University of Southern California Dissertations and Theses

Context-dependent role of androgen receptor (AR) in estrogen receptor-positive (ER+) breast cancer

(USC Thesis Other)

Context-dependent role of androgen receptor (AR) in estrogen receptor-positive (ER+) breast cancer

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

CONTEXT-DEPENDENT ROLE OF ANDROGEN RECEPTOR (AR) IN
ESTROGEN RECEPTOR-POSITIVE (ER+) BREAST CANCER

By

Sathish Kumar Ganesan

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

December 2022

Copyright 2022 Sathish Kumar Ganesan

ii
DEDICATION

Dedicated to the memory of Professor. Baruch Frenkel and Professor. Neil Segil

I will always be grateful to them for sharing their time with me. I hope to be as kind and
helpful as they were to people around them.

iii
ACKNOWLEDGEMENTS
To my family and friends for all their love and support through the years.
I would like to acknowledge my mentor, Dr. Min Yu, for always pushing me to be
a better scientist. I am grateful for her continued support, encouragement, but above all
for her empathy. I couldn’t have asked for a better mentor or a lab to spend an important
part of my scientific career.
I would like to thank Dr. Stallcup for serving as chair of my qualifying exam
committee and the initial years of my dissertation committee. I am grateful for his wisdom,
scientific curiosity, and kindheartedness.
To Dr. Jian Xu, for her time, resourcefulness, and helpfulness. I will always be
grateful for her help with the training grant.
To Dr. Ite Offringa for being the best cheerleader a student can ask for
I would like to thank: the PIBBS admin, Dr. Josh Neman, Dr. Junji Watanabe, and
Bernadette.
THANK YOU to members of Yu lab, you all are the BEST!

iv
TABLE OF CONTENTS

DEDICATION ................................................................................................................... ii

ACKNOWLEDGEMENTS ................................................................................................ iii

LIST OF FIGURES .......................................................................................................... vi

ABSTRACT ..................................................................................................................... ix

CHAPTER 1: ESTROGEN RECEPTOR-POSITIVE (ER+) BREAST CANCER .............. 1
1.1 Breast cancer epidemiology ............................................................................ 1
1.2 Tumor heterogeneity and classification ........................................................... 2
1.3 Estrogen receptor in breast cancer ................................................................. 4
1.4 Androgen receptor in ER+ breast cancer ........................................................ 9
1.5 Resistance to endocrine therapy and the need for new therapeutic
approaches ………..……………………………………………………………….12

CHAPTER 2: CONTEXT-DEPENDENT ROLE OF ANDROGEN RECEPTOR (AR)
IN ESTROGEN RECEPTOR-POSITIVE (ER+) BREAST CANCER ............................. 16
2.1 Introduction .................................................................................................... 16
2.2 Results .......................................................................................................... 18
2.2.1 AR is upregulated in breast cancer cell lines harboring ESR1
Y537S mutation ........................................................................................ 18
2.2.2 AR plays a context-dependent role in tumorigenesis of ER+
breast cancer cells ................................................................................... 25
2.2.3 Temporal upregulation of AR in long-term estrogen deprived
cells .......................................................................................................... 32
2.2.4 Androgen receptor supports anchorage-independent survival
in sodium pyruvate-free environment ...................................................... 44
2.3 Discussion ..................................................................................................... 50
2.4 Methods ......................................................................................................... 55

CHAPTER 3: EXPLORING THE ROLE OF AR IN BONE METATSTASIS OF
ER+ BREAST CANCER ................................................................................................. 64
3.1 Introduction .................................................................................................... 64
3.2 Results ........................................................................................................... 65
3.2.1 Identification of MBOAT2 as a novel, potential player in ER+
breast cancer bone metastasis ................................................................. 65
3.2.2 Assessing the significance of androgen receptor in ER+
bone metastasis models ........................................................................... 70
3.3 Discussion ..................................................................................................... 76
3.4 Methods ......................................................................................................... 78

CHAPTER 4: DISCUSSION ………………………………………………………………….81
v
4.1 Context-dependent divergent roles of AR ..................................................... 81
4.2 Future directions ............................................................................................ 84
4.3 Conclusions……………………………………………………………………… 85

References ..................................................................................................................... 86

vi
LIST OF FIGURES

CHAPTER 1: ESTROGEN RECEPTOR-POSITIVE (ER+) BREAST CANCER

Figure 1.1 Anatomy of the breast ..................................................................................... 2

Figure 1.2 Surrogate intrinsic classification of breast cancer ........................................... 3

Figure 1.3 Structure of ESR1, ERa, ESR2, and ERb ……………………………………….5

Figure 1.4 Estrogen and androgen receptor signaling pathways...…………………………7

Figure 1.5 Structure of androgen receptor gene and full-length protein ………………….10

Figure 1.6 ESR1 mutations in ligand binding domain ………………………………………14

CHAPTER 2: CONTEXT-DEPENDENT ROLE OF ANDROGEN RECEPTOR (AR)
IN ESTROGEN RECEPTOR-POSITIVE (ER+) BREAST CANCER

Figure 2.1: Elevated AR mRNA expression in mutant ER cell lines cultured in
full-serum media ............................................................................................................ 20

Figure 2.2: AR is upregulated in mutant ER cell lines cultured in full-serum media ....... 21

Figure 2.3: AR mRNA expression in mutant ER cell lines cultured in charcoal
stripped serum containing media (CSS media) .............................................................. 22

Figure 2.4: AR is upregulated in mutant ER cell lines cultured in charcoal-stripped
serum containing media (CSS media) ........................................................................... 23

Figure 2.5: T47D ER Y537S mutant cells show elevated nuclear expression of AR ..... 24

Figure 2.6: AR knockout in ESR1 mutant MCF7 cell lines ............................................. 28

Figure 2.7: Androgen receptor suppresses growth in presence of E2 and DHT ............ 29

Figure 2.8: AR-sg1 cells exhibit lower colony area in anchorage-independent
conditions .……………………………………………………………………………………...30

Figure 2.9: AR knockout decreases metastasis of ER Y537S cells……………………….31

Figure 2.10: AR is upregulated during long-term estrogen deprivation ……………..…...35

Figure 2.11: LTED 4 weeks: Androgen receptor suppresses growth in presence of E2
and DHT ………………………………………………………………………………………..36

vii
Figure 2.12: LTED 8 weeks: Androgen receptor suppresses growth in presence of E2
and DHT ………………………………………………………………………………………..38

Figure 2.13: LTED 12 weeks: Androgen receptor suppresses growth in presence of
E2 and DHT …..………………………………………………………………………………..40

Figure 2.14: LTED 4 weeks. AR is important for anchorage independent growth of
LTED MCF7s in soft agar assa………………………………………………………………..41

Figure 2.15: LTED 8 weeks. AR is important for anchorage independent growth of
LTED MCF7s in soft agar assay …………….………………………………………………..42

Figure 2.16: LTED 12 weeks. AR is important for anchorage independent growth of
LTED MCF7s in soft agar assay …………….………………………………………………..43

Figure 2.17: AR is important for anchorage-independent growth in the absence of
sodium pyruvate ………...……………………………………………………………………..46

Figure 2.18: Reintroduction of AR in AR KO cell lines …………………………………….47

Figure 2.19: Reintroduction of AR partially restores growth of soft agar colonies in
full-serum media with sodium pyruvate ………..…………………………………………….48

Figure 2.20: Reintroduction of AR partially restores growth of soft agar colonies in
full-serum media without sodium pyruvate …………….……………………………………49

Figure 2.21: A schematic of colony formation assay ……………………………………….59

Figure 2.22: A schematic of soft agar assay ………………………………………………..61

CHAPTER 3: EXPLORING THE ROLE OF AR IN BONE METASTASIS OF
ER+ BREAST CANCER

Figure 3.1: MBOAT2 is upregulated in bone metastatic derivatives of BRx68 ………….67

Figure 3.2: MBOAT2 RNA and protein expression from THE HUMAN PROTEIN
ATLAS ……..…………………………………………………………………………………...68

Figure 3.3: Higher MBOAT2 expression associated with lower metastasis-free
survival in breast cancer patients with bone relapse ….……………………………………69

Figure 3.4: Androgen receptor is expressed in BRx68 ……………………………………..72

Figure 3.5: Androgen receptor is expressed in BRx68-BoM ……………………………...72

Figure 3.6: Nuclear-AR positive MCF7 in bone from caudal artery mouse model………..73
viii

Figure 3.7: MCF7s in an ex vivo bone – cancer cell co-culture
system …………..……………………………………………………………………………...74

Figure 3.8: AR knockout did not affect bone colonization in ex vivo – cancer cell
co-culture assay ……….………………………………………………………………………75

ix

ABSTRACT

Metastasis is the leading cause of breast cancer deaths. Role of AR in ER+ breast
cancer is vigorously debated. Mutations in ESR1 is a common mechanism of resistance
to therapy. Identifying new therapeutic targets in resistant models is the main challenge
in the field. In mutant ER cells, AR was upregulated and localized in the nucleus. Utilizing
CRISPR-Cas9, AR knockout models were generated to precisely investigate role of AR
in mutant ER cells. AR suppresses growth in presence of E2 and DHT in adherent culture.
On the contrary, knockout of AR inhibits growth in anchorage-independent conditions. AR
knockout diminished total metastatic burden in mice. In a postmenopausal model of long-
term estrogen deprivation, AR expression increases over time in WT and mutant cells.
We hypothesized antioxidants in cell culture may be important to survive anchorage-
independent growth. Intriguingly, AR knockout abolishes anchorage-independent growth
in a sodium pyruvate-free environment. Moreover, reintroduction of AR partially rescued
growth inhibition of anchorage-independent culture in sodium pyruvate-free environment.
Collectively, our data indicate a context-dependent divergent role for AR in ER+ breast
cancer cells.

1
CHAPTER 1
ESTROGEN RECEPTOR-POSITIVE (ER+) BREAST CANCER

1.1 Breast cancer epidemiology
The most common type of malignancy in women is breast cancer. In the United
States, an estimated 287,850 cases of female breast cancer will be newly diagnosed in
2022, representing 31% of all new cancer cases in women. An estimated 15% of all
cancer mortality in women will be from breast cancer, contributing to second leading
cause of cancer-related deaths in women (Siegel et al., 2022). According to GLOBOCAN
2020 estimates of worldwide cancer incidence and mortality, breast cancer in woman has
overtaken lung cancer as the most diagnosed cancer with an estimated 2.3 million new
cases (Sung et al., 2021). According to the American Cancer Society, there is a 1 in 8
chance of a woman in the United States to develop breast cancer in her lifetime (Siegel
et al., 2022).
A great number of risk factors have been attributed to the development of breast
cancer. Non-modifiable risk factors include family history, race, age, sex, menstrual cycle,
pregnancy and genetic mutations. Modifiable factors include smoking, alcohol intake,
physical activity and obesity (Łukasiewicz et al., 2021). Next-generation sequencing
analysis of samples from a cohort of more than 113,000 women identified protein-
truncating variants in 9 genes – ATM, BRCA1, BRCA2, CHEK2, PALB2, BARD1,
RAD51C, RAD51D and TP53 – that were significantly associated with a risk of breast
cancer (Dorling et al., 2021).
2

1.2 Tumor heterogeneity and classification
The female breast consists of glands, ducts, fat and connective tissue. The
glandular tissue is composed of 15 – 20 lobes. The lobes contain smaller units, lobules,
that produce milk in response to hormonal changes in the body during pregnancy. Ducts
are present throughout the breast tissue and function to carry milk to the nipple (Figure
1.1). The terminal ductal lobular units are the functional units of breast, and most breast
cancers originate from the cells that make up these functional units (Yang et al., 2016).

Figure 1.1 Anatomy of the breast. The breast tissue consists of 15-20 lobes with smaller
milk producing units called lobules. The ducts carry milk and together with lobules form
the functional units of the breast. Majority of breast cancers originate from these functional
units. Created with BioRender.com.

3
Breast cancer is a highly heterogenous disease and can be classified based on
the histological and molecular characteristics of primary tumor. Pre-invasive breast
cancers include two histological subtypes: Ductal carcinoma in situ (DCIS) and Lobular
carcinoma in situ (LCIS). Invasive breast cancers are comprised of two histological
subtypes: Ductal carcinoma no special type (NST) and Lobular carcinoma (ILC) (Harbeck
et al., 2019). Molecular classification of breast cancer has evolved over the years. In
2000, Perou and Sorlie, generated cDNA microarray data from 65 human breast cancer
samples to identify four subtypes: ER+/luminal-like, basal-like, Erb-B2+ and normal
breast (Łukasiewicz et al., 2021). This intrinsic classification has undergone further
changes and now contain five subtypes: Luminal A, Luminal B, HER2-enriched, Basal-
like and Claudin-low. Clinically, breast cancer is classified into five surrogate subgroups
based on the expression of hormone receptors, HER2 status and proliferation markers
(Figure 1.2).
4

Figure 1.2 Surrogate intrinsic classification of breast cancer. The five subtypes of
breast cancer are based on the expression of hormone receptors, HER2, and proliferation
marker Ki67. Created with BioRender.com.

1.3 Estrogen receptor in breast cancer
Luminal breast cancers are characterized by the expression of estrogen receptor
(ER). ER+ tumors account for 60-70% of all breast cancers. There are two forms of ER –
represented as a and b. ERa is encoded by ESR1 and ERb is encoded by ESR2. Located
on chromosome 6, ESR1 consists of 8 exons that encode protein-coding domains of ERa.
ESR2 is located on chromosome 14 and consists of 8 exons that encode protein-coding
domains of ERb. There are many isoforms of ERa and ERb. The full-length ERa contains
595 amino acids and a molecular weight of 67kDa. The full-length ERb consists of 530
amino acids and a molecular weight of 59kDa. ERa and ERb are members of nuclear
receptor family. The two forms of ER share several domains in their structure. The full-
length isoform of ERa and ERb consists of four domains: A/B, C, D and E/F. The
functional region A/B represents the N-terminal domain (NTD); region C represents the
DNA binding region (DBD); region D represents hinge region; and region E/F represents
ligand binding domain (LBD). Moreover, there are two activation domains: AF1 within
NTD; AF2 within LBD (Fuentes et al., 2019) (Figure 1.3)
ERa and ERb are steroid hormone receptors activated by binding of estrogens.
ERa is the estrogen receptor of interest in the work presented here and henceforth
referred to as ER. 17b-estradiol, also referred to as estradiol or E2, is the main circulating
estrogen in humans. E2 is commonly used as ligand to activate ER in preclinical studies.
5
Estrogens are synthesized from dietary cholesterol, predominantly in the ovaries. Adrenal
glands and adipose tissue are secondary sites for estrogen synthesis (Samavat et al.,
2015). Aromatase enzyme in peripheral tissues metabolizes testosterone to estradiol.
The principal mode of action of estrogens is binding to estrogen receptors to
facilitate signaling events in the genome. E2 binds to ligand binding domain inducing a
conformational change that leads to receptor dimerization (Le Dily et al., 2019). Ligand
binding exposes nuclear localization signal in the hinge region of ER. Ligand-bound ER
translocate to nucleus and exert their genomic effects by binding to estrogen response
elements (EREs) in promoters, distal enhancers, and other regulatory components of
target genes (Figure 1.4).

Figure 1.3 Structure of ESR1, ERa, ESR2 and ERb. ESR1 has 8 exons and encodes
for ERa. ESR2 has 8 exons and encodes for ERb. Domains of ERa and ERb: A/B
containing N-terminal domain and AF-1; DNA binding domain in C; hinge region in D;
ligand binding domain and AF-2 in E/F. Created with Biorender.com.

6

The A/B region in NTD domain of the receptor facilitates binding to ERE sites.
Activation domains AF1 and AF2 are important for ER transcriptional regulation of its
target genes. In contrast to direct genome binding, ER can also exert effects on target
genes indirectly without binding to DNA. Interaction between ER and other transcription
factors and response elements is one of the mechanisms by which indirect genomic
signaling of ER is mediated (Siersbaek et al., 2018).
Gene expression regulated by ER is important for mammary gland development
and maintenance of bone mass. However, dysregulated ER signaling drives breast
cancer progression. Many E2-responsive genes have been implicated in tumorigenic
properties of breast cancer cells. c-MYC (MYC), cyclin D1 (CCND1), GREB1 are E2-
responsive genes that affect cell proliferation in breast cancer (Manavathi et al., 2013).
Scaffolding protein LLGL2 is overexpressed in ER+ breast cancer. LLGL2 regulates
leucine transporter SLC7A45 expression on cell surface and promotes leucine uptake to
rescue nutrient stress induced growth suppression (Saito et al., 2019). E3 ubiquitin ligase
TRIM56 is associated with poor endocrine treatment outcome. siRNA mediated
knockdown of TRIM56 decreased ER driven cell proliferation. TRIM56 enhances ER
signaling activity by interacting with and stabilizing ER (Xue et al., 2019). Hippo signaling
pathway is a highly conserved serine/threonine kinase cascade involved in regulation of
cell proliferation and apoptosis. CRISPR-Cas9 mediated double knockout of Hippo-
pathway kinases LATS1 and LATS2 abrogates ER expression and results in growth
suppression of ER+ breast cancer cells (Ma et al., 2021).
7
Abnormal ER signaling drives ER+ breast cancer and targeting of this signaling
mechanism is the principal goal of endocrine therapy. There are three broad categories
of endocrine therapy based on their mechanism of action: selective estrogen receptor
modulators (SERMs), selective estrogen receptor degraders (SERDs), aromatase
inhibitors (AIs).

Figure 1.4 Estrogen and androgen receptor signaling pathways. E2 activated ER
translocates to nucleus and regulates gene transcription. Fulvestrant, tamoxifen and AI
are endocrine therapies targeting ER signaling in ER+ breast cancer. DHT activated AR
translocates to nucleus and regulates gene transcription. Enzalutamide targets AR
signaling and is under clinical investigation for breast cancer. Created with biorender.com.

Choice of endocrine therapy is determined by multiple factors including stage of
tumor and menopausal status. Tamoxifen and raloxifene are two examples of SERMs
used for various indication of ER+ breast cancer. Since its approval in 1998, tamoxifen
8
has been the gold standard for treatment of ER+ breast cancer (Sayed et al., 2019).
SERMs are antiestrogens designed to distinctly modulate estrogen receptor signaling by
altering cofactors with which it associates. SERMs compete with estrogen and engender
varying effects depending on the tissue. Tamoxifen causes opposing effects: antagonistic
effect in breast tissue; agonistic effect in uterus and bone (Fuentes et al., 2019). Estrogen
binding-induced conformational change causes helix 12 (H12) of ER to secure the ligand
into the ligand-binding pocket. Consequentially, activation domain AF-2 cleft is unmasked
to allow for binding of coregulator through binding motifs. In contrast, tamoxifen binding
causes H12 to extend into AF-2 cleft and prevents coactivators from binding and
activating ER (Patel et al., 2018).
SERMs have a context-dependent effect on ER signaling activity. In contrast,
SERDs are designed to be antiestrogens with global ER antagonist activity. Fulvestrant
is the only FDA approved SERD. Originally approved for postmenopausal women with
hormone receptor positive (HR+) breast cancer progressing on tamoxifen or AI treatment,
it has since been approved for first line endocrine therapy for HR+ metastatic breast
cancer (Robertson et al., 2016). Binding of fulvestrant to ER destabilizes H12, rendering
AF1 and AF2 inactive. Fulvestrant binding also reduces receptor dimerization and
accelerates degradation of ER through ubiquitin-proteasome pathway (Patel et al., 2018).
In contrast to SERMs and SERDs, AIs attenuate ER signaling by inhibiting
aromatase activity, which results in lower plasma estrogen levels. Additionally, unlike
tamoxifen, AIs have no agonist activity on ER. In postmenopausal women with HR+
metastatic or locally advanced breast cancer, AI letrozole is approved as first-line therapy
(Mouridsen et al., 2003). Letrozole is also approved for treating advanced breast cancer
9
in postmenopausal women who have progressed on tamoxifen treatment (Mauri et al.,
2006). In premenopausal women, compensatory estrogen production in ovaries renders
AIs ineffective. A meta-analysis of data from four randomized trials in premenopausal
women with ER+ early-stage breast cancer receiving ovarian suppression/ablation, AI
treatment led to reduction in recurrence rates over tamoxifen (Early Breast Cancer
Trialists’ Collaborative. Group., 2022).

1.4 Androgen receptor in ER+ breast cancer
More than 70% of all breast cancers express AR. Approximately 90% of ER+
breast cancer is also AR+ (Collins et al., 2011; Niemeier et al., 2010). Located on
chromosome X, AR contains 8 exons. Full-length AR contains 919 amino acids with a
molecular weight of 110-kDa. Like ER, AR is a member of nuclear receptor family. The
full-length isoform of AR consists of three functional domains: N-terminal domain (NTD);
DNA binding region (DBD); ligand binding domain (LBD). Additionally, there are two
activation domains: AF1 within NTD; AF2 within LBD. A flexible hinge region connects
LBD and DBD (Figure 1.5) (Tan et al., 2015). Intracellularly, 5a-reductase metabolizes
testosterone to a potent 5a-dihydrotestosterone (DHT). Ligand binding to AR induces
conformational change that dissociates the receptor from accessory proteins. Activated
receptor translocate to nucleus, dimerize, and bind to androgen response elements
(AREs) in the genome (Figure 1.4). Activation domain AF-1 functions in a ligand-
independent manner and is essential for maximal activity of AR. In contrast, AF2 is
ligand-dependent and is essential for binding of coregulators.
10
AR is expressed in normal breast epithelial cells and is vital for development of
mammary glands. Female AR null (AR
-/-
) mice exhibit defects in mammary glands during
various stages of development (Yeh at al., 2003).

Figure 1.5 Structure of androgen receptor gene and full-length protein. Androgen
receptor gene contains 8 exons. Full-length AR protein contain several functional
domains: N-terminal domain (NTD); DNA binding domain (DBD); ligand binding domain
(LBD). Activation domain AF-1 is in NTD, and AF-2 in LBD. Hinge region connects DBD
and LBD. Created with biorender.com.

Role of AR in ER+ breast cancer is controversial. Many studies provide evidence
for both tumor promoting and suppressive activities of AR. High AR:ER ratio (>=2) in a
cohort of ER+ breast cancers indicated a four times increased risk for failure on tamoxifen
(Cochrane 2014). In another cohort of breast tumor samples, high AR-D-value (indicative
of higher nuclear AR levels) was a poor prognostic factor in nearly all subtypes (Feng et
al., 2017). In contrast, an analysis of ER+ cases from Nottingham Tenovus Primary Breast
Cancer series and METABRIC, AR was a favorable prognostic factor for breast cancer
survival (Hickey et al., 2021).
In vitro, enzalutamide synergizes with tamoxifen and fulvestrant to inhibit growth
of ER+ cell lines. (D’Amato et al., 2016). AR null (AR
-/-
) MCF7 cells show impaired
proliferation in cell culture, and abrogated growth in soft agar colony formation assay (Yeh
at al., 2003). MCF7s exhibit decreased proliferation and fewer soft agar colonies in
11
response to enzalutamide, an AR antagonist. The phenotype is recapitulated in AR
knockdown and with a different AR antagonist MJC13 (D’Amato et al., 2016). DHT
treatment of MCF7s exhibit a mild inhibitory effect on growth. However, cells display
profound increase in migratory and invasive abilities in response to DHT. These
phenotypes were accompanied by downregulation of epithelial markers and upregulation
of mesenchymal markers, indicating epithelial to mesenchymal transition (EMT).
Additionally, DHT treatment also increased CD24
low
/CD44
high
population and the ability to
form mammospheres (Feng et al., 2017). DHT-induced EMT program was abrogated by
AR knockdown and potentiated by AR overexpression, indicating AR is indispensable for
the phenotype. Lysine-specific demethylase 1A (LSD1), together with AR, epigenetically
regulates genes to induce EMT program. AR and LSD1 are essential for DHT-induced
lung metastasis in vivo (Feng et al., 2017). Ligand-activated AR binds to E-cadherin
promoter and represses expression to induce EMT-like phenotype. AR represses E-
cadherin expression by cooperating with HDAC1. Activated AR promotes migratory
characteristics of breast cancer cells in vitro and in vivo (Liu et al., 2008).
In a study utilizing patient-derived ER+ xenograft models and cell lines
supplemented with E2, DHT activation of AR suppresses tumor growth in vivo. In these
models, treatment with enzalutamide had no effect, indicating that in presence of E2,
together with AR, agonist leads to suppression of tumor growth (Hickey et al., 2021).
However, in another study, Enzalutamide inhibited DHT-induced growth of MCF7 tumors
in an orthotopic model (Cochrane 2014). Growth of primary tumors (driven by E2 alone
or DHT alone) is diminished by Enzalutamide. Additionally, Enzalutamide also weakened
metastatic burden of estrogen sensitive ER+ breast cancer cells (D’Amato et al., 2016).
12
Wide-ranging mechanistic insights into tumor promoting or inhibitory activities of
AR have been reported. Chromatin immunoprecipitation (ChIP) assay of MCF7 cells
treated with E2 and Enzalutamide or MJC13 indicate that AR antagonists decrease ER
binding but doesn’t affect its distribution across the genome. Enzalutamide decreases ER
binding to traditional ER targets. In ER+ breast cancer cells, E2 treatment induces nuclear
localization of AR, and improves its colocalization with ER in the nuclei. In response to
E2 there is a significant overlap between AR and ER genomic binding sites. Collectively
these data suggest AR is important for maximum genomic binding of ER (D’Amato et al.,
2016). In another study, ChIP of AR and ER indicates a genomic redistribution of ER
binding by activated AR. Moreover, ChIP for p300 and SRC-3 suggests that AR
sequesters these proteins to antagonize ER signaling (Hickey et al., 2021).
Some of the inconsistencies observed in data from different in vitro and in vivo
studies can be attributed to presence or absence of hormone treatments (DHT, E2),
differences in genetic or pharmacological interventions (AR
knockdowns/overexpression/antagonists), cell line models, and variations in mouse
models.

1.5 Resistance to endocrine therapy and the need for new therapeutic approaches
Endocrine therapy (SERMs, SERDs and AIs) has greatly benefited ER+ breast
cancer cases. In patients with early-stage HR+ breast cancer, adjuvant treatment of
tamoxifen for 5 years reduces risk of disease associated mortality at 15 years by about
a third (Early Breast Cancer Trialists’ Collaborative Group, 2011). In postmenopausal
women with ER+ early breast cancer, AIs were more effective than tamoxifen with a
13
proportional reduction in recurrence rates by 30% (Early Breast Cancer Trialists’
Collaborative Group, 2015). However, recurrence in ER+ disease is not uncommon. In a
meta-analysis of patients who underwent 5 years of adjuvant endocrine therapy, the risk
of distant recurrence ranged from 10 to 41% in a follow up period from 5 to 20 years (Pan
et al., 2017). While recurrent or metastatic disease may initially respond to endocrine
therapy, in most cases progression is inevitable. Resistance to endocrine therapy can be
intrinsic (de novo) – tumors that never exhibit sufficient response to therapy; or acquired
– tumors that initially respond but eventually develop resistance to therapy. Intrinsic
resistance is observed in 20 to 30% of ER+ breast cancers (Haricharan et al., 2017;
Robinson et al., 2013). Different mechanisms of acquired resistance to endocrine therapy
have been documented. Gain-of-function mutations in ER, ESR1 amplifications and
fusions, altered PI3K pathway, crosstalk between ER and growth factor pathways, and
alterations in transcriptional regulators are some of the mechanisms of resistance to
endocrine therapy (reviewed in Hanker et al., 2020)
Gain-of-function mutations in ER gene ESR1 have been identified by multiple
studies in patients with metastatic ER+ breast cancer, particularly after undergoing long-
term therapy with AIs. Frequency of ESR1 mutations detected in patient samples ranged
from 10.5% to 39% depending on the type of sample, method of detection, stage of
disease, and treatment regimen (Toy et al., 2013; Jeselsohn et al., 2014; Fribbens et al.,
2016; Schiavon et al., 2015). Clinically important point mutations in ESR1 are commonly
found in a hotspot region of the LBD (Figure 1.6). The gain-of-function mutations render
ER constitutively active in the absence of the ligand and confer on tumor cells an acquired
resistance to endocrine therapy. Y537S and D538G are among the most prominent
14
hotspot mutations in the ligand binding domain (LBD) of ER. In Y537S and D538G
mutations, helix H12 is stabilized in an active conformation that enables binding of
coactivators in a ligand-independent fashion (Jesselsohn et al., 2018). Characterization
of cell line models harboring these mutations have identified pathways and phenotypes
important for endocrine resistance and metastatic breast cancer (Bahreini et al., 2017; Yu
et al., 2019; Li et al., 2022; Williams et al., 2021).

Figure 1.6 ESR1 mutations in ligand binding domain. Mutations in LBD of ESR1
contribute to acquired resistance of endocrine therapy. Y537S and D538G are among the
most prominent mutations.

Recent studies have attempted to define the role of AR in endocrine resistant
models of ER+ breast cancer. Tamoxifen-resistant (TamR) MCF7s and long-term
estrogen deprived (LTED) cell lines have commonly been used in this context. AR was
upregulated in TamR and LTED MCF7 models. In these models, AR knockdown
diminished proliferation and restored sensitivity to tamoxifen treatment (Chia et al., 2019).
AR knockdown suppressed soft agar colonies, and the response improved further on
tamoxifen treatment. Interestingly, lower pAkt-S473 and increased FOXO3a levels
observed in TamR-AR knockdown cells suggest AR may be mediating effects through
PI3K/Akt pathway. However, treatment with AR antagonist Enzalutamide did not
15
phenocopy the effect of AR knockdown. Since Enzalutamide mainly targets canonical
activities of AR, it indicates endocrine resistance in TamR and LTEDs may be mediated
through non-canonical AR activities (Chia et al., 2019). In AR deficient MCF7 cell and
mouse models, effect of AR loss on growth was mediated through Ras/Raf/MAPK
cascade and ER signaling. Intriguingly, restoration of MAPK activity by AR NTD/DBD
domains, but not LBD, suggests non-canonical signaling of AR may be responsible for
the observed growth defect phenotype (Yeh at al., 2003). However, other studies have
demonstrated efficacy of enzalutamide in abrogating growth of naïve and endocrine
resistant models in vitro and in vivo (Cochrane et al., 2014; D’Amato et al., 2016; Creevey
et al., 2019).
Recent generation of genome-edited breast cancer cell models with ESR1
mutations provide new avenues to investigate mechanism of endocrine resistance.
Currently, literature is limited on the role of AR in this context. In T47D and MCF7 cells
expressing WT or mutant ER cells, AR was upregulated in full serum media and in LTED
cells. AR was upregulated, and increased nuclear AR was observed when cells were
grown in suspension. Treatment with enzalutamide or CYP17a1 lyase inhibitor
sevitreonel inhibited soft agar colonies of WT and mutant ER cells. In ovariectomized
NSG mice, D538G mutant T47D cells generated higher whole mouse disease burden
and lung disease burden compared to WT (Williams et al., 2021)
Thesis work described herein will focus on delineating the tumor growth promoting
and suppressive properties of AR in ESR1 mutant models.

16
CHAPTER 2

Context-dependent role of androgen receptor (AR) in estrogen
receptor-positive (ER+) breast cancer

2.1 Introduction
More than two-thirds of breast cancers express estrogen receptor-a (ER). While
70-90% of all ER+ tumors also express androgen receptor (AR), AR’s role in
tumorigenesis of ER-driven breast malignancies remains controversial. Endocrine
therapy targets ER signaling and is the mainstay of treatment for ER+ breast tumors.
Selective estrogen receptor modulators (SERMs), selective estrogen receptor
downregulators (SERDs), and aromatase inhibitors (AIs) are different classes of
endocrine therapy that have long been used as the standard of care for patients with ER+
breast cancer. AIs block aromatase enzyme to prevent estrogen synthesis from
androgens and are used in first-line treatment of post-menopausal women with ER+
breast cancer. Resistance to endocrine therapy is a well-documented phenomenon.
Intrinsic or de novo resistance to endocrine therapy at the time of diagnosis occurs in up
to 20% of ER+ HER2- breast cancers (Haricharan et al., 2017). Moreover, acquired
resistance to endocrine treatment occurs in up to 39% of patients over the duration of
long-term treatment in ER+ advanced breast cancer (Fribbens et al., 2016; Turner et al.,
2016; Clatot et al., 2016). Gain-of-function mutations in ER gene ESR1 had been
identified by multiple studies in patients with metastatic ER+ breast cancer, particularly
17
after undergoing long-term therapy with AIs (Toy et al., 2013; Jeselsohn et al., 2014;
Fribbens et al., 2016; Schiavon et al., 2015). These mutations render ER constitutively
active in the absence of the ligand and confer on tumor cells an acquired resistance to
endocrine therapy. Y537S and D538G are among the most prominent hotspot mutations
in the ligand binding domain (LBD) of ER. Characterization of cell line models harboring
these mutations has identified pathways and phenotype important for endocrine
resistance and metastatic breast cancer (Bahreini et al., 2017; Li et al., 2022; Yu et al.,
2019). AR is a nuclear hormone receptor activated by androgens and plays a crucial role
in the progression of prostate cancer. In contrast, role of AR in breast cancer is unclear.
Evidence supporting tumor promoting and antitumor activities of AR have been reported
(Hickey et al., 2021; Williams et al., 2021; Feng et al., 2017; Chia et al., 2019). Role of
AR in the context of postmenopausal breast cancer harboring ESR1 mutations is largely
unexplored. In postmenopausal patients treated with AI, estrogen levels are largely
depleted, but androgen levels remain elevated (Creevey et al., 2019). This change in
steroid hormonal milieu suggest AR may become more important for ER+ tumor cells
harboring ESR1 mutations in estrogen deprived environment. In our studies, we
examined the role of AR in ER+ cell lines harboring Y537S and D538G mutations in short-
term and long-term estrogen deprived conditions. We generated AR knockout models in
ER mutant cells and assessed the role of AR in adherent and anchorage-independent
conditions. Moreover, we evaluated role of AR in an in vivo experimental metastasis
model for postmenopausal breast cancer. We further provide evidence indicating a
potential role for AR in supporting survival of anchorage-independent cells in a sodium
pyruvate-free environment.
18
2.2 Results

2.2.1 AR is upregulated in breast cancer cell lines harboring ESR1 Y537S mutation

Androgen receptor has multiple transcript variants resulting from alternative
splicing. As many as 22 variants have been reported in literature for prostate cancer (Lu
et al., 2020). AR mRNA level in WT, Y537S and D538G mutant breast cancer cells was
measured by qPCR using primers targeting full-length variant (AR-FL). In MCF7s cultured
in full-serum media, AR was upregulated in Y537S, but not in D538G, compared to WT
(Figure 2.1.A). In T47Ds cultured in full-serum media, AR was upregulated in Y537S
clone#2. While a trend towards upregulation was observed in Y537S clone#1, AR
expression was lower in D538G (Figure 2.1.B). To assess the status of AR at protein level
in MCF7s and T47Ds cultured in media containing full-serum, immunoblot assay was
performed using antibody targeting amino terminal region of human AR protein. In
MCF7s, AR was upregulated in Y537S compared to WT, but not in D538G cells (Figure
2.2.A and 2.2.B). ER was downregulated in mutant cells compared to WT. No changes
were observed in NCOA3 - a common coactivator of both AR and ER. In T47Ds, AR
protein levels were higher in Y537S#1 and Y537S#2 compared to WT, but not in D538G
cells (Figure 2.2.C and 2.2.D). ER expression was lower in Y537S#1 and Y537S#2
compared to WT. ER levels in D538G didn’t change in comparison to WT. There were
no changes in NCOA3 expression in mutant cells compared to WT.
To assess AR status in conditions mimicking breast tumor cells in post-
menopausal patients, MCF7s and T47Ds were cultured in phenol-red free culture media
19
supplemented with charcoal-dextran stripped serum (CSS media) for 3 days. In MCF7s,
AR full-length mRNA was upregulated in Y537S, but not in D538G, compared to WT
(Figure 2.3.A). In T47Ds, there was no change in AR full-length mRNA levels in Y537S#1
or Y537S#2, however it was lower in D538G compared to WT cells (Figure 2.3.B). Next,
AR, ER and NCOA3 protein levels were assessed in MCF7s and T47Ds cultured in CSS
media for 3 days and treated with vehicle or 10nM E2 for 24 hours. In MCF7s, AR was
upregulated in Y537S-vehicle compared to WT-vehicle (Figure 2.4.A and 2.4.B).
Increased AR levels in Y537S were maintained on addition of E2, compared to WT-
vehicle. No change was observed in D538G-vehicle AR levels compared to WT-vehicle,
but AR levels increased on addition of E2 (Figure 2.4.A and 2.4.B). ER levels were lower
in all conditions when compared to WT-vehicle. On addition of E2, NCOA3 levels were
lower in WT compared to vehicle, but there were no changes in mutant cells. In T47Ds,
Y537S#1 and Y537S#2 mutant cells had higher baseline levels of AR compared to WT-
vehicle (Figure 2.4.C and 2.4.D). On addition of E2, higher AR protein levels were still
maintained in both Y537S#1 and Y537S#2, compared to WT-vehicle (Figure 2.4.C and
2.4.D).
AR is a transcription factor, and its genomic activity is dependent on nuclear
localization. Nuclear AR levels were assessed by its co-expression with DAPI in
immunofluorescence staining assay. In T47Ds cultured in standard media, higher nuclear
AR levels were observed in Y537S#1 compared to WT cells (Figure 2.5.A). Y537S#2
exhibits a trend towards increased nuclear AR levels compared to WT. Nuclear AR levels
were lower in D538G compared to WT cells. Next, nuclear AR levels were assessed in
T47D WT and Y537S#1 cultured in CSS media. Baseline levels of nuclear AR were higher
20
in Y537S#1 compared to WT. E2 treatment increased nuclear AR levels in both WT and
Y537S#1 mutant cells (Figure 2.5.B).

Figure 2.1 Elevated AR mRNA expression in mutant ER cell lines cultured in full-
serum media. (A) qPCR of cDNA from MCF7 ER WT, Y537S and D538G cells cultured
in full-serum media showed an increased expression of full-length AR in Y537S compared
to WT cells. No change observed in D538G compared to WT cells. Data from four
experiments is displayed as box and whiskers plot. Data were analyzed by one-way
ANOVA followed by Tukey’s multiple comparisons test. P values are indicated by
asterisks: ** P£ 0.01. (B) qPCR of cDNA from T47D ER WT, Y537S#1, Y537S#2, and
D538G cells cultured in full-serum media showed an increased expression of full-length
AR in Y537S#2 compared to WT. No change observed in Y537S#1 and D538G compared
to WT cells. AR expression in D538G was lower compared to Y537S#1 and Y537S#2.
Data from four experiments is represented as box and whiskers plot. Data were analyzed
by one-way ANOVA followed by Tukey’s multiple comparisons test. P values are
indicated by asterisks: * P£ 0.05; ** P£ 0.01.

A B
MCF7 T47D
21

Figure 2.2 AR is upregulated in mutant ER cell lines cultured in full-serum media.
(A) and (B) Lysates from MCF7 ER WT, Y537S and D538G were immunoblotted for AR,
ER, NCOA3 and loading control GAPDH. Representative images from western blot are
shown in (A) and quantification shown in (B). AR was upregulated in Y537S compared to
WT. No change observed in D538G AR expression compared to WT. ER expression was
lower in both Y537S and D538G. No change observed in NCOA3 expression in mutant
cells compared to WT. Data from seven experiments is displayed as box and whiskers
plot. Data were analyzed by one-way ANOVA followed by Dunnett’s multiple comparisons
test. P values are indicated by asterisks: * P£ 0.05; **** P£ 0.0001. (C) and (D) Lysates
from T47D ER WT, Y537S#1, Y537S#2 and D538G were immunoblotted for AR, ER and
NCOA3 and loading control GAPDH. Representative images from western blot is shown
in (C) and quantification in (D). AR was upregulated in Y537S#1 and Y537S#2 compared
to WT. No change observed in D538G AR levels compared to WT. ER expression was
lower in both Y537S#1 and Y537S#2, but not in D538G. No change observed in NCOA3
expression in mutant cells compared to WT. Data from four experiments is displayed as
box and whiskers plot. Data were analyzed by one-way ANOVA followed by Dunnett’s
multiple comparisons test. P values are indicated by asterisks: ** P£ 0.01; **** P£ 0.0001.
A B
AR
GAPDH
ER
NCOA3
WT Y537S D538G
C D
0
200
400
600
MCF7 - STANDARD MEDIA - AR
AR
Normalized Western Signal
WT
Y537S
D538G
✱
0
125
250
375
500
ER
Normalized Western Signal
MCF7 - STANDARD MEDIA - ER
WT
Y537S
D538G
✱✱✱✱
✱✱✱✱
0
400
800
1200
NCOA3
Normalized Western Signal
MCF7 - STANDARD MEDIA - NCOA3
WT
Y537S
D538G
AR
GAPDH
ER
NCOA3
WT Y537S#1 Y537S#2 D538G
0
250
500
750
1000
AR
Normalized Western Signal
T47D - STANDARD MEDIA - AR
✱✱
✱✱
WT
Y537S#1
Y537S#2
D538G
0
50
100
150
200
250
ER
Normalized Western Signal
T47D - STANDARD MEDIA - ER
✱✱
✱✱✱✱
WT
Y537S#1
Y537S#2
D538G
80
100
120
140
160
180
NCOA3
Normalized Western Signal
T47D - STANDARD MEDIA - NCOA3
WT
Y537S#1
Y537S#2
D538G
0
250
500
750
1000
AR
Normalized Western Signal
T47D - STANDARD MEDIA - AR
✱✱
✱✱
WT
Y537S#1
Y537S#2
D538G
0
50
100
150
200
250
ER
Normalized Western Signal
T47D - STANDARD MEDIA - ER
✱✱
✱✱✱✱
WT
Y537S#1
Y537S#2
D538G
80
100
120
140
160
180
NCOA3
Normalized Western Signal
T47D - STANDARD MEDIA - NCOA3
WT
Y537S#1
Y537S#2
D538G
0
250
500
750
1000
AR
Normalized Western Signal
T47D - STANDARD MEDIA - AR
✱✱
✱✱
WT
Y537S#1
Y537S#2
D538G
0
50
100
150
200
250
ER
Normalized Western Signal
T47D - STANDARD MEDIA - ER
✱✱
✱✱✱✱
WT
Y537S#1
Y537S#2
D538G
80
100
120
140
160
180
NCOA3
Normalized Western Signal
T47D - STANDARD MEDIA - NCOA3
WT
Y537S#1
Y537S#2
D538G
0
50
100
150
200
NCOA3
Normalized Western Signal
T47D - STANDARD MEDIA - NCOA3
WT
Y537S#1
Y537S#2
D538G
MCF7 MCF7
T47D T47D
22

Figure 2.3 AR mRNA expression in mutant ER cell lines cultured in charcoal-
stripped serum containing media. (CSS media). (A) qPCR of cDNA from MCF7 ER
WT, Y537S and D538G cells cultured in CSS media showed an increased expression of
full-length AR in Y537S compared to WT cells. No change observed in D538G compared
to WT cells. Data from three experiments are displayed as box and whiskers plot. Data
were analyzed by one-way ANOVA followed by Dunnett’s multiple comparisons test. P
values are indicated by asterisks: * P£ 0.05 (B) qPCR of cDNA from T47D ER WT,
Y537S#1, Y537S#2, and D538G cells cultured in CSS media showed no change in full-
length AR in mutant cells compared to WT. AR expression was lower in D538G compared
to WT. Data from four experiments is displayed as box and whiskers plot. Data were
analyzed by one-way ANOVA followed by Dunnett’s multiple comparisons test. P values
are indicated by asterisks: * P£ 0.05.

AR
0.8
1.0
1.2
1.4
1.6
Normalized AR-FL mRNA level relative to WT
WT
Y537S
D538G
✱
ns
A
AR
0.0
0.5
1.0
1.5
Normalized AR-FL mRNA level relative to WT
WT
Y537S#1
Y537S#2
D538G
ns
ns
✱
B
MCF7 T47D
23

Figure 2.4. AR is upregulated in mutant ER cell lines cultured in charcoal-stripped
serum containing media. (CSS media). (A) and (B) MCF7 ER WT, Y537S and D538G
cells cultured in CSS media for 3 days and treated with vehicle or 10nM E2 for 24 hours,
and lysates were immunoblotted for AR, ER, NCOA3 and loading control GAPDH.
Representative images from western blot are shown in (A) and quantification is shown in
(B). AR was upregulated in vehicle treated Y537S compared to vehicle treated WT. No
change observed in AR protein expression in D538G-vehicle compared to WT-vehicle.
ER expression was lower in mutant cells compared to WT in both vehicle and E2
treatments. NCOA3 expression was lower in E2 treated WT compared to WT-vehicle.
Data from five experiments is shown as box and whiskers plot. Data were analyzed by
one-way ANOVA followed by Dunnett’s multiple comparisons test. P values are indicated
by asterisks: * P£ 0.05; ** P£ 0.01; *** P£ 0.001. (C) and (D) T47D ER WT, Y537S#1,
Y537S#2 and D538G cells cultured in CSS media for 3 days and treated with vehicle or
10nM E2 for 24 hours, and lysates were immunoblotted for AR, ER, NCOA3 and loading
control GAPDH. Representative images from western blot are shown in (C) and
quantification is shown in (D). AR was upregulated in vehicle treated Y537S#1 compared
to vehicle treated WT. AR was also upregulated in Y537S#2 compared to WT-vehicle. No
change observed in AR protein expression in D538G-vehicle compared to WT-vehicle.
No changes observed in ER and NCOA3 expression in mutant cells compared to WT in
both vehicle and E2 treatments. Data from three experiments is shown as box and
whiskers plot. Data were analyzed by one-way ANOVA followed by Dunnett’s multiple
comparisons test. P values are indicated by asterisks: * P£ 0.05; *** P£ 0.001.
A B
AR
GAPDH
ER
NCOA3
WT WT+E2 Y537S Y537S+E2 D538G D538G+E2
C D
AR
GAPDH
ER
NCOA3
WT
WT E2
Y537S#1
Y537S#1 E2
Y537S#2
Y537S#2 E2
D538G
D538G E2
0
200
400
600
AR
Normalized Western Signal
MCF7 CSS MEDIA - AR
✱✱✱
✱✱✱
✱
WT + Vehicle
WT + E2
Y537S + Vehicle
Y537S + E2
D538G + Vehicle
D538G + E2
0
200
400
600
800
MCF7 CSS MEDIA - ER
ER
Normalized Western Signal
✱✱
✱✱
✱✱✱
✱
✱✱✱
WT + Vehicle
WT + E2
Y537S + Vehicle
Y537S + E2
D538G + Vehicle
D538G + E2
0
200
400
600
800
NCOA3
Normalized Western Signal
MCF7 CSS MEDIA - NCOA3
WT + Vehicle
WT + E2
Y537S + Vehicle
Y537S + E2
D538G + Vehicle
D538G + E2
✱
0
200
400
600
T47D - CSS MEDIA - AR
AR
Normalized Western Signal
✱
✱✱✱
✱
✱
WT + Vehicle
WT + E2
Y537S#1 + Vehicle
Y537S#1 + E2
Y537S#2 + Vehicle
Y537S#2 + E2
D538G + Vehicle
D538G + E2
0
50
100
150
T47D - CSS MEDIA - ER
ER
Normalized Western Signal
WT + Vehicle
WT + E2
Y537S#1 + Vehicle
Y537S#1 + E2
Y537S#2 + Vehicle
Y537S#2 + E2
D538G + Vehicle
D538G + E2
0
50
100
150
T47D- CSS MEDIA - NCOA3
NCOA3
Normalized Western Signal
WT + Vehicle
WT + E2
Y537S#1 + Vehicle
Y537S#1 + E2
Y537S#2 + Vehicle
Y537S#2 + E2
D538G + Vehicle
D538G + E2
0
200
400
600
T47D - CSS MEDIA - AR
AR
Normalized Western Signal
✱
✱✱✱
✱
✱
WT + Vehicle
WT + E2
Y537S#1 + Vehicle
Y537S#1 + E2
Y537S#2 + Vehicle
Y537S#2 + E2
D538G + Vehicle
D538G + E2
0
50
100
150
T47D - CSS MEDIA - ER
ER
Normalized Western Signal
WT + Vehicle
WT + E2
Y537S#1 + Vehicle
Y537S#1 + E2
Y537S#2 + Vehicle
Y537S#2 + E2
D538G + Vehicle
D538G + E2
0
50
100
150
T47D- CSS MEDIA - NCOA3
NCOA3
Normalized Western Signal
WT + Vehicle
WT + E2
Y537S#1 + Vehicle
Y537S#1 + E2
Y537S#2 + Vehicle
Y537S#2 + E2
D538G + Vehicle
D538G + E2
0
200
400
600
T47D - CSS MEDIA - AR
AR
Normalized Western Signal
✱
✱✱✱
✱
✱
WT + Vehicle
WT + E2
Y537S#1 + Vehicle
Y537S#1 + E2
Y537S#2 + Vehicle
Y537S#2 + E2
D538G + Vehicle
D538G + E2
0
50
100
150
T47D - CSS MEDIA - ER
ER
Normalized Western Signal
WT + Vehicle
WT + E2
Y537S#1 + Vehicle
Y537S#1 + E2
Y537S#2 + Vehicle
Y537S#2 + E2
D538G + Vehicle
D538G + E2
0
50
100
150
T47D- CSS MEDIA - NCOA3
NCOA3
Normalized Western Signal
WT + Vehicle
WT + E2
Y537S#1 + Vehicle
Y537S#1 + E2
Y537S#2 + Vehicle
Y537S#2 + E2
D538G + Vehicle
D538G + E2
0
200
400
600
T47D - CSS MEDIA - AR
AR
Normalized Western Signal
✱
✱✱✱
✱
✱
WT + Vehicle
WT + E2
Y537S#1 + Vehicle
Y537S#1 + E2
Y537S#2 + Vehicle
Y537S#2 + E2
D538G + Vehicle
D538G + E2
0
50
100
150
T47D - CSS MEDIA - ER
ER
Normalized Western Signal
WT + Vehicle
WT + E2
Y537S#1 + Vehicle
Y537S#1 + E2
Y537S#2 + Vehicle
Y537S#2 + E2
D538G + Vehicle
D538G + E2
0
50
100
150
T47D- CSS MEDIA - NCOA3
NCOA3
Normalized Western Signal
WT + Vehicle
WT + E2
Y537S#1 + Vehicle
Y537S#1 + E2
Y537S#2 + Vehicle
Y537S#2 + E2
D538G + Vehicle
D538G + E2
MCF7
MCF7
T47D
T47D
24

Figure 2.5. T47D ER Y537S mutant cells show elevated nuclear expression of AR.
(A) and (B) T47D ER WT, Y537S#1, Y537S#2 and D538G cells cultured in full-serum
media (A) or CSS media (B) and were probed for nuclear AR by co-expression with DAPI
in immunofluorescence assay. (A) In full-serum media, Y537S#1 cells showed increased
nuclear AR levels compared to WT. Y537S#2 cells showed a trend towards increased
nuclear AR levels compared to WT. D538G cells showed lower levels of nuclear AR
compared to WT. Representative images are shown on left and quantification is shown
on right. Data from three experiments are shown as scatter dot plot. Each point represents
fluorescence intensity from a cell. Data were analyzed by one-way ANOVA followed by
Dunnett’s multiple comparisons test. P values are indicated by asterisks: **** P£ 0.0001.
(B) In CSS media, Y537S#1 cells treated with vehicle showed higher baseline levels of
nuclear AR compared to WT cells treated with vehicle. On addition of E2, there was an
increase in nuclear AR levels in both WT and Y537S#1 compared to cells treated with
vehicle. Representative images are shown on left and quantification is shown on right.
Data from two experiments are shown as scatter dot plot. Each point represents
fluorescence intensity from a cell. Data were analyzed by one-way ANOVA followed by
Dunnett’s multiple comparisons test. P values are indicated by asterisks: ** P£ 0.01; ****
P£ 0.0001.
AR DAPI MERGE
WT
Y537S #1
Y537S #2
D538G
AR DAPI MERGE
WT
VEHICLE
Y537S #1
VEHICLE
WT
E2
Y537S #1
E2
WT_Vehicle
WT_E2
Y537S#1_Vehicle
Y537S#1_E2
0
2×10
6
4×10
6
6×10
6
8×10
6
1×10
7
T47D CSS Media AR 40X R Integration 02.06.21 02.12.21
Nuclear AR fluorescence Intensity
✱✱
✱✱
✱✱✱✱
✱✱✱✱
✱✱✱✱
A
B
WT
Y537S#1
Y537S#2
D538G
0.0
5.0×10
5
1.0×10
6
1.5×10
6
2.0×10
6
Nuclear AR fluorescence Intensity
T47D Std Media AR 40X R Integration - 01.06.21 01.10.21 01.18.21
✱✱✱✱
✱✱✱✱
25
2.2.2 Androgen receptor plays a context-dependent role in tumorigenesis of ER+
breast cancer cells

The association between ESR1 mutations and elevated AR protein level possibly
suggests a reliance of these cells on AR. To investigate contribution of AR to malignant
properties of cells harboring ESR1 mutations, MCF7s with AR gene knockout was
generated. As described in the methods section, CRISPR/Cas9 was utilized to achieve
knockout of AR in WT or mutant ER MCF7 cells. Two guide RNAs (sg1, sg2) targeting
exon 1 of AR were designed using the Broad Institute GPP Web Portal (Figure 2.6.A).
Depletion of AR resulting from successful gene knockout was confirmed in individual
clones by immunoblot assay with AR specific antibody (Figure 2.6.B). For each guide
RNA, three clones were pooled and used in subsequent experiments. Similarly, for control
cells (EV), two clones were pooled and used in subsequent experiments. AR knockout
cell lines will hereafter be referred as AR KO.
A recent study provided evidence for the role of AR as a tumor suppressor in ER+
breast cancer in the presence of E2 and DHT in both in vitro and in vivo experiments.
(Hickey et al., 2021). In contrast, another recent study provides evidence suggesting a
dependence on AR for survival in breast cancer cells harboring ESR1 mutations (Williams
et al., 2021). To delineate these opposing and contradictory roles of AR, CRISPR-edited
cells generated from our study were examined in three assays to test for phenotypes
indicative of malignant cells: ability to form colonies when plated at low density in
adherent condition (colony formation assay); ability to grow in anchorage independent
conditions (soft agar assay); and ability to form metastases in vivo. ESR1 mutations in
26
the CRISPR-edited cells are heterozygous, and to clearly attribute the role of mutations,
cells were cultured in CSS media for three days before utilizing in aforementioned assays.
MCF7 AR KO cell lines were seeded at low density and tested for growth in colony
formation assay (Figure 2.7). Cells were cultured in CSS media for three days and then
plated in 24-well plates. Wells were treated with vehicle, 1nM E2, 1nM DHT or 1nm E2 +
DHT for seven days. WT EV cells show a trend towards growth suppression in presence
of E2 + DHT, when compared to E2 alone (Figure 2.7.A). AR-KO WT-sg1 and WT-sg2
cells didn’t exhibit this growth suppression. In ER mutant cells Y537S-EV and D538G-
EV, suppression of growth was obvious in E2 + DHT group, compared to E2 alone (Figure
2.7.B and 2.7.C.). Moreover, DHT treatment alone induces a trend towards growth
suppression in mutant ER cells when compared to vehicle. In mutant ER cells with AR
knockout, there was no change in E2 + DHT treatment compared to E2 alone (Figure
2.7.B and 2.7.C). These data indicate AR suppresses cell proliferation in adherent cell
culture in presence of E2 and DHT.
Anchorage-independent growth is characteristic of cells that exhibit resistance to
anoikis. Detachment from extracellular matrix and survival in anchorage-independent
conditions are hallmarks of metastatic cells. MCF7 control and AR KO cells were cultured
in CSS media for three days and plated in soft agar assay. After 4 weeks, wells were
analyzed for area covered by colonies (percent colony area) and area covered by
colonies together with density of colonies (percent colony intensity). MCF7 WT and
D538G cells with sg1 show fewer colonies, as measured by percent colony area and
percent colony intensity, compared to EV group (Figure 2.8). Y537S sg1 and sg2 shows
a trend towards lower colony area and colony intensity compared to EV. AR sg2 exhibits
27
higher colony area and colony intensity in D538G. AR sg2 exhibits a trend towards lower
colony area compared to EV in WT cells. Differences between sg1 and sg2 in soft agar
assay suggests there may be some interference from assay-specific off-target effects
from different guides targeting AR.
In vitro assays to test anchorage-independent growth, such as soft agar assays,
may not full capture tumor cell behavior in vivo. To examine the role of AR in the ability
of mutant ER cells to survive in anchorage-independent conditions and to establish
metastasis in vivo, an experimental metastasis assay was performed. Furthermore, to
mimic conditions in postmenopausal women where ESR1 mutations arise, GFP-LUC
labelled cells were cultured in CSS media prior to injecting intracardially into
ovariectomized NSG mice, and without E2 supplementation. Mice were imaged weekly
for 4 weeks by IVIS imaging. Images from all mice immediately after injection (Day 0,
Figure 2.9.A) and mice alive at 4 weeks (Figure 2.9.B) are shown. Total IVIS signal at 4
weeks show that tumor burden was higher in Y537S compared to both Y537S AR KO
and WT groups (Figure 2.9.C).
Taken together, these results suggest AR suppresses cell proliferation in adherent
culture in the presence of E2 and DHT. However, under anchorage-independent
conditions, AR supports survival in an estrogen-deprived environment.

28

Figure 2.6 AR knockout in ESR1 mutant MCF7 cell lines. (A)CRISPR-Cas9 was
utilized to generate single cell clones of WT or mutant ER MCF7 cells. Two guide RNAs
(sg1, sg2) targeting exon 1 of AR were designed using the Broad Institute GPP Web
Portal. Image created with Biorender.com (B) Immunoblot to detect AR in lysates from
individual clones with EV, sg1 or sg2 confirmed successful knockout of AR.

AR
GAPDH
EV1
EV2
sg1-2
sg1-3
sg1-4
sg2-2
sg2-10
sg2-11
MCF7 ER WT
EV1
EV2
sg1-3
sg1-5
sg1-6
sg2-1
sg2-3
sg2-6
MCF7 ER Y537S
AR
GAPDH
EV2
EV3
sg1-1
sg1-2
sg1-4
sg2-2
sg2-4
sg2-6
MCF7 ER D538G
AR
GAPDH
A B
29

Figure 2.7 Androgen receptor suppresses growth in presence of E2 and DHT. (A),
(B), (C) MCF7 ER WT, Y537S or D538G cells were cultured in CSS media for 3 days and
plated at low density in colony formation assay. Wells were treated with vehicle, 1nM
DHT, 1nM E2 or 1nM E2 + 1nM DHT for seven days. Representative images of crystal
violet-stained colonies are shown on left, and quantification shown on right. Data
represent the mean ± s.e.m. of three experiments, with 4-6 replicates per condition per
experiment. Data were analyzed using a two-way ANOVA followed by Tukey’s multiple
comparisons test. P values are indicated by asterisks: * P£ 0.05; ** P£ 0.01; *** P£ 0.001;
**** P£ 0.0001. (A) ER WT in presence of E2 and DHT showed a trend towards
suppressed growth, when compared to E2 treatment alone. The growth suppression was
not observed in AR knockout cell lines, when E2+DHT treatment group was compared to
E2 alone group. (B) and (C) In presence of AR, mutant ER containing cells showed a
suppressed growth in presence of E2 and DHT, compared to E2 alone. Moreover,
Vehicle DHT E2 E2 + DHT
WT-EV WT-sg1 WT-sg2
Vehicle DHT
E2 E2 + DHT
Y537S-EV Y537S-sg1 Y537S-sg2
Vehicle DHT
E2 E2 + DHT
D538G-EV D538G-sg1 D538G-sg2
EV sg1 sg2
0.000000
0.000005
0.000010
0.000015
WT
Normalized Absrobance 590nm
Vehicle
DHT
E2
E2+DHT
✱✱✱✱
✱
✱✱✱✱
✱ ✱✱✱✱
✱✱✱✱
✱✱✱
✱✱✱✱
✱✱
✱✱
✱✱
✱✱
EV sg1 sg2
0.000000
0.000005
0.000010
0.000015
0.000020
0.000025
YS
Normalized Absrobance 590nm
Vehicle
DHT
E2
E2+DHT
✱✱✱
✱
EV sg1 sg2
0.000000
0.000005
0.000010
0.000015
0.000020
0.000025
DG
Normalized Absrobance 590nm
Vehicle
DHT
E2
E2+DHT
✱✱✱✱
✱✱✱✱
✱✱
✱ ✱
A
B
C
MCF7 ER WT
MCF7 ER Y537S
MCF7 ER D538G
30
treatment with DHT alone induced a trend towards suppressive growth phenotype,
compared to vehicle.

Figure 2.8 AR-sg1 cells exhibit lower colony area in anchorage-independent
conditions. Representative images of soft agar assay on left, quantification of wells for
colony area and colony intensity on right. Data represent the mean ± s.e.m. of three
experiments, with 2-3 replicates per condition per experiment. Data were analyzed using
one-way ANOVA followed by Dunnett’s multiple comparisons test. P values are indicated
by asterisks: * P£ 0.05; ** P£ 0.01. WT-sg1 showed lower colony area and colony
intensity compared to EV. Y537S-sg1 and Y537S-sg2 showed a trend towards lower
colony area and colony intensity compared to EV. D538G-sg1 exhibited lower colony area
and colony intensity compared to EV. CRISPR guide sg2 exhibited mixed results.
WT-EV
WT-sg1
WT-sg2
0
2
4
6
8
WT 3DAY CSS - % Colony Area
Normalized % Colony Area
✱✱
WT-EV
WT-sg1
WT-sg2
0.0
0.5
1.0
1.5
2.0
2.5
WT 3DAY CSS - % Colony Intensity
Normalized % Colony Intensity
✱✱
YS-EV
YS-sg1
YS-sg2
0
2
4
6
8
YS 3DAY CSS - % Colony Area
Normalized % Colony Area
YS-EV
YS-sg1
YS-sg2
0.0
0.5
1.0
1.5
2.0
2.5
YS 3DAY CSS - % Colony Intensity
Normalized % Colony Intensity
DG-EV
DG-sg1
DG-sg2
0
2
4
6
8
10
DG 3DAY CSS - % Colony Area
Normalized % Colony Area
✱✱
✱✱
DG-EV
DG-sg1
DG-sg2
0
1
2
3
DG 3DAY CSS - % Colony Intensity
Normalized % Colony Intensity
✱
✱✱
EV sg1 sg2
WT
Y537S
D538G
Colony Area Colony Intensity
31

Figure 2.9 AR knockout decreases metastasis of ER Y537S cells. (A), (B), (C)
Ovariectomized NSG mice were injected intracardially with GFP-Luc labelled MCF7 ER
WT, Y537S or Y537S AR KO and monitored for four weeks by IVIS imaging. (A) IVIS
images of mice from day 0 immediately after injection (B) IVIS images of mice at 4 weeks
(C) Total flux data from IVIS image from weeks 1 to 4, normalized to day 0 . Y537S AR
KO and WT groups showed lower total metastatic burden compared to Y537S. Data
represent mean ± s.e.m. Data were analyzed using a one-way ANOVA followed by
Tukey’s multiple comparisons test. P values are indicated by asterisks: * P£ 0.05.
Y537S
AR KO
Y537S WT
Y537S
AR KO
Y537S WT
0 1 2 3 4 5
0
2×10
3
4×10
3
6×10
3
Weeks
Total flux [p/s] (normalized to Day 0)
FRONT ROI 1 TO 4 WEEKS
WT
Y537S
Y537S AR KO
WT
YS
YS - AR KO
0
1×10
4
2×10
4
3×10
4
4×10
4
Total flux [p/s] (normalized to Day 0)
05.10.22 IC 4 weeks
✱ ✱
WT
YS
YS - AR KO
0
1×10
4
2×10
4
3×10
4
4×10
4
Total flux [p/s] (normalized to Day 0)
05.10.22 IC 4 weeks
✱ ✱
DAY 0 4 WEEKS
A B
C
32
2.2.3 Temporal upregulation of AR in long-term estrogen deprived cells

MCF7 is an ER+ breast cancer cell line that is maintained in culture with full-serum
media containing hormones including estrogens and androgens. To model conditions in
postmenopausal women with resistance to endocrine therapy where ESR1 mutations
mainly arise, MCF7s were cultured in CSS media for long-term estrogen depravation
(LTED). Changes in AR protein levels were monitored over time in LTED cells.
Immunoblot of lysates from LTEDs at 3 days, 4 weeks, 12 weeks, and 24 weeks indicate
a temporal upregulation of AR (Figure 2.10). Progressive increase in AR suggests LTED
MCF7s may become increasingly dependent on AR over time. LTED MCF7 AR KOs were
examined for role of AR in cell proliferation and survival in colony formation and soft agar
assays at 4 weeks, 8 weeks, and 12 weeks.
LTED MCF7s at 4-weeks were plated for colony formation assay. WT EV cells
show a growth suppression in presence of E2 + DHT, when compared to E2 alone (Figure
2.11.A). WT-sg1 and WT-sg2 cells didn’t exhibit this suppressive phenotype. In ER
mutant Y537S-EV cells, a growth suppression was obvious in E2 + DHT group compared
to E2 alone (Figure 2.11.B). Moreover, DHT treatment alone suppresses growth in mutant
ER cells when compared to vehicle group (Figure 2.11.B). In D538G-EV cells, a trend
towards growth suppression was observed in E2 + DHT group compared to E2 alone
(Figure 2.11.C). Additionally, DHT treatment alone induces a trend towards growth
suppression D538G-EV compared to vehicle group (Figure 2.11.C). In Y537S and D538G
AR KO cells, no growth suppression was observed in response to E2+DHT treatment
(Figure 2.11.B and 2.11.C).
33
LTED MCF7s at 8-weeks were plated for colony formation assay. WT EV cells
show a trend towards growth suppression in presence of E2 + DHT, when compared to
E2 alone (Figure 2.12.A). WT-sg1 and WT-sg2 cells didn’t exhibit this suppressive
phenotype. In ER mutant Y537S-EV cells, a growth suppression was obvious in E2 +
DHT group compared to E2 alone (Figure 2.12.B). Moreover, DHT treatment alone
suppresses growth in mutant ER cells when compared to vehicle group (Figure 2.12.B).
In D538G-EV cells, a trend towards growth suppression was observed in E2 + DHT group
compared to E2 alone (Figure 2.12.C). Additionally, DHT treatment alone induces a trend
towards growth suppression in D538G-EV compared to vehicle group (Figure 2.12.C). In
Y537S and D538G AR KO cells, no growth suppression was observed in response to
E2+DHT treatment (Figure 2.12.B and 2.12.C).
LTED MCF7s at 12-weeks were plated for colony formation assay. WT EV cells
show a growth suppression in presence of E2 + DHT, when compared to E2 alone (Figure
2.13.A). WT-sg1 and WT-sg2 cells didn’t exhibit this suppressive phenotype. In ER
mutant Y537S-EV cells, a growth suppression was obvious in E2 + DHT group compared
to E2 alone (Figure 2.13.B). Moreover, DHT treatment alone suppresses growth in mutant
ER cells when compared to vehicle group (Figure 2.13.B). In D538G-EV cells, a trend
towards growth suppression was observed in E2 + DHT group compared to E2 alone
(Figure 2.13.C). Additionally, DHT treatment alone induces growth suppression D538G-
EV compared to vehicle group (Figure 2.13.C). In Y537S and D538G AR KO cells, no
growth suppression was observed in response to E2+DHT treatment (Figure 2.13.B and
2.13.C).
34
LTED MCF7s at 4-weeks were plated for soft agar assay to test for role of AR in
anchorage-independent growth of LTED cells. WT-sg1 show lower percent colony area
and percent colony intensity compared to EV group (Figure 2.14). WT-sg2 shows a lower
percent colony intensity and a trend towards lower colony area compared to EV. Y537S-
sg1 and sg2 show a lower percent colony area and percent colony intensity compared to
EV. D538G-sg1 exhibits a trend towards lower percent colony area and percent colony
intensity compared to EV. D538G-sg2 exhibits no changes compared to EV.
Soft agar assay for LTED MCF7s at 8-weeks indicate lower percent colony area
and colony intensity in WT-sg1, Y537S-sg1 and D538G-sg1 compared to EV (Figure
2.15). WT-sg2 and YS-sg2 shows a trend towards lower percent colony area and percent
colony intensity compared to EV. D538G-sg2 exhibits an increase in percent colony area
and percent colony intensity compared to EV.
Soft agar assay for LTED MCF7s at 12-weeks indicate lower percent colony area
and colony intensity in WT-sg1 and D538G-sg1 compared to EV (Figure 2.16). Y537S-
sg1 shows a lower percent colony area and a trend towards lower colony intensity
compared to EV. WT-sg2 and Y537S-sg2 show a trend towards lower percent colony
area and percent colony intensity compared to EV. D538G-sg2 exhibits a trend towards
increase in percent colony area and percent colony intensity compared to EV.
In summary, results from colony formation assay and soft agar assay of LTEDs
suggest cells behave similarly to those from short term estrogen depravation of 3 days.
An upregulation of AR in LTEDs over time doesn’t alter the phenotypes tested here.

35

Figure 2.10 AR is upregulated during long-term estrogen deprivation. MCF7 ER WT,
Y537S or D538G cells were maintained in CSS media for 3 days, 4 weeks, 12 weeks, 24
weeks and immunoblotted for AR and loading control GAPDH. WT and mutant cells
showed a upregulation of AR over the duration of estrogen deprivation.

3d 4w 12w 24w
AR
WT
3d 4w 12w 24w
Y537S
3d 4w 12w 24w
D538G
GAPDH
36

Figure 2.11 LTED 4 weeks: Androgen receptor suppresses growth in presence of
E2 and DHT. (A), (B), (C) MCF7 ER WT, Y537S or D538G cells were cultured in CSS
media for 4 weeks and plated at low density for colony formation assay. Wells were
treated with vehicle, 1nM DHT, 1nM E2 or 1nM E2 + 1nM DHT for seven days.
Representative images of crystal violet-stained colonies are shown on left, and
quantification shown on right. Data represent the mean ± s.e.m. of three experiments,
with 4-6 replicates per condition per experiment. Data were analyzed using a two-way
Vehicle DHT E2 E2 + DHT
WT-EV WT-sg1 WT-sg2
Vehicle DHT E2 E2 + DHT
Y537S-EV Y537S-sg1 Y537S-sg2
Vehicle DHT E2 E2 + DHT
D538G-EV D538G-sg1 D538G-sg2
EV sg1 sg2
0.00000
0.00001
0.00002
0.00003
WT
Normalized Absrobance 590nm
Vehicle
DHT
E2
E2+DHT
✱✱✱✱
✱✱
✱✱✱✱
✱✱
✱
✱✱
✱✱✱✱
✱✱
✱✱✱✱
✱
✱✱✱
✱
✱✱✱
EV sg1 sg2
0.00000
0.00001
0.00002
0.00003
0.00004
YS
Normalized Absrobance 590nm
Vehicle
DHT
E2
E2+DHT
✱
✱✱✱
✱ ✱
EV sg1 sg2
0.00000
0.00001
0.00002
0.00003
0.00004
0.00005
DG
Normalized Absrobance 590nm
Vehicle
DHT
E2
E2+DHT
✱
✱✱
✱
✱
A
B
C
MCF7 ER WT
MCF7 ER Y537S
MCF7 ER D538G
37
ANOVA followed by Tukey’s multiple comparisons test. P values are indicated by
asterisks: * P£ 0.05; ** P£ 0.01; *** P£ 0.001; **** P£ 0.0001. (A) ER WT EV in presence
of E2 and DHT exhibited suppressed growth, when compared to E2 treatment alone. The
growth suppression was not observed in AR knockout cell lines when E2+DHT treatment
group was compared to E2 alone group. (B) ER Y537S EV in presence of E2 and DHT
showed suppressed growth, when compared to E2 treatment alone. The growth
suppression was not observed in AR knockout lines when E2+DHT treatment group was
compared to E2 alone group. Moreover, DHT alone mediates a suppressive growth
phenotype when compared to vehicle group. And (C) ER D538G EV showed a trend
towards suppressed growth in presence of E2 and DHT, compared to E2 alone.
Moreover, treatment with DHT alone induced a trend towards suppressive growth
phenotype compared to vehicle.

38

Figure 2.12 LTED 8 weeks: Androgen receptor suppresses growth in presence of
E2 and DHT. (A), (B), (C) MCF7 ER WT, Y537S or D538G cells were cultured in CSS
media for 8 weeks and plated at low density for colony formation assay. Wells were
treated with vehicle, 1nM DHT, 1nM E2 or 1nM E2 + 1nM DHT for seven days.
Representative images of crystal violet-stained colonies are shown on left, and
quantification shown on right. Data represent the mean ± s.e.m. of three experiments,
Vehicle DHT E2 E2 + DHT
WT-EV WT-sg1 WT-sg2
Vehicle DHT E2 E2 + DHT
Y537S-EV Y537S-sg1 Y537S-sg2
Vehicle DHT E2 E2 + DHT
D538G-EV D538G-sg1 D538G-sg2
EV sg1 sg2
0.000000
0.000002
0.000004
0.000006
0.000008
WT
Normalized Absrobance 590nm
Vehicle
DHT
E2
E2 + DHT
✱✱✱
✱
✱✱
✱ ✱
✱✱
✱
✱✱
EV sg1 sg2
0.000000
0.000002
0.000004
0.000006
0.000008
YS
Normalized Absrobance 590nm
Vehicle
DHT
E2
E2 + DHT
✱
✱✱✱
✱ ✱✱
EV sg1 sg2
0.000000
0.000002
0.000004
0.000006
0.000008
0.000010
DG
Normalized Absrobance 590nm
Vehicle
DHT
E2
E2 + DHT
✱
✱✱✱
✱✱✱✱
✱✱✱✱
✱✱
✱✱✱✱
A
B
C
MCF7 ER WT
MCF7 ER Y537S
MCF7 ER D538G
39
with 4-6 replicates per condition per experiment. Data were analyzed using a two-way
ANOVA followed by Tukey’s multiple comparisons test. P values are indicated by
asterisks: * P£ 0.05; ** P£ 0.01; *** P£ 0.001; **** P£ 0.0001. (A) ER WT EV in presence
of E2 and DHT exhibited a trend towards suppressed growth, when compared to E2
treatment alone. The growth suppression was not observed in AR knockout cell lines
when E2+DHT treatment group is compared to E2 alone group. (B)ER Y537S EV in
presence of E2 and DHT showed suppressed growth, when compared to E2 treatment
alone. The growth suppression was not observed in AR knockout lines when E2+DHT
treatment group was compared to E2 alone group. Moreover, DHT alone mediated a
suppressive growth phenotype when compared to vehicle group. and (C) ER D538G EV
showed a trend towards suppressed growth in presence of E2 and DHT, compared to E2
alone. Moreover, treatment with DHT alone induced a trend towards suppressive growth
phenotype compared to vehicle.

40

Figure 2.13 LTED 12 weeks: Androgen receptor suppresses growth in presence of
E2 and DHT. (A), (B), (C) MCF7 ER WT, Y537S or D538G cells were cultured in CSS
media for 12 weeks and plated at low density for colony formation assay. Wells were
treated with vehicle, 1nM DHT, 1nM E2 or 1nM E2 + 1nM DHT for seven days.
Representative images of crystal violet-stained colonies are shown on left, and
quantification shown on right. Data represent the mean ± s.e.m. of three experiments,
with 4-6 replicates per condition per experiment. Data were analyzed using a two-way
Vehicle DHT E2 E2 + DHT
WT-EV WT-sg1 WT-sg2
Vehicle DHT E2 E2 + DHT
Y537S-EV Y537S-sg1 Y537S-sg2
Vehicle DHT E2 E2 + DHT
D538G-EV D538G-sg1 D538G-sg2
EV sg1 sg2
0.000000
0.000001
0.000002
0.000003
0.000004
Normalized Absrobance 590nm
04.08.22.Normalized LTED 12WEEKS WT CF Assay - plated 01.14.22 01.27.22 01.26.22
Vehicle
DHT
E2
E2+DHT
✱✱
✱✱✱
✱
✱✱✱
✱✱
✱✱✱
✱
EV sg1 sg2
0.000000
0.000002
0.000004
0.000006
0.000008
04.08.22Normalized LTED 12 WEEKS YS CF Assay - plated 01.14.22 01.27.22 01.26.22
Normalized Absrobance 590nm
Vehicle
DHT
E2
E2+DHT
✱✱✱✱
✱✱✱✱
✱✱✱✱
✱✱✱✱
EV sg1 sg2
0.000000
0.000005
0.000010
0.000015
04.08.22Normalized LTED 12 WEEKS DG CF Assay - plated 01.14.22 01.27.22 01.26.22
Normalized Absrobance 590nm
Vehicle
DHT
E2
E2+DHT
✱✱✱
✱
✱✱ ✱✱✱✱
✱✱✱✱
✱✱✱✱
✱✱✱✱
A
B
C
MCF7 ER WT
MCF7 ER Y537S
MCF7 ER D538G
41
ANOVA followed by Tukey’s multiple comparisons test. P values are indicated by
asterisks: * P£ 0.05; ** P£ 0.01; *** P£ 0.001; **** P£ 0.0001. (A) ER WT EV in presence
of E2 and DHT exhibited a suppressed growth, when compared to E2 treatment alone.
The growth suppression was not observed in AR knockout cell lines when E2+DHT
treatment group is compared to E2 alone group. (B)ER Y537S EV in presence of E2 and
DHT showed suppressed growth, when compared to E2 treatment alone. The growth
suppression was not observed in AR knockout lines when E2+DHT treatment group is
compared to E2 alone group. Moreover, DHT alone mediated a suppressive growth
phenotype when compared to vehicle group. and (C) ER D538G EV showed a trend
towards suppressed growth in presence of E2 and DHT, compared to E2 alone.
Moreover, treatment with DHT alone induced suppressive growth phenotype compared
to vehicle.

Figure 2.14 LTED 4 weeks - AR is important for anchorage independent growth of
LTED MCF7s in soft agar assay. Representative images on left, quantification of wells
for colony area and colony intensity on right. Data represent the mean ± s.e.m. of three
experiments, with 2-3 replicates per condition per experiment. Data were analyzed using
one-way ANOVA followed by Dunnett’s multiple comparisons test. P values are indicated
by asterisks: * P£ 0.05; ** P£ 0.01, *** P£ 0.001. WT-sg1 showed lower colony area and
colony intensity compared to EV. Y537S-sg1 and Y537S-sg2 exhibited lower colony area
and colony intensity compared to EV. D538G-sg1 exhibited a trend towards lower colony
area and colony intensity, compared to EV. D538G-sg2 showed no change compared to
EV.

WT-EV
WT-sg1
WT-sg2
0
2
4
6
8
WT 4WEEKS - % Colony Area
Normalized % Colony Area
✱
WT-EV
WT-sg1
WT-sg2
0.0
0.5
1.0
1.5
2.0
2.5
WT 4WEEKS. - % Colony Intensity
Normalized % Colony Intensity
✱
✱
Y537S-EV
Y537S-sg1
Y537S-sg2
0
5
10
15
YS 4WEEK - % Colony Area
Normalized % Colony Area
✱✱
✱
Y537S-EV
Y537S-sg1
Y537S-sg2
0
1
2
3
4
YS 4WEEKS - % Colony Intensity
Normalized % Colony Intensity
✱✱✱
✱✱
D538G-EV
D538G-sg1
D538G-sg2
0
1
2
3
4
DG 4WEEKS - % Colony Area
Normalized % Colony Area
D538G-EV
D538G-sg1
D538G-sg2
0.0
0.5
1.0
1.5
DG 4 WEEKS - % Colony Intensity
Normalized % Colony Intensity
EV sg1 sg2
WT
Y537S
D538G
Colony Area Colony Intensity
42

Figure 2.15 LTED 8 weeks - AR is important for anchorage independent growth of
LTED MCF7s in soft agar assay. Representative images on left, quantification of wells
for colony area and colony intensity on right. Data represent the mean ± s.e.m. of four
experiments, with 2-3 replicates per condition per experiment. Data were analyzed using
one-way ANOVA followed by Dunnett’s multiple comparisons test. P values are indicated
by asterisks: * P£ 0.05; ** P£ 0.01, *** P£ 0.001. WT-sg1 showed lower colony area and
colony intensity compared to EV. WT-sg2 exhibited a trend towards lower colony area
and colony intensity compared to EV. Y537S-sg1 exhibited lower colony area and colony
intensity compared to EV. Y537S-sg2 exhibited a trend towards lower colony area and
colony intensity compared to EV. D538G-sg1 showed lower colony area and colony
intensity, compared to EV. D538G-sg2 showed higher colony area and colony intensity
compared to EV.

EV sg1 sg2
WT
Y537S
D538G
WT-EV
WT-sg1
WT-sg2
0
2
4
6
8
10
LTED 8WEEKS - WT % Colony Area
Normalized % Colony Area
WT-EV
WT-sg1
WT-sg2
✱
WT-EV
WT-sg1
WT-sg2
0
1
2
3
4
LTED 8WEEKS - WT % Colony Intensity
Normalized % Colony Intensity
WT-EV
WT-sg1
WT-sg2
✱
YS-EV
YS-sg1
YS-sg2
0
5
10
15
LTED 8WEEKS - YS % Colony Area
Normalized % Colony Area
YS-EV
YS-sg1
YS-sg2
✱✱✱
YS-EV
YS-sg1
YS-sg2
0
1
2
3
4
5
LTED 8WEEKS - YS % Colony Intentsity
Normalized % Colony Intensity
YS-EV
YS-sg1
YS-sg2
✱✱
DG-EV
DG-sg1
DG-sg2
0
1
2
3
4
5
LTED 8WEEKS - DG % Colony Intensity
Normalized % Colony Intensity
DG-EV
DG-sg1
DG-sg2
✱✱
✱
DG-EV
DG-sg1
DG-sg2
0
5
10
15
LTED 8WEEKS - DG % Colony Area
Normalized % Colony Area
DG-EV
DG-sg1
DG-sg2
✱✱
✱
43

Figure 2.16 LTED 12 weeks: AR is important for anchorage independent growth of
LTED MCF7s in soft agar assay. Representative images on left, quantification of wells
for colony area and colony intensity on right. Data represent the mean ± s.e.m. of three
experiments, with 2-3 replicates per condition per experiment. Data were analyzed using
one-way ANOVA followed by Dunnett’s multiple comparisons test. P values are indicated
by asterisks: * P£ 0.05; ** P£ 0.01. WT-sg1 showed lower colony area and colony
intensity compared to EV. No change observed in Y537S-sg1, Y537S-sg2 compared to
Y537S-EV. D538G-sg1 showed lower colony area and colony intensity, compared to EV.
D538G-sg2 showed no change compared to EV.

EV sg1 sg2
WT
Y537S
D538G
WT-EV
WT-sg1
WT-sg2
0
2
4
6
WT 12WEEKS - % Colony Area
Normalized % Colony Area
✱✱
WT-EV
WT-sg1
WT-sg2
0.0
0.5
1.0
1.5
2.0
WT 12WEEKS - % Colony Intensity
Normalized % Colony Intensity
✱✱
DG-EV
DG-sg1
DG-sg2
0
2
4
6
8
10
DG 12WEEKS. - % Colony Area
Normalized % Colony Area
✱✱
DG-EV
DG-sg1
DG-sg2
0
1
2
3
4
DG 12WEEKS - %Colony Intensity
Normalized % Colony Intensity
✱✱
Colony Area Colony Intensity
YS-EV
YS-sg1
YS-sg2
0
5
10
15
YS 12WEEKS - %ColonyArea
Normalized % Colony Area
✱
0
1
2
3
4
5
YS 12WEEKS - %Colony Intensity
Normalized % Colony Intensity
44
2.2.4 Androgen receptor supports anchorage-independent survival in sodium
pyruvate-free environment.

Tumor cells that undergo metastatic cascade periodically experience detachment
from ECM. To successfully metastasize, cells must adapt new mechanisms to survive in
anchorage-independent conditions. Elevated levels of reactive oxygen species (ROS) in
matrix-detached cells have been shown to induce cell death (Labuschagne et al., 2019;
Hawk et al., 2018; Jiang et al., 2016). Moreover, treatment of detached cells with
antioxidants improves cell survival (Schafer et al., 2009), circulating tumor cell (CTC)
survival in vitro (Teng and Kamal et al., 2021); and promotes metastasis in vivo (Gal et
al., 2015; Piskounova et al., 2015). Since matrix-detached cells can experience elevated
ROS levels, we hypothesized AR may counteract this effect to improve cell survival.
Results from soft agar assay with AR KO cells indicate a growth suppression in sg1 but
not always in sg2. This suggests there maybe interference from antioxidants present in
cell culture media. Sodium pyruvate, a key component of most cell culture media, is an
antioxidant and protects cells from ROS-induced oxidative stress (Wang et al., 2007;
Ramso-Ibeas et al., 2017). To test if presence of sodium pyruvate in cell culture media
affects the role of AR in survival of anchorage-independent conditions, AR KO cells were
plated in soft agar assay with and without sodium pyruvate (Figure 2.17). In presence of
full-serum media and sodium pyruvate, WT-sg1 and D538G-sg1 show lower percent
colony area and percent colony intensity compared to EV. Y537S-sg1 show no change
compared to EV. AR KO sg2 shows no change compared to EV in WT or mutant ER cells.
However, when plated in full-serum media without sodium pyruvate, both sg1 and sg2
45
show significant reduction in percent colony area and percent colony intensity compared
to EV. This suggests AR supports survival by alleviating oxidative stress from elevated
ROS levels during matrix-detached growth in the absence of antioxidant in cell culture.
To rescue AR-knockout-mediated inhibition of soft agar colonies, androgen
receptor was reintroduced into AR KO cells by lentiviral-mediated overexpression of AR
in MCF7 (WT, Y537S and D538G) sg1 and sg2 cell lines. Detection of AR by western blot
assay in whole cell lysates indicate successful reintroduction of AR in sg1 and sg2 AR
knockout cell lines (Figure 2.18).
AR-rescue cell lines were plated in soft agar assay with sodium pyruvate (Figure
2.19). Reintroduction of AR in WT-sg1 or sg2 cells had no impact on growth in soft agar
assay. However, in Y537S-sg1 and D538G-sg1, reintroduction of AR improved soft agar
colony growth. Reintroduction of AR didn’t markedly impact soft agar colony growth in
Y537S-sg2 and D538G-sg2.
To examine the effect of AR reintroduction on anchorage-independent growth in
full-serum media without sodium pyruvate, AR-rescue cell lines were plated for soft agar
assay (Figure 2.20). Intriguingly, reintroduction of AR in WT-sg1 or sg2 cells had no
impact on growth in soft agar assay. However, reintroduction of AR improved soft agar
colony growth in Y537S-sg1 and D538G-sg1. Moreover, reintroduction of AR improved
soft agar colony growth in D538G-sg2, but not in Y537S-sg2.
Collectively, these data suggests that growth in anchorage-independent conditions
are partially rescued by reintroducing AR in knockout cell lines, and the impact is more
pronounced in the absence of antioxidant sodium pyruvate.

46

Figure 2.17 AR is important for anchorage-independent growth in the absence of
sodium pyruvate. (A) (B) MCF7 EV, sg1, sg2 cultured in standard media and plated for
soft agar assay with sodium pyruvate (A) or without sodium pyruvate (B). Representative
images on left, quantification of wells for colony area and colony intensity on right. Data
represent the mean ± s.e.m. of three experiments, with 2-3 replicates per condition per
experiment. Data were analyzed using one-way ANOVA followed by Dunnett’s multiple
WT-EV
WT-sg1
WT-sg2
0
5
10
15
20
STD WITH SOPY - WT % Colony Area
Normalized % Colony Area
✱✱✱
WT-EV
WT-sg1
WT-sg2
0
2
4
6
8
STD WITH SOPY - WT % Colony Intensity
Normalized % Colony Intensity
✱✱✱
✱
YS-EV
YS-sg1
YS-sg2
0
2
4
6
8
STD WITH SOPY - YS % Colony Area
Normalized % Colony Area
YS-EV
YS-sg1
YS-sg2
0
1
2
3
4
5
STD WITH SOPY - YS % Colony Intensity
Normalized % Colony Intensity
DG-EV
DG-sg1
DG-sg2
0
5
10
15
20
STD WITH SOPY - DG % Colony Area
Normalized % Colony Area
✱✱
DG-EV
DG-sg1
DG-sg2
0
2
4
6
8
STD WITH SOPY - DG % Colony Intensity
Normalized % Colony Intensity
✱✱
WT
Y537S
D538G
EV sg1 sg2
Colony Area Colony Intensity
WT
Y537S
D538G
EV sg1 sg2
WT-EV
WT-sg1
WT-sg2
0
1
2
3
4
STD NO SOPY - WT % Colony Area
Normalized % Colony Area
✱
WT-EV
WT-sg1
WT-sg2
0.0
0.5
1.0
1.5
STD NO SOPY - WT % Colony Intensity
Normalized % Colony Intensity
✱
YS-EV
YS-sg1
YS-sg2
0
2
4
6
8
10
STD NO SOPY - YS % Colony Area
Normalized % Colony Area
✱✱
✱✱
YS-EV
YS-sg1
YS-sg2
0
1
2
3
STD NO SOPY - YS % Colony Intensity
Normalized % Colony Intensity
✱
✱✱
DG-EV
DG-sg1
DG-sg2
0
2
4
6
8
STD NO SOPY - DG % Colony Area
Normalized % Colony Area
✱✱✱✱
✱✱✱✱
DG-EV
DG-sg1
DG-sg2
0
1
2
3
STD NO SOPY - DG % Colony Intensity
Normalized % Colony Intensity
✱✱✱✱
✱✱✱✱
Colony Area Colony Intensity
A
B
47
comparisons test. P values are indicated by asterisks: * P£ 0.05; ** P£ 0.01. (A) In the
presence of sodium pyruvate in the media, WT-sg1 exhibited lower colony area and
colony intensity compared to WT-EV. WT-sg2 exhibited a trend towards lower colony
area and colony intensity compared to WT-EV. YS-sg1 and YS-sg2 showed no changes
compared to YS-EV. DG-sg1 showed lower colony area and colony intensity compared
to DG_EV. DG-sg2 showed no changes compared to DG-EV. (B) In the absence of
sodium pyruvate in the media, WT-sg2 showed a lower colony area and colony intensity.
WT-sg1 exhibited a trend towards lower colony growth compared to EV. In Y537S and
D538G cells, both sg1 and sg2 showed diminished colony growth compared to EV.

Figure 2.18 Reintroduction of AR in AR KO cell lines. Androgen receptor was
reintroduced into AR KO cells by lentiviral-mediated overexpression of AR in MCF7 WT,
Y537S and D538G cell lines. Detection of AR by western blot assay in whole cell lysates
indicated successful reintroduction of AR in sg1 and sg2 knockout cell lines.

AR
GAPDH
EV AR
ER WT ER Y537S ER D538G
EV AR
sg1 sg2
EV AR EV AR
sg1 sg2
EV AR EV AR
sg1 sg2
48

Figure 2.19 Reintroduction of AR partially restores growth of soft agar colonies in
full-serum media with sodium pyruvate. AR was reintroduced in MCF7 (WT, Y537S,
D538G) sg1 or sg2 cells and grown in full-serum media with sodium pyruvate for soft agar
assay. Representative images and quantification for sg1 on left and for sg2 on right. Data
represent the mean ± s.e.m. of three experiments, with 2-3 replicates per experiment.
Data were analyzed by paired t-test, two-tailed. Reintroduction of AR in WT-sg1 or sg2
cells had no impact on growth in soft agar assay. Reintroduction of AR improved soft agar
colony growth in Y537S-sg1 but not in Y537S-sg2. Reintroduction of AR improved soft
agar colony growth in D538G-sg1 but not in D538G-sg2.

EV AR
0
2
4
6
8
WT-sg1-ColonyArea
WT-sg1
Normalized % Colony Area
EV AR
WT-sg1
EV AR
0
5
10
15
20
YS-sg1-ColonyArea
YS-sg1
Normalized % Colony Area
✱
Y537S-sg1
EV AR
0
5
10
15
20
DG-sg1-ColonyArea
DG-sg1
Normalized % Colony Area
✱
D538G-sg1
EV AR
0
2
4
6
8
10
WT-sg2-ColonyArea
WT-sg2
Normalized % Colony Area
EV AR
WT-sg2
EV AR
0
5
10
15
20
YS-sg2-ColonyArea
YS-sg2
Normalized % Colony Area
Y537S-sg2
EV AR
0
5
10
15
20
DG-sg2-ColonyArea
DG-sg2
Normalized % Colony Area
D538G-sg2
sg1 sg2
49

Figure 2.20 Reintroduction of AR partially restores growth of soft agar colonies in
full-serum media without sodium pyruvate. AR was reintroduced in MCF7 (WT,
Y537S, D538G) sg1 or sg2 cells and grown in full-serum media without sodium pyruvate
for soft agar assay. Representative images and quantification for sg1 on left and for sg2
on right. Data represent the mean ± s.e.m. of three experiments, with 2-3 replicates per
experiment. Data were analyzed by paired t-test, two-tailed. Reintroduction of AR in WT-
sg1 or sg2 cells had no impact on growth of soft agar colonies. Reintroduction of AR
improved growth of soft agar colonies in Y537S-sg1, but not in Y537S-sg2. Reintroduction
of AR improved growth of soft agar colonies in both D538G-sg1 and D538G-sg2.

EV AR
0.0
0.5
1.0
1.5
2.0
WT-SG1- Colony AREA
WT-sg1
% Colony Area
EV AR
0.0
0.5
1.0
1.5
2.0
2.5
YS-SG1-Colony Area
YS-sg1
% Colony Area
✱✱
EV AR
WT-sg1 Y537S-sg1
EV AR
0.0
0.5
1.0
1.5
2.0
DG-SG1-Colony Area
DG-sg1
%Colony Area
✱✱
D538G-sg1
EV AR
0.0
0.1
0.2
0.3
WT-SG2- Colony Area
WT-sg2
% Colony Area
EV AR
0.0
0.5
1.0
1.5
2.0
2.5
YS - SG2-Colony Area
YS-sg2
% Colony Area
EV AR
WT-sg2 Y537S-sg2
EV AR
0.0
0.5
1.0
1.5
2.0
DG-SG2-Colony Area
DG-sg2
% Colony Area
✱✱
D538G-sg2
sg1 sg2
50
2.3 Discussion
Role of AR in ER+ positive breast cancer is vigorously debated, with evidence for
both protumorigenic and antitumor activities. Gain-of-function mutations in ligand binding
domain of ER are a common mechanism of endocrine resistance to AI therapy in
postmenopausal women with ER+ breast cancer. In this study using MCF7 and T47D
with ESR1 mutations Y537S and D538G, we evaluated the role of AR. In full-serum
media, AR was upregulated at both transcript and protein levels in MCF7 and T47D
Y537S compared to WT. In hormone-depleted CSS media, at transcript level, AR
expression was moderately upregulated in MCF7 Y537S but not in D538G. There were
no changes at transcript level of AR in T47D Y537S compared to WT, but expression was
lower in D538G compared to WT. At protein level, AR was upregulated in both MCF7 and
T47D Y537S compared to WT, both in vehicle and in presence of E2. There was no
change in D538G compared to WT. Moreover, Y537S was associated with elevated
levels of nuclear AR. To precisely investigate the role of AR in mutant ER cells, CRISPR-
Cas9 was utilized to generate AR KO models in MCF7 ER WT, Y537S and D538G cells.
In short-term estrogen deprived conditions, AR suppressed growth in adherent conditions
in presence of E2 and DHT in both WT and ER mutant cells. Of note, just deletion of AR
did not show any significant growth promoting activity in the absence of E2. On the
contrary, AR KO inhibited growth in anchorage-independent conditions, and knocking out
AR diminished total metastatic burden of MCF7 Y537S cells in ovariectomized immune
deficient mouse models. In long-term estrogen deprived (LTED) MCF7 models, AR
expression progressively increased over time in WT and mutant cells. AR KO LTED
models show AR suppresses proliferation in adherent growth conditions and promotes
51
survival in anchorage-independent conditions – similarly to what was observed in short-
term estrogen deprivation conditions. Intriguingly, role of AR in supporting growth in
matrix-detached cells is prominent in the absence of antioxidant sodium pyruvate in
media. Reintroduction of AR partially restores growth in anchorage-independent
conditions, which is more obvious in the absence of sodium pyruvate. Collectively, these
data point to a working model in which AR is upregulated in cells with mutant ER and
plays a context-dependent role during metastasis.
Role of AR in prostate cancer has been well characterized. Despite decades of
attempts to disentangle its role in ER+ breast cancer and metastasis, it remains poorly
understood. Recently, another study identified an association between increased AR
expression and ESR1 mutations in MCF7 and T47D cell lines, largely corroborating the
data presented in our study (Williams et al., 2021). Interestingly, in addition to AR, ER,
PR and GR were also upregulated in LTED cells with ESR1 mutations. Estradiol
treatment increased nuclear localization of AR in MCF7, ZR-75-1; and nuclear localization
is abrogated when treated with enzalutamide (D’Amato et al., 2016). Data presented in
our study indicate higher baseline nuclear AR levels in Y537S which improved further
with estradiol treatment. Utilizing multiple cell line and patient-derived models, DHT-
mediated activation of AR in presence of E2 suppressed tumor proliferation of ER+ breast
cancer in vitro and in vivo (Hickey et al., 2021). Our findings on AR-mediated growth
suppression in E2 and DHT treated Y537S and D538G mutant cells have not been
reported before. While DHT treatment suppresses proliferation of wildtype MCF7, it
increases migratory and invasive abilities in vitro (Feng et al., 2017). Ectopic expression
of AR increased these metastatic phenotypes, and by contrast short hairpin RNA (shRNA)
52
mediated knockdown of AR diminished it, suggesting these phenotypes are AR mediated.
In addition to LTED MCF7s, AR is also increased in tamoxifen-resistant derivative of
MCF7 (TamR) (Chia et al., 2019). Knocking down of AR suppressed proliferation in
adherent culture of TamR and MCF7 LTED models. Interestingly, the knockdown restored
sensitivity to tamoxifen in TamR cells. Letrozole is an aromatase inhibitor (AI) used in the
treatment of breast cancer (Mauri et al., 2006). AR is upregulated in letrozole-resistant
MCF7 and ZR75-1 models (Creevey et al., 2019).
Role of AR in growth in anchorage-independent conditions have been previously
reported in various contexts. DHT treatment improved mammosphere formation of MCF7
cells (Feng et al., 2017). This phenotype markedly increased on ectopic expression of
AR, and diminished with shRNA targeting AR. In another study, soft agar colony
formation of MCF7 TamR cells were reduced by siRNA targeting AR. Tamoxifen
treatment further improved the effect suggesting a restoration of sensitivity to the drug by
siAR (Chia et al.., 2019). Suspension culture of MCF7 and T47D in poly-hema coated
plates increased nuclear AR levels in WT and ESR1 mutant cells (Williams et al., 2021).
Enzalutamide is a nonsteroidal antiandrogen used in the treatment of prostate cancer
(Hussain et al., 2018; Davis et al., 2019). Enzalutamide activity in both pre-clinical and
clinical settings of ER+ and triple-negative breast cancer subtypes is an active area of
research (Traina et al., 2018; Krop et al., 2020; Cochrane et al., 2014; D’Amato et al.,
2016). Enzalutamide treatment of MCF7 and T47D reduced soft agar colonies of cells
harboring WT and mutant ER (Williams et al., 2021). This effect was more pronounced
when treated with seviteronel, a CYP17A1 lyase inhibitor that impedes synthesis of
androgens and estrogens. Increase in AR expression in endocrine-resistant models
53
(LTED, TamR, AI resistance, ESR1 mutations) and the effect of AR on anchorage-
independent growth have been previously reported. Evidence presented in our study
further supports these findings. Interestingly, our data indicate that AR is upregulated over
the duration of LTED, nonetheless maintaining its differential role in tumor growth under
adherent and anchorage-independent conditions. Recent studies have investigated the
role of AR in ER+ breast cancer via AR knockdown. However, AR knockout models of
ESR1 mutant cells have not been previously reported.
Similar to in vitro studies, conflicting roles for AR have also been reported from
various in vivo studies. Utilizing multiple patient-derived xenograft and cell line models
and E2 supplementation, DHT treatment diminished growth of primary tumor in vivo
(Hickey et al., 2021). However, its effect on metastasis was not evaluated. In an
orthotopic model of MCF7, DHT treatment had no effect on primary tumor, but greatly
enhanced lung metastasis. Furthermore, shRNA mediated AR knockdown abrogated
lung metastasis (Feng et al., 2017). In a patient-derived xenograft model, enzalutamide
treatment reduced primary tumor burden. Additionally, total metastatic burden of
intracardially injected cells was reduced on treatment with enzalutamide (D’Amato et al.,
2016). Tail vein injection of T47D in ovariectomized NSG mice demonstrated higher
whole mouse burden and lung metastatic burden in D538G compared to WT cells.
Elevated levels of AR in D538G point to a role in metastasis of these cells (Williams et
al., 2021). In our study, to model metastasis in post-menopausal women where ESR1
mutations arise, cells were cultured in hormone-depleted media and injected intracardially
into ovariectomized NSG mice without E2 supplementation. While Y537S cells were able
to establish higher metastatic tumor burden compared to WT, knockout of AR diminished
54
this effect – suggesting a role for AR in metastasis. Role of AR in this context has not
been reported previously. Collectively, these data suggest role of AR in primary tumor
and metastasis is context dependent.
Mechanisms to survive anchorage-independent conditions are crucial for
successful metastasis. Increased ROS levels in matrix-detached cells have been
previously reported (Labuschagne et al., 2019; Hawk et al., 2018; Jiang et al., 2016).
Elevated ROS levels in matrix-detached normal mammary epithelial cells were rescued
by overexpression of oncogene ERBB2. Additionally, treatment with antioxidant promoted
growth of soft agar colonies (Schafer et al., 2009) and promote metastasis in vivo (Gal et
al., 2015; Piskounova et al., 2015). Sodium pyruvate, a key component of cell culture
media, is an antioxidant and protects cells from ROS-induced oxidative stress (Wang et
al., 2007; Ramso-Ibeas et al., 2017). In our study, removal of antioxidant sodium pyruvate
from cell culture abrogated soft agar colonies of AR knockout cells. Ectopic expression of
AR in knockout cells partially rescued soft agar colonies, especially in mutant ER cells.
These data suggest there maybe additional factors affecting the role of AR in survival of
WT and mutant ER cells in anchorage-independent settings. Future studies to investigate
role of AR in counteracting ROS in matrix-detached cells may help identify mechanisms
of tumor cell survival during metastasis. All the evidence presented in the current work
point to differing roles of AR in adherent versus anchorage-independent contexts.
Additional studies in these contexts may help unravel the controversy in the field, and
explain contradictory roles attributed to AR.

55
2.4 Methods

Cell culture
MCF7 and T47D cell lines with knock-in ESR1 mutations (Y537S, D538G) were
kindly provided by Dr. Steffi Oesterreich, University of Pittsburgh, PA USA (Bahreini et
al., 2017). MCF7 cell lines were cultured in DMEM 10% FBS, 1% Penicillin-Streptomycin.
T47D cell lines were in RPMI, 10% FBS, 1% Penicillin-Streptomycin. For hormone
treatment experiments, MCF7 cell lines were cultured in phenol-free DMEM, 10%
Charcoal-Dextran Stripped FBS, 1% Penicillin-Streptomycin; T47D cell lines were
cultured in phenol-free RPMI, 10% Charcoal-Dextran Stripped FBS, 1% Penicillin-
Streptomycin. Cell lines were routinely tested for mycoplasma contamination with
MycoAlert (Lonza).

Construct generation
Guides for AR knockout were designed using Broad Institute GPP Web Portal to
target exon 1 of AR. Sequence for guide RNAs are as follows:
sg1 forward: 5’- CACCGAGGGTACCACACATCAGGTG-3’
sg1 reverse: 5’- AAACCACCTGATGTGTGGTACCCTC-3’
sg2 forward: 5’- CACCGCCTTAAAGACATCCTGAGCG-3’
sg2 reverse: 5’-AAACCGCTCAGGATGTCTTTAAGGC-3’
Guides were cloned into PX458 based on previously published protocol from Feng Zhang
lab (Ran et al., 2013). MCF7 WT, Y537S and D538G cell lines were transfected with
PX458 using Lipofectamine 3000. Two days later GFP+ single cells were sorted into 96-
56
well plates to generate clonal lines. AR knockout was verified by western blot assay. To
reintroduce AR in knockout lines, pLENTI6.3/AR-GC-E2325 (Addgene #85128).
Lentivirus of pLENTI6.3/AR-GC-E2325 were generated as described below in the section
on lentiviral mediated generation of stable cell lines. Cells were transduced with
pLENTI6.3/AR-GC-E2325 lentivirus and selected with blasticidin. AR expression was
verified by western blot assay.

Lenitviral mediated generation of stable cell lines
Lentivirus were produced utilizing protocol published by the RNAi Consortium
(TRC) of the Broad Institute. Accordingly, low passage 293T cells were cultured in DMEM
10% FBS without antibiotics for packaging lentivirus. In brief, TransIT-LT1 transfection
reagent (Mirus) was utilized to co-transfect above-mentioned expression constructs and
second-generation lentiviral packaging vectors into 293T cells. Next day, to promote high
titer virus production, media for 293T cells were replaced high serum growth medium.
Virus containing supernatants were collected at both 48- and 72-hours post-transfection.
The supernatants were concentrated with Lenti-X concentrator (Clonetech), as per
manufacturer’s instructions. To generate stable cell lines with lentivirus, 200,000 cells
were plated in a 6-well plate and transduced with 100uL of concentrated virus along with
8 ug/mL polybrene. Cells were selected by either antibiotic treatment or by FACS, as
appropriate.

57
Gene expression analysis by qRT-PCR
To extract RNA, cells were plated in 12-well or 6-well plates. Quick RNA Microprep
Kit (Zymo) was used to extract RNA, as per manufacturer’s instructions. 5X iScript
supermix (Bio-Rad) was used to reverse transcribe and generate cDNA from 100-1000ng
RNA. iQ SYBR Green Supermix (Bio-Rad) was used to run real-time quantitative PCR
reaction on a CFX iCycler real-time PCR machine (Bio-Rad) with 10 ng of cDNA per
reaction. Table# lists primer sequences used in qRT-PCR assays.

Western Blot
Cells were washed in PBS and whole cell lysates were collected by lysing in
Laemmli Buffer (50mM Tris pH=6.8, 1.25% SDS, 15% glycerol). Lysates were heated at
95C for 15 minutes for denaturation. Total protein concentration in the lysates was
quantificatied using Lowry protein assay (Bio-Rad). Beta-mercaptoethanol was added to
a final concentration of 5% (v/v) as a reducing agent. Bromophenol blue was added at a
final concentration of (0.01% v/v) as a loading dye. Samples were then heated at 95C for
5 minutes before loading on a 4-15% Mini-PROTEAN TGX Precast gels (Bio-Rad). Low
fluorescence PVDF membranes were used to transfer proteins from the gel by semi-dry
method utilizing Trans-Blot Turbo Transfer System (Bio-Rad) using the standard transfer
protocol. For protein detection, protocol from LI-COR Odyssey was used. In brief,
membranes were dried for a minimum of one hour, or overnight. Following activation with
ethanol, and a wash step with TBS, a 5% non-fat dry milk (NFDM) in TBS was used as
blocking solution. Primary antibody was diluted in blocking buffer containing 0.2%
Tween20 and incubated at 4C overnight on a rocker. Next day, membranes were washed
58
three times with 1X TBS containing 0.1% Tween20, ten minutes per wash. Meanwhile,
secondary antibodies were diluted in blocking buffer containing 0.2% Tween 20 and
0.01% SDS. Next, the membranes were incubated with secondary antibodies for one
hour at room temperature with gentle shaking, covered in dark. Next, membranes were
then washed three times, ten minutes each, with 1X TBS containing 0.1% Tween20.
Finally, the membranes were rinsed in TBS, and imaged using a LI-COR imaging
instrument. The following antibodies were used:

Colony formation assay
Cells cultured under hormone treatment conditions were washed once with DPBS.
Cells were dissociated by incubating with TrypLE express enzyme (Thermo Fisher, Cat#
12604013) at 37C for 3 minutes. The cell suspension was filtered using a 40 µm cell
strainer (Corning, Cat# 352340) and counted using TC-20 Automated Cell Counter (Bio-
Rad). Cells were plated at a density of 2000 cells/well in 1 mL volume per well in 24-well
tissue culture-treated plates. Plates were incubated at 37C overnight to enable cell
attachment. Next day, media was replaced with fresh media containing vehicle, 1nM E2,
1nM DHT or 1nM DHT + 1nM E2. Cells were cultured for 7 days with drug changes every
2-3 days. At the end of treatment, cells were washed twice with cold PBS containing 1
mM CaCl2 and 1 mM MgCl2. Freezer-cold methanol was used to fix cells for 10 minutes
at RT, followed by three washes with 1X PBS. Methanol and washes were disposed as
per lab safety instructions. Cells were incubated with crystal violet (0.5% wv crystal violet
in 20% ethanol) for 10 minutes at room temperature with gentle shaking on a rocker. To
remove excess crystal violet, plates were gently immersed and rotated in gentle circular
59
motion for one minute each in three 2L beakers of distilled water. Crystal violet and
washes were disposed as per lab safety instructions. Plates were allowed to dry overnight
and then imaged with Keyence BZ-X810 microscope at 4X magnification in stitching
mode. To recover crystal violet, 200uL of 100% methanol was added to each well for 15
minutes at room temperature with gentle shaking on a rocker. For quantification, 100uL
of the recovered crystal violet solution was transferred to clear flat-bottom 96-well plates.
BioTek plate reader was to measure absorbance at 590nm. A schematic of colony
formation assay setup is illustrated below (Figure 2.21).

Figure 2.21 A schematic of colony formation assay. Cells were plated at low density
in CSS media in 24-well plates. Wells were treated with E2, DHT or E2 + DHT for seven
days. Wells were fixed with methanol and stained with crystal violet. Plates were imaged
and crystal violet was eluted for quantification using a plate reader.

60
Soft agar assay
Soft agar assay was performed using previously published protocol (Borowicz et
al., 2014). To prepare materials for the assay, a 1% agar stock solution was prepared by
adding 1g of agar to 100 mL of deionized water. A 0.6% agar stock solution was prepared
by adding 0.6g of agar to 100 mL of deionized water. The agar stock solutions were
autoclaved to dissolve agar, and then cooled to RT and stored at 4C until ready to use.
To plate bottom layer, 1% agar stock solution was microwaved for 1-2 minutes until all
the agar is completely melted. The agar solution was placed in a large beaker filled with
hot tap water and allowed to cool until the temperature reached 42C. A 50 mL tube
containing culture media was placed in the bucket and allowed to equilibrate. Once the
agar solution reached 42C, equal volumes of media and agar were mixed. Next, 1.5 mL
of media-agar mixture was added to each well and allow to solidify at RT for 30 minutes.
The final concentration of bottom layer is 0.5% agar. Meanwhile, cells were washed,
dissociated (Trypsin for cells cultured in standard media, TrypLE express enzyme for cells
cultured in hormone-treatment conditions), filtered using a 40 µm cell strainer, and
counted. Aliquot 80,000 cells into a 15 mL tube containing 3 mL of media. To plate top
layer, 0.6% agar stock solution was microwaved for 1-2 minutes until all the agar is
completely melted. The agar solution was placed in a large beaker filled with hot tap water
and allowed to cool until the temperature reached 42C. When the temperature reached
42C, the 15 mL tube containing cells was placed in the beaker. 3 mL of 0.6% agar was
mixed with 3 mL of media containing cells in the tube. 1.5 mL of this mixture was added
on top of the solidified bottom layer of agar. The top layer was allowed to solidify for 30
minutes at RT. Once the top layer is solidified, 200 uL of media was added on top of the
61
layer and incubated at 37C for three to four weeks. Twice a week wells were
supplemented with 200 uL of media. A stock staining solution of 1 mg/mL Nitroblue
Tetrazolium Chloride (NBT) was prepared by dissolving 50 mg of NBT in 50 mL of PBS.
The stock solution was kept protected from light and stored at 4C. To stain the cells, the
wells were incubated with 200 uL of NBT for overnight at 37C. Next day, plates were
imaged with Keyence BZ-X810 microscope. Colonies were quantified using Image J and
Colony Area plugin. A schematic of soft agar assay setup is illustrated below (Figure
2.22).

Figure 2.22 A schematic of soft agar assay. In 6-well plates, a bottom layer of 0.5%
agar and media was plated. Once cooled and solidified, a layer of 0.3% agar and media
containing cells were plated on top. Once cooled and solidified, media was added on top
to keep the gels hydrated. At the end of timepoint, wells were stained with nitroblue
tetrazolium, imaged with Keyence, and images were processed with Image J and Colony
Area plugin.

62
Immunocytochemistry
12mm or 18mm Coverglass No. 1 coverslips were seeded with cells and allowed
to attached overnight. Next day, prepare 4% paraformaldehyde (PFA) by diluting 16%
PFA (Electron Microscopy Sciences) in PBS. To fix cells, coverslips were washed with
1X DPBS, and then fixed for ten minutes at RT with 4% paraformaldehyde in a chemical
cabinet. Next, coverslips were washed twice in PBS. 4% PFA and washes were disposed
as per lab safety instructions. To permeabilize cells, coverslips were incubated for ten
minutes at RT with 0.25% Triton X-100 in PBS. Next, coverslips were washed with PBS
containing 0.1% Tween20, and blocked with 5% goat serum in PBS for one hour at RT.
Primary antibodies were prepared by diluting in blocking buffer to appropriate
concentration and then incubate for overnight in a humidifying chamber at 4C. Next day,
coverslips were washed three times with PBS containing 0.1% Tween20, 5 minutes each
wash. Then the coverslips were incubated with secondary antibodies diluted in blocking
buffer and incubated for one hour at RT in a dark humidifying chamber. Next, the
coverslips were washed three times with PBS containing 0.1% Tween20, 5 minutes per
wash. To stain nuclei, coverslips were incubated with 4,6-dianidino-2-phenylindole (DAPI)
for 5 minutes in dark, followed by a rinse with PBS. Coverslips were mounted using
ProLong Gold Antifade Mounting media (Thermo Fisher) and allowed to dry overnight in
the dark. Next day, Keyence BZ-X810 was used to image coverslips at 20X and 40X
magnifications. Images were quantified using BZ-X800 Analyzer software.

In vivo experiments
63
All animal experiments were performed in accordance with protocols approved by the
USC Institutional Animal Care and Use Committee. To prepare cells for in vivo metastasis
in mice, GFP-LUC-labeled MCF7s were cultured for 3 days in hormone-free CSS media.
Cells were dissociated and counted to achieve a cell density of 1 x 10
6
cell in 1mL of PBS.
100uL (1 x 10
5
) of cells were injected into the left cardiac ventricles of 6-8-week-old
ovariectomized female NSG mice (The Jackson Laboratory, Strain#005557). Mice were
monitored every week for metastasis by injecting 100uL of D-Luciferin (Syd Labs)
intraperitoneally and followed by imaging with Perkin Elmer IVIS Lumina III.

64
CHAPTER 3
Exploring the role of AR in bone metastasis of ER+ breast
cancer.

3.1 Introduction
Metastasis is the underlying cause of more than 90% cancer-related deaths. Bone,
lungs, liver, and brain are the common sites of distant metastasis for breast cancer. Bone
is the most common site of distant metastasis for breast cancer. Metastasis begins when
tumor cells leave primary site to enter circulation and reach distant sites to seed new
tumors. Tumor cells may also be shed from secondary sites back into circulation where
they continue metastatic cascade. Circulating tumor cells (CTCs) are the population of
tumor cells in circulation. Mounting evidence show that CTCs are an indicator of disease
progression and treatment responses (Larsson et al., 2018; Goldkorn et al., 2021).
Detection of higher number of CTCs has been linked to worse prognoses in patients with
breast cancers (Yu et al., 2011). In metastatic breast cancer patients, CTCs may serve
as a sample of the most active metastatic tumors in patients with late-stage disease.
CTCs are a form of liquid biopsy, with considerable potential for monitoring active tumor
biology and for drug screen. Ex vivo culture of CTCs can provide sufficient material for
genomic and transcriptomic analysis of these rare CTCs (Yu et al., 2014). We recently
showed that luminal breast cancer CTC-derived cell lines can recapitulate metastatic
tropism in mice (Klotz et al., 2020). We identified a strong bone tropism in mice for CTC-
derived cell line BRx68. Intriguingly, inoculation of bone metastasis derivative of BRx68
65
into a new cohort of mice further enhanced bone tropism. Moreover, the patient from
where BRx68 was derived also exhibited bone metastasis. These data suggest BRx68
and its bone derivatives (BRx68-BoM) can be useful as model to study biology of bone
metastasis of ER+ breast cancer. In this chapter, I will focus on the data generated in my
exploratory work to understand bone metastasis of ER+ breast cancer.

3.2 Results
3.2.1 Identification of MBOAT2 as a novel, potential player in ER+ breast cancer
bone metastasis.
Utilizing the data published in our previous work, we analyzed RNA-seq data of
BRx68 parental and BRx68-BoM cell lines to identify genes that may play a potential role
in bone metastasis of this model. Hierarchical clustering of RNA-seq data from BRx68
parental and BRx68-BoM samples showed the parentals and bone derivates clustering
separately, suggesting differential gene expression patterns between these two groups
(Figure 3.1.A). When ordered by adjusted p-value, MBOAT2 is the second most
significantly upregulated gene (Figure 3.1.B). RNA-seq tracks using UCSC genome
browser show higher expression of MBOAT2 in BRx68-BoM compared to parental (Figure
3.1.C). qPCR assay using cDNA from parental and bone metastatic derivates of BRx68
indicate upregulation of MBOAT2 in several BRx68 bone metastatic derivates compared
to parental lines (Figure 3.1.D). Next, I checked THE HUMAN PROTEIN ATLAS
(www.proteinatlas.org) for expression of MBOAT2 in public datasets. RNA expression
data from TCGA dataset show MBOAT2 is expressed in multiple cancers including breast
and prostate (Figure 3.2.A). Protein expression data indicate MBOAT2 is expressed in
66
many cancers including breast and prostate, with moderate to strong cytoplasmic
localization (Figure 3.2.B). To understand the clinical relevance of genes identified from
RNA-seq analysis of BRx68 Parental vs BRx68-BoM, I analyzed the top 25 genes
(ordered by adjusted p-value and a log2 fold-change of greater than 2) in a dataset that
contains 204 primary tumors from breast cancer patients with known sites of relapse (Bos
et al., 2009; GSE12276). Kaplan-Meier curves indicate higher MBOAT2 expression is
significantly associated with lower metastasis-free survival in breast cancer patients with
bone relapse, but not in patients with brain or lung relapse (Figure 3.3). Interestingly,
among all the genes analyzed, MBOAT2 was the only gene with significant association
for bone relapse. Taken together, these data suggest MBOAT2 may play a role in bone
metastasis of breast cancer.

67

Figure 3.1 MBOAT2 is upregulated in bone metastatic derivatives of BRx68. (A)
Hierarchical clustering of RNA-seq data from BRx68 and BRx68-BoM. (B) Results from
differential expression analysis of RNA-seq data comparing BRx68 and BRx68-BoM
showed MBOAT2 is the second most upregulated gene. (C)Tracks for MBOAT2
expression from RNA-seq data using UCSC browser. (D) qPCR assay for MBOAT2 gene
expression in BRx68 parental and BoM cell lines showed upregulation in bone derivates.

Principal)Investigator:))Yu,$Min$ 4.11$
Research$Scholar$Grant$Application$$
July$2019$

contribute$to$the$progression$of$castration$resistant$prostate$cancer,$which$often$leads$to$bone$metastases
51
.$A$
key$enzyme$in$the$lipid$synthesis$pathway$included$in$this$report$as$an$AR$downstream$gene$is$MBOAT2.$
Indeed,$MBOAT2$is$the$second$most$significantly$upregulated$gene$when$we$compared$the$RNA@seq$of$
BRx68@BoM$ with$ BRx68.$ Intriguingly,$ the$
promoter$region$of$MBOAT2$showed$AR$binding$
sites$only$in$cells$with$mutant$ER$(Figure'9).$In$
addition,$a$study$has$shown$that$in$E2$treated$
breast$cancer$cells,$AR$can$facilitate$ER’s$full$
genomic$binding$and$transcriptional$regulation
44
.$
Therefore,$based$on$our$preliminary$data$and$
the$emerging$insights,$we$hypothesize$that$AR$
could$be$a$cofactor$to$facilitate$mutant$ER$in$low$
estrogen$conditions$to$promote$bone$metastasis.$
Here,$ in$ this$ aim,$ we$ will$ determine$ the$
significance$of$AR$in$mutant$ER$context$for$bone$
metastasis$formation.)$
$
2A.$Evaluate$enhanced$bone$metastasis$from$
additional$CRISPR@edited$Y537S$mutant$lines.$$
We$recognized$that$although$BRx68$CTC$line$
has$high$bone$metastasis$frequency,$it$is$still$
limited$to$one$patient$sample.$In$this$aim,$we$will$
use$a$recently$published$caudal$artery$injection$
approach
71
$to$deliver$MCF7@Y537S$and$T47D@
Y537S$knock@in$cell$lines$to$female$NSG$mice$
and$ evaluate$ the$ bone$ metastasis$ formation$
compared$to$the$isogenic$lines$with$wild$type$
ER.$As$a$pilot$experiment,$we$have$injected$MCF7$cells$
via$caudal$artery$injection$into$male$ NSG$ mice.$ The$
result$showed$a$striking$bone$metastasis$formation$in$
both$lower$legs$of$the$male$mice$(Figure'10).$This$data$
not$only$confirmed$the$feasibility$of$this$approach$to$
evaluate$ bone$ metastasis,$ but$ also$ suggested$ a$
potential$involvement$of$ androgen$hormone$ from$ the$
male$mice.$Therefore,$we$will$use$the$same$approach$
to$deliver$the$wild$type$ER$cells$or$Y573S@ER$cells$into$
female$NSG$mice$and$evaluate$the$bone$metastasis$
formation.$ These$ cells$ will$be$ tagged$ with$ GFP$ and$
luciferase$and$the$tumor$signal$will$be$evaluated$via$live$
bioluminescence$ imaging.$ At$ the$ endpoint,$ bone$
metastasis$will$be$analyzed$for$AR$nuclear$levels$via$
immunohistochemistry$staining$for$AR$and$compare$between$wild$type$and$mutant$ER$groups.$$
$
2B.$Evaluate$the$effect$of$suppressing$AR$in$mutant$ER$cells$on$bone$metastasis$$
To$evaluate$the$role$of$AR$in$the$context$of$ER@Y537S$associated$bone$metastasis,$we$will$suppress$AR$via$
shRNAs$in$BRx68$cells$and$evaluate$the$effect$on$bone$metastasis$formation.$Via$caudal$artery,$we$will$inject$
the$BRx68$cells$with$shControl,$and$2$different$shRNAs$against$AR,$as$well$as$AR$rescue$construct$in$the$
shAR$cells.$The$bone$metastasis$growth$will$be$monitored$via$bioluminescent$imaging$once$a$week.$$
$
2C.$Evaluate$the$effect$of$suppressing$NCOA3$in$mutant$ER$cells$on$bone$metastasis.$$
Since$NCOA3$is$a$key$cofactor$for$both$ER$and$AR,$and$being$upregulated$by$ER/AR,$we$will$suppress$
NCOA3$via$shRNAs$in$BRx68$cells$and$evaluate$the$effect$on$bone$metastasis$formation.$Similarly,$as$above,$
we$will$use$caudal$artery$injection$and$monitor$the$signals$of$bone$metastasis$in$shNCOA3$cells$versus$cells$
with$the$control$shRNA,$or$the$NCOA3$rescue$construct.$
Figure'10.'Example'of'bone'metastasis'generated'
by'caudal'artery'injection'of'MCF7'cells'into'male'
NSG'mice.'A)'Example$of$the$bioluminescence$signal$
at$ 5$ weeks$ post@injection.$ B)$ A$ plot$ showing$
bioluminescence$signal$at$each$limb$over$time.$$
Figure' 9.' MBOAT2' is' an' AR' downstream' target' that'
upregulated'in'bone'metastasis.'A)'Hierarchical$clustering$of$
RNA@seq$results$from$BRx68$parental$(blue)$and$Brx68@BoM$
(red)$samples.$B)$Top$five$significantly$upregulated$genes$in$
BRx68$BoM$samples,$compared$to$the$parental$samples.$The$
red$arrow$highlights$the$second$gene$on$the$list,$MBOAT2.$C)$
Kaplan@Meier$curves$showing$metastasis@free$status$in$bone$
(top)$and$lung$(bottom)$with$high$or$low$MBOAT2$levels$in$204$
breast$cancer$patients.$D)$The$genome$browser$showing$the$
AR$ChIP@seq$peaks$located$in$the$promoter$of$MBOAT2$gene$
in$different$mutant$ER$cell$types$but$absent$in$wild$type$ER$
cells.$
Principal)Investigator:))Yu,$Min$ 4.11$
Research$Scholar$Grant$Application$$
July$2019$

contribute$to$the$progression$of$castration$resistant$prostate$cancer,$which$often$leads$to$bone$metastases
51
.$A$
key$enzyme$in$the$lipid$synthesis$pathway$included$in$this$report$as$an$AR$downstream$gene$is$MBOAT2.$
Indeed,$MBOAT2$is$the$second$most$significantly$upregulated$gene$when$we$compared$the$RNA@seq$of$
BRx68@BoM$ with$ BRx68.$ Intriguingly,$ the$
promoter$region$of$MBOAT2$showed$AR$binding$
sites$only$in$cells$with$mutant$ER$(Figure'9).$In$
addition,$a$study$has$shown$that$in$E2$treated$
breast$cancer$cells,$AR$can$facilitate$ER’s$full$
genomic$binding$and$transcriptional$regulation
44
.$
Therefore,$based$on$our$preliminary$data$and$
the$emerging$insights,$we$hypothesize$that$AR$
could$be$a$cofactor$to$facilitate$mutant$ER$in$low$
estrogen$conditions$to$promote$bone$metastasis.$
Here,$ in$ this$ aim,$ we$ will$ determine$ the$
significance$of$AR$in$mutant$ER$context$for$bone$
metastasis$formation.)$
$
2A.$Evaluate$enhanced$bone$metastasis$from$
additional$CRISPR@edited$Y537S$mutant$lines.$$
We$recognized$that$although$BRx68$CTC$line$
has$high$bone$metastasis$frequency,$it$is$still$
limited$to$one$patient$sample.$In$this$aim,$we$will$
use$a$recently$published$caudal$artery$injection$
approach
71
$to$deliver$MCF7@Y537S$and$T47D@
Y537S$knock@in$cell$lines$to$female$NSG$mice$
and$ evaluate$ the$ bone$ metastasis$ formation$
compared$to$the$isogenic$lines$with$wild$type$
ER.$As$a$pilot$experiment,$we$have$injected$MCF7$cells$
via$caudal$artery$injection$into$male$ NSG$ mice.$ The$
result$showed$a$striking$bone$metastasis$formation$in$
both$lower$legs$of$the$male$mice$(Figure'10).$This$data$
not$only$confirmed$the$feasibility$of$this$approach$to$
evaluate$ bone$ metastasis,$ but$ also$ suggested$ a$
potential$involvement$of$ androgen$hormone$ from$ the$
male$mice.$Therefore,$we$will$use$the$same$approach$
to$deliver$the$wild$type$ER$cells$or$Y573S@ER$cells$into$
female$NSG$mice$and$evaluate$the$bone$metastasis$
formation.$ These$ cells$ will$be$ tagged$ with$ GFP$ and$
luciferase$and$the$tumor$signal$will$be$evaluated$via$live$
bioluminescence$ imaging.$ At$ the$ endpoint,$ bone$
metastasis$will$be$analyzed$for$AR$nuclear$levels$via$
immunohistochemistry$staining$for$AR$and$compare$between$wild$type$and$mutant$ER$groups.$$
$
2B.$Evaluate$the$effect$of$suppressing$AR$in$mutant$ER$cells$on$bone$metastasis$$
To$evaluate$the$role$of$AR$in$the$context$of$ER@Y537S$associated$bone$metastasis,$we$will$suppress$AR$via$
shRNAs$in$BRx68$cells$and$evaluate$the$effect$on$bone$metastasis$formation.$Via$caudal$artery,$we$will$inject$
the$BRx68$cells$with$shControl,$and$2$different$shRNAs$against$AR,$as$well$as$AR$rescue$construct$in$the$
shAR$cells.$The$bone$metastasis$growth$will$be$monitored$via$bioluminescent$imaging$once$a$week.$$
$
2C.$Evaluate$the$effect$of$suppressing$NCOA3$in$mutant$ER$cells$on$bone$metastasis.$$
Since$NCOA3$is$a$key$cofactor$for$both$ER$and$AR,$and$being$upregulated$by$ER/AR,$we$will$suppress$
NCOA3$via$shRNAs$in$BRx68$cells$and$evaluate$the$effect$on$bone$metastasis$formation.$Similarly,$as$above,$
we$will$use$caudal$artery$injection$and$monitor$the$signals$of$bone$metastasis$in$shNCOA3$cells$versus$cells$
with$the$control$shRNA,$or$the$NCOA3$rescue$construct.$
Figure'10.'Example'of'bone'metastasis'generated'
by'caudal'artery'injection'of'MCF7'cells'into'male'
NSG'mice.'A)'Example$of$the$bioluminescence$signal$
at$ 5$ weeks$ post@injection.$ B)$ A$ plot$ showing$
bioluminescence$signal$at$each$limb$over$time.$$
Figure' 9.' MBOAT2' is' an' AR' downstream' target' that'
upregulated'in'bone'metastasis.'A)'Hierarchical$clustering$of$
RNA@seq$results$from$BRx68$parental$(blue)$and$Brx68@BoM$
(red)$samples.$B)$Top$five$significantly$upregulated$genes$in$
BRx68$BoM$samples,$compared$to$the$parental$samples.$The$
red$arrow$highlights$the$second$gene$on$the$list,$MBOAT2.$C)$
Kaplan@Meier$curves$showing$metastasis@free$status$in$bone$
(top)$and$lung$(bottom)$with$high$or$low$MBOAT2$levels$in$204$
breast$cancer$patients.$D)$The$genome$browser$showing$the$
AR$ChIP@seq$peaks$located$in$the$promoter$of$MBOAT2$gene$
in$different$mutant$ER$cell$types$but$absent$in$wild$type$ER$
cells.$
C
0
0.5
1
1.5
2
2.5
3
3.5
68 Parental 68Bone26.1 68Bone84.2 68 Bone Jaw 2.1
Gene expression relative to 68 Parental
MBOAT2 expression by qPCR
D
BRx68 BRx68 BoM
68

Figure 3.2 MBOAT2 RNA and protein expression from THE HUMAN PROTEIN
ATLAS. (A) RNA expression data from TCGA dataset showed MBOAT2 is expressed in
many cancers including prostate (Blue arrow) and breast (red arrow). (B) MBOAT2
protein was widely expressed in many cancers including prostate (Blue arrow) and breast
(red arrow). Protein expression was particularly present in cytoplasm.

A
B
69

Figure 3.3 Higher MBOAT2 expression associated with lower metastasis-free
survival in breast cancer patients with bone relapse. Kaplan-Meier survival curves for
MBOAT2 in primary tumors of breast cancer patients with bone, brain, or lung relapse.
Higher MBOAT2 expression was significantly associated with lower metastasis-free
survival in patients with bone relapse, but not in those with brain or lung relapse. Log-
rank and Wilcoxon-Gehan statistical analysis were performed. GSE12276 dataset was
used for this analysis (Bos et al., 2009).
Bone Relapse
Brain Relapse
Lung Relapse
p=0.008
p=0.22
p=0.73
High Expression
Low Expression
High Expression
Low Expression
High Expression
Low Expression
70
3.2.2 Assessing the significance of androgen receptor in ER+ bone metastasis
models

Mammalian membrane-bound O-acyltransferase containing protein 2 (MBOAT2)
is a member of MBOAT superfamily of enzymes. MBOAT2 is located on chromosome 2
and is a lysophospholipid acyltransferase (LPLAT). It is involved in lands cycle, a
phospholipid remodeling pathway to generate membrane asymmetry and diversity
(Hishikawa et al., 2008; Gijon et al., 2008). Literature is very limited on MBOAT2, with
fewer papers investigating its role in cancer. MBOAT2 is significantly upregulated in
pancreatic cancer and promoted proliferation and migration properties in vitro;
overexpression of MBOAT2 enhanced cell cycle progression and suppressed CD8
+
T-
cell infiltration (Li et al., 2022). In a castration-resistant prostate cancer model (CRPC),
MBOAT2 was identified as one of several AR-regulated lipid biosynthesis pathway genes
that were reactivated during progression to CRPC (Han et al., 2018). MBOAT2 as an AR-
regulated gene in prostate cancer is very intriguing owing to high frequency of bone as
metastatic site in prostate cancer. Additionally, my work and published literature
discussed in chapter 2 highlights the role of AR in ER+ breast cancer. Owing to limited
choice of reagents available to study MBOAT2, I explored AR in the context of bone
metastasis. Immunostaining assays indicate AR is expressed in BRx68, and to a lesser
extent in two other CTC lines BRx07 and BRx50 (Figure 3.4). Moreover, immunostaining
of BRx68 and its bone variants indicate AR is expressed in parental line and its bone
variants (Figure 3.5). In these immunostaining experiments it also appears AR staining is
possibly nuclear. Prior work by Dr. Yu on characterizing CTC parental lines identified
71
ESR1 Y537S mutation in BRx68 (Yu et al., 2014). Additionally, my work discussed in
chapter 2, and published work from Williams et al., 2021 indicate AR upregulation in
MCF7 and T47D Y537S compared to WT ER cells. Taken together, these data suggest
upregulation of MBOAT2 in BRx68-BoM is associated with upregulated AR expression in
cells that harbor Y537S. Additionally, strong literature support for AR in bone metastasis
of prostate cancer suggests AR may also play a role in bone metastasis of breast cancer.
In literature, intratibial and intracardiac injection of tumor cells are commonly used to
generate bone metastasis. However, successful intratibial injections can be challenging.
Intracardiac injection delivers tumor cells systemically, and mice may succumb to tumors
at other sites before bone metastasis is established. So, we explored a recently
established technique to inject cells via caudal artery to establish bone metastasis
(Kuchimaru et al., 2018). In a pilot study using one male NSG mouse, GFP-LUC labelled
MCF7 cells were successfully delivered to the hind limbs of the mouse (Figure 3.6.A).
Bioluminescence imaging of mouse showed signal in both hind limbs, and it continued to
increase for 15 weeks (Figure 3.6.B). Immunohistochemistry for human mitochondria
indicated MCF7 cells in the bone (Figure 3.6.C). Furthermore, immunohistochemistry for
human AR indicated a distinct nuclear AR staining of MCF7 in the bone. Colonization is
the rate-limiting step in metastasis cascade (Massague et al., 2016). Identifying factors
essential for cancer cell colonization can lead to new therapeutic approaches for
metastatic disease. To interrogate colonization process in a more controlled environment,
I tested a recently published model of ex vivo bone - cancer cell co-culture. In pilot study,
I was able to successfully inject and monitor ex vivo growth of GFP-LUC MCF7 in bones
from female NSG mice (Figure 3.7). In chapter 2, I describe AR knockout of MCF7 cells
72
harboring WT or mutant ER. Testing all AR knockouts in ER WT and mutant cell lines in
caudal artery bone metastasis model would require a large cohort of animals. Ex vivo
bone – cancer cell co-culture assay is an alternative to test for bone metastasis with fewer
animals, albeit in a different context. In co-culture assay, AR knockout did not affect
colonization of WT or mutant ER cells. Moreover, no difference was observed between
WT ER and mutant ER cells harboring AR (Figure 3.8). These data suggest AR had no
effect on bone colonization in the assay conditions tested here.

Figure 3.4 Androgen receptor is expressed in BRx68. Immunostaining for AR indicate
AR was expressed in BRx68 parental cell line. AR expression appeared weaker in BRx07
and BRx50 parental cell lines.

Figure 3.5 Androgen receptor is expressed in BRx68-BoM. Immunostaining
indicated AR expression in BRx68 parental cell line, and was maintained in bone
metastatic variants.

AR
DAPI
BRx68 BRx07 BRx50
AR
DAPI
BRx68 BRx68-BoM23.1 BRx68-BoM26.1 BRx68-BoM84.2
73

Figure 3.6 Nuclear-AR positive MCF7 in bone from caudal artery mouse model.
(A)GFP-LUC labelled MCF7 were successfully delivered to hind limbs via caudal artery.
(B)MCF7 continued to proliferate for 15 weeks in vivo. (C)&(D) Immunohistochemistry
for human mitochondria (C) and AR (D) indicated presence of AR+ MCF7 in the bone.

Principal)Investigator:))Yu,$Min$ 4.11$
Research$Scholar$Grant$Application$$
July$2019$

contribute$to$the$progression$of$castration$resistant$prostate$cancer,$which$often$leads$to$bone$metastases
51
.$A$
key$enzyme$in$the$lipid$synthesis$pathway$included$in$this$report$as$an$AR$downstream$gene$is$MBOAT2.$
Indeed,$MBOAT2$is$the$second$most$significantly$upregulated$gene$when$we$compared$the$RNA@seq$of$
BRx68@BoM$ with$ BRx68.$ Intriguingly,$ the$
promoter$region$of$MBOAT2$showed$AR$binding$
sites$only$in$cells$with$mutant$ER$(Figure'9).$In$
addition,$a$study$has$shown$that$in$E2$treated$
breast$cancer$cells,$AR$can$facilitate$ER’s$full$
genomic$binding$and$transcriptional$regulation
44
.$
Therefore,$based$on$our$preliminary$data$and$
the$emerging$insights,$we$hypothesize$that$AR$
could$be$a$cofactor$to$facilitate$mutant$ER$in$low$
estrogen$conditions$to$promote$bone$metastasis.$
Here,$ in$ this$ aim,$ we$ will$ determine$ the$
significance$of$AR$in$mutant$ER$context$for$bone$
metastasis$formation.)$
$
2A.$Evaluate$enhanced$bone$metastasis$from$
additional$CRISPR@edited$Y537S$mutant$lines.$$
We$recognized$that$although$BRx68$CTC$line$
has$high$bone$metastasis$frequency,$it$is$still$
limited$to$one$patient$sample.$In$this$aim,$we$will$
use$a$recently$published$caudal$artery$injection$
approach
71
$to$deliver$MCF7@Y537S$and$T47D@
Y537S$knock@in$cell$lines$to$female$NSG$mice$
and$ evaluate$ the$ bone$ metastasis$ formation$
compared$to$the$isogenic$lines$with$wild$type$
ER.$As$a$pilot$experiment,$we$have$injected$MCF7$cells$
via$caudal$artery$injection$into$male$ NSG$ mice.$ The$
result$showed$a$striking$bone$metastasis$formation$in$
both$lower$legs$of$the$male$mice$(Figure'10).$This$data$
not$only$confirmed$the$feasibility$of$this$approach$to$
evaluate$ bone$ metastasis,$ but$ also$ suggested$ a$
potential$involvement$of$ androgen$hormone$ from$ the$
male$mice.$Therefore,$we$will$use$the$same$approach$
to$deliver$the$wild$type$ER$cells$or$Y573S@ER$cells$into$
female$NSG$mice$and$evaluate$the$bone$metastasis$
formation.$ These$ cells$ will$be$ tagged$ with$ GFP$ and$
luciferase$and$the$tumor$signal$will$be$evaluated$via$live$
bioluminescence$ imaging.$ At$ the$ endpoint,$ bone$
metastasis$will$be$analyzed$for$AR$nuclear$levels$via$
immunohistochemistry$staining$for$AR$and$compare$between$wild$type$and$mutant$ER$groups.$$
$
2B.$Evaluate$the$effect$of$suppressing$AR$in$mutant$ER$cells$on$bone$metastasis$$
To$evaluate$the$role$of$AR$in$the$context$of$ER@Y537S$associated$bone$metastasis,$we$will$suppress$AR$via$
shRNAs$in$BRx68$cells$and$evaluate$the$effect$on$bone$metastasis$formation.$Via$caudal$artery,$we$will$inject$
the$BRx68$cells$with$shControl,$and$2$different$shRNAs$against$AR,$as$well$as$AR$rescue$construct$in$the$
shAR$cells.$The$bone$metastasis$growth$will$be$monitored$via$bioluminescent$imaging$once$a$week.$$
$
2C.$Evaluate$the$effect$of$suppressing$NCOA3$in$mutant$ER$cells$on$bone$metastasis.$$
Since$NCOA3$is$a$key$cofactor$for$both$ER$and$AR,$and$being$upregulated$by$ER/AR,$we$will$suppress$
NCOA3$via$shRNAs$in$BRx68$cells$and$evaluate$the$effect$on$bone$metastasis$formation.$Similarly,$as$above,$
we$will$use$caudal$artery$injection$and$monitor$the$signals$of$bone$metastasis$in$shNCOA3$cells$versus$cells$
with$the$control$shRNA,$or$the$NCOA3$rescue$construct.$
Figure'10.'Example'of'bone'metastasis'generated'
by'caudal'artery'injection'of'MCF7'cells'into'male'
NSG'mice.'A)'Example$of$the$bioluminescence$signal$
at$ 5$ weeks$ post@injection.$ B)$ A$ plot$ showing$
bioluminescence$signal$at$each$limb$over$time.$$
Figure' 9.' MBOAT2' is' an' AR' downstream' target' that'
upregulated'in'bone'metastasis.'A)'Hierarchical$clustering$of$
RNA@seq$results$from$BRx68$parental$(blue)$and$Brx68@BoM$
(red)$samples.$B)$Top$five$significantly$upregulated$genes$in$
BRx68$BoM$samples,$compared$to$the$parental$samples.$The$
red$arrow$highlights$the$second$gene$on$the$list,$MBOAT2.$C)$
Kaplan@Meier$curves$showing$metastasis@free$status$in$bone$
(top)$and$lung$(bottom)$with$high$or$low$MBOAT2$levels$in$204$
breast$cancer$patients.$D)$The$genome$browser$showing$the$
AR$ChIP@seq$peaks$located$in$the$promoter$of$MBOAT2$gene$
in$different$mutant$ER$cell$types$but$absent$in$wild$type$ER$
cells.$
MCF7 Bone Metastasis
Human Mitochondria AR
C D
74

Figure 3.7 MCF7s in an ex vivo bone – cancer cell co-culture system. GFP-LUC
MCF7s were injected into femur and tibia of female NSG mice and monitored for 3
weeks. Bioluminescence images from day1 to day 22 are shown in (A). Quantified data
is shown in (B)

Day 1 Day 8 Day 15 Day 22
Plate 1 Plate 3 Plate 4 Plate 2
A
B
5.00E+00
5.00E+07
1.00E+08
1.50E+08
2.00E+08
2.50E+08
3.00E+08
3.50E+08
4.00E+08
4.50E+08
5.00E+08
Day 1 Day 8 Day 15 Day 22
Total Flux [p/s]
Time (Days)
Ex vivo Bone - MCF7 co-culture
75

Figure 3.8 AR knockout did not affect bone colonization in ex vivo bone – cancer
cell co-culture assay. GFP-LUC labelled AR knockout MCF7 WT and mutant ER cells
were injected into femur and tibia from female NSG mice in an ex vivo bone – cancer
cell co-culture assay. At four weeks, no changes were observed between cells with and
without AR. No changes were observed in mutant ER cells compared to WT.
Bioluminescence images from day 0 and at 4 weeks are shown in (A). Quantification of
data is shown in (B)

Day 0 Day 0 Day 28 Day 28
WT EV WT – sg2 YS EV WT – sg1
DG EV YS – sg2
YS – sg1
CONTROL DG – sg2 DG – sg1
WT EV WT – sg2 YS EV WT – sg1
DG EV YS – sg2
YS – sg1
CONTROL DG – sg2 DG – sg1
A
B
WT EV
WT -sg1
WT-sg2
YS EV
YS-sg1
YS-sg2
DG EV
DG-sg1
DG-sg2
0
2×10
2
4×10
2
6×10
2
8×10
2
1×10
3
Increase in Total Flux (photon/s)
Ex vivo Bone Colonization Assay
ns
ns
ns
ns
ns
ns
ns
ns
76
3.3 Discussion

Bone is the most common site of breast cancer metastasis. 5-year survival rate for
women with metastatic breast cancer is 29%, and 22% in males (Siegel et al., 2022). In
a cohort of 264 metastatic breast cancer patients, 73% developed bone metastasis
(Kuchuk et al., 2013). In a larger cohort of 7064 patients with mostly stage I-III breast
cancers, 22% developed bone metastasis in a median follow-up of 8.4 years (Harries et
al., 2014). I identified MBOAT2 from RNA-seq analysis of parental and bone variant
derivates of a CTC line. In my analysis, higher expression of the gene was associated
with poor survival in patients with bone relapse. While literature is limited for MBOAT2,
the gene is identified in a prostate cancer-induced osteoblastic bone metastasis-
associated stroma transcriptome (OB-BMST) (Ozdemir et al., 2014). This implicates
MBOAT2 in the reciprocal interaction between cancer cells and stroma in a prostate
cancer bone metastasis mouse model. No experimental evidence was provided to
validate the role of MBOAT2; additional studies are needed to confirm the data. A few
studies support a role for MBOAT2 in pancreatic cancer. Circular RNA of MBOAT2 (Cir-
MBOAT2) is upregulated in various prostate cancer cell lines. Silencing of MBOAT2 with
siRNA attenuated cell proliferation, migration, invasion, and glutamine metabolism in
pancreatic cells (Zhou et al., 2021). CirMBOAT2 promotes progression and metastasis in
a prostate cancer model (Shi et al., 2020). In my work, I used primers against the linear
form of MBOAT2. It will be interesting to see if primers against circular RNA of MBOAT2
identify differences between parental and bone variants. MBOAT2 is identified as an AR
regulated gene in literature. My work and published data in WT and mutant ER cells show
77
upregulation of AR in mutant ER cells. This would imply MBOAT2 may be upregulated in
mutant ER cells. I have not verified if this is true. AR staining in BRx68 and bone variants
appear nuclear, but additional experiments are needed to verify this. Due to limited
material of bone variant CTCs, cellular fraction for western may not be feasible. Confocal
microscopy can help identify subcellular localization of AR. In the pilot data from caudal
artery bone metastasis model, we used male NSG mice. Intriguingly, we see a distinct
AR localization in nucleus. This data suggests androgens in male could play a role.
Circulating estrogen levels are low in male mice and we did not provide E2
supplementation. MCF7s were able to successfully establish signal and grow without E2.
Additional experiments are needed to assess if androgen in male mice had any influence
on MCF7 tumor cell growth in bone. A validated reagent for detecting MBOAT2 by IHC
can indicate its role in bone metastasis. In ex vivo bone – cancer cell co-culture model,
AR KO didn’t affect tumor cell growth in bone. No differences were observed in WT vs
ER mutant cells. In this experiment, ex vivo co-cultures were maintained in hormone-
depleted CSS media. Since we observe nuclear AR staining in bone metastasis model
from male mice, it would be interesting to see the effect of adding DHT in CSS media for
co-culture experiments. The preliminary work presented herein provides groundwork to
explore MBOAT2 and AR in bone metastasis of ER+ breast cancer.

78
3.4 Methods
Cell culture
MCF7 cell lines were maintained as described in methods section of chapter 2. CTC lines
were maintained as previously described (Yu et al., 2014; Klotz et al., 2020). In brief, CTC
lines were cultured in suspension using ultra low attachment plates in 4% O2, 5% CO2.
CTC media is prepared using RPMI 1640 media supplemented with 1x B27, 20ng/mL
EGF, 20ng/mL bFGF, and 1x antibiotic/antimycotic. All cell lines were routinely tested for
mycoplasma with Mycoalert kit (Lonza).

RNA-seq analysis
Differential gene analysis was performed by Amal Thomas (The Smith lab, USC) as
previously described (Klotz et al., 2020).

Kaplan-Meier survival curves
To compute survival analysis, Partek Genomics Suite 6.6 was utilized to analyze gene
expression data GSE12276 (Bos et al., 2009). Kaplan-Meier survival curves were
calculated by splitting patients into high (top 50%) and low (bottom 50%) quartiles based
on median value of MBOAT2. Log-Rank and Wilcoxon-Gehan statistics were used to
generate p-values. P < 0.05 were considered as statistically significant.

Gene expression by qPCR
qPCR was performed as described in chapter 2, methods section. Primers for MBOAT2:
Forward 5’ – TGGCTCAAAAGGGTGTGTTA – 3’
79
Reverse 5’ – CAAATGGCAGAGAGAATGAACG-3’

Immunofluorescence assay
Immunofluorescence assay was performed as described in chapter 2, methods section.

Ex vivo bone – cancer co-culture assay
Ex vivo bone – cancer co-culture assay was performed as described previously (Romero-
Moreno et al., 2019). In brief, femur and tibia were collected from 8-week old female NSG
mice. The bone samples were cut in half at mid-diaphysis and cleaned with scissors,
followed by guaze to remove connective tissue. A 25-gauge needle was used to pierce
hole at epiphysis and placed immediately in cell culture media. 250K GFP-LUC labelled
cells were injected in 10µL volume using 27-gauge needle into open end of diaphysis.
The bone – cancer cell co-cultures were maintained at 5% CO2 in ultra-low attachment
plates for 4 weeks. D-luciferin (Syd Labs) was added to culture media and gently swirled
to enable complete mixing in the media and imaged with Perkin Elmer IVIS Lumina III.

In vivo experiments
All animal experiments were performed in accordance with protocols approved by the
USC Institutional Animal Care and Use Committee. Caudal artery bone metastasis model
was performed as previously described (Kuchimaru et al., 2018). Briefly, 5 x 10
5
GFP-
LUC labelled cells were prepared in 100µL of PBS and injected into caudal artery of
anesthetized 10-week-old male NSG mice (The Jackson Laboratory, Strain#005557)
using 29-gauge needle quickly. Mice were monitored weekley for metastasis signal by
80
injecting 100uL of D-Luciferin (Syd Labs) intraperitoneally and followed by imaging with
Perkin Elmer IVIS Lumina III.

Immunohistochemistry
Femur and tibia samples from caudal artery bone metastasis model were extracted and
fixed in 10% neutral buffered formalin overnight. Samples were washed three times with
DPBS, 5 minutes per wash on a rocker. Samples were decalcified with 20% EDTA for 2
weeks at room temperature on a rocker. Samples were washed three times with DPBS
and processed through USC’S histology core service. Immunohistochemical staning was
performed as described previously (Klotz et al., 2020). Sectioned samples were
deparaffinized, dehydrated and antigen retrieved with 10mM citrate buffer pH6 for 15
minutes. Sections were incubated for 15 minutes at room temperature with primary
antibodies (Mitochondria, Abcam ab92824; AR, Santacruz sc-7305) diluted in DAKO
antibody diluent. Sections were incubated with Dako Envision anti-mouse or anti-rabbit
reagent for 5 minutes with DAB (vector laboratories). Hematoxylin was used as
counterstain by incubating in staining solution for 45 seconds and washed thoroughly
before mounting.

81
CHAPTER 4
DISCUSSION
4.1 Context-dependent divergent roles of AR
Approximately 70% of all breast cancers express ER. Tamoxifen is the standard
of care for treating premenopausal ER+ breast cancer. Aromatase inhibitors are the
primary adjuvant therapy for postmenopausal patients. After 5 years on endocrine
therapy, the risk of distant recurrence ranges from 10 to 41% in follow up period of up to
20 years (Pan et al., 2017). CDK4/6 inhibitor plus fulvestrant is the standard of care for
hormone receptor-positive (HR+) advanced breast cancer patients who progressed or
relapsed on previous endocrine therapy (Turner et al., 2018). Resistance to treatments
eventually occurs. Mutations in LBD of ESR1 is a common form of acquired resistance to
endocrine therapy.
Major goal of the field is to identify new therapeutic targets in endocrine therapy-
resistant metastatic breast cancer. To achieve this goal, appropriate models must be
designed taking into consideration how they arise in clinic. Genetically engineered cell
line models harboring ESR1 mutations are ideal tools to study mechanisms of mutant ER
mediated resistance to therapy. However, these mutations primarily arise during long-
term aromatase inhibitor treatment in postmenopausal ER+ breast cancer patients. To
effectively recapitulate this setting in vitro, cells can be cultured under long-term estrogen
deprivation (LTED). My dissertation work presented here, together with recently published
work, indicate upregulation of AR in LTED ESR1 mutant cells. AR is also upregulated in
WT LTED cells suggesting a compensatory signal for estrogen deprivation.
82
Data from E2 and DHT treatment experiments in adherent culture indicate that AR
inhibits growth in this hormonal milieu. This agrees with data from (Hickey et al., 2021)
proposing a ‘tumor suppressor’ role for AR. This feature is maintained in adherent culture
of LTED WT and mutant ER cells. In contrast, data from soft agar assay of LTED cell
lines suggest a tumor promoting role for AR. This phenotype from our model agrees with
data from (Williams et al., 2021) in mutant cells treated with enzalutamide. E2 and DHT
treatment resembles circulating estrogen and androgens in a premenopausal setting. In
this hormonal milieu, E2 activates ER and DHT activates AR. Utilizing ChIP-seq data,
(Hickey et al., 2021) propose a model in which DHT activated AR sequesters p300 and
SRC-3 away from E2 activated ER, attenuating ER-driven cell proliferation. However,
data from (D’Amato et al., 2016) indicate that AR is required for maximum genomic
binding of E2 activated ER. Enzalutamide treatment reduced genomic binding,
proliferation, and transcriptional activity of ER. Additionally, their data indicate E2 induces
an ER-dependent genomic binding of AR. Collectively, these data suggest two different
models in two different hormonal environments. In a premenopausal setting, E2 and DHT
activated receptors compete for similar regions and lead to attenuated ER-driven
proliferation. On the contrary, in a postmenopausal setting of no or low E2, DHT activated
AR drives proliferation. The controversy surrounding AR as a tumor suppressor or
oncogene is much more nuanced. They are not mutually exclusive properties, but rather
influenced by factors like hormonal environment and anchorage independent conditions.
Data from anchorage-independent assay suggest a role for antioxidant properties of
sodium pyruvate. However, sodium pyruvate plays other roles in cellular metabolism.
Pyruvate dehydrogenase oxidizes pyruvate to generate acetyl-coA which then enters
83
citric acid cycle (TCA or Kreb’s cycle) to generate ATP via oxidative phosphorylation.
Additionally, pyruvate can undergo carboxylation to generate oxaloacetate to replenish
precursor molecule in citric acid cycle. These reactions are important in synthesis of
amino acids, lipids, and nuclei acids. Multitude of pyruvate’s role in metabolic reactions
suggest additional contributions from sodium pyruvate for cell survival and proliferation
(Yako et al., 2021; Rossi et al., 2020; Phannasil et al., 2015).
Historically, both androgens and estrogens have been tested clinically to treat for
breast cancer. With advances in development of new selective androgen receptor
modulators (SARM), it is important to clearly define role of AR in ER+ breast cancer. AR
positivity is reported as a positive prognosis factor for breast cancer survival. While ER
positivity is also a good prognosis factor, targeting ER signaling is the basis of endocrine
therapy (Caswell-Jin & Curtis, 2021). In a cohort of 1467 postmenopausal women with
stage I – III breast cancer, AR expression was associated with more favorable prognosis
among cases with ER+ tumors (Hu et al., 2011). In a meta-analysis of more than 7600
early-stage breast cancer cases, AR expression is associated with better overall survival
and disease-free survival, independent of ER co-expression (Vera-Badillo et al., 2013).
In a cohort of 192 ER+ breast cancers, AR:ER ³2.0 indicated a four-fold increased risk
for failure on tamoxifen (Cochrane et al., 2014). In another study, ER+ breast cancers
with ratio AR/ER³2.0 have upregulated expression of cell proliferation markers, compared
to tumors with ratios less than 2. However, in a larger validation cohort, ratios of
AR/ER³2.0 were mostly assigned to luminal B and HER2-enriched subtypes which are
characterized by higher Ki67+ and poorer prognosis. (Rangel et al., 2020). These data
84
suggest higher percentage of AR+ cells are well-equipped to develop endocrine
resistance and sustain proliferation.
My thesis work provides new and supporting evidence for context-dependent roles
played by AR in ER+ breast cancer. To the best of my knowledge, I provide the first AR
knock out models in the mutant ER background. The data generated herein shows both
growth promoting and inhibiting activities of AR in the same model, however in different
context. Additionally, the rescue cell lines provide precise tools to interrogate AR in
multiple contexts. AR’s role in rescuing anchorage independent colonies in the absence
of an antioxidant has not been reported before, especially in the context of ER mutant
cells.
4.2 Future directions
There are a few immediate directions for this project. Investigating mechanisms
that support survival during anchorage-independent conditions can provide insights into
tumorigenic properties of the cells. Changes in antioxidant enzymes, and detection of
ROS levels can provide clues about what pathways are involved. RNA-seq of AR-EV and
knockout lines in suspension culture can provide identity of pathways and genes involved
in anchorage-independent growth. Since AR is a transcription factor and is upregulated
in mutant ER cells and during LTED, AR binding profile in these different contexts can
provide mechanistic details into which regulatory regions are bound by AR in those
contexts. Additionally, it may help identify pathways important for surviving matrix-
detached conditions. LTEDs have long been used as models of endocrine resistance.
RNA-seq of LTED cell lines may provide information on how AR is upregulated in these
lines. In the in vivo data presented here, whole mouse tumor burden in AR knockdown
85
group was diminished. AR-rescue in this context would confirm that metastasis in this
model is AR driven. These experiments can help provide some mechanistic aspects of
AR in ER+ breast cancer.

4.3 Conclusions
Resistance to endocrine therapy in advanced breast cancer is the main challenge
facing the field. The dissertation work presented here provided evidence of the context-
dependent role played by AR in ER+ breast cancer. Knockout models and rescue lines
generated in this work can be valuable tools to further investigate AR in the context of ER
mutant driven resistance to endocrine therapy.

86
REFERENCES

1. Bahreini A, Li Z, Wang P, Levine KM, Tasdemir N, Cao L, Weir HM, Puhalla SL,
Davidson NE, Stern AM, Chu D, Park BH, Lee AV, Oesterreich S. Mutation site
and context dependent effects of ESR1 mutation in genome-edited breast cancer
cell models. Breast Cancer Res. 2017 May 23;19(1):60. doi: 10.1186/s13058-
017-0851-4. PMID: 28535794; PMCID: PMC5442865.

2. Borowicz S, Van Scoyk M, Avasarala S, Karuppusamy Rathinam MK, Tauler J,
Bikkavilli RK, Winn RA. The soft agar colony formation assay. J Vis Exp. 2014
Oct 27;(92):e51998. doi: 10.3791/51998. PMID: 25408172; PMCID:
PMC4353381.

3. Bos PD, Zhang XH, Nadal C, Shu W, Gomis RR, Nguyen DX, Minn AJ, van de
Vijver MJ, Gerald WL, Foekens JA, Massagué J. Genes that mediate breast
cancer metastasis to the brain. Nature. 2009 Jun 18;459(7249):1005-9. doi:
10.1038/nature08021. Epub 2009 May 6. PMID: 19421193; PMCID:
PMC2698953.

4. Breast Cancer Association Consortium, Dorling L, Carvalho S, Allen J, González-
Neira A, Luccarini C, Wahlström C, Pooley KA, Parsons MT, Fortuno C, Wang Q,
Bolla MK, Dennis J, Keeman R, Alonso MR, Álvarez N, Herraez B, Fernandez V,
Núñez-Torres R, Osorio A, Valcich J, Li M, Törngren T, Harrington PA, Baynes
C, Conroy DM, Decker B, Fachal L, Mavaddat N, Ahearn T, Aittomäki K,
Antonenkova NN, Arnold N, Arveux P, Ausems MGEM, Auvinen P, Becher H,
Beckmann MW, Behrens S, Bermisheva M, Białkowska K, Blomqvist C,
Bogdanova NV, Bogdanova-Markov N, Bojesen SE, Bonanni B, Børresen-Dale
AL, Brauch H, Bremer M, Briceno I, Brüning T, Burwinkel B, Cameron DA, Camp
NJ, Campbell A, Carracedo A, Castelao JE, Cessna MH, Chanock SJ,
Christiansen H, Collée JM, Cordina-Duverger E, Cornelissen S, Czene K, Dörk
T, Ekici AB, Engel C, Eriksson M, Fasching PA, Figueroa J, Flyger H, Försti A,
Gabrielson M, Gago-Dominguez M, Georgoulias V, Gil F, Giles GG, Glendon G,
Garcia EBG, Alnæs GIG, Guénel P, Hadjisavvas A, Haeberle L, Hahnen E, Hall
P, Hamann U, Harkness EF, Hartikainen JM, Hartman M, He W, Heemskerk-
Gerritsen BAM, Hillemanns P, Hogervorst FBL, Hollestelle A, Ho WK, Hooning
MJ, Howell A, Humphreys K, Idris F, Jakubowska A, Jung A, Kapoor PM, Kerin
MJ, Khusnutdinova E, Kim SW, Ko YD, Kosma VM, Kristensen VN, Kyriacou K,
Lakeman IMM, Lee JW, Lee MH, Li J, Lindblom A, Lo WY, Loizidou MA,
Lophatananon A, Lubiński J, MacInnis RJ, Madsen MJ, Mannermaa A,
Manoochehri M, Manoukian S, Margolin S, Martinez ME, Maurer T, Mavroudis D,
87
McLean C, Meindl A, Mensenkamp AR, Michailidou K, Miller N, Mohd Taib NA,
Muir K, Mulligan AM, Nevanlinna H, Newman WG, Nordestgaard BG, Ng PS,
Oosterwijk JC, Park SK, Park-Simon TW, Perez JIA, Peterlongo P, Porteous DJ,
Prajzendanc K, Prokofyeva D, Radice P, Rashid MU, Rhenius V, Rookus MA,
Rüdiger T, Saloustros E, Sawyer EJ, Schmutzler RK, Schneeweiss A,
Schürmann P, Shah M, Sohn C, Southey MC, Surowy H, Suvanto M,
Thanasitthichai S, Tomlinson I, Torres D, Truong T, Tzardi M, Valova Y, van
Asperen CJ, Van Dam RM, van den Ouweland AMW, van der Kolk LE, van Veen
EM, Wendt C, Williams JA, Yang XR, Yoon SY, Zamora MP, Evans DG, de la
Hoya M, Simard J, Antoniou AC, Borg Å, Andrulis IL, Chang-Claude J, García-
Closas M, Chenevix-Trench G, Milne RL, Pharoah PDP, Schmidt MK, Spurdle
AB, Vreeswijk MPG, Benitez J, Dunning AM, Kvist A, Teo SH, Devilee P, Easton
DF. Breast Cancer Risk Genes - Association Analysis in More than 113,000
Women. N Engl J Med. 2021 Feb 4;384(5):428-439. doi:
10.1056/NEJMoa1913948. Epub 2021 Jan 20. PMID: 33471991; PMCID:
PMC7611105.

5. Caswell-Jin JL, Curtis C. Androgen receptor agonists as breast cancer
therapeutics. Nat Med. 2021 Feb;27(2):198-199. doi: 10.1038/s41591-021-
01242-8. PMID: 33558723.

6. Chia K, Milioli H, Portman N, Laven-Law G, Coulson R, Yong A, Segara D,
Parker A, Caldon CE, Deng N, Swarbrick A, Tilley WD, Hickey TE, Lim E. Non-
canonical AR activity facilitates endocrine resistance in breast cancer. Endocr
Relat Cancer. 2019 Feb 1;26(2):251-264. doi: 10.1530/ERC-18-0333. PMID:
30557851.

7. Clatot F, Perdrix A, Augusto L, Beaussire L, Delacour J, Calbrix C, Sefrioui D,
Viailly PJ, Bubenheim M, Moldovan C, Alexandru C, Tennevet I, Rigal O,
Guillemet C, Leheurteur M, Gouérant S, Petrau C, Théry JC, Picquenot JM,
Veyret C, Frébourg T, Jardin F, Sarafan-Vasseur N, Di Fiore F. Kinetics,
prognostic and predictive values of ESR1 circulating mutations in metastatic
breast cancer patients progressing on aromatase inhibitor. Oncotarget. 2016 Nov
15;7(46):74448-74459. doi: 10.18632/oncotarget.12950. PMID: 27801670;
PMCID: PMC5342678.

8. Cochrane DR, Bernales S, Jacobsen BM, Cittelly DM, Howe EN, D'Amato NC,
Spoelstra NS, Edgerton SM, Jean A, Guerrero J, Gómez F, Medicherla S, Alfaro
IE, McCullagh E, Jedlicka P, Torkko KC, Thor AD, Elias AD, Protter AA, Richer
JK. Role of the androgen receptor in breast cancer and preclinical analysis of
enzalutamide. Breast Cancer Res. 2014 Jan 22;16(1):R7. doi: 10.1186/bcr3599.
PMID: 24451109; PMCID: PMC3978822.
88

9. Collins LC, Cole KS, Marotti JD, Hu R, Schnitt SJ, Tamimi RM. Androgen
receptor expression in breast cancer in relation to molecular phenotype: results
from the Nurses' Health Study. Mod Pathol. 2011 Jul;24(7):924-31. doi:
10.1038/modpathol.2011.54. Epub 2011 May 6. PMID: 21552212; PMCID:
PMC3128675.

10. Creevey L, Bleach R, Madden SF, Toomey S, Bane FT, Varešlija D, Hill AD,
Young LS, McIlroy M. Altered Steroid Milieu in AI-Resistant Breast Cancer
Facilitates AR Mediated Gene-Expression Associated with Poor Response to
Therapy. Mol Cancer Ther. 2019 Oct;18(10):1731-1743. doi: 10.1158/1535-
7163.MCT-18-0791. Epub 2019 Jul 9. PMID: 31289138.

11. D'Amato NC, Gordon MA, Babbs B, Spoelstra NS, Carson Butterfield KT, Torkko
KC, Phan VT, Barton VN, Rogers TJ, Sartorius CA, Elias A, Gertz J, Jacobsen
BM, Richer JK. Cooperative Dynamics of AR and ER Activity in Breast Cancer.
Mol Cancer Res. 2016 Nov;14(11):1054-1067. doi: 10.1158/1541-7786.MCR-16-
0167. Epub 2016 Aug 26. PMID: 27565181; PMCID: PMC5107172.

12. Davis ID, Martin AJ, Stockler MR, Begbie S, Chi KN, Chowdhury S, Coskinas X,
Frydenberg M, Hague WE, Horvath LG, Joshua AM, Lawrence NJ, Marx G,
McCaffrey J, McDermott R, McJannett M, North SA, Parnis F, Parulekar W, Pook
DW, Reaume MN, Sandhu SK, Tan A, Tan TH, Thomson A, Tu E, Vera-Badillo
F, Williams SG, Yip S, Zhang AY, Zielinski RR, Sweeney CJ; ENZAMET Trial
Investigators and the Australian and New Zealand Urogenital and Prostate
Cancer Trials Group. Enzalutamide with Standard First-Line Therapy in
Metastatic Prostate Cancer. N Engl J Med. 2019 Jul 11;381(2):121-131. doi:
10.1056/NEJMoa1903835. Epub 2019 Jun 2. PMID: 31157964.

13. Early Breast Cancer Trialists' Collaborative Group (EBCTCG), Davies C, Godwin
J, Gray R, Clarke M, Cutter D, Darby S, McGale P, Pan HC, Taylor C, Wang YC,
Dowsett M, Ingle J, Peto R. Relevance of breast cancer hormone receptors and
other factors to the efficacy of adjuvant tamoxifen: patient-level meta-analysis of
randomised trials. Lancet. 2011 Aug 27;378(9793):771-84. doi: 10.1016/S0140-
6736(11)60993-8. Epub 2011 Jul 28. PMID: 21802721; PMCID: PMC3163848.

14. Early Breast Cancer Trialists' Collaborative Group (EBCTCG). Aromatase
inhibitors versus tamoxifen in premenopausal women with oestrogen receptor-
positive early-stage breast cancer treated with ovarian suppression: a patient-
level meta-analysis of 7030 women from four randomised trials. Lancet Oncol.
89
2022 Mar;23(3):382-392. doi: 10.1016/S1470-2045(21)00758-0. Epub 2022 Feb
3. Erratum in: Lancet Oncol. 2022 Apr;23(4):e161. PMID: 35123662; PMCID:
PMC8885431.

15. Early Breast Cancer Trialists' Collaborative Group (EBCTCG). Aromatase
inhibitors versus tamoxifen in early breast cancer: patient-level meta-analysis of
the randomised trials. Lancet. 2015 Oct 3;386(10001):1341-1352. doi:
10.1016/S0140-6736(15)61074-1. Epub 2015 Jul 23. PMID: 26211827.

16. El Sayed R, El Jamal L, El Iskandarani S, Kort J, Abdel Salam M, Assi H.
Endocrine and Targeted Therapy for Hormone-Receptor-Positive, HER2-
Negative Advanced Breast Cancer: Insights to Sequencing Treatment and
Overcoming Resistance Based on Clinical Trials. Front Oncol. 2019 Jun
21;9:510. doi: 10.3389/fonc.2019.00510. PMID: 31281796; PMCID:
PMC6597942.

17. Feng J, Li L, Zhang N, Liu J, Zhang L, Gao H, Wang G, Li Y, Zhang Y, Li X, Liu
D, Lu J, Huang B. Androgen and AR contribute to breast cancer development
and metastasis: an insight of mechanisms. Oncogene. 2017 May
18;36(20):2775-2790. doi: 10.1038/onc.2016.432. Epub 2016 Nov 28. PMID:
27893717.

18. Fribbens C, O'Leary B, Kilburn L, Hrebien S, Garcia-Murillas I, Beaney M,
Cristofanilli M, Andre F, Loi S, Loibl S, Jiang J, Bartlett CH, Koehler M, Dowsett
M, Bliss JM, Johnston SR, Turner NC. Plasma ESR1 Mutations and the
Treatment of Estrogen Receptor-Positive Advanced Breast Cancer. J Clin Oncol.
2016 Sep 1;34(25):2961-8. doi: 10.1200/JCO.2016.67.3061. Epub 2016 Jun 6.
PMID: 27269946.

19. Fuentes N, Silveyra P. Estrogen receptor signaling mechanisms. Adv Protein
Chem Struct Biol. 2019;116:135-170. doi: 10.1016/bs.apcsb.2019.01.001. Epub
2019 Feb 4. PMID: 31036290; PMCID: PMC6533072.

20. Gijón MA, Riekhof WR, Zarini S, Murphy RC, Voelker DR. Lysophospholipid
acyltransferases and arachidonate recycling in human neutrophils. J Biol Chem.
2008 Oct 31;283(44):30235-45. doi: 10.1074/jbc.M806194200. Epub 2008 Sep 3.
PMID: 18772128; PMCID: PMC2573059.

90
21. Goldkorn A, Tangen C, Plets M, Morrison GJ, Cunha A, Xu T, Pinski JK, Ingles
SA, Triche T, Harzstark AL, Kohli M, MacVicar GR, Vaena DA, Crispino AW,
McConkey DJ, Lara PN Jr, Hussain MHA, Quinn DI, Vogelzang NJ, Thompson
IM Jr, Agarwal N. Baseline Circulating Tumor Cell Count as a Prognostic Marker
of PSA Response and Disease Progression in Metastatic Castrate-Sensitive
Prostate Cancer (SWOG S1216). Clin Cancer Res. 2021 Apr 1;27(7):1967-1973.
doi: 10.1158/1078-0432.CCR-20-3587. Epub 2021 Jan 26. PMID: 33500355;
PMCID: PMC8026618.

22. Han W, Gao S, Barrett D, Ahmed M, Han D, Macoska JA, He HH, Cai C.
Reactivation of androgen receptor-regulated lipid biosynthesis drives the
progression of castration-resistant prostate cancer. Oncogene. 2018 Feb
8;37(6):710-721. doi: 10.1038/onc.2017.385. Epub 2017 Oct 23. PMID:
29059155; PMCID: PMC5805650.

23. Hanker AB, Sudhan DR, Arteaga CL. Overcoming Endocrine Resistance in
Breast Cancer. Cancer Cell. 2020 Apr 13;37(4):496-513. doi:
10.1016/j.ccell.2020.03.009. PMID: 32289273; PMCID: PMC7169993.

24. Harbeck N, Penault-Llorca F, Cortes J, Gnant M, Houssami N, Poortmans P,
Ruddy K, Tsang J, Cardoso F. Breast cancer. Nat Rev Dis Primers. 2019 Sep
23;5(1):66. doi: 10.1038/s41572-019-0111-2. PMID: 31548545.

25. Haricharan S, Punturi N, Singh P, Holloway KR, Anurag M, Schmelz J, Schmidt
C, Lei JT, Suman V, Hunt K, Olson JA Jr, Hoog J, Li S, Huang S, Edwards DP,
Kavuri SM, Bainbridge MN, Ma CX, Ellis MJ. Loss of MutL Disrupts CHK2-
Dependent Cell-Cycle Control through CDK4/6 to Promote Intrinsic Endocrine
Therapy Resistance in Primary Breast Cancer. Cancer Discov. 2017
Oct;7(10):1168-1183. doi: 10.1158/2159-8290.CD-16-1179. Epub 2017 Aug 11.
PMID: 28801307; PMCID: PMC5733075.

26. Harries M, Taylor A, Holmberg L, Agbaje O, Garmo H, Kabilan S, Purushotham
A. Incidence of bone metastases and survival after a diagnosis of bone
metastases in breast cancer patients. Cancer Epidemiol. 2014 Aug;38(4):427-34.
doi: 10.1016/j.canep.2014.05.005. Epub 2014 Jun 10. PMID: 24923857.

27. Hawk MA, Gorsuch CL, Fagan P, Lee C, Kim SE, Hamann JC, Mason JA,
Weigel KJ, Tsegaye MA, Shen L, Shuff S, Zuo J, Hu S, Jiang L, Chapman S,
Leevy WM, DeBerardinis RJ, Overholtzer M, Schafer ZT. RIPK1-mediated
induction of mitophagy compromises the viability of extracellular-matrix-detached
91
cells. Nat Cell Biol. 2018 Mar;20(3):272-284. doi: 10.1038/s41556-018-0034-2.
Epub 2018 Feb 19. PMID: 29459781.

28. Hickey TE, Selth LA, Chia KM, Laven-Law G, Milioli HH, Roden D, Jindal S, Hui
M, Finlay-Schultz J, Ebrahimie E, Birrell SN, Stelloo S, Iggo R, Alexandrou S,
Caldon CE, Abdel-Fatah TM, Ellis IO, Zwart W, Palmieri C, Sartorius CA,
Swarbrick A, Lim E, Carroll JS, Tilley WD. The androgen receptor is a tumor
suppressor in estrogen receptor-positive breast cancer. Nat Med. 2021
Feb;27(2):310-320. doi: 10.1038/s41591-020-01168-7. Epub 2021 Jan 18. PMID:
33462444.

29. Hishikawa D, Shindou H, Kobayashi S, Nakanishi H, Taguchi R, Shimizu T.
Discovery of a lysophospholipid acyltransferase family essential for membrane
asymmetry and diversity. Proc Natl Acad Sci U S A. 2008 Feb 26;105(8):2830-5.
doi: 10.1073/pnas.0712245105. Epub 2008 Feb 20. PMID: 18287005; PMCID:
PMC2268545.

30. Hu R, Dawood S, Holmes MD, Collins LC, Schnitt SJ, Cole K, Marotti JD,
Hankinson SE, Colditz GA, Tamimi RM. Androgen receptor expression and
breast cancer survival in postmenopausal women. Clin Cancer Res. 2011 Apr
1;17(7):1867-74. doi: 10.1158/1078-0432.CCR-10-2021. Epub 2011 Feb 15.
PMID: 21325075; PMCID: PMC3076683.

31. Hussain M, Fizazi K, Saad F, Rathenborg P, Shore N, Ferreira U, Ivashchenko
P, Demirhan E, Modelska K, Phung D, Krivoshik A, Sternberg CN. Enzalutamide
in Men with Nonmetastatic, Castration-Resistant Prostate Cancer. N Engl J Med.
2018 Jun 28;378(26):2465-2474. doi: 10.1056/NEJMoa1800536. PMID:
29949494; PMCID: PMC8288034.

32. Jeselsohn R, Yelensky R, Buchwalter G, Frampton G, Meric-Bernstam F,
Gonzalez-Angulo AM, Ferrer-Lozano J, Perez-Fidalgo JA, Cristofanilli M, Gómez
H, Arteaga CL, Giltnane J, Balko JM, Cronin MT, Jarosz M, Sun J, Hawryluk M,
Lipson D, Otto G, Ross JS, Dvir A, Soussan-Gutman L, Wolf I, Rubinek T,
Gilmore L, Schnitt S, Come SE, Pusztai L, Stephens P, Brown M, Miller VA.
Emergence of constitutively active estrogen receptor-α mutations in pretreated
advanced estrogen receptor-positive breast cancer. Clin Cancer Res. 2014 Apr
1;20(7):1757-1767. doi: 10.1158/1078-0432.CCR-13-2332. Epub 2014 Jan 7.
PMID: 24398047; PMCID: PMC3998833.

92
33. Jiang L, Shestov AA, Swain P, Yang C, Parker SJ, Wang QA, Terada LS, Adams
ND, McCabe MT, Pietrak B, Schmidt S, Metallo CM, Dranka BP, Schwartz B,
DeBerardinis RJ. Reductive carboxylation supports redox homeostasis during
anchorage-independent growth. Nature. 2016 Apr 14;532(7598):255-8. doi:
10.1038/nature17393. Epub 2016 Apr 6. PMID: 27049945; PMCID:
PMC4860952.

34. Klotz R, Thomas A, Teng T, Han SM, Iriondo O, Li L, Restrepo-Vassalli S, Wang
A, Izadian N, MacKay M, Moon BS, Liu KJ, Ganesan SK, Lee G, Kang DS,
Walmsley CS, Pinto C, Press MF, Lu W, Lu J, Juric D, Bardia A, Hicks J, Salhia
B, Attenello F, Smith AD, Yu M. Circulating Tumor Cells Exhibit Metastatic
Tropism and Reveal Brain Metastasis Drivers. Cancer Discov. 2020
Jan;10(1):86-103. doi: 10.1158/2159-8290.CD-19-0384. Epub 2019 Oct 10.
PMID: 31601552; PMCID: PMC6954305.

35. Krop I, Abramson V, Colleoni M, Traina T, Holmes F, Garcia-Estevez L, Hart L,
Awada A, Zamagni C, Morris PG, Schwartzberg L, Chan S, Gucalp A, Biganzoli
L, Steinberg J, Sica L, Trudeau M, Markova D, Tarazi J, Zhu Z, O'Brien T, Kelly
CM, Winer E, Yardley DA. A Randomized Placebo Controlled Phase II Trial
Evaluating Exemestane with or without Enzalutamide in Patients with Hormone
Receptor-Positive Breast Cancer. Clin Cancer Res. 2020 Dec 1;26(23):6149-
6157. doi: 10.1158/1078-0432.CCR-20-1693. Epub 2020 Sep 28. PMID:
32988969.

36. Kuchimaru T, Kataoka N, Nakagawa K, Isozaki T, Miyabara H, Minegishi M,
Kadonosono T, Kizaka-Kondoh S. A reliable murine model of bone metastasis by
injecting cancer cells through caudal arteries. Nat Commun. 2018 Jul
30;9(1):2981. doi: 10.1038/s41467-018-05366-3. PMID: 30061695; PMCID:
PMC6065368.

37. Kuchuk M, Addison CL, Clemons M, Kuchuk I, Wheatley-Price P. Incidence and
consequences of bone metastases in lung cancer patients. J Bone Oncol. 2013
Jan 17;2(1):22-9. doi: 10.1016/j.jbo.2012.12.004. PMID: 26909268; PMCID:
PMC4723355.

38. Labuschagne CF, Cheung EC, Blagih J, Domart MC, Vousden KH. Cell
Clustering Promotes a Metabolic Switch that Supports Metastatic Colonization.
Cell Metab. 2019 Oct 1;30(4):720-734.e5. doi: 10.1016/j.cmet.2019.07.014. Epub
2019 Aug 22. PMID: 31447323; PMCID: PMC6863392.

93
39. Larsson AM, Jansson S, Bendahl PO, Levin Tykjaer Jörgensen C, Loman N,
Graffman C, Lundgren L, Aaltonen K, Rydén L. Longitudinal enumeration and
cluster evaluation of circulating tumor cells improve prognostication for patients
with newly diagnosed metastatic breast cancer in a prospective observational
trial. Breast Cancer Res. 2018 Jun 8;20(1):48. doi: 10.1186/s13058-018-0976-0.
PMID: 29884204; PMCID: PMC5994056.

40. Le Dily F, Vidal E, Cuartero Y, Quilez J, Nacht AS, Vicent GP, Carbonell-
Caballero J, Sharma P, Villanueva-Cañas JL, Ferrari R, De Llobet LI, Verde G,
Wright RHG, Beato M. Hormone-control regions mediate steroid receptor-
dependent genome organization. Genome Res. 2019 Jan;29(1):29-39. doi:
10.1101/gr.243824.118. Epub 2018 Dec 14. PMID: 30552103; PMCID:
PMC6314164.

41. Le Gal K, Ibrahim MX, Wiel C, Sayin VI, Akula MK, Karlsson C, Dalin MG,
Akyürek LM, Lindahl P, Nilsson J, Bergo MO. Antioxidants can increase
melanoma metastasis in mice. Sci Transl Med. 2015 Oct 7;7(308):308re8. doi:
10.1126/scitranslmed.aad3740. PMID: 26446958.

42. Li Z, Wu Y, Yates ME, Tasdemir N, Bahreini A, Chen J, Levine KM, Priedigkeit
NM, Nasrazadani A, Ali S, Buluwela L, Arnesen S, Gertz J, Richer JK, Troness
B, El-Ashry D, Zhang Q, Gerratana L, Zhang Y, Cristofanilli M, Montanez MA,
Sundd P, Wallace CT, Watkins SC, Fumagalli C, Guerini-Rocco E, Zhu L, Tseng
GC, Wagle N, Carroll JS, Jank P, Denkert C, Karsten MM, Blohmer JU, Park BH,
Lucas PC, Atkinson JM, Lee AV, Oesterreich S. Hotspot ESR1 Mutations Are
Multimodal and Contextual Modulators of Breast Cancer Metastasis. Cancer
Res. 2022 Apr 1;82(7):1321-1339. doi: 10.1158/0008-5472.CAN-21-2576. PMID:
35078818; PMCID: PMC8983597.

43. Li Z, Zhuang H, Chen X, Zhang Y, Ma Z, Wang S, Yan Q, Zhou Z, Huang S,
Zhang C, Hou B. Identification of MBOAT2 as an Unfavorable Biomarker
Correlated with KRAS Activation and Reduced CD8
+
T-Cell Infiltration in
Pancreatic Cancer. J Oncol. 2022 May 4;2022:4269733. doi:
10.1155/2022/4269733. PMID: 35571489; PMCID: PMC9095372.

44. Liu YN, Liu Y, Lee HJ, Hsu YH, Chen JH. Activated androgen receptor
downregulates E-cadherin gene expression and promotes tumor metastasis. Mol
Cell Biol. 2008 Dec;28(23):7096-108. doi: 10.1128/MCB.00449-08. Epub 2008
Sep 15. PMID: 18794357; PMCID: PMC2593382.

94

45. Lu C, Brown LC, Antonarakis ES, Armstrong AJ, Luo J. Androgen receptor
variant-driven prostate cancer II: advances in laboratory investigations. Prostate
Cancer Prostatic Dis. 2020 Sep;23(3):381-397. doi: 10.1038/s41391-020-0217-3.
Epub 2020 Mar 5. PMID: 32139878; PMCID: PMC7725416.

46. Łukasiewicz S, Czeczelewski M, Forma A, Baj J, Sitarz R, Stanisławek A. Breast
Cancer-Epidemiology, Risk Factors, Classification, Prognostic Markers, and
Current Treatment Strategies-An Updated Review. Cancers (Basel). 2021 Aug
25;13(17):4287. doi: 10.3390/cancers13174287. PMID: 34503097; PMCID:
PMC8428369.

47. Ma S, Wu Z, Yang F, Zhang J, Johnson RL, Rosenfeld MG, Guan KL. Hippo
signalling maintains ER expression and ER
+
breast cancer growth. Nature. 2021
Mar;591(7848):E1-E10. doi: 10.1038/s41586-020-03131-5. Epub 2021 Mar 3.
PMID: 33658690; PMCID: PMC8725601.

48. Manavathi B, Dey O, Gajulapalli VN, Bhatia RS, Bugide S, Kumar R. Derailed
estrogen signaling and breast cancer: an authentic couple. Endocr Rev. 2013
Feb;34(1):1-32. doi: 10.1210/er.2011-1057. Epub 2012 Sep 4. PMID: 22947396;
PMCID: PMC3565105.

49. Massagué J, Obenauf AC. Metastatic colonization by circulating tumour cells.
Nature. 2016 Jan 21;529(7586):298-306. doi: 10.1038/nature17038. PMID:
26791720; PMCID: PMC5029466.

50. Mauri D, Pavlidis N, Polyzos NP, Ioannidis JP. Survival with aromatase inhibitors
and inactivators versus standard hormonal therapy in advanced breast cancer:
meta-analysis. J Natl Cancer Inst. 2006 Sep 20;98(18):1285-91. doi:
10.1093/jnci/djj357. PMID: 16985247.

51. Mouridsen H, Gershanovich M, Sun Y, Perez-Carrion R, Boni C, Monnier A,
Apffelstaedt J, Smith R, Sleeboom HP, Jaenicke F, Pluzanska A, Dank M,
Becquart D, Bapsy PP, Salminen E, Snyder R, Chaudri-Ross H, Lang R, Wyld P,
Bhatnagar A. Phase III study of letrozole versus tamoxifen as first-line therapy of
advanced breast cancer in postmenopausal women: analysis of survival and
update of efficacy from the International Letrozole Breast Cancer Group. J Clin
Oncol. 2003 Jun 1;21(11):2101-9. doi: 10.1200/JCO.2003.04.194. PMID:
12775735.
95

52. Niemeier LA, Dabbs DJ, Beriwal S, Striebel JM, Bhargava R. Androgen receptor
in breast cancer: expression in estrogen receptor-positive tumors and in estrogen
receptor-negative tumors with apocrine differentiation. Mod Pathol. 2010
Feb;23(2):205-12. doi: 10.1038/modpathol.2009.159. Epub 2009 Nov 6. PMID:
19898421.

53. Özdemir BC, Hensel J, Secondini C, Wetterwald A, Schwaninger R, Fleischmann
A, Raffelsberger W, Poch O, Delorenzi M, Temanni R, Mills IG, van der Pluijm G,
Thalmann GN, Cecchini MG. The molecular signature of the stroma response in
prostate cancer-induced osteoblastic bone metastasis highlights expansion of
hematopoietic and prostate epithelial stem cell niches. PLoS One. 2014 Dec
8;9(12):e114530. doi: 10.1371/journal.pone.0114530. PMID: 25485970; PMCID:
PMC4259356.

54. Pan H, Gray R, Braybrooke J, Davies C, Taylor C, McGale P, Peto R, Pritchard
KI, Bergh J, Dowsett M, Hayes DF; EBCTCG. 20-Year Risks of Breast-Cancer
Recurrence after Stopping Endocrine Therapy at 5 Years. N Engl J Med. 2017
Nov 9;377(19):1836-1846. doi: 10.1056/NEJMoa1701830. PMID: 29117498;
PMCID: PMC5734609.

55. Patel HK, Bihani T. Selective estrogen receptor modulators (SERMs) and
selective estrogen receptor degraders (SERDs) in cancer treatment. Pharmacol
Ther. 2018 Jun;186:1-24. doi: 10.1016/j.pharmthera.2017.12.012. Epub 2017
Dec 28. PMID: 29289555.

56. Phannasil P, Thuwajit C, Warnnissorn M, Wallace JC, MacDonald MJ,
Jitrapakdee S. Pyruvate Carboxylase Is Up-Regulated in Breast Cancer and
Essential to Support Growth and Invasion of MDA-MB-231 Cells. PLoS One.
2015 Jun 12;10(6):e0129848. doi: 10.1371/journal.pone.0129848. PMID:
26070193; PMCID: PMC4467472.

57. Piskounova E, Agathocleous M, Murphy MM, Hu Z, Huddlestun SE, Zhao Z,
Leitch AM, Johnson TM, DeBerardinis RJ, Morrison SJ. Oxidative stress inhibits
distant metastasis by human melanoma cells. Nature. 2015 Nov
12;527(7577):186-91. doi: 10.1038/nature15726. Epub 2015 Oct 14. PMID:
26466563; PMCID: PMC4644103.

96
58. Ramos-Ibeas P, Barandalla M, Colleoni S, Lazzari G. Pyruvate antioxidant roles
in human fibroblasts and embryonic stem cells. Mol Cell Biochem. 2017
May;429(1-2):137-150. doi: 10.1007/s11010-017-2942-z. Epub 2017 Feb 28.
PMID: 28247212.

59. Ran FA, Hsu PD, Wright J, Agarwala V, Scott DA, Zhang F. Genome engineering
using the CRISPR-Cas9 system. Nat Protoc. 2013 Nov;8(11):2281-2308. doi:
10.1038/nprot.2013.143. Epub 2013 Oct 24. PMID: 24157548; PMCID:
PMC3969860.

60. Rangel N, Rondon-Lagos M, Annaratone L, Aristizábal-Pachon AF, Cassoni P,
Sapino A, Castellano I. AR/ER Ratio Correlates with Expression of Proliferation
Markers and with Distinct Subset of Breast Tumors. Cells. 2020 Apr
24;9(4):1064. doi: 10.3390/cells9041064. PMID: 32344660; PMCID:
PMC7226480.

61. Robertson JFR, Bondarenko IM, Trishkina E, Dvorkin M, Panasci L, Manikhas A,
Shparyk Y, Cardona-Huerta S, Cheung KL, Philco-Salas MJ, Ruiz-Borrego M,
Shao Z, Noguchi S, Rowbottom J, Stuart M, Grinsted LM, Fazal M, Ellis MJ.
Fulvestrant 500 mg versus anastrozole 1 mg for hormone receptor-positive
advanced breast cancer (FALCON): an international, randomised, double-blind,
phase 3 trial. Lancet. 2016 Dec 17;388(10063):2997-3005. doi: 10.1016/S0140-
6736(16)32389-3. Epub 2016 Nov 29. PMID: 27908454.

62. Robinson DR, Wu YM, Vats P, Su F, Lonigro RJ, Cao X, Kalyana-Sundaram S,
Wang R, Ning Y, Hodges L, Gursky A, Siddiqui J, Tomlins SA, Roychowdhury S,
Pienta KJ, Kim SY, Roberts JS, Rae JM, Van Poznak CH, Hayes DF, Chugh R,
Kunju LP, Talpaz M, Schott AF, Chinnaiyan AM. Activating ESR1 mutations in
hormone-resistant metastatic breast cancer. Nat Genet. 2013 Dec;45(12):1446-
51. doi: 10.1038/ng.2823. Epub 2013 Nov 3. PMID: 24185510; PMCID:
PMC4009946.

63. Rossi A, Rigotto G, Valente G, Giorgio V, Basso E, Filadi R, Pizzo P. Defective
Mitochondrial Pyruvate Flux Affects Cell Bioenergetics in Alzheimer's Disease-
Related Models. Cell Rep. 2020 Feb 18;30(7):2332-2348.e10. doi:
10.1016/j.celrep.2020.01.060. PMID: 32075767.

64. Saito Y, Li L, Coyaud E, Luna A, Sander C, Raught B, Asara JM, Brown M,
Muthuswamy SK. LLGL2 rescues nutrient stress by promoting leucine uptake in
97
ER
+
breast cancer. Nature. 2019 May;569(7755):275-279. doi: 10.1038/s41586-
019-1126-2. Epub 2019 Apr 17. PMID: 30996345.

65. Samavat H, Kurzer MS. Estrogen metabolism and breast cancer. Cancer Lett.
2015 Jan 28;356(2 Pt A):231-43. doi: 10.1016/j.canlet.2014.04.018. Epub 2014
Apr 28. PMID: 24784887; PMCID: PMC4505810.

66. Schafer ZT, Grassian AR, Song L, Jiang Z, Gerhart-Hines Z, Irie HY, Gao S,
Puigserver P, Brugge JS. Antioxidant and oncogene rescue of metabolic defects
caused by loss of matrix attachment. Nature. 2009 Sep 3;461(7260):109-13. doi:
10.1038/nature08268. Epub 2009 Aug 19. PMID: 19693011; PMCID:
PMC2931797.

67. Schiavon G, Hrebien S, Garcia-Murillas I, Cutts RJ, Pearson A, Tarazona N,
Fenwick K, Kozarewa I, Lopez-Knowles E, Ribas R, Nerurkar A, Osin P,
Chandarlapaty S, Martin LA, Dowsett M, Smith IE, Turner NC. Analysis of ESR1
mutation in circulating tumor DNA demonstrates evolution during therapy for
metastatic breast cancer. Sci Transl Med. 2015 Nov 11;7(313):313ra182. doi:
10.1126/scitranslmed.aac7551. PMID: 26560360; PMCID: PMC4998737.

68. Shi J, Liu C, Chen C, Guo K, Tang Z, Luo Y, Chen L, Su Y, Xu K. Circular RNA
circMBOAT2 promotes prostate cancer progression via a miR-1271-5p/mTOR
axis. Aging (Albany NY). 2020 Jul 9;12(13):13255-13280. doi:
10.18632/aging.103432. Epub 2020 Jul 9. PMID: 32645691; PMCID:
PMC7377862.

69. Siegel RL, Miller KD, Fuchs HE, Jemal A. Cancer statistics, 2022. CA Cancer J
Clin. 2022 Jan;72(1):7-33. doi: 10.3322/caac.21708. Epub 2022 Jan 12. PMID:
35020204.

70. Siersbæk R, Kumar S, Carroll JS. Signaling pathways and steroid receptors
modulating estrogen receptor α function in breast cancer. Genes Dev. 2018 Sep
1;32(17-18):1141-1154. doi: 10.1101/gad.316646.118. PMID: 30181360; PMCID:
PMC6120708.

71. Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, Bray F.
Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality
Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin. 2021
98
May;71(3):209-249. doi: 10.3322/caac.21660. Epub 2021 Feb 4. PMID:
33538338.

72. Tan MH, Li J, Xu HE, Melcher K, Yong EL. Androgen receptor: structure, role in
prostate cancer and drug discovery. Acta Pharmacol Sin. 2015 Jan;36(1):3-23.
doi: 10.1038/aps.2014.18. Epub 2014 Jun 9. PMID: 24909511; PMCID:
PMC4571323.

73. Teng T, Kamal M, Iriondo O, Amzaleg Y, Luo C, Thomas A, Lee G, Hsu CJ,
Nguyen JD, Kang I, Hicks J, Smith A, Sposto R, Yu M. N-Acetyl-L-cysteine
Promotes Ex Vivo Growth and Expansion of Single Circulating Tumor Cells by
Mitigating Cellular Stress Responses. Mol Cancer Res. 2021 Mar;19(3):441-450.
doi: 10.1158/1541-7786.MCR-20-0482. Epub 2020 Dec 10. PMID: 33303691;
PMCID: PMC7925378.

74. Toy W, Shen Y, Won H, Green B, Sakr RA, Will M, Li Z, Gala K, Fanning S, King
TA, Hudis C, Chen D, Taran T, Hortobagyi G, Greene G, Berger M, Baselga J,
Chandarlapaty S. ESR1 ligand-binding domain mutations in hormone-resistant
breast cancer. Nat Genet. 2013 Dec;45(12):1439-45. doi: 10.1038/ng.2822. Epub
2013 Nov 3. PMID: 24185512; PMCID: PMC3903423.

75. Traina TA, Miller K, Yardley DA, Eakle J, Schwartzberg LS, O'Shaughnessy J,
Gradishar W, Schmid P, Winer E, Kelly C, Nanda R, Gucalp A, Awada A, Garcia-
Estevez L, Trudeau ME, Steinberg J, Uppal H, Tudor IC, Peterson A, Cortes J.
Enzalutamide for the Treatment of Androgen Receptor-Expressing Triple-
Negative Breast Cancer. J Clin Oncol. 2018 Mar 20;36(9):884-890. doi:
10.1200/JCO.2016.71.3495. Epub 2018 Jan 26. PMID: 29373071; PMCID:
PMC5858523.

76. Turner NC, Slamon DJ, Ro J, Bondarenko I, Im SA, Masuda N, Colleoni M,
DeMichele A, Loi S, Verma S, Iwata H, Harbeck N, Loibl S, André F, Puyana
Theall K, Huang X, Giorgetti C, Huang Bartlett C, Cristofanilli M. Overall Survival
with Palbociclib and Fulvestrant in Advanced Breast Cancer. N Engl J Med. 2018
Nov 15;379(20):1926-1936. doi: 10.1056/NEJMoa1810527. Epub 2018 Oct 20.
PMID: 30345905.

77. Vera-Badillo FE, Templeton AJ, de Gouveia P, Diaz-Padilla I, Bedard PL, Al-
Mubarak M, Seruga B, Tannock IF, Ocana A, Amir E. Androgen receptor
expression and outcomes in early breast cancer: a systematic review and meta-
99
analysis. J Natl Cancer Inst. 2014 Jan;106(1):djt319. doi: 10.1093/jnci/djt319.
Epub 2013 Nov 22. PMID: 24273215.

78. Wang X, Perez E, Liu R, Yan LJ, Mallet RT, Yang SH. Pyruvate protects
mitochondria from oxidative stress in human neuroblastoma SK-N-SH cells.
Brain Res. 2007 Feb 9;1132(1):1-9. doi: 10.1016/j.brainres.2006.11.032. Epub
2006 Dec 15. PMID: 17174285; PMCID: PMC1853247.

79. Williams MM, Spoelstra NS, Arnesen S, O'Neill KI, Christenson JL, Reese J,
Torkko KC, Goodspeed A, Rosas E, Hanamura T, Sams SB, Li Z, Oesterreich S,
Riggins RB, Jacobsen BM, Elias A, Gertz J, Richer JK. Steroid Hormone
Receptor and Infiltrating Immune Cell Status Reveals Therapeutic Vulnerabilities
of ESR1-Mutant Breast Cancer. Cancer Res. 2021 Feb 1;81(3):732-746. doi:
10.1158/0008-5472.CAN-20-1200. Epub 2020 Nov 12. PMID: 33184106;
PMCID: PMC7854521.

80. Xue M, Zhang K, Mu K, Xu J, Yang H, Liu Y, Wang B, Wang Z, Li Z, Kong Q, Li
X, Wang H, Zhu J, Zhuang T. Regulation of estrogen signaling and breast cancer
proliferation by an ubiquitin ligase TRIM56. Oncogenesis. 2019 Apr 18;8(5):30.
doi: 10.1038/s41389-019-0139-x. PMID: 31000690; PMCID: PMC6473003.

81. Yang J, Yu H, Zhang L, Deng H, Wang Q, Li W, Zhang A, Gao H, Yin A.
Overexpressed genes associated with hormones in terminal ductal lobular units
identified by global transcriptome analysis: An insight into the anatomic origin of
breast cancer. Oncol Rep. 2016 Mar;35(3):1689-95. doi: 10.3892/or.2015.4523.
Epub 2015 Dec 28. PMID: 26707748.

82. Yako H, Niimi N, Kato A, Takaku S, Tatsumi Y, Nishito Y, Kato K, Sango K. Role
of pyruvate in maintaining cell viability and energy production under high-glucose
conditions. Sci Rep. 2021 Sep 23;11(1):18910. doi: 10.1038/s41598-021-98082-
w. PMID: 34556698; PMCID: PMC8460646.

83. Yeh S, Hu YC, Wang PH, Xie C, Xu Q, Tsai MY, Dong Z, Wang RS, Lee TH,
Chang C. Abnormal mammary gland development and growth retardation in
female mice and MCF7 breast cancer cells lacking androgen receptor. J Exp
Med. 2003 Dec 15;198(12):1899-908. doi: 10.1084/jem.20031233. PMID:
14676301; PMCID: PMC2194158.

100
84. Yu L, Wang L, Mao C, Duraki D, Kim JE, Huang R, Helferich WG, Nelson ER,
Park BH, Shapiro DJ. Estrogen-independent Myc overexpression confers
endocrine therapy resistance on breast cancer cells expressing ERαY537S and
ERαD538G mutations. Cancer Lett. 2019 Feb 1;442:373-382. doi:
10.1016/j.canlet.2018.10.041. Epub 2018 Nov 9. PMID: 30419347; PMCID:
PMC6351074.

85. Yu M, Bardia A, Aceto N, Bersani F, Madden MW, Donaldson MC, Desai R, Zhu
H, Comaills V, Zheng Z, Wittner BS, Stojanov P, Brachtel E, Sgroi D, Kapur R,
Shioda T, Ting DT, Ramaswamy S, Getz G, Iafrate AJ, Benes C, Toner M,
Maheswaran S, Haber DA. Cancer therapy. Ex vivo culture of circulating breast
tumor cells for individualized testing of drug susceptibility. Science. 2014 Jul
11;345(6193):216-20. doi: 10.1126/science.1253533. PMID: 25013076; PMCID:
PMC4358808.

86. Yu M, Stott S, Toner M, Maheswaran S, Haber DA. Circulating tumor cells:
approaches to isolation and characterization. J Cell Biol. 2011 Feb 7;192(3):373-
82. doi: 10.1083/jcb.201010021. PMID: 21300848; PMCID: PMC3101098.

87. Zhou X, Liu K, Cui J, Xiong J, Wu H, Peng T, Guo Y. Circ-MBOAT2 knockdown
represses tumor progression and glutamine catabolism by miR-433-3p/GOT1
axis in pancreatic cancer. J Exp Clin Cancer Res. 2021 Apr 8;40(1):124. doi:
10.1186/s13046-021-01894-x. PMID: 33832516; PMCID: PMC8034179.

Abstract (if available)

Abstract Metastasis is the leading cause of breast cancer deaths. Role of AR in ER+ breast cancer is vigorously debated. Mutations in ESR1 is a common mechanism of resistance to therapy. Identifying new therapeutic targets in resistant models is the main challenge in the field. In mutant ER cells, AR was upregulated and localized in the nucleus. Utilizing CRISPR-Cas9, AR knockout models were generated to precisely investigate role of AR in mutant ER cells. AR suppresses growth in presence of E2 and DHT in adherent culture. On the contrary, knockout of AR inhibits growth in anchorage-independent conditions. AR knockout diminished total metastatic burden in mice. In a postmenopausal model of long-term estrogen deprivation, AR expression increases over time in WT and mutant cells. We hypothesized antioxidants in cell culture may be important to survive anchorage-independent growth. Intriguingly, AR knockout abolishes anchorage-independent growth in a sodium pyruvate-free environment. Moreover, reintroduction of AR partially rescued growth inhibition of anchorage-independent culture in sodium pyruvate-free environment. Collectively, our data indicate a context-dependent divergent role for AR in ER+ breast cancer cells.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Estrogen receptor-β characterization in breast cancer: development of a reliable assay for measuring expression

PDF

Modulation of Runx proteins by steroid hormone receptors

PDF

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

PDF

Molecular mechanisms of chemoresistance in breast cancer

PDF

Identification of novel androgen receptor target genes in prostate cancer

PDF

Microenvironmental and biomechanical regulation of mitochondrial membrane potential in cancer cells

PDF

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

PDF

The effects of tobacco exposure on hormone levels and breast cancer risk among young women

PDF

The role of endoplasmic reticulum chaperone protein GRP78 in breast cancer

PDF

Pathogenic variants in cancer predisposition genes and risk of non-breast multiple primary cancers in breast cancer patients

PDF

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

PDF

Ectopic expression of a truncated isoform of hair keratin 81 in breast cancer alters biophysical characteristics to promote metastatic propensity

PDF

Decreased levels of expression of transmembrane protein 56 (TMEM56) in breast cancer tissues

PDF

Predictive factors of breast cancer survival: a population-based study

PDF

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

PDF

Molecular role of EZH2 overexpression in Colorectal Cancer progression

PDF

Roles of circadian clock genes in cancer

PDF

The role of pesticide exposure in breast cancer

PDF

The role of glucose-regulated proteins in endometrial and pancreatic cancers

PDF

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

Asset Metadata

Creator Ganesan, Sathish Kumar (author)

Core Title Context-dependent role of androgen receptor (AR) in estrogen receptor-positive (ER+) breast cancer

School Keck School of Medicine

Degree Doctor of Philosophy

Degree Program Cancer Biology and Genomics

Degree Conferral Date 2022-12

Publication Date 09/15/2024

Defense Date 08/26/2022

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

Tag anchorage-independent,anchorage-independent assay,androgen receptor,AR in ER+ breast cancer,AR Knockout,breast cancer,colony formation assay,CRISPR-Cas9,ER+ breast cancer,ESR1,ESR1 mutation,estrogen receptor,immunofluorescence,MCF7,Metastasis,NSG,OAI-PMH Harvest,sodium pyruvate,soft agar assay,western blot

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Xu, Jian (committee chair), Stallcup, Michael (committee member), Yu, Min (committee member)

Creator Email sganesan@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC111996065

Unique identifier UC111996065

Legacy Identifier etd-GanesanSat-11214

Document Type Dissertation

Format application/pdf (imt)

Rights Ganesan, Sathish Kumar

Type texts

Source 20220917-usctheses-batch-981 (batch), University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright. The original signature page accompanying the original submission of the work to the USC Libraries is retained by the USC Libraries and a copy of it may be obtained by authorized requesters contacting the repository e-mail address given.

Repository Name University of Southern California Digital Library

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

Repository Email cisadmin@lib.usc.edu