University of Southern California Dissertations and Theses

RNA methylation in cancer plasticity and drug resistance

(USC Thesis Other)

RNA methylation in cancer plasticity and drug resistance

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content RNA METHYLATION IN CANCER PLASTICITY
AND DRUG RESISTANCE
by
Emmanuelle Joya Hodara
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 2023
Copyright 2023 Emmanuelle J. Hodara

ii
Epigraph
“The mind is like an umbrella- it functions best when open.” – Walter Gropius

iii
Dedication
To my grandparents and parents on whose shoulder I get to stand today and every day.
To Joshua, my love and my life. To Margot and Romy, our future.
In loving memory of Eileen Lavian, a special friend and pure soul departed too soon.

iv
Acknowledgments
I am deeply grateful to all those have who supported me throughout my academic journey and
contributed to the successful completion of my graduate studies and this dissertation. There were
long stretches of time in the past five years when none of this seemed possible, and I am truly
indebted to each and every member of my village who made it all possible.
First and foremost, I would like to thank my husband Josh for his patience, kindness, listening
ears, wisdom, steadfast moral compass, understanding and love through all the ups and downs
over the past 8 years. Un grand merci to my parents who have not only been extraordinary role
models of resilience, responsibility, and upstandingness, but have also been an enormous source
of support: emotional, intellectual, and also physical and practical. Merci Papa for all the morning
walks with Margot, broadening her horizons like you broadened mine, and giving me the time to
recharge and get extra work done. Merci Maman for sharing your wisdom, love, and humor during
the countless hours of commute spent on the phone together. Thank you to my mother-in-law
Aimee for her kindness and generosity, being our on-call support with Margot. Thank you to my
siblings, July-Caroline, Elisabeth and Samuel for their perspectives, love, timely help, unwavering
support, and their constant presence despite the physical distance that separates us at times.
Thank you to our beautiful Margot (whose presence in this world has marked a turning point in
my research project) for brightening our world, inspiring and motivating me every day with her
curiosity, enthusiasm, and zest for life. Thank you to Romy, the newest addition to our family for
reminding me to slow down and enjoy here and now.
Thank you to the rest of the Hodara and Neufeld families for their love and support. Thank you to
all my dear friends for their support, advice, and occasional distraction from work. Special shout
out to Angelina Aliav, Kim Finas, Morgan Goodman, Leah Naghi, Daniel Naftalovich, Becca

v
Sverdlov, and Sabrina Maoz. Each of you has been a special source of light, love, motivation and
support through this journey.
Thank you to Dr. Amir Goldkorn, my research advisor, mentor and friend. The last seven years
have been a rollercoaster to say the least and I am grateful to have gotten the opportunity to learn
from a true physician-scientist who so capably, elegantly and humanly juggles the trifecta of clinic,
research, and teaching. It has been a humbling journey of professional and personal growth, and
I thank you for challenging me every step of the way. I would like to express my sincere
appreciation to my lab members, both present and past, for their invaluable help, mentorship,
guidance, and friendships. Special thanks to Tong Xu, Aubree Mades, Daniel Bsteh, Lisa Swartz,
Smruthi Maganti, Maheen Iqbal, Nofar Schahaf, Alex Cunha, Gareth Morrison, and John Lu. I
have learned a lot from each of you as scientists and human beings. I also want to give a special
thanks to Ben Weekley, who has been an incredible friend, colleague and sounding board through
all these years. I am so grateful for the wonderful ideas, troubleshooting, coding help and writing
support. I would like to express my gratitude to my committee members, Dr. Ite Offringa, Dr.
Peggy Farnham and Dr. Suhn Rhie and CBG Program Director Dr. Josh Neman, for their
guidance, feedback, expertise in shaping this work. I feel so fortunate to have had the opportunity
to learn from each of you. I would also like to acknowledge Dr. Jorge Torres, my undergraduate
mentor at UCLA, for kickstarting my career in biomedical research and for his continued support
through all these years. Finally, I would like to end with a special thought for my late mentor and
role model, Professor Olga Radko at UCLA, whose warm mentorship and rigor I hope to emulate
in my own career.

vi
TABLE OF CONTENTS
Epigraph…………………………………………………………………………………………………. ii
Dedication……………………………………………………………………………………………….. iii
Acknowledgements…………………………………………………………………………………….. iv
List of Tables……………………………………………………………………………………………. ix
List of Figures………………………………………………………………………………….............. x
Abstract……………………………………………………………………………………………….. xii
Chapter 1: Introduction………………………………………………………………………………. 1
1.1 Genitourinary Malignancies……….……………..………………………………………. 1
1.1.1 Bladder Cancer………………………..………….………………..……………… 1
1.1.2 Prostate Cancer ……………………...……………………..…………………….. 2
1.2 Drug Resistance & Phenotypic Plasticity …………………………………..………….. 3
1.3 M6
A: Background ……………………………………….………………………………… 6
1.4 M6
A and Cancer ………………………………………………………………………….. 9
1.4.1 M6
A and Bladder Cancer ………….……………………………………………... 9
1.4.2 M6
A and Prostate Cancer ………….…………………..………………………… 10
1.5 Goals of Dissertation …………………………………………………..………………… 11
Part I: Urinary Bladder Cancer
Chapter 2: M6
A RNA Methylation and Phenotypic Plasticity in Bladder Cancer: Model 1. 14
2.1 Our Bladder Cancer Model of Phenotypic Plasticity ……….…………………………. 14
2.2 Low Input MeRIP-seq …………………………………..………………………………... 16
2.3 Differential MeRIP-seq Analysis of SP vs NSP BC Model …………………………… 19
2.4 The Role of M6
A Eraser FTO in SP vs. NSP ………………………………………….. 22
2.5 The Role of M6
A Eraser FTO in Cisplatin Resistance ………….…………………….. 24
Conclusion ………….…………………..…………………………………………...………… 26
Chapter 3: M6
A RNA Methylation and Cisplatin Resistance in Bladder Cancer: Model 2.. 27
Author Information …………………………………………………………………………….. 27
Introduction …………………………………………………………………………………….. 27
3.1 Cisplatin-sensitive and cisplatin-resistant cells have distinct m6
A profiles …………. 29
3.2 Filtering and validating candidate transcripts …………………………………………. 33
3.3 In cisplatin-resistant cells, hypomethylated SLC7A11 transcript has decreased
binding of its reader YTHDF3, resulting in decreased degradation, increased
in SLC7A11 protein and reduced ferroptosis………………………………………. 41
3.4 Short-term cisplatin treatment of BC cell lines reduces SLC7A11 transcript
methylation, resulting in decreased YTHDF3 binding, decreased degradation,
and increased mRNA and protein levels …………………………………………... 45
3.5 Epitranscriptomic regulation and cancer promoting role of SLC7A11 are
recapitulated in patient derived organoids (PDOs) and clinical outcomes …….. 51
Discussion ………………..…………………………………………...………………………. 55

vii
Part II: Prostate Cancer
Chapter 4: M6
A RNA Methylation in enzalutamide resistance in prostate…....................... 61
Author Information ……….……………………………………………………………………. 61
Introduction …………………………………..………………………………......................... 61
4.1 Enzalutamide-sensitive and enzalutamide-resistant cells have distinct m6
A profiles 63
4.2 Filtering and validating candidate transcripts ........................................................... 66
4.3 Evaluating the role of candidate transcripts in enzalutamide resistance................... 70
Discussion …………..…………………..…………………………………………...…………72
Chapter 5: Multiparametric liquid biopsy analysis in metastatic prostate cancer …......... 74
Author Information ……….……………………………………………………………………. 74
Introduction …………………………………..………………………………......................... 74
5.1 Patients cohort and liquid biopsies ……………………………………………………... 75
5.2 Multiparametric molecular profiles ……………………………………………………… 81
5.3 Comparative analysis of liquid biopsy versus tumor biopsy …………………………. 84
5.4 Comparison of CTC DNA versus matched plasma …………………………………… 87
5.5 Comparison of single CTCs from the same sample ………………………………….. 93
Discussion …………..…………………..…………………………………………...…………98
Conclusion
Chapter 6: Materials & Methods ……….………………………….……………………………… 99
Bladder Cancer Cell Culture ……….………………………….……………………………. 99
Bladder Cancer Organoid Culture ……….………………………….……………………... 99
Prostate Cancer Cell Culture ……….………………………….……………………………100
RNA Isolation ……….………………………….…………………………………………….. 101
RNA Fragmentation ……….………………………….……………………………………... 101
MeRIP ……….………………………….…………………………………………………….. 101
RT-qPCR and MeRIP-qPCR ……….………………………….…………………………… 102
Library Preparation and Sequencing ……….………………………….………………….. 103
RNA-seq Analysis ……….………………………….……………………………………….. 103
Differential m6
A Analysis ……….………………………….………………………………...104
siRNA Knockdown ……….………………………….………………………………………. 104
Cisplatin Resistance Assay ……….………………………….…………………………….. 105
Enzalutamide Resistance Assay ……….………………………….………………………. 106
Actinomycin-D Treatment ……….………………………….………………………………. 107
Lipid Peroxidation Detection ……….………………………….…………………………… 107
Western Blotting ……….………………………….…………………………………………. 108
Immunofluorescence ……….………………………….……………………………………. 108
Organoid Histology Staining ……….………………………….……………………………. 109
RNA Immunoprecipitation (RIP) ……….………………………….……………………….. 109
Clinical Databases ……….………………………….………………………………………. 110
Statistical Analysis for Chapter 2-4 ……….………………………….……………………. 111
Data Availability ……….………………………….………………………………………….. 111
Patient Sample Collection ……….………………………….……………………………….111
CTC enumeration by immunomagnetic enrichment ……….………………………….…. 112
Single CTC recovery by dielectric manipulation ……….………………………….………112
CTC copy number variant analysis ……….………………………….……………………. 113
Purification of cfDNA and Extraction of CTCs ……….………………………….………... 115
Sequencing analysis of CTC-DNA and cfDNA ……….………………………….………..116
RNA Extraction, RT-PCR and relative cfRNA expression ……….……………………… 116
Statistical Analysis for Chapter 5 ……….………………………….………………………. 117

viii
Chapter 7: Conclusion ...........................................................................................................123
Bibliography ..............................................................................................................................127

ix
List of Tables
Table 3.1 Summary table of functional validation for the 7 transcripts previously
validated by qPCR and MeRIP-qPCR ……………………………………….…… 40
Table 4.1 Summary table of functional validation for the 6 transcripts previously
validated by qPCR and MeRIP-qPCR ……………………………………………. 70
Table 5.1 Summary chart of patient clinical data ……………………………………………. 77
Table 5.2 Summary chart of liquid biopsy alterations ………………………………………. 80
Table 5.3 Gene panel for CNV analysis ……………………………………………………… 85
Table 5.4 Gene panel for SSNV analysis ……………………………………………………. 85
Table 6.1 RT-qPCR and MeRIP-qPCR primers for bladder cancer targets ……………… 118
Table 6.2 siRNA references & conditions for bladder cancer targets …………………….. 119
Table 6.3 Western Blot antibodies ……………………………………………………………. 120
Table 6.4 Immunofluorescence antibodies ………………………………………………..... 120
Table 6.5 RT-qPCR and MeRIP-qPCR primers for prostate cancer targets …………….. 121
Table 6.6 siRNA references & conditions for prostate cancer targets ……………………. 122

x
List of Figures
1.1 Overview of genetic and non-genetic factors of intrinsic and acquired drug
resistance ……….………………………….………………………………………………. 5
1.2 M6
A writers, erasers and readers dynamically regulate m6
A levels contributing to
diverse cellular functions ……….………………………….…....................................... 8
2.1 Overview of bladder cancer phenotypic plasticity model ……………………………… 15
2.2 Validation of low-input MeRIP-PCR and MeRIP-seq in A549 cells ………………….. 18
2.3 MeRIP-Seq Analysis of SP vs. NSP fails to identify robust differentially methylated
targets ………………………………………………………………………………………. 21
2.4 FTO Inhibitor meclofenamic acid (MA) increases transition to drug-resistant
phenotype in BC …………………………………………………………………………… 23
2.5 Decrease in FTO mRNA expression potentiates cisplatin resistance in BC ………... 25
3.1 Cisplatin-sensitive and cisplatin-resistant cells have distinct m6
A profiles ………….. 31
3.2 Differential MeRIP-seq optimization and annotation between T24 and T24R2 BC
cells …………………………………………………………………………………………. 32
3.3 Identification of cancer-relevant transcripts that are both differentially methylated
and differentially expressed in cisplatin-resistant cells …………………………………35
3.4 Differential RNA-seq between T24 and T24R2 BC cells ……………………………… 36
3.5 Validation of top 15 differentially methylated and expressed candidate transcripts .. 37
3.6 siRNA-knockdown efficiency and cisplatin viability in BC cells ………………………. 39
3.7 In cisplatin-resistant cells, hypomethylated SLC7A11 transcript has decreased
binding of its reader YTHDF3 and decreased degradation, resulting in increased
SLC7A11 protein and reduced ferroptosis ……………………………………………… 43
3.8 Immunoprecipitation efficiency for RIP-qPCR and mRNA decay negative controls
in T24 and T24R2 BC cells ………………………………………………………………. 44
3.9 Short-term cisplatin treatment of cisplatin-sensitive BC cell lines reduces SLC7A11
transcript methylation, resulting in decreased YTHDF3 binding, decreased
degradation and increased mRNA and protein levels ………………………………….47
3.10 Cisplatin treatment optimization, MeRIP-qPCR and immunoprecipitation efficiency
for RIP-qPCR for T24 and UM-UC-3 BC cells …………………………………………..49
3.11 Epitranscriptomic regulation and cancer promoting role of SLC7A11 are
recapitulated in patient derived organoids (PDOs) and clinical outcomes ………….. 52
3.12 Characterization, MeRIP-qPCR, SLC7A11 siRNA knockdown efficiency and cell
viability of PDOs …………………………………………………………………………… 54
3.13 Model Schema: m6
A regulates SLC7A11 levels, promoting rapid transition to
cisplatin resistance …………………………………………………………………………59
4.1 Enzalutamide-sensitive and enzalutamide-resistant cells have distinct m6
A profiles 65

xi
4.2 Differential RNA-seq between C4-2B and MDV-R PC cells ………………………….. 67
4.3 Identification of cancer-relevant transcripts that are both differentially methylated
and differentially expressed in enzalutamide-resistant cells ………………………….. 68
4.4 Validation of top 12 differentially methylated and expressed candidate transcripts .. 69
4.5 Validation of THBS1 m6
A methylation, mRNA expression, and effect on
enzalutamide treatment ……………………………………………………………………71
5.1 Multiparametric workflow …………………………………………………………………. 79
5.2 Multiparametric profiles of individual patients (ID numbers 3, 10) …………………… 83
5.3 Distribution of genomic alterations by tissue source within individual patients ……...86
5.4 CTC CNV Summary Data for Patients 2, 3, 4, 7, 15-T2 ………………………………. 89
5.5 CNV distribution in individual CTCs ……………………………………………………... 92

xii
Abstract
Cancer drug resistance is recognized to occur not only through selection of pre-existing
genetically resistant clones, but also through rapid induction of transcriptional programs that allow
some cells to adapt and persist. This phenotypic plasticity mechanism is exemplified by models
of emerging drug resistance in two genitourinary malignancies, bladder cancer and prostate
cancer. In bladder cancer (BC), we previously reported that bladder cancer cells can rapidly
transition to and from a chemo-resistant phenotype. In prostate cancer (PC), there is a welldocumented emergence of ARV7, an alternative splice variant of the Androgen Receptor (AR)
implicated in enzalutamide resistance. In both malignancies, one potential contributing factor to
phenotypic plasticity is N6-methyladenosine (m6
A) RNA modification. Deposited by “m6A writers”
and removed by “m6A erasers”, m6
A reversibly regulates key cellular processes, including cell
fate, differentiation and mRNA processing. Given the dynamic nature of m6
A and its function in
regulating alternative splicing, I hypothesized that m6
A RNA modifications play a role in
phenotypic plasticity and transition to drug resistance in established BC and PC models. To test
this, I used methyl-RNA-immunoprecipitation followed by sequencing (MeRIP-seq) in parallel with
RNA-seq to identify gene transcripts that were both differentially methylated and differentially
expressed between drug-sensitive and drug-resistant cancer cells. In chapter 2, I test our
hypothesis in a bladder cancer model of phenotypic plasticity established in our laboratory, in
which cells interconvert cyclically in and out of a drug-resistant tumorigenic state, but I was unable
to identify differential m6
A sites due to the significant heterogeneity of cells in the model. In chapter
3, I repeat our analysis in another well-established cell-line based BC model of chemoresistance,
and found that cisplatin-sensitive and cisplatin-resistant cells have distinct m6
A profiles with 130
transcripts that are both differentially methylated and differentially expressed. I filtered and
prioritized these genes using clinical and functional database tools, then validated several of the
top candidates via targeted qPCR and MeRIP-PCR. In cisplatin-resistant cells, SLC7A11
transcripts had decreased methylation associated with decreased m6
A reader YTHDF3 binding,

xiii
prolonged RNA stability, and increased RNA and protein levels, leading to reduced ferroptosis
and increased survival. Consistent with this, cisplatin-sensitive BC cell lines and patient-derived
organoids (PDOs) exposed to cisplatin for as little as 48 hours exhibited similar mechanisms of
SLC7A11 upregulation and chemoresistance, trends that were also reflected in public cancer
survival databases. In chapter 4, I repeat this analysis in a well-established prostate cancer model
of ARV7-mediated enzalutamide resistance and found that enzalutamide-sensitive and
enzalutamide-resistant cells have distinct m6
A profiles with 46 transcripts that are both
differentially methylated and differentially expressed. After filtering and validation, one transcript
THBS1 had increased methylation and decreased expression in resistant cells. Depletion of
THBS1 by siRNA knockdown potentiated drug resistance in enzalutamide-sensitive PC cell lines.
Finally in chapter 5, I present a pilot study pioneering a multiparametric approach of liquid biopsy
profiling of prostate cancer patient samples, integrating multiple blood-based tumor phenotypes
to yield a maximally informative disease profile including putative genomic and transcriptomic
markers of drug resistance. Collectively, these findings highlight epitranscriptomic plasticity as a
mechanism of drug resistance and a potential therapeutic target in both bladder and prostate
cancer.

1
Chapter 1: Introduction
1.1 Genitourinary Malignancies
Genitourinary malignancies refer to a group of cancers that affect both reproductive and urinary
systems, the two most common of which are prostate and urinary bladder cancers. Prostate
cancer, the most prevalent malignancy in men with an estimated 288,300 annual new cases,
claimed 34,700 lives in the United States in 2022; while, urinary bladder cancer, with an estimated
82,290 annual new cases, claimed 16,710 lives in 20221
. Whereas cancer treatments have
advanced dramatically in recent decades resulting in long-term survival of patients with localized
disease, the majority of patients with advanced metastatic disease will ultimately succumb to their
cancers. In both of these genitourinary malignancies, a major impediment to achieving lasting
cures is the emergence of resistance to treatment.
1.1.1 Bladder Cancer
Bladder cancer is the fourth most common malignancy in men and 11th most common malignancy
in women1
. The primary presentation is awareness of hematuria, or blood in the urine2
. Patients
without visible hematuria have a longer time from onset of symptoms to diagnosis, and typically
present with urgency or recurrent urinary tract infections3
. Most bladder cancers are urothelial
carcinomas. At presentation, 75% of patients have non-muscle-invasive bladder (NMIBC) and
25% have muscle-invasive (MIBC) or metastatic disease4
. For non-muscle-invasive bladder
cancer, the mainstay treatment is transurethral resection of bladder tumor (TURBT) followed by
induction and maintenance immunotherapy treatment with intravesical BCG vaccine (Bacillus
Calmette–Guérin vaccine used against tuberculosis) or intravesical chemotherapy4
. While NMIBCs
frequently recur (50-70%), they rarely progress to invasion (10-15%) and have a good prognosis5
.
For muscle-invasive bladder cancer, multimodal treatment involving radical cystectomy and
bilateral pelvic lymphadenectomy with neoadjuvant chemotherapy are the standard of care.
6 For
advanced disease, systemic cisplatin-based chemotherapy is the preferred treatment, with

2
immunotherapy emerging as a viable treatment in patients in whom first-line chemotherapy failed
to control the disease.7 MIBCs have a less favorable prognosis, with a five-year survival inferior
to 50% and common progression to metastasis due to treatment resistance8,9.
Mechanistically, bladder cancer is driven by mutational load and environmental factors (smoking,
chemical exposure, diet and prior infections)9
. Cigarette smoking, the most common risk factor,
is estimated to account for half of all bladder cancer cases10. There is also evidence of
associations between bladder cancer and diet low in fruits and vegetables, consumption of water
polluted with arsenic, ambient air pollution and occupational exposure to carcinogen8
. Genomic
studies have revealed that urothelial carcinoma develops via two distinctive pathways, giving rise
to papillary NMIBCs and non-papillary or solid MIBCs 8
. High expression of mutant HRAS, deletion
of chromosome of 9 and point mutation in FGFR3 (fibroblast growth factor receptor 3) in precursor
urothelial hyperplastic lesions contribute to NMIBCs11-13. In contrast, MIBCs are characterized by
inactivation of one or more tumor suppressor genes such as TP53, RB1 and PTEN in both mouse
models and human samples14-16. NMIBCs and MIBCs share similar mutational features; however,
the mutational load is significantly greater in MIBCs compared to NMIBCs (with an average of
302 mutations compared to 169-195)17,18. Additionally, NMIBCs are typically diploid with few copy
number alterations, while MIBCs are aneuploid with numerous alterations, rearrangements and
copy number variants19,20.
1.1.2 Prostate Cancer
Prostate cancer is the most common malignancy in men1
. Most patients are diagnosed in the
localized stage and are asymptomatic at that time. Clinical signs are elevated prostate-specific
antigen (PSA) on laboratory testing and abnormal prostate finding on digital rectal examination
(DRE). A central feature of prostate cancer is its hormone responsiveness. In the 1940s, Huggins
and Hodges reported that castration led to tumor regression in prostate cancer patients 21.
Androgen deprivation therapy (ADT) using agents that block the androgen pathways has since

3
become the standard of care for prostate cancer. Resistance to ADT can develop, resulting in
primary castration-resistant prostate cancer (CRPC) or metastatic castration-resistant prostate
cancer (mCRPC)22. In CRPC and mCRPC, second generation AR signaling inhibitors (ARSIs)
such as enzalutamide and abiraterone are well-established. Abiraterone is a CYP17A1 inhibitor
that blocks androgen biosynthesis in the adrenal gland, tumor and testis and improves survival
by several months per COU-AA-302 and COU-AA-301 trials23-27. Enzalutamide is an AR
antagonist that impairs AR nuclear translocation and binding to androgen response elements on
the DNA and has similar effect on overall survival as abiraterone per the PREVAIL and AFFIRM
trials28-31. In addition to AR-targeted therapy, chemotherapy, primarily docetaxel and cabazitaxel,
poly(ADP-ribose) polymerase (PARP) inhibitors, bone-targeting treatments and immunotherapy
are also used in patients with mCRPC; but crucial questions remain regarding optimal timing and
sequencing of treatment to minimize induction of therapy resistance 32.
Mechanistically, oncogenesis of prostate cancer is a complex interaction of germline
susceptibility, acquired somatic alterations, and environmental factors. Early genomic aberrations
include TMPRSS2-ERG fusions in 40-60% patients, loss of function mutations in SPOP in 5-15%
of patients, and gain of function mutations in FOXA1 in 3-5% of patients 33,34. PTEN deletions and
TP53 mutations are found in over 50% of cases with advanced disease. Progression to mCRPC
is associated with dysregulation of additional genes important in cell growth, genetic stability, and
DNA damage response32. Additionally, alterations in AR pathways in the form of amplification,
mutation, overexpression and increased signaling are common in mCRPC35. Several AR splice
variants have been identified and are implicated in resistance to ARSI. The best documented
example is constitutively active variant ARV7 which lacks the ligand binding domain and localizes
to the nucleus independently of androgens36. Clinical studies of advanced prostate cancer
patients demonstrated that nuclear ARV7 expression in circulating tumor cells (CTCs) is
associated with resistance to ARSI, and lower overall survival (OS)37,38.

4
1.2 Drug Resistance & Phenotypic Plasticity
Unlike intrinsic resistance which represents a failure to obtain an initial response due to preexisting factors, acquired resistance refers to mechanisms that develop in response to drug
exposure leading to relapse and progression39,40. As shown in Fig. 1.1, there are genetic and nongenetic factors that contribute to both intrinsic and acquired resistance. Traditionally, resistant
clones were thought to result from presence of rare genetically-resistant subpopulations of cells
in the original tumor or from random mutations or genomic alterations induced by drug treatment
and propagated due to Darwinian selection41. However, it is now apparent that even within a
genetically identical cell population, significant phenotypic differences can exist or arise in
response to environmental stimuli, demonstrating a duality of genetic and non-genetic factors
contributing to drug resistance42-46 (Fig. 1.1). Non-genetic factors include rapid, Lamarckian
induction of transcriptional programs that allow some cells to adapt and persist. Whereas this
phenotypic plasticity may be a direct response to an environmental trigger, it can also manifest
itself as phenotypic heterogeneity, the presence of two or more isogenic populations with different
properties, or as “bistability,” the existence of two distinct populations that reversibly transition to
one another (Fig. 1.1). Both can be considered a form of “bet-hedging,” an evolutionary strategy
to maximize fitness of a population in a dynamic environment42.
Epigenetic factors such as DNA methylation, histone modifications and nucleosome occupancy
are examples of epigenetic mechanisms that have been demonstrated to regulate phenotypic
plasticity by rewiring regulatory networks43,47-50. More recently, a novel mechanism implicated in
plasticity may be RNA modification, rather than DNA modification alone. Here, we explore the
mechanistic role of N6-methyladenosine (m6
A), a methyl modification of adenosine in RNA, in
modulating cancer phenotypic plasticity culminating in drug resistance.

5
Figure 1.1 Overview of genetic and non-genetic factors of intrinsic and acquired drug
resistance.
Drug resistance can develop due to genetic and non-genetic factors. Genetic factors include
expansion of pre-existing resistant clones (yellow) and development of new resistant genetic
alterations (green) upon exposure to drug treatment. Drug resistance from non-genetic factors
can present as bistability of phenotypes (blue ovals vs. blue circles) that confer survival
advantages in different environments, and as phenotypic plasticity in which a specific phenotype
is induced by drug exposure.
Figure 1. Genetic vs. non-genetic models of intrinsic and acquired drug resistance
Drug treatment
Darwinian selection
Expansion of
resistant clones
Resistant genetic
alterations
Lamarckian induction
of resistant phenotypes
Expansion of
phenotypic variation
Bistability of phenotypes
Survival due to existing
phenotypic variation
Restoring phenotypic
variation
GENETIC
Intrinsic
Resistance
Acquired
Resistance
NON
GENETIC
Intrinsic
Resistance
Acquired
Resistance

6
1.3 M6
A: Background
M6
A, the most abundant mammalian RNA modification, was first identified in the 1970s51,52.
However, functional interest in the modifications stemmed from a study conducted decades later
in Arabidopsis, demonstrating that depletion of the m6
A-synthesizing enzyme (METTL3 plant
homolog) resulted in significant developmental defects53. In 2012, the development of m6
A
transcriptome-wide analysis called m6
A-seq was a critical step in advancing the field54,55. This
protocol involves the use of RNA immunoprecipitation (RIP) performed on fragmented RNA
utilizing anti-m6
A antibodies followed by sequencing of the captured enriched m6
A RNA fragments
compared to non-immunoprecipitated background control or input. Using m6
A-seq, studies
revealed that m6
A modifications occur on thousands of mRNAs and hundreds of long non-coding
RNAs (lncRNAs) in mouse and human cells54. Topologically, m6
A was found to be enriched near
3’end and stop codons of mRNAs56. M6
A mapping revealed a methylation consensus sequence,
designated DRACH (D=A/G/U, R=A/G, H=A/C/U); however, this motif occurs in 1 in every ~57
nucleotides, while m6
A occurs in 1 in every ~1000 nucleotides suggesting motif consensus alone
is not enough to target methylation57,58. Rather, it is believed that m6
A is a low stoichiometry
modification, meaning that any given adenosine site only a fraction of mRNA contains m6
A57.
M6
A is considered a reversible modification. As shown in Fig. 1.2, M6
A modifications are
deposited by the METTL3-METTL14-WTAP complex and METTL16, RNA methyltransferases
dubbed “m6
A writers”59. FTO and ALKBH5, “m6
A erasers”, remove the modifications; although
recent studies suggest that FTO more efficiently demethylates another RNA modification, N6-2’-
O-dimethyladenosine or m6
Am 60-62. Additionally, there are “m6
A readers” namely from the YTHD
and hnRNP protein families that recognize m6
A-containing mRNA and serve as effectors of
diverse fundamental cellular functions including mRNA processing, splicing, nuclear export,
translation and degradation, in turn regulating essential features of cancer 63-68.

7
M6
A has also been called dynamic, suggesting a transcript could be methylated and demethylated
within one life cycle. This idea was evidenced by studies demonstrating tissue-specific
transcriptome-wide m6
A distribution69. Another definition of dynamic refers to an increase or
decrease in m6
A in response to a stimulus. In this model, the stimulus would only change the
methylation of newly made transcripts, rather than pre-existing mRNAs. This has been
demonstrated in a variety of conditions including heat shock and DNA damage 54,69. However, reanalysis of many such studies comparing m6
A sites between conditions has raised some
concerns due to confounders like using too few replicates or detecting differential expression
rather than differential methylation70. New evidence suggests that m6
A distribution is mostly “hardwired” by gene architecture and that only a small subset of master-regulator transcripts containing
m6
A might be variable71. These transcripts would be jointly regulated and have been implicated
in cell fate, differentiation, and morphogenesis72-74.
M6
A readers, the most established of which are YTHDF1-3 and YTHDC1-2, serve as effectors
which mediate the downstream sequelae of m6
A modifications. It was previously believed that a
specific reader was associated with a specific outcome. YTHDF1 was associated with increased
translation efficiency, YTHDF2 with mRNA degradation and YTHDF3 with either increased
translation efficiency or mRNA degradation. However, these readers appear to have redundant
functions in promoting mRNA degradation, and it is now clear that the major function of m6
A is to
confer mRNA stability tagging transcripts for mRNA degradation75,76. Additionally, m6
A appears
to have a major effect on splicing in a small number of transcripts63,77

8
Figure 1.2. M6
A writers, erasers, and readers dynamically regulate m6
A levels contributing
to diverse cellular functions.
METTL3-METTL14-WTAP, the primary m6
A writer complex, and METTL16, a secondary m6
A
writer, deposit m6
a modification on DRACH motif of transcripts. ALKBH5 and FTO, the m6
A
erasers, remove the modification. M6
A readers including YTHDF1-3, YTHDC1-2 and HNRNP,
serve as effectors of downstream sequelae including mRNA processing, alternative splicing,
nuclear export, translation and degradation. All of these processes affect cell fate and phenotype,
contributing to cancer initiation, cancer progression, metastasis, drug resistance and cancer
relapse.

9
1.4 M6
A and Cancer
M6
A has been called a double-edged sword in tumorigenesis, with presence of the modification
at one target and absence at another both contributing to cancer progression78. For example, in
glioblastoma, two studies have shown that m6
A in RNA is associated with tumorigenesis and cell
proliferation79,80. In acute myeloid leukemia (AML), the m6
A eraser FTO enhances oncogenemediated cell transformation and inhibits all-trans-retinoic acid (ATRA)-induced AML cell
differentiation81. More recently, FTO inhibitors have been shown to dramatically suppress
proliferation and promote apoptosis in AML cell lines82. Thus, m6
A writers, erasers, and readers
constitute putative therapeutic targets. Despite these recent advances in epitranscriptomics, little
attention has been directed towards urological malignancies at the time this project was debuted.
One analysis of The Cancer Genome Atlas (TCGA) database found an upregulation of key m6
A
modulators in both bladder and prostate cancers 83. Since, a number of studies have been
conducted in both bladder and prostate cancer, all focusing on specific m6
A regulators.
1.4.1 M6
A and Bladder Cancer
In bladder cancer, m6
A writers, erasers and readers have been reported to play a role in
tumorigenesis and cancer progression. Several studies reveal different mechanisms by which
METTL3 upregulation is oncogenic, including by regulating the expression of PD-L1 and primiR221/222 84,85. Additionally, overexpression of WTAP, a cofactor to METTL3, was found to be
associated with poor prognosis 86,87. Several studies outline the role of FTO in bladder cancer
progression, some reporting that downregulation of FTO promotes cancer progression and
proliferation88,89; while others suggest that it is the upregulation of FTO that is oncogenic90-93.
Another study reports how m6
A reader YTHDF2 is frequently upregulated in bladder cancer, and
how it promotes cancer progression by regulating RIG-I innate immune response 94. With respect
to cisplatin response, several studies implicate m6
A in bladder cancer. One study shows that
knockdown of m6
A eraser ALKBH5 results in decreased cisplatin chemosensitivity95. Another

10
study finds that m6
A reader YTHDC1 positively regulates PTEN and renders cells more sensitive
to cisplatin. A third study demonstrates that interaction between WTAP and circ0008399 promotes
methyltransferase activity and leads to cisplatin resistance87. These studies have focused on a
specific m6
A effector (writer, eraser, reader) and considered transcriptome-wide m6
A changes
induced by knock-out or overexpression of the effector of interest, significantly advancing the field.
At the same time, because effectors have a multitude of partners and targets, any specific m6
A
modifications or cancer phenotypes induced by effector manipulation are understood to occur in
the context of numerous other transcripts and pathways potentially perturbed by that same
effector manipulation. Because of the multiple pleiotropic roles of effectors, in this dissertation,
we take a different approach by directly mapping the epitranscriptomic landscape of
chemotherapy sensitive vs. resistant cells, thereby focusing our discovery on m6
A alterations
associated with these disease states rather than on specific effectors.
1.4.2 M6
A and Prostate Cancer
In prostate cancer, m6
A studies have focused primarily on METTL3 and its oncogenic role.
Notably, one study generated the first m6
A map in prostate cancer and examined how METTL3
regulates gene and protein expression, finding that METTL3 knockdown rendered the cells
resistant to androgen receptor antagonists via androgen-receptor independent mechanisms96.
Several studies outline how METTL3 promotes the progression, invasion and metastasis of
prostate cancer via different mechanisms including by increasing the expression of lncRNA
MALAT1, LEF1 and USP4 97-99. One study found that upregulation of METTL3 and YTHDF2
predicted worse overall survival rate and that knockdown of either m6
A effector inhibited
proliferation and migration of prostate cancer cells in vitro and in vivo. Mechanistically, they found
that METTL3 modified LHPP and NKX3-1 and YTHDF2 bound these transcripts to target them
for mRNA degradation, resulting in AKT-dependent tumor progression100. Another very recent
study reports the upregulation of m6
A writer cofactor METTL14 in prostate cancer. METTL14 was

11
found to downregulate THBS1 in a m6
A-YTHDF2-dependent manner, resulting in
tumorigenesis101. With respect to ARV7 modulated drug resistance, one study reports that
enzalutamide decreases the expression of ALKBH5 resulting in increased methylation of SIAH1,
promoting the expression of CPSF1 which in turn facilitated the generation of ARV7 isoform102.
Still no study has systematically compared the m6
A landscape in enzalutamide-sensitive vs.
enzalutamide-resistant cells as we set out to do.
1.5 Goals of Dissertation
In this dissertation, we examine the role of m6
A in modulating phenotypic plasticity and drug
resistance in two genitourinary malignancies, bladder cancer and prostate cancer. In bladder
cancer (Part I), we use two in vitro models of phenotypic plasticity and drug resistance of urinary
bladder cancer. The first (chapter 2) is a bladder cancer model of phenotypic plasticity established
in our laboratory, in which cells interconvert cyclically in and out of a drug-resistant tumorigenic
state103,104. The second model (chapter 3) is a well-established cell-line model of cisplatin
resistance generated via long-term exposure to cisplatin105. In prostate cancer (Part II), the model
is the well-documented emergence of ARV7, an alternative splicing variant of the Androgen
Receptor (AR) implicated in prostate cancer drug resistance (chapter 4). Bladder and prostate
cancers represent two different neoplastic mechanisms: (i) mutational load and environmentallydriven in bladder cancer and (ii) steroid-hormone driven in prostate cancer; yet both malignancies
have strong characteristics that implicate a potential role for m6
A modifications: (i) cell fate in
bladder cancer plasticity and (ii) alternative splicing in prostate cancer drug resistance. In
characterizing an epitranscriptomic role in both cancers, our aim is to provide strong evidence for
pursuing this mechanism to short-circuit drug resistance in these and other cancers.
Finally, one chapter of this dissertation (chapter 5) is dedicated to an auxiliary translational project,
geared at pioneering a “multiparametric” approach of liquid biopsy profiling of prostate cancer
patient samples, integrating multiple blood-based tumor phenotypes to yield a maximally

12
informative disease profile including putative genomic and transcriptomic markers of drug
resistance. In this chapter, we present a pilot study using blood samples from 20 men with
mCRPC, interrogating various liquid biopsy analytes including cell-free DNA (cfDNA), cell-free
RNA (cfRNA), circulating tumor cell (CTC) DNA and germline DNA (further described in chapter
5). The aim is to demonstrate the feasibility and potential utility of using this approach for minimally
invasive yet comprehensive monitoring of disease phenotype over time, helping better guide
therapy.

13
Part I:
Urinary Bladder Cancer

14
Chapter 2: M6
A RNA Methylation and phenotypic plasticity in bladder cancer
All the experiments described in this chapter were conducted by Emmanuelle Hodara and are
unpublished, serving as optimization of methods and model used in later chapters.
2.1 Our Bladder Cancer Model of Phenotypic Plasticity
Our laboratory previously established a cancer phenotypic plasticity model of drug resistance in
bladder cell lines in which a drug-resistant side population (SP) is separated from a non-side
population (NSP) lacking these properties using flow cytometry (FACS) with Hoechst dye 33342
exclusion103,104,106-108. SP cells express ATP-binding cassette (ABC) transporters which pump out
Hoechst dye, enabling their detection as a Hoechst-low population at the side of the FACS plot109.
When coupling SP analysis with green fluorescent labeling (GFP), we observed a 2-way dynamic
equilibrium between SP and NSP in both cell culture and tumor xenografts independent of
selection pressures103. Specifically, the SP cells first became depleted by differentiation into NSP
cells, and then re-emerged cyclically, not through expansion of existing cancer stem-like cells,
but rather through coordinated, spontaneous conversion of large numbers of NSP cells back to
the SP phenotype over the course of days (Fig. 2.1). SP cells, relative to NSP cells, were found
to have high tumorigenicity in mice, drug resistance to docetaxel and cisplatin and high expression
of pluripotency genes. In a subsequent publication, we demonstrated that this phenotypic
plasticity was mediated at least in part by PIK3CA/AKT signaling and CBP/ß-catenin
transcriptional activation, pathways implicated in carcinogenesis, pluripotency, and drug
resistance104.

15
Figure 2.1. Overview of bladder cancer phenotypic plasticity model.
Bladder cancer phenotypic plasticity model in which highly tumorigenic, drug resistant cancer
stem-like side population (SP) cells directly convert into non-side population (NSP) cells lacking
these properties, and vice versa. Figure was taken from Xu T. et al. Int J Cancer (2020)106.
SP cell
population
as % of
entire cell
population
10
20
Cell passages (P) in culture over time (days)
P P
0 4 6 2 4 6 2
P
SP
Cancer
Stem-Like
NSP
Non-Cancer
Stem-Like
Day 1-3
Day 3-6
Figure 5. Bladder cancer phenotypic
plasticity model, in which highly
tumorigenic drug resistant, cancer
stem-like side population (SP) cells
directly convert into non-side
population (NSP) cells lacking these
properties, and vice versa.
SP cell
population
as % of
entire cell
population
10
20
Cell passages (P) in culture over time (days)
P P
0 4 6 2 4 6 2
P
SP
Cancer
Stem-Like
NSP
Non-Cancer
Stem-Like
Day 1-3
Day 3-6
Figure 5. Bladder cancer phenotypic
plasticity model, in which highly
tumorigenic drug resistant, cancer
stem-like side population (SP) cells
directly convert into non-side
population (NSP) cells lacking these
properties, and vice versa.

16
These coordinated cyclical transitions are not driven by acquired genomic alterations; therefore,
we hypothesized that epigenetic mechanisms may be at play. To investigate this possibility, we
mapped global DNA methylation and chromatin accessibility between SP and NSP bladder cells.
Indeed, we found significant epigenetic shifts as cells switched between drug-resistant and drugsensitive phenotypes. Specifically, SP and NSP cells within a population exhibited distinct states
of DNA methylation and chromatin accessibility, associated with significant gene expression
changes106. When analyzing chromatin accessibility at regulatory elements between SP and NSP,
we found that aggressive drug-resistant SP cells exhibited enhancer accessibility changes. In
particular, we found that differentially accessible enhancers were enriched for the FOXC1
transcription factor motif and that FOXC1 was significantly overexpressed, and that cisplatin
treatment of BC cells led to increased FOXC1 expression 108. More recently, a novel mechanism
implicated in plasticity may be RNA modification, rather than DNA modification alone. We, thus,
expanded these studies by using our SP-NSP model to explore the role of reversible m6
A RNA
modifications in BC phenotypic plasticity and drug resistance.
2.2 Low Input MeRIP-seq
M6
A was discovered in the 1970s, but little was known about its functional significance until the
development of m6
A RNA-immunoprecipitation and sequencing (MeRIP-seq) methodology, a
novel antibody-based approach for unbiased localization of the modification across the entire
transcriptome by the Rechavi Laboratory (Tel Aviv University), where I spent the summer of 2018
learning some of these methodologies54,55,110. Briefly, MeRIP-seq utilizes m6
A-specific antibodies
to precipitate m6
A-rich 200nt RNA fragments, followed by high-throughput sequencing (Fig.
2.2A). Read pile-ups of m6
A-containing fragments produce a “peak” reflecting an underlying m6
A
residue55,111. Peaks are discriminated from background noise by comparing MeRIP samples to
input controls. In utilizing MeRIP-seq to map m6
A modifications for the proposed study, an
important limitation was the requirement for 300μg total RNA which is not readily generated from

17
the FACS-sorted cells in in our bladder cancer phenotypic plasticity model. We were able to
address this limitation by adopting a refined MeRIP-seq protocol with input as low as 500ng,
developed by Dr. Hansen He (Princess Margaret Cancer Center, Toronto, CA)111. In consultation
with Dr. He, we successfully established and validated low-input MeRIP-qPCR and MeRIP-seq
using A549 cells and m6
A -modified and m6
A-unmodified synthetic controls. Fig. 2.2B
demonstrates significant m6
A (250-500-fold) enrichment of highly-modified genes (SETD7) and
synthetic spike-in controls (GLUC) relative to ones with little or no m6
A modifications (GAPDH,
CLUC). Additionally, we compared my MeRIP-seq data to two replicates from published data111,
demonstrating significant peak overlap and analogous peak location repartition (Fig. 2.2C-E).

18
Figure 2.2. Validation of low-input MeRIP-PCR and MeRIP-seq in A549 cells.
(A) Schematic diagram of MeRIP-seq. (B) MeRIP-PCR relative enrichment of genes
(SETD7/GAPDH) and synthetic controls (GLUC/CLUC) with high/low levels of m6
A. (C-E)
Comparison of m6
A peaks in my data compared to published data, focusing on SETD7 peak
signal (C), overlap of m6
A peaks (D), and locations of m6
A peaks (E).
D
Figure 4. Schematic diagram of m6A RNAimmunoprecipitation sequencing (MeRIP-seq).
m m
m
m m m
m m
m m m
Input control
m
m
m
m m
m
m m
m
m
m m
m m m m
Purified RNA
~200nt RNA
fragments
Chemical fragmentation
Immunoprecipitation
with anti-m6A antibody
Elution with
low/high salt buffer
cDNA library preparation
and sequencing
A
C
B
E

19
2.3 Differential MeRIP-seq Analysis of SP vs. NSP BC Model
The dynamic and reversible nature of m6
A modification and its role in cell fate and pluripotency
make m6
A a potential epigenetic contributor to the phenotypic plasticity that drives drug resistance
in BC. While the exact function of m6
A modification in cell fate and differentiation remains unclear,
several studies have demonstrated that proper m6
A formation is essential for differentiated cells
to regain pluripotency72,73,112. Specifically, there are cell type-specific epitranscriptome profiles and
m6
A has differential distribution between pluripotent and differentiated cells, with overall greater
abundance of m6
A having a positive impact on reprogramming and pluripotency113. Furthermore,
recent studies demonstrated that the expression of ABC family transporters, responsible in part
for the drug-resistant phenotype differences between SP and NSP in our plasticity model, can be
regulated by m6
A modifications, directly on their transcripts or indirectly via upstream signaling
pathways114-116. We thus hypothesized, that transcripts that are both differentially methylated and
differentially expressed between drug- sensitive and drug-resistant isogenic cells drive the rapid
shift between these two phenotypic states in our bladder cancer phenotypic plasticity model.
We used MeRIP-seq and RNA-seq to identify differentially methylated and differentially expressed
genes in SP vs. NSP cells using biological triplicates. Since m6
A is present at one to three
modifications per transcript, in about 25% of mammalian mRNAs56, we expected to find
widespread changes in m6
A distribution in cells that shift to the SP phenotype. However, very few
transcripts were found to be differentially methylated between the two phenotypes when running
the analysis in three biological replicates. When running a primary component analysis per
condition (Fig. 2.3A), it was apparent that the replicates were not grouping per condition
suggesting the two populations were not sufficiently distinct based on Hoechst efflux, or that the
SP and NSP phenotypes do not have steady state m6
A profiles. When running the differential
MeRIP-seq analysis using all the data from all three replicates, only five peaks were considered
statistically significant: 4 m6
A peak gains and 1 peak loss in SP compared to NSP (Fig. 2.3BC).

20
Fig. 2.3D is an IGV visualization of one these m6
A peaks in EFNB2 in all six replicates. In order
to be very thorough, we repeated the analysis using different combination of replicates and
identified a few additional m6
A peaks that were statistically significant. However, none could be
successfully validated by gene-specific MeRIP-qPCR. The problem may have stemmed from the
variability in FACS sorting SP and NSP. These two groups are not pure populations and there
exists significant heterogeneity when evaluating pools of cells. In order to detect m6
A peaks more
robustly, we would need a better model of bladder cancer phenotypic plasticity and drug
resistance in which the analysis could be completed with ample concordant replicates.
In Chapter 3, we describe an alternative model of cisplatin resistance in BC and a new and
improved bioinformatics pipeline incorporating recently published best differential MeRIP-seq
practices (McIntyre A. et al. Scientific Reports 2020) to minimize confounding effect of differential
expression of methylated genes on true differential methylation. After confirming the success of
this new pipeline in our model, we reanalyzed this SP and NSP dataset, but found no detectable
differential m6
A peaks between the two phenotypes. This further reinforces our conclusion that a
better model that minimized heterogeneity and provides concordant replicates is necessary in
order to get meaningful MeRIP-seq results.

21
Figure 2.3 MeRIP-Seq Analysis of SP vs. NSP fails to identify robust differentially
methylated targets.
(A) PCA plot of SP and NSP triplicates. (B) Volcano plot of differential MeRIP-seq peaks between
SP and NSP. (C) Venn Diagram of m6
A gains and losses in SP compared to NSP. (D) IGV plot
of statistically significant m6
A peak in EFNB2 between SP and NSP cells.
PCA: Condition
Principal Component #1 [45%]
Principal Component #2 [25%] −100
−50
0
50
−100 −50 0 50
1
2
3
1
2
3
SP
NSP
Binding Site Overlaps
SP vs. NSP:DB:DESeq2
4 0 1
Gain Loss
0
1
2
3
−0.5 0.0 0.5 1.0
log Fold Change [log2(SP) − log2(NSP)]
−log10(FDR)
Legend
FDR >0.05
FDR<=0.05
Contrast: SP vs. NSP [5 FDR<=0.050]
NSP
SP
D
A
C
B

22
2.4 The Role of M6
A Eraser FTO in SP vs. NSP
In addition to systematically identifying differentially methylated transcripts that drive phenotypic
plasticity and drug resistance in bladder cancer, we sought to define the function of FTO (fat mass
and obesity- associated protein), a key m6
A eraser that affects plasticity in our SP-NSP model,
by measuring its effects on BC drug resistance and identifying target transcripts that mediate its
effects. Deregulation of m6
A-associated proteins, such as the m6
A eraser FTO, has been linked
with malignant phenotypes and potentiation to drug resistance116. In acute myeloid leukemia
(AML), FTO enhances oncogene-mediated cell transformation and inhibits all-trans-retinoic acid
(ATRA)- induced AML cell differentiation49. Additionally, FTO overexpression promotes tyrosine
kinase inhibitor resistance in leukemic cells113 and FTO inhibitors have been shown to
dramatically suppress proliferation and promote apoptosis in AML cell lines50. We hypothesized
that FTO regulates m6
A modifications and subsequent expression of transcripts, contributing to
phenotypic plasticity and drug resistance in BC.
To test this hypothesis, we first treated J82 BC cells with meclofenamic acid (MA), an FDAapproved FTO inhibitor, and evaluated the effect on global m6
A levels using an ELISA-based
assay, and on SP percentage by FACS (Fig. 2.4). As previously reported103,104,106-108 and as
shown here, SP percentage predictably increases as overall cell counts grow towards confluence
(Fig. 2.4A); therefore, SP percentage in MA-treated and control cells was normalized to overall
cell counts. Notably, MA-mediated inhibition of FTO produced an increase in global m6
A levels
and a concomitant increase in normalized SP percentage (Fig. 2.4BC), suggesting that the
inhibition of FTO promotes the transition to an aggressive drug-resistant phenotype.

23
Figure 2.4. FTO Inhibitor meclofenamic acid (MA) increases transition to drug-resistant
phenotype in BC.
(A) SP Percentage depends on cell count in untreated J82 cells. (B) SP percentage normalized
to cell count and (C) global m6
A abundance both increase after 24 hours of MA treatment in J82
cells compared to controls.
Untreated DMSO-Ctrl MA-Treated
0
10
20
30
14.3
11.5
21.0
SP Percentage (%)
per million cells
A B
C
0.10 0.20 0.40 0.80 1.00
0
5
10
15
20
25
0.4 1.6
5.0
13.3
16.8
Million of Cells Seeded
SP Percentage (%)
Untreated DMSO-Ctrl MA-Treated
0
100
200
300
146.9
100.0
200.2
Relative m6A Content (%)

24
2.5 The Role of M6
A Eraser FTO in Cisplatin Resistance
Next, we used UCSC Xena exploration to compare FTO mRNA expression in bladder tumor
samples vs. normal tissues from The Cancer Genome Atlas (TCGA) and found that FTO
expression was decreased two-fold in bladder tumor samples (n=408) compared to normal
bladder tissue (n=19) from GTEX (Fig. 2.5A). We also compared FTO mRNA expression between
cisplatin-sensitive T24 BC cells to cisplatin-resistant T24R2 BC cells, and found that although
FTO was decreased in cisplatin-resistant cells, the difference was not statistically significant (Fig.
2.5B).
Having previously demonstrated that inhibition of FTO using meclofenamic acid (MA) resulted in
a significant increase of global m6
A abundance and concomitant increase in cisplatin-resistant SP
cells, we hypothesized that FTO siRNA knockdown would increase cisplatin resistance, colony
formation, migration, and invasiveness. Indeed, we found that siRNA knockdown of FTO
significantly increases cisplatin resistance in T24 cells (Fig. 2.5CD). Another group published
similar data demonstrating that FTO knockdown effectively increases proliferation and migration
in T24 and 5637 bladder cancer cells88. While they did not directly show that siFTO knockdown
increases cisplatin resistance, they demonstrated that pharmacological inhibition of FTO using
MA2, an analog of MA, rescues cisplatin toxicity in T24 and 5637 BC cells. Taken together, my
results and those recently published in BMC Urology corroborates our hypothesis that m6
A eraser
FTO may play a key role in regulating expression of gene transcripts that promote transition to a
drug-resistant state in bladder cancer.

25
Figure 2.5. Decrease in FTO mRNA expression potentiates cisplatin resistance in BC.
FTO mRNA expression is decreased in (A) BC samples compared to normal bladder tissue from
TCGA (p=0.00001), and (B) cisplatin-resistant T24R2 cells compared to cisplatin-sensitive T24
cells (p>0.01). (C) siFTO knockdown efficiency in T24 cells. (D) Evaluation of the effect of FTO
on cisplatin resistance via siFTO knockdown (siFTO + CIS) followed by cisplatin treatment in T24
cells compared to untreated (Untreated + CIS) and scrambled siRNA controls (NC +CIS).
100.00%
76.65%
0.00%
30.00%
60.00%
90.00%
120.00%
T24 T24R2
Relative FTO mRNA Expression
Relative FTO Expression in T24 vs T24R2
p = 0.00001062 p> 0.01
FTO Expression (log2normcount+1)
mean centered
FTO Expression in TCGA Bladder Samples
0.49
0.43
0.64
0.00
0.10
0.20
0.30
0.40
0.50
0.60
0.70
Untreated + CIS NC + CIS FTO + CIS
Cell Survival (%)
siFTO KD and Cisplatin Co-Treatment in T24 Cells
0.10 0.11
0.00
0.05
0.10
0.15
Relative FTO mRNA
FTO 2 + CIS FTO 2 - CIS
Expression
siFTO KD Efficiency
C
D
A B
siFTO + CIS siFTO Untreated
Untreated + CIS NC + CIS siFTO + CIS
Relative FTO mRNA
Expression
Relative FTO mRNA Expression Cell Survival (%)
FTO Expression (log2normcount+1)
mean centered

26
Conclusion
In this chapter, we optimized and established a low-input MeRIP-seq protocol in our laboratory
by validating previously published results using A549 cell lines. We then applied this methodology
to our well-established BC cell line model of phenotypic plasticity and found no statistically
significant differences between the two subpopulations of cells, drug-resistant SP cells vs. drugsensitive NSP cells, despite optimization of the bioinformatics pipeline. We found that significant
heterogeneity existed between replicates making the model inadequate for the MeRIP-seq
protocol which requires large numbers of concordant replicates to obtain robust results. In order
to overcome this challenge, we opted to use a different model of cisplatin resistance presented in
chapter 3. Furthermore, we explored the role of m6
A eraser FTO in cisplatin resistance in BC
using genetic and pharmacologic inhibition. We found that FTO inhibitor meclofenamic acid
increases the transition to drug-resistant cells, and that siRNA knockdown of FTO potentiates
cisplatin resistance in vitro.

27
Chapter 3: M6
A RNA Methylation and cisplatin resistance in bladder cancer
The work described in this chapter has been published in Hodara, E. et al. 2023. “M6
A
epitranscriptome analysis reveals differentially methylated transcripts that drive early chemoresistance in bladder cancer.” NAR Cancer, Volume 5, Issue 4, December 2023,
zcad054, https://doi.org/10.1093/narcan/zcad054 117.
Authors
Emmanuelle Hodara1
, Aubree Mades1
, Lisa Swartz1
, Maheen Iqbal1
, Tong Xu1
, Daniel Bsteh1
,
Peggy J. Farnham2
, Suhn K. Rhie2
, & Amir Goldkorn1,2
1
Division of Medical Oncology, Department of Medicine, Keck School of Medicine of USC and
Norris Comprehensive Cancer Center, Los Angeles, CA 90033, USA
2
Department of Biochemistry and Molecular Medicine, Keck School of Medicine of USC, Los
Angeles, CA 90033, USA
Author Contributions
E.H. conceived and designed the study, conducted all experiments, interpreted the results and
wrote the manuscript. A.M. conducted the qPCR and MeRIP-qPCR validation experiments,
reviewed and edited the manuscript. L.S. performed the immunofluorescence microscopy
experiments and prepared the relevant figure panels. M.I. and T.X. generated and characterized
the patient derived organoids (PDOs). D.B. participated in bioinformatics analysis and data
visualization. P.J.F. and S.K.R. provided critical feedback for data analysis and reviewed the
manuscript. A.G. designed the study, analyzed the data and wrote the manuscript.
Introduction
Cancer treatment has advanced dramatically in recent decades, but the majority of patients with
advanced malignancies ultimately succumb to their disease. Urinary bladder cancer (BC) takes a
major societal toll in the U.S., with an estimated 82,290 annual new cases in 2022. The backbone
of therapy for locally advanced and metastatic BC is cisplatin chemotherapy, but a majority of
tumors develop resistance, resulting in 16,710 deaths in 20221
. Treatment resistance traditionally
has been attributed to rare genetically-resistant subpopulations of cells in the original tumor or to
adaptive genomic alterations induced by drug treatment propagated via Darwinian selection41.
However, we and others have demonstrated that within a genetically identical cell population,
significant epigenetic and transcriptional heterogeneity can exist or arise by phenotypic plasticity
in response to environmental factors and contribute to drug resistance42-46.

28
Another mechanism of phenotypic plasticity involves N6
-methyladenosine (m6
A) RNA
modification, which has been shown to dynamically regulate mRNA processing, differentiation,
and cell fate54,68,72,118,119. M6
A is deposited co-transcriptionally by a conserved core writer complex
of METTL3, METTL14 and WTAP120,121, whereas m6
A is removed by two demethylases (erasers),
ALKBH5 and FTO122,123, although the precise functions ascribed to these proteins continue to
evolve. M6
A modifications are recognized by binding proteins (readers), namely YTHDF1-3, which
serve as effectors of downstream sequelae. M6
A has a major effect on splicing in a small number
of transcripts77, influences epigenetic silencing, and – most importantly – affects mRNA
stability57,74-76. Although m6
A status is not universally malleable 71, it can vary by cell type,
explaining tissue-specific patterns of expression 69. Moreover, m6
A-containing mRNAs are
believed to be key regulators of specific signaling pathways71. For example, m6
A residues are
found on mRNAs encoding tumor suppressor APC, and m6
A-mediated degradation of APC lead
to cancer progression124. Indeed, deregulation of m6
A and its effector proteins (writers, erasers,
readers) has been implicated in cancer initiation, progression, drug resistance and relapse in a
variety of malignancies119.
In BC, one analysis of The Cancer Genome Atlas (TCGA) database showed that 80% of BC
samples were altered in one or more m6
A effectors, most commonly VIRMA, YTHDF1/3, METTL4,
and RBM1583. Another study demonstrated that the m6
A eraser, ALKBH5, inhibited cell
proliferation and sensitized BC cells to cisplatin95. More recently, a third study reported that
circRNA circ0009399 binds WTAP to modulate expression of target RNA through m6
A and reduce
cisplatin sensitivity in bladder cancer 87. These studies underscore the potential impact of
elucidating m6
A’s role in BC, calling for a more systematic analysis to identify epitranscriptomic
drivers of disease progression and drug resistance.

29
To address this need, in this study we set out systematically identify changes in m6
A RNA
methylation and RNA expression between cisplatin-sensitive and cisplatin-resistant bladder
cancer cells. Using unbiased transcriptome-wide m6
A profiling followed by targeted validation, we
demonstrated that m6
A modifications regulate expression of clinically relevant gene transcripts in
BC, and that a subset of these modifications may promote resistance to
chemotherapy. Specifically, m6
A modification regulated the levels of SLC7A11, a cystineglutamate antiporter that potentiates cisplatin resistance in BC. Decreased m6
A levels on
SLC7A11 mRNA led to decreased binding of the m6
A reader YTHDF3, decreased mRNA
degradation, and increased SLC7A11 expression in cisplatin-resistant cells compared to cisplatinsensitive cells. The same trend was observed with short-term cisplatin treatment, suggesting that
m6
A RNA modifications can regulate early phenotypic changes that promote transition to cisplatin
resistance in BC.
Results
3.1 Cisplatin-sensitive and cisplatin-resistant cells have distinct m6
A profiles
We undertook a broad discovery approach comparing cisplatin-sensitive T24 BC cells (IC50: 10uM
cisplatin) to established cisplatin-resistant T24R2 BC cells (IC50: 100uM cisplatin) (Fig. 3.2A). We
used a low-input methyl-RNA-immunoprecipitation and sequencing (MeRIP-seq) protocol111 and
a stringent differential analysis pipeline incorporating all recommended best practices published
to date70, including 5 replicates per condition and filtering criteria to mitigate the confounding effect
of differential expression on differential methylation (Fig. 3.1A-B). Without these critical additional
steps, differential MeRIP-seq analysis can erroneously identify differentially methylated
transcripts as a result of their underlying differential expression levels, as illustrated by the linear
relationship between log2 fold changes of RNA expression vs. log2 fold changes of RNA
methylation (Fig. 3.2B). This is further illustrated when visualizing specific differentially
methylated candidates (Fig. 3.2C-D). When the methylation amplitude is compared directly

30
across all samples, the differences in m6
A in this particular transcript between sensitive and
resistant cells appear significant (Fig. 3.2C); however, when first normalizing the samples for
underlying transcript expression levels, it becomes apparent that all samples contain methylation
peaks in this transcript and that the difference in amplitude between sensitive and resistant cells
is actually not significant for this particular peak (Fig. 3.2D).
As a first pass quality control measure, we applied principal component analysis (PCA) to all
43,259 peaks called by MACS2 based on pull down versus input in all 10 replicates (5 sensitive
and 5 resistant). The sensitive and resistant replicates clustered tightly into two distinct groups,
with PC1 and PC2 representing 82% and 3% variation, respectively, reflecting the power of the
study design to accurately discriminate the two conditions (Fig. 3.2E). Encouraged by these
preliminary results, we proceeded with differential methylation analysis using three statistical
models (DESeq2, EdgeR and QNB) to evaluate statistical significance (p<0.05). We identified
1,033 differentially methylated peaks between cisplatin sensitive and resistant cells, of which 605
remained after filtering for differential expression and 348 remained after selecting those with at
least 10 reads. 234 of these peaks were hypermethylated in resistant cells, and 114 were
hypermethylated in sensitive cells (Fig. 3.1C). The volcano plot represents the 1,033 statistically
significant peaks, highlighting the 348 peaks that meet all the filtering criteria and are unique to
either sensitive (blue) or resistant (orange) cells (Fig. 3.1D). Peak distribution differed between
sensitive and resistant cells, with an increase in 3’UTR peaks in T24R2 cells and increase in
intronic peaks in T24 cells (Fig. 3.2F). Filtering out peaks with fewer than 10 reads removed
primarily intronic peaks as those would be typically degraded quickly (Fig. 3.2G). The top 20
differentially methylated transcripts based on DESeq2 p-adj were visualized using a heatmap
(Fig. 3.1E).

31
Figure 3.1. Cisplatin-sensitive and cisplatin-resistant cells have distinct m6
A profiles.
(A) MeRIP-seq bench discovery workflow and (B) informatics filtering pipeline for differential
methylation analysis with summarized peak number for different filtering steps. Numbers in
parentheses refer to: peaks hypermethylated in T24, peaks hypermethylated in T24R2. (C) PCA
plot of MeRIP samples with 5 replicates per cell line. (D) Summary of filtering results comparing
T24 to T24R2, with 114 peaks hypermethylated in T24 and 234 peaks hypermethylated in T24R2.
(E) Volcano Plot of statistically significant differential MeRIP-results (p<0.05 by DESEQ2, edgeR,
and QNB). T24 m6
A represents the 114 peaks hypermethylated in T24 cells with log2FC <-1 and
T24R2 m6
A represents the 234 peaks hypermethylated in T24R2 with log2FC > 1. (F) Heatmap
of top 20 differentially methylated transcripts ranked by DESeq2padj.

32
Figure 3.2. Differential MeRIP-seq optimization and annotation between T24 and T24R2
BC cells.
(A) Dose-response curve for cisplatin treatment in T24 and T24R2 cells. (B) Plotting raw RNA
expression Log2FC vs RNA Methylation Log2FC demonstrates a pronounced y=x (black) linear
relationship, underscoring the critical need to filter for differential expression which would
otherwise confound differential methylation calling. Representative IGV plot of m6
A peak found
on y=x line in (B) scaling all samples (C) together or by condition (D). (E) PCA plot of MeRIP
samples with 5 replicates per cell line. Peak Annotation before (F, 605 peaks) and after (G, 348
peaks) removing peaks with fewer than 10 reads, which results in loss of primarily intronic
differential peak calls.

33
3.2 Filtering and validating candidate transcripts
Since we hypothesized that m6
A regulates the expression level of gene transcripts that promote
transition to a drug resistant state in BC, we overlapped the MeRIP-seq differential methylation
results with RNA-seq differential expression from the same cisplatin sensitive and resistant cell
lines (Fig. 3.4A-C). We then applied the filtering pipeline outlined in Fig. 3.3A in order to home in
on transcripts relevant to cancer progression, as follows: We identified 130 candidate transcripts
(162 out of 348 m6
A peaks) that were both differentially expressed and differentially methylated
(Fig. 3.3B). We used Gene Ontology (GO) and Gene Set Enrichment Analysis (GSEA) to filter
the 130 transcripts based on membership in GSEA Hallmarks of Cancer pathways (Fig. 3.3C)
and relevance to GO terms like response to chemotherapy and radiation therapy (Fig. 3.3D). We
further ranked the resulting transcripts based on log fold change and relevance to bladder cancer
in existing literature and clinical databases such as The Cancer Genome Atlas (TCGA), the
Oncology Research Information Exchange Network (ORIEN) Avatar and the Catalog of Somatic
Mutations in Cancer (COSMIC). After ranking, we selected the top 15 transcripts (Fig. 3.3E) for
in vitro validation by qPCR and by targeted immunoprecipitation and PCR (MeRIP-PCR). 9 of the
15 differentially expressed transcripts were downregulated in RNA-seq, and 7 of these 9 were
also downregulated by qPCR (Fig. 3.5A, left). 6 of the 15 differentially expressed transcripts were
upregulated in RNA-seq, and 2 of these were also upregulated by qPCR (Fig. 3.5A, right). Of
the 9 transcripts whose differential expression was validated by qPCR, 7 were also validated as
differentially methylated by MeRIP-qPCR (Fig. 3.5B). To determine if any of these 7 candidate
genes were drivers of cisplatin resistance, each was siRNA-depleted in cell lines treated with
cisplatin (Table 3.1). One candidate, SLC7A11/xCT, was identified as a driver of cisplatin
resistance in three BC cell lines, with siRNA knockdown exhibiting significant synergy with
cisplatin compared to scrambled control (Fig. 3.5C-D and Fig 3.6A-C). MeRIP-seq of SLC7A11
mRNA revealed decreased m6
A levels at the 5’UTR in cisplatin-resistant cells compared to

34
cisplatin-sensitive ones and no difference in m6
A levels at other regions, including one unchanged
peak at the 3’UTR (Fig. 3.5E).

35
Figure 3.3. Identification of cancer-relevant transcripts that are both differentially
methylated and differentially expressed in cisplatin-resistant cells.
(A) Discovery and filtering pipeline. (B) RNA expression Log2FC vs RNA Methylation (m6
A)
Log2FC of the 130 transcripts (162 peaks) that are both differentially methylated and differentially
expressed between T24 and T24R2. Peaks are colored based on filtering criteria: GO, GSEA,
both or neither. (C) Differentially methylated and expressed transcripts statistically significantly
associated with Gene Ontology (GO) Terms (p-adj <0.05). (D) Differentially methylated and
expressed transcripts statistically significantly associated with Gene Set Enrichment Analysis
(GSEA) Cancer Hallmarks (p-adj <0.05). (E) Summary table of mRNA expression, methylation
and m6
A peak location for the 15 top candidate transcripts. Arrows indicate expression or
methylation levels in T24R2 compared to T24. Each arrow indicates the directionality of changes
in individual m6
A peaks within the listed transcript in the order listed in the “Peak Location” row.

36
Figure 3.4. Differential RNA-seq between T24 and T24R2 BC cells.
(A) PCA plot of RNA-seq samples with 5 replicates per cell line. (B) Volcano Plot of statistically
significant differential RNA-seq results (p<0.00001). Downregulated represents the peaks with
log2FC < -0.5 and Upregulated with log2FC > 0.5. (C) Heatmap of top 20 differentially expressed
transcripts ranked by log2FC. (D) RNA expression Log2FC vs RNA Methylation Log2FC after
filtering for confounding differential expression and m6
A coverage plot for PML, a representative
transcript that has increased expression and methylation in T24R2 compared to T24.

37

38
Figure 3.5. Validation of top 15 differentially methylated and expressed candidate
transcripts.
(A) 9 of 15 candidate transcripts (boxed in red) were validated as differentially expressed using
qPCR, corroborating RNA-seq results. (B) 7 of the 9 candidate transcripts (boxed in red) were
validated by MeRIP-qPCR, corroborating MeRIP-seq results. Red dotted lines in (A) and (B)
represent the validation threshold. (C-D) Functional validation of SLC7A11 evaluating its effect
on cisplatin resistance via siSLC7A11 knockdown followed by cisplatin treatment in (C) T24R2
cells (CDI: 0.338469), (D) T24 cells (CDI: 0.515087), and UM-UC-3 cells (CDI: 0.661111). CDI <
0.7 indicates statistically significant synergy. Experiments were performed using three biological
replicates. Significant codes: ‘****’: p< 0.00001, ‘**’: p<0.01. (E) M6
A coverage plot for SLC7A11
in T24 and T24R2 cells with differential peak at the 5’UTR, and consistent (no difference) peak at
the 3’UTR.

39
Figure 3.6. siRNA-knockdown efficiency and cisplatin viability in BC cells.
SLC7A11 siRNA-knockdown efficiency and cell viability upon siSLC7A11 knockdown (two
independent siRNAs) with and without cisplatin treatment in (A) T24R2, (B) T24, (C) UM-UC-3
cells. NC= negative control (scrambled siRNA). (D) ALKBH5 and FTO siRNA knockdown
efficiency and SLC7A11 mRNA expression upon respective knockdown. Significant codes: ‘****’:
p< 0.00001, ‘**’: p<0.01.

40
Table 3.1. Summary table of functional validation for the 7 transcripts previously validated
by qPCR and MeRIP-qPCR.
Target Differential
Expression
Peak #
Location
Differential
Methylation
Expected
Effect of
KD on
Resistance
Actual Effect
of KD on
Resistance:
CDI
Corroborates?
FABP5 ↓ 1: utr5 ↑ Antagonism 0.5519
SYNERGISTIC OPPOSITE
SERPINE1 ↓ 1: utr3 ↑ Antagonism 0.9553 NOT
SIGNIFICANT
OSBP2 ↓ 1: exon ↑ Antagonism 0.7738 NOT
SIGNIFICANT
ANO9 ↓ 1: exon
2: exon ↑ Antagonism 0.5614
SYNERGISTIC OPPOSITE
PLA2R1 ↓ 1: exon ↓ Antagonism 0.7466 NOT
SIGNIFICANT
SLC7A11 ↑ 1: utr5 ↓ Synergy 0.4198
SYNERGISTIC YES
PML ↑ 1: exon ↑ Synergy 0.9549 NOT
SIGNIFICANT
*Arrows indicate expression or methylation levels in T24R2 compared to T24. CDI is a coefficient
of synergy where CDI < 0.7 indicates statistically significant synergy and CDI > 1 indicates
statically significant antagonism. CDI is the average of 3 biological replicates.

41
3.3 In cisplatin-resistant cells, hypomethylated SLC7A11 transcript has decreased binding
of its reader YTHDF3, resulting in decreased degradation, increased in SLC7A11 protein
and reduced ferroptosis
To identify an m6
A reader associated with SLC7A11 transcript methylation, we performed RNA
Immunoprecipitation (RIP) using antibodies specific for the most common m6
A readers, YTHDF1-
3 and YTHDC2, followed by qPCR for SLC7A11 (Fig. 3.7A, Fig. 3.8A-C). We identified YTHDF3
as the corresponding reader for SLC7A11 m6
A, with 12.6-fold relative enrichment of SLC7A11
mRNA upon YTHDF3 pull-down that was abrogated by prior YTHDF3 siRNA depletion.
Consistent with this, hypomethylated SLC7A11 transcript in cisplatin-resistant T24R2 cells had
reduced YTHDF3 binding, resulting in reduced SLC7A11 mRNA enrichment of only 5.4-fold (Fig.
3.7A, Fig. 3.8B). We examined the RNA-seq data and found no significant differences in mRNA
expression levels between the three YTHDF readers, suggesting this was not a factor in the
specificity of reader binding (Fig. 3.8D). Since YTHDF3 is an m6
A reader known to promote
mRNA degradation, we evaluated SLC7A11 mRNA stability in cisplatin-sensitive vs. cisplatinresistant cells. To do so, we blocked transcription using Actinomycin D and evaluated the rate of
degradation of SLC7A11 mRNA. We found that SLC7A11 transcripts which are relatively
hypomethylated in T24R2 cells degraded at a slower rate, whereas an unmethylated control
transcript (GAPDH) was not affected (Fig. 3.7B, Fig. 3.8E). Consistent with this, SLC7A11 mRNA
was increased 8-fold by RNA-seq in cisplatin-resistant cells (Fig. 3.7C, Fig. 3.8F), as was
SLC7A11 protein level by Western blot (Fig. 3.7D). Correspondingly, more prominent cytoplasmic
distribution of SLC7A11 can be observed in resistant cells on immunofluorescence microscopy
(Fig. 3.7E).
SLC7A11 serves as a suppressor of ferroptosis, an iron-dependent cell death program induced
by excessive lipid peroxidation at cellular membranes125-127. Therefore, we treated T24 vs. T24R2
cells with cisplatin and quantified lipid peroxidation, which showed that cisplatin-resistant T24R2

42
cells had significantly less ferroptosis (Fig. 3.7F). Taken together, cisplatin-resistant cells have
decreased SLC7A11 m6
A associated with decreased binding to the YTHDF3 m6
A reader and
decreased transcript degradation, resulting in increased mRNA and protein levels of SLC7A11,
which attenuates ferroptosis-mediated cell death (Fig. 3.7G).

43
Figure 3.7. In cisplatin-resistant cells, hypomethylated SLC7A11 transcript has decreased
binding of its reader YTHDF3 and decreased degradation, resulting in increased SLC7A11
protein and reduced ferroptosis.
(A) YTHDF3 binding to SLC7A11 mRNA in RIP-qPCR of T24, T24 siYHDF3 knockdown and
T24R2 cells. Representative RIP-qPCR data of experiments performed in three biological
replicates. (B) SLC7A11 mRNA decay in T24 and T24R2 via Actinomycin D RNA stability assay.
(C) Normalized RNA-seq count of SLC7A11 in T24 and T24R2 cells. (D) Western blot and
densitometric quantification, and (E) immunofluorescence of SLC7A11 protein in T24 and T24R2
cells with DAPI in blue and SLC7A11 in green. (F) Lipid peroxidation and ferroptosis flow
cytometry assay using BODIPY 581/591 stain in and T24 and T24R2 cells treated with cisplatin.
(G) Model schema illustrating reduced SLC7A11 m6
A which leads to decreased YTHDF3 binding,
increased mRNA stability, and in turn, increased SLC7A11 protein and decreased ferroptosis.
Experiments were performed in three biological replicates. Significance codes: ‘**’: p<0.01, ‘*’: p
<0.05.

44
Figure 3.8. Immunoprecipitation efficiency for RIP-qPCR and mRNA decay negative
controls in T24 and T24R2 BC cells.
Western blot showing immunoprecipitation efficiency for RIP-qPCR experiment for YTHDF3 in
(A) T24 and T24 siYTHDF3 KD (two independent siRNAs) and (B) T24R2. Negative RIP-qPCR
results and IP efficiency for additional m6
A readers, (C) YTHDF1-2 and YTHDC2 showing no
enrichment of SLC7A11 mRNA upon significant pull-down. (D) Normalized RNA-seq count of
YTHDF1-3 in T24 and T24R2 cells. (E) Negative control GAPDH mRNA decay in T24 and T24R2
via Actinomycin D RNA stability assay. (F) Western blot and densitometric quantification of
SLC7A11 protein expression between T24 and T24R2 in four additional replicates. L=ladder.

45
3.4 Short-term cisplatin treatment of BC cell lines reduces SLC7A11 transcript methylation,
resulting in decreased YTHDF3 binding, decreased degradation, and increased mRNA and
protein levels.
Having observed the role of SLC7A11 transcript methylation in established cisplatin-resistant cell
lines, we evaluated whether these effects can be induced with short-term treatment of cisplatinsensitive cells. We started by evaluating the expression of SLC7A11 mRNA in T24 and UM-UC3 BC cell lines upon treatment with different concentrations of cisplatin over time and found a
significant increase in SLC7A11 as early as 48 hours in T24 cells and 72 hours in UM-UC-3 cells,
using both 5uM and 10uM cisplatin (Fig. 3.10A). We selected the earliest time point and lowest
cisplatin concentration for each cell line to evaluate SLC7A11 m6
A, mRNA and protein expression
changes upon cisplatin treatment (5uM for 48 hours for T24 and 5uM for 72 hours for UM-UC-3).
SLC7A11 m6
A measured by MeRIP-qPCR was 90% and 80% depleted in T24 and UM-UC-3
cells, respectively, after cisplatin treatment (Fig. 3.9A, Fig. 3.10B). This m6
A decrease was
associated with increase in SLC7A11 mRNA in both cell lines (Fig. 3.9B). Consistent with these
findings, YTHDF3 binding was significantly decreased in cisplatin treated cells (Fig. 3.9C), which
was associated with reduced degradation and greater SLC7A11 mRNA stability compared to an
unmethylated control transcript (GAPDH) (Fig. 3.9D, Fig. 3.10E) and increased SLC7A11 protein
levels by Western blot (Fig. 3.9E). Correspondingly, more prominent cytoplasmic distribution of
SLC7A11 can be observed in resistant cells in cisplatin-treated cells compared to untreated on
immunofluorescence microscopy (Fig. 3.9F). Hence, SLC7A11 transcript methylation phenotypes
displayed by established cisplatin-resistant cells were recapitulated in cisplatin-sensitive cells as
early as 48hours after cisplatin treatment, suggesting that m6
A RNA modifications may regulate
early changes that contribute to cisplatin resistance. We next asked whether the new pro-survival
methylation profile is “remembered” by cells and hypothesized that that SLC7A11 transcripts may
no longer retain the hypomethylated state once the chemotherapeutic stressor is removed. To
test this hypothesis, we treated T24 and UM-UC-3 cells with short-term cisplatin, and then

46
discontinued treatment and re-analyzed SLC7A11 mRNA methylation and expression when the
cells fully recovered their proliferation after 2-3 weeks. Surviving cells indeed returned to their
baseline phenotype with increased SLC7A11 m6
A and corresponding reduced expression of
SLC7A11 (Fig. 3.9G-H, Fig. 3.10F).

47

48
Figure 3.9. Short-term cisplatin treatment of cisplatin-sensitive BC cell lines reduces
SLC7A11 transcript methylation, resulting in decreased YTHDF3 binding, decreased
degradation and increased mRNA and protein levels.
(A) SCL7A11 m6
A fold enrichment in cisplatin-treated cells relative to untreated cells, measured
by MeRIP-qPCR in T24 and UM-UC-3 cells. (B) Relative SLC7A11 mRNA expression by qPCR
of T24 and UM-UC-3 cells treated with cisplatin compared to untreated cells. (C) SLC7A11 mRNA
enrichment from YTHDF3 by RIP-qPCR, performed in cisplatin-treated vs cisplatin-untreated T24
and UM-UC-3 cells. Error bars reflect replicate measurements of SLC7A11 transcripts from the
enriched protein. (D) SLC7A11 mRNA decay in T24 and UM-UC-3 cells treated with cisplatin
compared to untreated cells. (E) Western blot and densitometric quantification, and (F)
immunofluorescence of SLC7A11 protein in T24 and UM-UC-3 cells treated with cisplatin
compared to untreated cells. DAPI is in blue and SLC7A11 in green. (G) SLC7A11 m6
A fold
enrichment in cisplatin-treated cells relative to untreated cells, measured by MeRIP-qPCR in T24
and UM-UC-3 cells after cisplatin treatment, and again after recovery from cisplatin treatment. (H)
Relative SLC7A11 mRNA expression measured by qPCR in T24 and UM-UC-3 cells after
cisplatin treatment, and again after recovery from cisplatin treatment. Experiments were
performed in three biological replicates. Significance codes: ‘****’: p< 0.0001, ‘***’: p< 0.001, ‘**’:
p<0.01, ‘*’: p <0.05.

49

50
Figure 3.10. Cisplatin treatment optimization, MeRIP-qPCR and immunoprecipitation
efficiency for RIP-qPCR for T24 and UM-UC-3 BC cells.
(A) Relative SLC7A11 mRNA expression in T24 and UM-UC-3 cells untreated vs treated with
5uM and 10uM cisplatin after 48 and 72 hours. (B) SLC7A11 m6
A fold enrichment for T24 and
UM-UC-3 untreated and treated with 5uM cisplatin for 48 hours using MeRIP-qPCR. (C-D)
Western blot showing immunoprecipitation efficiency for RIP-qPCR experiment for YTHDF3 in
(C) T24 +/- cisplatin and (D) UM-UC-3 +/- cisplatin including input, IgG control IP and YTHDF3
IP. (E) Negative control GAPDH mRNA decay in T24 and UM-UC-3 cells treated with cisplatin
compared to untreated cells. (F) SLC7A11 m6
A fold enrichment for T24 and UM-UC-3 cells at
baseline, after cisplatin treatment, and again after recovery from cisplatin treatment, using
MeRIP-qPCR. L=ladder.

51
3.5 Epitranscriptomic regulation and cancer promoting role of SLC7A11 are recapitulated
in patient derived organoids (PDOs) and clinical outcomes.
To validate the cell line findings in tumor tissues from BC patients, we generated and
characterized patient derived organoids (PDOs) from patient bladder tumor samples from freshly
resected transurethral resection of bladder tumor (TURBT) (Fig. 3.11A, Fig 3.12A). For the
experiments showed, we used one PDO generated in our lab (CTC-1044) and one obtained from
the NCI repository (CK-9151). After 48hours of cisplatin treatment, m6
A was significantly depleted
(Fig. 3.11B, Fig. 3.12B), which was associated with significantly increased levels of SLC7A11
mRNA and protein levels (Fig. 3.11C-D). Consistent with these findings, siRNA-mediated
depletion of SLC7A11 sensitized the PDOs to cisplatin (Fig. 3.11E, Fig. 3.12C-E).
To further evaluate the clinical significance of our findings in larger cohorts of patients, we
analyzed TCGA public database to evaluate associations between SLC7A11 expression and
cancer-relevant clinical variables and outcomes. In TCGA, high expression of SLC7A11 was
associated with lower overall survival, disease-specific survival, and progression free interval
across all malignancies (Fig. 3.11F-H). In BC specifically, patients with high histologic grade had
higher SLC7A11 mRNA expression compared to those with low histologic grade (Fig. 3.11I).
Moreover, patients with progressive disease had higher SLC7A11 mRNA expression compared
to patients with no progression (Fig. 3.11J). Among patients with progressive disease, those with
higher SLC7A11 mRNA expression had lower disease-specific survival (Fig. 3.11K).

52

53
Figure 3.11. Epitranscriptomic regulation and cancer promoting roles of SLC7A11 are
recapitulated in patient derived organoids (PDOs) and clinical outcomes.
(A) Schema for establishing BC PDOs and representative bright-field images of CTC-1044 and
CK-9151 at 10X magnification. (B) Ratio of SLC7A11 m6
A fold enrichment of cisplatin-treated to
untreated PDOs from MeRIP-qPCR of CTC-1044 and CK-9151. (C) Relative SLC7A11 mRNA
expression by qPCR of CTC-1044 and CK-9151 PDOs treated with cisplatin, compared to
untreated cells. (D) Western blot and densitometric quantification of SLC7A11 protein in CTC1044 and CK-9151 PDOs treated with cisplatin compared to untreated. (E) Functional validation
of SLC7A11 evaluating its effect on cisplatin resistance in PDOs via siSLC7A11 knockdown
followed by cisplatin treatment in CTC-1044 (CDI: 0.5431747) and CK-9151 (CDI: 0.5660335).
(F-H) Kaplan-Meier Curves for overall survival (F), disease-specific survival (G), and progressionfree interval (H) in patients with all cancer types from The Cancer Genome Atlas (TCGA), with
high vs. low SLC7A11 expression split at the median. (I-K) SLC7A11 expression in BC patients
with high vs. low histologic grade (I), and non-progressive vs. progressive disease after primary
therapy (J) in TCGA. (K) Disease specific survival in BC patients with progressive disease on
primary therapy in TCGA, with high vs. low SLC7A11 expression between highest and lowest
quartiles. Experiments were performed in three biological replicates. Significance codes: ‘****’:
0.00001, ‘***’: 0.001, ‘*’: 0.05.

54
3.12. Characterization, MeRIP-qPCR, SLC7A11 siRNA knockdown efficiency and cell
viability of PDOs.
(A) Characterization of patient derived organoids (PDOs) using histologic stains for H&E, CK5,
Ki67, CK20, TP63 and Uroplakin III at 10X. (B) SLC7A11 m6
A fold enrichment by MeRIP-qPCR
for PDO CTC- 1044 and CK-9151 untreated vs. treated with 50uM cisplatin for 48 hours. (C)
SLC7A11 siRNA-knockdown efficiency in CTC-1044 and CK-9151. Cell viability upon siSLC7A11
knockdown with and without cisplatin treatment in (D) CTC-1044 and (E) CK-9151. NC= negative
control. Significant codes: ‘***’: 0.001, ‘*’: 0.05.

55
Discussion
A major obstacle to achieving lasting cures for advanced malignancies is the emergence of
acquired resistance to treatment, not only due to genetically resistant clones but also through
phenotypic plasticity. In this study, we aimed to characterize the role of m6
A RNA modifications
in regulating the expression of genes that promote transition to cisplatin resistance in bladder
cancer. M6
A modifications play a dynamic role in cell fate and pluripotency, raising the possibility
that this epitranscriptomic mechanism contributes to phenotypic plasticity and drug resistance in
BC72,73,112. It has been proposed that while most m6
A sites are largely hard-wired, a smaller more
variable subset of sites may serve as master regulators of specific cellular pathways57. Writers
and erasers are responsible for depositing and removing m6
A, and readers in particular serve as
effectors of downstream events, including alternative splicing, translation, and degradation,
thereby regulating cellular phenotypes and impacting cancer progression119. Despite this
multiplicity of observed regulatory mechanisms, the dominant function of m6
A is to regulate RNA
stability, leading to increases or decreases in gene expression57.
In cancer, m6
A has been dubbed a “double-edged sword” – the presence of modification at one
site or its absence at another may both contribute to cancer progression46. Thus, it is critical to
take an agnostic approach within each cancer of interest to accurately discover and identify
transcripts associated with differential methylation and differential expression. To date, most
studies have focused on a specific m6
A effector (writer, eraser, readers) and considered
transcriptome-wide m6
A changes induced by knock-out or overexpression of the effector of
interest84,88,95,128. These studies have significantly advanced the field. At the same time, because
effectors have a multitude of partners and targets, any specific m6
A modifications or cancer
phenotypes induced by effector manipulation are understood to occur in the context of numerous
other transcripts and pathways potentially perturbed that same effector manipulation. In this study,
we took a different approach by directly mapping the epitranscriptomic landscape of

56
chemotherapy sensitive vs. resistant cells, thereby focusing our discovery on m6
A alterations
associated with these disease states rather than on specific effectors.
Using a well-established cell line model of cisplatin resistance in bladder cancer105, we analyzed
transcriptome-wide changes in m6
A RNA modifications and gene expression using MeRIP-seq
and RNA-seq. To do so, we adopted a rigorous informatics approach: taking into consideration
the limitations of MeRIP-seq, implementing the guidelines for best practices outlined in McIntyre
et al. 202070 and using 5 tightly clustered replicates per condition. As a result, this study is, to our
knowledge, the most well-powered differential MeRIP-seq analysis of its kind to-date, as reflected
by the high validation rate (78%) of MeRIP-seq targets by repeat targeted MeRIP-qPCR (Fig.
3.5). In all, we identified 130 transcripts that were both differentially methylated and differentially
expressed. It is possible that many of these candidates are involved, singly or in combination, to
varying degrees in the transition to cisplatin resistance. But here we sought to focus on a
candidate driver of resistance with readily interpretable downstream mechanism of action as proof
of concept for the discovery approach used in this study. Having demonstrated the feasibility of
this approach, it is our hope that it will be applied to additional transcript candidates and in other
disease models.
Using this strategy, we identified SLC7A11/xCT to be differentially hypomethylated in resistant
cells, a modification that decreased its binding to the reader YTHDF3. Reduced association with
YTHDF3 in turn decreased SLC7A11 transcript degradation, leading to increased levels at the
mRNA and protein level, reduced ferroptosis, and enhanced survival (Fig. 3.13). Notably, these
changes were apparent not only in established cisplatin-resistant cells, but also in cisplatinsensitive cell lines and PDOs after brief (48hours) treatment with cisplatin. Upon removal of
chemotherapeutic stressor, surviving cells eventually resumed proliferating and returned to their
baseline phenotype with high SLC7A11 m6
A and low SLC7A11 expression, underscoring the
rapid adaptative nature of phenotypic plasticity.

57
SLC7A11 is a 1:1 cystine-glutamate antiporter that replenishes cystine, a key precursor of
glutathione biosynthesis and antioxidant defense 125. SLC7A11 is overexpressed in multiple
human cancers and promotes tumor growth by suppressing ferroptosis, a form of iron-dependent
regulated cell death induced by excessive lipid peroxidation at cellular membranes125-127.
SLC7A11 was first linked to cisplatin resistance in ovarian cancer and has since been studied in
multiple other cancer types 129. SLC7A11 inhibitors such as sulfasalazine and sorafenib, as well
as other ferroptosis inducers are currently being tested in combination therapy with cisplatin and
doxorubicin in clinical trials for head and neck carcinomas, acute myeloid leukemia, glioblastoma,
ovarian cancer, hepatocellular carcinoma and renal cell carcinoma 126,130. In BC, we are aware of
two studies involving SLC7A11 127,131. Both found an association between SLC7A11
overexpression and poor clinical outcome, and one reported a link with cisplatin resistance 131.
Although these prior reports in BC did not address epitranscriptomic regulation, this mechanism
was observed in other malignancies. In lung adenocarcinoma, one study reported that the
YTHDC2 m6
A reader binds an SLC7A11 m6
A site at the 3’UTR, leading to decreased mRNA
stability and expression132. Another study in hepatoblastoma reported that m6
A-dependent
inhibition of SLC7A11 deadenylation promotes tumorigenesis 133. In our current study, in-silico
analysis of publicly available databases confirmed that SLC7A11 levels are associated with
clinical outcomes in BC and other malignancies. While databases like TCGA are not annotated in
a manner that can directly inform our questions about cisplatin resistance and epitranscriptomic
regulation, they nonetheless attest to the potential clinical relevance of these targets. Future
studies comparing large cohorts of tumor samples from transurethral resection of bladder tumors
(TURBTs pre-cisplatin) versus tumor samples from radical cystectomies (post-cisplatin) could
further expand on the clinical relevance of our mechanistic findings.

58
One potential limitation of MeRIP-seq in this setting are its 100-nucleotide resolution for
determining exact methylation site. While we demonstrate a statistically significant association
between this m6
A site and SLC7A11 transcript enrichment on YTHDF3 (Fig. 4A), we cannot rule
out effector interactions at additional sites on SLC7A11. Future studies could further elucidate
this using newly developed tools with single-nucleotide resolution such as GLORI 134. Another
limitation of MeRIP-seq is its inability to distinguish definitively between m6
A and m6
Am. To
address this, we conducted an adjunct experiment (Sup. Fig. 3D), individually knocking down
ALKBH5 (m6
A eraser) and FTO (m6
Am eraser). Only ALKBH5 knockdown altered SLC7A11
expression, suggesting that the differentially methylated site on SLC7A11 is more likely to be m6
A
rather than m6
Am. Additional genetic manipulations of various effectors and specific m6
A sites
could offer new insights about their potential roles in regulating specific methylation sites
associated with adaptive chemoresistance. It is also important to note that while YTHDF readers
are predominantly implicated in RNA stability and degradation, m6
A at 5’UTR have also been
associated with changes in translation through direct interaction with translation factor eIF3 135. It
is possible that changes in translation could be at play in addition to YTHDF3-mediated mRNA
degradation; however, teasing out the relative contribution of mRNA degradation, eIF3-mediated
translation, and other mechanisms of translation (elongation, termination, ribosome recycling)
would be technically challenging.
Our current study demonstrates that SLC7A11 mRNA methylation and degradation are
reduced in response to cisplatin treatment, contributing to a rapid increase in SLC7A11 antiporter
level that mitigates ferroptosis and promotes survival of BC cells. These findings highlight
epitranscriptomic regulation as an important contributor to phenotypic plasticity in cancer cells.
This capacity for rapid adaptation to chemotherapeutic stress may serve as an initial survival
tactic that “buys time” until clonal selection of adaptive mutations can occur. Therefore, targeting
such early adaptive mechanisms may be an effective strategy to derail the transition to therapy
resistance before it is firmly established.

59
Figure 3.13. Model Schema: m6
A regulates SLC7A11 levels, promoting rapid transition to
cisplatin resistance.
In cisplatin-sensitive cells: high SLC7A11 m6
A is associated with high YTHDF3 m6
A reader
binding, leading to increased mRNA degradation and subsequently low SLC7A11 protein. With
few SLC7A11 antiporters of cystine and glutamate, there is reduced intracellular cystine, a
precursor of glutathione (GSH) synthesis. Glutathione, in terms, serves as antioxidant defense
quenching reactive oxygen species (ROS). Decreased GSH leads to build-up of ROS upon
cisplatin treatment and increased cell death by ferroptosis. Conversely, in cisplatin-resistant cells:
low SLC7A11 m6
A is associated with low YTHDF3 binding and increased SLC7A11 mRNA
stability, leading to an increase in SLC7A11 protein. With high amounts of SLC7A11 proteins,
cells have high intracellular cystine and increased GSH production. These cells are then capable
of quenching the build-up of ROS species upon cisplatin treatment and survive better.

60
Part II:
Prostate Cancer

61
Chapter 4: M6
A RNA Methylation in enzalutamide resistance in prostate cancer
All the experiments described in this chapter were conducted by Emmanuelle Hodara and are
currently unpublished.
Authors
Emmanuelle Hodara1
, Aubree Mades1
, Daniel Bsteh1
, Peggy J. Farnham2
, Suhn K. Rhie2
, &
Amir Goldkorn1,2
1
Division of Medical Oncology, Department of Medicine, Keck School of Medicine of USC and
Norris Comprehensive Cancer Center, Los Angeles, CA 90033, USA
2
Department of Biochemistry and Molecular Medicine, Keck School of Medicine of USC, Los
Angeles, CA 90033, USA
Author Contributions
E.H. conceived and designed the study, conducted all experiments, interpreted the results and
wrote the manuscript. A.M. conducted the qPCR and MeRIP-qPCR validation experiments,
reviewed and edited the manuscript. D.B. participated in bioinformatics analysis and data
visualization. P.J.F. and S.K.R. provided critical feedback for data analysis and reviewed the
manuscript. A.G. designed the study, analyzed the data and wrote the manuscript.
Introduction
The primary standard of care for locally advanced and metastatic prostate cancer is AndrogenDeprivation Therapy (ADT), which reduces testicular androgen production, thereby reducing the
activation of the Androgen Receptor (AR) and its downstream disease-driving signaling related to
cell cycle and cancer progression136. While patients typically respond well to treatment, they
eventually progress to castrate-resistant prostate cancer (CRPC) within 2-3 years of starting
ADT137,138. It was originally posited that CRPC was “hormone refractory” – no longer mediated by
AR; however, it is now established that CRPC remains mainly driven by AR signaling, albeit via
adaptive mechanisms in the form of amplification, mutation, overexpression and alternative splice
variants of AR 35,36. Second generation AR signaling inhibitors (ARSIs) such as enzalutamide and
abiraterone, were developed to treat recurrent, castrate resistant disease, but these extend
median survival only 2-8 months due to additional acquired drug resistance mechanisms138.

62
One such mechanism restores AR signaling through alternative splicing of AR mRNA. A number
of AR splice variants have been identified in the clinic and cell lines and are implicated in
resistance to ARSIs139. Notably ARV7, the best documented AR splice variants, lacks the ligand
binding domain (LBD), is constitutively active and localized to the nucleus independent of AR
binding and is overexpressed in resistant tumors140. Clinical studies of advanced prostate cancer
patients demonstrated that nuclear ARV7 expression in Circulating Tumor Cells (CTCs) is
associated with resistance to ARSI therapy and decreased overall survival (OS)37,38,141.
While ARV7 expression is manifestly implicated in CRPC emergence, the precise genomic
mechanism remains unclear. Recent studies demonstrate that the long non-coding RNA (lncRNA)
MALAT1 contributes to ARSI resistance by promoting the expression of ARV7, and that its
silencing suppresses prostate cancer progression by inhibiting androgen receptor signaling142-145.
MALAT1 has also been showed to regulate alternative splicing by modulation serine/arginine (SR)
splicing factor phosphorylation146. Interestingly, MALAT1 is a well-studied lncRNA with high levels
of in m6
A RNA modifications. It is the classic example of m6
A structural switch, a mode of action
of m6
A in which the presence of the modification increases likelihood of single-strandedness,
structurally altering binding sites of RNA binding proteins147. Additionally, m6
A plays a key role in
regulating alternative splicing and splicing kinetics. Specifically, m6
A modifications at splice
junctions have been associated with fast, constitutive splicing; whereas, intronic m6
A
modifications promote slow and alternative splicing63,148.
Given the role of m6
A modifications in alternative splicing, we posited that m6
A may be involved
in modulating splicing of ARV7 and other AR splice variants, contributing to this specific
mechanism of drug resistant in CRPC. In this study, we set out to systematically identify changes
in m6
A RNA methylation and RNA expression between enzalutamide-sensitive vs. enzalutamideresistant prostate cancer cells. Using unbiased transcriptome-wide m6
A profiling followed by

63
targeted validation, we demonstrated that m6
A modifications are associated with changes in
expression of clinically-relevant gene transcripts in prostate cancer, and that a subset of these
modifications may promote resistance to enzalutamide. Specifically, high m6
A modification is
associated with significant depletion of THBS1, antiangiogenic factor thrombospondin-1,
potentiating enzalutamide resistance in PC. Decreased m6
A levels on THBS1 mRNA is
associated with increased THBS1 expression in enzalutamide-sensitive cells compared to
enzalutamide-resistant cells.
Results
4.1 Enzalutamide-sensitive and enzalutamide-resistant cells have distinct m6
A profiles
We undertook a broad discovery approach comparing enzalutamide-sensitive C4-2B PC cells
(IC50: 50uM enzalutamide) to established enzalutamide-resistant MDV-R PC cells (IC50: 200uM
enzalutamide) (Fig. 4.1A). Resistant MDV-R cells have a 2.5-fold increase in ARV7 mRNA
expression compared to sensitive C4-2B (Fig. 4.1A). We used a low-input methyl-RNAimmunoprecipitation and sequencing (MeRIP-seq) protocol111 and a stringent differential analysis
pipeline incorporating all recommended best practices published to date70, including 4 replicates
per condition and filtering criteria to mitigate the confounding effect of differential expression on
differential methylation (Fig. 4.1B).
As a first pass quality control measure, we applied principal component analysis (PCA) to all
36,393 peaks called by MACS2 based on pull down versus input in all 8 replicates (4 sensitive
and 4 resistant). The sensitive and resistant replicates clustered tightly into two distinct groups,
with PC1 and PC2 representing 76% and 5% variation, respectively, reflecting the power of the
study design to accurately discriminate the two conditions (Fig. 4.1C). Encouraged by these
preliminary results, we proceeded with differential methylation analysis using three statistical
models (DESeq2, EdgeR and QNB) to evaluate statistical significance (p<0.05). We identified

64
633 differentially methylated peaks between enzalutamide sensitive and resistant cells, of which
487 remained after filtering for differential expression and 146 remained after selecting those with
at least 10 reads. 72 of these peaks were hypermethylated in resistant cells, and 74 were
hypermethylated in sensitive cells (Fig. 4.1D). The volcano plot represents the 633 statistically
significant peaks, highlighting the 487 peaks that meet all the filtering criteria and are unique to
either sensitive (blue) or resistant (red) cells (Fig. 4.1E). The top 20 differentially methylated
transcripts based on DESeq2 p-adj were visualized using a heatmap (Fig. 4.1F).

65
Figure 4.1. Enzalutamide-sensitive and Enzalutamide-resistant cells have distinct m6
A
profiles.
(A) Profiling of C4-2B and MDV-R PC cells including cell viability curves with enzalutamide
treatment and mRNA expression of AR-FL and ARV7. (B) MeRIP-seq informatics filtering pipeline
for differential methylation analysis with summarized peak number for different filtering steps.
Numbers in parentheses refer to: peaks hypermethylated in C4-2B, peaks hypermethylated in
MDV-R. (C) PCA plot of MeRIP samples with 4 replicates per cell line. (D) Summary of filtering
results comparing C4-2B to MDV-R, with 72 peaks hypermethylated in C4-2B and 74 peaks
hypermethylated in MDV-R. (E) Volcano Plot of statistically significant differential MeRIP-results
(p<0.05 by DESEQ2, edgeR, and QNB). C4-2B m6
A represents the 72 peaks hypermethylated in
C4-2B cells with log2FC <-1 and MDV-R m6
A represents the 74 peaks hypermethylated in MDVR with log2FC > 1. (F) Heatmap of top 20 differentially methylated transcripts ranked by
DESeq2padj.

66
4.2 Filtering and validating candidate transcripts
Since we hypothesized that m6
A regulates the expression level of gene transcripts that promote
transition to a drug resistant state in PC, we overlapped the MeRIP-seq differential methylation
results with RNA-seq differential expression from the same enzalutamide sensitive and resistant
cell lines (Fig. 4.2A-C). We then applied the filtering pipeline outlined in Fig. 4.3A in order to
home in on transcripts relevant to cancer progression, as follows: We identified 46 candidate
transcripts (52 out of 146 m6
A peaks) that were both differentially expressed and differentially
methylated (Fig. 4.3B). We used Gene Ontology (GO) and Gene Set Enrichment Analysis
(GSEA) to filter the 46 transcripts based on membership in GSEA Hallmarks of Cancer pathways
(Fig. 4.3C) and relevance to GO terms like response to chemotherapy and radiation therapy (Fig.
4.3D). We further ranked the resulting transcripts based on log fold change and relevance to
prostate cancer in existing literature and clinical databases such as The Cancer Genome Atlas
(TCGA), the Oncology Research Information Exchange Network (ORIEN) Avatar and the Catalog
of Somatic Mutations in Cancer (COSMIC). After ranking, we selected the top 12 transcripts for
in vitro validation by qPCR and by targeted immunoprecipitation and PCR (MeRIP-PCR). 8 of the
12 differentially expressed transcripts by RNA-seq were also differentially expressed by qPCR
(Fig. 4.4A). Of the 8 transcripts (10 peaks) whose differential expression was validated by qPCR,
6 transcripts (8 peaks) were also validated as differentially methylated by MeRIP-qPCR (Fig.
4.4B).

67
Figure 4.2. Differential RNA-seq between C4-2B and MDV-R PC cells.
(A) PCA plot of RNA-seq samples with 4 replicates per cell line. (B) Volcano Plot of statistically
significant differential RNA-seq results (p<0.00001). Downregulated represents the peaks with
log2FC < -0.5 and Upregulated with log2FC > 0.5. (C) Heatmap of top 20 differentially expressed
transcripts ranked by log2FC.

68
Figure 4.3. Identification of cancer-relevant transcripts that are both differentially
methylated and differentially expressed in enzalutamide-resistant cells.
(A) Discovery and filtering pipeline. (B) RNA expression Log2FC vs. RNA Methylation Log2FC of
the 46 transcripts (52 peaks) that are both differentially methylated and differentially expressed
between C4-2B and MDV-R. Peaks are colored based on filtering criteria: GO, GSEA, both or
neither. (C) Differentially methylated and expressed transcripts statistically significantly
associated with Gene Ontology (GO) Terms (p-adj <0.05). (D) Differentially methylated and
expressed transcripts statistically significantly associated with Gene Set Enrichment Analysis
(GSEA) Cancer Hallmarks (p-adj < 0.05).

69
Figure 4.4. Validation of top 12 differentially methylated and expressed candidate
transcripts.
(A) 8 of 12 candidate transcripts (boxed in red) were validated as differentially expressed using
qPCR, corroborating RNA-seq results. (B) 6 of the 8 candidate transcripts (boxed in red) were
validated by MeRIP-qPCR, corroborating MeRIP-seq results. Red dotted lines in (A) and (B)
represent the validation threshold.

70
4.3 Evaluating the role of candidate transcripts in enzalutamide resistance
Next, we determined if any of the 6 differentially methylated and expressed genes identified in
enzalutamide resistant MDVR cells also played a role in the transition to enzalutamide resistance
in sensitive cell lines. To do this, each of the 6 genes was siRNA-depleted in enzalutamidesensitive cell lines treated with enzalutamide (Table 4.1). One candidate, THBS1, was identified
as an inhibitor of acute enzalutamide resistance in two PC cell lines, C4-2B and LNCaP (IC50:
5uM enzalutamide, Fig.4.5C).siRNA knockdown of THBS1 had significant antagonism with
enzalutamide; in other words, THBS1 depletion reduced cell death in response to enzalutamide
and potentiated resistance, recapitulating the associations observed earlier in established
resistant MDVR cells. (Fig. 4.5D-E). MeRIP-seq of THBS1 mRNA revealed increased m6
A levels
at an exonic region in enzalutamide-resistant cells compared to enzalutamide-sensitive ones and
no difference in m6
A levels at other regions, including one unchanged peak at the 3’UTR (Fig.
4.5A). THBS1 mRNA expression was decreased 41-fold by RNA-seq in enzalutamide-resistant
cells (Fig. 4.5B).
Table 4.1. Summary table of functional validation for the 6 transcripts previously validated
by qPCR and MeRIP-qPCR.
Target Differential
Expression
Peak #
Location
Differential
Methylation
Expected
Effect of
KD on
Resistance
Actual Effect
of KD on
Resistance:
CDI
Corroborates?
GRHL2 ↓ 1: utr5 ↓ Antagonism NA: could not
KD GRHL2 NA
THBS1 ↓ 1: exon ↑ Antagonism CDI: 1.09
Antagonism YES
FN1 ↓ 1: exon ↑ Antagonism CDI: 1.00
No Effect NO
PARP10 ↓ 1: utr3
1: exon ↑↓ Antagonism CDI: 0.94
No Effect NO
PAK4 ↓ 1: intron ↓ Antagonism CDI: 0.89
Synergy
NOT
SIGNIFICANT
SYT4 ↑ 1: utr3 ↓ Synergy CDI: 0.83
Synergy
NOT
SIGNIFICANT
*Arrows indicate expression or methylation levels in MDV-R compared to C4-2B. CDI is a
coefficient of synergy between siRNA knockdown and Enzalutamide treatment in enzalutamidesensitive LnCAP cells. CDI < 0.7 indicates statistically significant synergy and CDI> 1 indicates
statically significant antagonism. CDI is the average of 3 biological replicates

71
Figure 4.5. Validation of THBS1 m6
A methylation, mRNA expression, and effect on
enzalutamide treatment.
(A) M6
A coverage plot for THBS1 in C4-2B and MDV-R cells with differential exonic peak, and
consistent (no difference) peak at the 3’UTR. (B) Normalized RNA-seq count of SLC7A11 in C4-
2B and MDV-R cells. (C) LNCaP cell viability after 72 hours of enzalutamide treatment at
varying concentration. (D) THBS1 siRNA-knockdown efficiency for C4-2B and LNCaP cells.
NC= negative control (scrambled siRNA). (E) Functional validation of THBS1 evaluating its
effect on enzalutamide resistance via siTHBS1 knockdown followed by enzalutamide treatment
in C4-2B cells (CDI: 1.30) and LNCaP cells (CDI: 1.09). CDI < 0.7 indicates statistically
significant synergy. CDI > 1.0 indicates statistically significant antagonism. Experiments were
performed using three biological replicates. Significant codes: ‘*’: p< 0.05, ‘***’: p<0.001.

72
Discussion
In this study, we aimed to characterize the role of m6
A RNA modifications in regulating the
expression of genes that promote transition to enzalutamide resistance in prostate cancer. M6
A
modifications play a dynamic role in alternative splicing, raising the possibility that this
epitranscriptomic mechanism contributes to ARV7-mediated drug resistance in PC57,63,77,148. One
previously published study generated the first m6
A map in prostate cancer and examined how
METTL3 regulates gene and protein expression, finding that METTL3 knockdown rendered the
cells resistant to androgen receptor antagonists via androgen-receptor independent
mechanisms96. Other studies have focused on a specific m6
A effector (writer, eraser, readers)
and considered transcriptome-wide m6
A changes induced by knock-out or overexpression of the
effector of interest. In order to focus our discovery on m6
A alterations associated with drugresistant disease state rather than on specific effectors, we opted to directly map the landscape
of enzalutamide-sensitive vs. resistant cells.
Using a well-established cell line model of enzalutamide resistance in prostate cancer105, we
analyzed transcriptome-wide changes in m6
A RNA modifications and gene expression using
MeRIP-seq and RNA-seq. In all, we identified 46 transcripts that were both differentially
methylated and differentially expressed. It is possible that many of these candidates are involved,
singly or in combination, to varying degrees in the transition to enzalutamide resistance. But here
we sought to focus on a candidate driver of resistance with readily interpretable downstream
mechanism of action. Using this strategy, we identified Thrombospondin-1 (THBS1) to be
differentially hypermethylated and differentially downregulated in resistant cells. Functionally, we
found that knocking down THBS1 potentiated an acute shift to enzalutamide resistance in two
enzalutamide-sensitive PC cell lines, potentiating the associations observed earlier in the
established resistant MDVR model. These findings identified THBS1 as a gene that becomes
repressed as prostate cancer cells transition to resistant phenotype.

73
THBS1, also known as TSP1, is a well-characterized adhesive glycoprotein that mediates cell-tocell and cell-to-matrix interactions and is considered an antiangiogenic factor149. THBS1 is
downregulated in multiple human cancers including melanoma and breast cancer150. Consistent
with our findings, THBS1 has been reported to be downregulated in advanced prostate cancer
patient samples and negatively correlate with neuroendocrine marker151. Specifically, treatment
with enzalutamide activates EZH2 axis to epigenetically repress THBS1. While antiangiogenic,
THBS1 has also been reported to promote migration and development of advanced prostate
tumors152. One recent study links m6
A to THBS1 downregulation in prostate cancer. Starting with
the clinical observation that m6
A writer co-factor METTL14 upregulation is correlated with poor
prognosis in prostate cancer patients, this study identifies THBS1 as a downstream target of
METTL14, resulting in its YTHDF2-mediated mRNA degradation and downregulation101. Unlike
our approach which compared enzalutamide-sensitive and enzalutamide-resistant cells, their
study compares RNA-seq and MeRIP-seq data of PC cells with and without knockdown of
METTL14 to identify the same important driver of prostate cancer progression. While the study
elucidates the epitranscriptomic regulatory mechanism of THBS1 we were seeking to explore, it
does not note the link between THBS1 downregulation and enzalutamide resistance that we
establish. Future directions for this project may focus on exploring the link between THBS1
methylation, transcript repression, and ARV7-mediated enzalutamide drug resistance in CRPC.

74
Chapter 5: Multiparametric liquid biopsy analysis in metastatic prostate cancer
The work described in this chapter has been published in Hodara, E. et al. 2019. “Multiparametric
liquid biopsy analysis in metastatic prostate cancer.” JCI Insight. (PMID: 30702443) 153.
Authors
Emmanuelle Hodara,1 Gareth Morrison,1 Alexander Cunha,1 Daniel Zainfeld,1 Tong
Xu,1 Yucheng Xu,1 Paul W. Dempsey,2 Paul C. Pagano,2 Farideh Bischoff,3 Aditi
Khurana,4 Samuel Koo,4 Marc Ting,4 Philip D. Cotter,4 Mathew W. Moore,4 Shelly Gunn,4 Joshua
Usher,5 Shahrooz Rabizadeh,6
Peter Danenberg,7 Kathleen Danenberg,5 John Carpten,8 Tanya
Dorff,1 David Quinn,1 and Amir Goldkorn1
Affiliations
1
Department of Medicine, University of Southern California (USC) Keck School of Medicine
and Norris Comprehensive Cancer Center (NCCC), Los Angeles, California, USA.
2
Cynvenio Biosystems Inc., Westlake Village, California, USA.
3
Menarini Silicon Biosystems, San Diego, California, USA.
4
ResearchDx, Irvine, California, USA.
5
Liquid Genomics, Torrance, California, USA.
6
NantHealth, Culver City, California, USA.
7
Department of Biochemistry and Molecular Medicine and.
8
Department of Translational Genomics, USC Keck School of Medicine and NCCC, Los
Angeles, California, USA.
Author Contributions
EH analyzed the data and wrote the manuscript. GM, AC, DZ, TX, and YX contributed to sample
collection and processing. PWD, PCP, FB, AK, SK, MT, PDC, MWM, SG, JU, SR, PD, and KD
contributed to sample processing and analysis using their respective platforms. JC contributed to
the final data analysis and manuscript writing. TD and DQ provided patient samples and reviewed
the analysis and manuscript. AG designed the studies, analyzed the data, and wrote the
manuscript.
Introduction
Prostate cancer is the second most prevalent malignancy in the US and the third highest
cause of cancer mortality in men154. Next-generation sequencing (NGS) studies of primary tumors
and metastases have begun to map the genomic landscape of prostate cancer from early to late
disease by identifying characteristic somatic alterations as well as disease-relevant transcripts
and germline alterations 155-159. These studies have revealed extensive intratumor and interpatient
heterogeneity and have identified somatic alterations that increase in frequency with exposure to
therapies. Given this spatial and temporal heterogeneity, a single tissue biopsy from one disease

75
site at one point in time is unlikely to fully represent the molecular profile of the cancer, yet
obtaining sequential tissue biopsies from multiple sites is prohibitively invasive and costly.
Liquid biopsies enriched from a standard peripheral blood draw offer an alternative to solid
tissue biopsies that is repeatable and minimally invasive, allowing “real-time” monitoring of a
patient’s treatment response and disease evolution over time160. Blood-based biomarkers in
advanced prostate cancer can be broadly divided into 2 main sample types: (a) the cellular
component consisting of white blood cells (WBCs) and circulating tumor cells (CTCs) reflecting
germline and somatic phenotypes, respectively, and (b) the plasma component containing cellfree DNA (cfDNA) and cell-free RNA (cfRNA) released from normal and malignant cells
throughout the body.
To date, most studies have focused on the biomarker potential of individual liquid biopsies,
yet each of these approaches yields orthogonal and potentially valuable cancer-relevant data.
Therefore, a critical next step in liquid biopsy profiling is what we term a “multiparametric”
approach, which integrates multiple blood-based tumor phenotypes to yield a maximally
informative disease profile. To test the technical feasibility of such an approach, we undertook a
pilot study using blood samples drawn from a cohort of 20 men with metastatic castrate-resistant
prostate cancer (mCRPC). Each blood sample was analyzed simultaneously for tumor-relevant
cfDNA, cfRNA, CTC DNA, and germline DNA, and these data were used to generate patientspecific, multiparametric liquid biopsy tumor profiles at sequential points.
Results
5.1 Patient cohort and liquid biopsies.
Blood samples were collected from 20 men with mCRPC (Table 5.1) with median age of 70 years
(range: 46–81 years), median prostate-specific antigen (PSA) at blood draw of 48 ng/ml (range:
0–435 ng/ml,) and visceral metastases (poor prognostic factor) present in 60% of cases. At the
time of sampling, patients had received an average of 3 lines of standard therapy: 85% with

76
second-line hormonal therapy, 55% with chemotherapy (docetaxel, cabazitaxel), 40% with
sipuleucel-T, and 30% with radium-223. Thirty percent of patients received additional therapies
(e.g., poly [ADP-ribose] polymerase, or PARP, inhibitors; immune checkpoint inhibitors; tyrosine
kinase inhibitors; and other chemotherapeutic agents).

77
Table 5.1. Summary chart of patient clinical data.
*T1 = first blood draw, T2= second blood draw. For treatment status, P= progressing on treatment,
R= responding to treatment. Abi is abiraterone, Enza is enzalutamide, SipT is sipuleucel-T, and
DTX is docetaxel.

78
At each liquid biopsy time point, 3 tubes of blood were processed simultaneously for CTC count,
somatic single nucleotide variants (SSNVs) derived from CTCs and matched cfDNA, copy number
variant (CNV) analysis derived from single CTCs, and androgen receptor (AR) expression profiles
from cfRNA (Fig. 5.1). In 40% of patients, a blood sample at a second time point was collected to
assess changes in the tumor molecular profile; most of these samples were evaluated at disease
progression, but some were still responding to the treatment being given at the first draw. An
overview of data generated from all 20 patients at all time points is in Table 5.2. Seventy-five
percent of patients had detectable CTCs, with a median of 20 CTCs/7.5 ml (range: 1–692/7.5 ml).
Single CTCs were recovered by dielectric manipulation for CNV analysis in 6 patients, revealing
amplifications and deletions in multiple cancer-relevant genes, including AR, MYC, TP53,
and PTEN. Matched CTC and cfDNA SSNV analyses were performed in 19 of 20 patients.
Fourteen of nineteen (68%) had detectable SSNVs in CTC DNA or matched cfDNA, including
prostate cancer–relevant genes, such as TP53, PIK3CA, EGFR, and HRAS. Matched CTC DNA
and cfDNA were analyzed at a second time point in 8 patients, and these revealed previously
observed as well as new SSNVs at disease progression. Nineteen of twenty patients had a
detectable cfRNA AR transcript, 5 of whom had a detectable cfRNA ARV7 transcript
(ARV7:AR ratio range: 0.5%–33.2%).

79
Figure 5.1. Multiparametric workflow.
Three blood samples were collected and analyzed in parallel for cfDNA (SSNVs), CTC DNA
(enumeration, SSNVs, CNVs), and cfRNA (AR, ARV7 relative expression). Solid tissue biopsy
data (SSNV, CNV) were available in a subset of patients (FoundationOne). qPCR, quantitative
PCR.

80
Table 5.2. Summary chart of Liquid Biopsy Alterations.
* T1 = first blood draw, T2= second blood draw. For CTC CNV, two numbers separated by “/”
denote the number of CNVs detected in each individual CTC. ND= none detected, NA= not
available (assay could not be performed due to platform failure or sample not meeting eligibility
criteria (e.g. minimum CTC # or purity).

81
5.2 Multiparametric molecular profiles.
An integrated tumor profile was generated for 2 of the patients using their multiparametric
molecular data. For example, patient 3 (Fig. 5.2A) is a 76-year-old man with lymph node, bone,
and brain metastases whose disease progressed after treatment with androgen deprivation
therapy (ADT), abiraterone, enzalutamide, sipuleucel-T, and docetaxel. At the time of his liquid
biopsy, he was responding to treatment with radium-223, and his PSA was 60 ng/ml. When his
blood was analyzed, he was found to have a high CellSearch CTC count of 168/7.5 ml. Individual
CTCs were analyzed for CNVs and were found to have ARamplification (commonly observed with
progression on abiraterone or enzalutamide) as well as amplification in other cancer-related
genes — BCL6, CCND1, MYC, SOX2, STAT4, TERC, RUNX1T1, and TMPRSS2 — and losses
in several tumor suppressor genes — BRCA2, RB1, and TP53. Some of these CNVs were
concordant with FoundationOne genomic profiling (Foundation Medicine) from a concurrent
lymph node biopsy, while others were unique to the CTCs. In addition, SSNV analysis of CTCs
and matched plasma cfDNA revealed a concordant nonsense mutation in TP53 detected in CTC
DNA and solid tissue but not in cfDNA. Analysis of cfRNA was positive for AR transcripts but
negative for ARV7.
In another example, patient 10 (Fig. 5.2B) is a 46-year-old man with lymph node and bone
metastases whose disease progressed after treatment with ADT, abiraterone, radium-223, and
pembrolizumab. At the time of his first liquid biopsy, he was responding to treatment with a PARP
inhibitor, and his PSA was 55 ng/ml. When his blood was analyzed, he was found to have a low
CTC count of 3/7.5 ml by CellSearch. SSNV analysis of CTC DNA and cfDNA revealed a
concordant missense mutation in TP53 that was also detected in the FoundationOne profile of a
primary tumor biopsy performed 38 months earlier. Analysis of cfRNA was positive
for AR transcripts but negative for ARV7. A second liquid biopsy was drawn after progression on
PARP inhibitors and a change in therapy to cabozantinib. At that time, patient 10’s PSA had risen
to 435 ng/ml and his CTC count had increased to 11/7.5 ml by CellSearch, both associated with

82
poor prognosis. SSNV analysis revealed the same TP53 missense mutation in both CTC DNA
and cfDNA but also an additional TP53 nonsense mutation detected in cfDNA only. This mutation
was not present in the first liquid biopsy or in the prostate tumor biopsy from 38 months prior.
Analysis of cfRNA revealed a 200-fold increase in AR transcript compared with the liquid biopsy
performed before progression, as well as newly detectable ARV7.

83
Figure 5.2. Multiparametric profiles of individual patients (ID numbers 3, 10).
Blue = shared SSNVs in liquid and solid biopsies; green = shared amplifications or losses in liquid
and solid biopsies; red = alterations interrogated in liquid and solid biopsy panels but detected in
only 1; black = alterations or expression levels interrogated in only 1 panel. amp, amplification;
LN, lymph node; Bx, biopsy; PARP-I, PARP inhibitor.

84
5.3 Comparative analysis of liquid biopsy versus tumor biopsy.
Mutation profiles were compared between liquid biopsies and tumor biopsies in the subset of 10
patients with FoundationOne data (Fig. 5.3A). Patient 7 had no detectable mutations in either
panel. Among the other 9 patients, 3 had concordant mutations — all in TP53 — detected in both
tumor biopsy and liquid biopsy, and all other mutations were unique to either the tumor biopsy or
the liquid biopsy. For the 3 patients with concordant mutations, the tumor and liquid biopsy
samples were collected concurrently in patient 3, 5 months apart in patient 4, and 38 months apart
in patient 10. Similarly, CNV profiles were compared between liquid biopsies and tumor biopsies
in the subset of 6 patients with FoundationOne data (Fig. 5.3A). Given the large number of
potential genes assessed by whole genome amplification/low-pass (WGA/low-pass) sequencing
(entire genome) and by FoundationOne (>300 genes), we focused on a subset of 58 prostate
cancer–relevant genes curated from recently published prostate cancer genomic profiling
studies161-164 for these comparisons (Table 5.3). Using this gene panel to compare CNVs from
tumor biopsies and CTCs, we detected both shared and unique amplifications and deletions (Fig.
5.3A).

85
Table 5.3. Gene Panel for CNV analysis
Table 5.4. Gene panel for SSNV analysis.

86
Figure 5.3. Distribution of genomic alterations by tissue source within individual patients.
(A) Detection of SSNVs and CNVs in a patient’s solid or liquid biopsy or in both. Analysis includes
only alterations tested in both solid and liquid panels. Boxed numbers denote months elapsed
between solid and liquid biopsies. (B) Detection of SSNVs in a patient’s CTC DNA or cfDNA or in
both (18 patients analyzed). Bx, biopsy.

87
5.4 Comparison of CTC DNA versus matched plasma.
Mutation profiles were compared between CTC DNA and matched cfDNA fractions enriched from
the same blood tube in the subset of 18 patients with available matched data (Fig. 5.3B). We
detected alterations unique to cfDNA (65.5%), unique to CTC DNA (20.7%), and shared in both
(13.8%). For example, no PIK3CA mutations were detected in CTC DNA whereas 6 alterations
were found in matched cfDNA samples.
5.5 Comparison of single CTCs from the same sample.
CNV profiles were generated from multiple single CTCs recovered from 6 patients. For each,
individual single-CTC CNV profiles were plotted and compared using the prostate cancer–
relevant gene list described earlier (Table 5.3, Fig. 5.4A-E). For example, patient 20 is a 65-yearold man with lymph node and bone metastases who was progressing on abiraterone and had a
PSA of 82 ng/ml at the time of the liquid biopsy draw. The patient also had a concurrent biopsy of
a bony metastasis analyzed with FoundationOne testing. His CTC count by CellSearch was
31/7.5 ml, and 2 of these cells were recovered and further analyzed for CNV analysis relative to
a single WBC from the same sample (Fig. 5.5). As would be expected, the WBC-derived germline
DNA had no detectable CNVs in the cancer-related genes interrogated. In contrast, there were
multiple CNVs detected in the CTC-derived somatic samples. Deletion of exon 1 and exon 2 in
CHD1 was identified in the bone metastasis and in both CTCs, as a homozygous loss in CTC 1
and a heterozygous loss in CTC 2. The bone metastasis amplification in CCNE1 was not mirrored
in the CTCs because no call could be made at the gene locus. In addition, AR amplification was
detected in the bone lesion and in CTC 1, which contained 9 copies of AR, but not in CTC 2.
Additional CNVs were identified in the CTCs that were not detected in the bone tissue, including
copy number gains in BRCA1, MYC, PCA3, PIK3CA, TERC, and TP53 and copy number losses
in BRCA2, PDL1, PTEN, and RB1. The number of shared versus distinct CNVs detected in
individual CTCs from a single blood draw was analyzed for the 6 patients with CTC CNV data

88
(Fig. 5.5B). Shared CNVs were observed between individual CTCs in 5 of the patients, but the
majority of CNVs detected in these patients were unique to one of their CTCs.

89
A B

90
C

91
Figure 5.4. CTC CNV Summary Data for Patients 2, 3, 4, 7, 15-T2.
Chromosome spread with amplifications (red) and losses (blue) for individual CTCs and control
germline WBC. Table outlines CNVs for genes queried in our panel (Table 3.1) in solid biopsy,
CTCs, and WBC.
D E

92
Figure 5.5. CNV distribution in individual CTCs.
(A) CNVs identified in single CTCs, a WBC, and a bone metastasis biopsy obtained concurrently
in patient 20 (copy number in parentheses). (B) CNVs shared by 1, 2, or 3 CTCs in a patient’s
sample (number of CTCs analyzed in parentheses).

93
Discussion
Molecular analysis of liquid biopsies has become an important adjunct to more traditional
tumor biopsies, and in fact blood-based biomarkers are emerging as stand-alone assays that
provide unique information. CellSearch CTC enumeration was the first blood-based biomarker in
prostate cancer with prognostic value, where high or increasing CTC counts were associated with
worse outcomes and shorter overall survival 165-167. More recently, commercial assays such as
Guardant360 (Guardant Health) and FoundationACT (Foundation Medicine/Roche) have
emerged as circulating tumor DNA NGS-based liquid biopsies used to interrogate potentially
actionable genes for alterations relevant to risk stratification and treatment selection in specific
cancer settings 168,169. However, most liquid biopsy strategies have been developed, analytically
validated, and clinically tested in a manner that uses a single blood analyte (i.e., only circulating
tumor DNA). Thus, although CTCs and CTC-derived nucleic acids can provide clinically relevant
information, and plasma-derived cell-free nucleic acids (DNA and RNA) also yield valuable
molecular data, little is known about the optimal way to leverage and integrate this multisourced
information: To what degree can 1 assay corroborate the results of another or provide additional
unique or complementary information, and how do these data compare with the results of more
traditional tumor molecular profiling?
Here, we demonstrate the feasibility of a multiparametric approach that integrates and
simplifies several workflows using 3 blood tubes simultaneously processed for CTC count, SSNVs
derived from CTCs and cfDNA, CNVs from single CTCs, and AR expression profiles from cfRNA.
The liquid biopsy assays chosen for this study were meant to encompass a broad array of
analytically validated assays with well-developed CLIA workflows to generate as many unique
blood-based phenotypes as possible from each sample (CTC mutations and CNVs, cfDNA
mutations, and cfRNA transcripts). Our goal was to leverage our unique access to these
technologies and patient samples to integrate and compare these disparate approaches. We
aimed to test whether multiple liquid biopsy assays can be applied effectively to each sample to

94
analyze and track molecular disease profiles as they evolve. This approach presented the
challenge of integrating and streamlining workflows that had been developed independently and
required different types, amounts, and purities of starting material.
One of the informatics challenges encountered unexpectedly in this pilot was how best to
convey the large amounts of orthogonal data generated from each sample in a manner that was
clear, was intuitive, and reflected similarities and differences in alterations. To do this, we
generated pictorial profiles (Fig. 2) that group the multiparametric results by source (plasma,
whole blood, solid tissue), biopsy type (cfRNA, cfDNA, CTC DNA, FoundationOne), and assay
(AR expression, SSNVs, and CNVs), with color-coding to highlight shared versus unique
alterations. Whereas at this early stage, all the information is presented, one envisions an
evolution toward more streamlined reports as the clinical utility of certain assays or of shared
alterations is prospectively validated. The presence of specific alterations or transcripts may prove
prognostic or predictive individually or as group, guiding treatment decisions. For example, patient
3 has a high CTC count of 168/7.5 ml by CellSearch, associated with poor prognosis. At this
stage, this patient does not have detectable ARV7, suggesting he may still respond to ARtargeting therapy. At the same time, he exhibits losses in RB1 and TP53. Cooperative losses of
2 or more tumor suppressor genes (RB1, TP53, PTEN) have been linked to an aggressive variant
of castrate-resistant prostate cancer that is less susceptible to hormonal therapy and more
responsive to platinum-based treatment 170,171. Furthermore, the CTCs from this patient have
a BRCA2 loss, an alteration that has been associated with response to therapeutic PARP
inhibition 172. Although these genomic signatures are still being clinically validated in large
prospective trials, their detection in liquid biopsies such as these may in time help guide
appropriate therapy.
This pilot was performed on men with advanced disease because they would be more
likely to have detectable CTCs and genomic alterations and would also be more likely to progress
on treatment, allowing for repeat liquid biopsy during the study period. The cohort’s heterogeneity

95
in age, PSA levels, and presence of visceral metastases reflects a diversity of advanced disease
typical of a tertiary care center. As expected in this setting, patients were heavily pretreated with
an average of 3 lines of therapy consisting of standard as well as investigational agents, and 60%
of patients had visceral disease compared with 20% in patients participating in first-line studies
for mCRPC 173,174. The disease severity and extent of pretreatment in this cohort likely affects
tumor molecular profiles, because previous studies have demonstrated that heavily pretreated
patients with advanced disease are more likely to have alterations
in AR, ERG, TP53, RB1, SPOP, CHD1, and BZTB16 as well as copy number gains and losses
that emerge over time as clonal resistance adaptations 156,161.
We detected CTCs in 75% of patients (median: 20/7.5 ml, range: 1–692/7.5 ml) reflecting
a broad range also seen in prior larger cohort studies evaluating CTC counts as prognostic
markers 166,167,175.These same studies demonstrated that increasing numbers of CTCs from
baseline were associated with shorter overall survival and that changes in CTC count changes
may be an indicator of response to treatment and improved survival, specifically conversion from
greater than or equal to 5 to less than or equal to 4 CTCs or from greater than or equal to 1 to 0
CTCs at week 13 165,166,176. In our cohort, 2 patients underwent such conversion, patient 9
(favorable, from 6 to 2 CTCs) and patient 10 (unfavorable, from 3 to 11), mirrored by a fall and
rise in PSA, respectively. When matched CTC DNA and cfDNA were analyzed at a second time
point in 8 patients, these revealed both shared and distinct SSNVs at disease progression,
reflecting the potential emergence of resistant subclones. These findings are consistent with
several recent large-scale genomic profiling studies in localized and metastatic prostate cancer,
which identified alterations in TP53, RB1, PTEN, AR, FOXA1, MYC, ERG, PI3K, and WNT that
emerged with progression from localized disease to metastatic castrate-resistant disease and
increased in frequency with exposure to hormonal therapies155,157,158,161,175,177.
Ninety-five percent of our patients had a detectable cfRNA AR transcript, as expected
from mCRPC patients for whom persistent AR signaling is a major driver of disease progression

96
178. AR has been extensively studied in liquid biopsies as a potential biomarker in late-stage
studies. In metastatic disease, quantitative PCR detection of ARV7, a constitutively active
truncated AR splice variant lacking the ligand-binding domain, in CTCs was associated with
resistance to hormonal therapies, such as enzalutamide and abiraterone, but not chemotherapy,
such as docetaxel and cabazitaxel37,179,180. Subsequently, a variety of other liquid
biopsy ARV7 approaches were tested: ARV7 was detected using an immunofluorescent protein
staining assay, which showed that patients with nuclear ARV7+ CTCs had poor response to
second-generation antiandrogens and better overall survival with taxane chemotherapy 38. In
other studies, ARV7 was also detected in whole-blood RNA (PAXgene tube) and was associated
with poor prognosis 181-183. As demonstrated by this broad spectrum of assays, ARV7 analysis
continues to evolve rapidly, and recent data in fact suggest an imperfect concordance
between ARV7 positivity and resistance to AR-targeted therapy 184,185. In our pilot, we
evaluated AR and ARV7 transcripts in cfRNA and found that 25% of patients had
detectable ARV7 (ARV7/AR ratio range: 0.5%–33.2%). All the ARV7+ patients had been treated
with abiraterone at some point in their treatment regimen. However, the 2 patients who converted
to being ARV7+ at a second time point were not progressing on second-line hormonal therapies
but rather on PARP inhibition (patient 10) and cabazitaxel (patient 18), perhaps reflecting
expansion of ARV7+ subclones that had not been previously detectable.
The central aim of this study was to test the feasibility of integrating multiple liquid biopsy
workflows rather than to compare detection rates of blood-based versus tumor-based assays.
However, because such comparisons are often made across tissues, it is important to recognize
the biological and methodological factors that contribute to the profiles generated using these
approaches. When possible, we normalized such differences by including only those genes that
were queried by both assays (Tables 5.3 and 5.4). Nevertheless, some differences in detected
alterations may have been attributable to methodological differences between liquid biopsy–
derived versus solid tumor–derived assays, such as DNA starting amounts, allele frequencies,

97
and sequencing approaches. For example, whereas we used an amplicon-based enrichment
assay for the liquid biopsies and called only mutations recorded in Catalogue Of Somatic
Mutations in Cancer (COSMIC) using ×200 read depth, FoundationOne uses a probe-based
enrichment assay on the solid tissue samples using median depth coverage of greater than ×500,
and it calls additional mutations not identified in COSMIC. Beyond technical considerations,
differences in liquid versus tumor biopsies may also reflect true biological distinctions, such as
clonal evolution as disease progresses to new sites or develops resistance between the time of
the tumor biopsy and the liquid biopsy.
In contrast, it was more feasible to compare SSNV profiles between CTC DNA and
matched plasma cfDNA because the CTC fractions and plasma fractions were obtained
concurrently from the same collection tube and analyzed using the same AmpliSeq workflow.
Despite these similarities, cfDNA and enriched CTC DNA do present important differences:
Plasma offers abundant starting material but usually a relatively low allele frequency, especially
with lower volume disease 175; conversely, CTC DNA is less abundant but may have greater allele
frequency if the CTCs are highly enriched and make up a significant portion of cells used to extract
DNA. In this cohort, we detected no PIK3CA mutations in CTC DNA whereas we found 6
alterations in matched cfDNA samples. AR and PIK3CA alterations have been detected
previously in cfDNA of mCRPC patients progressing on enzalutamide186. Consistent with this, in
our study 5 of the 6 patients with cfDNA PIK3CA mutations had been treated with second-line
hormonal therapy — abiraterone, enzalutamide, or both.
We performed single-cell CNV analysis on individual CTCs recovered from a subset of
patient samples. Rare single-cell recovery and analysis from a standard tube of blood is
recognized as technically challenging. In this cohort, 7 of 23 patient samples met our
predesignated criteria of having a minimum of 5 CTCs by CellSearch as well as passing WGA
quality control requirements. The CNV profiles generated from single CTCs (Fig. 5.4A-E) were in
some cases highly concordant, with modest differences between cells likely reflecting technical

98
variability, such as amplification bias. In other cases, the inter-CTC differences far exceeded any
expected assay variability and more likely represented biologically distinct tumor subclones.
Ultimately, single-CTC CNV analysis may provide a valuable high-resolution snapshot of
advanced disease, but further studies characterizing larger numbers of single cells are needed to
clinically validate the significance of these profiles.
Conclusion
This pilot study is to our knowledge the first to integrate several liquid biopsy assays into a
multiparametric tumor profile that can be repeated over time, demonstrating the feasibility and
potential utility of this approach. The abundant CTC and cell-free DNA and RNA data generated
from each sample comprise shared as well as unique cancer-specific alterations, which together
produce a high-resolution snapshot of tumor biology on treatment and at progression. Though
further refinement and validation in large prospective studies are necessary, this multiparametric
liquid biopsy strategy may ultimately become a key instrument for minimally invasive yet
comprehensive monitoring of disease phenotypes over time, helping better guide therapy. TO do
so, more studies need to be conducted to validate specific analytes as prognostic and predictive
markers in prostate cancer.

99
Chapter 6: Materials & Methods
Bladder Cancer Cell Culture
Human bladder cancer cell lines, T24 and UM-UC-3, were cultured in RPMI 1640 and DMEM
(Mediatech Inc., Manassas, VA), respectively, supplemented with 10% heat-inactivated fetal
bovine serum (Omega) and 1% penicillin/streptomycin (100 units/mL, Invitrogen), at 37°C and 5%
CO2. Prior to conducting all experiments, the cell lines were authenticated using 9-marker short
tandem repeat (STR) profiling and testing interspecies and mycoplasma contamination
(CellCheck 9 Plus, IDEXX BioAnalytics, Columbia, MO). We maintain stringent, good cell culture
practice (GCCP), and keep extensive records of cell lines’ culture and harvesting condition,
including passage count and cell density. We performed all experiments below passage 20, after
which cultures were restarted using thawed cells from earlier passages. T24R2 cell lines were
generated by serial desensitization of T24 cells with cisplatin and were a generous gift from the
Byun Lab (Seoul National University)105. T24R2 cell lines were maintained in RPMI 1640 media
with 6.6µM cisplatin (Catalog# NSC 119875, SelleckChem, Houston, TX).
Bladder Cancer Organoid Culture
CK-9151 is a patient-derived bladder cancer organoid acquired from the NCI PDMR repository
(Model 772611-094-R-V3-organoid, Sample Number CK9151). CTC-1044 is a patient-derived
bladder cancer organoid generated in our laboratory. Through an IRB-approved protocol and
under informed consent, we received excess tissue not needed for diagnostic purposes from
transurethral resection of bladder tumor (TURBT) from patients with transitional cell carcinoma of
the bladder undergoing treatment at the Keck School of Medicine of USC. Cells were treated with
Dispase II (1.5mg/mL, Catalog# 17105-041, ThermoFisher) in DMEM/F12 (Catalog# 12634028,
ThermoFisher) for 2 hours at 37°C to break down the basement membrane. Cells were then
collected, pelleted by centrifugation at 300g for 5 minutes and resuspended in 35uL BME
(Catalog# 3533-001-02, Cultrex) per well. The plate was inverted and incubated at 37°C for 15

100
minutes. Once the BME solidified, human bladder organoid media (BOM) was added based on a
modified 6A Media Recipe provided by the NCI PDMR with the following concentrations:
Advanced DMEM/F12 (1X) (Invitrogen, Cat#: 12634-028), HEPES (10mM) (Invitrogen, Cat#
15630080), GlutaMax Supplement (1X) (Life Technologies, Cat#: 35050061), Primocin (0.1
mg/mL) (InvivoGen, Cat#: Ant-pm-2), L-WRN Conditioned Media (50%) (Sigma-Aldrich, Cat#
SCM105), N-acetylcysteine (1.25 mM) (Sigma, Cat# A9165-5G), Nicotinamide (10 mM)(Sigma,
Cat# N0636-100G), B-27 Supplement (1X) (Life Technologies, Cat# 17504044), N-2 Supplement
(1X)( Life Technologies, Cat# 17502048), Y-27632 dihydrochloride (10μM) (Tocris, Cat# 1254),
FGF10 (100 ng/mL) (Peprotech Cat# 100-26), FGF7 (25 ng/mL) (Peprotech, Cat# 100-19), FGF2
(12.5 ng/mL) (Peprotech, Cat# 100-18B) and A83-01 (5μM) (Tocris, Cat# 2939). The media was
filtered using a 0.22 µm filter and prepared fresh every three weeks. Organoids were used 7-10
days after passaging and were passaged every 21 days.
Prostate Cancer Cell Culture
Human prostate cancer cell lines, C4-2B and LNCaP, were cultured in RPMI 1640 and DMEM
(Mediatech Inc., Manassas, VA), respectively, supplemented with 10% heat-inactivated fetal
bovine serum (Omega) and 1% penicillin/streptomycin (100 units/mL, Invitrogen), at 37°C and 5%
CO2. Prior to conducting all experiments, the cell lines were authenticated using 9-marker short
tandem repeat (STR) profiling and testing interspecies and mycoplasma contamination
(CellCheck 9 Plus, IDEXX BioAnalytics, Columbia, MO). We maintain stringent, good cell culture
practice (GCCP), and keep extensive records of cell lines’ culture and harvesting condition,
including passage count and cell density. We performed all experiments below passage 20, after
which cultures were restarted using thawed cells from earlier passages. MDV-R cell lines were
generated by serial desensitization of C4-2B cells with enzalutamide and were a generous gift
from the Gao Lab (UC Davis)187. MDV-R cell lines were maintained in RPMI 1640 media with 5µM
enzalutamide (Catalog#MDV3100, SelleckChem, Houston, TX).

101
RNA Isolation
Total RNA was extracted from cell lines using Trizol reagent and Direct-zol RNA extraction kit
(R2071, Zymo Research), which included treatment with DNase I for 20min at 37 °C. Total RNA
was extracted from organoids using RNEasy Micro Kit (Catalog# 74004, Qiagen, Germany). The
concentration of total RNA was measured by Qubit RNA HS Assay Kit (Catalog# Q32855, Thermo
Fisher Scientific) or via nanodrop.
RNA Fragmentation
Total RNA was chemically fragmented into ~200nt fragments as previously described111. Briefly,
3-5 µg of purified RNA was adjusted to a volume of 18µL with RNAse-free water. 2µL of 10X RNA
Fragmentation Buffer (100 mM Tris-HCl, 100 mM ZnCl2 in nuclease free H2O) was added and
incubated in a preheated thermal cycler for 4 ~ 5 min at 70°C. The reaction was stopped by adding
2µL of 0.5M EDTA. The following were then added to the mixture: 178µL of H2O, 20µL of sodium
acetate (3 M, pH 5.2, S7899, Sigma-Aldrich, St. Louis, MO), 14.4µL of glycogen (5 mg/ml,
Catalog#AM9510, Thermo Fisher Scientific) and 500µL of 100% ethanol and incubated at −80°C
overnight. Fragmented RNA was pelleted by centrifuge (30min at 12,000g at 4°C), washed with
75% ethanol and resuspended in RNAse-free water (10µL H2O per 1 µg human total RNA). Size
distribution was assessed using RNA 6000 Pico Kit on BioAnalyzer (Catalog# 50671513, Agilent
Technologies, Santa Clara, CA).
MeRIP
MeRIP was performed as previously published111. Briefly, 30µL of protein-A magnetic beads
(Catalog# 10002D, Thermo Fisher Scientific) and 30µL of Protein-G magnetic beads (Catalog#
10004D, Thermo Fisher Scientific) were washed with 500μL IP buffer (150 mM NaCl, 10 mM TrisHCl, pH 7.5, 0.1% IGEPAL CA-630 in nuclease free H2O) and resuspended in 500μL of IP buffer,

102
and tumbled with 5 μg anti-m6
A antibody (Catalog# E1610, NEB, Ipswich, MA) at 4°C overnight.
The bead-antibody mixture was washed with 500μL IP buffer and resuspended in 500μL IP buffer
containing the fragmented RNA, 100μL of 5X IP buffer and 5μL RNasin Plus RNAse Inhibitor
(Catalog# N2611, Promega, Madison, WI), and incubated for 2hours at 4°C.
The RNA reaction mixture was washed with 1000μL IP buffer, 1000μL low-salt IP buffer (50 mM
NaCl, 10 mM Tris-HCl, pH 7.5, 0.1% IGEPAL CA-630 in nuclease free H2O), and high-salt IP
buffer (500 mM NaCl, 10 mM Tris-HCl, pH 7.5, 0.1% IGEPAL CA-630 in nuclease free H2O) for
10min each at 4°C. After the washes, the m6
A-enriched fragmented RNA was eluted from the
beads in 200μL of RLT Buffer supplied in the RNeasy Micro Kit (Catalog# 74004, Qiagen,
Germany) for 2 min at room temperature. Magnetic separation rack (Catalog# 1231D, Thermo
Fisher Scientific) was applied to pull beads to the side of the tube. Supernatant was collected to
a new tube and combined with 400μL 100% ethanol. The mixture was transferred to an RNeasy
MicroElute spin column (RNeasy Micro Kit) and centrifuged at >12,000rpm at 4°C for 1min. The
column membrane was washed with 500μL RPE Buffer (RNeasy Micro Kit) once, and with 500μL
80% ethanol once. The column was centrifuged at full speed for 5min at 4°C to remove residual
ethanol. The m6
A-enriched RNA was eluted with 14μL nuclease-free water.
RT-qPCR and MeRIP-qPCR
cDNA synthesis was performed using qScript cDNA SuperMix (Catalog# 95408-500, QuantaBio,
Beverly, MA). Real Time PCR was performed using PerfeCTa SYBR Green FastMix (Catalog#
95071-250, QuantaBio) using a BioRad CFX96 Real Time PCR Detection System. To determine
the expression percentage of a target gene in RIP sample relative to the input control sample:
%Input=2^(Ct of target gene in Input Ctrl – Ct target gene in RIP). The relative gene expression
of various target genes after altering global m6
A (using genetic and pharmacologic modulation)
was calculated using GAPDH as a housekeeping gene. Relative Expression = 2^(Ct of GAPDH

103
– Ct of target gene). Primer sequences for bladder cancer targets can be found in Table 5.1.
Primer sequences for prostate cancer targets can be found in Table 5.5. All experiments were
performed in biological triplicates, and additional technical triplicates were used for all RT-qPCR
experiments.
Library Preparation and Sequencing
RNA-seq libraries were prepared using NEBNExt Ultra II RNA Library Prep Kit for Illumina
(Catalog# E7420, NEB) according to manufacturer’s protocol. MeRIP libraries were prepared
using SMARTer Stranded Total RNA-Seq Kit v2- Pico Input Mammalian (Catalog # 634413,
Takara-Clontech, Japan) according to manufacturer’s instructions. Briefly, 3.5uL of the 14uL
eluted RNA and 50ng input RNA were used for library construction omitting the fragmentation
step. Libraries for IP RNA and input RNA were PCR amplified 16 and 12 cycles, respectively.
Purified libraries were quantified using Qubit dsDNA High Sensitivity kit (Catalog# Q32851,
ThermoFisher) using a Qubit fluorometer, and the size distribution was checked by BioAnalyzer
(Agilent Technology) using the Agilent High Sensitivity DNA kit (Catalog# 5067-4626, Agilent).
The samples were then sequenced using a Novaseq PE 150 (Illumina, San Diego, CA). Adapter
sequences were removed, and sequences were demultiplexed using the bcl2fastq software
(Illumina).
RNA-seq Analysis
RNA-seq reads were aligned to human genome hg38 with reference annotation GENCODE v39
and counted using STAR (version 2.7.0)188,189. Only uniquely mapped reads without duplicates
were selected using samtools (version 1.10)190. Read counts were assigned to genes using
Subread featureCounts191. Read counts were normalized using DESeq2 package in R (version
4.1.3). To generate more accurate log2fold change estimates, shrinkage of the LFC estimates
towards zero was applied using DESeq2192. Differentially expressed transcripts with absolute

104
|log2FC| > 0.5 and adjusted-p value < 0.05 were retained. Gene set enrichment analysis and
Gene Ontology analysis were implemented and visualized using clusterProfiler package193-195.
Differential m6
A Analysis
MeRIP-seq reads were aligned to human genome hg38 with reference annotation GENCODE
v39, and counted using STAR (version 2.7.0)188,189. Uniquely mapped reads without duplicates
were selected using samtools (version 1.10)190. IP over input peaks were called using MACS2
callpeak using the parameters “-nomodel -extsize100 -gsize300e6”196. Differential m6
A analysis
was performed using DEQ package in R as previously published70. Briefly, DEQ runs statistical
analysis using DESeq2, edgeR and QNB packages70,192,197,198. Results that were statistically
significant with adjusted p-value < 0.05 using all three packages were considered significant.
Gene and peak expression changes were estimated as log2 fold changes from DESeq2.
Additional filtering to mitigate the confounding effect of differential expression on determining
differential methylation was applied using |peak IP log2FC- gene input log2FC| ≥ 1. Peaks with
fewer than 10 read counts were also removed to mitigate the confounding effect of differential
expression.
siRNA Knockdown
DsiRNAs targeting transcripts of interest and corresponding negative control were
purchased from Integrated DNA Technologies (IDT, Coralville, IA). DsiRNA sequences,
concentrations and timepoints for bladder cancer targets and prostate cancer targets used can
be found in Table 5.2 and Table 5.6, respectively.
Bladder Cancer cell lines: T24, T24R2 and UM-UC-3 cells, transfection was performed 24 hours
after seeding. Lipofectamine 3000 (Catalog# L3000015, Invitrogen, Waltham, MA) and OPTIMEM I reagents (Catalog# 31985062, ThermoFisher) were used to transfect cells at 70%

105
confluence according to manufacturer’s protocol. Cells were harvested and total RNA was
extracted at 24 or 48 hours depending on the transcript. Knockdown efficiency was evaluated by
RT-qPCR and Western blot.
Bladder cancer Patient derived organoids: Transfections were performed 10 days after passaging
the organoids using Lipofectamine RNAiMAX (Catalog#13778030, Invitrogen) and OPTI-MEM I
reagents. For each well, 45uL RNAiMax and 30uL OPTI-MEM I were incubated for 5 minutes
before adding the corresponding siRNA. The siRNA complexes were then incubated for 20
minutes before making up the volume to 450uL using bladder organoid media (BOM) without
antibiotics and supplemented with 10% FBS. Organoids were harvested and total RNA was
extracted 72 hours after transfection.
Prostate Cancer Cell Lines: C4-2B and LNCaP cells, transfection was performed 24 hours after
seeding. DharmaFECT #3 (Catalog# T-2003-02, Horizon Discovery, Waterbeach, UK) and OPTIMEM I reagents (Catalog# 31985062, ThermoFisher) were used to transfect cells at 70%
confluence according to manufacturer’s protocol. Cells were harvested and total RNA was
extracted at 24 or 48 hours depending on the transcript. Knockdown efficiency was evaluated by
RT-qPCR and Western blot.
Cisplatin Resistance Assay
Cancer cell lines: T24, T24R2 or UM-UC-3 cells were seeded and transfected the following day
with corresponding siRNA or negative control for 24-48hours prior to treating the cells with 10uM
cisplatin (Catalog# S1166, Selleck Chemicals) for 48hours. Cisplatin was dissolved in 154mM
NaCl saline and stored in single use aliquots at -20°C. After 48hours, MTS proliferation assay
was performed using CellTiter 96 AQueous One Solution Cell Proliferation Assay
(Catalog#G3582, Promega) according to manufacturer’s protocol. Briefly, 20uL of MTS solution

106
was added to each well and incubated at 37°C for 2hours, after which absorbance at 490nm was
recorded using a 96-well plate reader. Cell viability was calculated using: Percentage Viability=
(Absorbance[Sample]/Absorbance[NC - Cisplatin])x100%. Coefficient of drug interaction (CDI) to
evaluate synergy was calculated using: CDI= Percentage Viability[Condition +
Cisplatin]/(Percentage Viability [ Condition – Cisplatin] x Percentage Viability [ NC – Cisplatin]).
CDI < 1.0 indicates synergy and CDI < 0.7 indicates significant synergy199. Experiments were
repeated in 3-4 biological replicates, using 6 technical replicates.
Patient derived organoids: CK9151 or CTC-1044 were passaged into 24-well plates and treated
7-10 days with 50uM cisplatin for 48 hours alone or transfected with siSLC7A11 for 48 hours
followed by treatment with 50uM cisplatin for 48 hours. After the treatment, organoids were
digested with Dispase II (1.5mg/mL) for 2 hours at 37°C, followed by 15 minutes incubation with
TrypLE (Catalog#12-605-010, Gibco) to yield single cell resuspension. Cells were then counted
using Tryptan Blue Exclusion Assay using Nikon Eclipse TS100 Inverted Routine Microscope.
Cell counts were used to calculate cell survival and CDI to evaluate synergy. Experiments were
performed in three biological replicates, using three technical replicates.
Enzalutamide Resistance Assay
Cancer cell lines: C4-2B or LNCaP cells were seeded and transfected the following day
with corresponding siRNA or negative control for 24-48hours prior to treating reseeding cells with
5uM enzalutamide (Catalog# S1166, Selleck Chemicals) for 72 hours. Enzalutamide was
dissolved in DMSO and stored in single use aliquots at -80°C. MTS proliferation assay was
performed using CellTiter 96 AQueous One Solution Cell Proliferation Assay (Catalog#G3582,
Promega) according to manufacturer’s protocol. Briefly, 20uL of MTS solution was added to each
well and incubated at 37°C for 2hours, after which absorbance at 490nm was recorded using a
96-well plate reader. Cell viability was calculated using: Percentage Viability=

107
(Absorbance[Sample]/Absorbance[NC - Cisplatin])x100%. Coefficient of drug interaction (CDI) to
evaluate synergy was calculated using: CDI= Percentage Viability[Condition +
Cisplatin]/(Percentage Viability [ Condition – Cisplatin] x Percentage Viability [ NC – Cisplatin]).
CDI < 1.0 indicates synergy and CDI < 0.7 indicates significant synergy199. Experiments were
repeated in 3-4 biological replicates, using 6 technical replicates.
Actinomycin D Treatment
T24, T24R2 and UM-UC-3 cells were seeded with and without 10uM cisplatin. After 48 or 72
hours of treatment, cells were washed and collected at t=0 in RLT Buffer supplied in RNeasy
Micro Kit. The remainder of the wells were treated with 30uL of Actinomycin D (Catalog#
11805017, Gibco) at a concentration of 1mg/mL for a working concentration of 10ug/mL in 3mL
of culture media. Actinomycin D was dissolved to 1mg/mL in DMSO and stored in single-use
aliquots at -20°C. Samples were collected at t=2, 4, and 6 hours. RNA was extracted using the
RNeasy Micro Kit, followed by cDNA synthesis and RT-qPCR analysis of transcripts of interest to
calculate half-life of each transcript. Statistical significance was assessed by two-way ANOVA.
Lipid Peroxidation Detection
Lipid peroxidation was detected using C11-BODIPY (581/591) dye (Catalog# D3861, Fisher
Scientific). T24, T24R2 and UM-UC-3 cells were incubated with 5mM C11-BODIPY with and
without 10uM cisplatin treatment for 48 or 72 hours. All cells were collected, spun down, washed
and with and resuspended in HBSS with calcium and magnesium (Catalog# 14025092,
ThermoFisher), and analyzed using FACS on a BD SORP FACSYMPHONY S6 cell sorter. At
least 10,000 events were recorded. All samples were performed in technical triplicates and
experiments were repeated in three biological replicates.

108
Western Blotting
Western blots were performed using whole cell lysates extracted from human BC cells and BC
PDO’s using a RIPA lysis buffer (Catalog# R0278, Sigma-Aldrich, St. Louis, MO). Total protein
concentration was determined by Lowry Assay using the BSA Protein Assay Kit (Catalog#
5000002, Bio-Rad, Hercules, CA). 10-20ug protein lysate samples were boiled in loading buffer
for 10minutes before running on 4-20% Tris-Glycine gradient gels (Catalog# EC6021BOX,
Invitrogen) along with Blue Pre-stained Protein Marker (11-250kDA) (Catalog# 59329, Cell
Signaling Technology, Danvers, MA). Western transfer was performed using the iBlot Dry Blotting
System (ThermoFisher) and PVDF iBlot Transfer Stack (Catalog# IB401031, Invitrogen).
Membranes were blocked in Odyssey blocking buffer (Catalog#927-40000, LI-COR, Lincoln, NE)
and incubated with primary antibodies overnight at 4°C. After washes in 1X TBST + 0.10% Tween
20 Buffer, membranes were incubated in corresponding secondary antibodies for 45 minutes at
room temperature. After additional washes in 1X TBST + 0.10% Tween 20 Buffer, membranes
were visualized using an Odyssey DLx Imaging System (LI-COR). Digital Images were processed
and analyzed using ImageStudio Version 5.0 software (LI-COR) and ImageJ was used to quantify
protein expression. Primary and secondary antibodies and their corresponding concentrations
can be found in Table 5.3.
Immunofluorescence
Cells were seeded over microscopy coverslips. At collection time, cells were washed with 1X
PBS, fixed using 10% formalin for 15 minutes at 37°C, and washed with 1X PBS. Cells were then
permeabilized with 0.1% Triton-X-100 in 1X PBS for 15 minutes at room temperature, washed
with 1X PBS, blocked with 2% BSA in 1X PBS for 1 hour at room temperature, and incubated with
primary antibody overnight at 4°C. After 1X PBS washes, cells were incubated with secondary
antibody for 45 minutes at room temperature and washed with 1X PBS. Coverslips were mounted

109
on microscopy slides with VectaShield Antifade Mounting Medium with DAPI (Catalog# H-1200-
10, Vector Laboratories, Newark, CA) and sealed with nail polish.
Representative cells were selected and imaged on a Zeiss 800 Axio Imager.Z2 upright laser
scanning confocal microscope, using an EC Plan-Neofluar 40x/1.30 Oil lens. 1024 x 1024 pixel
images were acquired using GaAsP-PMT detectors. Samples were illuminated with 405 and
488nm lasers at 0.4% and 0.8%, respectively, with pinhole size set to 1AU. Alexa Fluor 488
acquisition parameters were: 505-545nm (detection wavelength range), 4.12μs (pixel time), and
650V (gain). For z-stack imaging, slices were acquired with 1μm intervals. Image analysis was
then performed uniformly using ImageJ. Primary and secondary antibodies can be found in Table
5.4.
Organoid Histology Staining
Organoids and tissue were fixed in 4% paraformaldehyde for 3h, dehydrated, and paraffinembedded according to standard histology procedures. Sections were stained with H&E and the
following antibodies: Keratin 5 (AF138 COVANCE 160P-100), Ki67 (Monosan MONX10283),
Keratin 20 (KS20.8 Dako M7019), TP63 (4A4 Abcam ab735), and Uroplakin III (AU1 Progen
651108)200, according to the manufacturer’s protocols. IHC procedure was performed by the USC
Immunohistochemistry Core Images were acquired using Zeiss Axio Observer.A1 and Axio
Imager.Z1 microscopes.
RNA Immunoprecipitation (RIP)
RIP assay was conducted using the Magna RIP Kit (Sigma Millipore, #17-701) according to
manufacturer’s protocol. Briefly, RIP lysate was incubated magnetic beads coupled with 5ug of
the antibody of interest (YTHDC2, YTHDF1-3) or corresponding Mouse IgG or Rabbit IgG
overnight at 4ºC. After 6 washes and proteinase K digestion, RNA was eluted from the beads as

110
previously described in the “MeRIP” section. Further analysis was performed using RT-qPCR
using the primers in Table 5.1.
Each RIP fraction was normalized to the input to account for RNA sample differences:
∆Ct[normalized RIP] = (Ct [RIP]- Log2(RIP Dilution Factor)) – (Ct[Input] – Log2(Input Dilution
Factor)), where RIP Dilution Factor = fraction of RIP used for cDNA synthesis or 8/14 and the
Input Dilution Factor = fraction of input RNA saved x fraction of input RNA used for cDNA
synthesis. % Input for each RIP fraction was calculated using: % Input = 2(-∆Ct[normalized RIP]).
Enrichment was calculated relative to background antibody (Mouse IgG or Rabbit IgG) using:
Fold Enrichment= 2(-∆∆Ct[RIP/IgG]) ∆∆Ct[RIP/IgG]= ∆Ct [normalized RIP] - ∆Ct [normalized IgG].
Finally, relative fold enrichment was calculated based on negative control transcript that does not
bind the RIP protein using: Relative Fold Enrichment = Fold Enrichment [Transcript of
Interest]/Fold Enrichment[Negative Control Transcript]. RIP efficiency was evaluated by Western
Blot of the input RIP lysate and a fraction of the final RIP wash in probing for the RIP protein of
interest.
Clinical Databases
TCGA analysis was performed using the TCGA PANCAN and TCGA BLCA datasets for
all cancer types and for BC. SLC7A11 mRNA expression was assessed for an association with
histologic grade, primary therapy outcome, overall survival, disease-specific survival, and
progression-free interval. Analysis was performed using UCSC Xena exploration tool201.
Statistical Analysis for Chapter 2-4

111
P-values were calculated using two-tailed Student’s t-test between two groups. One-way
ANOVA and two-way ANOVA were used to compare multiple groups. Data are presented as
means +/- SEMs from at least three independent experiments. Significant codes: ‘****’: p<
0.00001, ‘***’: p< 0.001, ‘**’: p<0.01, ‘*’: p <0.05. GraphPad Prism version 9, Excel version 16.68,
or R version 4.1.3 were used for statistical analysis.
Data Availability
The sequencing data have been deposited in the GEO database and can be accessed
by GSE231836. All other relevant data are within the paper.
Patient sample collection
The study was conducted with IRB approval at the University of Southern California Norris
Comprehensive Cancer Center (NCCC) between January 2017 and February 2018. After
obtaining informed consent, blood samples were obtained from twenty men with mCRPC
encountered in the outpatient oncology clinics. For each patient, a total of three 7.5mL blood
collection tubes (EDTA, Cell-Free RNA Streck, CellSave) were collected, processed, stored and
transported according to each manufacturers’ instructions for analysis. Specifically, the EDTA
tube was mixed with 10mL LB-Fix, Cynvenio’s blood stabilization solution, which was left to sit for
an hour prior to shipping at ambient temperature to Cynvenio Biosystems CLIA/C AP lab for
cfDNA and CTC DNA SSNV analysis. Cell-free RNA Streck tubes were immediately inverted a
total of 8 times prior to shipping at ambient temperature to Liquid Genomics for cfRNA relative
expression analysis. CTC enumeration using the CellSave tubes was performed by the
Circulating Tumor Cell/Liquid Biopsy Core at the University of Southern California (see below for
more details). For a subset of patients, second time point samples were collected at treatment
resistance, defined as new metastases, increased PSA, or clinical progression. Liquid biopsies

112
were compared to tumor molecular profiles from primary tumors or metastases when available
(e.g. Foundation One).
CTC enumeration by immunomagnetic enrichment
CTC enrichment and enumeration were performed on the FDA-cleared CellSearch platform
(Menarini Silicon Biosystems) according to standard manufacturer’s instructions. Briefly, 7.5ml
blood was drawn by standard peripheral venipuncture into a CellSave preservative tube (Menarini
Silicon Biosystems) and delivered at room temperature to the laboratory for processing on the
same day. CellSearch enriches and identifies candidate CTCs using antibodies to epithelial-cell
adhesion molecule (EpCAM) coupled with magnetic beads. After magnetic enrichment, isolated
cells are stained with fluorescent nucleic acid dye (DAPI) to identify nucleated cells. Recovered
cells are stained with fluorescent monoclonal antibodies to CD45 and CK 8, 18, 19 to distinguish
epithelial cells from leukocytes. Cells that are EpCAM+, CD45-, CK+ and DAPI+ and fulfill
morphologic criteria are counted as CTCs by a certified technician167.
Single CTC recovery by dielectric manipulation
CellSearch enriched patient samples were further-processed for DEPArray per manufacturer’s
instructions (Menarini Silicon Biosystems). Briefly, pre-coated (2% BSA) gel loading tips were
used to aspirate samples from each Veridex cartridge and two additional washes were performed
with SB 115 buffer (200μL) to improve recovery. Samples were then washed twice with 1000μL
and 900μL of SB 115 buffer before volume reduction to 13μL for loading onto a DEPArray
cartridge and into the DEPArray V2 System (Menarini Silicon Biosystems) for analysis. The
DEPArray system utilizes a non-uniform electric field to exert forces on neutral, polarizable
particles, such as cells, suspended in a liquid. Dielectrophoresis (DEP) is used to trap cells in
dielectric cages by creating an electric field above a subset of electrodes in counter phase with
the electric field of adjacent electrodes. After imaging, individual cells of interest are gently moved

113
with the DEP cages into holding chambers for isolation and recovery. Thus, single CTCs
(CK+DAPI+CD45-) along with control single WBCs (CK- /DAPI+/CD45+) were isolated and
recovered into individual MicroAmp reaction tube (Thermo Fisher). To facilitate downstream
analysis, cells were pelleted (14,100xg, 30s) before a wash step was performed with 100μL PBS
(14,100xg, 10min). Finally, volume reduction was performed by carefully aspirating the wash
buffer leaving a final volume of ~1μL. Samples were stored at -20C until needed as input for
Ampli1 WGA kit (see below for more details).
CTC copy number variant analysis
Ampli1TM whole genome amplification, DNA library construction and whole genome
sequencing
DNA of isolated cells was amplified using the Ampli1TM WGA kit according to manufacturer’s
instructions (Menarini Silicon Biosystems, Ampli1 QC Kit, September 2015 User Manual
V1.2.WGQC4). Quality of Ampli1TM WGA products was checked using Ampli1TM QC kit (Menarini
Silicon Biosystems) and only products with at least 3 amplified bands were retained. 5 μL of
Ampli1TM WGA product was transferred into a new tube and purified with 1.8X SPRIselect Beads
(Beckman Coulter) according to manufacturer instructions and eluted in 12.5 μL TE. The Ampli1TM
LowPass kit, (Menarini Silicon Biosystems) was used to prepare libraries for low-pass WGS
(Menarini Silicon Biosystems, Whole Genome Amplification for Single Cells, June 2015 User
Manual V3.WG001U). In brief, 10–50 ng of purified primary Ampli1TM WGA product was reamplified using hybrid PCR primers, including barcoded adaptors compatible with the Ion
TorrentTM Systems on the 5’ end, and primary WGA universal adaptor on the 3’ end. Barcoded
libraries were quantified using Qubit dsDNA HS Assay kit on a Qubit 2.0 Fluorometer (Thermo
Fisher Scientific) and pooled in equimolar concentrations to obtain a concentration of ~34ng/μL.
Pooled libraries were size selected (300–450 bp) using E- Gel SizeSelectTM Agarose Gels, 2%

114
on a E-Gel Agarose Gel Electrophoresis System (Thermo Fisher Scientific) according to
manufacturer’s instructions (Menarini Silicon Biosystems, Ampli1 QC Kit, September 2015 User
Manual V1.2.WGQC4). Size selected library pool was cleaned up with 1.2X SPRIselect Beads
(Beckman Coulter) according to manufacturer’s instructions and quantified using Agilent High
Sensitivity DNA Kit using the Agilent Bioanalyzer 2100 instrument (Agilent). Sequencing was
performed using the Ion Torrent PGM TM (Thermo Fisher Scientific, Ion AMpliSeq DNA and RNA
Library Preparation User Guide, Publication Number MAN0006735, Rev C.0). Libraries from
gDNAs (100ng) were prepared using Ion Xpress Plus gDNA Fragment Library preparation kit
(Thermo Fisher Scientific). Briefly, samples were fragmented for 200-base-read libraries, end
repaired, ligated with adaptors, nick repaired and bead purified prior to amplification of size
selected (E-Gel SizeSelect TM, Thermo Fisher Scientific) fragments around 250 bp long. Fragment
sizes were assessed using the Bioanalyzer system and quantified using the Ion Library TaqMan
Quantitation Kit (Thermo Fisher Scientific). Pooled libraries were used for emulsion PCR
amplification (200bp) using the Ion PGM TM System (Thermo Fisher Scientific).
Sequence alignment, read counting and normalization
Signal processing, base calling and alignment to the Homo sapiens hg19 reference sequence
was performed with the Torrent Suite TM v4.6 with—g 0 parameter for the alignment step with
tmap. Genome binning was performed using WindowMaker tool from BEDTOOLS suite202-204.
Read counting and assignment to genomic bins were performed using the HTSeq library202-204.
Reads spanning more than one bin were assigned to the one with the longest overlap. Read
counting and assignment to MseI fragments were performed by BEDTOOLS IntersectBed tool,
filtering out reads with more than one fragment match. GC-based normalization was performed
by LOWESS fitting of per-bin GC content versus read count on each bin. Calculation of bin
mappability value was performed using bigWigAverageOverBed
(http://hgdownload.cse.ucsc.edu/admin/exe/) using mappability track for 100mers produced by

115
Encode/CRG (wgEncodeCrgMapabilityAlign100mer; downloaded from
https://genome.ucsc.edu/).
Copy Number Alteration (CNA) calling
Control-FREEC (Control-Free Copy number caller) software was used to obtain copy-number
calls, using the mode without control sample
(http://boevalab.com/FREEC/index.html#downloads)
203. Read counts were corrected by GC
content and mappability (uniqMatch option). Bin size was manually set in order to match the
desired resolution. To determine significant CNA calls, Wilcoxon test and Kolmogorov-Smirnov
test (p value < 0.01) were performed using the script to assess significance202-204. Table 3.1 lists
the individual cancer genes interrogated.
Purification of cfDNA and Extraction of CTCs
Blood samples were fractionated within 96 hours of fixation, by centrifugation (500xg, 10min). The
upper layer of plasma was collected and cfDNA isolated using Qiagen’s Circulating Nucleic Acid
kit per manufacturer’s instructions. The remaining blood cell fraction was then processed as
described previously 5,6. Briefly, initial incubation with biotinylated antibodies against EpCAM,
TROP2, HER2, and PSMA (Cynvenio Biosystems) was followed by incubation with streptavidin
magnetic particles. The pre-processed blood was then loaded onto the LiquidBiopsy® platform
which performs automated immunomagnetic enrichment and immunofluorescent staining on a
microfluidic flow cell. Cells were stained with DAPI and fluorescent antibodies targeting
cytokeratin, CD45, PDL1 (Cynvenio Biosystems). Cells captured in the flow cell were imaged and
enumerated using the Ariol system (Leica Biosystems). Following image analysis, cells were
eluted from the microfluidic device and processed for sequencing. CTC samples are processed
for sequencing irrespective of enumeration results. CTC-DNA extraction was performed by re-

116
suspending the eluted cell pellet in 6.5uL LB Digest containing Proteinase K followed by
incubation at 55oC for 3 hours, then 70oC for 1 hour as reported previously205. The whole cell
lysate was then used for a multiplex PCR reaction with Cynvenio’s 27-gene ClearID panel (Table
3.2) followed by clean-up using AMPureXP beads.
Sequencing analysis for CTC-DNA and cfDNA
Sequencing libraries for cell-free DNA, CTC, and germline DNA were then prepared as described
previously using Cynvenio’s 27-gene panel in Table 3.2 205,206. Libraries were quantified using the
Taqman Library Quantification Kit (Life Technologies) and an ViiA7 qPCR machine (Applied
Biosystems). Germline, CTC-DNA and cfDNA libraries were the sequenced on an Ion Torrent
S5XL sequencer. Single-nucleotide variants were called using Everest Software (Cynvenio),
which identifies variants in CTC- and cf-DNA that are not present in germline samples (casecontrol sequencing). Variants expressed at ≥1% allelic frequency are reported. Only amplicons
with greater than 2000 reads are reported. At ~1% detection threshold, this requires 20 reads to
be informative.
RNA Extraction, RT-PCR, relative cfRNA expression
Whole blood (7.5 ml) was collected in cfRNA Streck tubes and fractionated by centrifugation
(300xg, 20 min) to isolate the upper plasma layer. The plasma layer was centrifuged (5000 xg, 10
min). cfRNA was extracted from 2mL of plasma with a proprietary in-house developed protocol
207. All nucleic acids were kept in bar-coded matrix storage tubes. RNA was stored either at -80°C
or at +4°C as complementary DNA (cDNA) via random-primed reverse-transcription.
Expression of AR and ARV7 were measured by quantitative real-time PCR to detect their
expression from cfRNA using appropriate gene-specific primers. Amplification was performed
using a proprietary method (Liquid Genomics) in a 10 μL reaction mix containing 2 μL cDNA, the
primer and probe207. β-actin was used as an internal control. Delta Ct (dCT) was calculated from

117
the Ct value using β-actin as the internal control. dCT was the Ct value of AR and ARV7
subtracted by the Ct value of β-actin. K was calculated to produce whole number for relative gene
expression control using RNA isolated from universal human reference RNA (UHR) as a positive
control for AR and an ARV7 synthetic fragment spiked into UHR as a positive control for ARV7.
Using this value of K, relative gene expression = (2^- dCT)*K.
Statistical Analysis for Chapter 5
Distribution of genomic alterations by tissue source Fig. 4.3 was provided as simple percentages
(number from each tissue as a portion of total alterations). All other quantities in Fig. 4.3 and Fig.
4.5 are described as direct counts of occurrences. Statistical cutoffs for variant calling are
described for each assay above.

118
Table 6.1. RT-qPCR and MeRIP-qPCR primers for bladder cancer targets
RT-qPCR Primers FWD REV
GAPDH TCA AGG CTG AGA ACG GGA
AG
GGA CTC CAC GAC GTA CTC
AG
SETD7 GGGGTTCAGAGACCTGGAAT GCATGGTGAGAGGATGTGAC
CLUC GCTTCAACATCACCGTCATTG CACAGAGGCCAGAGATCATTC
GLUC CGACATTCCTGAGATTCCTGG TTGAGCAGGTCAGAACACTG
METTL3 CCA CTG ATG CTG TGT CCA
TC
GGA GAC CTC GCT TTA CCT
CA
FTO ACT TGG CTC CCT TAT CTG
ACC
TGT GCA GTG TGA GAA AGG
CTT
ALKBH5 TTC TCT TCC TTG TCC ATC TC ATC CTC AGG AAG ACA AGA
TTA G
YTHDC2 AGA TAA GCA GGG ATG GGC
AG
AAG ACG TTC CCA TAA CTG
GAG C
YTHDF1 GCA CAC AAC CTC CAT CTT
CG
AAC TGG TTC GCC CTC ATT
GT
YTHDF2 ACT TGA GTC CAC AGG CAA
GG
AAG CAG CTT CAC CCA AAG
AA
YTHDF3 TGA CAA CAA ACC GGT TAC
CA
TGT TTC TAT TTC TCT CCC
TAC GC
SLC7A11 GCG TGG GCA TGT CTC TGA C GCT GGT AAT GGA CCA AAG
ACT TC
ANO9 CTC CGA GCA GTG GGA CTA
TG
AGT GCG GTA CAG GCC AAA
G
ERCC1 CCT TAT TCC GAT CTA CAC
AGA GC
TAT TCG GCG TAG GTC TGA
GGG
SERPINE1 GCA CCA CAG ACG CGA TCT T ACC TCT GAA AAG TCC ACT
TGC
PML CGC CCT GGA TAA CGT CTT
TTT
CTC GCA CTC AAA GCA CCA
GA
FABP5 TGA AGG AGC TAG GAG TGG
GAA
TGC ACC ATC TGT AAA GTT
GCA G
SLC37A4 AGG TAG CTC CTA CAT GAG
TGC
GGG TTC CCG TAG TTG GAC
AG
PLAU GCT TGT CCA AGA GTG CAT
GGT
CAG GGC TGG TTC TCG ATG
G
POLRMT CGC CAC ATC CAC CCT GTT C GGA CCA TCG AAA GGT GTC
TGG
FUT8 AAC TGG TTC AGC GGA GAA
TAA C
TGA GAT TCC AAG ATG AGT
GTT CG
ARHGAP45 CTC CTG TCC ATC TAC TCG
CTG
CCT GCG CTT CTC GTG TTC
A
FOSL1 CAG GCG GAG ACT GAC AAA
CTG
TCC TTC CGG GAT TTT GCA
GAT
PLA2R1 TGG AGT GGC AGG ATA AAG
GAA
AGG GTC AGA ACC GAT TTA
CCT
OSBP2 GAA CCT GTG TCC GAG ACG
AC
CCT GAG CTT GAC TCT GAC
CC

119
SBNO2 ACT CCC TGT CGG ACA TCG T GAA CAG CTT ATC GTG GGT
GGA
MeRIP-qPCR
Primers
FWD REV
SLC7A11 5’UTR m6
A
region
CAG CGC TAT AGT GTT CAC
AGG T
AGT AGT AAT TAG ATC GCT
GTG AAG G
ANO9 Exon m6
A
region 1
AGC AAA GAG TCC CGA GGA
GA
ACA GTC CTT CGT CGG TTG
TT
ANO9 Exon m6
A
region 2
CGT TGA TCC TGG GGA GGA
AGG
GTC TTT CCT ACC CGC TTC
TGT C
SERPINE1 3’UTR
m6
A region
GCC CCT CTT TTT CCC CTT
GAT
ACT CCG TCC TTT TGA TCC
CC
PML Exon m6
A
region
GCC AGG TGG TAG CTC ACG ACT GGC CAT CTC CTC GTA
GT
FABP5 5’UTR m6
A
region
CTT TCC CTC CCT GTC GCA
TC
GAT TTC TGC GGG AAA CTG
CG
SLC37A4 5’UTR m6
A
region
CTC CCT TTA TAG CCG CCT
TCT
CAG TTT GGC GCT CAG TAA
TCT C
ARHGAP45 3’UTR
m6
A region
GTA CAC AGT GGG GTC TCT
CG
GCC ACA GAA AAC ACC CGA
TT
PLA2R1 Exon m6
A
region
CCC CCG ACA ATA GTC TGT
CAT
CCC CAC CTA TGG AGG TAT
GT
OSBP2 Exon m6
A
region
TAA CGG CAC CTG AGG AGC
AT
TCC CCA GGT TGC TGA TCC
AT
Table 6.2. siRNA references and conditions for bladder cancer targets
Target Company Catalog# Concentration Time
Negative
Ctrl
IDT 51-01-14-03 10nM 24-48HRS
SLC7A11 IDT hs.Ri.SLC7A11.13.3 20nM 24HRS
ANO9 SCBT sc-96721 20nM 24HRS
SERPINE1 IDT hs.Ri.SERPINE1.13.2 10nM 24HRS
PML SCBT sc-36284 20nM 48HRS
FABP5 IDT hs.Ri.FABP5.13.2 10nM 24HRS
PLA2R1 IDT hs.Ri.PLA2R1.13.5 10nM 24HRS
OSBP2 IDT hs.Ri.OSBP2.13.1
hs.Ri.OSBP2.13.3
20nM/each 48HRS
YTHDF3 IDT hs.Ri.YTHDF3.13.1 10nM 48HRS

120
Table 6.3. Western Blot antibodies
Protein Target Size
(kDa)
Dilution Company Catalog# Lot#
GAPDH (Mouse) 37 1:5000 Invitrogen 437000 XD350279
β-ACTIN (Mouse) 45 1:5000 Cell Signaling
Technologies
8H10D10
#3700S
Lot 20
α-TUBULIN
(Mouse)
55 1:4000 Invitrogen 62204 2407525
SLC7A11 mAb for
WB (Rabbit)
35 1:1000 Cell Signaling
Technologies
D2M7A
#12691
Lot 5
YTHDC2 (Rabbit) 160 1:1000 Abcam ab176846 GR256347-46
YTHDF1 (Rabbit) 60 1:4000 Proteintech 17479-1-AP 00105213
YTHDF2 (Rabbit) 62 1:1000 Proteintech 24744-1-AP 00110531
YTHDF3 (Mouse) 64 1:500 SCBT sc-377119 A0422
Goat anti-Rabbit
IgG Secondary
Antibody
- 1:10,000 Invitrogen SA535571 VA296084
Goat anti-Mouse
IgG Secondary
Antibody
- 1:10,000 LI-COR 926-68070 D20316-15
Table 6.4. Immunofluorescence antibodies
Protein Target Size
(kDa)
Dilution Company Catalog# Lot#
SLC7A11 pAb for
IF (Rabbit)
- 1:500
1:1000
Invitrogen PA1-16893 XJ3707839
Goat anti-Rabbit
Alexa Fluor 488
- 1:1000 Invitrogen A-11008 2420731

121
Table 6.5. RT-qPCR and MeRIP-qPCR primers for prostate cancer targets
RT-qPCR Primers FWD REV
CENPN TGA ACT GAC AAC AAT CCT
GAA GG
CTT GCA CGC TTT TCC TCA
CAC
GRHL2 TCA ATA CCC GAA GAG CCT
ACA
CTT GGC TGT CAC TTG CTT
TGC
PARP10 CAG CTC TAC CAT GAG GAC
CTT
CGA AAG CCA GTC ATA TCC
GGT
FN1 CGG TGG CTG TCA GTC AAA G AAA CCT CGG CTT CCT CCA
TAA
THBS1 AGA CTC CGC ATC GCA AAG G TCA CCA CGT TGT TGT CAA
GGG
PRKCAB TTC AGC AAC TCA CAG TTA
AGC A
GGC ACA CTC ATA CTT CTG
ACC
PAK4 GGA CAT CAA GAG CGA CTC
GAT
CGA CCA GCG ACT TCC TTC
G
ATP2C2 CCA GAG CGT TTT GTG TGG
ACT
GGG GTT CTT AAA CTG ATC
CAG G
H2AFY CGG ATG CTG CGG TAC ATC
AA
CTC CGC TGT CAG GTA TTC
CAG
FLRT1 AGC GAG ATG GAC GAG TGT
TTT
GGG TAG TCA ATG TTG GAG
TCG
TICRR AAG CTA TCA GCG ATC TCG
GC
TGA ACC TGG CAA GGA GCA
AA
SYT4 ATG GGA TAC CCT ACA CCC
AAA T
TCC CGA GAG AGG AAT TAG
AAC TT
MeRIP-qPCR
Primers
FWD REV
THBS1 exon m6
A
region
CCC TGG CTT CTC ATA GCC
AA
TGC TGA GCA AGT CCA GTA
GC
PARP10 3’UTR m6
A
region
GAG TCC AGG GTT TGA GGG
AG
GTC ACT TGC TTC ACG GAG
GT
GRHL2 5’UTR m6
A
region 1
TCC CCT TTG GGC CTT GAT
AG
GTA TTC ATC CAG GAG CGG
GAG
GRHL2 5’UTR m6
A
region 2
CCC TTA GGA ATG GTC TCA
GCT C
ATC GGT GGG AGC AGC TAA
AA
PAK4 intron m6
A
region
CCT GCG TTA TGC TGG AGT
GT
TGG CAT TTG GCA GGA GCT
TA
SYT4 3’UTR m6
A
region
TTA ACT TCT GGC TGC CGT
GA
GTA TCA TGG GCC GTG GAA
CA
PRKACB exon m6
A
region
AGC CAA AGC CAA AGA AGA
CT
CTG AGT TGG ATT CTC CCA
TTT TT
CENPN 5’UTR m6
A
region
ACT TTG TTG TGC TGT TTT
TGT TTT G
TTG ATG AAC TCA GCA ACA
GTC TC
FN1 exon m6
A region CCA CTC ATC TCC AAC GGC
AT
GGC TTG AAC CAA CCT ACG
GA

122
Table 6.6. siRNA references and conditions for prostate cancer targets
Target Company Catalog# Concentration Time
Negative
Ctrl
IDT 51-01-14-03 10nM 24-48HRS
THBS1 IDT hs.Ri.THBS1.13.3 25nM 48HRS
PARP10 IDT hs.Ri.PARP10.13.1 25nM 48HRS
PAK4 IDT hs.Ri.PAK4.13.3 25nM 48HRS
SYT4 IDT hs.Ri.SYT4.13.3 25nM 48HRS
FN1 IDT hs.Ri.FN1.13.1 25nM 24HRS

123
Chapter 7: Conclusion
A major obstacle in effective treatment of advanced malignancies is the emergence of therapy
resistance, not only by selection of genetically resistant clones but also through phenotypic
plasticity, the ability of genetically identical cells in a population to switch states or phenotypes in
response to environmental conditions. This phenotypic plasticity serves as a form of “bet hedging”
or evolutionary strategy to maximize fitness in a dynamic environment. In overcoming drug
resistance driven by non-genetic factors, the goal is to lock cells into drug sensitive phenotypes
in order to get better response to treatment and avoid recurrence or relapse. To do so, it is
fundamental to understand the mechanisms involved in regulating phenotypic plasticity. Having
previously explored epigenetic factors such as DNA methylation, histone modifications and
nucleosome occupancy implicated in phenotypic plasticity by rewiring regulatory networks106,107,
we explore another possible contributing mechanism: m6
A RNA modifications. In this dissertation,
we focus on two genitourinary malignancies: bladder cancer driven by mutational-load and
environmental factors, and prostate cancer driven by steroid-hormones. Both malignancies have
strong characteristics that implicate a potential role for m6
A modifications: (i) cell fate in bladder
cancer plasticity and (ii) alternative splicing in prostate cancer drug resistance.
We started by validating a low input MeRIP-seq protocol111 by reproducing previously published
data using lung adenocarcinoma A549 cells. Having demonstrated our ability to generate
comparable results in our own laboratory, we sought to apply this pipeline to our BC ad PC models
of interest. In chapter 2, we present a cell line model of phenotypic plasticity and drug resistance
in bladder cancer established in our laboratory, in which cells interconvert cyclically in and out of
a drug-resistant tumorigenic state103,104. After using FACS to sort the two sub-populations, we
mapped the m6
A-epitranscriptome profiles of the two phenotypes and found no statistically
significant differences that could be validated. After further optimization of the bioinformatics
pipeline, we were still unable to detect any differences due to the significant heterogeneity of cell

124
subpopulations, limiting our ability to have concordant replicates within each group. In order to
overcome this technical challenge, we opted to change model to a well-established cell line model
of cisplatin resistance generated via long-term cisplatin exposure: T24 and T24R2 cells105. In
chapter 3, we compare these cisplatin-sensitive and cisplatin-resistant cells using a robust and
stringent informatics pipeline, and we found that the two have distinct m6
A profiles.
Using unbiased transcriptome-wide m6
A profiling and genetic studies in vitro, we demonstrate
that m6
A modifications regulate expression of clinically relevant gene transcripts in BC, and that
a subset of these modifications may promote resistance to chemotherapy. Specifically, SLC7A11
was identified as a driver of m6
A-regulated cisplatin resistance in bladder cancer, with decreased
5’UTR m6
A leading to decreased mRNA degradation and increased SLC7A11 expression in
cisplatin-resistant cells compared to cisplatin-sensitive cells. The same trend was witnessed with
short-term cisplatin treatment of cells and patient-derived organoids suggesting that m6
A is
implicated in regulating early changes that promote transition to drug resistance. Interestingly,
these changes were not “remembered” following discontinuation of short-term treatment,
highlighting the phenotypic plasticity of the cells. In clinical databases, we found that high
expression of SLC7A11 associated with lower overall survival, disease-specific survival and
progression free interval in all cancer types, and higher grade, progressive disease and lower
disease-specific survival in BC. Collectively, these results elucidate a novel epitranscriptomic
mechanism underlying cancer resistance to cisplatin chemotherapy, and have the potential to
identify paths towards new therapeutic targets to short-circuit drug resistance.
Future studies comparing large cohorts of tumor samples from transurethral resection of bladder
tumors (TURBTs pre-cisplatin) versus tumor samples from radical cystectomies (post-cisplatin)
could further expand on the clinical relevance of our mechanistic findings. Additionally, genetic
manipulation of the specific m6
A sites could offer new insights about their potential roles in

125
regulating adaptive chemoresistance. New tools and technology are being developed to target
individual sites using programmable CRISPR-based m6
A editing208,209.. The biggest limitation of
this study and perhaps more broadly of the field is how to translate this information from benchtop
to bedside. For example, m6
A effectors (writers, erasers, readers) are generally recognized to be
promiscuous, interacting with a broad array of targets. This pleiotropism makes them challenging
clinical targets. Having established the association of m6
A modifications with SLC7A11-mediated
cisplatin resistance, we currently do not yet have the molecular tools to manipulate these methyl
sites in a transcript-specific manner at therapeutic scale. Such tools for targeting individual m6
A
sites specifically would need to be further optimized and made ready for delivery in vivo. For now,
we can leverage the insights gained from our m6
A mapping studies to focus on m6
A-modified
transcripts like SLC7A11, and to control their expression broadly (independent of m6
A) to reduce
cisplatin resistance. There are a number of ongoing clinical trials evaluating the utility of SLC7A11
inhibitors, sorafenib and sulfasalazine, in conjunction with chemotherapy in other malignancies
including glioblastoma multiforme and breast cancer126. Although sorafenib and sulfasalazine are
themselves pharmacodynamically non-specific, these are promising first steps, and such studies
could be expanded to bladder cancer.
In chapter 4, we apply our epitranscriptome-wide strategy to prostate cancer using a welldocumented model of resistance to enzalutamide, a second-generation androgen receptor
signaling inhibitor (ARSI). One mechanisim of resistance isthe emergence of a constitutively
active splice variant ARV7. This offers a titillating potential link to m6
A regulation, which is known
to impact genes like MALAT1 that mediate post-transcriptional regulation and alternative splicing
(REF). We used a cell line model of enzalutamide resistance generated via long-term
enzalutamide exposure: C4-2B and MDV-R187. As in BC, we found that enzalutamide-sensitive
and enzalutamide-resistant PC cells had distinct m6
A profiles. In resistant MDV-R cells, THBS1,
was found to have increased m6
A and decreased expression, suggesting that THBS1 repression

126
may play a role in enzalutamide resistance.. Indeed, siRNA depletion of THBS1 potentiated
enzalutamide-resistance in sensitive PC cell lines. THBS1 methylation and downregulation was
recently reported in prostate tumorigenesis (REF), but our findings of its role in therapy resistance
are novel. Additional studies in this project can focus on identifying specific m6
A effectors
implicated in this pathway (METTL14 and YTHDF2 are possible candidates101). Other studies can
also focus on the link between THBS1 downregulation and enzalutamide resistance, whether
involving alternative splicing (e.g. MALAT1, ARV7) or other mechanisms. Our laboratory, in
conjunction with the Norris Comprehensive Cancer Center, has set up a new clinical cohort study,
the Longitudinal Advanced Prostate Cancer Cohort (LAPCC), collecting patient samples before
and after treatment for matched comparison. This will enable us to explore clinical correlates of
the transition to drug resistance for this and other studies in a way that current clinical database
do not allow us to do for lack of adequate clinical annotation.
The final study presented in this dissertation (chapter 5) is an auxiliary translational project, aimed
at pioneering a “multiparametric” approach of liquid biopsy profiling of prostate cancer patient
samples, integrating multiple blood-based tumor phenotypes to yield a maximally informative
disease profile including putative genomic and transcriptomic markers of drug resistance. We
present a pilot study with a small cohort of 20 patients with metastatic castrate-resistant prostate
cancer (mCRPC), demonstrating the feasibility and potential utility of using this approach for
minimally invasive yet comprehensive monitoring of disease phenotype over time, helping better
guide therapy. Since the publication of this study in 2019, we have expanded this work to include
additional analytes including radiomic workflow210. Both studies set up a framework for the future
of precision oncology, and large NCI-funded translational studies validating the utility of each of
these analytes in prospective clinical cohorts are currently being led by our group.

127
Bibliography
1. Siegel, R.L., Miller, K.D., Wagle, N.S., and Jemal, A. (2023). Cancer statistics, 2023. CA
Cancer J Clin 73, 17-48. 10.3322/caac.21763.
2. Jones, R., Latinovic, R., Charlton, J., and Gulliford, M.C. (2007). Alarm symptoms in
early diagnosis of cancer in primary care: cohort study using General Practice Research
Database. BMJ 334, 1040. 10.1136/bmj.39171.637106.AE.
3. Månsson, A., Anderson, H., and Colleen, S. (1993). Time lag to diagnosis of bladder
cancer--influence of psychosocial parameters and level of health-care provision. Scand J
Urol Nephrol 27, 363-369. 10.3109/00365599309180448.
4. Kamat, A.M., Hahn, N.M., Efstathiou, J.A., Lerner, S.P., Malmström, P.U., Choi, W.,
Guo, C.C., Lotan, Y., and Kassouf, W. (2016). Bladder cancer. Lancet 388, 2796-2810.
10.1016/S0140-6736(16)30512-8.
5. Prout, G.R., Barton, B.A., Griffin, P.P., and Friedell, G.H. (1992). Treated history of
noninvasive grade 1 transitional cell carcinoma. The National Bladder Cancer Group. J
Urol 148, 1413-1419. 10.1016/s0022-5347(17)36924-0.
6. Gakis, G., Efstathiou, J., Lerner, S.P., Cookson, M.S., Keegan, K.A., Guru, K.A.,
Shipley, W.U., Heidenreich, A., Schoenberg, M.P., Sagaloswky, A.I., et al. (2013). ICUDEAU International Consultation on Bladder Cancer 2012: Radical cystectomy and
bladder preservation for muscle-invasive urothelial carcinoma of the bladder. Eur Urol
63, 45-57. 10.1016/j.eururo.2012.08.009.
7. Konala, V.M., Adapa, S., and Aronow, W.S. (2022). Immunotherapy in Bladder Cancer.
Am J Ther 29, e334-e337. 10.1097/MJT.0000000000000934.
8. Sanli, O., Dobruch, J., Knowles, M.A., Burger, M., Alemozaffar, M., Nielsen, M.E., and
Lotan, Y. (2017). Bladder cancer. Nat Rev Dis Primers 3, 17022. 10.1038/nrdp.2017.22.
9. Hayashi, T., Fujita, K., Hayashi, Y., Hatano, K., Kawashima, A., McConkey, D.J., and
Nonomura, N. (2020). Mutational Landscape and Environmental Effects in Bladder
Cancer. Int J Mol Sci 21. 10.3390/ijms21176072.
10. Freedman, N.D., Silverman, D.T., Hollenbeck, A.R., Schatzkin, A., and Abnet, C.C.
(2011). Association between smoking and risk of bladder cancer among men and
women. JAMA 306, 737-745. 10.1001/jama.2011.1142.
11. Chow, N.H., Cairns, P., Eisenberger, C.F., Schoenberg, M.P., Taylor, D.C., Epstein, J.I.,
and Sidransky, D. (2000). Papillary urothelial hyperplasia is a clonal precursor to
papillary transitional cell bladder cancer. Int J Cancer 89, 514-518.
12. van Oers, J.M., Adam, C., Denzinger, S., Stoehr, R., Bertz, S., Zaak, D., Stief, C.,
Hofstaedter, F., Zwarthoff, E.C., van der Kwast, T.H., et al. (2006). Chromosome 9
deletions are more frequent than FGFR3 mutations in flat urothelial hyperplasias of the
bladder. Int J Cancer 119, 1212-1215. 10.1002/ijc.21958.
13. Obermann, E.C., Junker, K., Stoehr, R., Dietmaier, W., Zaak, D., Schubert, J.,
Hofstaedter, F., Knuechel, R., and Hartmann, A. (2003). Frequent genetic alterations in

128
flat urothelial hyperplasias and concomitant papillary bladder cancer as detected by
CGH, LOH, and FISH analyses. J Pathol 199, 50-57. 10.1002/path.1259.
14. Gao, J., Huang, H.Y., Pak, J., Cheng, J., Zhang, Z.T., Shapiro, E., Pellicer, A., Sun, T.T.,
and Wu, X.R. (2004). p53 deficiency provokes urothelial proliferation and synergizes
with activated Ha-ras in promoting urothelial tumorigenesis. Oncogene 23, 687-696.
10.1038/sj.onc.1207169.
15. Zhang, Z.T., Pak, J., Shapiro, E., Sun, T.T., and Wu, X.R. (1999). Urothelium-specific
expression of an oncogene in transgenic mice induced the formation of carcinoma in situ
and invasive transitional cell carcinoma. Cancer Res 59, 3512-3517.
16. Puzio-Kuter, A.M., Castillo-Martin, M., Kinkade, C.W., Wang, X., Shen, T.H., Matos, T.,
Shen, M.M., Cordon-Cardo, C., and Abate-Shen, C. (2009). Inactivation of p53 and Pten
promotes invasive bladder cancer. Genes Dev 23, 675-680. 10.1101/gad.1772909.
17. Balbás-Martínez, C., Sagrera, A., Carrillo-de-Santa-Pau, E., Earl, J., Márquez, M.,
Vazquez, M., Lapi, E., Castro-Giner, F., Beltran, S., Bayés, M., et al. (2013). Recurrent
inactivation of STAG2 in bladder cancer is not associated with aneuploidy. Nat Genet
45, 1464-1469. 10.1038/ng.2799.
18. Nordentoft, I., Lamy, P., Birkenkamp-Demtröder, K., Shumansky, K., Vang, S., Hornshøj,
H., Juul, M., Villesen, P., Hedegaard, J., Roth, A., et al. (2014). Mutational context and
diverse clonal development in early and late bladder cancer. Cell Rep 7, 1649-1663.
10.1016/j.celrep.2014.04.038.
19. Hurst, C.D., Platt, F.M., Taylor, C.F., and Knowles, M.A. (2012). Novel tumor subgroups
of urothelial carcinoma of the bladder defined by integrated genomic analysis. Clin
Cancer Res 18, 5865-5877. 10.1158/1078-0432.CCR-12-1807.
20. Network, C.G.A.R. (2014). Comprehensive molecular characterization of urothelial
bladder carcinoma. Nature 507, 315-322. 10.1038/nature12965.
21. Huggins, C., and Hodges, C.V. (1972). Studies on prostatic cancer. I. The effect of
castration, of estrogen and androgen injection on serum phosphatases in metastatic
carcinoma of the prostate. CA Cancer J Clin 22, 232-240. 10.3322/canjclin.22.4.232.
22. Wang, G., Zhao, D., Spring, D.J., and DePinho, R.A. (2018). Genetics and biology of
prostate cancer. Genes Dev 32, 1105-1140. 10.1101/gad.315739.118.
23. de Bono, J.S., Logothetis, C.J., Molina, A., Fizazi, K., North, S., Chu, L., Chi, K.N.,
Jones, R.J., Goodman, O.B., Saad, F., et al. (2011). Abiraterone and increased survival
in metastatic prostate cancer. N Engl J Med 364, 1995-2005. 10.1056/NEJMoa1014618.
24. Fizazi, K., Scher, H.I., Molina, A., Logothetis, C.J., Chi, K.N., Jones, R.J., Staffurth, J.N.,
North, S., Vogelzang, N.J., Saad, F., et al. (2012). Abiraterone acetate for treatment of
metastatic castration-resistant prostate cancer: final overall survival analysis of the COUAA-301 randomised, double-blind, placebo-controlled phase 3 study. Lancet Oncol 13,
983-992. 10.1016/S1470-2045(12)70379-0.

129
25. Ryan, C.J., Smith, M.R., de Bono, J.S., Molina, A., Logothetis, C.J., de Souza, P., Fizazi,
K., Mainwaring, P., Piulats, J.M., Ng, S., et al. (2013). Abiraterone in metastatic prostate
cancer without previous chemotherapy. N Engl J Med 368, 138-148.
10.1056/NEJMoa1209096.
26. James, N.D., de Bono, J.S., Spears, M.R., Clarke, N.W., Mason, M.D., Dearnaley, D.P.,
Ritchie, A.W.S., Amos, C.L., Gilson, C., Jones, R.J., et al. (2017). Abiraterone for
Prostate Cancer Not Previously Treated with Hormone Therapy. N Engl J Med 377, 338-
351. 10.1056/NEJMoa1702900.
27. Ryan, C.J., Smith, M.R., Fizazi, K., Saad, F., Mulders, P.F., Sternberg, C.N., Miller, K.,
Logothetis, C.J., Shore, N.D., Small, E.J., et al. (2015). Abiraterone acetate plus
prednisone versus placebo plus prednisone in chemotherapy-naive men with metastatic
castration-resistant prostate cancer (COU-AA-302): final overall survival analysis of a
randomised, double-blind, placebo-controlled phase 3 study. Lancet Oncol 16, 152-160.
10.1016/S1470-2045(14)71205-7.
28. Tran, C., Ouk, S., Clegg, N.J., Chen, Y., Watson, P.A., Arora, V., Wongvipat, J., SmithJones, P.M., Yoo, D., Kwon, A., et al. (2009). Development of a second-generation
antiandrogen for treatment of advanced prostate cancer. Science 324, 787-790.
10.1126/science.1168175.
29. Berruti, A., Generali, D., and Tampellini, M. (2012). Enzalutamide in prostate cancer
after chemotherapy. N Engl J Med 367, 2448; author reply 2448-2449.
10.1056/NEJMc1212940.
30. Scher, H.I., Fizazi, K., Saad, F., Taplin, M.E., Sternberg, C.N., Miller, K., de Wit, R.,
Mulders, P., Chi, K.N., Shore, N.D., et al. (2012). Increased survival with enzalutamide in
prostate cancer after chemotherapy. N Engl J Med 367, 1187-1197.
10.1056/NEJMoa1207506.
31. Beer, T.M., Armstrong, A.J., Rathkopf, D.E., Loriot, Y., Sternberg, C.N., Higano, C.S.,
Iversen, P., Bhattacharya, S., Carles, J., Chowdhury, S., et al. (2014). Enzalutamide in
metastatic prostate cancer before chemotherapy. N Engl J Med 371, 424-433.
10.1056/NEJMoa1405095.
32. Sandhu, S., Moore, C.M., Chiong, E., Beltran, H., Bristow, R.G., and Williams, S.G.
(2021). Prostate cancer. Lancet 398, 1075-1090. 10.1016/S0140-6736(21)00950-8.
33. Armenia, J., Wankowicz, S.A.M., Liu, D., Gao, J., Kundra, R., Reznik, E., Chatila, W.K.,
Chakravarty, D., Han, G.C., Coleman, I., et al. (2018). The long tail of oncogenic drivers
in prostate cancer. Nat Genet 50, 645-651. 10.1038/s41588-018-0078-z.
34. Fraser, M., Sabelnykova, V.Y., Yamaguchi, T.N., Heisler, L.E., Livingstone, J., Huang,
V., Shiah, Y.J., Yousif, F., Lin, X., Masella, A.P., et al. (2017). Genomic hallmarks of
localized, non-indolent prostate cancer. Nature 541, 359-364. 10.1038/nature20788.
35. Quigley, D.A., Dang, H.X., Zhao, S.G., Lloyd, P., Aggarwal, R., Alumkal, J.J., Foye, A.,
Kothari, V., Perry, M.D., Bailey, A.M., et al. (2018). Genomic Hallmarks and Structural
Variation in Metastatic Prostate Cancer. Cell 175, 889. 10.1016/j.cell.2018.10.019.

130
36. Nelson, W.G., and Yegnasubramanian, S. (2013). Resistance emerges to secondgeneration antiandrogens in prostate cancer. Cancer Discov 3, 971-974. 10.1158/2159-
8290.CD-13-0405.
37. Antonarakis, E.S., Lu, C., Luber, B., Wang, H., Chen, Y., Zhu, Y., Silberstein, J.L.,
Taylor, M.N., Maughan, B.L., Denmeade, S.R., et al. (2017). Clinical Significance of
Androgen Receptor Splice Variant-7 mRNA Detection in Circulating Tumor Cells of Men
With Metastatic Castration-Resistant Prostate Cancer Treated With First- and SecondLine Abiraterone and Enzalutamide. J Clin Oncol 35, 2149-2156.
10.1200/JCO.2016.70.1961.
38. Scher, H.I., Graf, R.P., Schreiber, N.A., McLaughlin, B., Lu, D., Louw, J., Danila, D.C.,
Dugan, L., Johnson, A., Heller, G., et al. (2017). Nuclear-specific AR-V7 Protein
Localization is Necessary to Guide Treatment Selection in Metastatic Castrationresistant Prostate Cancer. Eur Urol 71, 874-882. 10.1016/j.eururo.2016.11.024.
39. Nikolaou, M., Pavlopoulou, A., Georgakilas, A.G., and Kyrodimos, E. (2018). The
challenge of drug resistance in cancer treatment: a current overview. Clin Exp
Metastasis 35, 309-318. 10.1007/s10585-018-9903-0.
40. Cheng, C., Cui, H., Zhang, L., Jia, Z., Song, B., Wang, F., Li, Y., Liu, J., Kong, P., Shi,
R., et al. (2016). Genomic analyses reveal FAM84B and the NOTCH pathway are
associated with the progression of esophageal squamous cell carcinoma. Gigascience
5, 1. 10.1186/s13742-015-0107-0.
41. Bedard, P.L., Hansen, A.R., Ratain, M.J., and Siu, L.L. (2013). Tumour heterogeneity in
the clinic. Nature 501, 355-364. 10.1038/nature12627.
42. Jolly, M.K., Kulkarni, P., Weninger, K., Orban, J., and Levine, H. (2018). Phenotypic
Plasticity, Bet-Hedging, and Androgen Independence in Prostate Cancer: Role of NonGenetic Heterogeneity. Front Oncol 8, 50. 10.3389/fonc.2018.00050.
43. Pisco, A.O., and Huang, S. (2015). Non-genetic cancer cell plasticity and therapyinduced stemness in tumour relapse: 'What does not kill me strengthens me'. Br J
Cancer 112, 1725-1732. 10.1038/bjc.2015.146.
44. Hölzel, M., Bovier, A., and Tüting, T. (2013). Plasticity of tumour and immune cells: a
source of heterogeneity and a cause for therapy resistance? Nat Rev Cancer 13, 365-
376. 10.1038/nrc3498.
45. Veening, J.W., Smits, W.K., and Kuipers, O.P. (2008). Bistability, epigenetics, and bethedging in bacteria. Annu Rev Microbiol 62, 193-210.
10.1146/annurev.micro.62.081307.163002.
46. Salgia, R., and Kulkarni, P. (2018). The Genetic/Non-genetic Duality of Drug
'Resistance' in Cancer. Trends Cancer 4, 110-118. 10.1016/j.trecan.2018.01.001.
47. Sharma, S.V., Lee, D.Y., Li, B., Quinlan, M.P., Takahashi, F., Maheswaran, S.,
McDermott, U., Azizian, N., Zou, L., Fischbach, M.A., et al. (2010). A chromatinmediated reversible drug-tolerant state in cancer cell subpopulations. Cell 141, 69-80.
10.1016/j.cell.2010.02.027.

131
48. Chaffer, C.L., Marjanovic, N.D., Lee, T., Bell, G., Kleer, C.G., Reinhardt, F., D'Alessio,
A.C., Young, R.A., and Weinberg, R.A. (2013). Poised chromatin at the ZEB1 promoter
enables breast cancer cell plasticity and enhances tumorigenicity. Cell 154, 61-74.
10.1016/j.cell.2013.06.005.
49. Liau, B.B., Sievers, C., Donohue, L.K., Gillespie, S.M., Flavahan, W.A., Miller, T.E.,
Venteicher, A.S., Hebert, C.H., Carey, C.D., Rodig, S.J., et al. (2017). Adaptive
Chromatin Remodeling Drives Glioblastoma Stem Cell Plasticity and Drug Tolerance.
Cell Stem Cell 20, 233-246.e237. 10.1016/j.stem.2016.11.003.
50. Roesch, A., Vultur, A., Bogeski, I., Wang, H., Zimmermann, K.M., Speicher, D., Körbel,
C., Laschke, M.W., Gimotty, P.A., Philipp, S.E., et al. (2013). Overcoming intrinsic
multidrug resistance in melanoma by blocking the mitochondrial respiratory chain of
slow-cycling JARID1B(high) cells. Cancer Cell 23, 811-825. 10.1016/j.ccr.2013.05.003.
51. Perry, R.P., Kelley, D.E., Friderici, K., and Rottman, F. (1975). The methylated
constituents of L cell messenger RNA: evidence for an unusual cluster at the 5' terminus.
Cell 4, 387-394. 10.1016/0092-8674(75)90159-2.
52. Desrosiers, R., Friderici, K., and Rottman, F. (1974). Identification of methylated
nucleosides in messenger RNA from Novikoff hepatoma cells. Proc Natl Acad Sci U S A
71, 3971-3975. 10.1073/pnas.71.10.3971.
53. Zhong, S., Li, H., Bodi, Z., Button, J., Vespa, L., Herzog, M., and Fray, R.G. (2008). MTA
is an Arabidopsis messenger RNA adenosine methylase and interacts with a homolog of
a sex-specific splicing factor. Plant Cell 20, 1278-1288. 10.1105/tpc.108.058883.
54. Dominissini, D., Moshitch-Moshkovitz, S., Schwartz, S., Salmon-Divon, M., Ungar, L.,
Osenberg, S., Cesarkas, K., Jacob-Hirsch, J., Amariglio, N., Kupiec, M., et al. (2012).
Topology of the human and mouse m6A RNA methylomes revealed by m6A-seq. Nature
485, 201-206. 10.1038/nature11112.
55. Dominissini, D., Moshitch-Moshkovitz, S., Salmon-Divon, M., Amariglio, N., and Rechavi,
G. (2013). Transcriptome-wide mapping of N(6)-methyladenosine by m(6)A-seq based
on immunocapturing and massively parallel sequencing. Nat Protoc 8, 176-189.
10.1038/nprot.2012.148.
56. Meyer, K.D., Saletore, Y., Zumbo, P., Elemento, O., Mason, C.E., and Jaffrey, S.R.
(2012). Comprehensive analysis of mRNA methylation reveals enrichment in 3' UTRs
and near stop codons. Cell 149, 1635-1646. 10.1016/j.cell.2012.05.003.
57. Murakami, S., and Jaffrey, S.R. (2022). Hidden codes in mRNA: Control of gene
expression by m. Mol Cell 82, 2236-2251. 10.1016/j.molcel.2022.05.029.
58. Linder, B., Grozhik, A.V., Olarerin-George, A.O., Meydan, C., Mason, C.E., and Jaffrey,
S.R. (2015). Single-nucleotide-resolution mapping of m6A and m6Am throughout the
transcriptome. Nat Methods 12, 767-772. 10.1038/nmeth.3453.

132
59. Wang, S., Sun, C., Li, J., Zhang, E., Ma, Z., Xu, W., Li, H., Qiu, M., Xu, Y., Xia, W., et al.
(2017). Roles of RNA methylation by means of N. Cancer Lett 408, 112-120.
10.1016/j.canlet.2017.08.030.
60. Lan, Q., Liu, P.Y., Haase, J., Bell, J.L., Hüttelmaier, S., and Liu, T. (2019). The Critical
Role of RNA m. Cancer Res 79, 1285-1292. 10.1158/0008-5472.CAN-18-2965.
61. Mauer, J., Luo, X., Blanjoie, A., Jiao, X., Grozhik, A.V., Patil, D.P., Linder, B., Pickering,
B.F., Vasseur, J.J., Chen, Q., et al. (2017). Reversible methylation of m. Nature 541,
371-375. 10.1038/nature21022.
62. Mauer, J., Sindelar, M., Despic, V., Guez, T., Hawley, B.R., Vasseur, J.J., Rentmeister,
A., Gross, S.S., Pellizzoni, L., Debart, F., et al. (2019). FTO controls reversible m. Nat
Chem Biol 15, 340-347. 10.1038/s41589-019-0231-8.
63. Louloupi, A., Ntini, E., Conrad, T., and Ørom, U.A.V. (2018). Transient N-6-
Methyladenosine Transcriptome Sequencing Reveals a Regulatory Role of m6A in
Splicing Efficiency. Cell Rep 23, 3429-3437. 10.1016/j.celrep.2018.05.077.
64. Du, H., Zhao, Y., He, J., Zhang, Y., Xi, H., Liu, M., Ma, J., and Wu, L. (2016). YTHDF2
destabilizes m(6)A-containing RNA through direct recruitment of the CCR4-NOT
deadenylase complex. Nat Commun 7, 12626. 10.1038/ncomms12626.
65. Kasowitz, S.D., Ma, J., Anderson, S.J., Leu, N.A., Xu, Y., Gregory, B.D., Schultz, R.M.,
and Wang, P.J. (2018). Nuclear m6A reader YTHDC1 regulates alternative
polyadenylation and splicing during mouse oocyte development. PLoS Genet 14,
e1007412. 10.1371/journal.pgen.1007412.
66. Xiao, W., Adhikari, S., Dahal, U., Chen, Y.S., Hao, Y.J., Sun, B.F., Sun, H.Y., Li, A.,
Ping, X.L., Lai, W.Y., et al. (2016). Nuclear m(6)A Reader YTHDC1 Regulates mRNA
Splicing. Mol Cell 61, 507-519. 10.1016/j.molcel.2016.01.012.
67. Fustin, J.M., Doi, M., Yamaguchi, Y., Hida, H., Nishimura, S., Yoshida, M., Isagawa, T.,
Morioka, M.S., Kakeya, H., Manabe, I., and Okamura, H. (2013). RNA-methylationdependent RNA processing controls the speed of the circadian clock. Cell 155, 793-806.
10.1016/j.cell.2013.10.026.
68. Wang, X., Zhao, B.S., Roundtree, I.A., Lu, Z., Han, D., Ma, H., Weng, X., Chen, K., Shi,
H., and He, C. (2015). N(6)-methyladenosine Modulates Messenger RNA Translation
Efficiency. Cell 161, 1388-1399. 10.1016/j.cell.2015.05.014.
69. Zhang, H., Shi, X., Huang, T., Zhao, X., Chen, W., Gu, N., and Zhang, R. (2020).
Dynamic landscape and evolution of m6A methylation in human. Nucleic Acids Res 48,
6251-6264. 10.1093/nar/gkaa347.
70. McIntyre, A.B.R., Gokhale, N.S., Cerchietti, L., Jaffrey, S.R., Horner, S.M., and Mason,
C.E. (2020). Limits in the detection of m 6 A changes using MeRIP/m 6 A-seq. Sci Rep
10, 6590. 10.1038/s41598-020-63355-3.
71. Garcia-Campos, M.A., Edelheit, S., Toth, U., Safra, M., Shachar, R., Viukov, S., Winkler,
R., Nir, R., Lasman, L., Brandis, A., et al. (2019). Deciphering the "m6A code" via

133
antibody-independent quantitative profiling. Cell 178, 731-747.e716.
10.1016/j.cell.2019.06.013.
72. Geula, S., Moshitch-Moshkovitz, S., Dominissini, D., Mansour, A.A., Kol, N., SalmonDivon, M., Hershkovitz, V., Peer, E., Mor, N., Manor, Y.S., et al. (2015). m6A mRNA
methylation facilitates resolution of naïve pluripotency toward differentiation. Science
347, 1002-1006. 10.1126/science.1261417.
73. Batista, P.J. (2017). m(6)A RNA modification controls cell fate transition in mammalian
embryonic stem cells. Genomics Proteomics Bioinformatics 15, 154-163.
10.1016/j.gpb.2017.03.002.
74. Wang, X., Lu, Z., Gomez, A., Hon, G.C., Yue, Y., Han, D., Fu, Y., Parisien, M., Dai, Q.,
Jia, G., et al. (2014). N6-methyladenosine-dependent regulation of messenger RNA
stability. Nature 505, 117-120. 10.1038/nature12730.
75. Lasman, L., Krupalnik, V., Viukov, S., Mor, N., Aguilera-Castrejon, A., Schneir, D.,
Bayerl, J., Mizrahi, O., Peles, S., Tawil, S., et al. (2020). Context-dependent functional
compensation between Ythdf m. Genes Dev 34, 1373-1391. 10.1101/gad.340695.120.
76. Zaccara, S., and Jaffrey, S.R. (2020). A unified model for the function of YTHDF proteins
in regulating m6A-modified mRNA. Cell 181, 1582-1595.e1518.
10.1016/j.cell.2020.05.012.
77. Wei, G., Almeida, M., Pintacuda, G., Coker, H., Bowness, J.S., Ule, J., and Brockdorff,
N. (2021). Acute depletion of METTL3 implicates. Genome Res 31, 1395-1408.
10.1101/gr.271635.120.
78. Wang, S., Chai, P., and Jia, R. (2018). Novel insights on m 6 A RNA methylation in
tumorigenesis: a double-edged sword. Mol Cancer 17, 101. 10.1186/s12943-018-
0847-4.
79. Cui, Q., Shi, H., Ye, P., Li, L., Qu, Q., Sun, G., Lu, Z., Huang, Y., Yang, C.G., Riggs,
A.D., et al. (2017). m6A RNA methylation regulates the self-renewal and tumorigenesis
of glioblastoma stem cells,. Cell Rep 18, 2622-2634. 10.1016/j.celrep.2017.02.059.
80. Zhang, S., Zhao, B.S., Zhou, A., Lin, K., Zheng, S., Lu, Z., Chen, Y., Sulman, E.P., Xie,
K., Bögler, O., et al. (2017). m6A demethylase ALKBH5 maintains tumorigenicity of
glioblastoma stem-like cells by sus- taining FOXM1 expression and cell proliferation
program. Cancer Cell 31, 591-606.e596. 10.1016/j.ccell.2017.02.013.
81. Li, Z., Weng, H., Su, R., Weng, X., Zuo, Z., Li, C., Huang, H., Nachtergaele, S., Dong, L.,
Hu, C., et al. (2017). FTO Plays an Oncogenic Role in Acute Myeloid Leukemia as a N.
Cancer Cell 31, 127-141. 10.1016/j.ccell.2016.11.017.
82. Huang, Y., Su, R., Sheng, Y., Dong, L., Dong, Z., Xu, H., Ni, T., Zhang, Z.S., Zhang, T.,
Li, C., et al. (2019). Small-Molecule Targeting of Oncogenic FTO Demethylase in Acute
Myeloid Leukemia. Cancer Cell 35, 677-691.e610. 10.1016/j.ccell.2019.03.006.

134
83. Lobo, J., Barros-Silva, D., Henrique, R., and Jerónimo, C. (2018). The Emerging Role of
Epitranscriptomics in Cancer: Focus on Urological Tumors. Genes (Basel) 9.
10.3390/genes9110552.
84. Han, J., Wang, J.Z., Yang, X., Yu, H., Zhou, R., Lu, H.C., Yuan, W.B., Lu, J.C., Zhou,
Z.J., Lu, Q., et al. (2019). METTL3 promote tumor proliferation of bladder cancer by
accelerating pri-miR221/222 maturation in m6A-dependent manner. Mol Cancer 18, 110.
10.1186/s12943-019-1036-9.
85. Ni, Z., Sun, P., Zheng, J., Wu, M., Yang, C., Cheng, M., Yin, M., Cui, C., Wang, G.,
Yuan, L., et al. (2022). JNK Signaling Promotes Bladder Cancer Immune Escape by
Regulating METTL3-Mediated m6A Modification of PD-L1 mRNA. Cancer Res 82, 1789-
1802. 10.1158/0008-5472.CAN-21-1323.
86. Wang, K., Wang, G., Li, G., Zhang, W., Wang, Y., Lin, X., Han, C., Chen, H., Shi, L.,
Reheman, A., et al. (2023). m6A writer WTAP targets NRF2 to accelerate bladder
cancer malignancy via m6A-dependent ferroptosis regulation. Apoptosis 28, 627-638.
10.1007/s10495-023-01817-5.
87. Wei, W., Sun, J., Zhang, H., Xiao, X., Huang, C., Wang, L., Zhong, H., Jiang, Y., Zhang,
X., and Jiang, G. (2021). Interaction with WTAP Promotes Assembly and Activity of the
m. Cancer Res 81, 6142-6156. 10.1158/0008-5472.CAN-21-1518.
88. Wen, L., Pan, X., Yu, Y., and Yang, B. (2020). Down-regulation of FTO promotes
proliferation and migration, and protects bladder cancer cells from cisplatin-induced
cytotoxicity. BMC Urol 20, 39. 10.1186/s12894-020-00612-7.
89. Yi, W., Yu, Y., Li, Y., Yang, J., Gao, S., and Xu, L. (2021). The tumor-suppressive effects
of alpha-ketoglutarate-dependent dioxygenase FTO via N6-methyladenosine RNA
methylation on bladder cancer patients. Bioengineered 12, 5323-5333.
10.1080/21655979.2021.1964893.
90. Song, W., Yang, K., Luo, J., Gao, Z., and Gao, Y. (2021). Dysregulation of
USP18/FTO/PYCR1 signaling network promotes bladder cancer development and
progression. Aging (Albany NY) 13, 3909-3925. 10.18632/aging.202359.
91. Tao, L., Mu, X., Chen, H., Jin, D., Zhang, R., Zhao, Y., Fan, J., Cao, M., and Zhou, Z.
(2021). FTO modifies the m6A level of MALAT and promotes bladder cancer
progression. Clin Transl Med 11, e310. 10.1002/ctm2.310.
92. Zhou, G., Yan, K., Liu, J., Gao, L., Jiang, X., and Fan, Y. (2021). FTO promotes tumour
proliferation in bladder cancer via the FTO/miR-576/CDK6 axis in an m6A-dependent
manner. Cell Death Discov 7, 329. 10.1038/s41420-021-00724-5.
93. Sun, J., Ma, X., Ying, Y., Wang, W., Shen, H., Wang, S., Xie, H., Yi, J., Zhan, W., Li, J.,
and Liu, B. (2022). SMAD3 and FTO are involved in miR-5581-3p-mediated inhibition of
cell migration and proliferation in bladder cancer. Cell Death Discov 8, 199.
10.1038/s41420-022-01010-8.
94. Zhang, L., Li, Y., Zhou, L., Zhou, H., Ye, L., Ou, T., Hong, H., Zheng, S., Zhou, Z., Wu,
K., et al. (2023). The m6A Reader YTHDF2 Promotes Bladder Cancer Progression by

135
Suppressing RIG-I-Mediated Immune Response. Cancer Res 83, 1834-1850.
10.1158/0008-5472.CAN-22-2485.
95. Yu, H., Yang, X., Tang, J., Si, S., Zhou, Z., Lu, J., Han, J., Yuan, B., Wu, Q., Lu, Q., and
Yang, H. (2021). ALKBH5 Inhibited Cell Proliferation and Sensitized Bladder Cancer
Cells to Cisplatin by m6A-CK2α-Mediated Glycolysis. Mol Ther Nucleic Acids 23, 27-41.
10.1016/j.omtn.2020.10.031.
96. Cotter, K.A., Gallon, J., Uebersax, N., Rubin, P., Meyer, K.D., Piscuoglio, S., Jaffrey,
S.R., and Rubin, M.A. (2021). Mapping of m. Mol Cancer Res 19, 1398-1411.
10.1158/1541-7786.MCR-21-0014.
97. Mao, Y., Li, W., Weng, Y., Hua, B., Gu, X., Lu, C., Xu, B., Xu, H., and Wang, Z. (2022).
METTL3-Mediated m. Cell Transplant 31, 9636897221122997.
10.1177/09636897221122997.
98. Ma, X.X., Cao, Z.G., and Zhao, S.L. (2020). m6A methyltransferase METTL3 promotes
the progression of prostate cancer via m6A-modified LEF1. Eur Rev Med Pharmacol Sci
24, 3565-3571. 10.26355/eurrev_202004_20817.
99. Chen, Y., Pan, C., Wang, X., Xu, D., Ma, Y., Hu, J., Chen, P., Xiang, Z., Rao, Q., and
Han, X. (2021). Silencing of METTL3 effectively hinders invasion and metastasis of
prostate cancer cells. Theranostics 11, 7640-7657. 10.7150/thno.61178.
100. Li, J., Xie, H., Ying, Y., Chen, H., Yan, H., He, L., Xu, M., Xu, X., Liang, Z., Liu, B., et al.
(2020). YTHDF2 mediates the mRNA degradation of the tumor suppressors to induce
AKT phosphorylation in N6-methyladenosine-dependent way in prostate cancer. Mol
Cancer 19, 152. 10.1186/s12943-020-01267-6.
101. Wang, Y., Chen, J., Gao, W.Q., and Yang, R. (2022). METTL14 promotes prostate
tumorigenesis by inhibiting THBS1 via an m6A-YTHDF2-dependent mechanism. Cell
Death Discov 8, 143. 10.1038/s41420-022-00939-0.
102. Xia, L., Han, Q., Duan, X., Zhu, Y., Pan, J., Dong, B., Xia, W., Xue, W., and Sha, J.
(2022). m. Mol Ther Nucleic Acids 28, 219-230. 10.1016/j.omtn.2022.03.008.
103. He, K., Xu, T., and Goldkorn, A. (2011). Cancer cells cyclically lose and regain drugresistant highly tumorigenic features characteristic of a cancer stem-like phenotype. Mol
Cancer Ther 10, 938-948. 10.1158/1535-7163.MCT-10-1120.
104. He, K., Xu, T., Xu, Y., Ring, A., Kahn, M., and Goldkorn, A. (2014). Cancer cells acquire
a drug resistant, highly tumorigenic, cancer stem-like phenotype through modulation of
the PI3K/Akt/β-catenin/CBP pathway. Int J Cancer 134, 43-54. 10.1002/ijc.28341.
105. Byun, S.S., Kim, S.W., Choi, H., Lee, C., and Lee, E. (2005). Augmentation of cisplatin
sensitivity in cisplatin-resistant human bladder cancer cells by modulating glutathione
concentrations and glutathione-related enzyme activities. BJU Int 95, 1086-1090.
10.1111/j.1464-410X.2005.05472.x.

136
106. Xu, T., Li, H.T., Wei, J., Li, M., Hsieh, T.C., Lu, Y.T., Lakshminarasimhan, R., Xu, R.,
Hodara, E., Morrison, G., et al. (2020). Epigenetic plasticity potentiates a rapid cyclical
shift to and from an aggressive cancer phenotype. Int J Cancer. 10.1002/ijc.32904.
107. Xu, T., Junge, J.A., Delfarah, A., Lu, Y.T., Arnesano, C., Iqbal, M., Delijani, K., Hsieh,
T.C., Hodara, E., Mehta, H.H., et al. (2022). Bladder cancer cells shift rapidly and
spontaneously to cisplatin-resistant oxidative phosphorylation that is trackable in real
time. Sci Rep 12, 5518. 10.1038/s41598-022-09438-9.
108. Lu YT, X.T., Iqbal M, Hsieh TC, Luo Z, Liang G, Farnham PJ, Rhie SK, Goldkorn A
(2022). FOXC1 Binds Enhancers and Promotes Cisplatin Resistance in Bladder Cancer.
109. Zhou, J., Wulfkuhle, J., Zhang, H., Gu, P., Yang, Y., Deng, J., Margolick, J.B., Liotta,
L.A., Petricoin, E., and Zhang, Y. (2007). Activation of the PTEN/mTOR/STAT3 pathway
in breast cancer stem-like cells is required for viability and maintenance. Proc Natl Acad
Sci U S A 104, 16158-16163. 10.1073/pnas.0702596104.
110. Zhang, C., Fu, J., and Zhou, Y. (2019). A Review in Research Progress Concerning
m6A Methylation and Immunoregulation. Front Immunol 10, 922.
10.3389/fimmu.2019.00922.
111. Zeng, Y., Wang, S., Gao, S., Soares, F., Ahmed, M., Guo, H., Wang, M., Hua, J.T.,
Guan, J., Moran, M.F., et al. (2018). Refined RIP-seq protocol for epitranscriptome
analysis with low input materials. PLoS Biol 16, e2006092.
10.1371/journal.pbio.2006092.
112. Chen, T., Hao, Y.J., Zhang, Y., Li, M.M., Wang, M., Han, W., Wu, Y., Lv, Y., Hao, J.,
Wang, L., et al. (2015). m(6)A RNA methylation is regulated by microRNAs and
promotes reprogramming to pluripotency. Cell Stem Cell 16, 289-301.
10.1016/j.stem.2015.01.016.
113. Yan, F., Al-Kali, A., Zhang, Z., Liu, J., Pang, J., Zhao, N., He, C., Litzow, M.R., and Liu,
S. (2018). A dynamic N 6 -methyladenosine methylome regulates intrinsic and acquired
resistance to tyrosine kinase inhibitors. Cell Res 28, 1062-1076. 10.1038/s41422-018-
0097-4.
114. Li, W., Zhang, H., Assaraf, Y.G., Zhao, K., Xu, X., Xie, J., Yang, D.H., and Chen, Z.S.
(2016). Overcoming ABC transporter-mediated multidrug resistance: Molecular
mechanisms and novel therapeutic drug strategies. Drug Resist Updat 27, 14-29.
10.1016/j.drup.2016.05.001.
115. Kathawala, R.J., Gupta, P., Ashby, C.R., and Chen, Z.S. (2015). The modulation of ABC
transporter-mediated multidrug resistance in cancer: a review of the past decade. Drug
Resist Updat 18, 1-17. 10.1016/j.drup.2014.11.002.
116. Li, B., Jiang, J., Assaraf, Y.G., Xiao, H., Chen, Z.S., and Huang, C. (2020). Surmounting
cancer drug resistance: New insights from the perspective of N. Drug Resist Updat 53,
100720. 10.1016/j.drup.2020.100720.
117. Hodara, E., Mades, A., Swartz, L., Iqbal, M., Xu, T., Bsteh, D., Farnham, Peggy J., Rhie,
Suhn K., and Goldkorn, A. (2023). m6A epitranscriptome analysis reveals differentially

137
methylated transcripts that drive early chemoresistance in bladder cancer. NAR Cancer
5. 10.1093/narcan/zcad054.
118. Zaccara, S., Ries, R.J., and Jaffrey, S.R. (2019). Reading, writing and erasing mRNA
methylation. Nat Rev Mol Cell Biol 20, 608-624. 10.1038/s41580-019-0168-5.
119. Huang, H., Weng, H., and Chen, J. (2020). m 6 A Modification in Coding and Non-coding
RNAs: Roles and Therapeutic Implications in Cancer. Cancer Cell 37, 270-288.
10.1016/j.ccell.2020.02.004.
120. Liu, J., Yue, Y., Han, D., Wang, X., Fu, Y., Zhang, L., Jia, G., Yu, M., Lu, Z., Deng, X., et
al. (2014). A METTL3-METTL14 complex mediates mammalian nuclear RNA N6-
adenosine methylation. Nat Chem Biol 10, 93-95. 10.1038/nchembio.1432.
121. Knuckles, P., Lence, T., Haussmann, I.U., Jacob, D., Kreim, N., Carl, S.H., Masiello, I.,
Hares, T., Villaseñor, R., Hess, D., et al. (2018). Zc3h13/Flacc is required for adenosine
methylation by bridging the mRNA-binding factor Rbm15/Spenito to the m. Genes Dev
32, 415-429. 10.1101/gad.309146.117.
122. Jia, G., Fu, Y., Zhao, X., Dai, Q., Zheng, G., Yang, Y., Yi, C., Lindahl, T., Pan, T., Yang,
Y.G., and He, C. (2011). N6-methyladenosine in nuclear RNA is a major substrate of the
obesity-associated FTO. Nat Chem Biol 7, 885-887. 10.1038/nchembio.687.
123. Zheng, G., Dahl, J.A., Niu, Y., Fedorcsak, P., Huang, C.M., Li, C.J., Vågbø, C.B., Shi, Y.,
Wang, W.L., Song, S.H., et al. (2013). ALKBH5 is a mammalian RNA demethylase that
impacts RNA metabolism and mouse fertility. Mol Cell 49, 18-29.
10.1016/j.molcel.2012.10.015.
124. Wang, W., Shao, F., Yang, X., Wang, J., Zhu, R., Yang, Y., Zhao, G., Guo, D., Sun, Y.,
Xue, Q., et al. (2021). METTL3 promotes tumour development by decreasing APC
expression mediated by APC mRNA N. Nat Commun 12, 3803. 10.1038/s41467-021-
23501-5.
125. Koppula, P., Zhuang, L., and Gan, B. (2021). Cystine transporter SLC7A11/xCT in
cancer: ferroptosis, nutrient dependency, and cancer therapy. Protein Cell 12, 599-620.
10.1007/s13238-020-00789-5.
126. Chen, X., Kang, R., Kroemer, G., and Tang, D. (2021). Broadening horizons: the role of
ferroptosis in cancer. Nat Rev Clin Oncol 18, 280-296. 10.1038/s41571-020-00462-0.
127. Liang, Y., Ye, F., Xu, C., Zou, L., Hu, Y., Hu, J., and Jiang, H. (2021). A novel survival
model based on a Ferroptosis-related gene signature for predicting overall survival in
bladder cancer. BMC Cancer 21, 943. 10.1186/s12885-021-08687-7.
128. Gu, C., Wang, Z., Zhou, N., Li, G., Kou, Y., Luo, Y., Wang, Y., Yang, J., and Tian, F.
(2019). Mettl14 inhibits bladder TIC self-renewal and bladder tumorigenesis through N.
Mol Cancer 18, 168. 10.1186/s12943-019-1084-1.
129. Okuno, S., Sato, H., Kuriyama-Matsumura, K., Tamba, M., Wang, H., Sohda, S.,
Hamada, H., Yoshikawa, H., Kondo, T., and Bannai, S. (2003). Role of cystine transport

138
in intracellular glutathione level and cisplatin resistance in human ovarian cancer cell
lines. Br J Cancer 88, 951-956. 10.1038/sj.bjc.6600786.
130. Lei, G., Zhuang, L., and Gan, B. (2022). Targeting ferroptosis as a vulnerability in
cancer. Nat Rev Cancer 22, 381-396. 10.1038/s41568-022-00459-0.
131. Drayton, R.M., Dudziec, E., Peter, S., Bertz, S., Hartmann, A., Bryant, H.E., and Catto,
J.W. (2014). Reduced expression of miRNA-27a modulates cisplatin resistance in
bladder cancer by targeting the cystine/glutamate exchanger SLC7A11. Clin Cancer Res
20, 1990-2000. 10.1158/1078-0432.ccr-13-2805.
132. Ma, L., Chen, T., Zhang, X., Miao, Y., Tian, X., Yu, K., Xu, X., Niu, Y., Guo, S., Zhang,
C., et al. (2021). The m 6 A reader YTHDC2 inhibits lung adenocarcinoma tumorigenesis
by suppressing SLC7A11-dependent antioxidant function. Redox Biol 38, 101801.
10.1016/j.redox.2020.101801.
133. Liu, L., He, J., Sun, G., Huang, N., Bian, Z., Xu, C., Zhang, Y., Cui, Z., Xu, W., Sun, F.,
et al. (2022). The N6-methyladenosine modification enhances ferroptosis resistance
through inhibiting SLC7A11 mRNA deadenylation in hepatoblastoma. Clin Transl Med
12, e778. 10.1002/ctm2.778.
134. Liu, C., Sun, H., Yi, Y., Shen, W., Li, K., Xiao, Y., Li, F., Li, Y., Hou, Y., Lu, B., et al.
(2023). Absolute quantification of single-base m 6 A methylation in the mammalian
transcriptome using GLORI. Nat Biotechnol 41, 355-366. 10.1038/s41587-022-01487-9.
135. Meyer, K.D., Patil, D.P., Zhou, J., Zinoviev, A., Skabkin, M.A., Elemento, O., Pestova,
T.V., Qian, S.B., and Jaffrey, S.R. (2015). 5' UTR m(6)A Promotes Cap-Independent
Translation. Cell 163, 999-1010. 10.1016/j.cell.2015.10.012.
136. Snow, O., Lallous, N., Singh, K., Lack, N., Rennie, P., and Cherkasov, A. (2019).
Androgen receptor plasticity and its implications for prostate cancer therapy. Cancer
Treat Rev 81, 101871. 10.1016/j.ctrv.2019.05.001.
137. Wadosky, K.M., and Koochekpour, S. (2016). Molecular mechanisms underlying
resistance to androgen deprivation therapy in prostate cancer. Oncotarget 7, 64447-
64470. 10.18632/oncotarget.10901.
138. Wadosky, K.M., and Koochekpour, S. (2017). Androgen receptor splice variants and
prostate cancer: From bench to bedside. Oncotarget 8, 18550-18576.
10.18632/oncotarget.14537.
139. Nakazawa, M., Antonarakis, E.S., and Luo, J. (2014). Androgen receptor splice variants
in the era of enzalutamide and abiraterone. Horm Cancer 5, 265-273. 10.1007/s12672-
014-0190-1.
140. Guo, Z., Yang, X., Sun, F., Jiang, R., Linn, D.E., Chen, H., Kong, X., Melamed, J.,
Tepper, C.G., Kung, H.J., et al. (2009). A novel androgen receptor splice variant is upregulated during prostate cancer progression and promotes androgen depletion-resistant
growth. Cancer Res 69, 2305-2313. 10.1158/0008-5472.CAN-08-3795.

139
141. Sharp, A., Coleman, I., Yuan, W., Sprenger, C., Dolling, D., Rodrigues, D.N., Russo,
J.W., Figueiredo, I., Bertan, C., Seed, G., et al. (2019). Androgen receptor splice variant7 expression emerges with castration resistance in prostate cancer. J Clin Invest 129,
192-208. 10.1172/JCI122819.
142. Stone, L. (2017). Prostate cancer: Escaping enzalutamide: Malat1 contributes to
resistance. Nat Rev Urol 14, 450. 10.1038/nrurol.2017.91.
143. Wang, R., Sun, Y., Li, L., Niu, Y., Lin, W., Lin, C., Antonarakis, E.S., Luo, J., Yeh, S.,
and Chang, C. (2017). Preclinical Study using Malat1 Small Interfering RNA or Androgen
Receptor Splicing Variant 7 Degradation Enhancer ASC-J9. Eur Urol 72, 835-844.
10.1016/j.eururo.2017.04.005.
144. Chang, J., Xu, W., Du, X., and Hou, J. (2018). MALAT1 silencing suppresses prostate
cancer progression by upregulating miR-1 and downregulating KRAS. Onco Targets
Ther 11, 3461-3473. 10.2147/OTT.S164131.
145. Dai, X., Liu, L., Liang, Z., Guo, K., Xu, S., and Wang, H. (2019). Silencing of lncRNA
MALAT1 inhibits cell cycle progression via androgen receptor signaling in prostate
cancer cells. Pathol Res Pract 215, 712-721. 10.1016/j.prp.2019.01.011.
146. Tripathi, V., Ellis, J.D., Shen, Z., Song, D.Y., Pan, Q., Watt, A.T., Freier, S.M., Bennett,
C.F., Sharma, A., Bubulya, P.A., et al. (2010). The nuclear-retained noncoding RNA
MALAT1 regulates alternative splicing by modulating SR splicing factor phosphorylation.
Mol Cell 39, 925-938. 10.1016/j.molcel.2010.08.011.
147. Coker, H., Wei, G., and Brockdorff, N. (2019). m6A modification of non-coding RNA and
the control of mammalian gene expression. Biochim Biophys Acta Gene Regul Mech
1862, 310-318. 10.1016/j.bbagrm.2018.12.002.
148. Adhikari, S., Xiao, W., Zhao, Y.L., and Yang, Y.G. (2016). m(6)A: Signaling for mRNA
splicing. RNA Biol 13, 756-759. 10.1080/15476286.2016.1201628.
149. Miyata, Y., and Sakai, H. (2013). Thrombospondin-1 in urological cancer: pathological
role, clinical significance, and therapeutic prospects. Int J Mol Sci 14, 12249-12272.
10.3390/ijms140612249.
150. Kaur, S., Bronson, S.M., Pal-Nath, D., Miller, T.W., Soto-Pantoja, D.R., and Roberts,
D.D. (2021). Functions of Thrombospondin-1 in the Tumor Microenvironment. Int J Mol
Sci 22. 10.3390/ijms22094570.
151. Zhang, Y., Zheng, D., Zhou, T., Song, H., Hulsurkar, M., Su, N., Liu, Y., Wang, Z., Shao,
L., Ittmann, M., et al. (2018). Androgen deprivation promotes neuroendocrine
differentiation and angiogenesis through CREB-EZH2-TSP1 pathway in prostate
cancers. Nat Commun 9, 4080. 10.1038/s41467-018-06177-2.
152. Firlej, V., Mathieu, J.R., Gilbert, C., Lemonnier, L., Nakhlé, J., Gallou-Kabani, C.,
Guarmit, B., Morin, A., Prevarskaya, N., Delongchamps, N.B., and Cabon, F. (2011).
Thrombospondin-1 triggers cell migration and development of advanced prostate
tumors. Cancer Res 71, 7649-7658. 10.1158/0008-5472.CAN-11-0833.

140
153. Hodara, E., Morrison, G., Cunha, A., Zainfeld, D., Xu, T., Xu, Y., Dempsey, P.W.,
Pagano, P.C., Bischoff, F., Khurana, A., et al. (2019). Multiparametric liquid biopsy
analysis in metastatic prostate cancer. JCI Insight 4. 10.1172/jci.insight.125529.
154. Siegel, R.L., Miller, K.D., and Jemal, A. (2017). Cancer Statistics, 2017. CA Cancer J
Clin 67, 7-30. 10.3322/caac.21387.
155. Network, C.G.A.R. (2015). The Molecular Taxonomy of Primary Prostate Cancer. Cell
163, 1011-1025. 10.1016/j.cell.2015.10.025.
156. Cooper, C.S., Eeles, R., Wedge, D.C., Van Loo, P., Gundem, G., Alexandrov, L.B.,
Kremeyer, B., Butler, A., Lynch, A.G., Camacho, N., et al. (2015). Analysis of the genetic
phylogeny of multifocal prostate cancer identifies multiple independent clonal
expansions in neoplastic and morphologically normal prostate tissue. Nat Genet 47, 367-
372. 10.1038/ng.3221.
157. Robinson, D., Van Allen, E.M., Wu, Y.M., Schultz, N., Lonigro, R.J., Mosquera, J.M.,
Montgomery, B., Taplin, M.E., Pritchard, C.C., Attard, G., et al. (2015). Integrative
Clinical Genomics of Advanced Prostate Cancer. Cell 162, 454.
10.1016/j.cell.2015.06.053.
158. Gundem, G., Van Loo, P., Kremeyer, B., Alexandrov, L.B., Tubio, J.M.C.,
Papaemmanuil, E., Brewer, D.S., Kallio, H.M.L., Högnäs, G., Annala, M., et al. (2015).
The evolutionary history of lethal metastatic prostate cancer. Nature 520, 353-357.
10.1038/nature14347.
159. Boutros, P.C., Fraser, M., Harding, N.J., de Borja, R., Trudel, D., Lalonde, E., Meng, A.,
Hennings-Yeomans, P.H., McPherson, A., Sabelnykova, V.Y., et al. (2015). Spatial
genomic heterogeneity within localized, multifocal prostate cancer. Nat Genet 47, 736-
745. 10.1038/ng.3315.
160. Morrison, G.J., and Goldkorn, A. (2018). Development and Application of Liquid Biopsies
in Metastatic Prostate Cancer. Curr Oncol Rep 20, 35. 10.1007/s11912-018-0683-0.
161. Kumar, A., Coleman, I., Morrissey, C., Zhang, X., True, L.D., Gulati, R., Etzioni, R.,
Bolouri, H., Montgomery, B., White, T., et al. (2016). Substantial interindividual and
limited intraindividual genomic diversity among tumors from men with metastatic prostate
cancer. Nat Med 22, 369-378. 10.1038/nm.4053.
162. Danila, D.C., Anand, A., Schultz, N., Heller, G., Wan, M., Sung, C.C., Dai, C., Khanin,
R., Fleisher, M., Lilja, H., and Scher, H.I. (2014). Analytic and clinical validation of a
prostate cancer-enhanced messenger RNA detection assay in whole blood as a
prognostic biomarker for survival. Eur Urol 65, 1191-1197.
10.1016/j.eururo.2013.07.006.
163. Ku, S.Y., Rosario, S., Wang, Y., Mu, P., Seshadri, M., Goodrich, Z.W., Goodrich, M.M.,
Labbé, D.P., Gomez, E.C., Wang, J., et al. (2017). Rb1 and Trp53 cooperate to
suppress prostate cancer lineage plasticity, metastasis, and antiandrogen resistance.
Science 355, 78-83. 10.1126/science.aah4199.

141
164. Cuzick, J., Swanson, G.P., Fisher, G., Brothman, A.R., Berney, D.M., Reid, J.E.,
Mesher, D., Speights, V.O., Stankiewicz, E., Foster, C.S., et al. (2011). Prognostic value
of an RNA expression signature derived from cell cycle proliferation genes in patients
with prostate cancer: a retrospective study. Lancet Oncol 12, 245-255. 10.1016/S1470-
2045(10)70295-3.
165. Scher, H.I., Heller, G., Molina, A., Attard, G., Danila, D.C., Jia, X., Peng, W., Sandhu,
S.K., Olmos, D., Riisnaes, R., et al. (2015). Circulating tumor cell biomarker panel as an
individual-level surrogate for survival in metastatic castration-resistant prostate cancer. J
Clin Oncol 33, 1348-1355. 10.1200/JCO.2014.55.3487.
166. de Bono, J.S., Scher, H.I., Montgomery, R.B., Parker, C., Miller, M.C., Tissing, H., Doyle,
G.V., Terstappen, L.W., Pienta, K.J., and Raghavan, D. (2008). Circulating tumor cells
predict survival benefit from treatment in metastatic castration-resistant prostate cancer.
Clin Cancer Res 14, 6302-6309. 10.1158/1078-0432.CCR-08-0872.
167. Goldkorn, A., Ely, B., Quinn, D.I., Tangen, C.M., Fink, L.M., Xu, T., Twardowski, P., Van
Veldhuizen, P.J., Agarwal, N., Carducci, M.A., et al. (2014). Circulating tumor cell counts
are prognostic of overall survival in SWOG S0421: a phase III trial of docetaxel with or
without atrasentan for metastatic castration-resistant prostate cancer. J Clin Oncol 32,
1136-1142. 10.1200/JCO.2013.51.7417.
168. Ossandon, M.R., Agrawal, L., Bernhard, E.J., Conley, B.A., Dey, S.M., Divi, R.L., Guan,
P., Lively, T.G., McKee, T.C., Sorg, B.S., and Tricoli, J.V. (2018). Circulating Tumor DNA
Assays in Clinical Cancer Research. J Natl Cancer Inst 110, 929-934.
10.1093/jnci/djy105.
169. Merker, J.D., Oxnard, G.R., Compton, C., Diehn, M., Hurley, P., Lazar, A.J., Lindeman,
N., Lockwood, C.M., Rai, A.J., Schilsky, R.L., et al. (2018). Circulating Tumor DNA
Analysis in Patients With Cancer: American Society of Clinical Oncology and College of
American Pathologists Joint Review. J Clin Oncol 36, 1631-1641.
10.1200/JCO.2017.76.8671.
170. Beltran, H., Tomlins, S., Aparicio, A., Arora, V., Rickman, D., Ayala, G., Huang, J., True,
L., Gleave, M.E., Soule, H., et al. (2014). Aggressive variants of castration-resistant
prostate cancer. Clin Cancer Res 20, 2846-2850. 10.1158/1078-0432.CCR-13-3309.
171. Aparicio, A.M., Harzstark, A.L., Corn, P.G., Wen, S., Araujo, J.C., Tu, S.M., Pagliaro,
L.C., Kim, J., Millikan, R.E., Ryan, C., et al. (2013). Platinum-based chemotherapy for
variant castrate-resistant prostate cancer. Clin Cancer Res 19, 3621-3630.
10.1158/1078-0432.CCR-12-3791.
172. Scott, C.L., Swisher, E.M., and Kaufmann, S.H. (2015). Poly (ADP-ribose) polymerase
inhibitors: recent advances and future development. J Clin Oncol 33, 1397-1406.
10.1200/JCO.2014.58.8848.
173. Petrylak, D.P., Tangen, C.M., Hussain, M.H., Lara, P.N., Jones, J.A., Taplin, M.E.,
Burch, P.A., Berry, D., Moinpour, C., Kohli, M., et al. (2004). Docetaxel and estramustine
compared with mitoxantrone and prednisone for advanced refractory prostate cancer. N
Engl J Med 351, 1513-1520. 10.1056/NEJMoa041318.

142
174. Tannock, I.F., de Wit, R., Berry, W.R., Horti, J., Pluzanska, A., Chi, K.N., Oudard, S.,
Théodore, C., James, N.D., Turesson, I., et al. (2004). Docetaxel plus prednisone or
mitoxantrone plus prednisone for advanced prostate cancer. N Engl J Med 351, 1502-
1512. 10.1056/NEJMoa040720.
175. Carreira, S., Romanel, A., Goodall, J., Grist, E., Ferraldeschi, R., Miranda, S., Prandi, D.,
Lorente, D., Frenel, J.S., Pezaro, C., et al. (2014). Tumor clone dynamics in lethal
prostate cancer. Sci Transl Med 6, 254ra125. 10.1126/scitranslmed.3009448.
176. Heller, G., McCormack, R., Kheoh, T., Molina, A., Smith, M.R., Dreicer, R., Saad, F., de
Wit, R., Aftab, D.T., Hirmand, M., et al. (2018). Circulating Tumor Cell Number as a
Response Measure of Prolonged Survival for Metastatic Castration-Resistant Prostate
Cancer: A Comparison With Prostate-Specific Antigen Across Five Randomized Phase
III Clinical Trials. J Clin Oncol 36, 572-580. 10.1200/JCO.2017.75.2998.
177. Romanel, A., Gasi Tandefelt, D., Conteduca, V., Jayaram, A., Casiraghi, N., Wetterskog,
D., Salvi, S., Amadori, D., Zafeiriou, Z., Rescigno, P., et al. (2015). Plasma AR and
abiraterone-resistant prostate cancer. Sci Transl Med 7, 312re310.
10.1126/scitranslmed.aac9511.
178. Ferraldeschi, R., Welti, J., Luo, J., Attard, G., and de Bono, J.S. (2015). Targeting the
androgen receptor pathway in castration-resistant prostate cancer: progresses and
prospects. Oncogene 34, 1745-1757. 10.1038/onc.2014.115.
179. Antonarakis, E.S., Lu, C., Luber, B., Wang, H., Chen, Y., Nakazawa, M., Nadal, R.,
Paller, C.J., Denmeade, S.R., Carducci, M.A., et al. (2015). Androgen Receptor Splice
Variant 7 and Efficacy of Taxane Chemotherapy in Patients With Metastatic CastrationResistant Prostate Cancer. JAMA Oncol 1, 582-591. 10.1001/jamaoncol.2015.1341.
180. Onstenk, W., Sieuwerts, A.M., Kraan, J., Van, M., Nieuweboer, A.J., Mathijssen, R.H.,
Hamberg, P., Meulenbeld, H.J., De Laere, B., Dirix, L.Y., et al. (2015). Efficacy of
Cabazitaxel in Castration-resistant Prostate Cancer Is Independent of the Presence of
AR-V7 in Circulating Tumor Cells. Eur Urol 68, 939-945. 10.1016/j.eururo.2015.07.007.
181. Todenhöfer, T., Azad, A., Stewart, C., Gao, J., Eigl, B.J., Gleave, M.E., Joshua, A.M.,
Black, P.C., and Chi, K.N. (2017). AR-V7 Transcripts in Whole Blood RNA of Patients
with Metastatic Castration Resistant Prostate Cancer Correlate with Response to
Abiraterone Acetate. J Urol 197, 135-142. 10.1016/j.juro.2016.06.094.
182. Seitz, A.K., Thoene, S., Bietenbeck, A., Nawroth, R., Tauber, R., Thalgott, M., Schmid,
S., Secci, R., Retz, M., Gschwend, J.E., et al. (2017). AR-V7 in Peripheral Whole Blood
of Patients with Castration-resistant Prostate Cancer: Association with Treatmentspecific Outcome Under Abiraterone and Enzalutamide. Eur Urol 72, 828-834.
10.1016/j.eururo.2017.07.024.
183. Liu, X., Ledet, E., Li, D., Dotiwala, A., Steinberger, A., Feibus, A., Li, J., Qi, Y.,
Silberstein, J., Lee, B., et al. (2016). A Whole Blood Assay for AR-V7 and AR. J Urol
196, 1758-1763. 10.1016/j.juro.2016.06.095.

143
184. Bernemann, C., Steinestel, J., Boegemann, M., and Schrader, A.J. (2017). Novel AR-V7
detection in whole blood samples in patients with prostate cancer: not as simple as it
seems. World J Urol 35, 1625-1627. 10.1007/s00345-017-2024-6.
185. Bernemann, C., Steinestel, J., Humberg, V., Bögemann, M., Schrader, A.J., and
Lennerz, J.K. (2018). Performance comparison of two androgen receptor splice variant 7
(AR-V7) detection methods. BJU Int 122, 219-226. 10.1111/bju.14146.
186. Wyatt, A.W., Azad, A.A., Volik, S.V., Annala, M., Beja, K., McConeghy, B., Haegert, A.,
Warner, E.W., Mo, F., Brahmbhatt, S., et al. (2016). Genomic Alterations in Cell-Free
DNA and Enzalutamide Resistance in Castration-Resistant Prostate Cancer. JAMA
Oncol 2, 1598-1606. 10.1001/jamaoncol.2016.0494.
187. Liu, C., Lou, W., Zhu, Y., Nadiminty, N., Schwartz, C.T., Evans, C.P., and Gao, A.C.
(2014). Niclosamide inhibits androgen receptor variants expression and overcomes
enzalutamide resistance in castration-resistant prostate cancer. Clin Cancer Res 20,
3198-3210. 10.1158/1078-0432.CCR-13-3296.
188. Dobin, A., Davis, C.A., Schlesinger, F., Drenkow, J., Zaleski, C., Jha, S., Batut, P.,
Chaisson, M., and Gingeras, T.R. (2013). STAR: ultrafast universal RNA-seq aligner.
Bioinformatics 29, 15-21. 10.1093/bioinformatics/bts635.
189. Frankish, A., Diekhans, M., Jungreis, I., Lagarde, J., Loveland, J.E., Mudge, J.M., Sisu,
C., Wright, J.C., Armstrong, J., Barnes, I., et al. (2021). GENCODE 2021. Nucleic Acids
Res 49, D916-D923. 10.1093/nar/gkaa1087.
190. Li, H., Handsaker, B., Wysoker, A., Fennell, T., Ruan, J., Homer, N., Marth, G.,
Abecasis, G., Durbin, R., and Subgroup, G.P.D.P. (2009). The Sequence Alignment/Map
format and SAMtools. Bioinformatics 25, 2078-2079. 10.1093/bioinformatics/btp352.
191. Liao, Y., Smyth, G.K., and Shi, W. (2014). featureCounts: an efficient general purpose
program for assigning sequence reads to genomic features. Bioinformatics 30, 923-930.
10.1093/bioinformatics/btt656.
192. Love, M.I., Huber, W., and Anders, S. (2014). Moderated estimation of fold change and
dispersion for RNA-seq data with DESeq2. Genome Biol 15, 550. 10.1186/s13059-014-
0550-8.
193. Ashburner, M., Ball, C.A., Blake, J.A., Botstein, D., Butler, H., Cherry, J.M., Davis, A.P.,
Dolinski, K., Dwight, S.S., Eppig, J.T., et al. (2000). Gene ontology: tool for the
unification of biology. The Gene Ontology Consortium. Nat Genet 25, 25-29.
10.1038/75556.
194. Consortium, G.O. (2021). The Gene Ontology resource: enriching a GOld mine. Nucleic
Acids Res 49, D325-D334. 10.1093/nar/gkaa1113.
195. Wu, T., Hu, E., Xu, S., Chen, M., Guo, P., Dai, Z., Feng, T., Zhou, L., Tang, W., Zhan, L.,
et al. (2021). clusterProfiler 4.0: A universal enrichment tool for interpreting omics data.
Innovation (Camb) 2, 100141. 10.1016/j.xinn.2021.100141.

144
196. Zhang, Y., Liu, T., Meyer, C.A., Eeckhoute, J., Johnson, D.S., Bernstein, B.E.,
Nusbaum, C., Myers, R.M., Brown, M., Li, W., and Liu, X.S. (2008). Model-based
analysis of ChIP-Seq (MACS). Genome Biol 9, R137. 10.1186/gb-2008-9-9-r137.
197. Liu, L., Zhang, S.W., Huang, Y., and Meng, J. (2017). QNB: differential RNA methylation
analysis for count-based small-sample sequencing data with a quad-negative binomial
model. BMC Bioinformatics 18, 387. 10.1186/s12859-017-1808-4.
198. Robinson, M.D., McCarthy, D.J., and Smyth, G.K. (2010). edgeR: a Bioconductor
package for differential expression analysis of digital gene expression data.
Bioinformatics 26, 139-140. 10.1093/bioinformatics/btp616.
199. Zhao, Y., Gao, J.L., Ji, J.W., Gao, M., Yin, Q.S., Qiu, Q.L., Wang, C., Chen, S.Z., Xu, J.,
Liang, R.S., et al. (2014). Cytotoxicity enhancement in MDA-MB-231 cells by the
combination treatment of tetrahydropalmatine and berberine derived from Corydalis
yanhusuo W. T. Wang. J Intercult Ethnopharmacol 3, 68-72.
10.5455/jice.20140123040224.
200. Mullenders, J., de Jongh, E., Brousali, A., Roosen, M., Blom, J.P.A., Begthel, H.,
Korving, J., Jonges, T., Kranenburg, O., Meijer, R., and Clevers, H.C. (2019). Mouse
and human urothelial cancer organoids: A tool for bladder cancer research. Proc Natl
Acad Sci U S A 116, 4567-4574. 10.1073/pnas.1803595116.
201. Goldman, M.J., Craft, B., Hastie, M., Repečka, K., McDade, F., Kamath, A., Banerjee,
A., Luo, Y., Rogers, D., Brooks, A.N., et al. (2020). Visualizing and interpreting cancer
genomics data via the Xena platform. Nat Biotechnol 38, 675-678. 10.1038/s41587-020-
0546-8.
202. Quinlan, A.R., Clark, R.A., Sokolova, S., Leibowitz, M.L., Zhang, Y., Hurles, M.E., Mell,
J.C., and Hall, I.M. (2010). Genome-wide mapping and assembly of structural variant
breakpoints in the mouse genome. Genome Res 20, 623-635. 10.1101/gr.102970.109.
203. Anders, S., Pyl, P.T., and Huber, W. (2015). HTSeq--a Python framework to work with
high-throughput sequencing data. Bioinformatics 31, 166-169.
10.1093/bioinformatics/btu638.
204. Boeva, V., Popova, T., Bleakley, K., Chiche, P., Cappo, J., Schleiermacher, G.,
Janoueix-Lerosey, I., Delattre, O., and Barillot, E. (2012). Control-FREEC: a tool for
assessing copy number and allelic content using next-generation sequencing data.
Bioinformatics 28, 423-425. 10.1093/bioinformatics/btr670.
205. Winer-Jones, J.P., Vahidi, B., Arquilevich, N., Fang, C., Ferguson, S., Harkins, D., Hill,
C., Klem, E., Pagano, P.C., Peasley, C., et al. (2014). Circulating tumor cells: clinically
relevant molecular access based on a novel CTC flow cell. PLoS One 9, e86717.
10.1371/journal.pone.0086717.
206. Strauss, W.M., Carter, C., Simmons, J., Klem, E., Goodman, N., Vahidi, B., Romero, J.,
Masterman-Smith, M., O'Regan, R., Gogineni, K., et al. (2016). Analysis of tumor
template from multiple compartments in a blood sample provides complementary access
to peripheral tumor biomarkers. Oncotarget 7, 26724-26738. 10.18632/oncotarget.8494.

145
207. Ishiba, T., Hoffmann, A.C., Usher, J., Elshimali, Y., Sturdevant, T., Dang, M., Jaimes, Y.,
Tyagi, R., Gonzales, R., Grino, M., et al. (2018). Frequencies and expression levels of
programmed death ligand 1 (PD-L1) in circulating tumor RNA (ctRNA) in various cancer
types. Biochem Biophys Res Commun 500, 621-625. 10.1016/j.bbrc.2018.04.120.
208. Liu, X.M., Zhou, J., Mao, Y., Ji, Q., and Qian, S.B. (2019). Programmable RNA N. Nat
Chem Biol 15, 865-871. 10.1038/s41589-019-0327-1.
209. Rauch, S., He, E., Srienc, M., Zhou, H., Zhang, Z., and Dickinson, B.C. (2019).
Programmable RNA-Guided RNA Effector Proteins Built from Human Parts. Cell 178,
122-134.e112. 10.1016/j.cell.2019.05.049.
210. Morrison, G., Buckley, J., Ostrow, D., Varghese, B., Cen, S.Y., Werbin, J., Ericson, N.,
Cunha, A., Lu, Y.T., George, T., et al. (2022). Non-Invasive Profiling of Advanced
Prostate Cancer via Multi-Parametric Liquid Biopsy and Radiomic Analysis. Int J Mol Sci
23. 10.3390/ijms23052571.

Abstract (if available)

Abstract Cancer drug resistance is recognized to occur not only through selection of pre-existing genetically resistant clones, but also through rapid induction of transcriptional programs that allow some cells to adapt and persist. This phenotypic plasticity mechanism is exemplified by models of emerging drug resistance in two genitourinary malignancies, bladder cancer and prostate cancer. In bladder cancer (BC), we previously reported that bladder cancer cells can rapidly transition to and from a chemo-resistant phenotype. In prostate cancer (PC), there is a well-documented emergence of ARV7, an alternative splice variant of the Androgen Receptor (AR) implicated in enzalutamide resistance. In both malignancies, one potential contributing factor to phenotypic plasticity is N6-methyladenosine (m6A) RNA modification. Deposited by “m6A writers” and removed by “m6A erasers”, m6A reversibly regulates key cellular processes, including cell fate, differentiation and mRNA processing. Given the dynamic nature of m6A and its function in regulating alternative splicing, I hypothesized that m6A RNA modifications play a role in phenotypic plasticity and transition to drug resistance in established BC and PC models. To test this, I used methyl-RNA-immunoprecipitation followed by sequencing (MeRIP-seq) in parallel with RNA-seq to identify gene transcripts that were both differentially methylated and differentially expressed between drug-sensitive and drug-resistant cancer cells. In chapter 2, I test our hypothesis in a bladder cancer model of phenotypic plasticity established in our laboratory, in which cells interconvert cyclically in and out of a drug-resistant tumorigenic state, but I was unable to identify differential m6A sites due to the significant heterogeneity of cells in the model. In chapter 3, I repeat our analysis in another well-established cell-line based BC model of chemoresistance, and found that cisplatin-sensitive and cisplatin-resistant cells have distinct m6A profiles with 130 transcripts that are both differentially methylated and differentially expressed. I filtered and prioritized these genes using clinical and functional database tools, then validated several of the top candidates via targeted qPCR and MeRIP-PCR. In cisplatin-resistant cells, SLC7A11 transcripts had decreased methylation associated with decreased m6A reader YTHDF3 binding, prolonged RNA stability, and increased RNA and protein levels, leading to reduced ferroptosis and increased survival. Consistent with this, cisplatin-sensitive BC cell lines and patient-derived organoids (PDOs) exposed to cisplatin for as little as 48 hours exhibited similar mechanisms of SLC7A11 upregulation and chemoresistance, trends that were also reflected in public cancer survival databases. In chapter 4, I repeat this analysis in a well-established prostate cancer model of ARV7-mediated enzalutamide resistance and found that enzalutamide-sensitive and enzalutamide-resistant cells have distinct m6A profiles with 46 transcripts that are both differentially methylated and differentially expressed. After filtering and validation, one transcript THBS1 had increased methylation and decreased expression in resistant cells. Depletion of THBS1 by siRNA knockdown potentiated drug resistance in enzalutamide-sensitive PC cell lines. Finally in chapter 5, I present a pilot study pioneering a multiparametric approach of liquid biopsy profiling of prostate cancer patient samples, integrating multiple blood-based tumor phenotypes to yield a maximally informative disease profile including putative genomic and transcriptomic markers of drug resistance. Collectively, these findings highlight epitranscriptomic plasticity as a mechanism of drug resistance and a potential therapeutic target in both bladder and prostate cancer.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Temporal and spatial characterization of cisplatin treatment and emerging acute resistance in bladder cancer cells

PDF

Cisplatin activates mitochondrial oxphos leading to acute treatment resistance in bladder cancer

PDF

Investigating the effects of polycomb repressive complex inhibitors on cancer cell phenotypic plasticity

PDF

Applying multi-omics in cancer liquid biopsy for improved patient monitoring and biomarker discovery

PDF

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

PDF

Evaluation of preservatives in blood collection tubes for cell-free RNA transcriptional profiles in human plasma

PDF

CpG methylation profiling in lung cancer cell lines, tumors and non-tumors

PDF

Identification of novel epigenetic biomarkers and microRNAs for cancer therapeutics

PDF

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

PDF

Exploring the effects of CXCR4 inhibition on circulating tumor cell populations in metastatic prostate cancer

PDF

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

PDF

Functional DNA methylation changes in normal and cancer cells

PDF

Mass spectrometry-based proteomic analysis of inhibitor of kappa b kinase beta and its role in cytokine-induced drug resistance in cancer

PDF

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

PDF

Application of tracing enhancer networks using epigenetic traits (TENET) to identify epigenetic deregulation in cancer

PDF

Molecular signature of aggressive disease and clonal diversity revealed by single-cell copy number analysis of prostate cancer cells across multiple disease states

PDF

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

PDF

The role of GRP78 in the regulation of apoptosis and prostate cancer progression

PDF

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

PDF

Role of integrin α4 in drug resistant acute lymphoblastic leukemia

Asset Metadata

Creator Hodara, Emmanuelle (author)

Core Title RNA methylation in cancer plasticity and drug resistance

School Keck School of Medicine

Degree Doctor of Philosophy

Degree Program Cancer Biology and Genomics

Degree Conferral Date 2023-12

Publication Date 05/30/2024

Defense Date 10/31/2023

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

Tag bladder cancer,circulating tumor cells,cisplatin,drug resistance,Enzalutamide,liquid biopsy,m⁶A,MeRIP-seq,OAI-PMH Harvest,prostate cancer,ran methylation,SLC7A11

Format theses (aat)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Offringa, Ite (committee chair), Farnham, Peggy (committee member), Goldkorn, Amir (committee member), Rhie, Suhn (committee member)

Creator Email ehodara@usc.edu,emmanuelle.hodara@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC113781070

Unique identifier UC113781070

Identifier etd-HodaraEmma-12510.pdf (filename)

Legacy Identifier etd-HodaraEmma-12510

Document Type Dissertation

Format theses (aat)

Rights Hodara, Emmanuelle

Internet Media Type application/pdf

Type texts

Source 20231201-usctheses-batch-1110 (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.

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