in the
other systems referred to above, appropriate regulation of ER! seems to be required for
74
the maintenance of normal breast epithelial cell growth and differentiation. These
mechanisms of estrogens’ carcinogenic activity are still not fully understood.
The pro-skeletal property of estrogens is multifaceted. Estrogens increase cell
proliferation and survival in the osteoblast lineage, resulting in a bone anabolic effect (78,
115, 139). Most importantly, however, estrogens attenuate bone resorption and turnover,
and this occurs via at least two mechanisms. First, estrogens directly promote osteoclast
apoptosis (139), at least in part by stimulating FASL expression (91). Second, estrogens
limit osteoblastogenesis from precursor cells, thereby attenuating bone turnover (23), a
process that is skewed away from formation and towards resorption in most adults. Upon
estrogen loss, increased osteoblast numbers fuel excessive bone turnover (78); and, in the
absence of estrogens, each of these osteoblasts expresses higher levels of
osteoclastogenic factors (139). Molecular mechanisms remain to be elucidated, by which
estrogens attenuate osteoblastogenesis and osteoblast-driven osteoclastogensis. However,
the effects of ER! on the two processes alluded to above (cancer progression and bone
density regulation) may even merge since cancer metastases from breast cancer have a
predilection for bone (51), which seems to provide a ‘fertile soil’ where the cellular
milieu allows cancerous cells to thrive.
These important physiological and pathological processes highlight a possible
functional link between Runx2 and ER! activities. Specifically, that one of the main
‘protective’ effects of ER! on bone density resides in the inhibition of Runx2 activity,
whereas strong ER! activity during breast cancer progression (including breast cancer
75
bone metastases) may be due to the prevention of the tumor suppressor activities of
Runx2. However, during later ER!-negative stages of breast cancer progression, the
unopposed and inappropriate oncogenic potential of Runx2 overactivity emerges and
contributes to the known poor progonosis of such tumors. In summary, both Runx2 and
ER! have been shown to have functions in bone, and both have been implicated in
cancer. In this study there was evidence that liganded ER!, via the physical interaction
with Runx2, modulated the activity of Runx2. A novel observation was made, that
estrogens strongly inhibited Runx2 in COS7 cells, breast cancer cells and late-stage
MC3T3-E1 osteoblast cultures. This inhibitory activity was dissected in molecular terms,
and will be discussed as to how it may contribute to two well-established activities of
estrogens – the restraining of bone turnover and the promotion of breast carcinogenesis.
76
2. Results
2.1 ER!, not ER", inhibits Runx2 activity in a ligand-specific manner.
Initially the influence of estradiol (E2) was investigated and its receptors ER! and
ER" on Runx2 after transfecting COS7 cells with the respective expression vectors along
with the Runx2 reporter 6XOSE2-luc (24). As shown in Figure 3.1A, E2 inhibited the
activity of Runx2 by 76% in the presence of ER!. The inhibition ranged from 43% to
78% in eight independent experiments. E2 did not inhibit Runx2 activity in the presence
of ER", although, as shown in Figure 3.1C, the latter was almost as potent as ER! in
stimulating a luciferase reporter driven by classical estrogen response elements (ERE).
In the absence of ligand, both ER! and ER" inhibited Runx2 only minimally (Figure
3.1A). The strong inhibitory activity of ER! was elicited by E2, but not by
dihydrotestosterone or dexamethasone (Figure 3.1A). Similar to human Runx2 (Figure
3.1), the mouse homolog was also inhibited by E2-bound ER! (Figure 3.2).
Interestingly, E2-activated ERa only slightly inhibited transcription from 6XOSE2-luc
when the reporter was stimulated by Runx1 rather than Runx2 (Figure 3.1B). These
results indicate a strong and specific inhibitory mechanism, whereby E2-bound ER!
suppresses Runx2’s transactivation activity. In a complementary experiment, Runx2 did
not affect ER!-mediated activation of an ERE-containing reporter (Figure 3.3).
77
Figure 3.1 ER! inhibits human Runx2
Figure 3.1 Legend. (A,B) COS7 cells were transiently transfected with the Runx reporter
6XOSE2-luc (firefly) along with expression vectors encoding human Runx2 (A), Runx1
(B), ER!, and/or ER" as indicated. Cells were treated with DHT, E2, Dex (10 nM each),
or ethanol vehicle (0.01%) for 24 h and subjected to luciferase assay as describe in the
Materials and Methods section. (C) COS7 cells were transiently transfected with an ER
reporter (ERE-luc) along with an expression vector encoding either ER! or ER", and
treated for 24-hr with E2, followed by the luciferase assay. In each experiment, the
firefly luciferase results were corrected for the expression of a co-transfected CMV-
renilla luciferase construct (internal control), except for Panel A, where the renilla
luciferase values are shown in the inset graph. The immunoblot in A shows that Runx2
expression was not inhibited in the presence of ER! and its ligand. All data are presented
as mean values ± SEM, with n=4 dish replicates of a representative experiment, repeated
at least 3 times.
78
Figure 3.2 ER! inhibits mouse Runx2.
Figure 3.2 Legend. The effects of ER! and ER" on 6XOSE2-luc were determined in the
presence of mouse Runx2. The experimental design is the same as for human Runx2
(Figure 3.1A).
79
Figure 3.3 Runx2 does not influence ER!’s transcriptional activation activity.
Figure 3.3 Legend. The classical transcriptional activation activity of ER! was assessed
as in Figure 1C in the presence and absence of Runx2, as indicated.
80
2.2 Functional mapping of ER! domains responsible for Runx2 repression indicates
independence of ER!’s transactivation activity.
To functionally map the domain(s) of ER! responsible for Runx2 repression, and
to test whether repression was dependent on ER-mediated stimulation of ERE-containing
promoters, the influence of the ER! deletion constructs was measured (illustrated in
Figure 3.4A) on Runx2 activity. Stimulation of ERE-luc was assessed in parallel
experiments. Most importantly, deletion of the ER!-LBD, which was associated with
near complete loss of ERE-luc activity in COS7 cells (Figure 3.4D), resulted in
constitutive full repression of Runx2 activity (Figure 3.4C). Further suggesting
dissociation between ER!’s transcriptional activation activity from Runx2 repression,
each of the ER!-NTD, -DBD and -LBD inhibited Runx2-mediated activation of 6XOSE-
luc (Figure 3.4E) while leaving ERE-luc (as well as CMV-renilla-luc and tk-renilla-luc)
essentially unaffected (Figure 3.4F). The ER! DBD-LBD also fully repressed Runx2
activity (Figure 3.4C); and, just like full-length ER!, repression by the DBD-LBD was
ligand dependent. These results suggest that ER! represses Runx2 without activating
ERE-containing target genes, and that the repression mechanism employs multiple
domains of ER!.
81
Figure 3.4 Functional mapping of ER! substructures and the dissociation between
ER!-mediated transcriptional activation and Runx2 repression.
Figure 3.4 Legend. (A) Schematic illustration showing full-length and ER! fragments
that were transiently expressed in COS7 cells. The NTD, DBD and LBD fragments were
FLAG-tagged to facilitate immunoblot detection shown in the inset of F. (B)
Immunoblot analysis of cells transfected with the indicated constructs using antibodies
against ER! NTD (sc-7207, left blot) or ER! LBD (sc-787, right blot). (C-F) COS7
cells were transiently transfected with either the 6XOSE2-luc (C,E) or the ERE-luc
reporter (D,F), along with the indicated expression plasmids. Luciferase activity was
measured 24 h after treatment with either ethanol (0.01%) or E2 (10 nM). All data are
presented as mean values ± SEM, with n=4 dish replicates of a representative experiment,
repeated at least 3 times.
82
83
2.3 ER! physically interacts with Runx2 in a ligand-dependent manner.
Because repression of Runx2 is independent of ER!’s transactivation activity, it
could occur via physical interaction between the two proteins. Indeed, ER! was present
in Runx2 immunoprecipitates; the ER!/Runx2 interaction was weak in untreated cells,
and increased dramatically upon E2 treatment (Figure 3.5A). A series of ER! deletion
constructs, including the DBD-LBD, NTD-DBD, as well as the individual domains, were
also employed in the co-immunoprecipitation (co-IP) assay, and each of them was
associated with Runx2 (Figure 3.5A). However, unlike the full-length receptor, the
receptor fragments tested did not require E2 to form complexes with Runx2 (Figure
3.5A). These data suggest that ER! interacts with Runx2 through multiple surfaces, and
that these surfaces are exposed only after the full-length protein has assumed a
permissive ligand-induced conformation.
To test if the physical interactions demonstrated by co-IP are direct, we asked
whether immobilized ER! domains (Figure 3.5C) could pull down in vitro transcribed
and translated Runx2. In these GST pull-down assays, the ER!-DBD strongly interacted
with Runx2, and this strong interaction was recapitulated with Runx2’s PST domain
(alone or with the Runt domain), but not with the QA-Runt or Runt domains (Figure
3.5D). The PST domain continued to interact with ER!’s DBD after deletion of the 515-
596, but not the 417-596 amino acid sequence (Figure 3.5D). Positive pull-down was
also observed after deletion of the PST’s N-terminal 303-407 amino acids sequence,
thereby mapping the interaction surface to Runx2’s amino acids 417-514 (Figure 3.5E),
84
which contain activation domain 3 (AD3) and the nuclear matrix targeting sequences
(NMTS). In contrast to ERa’s DBD, the NTD did not interact with full-length or
fragments of Runx2 (Figure 3.5D). The ER!-LBD displayed a weak interaction with
Runx2, which, like the ER!-DBD, mapped to Runx2’s PST domain (Figure 3.5D). This
interaction was not influenced by physiological estrogen concentrations (Figure 3.5D),
but was increased in the presence of 1 !M E2 (Figure 3.6). Thus, ligand-dependent
ERa/Runx2 interaction is mediated primarily by a contact between Runx2’s amino acids
417-514 within the PST and ER! ‘s-DBD, a potential secondary direct contact of Runx2
with the ER!-LBD (via the PST domain), and an indirect contact with the NTD.
Confocal fluorescence microscopy was then performed to test whether ER! co-
localized with Runx2 in living cells (Figure 3.7). Both transcription factors were
concentrated in the nucleus regardless of E2 treatment, and occupied distinct domains.
Quantitative analysis of 10 randomly selected untreated cells indicated 34% co-
localization (Figure 4D), where ER! and Runx2 could interact in vivo. Upon treatment
with E2, there was a decrease in areas occupied by either protein alone, with a
concomitant increase in areas containing both proteins together (Figure 3.7A-C).
Quantitative analysis of 10 randomly selected E2-treated cells indicated a highly
significant (p<8.3X10
-8
) 2.1-fold increase in co-localization compared to the 10 randomly
selected untreated cells (Figure 3.7D), likely reflecting the E2-induced physical
interaction between the two transcription factors (see Figure 3.6).
85
Figure 3.5 Interaction between ER! and Runx2 domain in Co-IP and GST pull-
down assays.
Figure 3.5 Legend. (A) COS7 cells were transiently transfected with plasmids encoding
each of the specified ER! fragments and Runx2, followed by 24-hr treatment with either
ethanol or 10 nM E2. The ER! and its fragments were detected in either whole cell
extracts as input, or IgG or Runx2 immunoprecipitates. (B) Schematic diagram of full-
length (FL) Runx2 and fragments transcribed and translated in-vitro. The scheme at the
top depicts the three Runx2 domains. Boxes with 1, 2, 3 or N mark the positions of the
respective activation domains and the NMTS. The thick line above the PST domain
represents the surface interacting with ER!. (C) Coomasie stained SDS-PAGE of the
bacterially expressed and purified GST or GST fusion proteins used as baits in the pull-
down assays. (D, E) A mixture of the indicated radiolabeled Runx2 fragments were
incubated with the depicted GST-fusion proteins used as baits. The positive control
GST-CBFß was used as a bait for the Runt domain. The autoradiograph shows the
fragments pulled down by the indicated baits. Parts C-E of this experiment were done by
Dr S Baniwal in collaboration.
86
Figure 3.6 Direct interaction of ER!-LBD with Runx2 in superphysiological
conditions.
Figure 3.6 Legend. (A) The indicated domains of ER! were produced in BL21DE-3
E.Coli as GST fusion proteins and used to pull down
35
S-labeled Runx2 transcribed and
translated in reticulocyte lysates. The GST-LBD was also tested in the presence of 1 !M
E2 as indicated. (B) SDS-PAGE and Coomassie blue staining of the GST and GST fusion
proteins used in Panel B. This experiment was done by Dr S Baniwal and Daniel J.
Purcell in collaboration.
87
Figure 3.7 Immunofluorescence of ER! and Runx2.
Figure 3.7 Legend. COS7 cells were transiently transfected with ER! and Runx2, and
treated for 24 h with ethanol or E2 (10 nM). (A-C) ER! (red) and Runx2 (green) were
visualized using confocal microscopy as described in Materials and Methods. Co-
localization (yellow in A) is demonstrated by surface plots (B) and by Red/Green profiles
(C). (D)Ten cells were randomly selected from each of a set of four untreated and four
treated cultures, and co-localization was quantified and plotted as Mean ± SEM.
*p=8.3X10
-8
.
88
89
2.4 Inhibition of Runx2 by estradiol in late stage MC3T3-E1 osteoblast cultures.
To address the ER!-Runx2 interaction in osteoblasts, it was confirmed by co-IP
assay the physical association between endogenous ER! and endogenous Runx2 in
MC3T3-E1 osteoblasts (Figure 3.8A). The influence of E2 on Runx2 was then examined
during the development of the osteoblast phenotype in cultures of MC3T3-E1 cells that
had been stably transfected with the 6XOSE2-luc Runx2 reporter construct (24). We
measured the expression of the mRNAs for OC, a classical Runx2 target (1, 26); MMP9,
also a Runx2 target (108); RANKL, which may be regulated by Runx2 (32); and
luciferase (driven by the six Runx2 binding sites). Whereas OC and MMP9 were at best
weakly stimulated on day 4, all mRNAs were suppressed on day 18 (Figure 3.8B-E).
Thus, while in early cultures Runx2 may be stimulated by E2, the results demonstrate E2-
mediaed repression in late MC3T3-E1 cultures, which is similar to the inhibition
observed in COS7 cells (Figure 3.1) and breast cancer cells (see below). Although E2
did not affect mineralization in the MC3T3-E1 cultures (data not shown), inhibition of
RANKL and MMP9 may attenuate their osteoclastogenic activity.
90
Figure 3.8 Developmental stage-specific inhibition of Runx2 by E2 in osteoblasts.
Figure 3.8 Legend. (A) MC3T3-E1 cells were treated with ethanol or E2 (10 nM) and the
presence of ER! in Runx2 immuno-complexes was examined as described in Figure 3A.
(B-F) MC3T3-E1 cells stably transfected with the 6XOSE2-luc Runx2 reporter were
subjected to differentiation conditions and treated with ethanol (white bars) or E2 (10
nM; black bars) commencing at confluence. Levels of the indicated mRNAs were
measured on days 4, 11 and 18 by RT-qPCR as described in Material and Methods. Data
were corrected for the expression of ribosomal protein L10A, which itself did not
significantly change during culture progression or in response to E2. RANKL was not
expressed on day 4, and the low expression of osteocalcin and luciferase on this day is
shown in the respective insets (Mean ± SD; n=3). Parts B-E of this experiment were done
by Dr S Baniwal and Dr LeClerc in collaboration.
91
92
2.5 Synthetic ER ligands inhibit or stimulate Runx2 in a compound- and cell type-
specific manner.
Selective estrogen receptor modulators (SERMs) are widely used for the
management of breast cancer (52). A variety of SERMs are available, with different
levels of partial agonism or antagonism with respect to the native ligand. Different
SERMs induce in the receptor a variety of conformational changes, reflected for example,
by increased sensitivity of helix 12 to trypsin digestion in the presence of ICI compounds
as compared to E2, and even higher sensitivity in the presence of OHT (141). We first
asked whether these SERMs mimicked E2 in inducing interaction with Runx2. As
demonstrated by co-IP assays, each of OHT and ICI 182780 strongly induced the
formation of ER! /Runx2 complexes (Figure 3.9A). Remarkably, however, the
functional consequences with regard to Runx2 activity were different compared to E2.
While OHT slightly inhibited Runx2 activity, ICI 182780 had the opposite effect,
stimulating Runx2 activity by 3.1-fold (Figure 3.9B). This contrasted with the activity of
these SERMs on an ERE-containing template, where ICI 182780 and OHT had weak and
stronger partial agonist effects, respectively (Figure 3.9C). Interestingly, OHT also had a
stimulatory effect on Runx2 when bound to the ER! DBD-LBD (Figure 3.9B), raising
the possibility that different domains of OHT-bound ER! have opposing effects on
Runx2.
We subsequently examined the effects of SERMs on Runx2 in breast cancer cells.
When ERa was expressed in the ER-negative MDA-MB-231 cell line, E2 strongly
93
inhibited the co-transfected 6XOSE2 Runx2 reporter (Figure 4.9D). Unlike COS7 cells,
OHT inhibited Runx2 in MDA-MB-231 cells as strongly as E2 (Figure 3.9D). ICI
182780, which stimulated Runx2 in COS7 cells (Figure 3.9B), had a slight if any
inhibitory effect on Runx2 in MDA-MB-231 cells (Figure 3.9D). In T47D and MCF7
beast cancer cells, ICI 182780 slightly stimulated Runx2 (Figure 3.9E-F), whereas the
effect of OHT was uniformly inhibitory across all three breast cancer cell lines (Figure
3.9D-F). Endogenous Runx2 gene expression in MCF7 cells was inhibited by E2 and
slightly upregulated by ICI 182780 as well (Figure 3.10). There was a similar effect
examining endogenous Runx2 target genes in MCF7 cells with inhibition of Runx2 target
gene expression upon E2 treatment, and slight activation with ICI 182780 treatment
(Figure 3.11). Thus, OHT mimics the strong E2-mediated inhibition of Runx2
specifically in breast cancer cells; and ICI 182780, which stimulates Runx2 in COS7
cells, only mildly regulates Runx2 activity in breast cancer cells, either upwards or
downwards. The various effects of SERM-bound ER! on Runx2 activity may be
relevant to their selective effects in different tissues in vivo.
94
Figure 3.9 Various effects of SERM-bound ER! on Runx2.
Figure 3.9 Legend. (A) Co-IP assays were performed as in Figure 3.5A after treatment of
COS7 cells with 100 nM OHT or 100 nM ICI 182780. (B,C) COS7 cells were
transfected with Runx2 and its 6XOSE2-luc reporter (B), or with ERE-luc (C) along with
expression vectors coding for the indicated ER isoforms or fragments. Cells were treated
for 24 h with ethanol, E2 (10 nM), OHT (100 nM), ICI 182780 (100 nM), or
combinations thereof, and then subjected to luciferase assays. (D-F) Three breast cancer
cell lines, MDA-MB-231 (D), T47D (E) and MCF7 (F) were transfected with the
6XOSE2-luc (left) or the ERE-luc (right) reporter, along with expression vectors for ER!
(D), Runx2 (E), or empty vector control, and treated with 10 nM E2, 100 nM OHT or 100
nM ICI 182780 as indicated. All data are presented as mean values ± SEM, with n=4
dish replicates of a representative experiment, repeated at least 3 times.
95
96
Figure 3.10 Modulation of Runx2 target genes in MCF7 cells by SERMS.
Figure 3.10 Legend. Inhibition of Runx2 target genes IBSP, TCF7, and MMP9 upon
10nM E2 treatment for 24hrs in MCF7 cells. Upon 24hrs of 100nM ICI 182780 treatment
there is no inhibition or slight activation in MCF7 cells.
0
1
TCF7
Gene Expression in MCF7 Cells
0
1
IBSP
0
1
MMP9
EtOH
E2
ICI 182780
97
2.6 Negative correlation between ERa and Runx2 target genes in breast cancer
tissues.
If ER! inhibits Runx2 in breast cancer, then one would expect to see an inverse
relationship between the expression of ER! and Runx2 target genes in breast cancer
biopsies. Therefore, three datasets were mined from Oncomine (112) for the expression
of 40 known Runx2 target genes. The datasets represented comprehensive gene
expression analyses of three cohorts of 286, 295, and 198 breast cancer biopsies (22, 147,
152). The 40 Runx2 target genes, listed in Table 3.1 were the top hits from each of three
unbiased studies designed to discover Runx2 targets (39, 146, 163). Remarkably,
expression of each of the Runx2 target genes tested was negatively correlated with the
expression of ER! across each of the three cohorts, and as a positive control we used
known ER target genes Table 3.1, which gave us an expected positive correlation to ER.
For example, in the series of 286 biopsies, the correlation coefficient between ER!
expression and that of the Runx2 target gene MCM5 (163) was -0.58 (Figure 3.11A).
This correlation was close to the positive correlation between the expression of ER! and
its classical target pS2 (R=0.69; Figure 3.11B). Similar observations were made in the
other two series (Figures 3.12 A,B and 3.13 A,B, respectively). Meta-analysis of the
three studies altogether revealed a statistically significant negative correlation between
ER! expression and each of the 40 Runx2 target genes (Figure 3.11C). The average
correlation coefficient was -0.29, with a standard error of 0.05 (n=41), compared with
0.59 ± 0.05 (n=4) for classical ERa targets (Figure 3.11C). A significant negative
98
correlation (p=0.03) was also observed between the expression of ER! and OC (not
shown). To visually demonstrate the correlation between the expression of ER! and
Runx2 target genes, the 41 Runx2 target genes were subjected to unsupervised clustering
and examined the expression of ER! and ER! target genes in each cluster. In each of the
three studies, the observed clusters had an inverse relationship between genes regulated
by Runx2 versus ER! and its targets. For example, the cohort of 286 biopsies was
clustered into two main branches, one with low expression of Runx2 target genes and
high expression of ER! and its targets (Figure 3.11D, cluster 1) and the other displaying
a mirror image (Figure 3.11D, cluster 2). Similar relationships were observed with the
other two cohorts (Figures 3.12C and 3.13C). The negative correlation between
expression of ER! and Runx2 target genes was consistent with the idea that inhibition of
Runx2 by ER! occurs not only in breast cancer cell lines (Figure 6) but also in human
breast epithelial cells in vivo.
99
Table 3.1 Runx2 and ER! Target Genes
Gene Ascesion # References
ABCF1 NM_001090 (39, 146)
ACAT2 NM_005891 (146)
AGC1 NM_001135 (39, 146)
ALPL NM_000478 (39, 146)
BGLAP NM_199173 (24)
CCNB2 NM_004701 (163)
CDC6 NM_001254 (163)
CNN2 NM_201277 (39)
Col9a1 NM_001851 (39)
Col9A3 NM_001853 (39)
DKK1 NM_012242 (39, 146)
Dlx5 NM_005221 (39, 146)
GTPbp2 NM_019096 (39, 146)
HCK NM_002110 (163)
HLA-E NM_005516 (163)
HSP105 NM_006644 (39, 146)
IBSP NM_004967 (39, 146)
MATN4 NM_003833 (39, 146)
MCM5 NM_006739 (163)
MMP13 NM_002427 (39, 146)
MMP9 NM_004994 (39, 146)
MX1 NM_002462 (146)
PCOLCE2 NM_013363 (39, 146)
Pim1 NM_002648 (146)
PLTP NM_006227 (39, 146)
Rpl39l NM_052969 (39, 146)
RSAD2 NM_080657 (146)
SLC16A3 NM_004207 (146)
Slc2a1 NM_006516 (39, 146)
SOX9 NM_000346 (39, 146)
SPP1 NM_000582 (39, 146)
TAGLN2 BC009357 (146)
TCF7 NM_201632 (39, 146)
100
Table 3.1 Continued
VDR NM_000376 (39, 146)
VEGFA NM_003376 (163)
VIM EF445046 (146)
WWP2 NM_007014 (39, 146)
USP18 NM_017414 (146)
TGFBI NM_000358 (39, 146, 163)
TRAM2 NM_012288 (39)
TUBB4 NM_006087 (146)
ER! NM_000125 (72)
PR NM_000926 (72)
PS2 NM_003225 (72)
GREB1 NM_033090 (72)
NRIP1 NM_003489 (72)
101
Figure 3.11 Meta-analysis of the correlation between expression of ER! and Runx2
target genes in breast cancer biopsies.
Figure 3.11 Legend. (A, B) Scatter plots of the expression of MCM5 (A) or pS2 (B)
versus ER! in 286 beast cancer biopsies analyzed previously (152). (C) Correlation
coefficients between the expression of ER! and each of 41 Runx2 target genes based on
meta-analysis of 779 breast cancer biopsies described in three published databases (39,
146, 163). Each of the correlation coefficients was significant with p < 10
-4
. (D)
Expression levels of the Runx2 target genes in one cohort of 286 breast cancer biopsies
(152) was subjected to an unsupervised cluster analysis, resulting in two major branches
of tumor samples designated in the heatmap as 1 and 2. The expression levels of ER!
and 4 ER! target genes in each of the 286 biopsies are represented as a heatmap on the
right.
102
Figure 3.12 Correlation between ER! and Runx2 target genes in BCa biopsies
Figure 3.12 Legend. (A, B) Scatter plots of the expression of MCM5 (A) or pS2 (B)
versus ER! in 295 beast cancer biopsies analyzed (147). (C) Expression levels of the
Runx2 target genes in the 295 breast cancer biopsies was subjected to an unsupervised
cluster analysis, resulting in two major branches designated in the heatmap as 1 and 2.
The expression levels of ER! and 4 ER! target genes in each of the 295 biopsies are
presented on the right.
103
Figure 3.13 Correlation between ER! and Runx2 target genes in BCa biopsies
Figure 3.13 Legend. (A, B) Scatter plots of the expression of MCM5 (A) or pS2 (B)
versus ER! in 198 beast cancer biopsies analyzed (22). (C) Expression levels of the
Runx2 target genes in the 198 breast cancer biopsies was subjected to an unsupervised
cluster analysis, resulting in two major branches designated in the heatmap as 1 and 2.
The expression levels of ER! and 4 ER! target genes in each of the 198 biopsies are
presented on the right
104
2.7 Understanding the mechanisms of interaction between Runx2 and ER!.
To understand the mechanism of interaction, immunoprecipitates of Runx2 and
ER! were subjected to treatments of Mnase (1 ug) or EtBR (5 ul) to disrupt any DNA
present in the immunoprecipitates (Figure 3.14A). The disruption of DNA also caused
Runx2/ ER! interactions to be disturbed. To alter the DNA present in the
immunoprecipitates in another way, UV treatment was done which caused a strong
increase in interaction in the ICI 182780 condition as compared to input (Figure 3.14B).
These data suggest that DNA may be involved during the interaction between Runx2 and
ER!, which was not of great surprise since these are both transcription factors.
105
Figure 3.14 Immunoprecipitation of ER! by Runx2 with Mnase and EtBR.
Figure 3.14 Legend. The presence of Mnase and EtBR as shown above decreased the
interaction between ER! and Runx2 in the presence of 10nM E2 (A). The expression of
Runx2 and ER! was not affected by Mnase and EtBR treatment. UV treatment was
performed during the immunoprecipitation of ER! by Runx2 under vehicle (EtOH), E2
or ICI 182780 conditions (B).
106
DNA
ER!
Runx2
Control
Mnase
EtBR
Control
Mnase
EtBR
Input 15% IP Runx2
ER!
Dimer
UV Treatment
ICI 182780
E2
EtOH
IP Runx2
ICI 182780
E2
EtOH
Input 15%
A
B
107
3. Discussion
Two ERs, ! and ", mediate the various actions of estrogens. In addition to their
classical action on ERE-containing promoters, the ligand-activated receptors form
complexes with, and modulate the activity of other regulatory proteins, including signal
transducers (62) and transcription factors. For example, estrogens have been shown to
activate AP-1 and SP1, resulting in increased expression of cyclin D1 (15). Here it was
shown that ER! interacts with and inhibits Runx2 in osteoblasts and breast cancer cells.
Transfection experiments comparing ER! and ER " in COS7 cells show that the
inhibition is specific for ER!, the same receptor that also mediates the physiological
effects of estrogens in bone (134, 135) and breast (103).
E2 was required for ER!/Runx2 interaction in both the co-IP and the functional
inhibition assays. Ligand was also required for ER interaction with transcriptional co-
activators, which resulted in stimulation of ERE-containing promoters. However, Runx2
repression by ER! did not require classical activation through EREs, because (i) the
NTD-DBD, as well as individual domains of ER! each suppressed Runx2 activity
without activating EREs; (ii) the transcriptionally-competent ER" did not inhibit Runx2;
and (iii) the partial agonist activity of ICI 182780 with regard to classical ERE activation
was associated with stimulation, rather than inhibition of Runx2. Thus, Runx2 repression
is dissociable from ER!’s canonical transcriptional activation activity, and appears to
occur via direct protein-protein interaction. Among possible mechanisms leading to
inhibition of Runx2 by ER!, the two proteins could be present together at OSE2-like
108
sites and form platforms for the further recruitment of co-repressors or co-activators,
depending on the ER! ligand.
According to the co-IP data, each of ER!‘s NTD, DBD and LBD interacts with
Runx2, although strong direct interaction was observed in the GST pull-own assay only
with the DBD. These results are consistent with those of McCarthy et al., who used a
two-hybrid assay to demonstrate that each of the ER! domains, and mostly the DBD,
interacted with Runx2 (81). In this study, however, the reciprocal two-hybrid analysis
mapped the interacting surfaces in Runx2 to both the QA and the PST domains, whereas
our GST pull-down assay indicated interaction only at the PST domain. Be that as it
may, the fine mapping of the interaction surface to Runx2’s amino acids 417-514 within
the PST domain raises the interesting possibility that ER! inhibited Runx2 by masking
its C-terminal activation domain and/or its nuclear matrix targeting sequence (124).
The suggested inhibition of Runx2 by ER! in breast cancer cells in vivo could be
either pro- or anti-oncogenic depending on the presiding function of Runx2. Initially,
Runx2 likely plays a tumor suppressive role, and thus the E2-mediated inhibition at this
stage would constitute a mechanism of hormonal carcinogenesis. However, Runx2
activity in later stages of cancer progression may promote expression of the metastatic
phenotype (2). Therefore, in advanced breast cancer, inhibition of Runx2 by ER! may
become beneficial. This could possibly explain why, despite the well-established
oncogenic activity of ER signaling in breast epithelial cells (27), breast cancers that
maintain ER! expression during tumor progression are generally less aggressive than
109
ER!-negative tumors; in these tumors Runx2 would promote metastasis without ER!
opposition.
In osteoblasts, estrogens influence Runx2 in a developmental stage-specific
manner. Inhibition of Runx2 and its target genes OC and MMP9 is observed in late
MC3T3-E1 cultures, when Runx2 activity and OC expression are maximal. However,
data presented here and elsewhere (58, 81) show that in early MC3T3-E1 and in primary
osteoblast cultures, estrogens do not inhibit, and may even stimulate Runx2 activity and
OC expression. Both the stimulation and inhibition of Runx2 by estrogens could
contribute to their pro-skeletal properties. During specific differentiation stages,
stimulation of Runx2 may promote osteoblast function and bone formation. During other
stages, inhibition of Runx2 may benefit the skeleton by (i) attenuating osteoblastogenesis
thereby restraining bone turnover (23, 50, 78); and (ii) decreasing the expression of
osteoclastogenic genes, possibly RANKL and MMP9, which were down-regulated along
with OC in our late MC3T3-E1 cultures. The skeletal benefits of keeping Runx2 in
check is suggested by the excessive endosteal bone resorption (reminiscent of
postmenopausal bone loss) and spontaneous fractures observed in transgenic mice over-
expressing Runx2 (35, 76), as well as the resistance of Runx2-DN transgenic mice to
ovariectomy-induced bone loss (79).
Inhibition of Runx2 is a potential novel mechanism of action of estrogen in breast
and bone, and may contribute to drug discovery. Specifically, if the inhibition were
indeed important for bone and breast epithelial cell growth and differentiation, then
110
SERMs would behave most “naturally” if they mimicked estrogens not only in
stimulating ERE-driven transcription, but also in restraining Runx2. Furthermore,
indications for SERM therapy for breast cancer should possibly take into account the
potential consequences on Runx2 activity.
In conclusion, this project dissected the strong physical interaction and inhibition
of Runx2 by ER!. Suppression of Runx2 in response to E2, and the various responses to
synthetic ligands, may mediate some of their effects on bone, breast and other organ
systems. In addition, Runx2 inhibition assays may prove useful in SERM evaluation and
development.
111
Chapter 4: Modulation of Runx2 by AR
1. Introduction
The androgen receptor (AR) plays pivotal roles in a myriad of physiological and
pathological processes including reproduction, bone development, as well as prostate
cancer progression. In bone the AR helps maintain proper bone turnover and overall
volume. In prostate cancer the AR is relevant in all aspects of prostate cancer progression
from the initial stages to later ablation-resistant stages. Nevertheless, there are still
questions regarding its role during metastasis of prostate cancer to bone.
Approximately 70-90% of prostate cancer patients with advanced disease develop
metastasis to bone (117). Prostate cancer finds a fertile soil within bone to grow heralding
back to Paget’s theorem known as the “seed soil theory”. In this theory Stephan Paget
suggests that the distribution of metastasis is not due to chance, but instead cancer (the
seed) metastasizes in specific locations (soil) that is fertile for growth, such as bone
(113). Once prostate cancer metastasizes to bone it is difficult to treat; patients have a
mean survival time of nine months to one year (18). At metastatic sites the formation of
osteoblastic lesions is characterized by new woven bone formation and varying degrees
of osteoclastic bone resorption (88). It has been shown that these metastatic cells express
similar genes to osteoblasts, including RANK ligand, osteoprotegrin, bone sialoprotein,
osteopontin, osteoclacin, and the transcription factor that controls the expression of these
genes, namely Runx2 (11, 161).
In PC3 cells, a prostate cancer cell line derived from bone metastasis,
overexpression of Runx2 led to an increase in MMPs; which breaks down stroma and
112
basement membranes of the extracellular matrix (ECM) (108). The breakdown of the
ECM allows invasion into other tissues. Similar findings were found with the breast
cancer cell line MDA-MB-231 (108). It therefore can be considered that Runx2 is pro-
metastastatic in these settings.
Yet, a natural duality in Runx proteins seems to operate depending on the cellular
context. Runx2 normally functions in osteoblast differentiation in one context, but may
be implicated in cancer in another context (60, 131). Runx2 is upregulated in cell lines
such as C4-2B, which has osteoblastic properties that allow C4-2B cells to mineralize
(11, 167). However, studies presented here (Chapter 2) show that Runx2 DNA binding
capacity is inactive in these cells for unknown reasons. However, the osteolytic PC3 cell
line expresses a transcriptionally active Runx2, as presented in this work as well as
previous studies (161). Due to this inconsistency in Runx2 activity and the osteoblastic
phenotype, there must be other mechanisms involved with C4-2B cells capacity to
mineralize. One that has been implicated is that of Notch signaling and ERK activation as
being important for the osteomimetic properties of prostate cancer metastatic cell lines
(11, 161, 167).
The AR has a similar duality. The AR is known to mediate differentiation in
many settings, yet in prostate cancer the AR is strongly involved in cancer progression by
increasing proliferation, and the development of resistance to androgen deprivation
during ablation therapies (49). The exact mechanisms of prostate cancer metastasis and
androgen dependence are still relatively unclear. However, it is known that a major
player in these processes is the AR, which has been known to interact with Runx2 and
113
have functional consequences on Runx2 activity (58, 81). Similarly, Runx2 also has the
capacity to alter AR activity (58, 81).
The roles of androgens in bone physiology are also not entirely clear. It has been
suggested that androgens mainly complement the actions of estrogens by being converted
to estrogen by aromatase. However, the importance of androgens has begun to emerge by
AR knockout mice, studies that overexpress the AR in osteoblasts, and androgens
influencing differentiation in osteoblasts (160).
The two major players in bone and prostate cancer, Runx2 and AR respectively,
have been shown to be involved in mutual inhibition (58). However, studies of Slp (sex-
limited protein) gene regulation revealed that Runx2 actually stimulated AR-mediated
transactivation, potentially as a result of direct interaction between the two proteins (95).
Regardless of the different functional outcomes – stimulation versus repression – which
could be cell type and/or gene-specific, the direct interaction between AR and Runx2
could account for their influence of each other’s activity and changes in cellular
distribution (58).
However, the causal relationships between the three reported levels of AR/Runx2
interactions – physical binding with each other, influence on each other’s nuclear
distribution, and influence of each other’s transcriptional activity – remain unknown.
Furthermore, because the two proteins are transcription factors, they could influence each
other through respective target genes. The present study further characterizes the mutual
transcriptional repression of AR and Runx2 that was seen by Kawate et al. The use of
transcriptionally dead AR mutants that inhibit Runx2 activity, and vice versa, rules out
114
the potential of these factors to influence each other through their respective target genes.
This phenomenon is attributed to the interaction between their respective DNA-binding
domains, which results in mutual inhibition of DNA binding as assessed by EMSA and
ChIP. The work presented will also show, to a lesser extent, how the AR was able inhibit
Runx1 and vice versa. FRAP studies were also performed to show how these proteins
behaved in living cells. Several cell lines were used to examine the functional inhibition
of Runx2 by AR, and the inhibition of AR by Runx2.
115
2. Results
2.1 Luciferase assays and EMSA reveal the inhibition of endogenous Runx1 and
Runx2 proteins by AR in PC3 cells.
It has been previously reported that several prostate cancer cell lines have
endogenous Runx proteins (161). To assess the effect of the AR on endogenous Runx
proteins PC3 cells were examined, since they are AR negative and AR can subsequently
be transfected into these cells. When 6XOSE2-luciferase reporter was transfected in PC3
cells, there was strong constitutive activity presumably by endogenous Runx proteins.
Cotransfection of the AR inhibited this activity weakly in the absence of added ligand
and strongly in the presence of the AR ligand, DHT (Figure 4.1A).
An EMSA was performed on PC3 whole cell extract to identify which Runx
proteins could bind DNA in these cells. This was necessary because certain prostate
cancer cell lines have been reported to have Runx proteins present but their ability to bind
DNA was inhibited for unknown reasons (presented in the material and methods
chapeter) The shifted band seen in Figure 4.1B lane 2 was able to be supershifted by both
Runx1 (Figure 4.1B lanes 3,4) and Runx2 antibodies (Figure 4.1B lanes 4,5). The entire
Runx/DNA shifted complex was supershifted by the addition of Runx1 and Runx2
antibodies together (Figure 4.1B, lane 7). Therefore, PC3 cells have active Runx1 and
Runx2 proteins, and the AR, in the luciferase reporter assay, inhibited either one or both
of these transcription factors.
116
Figure 4.1 Luciferase Reporter assay and EMSA reveal Runx1 and Runx2
presence in PC3 Cells.
Figure 4.1 LEGEND. A) PC3 cells have constitutive Runx activity based on the
6XOSE2-luciferase reporter assay. Expression of the AR interferes with the endogenous
Runx activity to drive luciferase expression of this reporter. Runx activity is inhibited by
the AR in a ligand independent and ligand dependent manner. B) PC3 cell whole cell
extract was incubated with an OSE2 radiolabeled probe. There was a shift in lane 2, a
supershift by Runx1antibody in lanes 3 & 4, a supershift by Runx2 antibody in lanes 5 &
6, a supershift by both Runx1 and Runx2 antibody in lane 7. Boiled antibody in lane 8
did not abrogate the shift, however competition with unlabeled probe (10x and 100x) did
in lanes 9 and 10. There was expression of Runx1 and Runx2 in PC3 cells that was
competent in binding DNA.
117
6XOSE2-Luc
0
5000
10000
15000
20000
25000
30000
35000
40000
45000
-
EtOH
DHT
AR
Probe
Extract
ab Runx1
ab Runx2
wt Competitor
+ + + + + + + + + +
+ + + + + + + + +
10x 100x
+ +
+ + +
+
Runx2
Runx1
+boiled ab
OSE2 probe
lane
1 2 3 4 5 6 7 8 9 10
A B
118
2.2 Runx1 and Runx2 are inhibited by the AR in a ligand specific and receptor
specific manner.
Since both Runx1 and Runx2 are expressed and active in PC3 cells it was
necessary to determine if the AR was inhibiting one or both of these transcription factors.
To assess which Runx proteins were inhibited by the AR cells expressing neither Runx
nor AR was necessary. Cos7 cells served this purpose, since they contain no endogenous
Runx proteins and no endogenous steroid hormone receptors. Coexpression of Runx1 and
the AR in Cos7 cells led to the repression of Runx1 activity assessed by the coexpressed
6XOSE2-luc reporter (32) in a ligand independent manner as well as a ligand specific
manner, with dihydroxytesterone (DHT) strongly repressing Runx1 (eliminating 90% of
the activity) and not dexamethasone (Dex) and estradiol (E2) (Figure 4.2A). However,
coexpression of Runx1 and GR did not inhibit Runx1 activity (Figure 4.2A). Similar
affects were seen with the coexpression of Runx2 and AR, as well as with the
coexpression of Runx2 and GR in Cos7 cells (Figure 4.2B). To assess if the AR and GR
receptor were active, ARE-luc and GRE-luc reporter activities were assessed in
respective receptors expressed in Cos7 cells (Figure 4.2C). Consequently, Runx1 and
Runx2 are both inhibited in a receptor- and ligand-specific manner.
119
Figure 4.2 Luciferase assays reveal that Runx1 & Runx2 are repressed in a reporter
and ligand specific manner
Figure 4.2 Legend. A,B) Coexpression of Runx1 or Runx2 with the AR respectively, and
treatment with vehicle, DHT, E2 or Dex. Similarly, Runx1 and Runx2 activity was
assessed by coexpression with GR. Runx activity was determined by using a 6XOSE2-
luc reporter.C) AR and GR activity was determined using an ARE and GRE reporter
respectively.
0
3000
6000
-
-
GR
+
-
+
AR
+
Runx2 activity (RLU)
EtOH
DHT
E2
Dex
6XOSE2-Luc
Runx2
SHR
Runx2
0
3000
6000
- - AR GR
- + + +
EtOH
DHT
E2
Dex
Runx1
SHR
6XOSE2-Luc
Runx1 activity (RLU)
Runx1
0
50
100
AR EV
ARE - luc
0
GR EV
GRE - luc
50
100
A
B C
120
2.3 Runx1 and Runx2 Cellular localization changes in the presence of the AR and
DHT.
Immunoflourescent confocal microscopy was used to further assess how
endogenous Runx1 and Runx2 localization was altered in PC3 cells by ectopically
expressed AR (Figure 4.3). At steady state levels the AR in vehicle conditions
colocalized with Runx1 and Runx2 in the cytoplasm and nucleus (Figure 4.3A). Upon
treatment with DHT the AR was predominantly nuclear, translocating not only itself, but
also endogenous Runx1 and Runx2 (Figure 4.3A). The controls revealed that Runx1 and
Runx2 did not translocate to the nucleus upon DHT treatment alone in the absence of the
AR (Figure 4.3B).
2.4 Runx1 and Runx2 was Constitutively nuclear in Cos7 cells, AR Colocalization
Upon DHT treatment.
Unlike in PC3 cells, in Cos7 cells transfected Runx1 and Runx2 were mainly
nuclear at steady state levels (Figure 4.4 A,B). Also unlike PC3 cells AR was
predominantly cytoplasmic in vehicle conditions that resulted in the translocation to the
nucleus upon DHT treatment, where it colocalized with Runx1 and Runx2, respectively
(Figure 4.4 A,B). Controls revealed that nuclear localization of Runx1 and Runx2 was
unaltered in the presence or absence of AR in the absence of ligand or in the presence of
ligand without AR (Figure 4.4 C). The AR nuclear localization upon DHT treatment was
identical whether Runx1 or 2 were present or not (Figure 4.4C). Therefore, it was no
surprise that Runx1 and 2 when coexpressed with AR, colocalized upon DHT treatment.
121
Figure 4.3 Immunoflourescence of PC3 Cells with Runx1 or 2, and AR
Figure 4.3 Legend. PC3 cells were transiently transfected with AR, and treated for 24 h
with ethanol or DHT (10nM) A,B) AR (red) and Runx1 (green) or Runx2 (green) were
visualized using confocal microscopy as described in Materials and methods. Merged
images show regions of colocalization (Yellow) C) Controls of endogenous Runx1
(green) and Runx2 (green) upon vehicle or DHT treatment in the absence of AR.
EtOH
AR Runx2 Merge
DHT
EtOH
AR Runx1 Merge
DHT
Runx1
EtOH
DHT
Runx2
A
B
C
122
Figure 4.4 Immunoflourescence of Cos7 cells with Runx1 or 2, and AR
Figure 4.4 Legend. Cos7 cells were transiently transfect with AR and Runx2 or AR and
Runx1, and treated for 24 h with ethanol or DHT (10 nM). A,B) Cellular localization of
Runx2 (green) or Runx1 (green) in the presence of AR (Red) in both vehicle and DHT
conditions. Merged images show regions of colocalization (yellow) C) Controls of
expression of AR, Runx2, or Runx1 separately upon ethanol or DHT treatment.
123
EtOH
AR Runx2 Merge
DHT
EtOH
AR Runx1 Merge
DHT
Runx1
Runx2 AR
EtOH
DHT
A
B
C
124
2.5 Mechanism of Runx2 Inhibition by the AR
The phenomenon that liganded AR abrogates Runx1 and Runx2 activities is clear.
However, the mechanism(s) is not. Two main hypotheses can be proposed that may lead
to this inhibition, 1) the AR directly interacts with either Runx1 or Runx2, thereby
inhibiting their activities (Figure 4.5A) or 2) the AR, being a transcription factor, can
induce the production of a protein (X) that in turn may inhibit Runx1 and Runx2 (Figure
4.5B).
One method used to distinguish between the two hypotheses was to assess if there
was a difference in the DHT dependence of AR activation and Runx2 repression. A log
difference was apparent in the DHT concentration between the 50% activation of the AR
(3 nM) and 50% inhibition of Runx2 (0.3 nM) (Figure 4.6A). A similar difference was
observed for Runx1 inhibition versus AR activation (Figure 4.6B). It is therefore, very
likely that AR-mediated transcription of an intermediary factor is not the cause of the
observed Runx1 and 2 inhibition with the caveat that transcription of all AR target genes
may not have the same DHT dependence as ARE-luc reporter used in these experiements.
We decided to address this ussue by using different AR constructs. Since SHR are
modular in nature the LBD of the AR (leaving the NTD/DBD) was truncated in one
construct and the NTD (leaving the DBD/LBD) was truncated in another construct. This
strategy did not resolve the issue since the transactivationally active (Figure 4.7A)
NTD/DBD construct was able to inhibit Runx activity in PC3 cells similar to wildtype
AR, while the transactivationally dead DBD/LBD did not (Figure 4.7B). Therefore,
subsequent constructs were made further dissecting the NTD/DBD (Figure 4.8A). The
125
transactivationally dead DBD inhibited Runx activity robustly, while the NTD did not
(Figure 4.8B). Furthermore, the transactivationally dead DeltaAF1 (Figure 4.9A) mutant
inhibited Runx2 (Figure 4.9B); the inhibition was not as strong as the wt AR, which may
be due to lower expression of the DeltaAF1 (inset Figure 4.9B). Taken together the data
suggest that the Runx1 and Runx2 inhibition by the AR is not due to AR’s
transactivational activity, but instead it is most likely due to direct interaction between
Runx proteins and the AR.
126
Figure 4.5 Models of how AR may be inhibiting Runx2
Figure 4.5 Legend. Two possible hypotheses are modeled explaining Runx2 inhibition by
AR. A) Model of Runx2 inhibition by AR where Runx2 and AR directly interact and
thereby inhibit Runx2 activity. B) Model of Runx2 inhibition by AR where AR
transcribes a protein X and that protein X inhibits Runx2 activity.
A
B
127
Figure 4.6 Activation/Repression Curves of AR and Runx1 or Runx2
Figure 4.6 Legend. A) Repression curve of Runx2 activity (assessed by a 6XOSE2-luc
reporter) coexpressed with AR with increasing concentrations of ligand (black circle) and
activation curve of AR (assessed by a ARE-luc reporter) with increasing concentrations
of ligand (open triangle). The curves are presented in a log scale with 50% inhibition of
Runx2 by AR occurring at 0.3 nM and 50% activation of AR occurring at 3 nM as
indicated by solid lines. B) Similar to A, repression and activation curves with respect to
Runx1 (black circle). The curves are presented in a log scale with 50% inhibition of
Runx1 by AR occurring at 0.2 nM and 50% activation of AR occurring at 2 nM as
indicated by solid lines.
128
% Activation % Activation
DHT
X Data
0.001 0.01 0.1 1 10 100
Y Data
0
20
40
60
80
100
120
DHT (nM) vs AR % Activation
DHT (nM) vs Runx2 % inhibition
A
RunX1/AR vs DHT
DHT Conc. [nM]
0.001 0.01 0.1 1 10 100 1000
% Activation
-20
0
20
40
60
80
100
120
DHT (nM) vs Runx1 % inhibition
DHT (units) vs AR % Activation
Col 11 vs Col 12
Col 18 vs Col 19
B
Runx1
Runx2
129
Figure 4.7 Inhibition of Runx in PC3 cells by different AR domains
Figure 4.7 Legend. A,B) PC3 cells were transiently transfected with AR, the NTD/DBD
of the AR, or the DBD/LBD of the AR to measure luciferase activity against either the
ARE-luc or the 6XOSE2-luc reporter. Luciferase activity was measured 24 hrs after
treatment with either ethanol (0.01%) or DHT (10 nM). All data are presented as mean
values + SEM, with n=4 dish replicates of a reperesentative experiment, repeated at least
3 times.
AF1 AF5
LBD DBD NTD
AF2
Androgen receptor
NTD/DBD
DBD/LBD
ARE Reporter
0
200000
400000
600000
800000
1000000
1200000
1400000
1600000
- AR NTD/DBD DBD/LBD
EtOH
DHT
AR activity (RLU)
p6XOSE2 Reporter
0
1000
2000
3000
4000
5000
6000
7000
8000
9000
10000
- AR NTD/DBD DBD/LBD
EtOH
DHT
Runx activity (RLU)
A
B
130
Figure 4.8 Inhibition of Runx in PC3 cells by the AR DBD
Figure 4.8 Legend. A,B) PC3 cells were transiently transfected with AR, the NTD of the
AR, or the DBD of the AR at the indicated molar equivalents to measure luciferase
activity against either the ARE-luc or the 6XOSE2-luc reporter. Luciferase activity was
measured 24 h after treatment with either ethanol (0.01%) or DHT (10 nM).
AF1 AF5
LBD DBD NTD
AF2
Androgen receptor
NTD
DBD
0
50000
100000
150000
200000
250000
300000
350000
400000
450000
- AR NTD NTD NTD DBD DBD DBD DBD
H
DBD
H
DBD
H
LBD LBD LBD
- 1 1 3 10 1 3 10 1 3 10 1 3 10
EtOH
DHT
ARE-
Luciferase
0
5000
10000
15000
20000
25000
30000
35000
40000
45000
50000
- AR NTD NTD NTD DBD DBD DBD DBD
H
DBD
H
DBD
H
LBD LBD LBD
- 1 1 3 10 1 3 10 1 3 10 1 3 10
EtOH
DHT
6XOSE2-
Luciferase
A
B
131
Figure 4.9 Delta AF1 inhbition of Runx2 in Cos7 Cells
Figure 4.9 Legend. A) Cos7 cells were transiently transfected with wt AR or Delta AF1
and an ARE-luc reporter to measure activity. Cells were treated for 24 hr with ethanol or
DHT (10 nM). B) Cos7 cells were transiently transfected with Runx2 and wt AR or Delta
AF1 and a 6XOSE2-luc reporter to measure activity. Cells were treated as in A.
AF1 AF5
LBD DBD NTD
AF2
Delta AF1
0
225000
450000
EV AR Delta
AF1
RLU
EtOH
DHT
ARE-Luciferase
0
5000
10000
15000
20000
25000
- - AR Delta
AF1
- + + +
RLU
EtOH
DHT
6XOSE2-Luciferase
Runx2
SHR
AR
Delta AF1
EtOH
DHT
A B
132
2.6 Direct Protein-Protein Interaction of Runx2 and AR
To assess direct physical interactions between Runx and AR proteins, GST-pull
down assays were performed using bacterially-expressed and purified GST fusion
proteins of the AR as bait (Figure 4.10A) and prey proteins from S
35
-labeled Runx2
(Figure 4.10B). The interaction domain that participated in the complex formation
between the AR and Runx2 was the AR DBD (Figure 4.11A), which was an ionic
interaction (Figure 4.11B). The integrity of the whole AR-DBD was necessary since
dissection of the AR at the end of the 1
st
zinc finger (540-580 and 580-647, Figure
4.12A,B) was not able to interact with Runx2 (Figure 4.12C). In vitro translation of
different proteins of Runx2 as bait led to the discovery that the AR DBD binds to the
Runt-PST region of Runx2 (Figure 4.13B).
133
Figure 4.10 Overview of GST pull-down assays
Figure 4.10 Legend. A) Different GST-AR constructs bacterially expressed and purified.
A commassie was done on the expressed proteins to ensure equal expression of the GST-
constructs. B) Schematic of invitro translation of Runx2 proteins. This experiment was
done by Dr S Baniwal in collaboration.
1 110 367 540
AF1 AF5
540 647
GST-DBD
670 920
AF2
647
GST-LBD
GST-NTD
1 110 367 540 647 917
AF1 AF5
LBD DBD NTD
AF2
-DBD
90
51
29
GST-
kDa
-NTD
-LBD
GST
Androgen receptor
A
B
134
Figure 4.11 Runx2 directly interacts with the AR DBD
Figure 4.11 Legend. A) Radiolabled Runx2 was incubated with the indicated GST-fusion
proteins used as baits. B) Radiolabeled Runx2 was incubated with GST-DBD at
increasing salt concentrations as indicated. The autoradiographs in A and B shows Runx2
pulled down by the indicated baits. This experiment was done by Dr S Baniwal in
collaboration.
A B
135
Figure 4.12 Full AR DBD is necessary for the interaction with Runx2
Figure 4.12 Legend. A) A model depicting the DBD region of the AR. B) Coomasie
stained SDS-PAGE of the bacterially expressed and purifed GST or GST fusion proteins
of the DBD constructs (540-647, 540-580, or 580-647) used as baits in the pull-down
assay. B) The autoradiograph shows Runx2 pulled down by the indicated baits. This
experiment was done by Dr S Baniwal in collaboration.
LETAR
18aa
GEAS
22aa
mRunX2*
GST
540-647
540-580
580-647
In
GST
540-647
540-580
580-647
43
34
55
26
A
B C
136
Figure 4.13 Runt-PST of Runx2 interacts with the AR DBD
Figure 4.13 Legend. The indicated radiolabeled Runx2 fragments were incubated with
GST-AR-DBD fusion proteins. The Runt-PST region strongly interacted with the GST-
AR-DBD as shown by the autoradiograph above. This experiment was done by Dr S
Baniwal in collaboration.
mRunX2*
QA
1 176 303 596
PST Runt WRPY
GST-
AR-DBD
QA-Runt*
Runt-PST*
Runt*
GST
In
PST
*
137
2.7 Runx2 does not interact with the A573D Point Mutant
Several AR mutations, that occur in patients, are detrimental to the health of that
patient (29). One such mutation is A573D known to be present in patients with androgen
insensitivity syndrome (29). To assess if this mutation affected Runx2/AR interactions, a
GST-pull down assay was performed; no Runx2 interaction was observed compared to
the wt DBD (Figure 4.14A). Functionally, Runx2 inhibition was also significantly
reduced by this mutation (Figure 4.14B). Mechanistically, it was observed that the AR
inhibited Runx2 by not allowing it to bind DNA (Figure 4.14C lane 3,4). However, the
point mutant (Figure 4.14C lane 5) could not totally eliminate Runx2 binding to its DNA
element. The AR DBD itself does not compete for binding to a Runx2 specific element
(Figure 4.14C lane 7).
2.8 Inhibition of AR activity by Runx Proteins
Since the AR DBD is interacting with Runx2 it is possible that the AR’s ability to
bind to its cognate DNA element is being affected by Runx2. An EMSA was performed
to measure possible changes in AR binding to an ARE element (Figure 4.15A). It is
evident that the presence of Runx2 inhibited AR’s ability to bind DNA (Figure 4.15A
lanes 5 and 6, compared to lanes 7and 8, as well as the supershifted lanes 11 and 12
versus 13 and 14). It is likely that a functional consequence of AR DBD interacting with
Runx proteins is the elimination of the AR’s ability to bind DNA. To assess this Runx2
and Runx1 proteins were expressed in LNCaP cells with increasing amounts of each,
leading to a subsequent decrease in the activity of endogenous AR (Figure 4.15B).
138
Similar to Runx2 repression by a transactivationally dead AR, AR activity was also
inhibited by a transactivationally dead Runx2 (Figure 4.16A,B). Therefore, a mutual
repression by Runx2 and AR was evident in the systems described above.
139
Figure 4.14 A573D mutant AR doesnot interact with or affect Runx2 DNA binding
Figure 4.14 Legend. A) Upper panel coomassie stained SDS-PAGE of the bacterially
expressed and purified GST, GST-AR DBD, or GST AR DBD A573D mutant. Lower
panel autoradiograph shows radiolabeled Runx2 pulled down with the wt AR DBD but
not the A573D point mutant. This part of the experiment was done by Dr S Baniwal in
collaboration. B) Cos7 cells were transiently transfected with plasmids encoding either wt
AR or the A573D point mutation, and Runx2, followed by 24hrs of treatment with either
ethanol or 10nM DHT. All samples were transfected with the 6XOSE2-luc reporter and
assayed for luciferase activity after treatment. C) MC3T3-E1 extracts were incubated
with the OSE2 probe and with the indicated purified proteins or Runx2 antibody. An
EMSA was performed. The autoradiograph shows the shift by WCE from MC3T3E-1 as
well as the abrogation of the shift with incubation of the AR-DBD. This part of the
experiment was done by Dr S Baniwal in collaboration.
140
Runx2
wt Mut
AR-DBD
72
43
26
KDa
GST
0
5000
10000
15000
20000
25000
- - AR A573D
- + + +
RLU
EtOH
DHT
6XOSE2-luc
Runx2
SHR
unspecific
Runx2
OSE2
x 3x 3x 3x - 3x
!-Runx2
+ boil
OB
cell extract
+ + - + + + + + -
purified
-
DBM DBD GST
- -
DBD
A B
C
141
Figure 4.15 AR activity is inhbited by Runx1 and Runx2
Figure 4.15 Legend. A) Cos7 cells were transiently transfected with the indicated
constructs, followed by 24 hrs of treatment with either vehicle or 10nM DHT. WCE
extracts were made and EMSA was performed in the indicated conditions. Supershifts
were done using 1 ug of the AR antibody (SC-441X). B) LNCaP cells were transiently
transfected with an ARE-luciferase reporter and increasing concentrations of Runx1 or
Runx2. After 24 hrs of 10nM DHT treatment cells were lysed and assayed for luciferase
activity. This part of the experiment was done by Dr S Baniwal in collaboration.
+ !-AR
pCMV-AR
- + - + - +
pCMV-
AR/Runx2 pCMV pCMV-AR
- + - + - +
pCMV-
AR/Runx2 pCMV
DHT -
no ext
AREIII
AR
(super -
shifted)
Non-
specific
band
0
50000
100000
150000
200000
250000
300000
350000
Runx1
+ +
RLU
+ + + +
Runx2
(DHT)
T
A
ARE
Luciferase
A B
142
Figure 4.16 Transactivationally dead Runx2 can inhibit AR activity
Figure 4.16 Legend. A,B) Cos7 cells were transiently transfected with plasmids encoding
Runx2, DBM and AR in the indicated conditions with either the 6XOSE2-luc reporter or
the ARE-luc reporter.
ARE (MMTV)-Luc
0
2000
4000
6000
8000
10000
12000
14000
16000
18000
- - Runx2
1ng
Runx2
3ng
Runx2
10ng
DBM
1ng
DBM
3ng
DBM
10ng
- AR AR AR AR AR AR AR
RLU
EtOH
DHT
6XOSE2-luc
0
5000
10000
15000
20000
25000
30000
35000
- Runx2 Runx2 DBM DBM
- AR AR AR AR
RLU
EtOH
DHT
A
B
143
2.9 Nuclear Runx2 mobility was increased by the presence of AR and vice-versa.
Flourescent recovery after photobleaching (FRAP) measures the mobility of GFP
tagged proteins within a living cell. The amount of time in takes for the recovery after
photobleaching as well as the amount of recovery in intensity helps determine how
mobile the proteins are within the cell. Mobility can be altered by the interaction of the
GFP-tagged protein to other proteins. Using this concept we wanted to assess how GFP
tagged Runx2 or GFP tagged AR mobility could be altered by interaction with AR or
Runx2 respectively.
Therefore, to understand how AR and Runx2 proteins behave in vivo, fused GFP-
Runx2 and GFP-AR constructs were expressed in Cos7 cells and FRAP was performed.
GFP-Runx2 coexpressed with the A573D mutant of the AR and treated with both vehicle
or DHT, caused no significant alteration in GFP-Runx2 mobility compared to GFP-
Runx2 alone (Figure 4.17A). GFP-Runx2, coexpressed with the wt AR however, caused
an increase in Runx2 mobility only after DHT treatment when compared to GFP-Runx2
alone (Figure 4.17A). Under DHT conditions, the mobility of nuclear GFP-AR was also
examined (Figure 4.17B). When GFP-AR was coexpressed with Runx2 an increase in
mobility was recorded (Figure 4.17B), however when the GFP-A573D (the point mutant
which does not interact with Runx2) was coexpressed with Runx2, the mobility was
identical to GFP-A573D alone. The data taken together suggest that when Runx2
interacts with AR a subsequent change in mobility of both Runx2 and AR occurred. One
possible interpretation is that the interaction of these two transcription factors does not
144
allow them to bind to their cognative DNA elements, causing an increase in subsequent
mobilities of each protein as depicted in Figure 4.18.
145
Figure 4.17 FRAP experiments of GFP-Runx2 in the Presence of AR and vice versa
Figure 4.17 Legend. A) Cos7 cells were transiently transfected with GFP-Runx2 and the
indicated AR constructs. Cells were allowed to grow for 24 hrs and subsequently treated
for 1 hr with either vehicle or DHT. After treatment FRAP was performed as described in
Chapter 2 (M&M). B) Similar to (A) Cos7 cells were transfected with GFP-AR or GFP-
A573D either with or without Runx2. Cells were assayed the same as in A.
GFP Runx2 versus AR
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 10 20 30 40 50 60 70
Time (s)
Intensity (Arbritary Units)
GFP Runx2 and AR EtOH
GFP Runx2 and AR DHT
GFP Runx2
GFP Runx2 and A573D EtOH
GFP Runx2 and A573D DHT
Poly. (GFP Runx2 and AR EtOH)
Poly. (GFP Runx2 and AR DHT)
Poly. (GFP Runx2)
Poly. (GFP Runx2 and A573D EtOH)
Poly. (GFP Runx2 and A573D DHT)
GFP AR versus Runx2
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0 20 40 60 80 100 120 140 160
Time after bleaching (s)
Intensity (Arbritary Units)
GFP AR and Runx2
GFP AR
GFP A573D and Runx2
GFP A573D
A
B
146
Figure 4.18 Model depicting the mechanism of mutual AR/Runx2 inhibition
Figure 4.18 Legend. Model depicting the mechanism of mutual inhibition of AR and
Runx2. Upon AR and Runx2 interaction there is mutual inhibition of transcriptional
activity by not allowing binding to its respective DNA element.
NTD
DBD
LBD
Runx2
NTD
DBD
LBD
DHT
Runx2
AR
DNA Runx2
response element
DNA AR
response element
147
2.10 Inhibition of Endogenous Runx Proteins by the AR
The model systems analyzed so far express only Runx proteins (PC3) or the AR
(LNCaP) endogenously or neither (Cos7 cells). To see if AR-mediated Runx inhibition
occurs in cells expressing both Runx proteins and AR, we obtained three cell types
known to have endogenous expression of both: MC3T3-E1 (mouse osteoblast cells),
SAOS-2 (human osteosarcoma cells), and Ros (rat osteosarcoma cells).
In MC3T3-E1 cells the AR is stabilized by DHT (Figure 4.19C) and colocalizes
to specific foci with Runx2 in the nucleus in differentiated cultures, unlike early
proliferating cultures where colocalization is both cytoplasmic and nuclear (Figure
4.19B). Endogenous Runx2 target gene expression was also downregulated by DHT
treatment in differntiated cultures where Runx2 and AR were predominantly nuclear
(Figure 4.19A,C). At the time points where osteocalcin expression decreased upon DHT
treatment, Runx2 occupancy at the OSE2 region of the osteocalcin promoter also
decreased as assessed by ChIP (Figure 4.22).
In SAOS-2 cells, Runx activity was downregulated by DHT (Figure 4.20A) as
well as an increase in colocalization of Runx2 and AR was observed upon DHT treatment
(Figure 4.20B). However, unlike MC3T3-E1 and SAOS-2 cells, Ros cells were not
inhibited upon DHT treatment (Figure 3.21A). This may be due to the fact that the AR
does not translocate to the nucleus upon DHT treatment in these cells (Figure 4.21B).
Consequently, we can conclude that for Runx activity to be inhibited by the AR, the AR
and Runx2 have to be nuclear.
148
Figure 4.19 Runx2 activity is inhibited by AR in MC3T3-E1 Cells
Figure 4.19 Legend A) MC3T3E-1 cells were subjected to differtiation conditions and
treated with ethanol (grey bars) or DHT (10 nM black bars) commencing at confluence.
Levels of the indicated mRNAs were measured on Days 3,9, and 12 by RT-qPCR as
described in Chapter 2. Data were corrected for the expression of ribosomal protein
L10A, which itself did not significantly change during time or in response to DHT (mean
+ SD; n=3) B,C) Early or late MC3T3E-1 cultures were treated for 24 hrs with ethanol or
DHT (10 nM). AR (red) and Runx2 (green) were visualized using confocal microscopy
as described in Chapter 2. Colocalization (yellow) is demonstrated in the merged image.
EtOH
DHT
RT-PCR
0
20
45
60
80
100
arbitrary units
OC
MMP9
arbitrary units
0
20
45
60
80
100
arbitrary units
0
20
45
60
80
100
L10A
Days 3 9 12
EtOH
DHT
Early MC3T3E1 Cells
AR Runx2 Merge
DAPI
Late MC3T3E1 Cells
AR Runx2 Merge
DAPI
EtOH
DHT
A B
C
149
Figure 4.20 Runx2 activity is inhibited by AR in SAOS-2 Cells
Figure 4.20 Legend A) Saos-2 cells were transiently transfected with 6XOSE2-luc and
treated for 24 hrs with ethanol or DHT (10nM). Subsequently, cells were lysed and
assayed for luciferase activity. B) Saos-2 cells were treated for 24 hrs in ethanol or 10nM
DHT. AR (red) and Runx2 (green) were visualized as in Figure 3.19. C) Immunoblot of
AR and Runx2 expression in SAOS-2 cells after 24h of treatment wit ethanol or DHT.
SDS-PAGE commassie is shown as a loading control.
6XOSE2-luc
0
200
400
600
800
1000
1200
1400
1600
EtOH
DHT
Runx activity (RLU)
EtOH
AR Runx2 Merge DAPI
DHT
EtOH
DHT
AR
Runx2
Commassie
A B
150
Figure 4.21 Runx2 activity is not inhibited by AR in ROS Cells
Figure 4.21 Legend. A) Similar to figure 4.20A but performed in ROS cells. B) Similar to
figure 3.19B but performed in ROS cells.
EtOH
AR Runx2 Merge DAPI
DHT
0
200
400
600
800
1000
1200
Ros
EtOH
DHT
6XOSE2-luc
Runx activity (RLU)
A
B
151
Figure 4.22 Decrease in Runx2 Occupancy on the OSE2 site by DHT treatment
4 hrs DHT treatment at 9 days
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
- +
DHT treatment
% Input
Runx2 ChIP OC site
IgG ChIP OC site
Runx2 ChIP Insulin
IgG ChIP Insulin
Figure 4.22 Legend. Decrease in Runx2 occupancy in at the OSE2 site on day 9 post
confluent mineralizing cells upon DHT treatment. MC3T3E-1 cells were subjected to
differentiation conditions and cells were treated with 10nM DHT for 4 hrs at day 9.
ChIP was performed for Runx2 and an IgG control. Occupancy was measured by q-
PCR at the OSE2 site as well as a non-specific insulin site. Runx2 occupancy decreases
approximately 50% upon DHT treatment on the OSE2 site. There was no occupancy
observed on the non-specific insulin site as well as the IgG controls for all sites.
152
2.11 Effects of SARMS (Selective Androgen Receptor Modulators) on the
modulation of Runx activity by the AR
SARMS are used clinically in the treatment of prostate cancer progression.
Therefore, the effect of SARMS on Runx2 activity as mediated by the AR may have
clinical implications. The T877A mutant of the AR arose during androgen ablation
therapy and becomes an active receptor to OHF (hydroxyflutamide), one of the SARMS
used in prostate cancer treatment (29).
In PC3 cells OHF and Casodex did not inhibit endogenous Runx activity when
the wt AR was expressed as DHT did. However, when T877A was expressed, OHF
inhibited endogenousRunx activity (Figure 4.23A). It is known that the T877A mutant
may be activated by OHF on an ARE reporter unlike the wt AR (29). (4.23B). Since PC3
cells have endogenous Runx1 and Runx2 the above-mentioned activities were further
explored in Cos7 cells. The T877A mutant inhibited both Runx1 and Runx2 after either
DHT or OHF treatment, and the wt AR did not inhibit Runx1 or Runx2 activity upon
OHF treatment as it did with DHT treatment (Figure 4.24 A,B). Once again the T877A
mutant was activated by OHF (4.24C) wheras wt AR was not. Casodex in all conditions
did not activate the AR or T877A and did not cause inhibition of either Runx1 or Runx2
activity. FRAP experiments revealed an increase in GFP-Runx2 mobility by the T877A
mutant under OHF conditions similar to that of wt AR under DHT conditions (Figure
4.25). A possible explanation for inhibition and increase in GFP-Runx2 mobility by
153
T877A and OHF is that OHF may allow the mutant AR to open up and allow appropriate
contacts of the mutant AR to interact with Runx1 and Runx2.
154
Figure 4.23 Inhibition of Runx activity in PC3 cells by T877A AR mutant and OHF
Figure 4.23 Legend. A,B) PC3 cells were transiently transfected with the Runx reporter
6XOSE2-luc or ARE-luc, along with expression vectors encoding AR or the T877A
mutation. Cells were treated with DHT (10 nM), OHF (10 nM), Casodex (1 uM), or
ethanol (0.01%) for 24 hrs and subjected to luciferase assay as described in Chapter 2.
0
1000
2000
3000
4000
5000
6000
7000
8000
9000
- wt AR T877A
EtOH
DHT
OHF
Casodex
0
50000
100000
150000
200000
250000
300000
350000
400000
450000
500000
- AR T877A
EtOH
DHT
OHF
Casodex
6XOSE2-Luciferase ARE-Luciferase A B
155
Figure 4.24 Inhibition of Runx1 and 2 in Cos7 cells by T877A AR mutant and OHF
Figure 4.24 Legend. (A,B) Cos7 cells were transiently transfected with the Runx reporter
6XOSE2-luc along with expression vectors encoding Runx1 (A), Runx2 (B), wt AR,
and/or T877A mutant as indicated. Cells were treated as in figure 3.22. C) Cos7 cells
were transiently transfected with an AR reporter (ARE-luc) along with an expression
vector encoding the AR or T877A mutant and treated for 24 hrs as in figure 3.22).
0
10000
20000
30000
40000
50000
60000
70000
80000
90000
- - wt AR T877A
- + + +
EtOH
DHT
OHF
Casodex
0
20000
40000
60000
80000
100000
120000
- - wt AR T877A
- + + +
EtOH
DHT
OHF
Casodex
Runx2
SHR
6XOSE2-Luciferase 6XOSE2-Luciferase
Runx1
SHR
PSA Reporter
0
200000
400000
600000
800000
1000000
1200000
1400000
1600000
1800000
2000000
- AR T877A
EtOH
DHT
OHF
Casodex
ARE-Luciferase
A B
C
156
Figure 4.25 FRAP Experiments with GFP-Runx2 and T877A AR mutant with OHF
Figure 4.25 Legend. Cos7 cells were transiently transfected with GFP-Runx2 and the
indicated AR construct. Cells were allowed to grow for 24 hrs and subsequently treated
for 1 hr with vehicle, 10nM DHT or 10 nM OHF.
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0 10 20 30 40 50 60
Time (s)
Intensity Arbritary Units
GFP Runx2
GFP Runx2 and AR with EtOH
GFP Runx2 with AR and OHF
GFP Runx2 with T877A with EtOH
GFP Runx2 and T877A with DHT
GFP Runx2 and T877A with OHF
GFP Runx2 and AR with DHT
157
2.12 Microarray Data Analysis of Runx2 target Genes in Prostate Cancer
The top five Runx2 target genes were analyzed in samples from patients that had
prostate cancer with primary cancer or androgen ablation therapy obtained from Sloan-
Kettering (Personal communication with Grant Buchanan) (41). A negative correlation
was observed to PSA of the top five Runx2 target genes (Figure 4.26 A) when examining
23 primary prostate cancers from patients not receiving therapy (primary), 17 primary
prostate cancers following 3-month neoadjuvant androgen ablation therapy (AAT). It was
expected that upon androgen ablation Runx2 target genes should all go up according to
our hypothesis, and that was the case with four of the five Runx2 target genes (Figure
4.26B).
158
Figure 4.26 Correlation of KLK3 with Runx2 target genes in PCa Tumor Biopsies
Figure 4.26 Legend. A) Correlation to KLK3 (PSA) of five top Runx2 target genes in
both primary prostate cancer and androgen ablation treated (AAT) prostate cancer. B)
Gene expression normalized to the primary prostate cancer set of five Runx2 target
genes, comparing the primary tumors samples to AAT treated samples. The data was
graciously provided from Grant Buchanan and the Sloan Kettering Group.
Corrleation of Runx2 Target
Genes to PSA
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1
1.2
KLK3 BGLAP IBSP MMP13 MMP9 SPP1
Ave Gene Expression
0
0.5
1
1.5
2
2.5
3
3.5
4
BGLAP IBSP MMP13 MMP9 SPP1
Normalized Gene Expression
Primary Samples
AAT Samples
A B
159
3. Discussion
Runx proteins are unliganded transcription factors and unlike SHR their activity
cannot conveniently be controlled by ligand manipulations both experimentally and
physiologically. The data reported in this chapter suggest that androgen effects on Runx
activity are mediated by direct protein-protein interactions of the AR with Runx2 proteins
(AR DBD interacting with Runx2 Runt-PST).
Since the AR DBD was implicated in this interaction it was no surprise that an
AR DBD point mutant (A573D) abolished both interaction and inhibition of Runx2
activity. This mutant helped us in defining the mechanism, since the A573D mutant DBD
was unable to inhibit Runx2 DNA binding in EMSA experiments like the wt AR DBD.
Similarly, in FRAP experiments the wt AR was able to alter Runx2 mobility and the
A573D mutant was not. Another AR mutant, the transcriptional dead Delta AF1 mutant,
helped us disassociate the transcriptional activity of AR and the AR’s ability to inhibit
Runx2.
When examining the mechanism of AR inhibition, the AR was unable to bind to
its DNA element with addition of Runx2 in EMSA experiments. This also should not be
surprising since the DBD of the AR was implicated in the interaction. In FRAP
experiments we also saw an increase in AR mobility upon the addition of Runx2,
possibly due to the decreased capacity of AR to bind DNA. This inhibition is also not due
to Runx2’s transcriptional capacity, since the transcriptionally dead Runx2 DBM was
able to inhibit AR activity. Our findings are consistent with a previous reports, yet we
160
identify the mechanism and examine the consequence of this interaction in vivo with
endogenous proteins and FRAP experiments.
However, finding a system in which both Runx2 and AR were present was
difficult. PC3 cells have endogenous Runx proteins but no functional AR, and LNCaP
cells have endogenous AR but no functional Runx proteins. Nevertheless, our hypothesis
is consitent in these systems. Upon addition of AR in PC3 cells, Runx activity decreased
significantly. Similarly, upon addition of Runx2 in LNCaP cells, AR activity in LNCaP
cells decreased. Since AR activity (LNCaP) was inhibited by increasing Runx2 levels it
may be possible to alter proliferation within these cells using the minimal regions of
Runx2 that interact with AR. This could have far reaching clinical implications and drug
development if proliferation is inhibited by this method.
Yet, we wanted to find a system in which both AR and Runx2 were present.
Subsequently, we examined several cell lines and found a few that had both the AR and
Runx2 . In these cells when the AR was able to go nuclear and colocalize with Runx2
there was a functional consequence. For example inhibition of a well known Runx2
target, osteocalcin, in late MC3T3E-1 cultures upon androgen treatment. And when we
examined Runx2 occupancy on the promoter region our in vitro mechanism was further
confirmed by in vivo ChIP experiments in MC3T3-E1 cells. In these cells upon four
hours of DHT treatment Runx2 occupancy on the OSE2 site of the osteocalcin promoter
was significantly inhibited at the same differentiated state as where the gene expression
of osteocalcin was inhibited.
161
When we examine cellular localization by immunoflourescence in MC3T3-E1
cells there is a decrease and redistrubtion of the Runx2 (green) punctated spots in the
nucleus at both the early and late cultures upon DHT treatment. Nevertheless, it is the late
cultures that Runx2 target genes are inhibited by DHT, possibly due to Runx2 and AR
predominantly being nuclear even though there is less in the cell of each protein as a
whole. Inhibition of Runx2 and its target genes OC and MMP9 was observed in these late
MC3T3E-1 cultures, when Runx2 activity and OC expression were maximal. However in
Saos-2 cells, an osteosarcoma cell line with osteoblastic properties, inhibition occurred
24 hrs after DHT treatment as there was no need for post confluency differentiation. Ros
cells however did not inhibit Runx activity and may need to differentiate. It is interesting
that there are certain cell types that will not allow AR to go completely nuclear, such as
Ros cells or early MC3T3-E1 cultures. That’s where there is no functional consequence.
What is keeping the AR from going to the nucleus upon DHT treatment is unknown, but
there is an additional level of regulation outside of ligand that keeps Runx2 and AR from
interacting and thereby inhibiting any functional consequence from this interaction. In
osteoblasts, androgens influence Runx2 in a developmental stage-specific manner.
The next thing examined were the known drugs for the AR, SARMS that have
been used routinely in clinical studies. They behaved similar to vehicle with the wt AR,
yet OHF did inhibit Runx activity with the T877A mutation in PC3 cells. Suggesting that
mutant receptors can inhibit Runx activity with certain SARMS, allowing a better insight
as to how certain mutants behave on Runx activity. How SARMS modulate Runx2
162
activity can give a new understanding outside of the known dogma of AR activity being
suppressed by these ligands.
In a more biological setting than cell lines, microarray expression data was
analyzed to compare the relationship between PSA (an AR target gene) to Runx2 target
genes in primary prostate cancer tumors and AAT samples. There was an inverse
correlation suggesting that Runx2 activity can also be suppressed by AR activity in
prostate cancer patients. However, this is a correlation and not necessarily cause and
effect.
In this part of the project it was shown that the AR functions in a way very similar
to ER!. It was observed that a DHT bound AR was able to inhibit Runx2. This inhibition
was observed in prostate cancer and bone-like cell lines. This inverse relationship in
Runx2 activity and AR was observed in prostate cancer patient samples as well. Runx1
activity also was inhibited by the AR. However, unlike the ER story, the AR was
repressed by increasing concentrations of Runx1 and Runx2. Using the SARMS AR was
not able to activate Runx1 or Runx2, like ER did with SERMS.
163
Chapter 5: Discussion
1. Summary of Research findings
The studies presented here tell of how one transcription factor can modify the
activity of another. We have examined how AR & ER! inhibit Runx2 in the presence of
DHT or E2, respectively. To a lesser extent we have examined how AR inhibits Runx1
activity in the presence of DHT, and ER! doesnot in the presence of E2. Clinical
correlations have been identified in both prostate and breast cancer studies that suggest
inhibition of Runx2 by AR and ER!, respectively, occurs in vivo. Mechanistically we
have discovered that AR and Runx2 directly interact, thereby, abrogating the ability of
both transcription factors to bind DNA. This has led to the discovery that Runx2 also
represses AR activity. The mechanism for inhibition of Runx2 by ER! is still unclear,
however we know that these two trancription factors also have a direct protein-protein
interaction. In summary we have shown that Runx activity can be modified in both a
ligand and receptor specific manner in various contexts.
2. Overview of Findings in the ER/Runx2 Project
We found that the ! form of ER inhibited Runx2 activity and " did not in
transient transfections. This difference may be due to the presence of differences in the
NTD and LBD of these receptors, since the DBD of ER! and " are very similar. Runx1
was not inhibited by either ! or " form of the estrogen receptors in our transient assays.
164
Next we wanted to confirm these observations within a more endogenous setting.
In both MC3T3-E1 and MCF7 cells, that contain Runx2 and ER!, Runx2 target
genes were inhibited by E2 treatment. Suggesting, that the relationship seen in transient
transfections was also present endogenously in cells. Our data mining of microarray
studies examining breast cancer patients further confirms the inverse relationship
between ER! activity and Runx2 activity. Suggesting that what we see in the test tube
may have some consequence in a more natural setting than cell lines. This may lead to
the development of biomarkers such as the MMPs, BSP, and SPP1 to name a few of the
40 genes examined in approximately 800 patient samples across three breast cancer
studies. These Runx2 target genes may predict breast cancer status and progression.
Further work needs to be done to correlate the expression of different Runx2
target genes with the life expectancy and progression of breast cancer patients that are
either ER! positive or negative. Runx2 expression may be biphasic during breast cancer.
In the first phase Runx2 may be inhibited by ER! thereby keeping Runx2 from
differentiating these cells into an osteoblast phenotype. In the second phase, in ER!
negative cancer cells, Runx2 is active and may become pro-metastatic (Figure 5.1) . The
ER negative breast cancer cell line, MDA-MB-231 may represent a model for the latter
phase.
The phenomenon of Runx2 inhibition by ER! is clearly evident from the data
presented in this thesis. One model that can be suggested by our data is that ER! directly
interacts with Runx2 and thereby inhibiting Runx2 activity but not vice-versa. The
165
problem with this model is that it does not take into account the varying activity that is
seen with the SERMS. A possible reason for this apparent discrepancy may be that the
interaction between the two transcription factors is molecularly different when SERMs
are present. Another more probable explanation that better accounts for the varying levels
of modulation of Runx2 by ER!, is that ER! and Runx2 are not the only molecules in
the complex and that differential recruitment of such other molecules result in differential
effects on RUNX2 activity. Such different coactivators and/or corepressors may
modulate the Runx2 activity dependent on what ligand is bound to ER. It is known that
different corepressor and coactivators assemble on the ER LBD dependent on what
ligand is bound. This is probable since Runx2 and ER share several coactivators
(unpublished data, Khalid, Purcell & Stallcup et al). We found that coactivators, once
known only as secondary or tertiary coactivators for ER, may also directly bind to, and
modify Runx2 structurally and functionally (Figure 5.2). We found that Runx2 could
interact with coactivators as GRIP1, COCOA, PGC1!, and PRMT1 (Figure 5.2). CCAR1
interaction was used as control to show that not all ER! coactivators could interact with
Runx2 (Figure 5.2). A detailed understanding of how ER! is modulating Runx2 most
likely will depend on an appreciation of the cellular context and a detailed understanding
of the molecular architecture of the interacting complexes in the nucleus and/or on DNA.
166
Figure 5.1 Cartoon depicting the development of ER negative Breast Cancer
Figure 5.1 Legend. ER and Runx2 are in balance in a normal cell. In ER positive cancers
Runx2 activity is inhibited, however in ER negative cells Runx2 activity becomes pro-
metastatic and Runx2 activity is not hindered since ER is not present.
167
Figure 5.2. Immunoprecipitation of Runx2 with Coactivators
Figure 5.2 Legend. Runx2 and the coactivator listed on the left was coexpressed in Cos7
cells and immunoprecipitate with Runx2 antibody and immunoblotted with an HA
antibody.
Input 15%
IgG
IP RUNX2
HA-GRIP1
HA- CCAR1
HA- COCOA
HA- PRMT1
HA- PGC1!
168
3. Overview of the Findings from the AR/Runx1,2 Project
The AR was able to inhibit both Runx1 and Runx2 activity. We subsequently
focused on Runx2 inhibition and discovered that the DBD of the AR interacted directly
with and Runx2. The interaction was ionic and required the Runt and PST domains of
Runx2 for the interaction. Even though there is a direct protein-protein interaction
between the AR and Runx2. It is important to note that it was the zinc finger region of the
AR that was able to interact with Runx2. The SARMS had no effect other than inhibition
of Runx2 activity. Runx2 could never be activated in the presence of AR in any
condition. The interaction between AR and Runx2 may be so strong that it may
disassociate any other coactivator complexes from binding, therefore not allowing the
activation of Runx2 but only inhibition. Unfortunately, in this study it was difficult to
find a strong correlation between inhibition of Runx2 target genes and activation of the
AR in prostate cancer patients. Further work needs to be done to clarify how Runx2
activity is being affected in prostate cancer patients.
Both Runx1 and Runx2 inhibited the AR. This was shown in LNCaP and Cos7
cells. This makes sense since the AR DBD is what directly interacts with Runx2. The
mechanism of how Runx2 and AR is being inhibited by each other has also been
observed. The data suggests that Runx2 is not able to bind to its DNA element by the
presence of AR, and vice versa. This was shown in several ways including:1) EMSAs for
both Runx2 binding being abrogated by AR and AR binding being abrogated by Runx2
to their respective DNA elements, 2) FRAP experiments showing an increase in mobility
when both proteins are present (increase in mobility for both the AR and Runx2), and 3)
169
ChIP experiments showing a decrease in Runx2 occupancy upon DHT treatment in cells
where both AR and Runx2 are present.
To further look at the complexity in this interaction two point mutants were
examined the A573D mutant, found in patients with androgen insensitivity syndrome
(AIS) and the T877A mutant found in prostate cancer patients that have become resistant
to ablation therapy, especially OHF. The A573D mutant was not able to inhibit Runx2
activity as the wildtype AR, and the T877A mutant was more promiscuous in its
inhibition since OHF was able to inhibit Runx2 activity in the presence of the T877A
mutant and not the wildtype. It would be interesting to examine studies of AIS patients or
Tfm patients to examine the status of Runx2 genes. Similarly, it would be interesting to
examine Runx2 target genes in OHF resistant prostate cancers. Thereby, linking a clinical
correlation from “the bench to the bedside.”
4. Future Directions
There are several directions that these projects can go. The information presented
here is only a start in a list of things that can be done to help develop and appreciate the
findings here in a molecular as well as a clinical setting. A brief overview how these
projects can be developed will be presented here.
4.1 Clinical Setting
The eventual goal of the research presented here, is to ultimately have a clinical
impact. Prostate and breast cancer, as well as malformative bone diseases may benefit
170
from a better understanding of steroid receptor and Runx activities and how they may
interrelate. The ideas developed in this thesis may impact clinic practice. For example
high-throughput screens for new SERMS and SARMS may reveal how they modulate
Runx activity. This will provide new perspectives of how these drugs may effect bone
and progression of both prostate and breast cancers.
Another more complex method is the design of new drugs based on the
interactions that have been presented here. It has been found that a minimal region of
Runx2 in the PST domain interacts with ER!, drugs can be designed based on that
domain to alleviate that interaction from occurring in ER! positive breast cancer cells.
The peptide region necessary for the Runx2/ER! can be taken and formed into a
peptidomimetic with the appropriate cyclization to allow for stability of the drug. These
drugs can all be tested in silico using supercomputers that model this interaction in
programs such as GOLD, to ensure that proper interactions are taking place to allow for
the inhibition (Figure 5.3).
A similar approach can be taken for the AR/Runx2 interaction, with a greater
consequence since AR itself can be inhibited by Runx2. A region known as the “F-
region” in Runx2 has been determined as both necessary and sufficient in the AR/Runx2
interaction. This “F-region” can be further developed into a drug. Preliminary data
suggests that Runx2 has the capacity to inhibit cancer cell proliferation when infected
within LNCaP prostate cancer cells (Figure 5.4). Leading to the possibility that this may
be a viable method in drug development and inhibition of cancers.
171
An aspect of this project that has not been examined and developed especially for
a clinical setting, is how the AR is regulating Runx1 and how that may impact patients
suffering from various leukemias. Since Runx1 activity is altered in both directions
during leukemia settings AR may play a role in how Runx1 is behaving in these patients.
Also how these steroid receptors are interacting with the Runx1 translocated isoforms
that have been implicated in cancers may be of great interest to us in gaining a different
perspective on how these cancers are behaving at a molecular level.
172
Figure 5.3. Cartoon of the Strategy in AR/Runx2 Drug Development
Figure 5.3. Cartoon of the strategy possible in developing drugs based on the AR/Runx2
interaction. Since the minimal region of AR/Runx2 interaction is known it is possible to
take this interaction and develop it insilico into compounds able to inhibit AR activity.
173
Figure 5.4 LNCaP cells overexpressing Runx2 Undergoing Cell Death
Figure 5.4 Legend. A) LNCaP cells overexpressing GFP (top left panel) were healthy
compared to LNCaP cells overexpression Runx2 (top right panel). Bottom panels show
the presence of GFP in GFP positive cells. B) Immunoblot assessing the expression of
Runx2 in LNCaP cells infected with Flag-Runx2. Actin is in the bottom panel.
7 days post infection
130
55
43
95
55
!-Flag
!-Actin
1 2
2 LNCaP with Flag-Runx2
GFP Runx2
GFP
A B
1 LNCaP with GFP
174
4. 2 Molecular Setting
On a molecular level a lot is unknown about Runx2 binding to its cognitive DNA
elements, especially in the presence of hormones such as E2 and DHT, especially in the
light of the work presented here. It would be exciting to discover how Runx2 binds to its
DNA elements on a genome-wide scale in the presence and absence of steroid hormone
activity. Runx2 may be regulating transcription in a very different way depending on the
cellular context. The different Runx2 target genes discovered in mice, may behave
completely differently in cancer cells such as PC3, MDA-MB-231, or MCF7, which all
express Runx2 activity. Illumina studies can be done on MC3T3-E1 cells in various
hormone treatments in conjunction with Runx2 knockdowns since these cells have ER!,
AR, and Runx2. With the proper control it will be possible to determine what Runx2
target genes are relevant in the different hormone conditions. Also, we can knockdown
Runx2 in PC3 and MDA-MB-231 cells to assess by Illumina what Runx2 target genes are
present in a prostate and breast cancer setting.
Another part of the project that has not been examined is the accessibility of
Runx2 DNA binding sites through epigenetic changes and how that accessibility may be
affected by hormone treatment. Studies show that as cells differentiate, Runx2 occupancy
increases as well as the histone acetylation status of these sites changes (Stein). It has
been shown here that C4-2B cells (cells that have endogenous active AR) upon DHT
treatment have the ability to modify the nuclear structure and alter the cellular location of
these epigenetic sites. It may be possible that the epigenetic status of Runx2 sites is also
being altered, such as histone methylation marks on H3K4 sites throughout the
175
cell that are implicated in the status of active and inactive genes. Therefore, examining
these same histone modifications at Runx2 sites may be of interest when treating cells
with hormone in cultures with the capacity to differentiate.
It has been shown that Runx2 activity can be modulated in various ways by the
activity of steroid receptors and by hormones specific to that receptor. We have not
examined how a Runx site adjacent to an ARE site may affect transcription of a gene
when both AR and Runx proteins are present. Reports have shown that the PSA promoter
has both AREs (which is well documented) and Runx binding sites, and that Runx1 may
have the capacity to regulate PSA expression. Similar studies can be done on other genes
that have both Runx sites and AREs or EREs.
5. Concluding Remarks
The inhibition of Runx proteins by steroid hormone receptors is examined in the
work presented here. Although much detail of the molecular mechanisms and
physiological consequences of this inhibition was revealed in the work of this thesis,
several other mechanisms are still not understood. They include how Runx2 is inhibited
by ER!, how Runx1 is inhibited by AR, and why only specific steroid hormone receptors
have the capacity to modulate Runx1 or Runx2? Also we haven’t examined how Runx3
may be affected by steroid hormone receptors and vice versa.
There are several layers of complexity that have been observed; for example the
difficulty to simply find a cell line that has both of these proteins present. This suggests
that when a cell evolves a particular pathway it is more beneficial to either
176
maintain their hormone receptor status and/or activity or to maintain its Runx status
and/or activity than to have both. And if both the steroid hormone receptor pathway and
Runx pathways are present within the cell, it is only under very specific conditions that
both can have a functional consequence on each other’s activity. Obviously, the cellular
environment has a lot to do with which path the cell will take. Overall the projects
presented here help in understanding several molecular aspects of the interaction between
Runx and steroid hormone receptors leading to the development of new ideas and
perspectives in both a molecular and clinical setting.
177
References
1. Banerjee, C., L. R. McCabe, J. Y. Choi, S. W. Hiebert, J. L. Stein, G. S. Stein,
and J. B. Lian. 1997. Runt homology domain proteins in osteoblast
differentiation: AML3/CBFA1 is a major component of a bone-specific complex.
J Cell Biochem 66:1-8.
2. Barnes, G. L., K. E. Hebert, M. Kamal, A. Javed, T. A. Einhorn, J. B. Lian,
G. S. Stein, and L. C. Gerstenfeld. 2004. Fidelity of Runx2 activity in breast
cancer cells is required for the generation of metastases-associated osteolytic
disease. Cancer Res 64:4506-13.
3. Bellido, T., A. A. Ali, L. I. Plotkin, Q. Fu, I. Gubrij, P. K. Roberson, R. S.
Weinstein, C. A. O'Brien, S. C. Manolagas, and R. L. Jilka. 2003. Proteasomal
degradation of Runx2 shortens parathyroid hormone-induced anti-apoptotic
signaling in osteoblasts. A putative explanation for why intermittent
administration is needed for bone anabolism. J Biol Chem 278:50259-72.
4. Bent, S., C. Kane, K. Shinohara, J. Neuhaus, E. S. Hudes, H. Goldberg, and
A. L. Avins. 2006. Saw palmetto for benign prostatic hyperplasia. N Engl J Med
354:557-66.
5. Berardi, M. J., C. Sun, M. Zehr, F. Abildgaard, J. Peng, N. A. Speck, and J.
H. Bushweller. 1999. The Ig fold of the core binding factor alpha Runt domain is
a member of a family of structurally and functionally related Ig-fold DNA-
binding domains. Structure 7:1247-56.
6. Berrevoets, C. A., A. Umar, and A. O. Brinkmann. 2002. Antiandrogens:
selective androgen receptor modulators. Mol Cell Endocrinol 198:97-103.
7. Blackwood, E. M., and J. T. Kadonaga. 1998. Going the distance: a current
view of enhancer action. Science 281:60-3.
8. Blyth, K., E. R. Cameron, and J. C. Neil. 2005. The RUNX genes: gain or loss
of function in cancer. Nat Rev Cancer 5:376-87.
9. Brenner, O., D. Levanon, V. Negreanu, O. Golubkov, O. Fainaru, E. Woolf,
and Y. Groner. 2004. Loss of Runx3 function in leukocytes is associated with
spontaneously developed colitis and gastric mucosal hyperplasia. Proc Natl Acad
Sci U S A 101:16016-21.
178
10. Brown, C. J., S. J. Goss, D. B. Lubahn, D. R. Joseph, E. M. Wilson, F. S.
French, and H. F. Willard. 1989. Androgen receptor locus on the human X
chromosome: regional localization to Xq11-12 and description of a DNA
polymorphism. Am J Hum Genet 44:264-9.
11. Brubaker, K. D., R. L. Vessella, L. G. Brown, and E. Corey. 2003. Prostate
cancer expression of runt-domain transcription factor Runx2, a key regulator of
osteoblast differentiation and function. Prostate 56:13-22.
12. Buchanan, G., P. S. Craft, M. Yang, A. Cheong, J. Prescott, L. Jia, G. A.
Coetzee, and W. D. Tilley. 2004. PC-3 cells with enhanced androgen receptor
signaling: a model for clonal selection in prostate cancer. Prostate 60:352-66.
13. Cameron, E. R., and J. C. Neil. 2004. The Runx genes: lineage-specific
oncogenes and tumor suppressors. Oncogene 23:4308-14.
14. Canon, J., and U. Banerjee. 2003. In vivo analysis of a developmental circuit for
direct transcriptional activation and repression in the same cell by a Runx protein.
Genes Dev 17:838-43.
15. Castro-Rivera, E., I. Samudio, and S. Safe. 2001. Estrogen regulation of cyclin
D1 gene expression in ZR-75 breast cancer cells involves multiple enhancer
elements. J Biol Chem 276:30853-61.
16. Chan, K. K., N. Wei, S. S. Liu, L. Xiao-Yun, A. N. Cheung, and H. Y. Ngan.
2008. Estrogen receptor subtypes in ovarian cancer: a clinical correlation. Obstet
Gynecol 111:144-51.
17. Chen, D., S. M. Huang, and M. R. Stallcup. 2000. Synergistic, p160
coactivator-dependent enhancement of estrogen receptor function by CARM1 and
p300. J Biol Chem 275:40810-6.
18. Cheville, J. C., H. Zincke, C. M. Lohse, T. J. Sebo, D. Riehle, A. L. Weaver,
and M. L. Blute. 2002. pT1 clear cell renal cell carcinoma: a study of the
association between MIB-1 proliferative activity and pathologic features and
cancer specific survival. Cancer 94:2180-4.
19. Coffman, J. A. 2003. Runx transcription factors and the developmental balance
between cell proliferation and differentiation. Cell Biol Int 27:315-24.
20. Cohen, R. J., B. A. Shannon, M. Phillips, R. E. Moorin, T. M. Wheeler, and
K. L. Garrett. 2008. Central zone carcinoma of the prostate gland: a distinct
tumor type with poor prognostic features. J Urol 179:1762-7; discussion 1767.
179
21. De Bosscher, K., W. Vanden Berghe, and G. Haegeman. 2001. Glucocorticoid
repression of AP-1 is not mediated by competition for nuclear coactivators. Mol
Endocrinol 15:219-27.
22. Desmedt, C., and C. Sotiriou. 2006. Proliferation: the most prominent predictor
of clinical outcome in breast cancer. Cell Cycle 5:2198-202.
23. Di Gregorio, G. B., M. Yamamoto, A. A. Ali, E. Abe, P. Roberson, S. C.
Manolagas, and R. L. Jilka. 2001. Attenuation of the self-renewal of transit-
amplifying osteoblast progenitors in the murine bone marrow by 17 beta-
estradiol. J Clin Invest 107:803-12.
24. Ducy, P., and G. Karsenty. 1995. Two distinct osteoblast-specific cis-acting
elements control expression of a mouse osteocalcin gene. Mol Cell Biol 15:1858-
69.
25. Ducy, P., M. Starbuck, M. Priemel, J. Shen, G. Pinero, V. Geoffroy, M.
Amling, and G. Karsenty. 1999. A Cbfa1-dependent genetic pathway controls
bone formation beyond embryonic development. Genes Dev 13:1025-36.
26. Ducy, P., R. Zhang, V. Geoffroy, A. L. Ridall, and G. Karsenty. 1997.
Osf2/Cbfa1: a transcriptional activator of osteoblast differentiation. Cell 89:747-
54.
27. Duss, S., S. Andre, A. L. Nicoulaz, M. Fiche, H. Bonnefoi, C. Brisken, and R.
D. Iggo. 2007. An oestrogen-dependent model of breast cancer created by
transformation of normal human mammary epithelial cells. Breast Cancer Res
9:R38.
28. Fannon, S. A., R. M. Vidaver, and S. A. Marts. 2001. An abridged history of
sex steroid hormone receptor action. J Appl Physiol 91:1854-9.
29. Farla, P., R. Hersmus, J. Trapman, and A. B. Houtsmuller. 2005.
Antiandrogens prevent stable DNA-binding of the androgen receptor. J Cell Sci
118:4187-98.
30. Fondon, J. W., 3rd, and H. R. Garner. 2004. Molecular origins of rapid and
continuous morphological evolution. Proc Natl Acad Sci U S A 101:18058-63.
31. Frasor, J., J. M. Danes, C. C. Funk, and B. S. Katzenellenbogen. 2005.
Estrogen down-regulation of the corepressor N-CoR: mechanism and implications
for estrogen derepression of N-CoR-regulated genes. Proc Natl Acad Sci U S A
102:13153-7.
180
32. Fu, Q., S. C. Manolagas, and C. A. O'Brien. 2006. Parathyroid hormone
controls receptor activator of NF-kappaB ligand gene expression via a distant
transcriptional enhancer. Mol Cell Biol 26:6453-68.
33. Gardiner-Garden, M., and M. Frommer. 1987. CpG islands in vertebrate
genomes. J Mol Biol 196:261-82.
34. Gelmann, E. P. 2002. Molecular biology of the androgen receptor. J Clin Oncol
20:3001-15.
35. Geoffroy, V., M. Kneissel, B. Fournier, A. Boyde, and P. Matthias. 2002. High
bone resorption in adult aging transgenic mice overexpressing cbfa1/runx2 in
cells of the osteoblastic lineage. Mol Cell Biol 22:6222-33.
36. Grunstein, M., A. Hecht, G. Fisher-Adams, J. Wan, R. K. Mann, S. Strahl-
Bolsinger, T. Laroche, and S. Gasser. 1995. The regulation of euchromatin and
heterochromatin by histones in yeast. J Cell Sci Suppl 19:29-36.
37. Habtemariam, B., V. M. Anisimov, and A. D. MacKerell, Jr. 2005.
Cooperative binding of DNA and CBFbeta to the Runt domain of the CBFalpha
studied via MD simulations. Nucleic Acids Res 33:4212-22.
38. Harris, H. A., L. M. Albert, Y. Leathurby, M. S. Malamas, R. E. Mewshaw,
C. P. Miller, Y. P. Kharode, J. Marzolf, B. S. Komm, R. C. Winneker, D. E.
Frail, R. A. Henderson, Y. Zhu, and J. C. Keith, Jr. 2003. Evaluation of an
estrogen receptor-beta agonist in animal models of human disease. Endocrinology
144:4241-9.
39. Hecht, J., V. Seitz, M. Urban, F. Wagner, P. N. Robinson, A. Stiege, C.
Dieterich, U. Kornak, U. Wilkening, N. Brieske, C. Zwingman, A. Kidess, S.
Stricker, and S. Mundlos. 2007. Detection of novel skeletogenesis target genes
by comprehensive analysis of a Runx2(-/-) mouse model. Gene Expr Patterns
7:102-12.
40. Hirai, H., I. M. Samokhvalov, T. Fujimoto, S. Nishikawa, J. Imanishi, and S.
Nishikawa. 2005. Involvement of Runx1 in the down-regulation of fetal liver
kinase-1 expression during transition of endothelial cells to hematopoietic cells.
Blood 106:1948-55.
41. Holzbeierlein, J., P. Lal, E. LaTulippe, A. Smith, J. Satagopan, L. Zhang, C.
Ryan, S. Smith, H. Scher, P. Scardino, V. Reuter, and W. L. Gerald. 2004.
Gene expression analysis of human prostate carcinoma during hormonal therapy
identifies androgen-responsive genes and mechanisms of therapy resistance. Am J
Pathol 164:217-27.
181
42. Hong, H., K. Kohli, M. J. Garabedian, and M. R. Stallcup. 1997. GRIP1, a
transcriptional coactivator for the AF-2 transactivation domain of steroid, thyroid,
retinoid, and vitamin D receptors. Mol Cell Biol 17:2735-44.
43. Ingvarsdottir, K., C. Edwards, M. G. Lee, J. S. Lee, D. C. Schultz, A.
Shilatifard, R. Shiekhattar, and S. L. Berger. 2007. Histone H3 K4
demethylation during activation and attenuation of GAL1 transcription in
Saccharomyces cerevisiae. Mol Cell Biol 27:7856-64.
44. Inoue, K., S. Ozaki, K. Ito, T. Iseda, S. Kawaguchi, M. Ogawa, S. C. Bae, N.
Yamashita, S. Itohara, N. Kudo, and Y. Ito. 2003. Runx3 is essential for the
target-specific axon pathfinding of trkc-expressing dorsal root ganglion neurons.
Blood Cells Mol Dis 30:157-60.
45. Ito, K., Q. Liu, M. Salto-Tellez, T. Yano, K. Tada, H. Ida, C. Huang, N. Shah,
M. Inoue, A. Rajnakova, K. C. Hiong, B. K. Peh, H. C. Han, T. Ito, M. Teh,
K. G. Yeoh, and Y. Ito. 2005. RUNX3, a novel tumor suppressor, is frequently
inactivated in gastric cancer by protein mislocalization. Cancer Res 65:7743-50.
46. Ito, Y. 2004. Oncogenic potential of the RUNX gene family: 'overview'.
Oncogene 23:4198-208.
47. Ito, Y. 2008. RUNX genes in development and cancer: regulation of viral gene
expression and the discovery of RUNX family genes. Adv Cancer Res 99:33-76.
48. Jeon, E. J., K. Y. Lee, N. S. Choi, M. H. Lee, H. N. Kim, Y. H. Jin, H. M.
Ryoo, J. Y. Choi, M. Yoshida, N. Nishino, B. C. Oh, K. S. Lee, Y. H. Lee, and
S. C. Bae. 2006. Bone morphogenetic protein-2 stimulates Runx2 acetylation. J
Biol Chem 281:16502-11.
49. Jia, L., H. C. Shen, M. Wantroba, O. Khalid, G. Liang, Q. Wang, E.
Gentzschein, J. K. Pinski, F. Z. Stanczyk, P. A. Jones, and G. A. Coetzee.
2006. Locus-wide chromatin remodeling and enhanced androgen receptor-
mediated transcription in recurrent prostate tumor cells. Mol Cell Biol 26:7331-
41.
50. Jilka, R. L., K. Takahashi, M. Munshi, D. C. Williams, P. K. Roberson, and
S. C. Manolagas. 1998. Loss of estrogen upregulates osteoblastogenesis in the
murine bone marrow. Evidence for autonomy from factors released during bone
resorption. J Clin Invest 101:1942-50.
182
51. Jones, D. H., T. Nakashima, O. H. Sanchez, I. Kozieradzki, S. V. Komarova,
I. Sarosi, S. Morony, E. Rubin, R. Sarao, C. V. Hojilla, V. Komnenovic, Y. Y.
Kong, M. Schreiber, S. J. Dixon, S. M. Sims, R. Khokha, T. Wada, and J. M.
Penninger. 2006. Regulation of cancer cell migration and bone metastasis by
RANKL. Nature 440:692-6.
53. Kahler, R. A., and J. J. Westendorf. 2003. Lymphoid enhancer factor-1 and
beta-catenin inhibit Runx2-dependent transcriptional activation of the osteocalcin
promoter. J Biol Chem 278:11937-44.
54. Kanno, T., Y. Kanno, L. F. Chen, E. Ogawa, W. Y. Kim, and Y. Ito. 1998.
Intrinsic transcriptional activation-inhibition domains of the polyomavirus
enhancer binding protein 2/core binding factor alpha subunit revealed in the
presence of the beta subunit. Mol Cell Biol 18:2444-54.
55. Karsenty, G. 2008. Transcriptional control of skeletogenesis. Annu Rev
Genomics Hum Genet 9:183-96.
56. Kato, S., T. Sato, T. Watanabe, S. Takemasa, Y. Masuhiro, F. Ohtake, and T.
Matsumoto. 2005. Function of nuclear sex hormone receptors in gene regulation.
Cancer Chemother Pharmacol 56 Suppl 1:4-9.
57. Katoh, H., Y. Ogino, and G. Yamada. 2006. Cloning and expression analysis of
androgen receptor gene in chicken embryogenesis. FEBS Lett 580:1607-15.
58. Kawate, H., Y. Wu, K. Ohnaka, and R. Takayanagi. 2007. Mutual
transactivational repression of Runx2 and the androgen receptor by an impairment
of their normal compartmentalization. J Steroid Biochem Mol Biol 105:46-56.
59. Kharode, Y., P. V. Bodine, C. P. Miller, C. R. Lyttle, and B. S. Komm. 2008.
The Pairing of a Selective Estrogen Receptor Modulator, Bazedoxifene, With
Conjugated Estrogens as a New Paradigm for the Treatment of Menopausal
Symptoms and Osteoporosis Prevention. Endocrinology.
60. Komori, T. 2006. Regulation of osteoblast differentiation by transcription factors.
J Cell Biochem 99:1233-9.
61. Komori, T., H. Yagi, S. Nomura, A. Yamaguchi, K. Sasaki, K. Deguchi, Y.
Shimizu, R. T. Bronson, Y. H. Gao, M. Inada, M. Sato, R. Okamoto, Y.
Kitamura, S. Yoshiki, and T. Kishimoto. 1997. Targeted disruption of Cbfa1
results in a complete lack of bone formation owing to maturational arrest of
osteoblasts [see comments]. Cell 89:755-64.
183
62. Kousteni, S., T. Bellido, L. I. Plotkin, C. A. O'Brien, D. L. Bodenner, L. Han,
K. Han, G. B. DiGregorio, J. A. Katzenellenbogen, B. S. Katzenellenbogen, P.
K. Roberson, R. S. Weinstein, R. L. Jilka, and S. C. Manolagas. 2001.
Nongenotropic, sex-nonspecific signaling through the estrogen or androgen
receptors: dissociation from transcriptional activity. Cell 104:719-30.
63. Kumar, R., and E. B. Thompson. 2005. Gene regulation by the glucocorticoid
receptor: structure:function relationship. J Steroid Biochem Mol Biol 94:383-94.
64. Kumaran, R. I., R. Thakar, and D. L. Spector. 2008. Chromatin dynamics and
gene positioning. Cell 132:929-34.
65. Lavery, D. N., and I. J. McEwan. 2005. Structure and function of steroid
receptor AF1 transactivation domains: induction of active conformations.
Biochem J 391:449-64.
66. Lee, H. W., J. H. Suh, A. Y. Kim, Y. S. Lee, S. Y. Park, and J. B. Kim. 2006.
Histone deacetylase 1-mediated histone modification regulates osteoblast
differentiation. Mol Endocrinol 20:2432-43.
67. Lee, M. H., Y. J. Kim, H. J. Kim, H. D. Park, A. R. Kang, H. M. Kyung, J. H.
Sung, J. M. Wozney, H. J. Kim, and H. M. Ryoo. 2003. BMP-2-induced Runx2
expression is mediated by Dlx5, and TGF-beta 1 opposes the BMP-2-induced
osteoblast differentiation by suppression of Dlx5 expression. J Biol Chem
278:34387-94.
68. Lee, N. K., and G. Karsenty. 2008. Reciprocal regulation of bone and energy
metabolism. Trends Endocrinol Metab 19:161-6.
69. Levanon, D., and Y. Groner. 2004. Structure and regulated expression of
mammalian RUNX genes. Oncogene 23:4211-9.
70. Li, L. C., R. Chui, K. Nakajima, B. R. Oh, H. C. Au, and R. Dahiya. 2000.
Frequent methylation of estrogen receptor in prostate cancer: correlation with
tumor progression. Cancer Res 60:702-6.
71. Lian, J. B., J. L. Stein, G. S. Stein, A. J. van Wijnen, M. Montecino, A. Javed,
S. Gutierrez, J. Shen, S. K. Zaidi, and H. Drissi. 2003. Runx2/Cbfa1 functions:
diverse regulation of gene transcription by chromatin remodeling and co-
regulatory protein interactions. Connect Tissue Res 44 Suppl 1:141-8.
184
72. Lin, C. Y., A. Strom, V. B. Vega, S. L. Kong, A. L. Yeo, J. S. Thomsen, W. C.
Chan, B. Doray, D. K. Bangarusamy, A. Ramasamy, L. A. Vergara, S. Tang,
A. Chong, V. B. Bajic, L. D. Miller, J. A. Gustafsson, and E. T. Liu. 2004.
Discovery of estrogen receptor alpha target genes and response elements in breast
tumor cells. Genome Biol 5:R66.
73. Lin, H. Y., R. Hopkins, H. J. Cao, H. Y. Tang, C. Alexander, F. B. Davis, and
P. J. Davis. 2005. Acetylation of nuclear hormone receptor superfamily members:
thyroid hormone causes acetylation of its own receptor by a mitogen-activated
protein kinase-dependent mechanism. Steroids 70:444-9.
74. Lindsay, R. 1991. Estrogens, bone mass, and osteoporotic fracture. Am J Med
91:10S-13S.
75. Liu, P., S. A. Tarle, A. Hajra, D. F. Claxton, P. Marlton, M. Freedman, M. J.
Siciliano, and F. S. Collins. 1993. Fusion between transcription factor CBF
beta/PEBP2 beta and a myosin heavy chain in acute myeloid leukemia. Science
261:1041-4.
76. Liu, W., S. Toyosawa, T. Furuichi, N. Kanatani, C. Yoshida, Y. Liu, M.
Himeno, S. Narai, A. Yamaguchi, and T. Komori. 2001. Overexpression of
Cbfa1 in osteoblasts inhibits osteoblast maturation and causes osteopenia with
multiple fractures. J Cell Biol 155:157-66.
77. Luppen, C. A., E. Smith, L. Spevak, A. L. Boskey, and B. Frenkel. 2003. Bone
morphogenetic protein-2 restores mineralization in glucocorticoid-inhibited
MC3T3-E1 osteoblast cultures. J Bone Miner Res 18:1186-97.
78. Manolagas, S. C., S. Kousteni, and R. L. Jilka. 2002. Sex steroids and bone.
Recent Prog Horm Res 57:385-409.
79. Maruyama, Z., C. A. Yoshida, T. Furuichi, N. Amizuka, M. Ito, R.
Fukuyama, T. Miyazaki, H. Kitaura, K. Nakamura, T. Fujita, N. Kanatani,
T. Moriishi, K. Yamana, W. Liu, H. Kawaguchi, K. Nakamura, and T.
Komori. 2007. Runx2 determines bone maturity and turnover rate in postnatal
bone development and is involved in bone loss in estrogen deficiency. Dev Dyn
236:1876-90.
80. Maston, G. A., S. K. Evans, and M. R. Green. 2006. Transcriptional regulatory
elements in the human genome. Annu Rev Genomics Hum Genet 7:29-59.
81. McCarthy, T. L., W. Z. Chang, Y. Liu, and M. Centrella. 2003. Runx2
integrates estrogen activity in osteoblasts. J Biol Chem 278:43121-9.
185
82. Mendelsohn, M. E., and R. H. Karas. 1999. The protective effects of estrogen
on the cardiovascular system. N Engl J Med 340:1801-11.
83. Meyers, S., J. R. Downing, and S. W. Hiebert. 1993. Identification of AML-1
and the (8;21) translocation protein (AML-1/ETO) as sequence-specific DNA-
binding proteins: the runt homology domain is required for DNA binding and
protein-protein interactions. Mol Cell Biol 13:6336-45.
84. Michaud, J., F. Wu, M. Osato, G. M. Cottles, M. Yanagida, N. Asou, K.
Shigesada, Y. Ito, K. F. Benson, W. H. Raskind, C. Rossier, S. E.
Antonarakis, S. Israels, A. McNicol, H. Weiss, M. Horwitz, and H. S. Scott.
2002. In vitro analyses of known and novel RUNX1/AML1 mutations in
dominant familial platelet disorder with predisposition to acute myelogenous
leukemia: implications for mechanisms of pathogenesis. Blood 99:1364-72.
85. Miyoshi, H., K. Shimizu, T. Kozu, N. Maseki, Y. Kaneko, and M. Ohki. 1991.
t(8;21) breakpoints on chromosome 21 in acute myeloid leukemia are clustered
within a limited region of a single gene, AML1. Proc Natl Acad Sci U S A
88:10431-4.
86. Morales, V., C. Giamarchi, C. Chailleux, F. Moro, V. Marsaud, S. Le
Ricousse, and H. Richard-Foy. 2001. Chromatin structure and dynamics:
functional implications. Biochimie 83:1029-39.
87. Moras, D., and H. Gronemeyer. 1998. The nuclear receptor ligand-binding
domain: structure and function. Curr Opin Cell Biol 10:384-91.
88. Morrissey, C., and R. L. Vessella. 2007. The role of tumor microenvironment in
prostate cancer bone metastasis. J Cell Biochem 101:873-86.
89. Mottet, D., and V. Castronovo. 2008. Histone deacetylases: target enzymes for
cancer therapy. Clin Exp Metastasis 25:183-9.
90. Muramatsu, M., and S. Inoue. 2000. Estrogen receptors: how do they control
reproductive and nonreproductive functions? Biochem Biophys Res Commun
270:1-10.
91. Nakamura, T., Y. Imai, T. Matsumoto, S. Sato, K. Takeuchi, K. Igarashi, Y.
Harada, Y. Azuma, A. Krust, Y. Yamamoto, H. Nishina, S. Takeda, H.
Takayanagi, D. Metzger, J. Kanno, K. Takaoka, T. J. Martin, P. Chambon,
and S. Kato. 2007. Estrogen Prevents Bone Loss via Estrogen Receptor alpha
and Induction of Fas Ligand in Osteoclasts. Cell 130:811-23.
186
92. Neckers, L., and S. P. Ivy. 2003. Heat shock protein 90. Curr Opin Oncol
15:419-24.
93. Nemeth, A., and G. Langst. 2004. Chromatin higher order structure: opening up
chromatin for transcription. Brief Funct Genomic Proteomic 2:334-43.
94. Nguyen, D., S. V. Steinberg, E. Rouault, S. Chagnon, B. Gottlieb, L. Pinsky,
M. Trifiro, and S. Mader. 2001. A G577R mutation in the human AR P box
results in selective decreases in DNA binding and in partial androgen insensitivity
syndrome. Mol Endocrinol 15:1790-802.
95. Ning, Y. M., and D. M. Robins. 1999. AML3/CBFalpha1 is required for
androgen-specific activation of the enhancer of the mouse sex-limited protein
(Slp) gene. J Biol Chem 274:30624-30.
96. Ogbourne, S., and T. M. Antalis. 1998. Transcriptional control and the role of
silencers in transcriptional regulation in eukaryotes. Biochem J 331 ( Pt 1):1-14.
97. Olefsky, J. M. 2001. Nuclear receptor minireview series. J Biol Chem
276:36863-4.
98. Oliva, J., S. El Messaoudi, F. Pellestor, M. Fuentes, V. Georget, P. Balaguer,
V. Cavailles, F. Vignon, and E. Badia. 2005. Involvement of HP1alpha protein
in irreversible transcriptional inactivation by antiestrogens in breast cancer cells.
FEBS Lett 579:4278-86.
99. Oshimo, Y., N. Oue, Y. Mitani, H. Nakayama, Y. Kitadai, K. Yoshida, Y. Ito,
K. Chayama, and W. Yasui. 2004. Frequent loss of RUNX3 expression by
promoter hypermethylation in gastric carcinoma. Pathobiology 71:137-43.
100. Otto, F., H. Kanegane, and S. Mundlos. 2002. Mutations in the RUNX2 gene in
patients with cleidocranial dysplasia. Hum Mutat 19:209-16.
101. Otto, F., M. Lubbert, and M. Stock. 2003. Upstream and downstream targets of
RUNX proteins. J Cell Biochem 89:9-18.
102. Otto, F., A. P. Thornell, T. Crompton, A. Denzel, K. C. Gilmour, I. R.
Rosewell, G. W. Stamp, R. S. Beddington, S. Mundlos, B. R. Olsen, P. B.
Selby, and M. J. Owen. 1997. Cbfa1, a candidate gene for cleidocranial
dysplasia syndrome, is essential for osteoblast differentiation and bone
development [see comments]. Cell 89:765-71.
187
103. Pearce, S. T., and V. C. Jordan. 2004. The biological role of estrogen receptors
alpha and beta in cancer. Crit Rev Oncol Hematol 50:3-22.
104. Pelletier, N., N. Champagne, S. Stifani, and X. J. Yang. 2002. MOZ and
MORF histone acetyltransferases interact with the Runt-domain transcription
factor Runx2. Oncogene 21:2729-40.
105. Persson, I. 2000. Estrogens in the causation of breast, endometrial and ovarian
cancers - evidence and hypotheses from epidemiological findings. J Steroid
Biochem Mol Biol 74:357-64.
106. Platet, N., A. M. Cathiard, M. Gleizes, and M. Garcia. 2004. Estrogens and
their receptors in breast cancer progression: a dual role in cancer proliferation and
invasion. Crit Rev Oncol Hematol 51:55-67.
107. Poujol, N., J. M. Lobaccaro, L. Chiche, S. Lumbroso, and C. Sultan. 1997.
Functional and structural analysis of R607Q and R608K androgen receptor
substitutions associated with male breast cancer. Mol Cell Endocrinol 130:43-51.
108. Pratap, J., A. Javed, L. R. Languino, A. J. van Wijnen, J. L. Stein, G. S.
Stein, and J. B. Lian. 2005. The Runx2 osteogenic transcription factor regulates
matrix metalloproteinase 9 in bone metastatic cancer cells and controls cell
invasion. Mol Cell Biol 25:8581-91.
109. Pratap, J., J. B. Lian, A. Javed, G. L. Barnes, A. J. van Wijnen, J. L. Stein,
and G. S. Stein. 2006. Regulatory roles of Runx2 in metastatic tumor and cancer
cell interactions with bone. Cancer Metastasis Rev 25:589-600.
110. Ray, P., S. K. Ghosh, D. H. Zhang, and A. Ray. 1997. Repression of
interleukin-6 gene expression by 17 beta-estradiol: inhibition of the DNA-binding
activity of the transcription factors NF-IL6 and NF-kappa B by the estrogen
receptor. FEBS Lett 409:79-85.
111. Redon, C., D. Pilch, E. Rogakou, O. Sedelnikova, K. Newrock, and W.
Bonner. 2002. Histone H2A variants H2AX and H2AZ. Curr Opin Genet Dev
12:162-9.
112. Rhodes, D. R., S. Kalyana-Sundaram, V. Mahavisno, R. Varambally, J. Yu,
B. B. Briggs, T. R. Barrette, M. J. Anstet, C. Kincead-Beal, P. Kulkarni, S.
Varambally, D. Ghosh, and A. M. Chinnaiyan. 2007. Oncomine 3.0: genes,
pathways, and networks in a collection of 18,000 cancer gene expression profiles.
Neoplasia 9:166-80.
188
113. Ribatti, D., G. Mangialardi, and A. Vacca. 2006. Stephen Paget and the 'seed
and soil' theory of metastatic dissemination. Clin Exp Med 6:145-9.
114. Rice, K. L., I. Hormaeche, and J. D. Licht. 2007. Epigenetic regulation of
normal and malignant hematopoiesis. Oncogene 26:6697-714.
115. Riggs, B. L., S. Khosla, and L. J. Melton, 3rd. 2002. Sex steroids and the
construction and conservation of the adult skeleton. Endocr Rev 23:279-302.
116. Rochefort, H., M. Glondu, M. E. Sahla, N. Platet, and M. Garcia. 2003. How
to target estrogen receptor-negative breast cancer? Endocr Relat Cancer 10:261-6.
117. Roudier, M. P., E. Corey, L. D. True, C. S. Hiagno, S. M. Ott, and R. L.
Vessell. 2004. Histological, immunophenotypic and histomorphometric
characterization of prostate cancer bone metastases. Cancer Treat Res 118:311-
39.
118. Rowe, D., S. J. Cotterill, F. M. Ross, D. J. Bunyan, S. J. Vickers, J. Bryon, D.
J. McMullan, M. J. Griffiths, J. T. Reilly, E. A. Vandenberghe, G. Wilson, A.
E. Watmore, and N. P. Bown. 2000. Cytogenetically cryptic AML1-ETO and
CBF beta-MYH11 gene rearrangements: incidence in 412 cases of acute myeloid
leukaemia. Br J Haematol 111:1051-6.
119. Sack, J. S., K. F. Kish, C. Wang, R. M. Attar, S. E. Kiefer, Y. An, G. Y. Wu,
J. E. Scheffler, M. E. Salvati, S. R. Krystek, Jr., R. Weinmann, and H. M.
Einspahr. 2001. Crystallographic structures of the ligand-binding domains of the
androgen receptor and its T877A mutant complexed with the natural agonist
dihydrotestosterone. Proc Natl Acad Sci U S A 98:4904-9.
120. Saelzler, M. P., C. C. Spackman, Y. Liu, L. C. Martinez, J. P. Harris, and M.
K. Abe. 2006. ERK8 down-regulates transactivation of the glucocorticoid
receptor through Hic-5. J Biol Chem 281:16821-32.
121. Salingcarnboriboon, R., K. Tsuji, T. Komori, K. Nakashima, Y. Ezura, and
M. Noda. 2006. Runx2 is a target of mechanical unloading to alter osteoblastic
activity and bone formation in vivo. Endocrinology 147:2296-305.
122. Saville, B., H. Poukka, M. Wormke, O. A. Janne, J. J. Palvimo, M. Stoner, I.
Samudio, and S. Safe. 2002. Cooperative coactivation of estrogen receptor alpha
in ZR-75 human breast cancer cells by SNURF and TATA-binding protein. J Biol
Chem 277:2485-97.
189
123. Saxonov, S., P. Berg, and D. L. Brutlag. 2006. A genome-wide analysis of CpG
dinucleotides in the human genome distinguishes two distinct classes of
promoters. Proc Natl Acad Sci U S A 103:1412-7.
124. Schroeder, T. M., E. D. Jensen, and J. J. Westendorf. 2005. Runx2: a master
organizer of gene transcription in developing and maturing osteoblasts. Birth
Defects Res C Embryo Today 75:213-25.
125. Schwabe, J. W., L. Chapman, J. T. Finch, and D. Rhodes. 1993. The crystal
structure of the estrogen receptor DNA-binding domain bound to DNA: how
receptors discriminate between their response elements. Cell 75:567-78.
126. Schwabe, J. W., D. Neuhaus, and D. Rhodes. 1990. Solution structure of the
DNA-binding domain of the oestrogen receptor. Nature 348:458-61.
127. Schweikert, H. U., U. W. Tunn, U. F. Habenicht, J. Arnold, T. Senge, H.
Schulze, F. H. Schroder, J. H. Blom, O. Ennemoser, W. Horniger, and et al.
1993. Effects of estrogen deprivation on human benign prostatic hyperplasia. J
Steroid Biochem Mol Biol 44:573-6.
128. Shannon, B. A., J. E. McNeal, and R. J. Cohen. 2003. Transition zone
carcinoma of the prostate gland: a common indolent tumour type that occasionally
manifests aggressive behaviour. Pathology 35:467-71.
129. Shao, W., E. K. Keeton, D. P. McDonnell, and M. Brown. 2004. Coactivator
AIB1 links estrogen receptor transcriptional activity and stability. Proc Natl Acad
Sci U S A 101:11599-604.
130. Shiau, A. K., D. Barstad, P. M. Loria, L. Cheng, P. J. Kushner, D. A. Agard,
and G. L. Greene. 1998. The structural basis of estrogen receptor/coactivator
recognition and the antagonism of this interaction by tamoxifen. Cell 95:927-37.
131. Shore, P. 2005. A role for Runx2 in normal mammary gland and breast cancer
bone metastasis. J Cell Biochem 96:484-9.
132. Shurtleff, S. A., A. Buijs, F. G. Behm, J. E. Rubnitz, S. C. Raimondi, M. L.
Hancock, G. C. Chan, C. H. Pui, G. Grosveld, and J. R. Downing. 1995.
TEL/AML1 fusion resulting from a cryptic t(12;21) is the most common genetic
lesion in pediatric ALL and defines a subgroup of patients with an excellent
prognosis. Leukemia 9:1985-9.
190
133. Sierra, J., A. Villagra, R. Paredes, F. Cruzat, S. Gutierrez, A. Javed, G.
Arriagada, J. Olate, M. Imschenetzky, A. J. Van Wijnen, J. B. Lian, G. S.
Stein, J. L. Stein, and M. Montecino. 2003. Regulation of the bone-specific
osteocalcin gene by p300 requires Runx2/Cbfa1 and the vitamin D3 receptor but
not p300 intrinsic histone acetyltransferase activity. Mol Cell Biol 23:3339-51.
134. Sims, N. A., P. Clement-Lacroix, D. Minet, C. Fraslon-Vanhulle, M.
Gaillard-Kelly, M. Resche-Rigon, and R. Baron. 2003. A functional androgen
receptor is not sufficient to allow estradiol to protect bone after gonadectomy in
estradiol receptor-deficient mice. J Clin Invest 111:1319-27.
135. Sims, N. A., S. Dupont, A. Krust, P. Clement-Lacroix, D. Minet, M. Resche-
Rigon, M. Gaillard-Kelly, and R. Baron. 2002. Deletion of estrogen receptors
reveals a regulatory role for estrogen receptors-beta in bone remodeling in
females but not in males. Bone 30:18-25.
136. Speirs, V. 2002. Oestrogen receptor beta in breast cancer: good, bad or still too
early to tell? J Pathol 197:143-7.
137. Stallcup, M. R. 2001. Role of protein methylation in chromatin remodeling and
transcriptional regulation. Oncogene 20:3014-20.
138. Sudarsanam, P., and F. Winston. 2000. The Swi/Snf family nucleosome-
remodeling complexes and transcriptional control. Trends Genet 16:345-51.
139. Syed, F., and S. Khosla. 2005. Mechanisms of sex steroid effects on bone.
Biochem Biophys Res Commun 328:688-96.
140. Takai, D., and P. A. Jones. 2002. Comprehensive analysis of CpG islands in
human chromosomes 21 and 22. Proc Natl Acad Sci U S A 99:3740-5.
141. Tamrazi, A., K. E. Carlson, and J. A. Katzenellenbogen. 2003. Molecular
sensors of estrogen receptor conformations and dynamics. Mol Endocrinol
17:2593-602.
142. Tau, K. R., T. E. Hefferan, K. M. Waters, J. A. Robinson, M. Subramaniam,
B. L. Riggs, and T. C. Spelsberg. 1998. Estrogen regulation of a transforming
growth factor-beta inducible early gene that inhibits deoxyribonucleic acid
synthesis in human osteoblasts. Endocrinology 139:1346-53.
191
143. Thirunavukkarasu, K., M. Mahajan, K. W. McLarren, S. Stifani, and G.
Karsenty. 1998. Two domains unique to osteoblast-specific transcription factor
Osf2/Cbfa1 contribute to its transactivation function and its inability to
heterodimerize with Cbfbeta. Mol Cell Biol 18:4197-208.
144. Thomas, M. C., and C. M. Chiang. 2006. The general transcription machinery
and general cofactors. Crit Rev Biochem Mol Biol 41:105-78.
145. Tomaru, T., T. Satoh, S. Yoshino, T. Ishizuka, K. Hashimoto, T. Monden, M.
Yamada, and M. Mori. 2006. Isolation and characterization of a transcriptional
cofactor and its novel isoform that bind the deoxyribonucleic acid-binding domain
of peroxisome proliferator-activated receptor-gamma. Endocrinology 147:377-88.
146. Vaes, B. L., P. Ducy, A. M. Sijbers, J. M. Hendriks, E. P. van Someren, N. G.
de Jong, E. R. van den Heuvel, W. Olijve, E. J. van Zoelen, and K. J.
Dechering. 2006. Microarray analysis on Runx2-deficient mouse embryos reveals
novel Runx2 functions and target genes during intramembranous and
endochondral bone formation. Bone 39:724-38.
147. van de Vijver, M. J., Y. D. He, L. J. van't Veer, H. Dai, A. A. Hart, D. W.
Voskuil, G. J. Schreiber, J. L. Peterse, C. Roberts, M. J. Marton, M. Parrish,
D. Atsma, A. Witteveen, A. Glas, L. Delahaye, T. van der Velde, H.
Bartelink, S. Rodenhuis, E. T. Rutgers, S. H. Friend, and R. Bernards. 2002.
A gene-expression signature as a predictor of survival in breast cancer. N Engl J
Med 347:1999-2009.
148. Verrijdt, G., E. Schoenmakers, A. Haelens, B. Peeters, G. Verhoeven, W.
Rombauts, and F. Claessens. 2000. Change of specificity mutations in
androgen-selective enhancers. Evidence for a role of differential DNA binding by
the androgen receptor. J Biol Chem 275:12298-305.
149. Walker, V. R., and K. S. Korach. 2004. Estrogen receptor knockout mice as a
model for endocrine research. Ilar J 45:455-61.
150. Walter, P. P., T. A. Owen-Hughes, J. Cote, and J. L. Workman. 1995.
Stimulation of transcription factor binding and histone displacement by
nucleosome assembly protein 1 and nucleoplasmin requires disruption of the
histone octamer. Mol Cell Biol 15:6178-87.
151. Wang, Q., T. Stacy, M. Binder, M. Marin-Padilla, A. H. Sharpe, and N. A.
Speck. 1996. Disruption of the Cbfa2 gene causes necrosis and hemorrhaging in
the central nervous system and blocks definitive hematopoiesis. Proc Natl Acad
Sci U S A 93:3444-9.
192
152. Wang, Y., J. G. Klijn, Y. Zhang, A. M. Sieuwerts, M. P. Look, F. Yang, D.
Talantov, M. Timmermans, M. E. Meijer-van Gelder, J. Yu, T. Jatkoe, E. M.
Berns, D. Atkins, and J. A. Foekens. 2005. Gene-expression profiles to predict
distant metastasis of lymph-node-negative primary breast cancer. Lancet 365:671-
9.
153. Weatherman, R. V., R. J. Fletterick, and T. S. Scanlan. 1999. Nuclear-receptor
ligands and ligand-binding domains. Annu Rev Biochem 68:559-81.
154. Weiderpass, E., I. Persson, H. Melhus, S. Wedren, A. Kindmark, and J. A.
Baron. 2000. Estrogen receptor alpha gene polymorphisms and endometrial
cancer risk. Carcinogenesis 21:623-7.
155. Westendorf, J. J., S. K. Zaidi, J. E. Cascino, R. Kahler, A. J. van Wijnen, J.
B. Lian, M. Yoshida, G. S. Stein, and X. Li. 2002. Runx2 (Cbfa1, AML-3)
interacts with histone deacetylase 6 and represses the p21(CIP1/WAF1) promoter.
Mol Cell Biol 22:7982-92.
156. Woolf, E., C. Xiao, O. Fainaru, J. Lotem, D. Rosen, V. Negreanu, Y.
Bernstein, D. Goldenberg, O. Brenner, G. Berke, D. Levanon, and Y. Groner.
2003. Runx3 and Runx1 are required for CD8 T cell development during
thymopoiesis. Proc Natl Acad Sci U S A 100:7731-6.
157. Xiao, Z., H. A. Awad, S. Liu, J. Mahlios, S. Zhang, F. Guilak, M. S. Mayo,
and L. D. Quarles. 2005. Selective Runx2-II deficiency leads to low-turnover
osteopenia in adult mice. Dev Biol 283:345-56.
158. Yaffe, K. 2001. Estrogens, selective estrogen receptor modulators, and dementia:
what is the evidence? Ann N Y Acad Sci 949:215-22.
159. Yanagida, M., M. Osato, N. Yamashita, H. Liqun, B. Jacob, F. Wu, X. Cao,
T. Nakamura, T. Yokomizo, S. Takahashi, M. Yamamoto, K. Shigesada, and
Y. Ito. 2005. Increased dosage of Runx1/AML1 acts as a positive modulator of
myeloid leukemogenesis in BXH2 mice. Oncogene 24:4477-85.
160. Yeh, S., M. Y. Tsai, Q. Xu, X. M. Mu, H. Lardy, K. E. Huang, H. Lin, S. D.
Yeh, S. Altuwaijri, X. Zhou, L. Xing, B. F. Boyce, M. C. Hung, S. Zhang, L.
Gan, and C. Chang. 2002. Generation and characterization of androgen receptor
knockout (ARKO) mice: an in vivo model for the study of androgen functions in
selective tissues. Proc Natl Acad Sci U S A 99:13498-503.
193
161. Yeung, F., W. K. Law, C. H. Yeh, J. J. Westendorf, Y. Zhang, R. Wang, C.
Kao, and L. W. Chung. 2002. Regulation of human osteocalcin promoter in
hormone-independent human prostate cancer cells. J Biol Chem 277:2468-76.
162. Yokomizo, T., K. Hasegawa, H. Ishitobi, M. Osato, M. Ema, Y. Ito, M.
Yamamoto, and S. Takahashi. 2008. Runx1 is involved in primitive
erythropoiesis in the mouse. Blood 111:4075-80.
163. Young, D. W., M. Q. Hassan, X. Q. Yang, M. Galindo, A. Javed, S. K. Zaidi,
P. Furcinitti, D. Lapointe, M. Montecino, J. B. Lian, J. L. Stein, A. J. van
Wijnen, and G. S. Stein. 2007. Mitotic retention of gene expression patterns by
the cell fate-determining transcription factor Runx2. Proc Natl Acad Sci U S A
104:3189-94.
164. Young, D. W., S. K. Zaidi, P. S. Furcinitti, A. Javed, A. J. van Wijnen, J. L.
Stein, J. B. Lian, and G. S. Stein. 2004. Quantitative signature for architectural
organization of regulatory factors using intranuclear informatics. J Cell Sci
117:4889-96.
165. Zaidi, S. K., A. Javed, J. Y. Choi, A. J. van Wijnen, J. L. Stein, J. B. Lian,
and G. S. Stein. 2001. A specific targeting signal directs Runx2/Cbfa1 to
subnuclear domains and contributes to transactivation of the osteocalcin gene. J
Cell Sci 114:3093-102.
166. Zaidi, S. K., S. Pande, J. Pratap, T. Gaur, S. Grigoriu, S. A. Ali, J. L. Stein, J.
B. Lian, A. J. van Wijnen, and G. S. Stein. 2007. Runx2 deficiency and
defective subnuclear targeting bypass senescence to promote immortalization and
tumorigenic potential. Proc Natl Acad Sci U S A 104:19861-6.
167. Zayzafoon, M., S. A. Abdulkadir, and J. M. McDonald. 2004. Notch signaling
and ERK activation are important for the osteomimetic properties of prostate
cancer bone metastatic cell lines. J Biol Chem 279:3662-70.

Abstract (if available)

Abstract Estrogen receptor α (ER α) and androgen receptor (AR) are master transcription factors in the breast and prostate, respectively. They are commonly known in development of sexual characteristics. However, both ERα and AR have been known to be involved in breast cancer (BCa) and prostate cancer (PCa) progression, respectively. The Runx family of transcription factors plays a role in hematopoiesis (Runx1), skeletogenesis (Runx2) and neurogenesis (Runx3). In addition, Runx proteins inhibit cell cycle progression, and have been assigned tumor suppressor roles in various contexts. Because both BCa and PCa cells metastasize to bone at high frequency, investigators have interrogated the possibility that they share characteristics with osteoblasts. Indeed, BCa and PCa cells were found to have "osteomimetic" properties, including expression of Runx2 and Runx2-target genes otherwise expressed by osteoblasts. Provoked by the reported physical interaction between AR and Runx2, we initiated a study to test whether ERα and AR might promote BCa and PCa progression (respectively) by inhibiting Runx2 activity through direct protein-protein interactions. We report here that ERα and AR both inhibit Runx2 transcriptional activity via interaction through specific domains in a receptor specific and ligand specific manner. In addition, it was revealed to a lesser extent that AR was able to strongly inhibit Runx1 transcriptional activity, but ERα was not. Immunohistochemistry analyses revealed that both ERα and Runx2 colocalize in the nucleus in specific subnuclear domains as well as AR and Runx2.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

RUNX2 & sex steroids: molecular mechanisms in regulating bone turnover

PDF

Runx2 interactions with the osteoblast genome

PDF

Inhibitory effects of estradiol and SERMs on RUNX2-driven osteoblast differentiation and gene expression

PDF

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

PDF

Preosteoblast-specific RUNX2 promotes RANKL membrane association: antagonism by sex steroids through a non-genomic mechanism

PDF

Functional analyses of androgen receptor structure pertaining to prostate cancer

PDF

Identification of novel androgen receptor target genes in prostate cancer

PDF

Co-chaperone influence on androgen receptor signaling and identification of androgen receptor genes in prostate cancer

PDF

The effect of chaperone/co-chaperone overexpression on steroid receptor activity

PDF

Synthetic biopolymers modulate cell signaling

PDF

Glucocorticoids and Runx2 synergistically stimulate Wif1 and compromise pre-osteoblats in vitro

PDF

The role of Hic-5 in glucocorticoid receptor binding to chromatin

PDF

Interaction of Hic-5 with different steroid receptors and its selective coregulator activity

PDF

Protein arginine methyltransferases in murine skull development

PDF

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

PDF

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

PDF

Flipping the switch on protein activity activity: elastin-like polypeptides assemble into cell switches and vesicles

PDF

Developing peptide and antibody-mimetic ligands for the cell surface receptors β2AR and DC-SIGN

PDF

The modulation of dynamin and receptor endocytosis machinery using elastin-like polypeptides

PDF

Systematic analysis of single nucleotide polymorphisms in the human steroid 5-alpha reductase type I gene

Asset Metadata

Creator Khalid, Omar (author)

Core Title Modulation of Runx proteins by steroid hormone receptors

School School of Pharmacy

Degree Doctor of Philosophy

Degree Program Pharmaceutical Sciences

Degree Conferral Date 2009-05

Publication Date 03/02/2009

Defense Date 12/03/2008

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

Tag androgen receptor,bone,breast cancer,estrogen receptor,OAI-PMH Harvest,prostate cancer,Runx,steroids

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Hamm-Alvarez, Sarah F. (committee chair), Coetzee, Gerhard A. (committee member), Frenkel, Baruch (committee member), Okamoto, Curtis Toshio (committee member)

Creator Email khalid@usc.edu,pageplant99@yahoo.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m1991

Unique identifier UC1416629

Identifier etd-Khalid-2586 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-153892 (legacy record id),usctheses-m1991 (legacy record id)

Legacy Identifier etd-Khalid-2586.pdf

Dmrecord 153892

Document Type Dissertation

Rights Khalid, Omar

Type texts

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

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu